145 58 4MB
English Pages 288 [282] Year 2022
Synthese Library 451 Studies in Epistemology, Logic, Methodology, and Philosophy of Science
Christopher J. Austin Anna Marmodoro Andrea Roselli Editors
Powers, Time and Free Will
Synthese Library Studies in Epistemology, Logic, Methodology, and Philosophy of Science Volume 451
Editor-in-Chief Otávio Bueno, Department of Philosophy, University of Miami, Coral Gables, USA Editorial Board Members Berit Brogaard, University of Miami, Coral Gables, USA Anjan Chakravartty, Department of Philosophy, University of Miami, Coral Gables, USA Steven French, University of Leeds, Leeds, UK Catarina Dutilh Novaes, VU Amsterdam, Amsterdam, The Netherlands Darrell P. Rowbottom, Department of Philosophy, Lingnan University, Tuen Mun, Hong Kong Emma Ruttkamp, Department of Philosophy, University of South Africa, Pretoria, South Africa Kristie Miller, Department of Philosophy, Centre for Time, University of Sydney, Sydney, Australia
The aim of Synthese Library is to provide a forum for the best current work in the methodology and philosophy of science and in epistemology, all broadly understood. A wide variety of different approaches have traditionally been represented in the Library, and every effort is made to maintain this variety, not for its own sake, but because we believe that there are many fruitful and illuminating approaches to the philosophy of science and related disciplines. Special attention is paid to methodological studies which illustrate the interplay of empirical and philosophical viewpoints and to contributions to the formal (logical, set-theoretical, mathematical, information-theoretical, decision-theoretical, etc.) methodology of empirical sciences. Likewise, the applications of logical methods to epistemology as well as philosophically and methodologically relevant studies in logic are strongly encouraged. The emphasis on logic will be tempered by interest in the psychological, historical, and sociological aspects of science. In addition to monographs Synthese Library publishes thematically unified anthologies and edited volumes with a well-defined topical focus inside the aim and scope of the book series. The contributions in the volumes are expected to be focused and structurally organized in accordance with the central theme(s), and should be tied together by an extensive editorial introduction or set of introductions if the volume is divided into parts. An extensive bibliography and index are mandatory.
More information about this series at https://link.springer.com/bookseries/6607
Christopher J. Austin • Anna Marmodoro Andrea Roselli Editors
Powers, Time and Free Will
Editors Christopher J. Austin Department of Philosophy University of Reading Reading, UK Andrea Roselli Department of Philosophy Durham University Oxford, UK
Anna Marmodoro Faculty of Philosophy Oxford University Oxford, UK Department of Philosophy Durham University Durham, UK
ISSN 0166-6991 ISSN 2542-8292 (electronic) Synthese Library ISBN 978-3-030-92485-0 ISBN 978-3-030-92486-7 (eBook) https://doi.org/10.1007/978-3-030-92486-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 Chapters [2] and [4] are licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/). For further details see licence information in the chapters. This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Acknowledgements
The research and editorial work carried out by the co-editors toward this publication was supported by the Leverhulme Trust as part of the project Part-whole relations within the fundamental potentialities in nature (RPG-2018-079).
v
Introduction
Causal powers fuel change, and change fuels time; causal powers also fuel the will, its freedom or its determination. This volume explores foundational questions around these topics, in the context of a powers ontology conception of the world. If powers are dynamic, what does their nature entail about time, and how can we conceive of the ‘dynamic’ along its temporal dimension? Further, if powers are dynamic, what does their nature entail about free will, and how can we conceive of the ‘dynamic’ along its normative dimension – especially in a world of natural laws? Various positions are explored, supported, criticised, in this volume, a part of a debate about what is fundamental; proposals are constructed to contribute the debate, and offer explanations that take it a step beyond. In the opening chapter, Anna Marmodoro argues that accounting for causation in terms of the exercise of causal powers introduces a challenge or even a limit to our understanding of the physical world; and yet, not a reason to dismiss tout court a power-based account of causation. Marmodoro identifies an irreducible element in the nature of powers, namely their dynamism – what makes powers dynamic and able to account for change in the physical world. She argues that on account of this irreducible element, powers are a black box to us, which we cannot define, despite multiple attempts having been made in the literature. What can and has been identified in the sciences is how powers behave. Marmodoro shows that this does not tell us what a power is, but it does allow us to classify powers teleologically, in terms of their manifestations; and this is of scientific utility. In the following chapter, William Simpson and Simon Horsley challenge the wide-spread idea that contemporary metaphysicians should endorse microphysicalism (the idea that the fundamental constituents of the world are very small, such as particles), and defend a pluralist perspective, according to which there could be fundamental objects at different scales. The authors present some case studies to challenge the idea that contemporary physics supports the claim that the causal powers of a system supervene upon the ‘lower-level’ laws (‘physical state monism’); on the contrary, they argue, quantum physics admits ontological readings
vii
viii
Introduction
in which there are fundamental objects both at the microscopic and at the cosmic scale, readings that are then compatible with a pluralist perspective. Simpson and Horsley claim that there are macroscopic objects that have ‘higher-level’ physical properties that make a causal difference to the temporal development of their lowerlevel physical properties: the causal powers of many systems, for example, are determined by higher-level macroscopic properties, that are neither reducible nor weakly emergent. The authors, therefore, deny the reducibility of these higher-level powers to a lower-level substrate described by our best physical theory, and claim that contemporary physics is compatible with ontological pluralism, which affirms that these entities are robustly real; physics, they claim, is compatible with a strong emergence that admits irreducible, higher-level causal powers. Finally, the authors consider the role of different representations in explaining some results of quantum physics, concluding that our best physical theories do not bind us to a microphysical structure which evolves according to one set of lower-level laws. In the third chapter, Andrea Roselli focuses on what happens when a power-based ontology is combined with the most discussed metaphysics of time (Eternalism, the Growing Block view, Presentism). While there are some philosophers who have explicitly supported a combination of Eternalism and dispositional essentialism, others have more recently claimed that powers ontologies may not be compatible with it. He argues, against Backmann (2018), Friebe (2017) and Donati (2018), who have recently claimed that dispositional essentialism cannot be combined with ‘static’ views of time, that a (full-fledged) metaphysics of powers is compatible with Eternalism. Roselli presents and discusses other objections moved in general to the combination of powers ontologies and the different metaphysics of time, and argues that these objections fail, and that the alleged incompatibility results merely from a misconception of the staticity of Eternalism, on the one hand, and of the productiveness of powers on the other. In particular, while the author agrees that some power-based ontologies go in the direction of a robust notion of production (powers producing their manifestations in the strongest possible ontological sense: making them coming into being in a universe where there is an objective present advancing), he points out how this notion is connoted precisely in presentist terms, and it is therefore trivial that it cannot be accommodated in an eternalist world. Finally, the author shows how, if we accept this presentist notion of production, it is actually impossible for powers to exist in every metaphysics of time: if only some things in the history of the universe are ‘now’, and only those that are objectively ‘now’ are real, what does it mean for something real to have been produced? Manifestations would not be manifestations of (substantial productive) powers, because these powers would simply not exist anymore. Roselli’s conclusion is that a robust notion of production is simply untenable for powers ontologies. The following chapter analyses from a different perspective – but with similar conclusions – this intersection between the metaphysics of powers and the metaphysics of time: Robert Koons puts forward a dynamic ‘B’ theory of time, and he argues that an Aristotelian powers ontology provides the best basis for libertarian free will, and that such an ontology is in fact compatible with the ‘B’ theory of time. In order for this compatibility to work, he puts forward an irreducible notion of
Introduction
ix
‘event-relative potentiality’, defined in terms of the existence of appropriate causal powers. The author sketches a ‘dynamic B Theory’ in which powers and change are metaphysically prior to time; he then argues that, on the other hand, when Presentism is combined with an Aristotelian powers ontology, it faces the challenge of metaphysically fundamental and temporally extended processes, which cannot persist by ‘wholly existing’ at each instant of time. This, Koons argues, strengthens the dynamic B Theory; Koons’s conclusion is that it is possible to construct a dynamic B Theory of time that is compatible with an Aristotelian ontology of causal powers and with free will. In the fifth chapter, Ben Page claims, contra many philosophers arguing against the combination of Eternalism a libertarian freedom, that this combination is indeed possible. This chapter focuses in on what he sees as the woeful lack of interaction between the philosophical literature on time and that of free will – something he believes has substantially hindered philosophical progress. There is not, he argues, a straightforward incompatibility between Eternalism and Libertarianism. Page identifies the worry that Libertarians have not with the standard modality-based culprit – the Principle of Alternate Possibilities – but with what he calls ‘fixity’, a conceptual consequence of what Baron and Miller call ‘entity everywhenism’: the fact that Eternalism entails that entities which exist now, exist in the past, and exist in the future exist simpliciter. In the chapter, Page aims to show that Libertarian freedom can be had in a universe where fixity holds. In one avenue of exploration, Page considers an interesting thought experiment which questions whether the theory of special relativity allows some Libertarian leeway in a ‘entity everywhenism’ universe. In another, Page addresses the concern of whether Libertarian free will requires a kind of causative power that is incompatible with eternalism. Distinguishing between event-causal and agent-causal forms of Libertarian acts of will, Page argues that the latter can take advantage of a powers-based metaphysics of causation to secure freedom, even in an Eternalist universe. In the sixth chapter, Janice Chik examines the temporal structure of agency from an Aristotelian perspective. Using Aristotle’s account of action as kinesis – an intrinsically temporal process which is constructed from sub-action phases of a single movement – Chik attempts to defend the Causal Theory of Action (CTA) from recent critiques. According to the CTA, true agential action requires that the agent possess mental items that mediate or constitute that agent’s reasons for performing that action. However, as Chik makes clear, the perineal problem of ‘deviant causal chains’ leads to a worry that the CTA might “alienate” agents from their own actions in a significant respect: such problems appear to show that it is possible to account for an action wholly in terms of simple event causation, leaving the relevant agent completely out of the explanans. The chapter discusses the particular problem of ‘disappearing agents’ as proposed by Velleman (1992), Hornsby (2011), and Lavin (2012). Although each of these authors highlights the importance of mediation as a key area of critique of CTA, Chik focuses on Lavin, who frames the worry with CTA’s notion of agent-control as primarily an issue with the temporality of action as a kind of change over time. By utilising an Aristotelian insight, Chik addresses Lavin’s critique and proposes a novel variation on CTA by
x
Introduction
proposing that the temporal phases of an action are end-directed because of agential control. In the following chapter, Erasmus Mayr utilises the concept of ‘rational abilities’ to defend a position of Compatibilism about free will. Rational abilities, Mayr claims, bring a number of advantages for the Compatibilist to strengthen her position which stem from properly distinguishing mutually irreducible levels of explanation. Mayr proposes adopting an ontology of powers to capture the metaphysics of abilities – in particular, abilities for ‘rational guidance’, agential abilities which guide purposive action via decision and deliberation. Using List’s (2014) agential possibilities account of Compatibilism wherein the freedom of will depends upon psychological freedom rather than physical freedom, Mayr notes that due to the multiple realisability of psychological states qua supervenient states, on this account, physical determinism is no threat to an agent’s freedom. However, even if agential control is a higher-level phenomenon, Mayr offers a case study from biochemistry to illustrate that explanation of the relevant action is not always exclusively at that level. The only way in which one might effectively disregard the microphysical states of an agent (i.e. her neurological states) with respect to causatively determining her actions, Mayr claims, is by adopting an account of rational guidance that makes use of a powers-based notion of ability. An agent’s ability for rational guidance is often indeterministic – which effectively uncouples it from the shackles of its microphysical states – and, more importantly, offers a respect in which the agent “could have done otherwise” which is not merely illusory. In the eighth chapter, Daniel De Haan focuses on how to re-think the problem of free will without the laws of nature. Are we free to break the laws of nature? Many philosophers, De Haan argues, defend conceptions of free will that have to deal with inviolable laws of nature, and the debate seems to remain bound to a nomological definition of the free will problem. Given the fact that the ‘causal powers revolution’ has, in a sense, overthrown the laws of nature, the author argues that there might be a way here to rethink the problem in terms of the powers metaphysics. For exponents of power realism nomological antirealism (¼PRNA), the author argues, the basic assumptions of the standard approach to free will are problematic, as there are no laws of nature for free or other agents to break. How can we re-think the problem of free will without the laws of nature, then? De Haan shows how an Aristotelian approach starts with the intentional and voluntary actions that put the question of powers like free will on the table in the first place. The author focuses on the problems of action itself, which, he claims, puts us on the right path to understanding the power of free will in a world without the laws of nature. In the ninth chapter of this book, Helen Steward deals with a closely related topic, i.e. the fact that a libertarian view of free will is committed to the idea that free will requires going somehow ‘beyond’ the reach of natural law. Laws of nature restrict the possibilities: according to the libertarian, Steward claims, laws are constrainers. When they hold in a domain, nothing can happen that contravenes them – they are universal and absolute. However, she argues that even the totality of all the laws leaves much unsettled. Steward then turns to the resources we can find in Aristotelian metaphysics, with powers that can explain how the resolution of possibility into
Introduction
xi
actuality is effected. On the Aristotelian view, the actions and interactions of the many different things which our world contains can explain what actually happens, while a Humean metaphysics contains powerless things that can only produce other particular facts if bound by the laws of nature. The key to arguing for Law Compatibility via this route, in her view, therefore involves ditching the Humean metaphysics which entails that there is nothing but laws to serve as the engines of reality. Her conclusion, then, is that if laws are all we have available to explain how the world evolves through time, then natural necessitation and explanation become the same thing. In the following chapter, Andreas Hüttemann deals with another closely related topic by utilising a neo-Aristotelian metaphysics. According to many relevant accounts of our moral practice whether or not we are able to do something is crucial to understanding whether we are responsible or excusable for having done that particular thing. In his chapter, Hüttemann shows how our explanations of an agent’s moral practice differ according to whether accepts a Humean account of laws in which there are no necessary relations between distinct events. Hüttemann starts by closely examining the premises of the Consequence Argument in an attempt to articulate the tension between free will and deterministic laws. He argues that a central premise of the argument which entails that the laws of nature are out of our control is more contentious than it might first appear. Whether one accepts that premise depends, Hüttemann argues, on a crucial philosophical choice: the choice between a Humean or non-Humean metaphysics. Hüttemann claims that the consequence argument is primarily a challenge for those who accept a non-Humean conception of laws of nature while Humeans possess the conceptual resources to easily avoid the challenging implications of the argument. However, the feature that allows Humeans to bypass the Consequence Argument in fact makes them vulnerable to another problem, that of ‘radical freedom’ – namely, Humean laws of nature fail to significantly constrain what we can do. Hüttemann’s conclusion is that Humeans cannot bypass the Consequence Argument and also solve the problem of radical freedom. Importantly however, Hüttemann also argues that non-Humeans face a similar conundrum: they can avoid the problem of radical freedom, but in doing so they have no easy way out of the problematic implications of the Consequence Argument. Therefore, while it is true that one’s conception of the metaphysics laws of nature can make a significant difference in the debate about free will, it is also true that both options are problematic in their own way, and there is no clear way out of this situation for either conception on the horizon. In the eleventh chapter of this book, Mario De Caro and Sofia Bonicalzi address the famous neurobiological experiments of Benjamin Libet and attempt to discern their import in the philosophical debate on the existence of free will. Libet’s work, De Caro and Bonicalzi point out, has often been understood as providing empirical evidence that free will is illusory: these experiments appear to show that agents’ choices originate not from independent acts of will, but from prior unconscious antecedents – e.g. neuronal activity characterised by what is termed ‘readiness potential’. De Caro and Bonicalzi make a careful study of the sorts of ‘Libet actions’ (LA) that are at the centre of the discussion about the experimental data. While the
xii
Introduction
authors agree that LA represent intentional and voluntary action, the caution that LA are only one sort of such actions, ones that might be captured by more discerning experimental set-ups. In general, De Caro and Bonicalzi argue that conceptual frameworks from philosophical scholarship should be integrated in to these set-ups: doing so will allow the investigation in to the neural correlates of intentional actions to be more precise. One central issue on this front that the authors address is the debate on whether the experimental findings involved in LA tell against conscious mental states being causally efficacious. ‘Readiness potentials’ being temporally antecedent to, and themselves richly causal with respect to purported acts of free will, De Caro and Bonicalzi point out, are contentious subjects in the neuroscience literature. The authors’ complex discussion of these subjects leads them to conclude that while it is true that the available experimental findings, in themselves, have not directly addressed the problem of whether free will is illusory or not, they should be taken as valuable assets within a more comprehensive theory of human action. In the following chapter, Mauro Dorato utilises an ontology of dispositions to take a fresh perspective on answering a well-established quandary in the debate between compatibilists and libertarians – the Consequence Argument (van Inwagen 1983). There have been many proposed answers to the Consequence Argument in the philosophical literature, and Dorato discusses two important ones in this chapter. Firstly, the ‘Humean compatibilist’ solution. Inspired by Lewis’ (1973) analysis of Laws as global patterns which supervene on local matters of fact amalgamated in a “best system” (one whose explanatory prowess strikes the correct balance between simplicity and strength), these solutions claim that the character and content of Laws is in an important sense “up to us” and further that they, in and of themselves, have no oomph of the sort that natural necessity seems to entail: we “make” the Laws, and those Laws cannot make anything happen. Dorato argues that, aside from the ‘best system analysis’ of Laws being implausible, it offers only a counterfactual form of freedom that does not deliver what advocates of free will want – a form of agent freedom that satisfies the ‘Principle of Alternate Possibilities’. Secondly, Dorato examines a solution which adopts an indeterministic metaphysics. Whether we find indeterminism in the quantum nature of the universe or the fundamentally chaotic structure of micro-physics, Dorato argues that without reasons qua causes, an agent’s freedom is still highly at risk. After discussing a number of nuanced ways in which one might solve the dilemma that the Consequence argument poses based on distinct notions of determinism (syntactic, epistemic, and ontic), Dorato proposes a tentative solution to the central worry. By utilising an ontology of dispositions, Dorato suggests that ‘second-order’ dispositions of the sort that control or otherwise constrain normal, law-like ones, offer a possible way out: these dispositions allow genuine agential freedom to function as a kind of tertium quid between compatibilism and incompatibilism. In the final chapter of this book, Michael Esfeld presents a perspective that is altogether unique in this volume: a world without powers. Esfeld adopts a ‘SuperHumean’ approach to endorse what he refers to as a ‘scientific realist’ ontology. Super-Humeanism is an ontology that is extremely sparse and minimalistic – even
Introduction
xiii
more so than the neo-Humeanism that Lewis (1986) championed. Whereas Lewis allowed natural, intrinsic properties instantiated at space-time points to be the foundation of his ontology, Super-Humeanism of the sort Esfeld endorses only point-like particles and the distance relations between them: everything else in the familiar ‘Humean mosaic’ supervenes upon these. Esfeld sees the sparse ontology of Super-Humeanism as not only capable of maintaining scientific realism, but also useful for allowing us to jettison the “heavy metaphysical baggage” that comes along with ontologies which countenance some form of primitive modality – principally powers, or dispositions. On Super-Humeanism, Esfeld claims, although the universe is fundamentally deterministic, agential free will can still be causative: our free choices cannot influence the past, but they can help determine what the Laws are, given the assumption of a ‘Best Systems’ analysis of Laws. Mental events – acts of will in particular – are not typically included in the ‘scientific image’ of the world, Esfeld admits, but the Super-Humean’s claim is only that the laws of physics and/or the initial values of the physical magnitudes that determine those laws have mental events as difference-makers: changes in the latter are counterfactually correlated with changes in the former. From the Humean perspective, this is causation. And although Esfeld admits that on his account the causal activity of mental events have a kind of global metaphysical role, his claim is that this is less counterintuitive than it at first seems, and that his system’s main benefit – of accommodating our roughly Libertarian conception of free will in an ontology that sufficiently captures the scientific image of the world – outweighs any misgivings one might have about the seemingly strange and simple desert landscape of his Humean picture.
University of Reading, Reading, UK Oxford University, Oxford, UK Durham University, Oxford, UK
Christopher J. Austin Anna Marmodoro Andrea Roselli
Contents
1
What’s Dynamic About Causal Powers? A Black Box! . . . . . . . . . . Anna Marmodoro
1
2
Toppling the Pyramids: Physics Without Physical State Monism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . William M. R. Simpson and Simon A. R. Horsley
17
3
Dispositional Essentialism in the Eternalist Block . . . . . . . . . . . . . . Andrea Roselli
51
4
A Dynamic B Theory of Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert C. Koons
69
5
Libertarian Freedom in an Eternalist World? . . . . . . . . . . . . . . . . . Ben Page
83
6
The Temporal Structure of Agency . . . . . . . . . . . . . . . . . . . . . . . . . Janice Chik
95
7
Freedom of the Will and Rational Abilities . . . . . . . . . . . . . . . . . . . 115 Erasmus Mayr
8
The Power to Will Freely: How to Re-Think About the Problem of Free Will Without Laws of Nature . . . . . . . . . . . . . 137 Daniel D. De Haan
9
Laws Loosened . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Helen Steward
10
The Problem of Radical Freedom . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Andreas Hüttemann
xv
xvi
Contents
11
How the Libet Tradition Can Contribute to Understanding Human Action Rather than Free Will . . . . . . . . . . . . . . . . . . . . . . . 199 Sofia Bonicalzi and Mario De Caro
12
The Consequence Argument and an Ontology of Dispositions . . . . . 227 Mauro Dorato
13
Super-Humeanism and Mental Causation . . . . . . . . . . . . . . . . . . . . 257 Michael Esfeld
Chapter 1
What’s Dynamic About Causal Powers? A Black Box! Anna Marmodoro
Abstract Modern science cannot do without Aristotelian powers – thus have argued Cartwright and Pemberton (2013) among many others. Aristotelian powers are essentially dynamic entities, which account for causal phenomena, and thus explain how change comes about in the world. In this chapter I argue that explaining causation in terms of interacting causal powers places causation . . . beyond the reach of our understanding(!) – because causal interaction shows us what powers do, and not what powers are. Metaphysicians by and large agree that the intrinsic nature of powers is to be dynamic entities. I contend here that their dynamism is irreducible, and crucially, unknowable, rendering what powers are ‘black boxes’ to us, despite multiple attempts of defining them in the literature. The sciences discover only how powers behave, and classify them teleologically to tell us what they do. Powers, however, are mysterious and unexplorable black boxes to us, even though they are indispensable in our scientific explanations of change in the world. Keywords Causal powers · Dynamism · Causation · Scientific explanation
Assuming that causation holds the universe together, as J. L. Mackie (1974) perspicuously claimed, what is the ‘engine’ that keeps causal mechanisms ticking on? This has been a challenging problem in the history of philosophy, so much so that
I am grateful for the feedback received on earlier versions of this paper presented (online) at the Universities of Siegen (Change and Change-makers network) and Tübingen in Germany, Stockholm in Sweden, Turin in Italy (Labont), and the Yale-NUS College in Singapore. Michael Esfeld, Elisa Paganini and Alberto Voltolini also offered helpful comments on previous written versions of the paper, and so did my co-editors and an anonymous referee, whom I thank all warmly. The research leading to this publication was supported by the Leverhulme Trust as part of the project Part-whole relations within the fundamental potentialities in nature (RPG-2018-079). A. Marmodoro (*) Faculty of Philosophy, Oxford University, Oxford, UK Department of Philosophy, Durham University, Durham, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_1
1
2
A. Marmodoro
some philosophers took the stance of denying the existence of causation as a real phenomenon in the world.1 Throughout the history of philosophy however, those who thought that causal regularities are metaphysically different from contingent regularities, sought to reify in the ontology the ‘engine’ of causation by positing the existence of causal powers.2 Causation, for causal realists, is the exercise of causal powers, which are dynamic entities in the ontology in virtue of which change happens in the world. This account of causation however introduces an epistemic ‘black box’, I argue, which is a challenge or even a limit to our understanding of the physical world; and yet, not a reason to dismiss tout court a power-based account of causation. This black box is a metaphysical primitive which we cannot define, but of which we can specify the behaviour in certain circumstances. In this paper I will introduce what I call the ‘black box problem’; discuss critically some positions in the recent literature that may be taken to be attempts at solving the problem; and assess the significance of the issue for the friends of causal powers. While there is no consensus in the literature on a general characterisation of what a causal power is, for present purposes I will use as a working assumption that causal powers are properties that exist in the world, characterizing objects; e.g. the magnetic power of a magnet.3 Such properties can be either inactive, e.g. when the magnet is not attracting or repelling anything; or they can be activated (exercising or manifesting) in appropriate circumstances, e.g. when the magnet attracts or repels something else in its proximity.4 When powers are activated, they (or their possessors) bring about change.5 (I should here note that the exercise of a power is metaphysically different from the result that comes about from this exercise. Thus, if a magnet is moved on account of being attracted by another magnet, the magnet’s attraction is the power’s exercise, while the change of location is the resulting change.) The crux of the phenomenon that interests us here is this: how does what is inactive bring about activity and change? This is a puzzle. Whatever it is, it has to be
Thus, David Hume famously argued that what ‘there is’, in causal happenings in the world, is only what we can observe, namely, regularity: ‘All that causation consists of in the objects themselves [i.e. in the empirical world] (which is knowable to cognizers like human beings) is regularity, contiguity, and temporal priority of the cause to the effect. That is all there is to causation [. . .]’ (Hausman on Hume, 1998: 39). 2 I do not differentiate here between powers, dispositions, capacities, etc., because my aim is to develop an argument of general applicability to dynamic properties. 3 Powers qua properties characterise objects, whether objects result from the composition of powers or are ontologically prior to powers. (I endorse the former view.) 4 I briefly present my own view concerning the alleged inactivity of a power when not exercising/ manifesting in footnote 24. 5 For present purposes, whether it is the power or its possessor that bring about change, is not an issue that makes a difference to my present argument. 1
1 What’s Dynamic About Causal Powers? A Black Box!
3
‘within’ a power, and not dependent on external circumstances making the power active.6 Power metaphysicians thus build change into the very definition of powers: a (type of) power is essentially defined by the (type of) change it (or its possessor) can bring about. For example, a magnetic power is essentially an attractor or repeller – that is what a magnetic power does; it attracts or repels. This provides an explanation of how change is brought about: on account of the presence of the relevant powers, exercising in appropriate circumstances. Vetter (2020) among others highlights the significance of this explanatory role of powers in our metaphysics thus: Thus we arrive at what I suggest is the core of the contrast between Humean and anti-Humean metaphysics: the order of explanation among the modal-causal phenomena. For the anti-Humean, dispositions are a central explanans rather than a problematic explanandum. This is the basic idea of what I will suggest under the label ‘explanatory dispositionalism’. (My emphasis). I want to focus on Vetter’s claim that ‘dispositions are a central explanans rather than a problematic explanandum’. My point here will be to argue that the explanation causal powers provide for change comes at a cost, namely building primitively into the definition of powers what we need them to do in our metaphysics: having the ability of bringing about change. Taking powers thus defined as explanans, explains at the cost of making the explanans a primitive in our ontology.7
The external circumstances only ‘enable’ the exercise of powers but cannot make powers active, as many have noted in the literature, with what is sometimes called the ‘passivism’ objection: if powers were made active by the circumstances they are in, they would be themselves essentially passive entities. Thus e.g. Groff writes: a power ‘has the inherent, non-metaphorically dynamic, wherewithal to M [its manifestation]’ (2013: 222, original emphasis). 7 To be clear: some explanatory work gets done, even if at some cost. Esfeld (2020; see also Esfeld and Deckert 2017) and I concur on the diagnosis that positing powers to explain change is problematic, in so far as it requires appealing to primitives, whose workings are as mysterious as those of Molière’s opium’s virtus dormitiva; but while Esfeld argues that admitting powers as ontological primitives ‘does not lead to a gain in explanation’, I argue that it does do explanatory work. He writes: 6
[. . . a theory] does not explain why there is attractive motion in the universe by attributing a mass to bodies, because mass is defined in terms of the role to make bodies attract each other [. . .]: like the dormitive power of [Molière’s] opium, [physical powers such as mass] are defined in terms of the effects that they bring about under certain conditions. That is why they cannot explain these effects. And that is why it does not lead to a gain in explanation to admit powers or dispositions as ontological primitives. They cannot explain why there is motion at all, for they are powers for specific motion. They cannot explain that specific motion either, for they are defined in terms of a causal or functional role that is precisely that specific motion. (2020: 78; my emphasis) My reply is that an ontological theory needs to explain two things: first, that there is a phenomenon in the world; and second, why/how this phenomenon happens. My claim is that positing powers explains that change happens in the world, even though is sheds no light on why/how it happens, which is primitive dynamism in the ontology. My further claim (which I only state here because it cannot be defended within the scope of this paper) is that there are no alternatives that serve this explanatory purpose, either in metaphysics or in physics. Metaphysical
4
A. Marmodoro
Let us take the example of the fragility of glass. What we can empirically observe is that the glass breaks when struck in certain circumstances – we observe a change of a certain type. To explain how this change occurred, we (anti-Humeans) posit a power of a certain type, which is by definition able to bring about this type of change – the power of fragility. Our assumption is that the presence of the power that gets activated upon striking explains how the relevant change, the breaking of the glass, came about. So here is the question: how does a power do this? What is it ‘within’ a power that enables it to bring about change? Powers are posited as engines of change in the world: dynamic entities that literally ‘power-up’ everything that happens. But what is it for a power to be dynamic? My point is that claiming that it is a power’s dynamism that powers-up change in the world gives a name to what we are looking for, but does not explain it at all. It looks as if it is a bit of magic that somehow makes the world an eventful place of changes. This is the black box problem of powers.
1.1
The Black Box Problem
What is, and what is the significance of the black box problem for the causal power theory? I believe that this is the same type of problem that we also face in trying to explain why the world consists of the fundamental properties that it does, e.g. magnetism, spin, etc. We accept these as fundamental primitives, which characterize the universe we happen to live in. Neither science nor philosophy have any inkling as to how these came to be the fundamental building blocks of reality, or why they were not different than they are, or if they necessarily had to be of the kinds they are. In any ontology of the world, some fundamental building blocks are posited as primitives. I submit that powers are ineliminable fundamental building blocks of reality. We posit them for a reason, and with a guiding criterion, often referred to as the so-called Eleatic Principle according to which what exists is causally powerful;8 but we posit them as primitives. I thus address the black box problem on the basis of four theses:
and scientific explanation is grounded on axioms, which are primitives that explain that certain phenomena happen. Explaining also the why/how would ‘undo’ the axiomatic status of primitiveness, rendering axioms into theorems. If however a theory does not posit any axioms, but is grounded on the mutual coherence of the claims of the theory, then it will have no primitive items in the ontology either. But then, ‘coherence’ will be primitive; it will not explain anything, but will be a precondition of all explanation; and the problem will become the ‘black box of coherence’. 8 The Eleatic Principle is Plato’s: ‘I suggest that everything which possesses any power of any kind, either to produce a change in anything of any nature or to be affected even in the least degree by the slightest cause, though it be only on one occasion, has real existence. For I set up as a definition which defines being, that it is nothing else but power.’ (Sophist 247d-e.) A modern formulation is e.g. Armstrong’s, ‘everything that we postulate to exist should make some sort of contribution to the causal/nomic order of the world’ (2004: 37).
1 What’s Dynamic About Causal Powers? A Black Box!
5
(a) Realism: Both metaphysics and science give us reasons to posit powers as fundamental dynamic entities in our ontology.9 (b) Metaphysical Humility: Neither metaphysics nor science offers a reductive account of the dynamism of powers.10 (c) Primitivism: Powers are thus posited to be primitively dynamic fundamental entities in the ontology. (d) Epistemic Commitment: Even if we cannot look into the black box that powers are to us, epistemologically, we can nevertheless be committed to specifying (as an epistemic task) what this black box needs to be like, in order to do the work it is posited to do, in both metaphysics and science. I take the dynamism of a power to be what enables it, from within, to transition from inactivity to activity/exercise/manifestation; e.g. for a magnetic charge, from not attracting or repelling anything to attracting and repelling something. In what follows I motivate (with representative examples from the literature) why I hold Metaphysical Humility regarding the dynamism of powers; and then articulate what I mean by Epistemic Commitment and why I recommend it as the direction to follow for future research on the metaphysics of powers.
1.2
Telic Properties
Some philosophers have tried to account for the dynamism of powers in terms of natural teleology – a conception of the world that we owe to Aristotle. Aristotle’s initial contribution was to distinguish natural teleology from purpose, which is linked to the agency of an (animate) subject (see Physics 199b28). Natural processes have a telos (goal, end), for which Aristotle was aiming to develop a naturalistic understanding. The challenge here is that processes do not wear their teleology on their sleeves. There is change in nature, where one state succeeds another in a process; but is there ever an end to any natural process, and does it have a distinctive position in the sequence? A clue to answering these questions is the repetition of a process in nature. Processes that repeat themselves are cyclical, as they are repeated over and over again. However, cycles, as such, do not have starting points or end points. So, for a cyclical process to be teleological, it needs to be developmental,
9
This point has been laboured by many, among the most recent publications where it is discussed see Vetter (2020). For present purposes, I am assuming that science and metaphysics give us reasons to posit a plurality of powers such as the ones that use as examples in the chapter; I do this for the reader’s sake, to facilitate connecting my arguments to the debates to which I make reference. I am however open to the possibility that there may possibly be a global holistic power that accounts for all change in the world. Such research hypothesis is currently pursued by Simpson (2020a, b) and contemplated for instance by Esfeld (2020). 10 For instance, virtual particles are primitively assumed to behave dynamically.
6
A. Marmodoro
towards a final goal. But development is a theoretical concept that cannot be simply read off observations of a cyclical process. This brings us to the second main feature of Aristotle’s natural teleology: the telos of a process need not be the end of the process. For example, the telos of a plant is the development from seedling to full organism. The telos of a developmental process can be diachronic, e.g. the telos of a plant is the diachronic stage in which the plant is seed-bearing. However, crucially for our purposes, the telos of a plant is not the plant’s death. Thus we have, in a sense, two cycles: one is from the beginning to the telos of the developmental process – here, from growth to maturity; while the other is from the beginning to the end of the process – here, the plant’s death.11 Of course, the teleological cycle needs to account for the generation of the natural cycle, as it does here with organisms. So, the telos of a cyclical process in nature is a theoretical discovery, rather than an observational one. The telos of a cyclical process is the theoretical backbone that explains the stages of the process and its developmental circularity. The telos of a process is the ground on which the individuation concept of the process is developed, setting the criteria for its occurrence. Thus, ‘plant’ is a sortal concept that individuates the cycle of the process of that organism. Identifying the telos of an organism explains the stages of the organism all the way to its end, not only to its telos; e.g. a plant achieves its telos through a series of developmental stages, and loses its telos through a series of deterioration stages. Aristotle’s teleology has inspired modern approaches to the question of the powers’ dynamism. Kroll (2017) is a representative of this teleological approach to powers,12 and distinguishes, as Aristotle did, between instantaneous manifestation versus processual manifestation. Kroll writes: If x has a disposition to M [i.e. to manifest], when C [i.e. the appropriate condition] is activated, then in virtue of x being in a state directed at the end that x Ms. when C, either x immediately Ms. or there is a process directed at the end that x Ms. (2017: 21).
Kroll explains with an example: Suppose the salt is placed in water and its disposition is activated [...] [Thinking of dispositions teleologically] tells that, in virtue of the salt being in a state directed at the end that it dissolves when in water, either the salt immediately dissolves or there is a process directed at the end that the salt dissolve (2017: 21).
Such teleological accounts of powers however, of which Kroll’s is a representative example, are not explanatory of the powers’ dynamic nature; as the above quotes show, the directedness of powers is assumed to only dictate their end state, and it is
A third sense of a ‘cyclical telos’ (but applying only to biological entities) is that whose end-point is the cyclical maintenance of the organism’s morphology – e.g. the essence of a human is to produce such and such cells, grow this particular bone structure, etc. and productively maintain (by continual replacement) these things; see e.g. in connection with the metaphysics of powers Austin and Marmodoro (2017). 12 Similarly for instance in the recent literature Austin (2019: 41), Koons (2017: 15–16), Koons and Pruss (2017), Oderberg (2017) among others. 11
1 What’s Dynamic About Causal Powers? A Black Box!
7
not put to use to explain what, from within the powers, makes them tend towards it. I part company with this approach; yes, powers are telic properties, but we are still owed an account for their dynamism. Teleology thus put to use does not explain dynamism; it only presupposes it. Identifying the telos of a power tells us much about what follows the power’s activation, but it does not tell us anything about how the power does what it does, and brings about change. We need to ‘look’ within the constitution of powers, and not only at the process by which their exercise/manifestation leads to a certain effect in the world. Are there alternative approaches in the literature that are more promising than this teleological account?
1.3 1.3.1
The Directedness of Powers Physical Intentionality
An alternative approach to the explanation of the nature of powers is the one according to which the internal complexity of powers is conceived as the ‘directedness’ of powers: a power (of type) P is directed towards its manifestation/exercise (of type) M. So, what is directedness? (And does it account for the dynamism of powers?) Some metaphysicians have described the directedness of powers as ‘physical intentionality’ (e.g., Martin and Pfeifer (1986), Place (1996), Molnar (2003), Heil (2003), Martin (2008), Borghini (2009)).13 Others however have objected to this characterization (e.g., Bird, 2007a, b; Mumford, 1999; Mumford & Anjum, 2011; Bauer, 2016 engages in replying to the critics of the view).14 I consider ‘physical intentionality’ a metaphor at best, and an unhelpful one. For starters, even within the philosophy of mind, intentionality is not by and large considered a clearly understood concept, and that by itself gives us reasons to be sceptical as to the explanatory work it might do within the metaphysics of powers. Further, the obvious consideration against attributing physical intentionality to powers, already voiced by many, is that ‘intentionality’ is essentially mental, qua representational. Thus, explaining the nature of physical powers by appealing to intentionality appears to be an incongruous idea; clearly e.g., the salt’s power to dissolve in water could not involve the salt having a ‘representation’ of water. My further objection to the concept of physical intentionality as applied to powers is that, whereas powers are dynamic, there is nothing dynamic about intentionality understood as Brentano (1874) originally did, as a feature of thoughts, namely their having a ‘direction [Richtung] toward an object’ – as being about an object. I want to Molnar uses the expression ‘physical intentionality’ (2003: 60–81), Heil ‘natural intentionality’ (2003: 221), and Martin (2008: 178), Place (1999, passim), and Borghini (2009, passim) use ‘intentionality’. 14 Mumford goes as far as to take this approach literally, rather than as a metaphor. He writes that those who conceive of powers as having physical intentionality are thereby committed to animism or panpsychism (1999: 220–1). 13
8
A. Marmodoro
distinguish here between the ‘aboutness’ of mental states such as “I believe you made the coffee strong today”, and “I want to grab a coffee before going to work”. The former type of thought does not embody any motivation or desire for the coffee; its ‘intentionality’ towards the coffee or ‘being about’ the coffee is irrelevant, even if used only as a metaphor, for understanding the nature of powers. On the other hand, those who want to use the ‘aboutness’ of the second type of thought to explain the nature of powers, fall prey of circularity: desiring a coffee is the exercise of a mental power.
1.3.2
Identity-Fixing Relations
There is a group of accounts of powers which posit that a power and its exercise/ manifestation are related by an identity-fixing relation. This approach tries to link more intimately powers and their exercise/manifestation, by treating a power’s exercise/manifestation as the essence of the power itself.15 Generally, this is what dispositional essentialists hold: that a power (type) is essentially defined by its causal role, which is its exercise.16 There are different versions of this position. One version is that presented (but not endorsed as his own) by Psillos (2006)‘s, where he relates each power to a further property, its manifestation-property. Thus, a power P is related to its exercise/ manifestation M, which is a numerically different property from P; M is the essence of P, because P is defined as a power to M, e.g. breaking. Furthermore, Psillos holds that M needs to be a categorical property, not a power, to avoid a regress of powers (e.g. Psillos 2006). In a sense, Psillos’ position is that the essence of a power is extrinsic to it, and is not powerful, too – only the power is powerful. To put it in perspective, Psillos is dividing, ontologically, the telic nature of a power from its dynamism; the telic nature of a power is the essence of the power, which is given in the exercise/manifestation (property M) of the power; its dynamism is assumed to reside primitively in the power (P), and not in its essence.17 A further account of the identity dependence of a power on further properties is given by Bird (2007a, b, and elsewhere). On his account, a power P is a relational property, whose essence is fixed by the relations that a power P holds to other relational properties; specifically, a power’s essence is determined by the power’s relation to its manifestation/exercise M and its stimulus S. According to this account, 15
I am not here claiming that those who posit an identity-fixing relation between a power and its exercise/manifestation, do so with the explicit aim to account for the dynamism of powers. My point is that is if we are looking for such an account, this approach may be thought to ‘hit’ closer to the target than others. 16 See e.g. how Yates puts it: “in addition to having R essentially, having R as opposed to some other causal role is what makes charge the property it is, rather than some other property . . . In other words, R individuates charge” (2012: 1). 17 I demonstrate this in Marmodoro (2007).
1 What’s Dynamic About Causal Powers? A Black Box!
9
the fundamental ontology of the world constitutes a structure of intrinsicallyinterrelated powers. Bird’s captures his worldview by means of a graph; the graph describes a world, whose ontology is built on intrinsically-interrelated powers. But what is needed and missing in Bird’s account (for the sake of overcoming the black box problem) is an explanation of how the fundamental powers account for what they are posited to explain, namely, how the world changes from one state to another. Bird’s graph describes the types of changes there are in the world, but not, I argue, the world as it is changing. Here, too, no account is given of the dynamic behaviour of powers. In conclusion, while there are specific problems that Psillos’ and Bird’s arguments are prey of, already discussed in the literature (e.g. Vetter, 2009; Marmodoro, 2010 respectively, among others); here we are focusing on one aspect of causal powers, their dynamism. Identity-fixing relations don’t deliver an account of what we are looking for, namely the dynamism of powers.
1.3.3
A Relation of Production?
Many philosophers implicitly assume that to cause is to produce, and that if powers are causes, they are productive ones. This is, from one perspective, progress, in the sense that this approach focuses more than any other on the dynamism of powers: producing is doing something. However, this is the crux of the problem: what is it that a power does (to become active) in order to do what it does (i.e. bringing about a change in the world)? How does this ‘doing’ happen? I (and others) have argued that problems arise if one conceives of the relation between a power and its exercise/manifestation as a relation of production. A clear recent statement of the production account is to be found e.g. in Backmann (2019), who writes that ‘a powers ontology [. . .] is supposed to be inherently dynamic and productive’ (2019: 1 and 2). He aims to generate a reductio ad absurdum, based on the argument that ‘the powers view [qua productive] is at least not straightforwardly compatible with any of the most prominent temporal ontologies, which, if correct, would be a serious argument against the powers view’ (2019: 1). His argument concerning (to put it in a nutshell) the alleged incompatibility between powers and time has already received replies in the literature.18 What is directly relevant to my overall argument here is how Backmann understands production: ‘The culprit [of this alleged incompatibility of powers with any of the most prominent ontologies of time] is . . . activity, which is so central to the neo-Aristotelian view’ (2019: 2). I take issue with Backmann’s assumptions, that the activity of powers is productivity, and that for a power to be dynamic is for it to be productive.19
18
See Giannini (forthcoming) and Roselli (2022, in this volume). The same characterization of power-based causation and the same assumptions are also e.g. in Groff (2016):
19
10
A. Marmodoro
Backmann assumes that all power metaphysicians invoke activity in the productive sense, myself included: Anna Marmodoro holds that powers can be characterised as a ‘readiness for action’ (Marmodoro, 2010: 29). Powers need not necessarily be expressed in order to exist, but they equip their bearer with the potential to act. The quasi-agential language here is no accident. (2018: 19, my emphasis). Indeed, it is no accident. My position however differs from how Backmann understands it, in that, on my account, the powers’ activity needs not be productive of anything. Aristotle was first to introduce the distinction between doing and making (praxis and poiein in his language). Powers for me are essentially doers, not makers. On my account, key to the conception of the activity of an exercising power is that a power that is exercising/manifesting is the same power that it was when inactive, only in a different (active) phase. For instance, when I exercise pressure on a football, it is the very same power I possessed in potentiality to pressure the ball that I am now exercising to pressure it; and when a magnetic charge is approached by another magnetic charge, it is the very same power in potentiality to repel a magnetic charge that is being exercised by the first charge to repel the second. A power’s exercise/manifestation is a different state of the numerically same power. This is the key difference between my account of the powers’ exercise/manifestation and the mainstream account in the literature. All versions of the mainstream account, including the productive account, share in common the idea that the exercise/ manifestation of a power is a numerically different power from the power that was in potentiality (i.e. inactive).20 Causes, from this perspective, literally produce the effects that they cause [. . .] I will refer to the dynamic view of causation as a “powers-based” approach, and to the sense of causation that it sustains as “productive” causation [. . .] in the history of Western philosophy, Aristotle is the paradigmatic powers theorist, such that accounts of this type are widely considered to be neo-Aristotelian (2016: 287) I disagree with the general assumption that powers have a productive activity, and with the specific one that that Aristotle paradigmatically thought so. 20 In the case of Mumford and Anjum, although their views are ‘in the ballpark’ of the mainstream account, it is hard to work out which version of it they hold precisely – if it is a single consistent view, across their multiple publications. They hold that (i) a power P1’s manifestation/exercise is the production of a new power, P2 (see e.g. 2011: 8); (ii) a power P1’s manifestation/exercise is its replacement with a new power, P2 (see e.g. 2011: 5); and (iii) a power P1’s manifestation/exercise is both the end point of the self-transformation of P1 and a new power P2 (see 2018: 69). I call this latter position their ‘hybrid’ account. However, it is an account hard to comprehend; they explain it thus: This [which I call the ‘hybrid’ account] allows us to answer Marmodoro’s (2013: 550) concern, which was raised against our account of powers. She says that we take the manifestation of a power to be another power, while she thinks that the manifestation is a different stage of the same, first power. But in a process view it can be both. Given that the manifestation of a power produces a continuous change in properties and that properties are understood to be powerful, the properties produced by causes are both the end point of the original power and a new power, available to form a new reciprocal partnership if circumstances allow. (2018: 69)
1 What’s Dynamic About Causal Powers? A Black Box!
11
A word about my motivation for the stance I take, concerning the oneness of a power and its exercise/manifestation. Treating the exercise of a power as a numerically different power is mistaking the exercise of a power (e.g. magnetic charge attracting metal) for the product of a power (e.g. the metal’s moving closer). When a magnet is attracting a piece of metal, it’s exercising on the metal a power (the magnetic charge) which the magnet possessed ‘in potentiality’ before exercising it. It would not make sense to hold that the moment the magnet attracts the metal, it is doing so with a different power than its power to do so. A power is posited to explain the change it brings about; we cannot divorce a power from change-making, by positing another power to do just that work. Severing ontologically a power from its exercise/manifestation can only be a source of metaphysical troubles.21
1.4
Intrinsic Complexity and Extrinsic Dependency of Powers
When a charge attracts another, the second charge is attracted by the first. I have hereby just described two different phenomena (within a more complex causal nexus): the exercise of a power, and the interaction between two powers. The first phenomenon reveals the intrinsic complexity in the nature of a power, namely, its ability to exercise/manifest, e.g. from the potentiality to attract to actually attracting; the second phenomenon reveals the extrinsic dependence of a power on its ‘partner power’22 in the environment, e.g. that the charge is attracting a power that is being attracted by it. The intrinsic complexity of a power and its dependence on its partner power are themselves interdependent; not relationally, through a relation between a power to its partner power, but in an ontologically sui generis way. Their interdependence is revealed to us as two features of a single activity – the change, when a power interacts with its partner, e.g. the single activity of attracting-andbeing-attracted. Aristotle described this Janus-faced interaction as an activity with two beings – e.g. attracting-and-being-attracted.23 However, even here, where we come to understand the metaphysical complexity of exercise/manifestation as a physical activity, we are still not thereby explaining the dynamism of the transition
As I argued above, the exercise of a power is not the same entity as the product of the power, which Mumford and Anjum take it to be in their ‘hybrid’ account. Identifying the two items also runs afoul of the difference between a power altering from one stage to another, and the power transforming into a different power. 21 As mentioned above, the exercise of a power (the magnet’s attracting metal) is metaphysically different from the result that comes about from this exercise (the change of location) of the metal. 22 I am here adapting C. B. Martin and Heil (1998)‘s expression. 23 I argued for this interpretation of Aristotle’s position in Marmodoro (2007) and developed a version of this position as my own theory in Marmodoro (2020).
12
A. Marmodoro
from potentiality to activity. Dependence is not dynamism.24 Dependence is an ontological condition; dynamism is activity.
1.5
Taking Stock
We set out to investigate the crux of the metaphysics of powers, namely their dynamism – how to account for the ‘inner workings’ of a power that enable it to bring about activity. I argued that the existing teleological accounts describe all but the dynamism of a power – they describe the behaviour a power exhibits when activated; but not what accounts for its becoming activated. I further argued that among the accounts that attempt to explain the inner complexity of a power, those which posit an identity-fixing relation between a power and its exercise/manifestation fail to capture the power’s dynamic nature. The productive accounts do focus on the behaviour of powers, but only on its target, rather than its ‘mechanism’; and face also additional difficulties (already discussed in the literature). My own account, whose key feature is the numerical oneness between the inactive power and the exercising, on the one hand avoids the difficulties mentioned for other accounts, but on the other, discovers and reveals an ontological predicament regarding the dynamic nature of powers, which I articulate in what follows.
1.6
Tu Quoque, but for a Reason
In previous work, I described the dynamic nature of powers by likening the directedness of powers to ‘readiness for action’. However, one may ask: What is it for a power to be ‘ready for action’? Clearly ‘readiness for action’ is merely a metaphor. However, what may not be clear is the reason why I resort to a metaphor. I have done so because there is nothing I have found, which is ontologically more primitive than powers, which could be used to explain their dynamic nature. Powers are at the bedrock of reality, its very building blocks, and therefore, their dynamic nature is (ontologically) a primitive given. I submit that the intense and ever-growing research program into the metaphysics of powers that have been developed in the past few decades, is revealing to us that the dynamic behaviour of powers, which drive all change in reality, is an epistemological black box for us. We cannot use more fundamental concepts to explain the nature of a power, since powers are the most fundamental elements of reality; we shall therefore need to focus our investigation in what the black box is capable of doing, rather than what it consists of and how it is made up. This approach will not tell us what the power is, but rather point towards the black box and tell us that it
24
E.g. 4 is dependent on 3 as 3’s successor, but they are not dynamic.
1 What’s Dynamic About Causal Powers? A Black Box!
13
should be such as to be able to deliver ‘this’ and ‘that’, even if it will never be manifest to us. The recognition that powers are a black box helps us understand the history of the definitions of the nature of powers, from Aristotle onwards. Powers have always been defined teleologically, in terms of their manifestations: A power φ aims at φing, and φ-ies when the appropriate conditions occur. This tells us what the power can do, rather than what it is; but we are now seeing the reason why this is so. The definition of a power is an epistemological tool that says that the power’s nature, whatever this is (i.e. a black box), enables it to do ‘this’ and ‘that’. I say that a power is in ‘readiness for action’ to indicate that the black box their nature is, is all that is needed for a power to spring into exercising/manifesting, when appropriate external conditions occur.25
1.7
Conclusion
We have been looking for the mechanism of potentiality, to explain how powers are able to dynamically initiate change in the world, and we found a black box. This is in a sense a result to be expected: the very notion of mechanism rests on arrangements of powers and thus it would be circular to use this notion to explain powers. Powers are the bedrock of reality. So what I have argued for here is that, although powers are needed to explain causal change in nature, it does not follow that we should be able to offer an ontologically reductive mechanism to explain how powers can do this. The nature of a power, inactive or exercising/manifesting, is a black box that we struggle to understand through metaphors. Powers are dynamic, but we cannot explain how. Accordingly, my main point has been to argue that these metaphors are not breakthroughs into the nature of powers. Rather they are epistemological tools that help us grasp, not what is in the black box, but rather, how this black box behaves, specifically for each type of causal power, which delineates its teleological role as posited in the ontology. We cannot access epistemically the black box but we can make epistemic progress with respect to our understanding of how the black box behaves. I have identified an irreducible element in the nature of powers, namely their dynamism – what makes powers dynamic. I have further explained why it is irreducible, and therefore not definable.26 However, what can and has been identified 25
In this sense, on my account a power is constantly ready to spring into action, given the chance. By this I mean that the exercise/manifestation of a power is blocked by the environment, unless the appropriate conditions obtain, in which case we observe its manifestation. 26 I here briefly introduce a further issue, which cannot however be developed in full here because it would take us astray from the main argument of this paper. The issue concerns the relationship between powers and time. Does positing that powers have a temporal structure explain their dynamism? The answer is “Yes and no”. There are two conceptions of time: on the Aristotelian,
14
A. Marmodoro
in the sciences is how powers behave. I showed that this does not tell us what a power is, but it does allow us to classify powers teleologically, in terms of their manifestations. That is, specifying the manifestation of a power does not tell us what is in the black box, but only how the power behaves. These conclusions express my Metaphysical humility and Epistemic commitment. Thus, we should recognise that the so called ‘directedness’ of a power is but a teleological description of what a power does, not of its nature. I argued that this is the best we can do, epistemologically, in understanding and describing powers in nature. It is in this sense that powers are future-directed and telic properties, in that they are understood by us and science as doers, the most primitive doers, where what they do is stated in the very definition of their kind. In this respect, a power’s nature is predictive of the type of possible future states of the world.
References Armstrong, D. (2004). Truth and Truthmakers. Cambridge University Press. Austin, C. J. (2019). Essence in the age of evolution. Routledge. Austin, C. J., & Marmodoro, A. (2017). Structural powers and the homeodynamic unity of organisms. In R. Koons, N. Teh, & W. Simpson (Eds.), Neo-Aristotelian perspectives on modern science (pp. 169–183). Routledge. Backmann, M. (2019). No time for powers. Inquiry, 62(9–10), 979–1007. Bauer, W. A. (2016). Physical intentionality, Extrinsicness, and the direction of causation. Acta Analytica, 31, 397–417. Bird, A. (2007a). The regress of pure powers? The Philosophical Quarterly, 57(229), 513–534. Bird, A. (2007b). Nature’s metaphysics: Laws and properties. Clarendon Press. Borghini, A. (2009). Dispositions and their intentions. In G. Damschen, R. Schnepf, & K. R. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology and philosophy of mind. Walter de Gruyter. Esfeld, M. (2020). Science and human freedom. Palgrave Macmillan. Esfeld, M., & Deckert, D.-A. (2017). A minimalist ontology of the natural world. Routledge. Giannini, G. (forthcoming). Powers, processes, and time. Erkenntnis. Groff, R. (2016). Causal mechanisms and the philosophy of causation. Journal for the Theory of Social Behaviour, 47(3), 286–305. Hausman, D. M. (1998). Causal asymmetries. Cambridge University Press. Heil, J. (2003). From an ontological point of view. Clarendon Press.
A-series (as we call it) conception of time, time dynamically moves from a ‘now’ to subsequent ‘nows’. On the B-series conception on the other hand time, is like space; distances in space are not dynamic, nor is distance in time; it’s all laid ‘out there’, past, present and future, and events have spatiotemporal distances between them. But nothing ‘happens’ on this conception. So, on former conception of time, the passage of time is dynamic; on the latter, time is non-dynamic. So on the former conception, the temporal structure of powers explains their dynamism too; only that, if that is the case, the black box problem transfers to the nature of time. On the latter conception of time, the dynamism of powers needs to be explained differently than their temporal structure. If we take metaphysics and physics to give us reasons to prefer the B-series conception of time, as I do, what accounts for the dynamism of powers cannot be their temporal structure.
1 What’s Dynamic About Causal Powers? A Black Box!
15
Koons, R. C. (2017). The ontological and epistemological superiority of hylomorphism. Synthese, 1–19. Koons, R. C., & Pruss, A. (2017). Must functionalists be Aristotelians. In J. D. Jacobs (Ed.), Causal powers. Oxford University Press. Kroll, N. (2017). Teleological dispositions. Oxford Studies in Metaphysics, 10, 1–37. Mackie, J. L. (1974). The cement of the universe. Oxford University Press. Marmodoro, A. (2007). The union of cause and effect in Aristotle: Physics III 3. Oxford Studies in Ancient Philosophy, 32, 205–232. Marmodoro, A. (2010). Do powers need powers to make them powerful? From Pandispositionalism to Aristotle. In A. Marmodoro (Ed.), The metaphysics of powers: Their grounding and their manifestations. Routledge. Marmodoro, A. (2013). Aristotle’s hylomorphism without reconditioning. Philosophical Inquiry, 37(1/2, winter/spring), 5–22. Marmodoro, A. (2020). Powers, activity and interaction. In A. S. Meincke (Ed.), Dispositionalism: Perspectives from metaphysics and the philosophy of science (pp. 55–66). Springer. Martin, C. B. (2008). The mind in nature. Clarendon Press. Martin, C. B., & Heil, J. (1998). Rules and powers. In Philosophical perspectives, vol. 12, language, mind, and ontology (pp. 283–312). Blackwell. Martin, C. B., & Pfeifer, K. (1986). Intentionality and the non-psychological. Philosophy and Phenomenological Research, 46, 531–554. Molnar, G. (2003). Powers: A study in metaphysics. Oxford University Press. Mumford, S. (1999). Intentionality and the physical: A new theory of disposition ascription. The Philosophical Quarterly, 49(195), 215–225. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford University Press. Mumford, S., & Anjum, R. L. (2018). Dispositionalism: A dynamic theory of causation. In D. J. Nicholson & J. Dupré (Eds.), Everything flows: Towards a processual philosophy of biology. Oxford University Press. Oderberg, D. S. (2017). Finality revived: Powers and intentionality. Synthese, 194(7), 2387–2425. Place, U. T. (1996). Intentionality as the mark of the dispositional. Dialectica, 50, 91–120. Place, U. T. (1999). Intentionality and the physical: A reply to Mumford. The Philosophical Quarterly, 49(195), 225–231. Psillos, S. (2006). What do powers do when they are not manifested? Philosophy and Phenomenological Research, 72(1), 137–156. Roselli, A. (2022). Dispositional essentialism in the eternalist block. In Powers, time and free will. Cham: Springer International Publishing (in this volume). Simpson, W. (2020a). What’s the matter? Towards a now-Aristotelian ontology of nature. PhD dissertation, University of Cambridge. https://www.repository.cam.ac.uk/handle/1810/304747 Simpson, W. (2020b). Cosmic hylomorphism: A powerist ontology of quantum mechanics. European Journal for the Philosophy of Science, 11(2), 1–25. Vetter, B. (2009). Book review: Alexander Bird: Nature’s metaphysics: Laws and properties. History of Philosophy & Logical Analysis, 12(1), 320–328. Vetter, B. (2020). ‘Explanatory dispositionalism. What anti-humeans should say. Synthese. Online first: https://doi.org/10.1007/s11229-020-02872-0. Yates, D. (2012). The essence of dispositional essentialism. Philosophy and Phenomenological Research, 1–36.
Chapter 2
Toppling the Pyramids: Physics Without Physical State Monism William M. R. Simpson and Simon A. R. Horsley
Abstract In this paper, we challenge a wide-spread assumption among philosophers that contemporary physics supports physical state monism. This is the claim that the causal powers of a system supervene upon the ‘lower-level’ laws and the lower-level state of the cosmos (as represented by our ‘best physics’). On this view, it makes sense to ignore a macroscopic system’s higher-level properties in determining its causal powers, since any higher-level powers are merely artifacts of our special interests. We argue that this assumption is common both to microphysicalism, which carves the cosmos into a set of microscopic constituents, and priority monism, which posits a single cosmic substance, but is incompatible with any form of physical pluralism that attributes irreducibly higher-level powers to entities of intermediate scales. We consider a number of case studies in contemporary physics which fail to support the thesis of state monism. We argue that the causal powers of many systems are (determined by) higher-level, macroscopic properties that are neither reducible nor weakly emergent, and that contemporary physics is compatible with some kind of pluralism that affirms that these entities are robustly real. A pluralist ontology is likely to have implications for discussions of free will and agency. Keywords Scientific pluralism · Quantum mechanics · Causal powers
2.1
Introduction
In contemporary metaphysics, macroscopic objects are not ordinarily counted among nature’s building blocks, particularly among philosophers who are seeking to put forward a scientifically respectable account of nature’s fundamental ontology.
W. M. R. Simpson (*) University of Cambridge, Cambridge, UK e-mail: [email protected] S. A. R. Horsley University of Exeter, Exeter, UK e-mail: [email protected] © The Author(s) 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_2
17
18
W. M. R. Simpson and S. A. R. Horsley
Although various forms of scientific pluralism are being actively explored by philosophers of science (Chang, 2017), it is more common for metaphysicians to endorse microphysicalism, in which the fundamental constituents are very small (such as particles), or even to adopt priority monism (Schaffer, 2010), in which there is only one fundamental entity (namely, the cosmos). In contrast, pluralist perspectives of nature that admit fundamental objects at different scales are typically taken to involve some form of ‘folk mereology’ that lacks scientific respectability (Rose & Schaffer, 2017), since they fail to accord with what we have good reason to believe about nature on account of the success of our ‘best physics’. In this paper, we are concerned with removing this impediment to the construction of pluralist accounts. An example of such an account is Inman’s (2017) recent proposal for a ‘neo-Aristotelian’ mereology that admits ‘ordinary, macroscopic composite objects’ (ibid, p.4) within its fundamental ontology. Our concern is not to establish any particular ontology of nature, however, but to argue that there is reason to think quantum physics admits an ontological reading that is compatible with some kind of physical pluralism, in which there are fundamental objects that exist between the microscopic and the cosmic scale. A good reason for adopting some kind of physical pluralism is the existence of macroscopic objects with top-down causal powers. Such objects have ‘higher-level’ physical properties that make a causal difference to the temporal development of their lower-level physical properties. Conversely, a good reason to reject physical pluralism is the supposed reducibility of any higher-level powers to (or their weak emergence from) some lower-level substrate described by our best physical theory, whose physical content is (supposedly) the set of worlds possible according to its laws. We argue that neither microphysicalism nor priority monism should be regarded as more scientifically respectable than physical pluralism, however, because there is good reason to think that physics is compatible with a form of strong emergence that admits irreducible, higher-level causal powers. Our paper is divided as follows. In Sect. 2.2, we claim that microphysicalists and priority monists presuppose the truth of physical state monism. This is the assumption that the causal powers of a system which bring about change supervene upon the lower-level state of the cosmos and the lower-level laws that govern this state. In Sect. 2.3, we discuss the role played by higher-level, macroscopic boundary conditions in physics in determining the physical content of a theory, in addition to any lower-level laws, observing that the reduction of the higher-level to the lower-level never goes all the way in practice. In Sect. 2.4, we consider the role of different representations in fulfilling the explanatory virtues of quantum physics, observing that our best physical theories fail to disclose a unique microphysical structure which evolves according to one set of lower-level laws. In Sect. 2.5, we discuss three physics-based arguments against pluralism: first, Schaffer’s argument from quantum entanglement, which claims the whole of nature is reducible to a single substance; secondly, the Quantum Darwinist’s argument for the weak emergence of the classical world from a purely quantum substrate, which appeals to the process of decoherence; thirdly, the argument from the causal closure
2 Toppling the Pyramids: Physics Without Physical State Monism
19
of the microphysical world, which claims that strong emergence demands the existence of ‘spooky’ forces. We show how each of these arguments begs the question against physical pluralism. In Sect. 2.6, we offer some concluding remarks on the prospects of physical pluralism.
2.2 2.2.1
Bricks Without Straw Microphysicalism and Priority Monism
Microphysicalism remains a popular position among analytic philosophers. Hüttemann (2015) distinguishes three forms that it may take: micro-determination, micro-government and micro-causation. According to micro-determinists, ‘the behaviour or the properties of compound systems are determined by the behaviour or the properties of their constituents and the relations among them but not vice versa’ (ibid., p. 7). For philosophers who uphold micro-government, ‘the laws of the micro-level govern the systems on the macro-levels’ (ibid.). Among those who are micro-causalists, it is claimed that ‘all causation takes place in virtue of the causation on the level of the (ultimate) parts – or the micro-level. Macro-causation, on this view, is entirely derivative and piggybacks on the causation of the microconstituents’ (ibid., pp. 7–8). Microphysicalism exerted a powerful influence over the philosophy of science of the last century. In Oppenheim and Putnam’s (1958) influential paper, “The Unity of Science as a Working Hypothesis”, nature is conceived as a hierarchy in which cells are composed of molecules, molecules of atoms, and atoms of whatever microscopic constituents are identified by physics at the ‘unique lower level’ (ibid., p. 9), which have to be described solely in terms of the language of our best physics. Likewise, microphysicalism has proven a hard taskmaster in the philosophy of mind. Kim (1997) famously argued that mental properties must be reduced to microphysical properties if they are to be considered causally efficacious,1 which raises doubts about free will given the alleged causal completeness of the lower-level physics. The background conception of nature that shapes these metaphysical theories about how the world is put together can be pictorially represented as a giant pyramid, in which everything rests upon a comprehensive lower-level of microscopic constituents. It is perhaps not so controversial these days, even among card-carrying physicalists, to confess to having doubts about microphysicalism in the light of contemporary physics (Papineau, 2008). In his comments on Hüttemann’s (2015) case against microphysicalism, Schaffer (2008, p. 256) identifies three reasons to think microphysicalism false: against micro-determinists, he notes that, ‘the properties of subsystems are determined by the properties of systems and not vice versa’, appealing to the ‘argument from quantum entanglement’; against the concept of 1
As Kim (1984, p. 100) puts it, the world is as it is ‘because the micro-world is the way it is’.
20
W. M. R. Simpson and S. A. R. Horsley
micro-government, he observes that ‘the laws of the macro-physical govern the microphysical systems’, appealing to the ‘argument from the universe being the only isolated system’ (ibid.); against micro-causation, he urges that ‘all causation takes place in virtue of the causation on the macro-physical level’, appealing to the ‘argument from macro-government plus cause-law connection’ (ibid.). As Schaffer points out, however, these arguments against microphysicalism do not uniquely favour the kind of physical pluralism that Hüttemann espouses, in which physical systems of all scales are regarded as interdependent equals. They also count as evidence in favour of priority monism, in which the only fundamental thing in existence is the cosmos, and everything else is grounded in the cosmic whole (Schaffer, 2010). The picture of nature, in this case, is of an inverted pyramid, in which everything rests on a single point. Yet there is another reason for favouring some kind of physical pluralism over priority monism: namely, the existence of systems of an intermediate scale, between the microscopic and the cosmic, that possess higher-level causal powers. According to the physicist George Ellis, such higher-level powers act ‘top-down’ on the lowerlevel properties of their environments, and are not reducible to (nor weakly emerge from) a system’s lower-level powers (Bishop & Ellis, 2020). However, microphysicalism and priority monism are themselves manifestations of a more entrenched orthodoxy concerning the relation between physics and metaphysics that characterises contemporary philosophy, which rules a priori against the existence of macroscopic entities with higher-level causal powers. To put the point more accusingly: the orthodox way of interpreting physical theories among philosophers seems to be rigged in favour of reductionism.
2.2.2
Physical State Monism
It is widely supposed among analytic philosophers that to offer a metaphysical interpretation of a physical theory is to identify the set of worlds that are possible according to that theory. On this view, a possible world is a complete and internally consistent possible state of affairs, and a physical theory contributes to our knowledge of nature by declaring some of these states permissible whilst excluding others. It is also widely supposed that the laws specified by a physical theory determine the set of possible worlds that it permits: they are those complete states of affairs that are consistent with its laws. For many philosophers, the task of interpreting a physical theory, such as quantum mechanics, is a matter of fixing upon some set of lower-level constituents and picking out their possible arrangements according to its lower-level laws (Dorr, 2011, p. 139).2 The basic lower-level constituents described by our ‘best physics’ may be conceived as a set of microscopic entities or as modifications of one
2
Button (2013, p. 12) has described this recipe as ‘orthodoxy for post-Quinean metaphysics’.
2 Toppling the Pyramids: Physics Without Physical State Monism
21
underlying cosmic substance. Either way, the total set of possible arrangements of the world’s lower-level constituents determines the state space within which the state of the cosmos evolves. Having identified these basic constituents, propositions about the world may be evaluated as true or false just in case they can be understood as referring to their possible arrangements. On this view, ‘everything that is physically possible must be possible in the same way’ (Ruetsche, 2011, p. 3). Ruetsche (2011) calls this the unimodal conception of physical possibility, which is embodied in the standard conception of physical content: Unimodal physical content The physical content of some theory T is determined by the set of physically possible worlds W in which the laws of T are true.
The notion of laws, in this case, is to be understood in the narrow sense of the lowerlevel laws specified by some physical theory T, rather than the broad sense of laws that govern physical entities in general. Likewise, the notion of possibility is to be understood in the narrow sense of what is possible regarding the arrangements of the lower-level constituents to which T refers, rather than the broad sense of what is considered to be possible in nature. This characterisation is narrowly concerned with the causal closure of the lower-level constituents of a system within the state space in which they evolve. The unimodal conception of physical content is affirmed by microphysicalists, who favour the ontological priority of the microscopic, and priority monists, who believe microscopic reality is grounded in the cosmos as a whole. Microphysicalists and priority monists are divided concerning the number of fundamental entities, but united in excluding from the fundamental ontology any entities that exist between the microscopic or the cosmic scale. In their hierarchical picture of physics, higher levels are supposed to be related to lower levels in such a way that the physical content of a higher level-theory can be derived from the physical content of a lowerlevel theory (Leggett, 1992). Such a claim has been supported by appealing to some version of Nagel’s (1961, pp. 353-354) account of how this reduction is supposed to take place, in which higher-level and more specialised theories are reduced to lower-level and more comprehensive theories by deducing the higher-level laws from the lower-level laws (plus boundary conditions). Whilst the Nagel-Schaffner model is not without critics, it has been staunchly defended as offering the best account of a successful reduction, and it qualifies Nagel’s original account in a number of helpful and significant ways.3 Consider the following two theories: Tf, is some candidate fundamental theory, such as our ‘best physics’, and Tt is a more specialised theory targeted for reduction. According to this model: Nagelian-Schaffner reduction Tt is reduced by Tf just in case the laws of T t are derivable from the laws of T f , and the terms of T t are associated via bridge laws with terms of T f ,
3
For a recent exposition and defence, see Dizadji-Bahmani et al. (2010).
22
W. M. R. Simpson and S. A. R. Horsley
where (a) T t is an analogous (or approximate) version of Tt, and (b) T f is derivable from Tf by means of auxiliary assumptions. The Nagel-Schaffner model recognises that exact derivability is an unrealistic requirement: it suffices that the laws deduced from the fundamental theory should be analogous to, or approximate, the laws of the targeted theory. Likewise, the auxiliary assumptions required to derive T f from Tf are typically taken to be idealisations and boundary conditions that have no ultimate significance concerning what is possible according to the reducing theory. Having obtained T t as an analogous version of Tt, and T f from Tf by appropriate idealisations and boundary conditions, they can be connected by bridge laws. The bridge laws that enable the derivation of T t from T f constitute rules of translation which connect the vocabulary of T t to that of T f . For Schaffner, a bridge law can be characterised as a ‘reduction function’, which offers a statement to the effect that some term we of T t is coextensional with some term wf of T f , and specifies the functional relationship between the magnitudes of the terms. We shall classify philosophers who adopt a hierarchical conception of physics, in which the physical content of higher-level theories is derivable from lower-level theories, as physical state monists, whether they are microphysicalists or priority monists. Physical state monism can be distinguished from the kind of physical pluralism whose scientific respectability we wish to defend by the way in which it determines the causal powers of a physical system: Physical state monism Let S be a physical system (smaller than the whole cosmos), with boundary conditions B delineating S from its environment. 1. The causal powers of S to change its physical properties supervene upon lower-level laws and the boundary conditions B. 2. The lower-level laws govern the lower-level state of the cosmos, and the boundary conditions B supervene upon this lower-level cosmic state.
We shall understand B to include the initial conditions of S. To accommodate the possibility of indeterminism, we may consider the powers of S to be probabilistic powers to bring about changes in the lower-properties of its components, which are weighted by a certain probability. Both a microphysicalist and a priority monist can affirm 1 and 2, although they conceive the lower-level differently. Both regard the higher-level properties of macroscopic objects as supervening upon lower-level properties. State monism entails the following: Physical state monism: corollary The higher-level properties of a physical system have no irreducible causal powers to change the lower-level physical properties of a physical system.
Our aim in this paper is to raise doubts about this claim. In doing so, we mean to call into question the wide-spread assumption that contemporary physics straightforwardly supports physical state monism, in which the world is pictured hierarchically as a pyramid (or an inverted pyramid) that rests upon a lower-level that is composed of one or more entities that exist at the same physical scale, over some more egalitarian form of physical pluralism, in which the world contains fundamental
2 Toppling the Pyramids: Physics Without Physical State Monism
23
entities at different physical scales that exercise higher-level causal powers. We think the evidence from physics can be parsed in a pluralistic way that is scientifically respectable, and that philosophers should feel free to adopt a pluralistic approach to physics in constructing their metaphysical models of nature. We will begin by considering how higher-level, macroscopic properties are invoked by physicists in practice.
2.3
The Role of Boundary Conditions in Physics
We live in a world in which macroscopic entities seem to play a role in settling how nature unfolds. Animals, for example, have causal powers to engage freely in a variety of different activities, such as nutrition and locomotion, which do not seem to be explicable without appealing to their inherent agency (Steward, 2012). According to physical state monism, however, the whole course of nature is settled by the lower-level state of the cosmos and the laws that govern the temporal development of this physical state. Yet what reasons might one have for thinking that physical state monism is true or scientifically respectable? One might try to support such a position on the historical premise that progress in physics involves the stripping away of macroscopic quantities, with physical state monism occupying the triumphant limit of this march; a position whose scientific respectability is justified by inductive reasoning. We shall argue that this ‘historical’ premise, however, is difficult to square with scientific practice.
2.3.1
A Question of Scale
In physics, it is commonplace to use microscopic phenomenological quantities that are experimentally measured, such as mass, charge, or spin, whose values are not themselves determined by the physical theory, in defining the physical state of a system. These are often assumed to come from a more fundamental and comprehensive theory of the physical world that we do not yet know. An example of this practice can be found in quantum electrodynamics. In this theory, the electric charge is ‘renormalized’ (Hollowood, 2013), where the experimentally measured value of the charge is understood to be the result of an infinite series of interactions between charge and field, which in turn rescales the measured strength of the charge compared to its ‘bare’ value. Summing this series of interactions predicts an infinite rescaling, revealing an incompleteness in our understanding of electric charge. It is hoped by some physicists that a microscopic theory such as string theory will one day eliminate this problem, providing some kind of fundamental basis for the number we call ‘charge’ (Wen & Witten, 1985). Yet there are also many cases where physical theories contain macroscopic phenomenological quantities. These are also not determined by the theory, but by
24
W. M. R. Simpson and S. A. R. Horsley
contrast, depend upon large scale properties of the physical system. The ubiquity of such quantities in physical models has to be squared with the hierarchical view of physics as a tower with different levels of description that sequentially ‘zoom in’ on some fundamental level of reality. In the hierarchical view, each lower level replaces the parochial and phenomenological description of the level above with an improved and more comprehensive lower-level description. One day, it is presumed, we will reach the ground floor and possess the language of Nature. There is a complication, however, in the way in which laws are applied to explain phenomena in practice: physicists typically use theories that contains quantities from levels ‘above’ whatever happens to be the level of interest, not merely from ‘below’ the level of interest. Such mixed-level approaches to scientific explanation occur in physical descriptions of (i) the boundary conditions of a system. Most physical theories are written in terms of differential equations, but a physical model of a system only offers testable predictions once boundary conditions for these equations have been specified. Yet these boundary conditions often have a higher-level, macroscopic origin. A simple example is a quantum particle incident onto a reflecting barrier (see Sect. 2.3.2). According to microphysicalists, the barrier is made of many microscopic constituents. The physical theory, however, instead of adopting a many-particle description of the barrier, deploys a single number that represents a macroscopic property of the multitude of particles (Harrison & Valavanis, 2016). (ii) systems in, or close to, thermodynamic equilibrium. Take a small collection of particles (positions xn 2 P) in equilibrium with a much larger system S (P ⊂ S) (see Sect. 2.4.3). The coarsest level of description would use classical thermodynamics, treating the whole system S using only the macroscopic variables of volume V and temperature T. To describe the particles in P , however, we require the theory of open quantum systems (Breuer & Petruccione, 2007), which is several levels of description more detailed than classical thermodynamics. A naïve supporter of the hierarchical view of physics might expect macroscopic variables such as temperature simply to be eliminated in such a quantum description. However, to capture the interaction between the particles P and the rest of the system it is essential to retain the temperature variable (ibid.), which must be experimentally measured at the macroscopic level. (iii) multi scale phenomena. The theory of Quantum Optics, for example, describes the behaviour of individual quanta of light (photons) inside transparent materials such as glass (Loudon, 2000). In this theory, light is treated microscopically and fully quantum mechanically, but the material enters the theory through the value of the refractive index, not as the collection of atoms making up the glass. This description arises due to a separation of scales: the wavelength of photons in the material is much larger than the size of an atom. Since the hierarchical reading of physics suggests that quantum mechanics is a ‘zoomed in’ account of reality, concerned with atomic scale phenomena, one might therefore expect macroscopic quantities such as the optical refractive
2 Toppling the Pyramids: Physics Without Physical State Monism
25
index – a property of solid objects made up of many billions of atoms – to disappear from the representation. However, this is not how quantum mechanics works in practice. Let us consider a couple of more detailed examples to illustrate how common such mixed scale quantities are in physics.
2.3.2
Quantum Mechanics of a Particle in a Box
The case of the ‘particle in a box’ is an elementary problem in basic quantum mechanics (Griffiths & Schroeter, 2018). This problem concerns a microscopic point particle, such as an electron, that can move in only one direction, x. It is represented as a wavefunction ψ(x), the square magnitude of which tells us the probability of finding it at the point x. The wavefunction obeys the differential equation,
ħ2 d 2 ψ ð x Þ ¼ Eψ ðxÞ, 2m dx2
ð2:1Þ
which is the time-independent Schrödinger equation. Here ħ is the reduced Planck’s constant, m is the particle mass, and E is the particle energy. These quantities are concerned with either the microscopic properties of the electron (m and E), or the quantum scale (ħ). At first sight it seems we are dealing with a microscopic theory of a point particle; the energy, mass, and reduced Planck’s constant may be phenomenological quantities, but if they are, they arise from a more ‘zoomed in’ theory of the electron. The problem is to find the allowed wave functions ψ(x) and energies E of the particle. Significantly, Eq. (2.1) has no physical content, on its own. There are an infinite number of solutions: propagating waves (positive energy) E>0:
pffiffiffiffiffiffiffiffiffi 2mE ψ ðxÞ ¼ exp i x , ħ
ð2:2Þ
and exponentially diverging solutions (negative energy) E LÞ ¼ 0,
ð2:4Þ
In this example, it is known with certainty that the particle is somewhere between x ¼ 0 and L. Imposing the boundary condition (2.4) restricts the energy to be both positive, and quantized: E ¼ n2π 2ħ2/2mL2 (n is an integer), and constrains the wavefunction to be of the form rffiffiffi 2 nπx ψ ð xÞ ¼ sin , L L
ð2:5Þ
RL which is now normalized so that all probabilities add to unity 0 jψ ðxÞj2 dx ¼ 1. It is only now that we are in a position to make any predictions that concern actual experiments: the quantized energy levels tell us the possible energies (frequencies) of light that can be emitted when the particle makes a transition between energy states, and the wave functions (2.5) tells us where the particle is likely to be found if the system is experimentally probed. The boundary conditions (2.4), however, are a proxy for an interaction between the microscopic particle and another macroscopic system. This other system is not treated using a lower-level theory like quantum mechanics, which is evident from the absence of any probabilistic aspect to Eq. (2.4). A reasonable approximation to the boundary conditions (2.4) could be realised by applying a spatially varying electric field to an electron (zero field between x ¼ 0 and L, rapidly increasing to a very large value at the edges). But this electric field would have to be produced by a large number of charges for it to be treated as a classical (non-probabilistic) charge density distribution, and thus give rise to these definite boundary conditions (2.4). Moreover, it is doubtful whether there is a self-consistently quantum mechanical way of explaining how classical properties emerge from a purely quantum substrate (see Sect. 2.5.2). From a naïve perspective, it may seem obvious that the time-independent Schrödinger Eq. (2.1) represents a microscopic theory of small objects, which does not depend upon any quantities from higher-level, macroscopic physics. However, in order to use Eq. (2.1) in any scientific practice that makes predictions we have to supplement it with additional boundary conditions, and in even the simplest of quantum systems these boundary conditions are specified by macroscopic, classical variables. It is far from obvious that these classical properties are governed by the unitary Schrödinger equation (Drossel, 2017, 2020).
2 Toppling the Pyramids: Physics Without Physical State Monism
2.3.3
27
The Arrow of Time in Electromagnetism
Let us consider another example from physics where macroscopic boundary conditions are essential to the physical content of a model. This example concerns the problem of the arrow of time in electromagnetism. Maxwell’s equations are perhaps the most successful set of laws in physics. They govern the behaviour of the electric E and magnetic B fields in space and time. These fields are produced by distributions of electric charge density ρ, or current density j. In the case of an oscillating current jω (oscillating at frequency ω radians per second), the electric field is governed by the differential equation — 3 — 3 E k 20 E ¼ iωμ0 jω ,
ð2:6Þ
pffiffiffiffiffiffiffi where i ¼ 1, k0 ¼ ω/c, μ0 is the permeability of free space, and c is the speed of light. Equation (2.6) governs the emission and reception of electromagnetic waves of a fixed frequency; be that radio wave reception, or the emission of light from an atom. Suppose someone with mathematical ability, but no knowledge of physics, attempts to solve this equation. They will find that the field produced by the current can be written as the following set of integrals: EðxÞ ¼
μ0 c2 iω
Z
d 3 x0 ℝ3
Z
d 3 k k 20 jω ðx0 Þ kðk jω ðx0 ÞÞ ikðx 2 x0 Þ e : 3 k20 k2 ℝ3 ð2π Þ
ð2:7Þ
Although this formula appears to be a precise, analytic equation for the electric field, it fails to determine any quantity for the field at any position x. In the denominator of the second integral, which ranges over all possible real values of the vector k, there is a division by the quantity k 20 k2, and since somewhere in the integral k20 k2 ¼ 0, formula (2.7) thus involves division by zero. This renders the whole expression indeterminate. The resolution of this problem is the same as in Sect. 2.3.2. Equation (2.6) is not enough for the theory to have physical content; we must impose boundary conditions (Jackson, 1998). To do this, we can make the following replacement in the denominator of Eq. (2.7):
k20 k2
¼ ðk 0 jkjÞðk 0 þ jkjÞ ð2:8Þ ! ðk 0 jkj þ iη1 Þðk 0 þ jkj þ iη2 Þ,
where η1 and η2 are infinitely small real numbers that are taken to zero at the end of the calculation, thus recovering a solution to Eq. (2.6). With this modification we no longer end up dividing by zero when we evaluate (2.7). In principle these real numbers η1,2 could be different, but to simplify the discusson we assume that η1 ¼ η2 ¼ η.
28
W. M. R. Simpson and S. A. R. Horsley
The number η serves to specify the behaviour of the field at infinite distances, which in turn also specifies the arrow of time. It picks out one of several possible solutions to (2.6). For instance, if η > 0 then (2.7) predicts that the oscillating current density jω produces electromagnetic waves which travel outwards from the source to infinity. This is the behaviour of nature with which we are familiar. Meanwhile, if η < 0 then the time reverse of this behaviour is predicted, in which every source of waves becomes a sink. In this unfamiliar picture of nature, electromagnetic waves are continually arriving from infinity to focus onto electric currents and disappear (Weinstein, 2011). In such a world, mobile phone would receive calls from infinity, which would be transformed into the sound of your voice and absorbed by your vocal chords. A model that deploys the laws of electromagnetism is thus incomplete without the specification of a boundary condition. This boundary condition sets the arrow of time, which is observed to be such that the universe as a whole is evolving towards an ever more probable configuration. Just as a gas spreads out to fill a box rather than localising to one region of it, the electromagnetic field spreads out to fill space rather than concentrating in particular regions. This is the thermodynamic arrow of time, which accords with our observation of entropy increasing with time. The origin of this boundary condition remains unclear (see discussion in Earman, 2011a) and at present has no agreed derivation from a microscopic theory. What is evident is that the boundary condition is essential to the theory of electromagnetism (quantum or otherwise), whilst representing macroscopic behaviour described by classical thermodynamics. We are again using macroscopic parameters to constrain a microscopic theory of physics. The naïve induction argument for state monism proceeds upon a false premise.
2.4
The Role of Representations in Physics
The more sophisticated apologist for physical state monism may downplay the role of higher-level properties in specifying boundary conditions because they believe the physical state of a system, however large, can be uniquely represented by our best theory of the properties of matter (quantum physics), and that the complete physical content of this theory concerning some parochial system P can be derived in principle from its representation of a larger system that includes it S (P ⊂ S ). According to this argument, the need for boundary conditions in defining the physical state of a system is just a practical concern that has no bearing upon the interpretation of a theory in terms of a set of possible worlds, since the set of possible worlds for the smaller subsystem P is merely a subset of the set of possible worlds for the larger system S. From this perspective, the macroscopic boundary conditions for the particle in a box could be replaced in principle with a theory of a manyparticle system that includes the walls of the box (Fetter & Walecka, 2003). Of course, this more inclusive physical model would also require boundary conditions, but we can reiterate this process in principle until we arrive at the state of the entire
2 Toppling the Pyramids: Physics Without Physical State Monism
29
cosmos and the complete set of physically possible worlds. This uniqueness premise, however, is also difficult to square with scientific practice, as we shall now explain (Glick, 2016).
2.4.1
A Question of Context
At the heart of any quantum theory are the canonical commutation relations between conjugate quantities such as position and momentum – or the anti-commutation relations which hold between the Pauli spin matrices – which encode the Heisenberg uncertainty principle. Any quantum theory which specifies a quantum state defined in a Hilbert space, and a set of bounded self-adjoint operators corresponding to b i must realise the Weyl algebra associated with these relations. observables H , O When the operators on a Hilbert space conform to these commutation relations, they are said to be a Hilbert-space representation (or ‘representation’, for short) of these quantum relations. Suppose we believe that a physical theory delineates what is possible for a system to which it is applied, and agree that the possibilities for a quantum system are given by an expectation value assignment to its set of observables. Unitary equivalence is widely considered the standard of physical equivalence for representations of the commutation relations: if two representations are unitarily equivalent, there is some b that transforms one representation into the other U : H ! H 0 , unitary operator U such that both representations determine the same expectation values for the various b i0U ¼ O b i . However, both quantum field theory observables which they define U 1 O and quantum statistical mechanics (in its thermodynamic limit) generate a continuum of unitarily inequivalent representations (Ruetsche, 2002, 2003, 2006, 2013). The cause of this plurality of inequivalent representations is the necessity of using quantum models which have infinite degrees of freedom in order to describe many kinds of phenomena. Finite quantum systems admit only one concrete irreducible Hilbert-space representation. The Stone-von Neumann Theorem establishes that any pairs of distinct representations will be unitarily equivalent to the irreducible Schrödinger representation, since there is a unitary operator that transforms one into the other. Infinite quantum systems, however, admit infinitely many concrete Hilbert-space representations that fall outside of the scope of the Stone-von Neumann theorem (Ruetsche, 2011, chs. 2–3). The use of unitarily inequivalent representations is manifest in physical descriptions of: (i) relativistic effects in quantum field theory. A non-interacting quantum field can be described in terms of a Fock space in which creation and annihilation operators are defined, which attributes different numbers of ‘quanta’ (or, different levels of ‘excitation’) to the various possible states of the quantum field. In the vacuum state of a quantum field, no quanta have been created. The Unruh Effect, however, predicts that an observer who is accelerating uniformly
30
W. M. R. Simpson and S. A. R. Horsley
within a quantum field in the vacuum state of the Fock space associated with unaccelerated motion, would observe a thermal bath of Rindler quanta instead of a vacuum state (Earman, 2011b). The two representations of the quantum field that feature in this physically significant prediction are unitarily inequivalent. Likewise, unitarily inequivalent representations are also implicated in descriptions of interacting quantum fields and the interpretation of scattering theory (Earman & Fraser, 2006). (ii) symmetry breaking, and systems close to a critical point. Quantum statistical mechanics provides a quantum mechanical underpinning for bulk properties associated with macroscopic physical systems, such as their temperature, pressure, and entropy. In the thermodynamic limit, a system with N particles and volume V is replaced by one in which the parameters N and V are taken to infinity, whilst the density N/V is kept constant. This procedure furnishes a model in which thermodynamic quantities, such as pressure and energy, can be represented as closed functions of thermodynamic variables, such as temperature and density, by treating the matter of the system as a continuum which has infinite degrees of freedom. Quantum theories that describe cases of spontaneous symmetry breaking and phase transitions in thermodynamic systems, however, deploy physically significant operators that are defined within unitarily inequivalent representations (see Sect. 2.4.2) (Ruetsche, 2011). (iii) systems in, or close to, thermodynamic equilibrium. Consider two macroscopic systems which are in thermodynamic equilibrium with one another. Such materials both radiate and absorb electromagnetic energy, and can experience attractive (or repulsive) forces that are mediated by quantum fluctuations in the electromagnetic field (see Sect. 2.4.3). In order to represent the phenomenon of absorption within the theory of quantum electrodynamics, which occurs in any realistic physical system, a heat sink must be incorporated with uncountably many more degrees of freedom than the electromagnetic field (Philbin, 2010; Horsley & Philbin, 2014). Such systems do not admit a unique representation of the ground state of the electromagnetic field, and incorporate classical properties which have top-down causal powers: the temperature of the system, for example, affects both the amplitude of the field and the magnitude of the microscopic currents in the media. Let us consider two examples of quantum systems in more detail in which the choice of physical representation is context-dependent.
2.4.2
Representing Phase Transitions
When a material undergoes a phase transition, certain classical properties of the material undergo change – typically, discontinuous change – due to some macroscopic change in their immediate external conditions. For example, an iron bar that is at thermal equilibrium, above a critical temperature T Tc, exhibits a paramagnetic
2 Toppling the Pyramids: Physics Without Physical State Monism
31
phase, in which it experiences no net magnetization, and below this critical temperature exhibits a ferromagnetic phase, in which it experiences spontaneous magnetization. In the presence of an external magnetic field, the ferromagnet admits two possible metastable states which are characterised by opposite magnetic polarisations.4 The statistical physics of finite systems, however, identifies equilibrium states with unique Gibbs states (Ruetsche, 2011, p. 3), implying that the phase available to a system at temperature T is unique for all T. This is contrary to what we observe in experiments. As Ruetsche (ibid.) points out, it is ‘only in the thermodynamic limit [that] can one introduce a notion of equilibrium that allows what the Gibbs notion of equilibrium for finite systems disallows: the multiplicity of equilibrium states at a finite temperature implicated in phase structure’. We may think of the simple ferromagnet in our example as consisting of magnetic moments arranged in a 1d lattice; eg. atoms with spin 1/2. To build a quantum theory for this composite system, we begin by assigning an atom at location k in the lattice a Pauli spin, σ k ¼ σ kx , σ ky , σ kz , where k ranges from N to N, which satisfies the canonical anti-commutation relations: h i k k k σ kx , σ k0 y ¼ iδkk0 σ z , etc: σ σ ¼ 3I:
ð2:9Þ
This expression evaluates as zero for k 6¼ k0, as measurements of different sites commute with one another. One way to set up a Hilbert space for a system in the thermodynamic limit N ! 1, without departing from the tradition of using separable Hilbert spaces in quantum mechanics, is to begin with a ground state characterised by a sequence sk ¼ + 1 for k 2 ℤ, then add all the sequences that can be obtained by making finitely many local modifications to this sequence which replace some of the entries with 1. Let us label this Hilbert space H þ . A set of operators b σ zj may then be introduced such that sequences sk whose j th entry is 1 correspond to the eigenvector in the Hilbert space associated with the eigenvalue 1. b can be defined with A magnetic polarisation observable for the composite system m the components: b Ni ¼ m
k ¼N X 1 b σk , 2N þ 1 k¼N i
i 2 fx, y, zg,
ð2:10Þ
which has a limit N ! 1 in the weak topology of H þ, and is thus an element of the observable algebra of H þ. Let [sk]j 2 {1, +1} denote the j th entry in the sequence b Nz , in the basis sequence sk, is: sk. The expectation value of m
4 We mostly follow Ruetsche’s (2003, 2006, 2013) philosophical discussion, but Schwabl’s (2006) physics.
32
W. M. R. Simpson and S. A. R. Horsley k ¼N X 1 ½s : 2N þ 1 k¼N k j
ð2:11Þ
b will be oriented along the z axis and will For every state, the expectation value of m take the value of +1 in the thermodynamic limit, breaking the rotational symmetry of the system’s dynamics, since only a finite number of spins take the value 1. But what about states that break this symmetry in the opposite direction? For these cases, we begin with a ground state characterised by a sequence sk ¼ 1 for k 2 ℤ, adding all the sequences that can be obtained by making finitely many local modifications which replace some of the entries with +1. Let us label this alternative Hilbert space H . It is obvious that our two representations of the ground state of the ferromagnet are inequivalent. The proof may proceed by contradiction. Suppose there is some unitary transfor1 mation, U : H þ ! H , such that Ub σþ ¼b σ nU n for all n, which implies that b N ¼ U m b Nþ U 1 , and suppose that jψ +i and jψ i are unit vectors in the Hilbert m spaces H þ and H respectively. Assuming that these two vectors are related by the transformation jψ +i ¼ U j ψ i, it follows that þ Nþ þ N
ψ jmz jψ ¼ ψ jmz jψ :
ð2:12Þ
However, this identity does not hold in the thermodynamic limit N ! 1, since the right hand side evaluates as +1 and the lefthand side as 1. These representations are not physically equivalent, and Hamiltonians defined on these two models describe physically different dynamical situations. Nonetheless, both of these states are physically significant. In the 1d Ising model of ferromagnetism, the pairwise interactions between the different spin sites contribute to the total Hamiltonian (energy) of the system (Schwabl, 2006, pp. 287–307): H ¼ J
X
0
σk σk ,
ð2:13Þ
k, k 0
where J is a positive real number that depends on the distance between the spins. This model can be extended to include an external magnetic field B, which is directed along the preferred axis of the uniaxial magnet: H 0 ¼ J
X X 0 σk σk b σk , k, k 0
k
1 b ¼ gμB B, 2
ð2:14Þ
where μB is the Bohr magneton, which expresses the magnetic moment of an electron caused by its spin, and g is a dimensionless factor. Such a system energetically favours states in which the spins are all aligned with the field. In the H þ representation, this is the ground state that corresponds to the sequence sk ¼ + 1 for all k. Likewise, in the H representation, this is the ground state corresponding to
2 Toppling the Pyramids: Physics Without Physical State Monism
33
sk ¼ 1 for all k. Both the state in which the magnetic polarisation of the composite system is aligned in the positive direction +1, captured by the H þ representation, and the state in which it is aligned in the opposite direction 1, included in H , are available to the system as metastable states below the critical temperature. If the physical content of a theory is simply the set of possible worlds determined en bloc by its laws, then these two quantisations of the system constitute rival theories that contradict one another about what is physically possible. In practice, however, physicists frequently deploy inequivalent representations in their models of a system where the correct choice is context-dependent. It does not follow that they take seriously the infinite number of particles apparently introduced in the thermodynamic limit N ! 1 (Ruetsche, 2003). Rather, it is the plurality of equilibrium states implicated in phase structure that matters. Significantly, in cases like these, we find ourselves relying upon macroscopic properties to determine the appropriate representation ðH þ , H Þ.
2.4.3
Representing Chemical Forces
Let us now consider the case of the van der Waals force between two dipoles at positions r1 and r2, where each dipole is comprised of electrons in motion around a positive ion, and each has an overall neutral charge. The chemical bonds between molecules have sometimes been described as being nothing but a sum of van der Waals forces that is too large to compute in practice, but is explicable in principle in terms of a microscopic quantum field theory. This reductionist claim, however, glosses over some highly significant details. The Casimir-Polder expression for the energy of a dipole-dipole system, derived using normal-mode quantum electrodynamics, is of the form U/
α1 α2 , d7
ð2:15Þ
which depends upon the microscopic polarisabilities of the two dipoles, α1 and α2, that describe their linear response to the electromagnetic field, and upon the distance d between them (Casimir & Polder, 1948). It is sufficient for our purposes to note that this interaction energy can be derived using a quantum mechanical model of the system whose dynamics is characterised by a Hamiltonian function, and that it determines a mechanical force which arises from quantum fluctuations that occur within the system even at zero temperature. These fluctuations polarise the otherwise neutral dipoles and cause them to attract one another.5 However, standard text-book expressions of the van der Waals force, which are highly idealised, tacitly depend upon assumptions about the macroscopic
5
For an introduction to quantum fluctuation forces, see Simpson and Leonhardt (2015).
34
W. M. R. Simpson and S. A. R. Horsley
environment of the system, which are made explicit in models that make more realistic predictions. For example, the Casimir-Polder expression assumes the idealised geometry of free space, but a more general expression can be derived for an environment with an arbitrary geometry, from which the Casimir-Polder expression may be recovered as a limiting case. In reality, the electromagnetic field is not free but is scattered by materials comprising the surrounding environment. This scattering affects the micro-forces experienced at a given point in space, which determines the strength of the dipole-dipole interaction. In addition to scattering the fields, the surrounding materials serve as a heat sink that absorbs the fields, and are characterised by thermal, macroscopic properties. As electromagnetic waves propagate through these materials, they displace electromagnetic charges, and in so doing, they induce electric currents in the materials. Significantly, the dispersive (scattering) properties of a material are directly related to its dissipative (absorbing) properties:6 for macroscopic media they are typically modelled by complex-valued electric permittivity E(ω) and magnetic permeability μ(ω) functions, in which the real part of the function quantifies the way the field is dispersed by the medium, and the imaginary part the rate at which it is absorbed. Let’s consider this absorption mechanism in a little more detail. It is convenient to build the quantum model for such a system by beginning with a classical Lagrangian density broken up into a sum of three parts: L ¼ LF þ LH þ LI ,
ð2:16Þ
where L F is a contribution due to the electromagnetic field, L H is a contribution due to the surrounding materials, and L I accounts for the interaction between the matter and the fields. The part due to the field is the same as in empty space. In order to represent the phenomenon of absorption within the theory of quantum electrodynamics, physicists have modelled the heat sink of the materials as an infinite continuum of oscillators (see Philbin, 2010, then Horsley & Philbin, 2014). For a one-dimensional system: 1 LH ¼ 2
Z 0
1
" # 2 ∂X ω ðx, t Þ 2 2 ω X ω ðx, t Þ dω: ∂t
ð2:17Þ
This is a system with uncountably many more degrees of freedom than the field: at each point in space there is a continuum of oscillators of amplitude Xω. Even after the quantised field has been ‘renormalised’ to remove the divergences familiar to conventional quantum field theory, the thermalised system retains infinitely many degrees of freedom. Nonetheless, the inclusion of every possible natural frequency of oscillation at each point in space (2.17) is necessary to reproduce the wave equation in an absorbing material. Each oscillator in the heat sink contributes an
6
The real and imaginary parts are connected by the Kramers-Kronig relations.
2 Toppling the Pyramids: Physics Without Physical State Monism
35
b ω to the total polarisation of the medium, where α(ω) is a function that amount αðωÞX is related to the electric permittivity of the medium E(ω). For a system in which μ ¼ 1 and the electric field is Ez, the interaction of the medium and the field is given by Simpson and Leonhardt (2015, ch. 4): Z
1
L I ¼ Ez
αðωÞX ω ðx, t Þdω:
ð2:18Þ
0
The equation of motion for the oscillators in the heat sink is: 2 ∂ þ ω2 X ω ¼ αðωÞEz , ∂t
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2ωE0 Im½EðωÞ : αðωÞ ¼ π
ð2:19Þ
This results in a current in the medium: ∂ Jz ¼ ∂t
Z
1
αðωÞX ω dω:
ð2:20Þ
0
Energy does not flow out of the electromagnetic field and into the materials without coming back again. In fact, the materials heat up due to the current generated by the field, then radiate the energy they have absorbed, thus contributing to the surrounding field. In this environment, the quantum energy levels of the microscopic molecules comprising the dipole-dipole system are thus permanently ‘dressed’ by the thermalised electromagnetic field, due to their interaction with their immediate environment. In the general theory, the assumption of thermodynamic equilibrium is made explicit, in which the temperature of the total system is introduced as a ‘classical’ variable that affects both the amplitude of the fields and currents. Even for the case of fluctuation-induced forces at zero temperature, however, such as the van der Waals force between two dipoles in a vacuum, which can be recovered from the general theory as a limiting case, the ground state of the total system remains one in which the thermal properties of the materials and the electromagnetic properties of the fields are inextricably coupled. The ground state of the thermalised field in the general description is not the ground state of the free field, but belongs to a unitarily inequivalent representation. Fundamentally, it is the ground state of a polariton field, in which the thermalised matter and the electromagnetic fields are mixed (Simpson, 2014, 2015). It would be question-begging, then, to infer the reduction of thermalised chemical systems to aggregates of microscopic physical constituents by appealing to the reducibility of intermolecular forces to quantum fluctuation forces, when our best description of any of the forces that we can measure is one in which macroscopic features of the system are already implicated. To accomplish such a reduction, we would have to demonstrate how to remove these background features, by showing that higher-level properties like temperature can be consistently represented within a quantum mechanical model that admits a unique microphysical representation,
36
W. M. R. Simpson and S. A. R. Horsley
which applies to physical systems whether or not they are in thermodynamic equilibrium.
2.5
Saving the Macroscopic
According to physical state monists, the content of a physical theory is simply the set of worlds that are possible according to that theory’s laws, whilst our best physical theory is typically taken to describe some fundamental set of lower-level constituents which compose the whole cosmos. Since the arrangements of these lower-level constituents are determined solely by lower-level laws, the causal powers of any physical system to change its physical properties, however large or small, are supposed to supervene upon the lower-level laws and the lower-level state of the cosmos, as described by our best physics. We have considered a number of case studies in contemporary physics, however, in which macroscopic properties are seen to play an essential role in specifying the boundary conditions of a system and in characterising its environment in a way that is relevant to determining a theory’s physical content. Instead of progressively ‘zooming in’ upon some lower level of reality that admits a unique microphysical representation, physicists deploy physically inequivalent representations of a system’s states and observables in order to capture a wide variety of phenomena in different macroscopic contexts. This raises an important question for standard philosophical approaches to the interpretation of physics: if we cannot specify the physical content of a theory without invoking macroscopic properties of the system it describes, and if the dynamics that governs the micro-physics depends upon its macroscopic context, why should we be obliged to choose between microphysicalism or priority monism in carving up nature? We suggest that this lack of independence of the microscopic world from the macroscopic world, and this element of context-dependence, should encourage rather the development of *a* pluralistic parsing of physics, which admits the existence of fundamental entities at intermediate physical scales, and where the physical possibilities of nature are not determined en bloc by lower-level laws. In what follows, we consider three arguments against physical pluralism, each of which begs the question in what it sets out to prove. We argue that contemporary physics is compatible with a strong form of emergence that affirms the existence of higherlevel properties and top-down causal powers.
2.5.1
Quantum Entanglement and Reductionism
A key part of the apologetic for priority monism in contemporary metaphysics, which claims that the whole of physical reality is reducible to modifications of a single cosmic substance, is the argument offered by Schaffer (2010, section 2.2) that
2 Toppling the Pyramids: Physics Without Physical State Monism
37
appeals to the phenomenon of quantum entanglement. Since this claim is logically inconsistent with physical pluralism, such an argument counts as a reason for rejecting pluralism. It can be reconstructed as follows: 1. Quantum-entangled systems that evolve unitarily according to lower-level quantum mechanical laws are fundamental physical wholes. 2. The cosmos is a quantum-entangled system that evolves unitarily according to lower-level quantum mechanical laws. 3. Therefore, the cosmos is a fundamental physical whole. An entangled system of particles is one in which the outcomes of measurements of the particles are correlated in such a way that the quantum state describing this system cannot be decomposed into the product of spatially separate states associated with each particle (Bell, 1964). The fact of their ‘non-separability’ offers a prima facie reason for considering the entangled system to form an irreducible whole. Let us accept the first premise, for the sake of argument. Regarding the argument’s second premise, Schaffer (2010, p. 52) claims that ‘one gets initial entanglement from the assumption that the world begins in one explosion (the Big Bang) in which everything interacts. This initial entanglement is then preserved thereafter on the assumption that the world evolves via Schrödinger’s equation’. In other words, to get from the fact that some particles are quantum entangled in some situations to the inference that everything is mutually entangled and comprises an irreducible whole, the cosmos must have a universal wave function defined in a single Hilbert space that evolves unitarily according to lower-level laws, and must therefore admit a single unique representation. Every higher-level system must be reducible to this lower-level system. The reduction relation, in this case, is cashed out in terms of metaphysical grounding. Schaffer admits macroscopic objects within his ontology, but such objects are not fundamental, being grounded in the cosmos as a whole. The conclusion of his argument is inconsistent with the kind of physical pluralism whose scientific respectability we are seeking to defend. But why should we suppose the physical cosmos to be reducible to a single quantum-entangled system? Presumably, because any higher-level theory’s laws are said to be reducible to a lower-level theory’s laws. The Nagelian-Schaffner model of reduction, however, implicitly assumes the truth of physical state monism, by trivialising the role of higher-level properties that we find indispensable for generating physical models that make testable predictions. There are two problems with this claim. First, there is reason to doubt that there is a cosmic state that evolves unitarily according to Schrödinger’s equation, because of the role that macroscopic, classical properties appear to play in modifying the behaviour of quantum systems. We shall discuss this difficulty concerning the quantum dynamics in Sects. 2.5.2 and 2.5.3. Secondly, it is not sufficient to establish a nomological translation between two physical theories to show that one theory can capture all of the physical possibilities described by the other theory, and hence derive all of its physical content. The physical possibilities of a quantum system are supposed to be given by assignments of expectation values to sets of observables. Yet thermal systems, as
38
W. M. R. Simpson and S. A. R. Horsley
we have seen, require unitarily inequivalent Hilbert-space representations of the canonical commutation relations, which are disjoint in their assignments of expectation values (see Sect. 2.4.2), and two quantum states that belong to unitarily inequivalent representations are ipso facto not quantum-entangled. Yet if quantum states of physical systems are properly represented by Hilbert space representations, and if there is no unique representation in which a universal wave function for the cosmos can be defined, then the cosmos is not a single quantum-entangled system which evolves unitarily according to lower-level laws. In that case, the second premise of the argument for priority monism is false, and the conclusion that the cosmos is the only fundamental entity in existence has not been established. There is no obvious way to dismiss this troubling element of pluralism in the heart of our best quantum theories. On the one hand, to privilege the physical content of one particular Hilbert space representation – a move that Ruetsche (2011) calls ‘Hilbert space conservatism’ – would be to reduce the number of physically significant states to a subset of those that are generally accepted by scientific practices. On the other hand, to confine the physical content of a quantum theory to the algebraic structure shared by different Hilbert space representations – ‘algebraic imperialism’ – would be to reduce the number of physically significant observables, since only a proper subset of the bounded self-adjoint operators that are deployed in quantum mechanical explanations instantiate the Weyl algebra. Either move fails to support the explanatory agenda of our best quantum theories and is inconsistent with adopting a realist stance toward theories in virtue of their explanatory successes. Suppose instead of relying upon the algebraic approach to quantum field theory that is typically invoked by philosophers to discuss questions of interpretation, we focus on the ‘conventional’ Lagrangian version deployed by working physicists. According to Wallace (2011), the problem of unitarily inequivalent representations is circumvented in conventional quantum field theory by the introduction of cutoffs which limit its application to systems with only finite degrees of freedom. In this context, we might try to interpret the thermodynamic limit as merely a bridging law between a higher-level theory that describes thermal quantum systems, and a lowerlevel theory that describes a system in terms of a many-particle wave function with only finitely many degrees of freedom (Butterfield, 2011) (see Sect. 2.5.2). However, as Fraser (2009) points out, for a quantum field in spacetime to be considered to have only finite degrees of freedom, it would have to model spacetime as both finite and discrete. To waive this theoretical cost, conventional quantum field theory would have to be viewed as an effective field theory that is limited in its application within a specific energy scale, and thus unsuited to specifying the fundamental lower-level laws of nature, much less to determining all the physical possibilities. In trivialising the role of boundary conditions in determining physical possibilities, it seems the Nagelian-Schaffner theory of reduction is not only insensitive to scientific practices, but undermines scientific realism. As Ruetsche (2011, p. 5) observes: ‘there often isn’t a single interpretation under which a theory enjoys the full range of virtues realists are wont to cite as reasons for believing that theory’. Yet perhaps the decisive factor in determining whether there are macroscopic features of the world which are on an ontological par with its microscopic features is
2 Toppling the Pyramids: Physics Without Physical State Monism
39
the impossibility of recovering certain macroscopic features of the manifest image upon which we depend for the possibility of scientific inquiry from a lower-level quantum mechanical description. We consider this element of top-down dependency further in the following two sections.
2.5.2
Quantum Darwinism and Weak Emergence
A second argument against physical pluralism could be based on the ‘Quantum Darwinist’s’ claim that the higher-level classical properties of a physical system only weakly emerge from a lower-level quantum substrate which evolves according to lower-level laws. For weak emergentists, the causal powers conferred by the higherlevel properties of a system are a subset of the causal powers of the lower-level properties comprising the emergence base. These powers are generated by abstracting from the details of the lower-level physics, rather than something over and above the lower-level (Bedau, 1997). Since the conclusion of this argument undermines one of the key reasons for admitting higher-level entities into the fundamental ontology, in virtue of their possessing top-down causal powers, it counts as a reason for rejecting the kind of physical pluralism we are defending. The argument might be formulated as follows: 1. If higher-level properties weakly emerge from a lower-level substrate, their powers are subsets of the powers conferred by the lower-level properties. 2. If the powers of higher-level properties are subsets of the powers of lower-level properties, they do not confer top-down causal powers. 3. Higher-level properties weakly emerge from a lower-level substrate. 4. Therefore, higher-level properties do not confer top-down causal powers. We take the first premise to be true by definition: according to weak emergentists, ‘less is different’ (Butterfield, 2011).7 In other words, the novel behaviour that is associated with the phenomenon of emergence is to be explained in terms of what we have left out of the description and our own epistemic limitations, rather than in terms of an expanded ontology. In support of the second premise, it may be argued that, if the causal powers conferred by a system’s higher-level properties are only a subset of the system’s lower-level powers, generated by a coarse-grained description of the system in question, then ‘top-down causation’ is merely part of a phenomenological description of nature, and should not be part of any fundamental description. In defence of the third premise, Quantum Darwinists claim that one can eliminate any element of context-dependence by treating a thermal system as an open quantum system embedded in a cosmic environment, and the combined system that includes
7
To be contrasted with regard to the characterisation of emergence famously offered by Anderson (1972), for whom ‘more is different’.
40
W. M. R. Simpson and S. A. R. Horsley
its environment as an isolated, non-thermal quantum system. On this view, the universal reality underlying the quantum statistical mechanics of thermalised systems admits a unique physical representation in the form of a many-particle wave function, which evolves unitarily according to the Schrödinger equation, whilst the macroscopic property of temperature weakly emerges as part of a coarse-grained representation of the environment with which the open system is interacting. The quantum substrate from which macroscopic systems are said to emerge is described solely in terms of lower-level quantal properties. Nothing more is needed. The conclusion of the argument logically follows from the premises. It is the third premise that we wish to challenge. A difficulty arises in specifying the boundary between a microscopic quantum system that is to be measured by a macroscopic system from any macroscopic system that is used to measure it. According to advocates of the Quantum Darwinism programme, the macroscopic objects with higher-level classical features that scientists depend upon for making measurements are supposed to emerge from a lower-level quantum substrate that can be characterised without invoking macroscopic properties, due to the physical process of decoherence (Zurek, 1982, 2003).8 The theory of decoherence is supposed to explain how the reduced density matrix of a quantum system, which encodes all of the statistical information that can be extracted from the system by an observer, evolves from being a superposition of components that can ‘interfere’ with one another in a non-local way (due to the welldefined phase relations between these components), to a ‘mixed state’ in which the effects of interference between any of these components becomes negligible (due to the leakage of this phase coherence into the environment). The reduced density matrix represents the state of a system obtained from the composite state of a system and its environment þ by ‘tracing out’ the environment of the system (that is, by averaging over the environmental degrees of freedom). The goal of decoherence theory is to explain the vanishing of the ‘off-diagonal’ terms in the reduced density matrix, which corresponds to the vanishing of interference. According to the theory of environmentally induced superselection (also known as einselection), a large system such as a measuring device is steered towards having those sensible and determinate ‘pointer observables’ that are familiar to ordinary experience due to its continual interaction with an environment with many more degrees of freedom than itself. Such interactions are typically modelled as scattering processes, in which the system interacts with a swarm of surrounding particles, and they favour states that are well localised in position, since the interaction Hamiltonians describing such processes typically embody a force law that is a function of position. In the case of such a composite system þ , the off-diagonals in the reduced density matrix of are periodic functions that oscillate as a function of time. To secure the disappearance of interference, for all practical purposes, the off-diagonal function will have to be small in value, and have a long recurrence time before the process of recoherence is allowed to take place.
8
For a more complete account of QD see Zurek (2009).
2 Toppling the Pyramids: Physics Without Physical State Monism
41
Technically speaking, the diagonalised reduced density matrix of a system that has ‘decohered’ is an improper mixture, since the system that it describes remains quantum-entangled with its environment. It is simply that the phase coherence between its components that endangers classical behaviour at the macroscopic level has ‘leaked’ into the environment through the process of decoherence in such a way that there are no local observations that can reveal it. Whilst the reduced density matrix of a system (in which the off-diagonals have effectively vanished) predicts the correct measurement statistics, quantum events are not always averaged when considering the behaviour of macroscopic systems, and individual events can change the behaviour of macroscopic systems (consider, for example, the case of an atom undergoing radioactive decay). Consequently, decoherence theory does not obviate the need for an interpretation of quantum mechanics that can explain the existence of measurement outcomes. This fact is generally acknowledged. Nonetheless, decoherence theory remains the linchpin of the Quantum Darwinist programme, which seeks to offer an observer-independent explanation of the emergence of a ‘classical’ world of objects from the unitary quantum dynamics. These macroscopic objects mirror, for the most part, the behaviour predicted by classical NewtonMaxwell physics. Yet there is good reason to think that the process of decoherence depends upon the existence of a fundamental division between the microscopic and macroscopic (Kastner, 2014), and thus the theory of decoherence presupposes what the Quantum Darwinist programme sets out to derive. Specifically, in order for the off-diagonal elements in the reduced density matrix of the system to vanish, and for the recoherence time of the system to be sufficiently long to justify neglecting non-local effects, we must assume the initial states of the subsystems comprising its environment and their associated couplings with the system to be random. Yet the assumption of randomness is inconsistent with any purely quantum mechanical framework in which the universe is modelled as a closed system with a universal wave function that evolves unitarily according to Schrödinger’s dynamics. Of course, no observed system is really closed, and the assumption of randomness is consistent with what we observe. Every system that scientists probe in their experiments is interacting with a much larger environment that appears to be random. However, as Kastner (2014, p. 57) points out, ‘the “openness” of the system is not actually available in the. . . unitary-only picture’ of the Quantum Darwinist. In the case of the composite system, þ P þ , in which P is a macroscopic system corresponding to a measuring device that is monitoring a microscopic system S, the total system þ P þ is a closed system that is characterised by a single quantum state which evolves deterministically, however things may otherwise appear. In such a picture everything is always coherently entangled. Thus the Quantum Darwinist ‘can only make an arbitrary division into system+pointer+environment’ (ibid.), since there is no objective division that can be supported self-consistently from within the unitary-only picture. In the absence of any natural division of the world into microscopic systems, their measuring devices, and their environments, however, Quantum Darwinists cannot appeal to the appearance of randomness – as if the environment were in fact
42
W. M. R. Simpson and S. A. R. Horsley
objectively carved up into largely uncorrelated classical systems – without crucially begging the question. Hence the appeal to decoherence in this context is viciously circular, as Kastner and others have argued (see also Fields, 2010, 2011). Kastner (2014, p. 58) writes: ‘The problem is not so much a lack of observer-independence as it is failure to account for the initial independence of the environment from the system’, which is assumed by the adoption of random phases between the sub-systems of the environment.
2.5.3
Causal Closure and Strong Emergence
There is a third misgiving a modern philosopher may entertain about admitting pluralist parsings of physics, which concerns the violation of the causal closure of the microphysical world entailed by stronger forms of emergence. Strong emergentists can be distinguished from weak emergentists by their denial of the unimodal conception of physical content that unites physical state monists, in which the laws of our best physics determine the set of physically possible worlds en bloc. According to strong emergentists, there are systems with higher-level causal powers that influence the temporal development of lower-level properties, which do not merely supervene upon lower-level laws and the lower-level state of the cosmos, and the physical content of higher-level theories often is not derivable from the content of lower-level theories (Noble, 2011). According to contemporary physics, however, there are only four physical ‘forces’ in nature: gravitation, electromagnetism, the weak interaction, and the strong interaction. If there are higher-level properties which are neither reducible to lower-level properties, nor weakly emergent, then how are higher-level properties supposed to make a causal difference to the lower-level physics without introducing ‘spooky’ forces into nature – such as the discredited notion of vital forces in biology – in addition to the four physical forces? An argument against strong emergence might be formulated as follows: 1. There are no more than four basic forces in nature. 2. If top-down causation occurs, then the lower-level is not causally closed. 3. If the lower-level is not causally closed, then there are more than four basic forces in nature. 4. Therefore, top-down causation does not occur. The first premise can be taken as a methodological assumption governing the practice of physics, and as reflective of the confidence of scientists in the power of physical models to explain and predict natural phenomena. Philosophers should not posit more physical forces than are admitted by our best physics. We take the second premise to be a statement of the obvious: the lower-level physics cannot be causally closed if there are higher-level properties in nature which have causal powers to change lower-level properties, and if these higher-level properties are not ultimately reducible to lower-level properties. The third premise makes a claim about the
2 Toppling the Pyramids: Physics Without Physical State Monism
43
conditions under which the lower-level physics might fail to be causally closed. If we reject the causal closure of the lower-level, then we must appeal to the existence of additional and mysterious forces that are unknown to our best physics, such as the discredited notion of ‘vital forces’ in biology, to accommodate the causal powers of macroscopic entities such as plants and animals. The conclusion follows from the premises. The third premise of the argument is false, however, since the consequent is a non sequitur. There is another way in which higher-level properties can make a causal difference to lower-level properties. According to Ellis (2012, p. 1896): ‘the upper levels exercise crucial influences on lower level events by setting the context and boundary conditions for the lower level actions’. This context-dependence can be secured by adopting an appropriate quantum dynamics, in response to the measurement problem of quantum mechanics. The measurement problem is an open problem in the interpretation of quantum mechanics precisely because of the role that macroscopic measurements play in modifying microscopic behaviour (Schlosshauer, 2005). As we discussed in Sect. 2.3.2, the fundamental mathematical object within quantum theory is the wave function; or, more properly, the quantum state.9 This mathematical object encodes the probability of an arbitrarily complicated system having any particular configuration that can be specified in whatever way we choose. Prior to any measurement – or any other collapse-inducing event – the wave function evolves according to the time-dependent Schrödinger equation b j ψi ¼ iħ ∂ j ψi , H ∂t
ð2:21Þ
b is the Hamiltonian of the system, representing its energy. Equation (2.9) has where H b a formal solution in terms of a unitary operator U, b ðt Þ j ψ ð0Þi, j ψ ðt Þi ¼ U
ð2:22Þ
i.e. the wave function at some arbitrary time t can be obtained from the wave function at time t ¼ 0 through the action of this unitary ‘time evolution’ operator b .10 The theory tells us how to start from a given state of a system, perhaps a U configuration of particles or a field, and evolve the probability amplitudes for all the possible configurations of the system in time. But now suppose we perform a ‘nondemolition’ measurement on the system – perhaps the number of photons in an electromagnetic wave (Xiao et al., 2008) – which does not destroy the quantum
Here we write the wave function as jψi, to indicate that this object is not necessarily a function of position. 10 The meaning of ‘unitarity’ is that the probabilities computed from jψi always sum to unity; the b simply re-distributes the probabilities between different possibilities as time goes on. operator U 9
44
W. M. R. Simpson and S. A. R. Horsley
system being measured. After this measurement we know more about the state of the system than the wave function (2.22) lets on. For example, the measurement outcome of the experiment may, with certainty, have ruled out some of the states to which jψi assigns non–zero probability. To obtain the correct results for future experiments, we must therefore update the wave function with the empirical knowledge we have gained. Yet this updating is not b For instance suppose at time t we find performed by the time evolution operator U. an electromagnetic field has n photons in it, jψi¼jni. The wave function has to undergo the following discontinuous modification: b ðt δt Þjψ ð0Þi, jψ ðt δt Þi ¼ U jψ ðt þ δt Þi ¼ jni:
ð2:23Þ
b ðt δt Þ j ψ ð0Þi, to jni) is This discontinuous change of the wave function (from U known as the ‘collapse of the wave function’, and it is necessary to properly account for any non-demolition experiment. There is no agreed understanding of this process (Omnès, 1994). Even if the phenomenon of decoherence is taken into account (as suggested in e.g. Omnès, 1994), the time evolution operator must still be supplemented with a discontinuous change of state. According to Bell (1987), any realist approach to quantum mechanics that seeks to explain the existence of determinate measurement outcomes, such as the number of photons in an electromagnetic field, must come to terms with a dilemma: either the dynamics of standard quantum mechanics is wrong, and the wave function evolves according to a non-linear Schrödinger equation that permits the wave function to collapse, or standard quantum mechanics is incomplete, and there are ‘hidden variables’ (ibid, p.1), in addition to the wave function, that evolve according to some non-linear dynamics of their own. Maudlin (1995) has argued that the choice comes down to two possibilities: either we should adopt something like the GRW theory, or something like the theory of Bohmian mechanics. The GRW theory seizes the first horn of the dilemma by incorporating a stochastic mechanism which produces random hits on the wave function that occur universally for microscopic particles and result in an objective collapse of the wave function (Ghirardi et al., 1986). The effects of this non-linear modification to the Schrödinger equation become significant when a large number of quantum-entangled particles are involved, such as the particles that compose a macroscopic instrument of measurement. The theory of Bohmian mechanics seizes the second horn of the nomological dilemma by positing a global configuration of particles whose trajectories are choreographed by the universal wave function (de Broglie, 1928; Bohm, 1951, 1952). The guiding equation for the particles depends in a non-linear way upon the universal wave function, which evolves according to the standard Schrödinger equation. The collapse of the wave function, according to Bohmians, is not an objective event, but merely an artifact of using an effective wave function to
2 Toppling the Pyramids: Physics Without Physical State Monism
45
model a system, instead of the universal wave function. The only relevant part of the universal wave function is the part that supports the particle configuration. An alternative ‘contextual’ model of the quantum dynamics is available, however, proposed by Drossel and Ellis (2018), in which the interaction of a quantum system with the intrinsic heat bath of a measuring instrument plays a key role in solving the measurement problem. In this system, the macroscopic, thermal properties of an instrument have the power to collapse the wave function of a microscopic system. The CWC model (contextual wave function collapse) drops the assumption of physical state monism that underpins Maudlin’s ultimatum, by incorporating the notion of top-down causal powers. As in the GRW modification of quantum mechanics, the CWC model seizes the first horn of Bell’s dilemma, allowing the wave function to become localised with respect to position. It also distinguishes measurements from localisation events. Unlike GRW theory, however, the stochastic corrections that achieve these localisations depend upon the local macroscopic context of a system, which includes the measuring device. In short, the CWC model incorporates a feedback loop – from a particle, via the intrinsic heat bath of the measuring device, back to the particle – which introduces non-linear terms in the Schrödinger equation. These terms are physically motivated: they can be accounted for in terms of thermodynamics and solid-state physics (Drossel & Ellis, 2018, pp. 13–19). As in Bohmian mechanics, the CWC model relies upon the effects of the environment upon the measuring process to explain why the outcomes of quantum experiments conform to standard quantum statistics and Born’s rule for quantum observables. Unlike Bohmian mechanics, the CWC model does not conceive any part of the environment that is relevant to the measuring process as a many-particle quantum system that is subject to unitary and reversible time evolution. Rather, the heat bath of an instrument is characterised as having non-zero temperature and only a limited ‘memory’, since it radiates irreversibly into the heat sink of its surroundings. Consequently, the CWC model does not leave the quantum system entangled with any part of its environment beyond the usual time scale of decoherence. According to the CWC model, the heat bath of the measuring instrument can serve as a bridge between quantum systems and their ‘classical’ environment, just so long as we are willing to reject ‘the untestable and implausible claim that the environmental heat bath can be described by an infinite-precision wave function that is subject to unitary time evolution’ (Drossel & Ellis, 2018, p. 4). The CWC model does not introduce any spooky new forces in nature, nor does it assume state monism by postulating the existence of a universal wave function defined in a single Hilbert space. According to Ellis (ibid., p.25): ‘Quantum theory per se does not tell us what Hilbert spaces to use. This requires the classical, macroscopic context’. In this model, the higher-level classical properties derive their causal powers from the role they play in specifying the boundary conditions within which the micro-evolution of a system takes place.11
11
Simpson (2021) has proposed an ontology of thermal substances for the CWC model.
46
2.6
W. M. R. Simpson and S. A. R. Horsley
Concluding Remarks
Microphysicalism and priority monism are often regarded as antagonistic positions in contemporary philosophy. In this paper, however, we have argued that they are both manifestations of a more entrenched orthodoxy that rules against fundamental ontologies that admit macroscopic entities. According to physical state monists, the causal powers of a physical system supervene upon the lower-level laws and lowerlevel state of the cosmos, as they are represented by our best physics, because the physical content of a theory is assumed to be specified by a set of possible worlds which are determined by its laws (Sect. 2.2). This presumption is widely in evidence in philosophy – in discussions of free will, for example, and mental causation. Against this orthodoxy, we have argued that there is a striking inconsistency between the pretensions of state monism and scientific practices. We considered models of quantum systems in which the boundary conditions are specified by higher-level, macroscopic properties (Sect. 2.3). It is far from evident that our best physics ‘zooms in’ upon a basic level that can be described solely in terms of lowerlevel laws and properties. We also considered models which apply physically inequivalent representation to the same system to explain their behaviour (Sect. 2.4). It is highly doubtful that physics provides a unique microphysical representation of a lower-level cosmic state, which evolves according to one set of physical laws, from which the physical content of any model of a macroscopic system could be derived. Rather, physical laws seem to be patchy and context dependent, as Cartwright (1999) has famously argued. We broached the possibility of some kind of physical pluralism instead, which admits fundamental entities between the microscopic and cosmic scales, and refuted three arguments that attempt to impugn the scientific respectability of pluralism (Sect. 2.5): an argument in favour of priority monism, on the basis of quantum entanglement; an argument in favour of weak emergence, which appealed to the phenomenon of decoherence; and an argument in favour of the causal closure of the lower-level world, which invoked the bogeyman of mysterious forces. Each of these arguments presupposes a unimodal conception of physical possibilty and begs the question against physical pluralism. There is good reason to think that the physical possibilities encompassed by our best physical theories are not specified en bloc by lower-level laws, and hence to reject a unimodal conception of physical content. The framework of algebraic quantum mechanics, as we have seen, admits two levels in specifying the content of theory: at the level of the algebra, which is common to different microscopic systems, and at the level of the Hilbert space representation, which takes into account macroscopic contingencies of particular systems, such as equilibrium temperature (Ruetsche, 2003). A more nuanced conception of content is therefore needed to capture the full range of explanatory virtues realists typically cite as reasons for believing this theory. There is also good reason to abandon the orthodox belief that the lower-level physics is causally closed. The measurement problem in quantum mechanics admits a well-motivated solution in which higher-level properties set the
2 Toppling the Pyramids: Physics Without Physical State Monism
47
context and boundary conditions for the lower-level behaviour, which is consistent with actual scientific practices (Drossel & Ellis, 2018). This contextual model of the quantum dynamics opens the conceptual space for a fundamental ontology that includes macroscopic entities with top-down causal powers. We note in closing that there are other ways of interpreting the concept of physical possibility than in terms of a flat set of possible worlds. In Inman’s (2017) neo-Aristotelian mereology, for example, possibilities are understood rather in terms of things’ natures. According to Koons (2021) and Simpson (2021), the fundamental entities are ‘thermal substances’, which have both quantal and classical properties, whose natures are defined at the macroscopic level. Our object in this paper has not been to argue for any particular fundamental ontology, however, but rather to question the scientifically privileged status of microphysicalism and priority monism, which has been taken for granted by many philosophers, and to question the view that the laws specified by our best physics determine a set of possible worlds, rather than simply constraining the possibilities. In the absence of any good reason for believing state monism to be the default position endorsed by our best physics, we think it is reasonable for philosophers to construct ontologies that include physical entities at a variety of scales, which have the freedom to exercise top-down causal powers. Such ontologies are likely to have far-reaching implications for our understanding of free will and agency (topics which Helen Steward and Daniel De Haan explore in their own chapters in this volume), as well as for artificial intelligence (Ellis, 2016; Ellis & Drossel, 2019). Acknowledgements The authors would like to thank (in alphabetical order) Barbara Drossel, George Ellis, Robert Koons, and Jonathan Schaffer for their critical feedback on this paper. William Simpson would like to acknowledge the support of the John Templeton Foundation via the international project, God and the Book of Nature, Grant Id: 61507, and to thank Marcus Ackermann for his assistance in reformatting the final version of this paper.
References Anderson, P. W. (1972). More Is Different. Science, 177(4047), 393–396. Bedau, M. A. (1997). Weak emergence. Philosophical Perspectives, 11, 375–399. Bell, J. S. (1964). On the Einstein Podolsky Rosen paradox. Physics Physique Fizika, 1(3), 195–200. Bell, J. S. (1987). Speakable and unspeakable in quantum mechanics: Collected papers on quantum philosophy. Cambridge University Press. Bishop, R. C., & Ellis, G. F. R. (2020). Contextual emergence and physical properties. Foundations of Physics, 50(5), 481–510. Bohm, D. (1951). Quantum theory. Prentice-Hall, Incorporated. Bohm, D. (1952). A suggested interpretation of the quantum theory in terms of “hidden” variables. I. Physical Review, 85(2), 166–179. Breuer, H.-P., & Petruccione, F. (2007). The theory of open quantum systems. Oxford University Press. Butterfield, J. (2011). Less is different: Emergence and reduction reconciled. Foundations of Physics, 41(6), 1065–1135.
48
W. M. R. Simpson and S. A. R. Horsley
de Broglie, L. (1928). La nouvelle dynamique des quanta [The new dynamics of Quanta]. In H. A. Loretz (Ed.), Électrons et photons: Rapports et discussions du cinquième Conseil de physique tenu à Bruxelles du 24 au 29 octobre 1927 sous les auspices de l’Institut international de physique Solvay (pp. 105–132). Gauthier-Villars et Cie. English translation: Bacciagaluppi, G., & Valentini, A. (2009). The new dynamics of Quanta. In G. Bacciagaluppi & A. Valentini (Eds.), Quantum theory at the crossroads: Reconsidering the 1927 Solvay conference (pp. 341–371). Cambridge University Press. Button, T. (2013). The limits of realism. Oxford University Press. Cartwright, N. (1999). The dappled world: A study of the boundaries of science. Cambridge University Press. Casimir, H. B. G., & Polder, D. (1948). The influence of retardation on the London-van der Waals forces. Physical Review, 73(4), 360–372. Chang, H. (2017). Is pluralism compatible with scientific realism? In J. Saatsi (Ed.), The Routledge handbook of scientific realism (pp. 176–186). Routledge. Dizadji-Bahmani, F., Frigg, R., & Hartmann, S. (2010). Who’s afraid of Nagelian reduction? Erkenntnis, 73(3), 393–412. Dorr, C. (2011). Physical geometry and fundamental metaphysics. Proceedings of the Aristotelian Society, 111(1pt1), 135–159. Drossel, B. (2017). Connecting the quantum and classical worlds. Annalen der Physik, 529(3), 1600256. Drossel, B. (2020). What condensed matter physics and statistical physics teach us about the limits of unitary time evolution. Quantum Studies: Mathematics and Foundations, 7(2), 217–231. Drossel, B., & Ellis, G. (2018). Contextual Wavefunction collapse: An integrated theory of quantum measurement. New Journal of Physics, 20(11), 113025. Earman, J. (2011a). Sharpening the electromagnetic arrow(s) of time. In C. Callender (Ed.), The Oxford handbook of philosophy of time (pp. 485–528). Oxford University Press. Earman, J. (2011b). The Unruh effect for philosophers. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 42(2), 81–97. Earman, J., & Fraser, D. (2006). Haag’s theorem and its implications for the foundations of quantum field theory. Erkenntnis, 64(3), 305–344. Ellis, G. (2012). On the limits of quantum theory: Contextuality and the quantum–classical cut. Annals of Physics, 327(7), 1890–1932. Ellis, G. (2016). How can physics underlie the mind? Top-down causation in the human context. Springer. Ellis, G., & Drossel, B. (2019). How downwards causation occurs in digital computers. Foundations of Physics, 49(11), 1253–1277. Fetter, A. L., & Walecka, J. D. (2003). Quantum theory of many-particle systems. Dover Publications. Fields, C. (2010). Quantum Darwinism requires an extra-theoretical assumption of encoding redundancy. International Journal of Theoretical Physics, 49(10), 2523–2527. Fields, C. (2011). Classical system boundaries cannot be determined within quantum Darwinism. Physics Essays, 24(4), 518–522. Fraser, D. (2009). Quantum field theory: Underdetermination, inconsistency, and idealization*. Philosophy of Science, 76(4), 536–567. Ghirardi, G. C., Rimini, A., & Weber, T. (1986). Unified dynamics for microscopic and macroscopic systems. Physical Review D, 34(2), 470–491. Glick, D. (2016). The ontology of quantum field theory: Structural realism vindicated? Studies in History and Philosophy of Science Part A, 59, 78–86. Griffiths, D. J., & Schroeter, D. F. (2018). Introduction to quantum mechanics (3rd ed.). Cambridge University Press. Harrison, P., & Valavanis, A. (2016). Quantum Wells, wires and dots: Theoretical and computational physics of semiconductor nanostructures (4th ed.). John Wiley & Sons, Incorporated. Hollowood, T. J. (2013). Renormalization group and fixed points in quantum field theory. Springer.
2 Toppling the Pyramids: Physics Without Physical State Monism
49
Horsley, S. A. R., & Philbin, T. G. (2014). Canonical quantization of electromagnetism in spatially dispersive media. New Journal of Physics, 16(1), 013030. Hüttemann, A. (2015). What’s wrong with microphysicalism? Routledge. Inman, R. D. (2017). Substance and the fundamentality of the familiar: A neo-Aristotelian mereology. Routledge. Jackson, J. D. (1998). Classical electrodynamics (3rd ed.). Wiley. Kastner, R. E. (2014). ‘Einselection’ of pointer observables: The new H-theorem? Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 48(1), 56–58. Kim, J. (1984). Epiphenomenal and Supervenient causation. Midwest Studies in Philosophy, 9(1), 257–270. Kim, J. (1997). Does the problem of mental causation generalize? Proceedings of the Aristotelian Society, 97(1), 281–298. Koons, R. C. (2021). Thermal substances: A neo-Aristotelian ontology of the quantum world. Synthese, 198(Suppl 11), 2751–2772. Leggett, A. J. (1992). On the nature of research in condensed-state physics. Foundations of Physics, 22(2), 221–233. Loudon, R. (2000). The quantum theory of light (3rd ed.). Oxford University Press. Maudlin, T. (1995). Three measurement problems. Topoi, 14(1), 7–15. Nagel, E. (1961). The structure of science: Problems in the logic of scientific explanation. Routledge and Kegan Paul. Noble, D. (2011). A theory of biological relativity: No privileged level of causation. Interface Focus, 2(1), 55–64. Omnès, R. (1994). The interpretation of quantum mechanics. Princeton University Press. Oppenheim, P., & Putnam, H. (1958). Unity of science as a working hypothesis. In H. Feigl, M. Scriven, & G. Maxwell (Eds.), Minnesota studies in the philosophy of science (Vol. 2, pp. 3–36). University of Minnesota Press. Papineau, D. (2008). Must a physicalist be a microphysicalist? Oxford University Press. Philbin, T. G. (2010). Canonical quantization of macroscopic electromagnetism. New Journal of Physics, 12(12), 123008. Rose, D., & Schaffer, J. (2017). Folk mereology is teleological. Noûs, 51(2), 238–270. Ruetsche, L. (2002). Interpreting quantum field theory*. Philosophy of Science, 69(2), 348–378. Ruetsche, L. (2003). A matter of degree: Putting unitary Inequivalence to work. Philosophy of Science, 70(5), 1329–1342. Ruetsche, L. (2006). Johnny’s so long at the Ferromagnet. Philosophy of Science, 73(5), 473–486. Ruetsche, L. (2011). Interpreting quantum theories. Oxford University Press. Ruetsche, L. (2013). Unitary equivalence and physical equivalence. In R. Batterman (Ed.), The Oxford handbook of philosophy of physics (pp. 489–521). Oxford University Press. Schaffer, J. (2008). Alfred Hüttemann what’s wrong with microphysicalism? Review. British Journal for the Philosophy of Science, 59(2), 253–257. Schaffer, J. (2010). Monism: The priority of the whole. Philosophical Review, 199(1), 31–76. Schlosshauer, M. (2005). Decoherence, the measurement problem, and interpretations of quantum mechanics. Reviews of Modern Physics, 76(4), 1267–1305. Schwabl, F. (2006). Statistical mechanics (2nd ed.). Springer. Simpson, W. M. R. (2014). Ontological aspects of the Casimir effect. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 48(1), 84–88. Simpson, W. M. R. (2015). Surprises in theoretical Casimir physics: Quantum forces in inhomogeneous media. Springer. Simpson, W. M. R. (2021). From quantum physics to classical metaphysics. In W. M. R. Simpson, R. C. Koons, & J. Orr (Eds.), Neo-Aristotelian metaphysics and the theology of nature. Routledge.
50
W. M. R. Simpson and S. A. R. Horsley
Simpson, W. M. R., & Leonhardt, U. (2015). Forces of the quantum vacuum: An introduction to Casimir physics. World Scientific. Steward, H. (2012). A metaphysics for freedom. Oxford University Press. Wallace, D. (2011). Taking particle physics seriously: A critique of the algebraic approach to quantum field theory. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 42(2), 116–125. Weinstein, S. (2011). Electromagnetism and time-asymmetry. Modern Physics Letters A, 26(11), 815–818. Wen, X.-G., & Witten, E. (1985). Electric and magnetic charges in superstring models. Nuclear Physics B, 261, 651–677. Xiao, Y.-F., Özdemir, Ş. K., Gaddam, V., Dong, C.-H., Imoto, N., & Yang, L. (2008). Quantum nondemolition measurement of photon number via optical Kerr effect in an ultra-high-Q microtoroid cavity. Optics Express, 16(26), 21462–21475. Zurek, W. H. (1982). Environment-induced superselection rules. Physical Review D, 26(8), 1862–1880. Zurek, W. H. (2003). Decoherence, einselection, and the quantum origins of the classical. Reviews of Modern Physics, 75(3), 715–775. Zurek, W. H. (2009). Quantum Darwinism. Nature Physics, 5(3), 181–188.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Chapter 3
Dispositional Essentialism in the Eternalist Block Andrea Roselli
Abstract The connection between the metaphysics of time and the metaphysics of powers is a relatively new debate in the philosophical literature. Friebe (Journal for Theory, History and Foundations of Science, 33, 77–89, 2017, Backmann (Inquiry, 62, 979–1007, 2018), Donati (No time for powers, Phd Dissertation, University of Nottingham, 2018) have argued that dispositional essentialism may encounter some problems when combined with (in their words) ‘static’ views of time, such as Eternalism. I believe this is a challenge that it is important to address. I will first briefly present (the standard version of) the four main metaphysics of time; I will then present and discuss the main objections moved to the combination of powers ontologies and the metaphysics of time; finally, I will argue that these objections fail, and that the alleged incompatibility results merely from a misconception of the staticity of Eternalism, on the one hand, and of the productiveness of powers on the other; in particular, I will show how the ‘incompatibility argument’ is either false or trivial. Keywords Powers ontology · Eternalism · Metaphysics of time · Dispositional essentialism
I am grateful for the feedback received on earlier versions of this paper presented (online) at the 2021 Conference of the Society for the Metaphysics of Science The research leading to this publication was supported by the Leverhulme Trust as part of the project Part-whole relations within the fundamental potentialities in nature (RPG-2018-079). A. Roselli (*) Department of Philosophy, Durham University, Oxford, UK © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_3
51
52
A. Roselli
Recent years have seen the rise of power realism among metaphysicians. Ontic1 dispositional properties are now commonplace features in many prominent ontologies proposed by philosophers in the most diverse fields – from the philosophy of mind to Quantum Mechanics. With an increased interest in powers, the characterisation of their identities and their application in a number of philosophical debates have become increasingly complex. In parallel with the debate about the exact identity or essence2 of powers, there is also a lively debate concerning what powers ontologies can accomplish and what they are compatible with: if powers can be shown to be incompatible with some fundamental metaphysical frameworks most philosophers deal with, their appeal would be significantly reduced. In this paper, I focus on a relatively recent attack at powers ontologies along these lines: the claim that powers ontologies are incompatible with certain metaphysics of time. Powers ontologies do not look immediately at odds with any of the main metaphysics of time (I focus here on Eternalism, Presentism and the Growing Block view). However, some philosophers recently put forward what I call an ‘incompatibility argument’: they intend to show that powers ontologies must presuppose an existentially dynamical view of temporal existence (i.e. the growing block view or presentism), and are therefore incompatible with Eternalism – which is however very popular among philosophers nowadays. In this paper, I argue that the argument fails (it is either trivial or false), and that the alleged incompatibility results merely from a misconception of the staticity of Eternalism, on the one hand, and of the productiveness of powers, on the other. In the second section of the paper, I will present the three main metaphysics of time, highlighting some interesting features that are particularly relevant for the present debate; in the third section, I will analyse the arguments put forward for the incompatibility of powers ontologies and certain metaphysics of time, and I will outline what I call the incompatibility argument; in the fourth section, I will present my counterarguments and show that the incompatibility argument is either trivial or false.
In a ‘predicatory’ sense, whatever we say in our natural language about an object defines a property: an object has the property of being such that ‘x’. However, we might have good reasons to think that such a predicative property is not necessarily a full-fledged component of our ontology; powers are ‘ontic dispositional properties’ in the sense that are ontological components of the world (see e.g. Bird, 2016 for the difference between predicatory and ontic use of the term property). 2 In order for a dispositional property to be a power, its dispositionality needs to be a non-reducible, ineliminable part of what that property is. A nice and direct way to understand what powers are, then, would be to say that powers are properties which are essentially dispositional – as in e.g. Shoemaker (1969), Ellis (2001), Heil (2003), Bird (2007), Martin (2007); this might however be controversial, as some power ontologists, such as Molnar (2003), Mumford (2004), prefer to avoid the reference to ‘essences’, and simply claim that the dispositionality of powers is what fixes their identity, instead of representing their essence. 1
3 Dispositional Essentialism in the Eternalist Block
3.1
53
Temporal Dynamism in the Metaphysics of Time
The three main metaphysics of time3 are (1) Presentism, (2) the Growing Block view, and (3) Eternalism; these are the most discussed alternative metaphysical conceptions of the large-scale composition of the universe from a temporal point of view; in other words, they are different answers to the question “what does exist from a temporal point of view?” (1) Presentists hold4 that only what exists ‘now’ is real: past events were real, future events will be real; therefore, they hold that there is an ontological difference between the present and the two other temporal forms: there is a privileged global present that separates ‘the past’ and ‘the future’. In this sense, there is a fundamental difference between space and time:5 while everyone typically concede that other spatial locations exist and are ‘equally real’, Presentists deny that temporal instances other than the present exist, and maintain that their view accounts better for many of our fundamental intuitions and, as Wuthrich (2012: 442) puts it, “capture salient features of our experience of temporality”, such as the openness of the future, the passage or ‘flow’ of time, and the fact that events slip away from us. The main problem for Presentists comes from what seems to be a fundamental piece of knowledge of our best physics – the relativity of simultaneity; how can there be a global present if every observer has his or her reference frame? Relativity implies that there are no ‘distinguished’ observers, i.e., that none can claim to have a privileged temporal view of the universe. In particular, no such observer can claim to be ‘at rest’ while the others are moving; they are all simply in relative motion. The problem with presentism, then, is that this metaphysics of time seems to be bound to the concept of a privileged ‘absolute’ present, to be distinguished from all the other ‘relativistic’ presents, and this is simply not possible according to our best physics. Despite the problems surrounding a global plane of simultaneity, Presentists believe there is an intuitive pull which is too strong to be ignored: the present just feels special; there must be something temporally different between ‘me now’ and ‘me ten years ago’ – and this difference, Presentists claim, can be understood only if we admit that there is something objectively different (i.e. not relative to a reference frame) between the two.
3 See for example Callender 2002, Maudlin 2007, Dainton 2010, Markosian 2014 for an overview of the different positions in the debate. Presentism and Eternalism are, by far, the most popular metaphysics of time; the Growing Block view (Ellis, 2006, 2014) and the Moving Spotlight view (Skow, 2012, 2015), however, are quite discussed in recent literature. There are also other possibilities, such as the Burning Fuse model (Norton, 2015) or various combinations of the main four metaphysics, but it is not relevant for the present purposes. I will consider here the three most popular views: Presentism, Eternalism, and the Growing Block Theory (GBT). 4 There are of course many different ways to be a Presentist – and the same goes with the other metaphysics of time; what I am here presenting is a simplified, standard introduction to these positions to help a potential reader who is not acquainted with this temporal debate. 5 One of the interesting consequences of Presentism, which I will not however discuss here, is that it is difficult to see space and time as different aspects of the same entity ‘spacetime’.
54
A. Roselli
(2) Growing Block theorists think however that this description fails to recognize a fundamental asymmetry between the past and the future: we know how the World War II ended, but we don’t know if and when there will be a World War III; we can’t change the past, while there are many open alternatives in the future. Therefore, they think that a ‘Growing Block Theory’ (GBT) illustrates better this situation: the past and the present are real and fixed, while the future is open; the total sum of what is real is gradually increasing, by a process of moment-by-moment ‘absolute becoming’. On this view the present is the most recent addition to reality, and the interface, so to speak, between being and non-being. C.D. Broad, arguably the theoretical father of this approach, wrote (1923: 68–69) that “a change from future to present [. . .] is a matter of the event becoming or coming into existence”, and “when an event becomes, it comes into existence; and it was not anything at all until it had become [. . .]. There is no such thing as ceasing to exist; what has become exists henceforth for ever”.6 The future, then, simply does not exist, while the past exists and remains fixed: the present is a sort of hyperplane that separates what is from what is not: an ‘edge of Being’. Such a view shares with Presentism the difficulties concerning the global and absolute ‘now’, but has some on its own. For example: in describing the nature of this “genuine becoming”, it seems that there are two different kinds of time:7 time as location in the four-dimensional spacetime manifold that is the present and the past, and objective time which tells you where the border is. (3) Eternalists, finally, maintain that the stance that there is a ‘global present’ cannot be maintained after the Special Theory of Relativity. The fact that it is impossible to identify an absolute now, identical for all the possible reference frames in the universe, entails – they claim – a Block Universe: observers have their own nows, their pasts and their futures, but there isn’t any general, ontological and absolute distinction between the three temporal forms. The simultaneity of – or the temporal interval between – two points in spacetime is not defined until a coordinate system is arbitrarily chosen: as Savitt (2017: 3) puts it, “we have no way to distinguish the present from amongst the multitude of presents”, since there is a potential infinity of simultaneity planes passing through any spacetime point. As far as their ontological status is concerned, then, Eternalists think that all events are equally real, regardless of the time at which they occur; there is no unique privileged present time, and there is no such thing as a moving present.8
6
Broad’s Growing Block has received more and more attention starting from Tooley (1997) and many physicists are fascinated by the challenge of merging contemporary physics with a Growing Block account of the world, most noticeably George Ellis (see e.g. Ellis, 2006 and 2014). 7 See e.g. Braddon-Mitchell (2004) for a discussion of this topic. 8 I am here simplifying: Craig Callender, for example, consider himself an Eternalist and yet argues in favour of a stronger notion of present (see e.g. Callender, 2002, 2008). Another interesting way to present the debate would be to talk about tense theories and tenseless theories; however, for our present purposes, this would merely lead us astray, as I am already going to propose considering the (very similar) opposition between dynamic and static views of time.
3 Dispositional Essentialism in the Eternalist Block
55
The opposition between the first two stances and the last one could also be described as the opposition between ‘dynamists’ and ‘eternalists’, opposition which is based on different conceptions of the passage of time; as Dainton (2010: 12) puts it, “the dispute between dynamists and eternalists is over the nature of time, how time is in and of itself”. In other words, is time a real feature of our physical reality? Is it just a useful way of expressing ourselves? Typically, philosophers who defend an objective, advancing present which is global and absolute, believe that the flow of time is real and defines the difference between what is present and what is not. If Eternalists deny that there is ‘a’ present (and, typically, also that there is a real flow of time), how can they explain the passage of time we all seem to experience? In this sense, those who deny that the passage of time is a real feature of the physical world are considered defenders of a ‘static’ view of time. Here’s how Barry Dainton (2010: 7) describes the opposition between dynamic and static views: There are those who believe that temporal passage is a real phenomenon, an objective feature of reality that is independent of the perspective that conscious beings such as ourselves have on things. Anyone who believes this subscribes to a dynamic view of time. [. . .] In the opposing camp are the many philosophers who reject the claim that temporal passage is a real and objective feature of the world.
Dynamists, in other words, believe that the Eternalist description of the world is ‘too static’ to be true. A very common and influential argument in favour of the objective present and objective flow of time rests on our own experience of time: it seems to us as if time flows, the argument runs, and surely the most reasonable explanation of this is that there is some genuine motion of time which we experience, or in which we partake. But how would things seem if time did not objectively flow? George Ellis (2006) strongly believes that the Eternalist Block is unrealistic because “the irreversible flow of time is one of the dominant features of biology, as well as of the physics of complex interactions” (Ellis 2006: 1798); or, as Abner Shimony (1993: 284) puts it, “something fleeting does indeed traverse the world line, but that something is not subjective; it is the transient now, which as a matter of objective fact is momentarily present and thereafter is past”.9 The Eternalists’ usual description of the passage of time, that is, seems too ‘atemporal’ to describe dynamical phenomena. However, events in the Block are always at temporal locations; events in series occur at different temporal locations: for this reason, Huw Price (1996) claims that “things would seem this way, even if we ourselves were elements of a block universe” (Price 1996: 15). In the Special Theory of Relativity, the histories of material objects, always moving with speeds less than that of light, are represented in Minkowski diagrams by time-like world lines. As Dieks (2006) suggested, we can think of becoming in Minkowski spacetime as proper time along a time-like world line: the present for a point particle on a time-like world line would coincide with the particle,
9
Shimony, 1993: 284.
56
A. Roselli
and a succession of presents would be the successive occurrence of events along that world line. The metaphor of a Block Universe could be responsible for the (unfair) accusation of having a static nature, in opposition to the ‘moving present’ view of time. This is misleading, as it suggests that there is a timeframe in which the fourdimensional block stays the same: there isn’t. Simply, as Price (1996: 13) puts it, “defenders of the block universe deny that there is an objective present, and usually also deny that there is any objective flow of time”. In a sense, then, Presentists and Growing Block theorists believe in a different sort of Minkowski diagram: an animated one, in which what is moving is not the light cone of the observer, but the global and observer-independent transient now advancing. However, as already David Park (1972) claimed, it is arguable that there is no benefit in adding the animation: “the animated diagram may be more intuitive [. . .] than the atemporal one, but it contains no more specific, verifiable information. All of the science of dynamics, that is, all we know about how complex systems (including ourselves) behave and interact, is already represented on the atemporal Minkowski diagram” (Park, 1972: 115). The non-animated Minkowski diagram may be ‘static’, but the static diagram represents the evolution of systems along their world lines. The diagram, then, need not itself be animated to represent dynamical phenomena. This means that for dynamism to be intelligible we don’t need a real flow in physical reality over and above the variation in time, accommodated by the Block universe. As static as it may seem, this picture represents the evolution of systems along their world lines. This is a crucial point for the arguments we will consider in this paper: what are dynamic events like in a Block Universe? I believe there are very good reasons to believe that a Block Universe could account for them; in order to give you an intuitive pull towards this direction, consider this thought experiment proposed by James Hartle (2005): the American physicist conceptualised a simplified model of our perception and understanding of notions such as past, present and future, describing the processing of external informations by a robot – an IGUS, Information Gathering and Utilizing System –, complex enough to have (something that we can call) a sufficient distinction between past, present and future. Every time interval T*, the robot captures an image of the external environment, and stores it in its (say) four registers (R1, R2, R3 and R4). The new image is stored in register R1 (the robot’s ‘present’); after T*, a new image is stored in R1, and the old image shifts in R2; R2, R3 and R4 constitute the robot’s memory of the past. After four T*, all the registers are full. The new image is stored in R1, the other images shift of one position, and the image that was in R4 is erased (forgotten). The robot uses the images to build a simplified scheme of the external environment, to make predictions, take decisions, etc. The robot may then be said to predict the future, experience the present (in R1), remember the past (in R2, R3 and R4). There is a present at each instant along the robot’s world line, consisting of its most recently acquired data about the external world. The flow, or passage, of time is then nothing over and above the motion of information into the registers.
3 Dispositional Essentialism in the Eternalist Block
57
Fig. 3.1 The motion of information into the registers
It is easy to describe in four-dimensional terms10 this simplified model, without betraying any aspects of its ‘subjective experience’. As Hartle (2005, p. 103) puts it, “the center of mass of a localized IGUS [the robot] describes a timelike world line in spacetime. At each point along the world line, any tangent to it lies inside the light cone so that the IGUS is moving at less than the speed of light in any inertial frame”. Consider Fig. 3.1 (Hartle, 2005, p. 104): The world lines of the robot and of an object that appears in its stored images are represented. The contents of the four registers change after T* as described above. The most recently acquired image defines the (four-dimensional description of the) present. The object changes its shape (c, d, e, f, . . .) every T^ 6¼ T*; the dotted lines indicate the light rays conveying the image of the object. What is crucial, then, is that “the present is not one instant along the robot’s world line, much less a spacelike surface in spacetime. Rather, there is a ‘now’ for each instant along the robot’s world line extending over proper time T*” (Hartle, 2005, p. 104). The contents of R2, R3 and R4 constitute a four-dimensional representation of the past, while the contents of R1 (the boxes in bold) give a four-dimensional description of the robot’s present consistent with special relativity. Thus, Hartle conclusion is that “there is no conflict between the four-dimensional reality of physics and the subjective past, present, and future of an IGUS. Indeed [. . .] the subjective past, present, and future are fourdimensional notions. They are not properties of spacetime but of the history of a particular IGUS” (Hartle, 2005, p. 104). This is in line with our best physics: there is not an objective, global spacelike surface that stretches over the whole universe; this is well represented by the robot’s ‘present moment’, which is not a spacelike surface, but rather refers to the most recently acquired data; also, this notion is not restricted to one point of its world line, but it is valid for every point. There is no reference whatsoever to a global surface that cuts the spacetime in future and past; on the contrary, the reference of the ‘present moment’ is directed solely to the last data acquired. “For simplicity we consider the flat spacetime of special relativity (Minkowski space). But with little change it could be a curved spacetime of general relativity”. Hartle (2005), p. 103.
10
58
A. Roselli
What I find interesting about this simplified model is that we humans are also actually continuously11 scanning the environment and storing information. Even if this picture is described as somehow ‘static’ by Dynamists, it represents dynamically the evolution of systems along their world lines. The picture, in other words, does not need to be dynamical to represent dynamical phenomena: for dynamism to be intelligible we don’t need a real flow in physical reality over and above variation in time (which is what the Block universe predicts). As static as it may seem, this picture represents the dynamic evolution of systems along their world lines. An interesting corollary to this, which I will here mention only in passing, is that this picture seems also able to account for our ‘experience of something passing’: even if what we experience remains apparently identical (say: we stare at the same painting for ten minutes), the data we collect are always numerically distinct; they are another piece of information (we are well able to tell the difference between the experience “looking at the painting for ten seconds” and the experience “looking at the painting for ten minutes”). Hartle’s IGUS sees no difference in a series of images (R1, R2, R3. . .) in which we experience the same thing or a series in which we experience different things; the relevant point is that it acquired different images, then R1 is ‘the present’ and R2, R3, etc. are the past.12 Crucially, then, what is dynamical about such experiences does not seem to be necessarily a feeling of a passage of something external (an objective flow of time), but the awareness of the shifting of the experiences contained in R1 towards R2, and then R3, etc.13 In the next section, I present the incompatibility argument against the combination of powers metaphysics with ‘static’ views of time. What has been said so far will serve not only as an introduction, but first and foremost as a conceptual background for the counterargument presented in the last section, where I will argue that the so-called ‘static’ views of time are not, in fact, too static for powers metaphysics.
11
I am aware of the existence of different models to explain time estimation and duration perception (the most famous are probably the “Internal clock model” and Ian Phillips’ “Mental activity model”); but this querelle is not relevant here, since everything I need to claim is that – whichever model we choose to represent our internal temporal processing – more images, even if visually identical, are at least numerically distinct; that our brain is able to distinguish between one picture and the successive acquisition of ten identical pictures. 12 The situation is complicated a little by the fact that there are different models to account for our temporal phenomenology (the three main models of our temporal understanding are the Extensionalist, the Retentionalist and the Cinematist); the accumulation of experiences, however, is something that these models have in common. Whichever model we choose to describe our temporal phenomenology, our ‘dynamical feeling’ in front of the same painting is described by the continuous acquisition of data. 13 Dynamists might well argue that this awareness is impossible without the feeling of an objective flow of time, or that the nature of this awareness is unclear, or similar arguments; but I find this line of thought quite promising for Eternalists that want to explain our subjective temporal experience. As the present purposes have more to do with metaphysics than phenomenology, however, this debate is here not crucial.
3 Dispositional Essentialism in the Eternalist Block
3.2
59
The Incompatibility Argument
A question is now emerging in the philosophical debate: is the Eternalist Block too ‘static’ for the powers view? While there are philosophers who have explicitly supported a combination of Eternalism and powers ontologies,14 others have more recently claimed that powers may not be compatible with Eternalism. For instance, Cord Friebe (2017) holds that “humeanism requires eternalism and Power metaphysics must presuppose an existentially dynamical view of temporal existence” (2017: 77); Donatella Donati maintains that “powers theory and standard Eternalism are not compatible” (2018: 1); Marius Backmann claims that “the inherently dynamic and productive powers ontology is incompatible with static eternalism” (2018: 981). In their view, the problem arises from the fact that dispositional essentialism15 requires a strong kind of productive dynamism, which would only be possible in a ‘dynamical’ metaphysics of time, while Eternalism would be too ‘static’: The inherently dynamic and productive powers ontology is incompatible with static eternalism, which does not allow for productivity or activity, since there is no progression of an objective present and all past, present and future facts are equally real on such a view. (Backmann, 2018: 981) According to dynamical views of time, it makes an essential difference if something is (formerly) presently located at p or not. Likewise, according to the dispositionalists, it makes an essential, ontological difference if something has been produced at p or not. The connection is not casual: for, the Power metaphysicians must argue that “to be present” has productive sense. Thus, only in a dynamical world the dispositions can exercise their productive power. (Friebe, 2017: 88)
What they claim, then, is that there are metaphysics of time which are more dynamic than others – the Presentist and the Growing Block metaphysics of time envisage an objective present which advances, and new things, facts and events continuously coming into being. On the contrary, Eternalism does not envisage a privileged present advancing, and therefore there aren’t things or facts or events objectively coming into being. The first two alternatives would then be more dynamic in the sense that the present is the edge of being, and different presents come successively into existence with the passing of time. What these philosophers claim, then, is that only this (strongly, genuinely) dynamic metaphysics of time can accommodate the notion of activity that powers ontologies are supposed to capture. But what, exactly, is this dispositional notion of activity/productivity that cannot be accommodated in the (so-called) ‘static’ metaphysics of time? Backmann and Friebe hold that there is here an ontologically substantial notion of production in (some 14
See e.g. Bird (2007), Esfeld (2008). There are different possible terminologies in the literature for metaphysics which contain irreducible dispositional properties, each of which has different implications (some philosophers, for example, want to avoid the reference to essences, etc.). For reasons of simplicity, I will here use “dispositional essentialism” as a synonym of “powers metaphysic”.
15
60
A. Roselli
relevant versions of) dispositional essentialism: for an event x, to be located at a certain instant in time means to have been produced at that instant. It has not existed before, but it ‘came into being’ when another object exercised its powers. The problem then, as they see it, is that endorsing this ‘non-passivist’ reading, powers are non-metaphorically active properties with a dispositional essence, and they cannot be fitted into the Eternalists Block Universe where events are timelessly given, merely ordered in an earlier-later relation: they do not ‘come into existence’ as an objective present advances. Genuine dispositional properties, they hold, must ‘produce’ their manifestations, in the sense that manifestations depend existentially on the dispositions; Eternalism, then, does not fit with the requirement that manifestations are ‘products’ resulting from the exercise of dispositions, as e.g. Friebe (2017: 86–87) puts it here: Dispositions would be powerless if they existed merely perspectively at t; for, then, their manifestations would likewise exist merely perspectively at t’, and, hence, both the dispositions and their manifestations would be ‘given’ anyway. [. . .] Thus, according to Power metaphysics, something substantial has in fact to be added to ‘being located at p’: entity e has been produced at p.
At this point, it is legitimate to ask ourselves: is this strong notion of productiveness (manifestations coming into being) a feature that every powers ontology has, or should have? Either they are claiming A) that every powers ontology, in order to be a powers ontology, has to have such a strong notion of productiveness as ‘making something coming into being’; or B) they are claiming that some powers ontologies have such a notion. The first claim is too strong: it is simply not true that every powers ontology, in order to be a powers ontology, has to have such a strong notion of productiveness (I have already mentioned Bird (2007) or Esfeld (2008) as examples of a possible combination of eternalism and dispositional essentialism, and counterexamples to A). However, and crucially, in endorsing B) they add that powers ontologies which can be accommodated with Eternalism do not take seriously the ‘productive sense’ of powers: For most of the powers metaphysicians though, Bird is no true believer. The crucial split between Bird’s view and that of the neo-Aristotelian powers metaphysicians seems to be the latter’s notion of activity. The most common criticism from proponents of the powers view seems to be that Bird’s view is still reductive. (Backmann 2018: 986) There is the widespread intuition that Power metaphysics is compatible with the block universe view (see, e.g., Esfeld, 2008, chap. 5). The idea firstly seems to be that the crucial difference between categoricalism and dispositionalism is that the latter recognizes metaphysically necessary connections among distinct physical tokens, whereas the former denies them. These modal relations allegedly can hold between (or, even require) ‘already’ existing relata. So understood, Power metaphysics lacks productive sense and is, therefore, compatible with eternalism. But de-re modal relations of necessity or tendency are underestimated if one denies their productive sense. If a necessitates (or, is a tendency for) b, b cannot exist ‘anyway’ [. . .]. Rather, b only is located at its time t when necessitated or brought about. (Friebe, 2017: 87)
3 Dispositional Essentialism in the Eternalist Block
61
The idea driving such criticisms is that dispositional properties ‘produce’ their manifestations, in the sense that manifestations depend existentially on the dispositions. The ‘ontologically substantial’ notion of production that Backmann, Friebe and Donati have in mind takes after some already known ideas sketched in the powers literature by, among others, Groff (2013), Mumford (2009), and Ellis (2002): a power produces its manifestation in the robust sense that that manifestation is coming into being, it did not ‘already’ exist at a later spacetime point in the quadridimensional Block Universe. Otherwise, as Friebe (2017: 87) observes, dispositions would be powerless, since both the dispositions and their manifestations would be ‘given’ anyway. Friebe’s argument, then, is that dispositional essentialism – at least according to this ‘non-passivist’ reading – requires that events are existentially dependent on one another, and not merely temporally ordered. Powers, then, must bring new physical entities into existence, and this is at odds, he claims, with Eternalism, wherein a physical entity may only be temporally located later than another entity. Backmann maintains that in a non-passivist powers ontology “the fact did not exist before it was produced. This is not consistent with Eternalism, where all the facts exist equally, regardless of their temporal status, [. . .] and dynamism and activity can only be reduced to sequences of facts which all exist equally” (2017: 991–992). Their arguments stem from the idea that genuine power ontologies should envisage a strong notion of production – which presupposes in turn a dynamic (as opposed to static) metaphysic of time, as in the latter “future facts already timelessly exist” (Backmann, 2018: 986); even Friebe (2017: 87) maintains that “a power produces its manifestation in the robust sense that that manifestation is coming into being, it did not ‘already’ exist at a later spacetime point in the quadridimensional Block Universe”. Dispositional essentialism – according to this ‘non-passivist’ reading – requires that events are existentially dependent on one another, and not merely temporally ordered. Powers, then, must bring new physical entities into existence, and this is at odds with Eternalism, wherein a physical entity may only be temporally located later than another entity. The authors I mentioned hold that a ‘genuine’ dispositionalism needs an ontologically substantial notion of production: powers need to be non-metaphorically active properties with a dispositional essence. Eternalism would not fit with the requirement that manifestations are ‘products’ resulting from the exercise of dispositions, as this metaphysic of time would not allow powers to really bring about manifestations – powers are not really powerful if manifestations are not brought into existence. Let’s analytically sketch this Incompatibility Argument: 1. There is an ‘anti-reductivist’ powers metaphysic (for ‘true believers’!) which is temporally connoted: powers – and the activity they bestow upon their bearers – are ‘irreducibly dynamic’ from a temporal point of view. 2. This is a full-fledged powers metaphysics, in the sense that the productiveness of powers is taken seriously, and it is incompatible with Eternalism.16 16 Backmann (2018) mentions Groff (2013), Mumford (2009), and Ellis (2002) as paradigmatic cases of full-fledged powers metaphysics.
62
A. Roselli
3. In a full-fledged powers metaphysics, there is at least a feature which is incompatible with ‘static’ views of time.
3.3
The Argument Is Either Trivial or False
I take seriously this challenge, not only because I think Eternalism is a promising metaphysic of time, but also because it would be important for powers ontologies to be open to different metaphysical possibilities – as temporally liberal as possible, so to speak. Is it true that a full-fledged powers ontology is not compatible with a ‘static’ view of time? In what follows, I will try to answer this question. My claim is that the incompatibility argument is either trivial or false: • The argument is trivial if what it actually boils down to is the claim that a combination of a powers ontology and a presentist (or GBT) notion is not compatible with Eternalism. In this case, what the argument would say is merely that an anti-Eternalist theory is not compatible with Eternalism, but this is trivial. • The argument is false if the advocated non-passivist reading of powers is not connoted in presentist (or GBT) terms, but it is only ‘incompatible with eternalism’. It is possible for powers ontologies to ‘take productiveness seriously’ and be still compatible with Eternalism. The crucial point here is that it is one thing to claim that there are powers ontologies that focus on the role of productivity and combine this with a certain temporal view – this is perfectly legitimate, and I have no problem with that (it is arguably what Groff (2013), Mumford (2009), and Ellis (2002) do in their theories); but a totally different thing to claim that these powers ontologies can add something to the dispositionalist frame because of a different temporal view of the world. In other words, I believe a philosopher interested in powers can combine her theory with whatever metaphysic of time she prefers: she is wrong only if she claims that only a certain combination (in this case: powers ontology + Presentism or GBT) can take the productiveness of powers seriously. Consider this passage by Neil Williams (2019: 66), who goes so far as to say that the productiveness of powers is what distinguishes these properties from categorical properties:17
17 Causally inert properties, known in literature as ‘categorical’, are typically understood as geometrical/structural properties like shape and size (as well as any irreducible qualitative properties). Neo-Humeans take all properties to be causally inert categorical properties, lacking modal or causal characteristics; therefore, whatever causal or modal fact obtains in the world, it does so because it is determined by the limited base of categorical properties. An ontology in which all fundamental properties are categorical is known in literature as categorical monism; categorical monism is a key aspect of the neo-humean ontology, which eschews any fundamental property whose essence is at all powerful.
3 Dispositional Essentialism in the Eternalist Block
63
A way that powers are distinguished from categorical properties concerns their role in bringing about effects. In and of themselves, categorical properties are incapable of producing effects. Their essences, are the thin features of identity and difference. They simply do not have in them—on their own—the ability to bring about states in the world. They can only do so with the help of the laws. Powers, on the other hand, are effect-producers by nature. Nothing needs to be added for powers to produce some effect; their causal readiness and ability is part of what makes them what they are.
Passivism is typically associated with categorical properties: add some governing laws of nature, the story usually goes, and categorical properties can work with those laws to generate effects; take away the laws, and you have nothing causal; categorical properties have their causal profiles contingently; in and of themselves, categorical properties have no causal profile: any causal profile they acquire comes from outside them, via the addition of laws of nature. Powers, on the other hand, are those properties that bind most tightly to their causal profiles. So tightly, in fact, that their causal profiles determine their essences. This is to say that the ability to bring about change in the world, their productivity, is what defines powers, what distinguishes them from categorical properties. The point we are interested in is the following then: what does it mean to produce, to bring about manifestations, in an Eternalist metaphysic? If every event is ‘already there’ in the Block, it doesn’t seem it needs to be brought about by powers; nothing comes into being, and there is therefore no real production; this seems to be a bad situation for properties which are sometimes even defined in terms of their ability to bring about manifestations. However, as I have shown in the second paragraph, there is a perfectly reasonable way to talk about the past, present and future in the Block, as well as a way to reconstruct diachronic causal relations, where a cause brings about an effect: if we choose a reference frame and follow the time line of a particular set of events, ‘coming into being’ has a definite meaning even in an Eternalist metaphysic – simply, it is not happening at the same moment for every possible observer. But this holds in general, and for a sensible reason: according to relativistic physics, there is not a right way to answer the question ‘what time is it in the universe’ without specifying a position and a reference frame. The ‘coming into being’ of a manifestation is not ‘objective’ merely in the (very weak!) sense that it does not happen at the same time for every possible observer in the universe: there are more-than-one possible hyperplanes cutting the four-dimensional Block and defining what is future, present and past. However, and this is crucial, the fact that the manifestation happens after the power bringing it about is absolutely objective. According to Special Relativity, there is no possibility of disagreeing on temporal relations about events at time-like distance (where one event is in the past or future light-cone of another): in other words, the relativity of simultaneity disappears if we restrict our consideration to local regions. Let me stress this concept: relativistic physics tells us that for every possible observer, even without specifying a reference frame, there is an objective temporal order for objects at time-like distance (if A is in the causal past of B): therefore, we absolutely cannot make sense of the idea that, when a power is going to bring about a manifestation, that future manifestation is ‘already’ there.
64
A. Roselli
Since in relativistic physics everything travels slower than light,18 every manifestation is objectively in the future of a power, brought about by it according to every possible observer. Moreover, even if causation travelled faster than light, for every observer that chooses a particular reference frame (a global foliation) it is possible to distinguish every event as being objectively past, present or future. As Callender (2002), Dorato (2006, 2012), Meyer (2005), and Savitt (2006) among others suggested, the debate between a ‘static’ and a ‘dynamic’ metaphysic of time is less meaningful than it seems at first: as Dorato puts it, “what, exactly, is being denied by the presentist’s implication that the future ‘is not real’ or simply ‘does not exist’ above and beyond than the platitude that it does not exist now?” (2006: 561). Eternalists and Presentists have different temporal perspectives in mind: when the Eternalist claims that future events exist, she doesn’t mean that they exist simultaneously with her present and past (relatively to her now) objects; rather, that they will exist relatively to the point in which she is located. It is all relative to the perspective we choose to ‘temporally look’ at the events: if we take a tensed (internal) temporal perspective, and refer to reality from our particular point of view in time, the present really has a special status; if, on the other hand, we take the tenseless (external) temporal perspective – God’s eye view –, we can imagine that all events are four-dimensionally spread out in front of us, and that we look at reality from ‘nowhen’. In the latter case, the whole quadridimensional universe is there: every event in the entire history of the universe is in front of us. This perspective might seem odd, but it is useful as our best physics tells us there is not a privileged now: it would be unjustified to privilege one particular coordinate system F and call its simultaneity plane ‘reality’. This however doesn’t mean that once you are in a particular coordinate system F everything is simultaneously real relative to F. When a particular inertial frame is held fix, even the Eternalist distinguishes between events that, relative to that reference frame, objectively already occurred (‘came into being’) from those that will occur. When a philosopher adopts the external temporal perspective (God’s eye view), she is imagining seeing all the events in the history of the universe spread out in front of her; in this case, it has absolutely no meaning asking if something is ‘already’ there: not only because literally everything is there, but first and foremost because there is no ‘already’ until we specify relatively to what (an object and its reference frame). When we do that, however, the Eternalist is well able to tell if something is ‘already’ there or not. But the disagreement, the Presentist might say, regards what one would see from this external perspective. If there were an objective present advancing and making things objectively coming into being in a robust sense, one might say, even from an external perspective what we would see are just present events (or present and past events, if we endorse GBT). But this sort of objection completely misses the kind of intellectual effort the Eternalist is asking to make here:
It is well-known the fact that in QM there seem to be ‘spooky actions at a distance’, as Einstein used to say. Entangled particles seem to be showing that Quantum information could travel faster than light; I take this into consideration in the next sentence.
18
3 Dispositional Essentialism in the Eternalist Block
65
the external temporal perspective (God’s eye view) is not merely a look on ‘what there is’, but instead on everything that happened in the whole history of the universe, past present and future: what there is, there was and there will be; in no sense is it implied here that these events are simultaneous. It is simply false that, for the Eternalist, “future facts in a sense already timelessly exist” (Backmann, 2018: 3): ‘already’ does not have a meaning in a perspective from ‘nowhen’. Saying that future facts timelessly exist is patently a conflation of the two different perspectives. It is wrong to think that all facts and events are ‘already’ in the quadridimensional block, which is then somehow static; what does ‘already’ mean, when we consider the whole quadridimensional universe? Either we consider the world from an internal point of view, and in this sense there are indeed past and future facts, or we consider the world from an external point of view – the quadridimensional block in its entirety, as God would see it – and in this sense every event timelessly exists, but there is no ‘already’. For every event in the Block, there is a reason why that event is there: it has been caused by another event – if you are a powers metaphysician, a power; but even if you are not, there is a clear sense in which there is a causal connection in the Eternalist Block; even if you are a neo-Humean, there is a succession, one thing after another. If we do not conflate the internal and the external temporal perspective of the Block Universe, then, we can see that the Incompatibility Argument is false, because there is a clear sense in which powers in the Eternalist Block cause, or bring into existence, manifestations: when we follow the time line of the power and its reference frame we can define a present, a past, and a future: when the power is in the present, the manifestation is in the future; therefore, it is not in the present; therefore, it is not ‘already there’. It is the power that brings into existence/produces manifestations, that are there for a reason – that they have been produced by a power. The only way to say that this is not ‘productive’ enough is to claim that this productivity does not bring manifestations into being in the sense that there is not a privileged, advancing present that represents the edge of being. The argument would be that powers ontologies presuppose an advancing, objective present that makes things, facts and events, coming into being into an anti-relativistic sense – thing that presupposes a dynamic (as opposed to static) metaphysics of time. But then the arguments becomes trivial: if we claim that the distinctive feature of this ‘strong notion of production’ needs to be an objective present, an ‘edge of being’ which advances, this is necessarily only compatible with the so-called dynamic views (Presentism and GBT). In other words, if powerful production should be intended strictly as bringing events into being as an objective present advances, the argument becomes trivial: such a notion of productiveness would straightforwardly be a Presentist/GBT notion: it is no mystery it isn’t compatible with Eternalism. It seems this is what the proponents of the Incompatibility Argument have in mind; consider for example Backmann (2018: 992–993). That some event, fact or physical token is located at a certain instant in time for a powers ontologist means that it has been produced at that instant. It has not existed before and came into being when another object expressed its powers. [. . .] What dynamism could we
66
A. Roselli reasonably expect in a B-theoretic, static eternalism? Dynamism, if to be understood non-reductively, either involves production in the sense of bringing into existence or temporal progression: bringing a token into the objective present.
I take the argument expressed here to be that a combination of powers ontologies with a presentist/GBT notion is not compatible with Eternalism. But this is trivial. I have shown before how a strong notion of productiveness could be accommodated in the Block Universe; if one insists that the notion should be characterized in Presentist/GBT terms, it is no mystery that such a notion wouldn’t be compatible with Eternalism. In other words, the combination of a powers ontology with a specific temporal view does not add anything to the metaphysics of powers beyond the trivial fact that it is combined with this and not that temporal view. Productiveness can be strong, robust, genuine, both in an Eternalist frame and in a Presentist (or GBT) frame: what productiveness cannot do is be characterised in Presentist (or GBT) terms in an Eternalist frame, and vice-versa: but this is absolutely trivial. Therefore, either one argues that there cannot be a strong notion of productiveness in the Eternalist Block, since when powers are ready to cause manifestations, these manifestations are ‘already there’ – and this is false; or one argues that the notion of productiveness at play must be connected with the anti-relativistic idea of an objective present advancing, and this is not possible in the Eternalist Block – and this is trivial. Before concluding, I must mention the fact that the notion of production as ‘making things coming into being as an objective present advances’, far from being the only genuine possibility for true believers in powers, seems actually to be problematic on its own. Quite ironically, in fact, when we think about the combination of such a notion and powers ontologies – where there seems to be an intrinsic diachronic relation between a power and a successive manifestation – we discover that it is not easy to coherently make sense of it in any metaphysic of time. GBT shares with Eternalism the idea that the past is as real as the present. As Backmann (2018: 998) admits, it seems that in the Growing Block properties lose their productive power when becoming past, which contradicts the idea that they are essentially dispositional: A readiness to action, or a directedness towards a manifestation, does not require that the action will ever be performed. But in the past block, no action in the productive sense the powers metaphysicians invoke can ever be performed. So there cannot be a readiness for action in the past block. The unexercised powers at any past instant in the block cannot ever express themselves in such a productive manner. If there is no readiness for action in the past block, there cannot be any powers in the past block, given a productive, active reading of what a power is. [. . .] the past block has to be powerless for the reasons discussed above. So the powers metaphysicians have to explain to us which properties the objects in the past block bear”.
The problem is that there cannot be an exercise of a power in the past block. The past block is static, just as the Eternalist Block. If the exercise of a power is inherently productive, if it brings new facts into existence at the brink of the block, as Friebe (2017: 88) argues, then there cannot be an exercise of a power in a block of temporally ordered past facts. Therefore, this notion of production as
3 Dispositional Essentialism in the Eternalist Block
67
‘making things coming into being as an objective present advances’ is problematic even in the Growing Block, where powers would keep existing without being powerful any more (without making things coming objectively into being), which is arguably difficult to make sense of. Even more ironically, however, there are problems with Presentism as well. According to Presentism, there is nothing in the past or in the future, only the present and things in the present exist. However, manifestations in this case would not be manifestations of (substantial productive) powers, because these powers would simply not exist. How could the manifestation of a power be the manifestation of something that doesn’t exist? Recall that, in Friebe’s words, “according to Power metaphysics, something substantial has in fact to be added to ‘being located at p’: entity e has been produced at p” (Friebe, 2017: 86–87). If, however, only some things in the history of the universe are now, and only those that are objectively now are real, what does it mean for something real to have been produced? How can we add ‘something substantial’ to what exists now, if what exists now is all there is? There must be something in its causal past to produce it. But if past things literally do not exist, in this strong sense, there wouldn’t be anything productive, because everything would simply be – and not be produced. It would not be possible, in fact, to refer to past events as existing at time ‘t-1’ (in the past), since this would clearly be an Eternalist picture of the state of affairs.
References Backmann, M. (2018). No time for powers. Inquiry, 62(9–10), 979–1007. Bird, A. (2007). Nature’s metaphysics: Laws and Properties. Oxford University Press. Bird, A. (2016). Overpowering: How the powers ontology has overreached itself. Mind, 125(498), 341–383. Braddon-Mitchell, D. (2004). How do we know it is now now? Analysis, 64(283), 199–203. Broad, C. D. (1923). Scientific Thought. Kegan Paul. Callender, C. (2002). Time, reality and experience. Cambridge University Press. Callender, C. (2008). The common now. Philosophical Issues, 18, 339–361. Dainton, B. (2010). Time and space (2nd ed.). Acumen Publishing Limited. Dieks, D. (2006). The ontology of spacetime (pp. 111–127). Elsevier. Donati, D. (2018). No time for powers, Phd Dissertation, University of Nottingham. Dorato, M. (2006). The irrelevance of the presentism/Eternalism debate for the ontology of Minkowski Spacetime. In D. Dieks (Ed.), The ontology of spacetime (pp. 93–109). Elsevier. Dorato, M. (2012). Presentism/eternalism and endurantism/perdurantism: Why the unsubstantiality of the first debate implies that of the second. Philosophia Naturalis, 49(1), 25–41. Ellis, B. (2001). Scientific essentialism. Cambridge University Press. Ellis, B. (2002). The philosophy of nature. A guide to the new essentialism. Mc Gill-Queen’s University Press. Ellis, G. (2006). Physics in the real universe. General Relativity and Gravitation, 38, 1797–1824. Ellis, G. (2014). The evolving block universe and the meshing together of times. Annals of the New York Academy of Sciences, 1326, 26–41. Esfeld, M. (2008). Naturphilosophie als Metaphysik der Natur. M. Suhrkamp. Friebe, C. (2017). Metaphysics of laws and ontology of time, Theoria: An international. Journal for Theory, History and Foundations of Science, 33(1), 77–89.
68
A. Roselli
Groff, R. (2013). Whose powers? Which agency? In Powers and capacities in philosophy, edited by Ruth Groff and John Greco (pp. 207–227). Routledge. Hartle, J. (2005). The physics of now. American Journal of Physics, 73, 101–109. Heil, J. (2003). From an ontological point of view. Clarendon Press. Markosian, N. (2014). A defence of presentism. Oxford Studies in Metaphysics, 1(3), 47–82. Martin, C. (2007). The mind in nature. Oxford University Press. Maudlin, T. (2007). The metaphysics within physics. Oxford University Press. Meyer, U. (2005). The presentist’s dilemma. Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 122(3), 213–225. Molnar, G. (2003). Powers: A study in metaphysics. Oxford University Press. Mumford, S. (2004). Laws in nature. Routledge. Mumford, S. (2009). Powers and persistence. In L. Honnefelder, E. Runggaldier, & B. S. de Gruyter (Eds.), Unity and time in metaphysics (pp. 223–236). Norton, J. (2015). The burning fuse model of unbecoming in time. Studies in History and Philosophy of Modern Physics, 52(part A), 103–105. Park, D. (1972). The myth of the passage of time. In J. T. Fraser, F. Haber, & G. Muller (Eds.), The study of time (pp. 110–121). Springer-Verlag. Price, H. (1996). Time’s arrow and Archimedes’ point. Oxford University Press. Savitt, S. (2006). Presentism and eternalism in perspective. In D. Dieks (Ed.), The ontology of spacetime (pp. 111–127). Elsevier. Savitt, S. (2017). Being and becoming in modern physics. Stanford encyclopedia of philosophy. https://plato.stanford.edu/entries/spacetime-become Shimony. (1993). Search for a naturalistic world view (p. 1993). Cambridge University Press. Shoemaker, S. (1969). Time without change. The Journal of Philosophy, 66(12), 363–381. Skow, B. (2012). One second per second. Philosophy and Phenomenological Research, 85(2), 377–389. Skow, B. (2015). Objective becoming. Oxford University Press. Tooley, M. (1997). Time, tense and causation. Oxford University Press. Williams, N. (2019). The powers metaphysic. Oxford University Press. Wuthrich, C. (2012). Demarcating presentism. In H. de Regt, S. Hartmann, & S. Okasha (Eds.), EPSA philosophy of science: Amsterdam 2009. The European philosophy of science association proceedings (pp. 441–450). Springer.
Chapter 4
A Dynamic B Theory of Time Robert C. Koons
Abstract An adequate account of free will requires a robust powers ontology, but does it also require a “dynamic” A Theory of time? I argue that it is possible to construct a dynamic B Theory of time that is compatible with an Aristotelian ontology of causal powers and with free will. I conclude by considering whether extreme forms of A Theory, including a strict Presentism, are incompatible with a powers ontology. Keywords Powers ontology · Presentism · Eternalism · Free will
4.1
Free Will and Causal Powers
I begin with the assumption, well-defended by Peter van Inwagen (1983), that free will is incompatible with determinism. For an act to be free, there must be more than one option potentially choosable by the free agent, which entails that there be multiple possible futures, each consistent with the conjunction of past facts with the causal laws of nature or the distribution of causal powers and potentialities. Although necessary for free choice, such causal indeterminism is not sufficient. If we combine causal indeteminism with a neo-Humean theory of event causation, there is no room for the responsibility of the agent for his actions. The Humean tradition is replete with those (Hume himself, John Stuart Mill, and several authors published in Mind in the 1930’s – see Hume, 2000; Hobart, 1934; van Inwagen, 1983, pp. 142–50) who have argued that responsibility is incompatible with indeteminism, since in an indeterministic world it is chance, rather than the agent, that is responsible for the action actually taken. The best response to this Humean challenge takes the form of agent causation, a rejection of the Humean focus on event-to-event causation. Instead, it is supposed to be the agent himself that causes the action freely taken. The main drawback to this
R. C. Koons (*) University of Texas at Austin, Austin, TX, USA e-mail: [email protected] © The Author(s) 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_4
69
70
R. C. Koons
response is that it seemed to open up a deep fissure between causation in the natural world and causation in the domain of free agents. The recent revival of the ontology of causal powers closes this fissure, making agent causation simply a special case of ubiquitous causation by powerful particulars (Lowe, 2008; Swinburne, 2013; Jacobs & O’Connor, 2013). In this paper, I will concentrate on an Aristotelian version of powers ontology. In this version, the world consists fundamentally of substances and their accidents and relations. Each substance has a kind-essence, from which its fundamental causal powers and potentialities, both active and passive, flow with metaphysical necessity. The exercise of a power is to be identified with the actualization of a certain joint potentiality involving both an agent and a patient. In the metaphysical approach that I favor, both modality (possibility and necessity) and time are to be defined and explained in terms of causal powers. I follow Alexander Pruss’s advocacy of an Aristotelian conception of possibility (Pruss, 2011). A state S is possible at time t just in case either S is already actual at t, or there exist at t substances with the power to be bring it about that S be actual (either directly or indirectly).1 Similarly, I follow Aristotle in taking time to be the measure of change, and in taking change or motion to be the actualization of powers (potentialities) as such. Every power has its own intrinsic telos, and all change consists in a movement from actualization of a power toward that telos. Hence, each process and each interval of time has an intrinsic direction, a direction that does not depend on global patterns (like the increase of entropy) or asymmetries in the laws of nature.
4.2
Does Free Will Require the a Theory of Time?
J. M. E. McTaggart introduced in 1908 a distinction between two theories of time, the A and B Theories. The A Theory can be defined in either of two ways, one entailing the other. Hence, there is actually a trichotomy of theories of time: strict A, strict B, and intermediate theories. The narrow or strict definition of A Theory requires that the theory designate a single moment of time as metaphysically privileged, as the absolute Present moment. The broader definition of A Theory requires only that there be a real passage of time: that is, that there exist some metaphysically fundamental Clock whose successive states mark out the passage of time. It is clear that any strict A Theory must also be a broad A Theory, since the movement of the absolute Present would be a
1
In the Aristotelian framework, a chance event is a compound event that is the result of the exercise of two independent causal powers. So, for example, if a creditor and debtor meet by chance in the marketplace, each component of the meeting is the result of the exercise of certain locomotive powers. Thus, the meeting is in one sense fully explained in causal terms, even if there is no explanation of the coincidence as such. (Thanks to an anonymous reviewer for pressing this point.)
4 A Dynamic B Theory of Time
71
metaphysical Clock of the kind required. However, as we shall see, there are alternative Clocks that could be posited. Similarly, strict B Theory rules out the existence of a metaphysical Clock, while broad B Theory requires only that there be no metaphysically privileged Present. Hence, there are three possible positions: strict A Theory, strict B Theory, and the Intermediate Theory (which posits a metaphysical Clock but no privileged Present). There is also a super-strict form of the A Theory, namely, Presentism, which proposes that the Present moment exhausts all of reality. For the Presentist, absolutely Everything exists in the Present (in every sense of ‘exists’). In the Past there were things that do not exist now and in the Future there will be such things, but Nothing has such merely-past or merely-future existence. Consequently, many truths about the past and future lack truthmakers. Truth does not supervene on Being. I will argue that B Theory, even strict B Theory, is compatible with powers ontology and with free will. So far, I am in agreement with Benjamin Page in Chap. 10 and Andrea Roselli in Chap. 12 of this volume. However, in contrast to both Page and Roselli, I do not think that just any form of the strict B Theory is compatible with a powers ontology or with a powerist conception of free agency. In order to overcome the challenge of McTaggart’s arguments, the Aristotelian B Theorist must posit a plurality of modes of existence, following Aristotle’s dictum that ‘being’ is said in many ways (Metaphysics Gamma 2, 1003b5). Although the dynamic B Theory does provide an adequate foundation for free will, I will propose an Intermediate Theory which better respects our experience of the fleetingness and inexorable motion of time. And I will argue that Presentism is compatible with powers ontology and with free will only if it also recognizes the existence of multiple modes of being.
4.3
The Challenges of Logical Fatalism and the Unreality of Change
There are two reasons for thinking that the B Theory might be incompatible with both a causal-powers ontology and free will. These are the challenge of logical fatalism, and McTaggart’s charge that the B Theory is inconsistent with a robust conception of change (of the kind required by Aristotelianism). The logical fatalist challenge consists in pointing out that the B Theory entails that it is metaphysically necessary that the future be exactly as it is now. For B Theorists, the future is “out there,” existing in some fully determinate condition in the direction of later time. Hence, there can be no sense in which multiple, mutually inconsistent futures could all be (here and now) really possible. Consequently, all of our future supposedly free actions are pre-determined, and no one is free. Similarly, the causal powers account of modality collapses to triviality: whatever will be is necessarily so, and there never has or has been any contingency. But, in the absence
72
R. C. Koons
of contingency, how could the exercise of power (understood as the actualization of potentiality as such) be possible? The Aristotelian B Theorist can simply reject the inference from the fixity of the actual future to its necessity (as Benjamin Page also recognizes in Chap. 10). At each moment of time, reality consists of a branching tree, with a single trunk leading to the present moment (corresponding to the necessity of the past) but with multiple branches breaking off from the present moment and from many possible future events. These alternative branches represent alternative possibilities (potentialities). Potentiality is time-relative. Certain events are potential-relative-to-t1 but not-potential-relative-to-t2, where t2 is later than t1. To be still more precise, we need to relativize potentiality to an event. One event E is potential relative to an event F just in case F includes substances with the power to actualize E (directly or indirectly). We can also define relative actuality: an event E is actual relative to event F just in case, for every actual event G, there is no potentiality-relative-to-G of the joint state of F and not E. In other words, the actuality of F necessitates the actuality of E. Every possible state or event E is actual-at-E. An event E is also actual at all events later than E but may be merely-potential-at some events earlier than E. An event E is actual at time t iff there is an actual event F occurring entirely at t such that E is actual-relative-to F. An event E occurs entirely at t iff all standard clocks registering time later than t are potential relative to E. A potentiality E is actualized at t if E is actual-at-t but not actual at all times earlier than t. But, if every event E is actual-at-E, how can we differentiate between those potential future events that will in fact be actual from those that will not? We must remember that the Aristotelian B Theorists has, in addition to the event- and timerelative notions of actuality we just defined, a more fundamental, absolute notion of actuality. Crucially, we did not define actual-relative-to-t as absolutely actual and occurring at t. Something need not occur at t in order to be actual relative to t. We do get the following relationships between absolute and relative actuality: (1) Event F is absolutely actual if there is some event E such that (i) E’s occurrence is absolutely actual, and (ii) F is actual relative to E. (2) Event F is absolutely actual iff there is some time t such that F is actual relative to t. But, absolute actuality cannot be defined in terms of either form of relative actuality. Instead, the definition runs in the opposite direction. But how can a substance actualize a potential event E at t if that event is already (at t) either absolutely actual or absolutely non-actual? The answer, of course, is that merely potential future events that are absolutely actual are not already actual (that is, they are not actual relative to the present). The domain of absolutely actual events is partially ordered by time and causation. An absolutely actual event E that becomes actual at some point in time is absolutely actual because it was made actual by the exercise of some absolutely actual power, a power that was itself already actual at
4 A Dynamic B Theory of Time
73
times and events prior to E. It is not the case that it is actual at that time because it was already absolutely actual. But how can the same event be merely-potential-at-t while at the same time being absolutely actual? Aren’t mere potentiality and actuality contraries? Yes, but beingmerely-potential-at-t is contrary to being-actual-at-t, not to being actual absolutely. But how can it be true (at t and at all times) that event E is absolutely actual, while false that E is actual-at-t? Here we have to rely on Aristotle’s dictum that “being is said in many ways.” This entails that ‘actual being’ is also “said in many ways.” That is, there are multiple modes of actual being, one for each moment in B time, and another one for eternity. The time- and event-relative modes of actuality are derived and not fundamental, and so their behavior is determined by those definitions, which include a modal dimension. Consequently, being absolutely actual at time t need not entail being-actual-at-t. McTaggart offered the second challenge to a B-Theoretic version of Aristotelianism. McTaggart argued that the B Theory is incompatible with the reality of change, as conceived of in classical or Aristotelian terms. The B Theory seems to be committed to Bertrand Russell’s At-At theory of change: a substance S undergoes change at time t iff there is are intervals of time I1 and I2 with I1 ending at t, and I2 beginning at t, and S has some property P throughout I2 and at no time in I1 (except for t itself). That is, something is undergoing change just in case it has some properties at certain times that it lacks at other times. The At-At theory is incompatible with Aristotelianism, since it implies that time is metaphysically prior to change, and not vice versa. However, Aristotelian B Theorists need not embrace Russell’s theory. They can insist that change be identified with the exercise of causal powers. If a substance changes from cold to hot, there are B moments at which it is hot because some actual causal power was exercised. This differentiates real change from simple variation, like McTaggart’s rod, which is simultaneously hot on one end and cold on the other. If we say that McTaggart’s rod “changes” from being hot on the left to being cold on the right, we are merely speaking metaphorically (as McTaggart insisted upon). However, the Aristotelian B Theorist does not conceive of real change as being at all analogous to the rod. McTaggart’s rod is not hot on one end because some causal power as acted upon its cold end. However, a brick that changes from being cold to being hot is actually hot at the end because some causal power was exercised upon it in its cold state. But isn’t there still something unacceptably static about the strict B Theory? I have some sympathy to this response, but as an objection to the strict B Theory, this disquietude has nothing to do with Aristotelianism or with free will. And, as I will argue in Sect. 4.6, we can dispel this disquietude by embracing an intermediate theory of time, without going all the way to a strict A Theory.
74
4.4
R. C. Koons
A Formal Model of Dynamic B Theory
A formal theory of dynamic B Theory already exists – it is the Ockhamist branching time model created by Arthur Prior (Prior, 1967, 2003 and further developed by Hans Kamp (1968), John Burgess (1980), and others (Reynolds, 2002; Øhrstrøm, 2009; Goranko & Rumberg, 2020). (See also Plantinga, 1986.) I am going to re-package the logic here, in order to make clearer its connections with an Aristotelian B Theory. To do so, I first draw on Storrs McCall’s falling branches version of A Theory (McCall, 1976). McCall’s theory fits very well with Pruss’s Aristotelian theory of modality. At each moment of time, McCall models reality as a branching tree, with a single trunk. The trunk represents the past, and the present moment is the first point in the trunk from which two or more paths branch. Possible future events occur on the branches. As real (A) time passes, the future moves out to later points in time, and as the present moves, paths that branch from the trunk at earlier times disappear (“fall off the tree”). We will need two pairs of tense operators: potentially-past and actually-past, and potentially-future and actually-future. The branching structure of time will guarantee that the potentially-past events will coincide exactly with the actually-past ones. In my B-theoretic version of this model, we will have to evaluate tensed propositions relative to three parameters: a model M (representing the actual world as a whole), a tree T (representing an actual moment in B time, past, present or future), and a time t (representing a possible moment in the A future, relative to a tree). Tensed propositions are evaluated according to Arthur Prior’s semantics: M, T, t ‘ P pot ðϕÞ , ∃s s 2 T, s t∶ M, T, s ‘ ϕ M, T, t ‘ F pot ðϕÞ , ∃s s 2 T, t s∶ M, T, s ‘ ϕ: M, T, t ‘ P act ðϕÞ , ∃S 2 M T ⊂ S∶ M, S, NðSÞ ‘ ϕ M, T, t ‘ F act ðϕÞ , ∃S 2 M S ⊂ T∶ M, S, NðSÞ ‘ ϕ The N(S) term refers to the present moment of tree S, that is, the latest moment prior to which there are no branches in S. The resulting model is simply a B-Theoretic version of McCall’s. Instead of having a single, changing tree, the model consists of an infinite number of trees, each tree representing a single moment of B time. We stipulate that the trees in a given model fall into a unique linear ordering of time. One tree T1 is earlier than another tree T2 just in T2 is a proper sub-graph of T1 (i.e., T2 results from pruning off some of T1’s branches), or, equivalently, the present time of T2 corresponds to some possible future time of T1. Thus, we get two temporal dimensions, one within each B moment (as a counterpart to A time), and one that orders all of the B moments into a B series. This is equivalent to the Ockhamist branching-time model. The model theory will guarantee that Ppot and Pact will necessarily coincide, since trees branch toward the future and not toward the past. The definition of precedence
4 A Dynamic B Theory of Time
75
between trees ensures that the interpretation of Fact is a subset of Fpot, i.e., that the actual future is among the potential futures. The Fact operator will rigidly refer to what actually willWfollow W the present time, and so PpotFact φ will be equivalent to the disjunction ðF φ φ P φÞ (Kamp, 1971). This captures the linearity of B time – its non-branching character. It would be possible to add a linear ordering of trees to each node of each tree, which would give us a Stalnaker-like logic for counterfactual conditionals, but this would be metaphysically dubious. Consequently, I will stick with the simpler model, consisting of a linear ordering of McCall trees. As is generally the case with the B Theory, it is easy to adapt this model for consistency with the special theory of relativity. Instead of having a tree for each instant of time, we will need to have one for each point in spacetime. The branching futures in each tree will represent every future that is contained within the forward time cone of the spacetime point. We can use the same definition of ‘later than’, which will now hold of two trees when the first lies within the forward time cone of the first. The trees will no longer be in a linear ordering in the model, but they will have a strict partial ordering that corresponds to the earlier/later relation between time-like separated events. And we can assume that any two trees (even space-like separated ones) have both a common tree in the past and a common tree in the future.
4.5
McTaggart’s Challenge to the a Theory Applied to Dynamic B Theory
J. M. E. McTaggart argued in 1908 that the A Theory is inconsistent, since it entails that each event and each moment of time have contradictory properties – namely, the properties of pastness, presentness, and futurity. McTaggart argued that the obvious A Theoretic response, pointing out that each event has these properties at different times, involved a vicious circularity since the distinction between different times depended on the coherency of the A Theory. Regardless of whether this argument is successful against the A Theory, a version of it can be deployed against my dynamic B Theory with some real force. Each event belongs to infinitely many trees: in one it occurs at the present moment, in some it is in the past, and in others it is among the possible futures. But, as McTaggart pointed out, presentness, pastness, and futurity are contrary properties. So, for that matter, are actuality and mere potentiality, or necessity and possible non-existence. How can the same event bear such contraries? The Aristotelian cannot give the standard reply given by A Theorists, namely, that the contrary properties are had “at” different times, because all of the B moments eternally co-exist according to B Theory. If we suppose that presentness, pastness, and futurity are binary relations between events and B moments, then we would have to give an account of the nature and identity of B moments that is prior to the modalities of tense (which was the gist of McTaggart’s charge of vicious circularity). However, dynamic B Theorists cannot posit B times as existing prior to causation
76
R. C. Koons
and change, since this would be to embrace an unacceptably static and spatialized conception of time, as if time were a metaphysically prior framework into which events can be placed. For similar reasons, dynamic B Theorists cannot resort to events’ having disjoint temporal parts, as if the death of Julius Caesar had a part that has futurity in 100 B.C. and another part that has pastness in 100 A.D. There is a plausible response to this circularity objection. Instead of relativizing presentness, futurity (potentiality), and pastness to moments of B time, we could instead make them relative to events. One event is potential relative to another just in case the second involves the appropriate causal powers. Moments of B time could be identified with classes of events – maximal classes such that no member is potential relative to the other. This raises a question for the Aristotelian: what are events? Obviously, it would be problematic to identify events with triples of entities, properties, and times, since this would re-introduce B times into the picture at a more fundamental level. We could identify events with substances and accidents: the substance Socrates corresponding to the “event” of Socrates’ existing, and the accident of Socrates’ paleness corresponding to the “event” of Socrates’ being pale. This will work so long as accidents are not repeatable. Suppose, for example, that Socrates was pale every winter and ruddy every summer. There are clearly distinct episodes of Socrates’ paleness, so we could not identify one of these events with the accident of Socrates’ paleness if that accident recurred (i.e., if it had intermittent existence). So, Aristotelians should deny that accidents can recur, positing that the origin of each accident is essential to it. Nonetheless, this notion of event-relative potentiality ultimately falls prey to a version of McTaggart’s objection. This version of the objection relies on simple observation: potentiality is a monadic property of events---it is not a relation to anything. I have been guilty so far of confusing two things: the fact that the potentiality of certain facts is grounded in the existence of causal powers, and the supposed fact that potentiality is a relation between one fact and another. An event is either potential or it is not – it makes no sense to ask whether it is potential or not in relation to some other event, just as it is nonsense to ask whether an event is future in relation to another event (as opposed to later than that event). What options then would the dynamic B Theorists have? I think we must embrace the thesis that the copula itself must be tensed (Johnston, 1987). As I mentioned above, ‘being’ is said in many ways. There are distinct modes of present/actual being, one for each B moment. Events (including substances and their accidents) have different modes of actual being, depending on when they exist. Pastness and futurity are modalized derivatives of these modes of being. Each B moment involves only a single mode of actual being, plus an infinite number of tensed derivatives. Consequently, if m is the mode of being corresponding to 100 B.C. and n the mode corresponding to 100 A.D., then the death of Caesar ism possible-in-the-future, and it isn necessary-in-the-past. If o is the mode corresponding to the moment of Caesar’s death, then that death iso simpliciter. In addition to such temporal modes, there is a single mode of eternal being, e. All substances and accidents aree simpliciter, but none aree past or future. Eternal being is the focal meaning of ‘being’.
4 A Dynamic B Theory of Time
77
There is now no contradiction in supposing that Caesar’s death ism possible-inthe-future but isn not possible-in-the-future, nor ise it possible-in-the-future, since distinct modes of being are involved. The various temporal modes of being are causally ordered, as mentioned above. If m is prior to n, then that things aren as they are is causally dependent on how things arem. The principal objection to tensing the copula, pressed by David Lewis (2002), is the charge that it alienates bearers from their intrinsic properties, in much the same way as does treating tensed being as a binary relation between an entity and a time. If the rod ism hot, then it is not hot simpliciter but only in relation to the mode m. This seems inconsistent with heat being an intrinsic attribute of the rod. However, it is a mistake to read x ism as involving a relation between x and some entity m. The index ‘m’ is merely being used to distinguish ‘ism’ from other modes of being. It is not in any sense a relational predicate. To say that ism is a mode of being is precisely to insist that it be non-alienating. To say that x ism P is to attribute P to x intrinsically. There is an epistemological version of the alienation problem: what is the content of one’s thought when one thinks, S is now P? If, as in standard B Theory, we take the present tense and the now-adverb to be indexical in character, then we seem to be forced to suppose that we are thinking that S and P stand in some relation to this moment (i.e., the moment at which we are thinking). How, then, can we think that S is P, using a non-alienating copula? The solution lies in an often-neglected suggestion of Gottlob Frege (1997), an idea that has been developed by Maxwell Goss in his 2006 dissertation (Goss, 2006). When I think S is now P, I am using the only temporal mode of being that is directly accessible to my thought at this moment in time. At each moment in time, I am acquainted with only the mode of being corresponding to that moment. All other moments must be designated by description: as the moment of time n seconds earlier (later) than now. Hence, I never relate a predication to the present moment; instead, I simply use the relevant mode of being. Thus, the present tense involves no cognitive alienation of an entity from its intrinsic attributes (whether essential or accidental). We can also grasp the eternal mode of being. We can assert, for example, that 2 + 2 ise 4. Similarly, we can assert that Bucephalus existse and even that Bucephalus ise a horse. We can say that Socrates’ musicality or his paleness aree, but we cannot say that Socrates ise pale. In this sense, there is a certain degree of alienation between a substance and its transient accidents.
4.6
Moving from the Dynamic B Theory to the Intermediate Theory
The dynamic B Theory developed in Sects. 4.3, 4.4, and 4.5 suffices as a justification for the claim that Aristotelian powers ontology, the B Theory, and libertarian free will are mutually compatible. However, once we have the distinct modes of being in
78
R. C. Koons
place, there is room for adding yet another dimension of dynamism to the account, moving from the strict B Theory to the Intermediate account. On the Intermediate account, there is no privileged Present but there is fundamentally moving metaphysical Clock. Suppose that the modes of being are not permanently attached to B moments of time but are in continuous motion through the B series. In other words, the mode of being m, which is the mode associated with the present moment, was once associated with 100 B. C. and will one day be associated with 2100 A.D. A function that relates modes of being to the B moments to which they belong would then constitute a metaphysical Clock, keeping track of the real passage of time without introducing a uniquely privileged Present. If we relate this to the Fregean idea of direct acquaintance with the present moment, we could imagine that all of my memories and all of my anticipations of the future involve the very same mode of being. I remember when this mode, say m, was associated with the year 1980, and I look forward to when it will be associated with the year 2030. Thus, the temporal phenomenology of the fleetingness of the present and of the inexorable forward motion of time would both be vindicated, if we think of my present consciousness as borne along by the same mode of being. This degree of dynamism can be squared with a rejection of strict A Theory, since there is a future self who remembers when he wasn a child, and a past self that looks forward to beingo a philosopher, where n and o are modes of being associated now with future and past times, modes with which I am not now acquainted. My present mode m is not metaphysically privileged, even though it does move through the B series.2 Is the Intermediate Theory, like the strict B Theory, consistent with the standard interpretation of special relativity, that is, with the non-existence of absolute simultaneity? Yes, it can be. Instead of supposing that the real passage of time is globally synchronized, we could suppose that the passage at one point in space cannot be objectively coordinated with that at a different point. In other words, instead of introducing a single, global metaphysical Clock, we could introduce a multiplicity of such Clocks, one for each enduring entity (including substances and their integral parts and remnants). This would mean introducing a further multiplicity to the modes of being, with each enduring entity possessing its own private succession of modes. One last advantage to the Intermediate Theory: it can offer a principled answer to J. C. C. Smart’s challenge (Smart, 1949), how fast does time flow? We can measure the speed of the flow of time for a substance at a point in B time by comparing its motion with the flow of time for that same substance at other B moments. It is possible for one of my modes m to travel x seconds (currently at 1980) for every y seconds traveled by a different mode n (located in 1990). If so, my proper time at 1980 would be flowing at x/y 1980-seconds/1990-second. A particular mode can
2 My talk of ‘past selves’ and ‘future selves’ shouldn’t be taken too literally. I do not want to deny that substances endure through time. If there is some tension between this commitment to persisting substances and instantaneous modes of being, I could instead suppose that modes of being correspond to periods of time, rather than to instants.
4 A Dynamic B Theory of Time
79
accelerate or decelerate in its velocity through time (as measured by the movement of a different mode), moving at different relative velocities at different moments in A time (as registered by the enduring entity’s metaphysical Clock).
4.7
Problems for Aristotelian A Theory
The most plausible version of strict A Theory is the super-strict version, Presentism. Presentism provides A Theorists with its best solution to the problem of intrinsic change. According to Presentism, if a thing x is now P, then x is P simpliciter. To have been P or to be going to be P, in contrast, involve the application of non-factive modal operators. These non-factive operators could be expressed as counterfactual conditionals. E.g.: x was P n seconds ago means x would be P if n seconds ago were now. Presentism is typically combined with Actualism, the view that everything exists in actuality. Actualists refuse to countenance non-existing things within their domain of quantification. The combination of Actualism and Presentism banishes all merely past or merely future things from being values of quantified variable. We cannot say that some things exist only in the past. Instead, we have to say that in the past there were things that don’t exist now (embedding the quantifier within the tense operator). In fact, even this is problematic, since it seems to involve attributing the impossible attribute of non-existence to presently-existing things within the scope of the past operator. We would have to follow Robert Adams (1981) in distinguishing between not-existing-in a time (which is impossible) and not-existing-at a time (which is possible). There is a problem about combining Actualist-Presentism with an Aristotelian account of causal powers. A powers account of causation must suppose that temporally extended processes (kineses) are at least as fundamental as their instantaneous parts. Anti-powers theorists can embrace a Russellian At-At account of motion and change, and so they may say that the instantaneous states of being are prior to the whole process, since the laws of nature determine the character of each such instantaneous state. However, a powers theorist cannot suppose that the process is a result of a non-denumerable infinity of power-exercises, one for each instant, since the aggregate of such instantaneous exercises would leave the diachronic character of the whole process unexplained. Instead, powers theorists must suppose that the exercise of a causal power initiates a process or enduring state, whose existence and character are more fundamental than those of its constituent instantaneous states. There must be a top-down, whole-to-part direction of explanation. This is also necessary to avoid Zeno-like paradoxes and paradoxical super-tasks (like Thomson’s lamp). (See Chap. 9 of Koons & Pickavance, 2015, and Chap. 28 of Koons & Pickavance, 2018.) If reality consists fundamentally, at least in part, in temporally extended things (processes, motions, activities), then it cannot be the case that everything that exists exists now and at no other time. If an ongoing process does not consist in a mere
80
R. C. Koons
aggregation of instants, then it cannot exist simply by virtue of the existence of the part of the process that corresponds to the present moment. The whole, temporally extended process must exist in actuality, and that means that it must actually exist at other times as well. If the process existed only in the instantaneous present, then it wouldn’t be temporally extended, at least not fundamentally. The only kind of temporal extension that is possible under Presentism is a kind of modalized extension. So, I could say that Homer’s present activity of walking across the stadium is temporally extended in the sense that there were times in the past at which it existed, and there will be times in the future at which it will exist. However, this kind of temporal extension depends on the process being entirely composed, over time, of instantaneous parts, each of which exists for just a moment (when it is present). Here is an argument for the falsity of Presentism on Aristotelian grounds: (a) Some things (i.e., processes) are fundamentally extended in time. (b) If something that exists is fundamentally extended in time, then it does not exist at a time simply by virtue of having an instantaneous part that exists then. (c) If Presentism is true, then any process can exist at a time only by virtue of having an instantaneous part that exists then. (d) So, Presentism is false. The best response by an Aristotelian A Theorist would be to turn once again to the thesis that ‘being’ is said in many ways. The A Theorist can adopt Thomas Aquinas’s four-way distinction from his commentary on Aristotle’s Metaphysics. Thomas Aquinas theorizes four levels or degrees of being: substances, accidents, motions (processes), and negations (In Meta IV, lesson 1, 540–3). We could precisify the definition of Presentism to make clear that we should limit it to the existence of substances (existence1) and accidents (existence2), excluding motions (existence3) and privations (existence4): Presentism Everything that actually exists1 or exists2 exists (1 or 2) in the present, and nothing exists1 or exists2 (in actuality) at any other time. The Aristotelian A Theorist can now challenge premise (c) of my first argument: (c) If Presentism is true, then any process can exist at a time only by virtue of having an instantaneous part that exists then. Our modified version of Presentism allows processes to exist3 over a period of time, without fundamentally existing (1 or 2) at the intervening instants of time. Modified Presentism simply doesn’t apply to processes at all: they are free to exist in part in the past or future, as well as the present, since they never exist in either of the two most fundamental ways (existence1 or existence2). Presumably this can be done without compromising Presentism’s explanation of the possibility of intrinsic change, since processes do not undergo the same kind of intrinsic change that substances do. For processes, an At-At theory of change seems appropriate.
4 A Dynamic B Theory of Time
4.8
81
Conclusion
Aristotelian powers ontology provides the best basis for libertarian free will. Such an ontology is compatible with the B Theory of time, so long as the latter takes a sufficiently dynamic form. This requires a real and irreducible notion of eventrelative potentiality, which is, in turn, defined in terms of the existence of appropriate causal powers. The dynamic B Theory rejects the spatializing of time and Russell’s At-At theory of change in favor of an account in which powers and change are metaphysically prior to time. In order to overcome the challenge that McTaggart lodged against the A Theory, dynamic B theory must posit a plurality of modes of being. These temporal modes of being must be thought of as tensed copulas and not as relations to moments of time. Once modes of being are in place, it becomes possible to move from a dynamic B Theory to a still more dynamic Intermediate theory, with real passage of time but no privileged Present. The compatibility of Presentism, the most plausible version of the A Theory, with Aristotelian powers ontology faces the challenge of metaphysically fundamental, temporally extended processes, processes which cannot persist by virtue of wholly existing at each instant of time. Overcoming this challenge requires the A Theorist to acknowledge multiple modes of being, with a special mode reserved for processes. This fact strengthens the dynamic B Theory and the Intermediate Theory, since the multiplicity of modes of being is common ground for all versions of Aristotelian powers ontology.
References Adams, R. M. (1981). Actualism and thisness. Synthese, 49, 3–41. Burgess, J. (1980). Decidability for branching time. Studia Logica, 39, 203–218. Frege, G. (1997). Thought. In M. Beaney (Ed.), The Frege reader (pp. 325–345). Blackwell Publishing. Goranko, V., & Rumberg, A. (2020). Temporal logic. The Stanford encyclopedia of philosophy (Summer 2020 Edition), Edward N. Zalta (Ed.). https://plato.stanford.edu/archives/sum2020/ entries/logic-temporal/ Goss, M. (2006). Time and human nature: A modest defense of eternalism. PhD dissertation in philosophy. University of Texas at Austin. Hobart, R. E. (1934). Free will as involving determination and inconceivable without it. Mind, 43, 1–27. Hume, D. (2000). In D. F. Horton & M. J. Horton (Eds.), A treatise of human nature. Oxford University Press. Jacobs, J. D., & O’Connor, T. (2013). Agent causation in a neo-Aristotelian metaphysics. In S. C. Gibb & R. Ingthorsson (Eds.), Mental causation and ontology (pp. 173–192). Oxford University Press. Johnston, M. (1987). Is there a problem about persistence? Proceedings of the Aristotelian Society Supplement, 61, 107–135. Kamp, J. (1968). Tense logic and the theory of linear order, PhD Thesis. University of California.
82
R. C. Koons
Kamp, J. (1971). Formal properties of ‘now’. Theoria, 37(3), 227–273. Koons, R. C., & Pickavance, T. H. (2015). Metaphysics: The fundamentals. Wiley-Blackwell. Koons, R. C., & Pickavance, T. H. (2018). The atlas of reality: A comprehensive guide to metaphysics. Wiley-Blackwell. Lewis, D. K. (2002). Tensing the copula. Mind, 111, 1–14. Lowe, E. J. (2008). Personal agency: The metaphysics of mind and action. Oxford University Press. McCall, S. (1976). Objective time flow. Philosophy of Science, 43, 337–362. McTaggart, J. M. E. (1908). The unreality of time. Mind, 17, 456–473. Øhrstrøm, P. (2009). In defense of the thin red line: A case for Ockhamism. Humana Mente, 8, 17–32. Plantinga, A. (1986). On Ockham’s way out. Faith and Philosophy, 3, 235–269. Prior, A. (1967). Past, present and future. Oxford University Press. Prior, A. (2003). In P. Hasle, P. Ohrstrom, T. Brauner, & J. Copeland (Eds.), Papers on time and tense (3rd ed.). Oxford University Press. Pruss, A. R. (2011). Actuality, possibility, and worlds. Continuum. Reynolds, M. (2002). Axioms for branching time. Journal of Logic and Computation, 12(4), 679–697. Smart, J. C. C. (1949). The river of time. Mind, 48(232), 483–494. Swinburne, R. (2013). Mind, brain, and free will. Oxford University Press. van Inwagen, P. (1983). An essay on free will. Clarendon Press. Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Chapter 5
Libertarian Freedom in an Eternalist World? Ben Page
Abstract My students sometimes worry that if eternalism is true then they can’t have libertarian freedom. They aren’t alone, as this sentiment is also expressed, albeit typically briefly, by various philosophers. However, somewhat surprisingly, those working within the free will literature have largely had nothing to say about libertarianism’s relationship to time, with this also being similar in the case of those working in the philosophy of time, apart from some work which has mainly focused on nonlibertarian views of freedom. In this short paper I note why I’m currently unconvinced that there’s an incompatibility between eternalism and libertarianism, and in doing so one will see why I think they are compatible. Keywords Libertarianism · Free will · Eternalism
My students sometimes worry that if eternalism is true then they can’t have libertarian freedom. They aren’t alone, as this sentiment is also expressed, albeit typically briefly, by various philosophers (Koperski, 2015, 112–117; Lockwood, 2005, 69; Levin, 2007; Petkov, 2005, 152).1 Surprisingly, those working within the free will literature have largely had nothing to say about libertarianism’s relationship to time,2 with this also being similar in the case of those working in the philosophy of time,
My thanks to James Read, Robin Le Poidevin, Tim Mawson, Damiano Costa, and an anonymous referee for comments on an earlier draft of this paper. I also wish to acknowledge my students for interesting discussions on this topic, which provided inspiration for writing something on this subject. 1
For additional references, see Oaklander (1998). Apparently, some have no interest in the matter either, with Koperski writing that ‘Timpe [who specialises in the philosophy of free will] has said that he is not terribly concerned about the metaphysics of time (private conversation).’ (2015, 115).
2
B. Page (*) Pembroke College & Oriel College, University of Oxford, Oxford, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_5
83
84
B. Page
apart from some work which has mainly focused on non-libertarian views of freedom (Oaklander, 1995, 1998; Le Poidevin, 2013; Hoefer, 2002; Brennan, 2007; Rychter, 2015; Buckareff, 2019). In this short paper I note why I’m currently unconvinced that there’s an incompatibility between eternalism and libertarianism, and in doing so one will see why I take them to be compatible.
5.1
Eternalism and Libertarianism
To begin, it will be useful to clarify the central tenets of both eternalism and libertarianism. Eternalism claims that all events, time-slices or entities that ever were, ever are, or ever will be, exist.3 Eternalism therefore affirms what Baron and Miller call ENTITY EVERYWHENISM: entities that exist now, exist in the past, and exist in the future, exist simpliciter (2019, 15). This contrasts other views, such as presentism, in which only the present moment exists, ENTITY NOWISM, and the growing block view, where only the present and past exist, ENTITY NOW AND THENISM. Eternalism, as I’m thinking about it, comes in two main varieties, one which is combined with the A-theory of time, namely the moving-spotlight, and the other which is partnered with the B-theory.4 For my purposes, most of what I say will be applicable to both types of eternalism. Turn now to libertarianism. Chiefly libertarians are incompatibilists who think that agents can act freely.5 As such they think causal determinism doesn’t allow for free will, contra compatibilists, and that when agents act freely they are not causally determined. We can understand causal determinism here as Kane characterises it when he says, ‘an event (such as a choice or action) is determined when there are conditions obtaining earlier (such as the decrees of fate or the foreordaining acts of God or antecedent causes plus laws of nature) whose occurrence is a sufficient condition for the occurrence of the event. In other words, it must be the case that, if these earlier determining conditions obtain, then the determined event will occur.’ (Kane, 2005, 5–6) As such the libertarian claims that when an agent acts freely by doing X, it is false to say that X is determined by any conditions obtaining prior to X.
Note that throughout the paper I will speak of time-slices. I do this since I find it easier to think and speak about the block being ‘sliced’ up in this way. One big slice after another. I take it that nothing substantial rests on this, and one can translate my talk into events or entities if one prefers. It may even be that translating this talk into events is preferable since, as my colleague James Read informs me, ‘the Gödel universe isn’t foliable into hypersurfaces of constant time --- but we can still (the thought typically goes) be eternalists about it through thinking about it in terms of events.’ (in personal Correspondence) I let the reader do so if they wish. 4 There might be some slightly harder varieties of eternalism to classify, such as Maudlin’s (2007, ch.4), and perhaps Koons’s (2022), but I take it that the vast majority will fit comfortably into my classification. In any case, what I will largely be concerned with is the claim of ENTITY EVERY WHENISM, and this is one all eternalists seem to embrace. 5 Rather than the other type of incompatibilists, hard determinists, who think agents can’t act freely. 3
5 Libertarian Freedom in an Eternalist World?
85
Given this characterisation of the two views, one way to see if eternalism rules out libertarian freedom is by seeing if it implies causal determinism about agential actions, since if it does then libertarianism will turn out to be false.6 However, eternalism, at least as I have characterised it, has nothing whatsoever to say about whether causal determinism is true. All eternalism says is that all time-slices exist. Eternalism says nothing about the relationship between these time-slices, nor the relationship between those entities contained within, or if you prefer, are parts of the time-slices.7 For it seems at least prima facie conceivable that the relationship between time-slice t1 and time-slice t2 or the entities contained in t1 and t2 be indeterministically related.8 Why think that eternalism entails that a fundamental particle in t1 must be causally determined to do what it does in t2? Or that my existing in t1 implies that I’m causally determined to do Y in t2? If eternalism were to imply this, then eternalism alone would imply that any possible world with the same initial time-slice and laws of nature would always bring about the same future.9 But eternalism doesn’t entail this. What would make this the case is if the laws of nature implied that everything is causally determined.10 But if that’s right, then unless one can point out why eternalism requires causal determinism it seems the libertarian can claim that eternalism is compatible with the falsity of causal determinism, which at least starts to give the libertarian what they need.
5.2
Alternative Possibilities and Fixity
What I’ve said so far seems fairly obvious, so it must be another aspect of libertarianism that is at odds with eternalism. Here is an obvious candidate. Libertarians often claim that in order to be free one must have the ability to have done otherwise,
6 One could read Petkov as thinking this, as he writes that four-dimensionalism, by which I take him to mean eternalism, implies that our actions are ‘predetermined’ (2005, 152). Similarly Rietdijk (1966), who seems to think that all time-slices exist, claims that because of this there is no indeterminism, a view Putnam (243–247) also seems to hold. There is a little ambiguity, however, in what both mean when they speak of determinism, with Rietdijk saying that ‘we may conclude that there is determinism (which is, of course, not the same thing as causality)’ (1967, 343). Perhaps then what is meant is not causal determinism, but that events in the future are ‘set’ or what I will call later ‘fixed’, and it is for this reason that they call them determined. However, if this is all they mean I see no reason why this would imply that there couldn’t be an indeterministic link between two time-slices, even though it is set or fixed what will occur in our future. 7 For instance, my actions, since I take it that my actions do not exhaust the time-slice. There are, after all, your actions as well. 8 Instead of entities you could read this as parts of the time slices. 9 Defining determinism in terms of possible worlds makes this clear. Such as Clarke’s definition, where ‘one event is taken to deterministically cause another just in case, in every possible world in which the actual laws of nature obtain and in which the first event occurs, it causes the second.’ (2003, 4). 10 Assuming that the laws of nature were the only determining factor.
86
B. Page
with this being known as the principle of alternative possibilities (PAP).11 Those libertarians who take this to be the key feature of libertarianism have become known as ‘leeway incompatibilists’. The worry then may be that eternalism rules out PAP, since my acting in time-slice t500, which is in my future, currently exists given that eternalism implies ENTITY EVERYWHENISM , and hence at t500 I cannot act otherwise. As such, I cannot be libertarianly free (Levin, 2007, 452; Petkov, 2005, 152; Koperski, 2015, 115). I will say something about this shortly, but first let me mention that even if eternalism does rule out PAP, this need not mean that it rules out every type of libertarianism. The reason for this is that a leeway conception of libertarianism is not the only type of libertarianism there is. Another type is known as ‘sourcehood incompatibilism’, which holds that the key feature of libertarian action is that the agent is the ultimate source or originator of their action.12 There are several forms of this type of libertarianism (Tognazzi, 2011), but for our purposes we need only distinguish between two. First is ‘wide-sourcehood libertarianism’ which says that whilst sourcehood is the key feature for libertarian action, being able to act otherwise, PAP, is an essential secondary component (Kane, 1996; Timpe, 2013, 155–161; Pereboom, 2001). The second is ‘narrow-sourcehood libertarianism’, which claims that being the source is the key feature of libertarian action and that alterative possibilities, PAP, are not required for an act to be free (Zagzebski, 2000; Stump, 1999, 2003; Hunt, 2000; Shabo, 2010).13 Suppose that narrow-sourcehood libertarianism is a viable option. Then at least one form of libertarianism is compatible with eternalism, supposing that eternalism does not bring with it any other concerns for libertarianism.14 For this form of libertarianism does not require PAP, but only indeterminism, which as I have already suggested, is compatible with eternalism, and sourcehood, which I shall comment on further later in this article. Given this, if PAP is the reason why eternalism and libertarianism are incompatible, one must show that narrow-sourcehood libertarianism isn’t in fact a viable option. Whether it is or isn’t is not something I shall comment on here.15 Nevertheless, it’s fair to say that many, perhaps most, libertarians do hold PAP and it would therefore be beneficial if eternalism was open to them as well. So is PAP incompatible with eternalism? I think not. To see why note that one way a libertarian might understand PAP is as follows:
11
PAP sometimes includes a claim about moral responsibility, however here I just mean it to refer to having alternative possibilities. 12 Note that there are also sourcehood versions of compatibilism (Timpe, 2017), but I focus on libertarianism here. 13 This position is also attributed to Clarke in Fischer (1999, 129, footnote 66). 14 In fairness, and unlike others, Koperski (2015, 115) does mention this, although he moves past it fairly promptly. 15 Timpe (2013, 155–161), for instance, has suggested that narrow-sourcehood does require PAP. If he is correct, then it isn’t a viable option and this way to make eternalism and libertarianism compatible fails. By contrast Shabo thinks all sourcehood accounts should reject PAP (2010, 350, n.4). Who is in fact right about this, I leave undiscussed here.
5 Libertarian Freedom in an Eternalist World?
87
An agent S has the ability to choose or do otherwise than ϕ at time t if and only if it was possible, holding fixed everything up to t, that S choose or do otherwise than ϕ at t. (O’Connor & Franklin, 2020)16
Key here is that what is required is that it is possible for an agent to perform an action different from the one she does.17 This is a modal clause. One way to see if this condition is met is to think about whether an agent has access to some possible worlds beside our actual one (van Inwagen, 1983, 91).18 To do this think of all the closest possible worlds which have the same worldly history and the same laws of nature up until the moment of choice and then see if the agent acts differently in any of them. If the agent does act differently in some then we can say that PAP is met. Given what has been said above about eternalism and indeterminism I cannot see why eternalism would rule it out that there are possible worlds in which I act differently. After all, eternalism is not a modal thesis.19 Perhaps the worry instead isn’t PAP but what I’ll call fixity, with this being the concern that because eternalism implies ENTITY EVERYWHENISM how I act in my future is fixed and in virtue of this I cannot be free. One might therefore think that in order to be free fixity must be rejected (Kane, 2011, 105). My reply here is that fixity doesn’t seem to me to rule out free action. Before seeing why let me note that fixity differs importantly from PAP, for fixity is concerned with my actual future being fixed. By contrast PAP allows that what I do in the actual world is fixed, even though I still have access to some possible worlds besides my own, and as such it is possible that I acted differently. Let me first suggest then that fixity doesn’t seem to rule out libertarian freedom, since something in my future can be fixed, and yet I suggest a libertarian should still claim that it is free. To see this note that the special theory of relativity (STR) implies that I can be in a reference frame that allows me to see how Eran freely acts earlier than you do in your reference frame. Now consider a presentist who to overcome worries about STR implying that there is no absolute privileged present, holds that presentness is frame dependent (Hawley, 2009, 510; Baron & Miller, 2019, 108–110; Leftow, 1991, 232–234; 2018).20 Having assumed this theory, think 16
This is what O’Connor and Franklin call the categorical analysis of PAP. Another way PAP is sometimes put is in terms of agents having the power to act otherwise than they in fact do. Since I take it that powers are inherently modal, with reductive analyses failing, this will also make PAP inherently modal. 18 I understand ‘access’ talk, the way van Inwagen does, as providing ‘a way of organizing our talk about unexercised abilities by reference to unrealized possibilities: an unexercised ability is treated as an ability to realize some unrealized possibility.’ (1983, 87). 19 I have spoken here, and will speak elsewhere in this paper, of possible worlds. I take it that whilst modal realism might be seen as a natural analogue to eternalism, eternalism by no means requires it. In fact I see no reason why eternalism would require any specific view on the metaphysical status of possible worlds. Given this, and since I want eternalism and libertarianism to be compatible with as many views as possible, I take no particular view on the ontology of possible worlds and therefore one can understand them as they wish. 20 Hinchliff (2000) also seems to give some views of presentisms relationship to relativity which would also allow me to tell a similar story. 17
88
B. Page
about a world in which there are only two frames of reference, one which Eran and I share, Rf1, and a different one you are in Rf2. Consider what we should say about a case where Eran freely does X as opposed to Y. In this story, given Rf1, which is the reference frame Eran and I share, we can say that relative to Rf1, I see Eran make his free choice before you do in Rf2, as Rf1 and Rf2 are not simultaneous. If we relativise actuality to reference frames, this will mean that in Rf1 it is actual that Eran does X, even though Eran doing X is not currently actual in Rf2. Nevertheless, Eran freely doing X will become actual in Rf2, and it is fixed that he will do so because of what is actual in Rf1. We could also add that Eran’s future was open in Rf1, despite his future being fixed, and therefore closed, in Rf2. This being the case, it seems strange to say that whilst Eran acted freely in Rf1 he didn’t act freely in Rf2. After all, it is the one and the same Eran performing the same action in both reference frames. Again, the story allows that there are other possible worlds that Eran could access in Rf2, even though if he had accessed them he would have acted differently in Rf1. But nothing determines that Eran acts as he does other than himself in both frames of reference. Given that fixity by itself doesn’t seem to rule out free action, I suggest that if one can still affirm PAP and incompatibilism then even given fixity, a libertarian should think of an agent as libertarianly free.21 Let me try to show that one can through a thought experiment.22 Suppose that God exists and is timeless, that is, He exists outside of time, and so all worldly time-slices are timelessly laid out all at once before God.23 Given this God knows what occurs in our future, as well as what has occurred in our past and is occurring in our present, despite none of this being past or future to Him. As such, we can say that what occurs in our future is fixed.24 How then does God know a person’s, such as Tahira’s, future? On this story, it’s because God sees what Tahira does in her future.25 Tahira’s acting at t500, which is future to her, is present to God. God knows what Tahira does at t500 because He sees her do it. God’s knowledge about what Tahira does at t500 depends upon what she does in t500. How Tahira acts at t500, which is future to her, is therefore fixed. But it still seems to me to make 21
I would also add sourcehood which I will turn to shortly. This is also related to the problem of logical fatalism, which I don’t intend to contribute to here. The argument I’m responding to is sometimes purposefully distinguished from fatalism and as such is taken to be distinct (Koperski, 2015, 113–116). Nevertheless, let me say that some of the moves I make here, also speak against fatalism (Merricks, 2009, 2011) and that some have argued that eternalism grants additional resources for responding to fatalism (Finch & Rea, 2008). 23 Some might think that this requires eternalism, but it doesn’t and is compatible with presentism. See Leftow (2018, 1991, esp. 230–245). 24 Anselm, for instance seems to say just this in his De Concordia (The Compatibility of God’s Foreknowledge, Predestination, and Grace with Human Freedom), whilst also thinking that our actions are still free (Davies & Evans, 1998, 435–474). 25 This relation might be causal, with Tahira’s acts having a causal effect on what God believes. Some theists might worry about this. However, my purpose here isn’t to provide an orthodox theism but rather just to show that a closed future is compatible with free action. Nevertheless, a suggestion which may make what I say here non-causal is found in Merricks (2009, 41, 54–55; 2011, 574–579, 584, n.11) and Leftow (1991, 243–244, 255–265). 22
5 Libertarian Freedom in an Eternalist World?
89
sense to say that Tahira could have acted differently at t500, even though it’s true that she doesn’t. That is, there are different possible worlds Tahira could access where she does something different at t500. Had Tahira acted differently God would have known that timelessly, but again this is because God would have seen Tahira acting in a different way. It seems to me then, that despite fixity we can still think Tahira is free. Now return to the eternalist who is also a libertarian. They think that acts which are in your future are free even though they are fixed. What fixes them? You do. You determine what you do in each time-slice. Could you have acted differently? Sure. That is whilst you in t500 do Z, you could have done Y. Nothing makes you do Z rather than Y. But you don’t do Y in t500, you freely do Z. Hence what you freely do at t500 is fixed, due to what you do in this time-slice. Whilst the future is therefore closed, the eternalist can still embrace PAP, and so intuitively it seems to me that freedom is maintained. One of the key point’s here is a form of what Merricks calls Origen’s insight (2009, 52; 2011, 583–584), which for our purposes would say that what occurs in the eternalist block depends on what you do. It is your choice that determines what takes place in the slices of the block that you are in. The eternalist block could have been different, since there is nothing that determines what you do at each time-slice. Hence the block is fixed in the way it is only because at every time-slice you freely chose to do X rather than something else.
5.3
Causation and Change
Here’s a different type of worry that one might have about eternalism and libertarianism, namely that libertarianism requires a view of causation which eternalism cannot allow for. Perhaps the thought is that libertarianism requires that an agent brings about in a strong sense what occurs, which is at odds with eternalism.26 What should one say about this? Once again it will be important to distinguish between different types of libertarianism. We saw previously that there was a distinction between leeway and sourcehood libertarianism, but there is also a distinction between non-causal and causal libertarianism. This is important because non-causal accounts ‘require neither that a free action have any internal causal structure nor that it be caused by anything at all.’ (Clarke, 2003, 17) I suggest then that if there are any viable non-causal libertarian theories then these theories will bypass this type of objection. Nevertheless, many libertarians are causal libertarians, so how would they respond to this challenge? Causal libertarianism typically comes in two varieties. Event-causal and agentcausal libertarianism. One distinction here between the two views concerns the type of relata that is involved when an agent freely acts. Event-causalists say the cause is
26
Le Poidevin (2013, 540–542) also provides a response to this worry.
90
B. Page
an event,27 whilst agent-causalists say the cause is an agent. Let’s first think about one option which would allow event-causalism to be compatible with eternalism. We could adopt a reductive view of event causation, such as a counterfactual analysis of causation. Since this type of causation seems possible in an eternalist world,28 then so too will be event-causal libertarianism. Furthermore, at least one event-causalist, Ekstrom (1998), has suggested that incompatibilists have extra reason to endorse a counterfactual analysis of causation, since it allows them to block a powerful reply that compatibilists can give to the consequence argument. As such it seems event-causal accounts of libertarianism are prima facie possible given an eternalist world. Turn now to agent-causalist accounts. These are typically taken to adopt non-reductive accounts of causation (Clarke, 2003, 187), and therefore the move I made previously is ruled out for them.29 As such they sometimes claim to embrace a stronger view of ‘bringing about’ than reductive theories do. Is this type of causation compatible with eternalism? Here’s a quick appeal to authority to give one prima facie reason that they are. Skow defends an eternalist view of the world (2015) and also that the most basic use of cause relates ‘things’ rather than events, resulting in what he calls agent causation (2018, 137–179). I suppose that Skow wouldn’t defend this view of causation if he thought it incompatible with his view of time. Hence, if Skow is right, agent causation is compatible with eternalism. One can argue against this by showing why Skow is wrong, or by demonstrating why his theory is not enough for agent-causal libertarianism. Yet until one has done so we have prima facie reason to think they are compatible.30 Let’s turn to a more substantial response on behalf of the agent-causalist. Many who opt for this view of freedom are what I call ‘power’ theorists.31 They hold that agents are related to powers in some way, and that powers are irreducibly causal entities, which when brought together with other powers bring about their manifestations, what we can think of as effects. Whilst there are many different accounts of the metaphysics of powers, many who work on power ontologies think that powers manifestations are simultaneous with their causes. For example, Marmodoro writes, that all there is to ‘causal ‘interaction’ [between powers] is their mutual and simultaneous manifestation (e.g. heating and being heated).’ (2017, 70; Martin, 2007, 48–51; Heil 2012, 118–126; Mumford & Anjum, 2011, 106–129) Suppose
27
See Clarke (2003, 34). Or we could perhaps say is reducible to events (O’Connor & Franklin, 2020). 28 As it’s been argued this type of causation is compatible with timeless worlds (Baron and Miller, 2014, 2015; Tallant, 2019) it would seem it ought to be compatible in eternalist worlds. 29 This is not to say that event-causal accounts must be reductive. If they are not, then another response will need to be given. 30 I can also produce an argument from authority for the claim libertarianism is compatible with eternalism. Van Inwagen is a contemporary defender of libertarianism, and yet he also doesn’t seem to be at all a fan of dynamic theories of passage (2015, ch.4). Yet never, as far as I’m aware, has he suggested that the negation of dynamic theories rules out libertarianism. 31 I do not mean to imply here that those who adopt a powers view must be agent-causalists.
5 Libertarian Freedom in an Eternalist World?
91
this is right. Then an eternalist can adopt a powers view of causation, since causal interaction, can occur in individual time-slices.32 It is true that the manifestation of the powers might be spread out over a number of time-slices, but in each of those time slices the manifestation will be simultaneous with the continual causal interaction. It seems to me an agent-causalist about free will could say much the same, that is that one’s willing X is simultaneous with X occurring. Obviously if simultaneous causation is not possible, then neither is this reply, and I note here that some power theorists think it is not possible (Williams, 2019, 169–172), although I remain unconvinced. Nevertheless, even if it is impossible, I suspect there will be other avenues open for the power theorist to make eternalism compatible with a powers view of causation, such as those explored in Roselli (2022); Koons (2022); Giannini (2021), and with that in hand I suspect they will be able to translate this work into the free will discussion.33 For now, I’m fairly confident this objection can be overcome. Let me finish with a final concern that one might have with libertarianism and eternalism, namely a problem to do with change. The idea here stems from the thought that libertarianism requires that an agent exists through the different stages of their free actions and eternalism cannot allow for this. Lockwood might be thought to raise this concern when he implies that eternalism ‘rules out any conception of free will that pictures human agents, through their choices, as selectively conferring actuality on what are initially only potentialities.’ (2005, 69)34 Perhaps the eternalist is forced to say that an agent doesn’t really exist through their actions,35 if we suppose that these actions take place across multiple-time slices, but rather they are merely distinct temporal parts of an agent existing at each time-slice which jointly together constitute the agent.36 What is therefore being asked for is endurance, rather than perdurance, such that an agent is wholly present at each time they exist. Suppose then, that this is right, and that libertarianism does require endurance rather than perdurance,37 is this a problem for eternalist libertarianism? I think not, since it seems to me that eternalism is compatible with endurantism, with this being 32
I assume here that STR is compatible with simultaneous causation, but for some pushback on this see (Hansson Wahlberg, 2017). 33 At present there has been little published on how powers relate to time. Of what has been published, Vetter (2015, 290–299) thinks a powers theorist should prefer eternalism over presentism, whilst Backmann (2019) thinks the opposite. 34 Lockwood’s comment is also related to a powers theory of causation noted above and will be addressed further below. It is therefore also worth pointing out that many power theorists will deny that powers turn from being potentially X to being actually X. Instead they think of a power manifesting as jumping from being actually X to being actually Y (Mumford & Anjum, 2011; Bird, 2007). Whether they should do so, however, is another matter, since it might be that this theory denies change since it gives rise to the notorious always packing never travelling worry (Armstrong, 1997, 80). 35 Or in other words they do not ‘persist’. 36 This is ‘worm theory’ or standard perdurance. One could also adjust what I say here for ‘stagetheory’ if they wished. For more on this distinction see Hawley (2020). 37 This isn’t to say that I think this is in fact the case.
92
B. Page
something that has been argued for by numerous metaphysicians (Costa, 2020; Hansson Wahlberg, 2009; Sider, 2001, 68–70; Haslanger, 2003, 320–326; Rea, 1998; Sattig, 2006, 47–65). Yet if that’s right then eternalism does allow that an agent can be wholly present, and therefore really, exist through their free actions. Suppose then that this time we adopt another view of powers, in which the very same power can be in a state of potentiality when it isn’t manifesting, and a state of actuality when it is (Marmodoro, 2017).38 We can then say that an agent can be wholly present in one time-slice when willing Y is in potentiality, and also wholly present in the next time-slice when willing Y is in actuality. The agent themselves, given what I’ve said above, is the one that determines that Y will become actual in the second time-slice, since the contents of the time-slices depends on what they do, not the other way around.39 As such, this worry can also be overcome.40
5.4
Conclusion
Given this discussion, it seems to me that there are some plausible reasons to think that libertarian views of freedom are compatible with eternalism, or at the very least some variants of libertarianism are. As such, I take it that my students, and those like them who worry that there is such an incompatibility, are wrong to do so.
References Armstrong, D. M. (1997). A world of states of affairs. Cambridge University Press. Backmann, M. (2019). No time for powers. Inquiry, 62, 979–1007. Baron, S., & Miller, K. (2014). Causation in a timeless world. Synthese, 191, 2867–2886. Baron, S., & Miller, K. (2015). Causation “Sans” Time. American Philosophical Quarterly, 52, 27–40. Baron, S., & Miller, K. (2019). An introduction to the philosophy of time. Polity Press. Bird, A. (2007). Nature’s metaphysics. Oxford University Press. Brennan, J. (2007). Free will in the block universe. Philosophia, 35, 207–217. Buckareff, A. A. (2019). Time, leeway, and the Laws of nature. Metaphysica, 20, 51–71. Clarke, R. (2003). Libertarian accounts of free will. Oxford University Press. Costa, D. (2020). Mereological endurantism defined. Inquiry, 1–12. https://www.tandfonline.com/ doi/abs/10.1080/0020174X.2020.1851759?journalCode¼sinq20 Davies, B., & Evans, G. R. (1998). Anselm of Canterbury: The major works. Oxford University Press.
38 This view of powers isn’t prey to the always-packing-never-travelling worry (Marmodoro, 2017, 60–62). 39 Of course, the whole time-slice doesn’t depend on what they do, since it contains much besides the agent. But I take it you get the picture. 40 Obviously if libertarianism does not require endurantism and is compatible with other theories of persistence, then this objection will be overcome more easily.
5 Libertarian Freedom in an Eternalist World?
93
Ekstrom, L. W. (1998). Causation, and the consequence argument. Synthese, 115, 333–354. Finch, A., & Rea, M. (2008). Presentism and Ockham’s way out. Oxford Studies in Philosophy of Religion, 1, 1–17. Fischer, J. M. (1999). Recent work on moral responsibility. Ethics, 110, 93–139. Giannini, G. (2021). Powers, processes, and time. Erkenntnis, 1–25. https://link.springer.com/ article/10.1007/s10670-020-00327-z Hansson Wahlberg, T. (2009). Endurance per se in B-time. Metaphysica, 10, 175–183. Hansson Wahlberg, T. (2017). Meso-level objects, powers, and simultaneous causation. Metaphysica, 18, 107–125. Haslanger, S. (2003). Persistence through time. In M. J. Loux & D. Zimmerman (Eds.), The Oxford handbook of metaphysics. Oxford University Press. Hawley, K. (2009). Metaphysics and relativity. In R. Le Poidevin, P. Simons, A. McGonigal, & R. P. Cameron (Eds.), The Routledge companion to metaphysics. Routedge. Hawley, K. (2020). Temporal parts. In E. N. Zalta (Ed.), The Stanford Encylopedia of philosophy. URL: https://plato.stanford.edu/archives/sum2020/entries/temporal-parts/ Heil, J. (2012). The universe as we find it. Oxford University Press. Hinchliff, M. (2000). A defense of presentism in a relativistic setting. Philosophy of Science, 67, S575–S586. Hoefer, C. (2002). Freedom from the inside out. Royal Institute of Philosophy Supplement, 50, 201–222. Hunt, D. (2000). Moral responsibility and unavoidable action. Philosophical Studies, 97, 195–227. Kane, R. (1996). The significance of free will. Oxford University Press. Kane, R. (2005). A contemporary introduction to free will. Oxford University Press. Kane, R. (2011). The causal closure of physics and free will. In R. Kane (Ed.), The Oxford handbook of free will (2nd ed.). Oxford University Press. Koons, R. C. (2022). A dynamic B theory of time. In Powers, time and free will. Springer. (this volume). Koperski, J. (2015). The physics of theism. Wiley-Blackwell. Le Poidevin, R. (2013). Time and freedom. In H. Dyke & A. Bardon (Eds.), A companion to the philosophy of time. Wiley-Blackwell. Leftow, B. (1991). Time and eternity. Cornell University Press. Leftow, B. (2018). Presentism, Atemporality, and Time’s way. Faith and Philosophy, 35, 173–194. Levin, M. (2007). Compatibilism and special relativity. Journal of Philosophy, 104, 433–463. Lockwood, M. (2005). The labyrinth of time. Oxford University Press. Marmodoro, A. (2017). Aristotelian powers at work: Reciprocity without symmetry in causation. In J. D. Jacobs (Ed.), Causal powers. Oxford University Pres. Martin, C. B. (2007). The mind in nature. Oxford University Press. Maudlin, T. (2007). The metaphysics within physics. Oxford University Press. Merricks, T. (2009). Truth and freedom. Philosophical Review, 118, 29–57. Merricks, T. (2011). Foreknowledge and freedom. Philosophical Review, 120, 567–586. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford University Press. O’Connor, T., & Franklin, C. (2020). Free will. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. URL: https://plato.stanford.edu/archives/fall2020/entries/freewill/ Oaklander, L. N. (1995). Part II: Time and identity. In Q. Smith & L. N. Oaklander (Eds.), Time, change and freedom. Routledge. Oaklander, L. N. (1998). Freedom and the new theory of time. In R. Le Poidevin (Ed.), Questions of time and tense. Clarendon Press. Pereboom, D. (2001). Living without free will. Cambridge University Press. Petkov, V. (2005). Relativity and the nature of Spacetime. Springer. Putnam, H. (1967). Time and physical geometry. The Journal of Philosophy, 64, 240–247. Rea, M. C. (1998). Temporal parts unmotivated. Philosophical Review, 107, 225–260. Rietdijk, C. W. (1966). A rigorous proof of determinism derived from the special theory of relativity. Philosophy of Science, 33, 341–344.
94
B. Page
Roselli, A. (2022). Dispositional essentialism in the eternalist block. In Powers, time and free will. Springer. (this volume). Rychter, P. (2015). Classical compatibilism and temporal ontology. In A. Buckareff, C. Moya, & S. Rosell (Eds.), Agency, freedom, and responsibility. Palgrave Macmillan. Sattig, T. (2006). The language and reality of time. Oxford University Press. Shabo, S. (2010). Uncompromising source Incompatibilism. Philosophy and Phenomenological Research, 80, 349–383. Sider, T. (2001). Four-dimensionalism. Oxford University Press. Skow, B. (2015). Objective becoming. Oxford University Press. Skow, B. (2018). Causation, explanation, and the metaphysics of aspect. Oxford University Press. Stump, E. (1999). Alternative possibilities and moral responsibility: The flicker of freedom. The Journal of Ethics, 3, 299–324. Stump, E. (2003). Moral responsibility without alternative possibilities. In M. McKenna & D. Widerker (Eds.), Moral responsibility and alternative possibilities: Essays on the importance of alternative possibilities. Ashgate. Tallant, J. (2019). Causation in a timeless world? Inquiry, 62, 300–316. Timpe, K. (2013). Free will (2nd ed.). Bloomsbury. Timpe, K. (2017). Leeway vs Sourcehood conceptions of free will. In K. Timpe, M. Griffith, & N. Levy (Eds.), The Routledge companion to free will. Routledge. Tognazzi, N. (2011). Understanding source Incompatibilism. The Modern Schoolman, 88, 73–88. Van Inwagen, P. (1983). An essay on free will. Clarendon Press. Van Inwagen, P. (2015). Metaphysics (4th ed.). Westview Press. Vetter, B. (2015). Potentiality. Oxford University Press. Williams, N. E. (2019). The powers metaphysic. Oxford University Press. Zagzebski, L. (2000). Does libertarian freedom require alternative possibilities? Philosophical Perspectives, 14, 231–224.
Chapter 6
The Temporal Structure of Agency Janice Chik
Abstract Aristotle’s account of action as kinesis, an intrinsically temporal process, posits that action is a kind of change over time, with parts or sub-actions that occupy distinct subordinate phases of a movement. This general idea has produced conflicting accounts of agency, however. Proponents of the causal theory of action ((CTA), causalism, or ‘the standard account’) have taken it as evidence of a reductionist or decompositional programme that supports the concept of basic action (Davidson, 1980; Coope, 2007). In contrast, recent critiques of causalism have relied explicitly on Aristotle’s idea of kinesis to defend an anti-decompositional approach, arguing that the sub-phases of an action must be known to its agent for such proceedings to qualify as agential (Thompson, Life and action: Elementary structures of practice and practical thought, Harvard University Press, 2008; Lavin, Noûs 00(0):1–32, 2012, 2016). This essay assesses these challenges to (CTA) as representing variations on the problem of disappearing or alienated agency, and argues that (CTA)‘s strongest critics fail to show that basic action indeed lacks temporal structure. Instead, we propose a revision in the conventional account of action as either complex or basic, as either taking much time or no time at all. On our account, the analysis of action as possessing temporal sub-phases is reconcilable with an anti-decompositional approach: the latter’s success depends on a proper assessment of the temporal structure of agency, not only in the so-called ‘complex’ actions which have an obvious means-end rational character, but also in the metaphysically simple and spontaneous movements that occur in the sub-phases of performance, which some identify as ‘basic’. Keywords Aristotelian kinesis · Agency · Temporal structure
J. Chik (*) Ave Maria University, Ave Maria, FL, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_6
95
96
J. Chik In their parts and during the time they occupy, all movements are incomplete, and are different in kind from the whole movement and from each other.1 (Aristotle, Nic. Eth. 1174a 22–24)
6.1
Introduction: Agency and Temporality
Does action have a uniform temporal structure? Must action always occupy time? Aristotle’s remark in Nicomachean Ethics suggests that action is always a kind of change over time, with parts or sub-actions that occupy distinct sub-phases of ‘the whole movement’ (kinēsis) itself. Analysis of action as temporally structured in this way is double-edged, however. The possibility of such analysis has been interpreted by proponents of the causal theory of action (CTA, causalism, or ‘the standard account’) as evidence of a reductionist or decompositional programme, and in particular, as support for the concept of basic action.2 At the same time, some of the most compelling critiques of the causal theory of action have relied explicitly on Aristotle’s idea of kinēsis as entailing an anti-decompositional approach (Lavin, 2012). This essay proposes a compromise: to reconcile a conception of action analysable into sub-actions and possessing temporal sub-phases, with a commitment to an anti-decompositional approach. The latter’s success depends on a proper assessment of the temporal structure of agency, not only in the so-called ‘complex actions’ which have an obvious means-end rational character, but also in the metaphysically simple and spontaneous movements that occur in the sub-phases of human action, which some identify as ‘basic action’. Although there are exceptions, much discourse on human action proceeds either as if the temporal structure of agency is simply given, that it can be assumed in the background of action and activity – or that its temporality is simply irrelevant to the real analysis concerning the concept of action. This essay suggests otherwise: unless action is recognized in the first place as a kind of change over time,3 the apparent 1
Translation by W.D. Ross, revised by J.O. Urmson. Davidson, for instance, seems to regard causalism as an extension of Aristotle’s approach to agency, writing: ‘[T]he best argument for a scheme like Aristotle’s is that it alone promises to give an account of the ‘mysterious connection’ between reasons and actions’ (Davidson, 1980: 11). On Davidson’s interpretation, the causal relation promised by Aristotle’s account specifically involves the concept of a reason which, as the cause of action, can be broken down into compound mental states, a belief and a desire (or, in lieu of ‘desire’, what Davidson calls a ‘pro-attitude’). Other causalist accounts, such as Smith (1987, 2004, 2012), approach the matter from a Humean perspective. 3 The definition of action as change follows Coope (2007): the exercise of a certain kind of causal power of an agent, also understood as the partial fulfillment of that capacity (in Aristotelian terms, ‘the incomplete actuality of the moveable’). The present essay does not address the specific concerns Coope raises for Aristotle’s account, which she clarifies cannot be a conception of action that causes a change or a movement: instead, it is the change wherein the movements of one’s body are identified with the action itself. Action is the causing of the end state of affairs – it causes the end state to be – rather than of the changes that result in the end state. 2
6 The Temporal Structure of Agency
97
extremes of agential simplicity and complex activity cannot be reconciled as occupying positions within the same metaphysical domain of human agency. The challenges facing this proposal, that action has a distinctively agential temporal structure, are particularly burdensome for the causal theory of action (CTA). Let us take Aguilar and Buckareff’s definition as standard: (CTA) Any behavioral event A of an agent S is an action if and only if S’s A-ing is caused in the right way and causally explained by some appropriate nonactional mental item(s) that mediate or constitute S’s reasons for A-ing. (Aguilar & Buckareff, 2010: 1)
This definition of (CTA) appears deliberately broad. It allows for a variety of different conceptions of the mental causes of S’s A-ing, while following the Davidsonian principle that such mental items always constitute S’s reasons for A-ing. The broadness of (CTA), however, leaves open several critical questions. Foremost among them is the issue of what being ‘caused in the right way’ requires, and in particular, whether it requires more than merely the ‘appropriate nonactional mental items’ that constitute the agent’s reasons for acting.4 Another problem that arises from the definition of (CTA) is that its descriptions of a ‘behavioral event A of an agent S’ and ‘S’s A-ing’ are used as if interchangeable. Such usage, however, overlooks an important difference between them: a ‘behavioral event A of an agent S’ appears to entail that the event A belongs to an agent S, whereas ‘S’s A-ing’ places emphasis on an event of which the agent S is a logical ‘constituent’ or element.5 These formulations are, in fact, incompatible with one another. The former description suggests that events belong to agents, with the potential implication that agents are the primary causes of action (in which events may be a part or aspect), whereas Davidsonian causalism assumes that the relevant causal relata are events, of which agents are merely logical constituents and therefore secondary in explanatory relevance. Where action is described in terms of the agent, to whom behavioral events belong or are attributed, there may be hope that a sense of the agent as primary cause of her own movements may be retained. In contrast, action descriptions of the form ‘S’s A-ing’, in which any mention of S’s involvement might be easily left out so as to produce abstract descriptions such as ‘walking’, ‘cooking’, and ‘writing’, potentially imply that the agent’s involvement in such entities is explanatorily secondary to the status of events, states of affairs, or facts as the primary causal relata of action. Should one insist that one form of action description, ‘a behavioral event of an agent S’, is interchangeable with or reduces to the other, ‘S’s A-ing’, it cannot be assumed that such interchangeability or reductive 4
Relatedly, (CTA) raises the fundamental question of how a mental item can cause a physical event. This latter problem is understood by philosophers of mind as the problem of ‘causal exclusion’, not explored in this paper. 5 Such use of the term ‘constituent’ expresses a logical relation between agent and event, comparable to the way that ‘the vase is on the table’ may be said to have the logical constituents of ‘the vase’, the ‘being on’ relation, and ‘the table’. Even if this is a loose way of describing the relation between agent and events (or states of affairs, or facts), but it should be clear that the constituent relation employed here bears no significant similarities (except analogously) to metaphysical parthood, e.g., as when the legs of the table are understood as its constituent parts, among others.
98
J. Chik
identity holds likewise of the respective conceptions of the causal relata of action, which may be derived from these distinct forms of action explanation: i.e., (i) the agent as primary cause of her own movements, versus (ii) the agent as a logical part or mere explanatory addendum to events (or states of affairs, or facts), where these, rather than the agent, hold primary causal status. Only the first explanatory form, which posits the agent as the primary cause of what happens, is capable of accommodating an account of action as change over time, as having a temporal structure that is distinctively agential. This takes shape as a major critique of (CTA), as the next part shows.
6.2
The Problem of Alienated or Disappearing Agents
Critics of (CTA) frequently take aim at what is called ‘the problem of disappearing or alienated agents’: the concept of an agent, it is alleged, is diminished or goes missing altogether from causal explanations of action. As seen above, the worry is that the formulation ‘S’s A-ing’ makes too easily available certain abstract descriptions of action (e.g. ‘walking’), in which the causal relevance of an agent as principle source and performer of the action, is rendered otiose. The threat of disappearing agents can be construed as an extension of the problem of deviant causal chains, in which the sufficiency of the causalist’s requirements for action is questioned in light of exceptional hypothetical scenarios, such as Donald Davidson’s hapless rock climber: A climber might want to rid himself of the weight and danger of holding another man on a rope, and he might know that by loosening his hold on the rope he could rid himself of the weight and danger. This belief and want might so unnerve him as to cause him to loosen his hold, and yet it might be the case that he never chose to loosen his hold, nor did he do it intentionally. (Davidson, 1963: 79)
There are other creative iterations of deviant causal chains, but the main thrust of the problem is the same: an agent’s intended performance is interrupted or sabotaged in some way, so that there is a bodily occurrence that would appear to be the intended action, but is in fact not a voluntary performance. Although a causalist himself, Davidson speculated that the problem could not be solved satisfactorily by the causal theory. Anti-causalists allege that the problem not only reveals the insufficiency of the causal criterion of a belief-desire compound: deviant causal chains suggest the possibility of accounting for action wholly in terms of event causation without including an agent in the story at all.
6 The Temporal Structure of Agency
6.2.1
99
Velleman
The original problem of disappearing agents can be credited to David Velleman (1992), who criticized the causal theory of action for failing ‘to include an agent or, more precisely, fails to cast the agent in his proper role. In this story, reasons cause an intention and an intention causes bodily movement, but nobody—that is, no person—does anything’ (Velleman, 1992: 189). Deviant causal chains further accentuate the worry that we might fail to recognize ‘an agent at work amid the working of the mind’ (Velleman, 1992: 131). Velleman also clarifies that his objection to the causal story ‘is not that it mentions mental occurrences in the agent instead of the agent himself [but] that the occurrences it mentions in the agent are no more than occurrences in him, because their involvement in an action does not add up to the agent’s being involved’ (Velleman, 1992: 463). It appears, however, that the causalist has an obvious reply to these objections. Being a reductive account, the causal theory’s self-described aim is to cite occurrences below the level of the agent in its explanation of action. On its own terms, it would be recursive to include the agent herself in such explanation. But the fact that sub-agential occurrences are cited, without mention of the agent in whom they occur, does not entail a disappearing agent: why indeed should it? On the causalist’s account, the sub-agential occurrences are no less occurrences of the agent, a sub-agential exemplification of her ability to perform actions. As Michael Bratman argues, ‘Explanations of occurrences normally appeal to other occurrences—other events, states, processes, and the like. That is a reason why appeals to the agent as cause seem problematic when offered as explanations’ (Bratman, 2000: 39).
6.2.2
Hornsby
Jennifer Hornsby’s complaint against the causalist is similar to Velleman’s. She argues: ‘The idea of an agent at work goes entirely missing from Davidson’s story of human agency. Davidson spoke of causality as “central to the concept of agency”, but he said that “it is ordinary causality between events that is relevant, and it concerns the effects. . .of actions”’ (Hornsby, 2011: 108). Like Velleman, Hornsby objects to the idea that the causation of action involves a relation between events (or states) rather than agents. For her, however, the problem of the disappearing agent is framed as an issue in which the verbs used in action description reveal the implausibility of the causal theory. As counterexamples to the causalist convention of positing a ‘behavioral event A of an agent S’, she argues that transitive verbs such as ‘carry’, ‘push’, and ‘squash’, are of the kind in which their passive form (‘being carried’, ‘being pushed’, ‘being squashed’) ‘cannot be pried apart’ from the active
100
J. Chik
form that includes the agent as subject, such as S’s ‘carrying’ or ‘pushing’ an object o. By stating that these forms ‘cannot be pried apart’, Hornsby suggests that there exists no possible temporal or spatial gap between agent and action in which causal intermediaries, such as events or states, can be inserted, such that ‘S pushes o’ cannot be rendered as ‘“the event of o’s being pushed” is caused by “the event of S’s having an intention (or belief and desire) to push o”’. Descriptions such as these, Hornsby argues, involve the use of events as ‘mediating’ between their constituents, S and o. She remarks: ‘We might say that in such cases the person does the thing—carry something, or, as it might be, push, or squash something—non-mediately’ (Hornsby, 2011: 108). Hornsby’s problem with the causal theory, it appears, is that the latter fails to countenance a schema in which agency is mediated by anything other than events or states. It may be that Hornsby’s objection is subject to the same causalist reply as was addressed to Velleman’s concern, but there is another interesting feature of her objection to consider. This is the fact that Hornsby chooses to center her critique on the notion of action without intermediaries. If it is true that the causalist requires the mediation of events or states for the causation of action, then there may indeed be a problem about whether the agent mediates them, and if so, how? If the agent does not mediate them, then how do they arise, and how then could it be said that the agent is truly ‘at work’? Performance that is non-mediated proves that Davidson’s account of the spatial extent of an action is wrongly conceived, argues Hornsby; however, ‘examples of different sorts are needed to show that Davidson’s treatment of actions as eventcauses also gets their temporal extents wrong’ (Hornsby, 2011: 121). Her treatment of non-mediation entailed that nothing should intervene between the object o affected by an agent and the acting agent herself. In addressing the individuation of action based on its temporality, Hornsby now offers for consideration the possibility that an agent may act without contact with an object o, and yet be said to be acting on o: ‘A person may start or stop acting on an object before or after she makes contact with it’ (Hornsby, 2011: 121). The coherence of this account apparently relies on the generation and perdurance of the agent’s intended end. In the case of pushing a button to print a long document, ‘the explanation of why the event of your printing the document should be ongoing when you are not patently active is that the printer remains somewhat under your control, and your intention to print the document survives long enough for you to be able to deal with paper jams or the like. Your action of printing comes to an end only when your intended end is reached and you have a printed document’ (Hornsby, 2011: 121–22). Such an example shows that ‘there can be activity in a matter which is not bodily activity’ (Hornsby, 2011: 122). If this is granted, then it cannot be the case that, as Davidson famously remarked, ‘we never do more than move our bodies’; or that basic action should be the sole basis of action’s temporal extent.
6 The Temporal Structure of Agency
101
Do not such examples showcase action as mediated, however? Is not the completion of printing an action of mine mediated by some causal work that the printer does? In the case of someone who closes a door by pushing it part of the way, Hornsby remarks: ‘We might well treat this as another example of non-mediate closing, thinking that, even though the door acquires a momentum of its own, the causal work involved in its closing is [the agent’s] work. But even someone who insists that the door itself does causal work, and who thinks that this shows that the bit of the door’s closing[intrans] which takes place after [his] hand has left the door is not [his] work, will not think, as Davidson would, that the closing[trans] caused the closing[intrans]’ (Hornsby, 2011: 121). Presumably, this is because the temporal extent of closing[trans] includes closing[intrans], in order to accommodate the undeniable sense in which any putative causal work done by the door itself is intrinsically related to the causal work done by the agent. Like the paradigm cases of non-mediated performances, the causal work of a door (or a printer) does not stand in relation to the action as an event that mediates the agent’s performance: rather, the ‘work’ itself (of the door or printer) is in some sense the object o of the agent’s action, a complex object with which the agent continues to have a kind of temporal, non-bodily contact. According to Hornsby, action as non-mediated is therefore compatible with Anscombe’s view that ‘both the spatial and temporal extents of actions fail to coincide with those of bodily movements’ (Hornsby, 2011: 124). Again, this view is defended on the basis that one’s intentional action need not always correspond precisely to some bodily activity, for what is intentional necessarily refers to some agential end: ‘In Anscombe’s treatment, when there is a compound action, the series of means the agent takes are united by reference to the agent’s end, not by summing the several, separately conceived components of the action (still less to summing such bodily movements as those components might involve)’ (Hornsby, 2011: 124). The temporal extent of action has limits set by agential intention at both ends: at the commencement of action, which need not coincide with any bodily stirrings, and at its conclusion, which likewise may outlive the cessation of bodily motion. There is much to qualify concerning this account, of course. First, it is not always the case that the temporal extent of action precedes bodily movement, or that it always outlasts bodily activity or its effects. An example of the first possibility is that of breathing, a continuous bodily movement of which anyone can become aware and intentionally control. An example of the second is countenanced by Hornsby: ‘It can be allowed that if you dropped dead immediately after you used a bit of your body to set the printer in motion, then “You are printing the document” would be false while the printer did its work. It would then be false because death extinguishes intentions as well as the makings of movements’ (Hornsby, 2011: 123). It seems crucial to Hornsby’s account that we clarify the concept of agential end, as unifying cause for the series of means taken by the agent over the temporal phase that demarcates an action. Bratman (2000) proposes a Lockean explanation of agency as temporally extended, with an emphasis on what Parfit calls ‘overlapping
102
J. Chik
strands of psychological connectedness’ – in particular, conventional Lockean categories involving ‘backward-looking memory’, but also contemporary additions of ‘forward-looking connections like that between a prior intention and its later execution, and continuities in desire and the like’ (Bratman, 2000: 43–44). Bratman prefers the term ‘continuities of belief or desire’ over ‘connections’, arguing that the latter is misleading. It could be that the execution of one’s intention is connected with a prior intention, in what Bratman calls ‘temporal cross-reference’, but there can also be mere continuity of a desire without such connectedness or temporal crossreference. On Bratman’s view, ‘one and the same agent—me—begins, develops, and completes temporally extended and coordinated activities and projects; my agency is, in this sense, temporally extended’ (Bratman, 2000: 43). The emphasis within any analysis concerning temporal extension is on agency rather than the agent, for he apparently rejects at the outset any notion of the agent as cause of what happens: ‘[I]t is difficult to know what it means to say that the agent, as distinct from relevant psychological events, processes, and states, plays such a basic role in the etiology and explanation of action. Second, and relatedly, in seeing the agent as a fundamentally separate and distinct element in the metaphysics of our action we seem to abandon the idea that our agency is as fully embedded in the event causal order as is the agency of purposive agents like dogs and cats’ (Bratman, 2000: 39). It is doubtful that Hornsby has in mind a Lockean account of agency as temporally extended. Certainly, Bratman’s account of psychological continuity (rather than ‘connection’) cannot be the only plausible explanation for agency as being extended in time. First, it isn’t clear what continuity without connectedness entails: without temporal cross-reference of some kind to be given by the agent, how is a desire to φ at t2 related to a desire to φ at t1 as agential phenomena that is temporally continuous, if the agent herself would reject any sort of temporal cross-reference between them? If action is to be understood as something temporal, i.e. as a distinctively agential kind of change across time, then it must be the sort of occurrence whose starting- and ending-points are transparently cross-referenceable to the agent herself, as tied inherently to her agential end, or what unifies the temporal sub-phases of action, the means undertaken. Otherwise, what grounds agential continuity across time? Second, it is unclear why a conception of agent as cause should entail that (i) the agent is ‘a fundamentally separate and distinct element in the metaphysics of. . .action’, and (ii) that we thereby ‘abandon the idea that our agency is as fully embedded in the event causal order as is the agency of purposive agents like dogs and cats’. Neither worry is justified: as we see from Hornsby’s account of unmediated action, the agent is placed squarely at the center of what happens, fundamentally inseparable from her action and its immediate object. And there is no reason whatsoever to think that our agency is therefore so dissimilar from that of non-human animals so as to be causally unnatural. If we assume Hornsby’s account, then all animal agents – whether dogs or cats or human beings – are fully embedded in the natural causal order as the principal causes of their own doings, without the mediation of events, states, or facts.
6 The Temporal Structure of Agency
103
Admittedly, one drawback to Hornsby’s account is the potential confusion over the concept of mediation as she employs it. She appears to use the term ‘mediation’ simply wherever events are taken to be the causal relata, so that an ‘unmediated’ action expresses a concept of agent (rather than event) as cause.6 But it may be argued that events and states are not typically conceived by causalists as playing a mediating role, in any conventional sense of mediating: rather, these are cited as the reductive components of a complex that needs to be analysed, such as ‘S pushes o’. Thus, the questions raised by Hornsby’s account7 may be inconsequential if the causalist does not require a concept of action as mediated at all.8
6.2.3
Lavin
Douglas Lavin (2012), another critic of the causal theory, avoids the latter difficulty for Hornsby’s account by arguing that action must be understood as ‘mediated’, if agents are not to be forced into a position of alienation from their actions. Lavin’s notion of ‘mediation’ is clearly different from Hornsby’s: he argues that the concept of ‘basic action’, specifically an unmediated event that is done ‘just like that’, is ‘vital to the intelligibility of the causal theory of action’ (Lavin, 2012: 2–3). Danto’s original concept sought to reflect the structure of a basic idea: he described basic actions as ‘not compounded out of anything more elementary than themselves, but [are] instead the ultimately simple elements out of which [actions are] compounded’ (Danto, 1965: 147).9 An implication of Danto’s ‘practical atomism’ is that basic action is therefore ‘barren of means-end structure’ (Lavin, 2012: 2). What this means, Lavin explains, is that the performance of basic actions is never mediated ‘by any such thought—thought about how to bridge the gap between here, where one has not yet done what one aims to do, and there, where one has brought things to completion. With basic action, thought about how to realize the end in question gives out—one acts immediately, directly or “just like that”’ (Lavin, 2012: 4).
6 On the other hand, perhaps Hornsby’s concept of ‘mediation’ is meant to play a role in our understanding of actions: i.e., that we can understand ‘S pushes or squashes o’ without our understanding being mediated by the assumption that the causal relata involved in the description of S and o are mental events and physical movements. It remains unclear which interpretation is implied by her concept. Either way, Hornsby’s remarks amount to little more than a rejection of (CTA). It may be that she simply wishes to remind us of the naturalness of thinking in agent causation terms, in which ‘the person does the thing’, rather than a causal relation between events. 7 Further related texts offering depth to this question are Hornsby (2007, 2004a, 2004b) and Hornsby & Stanley (2005). 8 That is not to say, however, that Hornsby’s objection against event or state causation assumed by the causalist is of practical irrelevance: merely that her usage of the concept of ‘mediation’ is arguably too loose to correspond precisely to that objection. 9 See also Danto (1963, 1973).
104
J. Chik
Causalism is a ‘decompositional theory’: it reduces any complex action to something basic – ‘brief or even quasi-instantaneous actions in which there is no discernible intentional structure to speak of’ – and assumes that such actions are completed and in the temporal past (Lavin, 2016: 623). In the first place, the object of attention is typically a completed action (Jones buttered the toast, Shem kicked Shaun, the officer pushed the button that fired the missile that sank the Bismark), one whose real constitution from any parts belonging to it has already occurred and been settled once and for all, one that can be unproblematically referred to by a singular term denoting a concrete particular, a wholly determinate historical event (Jones’s buttering the toast, The officer’s sinking the Bismark). Secondly, the central focus of theoretical attention is characteristically on brief or even quasi-instantaneous actions in which there is no discernible intentional structure to speak of: such things as arm-raisings, button-pushings, trigger-pullings. To be sure, decompositional theorists acknowledge the existence of complex, temporally extended actions (dancing a tango, directing a military campaign), but they characteristically seek to reduce such complex actions to coordinated sequences of ‘basic actions’ of which it is true that they are severally brief or quasi-instantaneous and lacking in discernible intentional structure.10
Lavin’s critique of causalism is distinctive in its focus on temporal structure as providing the basis for discerning what is or is not agential. An occurrence has intentional structure only if it has temporal structure; without the latter, basic action is not in fact agential, something done by the agent. And therefore, he concludes, ‘the entire orientation of the decompositional approach points us away from the standpoint of the agent, and toward a consideration of actions as definite, achieved realities: realities in the life of some agent, to be sure, but considered in such a way that the practical point of view of this agent toward the relevant event does not come into the foreground, and the question of what binds the parts or phrases of this event into a unity does not easily arise’ (Lavin, 2016: 624). In other words, the agent disappears. We may summarize his argument as follows: (i) (ii) (iii) (iv)
Intentionality in action requires a temporal structure. Basic action, being quasi-instantaneous, lacks temporal structure. Basic action therefore lacks intentionality. Basic action is therefore not agential, since a doer of intentional actions must be acquainted with what is intentionally done.
Lavin’s first task is to show that basic action is central to the causal theory. Basic action is unmediated because it ‘is not the end of any other action; nothing else is done in order to do it; it is not an answer to “Why?” when asked about any other action’ (Lavin, 2012: 3–4). Lavin argues that the causal theorist in particular needs to isolate an event that lacks this sort of intrinsic intentionality11 in order to make good
Lavin, ‘Action as a Form of Temporal Unity: On Anscombe’s Intention’ (Canadian Journal of Philosophy, 2016), p. 623. 11 It could be added that the kind of ‘intrinsic intentionality’ lacking in basic action (as Lavin wishes to conceive the latter) not only does not offer an answer to the question ‘Why?’ but also precludes the agent’s own question ‘How?’. For if an action is basic, according to Danto’s definition, then one does not need to ask ‘How?’ in doing it: there is no prior action that the agent must take in order to perform a basic action, that would necessitate the question ‘How?’. 10
6 The Temporal Structure of Agency
105
on (CTA), or the thesis that action consists of a mere event joined by a condition of mind. A closer analysis of Davidson’s example of turning on the light reveals a regressive series: Jones turned on the light via his intention to turn on the light, but the causalist’s basic causal relation depends on a further connection: Jones’s flipping the switch resulted in the light turning on (Lavin, 2012: 8). But Jones’s flipping the switch is yet another intentional item, so the causalist has not yet successfully isolated the necessary constituent of his equation, the ‘not-intrinsically-intentional event whose causation in the right way can constitute intentional action’ (Lavin, 2012: 8–9). Lavin concludes: Basic action is what stops this regress. If ‘X did A’ reports a basic action, then, ex hypothesi, X’s doing A does not involve X’s having done anything else intentionally. Here, finally, we have an action which might itself be identified with non-intentional movement caused by thought. Here, we reach a practical atom, something that might be shown to stand as the correlate of a mere happening. Of course, with the basic case in place the causal theory will extend the analysis by describing how non-basic action ultimately decomposes into an aggregate of basic actions, caused by intentions with such-and-such contents, which can themselves be decomposed into independently intelligible, metaphysically independent, internal and external elements, the mind of the acting subject and what merely happens. (Lavin, 2012: 9)
The concept of basic action thus expresses a fundamental commitment to a notion of material processes as metaphysically independent of the mental. Complex (non-basic) action, understood as intentional movement, requires ‘atomic’ decomposition in order to reflect this fundamental commitment. Lavin’s argument is that (CTA) depends, for the sake of the causalist arithmetic, on the existence of basic action ‘as the correlate of a mere happening’. For the causal theorist shares the assumption that ‘thought or reason must stop short of movement itself: it cannot reach all the way into the constitution of what happens. If the jurisdiction of practical thought must be so limited, so must the jurisdiction of specifically instrumental thought, the rational ordering of means to ends—there must be basic action’ (2012: 9). Lavin’s analysis discharges the general thrust of (one interpretation of) Hornsby’s criticism: that the causal theory implicates the alienation of an agent from her action by interposing causal intermediaries, both spatial and temporal. The problem with (CTA), on his view, is that it fails to recognize the necessary mediation of agential thought at all levels of action, whether simple or complex. Lavin’s more ambitious aim is to show that the notion of basic action cannot in fact exist.12 He begins by arguing that certain features of contemporary discourse
12
Lavin offers a substantial preliminary discussion in support of a weaker claim, that basic action need not exist in consequence of certain regress arguments that might variously be given by contemporary theorists (see Lavin, 2012: 9–17). These regress arguments may take different approaches, but for instance, one may believe that there must be simple skills if there are complex ones. Such belief may correspond to Hornsby’s argument, as cited by Lavin, that ‘[a]mong the things a person knows how to do, some of them he must know how to do “just like that”, on pain of needing to ascribe to him indefinitely many distinct pieces of knowledge to account for his ability’ (Hornsby, 1980: 88). Lavin’s discussion of the regress arguments is worth considering, but for present purposes I focus only on his stronger argument that basic action cannot exist and function as contemporary action theorists believe that it does.
106
J. Chik
tend to obscure reflection on the temporal structure of action. In particular, ‘the tremendous reliance on abstract noun phrases (“action”, “event”) and name-like phrases (“X’s doing A to Y”, “Y’s A-ing”) in the formulation of general philosophical theses—conceal the bad fit with most any pretheoretical characterization of the phenomenon of which a theory is wanted’ (Lavin, 2012: 19). Lavin proposes that philosophers focus on the aspectual distinctions between the schematic sentences, ‘X is doing A’, ‘X was doing A’, and ‘X did A’, and as represented concretely in action descriptions such as ‘I am walking across the street’, ‘I was walking across the street’, and ‘I walked across the street’. The aspectual opposition between the perfective and the imperfective suggests a sense of completion and incompletion, of something that may at one time be still underway, and at another is underway no more.13 Thus ‘the poles of the aspectual opposition are not simply at odds. Here doing looks forward to having done: when walking across the street is in progress, it is not then finished, but nevertheless it reaches ahead to completion’ (Lavin, 2012: 19–20). The imperfective ‘reaches ahead to completion’ even in cases where the walker is interrupted, for it ‘has intrinsic direction; it is a specification of what is to be, even if not what will be in fact’ (Lavin, 2012: 20). The aspectual opposition so conceived clearly draws on Aristotle’s concept of kinēsis, a movement that is performed to completion over time.14 These considerations lead to an argument against basic action given its apparent lack of temporal structure: ‘When described in light of this temporal structure basic action will not be recognizable as action’ (Lavin, 2012: 21). Lavin’s argument assumes that the temporal structure of processes so described, following Aristotle’s concept of kinēsis, is true. He proceeds: 1. Suppose X did A, which is a basic action. 2. From the temporal structure of processes, it follows that earlier X was doing A and had not yet done it. 3. It also follows that at the time X is doing A, other things have already happened, i.e., X did A**, and at the moment that X is doing A, still others are underway, i.e., X is doing A*. The latter subordinate phase can be subject to the same temporal analysis as ‘X is doing A’.15 4. Since doing A is basic, it follows that any subordinate phases to A do not themselves involve anything intentional, for they ‘are not themselves undertaken in pursuit of the goal’ (Lavin, 2012: 22).
13 For earlier discussion on verb aspect and verb types, see Mourelatos (1978, 1993) and Vendler (1957). 14 As also quoted at the beginning of this essay, Lavin cites Aristotle at Nic. Eth. 1174a 22–24: ‘In their parts and during the time they occupy, all movements are incomplete, and are different in kind from the whole movement and from each other’. 15 ‘And so, earlier when X was doing A*** and had not yet done it, other things had already happened in progress (X is doing A) is at once an ever increasing stack of have done’s and ever shrinking list of still to do’s’ (Lavin, 2012: 22).
6 The Temporal Structure of Agency
107
5. Therefore, the subordinate phases that constitute the progress of A’s being done ‘towards its completion is thus wholly opaque to its subject, except in the way it might be known to an observer or to someone with general knowledge of how such things happen. In short, the subject of basic action is alienated from the progress of her deed’ (Lavin, 2012: 22). The subject of basic action, with regard to the progress of her deed, stands in the same position as one who is unaware of the unfolding of a causally dependent action, i.e., one that need only be set in motion to progress independently of the agent’s awareness or will. Davidson’s example of the naval officer who sinks the Bismarck aptly illustrates the concept of a causally dependent action. Having launched the torpedo (perhaps by merely pushing a button), the naval officer relinquishes control of his intended deed, which progresses even after his death. ‘Mere movements of the body’, Davidson stated, ‘are all the actions there are’ (Davidson, 1971: 59).16 Lavin’s argument purports to show that basic action is itself a similar kind of kinēsis, the nature of whose progress is comparable to the progress of causally dependent actions such as sinking the Bismarck. The completion of the latter may be satisfied entirely outside of the naval officer’s control and awareness, so long as he initiates the process in the right way. Likewise, the completion of a basic action, defined as a mere movement in which there is no intrinsic intentionality, apparently proceeds entirely outside of the intentionality and awareness of the one who triggers it. Precisely because the movement is basic, it cannot be constituted by subordinate phases any of which themselves are intentional or possessed of rational thought. Thus, the concept of basic action renders the agent superfluous with regard to the subsequent unfolding of sub-events that constitute the basic event. Lavin remarks: ‘It looks like performing a basic action is just being the subject of a mindless, automatic process which the subject has somehow initiated, triggered, or launched’ (Lavin, 2012: 23). Having triggered the automatic sequence, the agent is relegated to a position of passive observer of what happens, much like Marx’s Nichtarbeiter or non-worker, who is ‘alienated from the material processes that realize his own ends and fortunes’ (Lavin, 2012: 18). Although the non-worker may have awareness of these material processes, ‘such awareness would be of something alien, outside and external, something given through observation—a matter of being put into contact with what is already there’ (Lavin, 2012: 18). Likewise, such is the condition of the agent, whose alienation is a consequence of basic action’s failing to be action at all, for there is no subordinate part of its progress that is guided or purposive.
16
In another paper, Davidson considers a different situation in which his right arm is paralysed, but it is placed in a pulley system wherein he can raise it by pulling on a rope with his left arm. ‘Raising an arm is usually done without doing anything else, but not always’, he comments (Davidson, 2004: 103). In this case, Davidson claims that he has raised his paralysed right arm by pulling on the rope with the other. Although raising an arm is usually something that one can do ‘just like that’, he has accomplished it only by doing something else with his body, an extra causal intermediary. Having pulled on a rope in order to raise his other arm, however, Davidson denies that achieving this latter end is part of the action at all: ‘[T]he answer to Wittgenstein’s question for a case like this is that nothing is added to the rising of my arm that makes it a case of my raising my arm because the rising of my arm is not part of my action at all’ (Davidson, 2004: 104).
108
6.3
J. Chik
The Time of an Action
As already noted, Lavin’s critique of causalism is distinctive in framing its challenge as a primarily temporal issue. In explicitly referring to Aristotle’s concept of kinesis, Lavin’s account recognizes that action is a special kind of change over time: one that necessarily has parts or sub-actions that occupy distinct sub-phases of a movement understood as a whole, where each temporal sub-phase must be known to and controlled by the agent. His analysis is responsive to the crucial question of what unifies the movement understood as a whole, which grounds our conception of agency as diachronically continuous – a problem unresolved by Hornsby and Bratman, as argued earlier. We can assume that causal theorists will be reluctant to entirely jettison the concept of basic action, since doing so undermines the causal theory’s fundamental structure (i.e. that an action is constituted by a mere movement joined by certain mental states), thereby forcing a revisionary definition of action in terms of the causal conjunction of mental states and non-basic movement. But this would require the causalist to define action circularly, in terms of something already intrinsically intentional. Plausibly, causalists might critically address the transition from steps (4) to (5) in Lavin’s argument above. The lack of intrinsic intentionality in basic action, it may be argued, does not in all cases entail that the agent is faced by utter opacity with regard to her movements. Even certain involuntary or non-intentional movements, such as one’s knee jerk reflex, are bodily occurrences of which subjects can claim awareness. However, the causalist is still burdened to show that basic action is at least movement of this kind (i.e. lacking in intrinsic intentionality and yet something of which the agent may be aware). And it remains unclear how the progress of such movements (qua movements intrinsically lacking intentionality) can ever exist as transparent processes to their possessors, as subjects of the non-observational type of awareness that seems essential to action. If action is indeed a kind of change over time – a special kind of change initiated and overseen by an agent throughout its continuous diachronic proceedings – skepticism about ‘basic action’ may well be justified. There are at least two ways in which the definition of action as change over time holds relevance for the issue at hand. First, it posits that the agent has a unique kind of control or possession of the diachronic proceedings of an action. Second, it asserts that all action – if it is action at all – must have a distinctive kind of temporal structure in light of this agential control. But are either or both of these points forfeited by causalism? A problem with Lavin’s argument is his conflation between the concept of ‘basic action’ and things we do spontaneously (‘just like that’) or quasi-instantaneously. As the problem of causal deviance (§2) suggests, one may conceive of a number of counterexamples to the causalist’s claim that to act is to have a bodily movement caused by something like a prior intention, decision, or choice: surely these counterexamples include some things that we do spontaneously and quasiinstantaneously, without deliberation about means and ends. However, Lavin seems to deny this important feature of agency. He argues:
6 The Temporal Structure of Agency
109
The capacity to perform a basic action ‘boils down to a capacity rational agents have of getting [something] done without needing to cognitively control how it is done.’ The passage is from Berent Enç, who is admirably clear-sighted about the need for this sort of commitment. To Enç it seems ‘intuitively clear’ that we have this capacity. To me it seems intuitively clear that we do not have an intelligible conception of rational agency, if we just have something that announces the thing to do, leaving the execution of the task to another power. Reason is not practical (i.e. efficacious) if it is simply agenda setting in this way: it must also be sufficient to the realization of its ends, and thus must be able to constitute the progress of the deed, the getting done of what gets done. (Lavin, 2012: 23)
Lavin may be right about these latter commitments to the concept of rational agency, but it is not clear that Enç’s remark really entails what Lavin assumes of it.17 Indeed there appears to be an ambiguity about what is meant by ‘cognitive control’. The term may imply the sort of conscious guidance of one’s movements that is minimally expected of agents in everyday cases of intentional activity (e.g., baking a cake, walking to market, swimming laps). Construed differently, ‘cognitive control’ could also imply that an agent must deliberate about the means – the ‘how’ and perhaps also the ‘why’ – of what she would like to achieve. The latter construal could include the requirement that mental antecedents must be in place for the agent to do anything intentionally. If Enç assumes the latter definition of ‘cognitive control’, then he is surely right that agents – arguably, human or not – have the capacity of doing something ‘without needing to cognitively control how it is done’. Why indeed, contra Lavin, should this plausible proposition entail a conception of agency on which one simply ‘announces the thing to do, leaving the execution of the task to another power’? Lavin apparently assumes the first construal of ‘cognitive control’, the denial of which would be absurd even for the most unreflective or mundane of activities: for even occasionally stirring the soup in the pot requires some guidance from the agent, throughout the temporal phases in which he stirs. His critique targets a view that would deny instrumental means-end structure to seemingly mundane movements like stirring the contents of the soup pot once around. For – to reiterate his argument above – if stirring the contents of the pot once around is a basic action in the sense that there is nothing cognitive involved at all, then the subordinate phases of stirring the contents once around (i.e., stirring the contents half-way around, stirring them a quarter of the way around, and so on) likewise cannot involve anything cognitive on the part of the agent. If so, then it appears that the progress of such basic action is unknown to (and alienated from) the agent: for knowledge of one’s own movement surely requires some sort of cognitive involvement over its temporal phases. The potential challenge to Lavin’s analysis concerns whether the causal theorist’s denial of ‘cognitive control’ implies that basic action involves nothing cognitive at all. The causal theorist may well protest that Lavin simply assumes a notion of basic action whose primary characteristic is its absence of rationality and intentionality, i.e., an event that can hardly be identified as ‘action’ at all. In contrast, one might presume (following Enç) that ‘cognitive control’ denotes deliberation on the means 17
See Enç (2006).
110
J. Chik
by which the agent sets out to act: in which case one’s denial of the latter would hardly entail the thesis that the agent’s movements lack cognitive involvement entirely. As for the construal of ‘cognitive control’ as agential guidance and awareness of her own movements, it is hard to believe that the causalist would really insist on a conception of basic action as something that agents do which nonetheless amount to utterly unguided movements of which they are wholly unaware.18 It seems undeniable that (CTA) implicates a concept of mere movement devoid of rationality – where ‘rationality’ is presumably some variant of instrumental rationality, but not agential awareness tout court – as part of the causalist’s decompositional programme. ‘Basic action’ is therefore a term of art that enters only later into the analysis of complex action, i.e., as involving several steps or stages, versus simple or basic ones. Is basic action compatible, then, with the temporal structure of processes characteristic of Aristotle’s kinēsis? In light of the foregoing, the answer seems to depend on which definition of ‘cognitive control’ one assumes. It may be possible to accept Enç’s concept of basic action without generating a problem of alienated agents. Indeed, non-complex action – depending on one’s construal of ‘cognitive control’ – may be understood as one way to oppose the decompositional thrust of (CTA). Many of our actions are metaphysically simple, having few or no spatial and temporal parts; apparently unreflective but not non-rational, they fail the mental states requirement of (CTA), for they involve movements that are undeniable intentional and yet uncaused by any identifiable cognitive event. Such ‘simple actions’ (if we may coin a novel term that contrasts with ‘basic action’) resist atomistic analysis, for they lack decomposable elements: they are themselves non-decomposable.19 This concept of ‘simple action’ relies on the possibility of agential performance not preceded by (nor contemporaneous with) deliberation about means, or anything resembling instrumental rationalisation. Within Aristotle’s framework, there can be action done spontaneously, apparently ‘just like that’, without any cognitive processes or states identifiable as deliberation, whether causally antecedent or concurrent to the action. Yet such action is no less intentional, for it may be done for some end: in Book II of Physics, Aristotle calls it ‘absurd to suppose that purpose is not present because we do not observe the agent deliberating’ (199b27).
18 By the same token, Lavin’s assumption that action is either ‘bound together by instrumental rationalisation’ or wholly lacking in rationality altogether, could be criticised on the grounds that he cannot conceive of rational action as ever falling in between these polar opposites (Lavin, 2012: 13). However, it may be that his conception of ‘instrumental rationalisation’ is sufficiently broad to admit actions done as a matter of habit, if not ‘just like that’. It is difficult to see the coherent application of ‘instrumental rationalisation’ to habitual actions if one assumes that ‘instrumental rationalisation’ involves cognitive control requiring means-end deliberation. A broad conception is conceivable only if the latter assumption is rejected. 19 This sort of reply is not incompatible with Lavin’s account, since he countenances actions that are done unreflectively, as a matter of habit or a part of routine activity. His account, however, specifies that such actions must be ‘bound together by instrumental rationalisation’ (Lavin, 2012: 13).
6 The Temporal Structure of Agency
111
The swallow builds its nest, the spider spins its web – all without deliberation – and yet, he argues, such creaturely action is performed no less for an end (199a10–34).20 In the human context, we may envision examples both familiar and extraordinary: Penelope’s skillful weaving and unraveling at cloth and loom, the deft use of a knife to dice onions for dinner, Heifetz’s rendition of Bach’s Partita in E. What is interesting about much of human action is how extremely accomplished it is: the amount of effort, preparation, and deliberation that generally informs what we do, even the first steps of a child learning to walk. Yet this action is simultaneously simple – think of the quasi-instantaneous shift of a violinist’s finger on the string – for accomplishment is essentially defined by the fact that any deliberation, preparation, and effort is merely historical within the life of an agent. Accomplished performance necessarily includes simple action because the effort and deliberation typically required of ‘complex action’ is at so great a temporal distance from what is done that it is all but forgotten in the moment of acting. And yet this does not entail that simple action is without any temporal structure whatsoever: although the movement of a finger shift on a violin string in the course of performance may be instantaneous, it occurs within time, and constitutes action only in the context of an agential end pursued over time (in Heifetz’s case, performing the whole Partita). These considerations lead to a revision in the conventional account of action as either complex or basic, as either taking much time or no time at all. Aristotle’s framework complicates this picture, making it impossible for us to neatly divide action into such facile categories. Relative to the whole Partita, a single finger shift is both temporally and spatially simple; but the capacity to execute the finger shift, in light of agential accomplishment, is neither basic or simple, whether temporally or spatially. The capacity itself requires a specific agential history: for a professional virtuoso it will have taken years of practice to perform such an action simply, with little or no thought and effort, with instantaneous ease. For the young beginner, like the child learning to walk, a single finger shift can be monumentally difficult, physically complex, and temporally extended in performance, with distinct sub-phases structuring the shift itself. These discernible temporal sub-phases disappear gradually with years of training, suggesting that agential capacities (as Aristotle argued) are malleable, but no less temporally structured in their actualisations. Lavin’s critique is therefore partly vindicated, partly revised. Action cannot be an occurrence whose temporal sub-phases proceed without the agent’s awareness, for its temporal parts are known intimately well by an accomplished agent whose performative fluency necessarily renders action’s sub-phases indiscernible, metaphysically irrelevant to the successful execution of corresponding powers. Such actions, done ‘just like that’, do (contra Lavin) exist: they are indeed pervasive in 20 Aristotle offers a more extensive treatment of this topic in De Motu Animaliam, especially §7. Imagine the following syllogism that proceeds to an action: ‘I need a covering, a coat is a covering: I need a coat. What I need I ought to make, I need a coat: I make a coat. And the conclusion “I must make a coat” is an action’. Though this style of practical reasoning reveals a general structure, Aristotle argues that an agent need not dwell on such propositions, nor must they occur to him at all (701a6–701b1).
112
J. Chik
the lives of agents who are accomplished in one way or another, in activities both familiar and extraordinary.21 Contra (CTA), however, their temporal structure is not defined by the occurrence of temporally prior or concurrent mental states or events that causally determine one’s movement, for what makes much of our intentional agency ‘simple’, I have argued, is the absence of such contemporaneous or antecedent mental occurrences. There may be lingering objections: are simple actions really empirical counterexamples to (CTA)? That our third-person perspective suggests an apparent lack of efficacious mental states does not definitively prove that there is none. A rational agent can give reasons for action, usually formulable into a specific desire and a relevant belief (as Davidson noted, it is otiose to cite both). Swallows and spiders cannot articulate reasons, and since language proves the presence of thought, they and other non-human animals are excluded from Davidsonian variants of (CTA). This argument has several disadvantages. First, it assumes that the articulation of reasons for acting are always reducible to mental states or events that are necessary for etiological explanations of action. Second, it confirms Bratman’s earlier worry, that human agency should be causally at odds with ‘the agency of purposive agents like dogs and cats’. Finally, such an argument asserts the truth of one’s theory regardless of the empirical evidence, and in ways that potentially contradict our firstperson phenomenological perspectives on agency. But, on principle, we should always refuse to deny facts for the sake of upholding one’s theory. ‘Every movement involves time, and relates to some goal’, argues Aristotle (1174a20). Our proposal has been that it is impossible to comprehend this conception of agency, unless we accept the agent as the primary cause of what happens, as the principal architect of action as a special kind of change over time. The temporal phases of action, whether simple or complex, relate to an end because the agent makes it so; therefore the time of an action, whether instantaneous or prolonged, occupies a place within proceedings that are consciously well ordered.22
21 Santiago Amaya (2016) proposes the paradigm of ‘slip-proof actions’ – bodily movements analogous to verbal slips that simply never happen, such as talk of ‘Sruedian slips’ – as a way of defining basic action. However, this definition seems to be an unnecessary implication of things we do ‘just like that’, without needing to do anything else. What seems undeniably simple or basic may not be absolutely fail-safe: even Heifetz can still make mistakes, or a single step towards the door can go awry. 22 This account is sympathetic to a metaphysics of processes rather than events, as argued e.g. by Helen Steward (2016). Although not explored in this essay, it is possible to construe the account of action as agential change as a variant of Steward’s account of actions as processes ultimately alterable by the agent, a view that resists the event causation assumed by (CTA), as argued by Steward (2012).
6 The Temporal Structure of Agency
113
References Aguilar, J. H., & Buckareff, A. A. (2010). Causing human actions: New perspectives on the causal theory of action. MIT Press. Amaya, S. (2016). Slip-proof actions. In R. Altshuler & M. J. Sigrist (Eds.), Time and the philosophy of action (Routledge studies in contemporary philosophy). Routledge. Aristotle. De Motu Animalium. Aristotle. De Generatione Animalium. Aristotle. Nicomachean Ethics. Aristotle. Physics. Bratman, M. (2000). Reflection, planning, and temporally extended agency. The Philosophical Review, 109(1), 35. Coope, U. (2007). Aristotle on Action. Aristotelian Society Supplementary Volume, 81(1), 109–138. Danto, A. With Sidney Morgenbesser. (1963). What we can do. The Journal of Philosophy, 60(15), 435–445. Danto, A. (1965). Basic actions. American Philosophical Quarterly, 2(2), 141–148. Danto, A. (1973). Analytical philosophy of action. Cambridge University Press. Davidson, D. (1963). Actions, reasons and causes. Journal of Philosophy, 60, 685–700. Davidson, D. (1971). Agency. In R. Binkley, R. Bronaugh, and A. Marras (Eds.), Agent, Action and Reason. University of Toronto Press. Davidson, D. (1980). Essays on Action and Events. Oxford University Press. Davidson, D. (2004). Problems in the explanation of action. In D. Davidson, Problems of Rationality. Oxford University Press. Enç, B. (2006). How we act: Causes, reasons, and intentions. Oxford University Press. Hornsby, J. (2007). Knowledge and abilities in action. In Proceedings of the 29th International Ludwig Wittgenstein Symposium. Hornsby, J. (1980). Actions. Routledge & Kegan Paul. Hornsby, J. (2004a). Agency and actions. In H. Steward & J. Hyman (Eds.), Agency and action. Cambridge University Press. Hornsby, J. (2004b). Alienated agents. In M. De Caro & D. Macarthur (Eds.), Naturalism in question. Harvard University Press. Hornsby, J. (2011). Actions in their circumstances. In A. Ford, J. Hornsby, & F. Stoutland (Eds.), Essays on Anscombe’s intention. Harvard University Press. Hornsby, J., & Stanley, J. (2005). Semantic knowledge and practical knowledge. Aristotelian Society Supplementary, 79(1), 107–130. Lavin, D. (2012). Must there be basic action? Noûs, 00(0), 1–32. Lavin, D. (2016). Action as a form of temporal unity: on Anscombe’s Intention. Canadian Journal of Philosophy, 45(5-6), 609–629. Mourelatos, A. (1978). Events, processes, and states. Linguistics and Philosophy, 2, 415–434. Mourelatos, A. (1993). Aristotle’s Kinêsis/Energeia distinction: A marginal note on Kathleen Gill’s paper. Canadian Journal of Philosophy, 23(3), 385–388. Smith, M. (1987). The humean theory of motivation. Mind, 96, 36–61. Smith, M. (2004). The structure of orthonomy. Royal Institute of Philosophy Supplement, 55, 165–193.
114
J. Chik
Smith, M. (2012). Four objections to the standard story of action (and four replies). Philosophical Issues, 22, 387–401. Steward, H. (2012). A metaphysics for freedom. Oxford University Press. Steward, H. (2016). Making the agent reappear: How processes might help. In R. Altshuler & M. J. Sigrist (Eds.), Time and the philosophy of action (Routledge studies in contemporary philosophy). Routledge. Thompson, M. (2008). Life and action: Elementary structures of practice and practical thought. Harvard University Press. Velleman, D. (1992). What happens when someone acts? Mind, 101(403), 461–481. Vendler, Z. (1957). Verbs and times. The Philosophical Review, 66(2), 143–160.
Chapter 7
Freedom of the Will and Rational Abilities Erasmus Mayr
Abstract Since the 1990s, agential capacities and abilities have played an increasingly prominent role in the debate about free will, especially among compatibilists. But it is still not fully clear what precisely compatibilists will gain by relying on capacities and abilities. In this paper, I try to argue that at least for compatibilists of a certain stripe an appeal to abilities is of significant use. These are compatibilists who try to argue for the compatibility of agential freedom and microphysical determinism by distinguishing between different, mutually irreducible levels of explanation (as, e.g., Christian List has recently done). An abilities account of free will provides the grounds for a much-needed principled argument why we can and should disregard certain microphysical features when providing the kind of explanation which is the primary and salient one for explaining intentional actions. Keywords Free will · Abilities · Compatibilism · Microphysical determinism · Levels of explanation · Christian list
7.1
Introduction
Since the 1990s, agential capacities and abilities have gained an increasingly prominent role in the debate about free will and moral responsibility. A steadily growing number of philosophers have come to argue that the conditions for free will and moral responsibility can be spelled out, at least in part, by reference to the agent’s abilities to act or to her rational capacities. This move has been especially attractive to philosophers who have also espoused an ontological realism about powers and rejected traditional analyses of power-ascriptions, especially of a simple conditional form. It thus fits in tightly with the major revival of interest in powers in contemporary metaphysics.
E. Mayr (*) Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_7
115
116
E. Mayr
For the majority of philosophers who have appealed to abilities and capacities, this move has been part of a defence of compatibilism in the free-will debate.1 It is easy to see why such a move is attractive to those compatibilists who believe that free will requires that the agent could have acted otherwise and want to show how this is possible in a deterministic world. Even if it is antecedently determined that I will not do X, this does not necessarily deprive me of my ability to do X, if we understand the latter as a general ability. I can have the ability to play the piano, even if I cannot at this moment exercise it, because there is no piano around, or because the only piano available at the moment is standing in a packed concert hall and I’m frozen stiff by the prospect of playing in front of a large crowd.2 Even though it may be determined, given my history and the laws of nature, that I will not play the piano under such circumstances, there is still a perfectly good sense of ‘can’ in which I can play the piano: I have the ability to do so (I have learned to play when I was a kid). This may be enough to say, afterwards, that, I could, in this situation, have acted otherwise: I could have played the piano (though I didn’t). Compatibilist accounts which appeal to abilities and capacities face a number of well-known challenges, though. Let me just mention one fairly obvious one. Even though there is incontestably one sense of ‘could’ in which it serves to ascribe a general ability, this is hardly the only sense of ‘could’. The following alternative description of the piano situation would make equally good sense: ‘Erasmus didn’t play, because he couldn’t play, given there was no piano (or given his fear of large audiences)’. So, why should the general abilities ascription sense be the relevant one for free will and moral responsibility? If anything, the piano case suggests that it isn’t, since – my general ability to play notwithstanding – we would hardly say here that I chose freely not to play or could be held responsible for not playing.3 This worry pushes abilities accounts towards interpreting free will not just in terms of the possession of general abilities, but either in terms of abilities which are more specifically tied to the action-situation, or else towards combining the possession of general abilities with further conditions. But the more they do that, the more the characteristic advantages ability or capacities accounts promise to deliver to compatibilists threaten to get lost. For instance, when we adopt an extremely specific view of abilities such that an agent lacks the ability to do X under maximally specific 1
E.g. Vihvelin (2004); Fara (2008); Smith (2003). An important precursor of such a view is Kenny (1978). But also incompatibilists can draw on abilities in their accounts of free will. Especially philosophers who hold that two-way powers whose exercise is ‚up to us‘ are crucial for our agency – e.g., Alvarez (2013) – can use such an understanding to argue for incompatibilism. 2 Vihvelin (2004), 443, in discussing the latter scenario, concedes “that there is a temptation” to regard it as one where the person lacks the ability to play the piano, but argues – convincingly, I believe – that we should resist this temptation and that my ability to play may still be present, even if my fear prevents me from using my abilities of deliberation and choice. Similarly, Fara argues that in such cases my ability would be masked, but not lost; see Fara (2008), 844–848, for a more extensive discussion of such cases. 3 Whittle (2010), 10 f., forcefully makes this point: Distinguishing more ‘global’ and more ‘local’ abilities, she argues that, at least in typical (if not all) cases, it is the latter rather than the former that seem pertinent for moral responsibility.
7 Freedom of the Will and Rational Abilities
117
circumstances Y unless it is possible that she does X under these very circumstances,4 the ‘could do X’ used in ascribing the ability to do X seems no longer very different from the ‘could do X’ used in stating that, under these circumstances, the totality of conditions didn’t exclude that the agent did X. But, on such an understanding, the truth of determinism seems to rule out that the agent could have acted otherwise than she did. So the concerns which speak against an initial understanding of ‘could’ in terms of ability-ascriptions – i.e. concerns about which sense of ‘could’ is relevant for free will – threaten to push abilities accounts towards a point where bringing in abilities is no longer of any real use to compatibilists. In this paper, I want to make some progress towards answering the question of what specific advantages compatibilists can expect from bringing in rational abilities. It is only to be expected that these advantages (if there are any) will be different ones for the different sub-debates of the free will debate.5 In this paper, I will try to show that relying on abilities brings a significant advantage to those compatibilists who try to establish the compatibility of free will and microphysical determinism by distinguishing between the subject-matters of different sciences and between different, mutually irreducible levels of explanation. Free will, this compatibilist approach maintains, only shows up at a ‘personal’ level of reality or description, and, for this reason, would not be undermined by determinism at a microphysical level. This approach, which had prominent proponents in the 1960s and 1970s,6 has enjoyed a major revival in the last years, especially through the recent work of Christian List.7 Since I think List’s presentation brings out especially clearly much of what is attractive about the general idea, I will be using it as a foil for my following argument. However, List’s account is left open to a major problem, which can be avoided if one uses a powers framework to spell out the ‘different levels’ idea – or so I will argue. So, my aim in the paper is twofold: On the negative side, I will argue against List’s version of the ‘different levels’ idea. On the more positive side, I will argue that this idea can be made to work much better when one relies on powers and abilities. For (1) at least when we are asking whether an intentional action is free, we consider an explanation of the action in terms of the agent’s rational abilities as the salient one,
Cf. what Whittle would call “‘all-in’ local dispositions” which “concern what the object is likely to do in just one type of C-case, which specifies every variable concerning the state of the object, its circumstances, and the stimulus it is subject to”, (2010), 4. Incompatibilists take ‘all-in’ local abilities to be necessary for freedom, (2010), 9. 5 See, e.g., Fara (2008), 862 f., for what advantages can be expected with regard to answering Van Inwagen’s Consequence argument, and Chap. 8 of this volume for a discussion of how a powers ontology itself might affect our assessment of this argument. 6 Esp. Melden (1961) and Kenny (1978). 7 Another major contemporary defender is Dennett, esp. in his (2004). 4
118
E. Mayr
(2) for this reason the crucial question for free will and responsibility is whether these abilities and their manifestations could have led to a different result under the circumstances, and (3) the preceding question can be answered in the affirmative even if microphysical determinism is true. I will proceed as follows: I will begin by briefly describing some key assumptions of the kind of abilities account of free will I will be relying on (Sect. 7.2). I will then give a brief description of List’s position, noting one key problem for it (Sect. 7.3), before I argue for steps (1), (2) and (3) (Sect. 7.4) above, and show why an abilities account of free will should be particularly attractive to compatibilists who are sympathetic to the general approach which List’s account represents.
7.2
Which Abilities and How to Understand Them
In this section, I’ll set out a list of assumptions which form the background of the following arguments. The first three concern the nature of abilities, the next three relate more directly to the free will debate. 1. First, I will assume a fairly robust realism about powers.8 Powers are genuine properties of objects, which these objects possess fairly stably over time, even when they do not currently exercise them.9 Furthermore, I will assume that power-ascriptions cannot be reductively analysed in the ways proposed by conditional and causal analyses of power-ascriptions and that powers are distinct properties from non-power, ‘categorical’ properties. 2. Second, there are very different sorts of powers. In particular, I will presuppose that there are ‘multi-track’ powers, i.e. powers whose characteristic manifestation is not just one type of behaviour, but which have several different kinds of manifestation. Furthermore, though some powers are necessarily exercised when the conditions for their exercise obtain, unless an intervening factor occurs, this is not true for all powers. There are merely probabilistic powers and ‘tendencies’ (in Ryle’s sense of the term10), for which the obtaining of the conditions for exercise makes a certain behaviour more likely (or the thing we could expect to happen regularly) without necessitating it (even when no intervening factors occur). Arguably, there are also two-way powers, whose exercise depends on the agent’s choice.11 An example for a tendency (in Ryle’s sense) is irascibility: If I am irascible, that doesn’t imply that I have to fly into a temper whenever I am provoked, but only makes it much more likely that I do. An example for a
8
A realism of the sort proposed by Molnar (2003). As Vihvelin (2004), 430 f., rightly stresses, this is the ‚commonsense‘ view about powers, and abilities; as far as possible, philosophers should respect it. 10 Ryle (1949), 126. 11 For the notion of ‚two-way powers‘ and its Aristotelian ancestry, see Kenny (1976), 53 f. 9
7 Freedom of the Will and Rational Abilities
119
two-way power is my ability to play the piano – even when a piano is at hand, it still depends on my choice whether I play it or not. While I consider it as highly plausible that there are genuine two-way powers, the following arguments will not rely on their existence. It is only important that not all powers are such that they have to be exercised (or manifested), or that their exercise must lead to a specific result, even when the conditions for their exercise obtain. I will further assume that for the individuation of a power it is not only relevant what it is a power for and what the conditions for its exercise (if it has any) are, but also how the manifestation of the power is connected to the obtaining of these conditions. I.e. a tendency to laugh on hearing a joke is a different power, I will suppose, from a fixed disposition to do so (i.e. a disposition such that you necessarily laugh in such a scenario). 3. Third, abilities are a sub-set of powers. What precise kind of powers they are, and how they are related to dispositions to do something under given circumstances, is a matter of dispute, but its details need not interest us here.12 What I will assume, however, is that abilities to do X cannot be generally understood as dispositions to (successfully) do X if one tries.13 For there are many abilities, such as the abilities to see, hear or understand a language, whose exercise “depends not on our trying to exercise them but on the presence of stimuli to which our cognitive systems respond”.14 4. Let us now turn to the assumptions pertaining more directly to the free will debate. I will assume, fourth, that free will requires that the agent could, in some sense, have acted otherwise than she did. With regard to moral responsibility, a connection to alternative possibilities is notoriously disputed;15 I will therefore restrict my following discussion to free will, without presupposing that free will and moral responsibility necessarily go together (though clearly they are closely related). 5. Fifth, how should we best understand free will in terms of abilities? Let us return to our earlier example, where I had the general ability to play the piano, but did not play because I was paralyzed by fear of a large audience. Even though there is a sense of ‘could’ in which I could have played the piano (I had the general ability to do so), it seems fairly clear in this case that being able to play the piano in that sense is not sufficient for free will (or, for that matter, for moral responsibility): I did not freely decide not to play the piano. What do we need in addition? As a first step, in addition to my possessing the ability, external conditions must have been such that I could exercise it. Take a trivial example: Even a perfect piano player cannot play if there is no piano at hand or if her hands are tied to a chair. Following Anthony Kenny, I will say that the agent needs to also have
12
For an excellent recent overview of the debate see Jaster and Vetter (2017). As Fara (2008), 848, suggests for abilities to act (intentionally). (Fara’s focus is, from the start, only on abilities to act (intentionally), rather than on abilities in general, see p. 844, fn. 4, and p. 849.) 14 Jaster and Vetter (2017), 6. 15 The locus classicus for the attack on this connection is Frankfurt (1969). 13
120
E. Mayr
an ‘opportunity’ for exercising her ability.16 But, plausibly, we need even more than that. For agents can have abilities and be prevented from exercising them by psychological rather than external obstacles: Addictions, phobias, and my paralysing fear in the piano case can all remove my freedom and the possibility that I act otherwise than I actually do, even though they neither remove my general ability to act in a certain way nor change the external circumstances in such a way that I cannot exercise it. What the agent lacks, or cannot exercise, in such cases, many philosophers have thought, is an (additional) rational ability, i.e. the ability to properly deliberate or decide on the basis of perceived reasons, or to respond to (or rationally guide herself in the light of) the reasons that apply to her.17 Without going into the details of their different views, I will broadly follow suit: I will assume that what the agent needs for free will is (also) the ability to guide his action, decision and deliberation in the light of (some) perceived reasons and the opportunity to exercise this ability. For the sake of brevity, let us call this the ability for rational self-guidance.18 This ability is much more modest than an ability to do the right thing, or to do what one takes to be the right thing. Exercising this ability only requires that we deliberate, decide and act on the basis of some perceived reasons, even if we are aware that there are better reasons we are ignoring. The ability is therefore also exercised when we act akratically (i.e. act against our best judgement), but still act (in what we consider to be a wrong way) for reasons (e.g. because we think we will enjoy it very much). I hope that most readers generally sympathetic to an ability account of free will can agree that some such ability is needed, though they are bound to disagree on the details and may well think that further rational abilities are needed as well.19 But for the sake of simplicity, I will only be concerned with the ability for rational self-guidance here. The ability for rational self-guidance has some interesting features: On the one hand, it is an ability whose exercise does not typically depend on the person’s trying to exercise it. We often deliberate, decide and act spontaneously; and we
Kenny (1976), 133: „An ability is something internal to the agent, and an opportunity is something external.“ As Kenny himself immediately goes on to stress, it is hard to draw this distinction precisely. 17 Suggestions along these lines can be found, e.g., in Wallace (1994), Fischer and Ravizza (1998), Raz (1999), Smith (2003), Vihvelin (2004). There are many important differences between their views; but the details do not matter for us here. 18 This label is used in different ways in the debate, and my usage is not meant to reflect a generally accepted one. 19 Fischer and Ravizza’s “moderate reasons-responsiveness”, (1998, chap. 3), which they take to be necessary for moral responsibility, is a more demanding ability. Vihvelin’s “ability to make choices on the basis of reasons”, (2004), 429, is fairly close, though the ability for rational self-guidance need not manifest itself in decisions, but can directly issue in actions done for reasons. This fits better the fact that it is natural to consider many of our actions as intentional and free even when we haven’t decided to act beforehand. 16
7 Freedom of the Will and Rational Abilities
121
don’t normally try to deliberate or to decide. On the other hand, the exercise of the ability for rational guidance can lead to very different results: Both a decision to go to the cinema tonight and a decision to stay at home tonight can be results of this ability’s exercise. This is something this ability has in common with multitrack powers. But, at the same time, the former ability is very different from typical instances of multi-track powers, such as elasticity. Elasticity is multi-track because it can be manifested both by the object’s contracting and expanding, depending on whether it was first compressed or stretched out.20 I.e. the way the elastic object behaves varies according to the kind of stimulus applied on that occasion. This doesn’t mean that, on one and the same occasion, the elastic object could behave in different ways. The ability for rational self-guidance is importantly different in this respect: Different decisions could issue from one deliberation process on a particular occasion, even if the ‘starting-conditions’ and the features prompting the deliberation are fixed. Imagine that I have reasons to go to the cinema, but also reasons to stay at home this evening, and that both these sets of reasons are, from my deliberative perspective, incommensurable. Then my decision might go either way – as far as my deliberation process itself is concerned, its result is not fixed. (This does not mean that there may not be sub-personal features which determine what the outcome will be; we will return to this point below in Sect. 7.4). 6. My last assumption is a fairly common one in the debate about free will: Mental states and properties are neither identical with nor reducible to physical ones. This is compatible with a strong degree of determination of mental states by underlying physical ones. The physical state of my body and my environment may determine exactly which mental states I am in and which mental properties (including mental powers and abilities) I have. Still, mental states and properties are not identical to physical ones, nor is there a one-to-one match between the two. For the purposes of the following arguments, I will presuppose this broadly physicalist, but non-reductionist framework.21 I will further assume that mental properties and states are causally relevant rather than mere epiphenomena.
7.3
List on Free Will and Agential Possibility
Let us now turn to Christian List’s compatibilist account of free will. List argues that though free will requires the possibility of doing otherwise, this possibility is consistent with microphysical determinism since what is at stake for the question of free will is ‘agential’ possibility rather than microphysical possibility.
20
Ryle (1949), 113. For a more detailed overview of the versions of non-reductive physicalism see Kim (1996), chap. 9. 21
122
E. Mayr
When we are interested in whether a particular action is possible for an agent, [. . .] the appropriate frame of reference is not the one given by fundamental physics, but rather the one given by our best theory of human agency. Thus the description of the world that matters here is not a (microscopic) physical one, but a (macroscopic) psychological one. (List, 2014, 161)
The different psychological or ‘agential’ states that matter for ‘agential’ possibility are more coarse-grained than the underlying microphysical states, because, as List argues, even if we assume that the former supervene on the latter states, they can be multiply realized. “There is typically more than one physical state that gives rise to the same agential state; not every variation in the physical state needs to bring about a variation in the agential state.”22 While the microphysical state of the agent’s body and environment fully determines what psychological state she is in, the converse does not hold true, since one and the same psychological state (call it: supervenient state) could, on different occasions be due to different microphysical states (call them: subvenient states). In consequence, we cannot infer from the fact that only one subsequent course of events is (micro-)physically possible – assuming microphysical determinism to be true – that only one course of action is ‘agentially’ possible for the agent, too.23 For whether and which events are determined by a microphysical state might (and presumably will) depend on the specific and complete microphysical properties of that state. Not all differences in these properties will be reflected in differences of the properties of the more coarse-grained psychological state that supervenes on this microphysical state. Even if it is necessary, given the laws of nature, that micro-state A at t1 is followed by micro-state B at t2, it need not be similarly necessary that psychological state A* which supervenes on A be followed by B. For A* might have equally supervened on a quite different micro-state A’ which would not have been followed by B, but rather B0 . Thus, though the occurrence of A determines the occurrence of B, the occurrence of A* need not do so. The same point applies, of course, to the determination of future psychological or agential states by earlier psychological states. I think that in at least one important respect List’s argument is successful: As long as we do not consider psychological and microphysical states and properties to be either type-identical or at least to be matched one-to-one as types, we can reject the inference from determinism at the microphysical level to a corresponding determinism at the psychological or agential level. For even if the microphysical states fully determine which psychological states the agent is in, the determinism on the microphysical level might depend on specific microphysical features which just don’t ‘show up’ on the psychological level, because one and the same psychological state could have different microphysical states as its subvenient states. In that case, the agent’s choices and actions may not be necessitated by antecedent agential and psychological phenomena.
22 23
List (2014), 162. What List calls the rejection of the ‘linking assumption’, (2014), 161.
7 Freedom of the Will and Rational Abilities
123
The question remains, however, why we should abstract away from determination at the microphysical level when the question of free will is at issue. Why should the ways in which the psychological states of the agent are realized be ‘bracketed’, as it were, when we ask, in the context of the free will debate, whether the agent could have acted otherwise? List’s answer relies on the distinction between different ontological and explanatory levels. It is the coarse-grained psychological-agential level which provides the framework for questions about whether the agent is free rather than the fine-grained microphysical level. For free will is a phenomenon that only ‘shows up’ at the coarse-grained level. Considering the best theories which explain the phenomena at this level, for List, gives us even positive reason for thinking free will exists. Free will is “a special science phenomenon: Free will, in the technical sense of an agent’s having a choice between more than one course of action in many situations, is a key presupposition of our best scientific theories of agency, at least when these theories are understood literally.”24 And when it comes to picking our ontology with regard to a certain subject-area, we should adopt what List calls a “naturalistic ontological attitude”, and accord to our best scientific theories of this subject-area the decisive verdict.25 With regard to agential phenomena, these will be our best developed theories of decision theory. Since they presuppose free will, we should regard free will as a real phenomenon, too. It merits emphasis that List’s answer is not intended to establish the existence of alternative agential possibilities as a mere reflection of “the epistemic limitations of the special sciences”.26 The result is not meant to be just that, within the special science relevant for explaining human actions, we are unable to state how the agent must behave. Rather, List espouses a robust explanatory realism: Since the best theories in this science presuppose free will and alternative agential possibilities, and since this presupposition is compatible with deterministic theories at a micro-level, we should accept these phenomena as fully real. But incompatibilists are unlikely to be fully convinced by this answer.27 To begin with, they will insist that the notion of ‘having a choice’ they are concerned with is different from the one presupposed by advanced decision-theories. Decision-theory is just not interested in many of the features that are pertinent to the question of whether the agent has free will. It will consider the agent as ‘able to choose otherwise’ even when we wouldn’t do so in the context of the free will debate, where considerations about moral responsibility and the fairness of holding persons
24
List 2014, 168. List 2014, 167, and (2019), 7 ff.. 26 List 2014, 173. 27 There are, of course, further ways in which List’s proposal can be criticised (see, e.g., Chap. 3 of this volume), which I cannot go into here. 25
124
E. Mayr
responsible come in as well.28 So, even if decision-theory assumes that agents can choose between alternative courses of action, it is fairly unlikely that the sense of ‘can choose’ they are relying on is the same as the one at issue in the free will debate. By consequence, from the fact that decision theory makes the assumption that agents ‘can choose otherwise’ we cannot (without much further argument) infer that agents act ‘freely’ in the sense at stake in the free will debate.29 This is not merely a fact about the particular scientific theories List is relying on. There is a more general worry here about List’s strategy. List, in crucial places of his argument, seems to treat the question of ‘whether free will is real’ as an ontological question, to be decided by looking at the ontological presuppositions made by the relevant scientific theories – just as we would, according to him, decide whether, say, psychological states are real, on that basis.30 At the same time, he insists, that what he is trying to vindicate is a modal, rather than a conditional or dispositional, interpretation of the ability to do otherwise connected to free will, since the latter interpretations fail to capture that the agent must have genuine “alternative possibilities in the given situation”.31 But even if one is sympathetic to List’s approach to deciding ontological questions about what entities are real, it doesn’t follow that the modal question of whether the agent can really act otherwise in the action-situation can be decided in the same way. Even if decision-theoretic explanations (be it in their best currently available or further improved versions) presuppose that the agent has more than one option he could choose, this is only (and will only be) ‘could choose’ in some sense. There are bound to be several interpretations of this presupposition, all of which will be compatible with the role this presupposition is meant to play for the decision-theoretic explanations. Some of these interpretations are bound to be conditional or dispositional ones.32 So, even when one adopts List’s robust explanatory realism, a look at decision theory doesn’t seem, in principle, a promising way to show that we can act otherwise in the sense proposed by the modal interpretation (rather than in some other sense). If this is correct, it shows that List’s positive argument for the existence of free will is on much shakier grounds than he supposes. But this observation leaves a key part of List’s argument intact, namely the one about the in-principle compatibility of microphysical determinism with the existence of alternative ‘agential’ possibilities.
28
This objection is also raised by Menges (2021). It merits emphasis that this worry does not rest on assuming that free will and moral responsibility necessarily go together. And it is undeniable that concerns about moral responsibility are intimately tied to the understanding of free will in the traditional debate, and this connection plays a major role in giving the free will debate the importance that it has for many of its participants. 29 List himself comes close to accepting this, when he writes that we must assume, in decision theory, “that there is at least a thin, technical sense, in which [different] options could have been chosen”, 2014, 168. Why should this ‘thin, technical sense’ be the one at stake in the free will debate? 30 See List (2014), 167 f. 31 List (2019), 85. See also List (2014), 158. 32 E.g., Bok’s interpretation of ‘freedom’, as it is relevant to the practical point of view, in her (1998), ch. 3, seems likely to fit the bill.
7 Freedom of the Will and Rational Abilities
125
Should we accept at least this part of his argument? We can do so only if the higherlevel character of free will implies that, in asking whether the agent has free will, we should ‘bracket’ the specific microphysical ways in which the agent’s psychological states leading to her action are realized. The main reason List offers for such ‘bracketing’ is the multiple realizability of these psychological states. But multiple realizability in itself may seem too slim a base for abstracting from what microphysical state underlies the psychological state on the particular occasion: For whenever the agent has a relevant psychological state, this will be realized by some specific microphysical state (even though, on other occasions, it can be realized by others). If that specific microphysical state can only lead to one result on that occasion, would we not most naturally say that on that occasion the realized psychological state could only be followed by one result, too – and thus there were no alternative ways in which the agent could have acted after all?33 (Of course, we may not know what this realizer state is, and therefore be unable to predict which result will follow; but since List’s argument for the ‘reality of free will’ is not meant to be based on such epistemic limitations, such ignorance is irrelevant here.) Presumably,34 List would respond to this worry by insisting it underestimates the status of psychological and intentional explanations in making actions intelligible. These explanations are not a ‘second-best’ to microphysical explanations, which we adopt only faute de mieux because we are ignorant of which microphysical states underlie our psychological and intentional states on each occasion. Rather, they are the crucial explanations when it comes to understanding actions, and what decides the question of whether the agent ‘could have acted otherwise’ is whether these explanations are deterministic or not.35 There is a stronger and a weaker reading of the view that these explanations are ‘the crucial ones’, which I will take on in sequel. On the stronger reading, these explanations are the ‘crucial ones’ because they are the only ‘real’ causal explanations, i.e. the only ones which cite the causes of the action we are interested in, while the subvenient microphysical states are not really causes of this action at all. While such a claim may be prima facie surprising, a parallel claim is in fact made, for an important group of cases, by List and Menzies (2009) who argue that a higher-level causal relation rules out a corresponding lowerlevel causal relation if the former relation is “realization-insensitive”, which means that the result would obtain “even under some small perturbations in the realization” of the higher-level phenomenon which is the candidate cause.36 The intuition they
33 Menges (2021), 5 f. As Menges puts it (2021, 6): “Assuming that the realizing physical state is part of a deterministic history, it is very plausible that the realized psychological states follow this deterministic history.” Thanks also to an anonymous referee for pressing me on that point. 34 Apart from the considerations about the ‘naturalistic ontological attitude’ and multiple realizability I have discussed above, List does not, in his (2014) and (2019) offer an explicit argument for why the specific realizer state should be ‘bracketed’. What follows is, therefore, an attempted rational reconstruction of how his argument would presumably go. 35 I call an explanation ‘deterministic’ when the obtaining or occurrence of the explanantia make it necessary that the explanandum obtains or occurs. Mutatis mutandis for ‘indeterministic’ explanation. 36 List and Menzies (2009), 496.
126
E. Mayr
are relying on here is that a cause is ‘what makes the difference’, and that in cases of realization-insensitivity the precise lower-level realization of the higher-level phenomenon ‘didn’t matter’ for the result. Since the counterfactuals required for realization-insensitivity could obtain even when the lower-level is deterministic and the higher-level indeterministic, this might lead one to conclude that – if List’s and Menzies’ arguments from their (2009) are successful – at least for a significant class of cases a purported causal explanation in terms of microphysical phenomena is simply false, because the higher-level, e.g. psychological, phenomena exclude the lower-level ones from being causes. But quite independently from the question of how to assess List’s and Menzies’ argument in their (2009), the transfer of their result to the case where the lower-level is deterministic and the higher-level indeterministic would seem highly problematic.37 Remember, the intuition they are building on is that sometimes the lowerlevel state doesn’t make the difference. In the scenario we are considering here the psychological phenomenon does not necessitate the action, but only makes it more likely. The occurrence of the underlying microphysical event, by contrast, does make it (at least physically) necessary (together with other factors) that the action will occur. But then, the occurrence of the microphysical event did make some difference – at the very least, it made something physically necessary which otherwise wouldn’t have been so. So, I don’t think List can convincingly respond that the psychological explanation is the ‘crucial’ one because it is the only real causal one here. But there is a weaker reading of the ‘crucial explanation claim’, which only states that the psychological explanation is the ‘salient’ explanation (or the one telling the ‘salient causal story’) we are interested in when it comes to making actions intelligible. Why should this be so? Well, free will is a ‘higher-level’ phenomenon, in the sense that what we are concerned with when we ask whether the agent was free is understanding and explaining psychological or ‘agential’ phenomena, such as actions or decisions. These phenomena do not ‘show up’ on a microphysical level of reality. So, it is only natural to expect that such psychological or ‘agential’ phenomena must be explained in terms of other psychological or ‘agential’ phenomena, and (given a robust explanatory realism) we must consider these explanations to pick out the salient causal story of why these phenomena occur. If these higher-level explanations are not deterministic, should we not conclude that this causal story is one of indeterministic rather than of deterministic causation? And if the salient causal story is an indeterministic one,38 would this not provide a strong argument39 for concluding that the agent ‘could have acted otherwise’ in the sense primarily relevant for 37
And List and Menzies only consider deterministic causation, i.e. not the scenario at issue here, (2009), 477 f. 38 In Menges’ Alfia case, (2021), 5 f., List could argue that the deterministic story is only the microphysical and not the salient psychological one. 39 Not a necessarily decisive one, though: There might be additional considerations (e.g. about fairness to punish and reward) which might still speak in favour of stricter ‘could have done otherwise’ conditions. But there would be a strong prima facie case for considering the agent as free if the salient causal story leading to her action did not involve deterministic causation, and the
7 Freedom of the Will and Rational Abilities
127
understanding agential phenomena and thus in the sense that has the best prima facie claim to being the relevant one for the free will debate? I think this line of argument has some prima facie plausibility. But even if we grant List that the ‘salient’ explanations for understanding actions are those in terms of psychological phenomena, the argument does not ultimately succeed. The reason is the following: That a phenomenon is a higher-level one which can (and maybe: must) be explained in terms of other higher-level phenomena does not mean that it can (or must) be explained exclusively in higher-level terms. Nor does it mean that, in an explanation of the phenomenon which appeals to other higher-level phenomena, appeal to lower-level phenomena or features is necessarily inadmissible. For talk of different ‘levels’ of domains and explanation is always a simplification and overidealization. We cannot assume that natural phenomena come tidily ‘stratified’ in a hierarchy of different levels such that the different levels are ‘closed off’ against interference from each other and that explanations of phenomena of one level cannot include features from ‘lower levels’ as well. On the contrary, we should expect that there will often be partial ‘cross-explanations’ which mix explanantia from the same level as the explanandum and ones from lower levels. For illustration, take the following case involving biology and chemistry.40 Imagine that infection by a certain virus N (a biological phenomenon) can be realized by very different chemical states, N1 and N2, in agents with clinical pre-condition P. When it is realized by N1 (and only then), N can interact with the pre-condition P to produce a further chemical feature F (e.g. a massive rise in the oxidation rate in a relevant group of molecules), though it does not invariably do so. If F has been produced, then there will be further interaction with N that leads with full certainty to an inflammation of the patient’s liver. If, by contrast, F is not produced, N, when realized by N1, only leads to liver inflammation with a likelihood of 0.7, which is the same likelihood with which a realization of N by N2 leads to this effect. Now consider the point of time where F has been produced in a patient, and take a doctor who knows both this chemical fact and the biological fact that this patient with precondition P is infected by N. Should this doctor, in considering ex ante the likelihood of an inflammation of the patient’s liver, say that such inflammation may, but need not occur, given the patient’s condition? This would be a hardly credible result. The appropriate assessment, if the doctor knows that the patient is both infected by N and chemical feature F has been produced, seems to be simply that the inflammation will certainly occur. The same point holds when ex post explanation of the inflammation is at issue. Both the prediction and the explanation are based on a combination of biological and chemical features of the situation, and on how
burden of proof would lie with the incompatibilist to show that a different sense of ‘could have done otherwise’ was relevant. 40 The relation between these two sciences and their domain is often regarded as one instance of the ‘higher-lower level’ relationship, see, e.g., List (2019), 33.
128
E. Mayr
these features interact, which shows that there is no in principle obstacle to combining such features within one and the same explanation.41 It merits emphasis that the possibility of such ‘mixing’ is fully compatible with the status of biology as a ‘higher-level’ science and one that is not reducible to chemistry. Nor does it make biological phenomena (as distinct from chemical ones) causally inert or explanatorily redundant. This would only happen if (to stick with the context of our example) the full explanation of the connections between infections by virus N and liver inflammation could be spelled out in chemical terms, or if the biological laws which explain these connections could be completely derived from chemical laws. But nothing like this is implied by our example. What is implied is only that the higher-level phenomena and laws do not, in all cases, provide a full explanation of all higher-level phenomena and their features. This leaves ‘gaps’ in the higher-level explanations, which sometimes can be filled by appeal to lowerlevel phenomena. By the latter appeal, we can, as it were, ‘enrich’ the original explanation. Such mixed explanations are also possible when psychological or intentional features are essentially included in them. Imagine that an agent is trying to figure out whether to do X, and that figuring out whether this is a good idea is an extremely tedious task. Since the agent has a strong aversion to doing X, she will only do it if she positively decides that doing X is better than not doing it. Her desires and beliefs in the situation do not make it impossible that she will go through the figuring-out process and end up with the positive result that she should do X – and thus, leave it open whether she does X eventually. But the agent also has a neurophysiological condition which make it impossible for her to concentrate when too dehydrated, and, alas, that’s the state she is currently in. For that reason, the combination of her beliefs and desires, her neurophysiological condition and her dehydration, make it impossible that she does X on that occasion, and we would naturally combine all these features within one explanation of why she didn’t do X. If this point about the possibility of mixed explanations is true, List’s argument for compatibilism has a crucial lacuna. Even if agential phenomena are not determined by other agential phenomena (and psychological laws), and even if agential phenomena should be explained (also) in terms of other psychological phenomena, they might still be subject to psychological-cum-microphysical determination. For ‘gaps’ in the psychological explanation might always be capable of being filled by appeal to microphysical phenomena which ‘interact with’ the agential and psychological phenomena in the right way and which can therefore be cited within one and the same explanation as the latter. If that happens, why should we ignore this psychological-cum-microphysical determination? Why should we, in assessing whether the agent could act otherwise, abstract from the ‘cum-microphysical’ part, instead of concluding (similarly to the doctor in our example), that the agent could
41
A reductionist about biological phenomena might say there is no cross-level interaction here, because only chemical features interact. But I presuppose that the biological phenomena are not reducible to chemical ones (precisely as the special sciences realism List espouses suggests).
7 Freedom of the Will and Rational Abilities
129
not act otherwise, after all? At least, we are missing an argument why we should do the former: The fact that free will is a higher-order phenomenon does not yet establish that we should.
7.4
Bringing in Abilities
In this section, I am going to argue that adopting a rational abilities account helps the compatibilist to make significant progress on this last issue. In particular, I want to argue that such an account provides the compatibilist with an argument that certain lower-level features can (and should) be licitly disregarded when we ask whether it was possible for the agent to act otherwise, at least when we are concerned with intentional actions done for reasons. In order to make this argument, I will narrow my focus to the latter cases: I.e. to cases where an agent has intentionally performed an action (say stolen a book) for a reason (because he thought it would be a wonderful read), and the question is whether he did so freely and could have acted otherwise. Let us restate the difficulty for List’s proposal, as diagnosed in the last section: List can say that the agent’s intentional action is a higher-level rather than a microphysical phenomenon, and that we can expect a higher-level (i.e. agentialpsychological) explanation for it. But he cannot exclude, I have argued, that we might combine, within our explanation, higher-level factors with lower-level ones. For this reason, we could not infer that the salient explanation of the action was not a deterministic one from the (putative) fact that the higher-level factors did not determine the agent’s action. Why was such a cross-level combination possible? When we explain an occurrence in terms of the occurrences of antecedent states or events there seems no restriction on which states or events we can combine as partial explanantia within one and the same explanation, beyond a ban on explanatory redundancy and the requirement that these states and events were relevant to the occurrence of the effect. There is no obstacle to combining both psychological and physical or sub-personal features within one explanation as long as all these features are needed to explain the explanandum. Let me put this point in terms of standard Humean event-causation. (List’s remarks are fully compatible with thinking that the causal relations which make most scientific explanations true are all of this kind.) When you ask which events caused a certain result, there seems no real restriction on which events you can put together as (partial) event-causes, except a ban on ‘double-counting’, as long as there is some lawlike connection between these events and (features of) the effect.42 Both psychological and lower-level events can figure in the set.
42
Some theorists would insist that causes have to precede the effects; but this restriction doesn’t impact on my argument.
130
E. Mayr
But – you might object on List’s behalf – only if the latter aren’t subvenient events for the supervenient psychological ones we include! If we included both kinds of events in our causal explanation of what occurred, we would have doublecounting and explanatory redundancy. Which one should we choose if both are prima facie candidates for being included? If we follow Stephen Yablo, we should choose the more “proportional” factor as cause, i.e. the one that ‘made the difference’.43 This may well be the psychological event rather than the specific microphysical event which underlies it. And as long as we cannot include both events within the salient explanation of the action, doesn’t that already give List a principled account of why some microphysical events have to be disregarded in the salient explanation? Isn’t that observation enough to salvage List’s argument? I don’t think so. For, remember, the relevant scenario for List is one where we have microphysical determinism without necessitation of the action by antecedent psychological events. I.e. a scenario where the agential/psychological event at t1 makes the action more likely to occur at t2, but does not necessitate it, though the occurrence of the microphysical event at t1 makes it necessary that the action occurs at t2. In such a scenario, we cannot say that only the psychological event at t1 ‘made a difference’ to the action’s occurrence.44 Both events did. There is genuinely extra explanatory value in appealing to the occurrence of the subvenient state at t1 (or at least to those features of the subvenient state not captured by the description of the supervenient state) as compared to explaining the action’s occurrence solely in terms of the occurrence of the supervenient event at t1. The latter explanation cannot make it intelligible why the action had to occur, while the former can. Because of this, we shouldn’t rule out that both the supervenient and the subvenient events can be considered as causes of the resultant action45 and that both can be licitly cited within one explanation of this action. The situation crucially changes, I want to suggest, when we switch from Humean event-causation to an explanation of the resulting action in terms of the manifestation of powers or abilities. When we explain an occurrence in terms of the manifestation of a power, we cannot, within one and the same explanation, add any old further factors or events as additional explanantia, even if these factors are somehow relevant to the occurrence. We can only add factors which are suitably related to the manifestation of this power; otherwise we switch to a different kind of explanation where we no longer make the occurrence intelligible as (the result of) a manifestation of this power. Since abilities are a sub-type of powers, the same holds for abilities and explanations in terms of ability-exercise. What factors are suitably related to a manifestation of a power or an ability? The fact that the object possesses this power or ability, the presence of the external circumstances needed for their manifestation or exercise (the ‘opportunity’, for an ability) as well as the
43
Yablo (1992), 227. The case is therefore different from the ones that Yablo considers to illustrate the possibility that the mental event can be the more proportional one. 45 Vide p. 126 supra. 44
7 Freedom of the Will and Rational Abilities
131
absence of internal ‘blockers’ which would prevent the beginning of the exercise (e.g. the absence of stage-fright in our example in the introduction), features that led to the beginning of the exercise (e.g. the motives which led the agent to exercise her ability), features which can interact positively or negatively (interfere) with the power’s exercise once it is started (like some masks), and features which influence the further causal impact of this exercise. This list is not meant to be exhaustive, but any other factors which could be included must fit into the explanatory scheme in which the manifestation or exercise of the power or ability can figure, too. Take, for illustration, the following example: Imagine that a piece of sugar is put into water and you explain its dissolving by referring to its water-solubility which is being manifested in this process. There are many further features which you can add in this explanation, such as the action of the person who put the sugar into water or the rise in temperature which had just melted the water from an ice-block. These features explain why the sugar’s water-solubility came to be manifested and so can be combined with the manifestation-explanation. But you cannot, within this explanation, ‘add’ to the power-manifestation as further explanantia the microphysical processes and events which were necessary parts of the power-manifestation (such as the water’s breaking up the surface tension of the piece of sugar etc.). These microphysical processes and events didn’t interact with the manifestation, nor were they part of the external circumstances necessary for it or leading up to it. We can neither say that they caused the result jointly with the power’s manifestation nor that they caused it via the power’s manifestation (as the action of putting the sugar into water did). These microphysical processes can appropriately be referred to in a further explanation of how the manifestation ‘worked’; i.e. they can be cited in an explanation with a different explanandum.46 But they cannot be referred to in an explanation of why the result occurred which already explains this result in terms of the manifestation itself. Which of the two explanations we pick – explanation by reference to the power’s manifestation or by reference to the sub-steps – depends on the question we want to answer, and which factor we consider to be more ‘proportional’ to the result we want to explain. But we cannot pick one factor alongside the other in one and the same explanation. The explanatory scheme of powermanifestation explanations doesn’t allow that, because it is more restrictive, in this respect, than the explanatory scheme we use in event-causal explanations. If this is correct, how does this impinge on the issue of whether the agent could have acted otherwise in the context of free will? Assume that the ‘salient’ explanation for an intentional action is in terms of the manifestation of a rational ability rather than in terms of the microphysical goings-on which underlie this manifestation. Assume that this ability is, or involves, a mere tendency to X, while the microphysical goings-on lead deterministically to a microphysical event upon 46
Or else, we might think they provide a better explanation of why the effect occurred than the power-manifestation explanation, and renounce the latter. In some cases, this certainly happens. But, the general framework I have presupposed here – which includes both realism about powers and anti-reductivism about the mental – rules out that this can happen for all explanations in terms of the manifestation of psychological powers.
132
E. Mayr
which the performance of an action of type X supervenes. Then, as long as some of the features which make the microphysical event necessary cannot be combined as explanantia within one explanation with the ability’s manifestation, the explanation of the action in terms of the ability’s manifestation will be an indeterministic one. (And remember: such a combination will not be possible unless these features can be conceived as initiating or interacting with the ability’s manifestation. It is not sufficient that they interact merely with some of the underlying microphysical states or processes.47) As far as this explanation is concerned, the agent ‘could have done otherwise’. This, then, is the general idea which I will try to flesh out a bit more in the remainder of the paper, thereby motivating claims (1) to (3),48 which I set out in the introduction. Take an intentional action performed for a reason, and ask whether the agent could have acted otherwise. Since this action was performed in the light of a perceived reason, it must necessarily have issued from the ability for rational selfguidance, i.e. the ability to deliberate, decide and act in the light of perceived reasons. In order to even understand it as an intentional action performed for a reason, we must (at least implicity) understand it as having issued from this ability. This gives an explanation of the action in terms, inter alia, of this ability’s exercise the best claim for being the salient explanation of the action and the one we are first and foremost interested in when we understand it as an intentional action performed for reasons. For we are, by so understanding it, committed to such an explanation being available, while we are not, just by that, equally committed to the availability of any other kind of explanation. By consequence, the former explanation has the best prima facie claim to setting the framework for determining which sense of ‘could have done otherwise’ is relevant for the free will debate (at least as far as intentional actions performed for reasons are concerned). This gives us claim (1). Within this salient explanation, more features than just the agent’s ability for rational self-guidance can be cited, e.g. the reasons the agent perceived or the different courses of action she considered as open to her. But two things must be noted: First, if what we have said earlier is correct, the lower level phenomena underlying the exercise of the agent’s ability for rational guidance cannot be similarly cited in this explanation. Second, on the face of it, the exercise of the ability for rational guidance can lead to different results even when the external circumstances as well as the reasons the agent perceives and other features which can figure within this explanation are fixed and when there is no outside interference with the exercise. Take a case of deliberation where the agent is aware of different 47
This requirement does not necessarily mean that no microphysical state can interact with an ability’s manifestation in the right way: some may still do so and therefore still be capable of figuring in the (or a) salient explanation of an action. But in order to avoid the implication from microphysical determinism to the general availability of deterministic psychological-cum-microphysical explanations, it is enough that at least a significant group of microphysical states are excluded from this. 48 Vide p. 117f. supra.
7 Freedom of the Will and Rational Abilities
133
relevant reasons which speak in favour of two options she takes to be open to her. To make a decision, she must assess the comparative weights of these considerations. How she weighs them need not be pre-determined by her antecedent perception of these reasons or the external circumstances of the action. She may have to settle for herself which reasons she regards as weightier, or she may change a previous assessment or think that both sets are incommensurable such that she must simply choose which course of action to take. In such cases, different outcomes can equally be results of the exercise of her ability for rational self-guidance on this occasion, even though there is no outside interference with the deliberation process. So, at least prima facie, the salient explanation in terms of the agent’s rational self-guidance is often indeterministic. Since this explanation is the salient one for intentional actions performed for reasons, we therefore have some reason to hold that the agent ‘could do otherwise’, in the sense which is relevant for determining what was possible for her as an agent. Of course, the prima facie impression may be mistaken: It might turn out that the explanation is really deterministic after all (because, e.g., all our reasons have a fixed importance for us and the relative weights determine in advance how we will decide and act). But whether this is the case is the crucial question for deciding whether the agent can act otherwise in the relevant sense. This is claim (2). But is this possibility of doing otherwise not illusory in a world with microphysical determinism, even if we focus just on the salient explanation in terms of the agent’s ability for rational guidance? For how can any explanation in such terms be indeterministic if we assume that determinism reigns on the microphysical level and that every psychological and agential state and event supervenes on a microphysical state or event? It can be because not all the microphysical phenomena which are relevant for determination at some level must ‘reverberate’ or make a difference at the personal level of action-explanation, and some may not ‘show up’ in explanations which cite the ability for rational self-guidance. Supervenience of the mental on the microphysical implies that we have no mental differences without microphysical differences; it doesn’t imply that, conversely, every microphysical difference is reflected on the mental level, too.49 One case in point is List’s case of the multiple realizability of psychological states, which can be applied in the same way to the possession of the ability of rational self-guidance, to the presence of an opportunity for its exercise, to the perception of reasons etc. But when we look at cases of power-manifestations, there are further possibilities: certain microphysical states and occurrences may not be (part of) any subvenient states and events for psychological phenomena and may not interact with the manifestation of any ability of the agent. Imagine that the agent’s possession of the ability of rational self-guidance supervenes on the microstates A, ... D. Assume further that the presence of an opportunity for the exercise of this ability and the perception of a set of reasons similarly supervenes on microstates E, . . .G, and that any masking factor for the ability would have to supervene on microstates H, . . . J (which are not present). 49
Vide p. 122 supra.
134
E. Mayr
Assume further that, on the level of the microstates, determinism is true, such that the presence of A, . . . D and E, . . . G, combined with the absence of H, . . . J makes it necessary, given the circumstances, that microstate K occurs, which, in turn is the subvenient state for the action L* of the agent. Does this mean that the agent’s exercise of her ability for rational guidance, given the presence of an opportunity for its exercise, her perception of reasons, and the absence of maskers, must issue in action L*? No: The determination on the microlevel may only be due to the presence of some further factors M . . . . P, which together with A, . . . D, and E, . . . G deterministically cause L* (given the absence of H, . . . J). If those further events M, . . . P did not occur, L* would not be causally necessitated. Now assume that the occurrence of M, . . . P is not correlated with the possession of the agent’s ability, nor with the occurrence of states and events which can be identified as (partly constituting) opportunities or the absence of maskers for the exercise of the rational ability etc. In other words, M, . . . P is not correlated to any factor, which would be appropriately related to the exercise of the rational capacity and which we could cite, within one explanation, alongside the ability-manifestation. In this case, there is no deterministic explanation of why the agent acted as she did which includes the exercise of the agent’s ability for rational self-guidance. Thus, the salient explanation of the agent’s action, which has the best prima facie claim for determining the sense of ‘could have done otherwise’ relevant for free will, is indeterministic. This gives us claim (3).
7.5
Conclusion
In this paper, I have examined one way in which conceiving of free will in terms of rational abilities brings compatibilists a genuine advantage. Especially compatibilists attracted to the idea that the compatibility of agential freedom and microphysical determinism can be defended by distinguishing between different, mutually irreducible levels of explanation, I have argued, will find that an abilities account allows them to make a much better case that, even if microphysical determinism is true, agents ‘can do otherwise’. For such an account provides the grounds for a principled argument why we can and should disregard certain microphysical features when providing the kind of explanation which is salient for explaining intentional actions. This later explanation, though, has the best prima facie claim to setting the framework for determining which sense of ‘could have done otherwise’ is relevant for the free will debate (at least as far as intentional actions performed for reasons are concerned).50
50
For very helpful comments on earlier version of the text I am indebted to Dorothee Bleisch, Stefan Brandt, Christian Kietzmann, Leo Menges and Konstantin Weber, and an anonymous referee.
7 Freedom of the Will and Rational Abilities
135
References Alvarez, M. (2013). Agency and two-way powers. Proceedings of the Aristotelian Society, 113, 101–121. Bok, H. (1998). Freedom and Responsibility. Princeton UP. Dennett, D. (2004). Freedom evolves. Penguin. Fara, M. (2008). Masked abilities and compatibilism. Mind, 117, 843–865. Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: A theory of moral responsibility. CUP. Frankfurt, H. (1969). Alternate possibilities and moral responsibility. Journal of Philosophy, 66(23), 829–839. Jaster, R., & Vetter, B. (2017). Dispositional accounts of abilities. Philosophical Compass. https:// doi.org/10.1111/phc3.12432 Kenny, A. (1976). Freedom, will and power. Blackwell. Kenny, A. (1978). Free will and responsibility. Routledge. Kim, J. (1996). Philosophy of mind. Westview Press. List, C. (2014). Free will, determinism, and the possibility of doing otherwise. Nous, 48(1), 156–178. List, C. (2019). Why free will is real. Harvard University Press. List, C., & Menzies, P. (2009). Nonreductive physicalism and the limits of the exclusion principle. The Journal of Philosophy, 106(9), 475–502. Melden, A. (1961). Free action. Routledge & Keegan Paul Ltd.. Menges, L. (2021). Free will, determinism, and the right levels of description. Philosophical Explorations. https://doi.org/10.1080/13869795.2021.1937679 Molnar, G. (2003). Powers: A study in metaphysics, S. Mumford (Ed.). OUP. Raz, J. (1999). Engaging reason. OUP. Ryle, G. (1949). The concept of mind. Penguin (repr. 1990). Smith, M. (2003). Rational capacities, repr. in Smith, Ethics and the a priori (pp. 114–135). CUP, 2004. Vihvelin, K. (2004). Free will demystified: A dispositional account. Philosophical Topics, 32(1/2), 427–450. Wallace, J. (1994). Responsibility and the moral sentiments. Harvard UP. Whittle, A. (2010). Dispositional abilities. Philosophers’ Imprint 10. Yablo, S. (1992). Mental causation. Philosophical Review, 101(2), 245–280.
Chapter 8
The Power to Will Freely: How to Re-Think About the Problem of Free Will Without Laws of Nature Daniel D. De Haan
Abstract The problem of free will is ubiquitously articulated in terms of the (in)compatibility of free will with laws of nature. Peter van Inwagen, David Lewis, Robert Kane, and nearly all other contributors to the problem of free will, conceive it as a potential conflict between the laws of nature and free will. Despite dispositional defenses of compatibilism, the putatively dispositional principle of alternative possibilities among defenders of free will, as well as increased sympathy for dispositional and agent-causation versions of source incompatibilist free will, the standard debates remain beholden to a nomological definition of the free will problem. This is surprising since one of the upshots of the causal powers revolution has been the overthrow of laws of nature. Many exponents of the metaphysics of powers either reject laws of nature altogether, or regard them as abstract descriptions of the real work performed by causal powers. What remains unexplored for dispositionalists is to address the question: how to re-think about the problem of free will without the laws of nature? That is the aim of my essay. I will argue the shift to powers requires not only major revisions to the definitions of determinism, compatibilism, etc. put forth by van Inwagen and others, but also a re-thinking of the power of free will itself in light its manifestations. Keywords Free will · Laws of nature · Causal powers · Determinism Are we free to break the laws of nature? This question encapsulates what the standard approaches to the problem of free will take for granted in how they think about different conceptions of free will that are incompatible or compatible with a conception of determinism underwritten by inviolable laws of nature. Despite dispositional defenses of compatibilism, the putatively dispositional principle of alternative possibilities among defenders of free will, as well as increased sympathy for dispositional and agent-causation versions of source incompatibilist free will, the standard debates remain beholden to a nomological definition of the free will D. D. De Haan (*) Blackfriars and Campion Hall, University of Oxford, Oxford, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_8
137
138
D. D. De Haan
problem. This is surprising since one of the upshots of the causal powers revolution has been the overthrow of laws of nature. Many exponents of the metaphysics of powers either reject laws of nature altogether, or regard them as abstract descriptions of the real work performed by causal powers (Mumford & Anjum, 2011; Cartwright, 1999, 2004). Hence, for exponents of power realism nomological antirealism (¼PRNA), the opening question, and the basic assumptions of the standard approach to free will it presupposes, are problematic. They hold there are no laws of nature for free or other agents to break. The question advocates of PRNA must address is: How to re-think about the problem of free will without the laws of nature? That is my aim here. I shall show why a re-thinking of the problem of free will is necessary for anyone who holds PRNA. First, I delineate the central positions that comprise PRNA, which I presume as my point of departure. Second, I introduce van Inwagen’s avuncular advice on how to think about the problem of free will and use his definitions as a foil for re-conceptualizing notions of free will and determinism compatible with PRNA. Third, I distinguish one version of power-indeterminism and two forms of powerdeterminism. Fourth, I then distinguish three versions of power-free-will. Fifth, I conclude with an overview of the versions of compatibilism and incompatiblism these PRNA notions of free will and determinism generate. My overall goal is to survey this new PRNA territory and identify the distinct contours of its free will problem. I leave for future settling the task of developing these positions and prosecuting its debates.
8.1 8.1.1
Power Realism & Nomological Antirealism The Ontological Reality of Causal Powers
By “power realism” I mean a family of views which acknowledge that powers are basic, irreducible, and indispensable denizens of our ontology, and that powers perform all the causal work required such that laws of nature are ontologically otiose. Of course power realism and nomological antirealism need not go together (Dumsday, 2019), but the views I am concerned with do hold power realism reveals the ontological redundancy of laws of nature. I join the ranks of Ellis, Mumford, Anjum, Groff, Steward, Mayr, and others who have shown that to endorse power realism is to reject in toto the Humean presumptions that have continued to infelicitously influence and constrain the problem of free will – even by exponents of agent causation (e.g., O’Connor, 2000). The family of power realist positions I have in mind are more Aristotelian in inspiration (e.g., Marmodoro, 2017; Groff & Greco, 2013; Groff, 2012a, 2012b; Steward, 2012; Mayr, 2011; Martin, 2008; Jacobs, 2017; Cartwright, 2019; Jaworski, 2016; Mumford & Anjum, 2011). For my purposes here, no significant difference hangs on the terminology of “powers,” “dispositions,” or “capacities,” which I use interchangeably.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
8.1.2
139
Power Realism Rejects Categoricalism
Power realism entails the rejection of categoricalism, which holds fundamental properties are non-dispositional, that is, powers are not fundamental. My arguments here can remain neutral on the power realist debate among pandispositionalists, who hold the only fundamental properties are powers, dualists who hold that categorical and dispositional properties are two different kinds of fundamental properties, and identity theorists, who hold there is one kind of fundamental property that has categorical and dispositional aspects.
8.1.3
Real Causation by the Mutual Manifesting Powers of Powerful Particulars
Power realism expounds a realist theory of causation which consists in the simultaneous and reciprocal manifestations among power partners. Contrary to neo-Humean conceptions of causation as temporally successive independent events, causation “is not a matter of two events, but of one and the same event—a reciprocal dispositional partnering as a mutual manifesting” (Martin, 2008, 46; Marmodoro, 2017; Mumford & Anjum, 2011). Powers are what causes what. Insofar it matters to free will, I share Marmodoro’s neo-Aristotelian positions (a) that the causal interaction of mutually manifesting powers, at the very least, comprises different types of activities, namely, active and passive causal roles, if not the different activities of active and passive causal powers. And, (b) “the activation of a power is an internal ‘transition’ from one state to another of the very same power: its manifestation is not the occurrence of a new power; rather it is simply a different state of the original power: an activated state” (Marmodoro, 2017, 59). Additionally, it is powerful particulars or substances that are the agents of action and passion in virtue of the manifestation of their active and passive powers. Following Helen Steward, I distinguish between the realization of one-way powers from the exercise of two-way powers. Agents’ one-way powers are not exercised but are realized by manifesting. An agent exercises its two-way power by φ-ing or refraining from φ-ing, that is, by manifesting one of the two ways in which its two-way power to φ or not to φ can be manifested given the satisfaction of its necessary conditions or opportunities for manifesting (Steward, 2012, 155–156).
8.1.4
Defeasibility of Manifesting Powers
Power realism explains the defeasibility of occurrences or manifestations of powers on account of subtractive and additive interferences, finks, blockers, antidotes, absences, and so forth (Martin, 2008; Mumford & Anjum, 2011; Groff & Greco,
140
D. D. De Haan
2013). A fragile vase will shatter if it falls to the floor, unless its fall is broken by a soft cushion. “In cases of causation, then, some further natural process could always interfere and often does: what actually occurs will typically be a result of many dispositions acting together.” (Mumford, 2018, 213) A power’s activity can be overpowered by more powerful powers, and even co-manifesting powers remain essentially liable to influence and interference.
8.1.5
Causal Pluralism
Power realism thereby endorses causal pluralism since there can be different kinds of powers, which are identified by their (a) different activities or active and passive causal roles, (b) conditions or opportunities for manifestation, (c) along with their different mutual manifestation partners. Causal pluralism also reveals that many forms of causal overdetermination are common and unproblematic insofar as additional causal powers – including mental causal powers like free will – are differently manifesting, contributing, and influencing the outcome of a nexus of reciprocally manifesting powers.1 Capacities have a canonical way of acting, which is to be distinguished from what happens when they act. For each capacity, there is a prescribed set of ways in which it can act. When the capacity ‘gravity’ acts, it pulls, no matter what happens to the object on which it pulls. What actually happens [i.e., the outcome or result,] depends on what other powers gravity cooperates with in the circumstances and what the arrangements are. When the arrangements are right, the activities of the powers give rise to regular behaviors, as in the orbits of the planets around the sun, or the browning of bread in a toaster, or the expulsion of magnetic fields in a superconductor. (Cartwright, 2019, ch. 2, my emphasis)
Insofar as there are regularities in nature, they are due to the regularities of the co-manifestations of power partners. This brings us to nomological antirealism.
8.1.6
Why Causal Powers Deliver Nomological Antirealism
Many exponents of a realist causal powers ontology argue powers entail the ontological redundancy if not irrelevance of laws of nature. Some hold that “it is not the laws that are fundamental, but rather the capacities.. . . Whatever associations occur in nature arise as a consequence of the actions of these more fundamental capacities. In a sense, there are no laws of association at all. They are epiphenomena.” (Cartwright, 1989, 181) Technically, if laws are epiphenomena, then they are real, but they have no causal or governing role to play in reality. Nomological antirealism, 1
On overdetermination and mental causation, see Steward, 2012; Jaworski, 2016; Groff, 2012a. For nuanced points about the “contributions” of powers without reference to overdetermination, see Cartwright, 2019, ch. 2.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
141
strictly speaking, goes beyond epiphenomenalism by arguing that because laws of nature are not causally efficacious and do not govern anything in reality they are not real. Brian Ellis holds the laws of nature are not things in the world, but are general propositions descriptive of the kinds of natural necessities that exist in it. They are true, have truthmakers, bear logical relationships to one another, and so on, just like other kinds of true propositions. But they are not items that should occur in anyone’s ontology. I am not a Humean. But I am in full agreement with the Humeans on this point. The laws of nature are descriptive of reality, not prescriptive of it. They describe the ways that things of various kinds must be or behave in the various kinds of circumstances in which they may find themselves. (French et al., 2006, 439)
Mumford presents arguments that go one step beyond this epistemological recognition of laws of nature to advocate for their elimination. If “governing” is essential to the concept of a law, and laws of nature completely fail to govern, then there does not seem to be anything that distinguishes their reduction from their elimination (Mumford, 2004, 121, 2018; Demarest, 2017).2 For our purposes, “nomological antirealism” denotes a family of views which agree that, insofar as we continue to speak about laws of nature, these laws either are nothing more than abstract descriptions of regularities among co-manifesting power partners, or are epistemologically dispensable.
8.1.7
Cartwright on the Laws of Nature
Cartwright and others have presented additional arguments for demoting laws of nature to a mere epistemic status. They have shown that the reason why laws of nature require a seemingly endless iteration of ceteris paribus clauses to cover all the exceptions to what many credulously believe are universal and exceptionless regularities, is because these “laws” are derived from and only accurately apply to highly restricted – and often experimentally designed – arrangements of phenomena. For “the laws of physics apply only where its models fit, and that, apparently, includes only a very limited range of circumstances.” (Cartwright, 1999, 4, 2004, 2019). In sum, laws of nature are not universal and exceptionless, they do not actually govern any occurrences in nature, they are only descriptive not prescriptive, and they are not real. Nature is dappled with a diverse distribution of powers which mutually manifest and influence each other in myriad ways and it is these powers that are the principal focus of scientific experimentation and theorization. Scientific enquiries actually disclose nature to be the lawless order of mutually manifesting, influencing, and interfering powers of powerful particulars. There are many more topics concerning PRNA that are relevant to free will, but this rough summary of presumptions will suffice.
2 For an account of laws of nature that defines them in terms of constraining rather than directing, see Steward, 2022, this volume.
142
8.2
D. D. De Haan
Re-Thinking the Free Will Problem
So far I have introduced the major positions of PRNA being presumed here which have been extensively debated and defended elsewhere. We must now address why PRNA is incompatible with the standard conception of the free will problem. Peter van Inwagen’s proposal for how to think about the problem of free will remains representative of the standard approach at odds with PRNA. After presenting the “free will problem” and how to define its terms, van Inwagen provides the following definitions. These definitions serve to explain the system of concepts everyone who thinks about the free-will problem should use – or so I say. 1. The free-will thesis is the thesis that we are sometimes in the following position with respect to a contemplated future act: we simultaneously have both the following abilities: the ability to perform that act and the ability to refrain from performing that act (This entails that we have been in the following position: for something we did do, we were at some point prior to our doing it able to refrain from doing it, able not to do it). 2. Determinism is the thesis that the past and the laws of nature together determine, at every moment, a unique future (The denial of determinism is indeterminism). 3. Compatibilism is the thesis that determinism and the free-will thesis could both be true (And incompatibilism is the denial of compatibilism). 4. Libertarianism is the conjunction of the free-will thesis and incompatibilism (Libertarianism thus entails indeterminism). 5. Hard determinism is the conjunction of determinism and incompatibilism (Hard determinism thus entails the denial of the free-will thesis). 6. Soft determinism is the conjunction of determinism and the free-will thesis (Soft determinism thus entails compatibilism). I must emphasize that I intend these definitions to be entirely neutral as regards competing positions about the relations between free will and determinism and entirely neutral as regards competing accounts of the nature of free will. In principle, of course, as a matter of logical theory, a definition cannot favor one thesis over its logical contraries. . . I contend that my definitions are not tendentious. (van Inwagen, 2008, 329–330). Notice first that van Inwagen, like us, is not concerned with logical determinism; he is presenting a nomological conception of physical determinism. In his attempt to present neutral definitions van Inwagen remains studiously silent here on the nature of those abilities for free will and on the laws of nature.3 The dispute between incompatibilists and compatibilists typically concerns how to fill out the “abilities” required for free will, and later we will explore three different ways exponents of PRNA can delineate these abilities. What is meant by laws of nature is assumed to be less controversial, but determinism does require a demanding notion of laws of 3
For the details of his own view, see van Inwagen, 2017, chs. 1 & 14.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
143
nature in order to make intelligible the disagreement between incompatibilists and compatibilists. The most straightforward proposals define laws of nature as deterministic (rather than probabilistic) and exceptionless regularities that are all encompassing (rather than confined to some limited domain of reality). One could endorse laws of nature that weakened any of these provisos, but doing so would either thereby take the bite out of determinism and its purported tensions with libertarianism, or require introducing additional contentions to shore up any revised soft laws of nature that were either probabilistic, or admitted of exceptions, or were confined to limited domains. Furthermore, in order to secure determinism, it seems the laws of nature must actually govern or cause all natural phenomena such that free will is excluded from breaking the laws that govern nature. This is why many have thought laws of nature govern by being instantiated by causal factors, thereby combining nomological determinism with causal determinism. In his introduction to the “historical free will debate,” Derk Pereboom presents nomological and theological accounts of causal determinism that focus on “free will and actions being causally determined by factors beyond the agent’s control.” An action will be causally determined in this way if causally relevant factors occurring or active prior to the agent’s coming to be, and thus beyond her control, together with facts about the laws of nature, also beyond her control, ensure the occurrence of the action by a causal process that begins with those preceding causal factors and ends in the occurrence of the action. An action will also be causally determined by factors beyond the agent’s control if its occurrence is ensured by a causal process that originates in God’s timeless willing and ends in the occurrence of the action. (Pereboom, 2014, 1)
Anti-Humean nomological determinism can be defined without introducing causation, but we are still owed an explanation for how the laws of nature govern phenomena. Humean nomological determinists, however, contend that the laws of nature do not govern (Beebee, 2000; Demarest, 2017). We need not enter into these internecine debates between exponents of laws of nature here as we are presuming nomological antirealism, which gets us to the heart of the matter. To justify the neutrality of his definitions of the free will problem, van Inwagen points out that the “finest essay that has ever been written in defense of compatibilism—possibly the finest essay that has ever been written about any aspect of the free-will problem—, David Lewis’s ‘Are We Free to Break the Laws?’, opens with definitions of ‘determinism,’ ‘soft determinism,’ ‘hard determinism,’ and ‘compatibilism’ that are equivalent to the definitions I have set out.” (van Inwagen, 2008, 330; Lewis, 1981) For exponents of PRNA, the obvious difficulty with van Inwagen’s definitions is his assumption that the existence of laws of nature which can determine phenomena is uncontroversial, but this is precisely what PRNA rejects. For PRNA, the causal commerce of the world is not governed by deterministic laws of nature but by the defeasible manifestations of causal powers, by probabilistic powers, or simply by hap. And among these powers there might be natural powers for free will whose manifestations do not break any laws of nature because there are no laws to violate. By liberating free will from the constraints of laws of nature, PRNA allows us to radically re-think the problem of free will.
144
D. D. De Haan
So far we have seen that van Inwagen completely overlooks the possibility of nomological antirealism, and so, by his own criterion, fails to provide neutral definitions – an ideal that might not be achievable when dealing with rival paradigms – for how to think about the problem of free will. While there is no universal consensus on the correctness of van Inwagen’s or Lewis’s definitions of the free will problematic, the standard definitions – even of so-called “revisionists” – do not systematically diverge from nomological conceptions of the problematic. This oversight is significant for it discloses the extent to which participants within the standard problem of free will have failed to recognize that their very conception of the debate over free will and determinism is in the grips of a post-Humean tradition of enquiry that takes for granted a powerless ontology and nomological worldview. What is often thought to be the least tendentious definition for the free will problem, namely, that of nomological determinism, is from the point of view of PRNA one of the most contentious features assumed by the standard account of the problem of free will. If determinism requires a unique future that is fixed by the past + “p,” whatever “p” is it cannot be laws of nature for exponents of PRNA. This is why the standard nomological conceptions of determinism and free will are a dead end for PRNA. What conceptions of determinism and free will are compatible with PRNA? I shall start with a version of PRNA that is essentially incompatible with determinism, before considering two conceptions of power-determinism compatible with PRNA.
8.3 8.3.1
Indeterminisms & Determinisms Compatible with PRNA Dispositional Modality Incompatibilism
Mumford and Anjum are among the most noteworthy exponents of PRNA. They also argue it entails a radical revision of the free will problem (Mumford & Anjum, 2014, 2015a, 2015b). Their critique of the standard problem of free will starts from their rejection of the modal dualism of pure contingency and pure necessity in favor of a middle way account of powers and their tendencies to manifest. This middle way is their “dispositional modality” view (¼DMV) of powers. The problem with pure necessity is that it excludes the “Principle of Alternate Possibilities” (¼PAP), which libertarian event-causationists argue is required if free agents are genuinely able to φ or not to φ at t. The rejection of deterministic-necessity leaves us with the pure contingency of indeterminism, but many argue indeterminism is also problematic for free will. This is because if there is pure contingency, then agents fail to have the exigent control over their actions which would exclude the “Principle of Ultimate Authorship,” which libertarian agent-causationists argue is key to free will. Free will seems to require both of these principles but modal dualism forces theorists with an unacceptable modal choice. Mumford and Anjum argue that the dispositional
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
145
modality proprietary to powers can accommodate both PAP and the principle of ultimate authorship. If Mumford and Anjum are correct, it would show the superiority of their view over the one-sidedness of the major positions defending free will. The basis of the dispositional modality is the thought that causal powers are essentially capable of prevention and interference. . ., which shows that they do not necessitate their effects or manifestations. (Mumford & Anjum, 2014, 22)
Mumford and Anjum argue that powers exhibit a sui generis dispositional modality that is revealed by the abiding possibility for interference. Even though powers tend toward their manifestations, the possibility of interference entails that the manifestation of a power is never necessitated. “When a power is in all the right conditions for its manifestation, it still ‘only’ tends, and no more than tends, to produce its effect. . .” (Mumford, 2014, 112). They contend DMV can secure PAP, which underwrites freedom, insofar as all powers – not just the power of will – realize causal manifestations which can always be interfered with and only tend toward their manifestations. Hence, alternative possibilities are omnipresent. The realization of powers in alternative ways does not involve an ad hoc mode of causation that requires stepping “outside the laws of nature to be free. Causation in general is consistent with alternative possibilities, since no causal set up offers only one possible effect. Any power can be overcome by counteracting powers, including those of the agent.” (Mumford & Anjum, 2015a, 7–8, 2015b) Against objections that this kind of indeterminism or pure contingency undermines the control required by free will and the principle of ultimate authorship, Mumford and Anjum respond that “If an agent has a power to F, and succeeds in performing F through exercise of this power, then the agent has ultimate authorship of F even if those actions did not produce F by necessitating it.” (Mumford & Anjum, 2015a, 2015b, 8). Mumford and Anjum also propose a “new argument” for free will based on DMV. 1. If causal determinism is true, all events are necessitated. 2. If all events are necessitated, there are no powers. 3. Free will consists in the exercise of an agent’s causal powers. Therefore, if causal determinism is true, there is no free will. (Mumford & Anjum, 2014, 21) They argue that “with the correct theories of powers and of free will, a powersbased solution should come down firmly on the side of incompatibilism. Free will as a power . . . should commit one to incompatibilism” (2014, 20–21). DMV excludes causal determinism since all causation is dispositional causation, and all dispositional causes are manifested according to a dispositional modality that excludes any deterministic necessity. What is surprising is the implicit conclusion implied by the first two premises, namely, that power realism is incompatible with causal determinism independent of the question of free will. Let us consider a few objections to DMV and its version of free will incompatibilism.
146
D. D. De Haan
The first challenge arises from within the power realist family. Many power realists endorse Aristotelian conditional necessity for the manifestations of powers and so reject premise (2). If all the conditions for the manifestation of any power obtain and no interferences are present, then its manifestation occurs necessarily.4 But Mumford and Anjum present arguments for rejecting conditional necessity and defend premise (2) by appealing to DMV. Marmodoro has recently shown that their arguments against conditional necessity miss their target (Marmodoro, 2016). For instance, Mumford objects that the “the very idea of necessity seems to mean unconditional necessity” (Mumford, 2014, 111). But Marmodoro points out that this conflates the physical necessity in the actual world that pertains to conditional necessity with metaphysical necessity in all possible worlds. Contrary to Mumford and Anjum, good old-fashioned Aristotelian conditional necessity nicely captures both the contingency of conditions required for the manifestations of powers and the necessity of their manifestations whenever they obtain and interferences are absent. Powers are not always manifesting because the conditions they depend on for their actualization are not always satisfied and even then manifestations can be interfered with. But insofar as their conditions for manifesting are actualized without interferences, powers necessarily manifest. Marmodoro’s numerous arguments demonstrate that Mumford and Anjum have not prosecuted a convincing argument against the default Aristotelian notion of conditional necessity, which can effortlessly accommodate all of the ways interferences can prevent or modify the manifestations of powers, without endorsing DMV’s peculiar position. Marmodoro also draws attention to serious problems with DMV which arise precisely because it demurs conditional necessity. Insisting that powers only tend towards their manifestations, but do not necessarily manifest, even when all their conditions obtain and interferences are absent, is a weakness, not a strength of DMV. This is because the manifestation of any disposition becomes ultimately entirely arbitrary and inexplicable. Even if Mumford and Anjum are correct that DMV is not equivalent to pure indeterminism insofar as powers tend towards certain manifestations and not others, and that probabilities presume dispositional modalities, it nevertheless remains the case that, for DMV, the manifestation of a power is always inherently contingent. This is because DMV cannot exclude the uncanny possibility that a power might tend towards a certain manifestation, have all of its conditions obtain with no actual interferences, and yet the power never manifests. How could we identify DMV powers that never manifest? Indeterminism, even if mitigated, is intrinsic to DMV’s account of powers. While this is certainly a problem for DMV’s general account of powers, it is especially troublesome for DMV free will and the exigency to secure the control required for the principle of ultimate authorship. This is because, for DMV, agents cannot actually control whether or not their powers will in fact manifest. So if I attempt to exercise my power of free will it can always fail to 4
While I do not have space to address it here, it is noteworthy that some exponents of power realism – without endorsing DMV – hold only some powers, but not all, manifest according to conditional necessity. These other powers might manifest spontaneously; their manifestations might tend to occur or be more likely given their conditions. See Lowe, 2008; Mayr, 2022, this volume. Some of the criticisms leveled against DMV would also apply to these views.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
147
manifest, not because of any unsatisfied conditions or interferences, but simply because dispositional modalities are defeasible in their own right and so they can fail to manifest arbitrarily. Hence, Mumford and Anjum’s DMV does not secure the control over the exercise of free will required for the principle of ultimate authorship. On a different front, Penelope Mackie argues that Mumford and Anjum fail to demonstrate that powers – independent of the question of free will – are incompatible with determinism (Mackie, 2014). It is not my aim to adjudicate this debate between Mackie and Mumford and Anjum, however, it seems that Mackie does not notice that DMV’s rejection of conditional necessity in favor of the inherent indeterminism of dispositional modalities essentially rules out determinism. What Mackie’s arguments for the compatibility of powers and determinism do illuminate is how to formulate versions of determinism compatible with versions of power realism that endorse conditional necessity.
8.3.2
Determinisms Compatible with PRNA
In order to formulate an account of determinism that is, contrary Mumford and Anjum, compatible with power realism, Mackie looks to fixed background conditions that determine the manifestations of all powers, but without exploring in detail different versions of power-determinism compatible with PRNA. Since there are no universal exceptionless laws of nature for PRNA, what factors could ensure that all of the necessary conditions for the co-manifestations of all powers are necessitated? Universal power-determinism requires fixed background conditions that somehow rule in the obtaining of all necessary conditions for the manifestations of powers while ruling out all interferences to these powers’ manifestations. But it is important to recognize that, as C.B. Martin pointed out, our common way of talking about background conditions is misleading “because so-called background conditions are every bit as operative as the identified dispositional entity” (Martin, 2008, 50). In other words, the fixed background conditions for PRNA will not be laws of nature but other powers with deterministic manifestations. Which powers are these? Unlike laws of nature realism, which naturally gravitates towards determinism, PRNA is not inherently deterministic. It matters which kinds of powers exist. Defending powerdeterminism therefore turns on establishing that only certain kinds of powers exist, namely, those with necessary and one-way manifestations, even if they be conditionally necessary. At least three different kinds of powers are excluded from the cosmos of universal power-determinism: DMV powers, spontaneous powers,5 and two-way powers. The first two kinds of powers are excluded for reasons we have already covered, namely,
E. J. Lowe argues that the will and “radium’s spontaneous power to undergo radioactive decay” are two kinds of “spontaneous powers” and the latter does not have any “non-deterministic or probabilistic cause” (Lowe, 2008, 50).
5
148
D. D. De Haan
even when their conditions obtain and no interferences are present neither DMV nor spontaneous powers necessarily manifest. Two-way powers are excluded for a similar reason. This is because even two-way powers that manifest necessarily when their conditions obtain and interferences are absent, nevertheless differ from one-way powers insofar as the agent that exercises its two-way power can exercise it in more than one way. So, unlike one-way powers, two-way powers’ manifestations are not fixed to any single outcome in virtue of its nexus of co-manifesting powers. Hence, any universe that contains either DMV, spontaneous, or two-way powers is one that is liable to a PRNA kind of indeterminism given these powers’ erratic manifestations, failures to manifest, or multiple ways to manifest. Since any one of these kinds of manifestations could prevent or modulate co-manifesting powers and so alter the settling of the resulting outcome. Hence, if the “core idea in determinism is fixity of the future by the past” (Mumford & Anjum, 2011: 75), then the existence of such powers would undermine universal power-determinism. Given these caveats, there are at least two basic forms of power-determinism compatible with PRNA. de Facto PD The only manifesting powers that happen to exist are one-way powers that manifest via conditional necessity. The total state of co-manifesting powers at t1 fixes the total state of all future co-manifesting powers at tn. de Jure PD The only manifesting powers that can exist are one-way powers that manifest via conditional necessity. The total state of co-manifesting powers at t1 fixes the total state of all future co-manifesting powers at tn. It seems the background conditions necessary and sufficient for securing universal power-determinism compatible with PRNA requires a cosmos that only contains one-way powers that manifest according to conditional necessity.6 This is because in such a universe all of the co-manifestations and interferences within the nexus of future manifesting powers are fixed by the co-manifestations and interferences of prior power partners whose necessary conditions have obtained. The total state of the world at t1 consists of all the co-manifesting one-way powers whose necessary conditions obtain, and since these co-manifesting powers only manifest in one way, the co-manifestation nexus of partners powers at t1 fix which necessary conditions will and will not obtain for all future co-manifesting powers.7 There are some important objections that these versions of power-determinism face. First, even if each version of power-determinism appears to leave intact the basic essential features of powers presumed by PRNA (realism, pluralism, defeasible
6
It is possible that a deterministic universe might exist which contained other kinds of powers, like spontaneous or two-way powers, but these powers never manifested either inexplicably or because they were systematically prevented from manifesting. But, as we will see, there are inherent difficulties positing the existence of such powers, and even if this could be developed as a viable model of power-determinism, it would be a variation on de facto power-determinism. 7 I am extraordinarily grateful to Simon Kittle for his constructive criticisms of an earlier version of this section.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
149
manifestations), can these features of powers be identified if determinism is true? Each version needs to explain why it does not undercut the very arguments employed by PRNA to establish these basic essential features of powers. If powerdeterminism is true, then all the powers that manifest are those whose necessary conditions do not fail to obtain and their possible interferences are excluded. How then could we discover that powers have defeasible manifestations? What evidential grounds remain in place for distinguishing actual interferences from the necessary conditions of a power’s manifestations? Second, even if each of these forms of determinism is coherent and compatible with PRNA, on what grounds could we establish the truth of either one of them? A passage from Mumford and Anjum nicely sets in relief the difficulty facing powerdeterminism. What delivers the necessity . . . is that, somehow, everything got fixed. That will include the fixedness of all the background conditions—including which dispositions do, and which do not, act to produce the necessitated outcome—but it was not those powers that necessitated that outcome. (Mumford & Anjum, 2011, 179)
Power-determinism maintains the only powers that exist are one-way powers that manifest according to conditional necessity. Critics of power determinism have both a posteriori and a priori ways for challenging this contention. Some might object that the thesis of power-determinism cannot be an empirical truth discoverable by scientific enquiry. Others, like Lowe and Mumford have argued, with great plausibility, that our best physics seems to disclose the existence of powers that do not manifest according to conditional necessity. Of course, most power-determinists would reject these arguments from Lowe and Mumford and would argue instead that there are other ways to interpret these empirical phenomena which fit the model of one-way powers that manifest via conditional necessity. But even if powerdeterminists can resist these particular arguments from empirical phenomena, this would not establish the truth of either form of universal power-determinism. Consequently, it seems the only justification for de facto power-determinism is it is just a brute fact about the universe that the only powers which exist are those that manifest by conditional necessity. So any form of a posteriori naturalism leaves de facto power-determinism without any explanatory justification for holding it. Unlike de facto power-determinism, de jure power-determinism claims to be justified by cogent a priori arguments. There are two different routes these a priori arguments might take. The first route would require demonstrating that DMV powers, spontaneous powers, two-way powers, and any other alternatives to conditional necessity one-way powers, are each incoherent or physically or metaphysically impossible. We had a taste of these kinds of arguments leveled against DMV powers; later we will briefly look at some of the objections facing two-way powers. It is fair to say these remain ongoing contentious debates among exponents of PRNA. Furthermore, such arguments would at best show the failure of all the models proposed so far for kinds of causal powers other than conditional necessity one-way powers. Such a priori arguments do not unequivocally rule out in principle
150
D. D. De Haan
the possibility of other kinds of causal powers. For that we need a different kind of a priori argument. A second route might aim to establish that theological determinism is true and that it is incompatible with other kinds of powers. Theological determinism maintains everything that happens in nature occurs as it does because God has fixed all of the conditions for nature to unfold as it does. Some theists, for instance, might contend that arguments for divine omnipotence or omniscience entail theological determinism. So if God is to exercise total control over the manifestations of every powerful particular, then the theological determinist might argue God necessarily only created one-way powers that manifest according to conditional necessity. God created the universe and fixed the total background conditions so as to preclude all interferences by excluding any interfering two-way, DMV, or spontaneous powers. Consequently, it seems that the strongest a priori case for de jure universal powerdeterminism is found in a form of theological determinism, which alone can rule out in principle the existence of any other kinds of powers. Apart from theological determinism it would be inexplicable why the only causal powers that exist are conditional necessity one-way powers. There is, however, a striking incongruence here. In order to secure universal determinism, some exponents of power realism and nomological antirealism might be driven to appeal to a divine command or law that excludes the existence of any powers aside from one-way powers that manifest by conditional necessity. It is not my aim to settle these debates about power-determinism; rather, my task is that of a surveyor, to stake out the territory for forms of determinism that are compatible with PRNA, which others might wish to settle and develop in light of these and other challenges power-determinism needs to overcome. Before turning to survey the territory of free will compatible with PRNA, I want to make one noteworthy point. Many believe that the truth of determinism can only be addressed by a posteriori scientific enquiries. I think there are substantive retorsion or transcendental arguments based on the constitutive requirements for conducting a scientific experiments that suggest otherwise (De Haan, 2022). But independent of such controversial arguments, if my presentation here of the varieties of powerdeterminism is correct, then it seems the only way to establish a form of universal power-determinism that is explanatorily justified, turns out to depend upon successful a priori arguments, including demonstrations for the truth of theological determinism, which is certainly not an issue that can be established by scientific enquiry.
8.4
Free Will Compatible with PRNA
Van Inwagen’s definition of the free will thesis included the claim that “we simultaneously have both the following abilities: the ability to perform that act and the ability to refrain from performing that act.” (van Inwagen, 2008, 329) Defenders of the principle of ultimate authorship provide cogent arguments for insisting that any “neutral” description of free will cannot preclude a control condition. And reasons-
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
151
responsive theorists argue similarly about the role of responding to reasons. So long as we recognize these abilities are compatible with reasons-responsiveness and control conditions for performing φ or refraining from performing φ, these abilities can come close enough to meeting this neutrality proviso. There probably is no neutral starting point for free will, but this is about as good as it is going to get. Let us start with the phrase “we simultaneously have both the following abilities.” First, while there are a number of ways to work out the ontology of abilities, given the present presumption of power realism whatever abilities are required for free will, these abilities will be identical to or grounded in powers (Mayr, 2022, this volume). Second, does free will require two powers or one power? This is the question that shall give point and purpose to demarcating the territory for versions of free will compatible with PRNA. There seem to be three basic ways to articulate a PRNA account of power-free-will(¼PFW). Hierarchy of One-Way PFW Free will comprises a hierarchy of one-way powers. Some powers explain the ability for φ-ing and other higher-order powers explain the ability to refrain from φ-ing. Both powers and their conditions for manifestation are defeasible by interferences. If there are no interferences, and the necessary conditions for the former power obtain, it manifests necessarily, and if the conditions for the latter power obtain, it manifests necessarily. Single Two-Way PFW Free will is a single two-way power that explains both the ability for φ-ing and the ability to refrain from φ-ing. This two-way power and its conditions for manifestation are defeasible by interferences. If there are no interferences, and its necessary conditions or opportunities obtain, free will manifests necessarily either by φ-ing or by refraining from φ-ing. Dispositional Modality PFW Free will can be a single two-way power or comprised of a hierarchy of one-way powers that explain the abilities for φ-ing or refraining from φ-ing. All powers tend toward their manifestations according to a sui generis dispositional modality, such that if there are no interferences and all conditions obtain, the power(s) of free will might nevertheless fail manifest. The first view endorses a straightforward understanding of van Inwagen’s phrase along with the position that conditional necessity is universal to all powers and their manifestations. Free will is to be analyzed in terms of, at least, two one-way powers, one power with the ability for φ-ing and another power with the ability to refrain from φ-ing. What distinguishes this first view from the second view, is it holds the only powers that exist are one-way powers; it rejects the existence of two-way powers. The second view maintains there are two kinds of powers: one-way powers and two-way powers, and both of them manifest necessarily if their conditions obtain and no interferences are present. Free will is a single two-way power that is manifested or exercised by φ-ing or refraining from φ-ing. We have already seen the third view, it rejects conditional necessity and endorses Mumford and Anjum’s position that a sui generis dispositional modality is common to all powers. The dispositional modality view is compatible with one-way and two-way power conceptions of free will, but Anjum and Mumford do not explicitly affirm free will is a
152
D. D. De Haan
single two-way power. In a few places they argue free will is comprised of a number of powers, and suggest these powers might be hierarchically arranged (Anjum & Mumford, 2018, ch. 10) We have already touched on some of the problems with DMV and the additional problems it confronts in securing the control conditions required for free will by the principle of ultimate authorship. If proponents of DMV free will can address these and other worries, they will have staked out a unique position on free will that is compatible with PRNA but incompatible with determinism. The two versions of power-determinism delineated above presuppose the conditional necessity rejected by DMV. We have seen that DMV’s inherent, even if mitigated, form of indeterminism entails its incompatibility with determinism. Unlike DMV free will, the two other PRNA accounts of free will endorse conditional necessity. Does endorsing conditional necessity render free will compatible with power-determinism? Before we can address this question, we need to sketch the contours of these two accounts of power-free-will.
8.4.1
Two-Way Power-Free-Will
The second position maintains free will is a single two-way power with the ability to φ or to refrain from φ-ing. To exercise a two-way power requires that the agent has both the opportunity to φ or not to φ and the ability to φ or not to φ (Kenny, 1975; Alvarez, 2013; Hacker, 2007). Opportunities are the necessary conditions for the manifestation or exercise of a two-way power, but even when a necessary opportunity obtains, it is only necessary that the two-way power manifests.8 Opportunities neither necessitate that a two-way power manifests by φ-ing nor necessitate that the same two-way power manifests by not φ-ing, that is, by refraining from φ-ing. A toddler has the ability to eat kasha or not to eat kasha, but she cannot exercise this ability in the absence of the opportunity to eat kasha when, for example, there is no kasha. But since she has the two-way power, when all the opportunities for eating kasha obtain, the toddler’s power manifests and she can exercise it by eating kasha or she can exercise the same power by not eating kasha.9 8
Not all exponents of two-way powers endorse conditional necessity. For instance, Lowe’s sui generis view combines his notion of spontaneous powers with an account of the will as a two-way power (Lowe, 2008, 8). I do not have the space to address the details of Lowe’s view here which has characteristics that make it similar to both the DMV and conditional necessity two-way powers, while being importantly different from both. 9 Some advocates of two-way powers might contend that “not φ-ing” and “refraining from φ-ing” are not manifestations of the power. Exponents of two-way powers might also argue that, given its necessary conditions, it is up to the agent whether or not to exercise and manifest its two-way power. I do not have space to address either view at length, but both views require rejecting conditional necessity since they propose that a two-way power might not manifest necessarily even when its conditions obtain without interferences. I thank Simon Kittle for drawing this issue to my attention.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
153
PRNA can help fill out this account of free will as a single two-way power that manifests by willing or refraining from willing. There are a number of ways to delineate this position. I shall draw on Thomas Aquinas’s theory of the will from Summa theologiae I-II.6–21 since Aquinas conceived of the will within a thoroughly PRNA ontology and a lot has been written about his account (Brock, 1998; Kenny, 1975). What follows is an appropriation, not exegesis. The will is a two-way power that tends toward the good and away from evil, but this power cannot be exercised in the absence of the opportunities provided by the co-manifestations of other psychological powers and the powers of objects in the agent’s surroundings. As with all necessary conditions, opportunities comprise the co-manifestations of the reciprocal power partners of the will, including practical reason, enactive perception, motivations and emotions, embodied motility, habits like the virtues and vices of these powers, and the powers of any relevant or needed affordances of objects in the environment. Opportunities are necessary since in their absence the human agent cannot exercise her two-way power by willing φ or nilling φ, that is, by refraining from willing φ. The will itself does not specify what any φ might be, this is provided by the opportunities or co-manifestations of other powers, and in particular by the teleological specifications of practical reason. The will cannot be exercised in a vacuum without the informational specification of actional opportunities by some combination of enactive perceptual registrations, motivational drives, imagery, memories, emotions, and practical reasoning. Unlike the necessary conditions of one-way powers, the obtaining of the opportunities for free will by the co-manifestations of any of these powers are insufficient to settle what to φ, whether to φ, when to φ, where to φ, how to φ and in what manner, or to nill and refrain from φ-ing in any of these ways. In other words, if the necessary conditions of a two-way power obtain, they necessitate the power manifests but they do not necessitate its exercise in one way or necessitate its exercise not in that one way. This is what distinguishes the realization of a one-way power from the exercise of a two-way power (Steward, 2012). For both one-way and two-way powers the causal manifestations of their reciprocal powers are necessary conditions or opportunities for their own manifestations, that is, they both have causes of their manifestations, but the agent with a two-way power of will can exercise its power by refraining from φ-ing just as well as it can by φ-ing. By willing or nilling the agent is thereby causing or not causing the manifestations of other powers. When the rational agent exercises its two-way power of will by willing, the will plays the active causal role in its interactions or co-manifestations with other powers, Many of are playing passive causal roles that are activated by the willing, qua causing or acting, of the exercised two-way power. But when the will is nilling it is not actively causing the manifestations of other powers. For Aquinas, human actions or intentional actions are comprised of the “internal act” of the will and the “external acts” of powers activated by the will, including the agent’s own psychological powers like practical reason, passions, enactive perceptions, motile powers, and the powers of bodily objects in the surroundings. Notice a human action, on Aquinas’s account, does not consist in two independent and successive events, as on the standard story or causal theory of action, but comprises the will’s active willing and other powers’ passive co-manifestation roles within a single process or event (Brock, 1998).
154
D. D. De Haan
Consider now four prima facie objections to this single two-way power of free will position. First, if two-way powers exist, then determinism is false, and it seems like determinism is an empirical matter. Second, it seems extraordinary that in a world where all causal commerce takes place between one-way powers we find an exception to this continuity in nature in the case of free will, wherein humans are uniquely empowered with a two-way power. Third, free will’s ability not to φ, that is, to nill φ, renders the will into a power to act randomly, irrationally, and without control. For if practical reason specifies the will with “φ-ing” as the most reasonable all things considered act to perform, the will as a two-way power can nevertheless not follow reason by refraining from “φ-ing,” which is irrational. In other words, the will exhibits its freedom only within the context of the possibility for being irrational, which conflicts with the idea that the will is a power for rational control, not irrational arbitrariness. Fourth, the proposal of a single two-way power explanation of an agent’s ability for willing φ or nilling φ can be reduced to the conjunction of two one-way powers, which is a more parsimonious explanation and does not require the extravagant postulation of two-way powers. In response to the first, we have already explained why the only form of universal determinism compatible with PRNA and capable of absolutely precluding two-way powers is theological determinism, which is established by philosophical or theological arguments. Hence, contrary to this first objection, because the case for universal power-determinism compatible with PRNA is not an empirical matter, one cannot object to a philosophical conception of free will simply because it rules out determinism. Furthermore, it is also no objection to nomological realism that it appears to favor determinism, but it is worth noting that it is the empirically informed commitments of PRNA that also appear to favor non-determinism (Steward, 2012; Cartwright, 2019). In response to the second, it is important to recognize that unlike the standard accounts of agent-causation which make humans an exception to otherwise ubiquitous event-causation, agency and power causation are ubiquitous in nature according to PRNA. Furthermore and following Aristotle, two-way powers are not unique to humans but are prevalent in many different species of higher animals (Steward, 2012; Alvarez, 2013). My earlier example of the toddler who exercises a two-way power to eat or not eat kasha did not presume the toddler was exercising powers of practical reason and free will in her eating or refusing to eat kasha. Free will is one kind of a two-way power, which has practical reason among its essential co-manifesting reciprocal power partners. Consequently, the power of free will is unique to humans only if humans in fact are the only agents with practical reason and will as co-manifesting power partners. The third objection confuses two-way powers with standard forms of agentcausation that either have agents be uncaused causes or self-causes of their actions and DMV free will which leaves agents always trying but never in control over their skittish power’s manifestations. Two-way powers, however, always co-manifest with other causal powers, which are sufficient conditions for their manifestation. Free will co-manifests with practical reason either by willing φ or by nilling φ even when practical reason proposes φ-ing as the most reasonable thing to do, for that is
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
155
what two-way powers can do. But practical reason is itself non-monotonic and so there are always additional reasons for not φ-ing either in this manner or here and now (Kenny, 1975; Steward, 2012). Some additional reasons are poor reasons, which is why akrasia is a common feature of human agency, and it is a strength of this model that it nicely accommodates akratic willing and nilling. More needs to be said on this front, but this should suffice to illustrate why these standard worries do not obviously confute two-way power-free-will. Maria Alvarez provides an incisive response to the fourth objection, “one-way powers are characterized by the fact that when the conditions for their manifestation obtain, the power will be necessarily manifested. But if an agent had the ability and opportunity to φ and also the ability and opportunity not to φ at t, and this were the conjunction of two one-way powers, then the agent would both φ and not φ at t—but that is impossible.” (Alvarez, 2013, 109) Alvarez’s argument not only confutes one reductive explanation of two-way powers, it also suggests that no two one-way powers model of free will can capture van Inwagen’s criteria for abilities to φ and to refrain from φ-ing.
8.4.2
Hierarchical Model of One-Way Powers for Free Will
Some advocates of PRNA have proposed a hierarchical model of one-way powers for free will (¼HMPFW), that is, the aforementioned first position, which is to be analyzed in terms of, at least, two one-way powers, one power with the ability for φ-ing and another power the ability to refrain from φ-ing. This first position accepts traditional conditional necessity as being universal to all powers and their manifestations, and this is held to be universal because the only powers that exist are one-way powers. Brian Ellis argues human agency manifested in deliberations, decisions, and intentional actions involves meta-causal powers that act on other powers. Even though Ellis does not mention free will, the territory he covers overlaps with most discussions of free will and I see no reason to object to Ruth Groff freely drawing on Ellis’s higher-order account of human agency and action to articulate her account of free will (Groff, 2019). Ellis argues that: From the point of view of a causal power realist, the power of deciding what to do, or what course of action to take, is a meta-causal power; i.e. a power, the exercise of which will create, modify or confirm a disposition to act in a certain kind of way, or take a course of action of certain kind. It is manifestly a mental power, but I call it a meta-causal power . . . because its outcome is to create, modify, or confirm a first order causal power. (Ellis, 2012, 198)
For most hierarchical views of action and free will, first-order and higher-order one-way powers are chained together in a vast web or network, typically involving an array of feedforward and feedback loops among power partners. We can “count as second-order not only the power to gain new powers, but the power to exercise or refrain from exercising currently-had powers. Certainly both our first and secondorder powers afford us a type of control over what we do that salt does not enjoy. . . .
156
D. D. De Haan
[M]embers of our kind just have the power, if we do, to intentionally expand upon and/or exercise or refrain from exercising our first-order powers” (Groff, 2019, 188). On a HMPFW, the ability to φ can be underwritten by first-order powers whose manifestations can initiate φ-ing. The ability to refrain from φ-ing can be underwritten by higher-order powers whose manifestations veto, cancel, deactive, or in some other ways interfere with the manifestations of the first-order powers that can initiate φ-ing. All of these powers are one-way powers that necessarily manifest in one way if their necessary conditions obtain and no interferences are present. So first-order powers necessarily manifest in φ-ing if their conditions obtain, but these manifestations can be inhibited if a higher-order power manifests to veto the manifestations of these lower-order powers. And this higher-order meta-causal power to refrain from φ-ing necessarily manifests if its own necessary conditions obtain via some network of co-manifesting powers. Besides the parsimony of an ontology that only has one-way powers, a HMPFW compatible with PRNA can also argue that its similarities to functionalist and neural network models of cognition speak in its favor. Indeed, this model could accommodate the notion of changing weights between nodes in a neural network of powers by either introducing more hierarchically ordered powers within a network or by endorsing a conception of powers that can acquire dispositions which modulate the critical thresholds for their necessary conditions for manifestation. Similarly, some might argue that the meta-causal powers for vetoing lower-order powers’ manifestations provides a straightforward PRNA explanation of Libet-style experiments on the neuroscience of free will, which conclude humans do not have free will, but free won’t, that is, the ability to veto volitions to act (Libet, 1999). Of course, for some advocates of Aristotelian approaches to PRNA, these similarities to functionalism and Libet-style neuroscience of free will provide additional grounds for rejecting HMPFW. What about Alvarez’s objection to the idea that the ability for φ-ing or refraining from φ-ing could be reductively explained by the conjunction of two one-way powers? HMPFW can avoid this problem by holding the powers do not conjunctively manifest either because their manifestations are distinct and successive, or because the higher-order meta-causal powers can veto, inhibit, or prevent the lower order powers from manifesting or achieving certain effects. There are some substantive objections against HMPFW. The most significant is the disappearance of the agent objection that has menaced all attempts to explain the reduction of agency in the standard causal story of action by introducing a hierarchy of desires (Mayr, 2011; Steward, 2012). While exponents of PRNA reject the standard story’s neo-Humean conception of causation and can eschew Frankfurt’s exclusive focus on desires independent from practical reasoning (Frankfurt, 1971), these caveats do not allow HMPFW to evade the particular perils of any attempt to explain agency by an ordered hierarchy of causal factors. On HMPFW, it seems it is never up to the agent to actively determine whether it will φ or refrain from φ-ing. What settles the outcome of what the agent does is whatever cause initiates the input manifestation of the agent’s complex and dynamically altering network of feedforward and feedback one-way powers. HMPFW paints a picture wherein the
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
157
human and any other “animal as subject and agent effectively disappears from the scene altogether, to be replaced with a mere body, an object unenlivened by any perspective or viewpoint on events, which is portrayed as a simple locus for the meeting of various interacting forces” (Steward, 2012, 96). HMPFW also fails to provide a plausible explanation of the role free will plays in an agent’s identification with its own purposes and actions as distinct from the akratic and especially alien desires, beliefs, and behaviors of the agent. Given the conflicting manifestations of different powers – some for φ-ing, others for refraining from φ-ing – which of these manifestations authentically express what the agent cares about, and which exhibit its alien, and sometimes overpowering, desires, beliefs, and behaviors? HMPFW faces the challenge of explaining both why the agent identifies with powers for φ-ing in some cases and with powers for refraining from φ-ing on other occasions, and providing such an explanation without re-introducing agential concepts within the explanans (Mayr, 2011). It seems that HMPFW inherits all of these problems concerning agential identity that plague its standard story equivalent. Much more needs to be said both to explicate further and to address challenges to these versions of free will compatible with PRNA. While I have resourced some of the problems that confront positions within the standard approach to free will, I believe any attempt to situate these forms of PRNA free will within the procrustean categories of event or agent-causation will result in unneeded obfuscation. Even Steward’s highly qualified notion of “Agency Incompatibilism” risks others mistakenly conflating her PRNA compatible notion of agency with the very positions she so deftly confutes.
8.5
Compatibilisms and Incompatiblisms
I have demarcated two versions of determinism compatible with PRNA as well as three versions of free will compatible with PRNA. Are any of these versions of determinism and free will compatible with each other? We have seen that DMV presupposes an inherently indeterminist construal of PRNA which excludes determinism independently from the problem of free will. Interestingly, as with indeterministic event-causation libertarianism, the inherent indeterminism of DMV’s skittish power manifestations is what generates the compatibility problems it faces in accommodating free will with sufficient agential control. On the other end of the spectrum is the hierarchy of one-way powers model of free will. Both versions of power-determinism are compatible with HMPFW. For HMPFW explains free will’s abilities by one-way powers that manifest necessarily if their conditions obtain, and this is simply what power-determinism tries to guarantee. The single two-way power model of free will is incompatible with power-determinism. It is also interestingly incompatible with the mitigated manifestation-tendency indeterminism of DMV free will. This is due to its commitment to conditional necessity, not despite it. It is incompatible with determinism,
158
D. D. De Haan
because when the power of free will’s manifestation is necessitated by the obtaining of its necessary conditions or opportunities, the exercise of free will – like all two-way powers – is intrinsically open to the alternative possibilities of manifesting by φ-ing or manifesting by refraining from φ-ing. Hence, “the falsity of determinism is a fairly straightforward consequence of commitments concerning the nature of agency” (Steward, 2012, 6) constituted by the exercise of two-way powers – independent of free will – that can manifest in alternative ways because their necessary conditions obtain. Similarly, it is incompatible with DMV’s mitigated indeterminism because it is intrinsic to such two-way powers to necessarily co-manifest with their causal power partners if their conditions obtain and all interferences are absent. Is it not conceivable on de jure power-determinism that God could arrange reality such that the manifestation conditions for two-way powers never obtain, and so two-way power-free-will could be compatible with de jure power-determinism? Even if this is conceivable, it would undercut the very grounds for positing two-way powers in the first place, for powers are identified by their manifestations and on this suggestion – or any others involving finks and masks – two-way powers never manifest. This is not a defensible conception of PRNA compatibilism.
8.6
Conclusion
I have argued that endorsing PRNA liberates its advocates from the conceptual straightjackets that hold captive the standard ways of thinking about the problem of free will. It also requires re-thinking about the concepts of determinism and free will. Determinism – where the past together with other factors fix all future outcomes – naturally follows from nomological realism, but it does not naturally follow from PRNA. We have seen that independent of free will, alternative possibilities are naturally built into the myriad ways the manifestations of different kinds of causal powers can be modulated and interfered with by their causal power partners. It is also noteworthy that the only form of de jure power-determinism that precludes the existence of these diverse kinds of powers depends on a priori arguments rooted in a form of theological determinism. There are not any principled empirically based arguments for power-determinism compatible with PRNA. Surprisingly, there is no a posteriori naturalistic account of de jure power-determinism – although this might also be the case for nomological determinism (Cartwright, 2004). PRNA is perfectly compatible with indeterminism, if indeterminism means the present and future are not wholly fixed by the co-manifestations of past powers. But PRNA is incompatible with the many standard portrayals of indeterminism, which, because they remain beholden to nomological conceptions of causation, contend the absence of deterministic causation leaves nothing to hold nature together besides chancy connections among agents and events (Steward, 2012). We have shown that many worries about indeterminism and arbitrary connections disappear for some versions of PRNA, due to the co-manifestations of causal powers, including free will, when their necessary conditions obtain without interferences.
8 The Power to Will Freely: How to Re-Think About the Problem of Free Will. . .
159
Many exponents of PRNA caution against falling into the trap of inheriting misconceptions of phenomena or taking on other burdens from the very paradigm PRNA aims to escape and set aside. Cartwright has revealed that one of those errors is adopting a view from nowhere that presents a vision of everything only by ignoring the concrete circumstances where we really encounter co-manifesting causal powers. The standard problem of free will also starts from a grand picture of everything and asks, can free will have a place in this picture? I have followed the standard approach here in order to show why endorsing the rival paradigm of PRNA liberates us from the seemingly innocent problematic commitments the standard approach imposes on the problem of free will. But a more thoroughly Aristotelian approach would start instead with the manifestations or rather intentional and voluntary actions that put the question of powers like free will on the table in the first place. What is required then is to start again, but this time by focusing on the problems of action, which provide a surer footing for getting off on the right path to understanding the power of free will in world without laws of nature.10
References Alvarez, M. (2013). Agency and two-way powers. Proceedings of the Aristotelian Society, 113, 101–121. Anjum, R. L., & Mumford, S. (2018). What tends to be: The philosophy of dispositional modality. Routledge. Beebee, H. (2000). The non-governing conception of laws of nature. Philosophy and Phenomenological Research, 61(3), 571–594. Brock, S. (1998). Action and conduct: Thomas Aquinas and the theory of action. T&T Clarke. Cartwright, N. (1989). Nature’s capacities and their measurement. Oxford University Press. Cartwright, N. (1999). The dappled world. Cambridge University Press. Cartwright, N. (2004). No god, no laws. In S. Moriggi & E. Sindoni (Eds.), God and the laws of nature (pp. 183–190). Angelicum Mondo X. Cartwright, N. (2019). Nature, the artful modeler: Lectures on laws, science, how nature arranges the world and how we can arrange it better. Open Court. De Haan, D. (2022). The power to perform experiments. In W. M. R. Simpson, R. C. Koons, & J. Orr (Eds.), Neo-Aristotelian metaphysics and the theology of nature. Routledge. Ch. 7. Demarest, H. (2017). Powerful properties, powerless Laws. In J. Jacobs (Ed.), Causal powers. Oxford University Press. Dumsday, T. (2019). Dispositionalism and the metaphysics of science. Cambridge University Press. Ellis, B. D. (2012). The power of agency. In R. Groff & J. Greco (Eds.), Powers and capacities in philosophy: The new Aristotelianism. Routledge. Frankfurt, H. (1971). Freedom of the will and the concept of a person. Journal of Philosophy, 68, 5–20. French, S., Bird, A., Ellis, B., Mumford, S., & Psillos, S. (2006). Looking for Laws. Metascience, 15, 437–469. Groff, R. (2012a). Ontology revisited: Metaphysics in social and political philosophy. Routledge.
10
I would like to thank an anonymous reviewer and Simon Kittle for enormously helpful feedback and criticisms on earlier versions of this essay.
160
D. D. De Haan
Groff, R. (2012b). Whose powers? Which agency? In R. Groff & J. Greco (Eds.), Powers and capacities in philosophy: The new Aristotelianism. Routledge. Groff, R. (2019). Sublating the free will problematic: Powers, agency and causal determination. Synthese, 196, 179–200. Groff, R., & Greco, J. (2013). Powers and capacities in philosophy: The new Aristotelianism. Routledge. Hacker, P. M. S. (2007). Human nature: The categorial framework. Blackwell Publishers. Jacobs, J. (2017). Causal powers. Oxford University Press. Jaworski, W. (2016). Structure and the metaphysics of mind. Oxford University Press. Kenny, A. (1975). Will, freedom, and power. Basil Blackwell. Lewis, D. (1981). Are we free to break the Laws? Theoria, 47(3), 113–121. Libet, B. (1999). Do we have free will? Journal of Consciousness Studies, 6(8–9), 47–57. Lowe, E. J. (2008). Personal agency: The metaphysics of mind and action. Oxford University Press. Mackie, P. (2014). Mumford and Anjum on incompatibilism, powers, and determinism. Analysis, 74(4), 593–603. Marmodoro, A. (2016). Dispositional modality Vis-à-Vis conditional necessity. Philosophical Investigations, 39(3), 205–214. Marmodoro, A. (2017). Aristotelian powers at work: Reciprocity without symmetry in causation. In J. Jacobs (Ed.), Causal powers. Oxford University Press. Martin, C. B. (2008). The mind in nature. Oxford University Press. Mayr, E. (2011). Understanding human agency. Oxford University Press. Mayr, E. (2022). Freedom of the will and rational abilities. In A. Marmodoro, C. Austin, & A. Roselli (Eds.), Powers, Time, and Free Will. Springer (this volume). Mumford, S. (2004). Laws in nature. Routledge. Mumford, S. (2014). The irreducibility of dispositionality. In R. Hüntelmann & J. Hattler (Eds.), New scholasticism meets analytic philosophy (pp. 105–128). Editiones Scholasticae. Mumford, S. (2018). Laws and their exceptions. In W. Ott & L. Patton (Eds.), Laws of nature. Oxford University Press. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford University Press. Mumford, S., & Anjum, R. L. (2014). A new argument against compatibilism. Analysis, 74, 20–25. Mumford, S., & Anjum, R. L. (2015a). Freedom and control: On the modality of free will. American Philosophical Quarterly, 52(1), 1–11. Mumford, S., & Anjum, R. L. (2015b). Powers, non-consent and freedom. Philosophy and Phenomenological Research, 91(1), 136–152. O’Connor, T. (2000). Persons and causes: The metaphysics of free will. Oxford University Press. Pereboom, D. (2014). Free will, agency, and meaning in life. Oxford University Press. Steward, H. (2012). A metaphysics for freedom. Oxford University Press. Steward, H. (2022). Laws loosened. In A. Marmodoro, C. Austin, & A. Roselli (Eds.), Powers, Time, and Free Will. Springer (this volume). van Inwagen, P. (2008). How to think about the problem of free will. The Journal of Ethics, 12, 327–341. van Inwagen, P. (2017). Thinking about free will. Cambridge University Press.
Chapter 9
Laws Loosened Helen Steward
Abstract In this paper, I shall consider a number of different ways in which philosophers in recent years have attempted to offer conceptions of natural law which in various respects suggest that the grip of law on reality might be less tight than has been traditionally supposed. One such loosening is represented by the suggestion that many laws might be best thought of as probabilistic rather than deterministic. A second kind of loosening has been the admission that certain laws relevant to human behaviour might hold only ceteris paribus. Yet a third is the suggestion that all laws - including even fundamental physical ones – might hold only ceteris paribus (Cartwright, The dappled world, Cambridge University Press, Cambridge, 1999). How, though, are these different suggestions related to one another? Which kinds of loosening might entail which other kinds? And which, if any, might be most promising as regards making room in the universe for free will? In this paper I shall try to suggest that the first and second strategies are far less useful than the third in making the kind of space which would be required to subserve the reality of free will; and that a fourth kind of loosening – compatible with but not entailing any of these other kinds – from laws as world-dictators to laws as worldconstrainers might yet be more useful than any of the other three in this respect. Keywords Free will · Laws of nature · Aristotelian metaphysics · Humean metaphysics It is sometimes suggested that a libertarian view of free will – that is to say, a view of free will according to which free will exists in such a way as to be inconsistent with universal determinism – is thereby committed to the idea that free will requires the
H. Steward (*) University of Leeds, Woodhouse Lane, Leeds, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_9
161
162
H. Steward
beings that possess it to be beyond the reach of natural law. Here, for example, is Jerry Coyne, an evolutionary biologist1: Our thoughts and actions are the outputs of a computer made of meat – our brain – a computer that must obey the laws of physics. Our choices, therefore, must also obey those laws. This puts paid to the traditional idea of . . . free will: that our lives comprise a series of decisions in which we could have chosen otherwise (Coyne, 2014).
Coyne does not explain in any detail why he thinks the fact that our brains must obey the laws of physics puts paid to free will. But few philosophers would agree these days that there is any straightforward route available to Coyne’s conclusion. Even if we interpret ‘traditional’ free will as something like ‘free will as typically understood by the libertarian’, the recognition that the fundamental realm to which the laws of physics pertain cannot be assumed to be identical with the deterministic world-system of Newtonian science2 – or any similarly deterministic system of all-encompassing physical laws – leaves Coyne’s claim in want of justification. If we are to be persuaded to believe him, we need to know what assumptions he is making about the scope and nature of the laws of physics, and whether there is any justification either for those assumptions, or for the conclusion that they warrant the denial of free will. Along with many libertarians, I believe that Coyne is wrong about the inconsistency of what I shall continue to call ‘traditional’ free will with physical law. And, like most of those libertarians, I also believe that the key to understanding how this is possible lies in a correct conception of the world as a globally indeterministic place, a place therefore in which laws have to be thought about as principles far more permissive of alternative futures than laws as they have often been conceived. But I part company with many libertarians on the question of what kind of adjusted understanding of the nature of the world’s subjection to law is needed for the job. We libertarians all agree, of course, that indeterminism is necessary for free will, because we all agree that determinism and free will are inconsistent. However, that doesn’t mean we all agree about how to conceive positively of the indeterminism that we all insist upon. If indeterminism is simply the negation of determinism, that still leaves plenty of room for debate about what kind of denial of determinism is needed for free will. And this is important, because some positive conceptions of the indeterministic universe might be a good deal more hospitable to the accommodation of free will than others. In this paper, I shall examine a range of possible ways of departing from a strictly deterministic vision of the Universe which have been claimed to provide a basis for free will.3 Each of the departures I shall consider makes a different suggestion about how our conception of the physical laws (or indeed more broadly, of the natural
1
For another prominent example of the same sort of claim, this time from a quantum physicist, see Hossenfelder (2019). 2 If indeed Newtonian science is deterministic. For a contrary view, see Earman (1986). 3 It may be important to say here, though, that I shall not here defend the libertarian line of thought in general – rather, for the purposes of the paper, I shall take it for granted. My aim here is to
9 Laws Loosened
163
laws) might be ‘loosened’ so as to make space for libertarian freedom. One of my aims is simply to forefront the fact that there is such a range of non-equivalent suggestions available to be considered – since far too often, discussion of libertarianism in the free will literature simply assumes without argument the version of indeterminism that I shall consider first, below. But a second aim is to suggest that the dominance in the literature of this first conception is a problem. The reason it is a problem is that the incursions into determinism which this first loosening strategy envisages are in fact the least useful to the libertarian of those that seem potentially to be available. As is frequently observed, the kind of loosening the first strategy envisages seems insufficient, without serious metaphysical supplementation, to support the existence of free will. Moreover, and though this is sometimes claimed on its behalf, it is not even necessary for the laws to be loosened in this first way in order to allow for the possibility of libertarian free will. This need not mean that the variety of indeterminism which this first conception of ‘loosening’ countenances could not, in principle, be part of a convincing account of how libertarians might secure what I shall, for the purposes of this paper, call Law-Compatibility (by which I mean the thesis that traditional free will is not inconsistent with the laws of nature). But it does mean that libertarians who take this route are left with a great deal of additional work to do. The best the first strategy unsupplemented can do, in my estimation, is afford a means of showing that it may not be absolutely clear that the laws, such as they are, rule out free will. It cannot show what the libertarian might really wish to show, namely, that it is absolutely clear (on that conception of law) that they do not do so. I have been (in one sense) baffled for years about why so few people seem to make what seems to me to be the obvious rejoinder to the charge that libertarian free will is inconsistent with the government of reality by law. But at the same time, I am also (in another way) not in the least surprised that the obvious rejoinder is not thought to be any kind of route to the establishment of Law Compatibility. The strategy I prefer may ultimately depend for its success on the rejection of a very widely-held view of reality – and persuading people to change their minds about such things is a difficult business. I shall not attempt in this paper, therefore, to complete the persuasive task – that is a job for a book, not a paper. I aim rather to show at least that the option is there for those who are prepared to ditch that widelyheld view – and to indicate some of the advantages that it has over its rivals. I shall explain (briefly) what I take to be the obvious but neglected rejoinder in section (iv). Before that, though, I shall explicate some of the strategies that others have adopted which I take to be wanting – and will try to show how and why they are wanting. I will also suggest that they share a feature which my own preferred solution is lacking – a feature which I believe is not unconnected with what I (and many others – mainly compatibilists) regard as their dim prospects of success. My hope is that because it employs a strategy of a notably different variety, my account
differentiate between some alternative ways of understanding what the falsity of universal determinism might consist in and adjudicate between them.
164
H. Steward
of how Law Compatibility is to be secured is a good deal more promising than the other accounts I shall consider.
9.1
Probabilistic Laws
Supposing that universal determinism is best formulated as a thesis about the entailment of facts about the future by facts about the past and present, together with the laws of nature,4 most strategies for understanding what the falsity of that thesis might consist in have focused on describing some possible alternatives to the universal reign of deterministic laws. And by far the most commonly touted alternative canvassed in the free will literature is the suggestion that some of the laws in question might be probabilistic, rather than deterministic. Probabilistic laws, in conjunction with facts about the past or the present, would not entail facts about the future, precisely because they are merely probabilistic, not deterministic. But this suggestion is a good deal less straightforward than it seems – and I will now argue that even if it is true (as I have no problem at all conceding) that many natural laws are probabilistic, it is very difficult (as is frequently pointed out by compatibilists) to show how this would make space for the exercise of free will. We must start by asking the question what a probabilistic law is. It is usually said that probabilistic laws are laws which fix the chances of a given kind of outcome, without actually settling which outcome will occur in any given case. But the mere fact that some laws are probabilistic in this sense is not in fact sufficient to rule out the reign of something that might perfectly well be called universal determinism. To see this, consider a law from a science in which there do indeed seem to be laws that are probabilistic in the sense suggested. The Mendelian genetic laws of Segregation and Independent Assortment result in there being a one in four chance that a dihybrid cross between two parents heterozygous for each binary trait yielding a child that is homozygous for each such trait (Glynn, 2010). There seems little reason to deny that this Mendelian principle is a law. Glynn argues convincingly that it supports counterfactuals and is confirmed by its instances and therefore meets the main criteria which it is usually said to be necessary for laws to meet. Nor is there any reason to deny that it is a probabilistic law in the sense specified above, since it is a
4
This is by far the most popular way of formulating universal determinism, and so I adopt it here, in deference to the literature, although I have argued elsewhere that this is not in fact the best way to formulate the thesis of determinism since it is damagingly neutral about the question whether there is any such thing as natural necessity (Steward, 2021). This may be the place also to note that for the purposes of this paper, I shall be assuming what some have called a ‘governing’ conception of laws of nature (see Beebee, 2000), since I am in agreement with many compatibilists that on a non-governing conception, there is no evident problem about free will at all. It is only if one assumes that the laws are such as to govern that there is any problem about the compatibility of free will with determinism (though see Huttemann, this volume, for an argument that the Humean faces different problems – an assessment with which I concur).
9 Laws Loosened
165
law which does not say what will happen in any given case, but only fixes the chances of different kinds of outcome in any given case of the relevant kind. However, it is a law which would seem to be perfectly consistent with universal determinism at the physical level. The precise underlying mechanisms which determine which phenotypic trait each individual offspring will have might perfectly well take place in accordance with fully deterministic laws, for all that is claimed by the probabilistic Mendelian law. If it were not so, indeed, we would have a much quicker route to the conclusion that indeterminism is generally true of our universe than is normally supposed to be available. We need, therefore, to be careful to specify more carefully which laws have to be probabilistic if we are even going to be certain that we have managed to characterise an indeterministic universe by invoking them. It might seem as though the obvious way to fix this problem is just to restrict our attention from the outset only to laws which are couched in terms of the fundamental level.5 If, even at the fundamental physical level, the laws only fix the chances that an event of a certain kind will happen and we cannot find any hidden variable that might explain differences between apparently identical scenarios in which different outcomes occur at different times, then it looks as though the Universe is such as to be irreducibly chancy in a way that means, surely, that universal determinism must be false. If we are already at the fundamental level, there is nowhere else to go for the detection of further underlying mechanisms which are nevertheless deterministic. But now we have a different problem – one that is often one of the main reasons why libertarianism is taken to be hopeless. The question now is how the existence of probabilistic laws at the fundamental level might make space, exactly, for free will. One view which has been often expressed, for example, is that the fundamental level is far too fundamental for probabilistic laws holding at that level to be relevant to a phenomenon like human freedom. One argument that has been frequently made is that such indeterminacy as seems to exist at the fundamental level might very well ‘cancel out’ at higher levels, leaving the motions of macroscopic objects such as human animals to all intents and purposes determined. Honderich (1988), for example, suggests the possibility that “an undetermined micro-event may be one of a specific and finite set of possible micro-events and further, that each member of the set would have had the same effect in the macro-world. In which case, the macroevent was fixed in so far as the micro-world is concerned, despite its being the effect of a chance event” (p. 328). Honderich himself does not fully endorse this argument, noting (with impeccable fair-mindedness) that we would need to be sure that all micro-indeterminisms cancelled out in this manner before we could be confident of macro-determinism – but he then moves on swiftly to offer, in addition, a second argument of the general kind which many philosophers still take to be the clincher. The question is how exactly indeterminism, whether or not it somehow ramified into an indeterminism that is still significant at the macro level, could possibly convert into agential control. Even if a given nexus is indeterministic at the fundamental
5
I do not intend to question, for the purposes of this paper, the widespread assumption that there is such a fundamental level, nor that the task of formulating its laws falls to physics.
166
H. Steward
level, this does not help us secure free will unless we can understand how such micro-level indeterminism delivers such control. We seem to face a dilemma: either the agent has some control over what happens at the micro-level, or she does not. If she does, we have the problem of understanding how this control is exercised. Certainly, any such control cannot be understood as an ordinary example of direct intentional control, since we definitely do not knowingly control microphysical events, under normal circumstances. But on the other hand, if the agent does not have control over the fundamental level, she seems to be subject to the problem of luck.6 Even if the relevant micro-level indeterminacies could somehow get amplified in such a way as to make indeterministic mental events possible, an agent whose options are controlled by which of a range of entirely random micro-level events occurs appears to be hopelessly at the mercy of chance events – and hence arguably is no better a candidate to be a possessor of free will than a fully determined agent. Some libertarians who have invoked this kind of ‘loosening’ have, of course, attempted to argue that there is a way out of the dilemma. Robert Kane, in particular, has offered a detailed and complex libertarian view based on the notion of a ‘selfforming willing’ (SFW), a kind of struggle-resolving event which arises in circumstances in which one is attempting to resolve moral or prudential conflict by battling against temptation. SFWs are, in Kane’s view, only indeterministically related to their prior mentalistic causes (which are things such as beliefs, principles, values, etc.), in the sense that those beliefs, principles, values, and so on do not guarantee that a given moral or prudential struggle will result in any particular outcome. This idea of undetermined choice is of course common to many libertarian accounts of the psychology of free willed decision; but Kane goes much further than most libertarian philosophers in the attempt to present a detailed vision of what might be going on, physically and neurologically speaking, in the brain of an agent, when such a selfforming willing event occurs. According to Kane, since alternatives to any given SFW are always metaphysically possible, micro-level indeterminism must be a necessary condition for their occurrence. But how does micro-level indeterminism make room for the possibility of the kind of macro-indeterminism which Kane supposes pertains to situations in which SFWs occur? One possibility, Kane explains, is that SFWs are the result of the amplification of quantum-level indeterminacies in neural networks. In chaotic systems, large and macroscopically detectable differences may eventually result from tiny variations in starting conditions. Kane’s idea is that systemic effects at the level of neural networks might amplify micro-level indeterminacies in such a way as to make it possible that more than one psychological-level outcome might result. What still seems difficult to understand, though, on Kane’s picture, is how the agent can possibly have any influence over which undetermined quantum-level event occurs and subsequently gets ‘amplified’. Even if neural networks can amplify minute variations at the microphysical level into major differences at the
6 For various versions of this worry, see e.g. Strawson (1994); Haji (1999); Almeida and Bernstein (2003); Mele (1995): 195–204; Mele (2006).
9 Laws Loosened
167
macroscopic level, the agent still seems to be at the mercy of luck, since without more additions to the story, the agent still seems to be subject to the chance occurrence of indeterministic event A happening (and then being amplified) as opposed to chance event B occurring (and then being amplified). At one point, Kane does get close to providing the sort of explanation which I think actually could help us see how this problem could be avoided, by appearing to suggest that top-down effects might be possible within neural networks: Of special interest are the potential effects . . . chaotic amplification might have on neural networks, which are systems of many functionally interconnected neurons. The operation of such networks is holistic in the sense that, as Gordon Globus (1995) puts it, “the influence of the whole net” of neurons affects each “individual node [i.e. each neuron] and the influence of the individual node [affects] the whole net”. As a consequence, such networks can be sensitive to variations of firings of individual neurons . . . . Similarly, the self-organization of the network can effect (sic) the firing potentials of its individual nodes (Kane, 1996: 130).
If we imagine that in some sense the workings of our conscious decision-making processes are to be identified with processes going on within the neural net, then perhaps we can see an argument for the view that for the purposes of thinking about moral and prudential decision-making, an agent is such a net, or perhaps an agglomeration of them; and so if a net can affect what happens in its parts, and those parts in turn can affect what happens in their parts, and so on, we have a way of understanding how an agent could ultimately come to have some control, by way of deliberation and decision-making, over events at the fundamental level. This would afford us a way of embracing the first horn of the dilemma – we could accept that the agent does have control over events at the fundamental level, because there is such a thing as top-down causation from whole neural net to quantum level event, via a cascading chain of whole-to-part determination relations. However, after briefly mentioning the capacity of the whole net to affect its parts, Kane lays no further stress on the idea that the whole net might affect individual nodes – and in particular, he never extends the suggestion that top-down influences might be at work in such a way as to allow it to be the case that a whole physical system might be able to have top-down effects on quantum-mechanical events. But without such supplementation, we seem to be left impaled on the ‘problem of luck’ horn of our original dilemma. Where, then, do we stand with respect to Kane’s proposed solution? To be as charitable as possible, perhaps it is just about conceivable that some mechanism such as that which Kane envisages could somehow underlie free will. But in order to escape the dilemma posed above, it is insufficient merely to postulate probabilistic laws; further heavyweight metaphysical posits such as top-down causation from system to sub-systemic part appears to be essential if a version of the dilemma’s first horn is to become available for occupation. Moreover, and perhaps more importantly for the purposes of this particular paper, I would like to point out that it does not seem at all necessary for anything that is important to Kane’s story that there be probabilistic laws at the fundamental level – what is essential is merely that some events and circumstances should be such as to escape the net of deterministic law. And it is important to see that these two suggestions are not equivalent. Recall that probabilistic laws (as they are generally defined) fix the chances that a given outcome
168
H. Steward
will result from certain conditions (in that respect, they resemble deterministic laws, which are special only in that they fix those chances at 0 or 1). But why must everything that happens either (a) be determined or (b) have its chances precisely fixed by some law, or laws? Why might there not be events which are simply (in certain respects) lawless? – neither determined to happen, nor likely to happen with any particular fixed probability of the sort which a probabilistic law, or set of such laws, might together dictate?7 The assumption that the world is indeterministic need not, surely, be the same thing as the assumption that at least some of its laws are probabilistic. One can, for instance, surely imagine indeterministic worlds all of whose laws are completely deterministic. For example, one can imagine a world in which there is just a single law to the effect that everything that is red at a given moment turns blue 10 s later and then back again to red 10 s after that – and that’s it. Everything else in that world, let us suppose, is sheer chaos, with neither deterministic nor chance-fixing laws holding sway.8 In trying to understand more fully how we might make the probabilistic law variety of loosening serve the purposes of the libertarian, then, we have effectively come to see that this idea is (a) useless without serious metaphysical supplementation; and (b) in any case represents by no means the only way we have of understanding how some events might fail to be determined by prior conditions together with the laws; and it is this negative condition, and not anything specific to chancefixing laws, that is important to the workability of a Kane-style solution. I shall return to this theme in section (iv).
9.2
Ceteris Paribus Laws
The shift from deterministic to probabilistic conceptions of law is one kind of ‘loosening’ of the grip of law which has been thought potentially helpful to defenders of libertarian free will. It is not, however, the only kind of loosening which has been attempted in recent years which has been thought potentially relevant to the free will debate. Another strategy which has been thought promising by some has taken note of the fact that many laws are apparently not ‘strict’ but are rather so-called ‘ceteris paribus’ laws – that is to say, laws which are not completely exceptionless, but hold true only absent interference, or for the most part, or only under idealised circumstances that never actually obtain. The literature on how precisely to think, in general, about ceteris paribus laws is vast and I shall not be able to do it justice to its richness here; nor can I take on the task of justifying the Cf Nancy Cartwright “For all we know, most of what occurs in nature occurs by hap, subject to no law at all” (1999: 1). 8 Recall that we are assuming, for present purposes, a governing conception of laws of nature (see note 4 above). The issue might admittedly present somewhat differently given a Lewisian ‘best system’ account, but as mentioned above, I believe libertarians have in any case nothing to fear from ‘best system’ laws. 7
9 Laws Loosened
169
assumption that such laws exist and deserve the appellation. What I shall do instead is to move directly to consider how ideas related to the observation that some laws have exceptions or are ‘non-strict’ has been thought potentially helpful to the defender of free will. The strategy I have in mind begins from the idea that it may be helpful to free will to observe that the laws, such as they are, of human behaviour, that is to say, the laws, as best we can find them, that are proper to psychology and perhaps to certain other of the social sciences, such as economics or sociology are, at best, merely ceteris paribus (CP) laws. Psychological examples, indeed, were prominent in the early arguments for the recognition of CP laws in general – things such as “if person X wants A and believes B to be an optimal means of achieving A, then X will attempt to do B” (Fodor, 1987). Fodor, one of the most influential advocates of the importance of recognising CP laws in the special sciences, noted that although principles like this were enormously useful, explanatory, supportive of counterfactuals, and so on, and thus seemed to be deserving of recognition as laws of some kind, it certainly could not be maintained that they held without exception – to take the present example, there are, after all, such things as weak-willed agents, who precisely do not attempt to take what they believe to be the best means to satisfy what seem to be their strongest all-things-considered desires. There are of course very different accounts in the philosophical literature of why principles formulated in folk-psychological terminology seems to resist the reach of universal law – for some, it is a matter of a fundamental distinction between reasons and causes9; for others, although it is conceded that reasons can be causes, the psychological is nevertheless supposed to be fundamentally anomalous in and of itself, there being neither psychological nor psychophysical laws.10 But we need not decide this issue for present purposes; what is important is merely that ceteris paribus laws do not hold without exception. Are ceteris paribus laws also probabilistic laws? Not in the sense described above, since they are not laws that fix the chances of anything. Rather, they are general principles which hold for the most part – and though that implies that they are in a sense indeterministic, it does not follow that they are probabilistic according to the definition of probabilistic law mooted above. However, it is noteworthy for present purposes that they appear to share with higher-level probabilistic laws like the Mendelian law I considered earlier the property of being ostensibly quite compatible with universal determinism at the fundamental level. For example, even if we accept as a merely CP law the suggestion above that “if person X wants A and believes B to be an optimal means of achieving A, then X will attempt to do B”, there seems no reason to suppose that the universe might not be deterministic at levels below the
9
For a range of representative examples, see Ryle (1949); Anscombe (1958); Kenny (1963); Stoutland (1986); Tanney (1995). 10 See in particular Davidson’s (1970), (1973) and (1974) for an influential development of the thesis of the ‘anomalism of the mental’.
170
H. Steward
psychological, in such a way that any particular departure from the relevant psychological principle could perfectly well be given a deterministic explanation. Having made this observation, it might now be wondered how ceteris paribus laws could possibly be of any use to the incompatibilist, given their apparent consistency with the rule of universally deterministic physical law, and so, one might have thought, with universal determinism itself. But those who have tried to exploit the ceteris paribus strategy would insist that we make a mistake in supposing that determinism and indeterminism can be sensibly characterised independently of any reference to levels of description. Kenny (1975), for example, argues that since it is only actions described in terms of human behaviour that libertarians claim to be free, and since there are no strict psychological laws which are couched in the terminology of human behaviour, we have a way to reconcile free will with determinism at other, lower levels.11 Kenny accepts that sociological or economic or psychological determinisms would be incompatible with human freedom. Any of these sorts of determinism would, in Kenny’s view, be incompatible with the idea of voluntary action – an agent cannot genuinely do something because she wants to, or because she sees a reason to do so, if her desire produced her action by way of a strict deterministic psychological law – and moreover if that desire was itself produced by such a law. In his view, though, so far as free will is concerned, we need not worry about the reign of deterministic laws only at lower levels – such as the physiological, for example – or (presumably) the level of fundamental physics – because this is not the level of description in terms of which human actions are singled out. The question is, though, whether this really helps much with the standard worries about determinism and freedom. Kenny himself raises the main issue that I imagine libertarians are likely to have in mind about his purported reconciliation of free will with physiological determinism: . . .surely, if every movement of a man’s hands, every twitch of every muscle was predictable; then surely his whole observable life would be predictable, no matter in what terms it was described. The untidy nature of the translation from physiological into intentional terms does not really count against this. The situation might be compared to a jigsaw puzzle. A man’s life, told in the terms which would appear in his biography, might be compared with the picture on the completed puzzle; the physiological events which make up his life might be compared to the pieces of the puzzle. There is no systematic correlation between pieces of the puzzles and details of the picture . . . . For all that, once the pictures are fitted together, there you have the picture; and anyone who knows how to put the pieces together can eo ipso lay down the picture. (Kenny, 1975: 150).
Kenny himself claims that this picture, though powerful, is misleading. However, he does not really tell us what exactly is supposed to be misleading about it. Rather, he reverts to four necessary and sufficient conditions for possession of ‘liberty of indifference’ which he takes himself to have established earlier in the book, and
Kenny speaks mainly of determinism at the ‘physiological’ level as being the potential threat to free will – but presumably, he would think that the same was true of determinism at (for example) the chemical or physical levels. 11
9 Laws Loosened
171
proceeds to try to argue that none of these conditions is incompatible with physiological determinism. The four conditions are these: 1. 2. 3. 4.
A has at t the ability to ø. A had at t the ability not to ø. A had at t the opportunity to ø at t. A had at t the opportunity not to ø at t.
The trouble, though, is that the way in which Kenny interprets the two ability conditions is highly controversial. He adopts a traditional compatibilist strategy that goes back at least to G.E. Moore in claiming that: whether at t A has the ability to ø and the ability not to ø – the two-way power of ø-ing – can be settled independently of the circumstances obtaining at t. Provided that A has in the past, and continues in the future, to satisfy the criteria for possessing this ability, ‘A can ø’ will be true of him at this present moment t. (Kenny, 1975: 151).
But the distinction between what an agent can do in a large range of circumstances (what are sometimes called ‘global abilities’), and what the agent is able to do now, in some very particular circumstances (what are often called ‘local’ abilities), is very often made in the free will literature, and it is hard to deny that it can be pertinent to the decision whether an agent was or was not free to act in some particular way. Features of specific circumstances can certainly be such as to intuitively prevent an agent from acting freely – and the incompatibilist will normally want to insist that it is the latter and not the former that are truly relevant to the question whether an agent could have done otherwise at some particular moment in time. Whittle (2010), for example, presses this point against the so-called ‘new dispositionalists’.12 She adopts what is a useful convention for distinguishing between the two kinds of ability – consider Sally, for example, who is generally an excellent singer and confident performer but for some reason finds herself floored in the presence of her forbidding aunt and cannot utter a note. We can then admit, with Kenny, that she still has the ability to sing (when her aunt is present) – her aunt’s presence cannot rob her of that general ability. But this point alone cannot suffice to show that she has the-ability-to-sing-when-her-aunt-is-present. And given that this distinction seems coherent, we need an argument for insisting that it is only relatively global abilities which are required for free will. This is something the incompatibilist is very unlikely to concede to Kenny – certainly it is very unlikely to be strategically useful for the defence of an incompatibilist position. In short, in the end, what Kenny offers is a variation on an old-style compatibilist point about the conditions under which abilities can be attributed – and this is not likely to recommend itself to the libertarian as a way of showing that free will is possible. The
According to new dispositionalists, an agent has the ability to φ if and only if they have a disposition to φ when they are trying (or are otherwise properly motivated) to φ (I borrow this characterisation from Vetter and Jaster (2017)). New dispositionalist positions are presented by Smith (1997); Vihvelin (2004); and Fara (2005). 12
172
H. Steward
libertarian will be too doubtful that under physiological determinism, the agent really does have the ability not to ø at t – and she will feel ultimately that the jigsaw puzzle worry above has not been dealt with. In recent years, Christian List has offered a somewhat different version of the ceteris paribus strategy. Like Kenny, List insists that free will depends only upon what the laws are like at the psychological level. List claims that free will is genuinely inconsistent with a thesis he calls ‘Agential Determinism’, which states that “in any situation, only one course of action is possible for the agent” (List, 2019: 87). But List says that it would only follow that free will and physical determinism were incompatible if something he calls ‘The Linking Thesis’ is true: The Linking Thesis If, given the complete physical state of the world at any point in time, only one future sequence of events is physically possible, then, in any situation, only one course of action is ever possible for an agent (List, 2019: 88).
But List claims that the Linking Thesis is false, on the grounds that when we are asking what an agent can or cannot do, the relevant level of description is not the fundamental level but the psychological one. List’s strategy is in some ways evidently akin to Kenny’s, in that he insists on assessing the question what courses of action are possible for agents with reference only to the psychological level of description. I have somewhat more sympathy with List’s particular version of the ceteris paribus strategy than Kenny’s, though, because of the emphasis he places on the phrase ‘possible for an agent’. It does indeed seem to me that the question what is possible for an agent in a given situation is a different question from the question what it is possible that such an agent will do, which may open up room for the observation that even if it is settled by physical laws and prior physical conditions that an agent will do such-and-such a thing, it need not follow that only one thing is possible for them. ‘Possible for’ suggests an intradeliberative perspective. If more than one thing is possible for an agent, we are invited to think, there is more than one thing they can coherently consider doing. But even if List can show that there is room here to make a promising distinction, it would only serve to open up space for free will if one was confident that there were no necessary conditions for free will which would be ruled out by physical determinism, other than the existence of these intra-deliberative possibilities-for-acting. Worries about sourcehood would represent one kind of anxiety here.13 Another kind, one which is of particular concern to me, would be that List’s move to distinguish ‘possibilities for’ from ‘possibilities that’ has no appeal for those who embrace the view called Agency Incompatibilism – the view that agency itself is incompatible 13 So-called ‘source incompatibilists’ stress that there is a necessary condition additional to the standard ‘alternate possibilities’ condition on acting freely – which specifies that we must be the origin or source of our free actions and decisions. According to source incompatibilists, we cannot meet this necessary condition in the way required, if universal determinism is true, since according to determinism, every condition of the world (except, presumably, the first) can be traced to a prior one. See for example Kane (1996, 1999, 2008, 2011) and Pereboom (2001, 2005, 2014).
9 Laws Loosened
173
with universal determinism14 – since List’s distinction already takes for granted that agents may exist and be confronted with choices in worlds in which physical determinism reigns. Here is not the place to expound and defend these particular arguments for incompatibilist positions – but we can at least observe, I think, that List’s view is at best a response to a certain particular kind of claim about what determinism would rule out – the existence of alternative possibilities for agents who can already be assumed unproblematically to exist. But if the existence of agents under determinism is itself in question, List’s strategy will be irrelevant, because it cannot be used to offer a defence of the changed point of contention between compatibilist, on the one hand, and incompatibilist, on the other. The Kenny-List, strategy, then, will not help to allay the concerns of those who believe that universal determinism at the fundamental level is inconsistent with such things as the sourcehood or agency requirements of free will. However, it might be suggested that the ceteris paribus strategy might meet with more success if we could show that all laws – including laws at the most fundamental level – might hold, at best, only ceteris paribus. Surely, one might think, if all laws allowed for exceptions, Law Compatibility might be safely accommodated? – even on the assumption that there are strong sourcehood requirements on agency. In the next section, therefore, I shall take a look at Nancy Cartwright’s argument for the view that all laws, including the laws of physics, hold only ceteris paribus and will try to cast doubt on the idea that this move alone could be sufficient to establish Law Compatibility.
9.3
Laws as Ceteris Paribus All the Way Down?
In the introduction to The Dappled World, Cartwright announces the three central theses of her book. The second of the three is this: Laws, where they do apply, hold only ceteris paribus. By ‘laws’, I mean descriptions of what regularly happens, whether regular associations or singular causings that occur with regularity, where we may, if we wish, allow counterfactual as well as actual regularities or add the proviso that the regularities in question must occur ‘by necessity’. Laws hold as a consequence of the repeated, successful operation of what, I shall argue, is reasonably thought of as a nomological machine.15 (Cartwright, 1999: 4).
What sorts of things fall under Cartwright’s conception of laws as outlined here? This seems to me to be an important question so far as the assessment of her arguments for the claim are concerned, and the answer is not altogether clear, despite the apparently explicit specification offered in the quotation above. That
14
See (Steward) 2012. A nomological machine is “a fixed (enough) arrangement of components, or factors, with stable (enough) capacities that in the right sort of stable (enough) environment will, with repeated operation, give rise to the kind of regular behaviour that we represent in our scientific laws” (Cartwright, 1999: 50).
15
174
H. Steward
specification is consistent with Cartwright’s indication a little later on that the laws she believes must hold only ceteris paribus are the laws of “the liberalised Humean empiricism of most post-logical-positivist philosophy of science: a law of nature is a necessary regular association by properties antecedently regarded as OK” (p. 49). Empiricists differ, she explains, about what properties they take to be OK – but “the usual favourites are sensible properties, measurable properties and occurrent properties” (p. 49). In reading this description (though with a pause for thought about ‘measurable’), I at first supposed that Cartwright must have in mind laws of the sort which can be thought of as universally quantified conditionals of roughly the ‘whenever you have a this you get a that’ variety. But these sorts of laws seem to me to deserve a somewhat different treatment from a second variety of law also much discussed by Cartwright. The second variety of law encompasses the sorts of things which are more likely actually to figure as laws in physics textbooks – laws such as Newton’s laws of motion, Coulomb’s law, and the like – which use precise, abstract concepts, such as ‘force’, ‘mass’ and ‘charge’, properties which are not straightforwardly observable by means of the unaided senses (though they may correlate in interesting ways with ones that are). However, with respect specifically to Newton’s second law of motion, ‘F ¼ ma’, Cartwright writes as follows: Most of us, brought up within the fundamentalist canon, read this with a universal quantifier in front: for any body in any situation, the acceleration it undergoes will be equal to the force exerted on it in that situation divided by its inertial mass. I want instead to read it, as indeed I believe we should read all nomologicals, as a ceteris paribus law. (p. 25).
It seems on the face of it, then, as though Cartwright wants to insist that laws of both varieties – the ‘empiricist’ kind and the ‘textbook’ kind, as I shall henceforth refer to them for convenience’s sake, are only ever true when modified by a ceteris paribus clause. I believe, however, that despite initial appearances, this is not the best way to understand Cartwright’s considered view of the ‘textbook’ laws. I shall therefore proceed by considering the two kinds of law separately in turn. My claim will be that it is much harder to show of the ‘textbook’ laws than the ‘empiricist’ ones that they can only be considered true when qualified by a CP clause (and despite appearances, as I shall explain shortly, it seems to me that Cartwright would actually agree). For this reason, I am doubtful whether the ‘ceteris paribus all the way down’ solution to the problem of Law Compatibility can be confidently endorsed, since it does not seem to me plausibly endorsable for a range of ‘textbook’ laws. However, thinking about the way in which the world seems definitively bound by the ‘textbook’ laws helps us see what we need to say instead about laws, in order to deliver what the libertarian really needs. Let us begin with empiricist laws which – let us suppose for argument’s sake – encode regularities amongst sensible properties or collections of such. Cartwright’s view is that even physical laws in this description-of-regularities sense hold only ceteris paribus. Take for instance the observable regularities that characterise the motions of balls across a billiard table. An experienced player might know from years of experience laws such as the following: ‘If the cue ball is hit with just this
9 Laws Loosened
175
force,16 in just this position, towards another ball at just this distance in this direction from the first, the second ball will move off in just this direction at this speed and come eventually to rest here’. But Cartwright’s point is that this kind of law can only ever be true for the most part, and hence ceteris paribus, unless we explicitly specify the absence of interference and prevention by other systems.17 A person might, for example, lean over the table and pick up the second ball. The light suspended overhead might break and crash down on the table, smashing the second ball. An earthquake might occur, cracking the table so that the balls fall on the floor and roll across it. And so on. And then the universally quantified conditionals which describe our empiricist regularities would fail to be borne out. No empiricist regularities of the sort that characterise well-behaved ‘nomological machines’ – like those composed of billiard tables, balls, cues and people equipped to use those cues, playing the game under the normal rules and conditions – are so secure that they cannot be rudely disrupted by external interference. Cartwright seems to be right about the fact that the possibility of prevention of, and interference with effects in one system by the machinations of another is ubiquitous. But on the face of it, what Cartwright says here about the regularities which characterise nomological machines appears perfectly consistent with the reign of global universal determinism. Even if universal determinism were true, it might still be that the observable regularities we specify and rely upon in everyday life, or even in science, are only ceteris paribus because we do not generally need to specify the descriptions which figure in the antecedents of our universally quantified conditionals to the level of detail that would be required to rule out the operation of any interferers. Moreover it seems likely that we could not do so even if we tried, for reasons of informational complexity. But in principle, one might think, had we but world enough and time, it could be done. For example, I might be able, by describing the area surrounding the billiard table in meticulous detail, to rule out the presence of a person who might pick up a ball or the possible breaking of the frayed cables which fix the lights to the ceiling. I would need to go further to rule out the earthquake, no doubt – and indeed as possible interference and prevention scenarios suggest themselves to one’s imagination, it becomes quickly apparent that the antecedent of any conditional with a chance of not being true merely ceteris paribus would need to pack in vast quantities of detail over huge regions of space. But in principle: why not? If we had the regularities properly stated, someone might think – while conceding that this will forever be quite impossible in practice – the regularities could in principle be strict laws, quite consistently with Cartwright’s point about the need to rely on ceteris paribus laws even in physics for all practical purposes. It is true that If necessary for the purposes of according with the wanted ‘empiricist’ conception of properties that are ‘OK’ we can treat ‘force’ here as pertaining to a phenomenologically available property, such as e.g., felt pressure, impact, effort, motion, or whatever, rather than the abstract concept of Newtonian physics. 17 Anscombe was also mindful of the importance of this point – see the final sentence of her (1971) “The most neglected of the key topics in this subject are: interference and prevention” (147). 16
176
H. Steward
what we would end up with via this route, were it possible to take it, would presumably be something that looked very little like a collection of laws – and much more like a collection of extremely specific conditionals of the ‘if you have exactly this global situation at t1, then you would get exactly this global situation at t2’ variety. These conditionals would lack the useful generality of laws, because although in principle they would be applicable to more than one situation sharing a certain general character, that general character would have become so specific that the chances of it arising more than once would be negligible.18 But nevertheless, if such highly specific laws governed every situation, they would imply the rule of universal determinism. And hence, the claim that even the laws of physics are ceteris paribus – if we are conceiving of them broadly as empiricist laws of the kind specified by Cartwright – seems not to imply the falsity of universal determinism. One might wonder, therefore, whether this is a view of laws that can really help the libertarian. I think, though, that this dismissal of the capacity of the ‘ceteris paribus all the way down’ strategy to make space for libertarian free will would be too quick. The charge that libertarian free will must flout Law Compatibility only has any power to undermine libertarianism if the laws whose compatibility with free will are in question are laws that we actually have reason to believe are scientifically established. We do have reason to believe in general regularity principles such as the ones for example, which I said above would be known by any experienced billiard player – and the thousands of other such principles on which we base our daily expectations; and so it is important that, as libertarians, we ensure that our doctrine can accommodate those regularities. But do we have reason to believe that we could in principle complete the empiricist-style laws so as to entirely exclude the need for any ceteris paribus clauses? Cartwright herself insists that we have no such reason. Her view is that it is merely an unjustified article of faith that they could be produced, even in principle. We actually possess no such empiricist laws of whose non-CP status we can be entirely certain at the present time. And it would be unreasonable to demand of the libertarian that she show that her view is consistent with a doctrine that is merely speculative – the doctrine that the CP regularities we all know and love can be successfully turned into the detailed specifications of deterministic relations between global world-states. So far so good for Law Compatibility, then, if Cartwright is correct. But someone might allege that this result has only been obtainable because we have focused on the wrong sorts of laws of physics to begin with. As well as the empiricist laws which describe the regular workings of Cartwright’s nomological machines, one might think, we must consider the underlying laws which one actually finds in physics textbooks – things like Newton’s second law of motion, F ¼ ma, say, or Coulomb’s Bertrand Russell makes something rather like this point: “In order to be sure of the expected effect, we must know that there is nothing in the environment to interfere with it. But this means that the suppose cause is not, by itself, adequate to insure (sic) the effect. And as soon as we include the environment, the probability of repetition is diminished, until at last, when the whole environment is included, the probability of repetition becomes almost nil” (1912: 179–80).
18
9 Laws Loosened
177
Law, which says that the force between two charges having magnitudes q1 and q2 and separated by a distance r is equal to kq1q2/r2, where k is a constant. There are questions, of course, about what the correct ‘underlying’ laws actually are – for example, it might be said that we already know that Newton’s second law is not universally applicable since it only holds within certain specifiable limits.19 But provided we can specify those limits, it does not seem to be exactly a ceteris paribus law in virtue of those limitations. And in any case, we need not fixate on the laws as we have them. The idea would rather be that some future completed physics might in principle deliver the list of laws that together govern the universe – and that there is no reason to think that these would have to be ceteris paribus. Does Cartwright think that these other sorts of laws – the physics textbook kind of laws – hold only ceteris paribus? In the quotation I gave above, she seemed to say so. But as her argument develops, it becomes clear, I think, that she is (or was then) prepared to countenance – and perhaps even prefers – a second possibility – one which indeed she appears to endorse eventually for the case of ‘F ¼ ma’. Here is what she eventually says about that case: If the laws of mechanics are not universal, but nevertheless true, there are at least two options for them. They could be pure ceteris paribus laws . . .And that’s it. Nothing follows about what happens in different settings or in case where other causes occur that cannot be brought under the concepts of the theory in question. Presumably this option is too weak for our example of Newtonian mechanics. When a force is exerted on an object, the force will be relevant to the motion of the object even if other causes for its motion not renderable as forces are at work as well, and the exact relevance of the force will be given by the formula ‘F ¼ ma’ . . . . For cases like this, the older language of natures is appropriate. It is in the nature of a force to produce an acceleration of the requisite size . . . even when other forces are at work, it will ‘try’ to do so. (p. 28).
A reasonable understanding of the view which apparently here constitutes Cartwright’s second option would seem to be that at least some of these textbook laws are not, on due reflection, to be considered ceteris paribus. They hold absolutely – but in order to find them to be absolute, we must not interpret them as universally quantified conditionals about what always actually happens. Rather, we must instead interpret them as claims made true by the tendencies and capacities of things – as claims about what things will ‘try’ to do, even though their tryings may be overwhelmed by the ‘tryings’ of other things whose powers also impinge on the situation. And thus construed, there seems no obvious reason to insist that these textbook laws are true only ceteris paribus. Rather, they state with some precision a certain kind of truth about what will always and without exception be found to be the case concerning the tendencies and capacities of things to contribute to an overall result. There has been a lively debate in the literature on powers and dispositions about the extent to which we might be justified in endorsing the reality of such component
19
One might also raise the question, of course, as does De Hahn (this volume), whether there are any such fundamental laws of physics at all. But if there are not, then the arguments of those who suppose that the reign of such laws is incompatible with free will will not even get off the ground.
178
H. Steward
‘contributions’ to effects as Cartwright appears to be envisaging in the passage just discussed (Molnar, 2003; Mumford, 1998, 2009; Wilson, 2009; McKitrick, 2010). Some contributors to that debate (and especially McKitrick, 2010) have questioned whether we should accept the literal existence of such ‘contributions’ and Cartwright herself has indeed seemingly become uneasy about them, conceived of as a way of understanding how capacities might act together in general (Cartwright & Merlussi, 2018). As the authors of that paper put it: Nature may assign each capacity its own role, a role that it has qua the capacity it is; and nature may fix what happens when capacities act in consort in given circumstances. But nature need not do this via a simple model where each capacity separately produces its own canonical effect, and what results overall just is all these separate effects piled up together (Cartwright & Merlussi: 240).
It appears, then, that Cartwright now thinks that it is a mistake to try to understand the way in which different laws and principles interact to produce effects in terms of realistically-construed ‘contributions’ which certain capacities always make to the scenarios in which they are exercised. But note that she still seems inclined to countenance the idea that it might, at any rate, be a serious possibility that “nature may fix what happens when capacities act in consort in given circumstances”. And such a view would still be perfectly compatible, one might think, with the existence of many textbook laws which were ‘absolute’ in my sense of that word – and depending on the nature of the ‘fixing’ that Cartwright here has in mind, may indeed entail the existence of many such laws. I cannot here enter the complex debate about the metaphysics of contributions – but for present purposes, we should not need to do so. The dialectic here is this: we are currently searching for an account of the fundamental ‘textbook’ laws of physics which might, if it were true, give us a prima facie reason for suspecting that their joint reign is incompatible with the existence of libertarian free will – a view according to which we are to think of them as ‘absolute’, rather than as ceteris paribus laws. And so far as this search is concerned, we can afford, I think, to be indifferent about how exactly the underlying metaphysics might provide for the absoluteness – whether by realistically-construed ‘contributions’ which are always identical, given that the identical capacity is at work, or in some other manner. The crucial point so far as our current interests are concerned is that that the view we consider should imply that a certain kind of absolute truth about the relationships between such things as the values of variables is given by the ‘textbook’ laws – whatever the underlying metaphysical explanation of why that representation is absolutely correct may be. If such a view can be maintained – and I do not think we have yet seen any reason for thinking it might not be the correct view of the laws of a ‘completed’ physics – it will not be appropriate to say that any such laws of physics would be true only ceteris paribus. The laws would simply hold without any qualification (at any rate once we have specified them correctly and with sufficient precision). It is this view whose compatibility with the existence of libertarian free will I now wish, for the purposes of the final section of the paper, to defend – because I think it is the view which best captures the vision of the rule of law which tempts
9 Laws Loosened
179
those such as Coyne to believe, as he puts it, that the view ‘puts paid’ to the traditional idea of free will. Suppose then – as I think many would likely find compelling – that the universe is indeed governed by a number of laws of this textbook variety which hold absolutely and universally across the domains of things which possess the tendencies and capacities to which those laws have relevance (however that absolute character is to be explained by the underlying metaphysics). What must the libertarian say to accommodate these absolute and universal fundamental physical laws? In the final section of this paper, I turn to outline what seems to me to be the right kind of answer to this question.
9.4
Laws as World-Constrainers vs Laws as World-Dictators
All of the strategies I have looked at so far have attempted to accommodate free will by offering an adjusted conception of what the laws say. The laws say something not about what will happen but about the chances of things happening; or they say that something is true only for the most part; or in idealised circumstances. My own suggestion – which in some ways, as I said at the outset of the paper, strikes me as the most obvious possibility for the defence of Law Compatibility – is different from any of these, in that it is focused not on what the laws say – but rather on the nature of their relation to reality. What I suggest is that the libertarian needs to make use of the idea that natural laws constrain without dictating the course of the world to which they apply. There is no reason to think that constraining laws of what I have been calling the ‘textbook’ variety might not be absolute – where they hold, they may hold without exception, thought of as abstractly rendered characterisations of the tendencies and capacities of things; or indeed as abstract facts expressing relations between universals; or in some other way. The crucial point is that they need not be probabilistic and they need not be ceteris paribus – one can imagine them to be as strict and exceptionless as one likes. But this still by no means implies that those laws in their totality dictate a single course for reality (as shown by my earlier example of the world in which there is just a single and absolutely strict law concerning colour changes). Of course, our own world contains many more laws than this imaginary one. But why need the totality of all the laws there are, even in our strongly law-governed world, be world-dictating, rather than worldconstraining? It is this point which I think represents the obvious and neglected means of securing Law Compatibility for the libertarian. This suggestion takes seriously the metaphor on which talk of laws is based. The laws of a nation-state constrain its inhabitants; there are many things they may not do without contravening the law. But there is no jurisdiction so severe that it dictates to its subjects or citizens a single and precise course of action, prescribed in every detail of timing and execution. Many laws say merely what may not be done – but
180
H. Steward
even where laws insist that citizens engage in particular positive actions – that they must vote, say, or register the births of their children, they leave a certain leeway to the citizens about when precisely they must cast that vote (perhaps within a certain 16 hour period, say) or register that birth (within the first 42 days). Laws of the societal kind are constrainers, not dictators. They hem us in in different ways, but much freedom remains for us to operate differently while remaining within the bounds they set. My suggestion is that the affordances of this metaphor present the most obvious way forward as regards the defence of Law Compatibility. We should regard the laws of nature as being like the laws of a nation in that they restrict the possibilities. Because of the collective heft of those laws, I cannot fly unaided, burst into flame spontaneously at will or send telepathic messages to Donald Trump. But what laws are, mainly, according to the libertarian, is constrainers. Moreover, there is no reason not to suppose that a good number of them are perfectly uniform constrainers and in that sense are more like deterministic than probabilistic laws – they determine what will happen, not as regards observable outcomes, to be sure – but as regards the tendencies and capacities at work in the world, or as regards other underlying regular quantitative relationships between variable factors which must be observed. When they hold in a domain, nothing can happen that contravenes them. In that sense, they are universal and absolute – neither probabilistic nor ceteris paribus. But the libertarian should say that even the totality of all the laws of this kind that there are leaves much unsettled, just as the totality of societal laws leaves many legal ways available in which to live one’s life. If this is how laws relate to the world, one must work very hard to develop even a prima facie case against Law Compatibility. One would need not merely to wave one’s hand vaguely in the direction of the rule of the laws of physics, but rather to show in great detail how the particular constraints imposed by the particular laws leave us no leeway of the appropriate kind. In a sense, as I said at the outset of this paper, I think it is baffling that this extremely straightforward move is almost never made. But in another sense, I am not baffled. I know why it is not made and I know only too well the thought process that leads there. The thought process is this: – suppose some situation in the world and suppose it is governed by merely constraining laws which do not limit to one outcome the possibilities which might occur at a given time, t. A variety of possibilities then remains as to what will occur at that time. What, then, determines the resolution of those multiple possibilities into one actuality? What makes it the case that the world goes the way it does? If the totality of laws does not settle things, it seems the settling must be at least to some extent a chancy matter – and we are back to the problem that we discussed earlier of how chanciness could be of any possible help to the libertarian. The problem is to understand the explanation of the actual – as we might say, the explanation of how actuality is determined – so as to make space for an answer outside the unappealing disjunctive possibility of ‘laws or chance’. The problem here, in my view, is that we have left ourselves without the resources for offering this explanation because for philosophical (though not for everyday) purposes, we have largely abandoned the Aristotelian metaphysics of things with
9 Laws Loosened
181
powers which might give us the means of explaining how the resolution of possibility into actuality is effected. On the Aristotelian view, we are able simply to say that it is the doings of substances, the actions and interactions of the many different things which our lumpy and deeply non-homogeneous world contains, that explains what actually happens. But if, with the post-Humean metaphysical consensus, we eschew things and their powers, we are left only with an ontology containing such powerless things as events or tropes or particular facts – things which can only produce other such events or tropes or particular facts if we suppose them corralled into doing so by the operative laws. The key to arguing for Law Compatibility via this route, in my view, therefore involves ditching the Humean metaphysics which entails that there is nothing but laws to serve as the engines of reality. If laws are all you have available to explain how the world evolves through time then natural necessitation and explanation become the same thing – and the failure of the first becomes the failure also of the second. I am not able to argue here for the reinstatement of a neo-Aristotelian ontology of things with powers. But I hope that there is at least an indirect argument for it lurking in the fact that it seems to be the only metaphysics which can properly accommodate both strict deterministic law of the ‘textbook’ variety and libertarian free will. What I hope to have argued for more directly is the claim that neither the resort to probabilistic laws, nor the Kenny-List invocation of the non-strict character of psychological laws is at all promising for the libertarian hoping to defend Law Compatibility. In Cartwright’s views, I believe, we have a range of more promising proposals – but what appears to be the best version of the Cartwrightian ‘loosening’ proposals needs supplementation with the explicit recognition of the idea that we need to move from the conception of laws as world-dictators to laws as worldconstrainers.
References Almeida, M., & Bernstein, M. (2003). Lucky libertarianism. Philosophical Studies, 113, 93–119. Anscombe, G. E. M. (1958). Intention. Blackwell. Anscombe, G. E. M. (1971). Causality and determination. In Metaphysics and the philosophy of mind: Collected philosophical papers (Vol. II, pp. 133–147). Blackwell. Beebee, H. (2000). The non-governing conception of laws of nature. Philosophy and Phenomenological Research, LXI 3, 571–594. Cartwright, N. (1999). The dappled world. Cambridge University Press. Cartwright, N., & Merlussi, P. (2018). Are laws of nature consistent with contingency’? In W. Ott & L. Patton (Eds.), Laws of nature. Oxford University Press. Coyne, J. (2014). What scientific idea is ready for retirement?: Free will. The Edge. https://www. edge.org/response-detail/25381. Accessed 23 Dec 2020. Davidson, D. (1970). Mental events. In L. Foster & J. W. Swanson (Eds.), Experience and theory. Duckworth. Reprinted in his Essays on actions and events (pp. 207–227). Oxford: Oxford University Press. Davidson, D. (1973). The material mind. In P. Suppes, L. Henkin, G. C. Moisil, & A. Joja (Eds.), Proceedings of the fourth international congress for logic, methodology and philosophy of
182
H. Steward
science. Reprinted in his Essays on actions and events (pp. 245–259). Oxford: Oxford University Press. Davidson, D. (1974). Psychology as philosophy. In S. C. Brown (Ed.), Philosophy of psychology. Macmillan. Reprinted with comments and replies in his Essays on actions and events (pp. 229–244). Oxford: Oxford University Press. De Haan, D. (2022). How to re-think about the problem of free will without laws of nature. In Powers, Time and Free Will. Springer, this volume. Earman, J. (1986). A primer on determinism. Reidel. Fara, M. (2005). Dispositions and habituals. Nous, 39, 43–82. Fodor, J. (1987). Psychosemantics: The problem of meaning in the philosophy of mind. MIT Press. Glynn, L. (2010). Deterministic chance. British Journal for the Philosophy of Science, 61, 51–80. Haji, I. (1999). Indeterminism and Frankfurt-type examples. Philosophical Explorations, 2, 42–58. Honderich, T. (1988). Mind and brain: A theory of determinism (Vol. 1). Clarendon Press. Hossenfelder, S. (2019). How to live without free will. Retrieved from https://backreaction. blogspot.com/2019/05/how-to-live-without-free-will.html Huttemann, A. (2022). The problem of radical freedom. In Powers, time and free will. Springer, this volume. Kane, R. (1996). The significance of free will. Oxford University Press. Kane, R. (1999). Responsibility, luck, and chance: Reflections on free will and indeterminism. Journal of Philosophy, 96(5), 217–240. Kane, R. (2008). Incompatibilism. In T. Sider, J. Hawthorne, & W. Zimmerman (Eds.), Contemporary debates in metaphysics (pp. 285–302). Blackwell Publishing. Kane, R. (2011). Rethinking free will: New perspectives on an ancient problem. In R. Kane (Ed.), The Oxford handbook of free will (pp. 381–401). Oxford University Press. Kenny, A. (1963). Action, emotion and will. Routledge. Kenny, A. (1975). Will, freedom and power. Blackwell. List, C. (2019). Why free will is real. Harvard University Press. McKitrick, J. (2010). Manifestations as effects. In A. Marmodoro (Ed.), The metaphysics of powers: Their grounding and their manifestations. Routledge. Mele, A. (1995). Autonomous agents. Oxford University Press. Mele, A. (2006). Free will and luck. Oxford University Press. Molnar, G. (2003). Powers: A study in metaphysics. Oxford University Press. Mumford, S. (1998). Dispositions. Oxford University Press. Mumford, S. (2009). Passing powers around. The Monist, 92, 94–111. Pereboom, D. (2001). Living without free will. Cambridge University Press. Pereboom, D. (2005). Defending hard incompatibilism. Midwest Studies in Philosophy, 29, 228–247. Pereboom, D. (2014). Free will, agency, and meaning in life. Oxford University Press. Russell, B. (1912). On the notion of cause. In Mysticism and logic (pp. 173–199). Unwin. Ryle, G. (1949). The concept of mind. Penguin. Smith, M. (1997). A theory of freedom and responsibility. In Ethics and the a priori (repr.) (pp. 84–113). Cambridge University Press. Steward, H. (2012). A metaphysics for freedom. Oxford University Press. Steward, H. (2021). What is determinism? Why we should ditch the entailment definition. In M. Hausmann & H. Noller (Eds.), Free will: Historical and analytical perspectives. Palgrave Macmillan. Stoutland, F. (1986). Reasons, causes and intentional explanations. Analyse and Kritik, 8, 28–55. Strawson, G. (1994). The impossibility of moral responsibility. Philosophical Studies, 75, 5–24. Tanney, J. (1995). Why reasons may not be causes. Mind and Language, 10, 105–128.
9 Laws Loosened
183
Vetter, B., & Jaster, R. (2017). Dispositionalist accounts of abilities. Philosophy Compass, 12(8), 1–11. Vihvelin, K. (2004). Free will demystified: A dispositional account. Philosophical Topics, 32, 427–450. Whittle, A. (2010). Dispositional abilities. Philosophers’ Imprint, 10, 1–23. Wilson, A. (2009). Disposition-manifestations and reference-frames. Dialectica, 63, 591–601.
Chapter 10
The Problem of Radical Freedom Andreas Hüttemann
Abstract Whether or not we are able to do x is on many philosophical accounts of our moral practice relevant for whether we are responsible for not doing x or for being excusable for not having done x. In this paper I will examine how such accounts are affected by whether a Humean or non-Humean account of laws is presupposed. More particularly, I will argue that (on one interpretation) Humean conceptions of laws, while able to avoid the consequence argument, run into what might be called “the problem of radical freedom”: Humean laws fail to constrain what we can do. By contrast, non-Humean laws (and Humean laws on a second interpretation) avoid the problem of radical freedom but have no easy way out of the consequence argument. Keywords Responsibility · Humean accounts of laws · Radical freedom · Consequence argument
10.1
Introduction
In the empiricist tradition it is sometimes argued that based on of what is now called a Humean conception of laws of nature the problem or at least the tension that is generated for free will by deterministic laws can be easily dissolved. Thus, Ayer wrote
This paper not only rehearses part of a joint paper with Christian Loew, the idea of the paper grew out of joint discussions (see also Hüttemann & Loew, 2019). So insofar as this paper contains convincing thoughts, they can also be attributed to Christian Loew. I would also like to thank audiences in Oxford, the DFG-Research Group on Inductive Metaphysics, the Colloquium in Cologne, Anna Marmodoro as well as an anonymous referee for helpful comments. A. Hüttemann (*) Philosophisches Seminar der Universität zu Köln, Köln, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_10
185
186
A. Hüttemann
But, I repeat, the fact is simply that when an event of one type occurs, an event of another type occurs also, in a certain temporal or spatio-temporal relation to the first. The rest is only metaphor. And it is because, of the metaphor, and not because of the fact, that we come to think that there is an antithesis between causality and freedom. (Ayer, 1954: 283)1
In this paper I will focus on the consequence argument – one attempt to articulate the tension between free will and deterministic laws. In particular I will examine how one premise of the argument, according to which the laws are not ‘up to us’, is affected by whether one adopts a Humean or a Non-Humean account of laws of nature. After contrasting Humean and non-Humean conceptions of laws of nature (Sect. 10.1) I will briefly examine why the consequence argument is a particular challenge for the non-Humean (Sect. 10.2). In Sect. 10.3, I will rehearse an argument (by Loew and Hüttemann (forthcoming) to the effect that the Humean has a way out of the consequence argument that is not open to the non-Humean. However, as I argue in Sect. 10.4, the very same feature that allows the Humean to bypass the consequence argument makes her vulnerable to an equally vexing problem, the problem of radical freedom. It will turn out that the Humean cannot claim both that she is able to bypass the consequence argument and to solve the problem of radical freedom. Depending on the Humean’s understanding of necessity or constraint, she can either avoid the consequence argument but will then face the problem of radical freedom, or she is in the same position as the non-Humean and can cope with the problem of radical freedom but has to face the full force of the consequence argument.
10.2
Humean and Non-Humean Conceptions of Laws
Humeanism (or Humean Supervenience) writes David Lewis “is named in honor of the greater denier of necessary connections.” What is rejected here is that there are relations of necessity or constraints that obtain between distinct events. This denial is the essential feature of Humeanism that will be appealed to in the remainder of the paper. Lewis elaborates his view: It is the doctrine that all there is to the world is a vast mosaic of local matters of particular fact, just one little thing and then another. [. . .] perfectly natural intrinsic properties, which need nothing bigger than a point at which to be instantiated. For short: we have an arrangement of qualities. And that is all. All else supervenes on that.” (Lewis, 1986: ix–x)
All there is an arrangement of non-modal categorical properties, the so-called Humean Mosaic. The claim that all else supervenes on the Humean Mosaic has to be understood as saying that whatever else there is it is nothing over and above the
1
Similar views have been voiced e.g. by Swartz (2003), Beebee and Mele (2002) and Perry (2004)
10
The Problem of Radical Freedom
187
Mosaic. Supervenience is not to be understood as mere modal covariation but rather as a claim of ontological reduction.2 I will, therefore, also use the term ‘Humean reductionism’. Humean accounts of laws of nature illustrate how Humean reductionism is supposed to work. The central idea is to characterise laws of nature as an elite class of true generalisations, where whether or not a generalisation is an element of the elite class is completely determined by the Mosaic. Thus, suppose we have the following generalisation: (G) 8s 8t (If a system s is in state S1 at t it will develop into state S2 at t + Δt) If (G) is a true generalisation it describes and is made true by a regularity or pattern in the Mosaic. According to Lewis’ account, (G) is not only a true generalisation but also a law of nature “if and only if it appears as a theorem (or axiom) in each of the true deductive systems that achieves a best combination of simplicity and strength.” (Lewis, 1973, 73) There is no need to go into the notions of simplicity and strength. The important point is that what constitutes the best combination of these theoretical virtues is fixed by the Mosaic. There are other Humean conceptions of laws which replace or complement simplicity and strengths by other features (e.g., Beebee (2000), Cohen and Callender (2009), Dorst (2018), Hall (ms), Hicks (2018) and Jaag and Loew (2020), however what is essential is that they all characterise laws exclusively in terms of the Humean Mosaic. Talking about laws does not introduce anything over and above the Mosaic. In this sense then, the laws have been reduced to the Mosaic. By contrast, according to Non-Humean conceptions, laws of nature cannot be reduced to the Humean Mosaic. Non-Humeans argue that modal connections between distinct events have to be introduced in order to account for laws of nature and their role in scientific practice. It is argued that modal connections between distinct entities have to be introduced. A law of nature that holds in virtue of these modal connections implies not only a true generalisation such as (G), it furthermore implies that, e.g., s being in state S1 at t constrains the system such that the system cannot but develop into state S2 at t + Δt, the state at t + Δt is necessitated by the law, given the initial state. Non-Humeans account for the constraint or the necessitation in terms of essences, dispositions, necessitation relations or in terms of primitive modal relations (Armstrong, 1983; Bird, 2007; Mumford & Anjum, 2011; Maudlin, 2007). The details of these accounts need not bother us for the purposes of this paper. What they share is the view that laws cannot be reduced to the Humean Mosaic because what is constitutive for laws is at least partly something over and above the Mosaic, namely relations of constraint or necessitation (I use these terms synonymously in this paper) between distinct events.
2
“Imagine a grid of a million tiny spots – pixels – each of which can be made light or dark. When some are light and some are dark, they form a picture, replete with intrinsic gestalt properties. [. . .] the picture and the properties reduce to the arrangement of light and dark pixels. They are nothing over and above the pixels.” (Lewis, 1994: 413).
188
A. Hüttemann
On the basis of these brief characterisations the essential disagreement between the Humean and the non-Humean can be illustrated as follows: If it is a law that 8s 8t (If a system s is in state S1 at t it will develop into state S2 at t + Δt) the non-Humean holds that if some s is indeed in S1 at t there is a relation of constraint or necessitation such that s will develop into state S2 at t + Δt, while the Humean denies that there is such a relation of necessitation or constraint and merely holds that as a matter of fact s develops into state S2 at t + Δt. Even though the Humean denies the existence of modal connections on the fundamental level of the Mosaic, this does not mean that she has to abstain from modal talk altogether. Modal notions can be reintroduced at levels further up – as something that can be reduced to the Mosaic. For instance, once we have introduced a Humean conception of laws we can define nomological necessity in terms of these laws. Events can be said to be nomologically necessary if they are implied by certain initial conditions and the laws of nature. On the basis of this conception of nomological necessity the Humean can explain what it means that one event constrains what is going to happen next (in virtue of a law that pertains to these events). The Humean may thus argue that nomological necessity has been reduced to the Humean mosaic and that she is able to account for the way we modally characterize what is going on in world. For the purposes of this paper we do not have to settle the question whether such a reduction (if feasible) shows that these higher-level modal features are real or that they are mere projections onto the world (or mere “metaphors” as Ayer put it). However, what is important for what follows, is that it cannot be the case that relations of necessity or constraint are on the one hand denied by the Humean but then on the other hand reintroduced and thus affirmed via reduction in the very same sense. The Humean must operate with two different senses of necessity and constraint. I will indicate this difference by contrasting fundamental and less fundamental (higher-level) necessity or constraint: necessityfun vs. necessityless-fun and constraintfun vs. constraintless-fun.
10.3
The Non-Humean Conception of Laws and the Consequence Argument
The consequence-argument is a serious challenge for the Non-Humean, precisely because she postulates relations of constraintfun. A well-known, informal presentation of the argument is due to van Inwagen: If determinism is true, then our acts are the consequence of laws of nature and events in the remote past. But it’s not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us. (van Inwagen, 1983: 56)
10
The Problem of Radical Freedom
189
Before I turn to the question of why this is a particular challenge for the non-Humean, let me briefly comment on the phrase up to us. According to van Inwagen this can be understood as us having the ability to render a proposition false, where: to be able to render a proposition false is to be able to arrange or modify the concrete objects that constitute one’s environment – shoes, ships, bits of sealing wax – in a way sufficient for the falsity of that proposition. (van Inwagen, 1983: 67).
To illustrate, the fact that the number of chairs in my room is two is up to me because I have the ability to render the proposition that the number of chairs in my room is two false by being able to bring in another chair – even if I don’t exercise this ability. By contrast that 2 plus 2 equals 4 is not up to me because I cannot render the proposition that 2 plus 2 equals 4 false. Similarly, a true generalisation of the form 8x (Ax!Bx) is up to me provided I can render it false, e.g. by being able to produce an x that is an A but not a B. The consequence argument attempts to show that if determinism is true, then our actions are not up to us in the sense that we cannot do otherwise. The argument presupposes that there are deterministic laws that apply to our actions (Helen Steward in her contribution to this volume explores some alternatives to this assumption). It furthermore depends on assumptions about the past (not up to us) and about laws (not up to us either).3 In what follows I will be exclusively concerned with whether the Humean conception of laws is relevant for this latter issue. I will thus bypass the question whether the rejection of the premise, according to which the past is not up to us, is helped by a Humean conception of laws. Furthermore, I will not discuss Lewis’ distinction of a strong and a weak reading of the laws being up to us, because Lewis’ distinction is motivated by his semantics of counterfactuals, not by his Humean account of laws (Lewis, 1981).4 Let us call the assumption that the laws are not up to us the ‘Fixity of the Laws’ Fixity of the Laws (FOL) If it is a law of nature that p, then it is not up to us whether p is true. (For a law not being up to us strictly speaking two conditions have to be met: It is neither up to us whether p is true nor whether p is a law. However, for the purposes of this paper all that is needed is that the truth of p is not up to us.)
3
These are by no means the only assumptions. There is an extensive debate about how the consequence argument should be understood that I don’t need to go into for the purposes of this paper. (Some of the options are reviewed in Mauro Dorato’s contribution to this volume. Esfeld (also this volume) explores whether the remote past may be up to us – contrary to what van Inwagen assumes.) What is relevant in this paper is that different conceptions of lawhood make a difference for whether or not the informal consequence argument discussed here comes out as plausible. (This question is also raised in the contribution by de Haan). 4 It has been suggested (in conversation both by Helen Beebee and by Thomas Blanchard) that Lewis’ semantics of counterfactuals may in turn be motivated by his account of laws. For the purposes of this paper I can bypass this issue.
190
A. Hüttemann
What would it mean to render a law-proposition false? Take the example of the law we considered above: if a system s is in state S1 at t it will develop into state S2 at t + Δt. It could, for example, be the case that I myself am system s and that I am in state S1 at t and subsequently in state S2 at t + Δt. To be able to render this law false, I would have to be able to bring about a state S3 that differs from S2 at t + Δt (even though I didn’t). For the Non-Humean the consequence argument is a particular challenge because she has to accept FOL. It is a defining feature of non-Humean conceptions of laws that the antecedent state constrainsfun what is a possible state for s. Thus, given the antecedent state S1, S2 is necessitatedfun. It is because of the necessity relation that I am unable to bring about S3 and thus unable to render the law-proposition false. This illustrates why the non-Humean is committed to FOL. By contrast, the Humean does not postulate a necessityfun relation and thus needs not commit herself to FOL (see the next section for an extended argument).
10.4
The Humean and the Consequence Argument
In what follows I will examine whether Humean compatibilism is a tenable option in the free will debate. Beebee and Mele (2002) who made this label prominent characterise a Humean compatibilist as holding the combination of a Humean account of laws and compatibilism in the free will debate. I will use the term more exclusively for someone who holds a compatibilist view in virtue of a Humean account of laws. This stricter conception excludes, for example, Lewis, who is both a Compatibilist and a Humean, but he is not a compatibilist in virtue of his Humean account of laws, but rather in virtue of his account of counterfactuals. Loew and Hüttemann (forthcoming), elaborating on a paper by Beebee and Mele (2002), have argued that the Humean is in a better position than the Non-Humean to deal with the consequence argument because she is not committed to FOL. In this section I will rehearse the argument. As I indicated above, Humean reductionists hold that the laws of nature reduce to the Humean Mosaic. Loew and Hüttemann understand Humean reduction in terms of grounding. According to this interpretation of Humean reductionism, the laws of nature are fully grounded in the Humean mosaic. However, the argument could just as well be presented in terms of supervenience as understood by Lewis or in terms of realisation. All that is needed is a synchronic, asymmetric dependence relation that allows for a distinction between difference-making and non-difference-making factors.5
5
Supervenience is generally understood to be a non-symmetric rather than an asymmetric relation. Lewis, however, seems to think of it as an asymmetric relation. Asymmetry will be relevant only when I briefly discuss Armstrong towards the end of this section.
10
The Problem of Radical Freedom
191
Loew and Hüttemann provide the following argument for Humean compatibilism. The first premise is that ‘up to us’ obeys a transfer principle: Transfer Principle (TP) If p is up to an agent A and p is a difference-making ground of q, then q is up to A. Difference-making grounds are grounds that are non-redundant. Suppose the set F contains the complete grounds of q. A fact p is then a difference-making ground of q, just in case: (i) p is part of F, and (ii) without p the remaining facts in F would no longer be complete grounds of q (Krämer & Roski, 2017: 1195). Given the analysis of the phrase ‘up to us’ in terms of the ability to render a proposition false, TP says that if you have the ability to render p false and p is a difference-making ground of q, then you have the ability to render q false. For example, suppose you can render false the proposition that you raise your hand. And suppose that raising your hand is a difference-making ground of the fact that a certain candidate is elected into office. Then, according to TP, you can thereby render the proposition that the candidate is elected false. TP is a plausible principle that explains why we often prevent one thing by not doing another.6 While TP is plausible independently of whether Humeanism about laws is true, the second premise is motivated by a Humean analysis of laws: Law-making (LM) Our actions are among the difference-making grounds of deterministic laws. Given the assumption – which is common in the consequence argument literature – that deterministic laws cover our actions, LM follows from the grounding-based formulation of Humean reductionism introduced above, viz., that laws of nature are grounded in the Humean mosaic. If Humean reductionism about laws is true, then laws of nature are elements of an elite class of regularities or generalisations. Like all universal generalisations, the truth of a generalisation like (G) is grounded in its instances (see, e.g., Fine, 2012). The generalisation ‘8s 8t (If a system s is in state S1 at t it will develop into state S2 at t+Δt)’ is grounded in: (i) all instances of S1; (ii) the instances of S2 that follow them after t; and (iii) the fact that these are all of the instances of S1. For example, suppose there are three instantiations of S1: S10 , S100 , and S1000 , and each is followed, within the relevant time period, by, say, a hand-raising, S20 , S200 , and S2000 . The
6
Formulating TP in terms of difference-making grounds rather than grounds simpliciter avoids certain counterexamples. Suppose you can render the proposition that you will raise your hand false and raising your hand is among the grounds of a certain candidate’s election. However, suppose the candidate received more votes than she needed to get elected. Even though raising your hand was then up to you and was also an ontological ground of the candidate’s being elected, you could not have rendered the proposition that she gets elected false. The candidate did not need your vote to be elected. Using difference-making grounds instead of grounds simpliciter bypasses this problem because in the scenario described raising your hand is not a difference-making ground. Without you raising your hand, the remaining votes still would be complete grounds of the candidate’s being elected.
192
A. Hüttemann
complete grounds of (G)’s truth then are: S10 , S100 , S1000 S20 , S200 , S2000 plus the fact that S10 , S100 and S1000 are the only instances of S1. Any particular hand raising of yours, S20 , say, is then among the grounds of (G). However, it remains to be shown that S20 is a difference-making ground of (G).7 Your hand raising, S20 , is not just a ground but a difference-making ground of (G). Without S20 , the remaining grounds are: S10 , S100 , S1000 , S200 , S2000 plus the fact that S10 , S100 and S1000 are the only instances of S1. These facts, however, are not complete grounds of (G) because they do not make (G) true: there is now an instance of S1 at t, viz., S10 , that is not followed by a hand raising within the requisite time period and so (G) is false. Hence, S20 is a difference-making ground of (G) because without it the remaining grounds are not complete grounds of (G). With TP and LM in place the following argument can now be formulated: 1. If p is up to an agent A and p is a difference-making ground of q, then q is up to A. (Premise (TP)). 2. If our actions (or rather: propositions characterising our actions) are up to us and these actions are difference making grounds of deterministic laws, then the laws are up to us. (from 1) 3. If the laws are not up to us, either our actions are not up to us or these actions fail to be difference making grounds of deterministic laws. (by contraposition from 2) Thus far Humeans and Non-Humeans will agree. If the laws are not up to us at least one of two conditions has to hold: Either our actions are not up to us or the actions fail to be difference making grounds for the laws. The relevant difference between the Humean and the Non-Humean is that the Humean affirms (by holding LM) that our actions are among the difference making grounds for laws whereas the Non-Humean denies this. So, for the Humean (3) implies that the laws fail to be up to us only if our actions are not up to us. The argument can thus be completed as follows: 4. Our actions are among the difference-making grounds of deterministic laws. (Premise (LM)) 5. If the laws are not up to us, then our actions are not up to us. (from 3 and 4) Given a Humean conception of laws it is a necessary condition for FOL to hold that our actions are not up to us. The above argument does not show that FOL is false, rather it shows that Consequence argument is dialectically ineffective given a Humean conception of laws. TP entails that if a fact q is not up to you, then no fact that is a difference-making ground of q is up to you. After all, if a difference-making ground of q were up to you, TP would entail that q itself is up to you. So, by assuming FOL as a premise and thereby assuming that deterministic laws are not up to us, the Consequence argument ipso facto presupposes that no difference-making grounds of the laws are up to us. But given that Humeans are committed to LM, this
7
Note that we are only concerned with the grounds of the fact that (G) is true. The additional fact that (G) is a law of nature has further grounds.
10
The Problem of Radical Freedom
193
assumption already entails (regardless of any of the other premises of the Consequence argument) that our actions are not up to us. Hence, by assuming FOL as a premise, the Consequence argument already presupposes what it is supposed to show. The Humean does not have to commit herself to FOL. Why couldn’t the Non-Humean employ the same kind of reasoning? The non-Humean is clearly entitled to argue that by assuming FOL as a premise and thereby assuming that deterministic laws are not up to us, the Consequence argument ipso facto presupposes that no difference-making grounds of the laws are up to us. However, these difference making grounds – in the case of the Non-Humean – are not the instances of the laws. Rather, the Non-Humean treats necessityfunrelations as difference-making grounds for laws. The laws obtain exclusively in virtue of essences, dispositions, necessitating relations, primitive modalities, etc.8 Our actions are not among the difference-making grounds of deterministic laws. In the context of the consequence argument it is not question-begging to assume that these features are not up to us. Whereas it would be question-begging to assume that our actions are not up to us. The Non-Humean – on standard accounts – does not and cannot accept LM and thus the above argument is not available to her. (One might think of two non-standard theoretical options for Non-Humeans to allow for embracing premise LM: (a) One might argue that the dispositions in virtue of which the relevant laws obtain are up to us after all because they are character dispositions which we form ourselves; (b) One might argue that the laws are only partly grounded in modalfun relations and partly grounded in facts that are up to us. I don’t think anybody has defended such views and will set them aside in the remainder of the paper.)
10.5
The Problem of Radical Freedom
One might worry that Humean Compatibilism has the implausible consequence that we have outlandish abilities, such as to travel faster than light. If we have the ability to render the laws false by doing otherwise than we are lawfully determined to do, why can we not also perform other law-violating actions? According to Beebee and Mele (2002: 212), “[t]his is a legitimate worry to have because the Humean View does indeed have those consequences.”9 Let us call the problem that we seem to have the ability to break the laws (by rendering them false) “the problem of radical freedom”.
8
Armstrong (1983, 82) requires that universals (and among them the necessitation relation) are instantiated, but still the instatiations, for instance our actions, obtain in virtue of the necessity relation, not vice versa. 9 Beebee and Mele go on to argue that while their Humean compatibilist arguments entail that there is a legitimate sense of ability according to which we do have outlandish abilities, there might be a different sense of ability according to which agents can do otherwise yet do not have outlandish abilities. But Humean compatibilists would then still have to show that this second sense of ability is indeed supported by a Humean metaphysics of laws.
194
A. Hüttemann
In the remainder of the paper I will argue that the problem of radical freedom is a serious problem for those who embrace the argument for Humean Compatibilism outlined in Sect. 10.3.10 However, the argument for Humean Compabilism (more precisely: the argument that shows that the consequence argument – an argument in favour of incompatibilism – is dialectically ineffective given a Humean conception of laws) does not entail that we have outlandish abilities such as moving faster than the speed of light. In fact, the argument by itself does not even show that we have the ability to do otherwise than we are lawfully determined to do. It only shows that if we have independent reasons for thinking that we are able to do otherwise, then the mere fact that doing so would falsify deterministic laws does not undermine this ability (henceforth: The Conditional). The argument implies a conditional claim and is as such not committed to the existence of any abilities (whether outlandish or not). However, even though the argument for Humean Compabilism is non-committal, the situation changes when it comes to the analysis of situations which are the dialectical starting point for discussions concerning the consequence argument. In such situations we (sometimes) do assume that agents have the ability to do otherwise, i.e. that certain actions are up to them. We then have both The Conditional plus a true antecedent claim, which then imply that the law is up to the agent. To illustrate how the above argument helps the Humean to get the right verdict on blaming, consider the following case: (NEWSPAPER) S, while standing close to his 4-year-old son is reading a newspaper. The son trips over a stone and harms himself. S continues to read the newspaper. We assume that S could have acted otherwise and we, thus, blame him for not helping his son. The consequence argument attempts to show that if the laws are deterministic (and a few further assumptions in place) this case of rightfully blaming S cannot be accounted for. Our actual practice (of blaming S) turns out to be undermined by deterministic laws, the incompatibilist concludes on the basis of the consequenceargument. The next move in the dialectic is that the Humean defends our original verdict (blaming S) by pointing out that the consequence-argument presupposes a Non-Humean conception of laws. The Humean need not commit herself to FOL, can thus reject the consequence argument and account for the moral practice of blaming S in a situation as described above. The context in which the defence of the original verdict takes place is a context in which we assume that S does have the ability to act otherwise than he actually does (and that we thus rightly blame him for not helping his son). In the defence of the original verdict the Humean commits herself to S having the ability to act otherwise. She then points out that this ability is not undermined by deterministic laws that cover S’s acts because the Humean holds that the law is grounded in the actions of the agents rather than in the existence of necessitating relations. According to the
10 As indicated above this is not a problem for Lewis’ proposal (Lewis, 1981) to deal with consequence argument.
10
The Problem of Radical Freedom
195
Humean it is not the case that the laws constrainfun and thus S has the ability to stop reading his newspaper even though, as a matter of fact, he does continue reading. In order to account for our practice of blaming S the Humean Compatibilist will endorse the claim that S can do otherwise because the law under which his continuing to read the newspaper falls, is up to S. By contrast the non-Humean appears to be forced by the consequence argument to conclude that S – given certain initial conditions that are not up to S and the deterministic laws – could not have done otherwise and thus ought to be exempted from blame (assuming the ability to act otherwise to be a necessary condition for responsibility). The Non-Humean cannot account for our practice of attributing responsibility in cases like the one described, while the Humean can. The consequence-argument is a problem for the Non-Humean but not for the Humean (though I will qualify this claim below). Let us now consider a second case. (SKID) S and his son visit a playground. They both use the skid in turn (a long and exciting slide). While S is sliding down the skid the son trips over a stone and harms himself. S continues to slide. Because S cannot simply stop, we would not blame S for not (immediately) helping his son. Assuming again that the ability to act otherwise is relevant for blame, it is now the Non-Humean who can explain why we exempt S. Given that S was already in the skid he couldn’t stop. Given the initial condition and the law of gravity S was necessitatedfun to continue. The Humean, might argue that in this case we do not assume that the antecedent of The Conditional is true. We assume that it is not up to S to stop sliding. Thus, in this case we have no reason to conclude that the law that covers the sliding is up to S. However, the question arises whether the Humean can explain in virtue of what the sliding (and thus the relevant law) is not up to S while in (NEWSPAPER) continuing to read (and the relevant law) were up to S. The Humean cannot appeal to a relation of necessityfun to account for the sliding not being up to S. There is no constrainfun that limits what S can do. It seems that given the Humeans’ analysis of laws S should be able to do whatever S likes – unconstrained by laws. This is what might be called the ‘problem of radical freedom’. The problem of radical freedom is a problem for the Humean but not for the Non-Humean. It now seems that the Humean and the Non-Humean score evenly. While the Humean can avoid the consequence argument she has to face the equally vexing problem of radical freedom. The Non-Humean does not run into the problem of radical freedom but has to face the full force of the consequence argument. The Humean might object that this assessment is not quite correct (see Loew and Hüttemann (forthcoming)). Humeans do allow for some notion of constraint and necessitation, and they could therefore try to resolve cases like SKID by relying on constraintnon-fun whilst still relying on the absence of constraintfun to respond to the consequence argument. According to this line of argument, there is no tension between having the ability to render the laws false and the laws constraining what we can do. This works because the first modal claim (about what we can or cannot do) concerns necessityfun
196
A. Hüttemann
or contstraintfun, while the second claim concerns constraintless-fun. By contrast, the Non-Humean, when discussing the necessity of laws or the laws’ constraining our action uses a univocal notion of necessity or constraint: constraintfun. It is the univocity that generates a problem for the Non-Humean: For her the laws cannot both be up to us and constrain what we can do. For the Non-Humean the laws constrainfun and are thus not up to usfun. For the Non-Humean there is indeed a tension between these two claims. However, the Humeans’ equivocal use of modalities generates a problem of its own. It gives rise to different senses of “ability”. Let us have a closer look at an argument by Beebee and Mele who explicitly use the strategy of distinguishing different abilities to reply to the problem of radical freedom (though they don’t use this latter phrase). Beebee and Mele concede that there is a sense of ability, supported by Humean Reductionism, according to which, agents can move faster than the speed of light and defy gravity. However, Beebee and Mele (2002: 213) maintain that: [t]he existence of the legitimate (by Humean lights) sense of ability already identified does not entail that there are no other legitimate senses of ‘ability’, consistent with Humeanism, according to which Fred is able to eat cake [even though he is nomically determined not to] but unable to raise his hand faster than the speed of light.
Humean Compatibilists are free to distinguish different senses of abilities. But doing so does not allow them to both maintain that the premise FOL of the Consequence Argument is false and avoid the problem of radical freedom. The core claim of Humean Compatibilism is that there is a sense of ability that can be motivated in light of a Humean metaphysics according to which FOL is false. Call this sense of ability “abilityfun.” According to this sense of ability humans can perform such outlandish actions as moving faster than the speed of light and jumping over large buildings – the problem of radical freedom. Beebee’s and Mele’s response is to invoke a second sense of ability according to which we can act otherwise than we are nomically determined to act yet we cannot perform these more outlandish violations of the laws of nature. Call this sense of ability “abilityless-fun.” (Beebee and Mele define abilityless-fun in terms of counterfactuals. But the details of the definition are irrelevant. What is essential – in order to avoid the tension that the Non-Humean has to confront – is that abilityless-fun needs to invoke relations of constraint that are non-fundamental, whether in terms of laws, causal structure, dispositions, and/or counterfactuals does not matter.) In a situation in which I do not raise my hand, there is no tension in saying on the one hand that I have the abilityfun to render false a law according to which I raise my hand, and on the other hand that the law constrains what I can do, and that I thus lack the abilityless-fun to raise my hand. Though there is no tension, there is still a problem. To illustrate, let us return to SKID. The question that needs to be raised is whether it is the abilityfun or abilitylessfun that we need to appeal to in order to account for not blaming S in SKID. S has the abilityfun to stop and help his son immediately but he does not have the abiltyless-fun to stop and help his son immediately. Since, clearly, we cannot both blame S and
10
The Problem of Radical Freedom
197
exempt him at the same time, we have to decide whether it is abilityfun or abilitylessfun that is morally relevant. If abilityfun is the morally relevant ability we should blame S for not having helped his son immediately because he could have done so. I take it that this is an implausible verdict in this situation. More generally, assuming that abilityfun is morally relevant yields the problem of radical freedom. That was why the second sense of ability had to be introduced. If abilityless-fun is morally relevant we get the ‘right’ verdict for SKID and more generally avoid the problem of radical freedom. But now the question arises why abilityfun rather than abilityless-fun was taken to be the morally relevant ability in NEWSPAPER. The Humean has her special way out of the consequence-argument only if she appeals to abilityfun. But why should different kinds of abilities be morally relevant in SKID and NEWSPA PER? It is hard to see that there is a relevant difference between SKID and NEWS PAPER that would motivate why we should appeal to different conceptions of ability in these two cases when we are evaluating whether or not we should blame S. It seems that the very same sense of being able to act otherwise is relevant in both cases. If the latter consideration is granted there remain only two options for the Humean: Either the morally relevant ability in both cases is abilityfun. If that is the case the Humean can bypass the consequence-argument but has to confront the problem of radical freedom. Or the morally relevant ability is in both cases abilitylessfun. In that case the Humean is very much in the same position as the Non-Humean: The problem of radical freedom can be avoided, but the consequence argument looms. Thus, whether one’s conception of natural modalities is Humean or non-Humean may make a significant difference for the debate of free will. The dialectic and the problems to be addressed crucially depend on this choice (as well as on the interpretation of the morally relevant abilities). Pace Ayer however, it does seem that the Humean has no distinctive advantage over the Non-Humean.
References Armstrong, D. (1983). What is a law of nature? Cambridge University Press. Ayer, A. J. (1954). Freedom and necessity. In Philosophical essays (pp. 271–284). Palgrave Macmillan. Beebee, H. (2000). The non-governing conception of laws of nature. Philosophy and Phenomenological Research, 61, 571–593. Beebee, H., & Mele, A. (2002). Humean compatibilism. Mind, 111, 201–223. Bird, A. (2007). Nature’s metaphysics: Laws and properties. Oxford University Press. Cohen, J., & Callender, C. (2009). A better best system account of lawhood. Philosophical Studies, 145, 1–34. Dorst, C. (2018). Toward a best predictive system account of laws of nature. British Journal for the Philosophy of Science.. https://doi.org/10.1093/bjps/axy016 Fine, K. (2012). Guide to ground. In F. Correia & B. Schnieder (Eds.), Metaphysical grounding (pp. 37–80). Cambridge University Press.
198
A. Hüttemann
Hall, N. (manuscript) Humean reductionism about laws of nature. http://philpapers.org/go.pl? id¼HALHRA&proxyId¼&u¼http%3A%2F%2Fphilpapers.org%2Farchive%2FHALHRA. pdf. Accessed 14 May 2020. Hicks, M. T. (2018). Dynamic Humeanism. British Journal for the Philosophy of Science, 69, 983–1007. Hüttemann, A., & Loew, C. (2019). Freier Wille und Naturgesetze – Überlegungen zum Konsequenzargument (with Christian Loew). In K. von Stoch, S. Wendel, M. Breul, & A. Langenfeld (Eds.), Streit um die Freiheit – Philosophische und Theologische Perspektiven (pp. 77–93). Schöningh. Jaag, S., & Loew, C. (2020). Making best systems best for us. Synthese, 197, 2525–2550. Krämer, S., & Roski, S. (2017). Difference-making grounds. Philosophical Studies, 174, 1191–1215. Lewis, D. (1973). Counterfactuals. Oxford University Press. Lewis, D. (1981). Are we free to break the laws? Theoria, 47, 113–121. Lewis, D. (1986). Causation. In Philosophical papers (Vol. 2, pp. 159–213). Oxford University Press. Lewis, D. (1994). Reduction of Mind. in S. Guttenplan (Ed.), A Companion ot the Philosophy of Mind, Cambridge MA, 412–431. Loew, C., & Hüttemann, A. (forthcoming). Are we Free to Make the Laws? Synthese. Maudlin, T. (2007). The metaphysics within physics. Oxford University Press. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford University Press. Perry, J. (2004). Compatibilist options. In J. K. Campbell, M. O’Rourke, & D. Shier (Eds.), Freedom and determinism (pp. 231–254). MIT Press. Swartz, N. (2003). The concept of a physical law (2nd ed.). Cambridge University Press. van Inwagen, P. (1983). An essay on free will. Oxford University Press.
Chapter 11
How the Libet Tradition Can Contribute to Understanding Human Action Rather than Free Will Sofia Bonicalzi and Mario De Caro
Abstract Experimental findings on the neurobiological roots of free will, pioneered by Benjamin Libet’s work in the 1980s, have been met with a mixture of acceptance and controversy. Discussions in both philosophy and cognitive neuroscience have indeed generated an active and at times polarized debate on whether such findings successfully disprove free will, which is the issue that is customarily considered to be at stake. In our view, this polarization often comes at the expense of genuine crossdisciplinary fertilization, which may turn into a mere attempt at bringing others round to one’s own positions. In this chapter, we argue for an alternative approach. In the first place, we claim that, in themselves, these findings do not address the problem of whether free will, as intended in the philosophical tradition, is illusory or not. However, we also claim that they should be taken as a valuable asset within a more comprehensive theory of human action, aiming to explain how individuals navigate the environment by means of a wide repertoire of more or less complex, flexible, and intelligent behaviors. Keywords Free will · Cognitive neuroscience · Action · Libet experiments Experimental findings on the neurobiological roots of free will, pioneered by Benjamin Libet’s work in the 1980s, have been met with a mixture of acceptance and controversy. Discussions in both philosophy and cognitive neuroscience have indeed generated an active and at times polarized debate on whether such findings successfully disprove free will, which is the issue that is customarily considered to be
S. Bonicalzi (*) Roma Tre University, Rome, Italy Ludwig-Maximilians-Universität München, Munich, Germany e-mail: sofi[email protected] M. De Caro Roma Tre University, Rome, Italy Tufts University, Medford, MA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_11
199
200
S. Bonicalzi and M. De Caro
at stake. In our view, this polarization often comes at the expense of genuine crossdisciplinary fertilization, which may turn into a mere attempt at bringing others round to one’s own positions. In this chapter, we argue for an alternative approach. In the first place, we claim that, in themselves, these findings do not address the problem of whether free will, as intended in the philosophical tradition, is illusory or not. However, we also claim that they should be taken as a valuable asset within a more comprehensive theory of human action, aiming to explain how individuals navigate the environment by means of a wide repertoire of more or less complex, flexible, and intelligent behaviors. The scope of the chapter is then twofold. First, we argue that the discussion regarding the philosophical import of Libet-like experiments should focus away from their alleged relevance for the free will issue. Second, we claim that these experiments can throw light on how humans navigate the environment by means of different action-types. This is a task that requires the contribution of both the philosophy and the cognitive neuroscience of action. More specifically, if actionstypes are then to be investigated experimentally, fitting and nuanced operational definitions have to be shaped – and philosophical theorizing could, or should contribute to providing, or eventually refining, such definitions. While this is the overarching approach we aim to argue for, the chapter moves forwards through the discussion of what appear to us as the most challenging findings and open questions in this field. The chapter is organized as follows. In Sect. 11.1, we will introduce the classic Libet experiment, aiming to dissect its underlying logic and most problematic conceptual steps. The following sections will then analyze these steps in turn, respectively focusing: on the key features of the action-type that is central to the Libet experiment, which will be contrasted with other action-types, such as behaviors that are planned, preference-loaded, or reason-based (Sect. 11.2); on the causal role of conscious intentions (Sect. 11.3); and on the causal role of the neural precursors of voluntary or intentional actions (Sect. 11.4).
11.1
The Logic of the Classic Libet Experiment
In the original Libet experiment (Libet et al., 1982, 1983), participants had to repeatedly make simple, self-paced bodily movements, such as flexing one’s finger while watching a clock with a dot circling around it. At the end of each trial, they had to report where the dot was when they consciously felt that they wanted to move. One key task instruction was that the decision about when to move the finger had to be spontaneous and endogenous (i.e., internally generated): participants were asked to act at their pace and base their decision about when to act on internal conscious states (i.e., the urge to move) rather than on the external sensory inputs they were exposed to (e.g., the clock itself). This combination of endogenous/spontaneous bodily movements associated with a mental chronometry task readily became a mainstay in the experimental literature on free will (Haggard, 2017). Discussing this literature, in the rest of the chapter we will refer to this type of actions simply as
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
201
LAs (an acronym for Libet’s actions), contrasting them, whenever this is required, with other action-types.1 The original Libet experiment included a combination of behavioral and neurophysiological measurements. As mentioned, participants were instructed to take a mental note of the time (W) when they became consciously aware that they wanted to move, reporting the time W to the experimenter at the end of each trial. In the meantime, an electromyogram (EMG) measured the onset of the bodily movement, and an electroencephalogram (EEG) recorded the underlying electrical brain activity from participants’ scalp. This combination of methods allowed the experimenters to detect the Bereitschaftspotential or Readiness Potential (RP) (Kornhuber & Deecke, 1965) that preceded LAs. The RP is a gradual increase in the electrical negativity in motor areas, and in particular in the supplementary and/or the pre-supplementary motor area, starting in the second or so that precedes muscular motion and peaking just before the movement begins. Traditionally, the RP has been identified as a reliable neural marker of action planning, preparation, and initiation (Shibasaki & Hallett, 2006). Libet’s central experimental finding was that the onset of the RP temporally precedes W, which seemingly implies that participants become aware of wanting to make a movement after the planning has already initiated, unconsciously, at the brain level. In the years following the initial experiment, new findings and techniques, including studies using fMRI (Soon et al., 2008, 2013) and electrodes implanted directly in the scalp during brain surgery (Fried et al., 2011; Maoz et al., 2013b), have lent support to the idea that LAs are initiated by unconscious neural activity. This activity is localized in pre-motor cortical areas, the anterior cingulate cortex (ACC), and frontal regions, such as the dorsolateral prefrontal cortex (DLPFC) (Cunnington et al., 2002; Jahanshahi et al., 1995). Emphasizing the linkage between prior neural activity and voluntary action, existing results have shown correlations between motor-related cortical potentials, and in particular the RP, and features such as the complexity of the action (Benecke et al., 1985) or the effort that it requires (Slobounov et al., 2004). Conversely, experimental critiques have focused on the technical imperfections of this line of research, such as its reliance on mental chronometry and participants’ shaky introspective skills (Mele, 2010; Moore et al., 2009; Ulrich et al., 2006; Vul et al., 2009), or the limited predictive ability in decoding the content of participants’ upcoming actions (Lavazza & De Caro, 2010). These imperfections testify to the limitations of existing experimental techniques, while also being potentially warning signs of more serious theoretical flaws. Although important differences exist between the findings of the various experiments inspired by Libet, we abstract away from details and focus on the conceptual reconstruction of the logic behind the classic Libet experiment.2 Based on the
1 For a meta-analysis of Libet-style experimental findings and a critical discussion of their solidity, see Braun et al. (2021). 2 For extensive analyses of the limits and the potentials of the research tradition initiated by Libet, see, e.g., the essays in Feltz et al. (2019), Pockett et al. (2006) and Sinnott-Armstrong and Nadel (2011) .
202
S. Bonicalzi and M. De Caro
temporal priority of the RP, Libet concluded that free will cannot be the factor that (causally) initiates the ensuing action: “free will cannot be viewed as an initiator of such a freely voluntary process [. . .] The initiation of the preparation to culminate in a freely voluntary movement arises unconsciously in the brain, preceding the conscious awareness of wanting or intending to ‘act now’ by about 400 msec or more” (Libet, 2005, p. 141).3 Behind the apparently straightforward connection between free will and prior awareness, claimed by Libet himself, it is useful to break down this logic into its constituents. Libet’s reasoning can be seemingly summarized as follows: 1. 2. 3. 4. 5. 6.
LAs exemplify free actions. Any free action must be initiated by an act of free will. Free will requires initial awareness of wanting to move. LAs originate from prior unconscious antecedents. Then LAs are not initiated by an act of free will. Therefore, LAs are not free actions.
This logic has already been extensively criticized elsewhere as flawed (Björnsson & Pereboom, 2014; Bonicalzi, 2015; De Caro et al., 2019; Lavazza & De Caro, 2010; Mele, 2010; Roskies, 2010).4 In this chapter, we aim at focusing on some of the most critical steps of this logic while turning them into research questions and invitations for further research. On the one hand, we will consider such critical steps in the light of more recent experimental findings in the Libet tradition. On the other hand, we will discuss open issues at the intersection between the philosophy and the cognitive neuroscience of volition and action. More specifically, we will single out three overarching research themes. In Sect. 11.2, we will devote our attention to the operational definition of free will or free action that is at play in LAs. We will consider how this operational definition makes it difficult to make sensible generalizations across different action-types. Indeed, LAs exemplify some necessary features of free actions, i.e., their being intentional or voluntary, without providing a full-fledged model of what a free action stands for. We will claim, however, that this limitation does not disqualify the relevance of the empirical results if these are understood within the framework of a more comprehensive theory of action covering different action-types. In this sense, we will suggest that different action-types, including LAs, are part of the behavioral repertoire that humans display and that is worth investigating philosophically and empirically, with fitting operational definitions and experimental tools. In Sect. 11.3, we will distinguish between the Libet-
In other passages, Libet is more cautious, e.g., he writes that his findings posit “certain constraints on the potentiality for conscious initiation and control of voluntary acts” (Libet et al., 1983, p. 623). As reported by Mele (2010), the skeptical conclusion has been drawn by many, including, e.g., Pockett (2006), Roediger et al. (2008), Shariff et al. (2008) and Wegner (2002). 4 The main criticism has been that the experimental results in the Libet tradition can be easily accommodated by the compatibilists, who have no problem in granting that the generation of free actions may be determined by several prior factors, some of which might be unconscious (see Sect. 11.3). 3
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
203
based claim that actions are preceded by unconscious neural antecedents (which is compatible with granting a causal role for conscious mental states) and the more worrisome claim that conscious mental states are fully bypassed by automatic and unconscious processes. Finally, in Sect. 11.4, we discuss the thorny issue of the causal status of the RP with respect to LAs. In the following, we adopt a pluralist approach in terms of the contribution that both philosophy and cognitive (neuro)science can provide to the understanding of human action (Bonicalzi & Haggard, 2019; De Caro, 2007). As one of us has extensively argued elsewhere, by defending a form of liberal naturalism, philosophical views on free will and action must be compatible with, although non-necessarily reducible to, our best scientific knowledge (De Caro, 2020). Philosophy contributes to the discussion by defining the conceptual contours of the research questions, e.g., how free will, free actions, and the different types of action should be thought of. Indeed, as mentioned, it is a task for theoretical analysis to provide more nuanced operational definitions and hypotheses, some of which can then undergo empirical investigation. Cognitive (neuro)science plays a role insofar as the relation between actions, cognitive skills, and brain functions must be addressed empirically. It is a task for empirical investigation to explain how we are able to perform the rich array of actions that enable us to navigate the complex environment we inhabit.
11.2
Types of Actions
One classic source of criticisms concerning the ecological validity of Libet’s findings stems from the alleged irrelevance of LAs, which has been termed their “Buridan’s ass quality” (Mele, 2010). The worry is that repetitive decisions about when to flex one’s wrist or finger are devoid of any practical significance, and therefore do not truly exemplify what free actions stand for. These decisions are tied neither to prior preferences nor to meaningful outcomes. As a result, they have little in common with the richer experiences of agency that are customarily discussed within the philosophical literature on free will. For example, Mark Balaguer (2009) associates free will with the experience of choosing between preference-based options, which he calls “torn decisions”; in turn, Robert Kane (1996, 2002) relates free will to the agent’s making “self-forming actions” that channel and define her personality and future behavior. Alfred Mele (2010) has argued that LAs do not harbor bona fide decisions. Indeed, bona fide decisions entail a prior element of undecidedness, followed by a deliberative phase, which is absent in the Libet’s task where participants have to mechanically repeat the bodily movement they were initially instructed to perform. It seems reasonable to suggest that the impoverished setting of LAs makes them substantially irrelevant for the free will issue as philosophers have defined it. Consequently, one might be tempted to conclude that no informative comparison can be legitimately drawn between LAs and the more meaningful, preferenceloaded, planned action-types that we make experience of in our everyday life. In
204
S. Bonicalzi and M. De Caro
this section, we discuss this inadequacy issue by focusing on different features of LAs, notably their content (Sect. 11.2.1), their presumptive habituality (Sect. 11.2.2), and their being associated with unplanned bodily movements (Sect. 11.2.3). Overall, while pointing at the limitations of LAs as a model of free actions, we suggest that they constitute a type of intentional or voluntary behavior that should find its proper place within a more general theory of action aiming to account for different action-types.
11.2.1 The Content of LAs The irrelevance of LAs in terms of their content is often indicated by philosophers as one of the fatal flaws in the empirical investigation of free will, in particular, and of volition and action, in general. However, little attention is usually devoted to discussing why most empirical paradigms opt for LAs and whether there would be anything to lose in doing otherwise. The Buridan’s ass quality of LAs is indeed a targeted, rather than unintended, form of irrelevance. As such, it is an aspect of LAs that is very central to how cognitive scientists conceptualize voluntary or intentional action, and thus also free actions as long as they are voluntary or intentional (Passingham et al., 2010).5 Therefore, criticizing LAs because of their Buridan’s ass quality amounts to rejecting the validity of the operational definition that underlies most neurophysiological experiments on volition and action. Within the experimental literature, free actions are essentially described as spontaneous and endogenous behaviors, i.e., self-paced behaviors elicited by the agent’s internal states rather than by clearly identifiable external cues (e.g., sensory prompts) that would bias, trigger, or instruct behavior.6 Spontaneous, endogenous actions are initiated from within and are thus irreducible to simple stimulus-response behaviors. In this vein, Gold and Shadlen discuss voluntary actions in terms of their freedom from immediacy (2003). The attempt at artificially creating the conditions for this freedom from immediacy to emerge is evident in how typical task instructions are framed, i.e., participants are explicitly asked to be spontaneous, to act at their pace, not to follow fixed patterns, not to decide based on the available sensory inputs.7 In
5
Whereas the notions of intentional and voluntary are usually used interchangeably in cognitive science, the distinction between the two is often much more nuanced in philosophy (see Anscombe’s classic essay Intention (1963)). Recently, the philosopher Yair Levy (2013) has distinguished between fully-fledged intentional actions (e.g., going shopping, cooking a meal) from mere voluntary behaviors (e.g., changing position, playing with one’s hair). 6 Spontaneity is not to be understood as an indication of a commitment to metaphysical indeterminism: spontaneous actions are simply actions whose causes (whether deterministic or indeterministic) are not immediately detectable in the environment. 7 For instance, as part of the instructions in Soon et al. (2013), participants were told that “the decision time and choice of task were completely up to them but that they should be as spontaneous as possible and execute their decision without hesitation once it was made”.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
205
the case of LAs, the less a decision is causally driven by clear determinants, the more it is believed to approximate a paradigmatic free choice. The dichotomy between spontaneous/endogenous and instructed actions traces back to Skinner’s distinction between operant and respondent behavior (1948). From the experimental point of view, this distinction has been extremely advantageous. Indeed, it provides the templates for two easily implementable experimental conditions contrasting kinematically analogous bodily movements. The problem is that spontaneity and endogenicity can be seen as relevant, even necessary, features of what a free action is but they do not suffice to single out what philosophers have classically thought about what a free action stands for, i.e., in particular free actions that have consequences for which we can be held responsible.8 The result is that philosophers and neuroscientists risk talking past each other. Indeed, without further qualifications, besides spontaneity/endogenicity, it remains difficult to account for actions that are not just intentional or voluntary but also free in some philosophically relevant way.9 That said, we don’t see any truly convincing reason why actions that are scarcely relevant for our future life, but that we can still execute at will,10 should not be included in the matrix of action-types that are worth investigating within the boundaries of a more global theory of intentional action. In this sense, we adopt an ecumenical approach: LAs represents one type of intentional or voluntary action, without this implying that LAs can account for all action-types that we may want to call voluntary or intentional and, more specifically, for actions that, in the light of philosophical analysis, we may want to call free. In particular, starting from a compatibilist model of free will, one would likely be uncomfortable with defining free actions as actions that are free from immediate, i.e., clearly identifiable, causes.11 The lack of clearly identifiable causes is not evidence that there are no prior causes, including internal causes that may instruct participants’ behaviors from within.
8
For example, Mele has indicated the lack of compelling motivational states or coercion, reasonbased deliberation, and the agent’s being a reliable deliberator as jointly sufficient for “psychological autonomy”, which is what is required for grounding responsibility (1995). 9 It seems worth noticing that Aristotle’s distinction (2000) between involuntary and voluntary actions in the Nicomachean Ethics bears some resemblance with the couple externally/internally generated actions. In particular, Aristotle highlights that one feature of involuntary actions is that they are originated from without, with the agent remaining passive with respect to them. 10 One historically controversial aspect of Libet’s view is the defense of our ability to consciously veto the action, just before execution. Libet described this as a power of “permitting or triggering the final motor outcome of the unconsciously initiated process” or “vetoing the progression to actual motor activation” (1985, p. 529). It has been plausibly suggested, however, that such last-moment inhibitory processes also depend on prior unconscious activity (Filevich et al., 2013), which makes negative volitions substantially analogous to positive volitions in terms of their potential for supporting free will (De Caro et al., 2019). For an interpretation of the veto power based on the metaphysics of power (Marmodoro, 2022, Chap. 1), see De Hann (2022, Chap. 8). 11 For a disambiguation of the compatibilist and incompatibilist understandings of the notion of freedom, see Dorato (2022, Chap. 12).
206
S. Bonicalzi and M. De Caro
While we agree that grounding free will in the distinction between internal and external causes of an action risks sounding arbitrary, we also acknowledge that there is no clear, unified philosophical understanding of what a free action, as produced by an act of free will, stands for. Different philosophical views highlight different, if not alternative, features as central to a coherent understanding of free will.12 In another guise, this notion of freedom as spontaneity/endogenicity, supplemented with other qualifications, may even play a role within some philosophical views on free will. For instance, consider again Balaguer’s view (2009) of free will as linked to torn decisions. Torn decisions are singled out as situations in which the respective weight of the individual determinants (e.g., prior preferences, future incentives, etc.) does not suffice to bias the agents’ response towards one of the options. In one of Balaguer’s examples, an agent (Ralph) struggles with making a life-changing decision between moving to New York and staying in Mayberry. Ralph spends quite some time ruminating on the pros and cons of each option but, in the end, he realizes that none of them outweighs the other. In making the final decision, Ralph is aware that the choice will have a considerable impact on his future life. So, in this sense, Ralph’s decisional process bears little resemblance to LAs’ apparent randomness. However, there is a sense in which Balaguer suggests that Ralph, exactly because no option outweighs the other, in the end just picks one option: since deliberation does not help Ralph to come out with a solution, he ends deliberation, somehow arbitrarily, and makes a choice. As a result, Ralph ends up with the feeling that he just decided, e.g., to go to New York (see Bonicalzi, 2019).13 If this reconstruction is accurate, free, torn decisions are those decisions that are taken in a sort of arbitrary manner (which bring them closer to LAs), but that also force the agent to engage in deliberation (which makes them different from LAs).14 In turn, the decoupling between spontaneous/endogenous and instructed behaviors, which is central to the operational definition of freedom underlying LAs, finds
12
Nor it seems easy to rely on folk intuitions of what free will, or free action, stands for. In this respect, the consensus seems to be that folk intuitions about what free will/action entails are quite mixed (Roskies & Nichols, 2008). 13 Here, we are just concerned with the phenomenology of Ralph’s decision-making process. Ralph sees the two options as equally attractive ending up with the feeling that the decision is not strongly biased by his prior preferences or future incentives. This does not imply that his final decision was not in fact determined by more subterranean determinants Ralph is not aware of. In this latter case, Balaguer’s reconstruction of indeterministic free choices would be incorrect in the sense that the choices can be determined by factors the agent is not aware of, independently of whether the agent has the subjective feeling of not being biased. If there is something that the literature on implicit biases and the automatic mind has shown, it is that people are not entirely aware of all the, more or less subterranean, determinants of their behaviors (see Holroyd, 2015). 14 That said, libertarian restrictivists, including Balaguer (2009), see libertarian free actions (i.e., thorn decisions that are indeterministic and sufficiently controlled) as just the opposite of irrelevant, Buridan ass-like, actions (see also the already mentioned Kane, 1996). Here, we discuss Ralph’s action as arbitrary in terms of the fact that it is not clearly driven by its prior causal antecedents (i.e., Ralph’s existing preferences), while Balaguer describes Ralph’s action as nonrandom since it represents the agent’s own decision (see also Dorato, 2022, Chap. 12).
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
207
some support in neurobiologically relevant constraints at the brain level. Indeed, existing neuroscientific evidence has distinguished between separate neural patterns enabling spontaneous/endogenous and instructed behaviors.15 Even if we restrict the focus to electrophysiological differences at the level of the RP, the RP is present with spontaneous/endogenous but not with instructed behaviors, such as when participants are responding to visual prompts (Jahanshahi et al., 1995). While we are suggesting that we should probably accept LAs as a type of intentional or voluntary behavior that is worth investigating, it remains an open question how to devise informative experimental paradigms that can study other action-types, thus reflecting the features of actions that philosophers would more comfortably consider free. In this sense, as mentioned, philosophy can contribute to this effort by providing operationalizable experimental definitions of different action-types and their distinctive features. Most experimentalists working on volition and action are not blind either to the pressing need to provide a more nuanced framework for the study of volition and action (Nachev & Hacker, 2014; Nachev & Husain, 2010). The dichotomy between spontaneous/endogenous vs. instructed behaviors has also been occasionally reframed by scientists more in terms of a continuum (Haggard, 2014). Within this debate, perhaps less attention has been devoted to the potential tension between the two features of endogenicity and spontaneity. Albeit these two notions are often paired in discussion (and when instructing experimental participants), in real life they do not necessarily mutually entail one another: endogenous (i.e., internally generated) actions can be quite short of spontaneity (i.e., lack of clear causes), especially when agents spend some excruciating time ruminating on different options. For example, slightly modifying Balaguer’s Ralph case, we can think of Ralph as finally deciding that his preference for New York outweighs his preference for Mayberry. While his decision may still count as endogenous (i.e., he was not forced to choose New York), it would definitely not count as spontaneous (i.e., Ralph is responding to specific reasons when he decides to go to New York). Thus, when moving from theory to experimental practice, one key challenge for current and future neurophysiological studies consists in devising experiments that can combine spontaneity and/or endogenicity with more high-level features of actions, such as their being goal-directed, guided by intentions, preference-oriented, and sensitive to external validation. Taken on their own, these elements are already extensively investigated within different domains in cognitive sciences, which have a long-standing interest in the human capacity to form mental representations and compute task-related costs and prospects (Haazebroek & Hommel, 2008); to choose courses of action in view of expected rewards or punishments (Dickinson & Balleine, 2002; Maoz et al., 2013a; Robinson et al., 2010); or to select bodily movements as especially apt to a given task (Ghahramani & Wolpert, 1997).
15
For a review of the different brain areas accounting for spontaneous/endogenous and instructed behaviors, see Bonicalzi & Haggard (2019). But see also Nachev & Husain (2010) about the possible problems inherent in this approach.
208
S. Bonicalzi and M. De Caro
Questions about how people make decisions in specific contexts have been consistently addressed in fields such as neuroeconomics (Sanfey et al., 2006) or moral decision-making (Crockett et al., 2014). Whereas LAs appear to be a piece of a multi-layered account of action-types, largely understudied areas of investigation still concern the connection between the philosophy and science of decision-making and the philosophy and science of volition and action. Recently, some new paradigms have attempted to fill the existing gaps by injecting some preference-loaded decisions into the Libet’s framework, to study meaningful actions that are not stimulus-triggered. This line of research has the potential to substantially enrich the Libet tradition, although the results are not yet conclusive and the operational definitions of what a deliberate action consists in are again multi-faceted. We will conclude this section by presenting a couple of related examples.16 In one case, Maoz et al. (2019) used the classic Libet’s task to directly compare the neural precursors of deliberate vs. arbitrary choices resulting in kinematically analogous bodily actions. In the deliberate choice condition, participants were repeatedly presented, on a computer screen, couples of charities about which they had previously expressed their likes and dislikes. Participants had then to choose, by pressing the right or left button, to what charity they wanted to donate a monetary bonus. Deliberate choices were thus operationalized as endogenous, rationalized decisions, based on overtly expressed prior likes/dislikes and having appreciable consequences in the external world. In the arbitrary choice condition, reproducing LAs, participants were instructed that, independently of the specific button they had pressed, the two charities would have received an equal monetary bonus. The most striking finding consisted in the absence of a detectable RP for the deliberate choice condition, while the RP remained visible in the arbitrary condition, even though the bodily movements in the two conditions were kinematically analogous. Conversely, in another recent experiment (Khalighinejad et al., 2018), the RP was regularly observed both in deliberate and instructed choices. Rather than using the classic Libet’s task, the study adopted a perceptual decision-making paradigm with two main conditions (i.e., kinematically analogous movements for deliberate vs. instructed choices). In both conditions, participants had to watch a display of moving dots and press a left/right button as soon as they noticed that the dots had started moving coherently to the left or the right side of the screen. Monetary incentives rewarded correct responses. Participants were informed that, in each trial, the dots would have started moving towards a coherent direction following a variable delay from trial onset, potentially resulting in extended waiting time. Since the total length of the experiment was fixed (600 ) and monetary incentives were awarded on a trial-by-trial basis, long waiting times within single trials would have resulted in a low total number of completed trials, and thus in a smaller
16
In a similar vein, Pornpattananangkul and Nusslock (2015) showed that the RP is sensitive to the anticipation of future rewards, with the presence of reward-anticipation cues associated with a more negative RP.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
209
cumulative reward, by the end of the experimental session. Participants were thus willing not to wait for too long, in every single trial, in order to maximize their cumulative reward. In the deliberate condition, participants had the opportunity to end the ongoing trial at will, whenever they felt that they had waited for too long, by pressing a skip button, which let them move on to the next trial. Deliberative choices were thus operationalized as reason-based decisions motivated by boredom or strategic playing. Although the results concerning the presence of the RP for deliberate choices are still debated, one clear advantage of both Maoz et al.’s and Khalighinejad et al.’s chosen paradigms is that there is no need to artificially instruct participants to act spontaneously/endogenously, as it was the case in the more classic Libet paradigm.
11.2.2 The Habituality of LAs In the classic Libet experiment, participants are asked to perform several repetitions of kinematically analogous bodily actions. As a result, questions might arise in terms of whether repetitive actions of the sort require to be initiated by an act of conscious will or should rather be categorized as habitual.17 Indeed, the literature on habitual actions and automatisms has suggested that certain actions, or even action sequences, become routinized through repetition until they can be finally executed in the absence of top-down, deliberative control or intentions to act. In the cognitive science of action, the distinction drawn between goal-directed and habitual actions tends to be quite profound. Specificities at the level, e.g., of neural realizers (Dolan & Dayan, 2013) and computational mechanisms (Keramati et al., 2016) underlie the overtly different characters of these action-types. While goal-directed actions require an active role of conscious awareness and monitoring, habitual actions may proceed smoothly while we navigate the environment absent-mindedly (Dezfouli et al., 2014). In given circumstances, automatic actions can be activated by the relevant contextual cues (Shiffrin & Schneider, 1977), independently or even contrary to conscious motivation (Neal et al., 2011). Attempts at exerting top-down control may even be detrimental to action performance, e.g., in terms of diminishing the action speed or accuracy (Logan & Crump, 2010). Therefore, one may reason that, if we can categorize LAs as habitual, then we can also stop worrying about the lack of awareness prior to the RP onset: habitual actions do indeed occur without the intervention of top-down awareness. However, although LAs, kinematically speaking, are of a type that lends themselves to habituality, in this section we will argue that it is rather problematic to simply assume that participants perform LAs in a habitual fashion.
17
In moral philosophy, there is a vast debate (particularly alive regarding virtue theories) on whether habitual actions can be said to encompass deliberation and count as rational and virtuous: see De Caro et al. (2021).
210
S. Bonicalzi and M. De Caro
To facilitate interaction between philosophical analyses and the experimental work, it seems again valuable to point out at some terminological discrepancies between disciplines. Analogously to what we have said about the dichotomy between spontaneous/endogenous and instructed actions (see Sect. 11.2.1), positing a rigid distinction between goal-directed and habitual actions could be misleading. Indeed, it would be mistaken to assume that habitual actions necessarily operate in the absence of a goal: the point is rather that the goal may inform the action without the agent’s being consciously aware of its content during action execution. For example, when we routinely brush our teeth, we do so in view of having our teeth cleaning done while potentially engaging absent-mindedly in the specific action sequence that brings us closer to the goal. Philosophical views make room more explicitly for actions that are both habitual and somehow intentional. For instance, according to the classic causal theory of action, actions are intentional whenever they are appropriately caused by the right mental states, such as intentions or plans (Bratman, 1987; Davidson, 1978).18 However, Davidson readily acknowledges that intentional behaviors do not necessarily include an element of conscious deliberation. For an action to be intentional, it suffices that the intention is consciously accessible, so that agents can access the intention, if circumstances so require, without necessarily consciously focusing on it prior to the action.19 Alfred Mele (2010) has discussed the role of intentions within habitual actions in terms of an intention’s being present without being actively formed through an exercise of practical deliberation.20 In his example, every morning an agent opens the office door without engaging in any practical deliberation about what to do: as a result, the agent has an occurrent intention to unlock the door that did not emerge through deliberation. Assuming a more or less liberal understanding of the linkage between intention and habituality in actions, one may conclude that people can perform intentional, routinized actions without prior awareness of wanting to move. Whenever circumstances so require, e.g., in case of troubleshooting, lack of prior awareness does not prevent the agent from quickly stepping back from the automatic mode and regaining conscious monitoring over the action.21 Returning to the example by Mele, had the agent overheard people’s fighting inside the office, she could quickly make up her mind and consciously deliberate as to whether she wanted to open the door or not. However, although people can often navigate the environment without being in a conscious monitoring mode, it seems unlikely that this is what actually happens when they perform LAs during a typical experimental session. As Mele’s example 18
For a discussion of recent challenges to the causal theory of actions based on an Aristotelian analysis of the temporal structure of agency, see Chik (2022, Chap. 6). 19 This does not necessarily mean that all habitual behaviors should be seen as intentional. It might be controversial, for example, whether habitual, stereotyped behaviors (Holroyd, 2015) that stem from implicit biases can count as somehow intentional. 20 On habitual actions being weakly intentional, see Martens and Roelofs (2019). 21 For a classic model of how the brain can continuously switch between the automatic and the goaldirected mode, see Norman and Shallice (1986).
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
211
shows, kinematically similar bodily movements (e.g., opening the office door) can be executed in the habitual or deliberating mode depending on circumstances. Thus, one might think that LAs can also be performed in the habitual or deliberating mode depending on circumstances. However, since participants in the Libet experiment are explicitly required to consciously monitor their intention, in order to report them later on, there seems to be little ground to assume that LAs, during a typical experimental session, are performed in the habitual mode, unless we conclude that the experimental instructions are systematically disregarded. It is true that there are some actions or even action sequences that, to be proficiently executed, must become routinized. This is typically the case for actions that must be performed at a high speed or with high precision so that it would be detrimental to the action’s success if top-down control enters the scene (Shepherd, 2017; but see also Fridland, 2016). However, again, this is hardly the case for simple LAs that participants can perform at their pace. In sum, if LAs are not performed in the habitual mode (e.g., because participants are asked to pay attention to their intentions), then the lack of prior conscious awareness might be not so easily set aside as unproblematic. Indirect support to the claim that experimental instructions are not simply disregarded comes from several experiments, including Lau et al.’s fMRI version of the Libet’s task (2004). The experiment compares the blood oxygenation leveldependent (BOLD) signal across two conditions: in one condition subjects were asked to report time W, as in the classic Libet experiment. In the other condition, subjects were instructed to focus on the time of the action. The comparison shows an enhancement in the BOLD signal in the pre-SMA region of the medial prefrontal cortex (where intentions are represented) in relation to the first condition, when people were explicitly told to pay attention to their intentions. More generally, several studies investigate the relationship between explicit or conscious attention and pre-movement brain activity (Bortoletto & Cunnington, 2010; Haggard & Eimer, 1999; Trevena & Miller, 2010), showing that significant differences can be detected at the level of preparatory brain activity preceding, respectively, attended and unattended spontaneous movements. For example, in an interesting experiment, Baker and Piriyapunyaporn (2012) distinguished between unattended and explicitly timed spontaneous bodily movements. The experimental paradigm consisted in a time reproduction task where participants had to reproduce the interval between two auditory tones. To do so, participants had to press a button twice in order to signal the beginning and the end of the time interval. Then, they received a feedback about how close they were to the original time interval. While the time of the first button press could be relatively unattended or arbitrary, the second button press, signaling the end of the interval, was necessarily to be chosen more carefully. The findings showed that the normal preparatory activity we observe in standard LAs was missing in unattended bodily movements. In comparison, explicitly timed actions show RPs that were topographically and structurally analogous to those that are observed in relation to standard LAs. Therefore, if we assume that participants are actually able to perform LAs in a goal-directed, i.e., non-habitual, manner, disqualifying LAs as irrelevant for illuminating features of intentional or voluntary behavior might be inappropriate. LAs are
212
S. Bonicalzi and M. De Caro
seemingly action-types that we can perform either in a goal-directed or in a habitual manner, depending on circumstances. Once again, a nuanced matrix of action-types is needed to appropriately locate LAs within a more general theory of action.
11.2.3 LAs as Bodily Movements Although the terminology may differ, many scholars draw a distinction between (abstract) intentions as plans or commitment to future actions and (implementation) intentions to do something now (Bratman, 1987, 2007; Mele, 1992; Searle, 1983). Since neurophysiological experiments focus on the former, the latter tend to be empirically understudied. Libet himself made some preliminary attempts at tackling the issue by distinguishing RPs between pre-planned (type I) and spontaneous (type II) LAs. Interestingly, the RP onset for type I-LAs was further anticipated, beginning around 1000 ms before the action onset vs. the standard 550 ms registered for type II-LAs (Libet et al., 1983). However, with both type I and type II-LAs, the action was tied, in terms of its endpoint, to a peripheral bodily motion. This raises interesting questions concerning what can be said about the neural determinants preceding purely mental actions, such as deciding, planning, or making a commitment for the future. In this section, we briefly introduce two issues related to abstract intentions and planning. One concerns how abstract intentions or planning may connect with the corresponding peripheral bodily motions in service of action coordination and execution. The other concerns the neural precursors of intentions or plans that do not correspond to any peripheral bodily movement. The first issue speaks to the need of modeling how abstract intentions or planning relate to more specific motor actions, e.g., how the general intention to bake a cake governs the more specific motor actions that result in the cake’s being finally baked.22 The integration of abstract goals and executive motor commands has often been explained with reference to the cognitive architecture of the lateral prefrontal cortex (LPFC). In particular, Koechlin and colleagues proposed a widely discussed model in which action control is achieved through a hierarchical, cascade process of top-down control, with anterior areas of the LPFC devoted to more deliberative, temporally extended, and abstract control and posterior areas devoted to the more executive, fine-grained aspects of action control.23 Taking on board the hierarchical view of action control, Elisabeth Pacherie (2008, 2015; see also Mylopoulos & Pacherie, 2017) has distinguished between three types of intentions (distal, proximal, motor), which coordinate different levels of action specification, from the more abstract to the more fine-grained, context-dependent and motoric
22
For a discussion on how abstract intentions interlock with motor representations, see Butterfill and Sinigaglia (2014). 23 See Koechlin et al. (2003) and Koechlin and Summerfield (2007). See also Schurger and Uithol (2015) for a non-hierarchical model of action control.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
213
details of the action. These motoric, fine-grained details, as well as the correction of small errors in ongoing bodily movements, can be regulated outside conscious awareness (Fourneret & Jeannerod, 1998). In the philosophical and empirical literature, this question about how we control and coordinate actions in view of intentions is problematically kept separate from discussions about free will as based on the Libet tradition. Since control (as the matching between intentions and actions) is usually thought to be central to free will in philosophical accounts (Fischer & Ravizza, 1998), it would be crucial to implement empirically tractable models of action that explain how features like spontaneity/endogeneity can combine with action control and coordination in view of intentions and goals. Indeed, computationally speaking, the literature on action control covers aspects that free will-theorists should also be concerned with, notably the selection and execution of goal-directed actions (Duncan & Owen, 2000; Miller & Cohen, 2001; Rowe et al., 2000). The second aspect that, we suggest, deserves further attention concerns the neural precursors of actions that are purely mental, e.g., in the case of abstract decisions that are disconnected from implementation intentions (see Gollwitzer, 2003). It has been suggested that these mental actions, in the form of high order planning, are the proper place for locating free will (see Gallagher, 2006). Building on this, mental actions thus constitute another terrain for the philosophical and empirical investigation on volition and action. Some studies are now starting to address questions about the neural precursors of purely mental actions. For example, in the already mentioned fMRI study by Soon et al. (2013), the experimental task consisted in asking participants to repeatedly make spontaneous mental decisions while being in the MRI scanner. In particular, participants had to watch a computer screen displaying a stream of letters surrounded by numbers. While doing this they had to memorize what letter was visible on the screen when they decided to perform a mental calculation task, by adding or subtracting the available numbers. At the end of each trial, participants had to indicate both what letter they had chosen and the solution to the mental calculation. Particularly relevant for the present discussion is that Soon and colleagues were interested in identifying activations corresponding to a task-independent brain network, which could be associated with both bodily and mental actions, where different action-types are made. The experimental findings show that the content of the mental action (i.e., what calculation to pursue) could be decoded, in a medial frontopolar and a posterior cingulate/precuneus region, up to 4 s before subjective awareness. In a previous study by Soon et al. (2008), these regions were already indicated as relevant for the decoding of intentions linked to motor actions. The two findings jointly hint at the existence of a shared network where the preparatory processes enabling both abstract and motor spontaneous actions are encoded before conscious awareness. Importantly, this network does not encode information related to the motoric response per se, which was rather associated with a different network (Soon et al., 2008), including the preSMA, the SMA, and the rostral cingulate zone, where information about the time (the when) of the action was encoded before conscious awareness. Immediately before and after conscious awareness, abstract
214
S. Bonicalzi and M. De Caro
and motor actions depart from one another, with different areas (e.g., the angular gyrus for the calculation task) being progressively involved in task-specific mental and motoric activities. In sum, we want to emphasize that, in view of building a coherent matrix of action-types, the experimental and philosophical literature on free action should connect more systematically with the literature on abstract intentions and planning. In this light, the attempt at investigating the neural correlates of purely mental actions, and the search for shared or distinguishable brain networks enabling these different action-types, represent a welcome addition to the field.
11.3
LAs and the (Causal) Role of Conscious Mental States
As mentioned earlier, the ability to act as one wanted to act, i.e., so that the action matches the corresponding intention, is central to many accounts of free will. This is the case in particular on the compatibilist side, which rejects the thesis that free will requires freedom from prior causes (but this is also true of libertarians who appeal to indeterministic causation (Kane, 1996)). Along these lines, Fischer and Ravizza (1998) propose that freedom essentially consists in guidance control, which considers the ability to act as one wanted or planned to act as a key component.24 Also outside of the debate on free will, action control is often understood in terms of the (causal) matching between the intention and the action (Shepherd, 2014).25 The matching between (different layers of) intentions and actions plays a role also within the hierarchical views of action control discussed in Sect. 11.2.3. By contrast, the (causal) role of intentions is far less clear within the experimental literature that is linked with the Libet tradition. One important contribution that philosophy has brought to this discussion, and that is often neglected by more experimentally minded scholars, consists in the clarification made by the compatibilist tradition that the existence of prior determinants, be they conscious or unconscious, does not make the ensuing action necessarily unfree to the extent that conscious mental states also play a (causal) role in action production.26 Moreover, since most philosophers accept a materialistic view
24
More precisely, Fischer and Ravizza discuss guidance control in terms of (i) the possession of the mechanism from which the action flows and (ii) the subject’s sensitivity and responsiveness to reasons. See Mayr (2022, Chap. 7) for a discussion of the relationship between free will and rational abilities. 25 For a non-causal view of action guidance, see Frankfurt (1978). 26 Although this is typically defended within compatibilist accounts of free will (see Mele, 2010), see Bernáth (2019) for a discussion about whether some forms of libertarianism can be compatible with the Libet’s results. See Steward (2022, Chap. 9) about the wider issue of how libertarian free will can be compatible with the laws of nature. Concerning the compatibility between free will and the laws of nature in the context of a Humean or non-Humean ontology, see the contributions by Dorato (2022, Chap. 12), Esfeld (2022, Chap. 13), and Huttemann (2022, Chap. 10).
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
215
of the human mind, it does not come as a surprise that such conscious mental states are somehow connected with the underlying neural states.27 Therefore, it would be mistaken to simplistically assume that the presence of unconscious neural antecedents rules out conscious causation. Assuming that conscious mental states play a (causal) role, action control might still be guaranteed in virtue of the matching between the conscious intentions and the action, or the subject’s identification with the underlying mental states (De Caro, 2014). In the case of motoric actions, this matching can be properly understood, as discussed in Sect. 11.2.3, as a linkage between the general intention to act and the bodily action, with the more fine-grained aspects of the action being possibly dealt with outside conscious awareness. However, a different experimental strand, which is compatible with a particular interpretation of Libet’s work, can lend support to the claim that conscious mental states do not play any (causal) role in action production (see Mele, 2018; Nahmias, 2014). Based on this line, it has been suggested that our first-person perspective is constantly prone to error, much more than we would customarily assume, the upshot being that any impression to know and control our mental life is illusory (Carruthers, 2011; Doris, 2015). Taking this view to the extreme, the overall coherence we experience between mental states and actions would emerge from post hoc rationalizations that people inferentially put together to make sense of their subjective life (Carruthers, 2011; Wegner, 2002). A growing body of empirical evidence supports the view that we are easy prey to manipulations in ways that are not amenable to conscious reflection and reason-based assessment (Bargh & Chartland, 1999). To take one classic example, Carver used a task, inspired by the classic Milgram’s experiment (1974), combining priming exposure to hostile words with the instruction of administering shocks to a co-player (a confederate receiving ersatz shocks) to show that actions can be influenced by automatic environmentperception-behavior links bypassing conscious awareness. Participants who were exposed to hostile words exhibited a more aggressive response pattern compared to the control group (1983). More disputably (Caspar et al., 2017), even propositional beliefs, such as our faith in the existence of free will, can be effectively manipulated, resulting in overt behavioral changes, e.g., in terms of a correlation between the disbelief in free will and a rise in anti-social tendencies (Baumeister et al., 2009). The overall message is that our ability to act on reasons is rather volatile, easy to override, and readily fooled by post hoc rationalizations (Johansson et al., 2005; Wegner, 2002). If conscious mental states were bypassed by their unconscious antecedents or ruled out by automatic processes (in technical terms, if the conscious mind were epiphenomenal), it would be pointless to discuss action control in terms of the matching between intention and action – and free will (even compatibilist free will) would be, undoubtedly, illusionary. This view is therefore much more radical than Libet’s. Indeed, in the case of LAs, participants may not consciously decide
27
This relationship is usually understood in terms of one of the following notions: supervenience, emergence, realization, or reduction (Kim, 2003).
216
S. Bonicalzi and M. De Caro
when to act but retain conscious control over what to do (e.g., consciously flexing their finger, thus acting according to the instructions) and over whether to act or not. However, although bypassing theorists are arguably right that action control is partially outsourced to automatic processes, it is problematic to generalize (i) from individual, heterogenous findings to the general claim that our mental life, as a whole, is an illusion; (ii) from single action-tokens to every action of that type or to other action-types (see Sect. 11.2); (iii) from single aspects of action control to action control as a whole (see Sect. 11.2.3). Very dubiously, Wegner assumes that actions are all of a kind (2002), i.e., all consciously controlled or not. On the contrary, in this chapter, we have argued, on the basis of evidence and theorization, that actions are not all of a kind. Actions can be planned or unplanned (e.g., Libet et al., 1983; Verleger et al., 2016), goal-directed or habitual (e.g., Dolan & Dayan, 2013), deliberate or arbitrary (e.g., Maoz et al., 2019), mental or motoric (e.g., Soon et al., 2013), unattended or timed (e.g., Baker & Piriyapunyaporn, 2012) just to mention some of the dichotomies we discussed, and called into question, throughout the chapter. Even within single action-tokens, action control is not realized through fully conscious or fully unconscious processes but is a multi-layered process, enabled by different cognitive and brain patterns (e.g., Pacherie, 2015). As a result, the value of individual experimental findings cannot depend on their telling us whether we are free or not, but on their tracing the boundaries of how different action-types may work. For instance, Wegner’s wellknown experiments on the illusion of conscious control (2002) show that our feeling of controlling actions and their consequences in the external world is partially driven by postdictive, inferential processes, rather than being granted exclusively by predictions of action-outcome contingencies. Providing insights into how predictive and postdictive processes are balanced in view of action control, enabling also the corresponding subjective feeling of control (Haggard, 2017), is one of the challenges for the current and future philosophy and cognitive science of action.
11.4
LAs and the Causal Role of the RP
Whereas the place of the RP is so central within the way neurophysiologists conceive of free will, its dubious causal status makes it difficult to establish what its causal contribution to action production actually is. The causal connection between the RP and the ensuing action has been routinely assumed on the basis of its temporal location and stable association with LAs. However, decades after the initial Libet experiment, several questions about how, and even whether, the RP causes both conscious awareness and the ensuing action remain unanswered.28 In
28
Haggard and Eimer (1999) showed that there is no structural relationship between the onset of the RP and the time of conscious awareness W. By contrast, they suggested that the time of conscious awareness might be causally determined by the so-called “lateralized RP” (LRP), which reflects the
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
217
this section, we briefly introduce some more specific discussion points referring to how recent findings and theoretical proposals may provide some guidance in addressing multiple causal questions. In particular, we indicate three currently debated issues as central to advancing the discussion on the neurobiological roots of intentional or voluntary actions: (1) whether the RP is necessary for action production; (2) whether an action occurs whenever the RP is present; (3) whether the RP must be understood as a gradual buildup of electrical activity deterministically culminating in the action’s being produced. As mentioned in Sect. 11.2.1, the RP has been indicated as specific to spontaneous/endogenous actions, while being absent prior to instructed behaviors (Jahanshahi et al., 1995). If the specificity of the RP in this respect has been quite reliably established, it remains considerably less clear whether the RP has to be necessarily present for a voluntary action to be produced (question 1). Answering this central question has proven to be considerably difficult since one-on-one matches between single trials and their corresponding RPs are difficult to observe. Indeed, based on Kornhuber and Deecke’s method (1965), traditional experimental protocols demanded that participants repeatedly perform LAs for the RP to be isolated. Indeed, the RP can usually be extracted by averaging together the EEG signals, time-locked to EMG onset, of several trials. Insufficient numbers of trials are associated with a low signal-to-noise ratio, which makes it difficult to isolate an RP-like signal (Shibasaki & Hallett, 2006). While this limitation might be just due to the structural features of the ERPs (including the RP), some have argued that the undetectability of the RP at the level of single trials must prevent one from concluding that the RP is necessary for the voluntary action to occur (Pockett & Purdy, 2011). While the issue remains debated, recent studies have suggested that, by means of more innovative experimental techniques, it is actually possible to extract RPs in real-time and, more importantly for the present discussion, from single trials. In particular, Schultze-Kraft et al. (2016) conducted a study in which a Brain-Computer Interface (BCI) was trained to detect the typical pattern exhibited by the RPs in individual experimental subjects. Following the training and while participants were performing a motor task, the BCI was then able to predict when the subject was about to move, by detecting the onset of the RP at the level of single trials. Whenever the BCI detected an RP-like signal, it could forward a stop alert to participants. The alert could then be sent to participants during the time window between when the BCI detected the onset of the RP signal and when participants started moving, based precisely on the detection of single trials-RPs. This experiment is particularly relevant for discussing causation because it also speaks to another important causal question, that is, whether an action necessarily occurs whenever the RP is present (question 2). This problem was initially addressed
preparation of a specific bodily movement after the action has been already selected. From this, Haggard and Eimer drew the conclusion that W is linked to the specific bodily movement the subject is willing to perform rather than to a more general, unconscious motor preparation.
218
S. Bonicalzi and M. De Caro
by Libet himself (1985), who postulated the existence of a veto power (see note 10). The veto power consists in the subject’s ability to refrain from acting after the RP has already initiated, via an act of top-down conscious intervention. Interestingly, this provides a direct insight on how Libet himself thought of the kind of free will that humans can afford, i.e., in terms of the ability to permit or prevent something that was already occurring on its own. Returning to Schultze-Kraft et al.’s experiment (2016): after the BCI was fully trained in recognizing participants’ RP signals, the proper experimental task begun. Participants were then required to repeatedly perform LAs unless they received, at the level of single trials, a stop alert from the BCI. Had they received a stop alert, their task was to exert their veto power and refrain from acting. The reasoning behind this procedure is that in case an action occurs whenever the RP is present, participants should not have been able to terminate the action upon receiving the stop alert. Indeed, the BCI might have missed some RP signals in trials where no stop alert was sent, but at least the RP must have been present whenever a stop alert was sent (subject to false alarms). By contrast, the experimental results indicated that participants were actually able to refrain from acting in some of the trials where they had received the stop alert. In particular, for participants to be able to refrain from acting, the stop alert had to be sent at least 200 ms (the so-called “point of no return”, after which the ongoing action became unavoidable) before the onset of the bodily movement. In other words, in trials in which the time interval between when the stop alert was forwarded and the bodily movement began was shorter than 200 ms, participants were unable to terminate the process. This indicates that participants’ top-down veto intervention can terminate the action, which is therefore not made unavoidable by the mere presence of the RP, at least until a point of no return after which the action will naturally occur. Schultze-Kraft et al. (2016) show that actions can be prevented, until the point of no return, when participants are informed by a stop alert that tells them to terminate the process. However, this does not suffice to clarify whether, in the absence of a top-down intervention terminating the process, an action occurs whenever the RP is present. In a widely influential paper, Schurger et al. (2012) indirectly address this question by rejecting the received model of the RP as a slow buildup of electrical negativity reliably reflecting action planning, preparation, and initiation (see also Schurger, 2018) (question 3). According to Schurger et al.’s alternative model, the distinctive shape of the RP signal, peaking just before the action onset, is a mere artifact that materializes in the process of averaging together multiple trials, all timelocked to when muscular motion begins. Once the artifact is removed, the signal per se would merely reflect spontaneous electrical fluctuations of internal physiological noise. When there is a general intention to act (e.g., to produce a LA), the ongoing physiological process ultimately leads to a LA’s being produced, without the action’s being tied to any specific preparatory process deterministically progressing until muscular motion begins. The exact time of the action depends on when the spontaneous, stochastic electrical fluctuations cross a threshold. This model, in turn, raises new causal questions concerning whether spontaneous, RP-like, fluctuations that do
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
219
not reach the threshold might be present throughout the EEG continuous signal, without any LA’s being actually produced. These RP-like signals would go missing with classic averaging methods, where only action-related epochs are isolated and extracted from the continuous EEG data. However, a recent search of RP-like signals within continuous EEG data (Travers et al., 2020) did not find any support for this hypothesis. In other words, no bona fide RPs were found in EEG segments where no LA was actually produced, which seemingly reinforces the view that the RP specifically arises in connection with voluntary or intentional (i.e., endogenous/ spontaneous) behaviors. Overall, while progress has been made, these lively conceptual and empirical debates indicate that to build a precise and empirically informed matrix of actiontypes, the neurobiological causes of intentional and voluntary actions (i.e., how exactly these action-types are produced and what their connection with conscious states are) are still in need of further investigation.
11.5
Conclusion
In this chapter, we have suggested that the neurobiological experiments on the roots of intentional or voluntary behaviors cannot provide answers as to whether free will, in general, exists. Their contribution is nonetheless very valuable within a comprehensive theory of action aiming to account for different action-types. Indeed, as humans, we perform a wide array of different actions, which can be encapsulated in different operational definitions and are plausibly enabled by different neurobiological patterns. In this light, we proposed that the philosophy and the cognitive science of volition and action should join forces to provide more nuanced questions and answers concerning how people navigate the environment by means of different action-types. In this respect, the Libet’s study (or even the Libet tradition), which worked as our central case study, represents one piece of a more complex puzzle whose internal connections are not yet entirely explored. Relatedly, we hope that this chapter contributed to highlighting some of these pressing open issues, which in particular concern the need for a richer matrix of action-types and for a more comprehensive understanding of how the underlying causal processes actually work. Funding: S. B. and M. D. C. benefitted from the PRIN grant 20175YZ855 from the Italian government. S.B. further benefitted from a fellowship at the Paris Institute for Advanced Study (France), with the financial support of the French State, programme “Investissements d’avenir” managed by the Agence Nationale de la Recherche (ANR-11-LABX-0027-01 Labex RFIEA+). M. D. C. further benefitted from the project ‘Dinamiche pubbliche della paura e cittadinanza inclusiva’ financed by Università Roma Tre in the framework of the call ‘Azione 4: azione sperimentale di finanziamento a progetti di ricerca innovativi e di natura interdisciplinare’.
220
S. Bonicalzi and M. De Caro
References Anscombe, G. E. M. (1963). Intention. Blackwell. Aristotle. (2000). In R. Crisp (Ed.), Nicomachean ethics. Cambridge University Press. Baker, K. S., & Piriyapunyaporn, T. (2012). Neural activity in readiness for incidental and explicitly timed actions. Neuropsychologia, 50(5), 715–722. Balaguer, M. (2009). Free will as an open scientific problem. MIT Press. Bargh, J. A., & Chartland, T. L. (1999). The unbearable automaticity of being. American Psychologist, 54, 462–479. Baumeister, R. F., Masicampo, E. J., & DeWall, C. N. (2009). Prosocial benefits of feeling free: Disbelief in free will increases aggression and reduces helpfulness. Personality and Social Psychology Bulletin, 35, 260–268. Benecke, R., Dick, J. P., Rothwell, J. C., Day, B. C., & Marsden, C. D. (1985). Increase of the Bereitschaftspotential in simultaneous and sequential movements. Neuroscience Letters, 62, 347–352. Bernáth, L. (2019). Why Libet-style experiments cannot refute all forms of libertarianism. In B. Feltz, M. Missal, & A. Sims (Eds.), Free will, causality, and neuroscience (pp. 97–119). Brill-Rodopi. Björnsson, G., & Pereboom, D. (2014). Free will skepticism and bypassing. In W. SinnottArmstrong (Ed.), Moral psychology: Freedom and responsibility (Vol. 4, pp. 27–35). MIT Press. Bonicalzi, S. (2015). Libet-like experiments and the efficacy of the will. Rivista Internazionale di Filosofia e Psicologia, 6(1), 230–144. Bonicalzi, S. (2019). Actions and intentions. In B. Feltz, M. Missal, & A. C. Sims (Eds.), Free will, causality and neuroscience (pp. 120–142). Brill – Rodopi. Bonicalzi, S., & Haggard, P. (2019). From freedom from to freedom to: New perspectives on intentional action. Frontiers in Psychology, 10, 1193. Bortoletto, M., & Cunnington, R. (2010). Motor timing and motor sequencing contribute differently to the preparation for voluntary movement. NeuroImage, 49(4), 3338–3348. Bratman, M. E. (1987). Intention, plans, and practical reason. Cambridge University Press. Bratman, M. E. (2007). Structures of agency: Essays. Oxford University Press. Braun, M. N., Wessler, J., & Friese, M. (2021). A meta-analysis of Libet-style experiments. Neuroscience and Biobehavioral Reviews, 128, 182–198. Butterfill, S. A., & Sinigaglia, C. (2014). Intention and motor representation in purposive action. Philosophy and Phenomenological Research, 88(1), 119–145. Carruthers, P. (2011). The opacity of mind. Oxford University Press. Carver, C., Ganellen, R., Froming, W., & Chambers, W. (1983). Modeling: An analysis in terms of category accessibility. Journal of Experimental Social Psychology, 19, 403–421. Caspar, E. A., Vuillaume, L., De Saldanha, M., da Gama, P., & Cleeremans, A. (2017). The influence of (dis)belief in free will on immoral behavior. Frontiers in Psychology, 8, 20. Chik, J. (2022). The temporal structure of agency. In Powers, time and free will. Springer, this volume. Crockett, M. J., Kurth-Nelson, Z., Siegel, J. Z., Dayan, P., & Dolan, R. J. (2014). Harm to others outweighs harm to self in moral decision making. PNAS, 111(48), 17320–17325. Cunnington, R. W., Windischberger, C., Deecke, L., & Moser, E. (2002). The preparation and execution of self-initiated and externally-triggered movement: A study of event-related fMRI. NeuroImage, 15, 373–385. Davidson, D. (1978). Intending. In Y. Yovel (Ed.), Philosophy of history and action (Philosophical studies series in philosophy) (pp. 41–60). Springer.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
221
De Caro, M. (2007). How to deal with the free will issue: The roles of conceptual analysis and empirical science. In M. Marraffa, M. De Caro, & F. Ferretti (Eds.), Cartographies of the mind studies. Philosophy and psychology in intersection (pp. 255–268). Springer. De Caro, M. (2014). Analisi concettuale e scienza: il dibattito contemporaneo sul libero arbitrio. In M. De Caro, M. Mori, & E. Spinelli (Eds.), Libero arbitrio. Storia di una controversia filosofica (pp. 365–382). Carocci. De Caro, M. (2020). Realtà. Bollati Boringhieri. De Caro, M., Lavazza, A., & Sartori, G. (2019). Introduzione. In M. De Caro, A. Lavazza, & G. Sartori (Eds.), Siamo davvero liberi? Le neuroscienze e il mistero del libero arbitrio (pp. VII–XXX). Codice Edizioni. De Caro, M., Marraffa, M., & Vaccarezza, M. S. (2021). The priority of phronesis: How to defend virtue theory from three potentially fatal challenges. In M. De Caro & M. S. Vaccarezza (Eds.), Practical wisdom. Philosophical and psychological perspectives. Routledge. De Hann, D. (2022). The power to will freely: How to re-think about the problem of free will without laws of nature. In Powers, time and free will. Springer, this volume. Dezfouli, A., Lingawi, N. W., & Balleine, B. (2014). Habits as action sequences: Hierarchical action control and changes in outcome value. Philosophical Transactions of the Royal Society B, 369(1655). Dickinson, A., & Balleine, B. (2002). The role of learning in the operation of motivational systems. In Stevens’ handbook of experimental psychology. Wiley. Dolan, R. J., & Dayan, P. (2013). Goals and habits in the brain. Neuron, 80(2), 312–325. Dorato, M. (2022). The consequence argument and an ontology of dispositions. In Powers, time and free will. Springer, this volume. Doris, J. M. (2015). Talking to our selves: Reflection, ignorance, and agency. Oxford University Press. Duncan, J., & Owen, A. M. (2000). Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends in Neurosciences, 23(10), 475–483. Esfeld, M. (2022). Super-Humeanism and mental causation. In Powers, time and free will. Springer, this volume. Feltz, B., Missal, M., & Sims, A. C. (Eds.). (2019). Free will, causality, and neuroscience. Brill – Rodopi. Filevich, E., Kuhn, S., & Haggard, P. (2013). There is no free won’t: Antecedent brain activity predicts decisions to inhibit. PLoS One, 8(2), e53053. Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: A theory of moral responsibility. Cambridge University Press. Fourneret, P., & Jeannerod, M. (1998). Limited conscious monitoring of motor performance in normal subjects. Neuropsychologia, 36, 1133–1140. Frankfurt, H. G. (1978). The problem of action. American Philosophical Quarterly, 15(2), 157–162. Fridland, E. (2016). Skill and motor control: Intelligence all the way down. Philosophical Studies, 174(6), 1539–1560. Fried, I., Mukamel, R., & Kreiman, G. (2011). Internally generated preactivation of single neurons in human medial frontal cortex predicts volition. Neuron, 69(3), 548–562. Gallagher, S. (2006). Where’s the action? Epiphenomenalism and the problem of free will. In: W. Banks, S. Pockett, & S. Gallagher, (Eds.), Does consciousness cause behavior? An investigation of the nature of volition (pp. 109–124). MIT Press. Ghahramani, Z., & Wolpert, D. M. (1997). Modular decomposition in visuomotor learning. Nature, 386(6623), 392–395.
222
S. Bonicalzi and M. De Caro
Gold J. I., & Shadlen, M. N. (2003). The influence of behavioral context on the representation of a perceptual decision in developing oculomotor commands. The Journal of Neuroscience, 23, 632–651. Gollwitzer, P. (2003). Why we thought the action mindsets affect illusions of control. Psychological Inquiry, 14, 259–267. Haazebroek, P., & Hommel, B. (2008). Anticipative control of voluntary action: Towards a computational model. In I. G. Pezzulo & M. V. Butz (Eds.), Anticipatory behavior in adaptive learning systems: Foundations, theories, and systems (pp. 31–47). Springer. Haggard, P. (2014). Intention and agency. In I. M. S. Gazzaniga & G. R. Mangun (Eds.), The cognitive neurosciences (Vol. V). MIT Press. Haggard, P. (2017). Sense of agency in the human brain. Nature Reviews Neuroscience, 18(4), 196–207. Haggard, P., & Eimer, M. (1999). On the relation between brain potentials and the awareness of voluntary movements. Experimental Brain Research, 126(1), 128–133. Holroyd, J. (2015). Implicit bias, awareness and imperfect cognitions. Consciousness and Cognition, 33, 511–523. Huttemann, A. (2022). The problem of radical freedom. In Powers, time and free will. Springer, this volume. Jahanshahi, M., Jenkins, I. H., Brown, R. G., Marsden, C. D., Passingham, R. E., & Brooks, D. (1995). Self-initiated versus externally triggered movements. I. an investigation using measurement of regional cerebral blood flow with PET and movement-related potentials in normal and Parkinson’s disease subjects. Brain, 118(4), 913–933. Johansson, P., Hall, L., Sikström, S., & Olsson, A. (2005). Failure to detect mismatches between intention and outcome in a simple decision task. Science, 310(5745), 116–119. Kane, R. H. (1996). The significance of free will. Oxford University Press. Kane, R. H. (2002). Some neglected pathways in the free will labyrinth. In R. H. Kane (Ed.), The Oxford handbook of free will (pp. 406–437). Oxford University Press. Keramati, M., Smittenaar, P., Dolan, R. J., & Dayan, P. (2016). Adaptive integration of habits into depth-limited planning defines a habitual-goal–directed spectrum. PNAS, 113(45), 12868–12873. Khalighinejad, N., Schurger, A., Desantis, A., Zmigrod, L., & Haggard, P. (2018). Precursor processes of human self-initiated action. NeuroImage, 165, 35–47. Kim, J. (2003). Supervenience, emergence, realization, reduction. In M. J. Loux & D. W. Zimmerman (Eds.), The Oxford handbook of metaphysics (pp. 556–584). Oxford University Press. Koechlin, E., & Summerfield, C. (2007). An information theoretical approach to prefrontal executive function. Trends in Cognitive Sciences, 11(6), 229–235. Koechlin, E., Ody, C., & Kouneiher, F. (2003). The architecture of cognitive control in the human prefrontal cortex. Science, 302(5648), 1181–1185. Kornhuber, H. H., & Deecke, L. (1965). Hirnpotentialänderungen bei Willkürbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pflügers Archiv, 1–17. Lau, H. C., Rogers, R. D., Haggard, P., & Passingham, R. E. (2004). Attention to intention. Science, 303, 1208–1210. Lavazza, A., & De Caro, M. (2010). Not so fast. On some bold neuroscientific claims concerning human agency. Neuroethics, 3(1), 23–41. Levy, Y. (2013). Intentional action first. Australasian Journal of Philosophy, 91(4), 705–718. Libet, B. (1985). Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavioral and Brain Sciences, 8(4), 529–566.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
223
Libet, B. (2005). Mind time. The temporal factor in consciousness. Harvard University Press. Libet, B., Wright, E. W., & Gleason, C. A. (1982). Readiness-potentials preceding unrestricted “spontaneous” vs. pre-planned voluntary acts. Electroencephalography and Clinical Neurophysiology, 54(3), 322–335. Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential): The unconscious initiation of a freely voluntary act. Brain, 106(Pt 3), 623–642. Logan, G. D., & Crump, M. J. (2010). Cognitive illusions of authorship reveal hierarchical error detection in skilled typists. Science, 330(6004), 683–686. Maoz, U., Rutishauser, U., Kim, S., Lee, D., & Koch, C. (2013a). Predeliberation activity in prefrontal cortex and striatum and the prediction of subsequent value judgment. Frontiers in Neuroscience, 7, 225. Maoz, U., Shengxuan, Y., Ross, I. B., Mamelak, A. N., & Koch, C. (2013b). Predicting action content on-line and in real time before action onset – An intracranial human study. In Advances in neural information processing systems 25: 26th annual conference on neural information processing systems (pp. 872–880). Red Hook. Maoz, U., Yaffe, G., Koch, C., & Mudrik, L. (2019). Neural precursors of decisions that matter— An ERP study of deliberate and arbitrary choice. eLife, 8, e39787. Marmodoro, A. (2022). What’s dynamic about causal powers? A black box! In Powers, time and free will. Springer, this volume. Martens, J., & Roelofs, L. (2019). Implicit coordination: Acting quasi-jointly on implicit shared intentions. Journal of Social Ontology, 4(2), 93–120. Mayr, E. (2022). Freedom of the will and rational abilities. In Powers, time and free will. Springer, this volume. Mele, A. R. (1992). Springs of action. Oxford University Press. Mele, A. R. (1995). Autonomous agents. Oxford University Press. Mele, A. R. (2010). Effective intentions. The power of conscious will. Oxford University Press. Mele, A. R. (2018). Free will, moral responsibility, and scientific epiphenomenalism. Frontiers in Psychology, 9, 2536. Milgram, S. (1974). Obedience to authority; an experimental view. Harpercollins. Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24, 167–202. Moore, J. W., Wegner, D. M., & Haggard, P. (2009). Modulating the sense of agency with external cues. Consciousness and Cognition, 18(4), 1056–1064. Mylopoulos, M., & Pacherie, E. (2017). Intentions and motor representations: The Interface challenge. Review of philosophy and psychology. Review of Philosophy and Psychology, 8(2), 317–336. Nachev, P., & Hacker, P. (2014). The neural antecedents to voluntary action: A conceptual analysis. Cognitive Neuroscience, 5(3–4), 193–208. Nachev, P., & Husain, M. (2010). Action and the fallacy of the ‘internal’: Comment on Passingham et al. Trends in Cognitive Sciences, 14(5), 193–194. Nahmias, E. (2014). Is free will an illusion? Confronting challenges from the modern mind sciences. In W. Sinnott-Armstrong (Ed.), Moral psychology (Freedom and responsibility) (Vol. 4, pp. 1–25). MIT Press. Neal, D., Wood, W., Wu, M., & Kurlander, D. E. (2011). The pull of the past: When do habits persist despite conflict with motives? Personality and Social Psychology Bulletin, 37(11), 1428–1437.
224
S. Bonicalzi and M. De Caro
Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behavior. In R. J. Davidson, G. E. Schwartz, & D. Shapiro (Eds.), Consciousness and self regulation: Advances in research (Vol. IV). Plenum Press. Pacherie, E. (2008). The phenomenology of action: A conceptual framework. Cognition, 107(1), 179–217. Pacherie, E. (2015). In T. Metzinger & J. M. Windt (Eds.), Conscious intentions. The social creation myth. MIND Group. Passingham, R. E., Bengtsson, S. L., & Lau, H. C. (2010). Medial frontal cortex: From selfgenerated action to reflection on one’s own performance. Trends in Cognitive Sciences, 14(1), 16–21. Pockett, S. (2006). The neuroscience of movement. In S. Pockett, W. Banks, & S. Gallagher (Eds.), Does consciousness cause behavior? MIT Press. Pockett, S., & Purdy, S. (2011). Are voluntary movements initiated preconsciously? The relationships between readiness potentials, urges and decisions. In W. Sinnott-Armstrong & L. Nadel (Eds.), Conscious will and responsibility: A tribute to Benjamin Libet (pp. 34–46). Oxford University Press. Pockett, S., Banks, W. P., & Gallagher, S. (Eds.). (2006). Does consciousness cause behavior? MIT Press. Pornpattananangkul, N., & Nusslock, R. (2015). Motivated to win: Relationship between anticipatory and outcome reward-related neural activity. Brain and Cognition, 100, 21–40. Robinson, O. J., Frank, M. J., Sahakian, B. J., & Cools, R. (2010). Dissociable responses to punishment in distinct striatal regions during reversal learning. NeuroImage, 51(4), 1459–1467. Roediger, H. M., Goode, M., & Zaromb, F. (2008). Free will and the control of action. In J. Baer, J. Kaufman, & R. Baumeister (Eds.), Are we free? Psychology and free will. Oxford University Press. Roskies, A. L. (2010). How does neuroscience affect our conception of volition? Annual Review of Neuroscience, 33, 109–130. Roskies, A., & Nichols, S. (2008). Bringing moral responsibility down to earth. Journal of Philosophy, 105(7), 371–388. Rowe, J. B., Toni, I., Josephs, O., Franckoviak, R. S., & Passingham, R. E. (2000). The prefrontal cortex: Response selection or maintenance within working memory? Science, 288(5471), 1656–1660. Sanfey, A. G., Loewenstein, G., McClure, S. M., & Cohen, J. D. (2006). Neuroeconomics: Crosscurrents in research on decision-making. Trends in Cognitive Sciences, 10, 108–116. Schultze-Kraft, M., Birman, D., Rusconi, M., Allefeld, C., Görgen, K., Dähne, S., . . . Haynes, J. D. (2016). The point of no return in vetoing self-initiated movements. Proceedings of the National Academy of Sciences USA, 113, 1080–1085. Schurger, A. (2018). Specific relationship between the shape of the readiness potential, subjective decision time, and waiting time predicted by an accumulator model with temporally autocorrelated input noise. eNeuro, 5(1), e0302. Schurger, A., & Uithol, S. (2015). Nowhere and everywhere: The causal origin of voluntary action. Review of Philosophy and Psychology, 6(4), 761–778. Schurger, A., Sitta, J. D., & Dehaene, S. (2012). An accumulator model for spontaneous neural activity prior to self-initiated movement. Proceedings of the National Academy of Sciences USA, 109(42), E2904–E2913. Searle, J. R. (1983). Intentionality. An essay in the philosophy of mind. Cambridge University Press. Shariff, A., Schooler, J., & Vohs, K. (2008). The hazards of claiming to have solved the hard problem of free will. In J. Baer, J. Kaufman, & R. Baumeister (Eds.), Are we free? Psychology and free will. Oxford University Press.
11
How the Libet Tradition Can Contribute to Understanding Human Action. . .
225
Shepherd, J. (2014). The contours of control. Philosophical Studies, 170, 395–411. Shepherd, J. (2017). Skilled action and the double life of intention. Philosophy and Phenomenological Research, 170(3), 395–411. Shibasaki, H., & Hallett, M. (2006). What is the Bereitschaftspotential? Clinical Neurophysiology, 117(11), 2341–2356. Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychological Review, 84(2), 127–190. Sinnott-Armstrong, W., & Nadel, L. (Eds.). (2011). Conscious will and responsibility: A tribute to Benjamin Libet. Oxford University Press. Skinner, B. F. (1948). “Superstition” in the pigeon. Journal of Experimental Psychology, 38, 168–272. Slobounov, S., Hallett, M., & Newell, K. M. (2004). Perceived effort in force production as reflected in motor-related cortical potentials. Clinical Neurophysiology, 115(10), 2391–2402. Soon, C. S., Brass, M., Heinze, H.-J., & Haynes, J.-D. (2008). Unconscious determinants of free decisions in the human brain. Nature Neuroscience, 11, 543–545. Soon, C. S., He, A. H., Bode, S., & Haynes, J.-D. (2013). Predicting free choices for abstract intentions. Proceedings of the National Academy of Sciences USA, 110(5), 6217–6222. Steward, H. (2022). Laws loosened. In Powers, time and free will. Springer, this volume. Travers, E., Khalighinejad, N., Schurger, A., & Haggard, P. (2020). Do readiness potentials happen all the time? NeuroImage, 206, 116286. Trevena, J. A., & Miller, E. K. (2010). Brain preparation before a voluntary action: Evidence against unconscious movement initiation. Consciousness and Cognition, 19(1), 447–456. Ulrich, R., Nitschke, J., & Rammsayer, T. (2006). Perceived duration of expected and unexpected stimuli. Psychological Research, 70(2), 77–87. Verleger, R., Haake, M., Baur, A., & Śmigasiewicz, K. (2016). Time to move again: Does the Bereitschaftspotential covary with demands on internal timing? Frontiers in Human Neuroscience, 10, 642. Vul, E., Harris, P., Winkielman, P., & Pashler, H. (2009). Puzzlingly high correlations in fMRI studies of emotion, personality, and social cognition. Perspectives on Psychological Science, 4(3), 274–290. Wegner, D. M. (2002). The illusion of conscious will. MIT Press.
Chapter 12
The Consequence Argument and an Ontology of Dispositions Mauro Dorato
Abstract In this paper I discuss naturalistic, transcendentalist and ethical approaches to the problem of free will. After a brief introduction to the libertarian/ compatibilist debate, I show in what sense the notion of determinism and indeterminism on which it is based presupposes a philosophical position on the nature of laws of nature. After introducing van Inwagen’s consequent argument, I argue that Humean attempt to solve it fails it because the humean approach to laws is untenable and because the laws of nature do not depend on us in the sense advocated by humeans. In the last part of the paper, I claim that and ontology of dispositions opens the way to what is most important in human freedom, namely the ability to act in a certain way as in virtue centered morality, and in the capability approach already defended by Sen and Nussbaum. Keywords Determinism · Indeterminism · Libertarian compatibilist · Consequent argument · Humean compatibilism · Dispositional free will
12.1
Introduction
Schematizing, in the current literature on the age-old problem of free will we find, two opposite attitudes. On the one hand, we read that “the problem [of free will] boils down to a straightforward and (wide-open) empirical question about the causal histories of certain neurons” (Balaguer, 2010, p. 1). On the other, we read that “if my mental processes are determined wholly by the motions of the atoms in my brain,
This work has been supported by the Italian Ministry of Education, University and Research through the PRIN 2017 program “The Manifest Image and the Scientific Image” prot. 2017ZNWW7F_004. M. Dorato (*) Department of Philosophy, Communication and Media Studies, University of Rome Three, Via Ostiense 234, 00146, Rome, Italy e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_12
227
228
M. Dorato
I have no reason to suppose that my beliefs are true . . . and hence I have no reason for supposing my brain to be composed of atoms” (Haldane, 1927, p. 209, my emphasis). In the last quotation, the crucial word is “reason”: to the extent that the normative dimension of our concepts is regarded as irreducible to facts (in virtue of the factvalue distinction), it seems plausible to claim that the evaluation and the justification of any scientific theory presuppose free will because they depend on a normative dimension that makes them possible. In a recent, thought-provoking book, this normative dimension has even been regarded as based on an ontology of persons! (Esfeld, 2020). In my paper, I will discuss various attempts to reject the conclusion of the consequence argument as it has been formulated by van Inwagen’s infamous paper (van Inwagen, 1983) by showing that neither a Humean compatibilist solution nor an indeterministic metaphysics can avoid the difficulties posed by the argument. Rather, I will suggest that, in order to advance the discussion, it is opportune to connect it to the problem of the status of the laws of nature, that is indispensable to understand the notion of determinism and indeterminism. In this respect, I will defend an ontology of dispositions and capacity (Cartwright, 1989). Not only are regularities irreducible to dispositions and capacities (so that the status of laws as regularities is undermined) but, more importantly, an ontology of dispositions applies also to human actions and offers a new perspective on the problem of free will. This claim has three main motivations. First, neither a Humean compatibilist solution based on the so-called best system analysis of the nature of laws (BSA), nor an indeterministic metaphysics can avoid the difficulties posed by the consequence argument.1 Second, by enabling mental events to matter at the level of brain processes (they can be regarded as identical), an ontology of capacities makes room for strategies that are causally effective in bringing about the desired effects of our actions. Third, such an ontology is strictly related to our capacity to cultivate the so-called “Aristotelian virtues”, a term to be explained in the following sections. The cultivation of these virtues in the right social environment makes our life worth living independently of questions related to the indeterminism/determinism debate, (which become irrelevant), and allows us to formulate the debate differently, thereby defending a type of free will “worth wanting”.2 The plan of the paper is as follows. In the first section, I will briefly present van Inwagen’s consequence argument by showing the highly idealized assumptions on which it is based. I will argue that the traditional debate between compatibilism and libertarianism has reached a stalemate and that from the viewpoint of the intuitive 1
The literature on the best system analysis is immense. Usually Mill and Ramsey are considered the originators of the account. More recently, the view has been expanded and articulated in particular by Lewis 1973, 1983, 1986 and 1994. By doing injustice to many important contributors, here I will just list Loewer (1996) and, more recently, Callender and Cohen (2009). For of one of the most complete accounts of Lewisian’s approach to chance, see Hoefer (2019). 2 The expression is Dennett’s (1983).
12
The Consequence Argument and an Ontology of Dispositions
229
concept of freedom generates a dilemma. In the second section, I will discuss three compatibilist objections to the consequence argument in order to show why, oft-discussed naturalistic arguments are also insufficient to reject van Inwagen’s conclusion. After a discussion of additional, possible ways out of the consequence argument, in the third section, I will discuss three different approaches to the notion of determinism (syntactic, epistemic and ontic) in order to show why only the ontic approach is relevant for the discussion. The ontic approach may justify the view that laws supervene on dispositions, so that, strictly speaking, there are no universal laws of nature: since laws are true only in the mathematical models which they define or in man-made “nomological machines”, I will adopt in a conditional form the claim that we live in “a dappled world” of causal processes and capacities (Cartwright, 1999) on which laws supervene. What matters to our dignity as human beings is not whether we are free to do otherwise, as the consequence argument has it, but whether we can organize our educational system and societies in such a way that we can develop dispositions and character traits that can make our life blossom and worth living.3 From this perspective, freedom consists in the causally determined natural and social dispositions and capacity to fulfill our potentialities to achieve our well-being as humans, which, as in the capability approach, is a matter of what we are able to do. Note that in my approach to freedom as the capacity to live, van Inwagen’s consequence argument is not defeated, but it becomes irrelevant.
12.2
Three Notions of Freedom and Van Inwagen’s Consequence Argument
Van Inwagen’s consequence argument implicitly refers to the dispute between compatibilists and incompatibilist libertarians, since it revolves around two different meanings of “freedom of action”. For the sake of clarity, I will refer to these two senses of freedom with the label freedom1 and freedom2. According to compatibilism, an agent is free1 when she has the power, the ability, the capacity to do what she wants, even if her will is fully determined by previous conditions (biological constraints, character, past desires, education, social inputs, etc.). Compatibilists claim that we have this sort of freedom1 and that it is sufficient to guarantee that we are not puppets in the hands of fate: the key point is that our desires cause our actions. The second notion of freedom (freedom2) is espoused by the so-called libertarian incompatibilists, according to whom determinism and free will are incompatible. In addition to the “negative” kind of freedom as expressed by freedom1, they regard an agent as free2 if and only if she could have acted differently given the same past,
3 In a sense, Sen’s capability approach presupposes that freedom to achieve well-being is the fundamental notion of freedom (Sen, 1985). See also Nussbaum, M. (1988).
230
M. Dorato
according to a principle known as the Principle of Alternative Possibilities (PAP for short). In this second sense, the fact that an agent will do x rather than y is not fully constrained by whatever preceded her act of will. Accordingly, human beings are “originators of values” (Nozick, 1981, p. 294) and in some important circumstances their choices are “up to them”. The position according to which freedom is a precondition for doing science is yet another position on free will and should be distinguished from the two above. Even if this position ends up defending a form of libertarianism or incompatibilism, it does so by following a different, non-naturalistic conceptual framework, based on the irreducibility of norms to facts (Esfeld, 2020) and therefore deserves a separate treatment. The target of the consequence argument is the compatibilist approach to human freedom. In Inwagen’s original wording: “If determinism is true, then our acts are the consequence of laws of nature and events in the remote past. But it’s not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us” (van Inwagen, 1983, p. 56). Since the argument is correct, let us spell out its premises in order to evaluate their truth: (1) If determinism is true, then our acts are the consequences of laws of nature and events or states of affairs in the past. (2) It is not up to us what went on before we were born. (3) It is not up to us what the laws of nature are. (4) It follows that our choices and actions, expressed by propositions that are the consequences of (1)–(3), are not up to us. (5) If our present choices and actions are not up to us, we must do what we actually do as a matter of nomological necessity, and PAP fails Conclusion (6) The failure of PAP (implied by determinism) entails that we have no free will (freedom2). Let us discuss the premises of the argument in turn Once the notion of “consequence” is analyzed in terms of the deductibility of propositions about present events and actions from propositions about laws and initial conditions (“events in the past”), the first premise of the argument expresses the commitment to a determinist conception of the universe. As far as (1) is concerned, it is of the utmost importance to note at the outset that the concept of determinism in question is metaphysical and ontological rather than scientific, since it does not take into account, for instance, the effective computability of the algorithms necessary for the deductions in question (the “consequences”). The assumption that determinism as used in premise (1) is a meaningful notion seems question-begging. If this were the case, the consequence argument would lose any relevance. However, I will grant van Inwagen, as it is usually done in the literature, that the effective deducibility of the propositions expressing our actions is irrelevant, since
12
The Consequence Argument and an Ontology of Dispositions
231
determinism is an ontological or metaphysical thesis, while the effective computability could be regarded as a mere byproduct of epistemic (human or machine) limitations. I will come back to the important difference between different formulations of determinism (ontic and epistemic) in later sections of the paper. In a word, premise (1) presupposes that we could live in a physically (and certainly logically) possible world in which determinism holds but the actual capacity (by human beings or computers) to deduce any physical states from laws and initial conditions is irrelevant, as it would be the case with Laplace’s famous “intellect” or demon4 (Laplace, 1951, p. 4).
12.3
The Deterministic Horn and the Humean Compatibilist Solution
The second and the third premise of the argument – invoking the impossibility of changing the laws and the initial conditions of the universe – can be discussed together, as they have been recently put into doubt for very similar reasons. In order to present the relevant objections, and with apologies to other philosophers, I will restrict my discussion to a few central arguments put forward by Beebee and Mele (2002), Ismael (2016) and Esfeld (2020), and Esfeld (2022 this volume) who draw from the seminal work of Lewis (1973). These authors share a so-called Humean conception of laws, where laws are to be conceived as global patterns in the universe that are supervenient on local matters of fact. In this conception, laws are the axioms or theorems of deductive systems which strike the best balance between simplicity and strength (BSA).5 One of the (alleged) advantages of this view of laws is the possibility to defend a Humean form of compatibilism, in particular, if one considers that in this view there is no room for any notion of natural necessity. Humean compatibilists usually insist that laws “do not force us to do anything” (Ismael, 2016, p. 225). This claim, however, advanced a long time ago by modern and contemporary compatibilists (see for instance Carnap, 1966, Chap. 22), does not cut ice, since it holds independently of the different contemporary conceptions of laws of nature. It is only via anthropomorphic views of nature that can one hold that laws “compel” our action in the same sense in which a human being can. The most interesting claim that Beebee and Mele (2002), Ismael (2016) Esfeld (2020) and Esfeld (2022, this volume) (among others) put forward is that there is a legitimate sense in which both the past initial conditions and the laws of nature are up to us. Let us examine the counterintuitive claims of these three authors together,
4 An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed. . .nothing would be uncertain . . . . (Laplace, 1951, p. 4, my italics) 5 This conception of laws as axiomatic systems combining simplicity and strength has been famously advanced by Lewis (1973, p. 73)
232
M. Dorato
given that there is no essential difference between them. What does “up to us” mean? Usually, “up to us” means very plausibly “causally depend on us”. In a related sense, it may mean “to be able to make a proposition P expressing a law false by determining its truth value”. In a crucial passage, and by referring to the notion of ability, Beebee and Mele claim that: To say that the laws are such that together with past facts they entail that Fred eats breakfast tomorrow is already to presuppose that Fred eats breakfast tomorrow; but this presupposition, as we explained, does not itself strip Fred of the ability to decide whether or not to skip breakfast and act accordingly (Beebee & Mele, pp. 207–208, italics in the original).
By supervening on all local matters of fact, laws presuppose that Fred atensionally has breakfast somewhere in spacetime since the “global pattern” constituting Humean laws also contains Fred’s highly local choice and, in this sense, depends on or includes it. Unfortunately, there seems to be is type-token distinction that intervenes in the arguments above that the authors did not consider. When I claim that “I can make a proposition expressing a law false” (it is up to me), I claim that I have this type of capacity or ability. Even if determinism holds, in circumstances that are very similar to those that led me to raise my hand, I can refrain from doing it, because I have the capacity or the ability not to raise my hand. However, on determinism, this is compatible with claiming that if in a particular (token) circumstance I raise my hand, I do so because that particular action could not fail to happen since it is part of the Humean mosaic from eternity. Analogously, Ismael tries to reject the consequence argument by stressing that there is a sense in which the laws “depend” or determine the laws: When we adopt a globalist perspective, our activities become part of the pattern of events that make up history. Since our activities partly determine the pattern, and the pattern determines the laws, our activities partly determine the laws. (Ismael, 2016, p. 111).6
Now, this very clear quotation amounts to the claim that had we acted differently, the global pattern (the laws) would have been slightly different. Remember that our spatiotemporally, infinitesimally small actions are part of spacetime since eternity – Humeans are committed to the block view, given that the laws/pattern have to be globally valid. This conclusion is exactly what determinism entails and what compatibilists usually claim: I would have acted differently if my desires had been different. For instance, I could have raised my hand, but in that particular circumstance, I did not raise it because I did not want to. I could have raised it only if my desire had been different. But on the current hypothesis of determinism, the desire qua cause of the action could have been different only if the laws and the initial condition would have been different. However, this is just another way of claiming that my desire to raise my hand causes the corresponding act. And the cause in question is deterministic as it can be. It is only in this counterfactual sense that the
6
See Esfeld 2020, p. 65.
12
The Consequence Argument and an Ontology of Dispositions
233
laws, qua global and supervenient patterns, “depend” on my decision to raise my hand. Also libertarians of course do not claim that we “cause” laws to be what they are: on the contrary, laws enable all of our actions: clearly, we cannot cause them. Furthermore, it seems implausible to claim that our “highly local” actions (highly local matters of fact) determine regularities that, qua world-patterns, must be distributed across the universe. Relatedly, it is hardly believable that the laws of nature are what they are depending on whether we exist or even act in a certain way or not. Since laws are universally valid, the arrows of dependence point in the opposite direction. The cosmologist Ellis writes: “A unique role of the universe is in creating the environment in which galaxies, stars, and planets develop, thus providing a setting in which local physics and chemistry can function in a way that enables the evolution of life on planets such the Earth. If the cosmological environment were substantially different, local conditions would be different and we would not be here” (Ellis, 2007, p. 1183). By presupposing metaphysical and ontic determinism, on Humean determinism our actions cannot be different from what they are and all state of affairs at any time (our actions included) are fixed since eternity. Consequently, also on Humean determinism PAP fails because laws and the initial conditions are up to us only counterfactually, in the same sense in which we can trivially say that “we can change the future”. The statement in inverted commas means that: had we acted differently (a physically impossible fact) the future would have been also different. The actual future is what it is also thanks to our negligible causal powers. In sum, by taking the deterministic horn, there is no freedom2, which, admittedly, is the only kind of freedom that corresponds to our pre-theoretic sense of what it means to be free. Before concluding my brief presentation of the impact of determinism on our intuitive conception of free will, I must add that two positions need to be distinguished, a compatibilist and a hard determinist one. According to the former position, on determinism the only notion of freedom that survives is freedom1, a compatibilist stance that typically defends the claim that we are free1 if and only if our desire to do x originates from us, and no obstacles prevent us from doing x, even if our desire is a ring in a causal chain that goes back to events before our birth. The authors discussed above are certainly compatibilists, but I have shown that their attempt to reject van Inwagen’s argument fails. The latter position is taken by the hard determinists, who hold that, on the assumption of determinism, freedom2 is ruled out. So far they agree with the compatibilists. However, they also agree with the libertarians that freedom1 is incompatible with our intuitive sense of freedom. Since determinism is incompatible with both types of freedom, free will is either an illusion, the impression that we have free will is illusory (see, among others, Pereboom (2001) and Wegner (2002)). Before abandoning the intuition that besides freedom1 we also have the possibility to do otherwise (PAP), let us analyze the indeterministic horn.
234
12.4
M. Dorato
The Indeterministic Horn
Indeterminism opens the way to PAP because it implies that there is more than one possible future compossible with the same past and the same present. The past can be likened to the trunk of a tree, the possible futures being its branches, while the node separating the trunk from the branches are undetermined events that also include our decisions. By discussing recent proposals advanced in a naturalistic conceptual framework, I will now show why the introduction of genuine, irreducible randomness is not sufficient to rescue the intuition that we are free in the sense referred to by PAP There are at least three arguments in favor of my claim. 1. An important, non-metaphysical, and naturalistic source of randomness is obviously quantum mechanics.7 However, currently, we cannot exclude that the probabilities entering quantum theory are epistemic (as is generally assumed in Bohmian mechanics).8 The problem of interpreting non-relativistic quantum mechanics is therefore quite open and drawing metaphysics lessons from the theory looks premature. 2. But let us grant for the sake of the argument that quantum mechanics can not be completed by a deterministic theory and is therefore irreducibly stochastic. Even if probabilities were ontic, contrary to Kane’s hypothesis (1966), currently we lack any evidence for the existence of chaotic amplification of quantum events in our brains. By defending a specific kind of causation most plausibly typical for human beings, Kane invokes these kinds of amplification of quantum processes to justify the hypothesis that, in what he calls self-forming actions, we randomly choose among different courses of actions, all of which can be recognized as motivated by two conflicting reasons that, importantly, we can both can ours. For instance,9 if we faced a dilemma between (i) going to an important job interview and (ii) helping someone that injured herself by sliding on a wet floor, both actions would be motivated by reasons that, separately considered, we wholeheartedly would identify with. In order to make room for PAP, however, the choice between the two reasons for acting must be somewhat “random and ours”, where the randomness comes from the fact the relevant quantum process is, to put it with Bohr, absolutely “uncontrollable”. 3. This third objection perhaps is the most difficult to counter. Although so far we don’t know whether or at which level there is a transition between the quantum and the classical realms, it has been claimed that “the collective effects of all molecules moving about is to smear out any quantum indeterminacy” in our brain
7
For an irreducibly stochastic interpretation of quantum theory, see Ghirardi, Rimini and Weber (1986). 8 For a recent treatment, see Dürr, Goldstein, and Zanghì, Nino (2013). Whether the probabilities of Bohmian mechanics are epistemic, see Hoefer (2019). 9 This example is Kane’s.
12
The Consequence Argument and an Ontology of Dispositions
235
(Koch, 2009, p. 44). The reason for this statement is the extremely quick intervention of decoherence (ibid.), which destroys the holistic coherence of quantum states referred to also by Penrose (1994). As far as we know from computational neurobiology, it is at the level of the network of neurons the brain can be regarded for all practical purposes as a deterministic system even if, as any other physical entity, it is by itself a quantum system too. Even if, despite these three arguments, some chancy event in the brain were partially responsible for our decisions, our choices could not be regarded as being fully up to us: hence, the relevant action would not be fully explainable. It must be admitted with Kane that a element of chance separating our present decision from the values that we identified with in the past could still express our deeper personality. That is, during an important deliberation, we can and often have two conflicting reasons (to do x and abstain from doing it), both springing from our wholehearted commitment to what matters more to us. And that this is not incompatible with the existence of a random element in our brain that intervenes in making us decide to act one way or another, grounding both PAP and freedom. But in the end, on indeterminism, we cannot fully explain why we choose to help by giving up our interview (supposing that it cannot be postponed). On both compatibilism and libertarianism, it is the process of deliberation that generates a stronger inclination toward doing x or abstaining from doing it. In this sense, the deliberating process is generated by the agent. However, while on determinism, at least in principle, we can explain the “tilting of the balance” and the prevalence of one of the two reasons or desires, on indeterminism the outcome of the deliberation has no cause, in the same sense in which the particular moment in which a radioactive substance emits a beta particle is not only unpredictable but has literally no cause. There is literally no reason why one half of a bunch of photons emitted all in the same way and oscillating in a plane inclined at 45 degrees is in average absorbed or not absorbed by a polaroid lens inclined at 90 degrees. In a indeterministic world, PAP would hold, but our choices, qua originated by chance, would be equivalent to tossing a fair coin.10 There is a further distinction that is worth making. Independently of whether determinism holds of not, we have seen that according to common sense we are basically able or have the capacity to scratch our arm (action x) or refrain from doing so. That is, in types of circumstances that are extremely similar to those discussed in the consequence argument, we are free to do x or abstain from doing it (which is what PAP requires) in force of the fact that our brain is a chaotic system. On this very plausible hypothesis, the brain must be described by a determinist evolution equation that is highly sensitive to the initial conditions of the deliberating process, whose outcome, therefore, is completely unpredictable and practically indistinguishable from an indeterministic evolution. On this hypothesis, for all practical purposes PAP holds also in a determinist world: before a difficult decision, we
10
Assuming that the evolution of the process of coin tossing is indeterministic, which is not.
236
M. Dorato
cannot predict what we will do! Why isn’t unpredictability of others’ action and of our won not enough for freedom2? The problem is that this purely epistemic sense of PAP (the chaotic evolution is ontically deterministic) is not sufficient for freedom2 since, on the very idealized assumptions presented at the beginning, it calls into play only freedom1. Once we distinguish between an epistemic from an ontological sense of determinism, we should hold that in each token circumstance, by causing the action of moving our finger to scratch our head, we could not do otherwise. In a word, taking the indeterministic horn does not solve the dilemma either. On Humean regularism, our actions occur in a ridiculously small part of the cosmic tapestry that the laws of nature consist in. It seems implausible to claim that such a pattern would be “different” without our presence on Earth, but even if this were true, the laws would not be up to us: we should say that they enable us to make a decision and act consequently. Consider this passage from the cosmologist Ellis: “And how “different” must the laws be with respect to those holding in our world in order to deny one of the premises of the consequence argument?” Ellis (2007, p. 1183) However, even if we granted the Humean regularist that her position is sufficient to refute the consequence argument, it would still have to be considered to be a type of compatibilism, which does not make room for the second, more important sense of freedom, i.e., the freedom2 to do otherwise or PAP. Even if in ontic indeterminism PAP holds, we could not claim that actions are fully up to us.
12.5
The First Way Out of the Dilemma
We have seen that Balaguer (2010) defends a naturalistic view of the human being and a physicalist view of the brain, by claiming that the question whether we have free will or not “is an open scientific question”. In other words, he claims that the problem cannot be solved just by philosophical or conceptual analysis. Since we still know too little about the brain, the task of philosophers is to set the conceptual stage in order to let neuroscience decide whether we have freedom2, which is the intuitive sense of freedom that she is presupposing. If the right empirical, neural conditions were verified, then we would be free2. His argument seems transcendental in a sense that is opposite to Kant’s: supposing that we have free will in the intuitive sense (freedom2), which kind of empirical results should the scientists find in order to confirm this intuition? While trying to argue in favor of freedom2, Balaguer proposes a sort of interesting tertium quid between the two positions illustrated above. More in detail, he rejects both the idea that freedom1 is sufficient for our intuitive sense of freedom and that randomness should be identified with pure chance. According to him, there are various senses of “randomness”, and therefore of non-randomness, a notion that is not to be equated with determinism. There is one sense of non-randomness that intervenes when the self is torn between two real alternatives (these are the typical
12
The Consequence Argument and an Ontology of Dispositions
237
situations when free will is involved,)11 and then decides in favor of one of them apparently with no reason. Therefore, a free decision must be appropriately nonrandom – in order to defeat the arguments given in Sect. 12.2, the decision cannot just ‘happen’ to us but must be our own – but at the same time, in order to respond to the challenge of van Inwagen’s argument, it must be uncaused or in some sense undetermined. In a word, Balaguer argues that a free decision is appropriately nonrandom because it is the agent’s decision, but the decision at the same time is undetermined because, in the topological structure of time presented above, it must correspond to a node in the branching tree. If these are the conceptual requirements for having free will in the intuitive sense, in every situation in which we introspectively or phenomenologically feel that are deciding freely, the mental decision leading to our action seems to us to be free. The fact that free will “is an open scientific problem” is based on the claim that, on Balaguer’s explicit physicalism, every token mental act (our decisions included) must be identical with very complicated token neural processes. It follows that in case future discoveries in the neurosciences correspond to, and can be identified with, the appropriate mental state, then we are free: and quantum indeterminacies reinforce Balaguer’s view of free will, since the brain is a quantum system as every other system: if we cannot find the above indeterminist processes in the brain, then we must conclude that we are not free2. In a word, a person’s decisions are at least sometimes free if and only if their torn decisions “(a) are both undetermined and appropriately nonrandom, and (b) the indeterminacy is relevant to the appropriate nonrandomness in the sense that it generates the nonrandomness, or procures it, or enhance it, or increases it, or something along these lines” (Balaguer, 2010, p. 10). There are two objections to this brilliant proposal, which I cannot present in more detail for lack of space. The first objection is that in order to justify the claim that the undeterminedness of quantum mechanics generates or procures the appropriate nonrandomness, Balaguer needs some sort of causal hypothesis. ‘To procure’, ‘to bring about’ ‘to generate’ etc. are synonymous with “to cause”. This causal link between undetermined events in our brain (causes) and the appropriate nonrandomness of our decisions (effects) can either be deterministic or probabilistic. The former possibility – causal determinism – must be excluded for reasons that should be already clear from what I just illustrated: it is incompatible with freedom2. The latter, which for Balaguer is the only option, amounts to the idea that causal undeterminedness reinforces or makes more probable the appropriate nonrandomness in which free will consists, namely our authorship and control of a torn decision. Of course, it is not unthinkable that undetermined, completely uncaused events (recall the example of the absorption of the photons presented above) can explain via unconscious indeterministic processes why I choose to do x rather y: so far we have no definite evidence to think that at the appropriate explanatory level the brain’s
11 Balaguer 2010, p. 73. There are some differences with Kane’s notion of self-forming actions on which I cannot enter.
238
M. Dorato
decisions are deterministic or indeterministic. From this undeniable fact, his insistence on the empirical side to the problem – with which I agree – is justified. There is an additional difficulty, however, which is independent of future discoveries in the neurosciences: even a small amount of undeterminedness jeopardizes the explainability of our action in terms of our conscious reasons or intentions. The fact that before the choice we have reasons both for helping and leaving the injured person on the ground, does not make “our” “choice” fully explainable in terms of our reasons. So we fall back on the problem that was raised in the previous section. The third problem is raised by his defense of two possible senses of ‘nonrandom’: (1) nonrandom ¼ “causally or nomologically determined”, which is the standard meaning of the word, and (2) nonrandom ¼ “appropriate for human decisions”, therefore nonrandom as “not purely chancy”. Of these two definitions, the latter seems ad hoc and therefore relying on a petitio principii, motivated by Balaguer’s attempt to make room for the possibility that, in torn circumstances, an agent can be the author of the decision (can cause it), which in his view is the appropriate kind of nonrandomness. Given these two senses of nonrandom, another dilemma arises. The first horn is as follows: if nonrandom means appropriate for human decision, then the hypothesis that my fully conscious mental state causes my decision presupposes the existence of a special kind of agent causation (Chisholm, 1976). However, since Balaguer is a physicalist, this form of dual type of causation must be ruled out, and neither can be accepted the stipulation that “nonrandom” means “caused by the agent”. If non-random means deterministic (even if unpredictable), then we fall back on van Inwagen’s argument.
12.6
The Second Way Out of the Dilemma: Freedom2 as a Presupposition of Scientific Activity
On the other hand of the spectrum occupied by Balaguer’s naturalistic proposal, we find an argument in favor of the existence of freedom2 that relies on the logical gap between facts and norms (see Esfeld, 2020). According to this position, there cannot be any conflict between scientific determinism and freedom2 because the former belongs to the realm of facts, while the latter to the realm of values: given the gap, the two realms are incommensurable. There cannot be any such conflict– i.e. between the scientific and the intuitive image of free will (Sellars, 1956, 1962) – because the evaluation of any scientific theory presupposes epistemic and social norms, and these in their turn presuppose the possibility of choosing freely2 among them. If this is the case, any scientific theory purporting to show that there is freedom2 would be impossible and self-refuting, since its rejection would presuppose PAP. However, why would this be the case? I will focus on three related explanations of this hypothesis that are at the same time three attempts at refuting the consequence argument.
12
The Consequence Argument and an Ontology of Dispositions
239
O1 Any evaluation of a theory presupposes epistemic values or norms belonging to the realm of reasons: the acceptance of any scientific theory T, determinism included, presupposes the possibility to reject it and therefore PAP: it cannot be the case that we cannot but accept it as a matter of nomological necessity. Otherwise, the rational justification of our belief in a scientific theory (determinism included) would be groundless. O2 In particular during revolutionary periods, the choice between empirically equivalent theories depends on epistemic “virtues” (norms) like consistency, scope, simplicity, experimental accuracy, and fruitfulness (Kuhn, 1977 pp. 322–324) that are not reducible to facts or causal processes. O3 The construction of a theory T and the relevant experimental apparatus is possible only via very complex, rule-governed social interactions characterizing any collective activity striving to knowledge. My reply to these objections will be concise, but for my purpose, it will be sufficient to show that none of them is decisive against the consequence argument. R1 The correctness of determinism or of any theory T can be justified by a pragmatic notion of truth: any mental state corresponding to a belief is true if and only if it is instrumental to achieve our aims via a causal chain involving our behavior and the external world. Unlike having false beliefs, true beliefs are much more effective for our survival, and a causal theory of knowledge might in principle be able to offer additional arguments in favor of this claim. In order to aim at having the largest number of effective beliefs based on empirical claims, an overarching belief in the value of objectivity is necessary. It is only once we are committed to this value – as Monod stressed (Monod, 1971) – that we can believe in the effectiveness of all our scientific beliefs. However, the value of objectivity has an evolutionary origin, depends on the opposition between our wishes and reality, and our acceptance or rejection of it can be explained by a psychological and a sociological theory. We are not free to reject the fact that 2 + 2 ¼ 4 and the same holds for all mathematical truths: we accept the axioms for practical reasons linked to our aim to deduce the maximum amount of truths from a minimal number of presuppositions: deductive rules are socially adopted because they transmit the truth from the premise to the conclusion. R2 The Kuhnian hypothesis that epistemic values intervene in choosing among empirically scientific theories does not seem plausible. In the history of science, and even during scientific revolutions, the empirical equivalence among ontologically different theories has always been temporary. This remains true even if one accepts the thesis that scientific theories are in principle underdetermined by data. The geostatic theory has eventually been superseded by the Copernican theory, in the same sense in which, even granting their initial empirical equivalence, the phlogiston theory has been abandoned after Lavoisier’s oxygen theory. These remarks can be regarded as unfair. Lorenz’s dynamic approach to special relativity and Einstein’s kinematical approach are still regarded as rival (Brown,
240
M. Dorato
2005), even if only by a minority of philosophers. Likewise, the case of alternative formulations of quantum theory (Bohmian mechanics, dynamical reductions models, many worlds theories) and the standard view of quantum mechanics involving a collapse are usually regarded as empirically equivalent. And one could insist that in all of these cases Kuhn’s epistemic virtues play an indispensable role. However, the question whether the problem of the underdetermination of theories by data and the evaluation of empirically equivalent theories is really so troublesome for the methodology of science is quite controversial (Laudan & Leplin, 1991) and it has been argued that in the history of science it has never been the case that even one theory has ever been compatible with all known data (Norton, 2008), let alone two.12And the crucial norms of experimental accuracy and coherence have a pragmatical explanation: the numerical agreement between an experiment and a hypothesis ensures a more effective way to predict and control the external world. Likewise, the desire to eliminate contradictory propositions belonging to a theory may also depend on the purpose at hand, as Bohr’s 1913 theory of the atom shows,13even though in most cases the elimination of inconsistencies in our web of beliefs maximizes the probability of an effective action. R3 An additional problem with an appeal to the underdetermination argument as evidence that the evaluation of all scientific theories presupposes PAP is that the relationship between Kuhn’s epistemic values and freedom2 needs a deeper analysis. The fact is that the choice of one of these (possibly conflictual), epistemic values on the part of an individual or groups of scientists can be explained in principle via a deterministic, or causal sociological hypothesis, at least in the idealized definition of determinism proposed above. This approach substantially weakens all the objections above, in particular O3. Freedom2 is not necessarily required in order to choose among the various epistemic norms. Deliberations leading scientists to accept one theory rather than another by appealing to certain epistemic values are not different in nature from those characterizing some important decisions in our life. If the latter are caused by previous deterministic factors, so might be the former, even if they are much more complex. From a conceptual point of view, which here is the only relevant one, both the single scientist’s deliberations and the collective choices advanced by a scientific community do not rule out a deterministic, causal explanation. In science, instances of such explanations abound. For instance, an individual scientist’s desire to conform to the opinion of the most authoritative scientists, to receive more funds from the government or from a private company, to become more famous, to belong to a collaborative group, etc. do imply choosing among values in the same sense in which we choose one course of action rather than another in order to reach a certain goal. The choice of the single scientist may be correctly regarded as irrelevant: science is a collective enterprise. However, the commitment
12 13
For an opposite position, see Stanford (2006). For the usefulness of inconsistent science, see Vickers (2013).
12
The Consequence Argument and an Ontology of Dispositions
241
to social scientific norms like Universalism, Communality, Disinterestedness and Organized Skepticism (Merton, 1942) can be regarded as morally compelling for each single scientist for the simple reason that these norms have been learned during a long social and training process, the main feature of which can be explained in principle by determinist causes. It follows that the acceptance of a scientific theory can be regarded as severely constrained by certain causal factors (influence of a particular teacher, of the education, of the research group that has been joined etc.) which limit to a significant degree the argument proposed by Esfeld (2020). A final difficulty of the claim that science presupposes freedom2 is that, like all “transcendental” arguments TA, it cannot be empirically refuted: since any scientific attempt at disproving TA would presuppose its truth, the claim we are currently discussing becomes immune to any criticism from an empirical viewpoint.
12.7
A Third Way Out of the Consequence Argument: Kant’s Transcendental Viewpoint
Kant, notoriously, is a libertarian, but his transcendental argument in favor of his position is different from those presented in the previous sections, even if, in contemporary language, he endorses a radical separation between causes and reasons. According to Kant, there are two irreducible ways to regard human beings, one in phenomenological and the other in noumenal terms. By synthesizing spatiotemporal phenomena via the categories of the intellect, we know (kennen) that our actions are fully determined by Newton’s mechanical, deterministic laws (transcendental analytics). In the extra-phenomenical, noumenal world, on the contrary, we are legitimated to think (denken) of ourselves as free2 and as moved by reasons, but only because our freedom2 or autonomy are postulates of the universal moral law. In his own words, Kant’s dualism about the human nature – which is indispensable to ensure freedom2 – can be synthesized thus: “I had to limit knowledge to make room for faith”. Knowledge here refers to Kant’s phenomenalism, where Newtonian determinism holds sway. On the contrary, we cannot know the noumena but only “think” them. Relatedly, “faith” in the Critique has a special meaning that should not be misunderstood since it has nothing to do with superstition, bigotry or religion: faith must be made room for because our freedom2 cannot be proved or refuted by a scientific theory. The force of Kant’s argument lies in the claim exposed in his transcendental dialectic: if we tried to prove or disprove in empirical ways the existence of freedom2, we would always fall into the traps of the antinomies of reason since we would two opposite and equally plausible reasons to believe and disbelief in freedom2. Exactly for the fact that we will never have definite arguments in favor or against the libertarian conception, we are entitled to believe in freedom2, given that otherwise, the moral law would be illusory. More in detail, Kant’s belief in freedom2 is justified by his belief that we are capable of making fully autonomous decisions
242
M. Dorato
because when it comes to the noumenal aspect of our nature, we are immediately aware of being subject to the moral law (he refers to this as “a fact of practical reason”). Being subject to the noumenal world of reasons, such an awareness justifies the maxim that we should act as if we were free and autonomous, even though at the phenomenical level Newtonian determinism openly contradicts this belief. Note that Kant does not endorse the view that freedom is a presupposition of our scientific activity, but only that it is a transcendental condition to make possible our practical and ethical behavior. In short, due to his philosophical analysis of scientific knowledge, for Kant regarding ourselves as possessors of free will is a matter of decision and not of discovery: if PAP were an illusion, the moral law would be an illusion too. The main objection to this argument, on which I cannot dwell, is that the moral law is not based on a fact of reason, but to a significant extent a contingent historical byproduct of our evolutionary and social past.
12.8
Why Van Inwagen’s Argument Needs an Ontic Form of Determinism
In order to criticize more effectively the consequence argument, we need a more precise understanding of “determinism”, which in its turn entails a careful analysis of the notion of laws, without which determinism cannot be defined. In order to take a step forward in the discussion, let me endorse the crucial distinction between laws of nature and laws of science or natural laws as clearly spelled out by Scriven (1959) and others.14 From this perspective, I will argue that since the formulation of determinism adopted in the consequence argument presupposes the existence of laws of nature. If there were no laws of nature – see van Fraassen 1989, Giere 1999, Ward 2002, Mumford 2004 for a defense of an antirealist stance - but only laws of science (scientific laws as they are postulated in mathematical models of the world), the propositions expressing the laws in the mathematical models would have no reference. This would solve automatically the question of the relation between determinism/indeterminism and free will. As far as I know, this argument has not been put forward before. In order to justify my thesis, we should distinguish among a syntactical (a), an epistemic (b), and (c) an ontic approach to laws and, as we have seen at the beginning, and show that it is only (c) that is relevant in the debate on free will.
14
See also Weinert (1995).
12
The Consequence Argument and an Ontology of Dispositions
243
12.8.1 Syntactic Approach In van Inwagen’s formulation, determinism is defined syntactically in terms of the notion of deductive consequence: “If determinism is true, then our acts are the consequence of laws of nature and events in the remote past.” (van Inwagen, ibid.) More precisely, a system S is deterministic if and only from the propositions expressing the evolution laws L of the system (of the world if such a system is the world) and the description of all states of S, from L and S we can deduce all the propositions describing any other state of the world, including those that involve our decisions. Since deductions in physics are typically possible only within mathematical models, given the distinction above the consequence argument refers to laws of science as they formulated in mathematical models, and not to laws of nature, which for present purposes are our central concern, since determinism is a theory about the physical world. In the hypothesis, to be discussed below, that the mathematical models defined by scientific laws are hyper-simplified maps of the physical world, so that there are no laws of nature but only laws of science, neither indeterminism nor determinism could be defined, and the whole debate would over.
12.8.2 Epistemic Approach For our problem, epistemic approaches to determinism do not take us very far. As anticipated above, in epistemic approaches predictability plays the central role; given that predictability in scientific theories presupposes deducibility,15 and deducibility is possible only among propositions defining certain aspects of mathematical models, the epistemic approach is different from the syntactic approach in name only. Furthermore, in the consequence argument, predictability is not what is at stake. This point is easily forgotten. For instance, in Ismael’s latest book, we read: “we could thumb our noses to any Laplacian [demon]who claimed to know how we will act. Let him tell me what he predicts, and we will see who wins the game” (Ismael, 2016, p. 171, italics added). Suppose the demon predicts and communicates to me that I will have tea. I will then decide not to have it: his prediction fails. The same argument could be run if he predicts and tells me that I will not have tea. Given that he can only predict P or not P, the prediction cannot be verified and according to Ismael, as much as I understand her argument, van Inwagen’s argument fails. If we could solve the dispute in this way, we would trivialize it. Clearly, Ismael is correct in claiming that the demon cannot predict that I will have tea for breakfast, for the simple fact that I may want to falsify his prediction by not having it, and “Scientific determinism is the doctrine that the state of any physical system at any given future instant of time can be predicted . . . by deducing the prediction from theories. . . .”(Popper 1982, p. 36)
15
244
M. Dorato
conversely. However, this epistemic argument has no force against van Inwagen’s argument for five related reasons. Firstly, the fact that we can thumb our nose to his prediction presupposes that we are free2 to do so and does not prove it. Secondly, why should we formulate determinism in such a way that the demon tells me what I am going to do? This assumption seems ad hoc, since without any information coming from him, my action would always confirm his prediction. Thirdly: by endorsing epistemic approaches to determinism, epistemic determinism and freedom become trivially compatible. Determinism is not icncompatible with the fact that when we choose what to do by deciding to perform an action, we don’t know in advance what we will do. we cannot predict our decisions. And they, even if they can be necessitated by laws and previous events. Actually, this is exactly what determinism typically claims: very often, the causes of our decisions are unknown to us, even if they determine our choices. Fourthly, an identification between determinism and predictability neglects the wellconfirmed fact that the evolution of deterministic systems depending sensibly on initial conditions ‘soon’ becomes unpredictable, where ‘soon’ depends on the physical system under consideration. Consequently, we must separate determinism from predictability, even if, for all practical purposes, chaotic determinism and indeterminism are indistinguishable. Fifthly, Relatedly, Hoefer reminds us (Hoefer, 2016, Sect. 3) that, on the basis of a formal result due to Ornstein (1974), “Deterministic metaphysicians can comfortably hold to their view knowing they cannot be empirically refuted, but so can indeterministic ones as well.” (Suppes, 1993, p. 254, my emphasis). If we follow Suppes and fail to distinguish epistemic from ontic determinism, the debate between compatibilism and incompatibilism is over: presumably, empirical arguments involve epistemic considerations. To the extent that PAP requires an indeterministic metaphysics, but this doctrine is empirically indistinguishable from a deterministic metaphysics, the difference between compatibilism and incompatibilism disappears: whether PAP holds or not can never be established with an empirical argument. Unfortunately, this is not an effective way to solve the consequence argument. The problem whether determinism or indeterminism holds or not depends on empirical evidence, but cannot be based only on evidence. For instance, it is only thanks to a careful philosophical analysis that can we claim that determinism (indeterminism) are theses about the physical world involving in particular the status of the laws of nature. There is no reason why science, when certain evidential conditions are satisfied, should not be interpreted as a form of metaphysics. For instance, on the basis of scientific realism, quantum mechanics is usually taken to support a indeterministic metaphysics. In a word, failing to distinguish between empirical conditions and metaphysical hypotheses can only be justified within a instrumentalist framework.
12
The Consequence Argument and an Ontology of Dispositions
245
12.8.3 Ontic Approach In order to characterize the ontic approach to determinism and indeterminism, it is useful to invoke a definition proposed by Earman (1986), which is based on the notion of agreement: Determinism: Any two physically possible worlds ‘agreeing’ at one time, agree at all other times (Earman, 1986, p. 13)16 – as in a one-to-one function. Indeterminism: in the evolution of the world, there is at least a time t at which two possible worlds agreeing before t disagree afterward (one-to-many function). This approach helps to formulate more clearly both the epistemic unpredictability of the temporal evolution of the universe and the ontic characterization of indeterminism, represented by the universe branching into a set of non-epistemic, real possibilities: it is only in this metaphysical framework that PAP is satisfied. The crucial notion of ‘agreement’ raises the natural questions: agreement (between possible worlds) about what? Dispositions, properties, local matters of facts, point particles with their changing relations (Esfeld, 2020). Earman has a very nice way of putting the difference between reductive and nonreductive views of laws of nature: “Do any two possible worlds agreeing in (e.g.) regularities also agree in laws, or conversely?” (Earman, 1986, p. 85).
12.9
Free Will and the Problem of the Existence of Laws of Nature
Earman’s question enables us to explore a new path and advance the discussion. I have just argued that the syntactic or the epistemic versions of determinism do not suffice to reject the consequence argument. However, given that ontic determinism (or indeterminism) is true or false depending on the ontological structure of the world, it follows that the consequence argument becomes a real threat to our free will only if laws, whatever they are (regularities, dispositions), are part of the ontological furniture of the world. Consider the following three hypotheses. Suppose that (i) there are no laws of nature (see the various ways to articulate instrumentalist positions as developed by (van Fraassen, 1989; Giere, 1999; Mumford, 2004; Bird, 2005) or (ii), less radically, that “dappled” dispositions and capacities are the supervenience basis of regularities so that laws of nature are not the primary ontological posits of the physical word (Cartwright, 1999), or (iii) there is no causal (non-nomic) determinism. If these three hypotheses were true the consequence argument would fail. To be clearer, disjunct (i) implies that if there were no laws of nature, we could not have ontic
16
Here I ignore the difference between stable determinism and chaotic determinism.
246
M. Dorato
determinism (or indeterminism). The second alternative (ii) is based on the hypothesis that mathematical models of the physical world are highly idealized: so that the primary ontology is one of dispositions and causal powers. The third alternative (iii) presupposes that causal determinism is not reducible to nomic determinisms and that van Inwagen’s argument cannot be formulated in terms of causal determinism.. Even though the arguments presented in this section are sketchy and must be regarded as indications for future research, I trust that they can go a sufficiently long way to justify my main conclusion. The structure of the argument in favor of the view that laws of nature are not ontologically prior has the form of a trilemma: (1) laws do not exist, or, if they do, (2) they are either reducible or supervenient on regularities or (3) they are irreducible and non-supervenient. Given hypothesis 1 (for instance laws are to be reduced to symmetries (van Fraassen, 1989) van Inwagen’s argument can be immediately rejected. Therefore, in the rest of the paper I will first show that laws of nature cannot be identified with regularities (i.e. 2), so that (3) the ontology of the world is one of dispositions, capacities, and properties, only very approximately described by simplified mathematical models of the physical world. In a word in all of these three hypotheses, there are no strict laws of nature. In (6), I will show that even if determinism can be formulated in terms of causes (causal determinism) the consequence arguments would not have undesirable consequences and since an ontology of dispositions, capacities, and properties would suffice to vindicate our dignity of human beings.
12.10
The Problems of the Humean Conception of Laws
The most prominent reductionist approach to the role of modality in a theory of laws of nature has been formulated in different ways, but the essential, common idea is that laws are regularities, which is why the approach has been baptized Humean. The claim that laws supervene on local matters of fact is known as Humean supervenience: in Lewis’s words, “all there is to the world is a vast mosaic of local matters of particular fact, just one little thing and then another”. (Lewis, 1986b, p. ix). The main reductionists/regularist positions discussed in the literature come in two different forms: Humeanism and Superhumeanism: the Humean view as it is understood today identifies a theory with a deductive structure, where true laws are distinguished from accidental generalizations by identifying the former with axioms or theorems that combine simplicity and strength (Best System Analysis, or BSA). “A contingent generalization is a law of nature if and only if it appears as a theorem (or axiom) in each of the true deductive system that achieves the best combination of simplicity and strength” (Lewis, 1973, p. 73, italics added). Superhumeanism, a more radical version of reductionism, is ontologically more parsimonious than Humeanism, since it restricts the primitive ontology of the world on which laws supervene on structureless matter points changing their spatial distances and therefore evolving in time (Esfeld & Deckert, 2017).
12
The Consequence Argument and an Ontology of Dispositions
247
Note that in principle, Humean conceptions of laws are not antirealist about laws but only reductionist about modality. Since epistemic virtues like simplicity and strength seem to be mind-dependent, one could argue that laws end up being minddependent too. Be that as it may, if I managed to show that – even though laws could be regarded as regularities in an ontic sense – no satisfactory theory of regularities is available so far, we can abandon Humean supervenience and therefore safely neglect its possible contribution to realism about laws. There are at least five arguments against the BSA account to which, if, as far as know its defenders so far have not elaborated a satisfactory reply. 1. Axiomatizations depend on language, and simplicity is vague and languagedependent: each time the best balance between simplicity and strength must be found pragmatically or case by case. Consequently, on this approach, whether a proposition belonging to a theory counts as a law also depends on the solution to a pragmatic question, and this seems in partial conflict with an ontological approach to laws. 2. Most of the fundamental physical theories (except for classical mechanics and special relativity that are both strictly speaking false) so far have not been rigorously axiomatized. BSA is a regulative idea (in the sense of Kant) that became in part realized in physics in virtue of the influence of the Hilbertian research program. But we are still a long way from a rigorous axiomatization of quantum field theory17 or general relativity, the most fundamental, wellconfirmed theories of the physical world. And yet there are scientific laws also in these two theories, that within the BA cannot be regarded as such. To my knowledge, this second difficulty of the BSA has not been discussed. 3. There are laws, the law of inertia for instance, that are unexemplified by regularities (local matters of fact in Lewis’sense) and seem to be made true by irreducible dispositions (the disposition of bodies to persevere in their state of motion): “Every body perseveres in its state of being at rest or of moving uniformly straight forward, except insofar as it is compelled to change its state by forces impressed.”18 The Humean can reply to this objection, and in general to the objection raised by the laws that are not instantiated by local matters of fact (vacuous laws), by stressing the fact that the BSA catches the greatest number of truths (strength) with the minimum amount of complexity (simplicity), a position that is certainly faithful to how Newton presented his three laws of motion: he regarded his laws of motion as axioms And yet, by interpreting laws in this way, there is a sense in which the empiricist spirit is lost, given that the Newtonian laws hold ceteris paribus (Cartwright, 1989) and cannot be regarded as supervenient on spatiotemporal local matters of facts. Given the ubiquity of gravity, Newton was well aware of the fact that there may no physical body in the universe that
The claim that the algebraic approach is sufficient to regard quantum field theory as axiomatized is not uncontroversial. 18 Newton (1726), Law 1, 416. 17
248
M. Dorato
perseveres in its state of motion or is not accelerated (unless it is very far from any mass). 4. There is a distinction between initial conditions and equations, that cannot be rendered by the view that all there is are contingent local matters of facts evolving in time. 5. some law requires a modal distinction between the actual and the nomically impossible (a perpetuum mobile, or the speed of light in vacuum) that cannot be accounted for by a mere absence of a pattern formed by a local matter of fact (for some of these objections, see Dorato 2005). Despite its conciseness,19 I take it that this criticism justifies the conclusion that regularities are to be construed as supervenient on modal properties of nature. Consequently, the BSA account of laws cannot be used against the consequence argument, because if these objections are correct, regularities are not part of the fundamental structure of the world. If these modal properties sufficed to explain the existence of regularities without implying the existence of laws, laws would not be fundamental ingredients of the natural world, and the consequence argument would be rejected (see point 2 above). This is the task that will be tackled in the next section.
12.11
The Antireductionist Conception of Laws
In this section I will briefly review three antireductionist accounts of lawhood in order to show how the third in particular is more suitable for our aim. Firstly, Maudlin’s (2007) primitivist view holds that laws of nature are not just conceptually prior in our linguistic framework (Carrol, 1994) but also ontically prior. From this hypothesis, he derives the notion of possible models of the theory, which then helps him to define counterfactuals, and other related nomic notions. The problem with this view is that the move consisting in making laws primitive turns a philosophical notion that is in need of clarification into something that can neither be defined nor analyzed. His proposal amounts to trying to solve a philosophical problem (what is a law of nature) by making it non-understandable. It is true that in philosophy posing unclear and analyzable primitive notions is unavoidable and non-arbitrary since it is to a certain extent justified by the number of consequences that one can draw from the choice of the primitive. But if the chosen primitive notion as in our case requires at least some form of understanding, the choice seems quite wrong: what does it mean to claim that “laws exist”? Note that even if this difficulty could be solved, the primitive existence of laws would give a good growing soil to the consequence argument, so that primitivism could not be used for our purpose.
19
For a more articulated justification of these five points, see Dorato (2005).
12
The Consequence Argument and an Ontology of Dispositions
249
The same objection applies to Lange’s (1989) non-reductionist claim, according to which laws are made true by subjunctive facts and additional abstract conditions concerning set membership (Lange, 1989, Chap. 5). Lange rejects primitivism about laws but his position is in line with Maudlin’s ontological stance, and therefore can be used as a necessary condition for ontic determinism to hold (independently of whether subjunctive facts can play the role of lawmakers Demerest, 2012). Thirdly, laws can be regarded as supervenient on dispositions capacities or causal powers (see for instance Cartwright (1989, 1999), Ward (2002) Mumford (2004), and Bird (2005).20 Being the primitive ingredient of the physical world, dispositions cannot be reduced to regularities or laws of associations.21 By putting forward the hypothesis that dispositional properties are ontologically more fundamental of regularities, we capture two important features of science. One has been stressed by Giere, who holds that the mathematical models used in physics are defined by scientific laws, but the latter are highly idealized and simplified descriptions of the physical world and as such, are neither true nor false: “For a physicist, the literal, exact truth, is not what matters. What matters is being close enough to the purpose at hand, whatever that might be” (Giere, 1999, p. 78). This aspect involves the practice of science. The second aspect is more metaphysical: an ontology of dispositional properties makes better sense of the fundamental theory of the physical world, namely quantum mechanics (Dorato, 2007). For instance, an electron has intrinsic properties like mass, spin and charge distinguishing it from the zoo of the other particles (bosons, other fermions, etc.). These properties are discovered thanks to experimental interactions (Hacking, 1983). that is causal laws, or “causally effective strategies” (Cartwright, 1989) which are made possible by our knowledge of low-level phenomenological laws that are largely independent of high-level theoretical laws. Furthermore, an electron also has dispositions to display a certain definitive position or momentum depending on the kind of experiment one performs, for instance spin + h o – h as a function of the orientation of a Stern Gerlack magnet. This form of dispositionalism holds not just for dynamical collapse models (Dorato & Esfeld, 2010), but also in Bohr’s or Rovelli’s interpretation of quantum mechanics (Dorato, 2020). In classical physics, Maxwell’s laws model the interactions between the properties of the magnetic and the electric field, which have a dispositional nature, while Einstein’s field equation model mathematically the relations between the properties of the metric field and those of the matter field, both of which have certain dispositions to interact with each other according to the Einstein field equations. Dispositions, in a word, are not the truth makers of the simplified mathematical models of the world defined by scientific laws, because the latter are only highly
20
For the current purpose, there is need to enter into details to distinguish among these notions. The antireductionist camp also includes views claiming that laws are contingent relations of necessity between properties, explaining the repeatable character of regularities (see Armstrong, 1983; Dretske, 1977; Tooley, 1977) Within this conception, laws exist, so that it cannot be used to attack the consequence argument 21
250
M. Dorato
idealized approximations of properties of physical systems. It is important to insist that these properties and relations are mind-independent, in the sense that they exist and have features that do not depend on the causal interactions that enable us to discover them.22 Modulo quantum mechanics, where measurement interactions do not reveal preexisting properties, when causally stimulated dispositional properties manifest themselves in events, among which particularly important for us are those revealed in our experimental practice. To summarize, the most powerful attack to the consequence argument consists in showing that the necessary conditions for ontic determinism presuppose a ontic conception of laws. In instrumentalist views, ontic determinism obviously fails. In Humean supervenience approaches, laws are to be regarded as supervenient on the Humean mosaic, but the various difficulties raised by these reductionist approaches make a dispositional approach to the ontology of the world more convincing. The thicket of dispositions and causal properties, measured by experiments and in which the world consists under this hypothesis, is not described by scientific laws but is rather just the supervenient basis of regularities. The fact that within the dispositional approach there are no laws of nature, opens a very promising way to reject the consequence argument: laws are true only in abstract models and, as Cartwright has it, in man-built nomological machines (for instance, frictionless plane).
12.12
Properties, Capacities and Human Capabilities
The hypothesis that within the dispositional approach there are no laws of nature might be insufficient to dispel the ghost of the consequence argument.23 Causal determinism is still an open possibility: even if the dispositionalist approach presupposes that laws of associations are not reducible to causes, causal determinism might be used to reformulate the consequence argument in terms of causes. However, even if we had to accept causal determinism (a big if), an ontology of dispositions and capacities, unlike other ontological assumptions involving laws, enables us to introduce a conceptual framework in which our sense of dignity as persons can be guaranteed. Since in the dispositional approach to laws, laws are the supervenient basis of regularities, the debate on free will can be reformulated in terms of (single) capacities and powers of which human beings are endowed. Within a naturalistic conception of human beings, these capacities are part of our nature, of our genetic dispositions, but manifest themselves in very different way according to the social environment. In increasing order of complexity, and given the assumptions above, quarks, electrons, molecules, cells, organs, individuals, human beings and societies have
22
Quantum mechanics requires a more detailed treatment, since quantum measurements typically do not reveal preexisting values of the measured system. 23 I owe this objection to Carl Hoefer.
12
The Consequence Argument and an Ontology of Dispositions
251
various degrees of intrinsic, relational and dispositional properties. An ontology of properties and dispositions seems general enough to apply to our actions, despite the infinitely more complex ontological level characterizing human beings and their actions. Which properties, intrinsic or relational, which capacities would be desirable to have in order to dispel the ghost of human beings transformed into puppets in the hands of brute necessity? The conceptual link between an ontology of dispositions and the problem that we have discussed so far is straightforward. Of the utmost importance for our dignity is to find out whether we can attribute to ourselves the capacity of self-control and restraint, which presupposes in its turn the capacity to have second-order will that be causally effective in steering, encouraging or inhibiting our first order dispositions or desires (Frankfurt, 1982, pp. 81–95). Suppose that I have the first order will, or the first order disposition, to be arrogant or impolite by training (not due to my genes, since genetic determinism is false)24 or for whatever reason one can imagine. The first piece of evidence in favor of the possibility to acquire a second-order will is grounded in the plasticity of the human brain. It follows that in favorable educational circumstances, I can learn to develop the will, or the second-order dispositions, not to be arrogant or impolite, if I want to want to become such a person. The crux of the matter is: is this second-order desire itself determined?25 My answer, following Balaguer, is that the question whether we are endowed with the capacity of “wanting to want” is empirical. The reason for wanting such second-order desires may be determined by various factors: maybe we develop stronger self-esteem by trying to act kindly, the amount of pleasure that this second-order gives us is greater than the affective fallouts of behaving like a wanton, etc. Obviously, obsessive-compulsive people tragically lack the capacity of wanting to want, and there plausibly is a continuous variation between this mental disorder and an impossible and undesirable perfect dominion over one’s first-order desire, and there are certainly some character traits that some individuals have that are hard to uproot even if one wanted to. And yet, if our freedom consists in, as Frankfurt has, in our being able or capable to will to will, or to desire to desire, at least from our experience, it does not seem unreasonable to attribute to ourselves the capacity of changing some character traits learned from culture or education or even passed on to us by our genetic inheritance. Of course, this capacity can be attributed only via a case by case analysis.
24
See among other text, Buchanan A, Brock D, Daniels N, Wikler D (2000). Frankfurt distinguishes between first order desires and first order will: “it is having second order volitions, and not having second-order desires generally, that I regard as essential to being a person (Frankfurt 1982 in Watson p. 86). Smith et al. prefer to naturalize the concept of will by identifying it with a concept of a desire than that is stronger than any other desire that a persona may have in a certain moment “I say that to be a value means to be that which we desire to desire. Then to be a value-to be good, near enough-means to be that which we are disposed, under ideal conditions, to desire” (Smith et al. 1989, p. 116). I treat ‘will’ and ‘desire’ as interchangeable.
25
252
M. Dorato
A dispositional approach to properties here is very useful. If we tend to be shy, irritable or fearful, one can effectively change these dispositions by acting as if we were outgoing, calm and courageous thanks to a second-order disposition, a suggestion that we learned from Aristotle and Pascal and that can be causally effective in Cartwright’s sense after a long training.26 In a word, since the possibility of these changes cannot be ruled out a priori, we should not that some evidence in its favor comes not only from our own experience but also from the history of philosophy. For instance, according to the Stoic philosophers and for Spinoza, our freedom consists in resisting the desire to want events to be different from what they are in virtue of an attempt to understand their causal role as part of an immutable lawlike-order. It remains to be seen whether a dispositionalist account of the laws of nature is sufficient to reject van Inwagen’s argument. Frankfurt’s position is usually described as compabilist: “It seems conceivable that it should be causally determined that a person is free to want what he wants to want. If this is conceivable, then it might be causally determined that a person enjoys a free will” (ibid, p. 92). However, the following quotation shows that what he is after is only an attempt to define what it means to have a free will “It seems to me both natural and useful to construe the question of whether a person’s will is free in close analogy to the question of whether an agent enjoys freedom of action. Now freedom of action is . . . freedom to do what one wants to do. Analogously, then, the statement that a person enjoys freedom of the will means . . . that he is free to want what he wants to want. More, precisely, it means that he is free to will what he wants to will, or to have the will that he wants. Just as the question about the freedom of an agent’s action has to do with whether it is the action he wants to perform, so the question about the freedom of the will has to do with whether it is the will that he wants to have.” (Frankfurt, 1982, p. 90). To conclude, I take it that Frankfurt’s formulation of the problem of free will, together with the postulation of an ontology of dispositions, which bridges the gap between the scientific and the manifest image of free will, enables us to overcome the traditional debate posed by van Inwagen by formulating a tertium quid between compatibilism and incompatibilism as described above. On the one hand, our second order disposition to be disposed toward a certain way of acting maybe causally determined. On other hand, if we have the capacity to want what we want to want (or desire what we want to desire), the question whether determinism holds or not is irrelevant for our dignity of persons, as long as we can organize our societies in such a way that we can develop the dispositions that make our life worth living. The indissoluble link between nature and culture suggests that becoming kinder, more loving, more knowledgeable, more generous persons makes our life and our societies “flourish “and blossom” in the sense of Aristotle’s Nichomachean Ethics, developed in contemporary philosophy especially by Martha Nussbaum (Nussbaum 1988, Nussbaum & Sen, 1993) in terms of the so-called capability approach to ethics and politics.
Via complex but effective “spiritual exercises”, Hadot (1995) shows how effective stoic teaching was in learning to want what we don’t want.
26
12
The Consequence Argument and an Ontology of Dispositions
253
References Armstrong, D. (1983). What is a law of nature? Cambridge University Press. Balaguer, M. (2010). Free will as an open scientific problem. The MIT Press. Beebee, H., & Mele, A. R. (2002). Humean compatibilism. Mind, 111, 201–223. Bird, A. (2005). The dispositionalist conception of laws. Foundations of Science, 10, 353–370. Bohr, N. (1913). On the constitution of atoms and molecules. Philosophical Magazines, 26(6), 1–25, 476–502, 857–875. Bradie, M., & Harms, W. (2020). Evolutionary epistemology. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2020 ed.) https://plato.stanford.edu/archives/spr2020/ entries/epistemology-evolutionary/ Brown, H. (2005). Physical relativity. Oxford University Press. Buchanan, A., Brock, D., Daniels, N., & Wikler, D. (2000). From chance to choice. Cambridge University Press. Callender, C., & Cohen, J. (2009). A better best system account of lawhood. Philosophical Studies, 145(1), 1–34. Carnap, R. (1966). Philosophical foundations of physics. Basics Books. Carrol, I. (1994). Laws of nature. Cambridge University Press. Carroll, J. (2016). Laws of nature. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Fall 2016 ed.) https://plato.stanford.edu/archives/fall2016/entries/laws-of-nature/ Cartwright, N. (1989). How the laws of physics lie. Oxford University Press. Cartwright, N. (1999). The dappled world: A study of the boundaries of science. Cambridge University Press. Chisholm, R. (1976). Persons and objects. Open Court. Davidson, D. (1970). Mental events. In L. Foster & J. W. Swanson (Eds.), Experience and theory (pp. 79–101). University of Massachusetts Press. Demerest, H. (2012). Do counterfactuals ground the laws of nature? A critique of Lange. Philosophy of Science, 79, 333–334. Dennett, D. (1983). Free will. The varieties of free worth wanting. Oxford University Press. Dennett, D. (2003). Freedom evolves. Penguin Books. Dorato, M. (2005). The software of the universe. Ashgate Dorato, M. (2007). Dispositions, relational properties and the quantum world. In M. Kistler & B. Gnassonou (Eds.), Dispositions and causal powers (pp. 249–270). Ashgate. Dorato, M. (2020). Bohr meets Rovelli: A dispositionalist account of the quantum limits of knowledge. Quantum Studies Mathematics and Foundations. https://doi.org/10.1007/s40509020-00220-y. ISSN: 2196-5609 (Print) 2196-5617. Dorato, M., & Esfeld, M. (2010). GRW as an ontology of dispositions. Studies in History and Philosophy of Modern Physics, 41, 41–49. Dretske, F. (1977). Laws of nature. Philosophy of Science, 44, 248–268. Dürr, D., Goldstein, S., & Zanghì, N. (2013). Quantum physics without quantum philosophy. Springer. Earman, J. (1986). A primer on determinism. Reidel Publishing Company. Ellis, G. (2007). Issues in the philosophy of cosmology. In J. Butterfield & J. Earman (Eds.), Philosophy of physics (Part B) (p. 1183, 1286). Elsevier. Esfeld, M. (2020). Science and human freedom. Palgrave. Esfeld, M. (2022). Super-Humeanism and mental causation. In A. Marmodoro, C. Austin, & A. Roselli (Eds.), Time, laws and free will. Springer (this volume). Esfeld, M., & Deckert, D. (2017). A minimalist ontology of the natural world. Routledge. Frankfurt, H. (1982). Freedom of the will and the concept of a person. In G. Watson (Ed.), Free will (pp. 81–95). Oxford University Press. Ghirardi, G. C., Rimini, A., & Weber, T. (1986). Unified dynamics for microscopic and macroscopic systems. Physical Review D, 34, 470–491. Giere, R. (1999). Science without laws. University of Chicago Press.
254
M. Dorato
Hacking, I. (1983). Representing and intervening. Cambridge University Press. Hadot, P. (1995). Philosophy as a way of life. Blackwell. Haldane, J. B. S. (1927). Possible worlds and other essays. Chatto and Windus. Hoefer, C. (2016). Causal determinism. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2016 ed.) https://plato.stanford.edu/archives/spr2016/entries/determinismcausal/ Hoefer, C. (2019). Chance in the world. A Humean guide to objective chance. Oxford University Press. Ismael, J. (2016). How physics makes us free. Oxford University Press. Kane, R. (1966). The significance fo free will. Oxford University Press. Koch, C. (2009). Free will, physics, biology, and the brain. In N. Murphy, G. F. R. Ellis, & T. O’Connor (Eds.), Downward causation and the neurobiology of free will. Springer. Kuhn, T. (1977). The essential tension. Chicago University Press. Lange, M. (1989). Laws and lawmakers. Oxford University Press. Laplace, P. S. (1951). In F. W. Truscott & F. L. Emory (Eds.), A philosophical essay on probabilities (Translated into English from the original French) (6th ed.). Dover Publications. Laudan, L., & Leplin, J. (1991). Empirical equivalence and underdetermination. Journal of Philosophy, 88, 449–472. Lewis, D. (1973). Counterfactuals. Harvard University Press. Lewis, D. (1983). New work for a theory of universals. Australasian Journal of Philosophy, 61, 343–377. Lewis, D. (1986a). Are we free to break the laws? In Philosophical papers (Vol. II, pp. 291–298). Oxford. Lewis, D. (1986b). Philosophical papers (Vol. II). Oxford University Press. Lewis, D. (1994). Humean supervenience debugged. Mind, 103, 473–390. Lewis, M., & Johnston, M. (1989). Dispositional theories of value. In Proceedings of the Aristotelian Society, London, Supplementary volume, Vol. 63, pp. 89–111+113–137+139. Loewer, B. (1996). Humean supervenience. Philosophical Topics, 24, 101–126. Maudlin, T. (2007). Metaphysics within physics. Oxford University Press. Merton, R. K. (1942). Science and technology in a democratic order. Journal of Legal and Political Sociology, 1, 115–126. Monod, J. (1971). Chance and necessity: An essay on the natural philosophy of modern Biology. A. Knopf. Mumford, S. (2004). Laws in nature. Routledge. Newton, I. (1726). Mathematical principles of natural philosophy (I. B. Cohen & A. Whitman, Trans.). University of California Press, 2002. Norton, J. (2008). Must evidence underdetermine theory? In M. Carrier, D. Howard, & J. Kourany (Eds.), The challenge of the social and the pressure of practice: Science and values revisited (pp. 17–44). University of Pittsburgh Press. Nozick, R. (1981). Philosophical explanations. Harvard University Press. Nussbaum, M. (1988). Nature, functioning and capability: Aristotle on political distribution. Oxford Studies in Ancient Philosophy (Supplementary volume), 6, 145–184. Nussbaum, M. (1992). Human functioning and social justice. In defense of Aristotelian essentialism. Political Theory, 20(2), 202–246. Nussbaum, M., & Sen, A. (1993). The quality of life. Clarendon Press. Ornstein, D. S. (1974). Ergodic theory, randomness, and dynamical systems. Yale University Press. Penrose, R. (1994). Shadows of the mind. Oxford University Press. Pereboom, D. (2001). Living without free will. Cambridge University Press. Popper, K. R. (1982). The open universe. Rowman and Littlefield. Scriven, M. (1959). The key property of physical laws – Inaccuracy. In H. Feigl & G. Maxwell (Eds.), Current issues in the philosophy of science – Proceedings of Section L of the American Association for the Advancement of Sciences. Holt Rinehart and Winston, 1961, pp. 91–104.
12
The Consequence Argument and an Ontology of Dispositions
255
Sellars, W. (1956). Empiricism and the philosophy of mind. In H. Feigl & M. Scriven (Eds.), The foundations of science and the concepts of psychology and psychoanalysis. University of Minnesota Press. Sellars, W. (1962). Philosophy and the scientific image of man. In R. Colodny (Ed.), Frontiers of science and philosophy (pp. 35–78). University of Pittsburgh Press. Sen, A. (1985). Well-being, agency and freedom: The Dewey lectures 1984. Journal of Philosophy, 82(4), 169–221. Smith, M., Lewis, M., & Johnston, M. (1989). Dispositional theories of value. In Proceedings of the Aristotelian Society, London, Supplementary volume, Vol. 63, pp. 89–111+113–137+139. Stanford, K. (2006). Exceeding our grasp: Science, history, and the problem of unconceived alternatives. OUP. Suppes, P. (1993). The transcendental character of determinism. Midwest Studies in Philosophy, 18 (1):242–257. Tooley, M. (1977). The nature of laws. Canadian Journal of Philosophy, 7, 667–698. Van Fraassen, B. (1989). Laws and symmetry. Oxford University Press. van Inwagen, P. (1983). An essay on free will. Oxford University Press. Vickers, P. (2013). Understanding inconsistent science. Oxford University Press. Ward, B. (2002). Humeanism without Humean supervenience: A projectivist account of laws and possibilities. Philosophical Studies, 107, 191–218. Wegner, M. D. (2002). The illusion of conscious will. Princeton University Press. Weinert, F. (Ed.). (1995). Laws of nature. Essays on the Philosophical, Scientific and Historical Dimension. Walter de Gruyter.
Chapter 13
Super-Humeanism and Mental Causation Michael Esfeld
Abstract The stance that has become known as Super-Humeanism is a minimal ontology of the natural world in the spirit of scientific realism. The paper enquires to what extent this ontology, due to its parsimony, can be employed to remove commitments from the scientific image that are the source of the clash with the view of ourselves as persons (the manifest image). The paper argues that SuperHumeanism provides the conceptual space to accommodate persons as irreducible to matter in motion, attribute to them free will (even in the libertarian sense) and recognize their volitions as causing some of their behaviour without coming into conflict with science. Keywords Free will · Humeanism · Libertarianism · Manifest image · Mental causation · Persons · Scientific image · Super-Humeanism · Laws and free will
13.1
(Super-)Humeanism, Laws and Free Will
At the core of how we conceive of ourselves as persons, there is free will and mental causation in the sense that a good deal of our behaviour is caused by free decisions. We could have done otherwise and are therefore responsible for what we actually do. This conception of ourselves is part and parcel of what is called – following Sellars (1962) – the manifest image. This paper seeks to make a new contribution to bring this conception of ourselves together with the scientific image. The idea is that an ontology that is minimally sufficient to accommodate what natural science (notably physics) tells us about the world leaves enough conceptual space open to include the central features of the view of ourselves as persons. What is known as Super-Humeanism is a minimal ontology in this sense. On the basis of what Super-Humeanism says about physical magnitudes, laws and causation, this paper points out the conceptual space left open
M. Esfeld (*) University of Lausanne, Lausanne, Switzerland e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 C. J. Austin et al. (eds.), Powers, Time and Free Will, Synthese Library 451, https://doi.org/10.1007/978-3-030-92486-7_13
257
258
M. Esfeld
for free will (this section) and then explores the account of persons and mental causation that can be built on this basis. Humeanism about laws of nature and causation in the natural world is, roughly speaking, the view that first comes the configuration of matter of the universe and its evolution, then come the laws and with them causal relationships between events as supervening on the evolution of the configuration of matter. Hence, it is not the laws or causation that fix, determine or produce that evolution. Quite the contrary, the evolution that actually happens fixes or determines the laws and with them causal relationships. Causal relationships between earlier and later stages of that evolution obtain, but they depend on the evolution that actually occurs instead of predetermining it. There are laws because this evolution exhibits some stable, but contingent regularities. Given such regularities, what is known as the best system is the system that achieves the optimal combination of simplicity and informational content in representing the evolution of the configuration of matter of the universe. Laws are theorems of the best system. The laws then specify which causal relationships obtain. The best system – and thus the laws – can be deterministic in the sense that given the propositions stating the laws and the specification of initial conditions about the configuration of matter at any given time, the propositions about the entire past and future evolution of the configuration of matter are entailed by these propositions. But there is no predetermination of anything. The supervenience basis for the laws is, strictly speaking, the entire evolution of the configuration of matter of the universe. Hence, if one formulates laws on the basis of the configuration of matter and its evolution that is accessible to us at the present day, one may be justified in taking these laws to be the laws of the universe (or approximations of them). But even if this is so, there is nothing in the present configuration of matter and the laws, thus conceived, that governs, predetermines or even produces the future evolution of the configuration of matter. That notwithstanding, the laws still are explanatory, namely in the sense of unification: they bring out the salient patterns in the evolution of the configuration of matter. Explanations in science end once such patterns are identified. For instance, suppose that a law of gravitation is part of the best system. Science cannot explain why there is gravitation, that is, why bodies attract each other. Science can only trace the particular phenomena of attractive motion back to the general pattern of gravitation. The same holds in case the best system contains no deterministic, but only probabilistic laws. Again, the laws and with them the objective probabilities for certain events to occur supervene, strictly speaking, on the entire evolution of the universe. This comes out in a clear manner in what is known as the Mentaculus account presented in Loewer (2012). Our assumptions about the objective probabilities for future events formulated on the basis of the configuration of matter and the evolution that is accessible to us at the present day can be correct. Yet, there is nothing in the present configuration of matter and the laws, thus conceived, that instantiates a tendency in nature for a particular future evolution to occur. In other words, there are no fundamental propensities, dispositions or power in nature that ground the objective probabilities.
13
Super-Humeanism and Mental Causation
259
Against this background, it has been established in the literature that Humeanism offers a distinct stance to make laws of nature – be they deterministic or probabilistic – compatible with human free will. Jenann Ismael, in her recent book How physics makes us free, writes: When we adopt a globalist perspective, our activities become part of the pattern of events that make up history. Since our activities partly determine the pattern, and the pattern determines the laws, our activities partly determine the laws.1
Hence, free will is compatible with the laws of nature because the physical events that are expressions of the free will of humans are part and parcel of the supervenience basis of the laws. By way of consequence, if persons had chosen to do otherwise, the laws of nature would have been slightly different. However, one can object that this stance gives us too much: our bodily movements are not that important in the universe that they have an influence on what the laws of nature are.2 Whatever humans choose to do, the laws will be the same. Quite the contrary, laws of nature are useful for our actions, because they set the frame for what we can do and what we cannot do. In other words, we need laws of nature and must keep them fixed to delimit the range within which we can freely choose our actions. For instance, a person can choose to walk slowly or quickly, but she cannot jump up to the roof of her house – gravitation prevents her from doing so. How fast and in which direction her body moves is an issue of the initial conditions and not of the laws. It is here that the position that has become known as Super-Humeanism in recent years makes a contribution to the free will debate.3 It thereby joins several proposals that have been set out recently and that all seek to vindicate free will not by taking human decisions to contribute to determining what the laws are, but to contribute to determining what exactly the initial or boundary conditions of the universe are.4 Super-Humeanism is in the first place a proposal for an even more parsimonious ontology of the natural world than standard Humean metaphysics as exemplified by David Lewis’s thesis of Humean supervenience, formulated in Lewis (1986, introduction). It deletes the commitment to natural intrinsic properties instantiated at space-time points and keeps only the commitment to one type of natural relation, namely distance relations individuating point particles. The ontology of the physical world is given by these two axioms conceived in Esfeld and Deckert (2017, p. 21):
1
Ismael (2016, p. 111; see also pp. 225–226). See also Beebee and Mele (2002), Swartz (2003, Chap. 11, in particular p. 127) and Perry (2004, pp. 237–239). 2 See the objection of Hüttemann and Loew (2019) and Huttemann (this volume, Chap. 10). 3 Super-Humeanism is set out in Esfeld and Deckert (2017, Chap. 2.3) and in Esfeld (2020a, Chap. 2). For critical discussions, see Wilson (2018), Marmodoro (2018), Darby (2018), Lazarovici (2018), Simpson (2019) and Matarese (2020a, b). The step from Humeanism to Super-Humeanism is inspired by the work of Hall (2009, § 5.2) and Loewer (2007); see also Loewer (2020b). 4 See Hoefer (2002), Loewer (2020a) and Thyssen and Wenmackers (2021).
260
M. Esfeld
(1) There are distance relations that individuate simple objects – namely, matter points. (2) The matter points are permanent, with the distances between them changing. Everything else supervenes on the Humean mosaic thus defined. Consequently, not only the laws, but also all the physical magnitudes apart from position supervene on the relative positions of the point particles and their evolution throughout the history of the universe. The reason is that in physical theories, only position has to be admitted as primitive and all the other magnitudes can be introduced in terms of their functional role for the evolution of the particle positions and thus be located in that evolution.5 That is to say: the values of such diverse magnitudes as mass, charge, constants of nature, the wave function of the universe, etc. are not instantiated in the initial configuration of matter of the universe, but are fixed only in the course of the evolution of that configuration. Super-Humeanism hence is a parsimonious ontology – arguably the most parsimonious one – that endorses scientific realism: it frees us not only from the heavy metaphysical baggage that comes with countenancing primitive modality (fundamental dispositions, powers, laws as primitives, etc.), but also from the problems that arise from subscribing to natural intrinsic properties: if they are not dispositions or powers, implementing a primitive modality, they are categorical properties with the ensuing commitments to quidditism and humility, as acknowledged by Lewis (2009). Furthermore, it solves the problem that the wave function in quantum mechanics poses for Humeanism: the wave function of the universe supervenes on the evolution of the configuration of matter of the universe that consists only in the positions of point objects (point particles in Bohmian quantum mechanics, flashevents in the GRW quantum theory).6 That notwithstanding, the propositions that ascribe masses, charges, a wave-function, etc. to physical systems as well as the propositions about laws and causation are true; but their truth-maker is only the mosaic of point particles that is constituted exclusively by the distance relations that individuate them and the change in these relations. The transition to Super-Humeanism has the following implication for the account of free will: not only the laws, as on standard Humeanism, but also the values of the magnitudes that enter as initial conditions into the laws are determined by the change that actually occurs in the relative particle positions throughout the history of the universe. This opens up the conceptual space for the following position: there is no need to maintain that human bodily motions contribute to determining what the laws are. But they contribute to determining what the precise values of some magnitudes that are part of the initial conditions – such as constants of nature, or the initial wave function – will turn out to be. As mentioned above, the laws set the frame within which we can act freely such that how precisely a person moves her body is an issue of the initial conditions, but not the laws. Assuming that a fundamental physical
5 6
See Esfeld and Deckert (2017, Chaps. 3–5) and Esfeld (2020a, Chaps. 1 and 2). See Miller (2014), Esfeld (2014), Callender (2015) and Bhogal and Perry (2017).
13
Super-Humeanism and Mental Causation
261
theory that is deterministic is part of the best system of the universe and that persons can freely choose, for instance, what drink to have at breakfast, something in the best system has to be different in the event a person chooses tea on a given occasion instead of choosing coffee. That difference does not affect the laws. It is a very tiny difference in initial values of some magnitudes, such as constants, or the wave function.7 However, there is no backward causation involved here: what humans choose to do – in general, what happens in the universe at a time t or after t – does not alter past observations or touch upon the validity of records of the past, since these are all position observations and are recorded as spatial configurations. The Super-Humean account of free will accommodates the common sense truism that the past is fixed and the future open: our free choices at t influence only particle positions that obtain later than t. For this to be possible under the assumption that the fundamental physical theory of the universe is deterministic, our free choices at t also have to contribute to determining what the exact initial values of some magnitudes are that figure in the initial conditions that enter into the laws. In order to establish this, Super-Humeanism plays out the distinction between position as primitive variable on the one hand and the additional magnitudes that enter into the initial conditions for a law of motion on the other hand. These latter are functionally defined in terms of the role that they play for the motion that actually occurs. Their initial values can therefore be dependent on future motions, including motions that are the result of free will, without any paradox arising. They are placed or located in the particle motion as a whole. This stance hence implies that such tiny differences in initial values of some magnitudes as are required, for instance, to accommodate the situation in which a person has tea at a certain time t instead of coffee do not show up in differences in particle motion before t, but only after t. Determinism does not rule this out; only what is known as super-determinism in the discussion about the free choice of measured variables in quantum mechanics would do so, as pointed out, for instance, in Esfeld (2015). Indeed, Bohmian quantum mechanics can serve as illustration of a deterministic physical theory that allows such a scenario, because the universal wave function can be considered as supervening on and thus being located in the particle motion as a whole – that this, the whole evolution of the particle positions of the universe serves as the supervenience basis that fixes the wave function of the universe modulo further constraints such as simplicity. Furthermore, what is arguably the fundamental version of Bohmian quantum mechanics, namely the theory known as identity-based Bohmian mechanics, allows for particles even to move in such a way that the values of mass change during their evolution, while the theory still is deterministic.8 Apart from Bohmian mechanics, Dowker and Herbauts (2005) have set out a concrete
7
See Esfeld (2020a, Chap. 2.4, 2020b). See Goldstein et al. (2005a, b) for the physical theory and Esfeld et al. (2017) for a philosophical assessment. 8
262
M. Esfeld
model of how the wave function supervenes on primitive positions and their evolution in the framework of the GRW flash ontology (which includes a probabilistic law, but this is irrelevant to the point at issue of the initial wave function supervening on the subsequent evolution of primitive positions that actually occurs in the universe). In general, an analogy with Kripke’s (1982) considerations about Wittgenstein (1953, §§ 138–242) on rule following may help to illustrate the point at issue here: any given sequence of items (here: particle positions up to a time t) instantiates many rules (here: wave functions) such that the difference between these rules shows up only in the further evolution of the sequence (here: the evolution of the particle positions after t). In sum, Super-Humeanism gives rise to a distinct proposal of how there can be free will in a universe with natural laws. More precisely, this proposal satisfies what is known as the principle of alternative possibilities: the agent could have done otherwise at t given exactly the same physical conditions, namely exactly the same evolution of particle positions up to t. It thereby accommodates also a libertarian conception of free will. However, this account of free will is only a half-built house: we also need an account of mental causation in this framework. This is what I seek to sketch out in the following. Section 13.2 recalls the problem of mental causation. Section 13.3 goes into how mental events can make a difference in the physical domain in the framework of (Super-)Humeanism. Section 13.4 enquires whether the Humean view of causation can be the appropriate one for mental causation as well.
13.2
The Problem of Mental Causation
The problem of mental causation is usually formulated as consisting in four principles each of which is plausible considered in isolation, but the conjunction of all four of them is inconsistent. Following Kim (1998, Chap. 1), the principles can be summed up in this way: (1) Distinction: Mental events are not identical with physical events. (2) Mental causation: Some mental events cause physical events. (3) Completeness: For any physical event p, there is a complete physical cause insofar as p has a cause at all. (4) No systematic overdetermination: Physical events are not systematically overdetermined by physical causes and additional mental causes that are distinct from the physical ones. Rejecting mental causation (2) would not be a solution to the problem, but denying that there is a problem, since mental causation would then not exist. Overdetermination (rejecting (4)) arguably does not yield a satisfactory solution either: it is central to the view of ourselves as persons that our volitions make a difference to our behaviour. But they do not make a difference if for every physical
13
Super-Humeanism and Mental Causation
263
movement that they cause there also is a complete physical cause that is distinct from the mental one. We thus get to the traditional opposition between physicalism on the one hand and dualism on the other that dominates the debate from the exchange of letters between Hobbes and Descartes to this day. On physicalism, mental events – at least those ones that are causally efficacious – are identical with physical events so that the causal completeness of the physical domain is respected and no systematic overdetermination through distinct causes arises. However, this solution to the problem of mental causation is not as such an argument for physicalism. To establish physicalism, one has to show how the features that characterize the mind can be part and parcel of the physical domain as described by natural science. Doubts in that respect remain notably as regards the qualitative features of conscious experience and normativity as being central to the deliberations about what one should think and do. However, dualism faces the objection of being outdated by modern science: the principle of completeness (3) is a philosophical principle. But it is not clear how one can reject this principle while endorsing realism with respect to what science tells us about the natural world. Most Humeans about laws of nature and causation are physicalists. The above mentioned recent book by Ismael (2016) makes an excellent case as to how a physicalist stance about the mind, including free will, can be combined with Humeanism. However, Humeanism does not do any specific service to physicalism: if one endorses physicalism, the problem of how the mind is related to the physical world in general – and the problem of mental causation in particular – are solved in terms of some form or other of the identity theory, independently of which stance on laws of nature and causation one adopts. By contrast, most dualists subscribe to an anti-Humean view of laws and causation that admits some sort of primitive modality, for instance, by endorsing fundamental dispositions or powers. However, this heavy metaphysical baggage makes it the more difficult (if not impossible) to solve the problem of mental causation while respecting scientific realism, because it is not clear how the mental powers could cause behaviour without clashing with the physical powers as they are conceptualized by the laws of physics.9 The rather unusual combination of dualism and Humeanism is therefore worth exploring, and be it only for the sake of the argument, to get clarity about the logical space of available positions: if the ontologically light view of laws and causation in Humeanism and, in particular, Super-Humeanism is to make a specific contribution to solving the problem of mental causation, this will be a contribution in the dualist framework and, as mentioned above, a contribution that accommodates also libertarianism about free will.
9
For a proposal in that respect that takes the laws of physics not to be strict laws see von Wachter (2015). See also Steward (this volume, Chap. 9) and De Haan (this volume, Chap. 8).
264
13.3
M. Esfeld
Difference Making
Given the assessment in the previous section, the psycho-physical dualism worth exploring here is one that takes mental events to make a difference in the physical realm without, however, coming into conflict with scientific realism.10 Abandoning scientific realism would be too cheap a solution to the problem of mental causation: at the heart of the problem is the apparent conflict between what science tells us about the physical realm and the common sense view of persons whose volitions make a difference in the physical realm while being distinct from physical events and being such that a person is free to choose what to do within the frame set by the physical laws. If (Super-)Humeanism is to contribute to solving this problem, its contribution stems from its parsimonious ontology. The idea is this one: by proposing an ontology that is minimally sufficient to accommodate scientific realism, it avoids a commitment to entities that entail a conflict with the view of ourselves as persons conceptualized in what Sellars (1962) calls the manifest image. The position we’re after hence is this one: mental events such as volitions, being distinct from physical events, make a difference in the course of physical events and thus in the particle motion. However, there is no need to amend (if not distort) physical theories in such a way that they include mental variables in their dynamical laws in order to cover the dynamics of physical events. These physical theories as they stand tell the truth about the physical realm (or theories in the same vein as our current ones do so). The expectation is that (Super-)Humeanism makes this position available. If (Super-)Humeanism can do this trick, then it does so based on the idea that the laws are fixed only by the entire evolution of the physical events. Suppose that this evolution includes changes that are caused by mental events such as volitions. Then, the laws are fixed by physical events some of which have irreducibly mental causes. But these causes do not figure in the supervenience basis for the laws. The laws include only physical magnitudes: the best system is the system that achieves the best balance between simplicity and strength with respect to the entire evolution of the physical events, whatever their causes may otherwise be. Science considers only physical events. It discovers general patterns or regularities in the evolution of these events and on this basis formulates general laws that apply to all physical systems. Even if there is mental causation and dualism is true, there is no reason for science to consider mental causes and include mental variables in its theories: doing so would compromise gravely the simplicity of the dynamical laws without leading to a gain in information. Only if the general patterns of interactions between physical events were altered in certain complex physical systems – such as, for instance, human brains – would it be necessary to amend the laws when it comes to these systems. However, pace the recent argument in Gillet (2016), there is no evidence for that: the neuroscientific research that we know is applied physics, namely applied classical mechanics and electrodynamics and in 10
See also Kroedel (2020, Chaps. 2 and 4).
13
Super-Humeanism and Mental Causation
265
the last resort applied quantum mechanics.11 In any case, the whole point of (Super-) Humeanism in this context is that we can countenance mental causation even in a dualist framework without having to search for a breakdown of the physical laws somewhere in the brain. If, nevertheless, mental events make a difference in the physical realm, the following has to be true: the course of physical events in a purely physical world is different from the course of physical events in a world with mental causes intervening in the physical world, although in both cases the dynamical laws contain only physical variables. Again, the difference between Humeanism and SuperHumeanism is relevant here: on Humeanism, in the dualist framework under consideration here, there has to be a difference on the level of the laws of physics between a world with only physical events and a world with mental causation. As explained in Sect. 13.1, on Humeanism, free will (free choice and thus the ability that the person could have acted otherwise) is possible on the condition that the laws of physics depend also on what humans actually do. On Super-Humeanism, by contrast, there only has to be a difference on the level of the initial values of some physical magnitudes that enter into physical theories through the functional role that they play for the evolution of the configuration of matter (such as slight differences in the precise values of some constants, or in the initial wave function of the universe). Again, the idea that the existence of mental causation makes a difference to the latter is much more palatable than having to attribute to mental causation such an importance that its existence makes a difference to the physical laws. Whatever we freely choose to do cannot affect the laws of physics; but it can contribute to determining the precise initial values of some magnitudes. In sum, the laws of physics and/or the initial values of some physical magnitudes have to be different in a world with mental causation from a world without mental causation. In a sense, the principle of completeness (3) as well as the principle of no systematic overdetermination (4) are both respected and violated by this stance. The principle of completeness is respected, since what enters into the best system in order to fix the dynamical laws and the causal relations derived from them are the physical events and their evolution only. Hence, insofar as a physical event has a cause given by a law of nature, it has a complete physical cause. However, the spirit or the idea that drives this principle is violated: some of the physical events upon which the laws of nature and the causal relations derived from them supervene have mental causes; what the natural laws and/or the initial values of some physical magnitudes are depends on the mental causes. In other words, the claim is that the physical causes do not exclude the mental ones, because the fact of some physical events having mental causes is already included in the basis of physical events on which the relations of physical causation supervene. Consequently, there is systematic overdetermination of some physical events by both mental and physical causes. More precisely, whenever a physical event has a
11
But see also the considerations about physics accommodating higher-level causation by Simpson and Horsley (this volume, Chap. 2).
266
M. Esfeld
mental cause, it also has a physical cause, and mental events are not identical with physical events. Admitting mental causation requires psycho-physical laws in the framework of Humeanism. These are not strict or exceptionless laws as in fundamental physics. Nevertheless, there is a salient pattern or regularity in the sense that whenever a person has the volition to do x with x being some form of behaviour that she is capable to do, x occurs, ceteris paribus. Hence, whenever there is mental causation, there are both psychophysical and physical laws and, based on them, both mental and physical causes. Nonetheless, these are not independent of one another: the supervenience basis for the physical laws and causes – the evolution of the physical events – includes already the fact that there is mental causation. Consequently, fixing the physical causes includes the mental ones. Although the letter of the principle of no systematic overdetermination is violated, its spirit is respected: mental causes make a difference. The course of physical events in a purely physical world is different from the course of physical events in a world in which some physical events have mental causes. However, assuming dualism, it seems that there can be an exact physical duplicate of the actual world, but in this duplicate, there are no mental events that cause some of the physical events. This would be a zombie world so to speak. Hence, the envisaged service that (Super-)Humeanism can do to dualism in making mental causation intelligible without coming into conflict with science seems to end up in a dilemma: if there is the metaphysical possibility of an exact physical duplicate of the actual world in which there are no mental causes of some of the physical events, then mental causation is abandoned also in the actual world and the position runs into epiphenomenalism with respect to mental events; these then do not make a difference. To exclude this metaphysical possibility, however, it seems that one has to drop dualism. Indeed, this dilemma makes evident that there is only a tiny conceptual space for dualism open if one seeks to employ (Super-)Humeanism about laws and causation in order to make mental causation intelligible while respecting scientific realism. If the metaphysical possibility of a physical duplicate of the actual world without mental events causing some physical events is to be excluded, this implies that the fact that there are mental events – more precisely, the fact that there are persons or minds that act –, is fixed by and thus supervenes on the configuration of the physical events and its evolution. In all metaphysically possible worlds, whenever there is a physical course of events like in the actual world, there are minds that think and act coming into existence in any such world. In other words, minds come into being during the evolution of the configuration of matter of the universe. There are purely physical sufficient conditions (that is, metaphysically sufficient conditions) for physical systems to be organisms including human beings with minds and free will. This is the limitation that we have to impose on the dualism that is admissible in this context. Nonetheless, once persons have come into being during the evolution of the configuration of matter of the universe, persons are such that they are free what to think and what to do such that, given exactly the same physical conditions, they could have thought and done otherwise. Consider what Kant says in the
13
Super-Humeanism and Mental Causation
267
Prolegomena about the transition from sensory impressions caused by stimulations of the sense organs to thought: If an appearance is given to us, we are still completely free as to how we want to judge things from it.12
For present purposes, we can assume that there are sufficient physical conditions for persons to come into being and to receive certain sensory impressions through their sense organs. The point that Kant makes then is that sensory impressions cannot impose on us what to believe, that is, what beliefs or thoughts to form that have a conceptual content. Given sensory impressions, a person has to make up her mind how to judge things as well as what to do. Hence, freedom is not only free will, but concerns thought in the same way as action. Accordingly, Kant (1996, p. 139) regards the concept of freedom “as the keystone of the whole structure of a system of pure reason” in the preface to the Critique of practical reason. Why is this so? Consider what Sellars (1956) denounces as the “myth of the given”: the myth is the idea that something that is simply given to the mind – such as sensory impressions – has as such an epistemic status in being in the position to justify beliefs and actions. However, sense impressions, construed as the effects of interactions of a person with her environment, cannot, qua being the result of physical causal processes, justify anything. The transition from sensory impressions and biological needs to beliefs and actions is such that the latter ones, in contrast to the former, are subject to a justification: it makes no sense to ask a person to justify the sensory impressions and biological needs that she has; but it does make sense to ask for a justification of the beliefs and actions that she forms on the basis of her sensory impressions and biological needs. This implies that the sensory impressions and biological needs cannot as such be the justification of her beliefs and actions. To put it differently, freedom consists in that as regards whatever is given to her mind, the person has to and is free how to position herself with respect to that input. The input cannot impose a certain positioning upon her: in being subject to a justification, the person is free how to form her beliefs and actions based on what is given to her mind. That is why she could have thought and done otherwise. In forming beliefs and actions, a person has to make up her mind as to what to endorse as a reliable source of knowledge and guide for actions, thereby seeking for and hence being responsive to reasons. In other words, the person has to decide herself in deliberating about what is given to her which beliefs she should adopt and which actions she should do. In this way, free will is tied to reason, normativity and justification and thereby distinct from arbitrariness or chance events. According to Sellars (1956, § 5), the myth of the given is on a par with the naturalistic fallacy in ethics, that is, to derive propositions about what should be the case from propositions about what is the case. For instance, smoking being unhealthy does not imply that one should not smoke without adding the normative premise that one should not do what is unhealthy; that normative premise cannot be
12
Prolegomena § 13, note III; quoted from Kant (2002, p. 85).
268
M. Esfeld
justified solely by invoking facts about health or any other biological facts. Hence, although the propositions about the physical events and their evolution entail propositions to the effect that there are persons, they do not entail propositions about what these persons then determine as the content of their intentional states. In other words, the referents of the beliefs and theories that a person adopts cannot impose the acceptance of the beliefs and theories in question on her. Consequently, any claims about the physical facts or events entailing the beliefs and intentions of persons miss the point about persons having in the first place to make up their minds about what they should think and do – otherwise they would not be persons. In a nutshell, thus, there are sufficient physical conditions for persons coming into being, but not for what they then determine as the content of their beliefs and intentions. Following the linguistic turn in twentieth century philosophy and, notably, the argument of Wittgenstein against private language in the Philosophical Investigations (1953, §§ 138–242) and the presentation of this argument by Kripke (1982), it is generally admitted that concept formation requires social interactions. Something can be a person only in a community of persons in which normative attitudes are developed into an exchange of giving and asking for reasons through which conceptual content is determined that then enables each person to freely form her beliefs and intentions to act. As notably Davidson (1984, essays 9–12) works out, what the conceptual content is that thus is determined is accessible only from within participating in the social interactions that determine it and thereby contributing to further shape it out. In contrast to the physical facts, it is not accessible from a third person perspective. Hence, again, the conclusion is that there are sufficient physical conditions for beings to have the capacity to participate in social practices of mutual interpretation, giving and asking for reasons; Tomasello (2014), for instance, formulates a biological explanation of this capacity in terms of the enhancement of fitness that cooperation between humans provides. Nonetheless, the content that is determined in these practices is not accessible from the physical facts. Complete knowledge of the physical facts would not entail the propositions about what the content is that is determined in these social practices. In this way, then, persons can come into being within the evolution of the configuration of matter of the universe such that there are sufficient physical conditions for their existence, while they are still free. Their free thoughts and actions are distinct from any arbitrariness or chance events and yet cause a good deal of their behaviour.
13.4
Mental Causation and the Humean View of Causation
If one accepts something like the sketch in the preceding section as spelling out what is distinctive of persons as conceived in the manifest image, then the Super-Humean ontology of the physical realm shows how thoughts and volitions, thus conceived, can make a difference to our bodily motions without coming into conflict with scientific realism. The only restriction to enable this view is the commitment to persons coming into being during the evolution of the configuration of matter in the
13
Super-Humeanism and Mental Causation
269
universe in the sense that there are sufficient physical conditions for the existence of persons. But there are no sufficient physical conditions that determine the content of the intentional states of persons. Their freedom consists in them determining themselves what to think and what to do through their weighing of reasons. Consequently, admitting persons in that sense requires an ontological commitment to persons as a further primitive over and above the physical primitives, although persons depend on there being certain sufficient physical conditions for them to come into existence. However, one may object that this attempt to bring the view of ourselves as persons as conceived in the manifest image together with the scientific worldview through a minimalist ontology of science in the vein of (Super-)Humeanism is misguided from the outset, since the Humean conception of causation is counterintuitive and opposed to what is assumed about causation in the manifest image. On Humeanism, causation comes together with laws as supervening on the mosaic of particular events. Even if the dualist view sketched out here is accepted with the commitment to persons entering into the ontology as a further primitive over and above the physical primitives, the Humean view of causation still is endorsed according to which the issue of whether two events are related as cause and effect is not a local affair; it does not depend only on the two events in question. But this is not as counterintuitive as it may seem at first glance: whether or not physical conditions obtain that are sufficient for persons to come into being is a global and not a local affair. The existence of persons depends on certain conditions being fulfilled in the overall evolution of the configuration of matter of the universe. When these conditions obtain, also conditions obtain that are such that there are psychophysical laws like the ones mentioned in the preceding section and with them causal relations between mental and physical events.13 The crucial point to bear in mind in this context is that being a global affair by contrast to a local one does not mean that it depends on the entire Humean mosaic: in particular, whether or not such conditions obtain in the universe at a time t does not depend on what happens after t. Such conditions can obtain at a time t given the universal configuration of matter at t and its evolution up to t. Something that happens after t may annihilate persons, but nothing that happens after t can remove the fact that there were persons and mental causation in the universe at t. By the same token, nothing that happens after t can alter the fact that attractive motion was a stable pattern or regularity in the universe up to t. The point at issue thus is that on Humeanism, as pointed out by Weatherson (2007, pp. 531–532), salient regularities that obtain in particular space-time regions are sufficient to entail propositions about
13
I thus change the view argued for in Esfeld (2007, Sect. 2), but still think that the following conditional is correct: if one endorses a metaphysics of causation in terms of powers, then the theory of the identity of mental with physical causes is the only plausible option to vindicate mental causation under scientific realism. By way of consequence, if one abandons the identity theory, one has to go for a metaphysics of causation that is not committed to causes being powers that produce their effects. So I switch from the modus ponens argued for in my (2007) to the respective modus tollens.
270
M. Esfeld
what causes what in these regions, and nothing that happens outside these regions can invalidate these causal relationships. One may object that we experience the feeling of the power of our mind making our limbs move in certain ways. However, also such a feeling can be there only against the background of stable physical regularities such as gravitation. Against this background, the Humean can account for such a feeling in terms of stable regularities that link conscious mental events with bodily movements. There is no valid argument from experiences to the commitment to primitive modality without adding premises about general metaphysical considerations that are a subject of dispute. In any case, the argument for endorsing the view of ourselves as persons as conceived in the manifest image has nothing to do with claims about feelings or direct experiences. It is entirely based on thoughts and actions admitting of a justification and the fact that nothing that is given to the mind – no sensory impressions, no biological needs, etc. – is as such in the position to justify anything. The project of bringing the manifest and the scientific image together is doomed from the start if on the one hand claims about feelings, direct experiences, etc. are taken at face value as the basis for ontological commitments and on the other hand all the magnitudes that figure in scientific theories are taken at face value in the sense that they are received as corresponding directly to something in the physical world. The issue is to achieve an overall reflective equilibrium that brings the core claims about persons in the manifest image together with an ontology that is minimally sufficient to accommodate what science tells us about the world in a scientific realist spirit. The argument of this paper is that Super-Humeanism can make a significant contribution to achieving that goal. Acknowledgements I am grateful to the editors for the invitation to contribute to this volume and to Anna Marmodoro and an anonymous referee for helpful comments on the draft of this paper.
References Beebee, H., & Mele, A. R. (2002). Humean compatibilism. Mind, 111, 201–223. Bhogal, H., & Perry, Z. (2017). What the Humean should say about entanglement. Noûs, 51, 74–94. Callender, C. (2015). One world, one beable. Synthese, 192, 3153–3177. Darby, G. (2018). A minimalist Humeanism? Meta, 27, 433–437. Davidson, D. (1984). Inquiries into truth and interpretation. Oxford University Press. De Haan, D. (2022). The power to will freely: How to re-think about the problem of free will without laws of nature. In Powers, time and free will. Springer (this volume). Dowker, F., & Herbauts, I. (2005). The status of the wave function in dynamical collapse models. Foundations of Physics Letters, 18, 499–518. Esfeld, M. (2007). Mental causation and the metaphysics of causation. Erkenntnis, 67, 207–220. Esfeld, M. (2014). Quantum Humeanism. The Philosophical Quarterly, 64, 453–470. Esfeld, M. (2015). Bell’s theorem and the issue of determinism and indeterminism. Foundations of Physics, 45, 471–482. Esfeld, M. (2020a). Science and human freedom. Palgrave Macmillan.
13
Super-Humeanism and Mental Causation
271
Esfeld, M. (2020b). Super-Humeanism and free will. Synthese, 198, 6245–6258. Esfeld, M., & Deckert, D.-A. (2017). A minimalist ontology of the natural world. Routledge. Esfeld, M., Lazarovici, D., Lam, V., & Hubert, M. (2017). The physics and metaphysics of primitive stuff. British Journal for the Philosophy of Science, 68, 133–161. Gillet, C. (2016). Reduction and emergence in science and philosophy. Oxford University Press. Goldstein, S., Taylor, J., Tumulka, R., & Zanghì, N. (2005a). Are all particles real? Studies in History and Philosophy of Modern Physics, 36, 103–112. Goldstein, S., Taylor, J., Tumulka, R., & Zanghì, N. (2005b). Are all particles identical? Journal of Physics A, 38, 1567–1576. Hall, N. (2009). Humean reductionism about laws of nature. Unpublished manuscript. http:// philpapers.org/rec/HALHRA Hoefer, C. (2002). Freedom from the inside out. Royal Institute of Philosophy Supplement, 50, 201–222. Huttemann, A. (2022). The problem of radical freedom. In Powers, time and free will. Springer (this volume). Hüttemann, A., & Loew, C. (2019). Freier Wille und Naturgesetze – Überlegungen zum Konsequenzargument. In K. von Stoch, S. Wendel, M. Breul, & A. Langenfeld (Eds.), Streit um die Freiheit – philosophische und theologische Perspektiven (pp. 77–93). Mentis. Ismael, J. (2016). How physics makes us free. Oxford University Press. Kant, I. (1996). In M. J. Gregor (Ed.), The Cambridge edition of the works of Immanuel Kant. Vol. 4, Practical philosophy. Cambridge University Press. Kant, I. (2002). In H. Allison & P. Heath (Eds.), The Cambridge edition of the works of Immanuel Kant. Vol. 3, Theoretical philosophy after 1781. Cambridge University Press. Kim, J. (1998). Mind in a physical world. An essay on the mind-body problem and mental causation. MIT Press. Kripke, S. A. (1982). Wittgenstein on rules and private language. Blackwell. Kroedel, T. (2020). Mental causation. A counterfactual theory. Cambridge University Press. Lazarovici, D. (2018). Super-Humeanism: A starving ontology. Studies in History and Philosophy of Modern Physics, 64, 79–86. Lewis, D. (1986). Philosophical papers (Vol. 2). Oxford University Press. Lewis, D. (2009). Ramseyan humility. In D. Braddon-Mitchell & R. Nola (Eds.), Conceptual analysis and philosophical naturalism (pp. 203–222). MIT Press. Loewer, B. (2007). Laws and natural properties. Philosophical Topics, 35, 313–328. Loewer, B. (2012). Two accounts of law and time. Philosophical Studies, 160, 115–137. Loewer, B. (2020a, Forthcoming). The consequence argument meets the Mentaculus. In B. Loewer, B. Weslake, & E. Winsberg (Eds.), Time’s arrow and the probability structure of the world. Harvard University Press. Loewer, B. (2020b). The package deal account of laws and properties (PDA). Synthese. https://doi. org/10.1007/s11229-020-02765-2. Marmodoro, A. (2018). Atomism, holism and structuralism: Costs and benefits of a minimalist ontology of the world. Metascience, 27, 421–425. Matarese, V. (2020a). A challenge for super-Humeanism: The problem of immanent comparisons. Synthese, 197, 4001–4020. Matarese, V. (2020b, Forthcoming). Super-Humeanism and physics: A merry relationship? Synthese. https://doi.org/10.1007/s11229-020-02717-w. Miller, E. (2014). Quantum entanglement, Bohmian mechanics, and Humean supervenience. Australasian Journal of Philosophy, 92, 567–583. Perry, J. (2004). Comptabilist options. In J. K. Campbell, M. O’Rourke, & D. Shier (Eds.), Freedom and determinism (pp. 231–254). MIT Press. Sellars, W. (1956). Empiricism and the philosophy of mind. In H. Feigl & M. Scriven (Eds.), The foundations of science and the concepts of psychology and psychoanalysis (pp. 253–329). University of Minnesota Press.
272
M. Esfeld
Sellars, W. (1962). Philosophy and the scientific image of man. In R. Colodny (Ed.), Frontiers of science and philosophy (pp. 35–78). University of Pittsburgh Press. Simpson, W. M. R. (2019, Forthcoming). What’s the matter with super-Humeanism? British Journal for the Philosophy of Science. https://doi.org/10.1093/bjps/axz028. Simpson, W. M. R., & Horsley, S. (2022). Toppling the pyramids. Physics without physical state monism. In Powers, time and free will. Springer (this volume). Steward, H. (2022). Laws loosened. In Powers, time and free will. Springer (this volume). Swartz, N. (2003). The concept of physical law. https://www.sfu.ca/~swartz/physical-law/index. htm. (First edition Cambridge University Press, 1985). Thyssen, P., & Wenmackers, S. (2021). Degrees of freedom. Synthese, 198. 10207–10235. https:// doi.org/10.1007/s11229-020-02714-z. Tomasello, M. (2014). A natural history of human thinking. Harvard University Press. von Wachter, D. (2015). Miracles are not violations of the laws of nature because the laws do not entail regularities. European Journal for Philosophy of Religion, 7, 37–60. Weatherson, B. (2007). Humeans aren’t out of their minds. Noûs, 41, 529–535. Wilson, A. (2018). Super-Humeanism: Insufficiently naturalistic and insufficiently explanatory. Metascience, 27, 427–431. Wittgenstein, L. (1953). Philosophical investigations (G. E. M. Anscombe, Trans.). Blackwell.