(Article originally published in 2005 Journal of Consciousness Studies 12(1) 1-19, with commentaries at 20-75:
journal web page: http://www.imprint.co.uk/jcs.html)
In my experience, plain persons (here meaning persons who are neither philosophers or cognitive scientists) tend to accept something like a libertarian position on free will, namely that free will exists and is inconsistent with determinism. That position is widely debunked by philosophers and cognitive scientists. My view at present is that something like this plain person’s position is not only defensible but is likely to be closer to the truth than opposing views. To put this to the test, I have written a simple and straightforward outline of what I hope is a philosophically and scientifically respectable version of the plain person’s position on free will, and have offered it for demolition by those who say such a view is untenable.
My account of free will is a robust one, explicitly inconsistent with determinism and intended to support equally robust views of personal responsibility for conduct. I see three broad areas of difficulty for this account.
(1) The randomness problem: how can there be an intelligible and plausible alternative to determinism that is not mere randomness? Cf. Smart (1961).
(2) The moral luck problem: we are products of genes and environment, so how can the way we are at any time and therefore the way we act be other than due to things outside our control, that is, be other than just a matter of luck? Cf. Strawson (1986, 1998, 2002).
(3) The supernaturalism problem: science has given us a successful and comprehensive naturalistic account of how the world works, so is it not unreasonable to propose that human beings are somehow outside this account and outside the causal order apparently demonstrated by this account?
I address these difficulties in this article.
I will proceed by asserting and explaining nine propositions. I see the first five as basic requirements for any intelligible account of indeterministic free will; while the remainder are further explanations and elaborations of my own particular account of free will and responsibility. All these propositions have some relevance to each of the three problems I have identified; but propositions 3 to 5 are particularly relevant to the randomness problem, propositions 5 and 7 to 9 are particularly relevant to the moral luck problem, and propositions 5 and 6 are particularly relevant to the supernaturalism problem.
1. The alternatives requirement: there is a pre-choice state such that the way the world is and the laws of nature leave open at least two post-choice states.
This is a minimum requirement for indeterministic free will. There must be a time before an exercise of free will by the doing of an action (or the making of some other choice or decision) when the action (or result of the choice or decision) is not uniquely pre-determined by the way the world then is and the laws of nature. There must be alternatives available: they may be the alternatives of two actions such as pushing button A and pushing button B; or of doing something and not doing it; or of shaping an action in one way and shaping it in another way. I say that the alternatives may also be the alternatives of making one judgment as to what to believe and making a different judgment, because I say that free will is exercised in making decisions or appraisals of this kind as well as in doing actions.
This requirement is intelligible, and is not implausible having regard to what quantum mechanics (QM) tells us about the world. Of course, there is the difficulty that, according to QM, any indeterminism is mere randomness, and I will return to this. And there is also the difficulty that QM indeterminism may be at scales of mass, distance and time such that it cannot account for macroscopic alternatives like doing or not doing an act, or deciding a question one way or another.
This second difficulty does not affect intelligibility, and the degree to which it affects plausibility is a matter of controversy: I will not consider it in detail here. However, there are plausible suggestions as to how QM indeterminacy and indeterminism (perhaps together with chaos theory) could give rise to macroscopic alternatives: see Stapp (1998), Penrose (1994), Eccles (1994), and Jibu and Yasue (1995).
It is not part of this requirement that the alternatives be equally open, or that there be a sudden jump from a single pre-choice state to a single post-choice state. In some or even all cases, there could be a period of transition in which the likelihood of all but one possibility is progressively reduced to zero; and there may sometimes be extended processes of decision-making in which the likelihood of the various possibilities fluctuates significantly: see Hodgson (1999).
2. The consciousness requirement: the transition from the pre-choice state to a single post-choice state is a conscious process, involving the interdependent existence of a subject and contents of consciousness.
This associates the exercise of free will with consciousness, and adopts a view of consciousness as involving the interdependent existence of a self or subject and contents of consciousness (cf. Honderich 1987), with the subject taken as continuing as the same subject throughout the process of transition from the pre-choice state to the post-choice state. I will later argue that the subject in fact continues longer than this; but I do not contend, and my account does not require, that it be a ‘substance’ distinct from the brain processes that support it, much less an immortal soul. My account is rather a dual-aspect account of physical processes and conscious processes, with the subject being considered as the bearer or experiencer of the contents of consciousness.
The contents of consciousness may be generally described as experiences, but should not be considered as limited to passive contents: it is essential to an account of free will that subjects be considered as capable of being active, and this activity must presumably be reflected in the contents of consciousness. Again, this is intelligible and plausible: indeed, it is widely accepted that voluntary behaviour is active conscious behaviour.
However, some have argued that experiments by Benjamin Libet (Libet et al 1983) demonstrate that the real decisions are made pre-consciously, so that conscious free will can be no more than an illusion. In these experiments, participants were asked to press a button at any time they wished and to note the time of deciding to do so; and neural preparations for the action were recorded as occurring some tens of seconds before the time noted by the participants as the time of deciding to push the button. But it should be recognised that these experiments are applicable only to one kind of choice, namely that between doing and not doing an action; and even in relation to that kind of choice, show no more than that unconscious preparation is required before a person has immediately available the alternatives of consciously doing or not doing an action; and that is neither surprising nor inconsistent with conscious free will.
In specifying this requirement, I am not suggesting that all our motivation is conscious: plainly this is not the case. Nor am I suggesting that consciousness is other than a matter of degree: I am content to adopt the ‘dimmer-switch’ view of consciousness advocated by Susan Greenfield (1999).
And I do not think I am succumbing to the myth of the homunculus in the Cartesian theatre (Dennett 1991, Part II). Daniel Dennett convincingly refutes the idea that there is a central headquarters in the brain where consciousness occurs; but my proposition does suppose that there is. I accept that the physical processes that correlate with conscious mental processes occur over spatially extended regions of the brain, but this by no means precludes the existence or causal efficacy of conscious processes involving the interdependent existence of a subject and contents of consciousness. In so far as Dennett’s rejection of the Cartesian theatre suggests the contrary, this depends on an assumption that causation must be local, as required by classical relativistic physics, an assumption which has been decisively undermined by QM (see Hodgson 1996, 2002a).
3. The grasping requirement: in this conscious transition process, the subject grasps the availability of alternatives and knows-how to select one of them.
Again, this is a minimum requirement for free will. For example, if the choice between doing or refraining from doing an action is to be considered an exercise of free will, the subject must to some minimum extent grasp the possibility of either doing an action or not doing it, and must know-how to do the action and also know-how to refrain from doing it. This again is intelligible and plausible.
Once we have learnt to control our bodily movements, we are during consciousness generally aware that we can make various movements or not make them, and we know-how to make them or not make them. This requirement accords with the Libet experiments mentioned above, namely that there must in general be non-conscious preparation before the choice process starts: it is plausible that there could not be conscious grasping of available alternatives unless this has been made possible by some preparation that must be largely unconscious.
In previous writings I have suggested that free will can also be exercised in the shaping of bodily movements, as distinct from their initiation. For example, when a pianist performs a well-learnt piece of music, consciousness comes too late to direct fingers the right keys, but not too late to make choices in the shaping of musical passages. In such a case, I suggest, the pianist grasps the possibility of shaping the music in a particular way, and also is at least faintly aware of the possibility of shaping it in another way or else not consciously shaping it at all, and knows-how to select either alternative; and thereby can respond consciously, and I would say freely, to sounds heard and emotions felt. This is part of the reason for the intense concentration that musicians report to be a requirement for their performances.
In those cases where we are faced with a choice or decision to be made between two or more explicitly-presented alternatives, whether they be alternative actions or alternative beliefs or appraisals, this requirement will plainly be satisfied.
4. The reasons requirement: in significant exercises of free will, the subject experiences reasons on which a selection can be based, reasons that are non-conclusive and thus can influence but not dictate the selection.
I suggest that in significant choices we are consciously aware of experiences, thoughts (including thoughts in which we attend to beliefs), and/or feelings, that provide reasons, generally inconclusive and often conflicting, for one or more of the available alternatives. As mentioned earlier, I do not suggest that all our motivation is consciously experienced, much less that it is all consciously understood by us: plainly, much of our motivation is unconscious, and even the reasons of which we are conscious have a basis in extensive non-conscious processing. However, we do become consciously aware of feelings like pain or hunger, and of ‘somatic markers’ (to use a phrase from Damasio 1996) associated with different alternatives, and also of beliefs and experiences relevant to our choices; and it seems that we are motivated by these feelings, beliefs and experiences.
It is plain that such feelings and other reasons are of diverse kinds, generally not measurable, and generally incommensurable. There is, for example, no common scale on which hunger for food can be measured against a feeling of obligation to carry out a promised task. That is one reason why it is a mistake to suggest that we act according to the preponderance of our desires: desires are not like forces in Newtonian physics that are commensurable and so can be combined to produce a resultant force. In general, the reasons experienced by a subject and relevant to a decision to be made by the subject do not dictate a conclusion. As I put it in Hodgson (1999), the reasons do not include a clincher; and as John Searle put it in Searle (2001), there is a gap between the reasons and the conclusion. The only clincher is the choice itself.
I cannot exclude the possibility that a choice between apparently incommensurable reasons is in fact wholly determined by unconscious processes, which are identical with physical processes that are in turn determined by measurable and commensurable physical properties and laws of nature; but the existence of that possibility does not justify disregard of this alternative account I am giving, or show that it is not intelligible and plausible.
5. The selection requirement: the subject makes an effective non-random selection between the available alternatives, based on these non-conclusive reasons, albeit not determined by rules or laws of nature.
This is a vital proposition, one that is necessary to overcome the alleged dichotomy of determinism and randomness. It is a proposition which I’ve been advocating since 1991, following ideas of Nozick (1981) and Putnam (1983) (see for example Hodgson 1991 Ch. 5, and particularly Hodgson 1999), but which is still generally overlooked. If it is true, it is of enormous significance, inter alia in that it would show how different human beings are from computing machines as presently understood, no matter how powerful such machines may be or become. The contrary position, that what is not wholly determined by initial conditions plus laws or rules must be random, is widely assumed but rarely examined. It is considered with some care in Strawson (1986), but my opposing arguments remain unanswered.
It is convenient to consider first those exercises of free will involved in deciding between competing hypotheses or appraisals on the basis of inconclusive evidence. What is often overlooked is that, apart from rules of reasoning such as those of mathematics, logic, and probability theory, there are no known rules (that is, strict rules as distinct from non-conclusive heuristics) governing good plausible reasoning. Of course, it is possible that plausible reasoning proceeds in accordance with evolution-selected computation-like procedures that we do not understand, and undoubtedly this is part of the story (I would say, that part concerning the determination of alternatives, reasons and tendencies); but there are powerful arguments for thinking that it is not the whole story, and that there is an element of judgment in plausible reasoning that is not accounted for by strict rules of any kind. These arguments include the following:
(1) If choices were in fact determined by algorithms, such as evolution-selected computation-like procedures, which as algorithms need no help from conscious judgment and could indeed be hindered by conscious interference, there could be no plausible explanation of why evolution selected in favour of brains that, at considerable expense in terms of complexity and energy-use, support conscious processes.
(2) In particular, there could in that event be no plausible explanation (a) of why we have feelings like pain to motivate us, when it would be absurd (even if possible) to use pain or any other feelings to motivate a computer to proceed in accordance with its program; or (b) of why are we so constituted that our conscious awareness is automatically called into play when we are faced with a novel situation calling for decisive action.
(3) Our rationality is well adapted to dealing with problems remote from the evolutionary tests that faced our evolutionary ancestors, and this makes it unlikely that it is no more than a matter of useful algorithmic processes selected through those tests: see Nagle (1986, p. 79).
(4) If we cannot rely on our plausible reasoning as the conscious non-algorithmic process that we instinctively take it to be, then any confidence that we could have in it would have to depend on the circumstance that it comprises computation-like processes whose reliability is assured by the evolutionary tests they have passed; yet any belief in this circumstance and accordingly any justified confidence would itself depend on extensive plausible reasoning, giving rise to a vicious circle: cf. Plantinga (1993, Ch. 12), Nagel (1997, Ch. 7).
(5) When we are conscious, our brain processes give rise to qualia (experiences or potential experiences) of various types and chunk them into unique particular global experiences of particular subjects: these are what I have called two tricks of consciousness, the qualia trick and the chunking trick (Hodgson 2002b). If a particular subject/experience combination produced by these tricks is to have a causal role in what happens, otherwise than through its general properties whose existence does not require this combination of tricks, then, because of the uniqueness and particularity of the subject/experience, this role cannot be one determined by generally applicable rules or laws of nature. Yet it seems clear that a particular gestalt or global experience, for example an experience of a unique and unprecedented work of art like Picasso’s Les Demoiselles d’Avignon by the artist when he created it or an early appraiser of the work, does have a role in aesthetic judgments, a role that is part of rational appraisal yet which cannot be rule-governed because there can be no general rules that engage with a particular subject’s particular global experience of a unique unprecedented object (see Hodgson 2001, 2002b).1
This fifth argument applies to judgments as to what to believe as well as to aesthetic appraisals. In deciding what to believe on the basis of uncertain evidence, we seem to take into account our assessments of whole particular gestalt experiences; yet these experiences cannot, as unique and particular wholes, engage with general rules. For example, in deciding whether an experience is an accurate experience of some aspect of the world or is some kind of illusion or otherwise inaccurate, we take into account the particular global experience, and have regard to its clarity, immediacy, vividness, internal coherence, coherence with accepted beliefs, relevant similarities to experiences accepted as real, and so on. In such cases, I suggest, the outcome is not completely determined by the pre-choice state plus rules or laws of nature, but by a process that depends in part on a particular subject’s non-algorithmic response to a whole particular experience. It is not reasonable to think that the outcome in such cases is either merely random or the result of some unique constraint that engages with the unique pre-choice state and no other: it is more plausible to think that it is the result of an indeterministic but non-random selection.
Roger Penrose (1994) has argued strongly for the view that human intelligence involves understanding of a kind that computers lack. My contention is that our access to and ability to use particular global experiences in a way that rules cannot determine is an important part of what is required for this understanding.
Similar arguments apply also to decisions about what to do, as well as about what to believe. What I suggest is that the ability of a conscious subject to take into account whole particular gestalt experiences, and to act upon judgments based on inconclusive reasons, has been selected by evolution just because it is more conducive to actions favourable to survival and reproduction than are purely algorithmic processes.
All this is confirmed by the powerful and ineradicable feeling we have that we are consciously making choices and making things happen by doing them. Suggestions, such as that by Crick (1994, p. 266), to the effect that we have this feeling just because we are not aware of the unconscious processes that are actually efficacious, provide no reason why any feeling of choosing or doing would be involved if this were the case. It would be as if I simply became aware of a thought and a movement of my body in accordance with that thought, without actually making a choice or doing an action. If this happened, surely I would not have any feeling of choosing or doing: rather, I would be puzzled by such an occurrence, in the same sort of way that the talking hemisphere of a split brain patient is puzzled by bodily movements initiated by the non-talking hemisphere
This fifth proposition is probably the most crucial and difficult of my nine propositions. It is difficult, because we are so accustomed to looking for reductionist explanations, in particular explanations in terms of law-governed processes, and because in all fields apart from those involving questions about consciousness, reductionist explanations have been spectacularly successful. But while acceptance of this fifth proposition may go against the grain, I would argue that nothing is more familiar to us than our non-algorithmic plausible reasoning, which we can understand, to some extent at least, without making the reductionist assumption that it is no more than a small part of a wholly algorithmic iceberg that somehow gives rise to the misleading illusion that the processes of its conscious tip are rational but non-algorithmic.
Indeed, unless and until there were to be some other explanation of why we have conscious experiences and of what is their causal role, and also a satisfactory account of plausible reasoning in terms of algorithms, an account that altogether dispenses with judgments based on feelings and particular gestalts, I believe acceptance of this fifth proposition is more reasonable than its rejection.
Before leaving this proposition, I should note one significant attempt to give an evolutionary explanation of conscious experiences and their causal role, notwithstanding an assumption that they are constituted by and/or wholly depend upon computational algorithms carried out by our brains (see for example Dennett 2003). This is to the effect that there have been evolutionary advantages for human beings and their close evolutionary ancestors in being able to monitor and communicate some of their own mental processes. Because these processes are too complex at the level of the computational algorithms (or any more basic level) to be grasped for monitoring or communication, evolutionary selection has developed brains able to produce simplified ‘user-friendly’ accounts of these processes, in terms of the existence of an integrated conscious subject or self that has conscious experiences, has goals and purposes, and chooses between available alternatives. These accounts are not exactly false, on this approach, because at the level at which our brains can grasp their own processes for monitoring and communication, they give about the best available approximation to the truth; but they are not exactly true either, in that they tend to suggest that the brain’s processes are other than the working out of computational algorithms, and to that extent they are false or at least misleading.
To anyone who strongly adheres to the view that our mental processes must be algorithmic, this approach may seem attractive. However, I think it has far too high a cost. A person’s agonising pain would be treated as an account of the person’s complex of dispositions to act in certain ways, produced by the person’s brain and thereby enabling the person to monitor and communicate relevant brain processes: it would have no other causal role, and in particular no efficacy in contributing to conduct by virtue of its subjective feel. Such a pain would be ‘felt’ no less by a Turing machine carrying out the same algorithms and thus producing the same account; whereas the brains of animals that do not monitor and/or communicate their mental processes would not produce such accounts, so that presumably those animals would not ‘feel’ pain even in this sense. I think pain has a reality, both as a feeling and as a motivator, that this approach denies; and I strongly disagree with the view (going back to Descartes) that no non-human animals feel any pain.
6. Naturalism: there is nothing supernatural and no violation of physical law involved in such selections.
A standard complaint about libertarianism is that it introduces a supernatural element in order to account for ‘contra-causal freedom’, and that it involves violation of physical law. However, as mentioned earlier, QM and chaos theory make it possible for there to be macroscopic alternatives for selection, and QM shows that causation can operate non-locally so that spatially extended conscious processes could be globally efficacious.
Even when these points are accepted, it is still argued that, unless in every case one of the alternatives that are possible according to QM occurs at random within the probability parameters established by the laws of QM, then physical law would be violated, in the sense that the statistical predictions of QM would be falsified. However, it is reasonable to think that the felt strength of reasons has some relationship to QM probabilities, so that selections are likely to approximate to QM statistics; and having regard to the uniqueness and complexity of pre-choice states, a demonstrated violation of QM statistical predictions is unlikely in the extreme (see Hodgson 1999).
More importantly, QM has not yet been applied to conscious systems, and it is an open question how its statistical predictions apply to such systems. The qualia trick and the chunking trick give rise to particular experiences of particular subjects; and if these whole particular experiences of particular subjects have an irreducible causal role in what happens, then, because of the uniqueness and particularity of the experiences and subjects, that causal role cannot be fully accounted for by any system of physical laws of general application, even those of QM. Indeed, my suggestion is that the capacity to respond to particular gestalts, to which rules cannot apply, has been selected by evolution just because this capacity is conducive to satisfactory choices and in that sense makes satisfactory choices more likely than they would be if choices occurred at random in accordance with QM statistics; so that, if it were possible to calculate them, the statistics of free choices would not be the same as if free will did not exist. This would not be a violation of physical law, but a limitation on the law’s applicability.
And this is not an appeal to the supernatural, but a recognition that the natural is not as narrow and limited as it is sometimes supposed to be; in particular, that it is not limited to the mechanistic development of systems in accordance with physical laws and randomness. I have previously argued (Hodgson 2001) that laws of nature may be of different kinds, in particular that there may, in addition to laws that constrain outcomes (C-laws), be laws that empower systems to direct or select outcomes (E-laws) and laws that guide systems in such selections (G-laws). I there suggested that E-laws provide systems such as human beings with both the capacity to make selections and the reasons on the basis of which selections are made; and that G-laws include moral principles that affect such selections.
On that approach, exercise of a capacity to select is not contra-causal, but in accordance with a wider concept of causation. And I do not suggest that the capacity to select appeared suddenly in human beings. Rather, I suggest it emerged very gradually in evolution along with the gradual emergence of consciousness. I believe that even primitive consciousness involves qualia and chunking, and may also involve the capacity to select. In primitive conscious systems, this might be exercised in such things as selecting in particular circumstances between getting food and avoiding predators. But I would not regard such capacity to select as amounting to free will unless and until combined with the self-conscious rationality of human beings.
7. Capacity to select: although differences between persons affect alternatives, reasons and tendencies, they do not otherwise affect capacity to select, which is the same for all persons.
It follows from the above that a person exercising free will does so subject to considerable pre-choice limitations. The person has no alternatives apart from those made available by the pre-choice state and grasped by the person in the process of selection. The person has no reasons apart from those presented by the pre-choice state, generally based on non-conscious processes; and those same processes largely determine how the reasons feel and appeal to the person, and also give rise to tendencies to act in various ways. Another aspect of the pre-choice state, for which the person can have no responsibility, is that the person has the capacity to make a selection between the alternatives on the basis of the reasons. But I suggest that nothing in the pre-choice state pre-determines the result of exercise of that capacity.
The totality of ‘the way a person is’, prior to the selection being made, is inconclusive as between the available alternatives: it gives rise to reasons and tendencies to act in one or other of the ways that are open, but does not pre-determine the outcome. Thus the person’s selection is influenced by the reasons and the tendencies, but not pre-determined by them. Indeed, it is reasonable to think that ‘the way a person is’, prior to the selection, does not affect the selection otherwise than through providing the alternatives, the reasons, the tendencies, and through the existence and exercise of the capacity to select. Different persons have different characters, and act differently because of these different characters. However, I am suggesting that this is because of the differences that pre-choice states make to alternatives, reasons, and tendencies, not because of any differences in the persons’ capacity to select. In relation this capacity, each person is entirely the same, unaffected by differences in pre-choice states, whether due to genes, environment, prior selections, or all three; and in relation to its exercise, to the extent that each person can notionally be considered apart from differences affecting alternatives, reasons and tendencies, each person is entirely the same.
Thomas Clark (1999, p. 286) has suggested that this approach makes the choosing subject an abstract entity devoid of character and motives. The reverse is the truth. The subject is the unique totality of all its properties, and it is precisely because this unique totality together with particular experiences enters into the causal process that outcomes are not predetermined by constituent properties which it may share in varying degrees with other entities and with which general laws can engage.
We can’t help having capacity to select, and nothing we can do at the time of selection can make us responsible for our particular characteristics that affect alternatives, reasons, and tendencies; but our particular characteristics do not otherwise affect the way we exercise our capacity to select. We do this by choosing which alternative occurs, thus providing the clincher; and there is an element (by which I do not mean a distinct or severable element) of this process that is entirely up to us, unaffected by any differences between different persons. And as asserted by the fifth proposition, this does not mean that the selection, or any part of it, is random or otherwise not rational.
8. Moral principles: to greater or lesser extents, persons grasp moral requirements that should guide selections.
It could be argued that, even if persons have the capacity to make selections, that does not make it fair to treat them well or badly because of selections they make, because this would pre-suppose some objectively valid and binding standards of behaviour, and an ability in persons generally to know these standards. But I suggest it is intelligible and plausible to say that there are such standards and that persons can to greater and lesser extents grasp them. I mentioned earlier my proposal in Hodgson (2001) for a classification of laws of nature so as to include G-laws, which guide systems in making selections from alternatives open to them.
As suggested there, the grasping of G-laws could begin with the emergence in evolution of conscious systems, having some marginal capacity and reason to select and bring about one future state of itself, among those left open by the C-laws, in circumstances where there was fuzziness or conflict in the disposition or motivation of such a system as to what state should occur. For example, suppose that such a system felt, in a primitive way, something like what we would feel as motivation to minimise the pain of an injury, and also something like what we would feel as motivation to satisfy hunger; and that it felt it could follow one feeling or the other, but that following one would preclude following the other (getting the food would exacerbate the pain). The system, having these conflicting feelings, and feeling itself motivated by them towards differing future states of itself, both of which were open to selection by it, could I suggest also feel something like a requirement to resolve them ‘rightly’, and to bring about one state of itself (that is, to act) in accordance with that resolution.
This suggests the most basic G-law, which would to some degree guide and be felt by even such a primitive conscious system, a law which I call ‘act rightly’:
Act so as ‘rightly’ to resolve fuzziness or conflicts of motivation.
I use ‘rightly’ at this stage without any moral implications, so that the law here simply means, do whichever of the conflicting possibilities is apt or fitting or appropriate or ‘to be done’. I say this law would be felt, because its guidance would, to some extent at least, take effect through its influence being felt and acted upon by the system itself. In more complex conscious systems, the basic G-law could come to separate out into two distinct aspects or sub-laws, which I call ‘decide rightly’ and ‘carry out’:
Decide what act would rightly resolve fuzziness or conflicts of motivation; and
Carry out that decision.
And in such systems, selections could be assisted by further G-laws and/or by principles associated with them. In particular, these systems could feel and apply a G-law, which I call ‘find out’:
Optimise the reasons (including information and feelings) on the basis of which to act.
In moderately primitive conscious systems, this could be felt as requiring attention to relevant information-and-feelings, as delivered by the senses and emotions. In more elaborate conscious systems, it could be felt as requiring such things as exploration of relevant information-and-feelings, verification by checking, looking for coherence and consistency, attending to analogies, and seeking an understanding of issues facing the system.
In conscious systems without the self-conscious rationality of human beings, the application of G-laws would not be a matter of morality, although analogies with human moral issues could be drawn. For example, some conflicts of motivation could be analogous to human moral conflicts, such as a conflict between an animal’s motivation to minimise its own pain and its motivation to protect its offspring; and some actions by animals may display ‘virtues’ of courage and determination in carrying out decisions as to what act would be ‘right’. However, in systems with self-conscious rationality, G-laws could have central moral significance. The basic G-law ‘act rightly’, as it applies to these systems, could be the fundamental moral prescription. Moral laws such as ‘do no harm’, ‘be fair’, ‘be honest’, ‘fulfil commitments’, and ‘do good’ could be further G-laws felt by these systems. Plainly, these further moral laws can conflict with each other, and they can also conflict with a system’s basic motivating feelings. In such cases, the basic G-law would require the conscious systems to resolve conflicts rightly, having regard to all relevant G-laws.
Another G-law that a rational self-conscious system could feel is the following moral law, which I call ‘improve oneself’:
Enhance one’s own ability to find out, decide rightly, carry out, and do good.
This law would require the cultivation of virtues associated with the seeking of truth, particularly in so far as the truth was relevant to one’s own actions; and of virtues associated with readiness, willingness, and ability to put decisions rightly made into effect, and also to enlarge one’s opportunities to do good.
Thus, in systems with self-conscious rationality, the G-laws, which I conjecture are laws of nature that are to some degree felt by all conscious systems, could come to be felt as being or including a system of guiding moral laws, which are truly existing features of the universe, ascertainable, and to be respected by us whether we like it or not. Although such a view is very unfashionable today in some circles, some such view is required if, for example, an opinion that it is wrong to torture children for amusement is to be considered a matter of truth rather than merely something we have been programmed to believe by evolution and education. And there is no greater problem with saying that the truth of such moral rules can be supported by plausible reasoning than, for example, with saying that the truth of factual inductive conclusions can be supported by plausible reasoning. There could be disagreement as to what it is that G-laws require in particular circumstances, and as to what, among rules which people claim to be moral laws, are G-laws or rightly derived from G-laws and what are merely fallible human inventions; although I don’t think there would be much room for disagreement about the G-laws I have identified, most or all of which are I suggest to some extent grasped by all persons of reasonable mental capacity and sanity.
9. Ultimate responsibility: accordingly, there is some ultimate responsibility for selections, and thereby for subsequent pre-choice states.
On this account, then, persons generally have alternatives open to them in their conscious behaviour, and feel reasons for selecting among these alternatives, including reasons associated with their grasp of moral principles such as those discussed above. ‘The way a person is’ provides alternatives, reasons (including the grasp of moral principles), tendencies, and capacity to select, but does not otherwise influence the selection. Thus I suggest that, in making selections, persons do have some ultimate responsibility, with degrees of responsibility affected by how hard it is, by reason of the pre-choice state, to make the right selection. And I do accept that these degrees of responsibility may vary widely, so that, for example, environmental disadvantages such as abuse in childhood may enormously reduce responsibility and blameworthiness for later conduct.
I earlier suggested that subjects continue throughout processes of selection; and now I suggest that it is reasonable to see this continuance as indicative of a more extensive continuance, throughout longer periods of deliberation on problems, and indeed throughout a whole integrated life-history that can be regarded as a progressive and continuous addressing of life’s challenges (Hodgson 1999, 2001). A conscious system comes into existence at or prior to the birth of a human being and continues as a system with the same subject, at least until this continuance is interrupted or terminated by significant brain injury or mental illness, or death. Thus there can be justice in treating a person differently according to what that person has done in the past, not just because the person has some ultimate responsibility for what was done, but also because the person is in a substantial sense the same person both at the time of the action and at the time of the subsequent treatment.
Furthermore, since prior selections, for which a subject has some ultimate responsibility, can in turn affect later pre-choice states of the same subject, the subject has some ultimate responsibility also for those later pre-choice states and thereby additional responsibility for what is done at later times.
II THREE QUESTIONS
I will conclude this essay by considering three questions about the account of free will given in my nine propositions.
1. Does this account involve agent causation?
One prominent version of indeterministic free will embraces a distinction between what is called event causation and what is called agent causation. The former is causation by events or happenings in the world, which is the causation dealt with by the physical sciences; and the latter is causation by agents rather than events, which is the causation supposed to be involved in exercises of free will.
In one sense my account of free will involves causation by agents, in that I suggest that a subject or agent persists throughout a process of selection (indeed, generally throughout a life) and actually makes the selection on the basis of inconclusive reasons: as I have put it, in selection the agent provides the clincher that finally determines which alternative out of those open actually occurs. However, I do not draw a sharp distinction between causation by events and causation by agents.
For one thing, I do not say that causal processes not involving subjects or agents must be analysed in terms of events: analysis in terms of things or processes or states of affairs may be equally or more appropriate for certain purposes. I do not think it is helpful to debate whether the world really is made up of things or events or processes or states of affairs, or whether causation is really causation by things or events or processes or states of affairs. There are things and events and processes and states of affairs, and consideration of questions of causation may involve any one or more of these categories: none I suggest needs to be considered as being more basic than the others.
More importantly, I suggest that events do have an important role in exercises of free will. Events have a role at least in limiting available alternatives and in providing reasons and tendencies; and the selection itself can be regarded as an event. The causation of the selection might be considered as being partly by other events and partly by the activity of the conscious subject or agent exercising its capacity to select; but even that analysis could be misleading, in that I regard the selection process as a global process, not divisible into distinct parts. So rather than distinguishing causation by events and causation by agents, I prefer to distinguish physical causation, which is that aspect of causation capable of being fully understood in terms of the operation of laws of nature and randomness, and volitional causation, in which the conscious activity of a subject or agent makes a contribution that can’t be fully understood in that way.2
On my account, then, the nature and degree of responsibility of an agent for the agent’s conduct can be a matter for rational consideration; whereas on the standard ‘agent causation’ account, causation by agents seems to be mysterious, absolute, and incapable of further explanation.
2. Does this account deal with the randomness problem and the moral luck problem?
I mentioned earlier the randomness problem, the problem of making intelligible an alternative to determinism that is not mere randomness, and the problem of giving an intelligible answer to the moral luck argument.
The substantial answer to the randomness problem is that given in the discussion of the fifth proposition: but to see if the problem is really answered, it is useful to look at two elaborations of the problem given by Peter van Inwagen (2002).
In the first, van Inwagen supposes that God repeats many times a person’s pre-choice state, and that (as required by libertarianism) on some occasions the decision goes one way and on others it goes another way. He supposes that, as the repetitions continue, the statistics of the choices appear consistent with a certain probability for each decision. He suggests that free choice is thus indistinguishable from random occurrences within probability parameters.
But as stated above, on my account of free will the statistics of many choices would not necessarily be the same as the statistics suggested by the laws of QM as applied to the pre-choice state that is repeated in this way, because the QM statistics would not take into account the subject’s particular gestalt experiences that are part of the pre-choice state. The subject’s selected response to the particular experiences is, for evolutionary reasons, more likely to be conducive to survival and reproduction than random occurrences within QM probability parameters,
In the second, van Inwagen supposes a choice between upholding public morality and betraying a friend on the one hand, and keeping silent on the other hand, in which the pre-choice state gives only just over 0.5 probability of keeping silent. Van Inwagen asks whether a person, knowing this probability, could in good conscience promise the friend to keep silent, when there is over 0.4 probability that the promise will not be kept.
On my account of free will, the probability given by the pre-choice state is at best a QM probability that does not take into account all relevant factors, including the particular gestalts of the pre-choice state; and in any event, the person will be able to freely choose what to do when the time comes. Furthermore, plainly the making of the promise will affect the pre-choice state, presumably making it more likely that what is promised will be done. But finally, if a person concluded that, even if the promise were made, he or she could not be confident of keeping the promise, that would be a strong reason for saying that the promise could not in good conscience be made. In this respect, libertarianism is in no worse case than competing views.
Turning to the moral luck argument, Galen Strawson elaborates on this argument in Strawson (2002), building on two premisses:
(1) We act as we do because of the way we are.
(2) We cannot be responsible (in the sense of ultimately responsible, the buck stopping with us) for the way we act unless we are responsible for the way we are.
And he goes on to argue that we cannot be responsible for the way we are when we first make decisions in life, so we cannot be responsible for actions based on those decisions, or for how those decisions and actions affect the way we are later on; and so on. Thus, we can never become responsible for the way we are later in life, or responsible for the way we act later in life.
He suggests that there can be no argument with his first premiss, and, relying partly on the first premiss, argues powerfully in support of the second premiss.
What I say to this is that it is necessary to bring out an ambiguity in the word ‘because’ in first premiss. It could mean that the way we are plus our circumstances plus the laws of nature pre-determine the way we act in those circumstances; and if it means that and is true, then it is hard to argue with the second premiss. However, on my account of free will, ‘because’ in the first premiss means that the way we are plus our circumstances plus laws of nature provide alternatives, inconclusive reasons, and tendencies, and also the capacity to select between the alternatives on the basis of the reasons; and what we do is what we select in exercise of that capacity, the selection not being influenced by any differentiating features of the way we are otherwise than through the alternatives, reasons, and tendencies. If the first premiss is true on that interpretation of ‘because’, the second premiss is untrue; and the Strawson argument collapses.3
3. Is this account believable?
As noted earlier, there is a question mark over the first proposition in terms of the availability of macroscopic alternatives. The fifth proposition is difficult, but in my contention difficult mainly because of ingrained habits of thought; and for reasons I have given I believe it is reasonable to accept it at the present time. The eighth proposition is highly unfashionable; but I think that at least some moral opinions are matters of fact and truth rather than mere evolutionary artefacts, and thus that it is reasonable to accept my eighth proposition.
Reductionist science has had enormous success in accounting for many aspects of the universe, but very little success in explaining consciousness and its role in the way events unfold in the world. I think it is reasonable to believe that consciousness does have an important and irreducible causal role, and I suggest that something like this version of free will is required to account for this role.
So all in all, I say my account is believable, indeed at present more so than any alternative.
1. This argument, introduced in the first of these two articles and developed in the second of them, is I think an original and just possibly even an important argument. It is one of very few attempts to give a positive and non-mysterious account of why and how conscious processes can contribute to rational decision-making in a way not available to law-governed machines. I would refer readers to those articles, particularly the second of them, for a full exposition of this argument.
2. My view may be closer to the standard agent-causation view than is that of Robert Kane (1996), and it may be that Kane does not accept my fifth proposition. I think the criticism of Kane by Dennett (2003, Ch. 4) in fact, rightly or wrongly, assumes that Kane does not accept this proposition.
3. In his recent impressive exposition of a compatibilist view of free will and responsibility, Dennett (2003) does not refer to Galen Strawson or the moral luck argument. He does set out a related argument, along the following lines: what we do is wholly determined by events in the distant past and laws of nature; those events and laws are not ‘up to us’; therefore what we do is not up to us. He claims (pp. 126-36) that this argument commits the same fallacy as an argument that there couldn’t be mammals: if there have been any mammals, there have been only a finite number of them; every mammal has a mammal for a mother; therefore if there have been any mammals, there have been an infinite number of them. However, the fallacy in this argument is that there is an indeterminate boundary between mammal-like reptiles and reptile-like mammals; and while it may be said that there is a similar indeterminate boundary between actions in our early lives that are in no sense ‘up to us’ and actions in our later lives that are in some sense up to us, this does not deal with Strawson’s premisses (1) and (2) and thus does not answer Strawson’s moral luck argument. I suggest that something like my propositions 5 and 7 to 9 is required to answer that argument.
Clark, T. (1999), ‘Fear of mechanism’: in Libet et al (1999).
Crick, F. (1994), The Astonishing Hypothesis (London: Simon & Schuster).
Damasio, A. (1996), Descartes’ Error (London: Macmillan).
Dennett, D. (1991), Consciousness Explained (New York: Allen Lane).
Dennett, D. (2003), Freedom Evolves (London: Allen Lane).
Eccles, John (1994), How the Self Controls its Brain (Berlin: Springer).
Greenfield, Susan (1999), ‘How might the brain generate consciousness?’: in Rose (ed), From Brains to Consciousness (Penguin: London).
Hodgson, D. (1991), The Mind Matters (Oxford: Oxford University Press).
Hodgson, D. (1996), ‘Nonlocality, local indeterminism, and consciousness’, Ratio, 9, 1-22.
Hodgson, D. (1999), ‘Hume’s mistake’: in Libet et al (1999).
Hodgson, D. (2001), ‘Constraint, empowerment, and guidance: a conjectural classification of laws of nature’, Philosophy, 76, 341-70.
Hodgson, D. (2002a), ‘Quantum physics, consciousness, and free will’: in Kane (2002).
Hodgson, D. (2002b), ‘Three tricks of consciousness’, Journal of Consciousness Studies, 9, 65-88.
Honderich, T. (1987), ‘Mind, brain and self-conscious mind’, in Blakemore, C. and Greenfield, S. (eds), Mindwaves (Oxford: Blackwell).
Jibu, M. and Yasue, K. (1995), Quantum Brain Dynamics and Consciousness: an Introduction (Amsterdam: Benjamin).
Kane, R. (1996), The Significance of Free Will (New York: Oxford University Press).
Kane, R. (ed.) (2002), Oxford Handbook of Free Will (New York: Oxford University Press).
Libet, B., Gleason, C., Wright, W., and Pearl, D. (1983), ‘Time of conscious intention to act in relation to onset of cerebral activities (readiness potential): the unconscious initiation of a freely voluntary act’, Brain, 106, 623-42.
Libet, B., Freeman, A., and Sutherland, K. (eds) (1999), The Volitional Brain (Thorverton: Imprint Academic).
Nagel, T. (1986), The View From Nowhere (New York: Oxford University Press).
Nagel, T. (1997), The Last Word (New York: Oxford University Press).
Nozick, R. (1981), Philosophical Explanations (New York: Oxford University Press).
Penrose, Roger (1994), Shadows of the Mind (Oxford: Oxford University Press).
Plantinga, A. (1993), Warrant and Proper Function (New York: Oxford University Press).
Putnam, Hilary (1983), Reason, Truth, and History (Cambridge: Cambridge University Press).
Searle, John R. (2001), Rationality in Action (Cambridge MA: MIT Press).
Smart, J. J. C. (1961), ‘Free-will, praise and blame’, Mind, 70, 483-94.
Stapp, Henry (1998), ‘Pragmatic approach to consciousness’: in Pribram, K. H. Brain and Values (Erlbaum).
Strawson, G. (1986), Freedom and Belief (Oxford: Oxford University Press).
Strawson, G. (1998), ‘Luck swallows everything’, Times Literary Supplement, 26 June, 8-10.
Strawson, G. (2002), ‘The bounds of freedom’: in Kane (2002).
Van Inwagen, P. (2002), ‘Free will remains a mystery’: in Kane (2002).