Does Neuroscience Threaten Freedom and Dignity?

Analysis and comment on "Neuroscience and the Human Spirit: Meeting the Challenges of Contemporary Brain Research", a 1998 conference hosted by the Ethics and Public Policy Center.

From the original conference description:

  • Free will, autonomy, and personal accountability: core values of western civilization.
  • Love, hope, faith, altruism: unique elements of the human spirit.
  • Advances in neuroscience point to significant biological, chemical, and genetic influences on human behavior. Does the steady advance of the brain sciences threaten the human spirit or western, Judeo-Christian views of it? Can science and human values be reconciled?

Introduction

In 1998, the Ethics and Public Policy Center (EPPC, www.eppc.org) in Washington D.C. hosted a two day conference, "Neuroscience and the Human Spirit," in which a number of scientists, philosophers, doctors, psychologists, journalists, and others met to discuss the implications of neuroscience for free will, morality, law, and public policy. What follows is an analysis of some of the central concerns expressed by those attending and their responses to those concerns. My perspective on the issues raised by the conference is naturalistic, and in what follows I hope to supply a critique that was mostly, but not entirely missing at the conference. I will only address a few issues here, for reasons of space and so as not to unduly tax the reader’s patience. I am grateful to Dr. Frederick Goodwin’s office for providing me transcripts of many of the sessions, and I’m grateful to the EPPC for hosting the event in the first place, since open, in-depth public dialog on these matters is long overdue. I hope the EPPC will keep the dialog going. The conference included:

Opening Remarks: Elliot Abrams, Dr. Frederick K. Goodwin

Keynote Address: "Defining the Human Spirit: Historical Perspectives and Contemporary Challenges" – William Bennett

Tutorials: 1) Behavioral Neuroscience, 2) Neuropharmacology, 3) Behavioral Genetics

Panels: 1) Ethics, Philosophy, and Religion, 2) Medicine and Science, 3) Law and Public Policy, 4) Media and Public Attitudes

Roundtable Discussion: "Reconciliation?"

I will boldface the names of conference participants, not all of whom get mentioned below. Not getting a mention is no reflection on particular contributions, rather just a matter of what themes I ended up addressing in these comments. Much of interest is not covered below. For more on the conference and the EPPC project on Neuroscience and the Human Spirit, visit http://www.eppc.org/programs/neuro.html, http://www.eppc.org/programs/neuroconf.html, and especially http://www.eppc.org/newsletters/newsf98.html.

The Threat of Science

The overarching theme of the conference is the threat, both real and imagined, of neuroscience and other sciences which have as their goal thorough explanations of human behavior, to our traditional ways of thinking about ourselves. It looks as though science, in showing that we are fully caused creatures, might rob us of the free will and autonomy which are normally taken to be the basis of morality and the law. As Dr. Frederick K. Goodwin, organizer of the conference puts it in his opening remarks, the question at the core of the gathering is "Do …scientific advances challenge the first principles that the majority of our citizens believe provide the very foundation upon which our civilization rests – free will and the capacity to make moral choices?" Put otherwise: "Does this growing understanding of genetic and environmental influences on human behavior leave any room for free will?" (opening remarks, p. 3) And again: "How can the ever-mounting discoveries of biological, genetic, and environmental factors shaping human behavior be integrated into our culture without contributing to further erosion of individual responsibility?" Goodwin expresses the intuition behind the threat: "To the extent that our choices are not truly free, it would seem that we have less moral responsibility for them." In other words, he, and he supposes most people, sense a deep incompatibility between what science tells us about ourselves and what is assumed to be the necessary basis for responsibility: a "truly free," originative capacity to make moral choices, or what philosopher Galen Strawson calls a "causa sui" (self-caused) self.[1]

Other variants of the threat come up throughout the conference, for instance, the threat to personhood: "…the more neuroscientists discover that our brains function in a manner analogous to some sort of biochemical robot, the more imperative it is that we as human beings not lose contact with that essence that prevents us from seeing ourselves as mere automatons" (Dr. Frederick J. Frese, panel 2, p. 14). Science may also threaten the basis for imposing accountability: "…will our greater ability to predict crime reduce our ability to condemn it?" (James Q. Wilson, panel 3, p. 3). The political and cultural battle lines of the threat are most vividly drawn by Charles Krauthammer: "…there’s an ideological agenda…, a sophisticated, elite culture that finds such notions as free will and human autonomy not just tiresome but [similar] to old religious superstitions. Instead, it would like to bring us all to the sophistication of a brave new mechanistic world in which moral judgments are personal and thus suspect; in which individual and personal responsibility is lost in a fog of a therapeutic culture whose ultimate sin is to be judgmental and whose most cherished ideal is healing" (panel 4, p. 15).

There are two components to the threat. First is the proposition that explanatory science undercuts the notion of a special human capacity for ultimately free choices, choices which have some essential aspect that owes nothing to either environment or heredity. The second is the proposition that this capacity is the necessary basis for morality, accountability, credit, blame, justice, and the whole nine yards of Western civilization. While the first proposition, from a naturalistic perspective, looks to be true, the second is not. That is, it is very possibly the case that nothing about us human beings escapes the net of natural causality, even the very highest capacities for deliberation and choice, but it is not the case that this leaves us bereft of justifications for moral judgments, social sanctions, praise, blame, and the whole nine yards, etc. Why? Because, in a nutshell, our moral vocabulary and practices find adequate justification simply as instruments of character formation and social control; they need no freely willing agent as a target; neither does human dignity require such freedom. Our intuitions and practices concerning credit, blame, retribution, merit, etc, may indeed change to some extent once it is seen that libertarian free will is insupportable, but we need not fear that the ethical and legal floor drops out from under us.

It should be emphasized early on that it isn’t determinism, buttressed by science (with quantum exceptions), that especially threatens the ultimate, originative freedom ordinarily thought necessary to ground moral judgments. Indeterminism, although it’s often brought in to rescue free will, can’t serve as the basis for this sort of freedom, since after all, it is behavior arising from our stable characters and dispositions, not behavior which has a random, indeterminate aspect, which is considered appropriate to praise and blame. (This does get mentioned at the conference: both philosopher Peter Caws and legal scholar Michael Moore (panel 3, p. 19) see that randomness doesn’t help in getting us the sort of free will, as philosopher Daniel Dennett puts it, worth wanting).[2] So it isn’t science which is at fault for invalidating the notion of our ultimate freedom, rather it’s that the very concept is suspect at its core.

The traditional, libertarian notion of free will holds that afs agents we very much cause our behavior, but that in some respect we are agents that are uncaused first causes or that cause ourselves (very much like God, actually). Therefore a robust notion of causality,[3] one very much countenanced and encouraged by science, is necessary to one side of this conception (the "output" side, let us say), while it is denied to the other, "input" side. Science merely highlights the difficulties of locating the libertarian agent - one that causes but is not fully caused in turn - within a world otherwise amenable to description and explanation. A major omission of the conference was not to have outlined (as it did admirably the advances in neuroscience) the basic philosophical positions on free will. If one supposes one’s free will is being threatened, and that such free will undergirds much that’s held dear, it would be helpful to describe exactly what this freedom consists in to see if indeed it’s worth defending, or capable of defense, or indeed whether such freedom is really necessary to undergird anything.

Strong Reductionism vs. Causal Explanation

The scientific threat to human freedom and dignity sometimes assumes the form, in this conference, as the encroachment of what might be called "strong explanatory reductionism": the idea that explanations of behavior at the level of physics, biology, and (eventually) behavioral neuroscience, might supplant explanations at higher, agent/personal levels. Since concepts of moral responsibility and personhood necessarily involve agent-level attributions of motives and rationality, replacing such attributions with reductionistic sub-personal accounts of behavior might indeed threaten concepts of morality and self. As William Bennett puts it in his keynote address, "Simply, people are not morally responsible if they are controlled by laws of biology, chemistry, and physics…", or as Dr. Frederick J. Frese expresses it, "…we must begin to take control of ourselves and not just be a function of scientific molecules happening…" (panel 2, p. 17).

The main proponent of strong reductionism mentioned at the conference is of course E.O. Wilson, whose book Consilience, as well as his earlier works on sociobiology, put him in the forefront of those (few) who imagine that science somehow (it’s not clear how) might eventually assimilate higher level moral vocabularies and practices. Bennett, not surprisingly, derides Wilson’s vaulting ambition, "the claim…that science will be able to replace the humanities, unify all knowledge, and be the unifier itself of that knowledge" (p 3, keynote address). The issue, Bennett says, "is whether [science] is the only truth, the only meaningful account of human activity" (keynote address, p. 6).

Bennett goes on to make the eminently reasonable point that of course science is not the only truth, that "electrons don’t have biographies" but people do, and that of course we need other sorts of higher level vocabularies and explanations to account for ourselves and our actions and to serve other purposes besides explanation altogether (e.g., pursuits in the humanities).  Strong reductionism applied to human affairs, espoused by only a radical few, is an obvious non-starter and thus a spurious threat to our dignity. E. O. Wilson is all too easy a target (he must be, if both Bennett and Rorty agree about him!), and we can rest assured that human biographies (except for episodes of brain-based mental illness) will always be told in terms of beliefs and desires, not neurotransmitters and peptides. Moral injunctions can only be issued to persons, not to their discrete motivational components, however well these components are understood at the neural and computational level.

But there is more to Bennett’s (and other conference attendees’) objection to the scientific program than the sensible rejection of strong reductionism, and this has to do with determinism. The fear seems to be that since science shows we are strictly natural, material creatures, subject to cause and effect, we cease under such a description to be the justifiable objects of what philosopher Peter Strawson called "reactive attitudes" such as admiration and resentment, gratitude and anger. [4] Scientific advances indeed pose a threat to libertarian agency (normally thought necessary to justify such attitudes), not because neuroscientific accounts will ever replace person-level accounts, but because it will be seen that our higher level capacities are instantiated in and arise out of strictly material substrates and networks. By showing that the embodied and socially situated brain is all that’s necessary for consciousness, rationality, [5] character, and the development of the moral sense, science shows us that there is no need to posit an immaterial ghost in the machine, or mental rider that controls the physical horse. There is no empirical or explanatory justification, therefore, to posit an originative agent that escapes being, in all respects, at the effect of the various conditions, physical and social, that constitute and surround the person.

That this troubles Bennett is illustrated by his quote from Tom Wolfe:

"The new generation of neuroscientists are not cautious for a second. In private conversations – the bull sessions, as it were – that create the mental atmosphere of any hot new science, they express an uncompromising determinism. Science for them becomes a court from which there is no appeal and is thus rendered a self-enclosed theory of human existence." [6]

Bennett thinks that accepting such Laplacean determinism, "the deterministic end-all, be-all argument," would undercut freedom and therefore moral responsibility (keynote address, pp. 5-6). More precisely, deterministic explanations that reach into the personal, agent-level realm of behavior conflict with the ordinary notion of moral responsibility: we can’t, it is commonly supposed, justifiably apply moral sanctions to behavior which results from choices that could not have been other than what they were, given the governing conditions. But why, precisely, can’t we?

First, even if strong reductionism is false, it may nevertheless be reasonable to suppose that human behavior, even at the highest levels, is still a function, albeit vastly complex and practically speaking unpredictable, of factors at all explanatory levels, from physics and biology, to neuroscience, to the neighborhood, to culture (television). This is indeed the naturalistic presumption: that there’s nothing about us that escapes the causal net, even if that net can’t be cashed out in terms of lower level mechanisms but instead must make use of higher level nomic generalizations, for instance those of folk and other slightly better qualified psychologies, behavioral choice theory, economics, etc.

I won’t defend such naturalistic inclusiveness here, but will simply make the point that if indeed we are conceived to be completely caught up in an explanatory, causal net, this fact does nothing whatsoever to lessen the necessity and wisdom of applying moral categories to behavior for purposes of molding character. There is simply no substitute for the various social and personal nudges that moral, evaluative language makes available to shape us into mature, reasonable adults. That the standard notions of free will and moral responsibility wrongly suppose (on my naturalistic account) that moral practices need an "ultimately" deserving target doesn’t mean that moral practices are ineffective. Indeed, they are effective precisely because we respond quite reliably to moral sanctions and blessings, which we wouldn’t if we had some sort of ultimate freedom to ignore them. And our dignity as rational, responsible, accountable agents survives the discovery that we have been caused to be such agents; we don’t need the non-naturalistic dignity of self-originating agent-hood.

Father Robert Sokolowski understands pretty well that persons are a function of conditions: "There is no other way to shape a person except by acting. And unless the circumstances for action are available – that is, the kind of cultural condition in which actions become possible and praised, evaluated properly – the characters won’t be formed" (panel 1, p. 18). But then, responding to the supposed threat of deterministic explanations to character and responsibility expressed by Goodwin, he says that people need to "feel that there is definite irreducibility to character, actions, and things like that." He wants to defend what he calls a "basic, standard vocabulary" of moral discourse. If this irreducibility means just that we can’t understand people (or treat them) just as collections of biological and neural processes, but that we must deal with them at the agent, personal level, then of course he’s right. Strong reductionism simply isn’t a viable approach to understanding our higher capacities or a basis for dealing with people as moral agents.[7] But if he means that there has to be some non-deterministic element or "first cause" within each person to support the standard moral vocabulary (what Goodwin implies from time to time during the conference), then he (Sokolowski) is wrong.

To repeat: We shouldn’t suppose that just because strong reductionism is false, that we are therefore special exceptions to naturalistic causality, that is, we shouldn’t suppose that its falsity somehow endows us with libertarian free will. Agent-level attributions of motives and rationality will always be necessary, but this doesn’t mean that it isn’t sensible to suppose that our characters, selves, choices, and actions arise out of physical and social conditions, which if we knew enough about, would enable us to construct plausible explanations of such agent-level phenomena, explanations which wouldn’t include appeals to libertarian free will. Further, and crucially, such explanations don’t reduce the need for and appropriateness of moral language and practices, although these may change in certain ways when we dispense with free will. Reactive attitudes – our naturally evolved and socially modulated responses to human transgressions and triumphs – won’t disappear under naturalism, but they may alter in emphasis and application (for the better, I would argue, see below).

Freedom of the Gaps

Given the fear that explanatory science does away with libertarian free will (true) and with it the basis for moral judgments (false), it’s not surprising to find in the conference many assertions about the limits of science. Freedom, again very much like God, perhaps resides in the gaps that science yet leaves in our self-understanding. But a good deal of the conference is dedicated to documenting the ominous (some think), rapidly accelerating advance in self-understanding ("Neuroscience…), which seems to leave less and less room - fewer gaps - for freedom (…the Human Spirit"). Goodwin, a highly respected psychologist, is himself no slouch in adducing the mounting evidence that behavior is a function of various factors; indeed he predicts that "as these genetic things come out, we’re going to be facing multiple correlations where you put environmental and genetic things into the regression and you’re going to find that regression scores are going to go well above.[8]" (panel 3, p. 27) That is, statistical models of behavior may eventually be able to account for over 90 percent of the variability in behavior as a function of the factors in the model. Goodwin acknowledges that this raises a serious challenge to the traditional basis of human dignity: is the residual 10 percent of unexplained variability free will, or is it just experimental noise, resolvable in principle by better science?

William Bennett is eager to show that science has "natural, inevitable limits" (keynote address, p. 9). What science can’t get at, ultimately, is the "human soul." Similarly, Robert Phillips, M.D. says in the panel on law and public policy, that "Intent… is something that perhaps finds itself rooted more deeply in the soul, and the ability to define intention and its application in the criminal law is a pursuit that I think will go unfulfilled" (panel 3, p. 13). Phillips, like Bennett, is claiming that a special, unfathomable personal "something" – the soul – will always elude explanation or prediction and therefore will continue to serve as the locus of freedom and responsibility, as traditionally conceived. Likewise, philosopher Peter Caws, who otherwise makes several good points against dualism and mentalism which go against the ideological grain of this conference, states rather categorically "I’m not only a knower but an agent, and my situation as a subject is out of the reach of scientific explanation, [which] means that no argument can show me that I’m not free. So I remain responsible. Nothing changes that." (panel 1, p. 8). And again, psychologist Martin Seligman, talking about regression models, hopes that "…the space between .4 and 1.0 might be the space of unpredictability, the...place in which decision and free will might occur" (panel 3, p. 18).

There are other examples, but the point is this: explanatory gaps and "in principle" limitations of science have a regrettable tendency (regrettable, that is, if you are a traditionalist about freedom and morality) to eventually be filled or transcended. To pin one’s hopes for libertarian freedom on the continued existence of such gaps and limitations is a poor bet. Moreover, and perhaps more tellingly, a gap in understanding isn’t anything like a positive demonstration of the existence of the radically free, libertarian agent that Bennett, Goodwin, et al. suppose we must be to sustain moral responsibility. But such a demonstration, like an existence proof of God, is what must be undertaken to establish, independent of the possible shortcomings or limitations of science, the reality of libertarian agency.

No one attempts this at the conference, although philosophers such as Robert Kane and neuroscientists such as Benjamin Libet have tried elsewhere, with little success, I’ve argued.9 To respond, as I suspect some libertarians might be tempted, that such a demonstration is unnecessary - to claim that our undeniable feeling of freedom establishes the metaphysical fact of freedom - would simply amount to a rather naïve intuitionism, quite at odds with the scientific, empirical ambitions of the conference.[9] And of course any empirical quest to discover a self that is in some respect not a function of surrounding and pre-existing conditions only ends up closing the explanatory gaps where such a self might survive. This is the poignant plight of the conference: how can we pursue science aggressively yet save the libertarian agent from the very knowledge we amass? Answer: with great difficulty.

Hard Compatibilism vs. Mitigation

The award for tough guy of this conference has to go to law professor Michael Moore, from the University of Pennsylvania, who wholeheartedly accepts the full inclusion of the individual in the causal network, but who nevertheless sees no incompatibility with this and traditional ascriptions of responsibility, with their full complement of punitive implications. We might call this "hard compatibilism," hard because while it sees causal explanations and ascriptions of responsibility as compatible, it countenances no softening of our reactive dispositions (e.g., the desire to retaliate) that we ordinarily think might follow when we discover the full causal story behind behavior. Hard compatibilism is opposed to what seems to me the widely shared, natural inclination to moderate retributive responses (or effusive, fawning praise) directed solely at the individual when it is clearly seen that the causes for behavior lie in external circumstances, which if they had been different, would have made the behavior less likely.

Moore himself provides a vivid example of his hard compatibilism. A faculty colleague of Moore’s had an annoying propensity to repeat World War II stories, and it was discovered that the mere mention of Harry Truman would trigger the story telling:

"It then became a faculty game, who could most subtly weave in something about Truman and see if you could [elicit] these same old boring stories. Now it is an offense, although not criminal, it is a moral offense to bore your colleagues at lunch. We held him responsible every time. Were we responsible? Of course, we caused him to do it but he also chose to do it – did it intentionally, fully blamable. Neither law or morality has a one-to-the-customer rule. You actually could have more blameworthy people than you have blameworthy results. We were both responsible. His isn’t diminished by ours." (panel 3, p. 9)

Moore understands his responsibility for eliciting the "moral offense" of being bored by his colleague. Without Moore’s deliberate incitement of his colleague’s story telling, it wouldn’t have happened, at least not as often. But strangely, Moore believes that his own manifest responsibility does nothing in the least to diminish the story teller’s. That is, for Moore there is no mitigation of an agent’s responsibility, even when it is seen to what extent (in this case, a rather great extent) the agent’s behavior is a function of conditions outside awareness and thus outside the possibility of control.

Why does Moore think there no mitigation? Moore might say, "Well, my colleague should have known he was boring me, even if I was the one to trigger his story telling." Perhaps he should have known, but he didn’t, and Moore could have easily filled this gap in self-awareness by pointing it out, instead of exploiting it for his own amusement.[10] Moore’s role as intentional instigator mitigates the responsibility of the story teller (at least on the occasions that Moore instigated) precisely because Moore knew what was going on, while the story teller did not. If we understand moral practice to be not the ascription of ultimate blame- or praise-worthiness, but rather the instrumental shaping of behavior via agent-level interventions to effect a desired outcome, then the discovery of the relative contributions of agents to an offense will calibrate our sanctioning (or praise) of those agents in proportion to their contribution. As responsibility is distributed, so is blame and credit, and these are instrumental, not ultimate.

Moore’s motive for espousing hard compatibilism may lie in supposing that mitigation puts us on a slippery slope, in which justifications for ascribing responsibility and accountability disappear altogether once the full causal story is told. But of course this doesn’t follow, as Moore’s own example illustrates. The causal story simply directs our interventions to the most efficient pressure points, those which when pushed have the best chance of redirecting (or reinforcing, if that’s what we want) the behavior in question. Responsibility gets ascribed to – that is, accountability is enforced upon – those agents which have the capacity to respond to sanctions, praise, advice, grant awards, defunding, or other agent-level interventions. So we sit the bore down and tell him (gently) about his regrettable tendency to repeat war stories, sometimes triggered by "Truman", and we sit Moore down and tell him (sternly) to stop exploiting this tendency, et voila, the boring behavior stops. This redirection of interventions to take into account the causality is exactly paralleled by the redirection of our sympathies in response to the causal story. Exculpatory mitigation is a natural response which tracks causality, which in turn prompts the pragmatic, efficient allocation of interventions.

That Moore evinces no such sympathies, and holds to the hard compatibilism that insists we are both "fully caused and fully responsible," is itself a curious behavioral outlier, which might be explained not just by fears of slippery slopes, but an interacting combination of ideological commitments, personality traits, and honest, hard to refute arguments. The latter, however, are not to be found in his presentation,[11] since he merely asserts that: "Fill out the [causal] story and I don’t think you’re going to end up saying well it’s only partial or inclining or predisposing or probabilistic cause. Fill out the story, he’s fully caused and fully responsible. There’s not an incompatibility. That, I think, should lead one to the happy conclusion, you can have your cake and eat it too." But why shouldn’t we end up thinking that the causal story does indeed distribute and thus mitigate responsibility, as illustrated above? What’s the counter argument?

Moore says simply that "the law…blames us for a razor point in time making a culpable choice and acting on it, i.e., having that choice, issuing an action that causes certain things in the world to occur that instantiate the representation under which we acted. It’s a razor point in time judgment" (panel 3, p.21). The law, conveniently for attributions of ultimate responsibility, blocks the consideration of causal explanations in favor of placing the agent fully, and freely (in the radical libertarian sense), in control at the moment of choice. But of course there’s no evidence for such a causally insulated "razor point in time" or of the libertarian agent that supposedly takes advantage of it. The tail of free will is simply wagging the dog of causal understanding.

Causation and Justice

Some of the other conference attendees do suppose, more sensibly, that not all the retributive cake survives our explanatory appetites, that mitigation does indeed follow from causal explanation. Dr. Robert Phillips notes that in capital cases that sometimes "the jury may come back and in fact find that the behavior met the test of mitigation and life should be the sentence as opposed to death" (panel 3, p. 25). Dr. David Rowe raises the question of mitigation starkly:

"My question is if we have a child who’s adopted at birth and raised by a happy middle class family who goes on to commit a number of crimes, including a heinous murder, and discover that in the biological family of that child there’s much criminality, prostitution…degeneracy, if we want to go back to a very old word. Would that be evidence to mitigate punishment? And that’s what we can do today. Suppose in ten or fifteen years we can also take the DNA and run it through the DNA chip… and that DNA tells us that people with that type of pattern, instead of having a one percent chance of crime or whatever the population base rate of these serious crimes, it’s actually eighty percent, would that be mitigating?" (panel 3, p. 24)

Of course, we need not imagine hypothetical advances in sociobiology to raise this question, since we already know that, leaving aside inherited dispositions, certain environments and certain sorts of upbringing are associated with far higher rates of violent criminality. It’s clear, in other words, that character and behavior are indeed a function of both environment and genetics, so that in seeking to remedy our moral failings, we need to address both. But the aim of the law is not fundamentally the scientific one of understanding behavior, nor of directing social (and eventually biological) engineering to alleviate crime, but is instead the assignment of criminal responsibility under the assumption of free will. Mitigation in sentencing is the law’s nod to obvious influences on an offender’s behavior, be they environmental or genetic, since all but the hardest compatibilists find their sympathies swayed by a clear causal story. But for the most part the law’s only unit of responsibility, the only pressure point it seeks to push, is the person, the most proximate cause of the offending behavior, and it applies this pressure with a very blunt, traditional instrument: the non-naturalistic assumption of our being first causes. Other pressure points, which take into account a naturalistic view of ourselves as a function of surrounding conditions, are policies concerning child welfare, nutrition, education, community development, mental and physical health, and drug abuse treatment, to name just a few.

Of course justice still ends up as an instrument of character adjustment and social change, even if it operates under antiquated suppositions about freedom of the will. The threat of criminal sanctions obviously works to shape behavior, and the ancient motives of retaliation and revenge – codified, smoothed out, and somewhat de-personalized as they are in the workings of the law – were obviously selected for by their essential role in repulsing and deterring aggression. Thus the courts play a necessary role in deterring wrongful behavior and incapacitating dangerous individuals, while expressing an (unfortunately) necessary component of human nature. But by being tied to the assumption of libertarian agency, they effectively block the admission of, and therefore action on, the fact that people are indeed caused creatures to the very core. The supposition that to have justice we have to have free will suppresses the deeper, scientific consideration of the causes of behavior, which of course guarantees that criminality and moral lapses will survive to be punished another day.

That the law may eventually have to face up to its central anachronism prompts some concern at the conference. As Judge John T. Noonan puts it "We have all the concepts of the moral moved into the legal world. They interact with each other, support each other and we can’t get rid of them without recasting everything we do" (panel 3, p. 14). Noonan sees that accepting a scientific account of ourselves, one which dispenses with libertarian agency, would necessarily undermine the current justification for assigning responsibility: that the agent could have done otherwise. Such "recasting" would instead place the person being judged in a causal context, a context in which it becomes obvious that imposing sanctions on the person is but one (although often necessary) response to the fact of criminality. Such a "court of causes" would find the "ultimate" responsibility for crime distributed over the multiple factors that shaped the person and behavior in question, in which case retributive justifications for punishment lose a good deal of their plausibility. Instead, myriad opportunities for ameliorative, preventive action would present themselves, and the emphasis would shift from the punitive to the restorative.

All this would indeed constitute a revolution in our concept of justice and in its aims, a revolution which hard compatibilists would be in no rush to support. After all, forgoing the satisfactions of retribution, (e.g., imposing the death penalty, hard time in prison) isn’t easy (and I speak from deep and ongoing personal experience on this), even though it’s probably the best course in the long run if we are serious about reducing criminality. And in the long run this is how our concept of justice must change if it is to co-exist with a naturalistic understanding which places us and our voluntary choices in a causal matrix of environment [12] and heredity. (Remember, such understanding won’t be, impossibly, strongly reductionistic, but will rather make the best heuristic use of the developing behavioral and social sciences.)

In Conclusion

But why should anyone agree to adopt a naturalistic understanding of ourselves in the first place? Certainly most of the conference speakers would vigorously dispute the naturalistic premise that we aren’t causal exceptions to the natural order. Free will is rather too much of a foundational tenet, a basic, presupposed value (as Goodwin puts it, "the very foundation upon which our civilization rests," opening remarks, p. 2) to question seriously in public discussions of human nature and social policy.

Nevertheless, the thrust of the entire conference is naturalistic insofar as it takes quite seriously the advances in neuroscience that are the crux of its concern, thus it necessarily embarks on such questioning. As sketched above, there are no empirical or explanatory grounds for granting the existence of an originative agent that acts as a miniature first cause - a local god, so to speak. Nor is it necessary to posit such an agent as the object of reactive attitudes in order to sustain moral practices. But it is just such an agent, Goodwin and others think, that is the necessary foundation for civilization, dignity, and responsibility. This points again to the inherent tension of pursuing explanatory science on the one hand, and with the other holding tightly to something (libertarian free will) which by it’s very definition defies inclusion in any thorough explanation.

In the end, the acceptance or rejection of naturalistic inclusiveness depends on a host of interlocking cognitive and emotional commitments, some of which may not admit of further justification (e.g., my commitment to knowledge as primarily intersubjective, not a matter of private intuition; or someone else’s commitment to finding an intrinsic purposive meaning to the totality of existence). Therefore it is nearly useless to argue for naturalism (or its contrary) per se. However, what can be done is to show the conceptual incoherence and empirical implausibility of libertarian agency, and make the case that our moral intuitions and practices, although they may change to some extent, don’t fall apart under a naturalistic understanding of ourselves (see www.naturalism.org for a good deal more along these lines). We don’t need to be radically free to be good, or to be held accountable, or to become responsible citizens in an open society, or to make the distinction between illness and evil.[13] Our values stay intact even as we come to understand their basis in our existence as strictly material creatures. This insight permits an accommodation between morality and science, an accommodation that the conference, by virtue of its overall non-naturalistic stance, could never have achieved.

Thomas W. Clark, November, 2000

Notes

[1] See Strawson’s "Luck Swallows Everything," Times Literary Supplement, June 26, 1998, on the web at http://world.std.com/~twc/strawson.htm

[2] Dennett, D. (1984), Elbow Room (Cambridge, MA: MIT Press).

[3] Taking a deflationary Humean view of causality, we can construe it as predictively reliable conjunctions of event classes.

[4] Strawson, P. (1962), ‘Freedom and resentment’, Proceedings of the British Academy, XLVIII, p. 1-25.

[5] It is sometimes supposed that our rational faculties, in order to do their job, must be insulated from causal influences. But carefully deliberating on reasons for action is precisely to be a finely tuned detector and predictor of causal relations, in service to our motives and desires. Something uninfluenced by such considerations would have no reason to move one way or another, and so would be functionally inert. Having libertarian free will would thus do nothing to make us more rational, or more effective agents.

[6] is from Wolfe’s 1996 Forbes article, "Sorry, but your soul just died", a must read, at http://www.orthodoxytoday.org/articlesprint/WolfeSoulDiedP.htm.

[7] For why, see Dennett, D. (1987), The Intentional Stance (Cambridge, MA: MIT Press).

[8] Clark, T., "Fear of Mechanism," in The Volitional Brain: toward a neuroscience of free will, Libet, B., Freeman, A., Sutherland, K. eds., Imprint Academic, UK, 1999, on the web at http://world.std.com/~twc/fearof.htm

[9] It is such differing epistemic commitments, reinforced by cultural norms and personality differences, which I think help explain the irresolvable disputes between science and religion. See, for instance the Humanist articles "Relativism and the Limits of Rationality" at http://world.std.com/~twc/relativi.htm and "Faith, Science and the Soul" at http://world.std.com/~twc/faithsci.htm.

[10] To expand: Given the existing conditions, Moore couldn’t have done otherwise but bait his colleague, but he had the capacity to do otherwise if conditions had been different. Thus it makes sense to change conditions via moral judgment with the aim of preventing future offenses, in this case to decrease Moore’s propensity for exploiting other’s weaknesses by letting him know it’s wrong. It is this capacity that marks him (and most of us) as moral agents: we are persons for whom the anticipation of judgment makes a difference in behavior.

[11] Moore has arguments elsewhere, for instance in his 1998 book, Placing Blame: A General Theory of the Criminal Law, which I critique at length on the Criminal Justice page.

[12] For a detailed analysis, in the context of addiction theory, of how voluntary choices are a function of environmental contingencies, see Gene Heyman's Resolving the contradictions of addiction, Behavioral and Brain Sciences, 1996 19, 561-610, with open peer commentary; and his chapter, "Is addiction a chronic relapsing disease?" in Drug Addiction and Drug Policy, Philip Heymann and William Brownsburger, eds., Harvard University Press.

[13] For a brief discussion of why the illness/evil distinction survives under naturalism, See "Materialism and Morality: the problem with Pinker," from the Humanist.