Mind-body dualists, perhaps the majority of folks in the West, suppose, as Steven Pinker once put it in Newsweek, that the brain is a something like a pocket PC for the soul. The brain isn’t enough on its own to support rationality, memory, emotion, the sense of self and all the other riches of mental life. There must be something else, something more than our physical selves, that uses the brain as a kind of interface with the material world. That they think this isn’t surprising, since 1) we’re likely natural-born dualists to some extent (as Paul Bloom has argued), 2) the meme of dualism is heavily reinforced in our culture by religion and new age thinking, and 3) it sure feels like there’s something immaterial in here, doing the mental work.

The same point often applies to such things as morality and responsibility: for us to be morally responsible agents people think there’s got to be something more than the brain in charge. If there weren’t, we could simply plead that, your honor, my brain made me do it, and brains, however complex, are physical, deterministic mechanisms. Without the freely willing, determinism-evading soul to take the ultimate credit and blame, on what grounds do people really deserve praise and punishment?

Dualism is also a refuge for many mainstream religionists, who see in the apparently immaterial self a reflection of god’s essential spiritual nature. Indeed, the soul or non-physical mental agent is often thought of as a virtual little god, which has special powers of choice and rationality that transcend what mere materiality could possibly accomplish. Theologians such as Alvin Plantinga and John Haught argue that in order to have trust in our reason, we must suppose we are more than physical beings. Evolution operating via natural selection could only assure that our beliefs are adaptive, not that they are true. Although they might not be theists, adherents of various New Age worldviews often join with religionists in celebrating the virtues and wonders of the categorically spiritual realm, one that ranks considerably above the physical in their estimation. 

Given the widespread prejudice in favor of dualism, we need a book that explains how the brain might be enough, one that makes a case for an unapologetic physicalism as sufficient for reason and responsibility. Were its authors religionists, that would add to its credibility, since they couldn’t be accused of grinding an axe against god and spirituality. So I’m pleased to report that there is at least one such book, Nancey Murphy and Warren Brown’s Did My Neurons Make Me Do It?. Both are card-carrying theists resident at Fuller Theological Seminary, but as the book makes abundantly clear, neither are dualists, at least with respect to human persons (god and ultimate reality are not discussed).

The title, looking to attract a wide lay audience, is a bit misleading, suggesting that the main focus is on moral, and in particular, criminal responsibility. Although the last two chapters take up moral responsibility and free will, the bulk of the book is a systematic defense of non-reductive physicalism about the mind. The authors want to construct a clear explanation of how “reason gets a grip on the brain” without resorting to any obscure dualisms or simplistic reductionisms, and it’s on this basis which viable notions of freedom and responsibility can then, finally, be introduced. So what we get is a very deliberate, detailed consideration of some of the latest thinking about thinking, meaning, reference, mental causation, reductionism, emergence, representation and rationality, all presented in an empirically-based and philosophically-informed theoretical context.

It’s a nicely layered presentation, building the relevant concepts by traversing the “hierarchy of complexity in the animal world,” from simple biological mechanisms, to ant colonies, to brain physiology, to neural nets, and ultimately to natural symbolic and semantic creatures such as ourselves. By the time we’re ready to consider what it means to be a moral agent, we’ve got a good overview of cognitive neuro-philosophy under our belts, at least one broad take on it. All told, it’s a very rich, well-referenced book, drawing on a wide range of experts who each contribute solutions to parts of the puzzle. Those of different philo-scientific[1] persuasions, not to mention dualists, will of course differ on whether its thesis holds water, but whatever its ultimate fate, it’s a good stab at a all-in-one, comprehensive theory of how strictly physical beings can have causally effective and rational minds. Having such minds, the authors argue, is sufficient for moral responsibility and free will of the compatibilist variety – that is, compatible with not having libertarian, contra-causal free will. We don’t have to be unmoved movers, just self-controlled movers that can appreciate reasons and be responsive to moral norms. All of this is perfectly possible for beings with complex, physically-instantiated and socially-situated minds like ours.

Physicalists and naturalists like myself will find little to object to in the overall approach, which eschews any appeal to supernatural, occult causation in accounting for higher-level human cognition. But having accepted the basic naturalistic premise about explaining persons, the fun begins, since there are interesting disagreements about how we get from mindless organic parts – neurons and neurotransmitters hooked up to muscles – to conscious beings that act on the basis of reasons. Murphy and Brown defend what’s become known as non-reductive physicalism, the idea that, although we’re strictly physical creatures, the complex organization of our parts gives rise to categorically new causal powers, such that “higher level patterns of action…do some real work, and thus [are] not …reducible to the mass effect of lower level constituents” (53, original emphasis). This move helps to certify the reality of persons, of reasons, and of higher-level mental and intentional phenomena (beliefs, desires, rationality) generally. If instead we are causal or atomistic reductionists, supposing that all the real work gets done at the lower physical levels and that causation is all “bottom up,” we might think persons and their mental lives are eliminable in our description of what really exists. They don’t count as real entities in our ontological catalogue, at least not as real as atoms, molecules and mechanisms. If so, the idea that we’re rational and responsible moral agents that make a difference in the world seems hard to sustain.

But Murphy and Brown make a good case that neurons properly assembled in brains, brains properly connected to bodies, and bodies properly situated in society do indeed entail real mental phenomena that make for persons with real causal powers. The reality comes from the fact that higher level organization, both in brains and in culture, constrains and directs the ends toward which lower level parts and processes in brains and bodies are put to work. This is so-called downward causation: parts of persons, such as neurons and muscles, are recruited for ends that none of the parts have in mind. Having a mind just is to be constituted by a coordinated collection of higher-level processes that operate in service to the continued existence of the collection. This means that “agents are causes of their own behavior,” as they put it, not the mere working out of physical laws via their constituent parts as a strict reductionist might claim.

Since mental phenomena such as beliefs and desires require not just a brain and body, but an environment (physical and social) to which they refer and which gives them content, Murphy and Brown argue, following philosopher Fred Dretske, that the mental qua mental isn’t found “in the head,” but is in effect distributed between the person and their environment: “A mental state is a brain-body event relevant to or directed toward a social or environmental context – past, present or future” (40).[2]

Filling in this claim, of how mental phenomena refer, of how words mean, of how it is that we can act for reasons and yet be still fully natural beings subject to laws of physics and chemistry, is a central burden of the book. Drawing on the work of Andy Clark, Donald MacKay, Terrence Deacon and Alicia Juarrero, among many others, the authors put together a systems theory of embodied cognition which purports to explain how properly contextualized brain events take on intentional properties. Such properties, they claim, are essential, irreducible elements in a perspicacious account of intelligent behavior. Their account does much to suggest how emergence, a notoriously fuzzy concept, might actually work. Not being an expert in all this – it’s a complex business indeed –  I won’t pronounce on its success. But it’s clear that on their view, there’s no conflict or competition between the physical story of what instantiates a cognitive system, the behavioral story of why cognition is needed in the first place, and the higher-level informational story about representation, meaning and understanding characteristic of human intelligence. All levels are real, which is to say they are all necessary for an explanatory, predictive theory of human action in the world (remember that basic physical elements and forces are also theoretical entities). To be a physicalist, therefore, is not necessarily to deny the existence of the mental, if one takes the mental to be a higher-order property of a sufficiently complex and flexible physical system that tracks its environment in service to self-preservation.

The mental, thus conceived, has real causal powers just insofar as the informational, intentional (representational, referential) properties of complex systems play an ineliminable explanatory role in accounting for behavior, and if Murphy and Brown are right (or something like their theory is right), they do. There’s no spooky mental force that acts on matter which violates the laws of physics; it’s all matter acting on matter in a way consistent with physical laws, but the pattern of action typical of complex cognitive systems like us isn’t something that can be accounted for or described by physical laws. The domain of law-like explanation that does account for it constitutes the mental domain, which although it’s constituted by the physical, can’t be reduced to it.

What’s less clear on Murphy and Brown’s account is whether the subjective phenomenal aspect of conscious mental states – qualia, the what it’s like to be in pain, to see red, etc. – are causally efficacious. Most of us suppose that consciousness is essential for flexible, foresightful, goal-oriented human behavior, for being (usually) in control in the way that makes us moral agents, but this claim is ambiguous. Certainly the mental (higher-order, informational, contextualized) brain-based processes that correlate with conscious states are necessary for such control, but what’s the causal role of qualia per se? As Jaegwon Kim has argued, if qualia are not reducible to something physical, functional, or representational, then it’s problematic to accord them a causal role, since from a systems-level view the causal work is already being carried out by the physical, functional or representational goings-on instantiated by the brain. Since it isn’t obvious how qualia could just be such goings-on (although conjectures abound), then it seems they might be non-functional epiphenomena. Indeed, in their discussion of Kim’s work, Murphy and Brown concede this point: “Relations between qualia, such as there being a difference between [the subjective experience of] red and green, can be functionalized and reduced; qualia themselves (the redness of red) cannot, and are therefore epiphenomenal. So, we would not categorically disagree with Kim…that some qualia are epiphenomenal…” (235, original emphasis). From this it appears Murphy and Brown think some qualia do play a causal role, but from their discussion of mental causation (193-237) it isn’t clear to me which, or how. But in any case, their theory admits the possibility that the phenomenal particulars of our conscious, subjective lives - the redness of red, the painfulness of pain - which likely supervene on causally efficacious neural processes (what we can justly call mental processes from a systems standpoint), may not themselves be causally efficacious.[3]

Should this worry us? Only if we suppose qualitative consciousness is the sine qua non of human freedom, responsibility and dignity. And many do. But this worry forgets that the neural processes that accompany consciousness (and that may in some sense constitute it) are causally effective and central to those higher-level capacities which support moral agenthood. Even if it turns out that qualia per se are epiphenomenal, their neural basis is not, so responsibility, like reason, gets a grip on the brain via purely physical means. We can and must hold mechanisms – organic, indefinitely recursive self-modifying mechanisms like ourselves – responsible, whatever is the case about qualia.[4]

The authors’ conception of moral agency draws heavily on Alasdair MacIntyre’s view that, as he puts it, the capacity for moral responsibility is “the ability to evaluate that which moves one to act in light of a concept of a good” (243). This requires a rather sophisticated cognitive architecture, involving the self-critical ability to evaluate one’s goals on the basis of social norms, and even to evaluate the norms themselves. Murphy and Brown make the case that their physicalist theory of mind and personhood has the resources to ground this conception of moral agency, but I think the conception itself is too narrow and high-falutin’; it’s a conception of an ideal moral agent, perhaps, but not the broader, general purpose conception that we ordinarily apply when holding each other responsible. For most practical purposes, we needn’t suppose that moral agents have the capacity or inclination to engage in a meta-level critique of their conception of the good, but only that they are standardly capable of internalizing norms and being responsive to the prospect of being held responsible. Given their elaborate defense of ideal moral agenthood, Murphy and Brown’s theory should have little difficulty supporting this simpler conception.

The final chapter provides a very useful critical survey of the current debate on free will, highlighting the implausibilities of libertarianism (the idea that we’re ultimately responsible for ourselves and our choices) exemplified by Robert Kane, but also taking issue with Daniel Dennett’s compatibilism as described in his book Freedom Evolves. The authors charge Dennett with a sort of reductive materialism that leaves us with only an as-if sort of freedom – free will in name only, not the robust freedom they think is possible on their non-reductive physicalism. I’m not convinced, however, that their compatibilism is all that different from Dennett’s, and indeed they acknowledge many similarities between his account of cognition and action and their own. The sticking point is their contention that Dennett is an instrumentalist, not a realist, about intentionality and consciousness, and that he’s a mechanist-reductionist about human agency, ignoring the role of top-down causation. So he only gives us “a very imaginative account of how complex machines could appear to have language, beliefs, morality, and free will” (298, original emphasis). However, Dennett doesn’t discount the reality of our capacity for self-reflective, socially-guided self-control – the constraining influence on behavior of the rational appreciation of reasons for ethical action. He too has inveighed against the “greedy reductionism” of those who imagine that higher level capacities are explicable on the basis of a strictly bottom-up view, even though we are complex machines (see note 4) composed of millions of “mindless robots” (the physicalist thesis). If Dennett is an anti-realist or eliminativist about some things, a robust, desert-entailing compatibilist freedom based in our real capacity for assessing choices in the light of normative considerations is not among them.

Although it isn’t discussed in their book, Murphy and Brown seem to agree with Dennett that a compatibilist conception of freedom and moral responsibility, that is, compatible with our being fully subject to causation (even if we gain a great deal of relative autonomy from simple biological and cultural determinism),[5] leaves our moral responsibility practices untouched. As they put it: “We noted (in the Introduction) that the most significant reason for wanting free will is to preserve traditional notions of moral responsibility and associated practices of social rewards and punishments” (266). This essentially conservative position puts the cart of culturally embedded notions of agency and responsibility practices before the horse of science-based theory. Yes, we are complex creatures who are proximately self-authoring on a naturalistic view of ourselves, but does this sort of non-ultimate autonomy support notions of desert, for instance desert-based retribution, that have historically been justified by the idea of supernatural, contra-causal free will? This is the important, real-world, policy-relevant question about free will and moral responsibility that Did My Neurons Make Me Do It? doesn’t address, which short-changed my expectations to some extent.[6] But that and a few other quibbles aside, this book is an illuminating contribution to a naturalistic understanding of mind and agency, well written and argued by supernaturalists, that I whole-heartedly recommend.

TWC, January, 2008

Notes