Does morality require a soul?
The Problem of the Soul: Two Visions of Mind and How to Reconcile Them, by Owen Flanagan, New York: Basic Books, 364 pages, $27.50
Humanists have always eyed science a bit warily. Just over a century after the 1687 publication of Newton's Principia, we find Wordsworth complaining, "Our meddling intellect / Mis-shapes the beauteous forms of things: -- / We murder to dissect." His Romantic fellow traveler Keats, disturbed that science had included even the rainbow in "the dull catalogue of common things," lamented that physics and philosophy would "clip an Angel's wings, / Conquer all mysteries by rule and line…"
Such concerns now seem a bit quaint: Few of us think rainbows any less pretty for being understood as refracted light. But if optics has lost its terrors, many remain uneasy when we begin to focus the scientific lens on ourselves. Do physics, neurology, and biochemistry leave any room for theology and spirituality? Can we continue to think of ourselves as beings endowed with free will and distinguished from the rest of the animal world by our immortal souls?
Well, no, we can't, for reasons laid out with brutal precision in Owen Flanagan's The Problem of the Soul. Flanagan, who holds professorships at Duke University in philosophy, neuroscience, and psychology, brings all three disciplines to bear in a tightly argued quest for rapprochement between the "manifest image" of the self inherited from millennia of philosophy and the very different image presented by contemporary philosophy and neuroscience. As scientists continue to confirm that human beings are just an especially clever species of animal, philosophers have hammered away at the old picture of the mind as an immaterial, unconstrained substance for which the body is merely a fleshy garment.
On the scientific front, souls seem increasingly in danger of being banished to the trash heap of redundant hypotheses. The problems on the philosophical end are, in a sense, still more disturbing to the conventional picture, because they go to the conceptual root of the manifest image. Many philosophers now believe that radical free will or fixed "true selves" are incoherent concepts.
It would be a mistake to suppose that the tension between the two worldviews presents a problem only for religious believers. As Flanagan notes, the manifest image has so permeated Western culture that even avowed secularists are likely to have at least implicitly adopted parts of it. Core tenets of liberalism, such as the emphasis on free, autonomous choice and individual rights, were originally closely tied to the Christian doctrine of the salvation of the individual soul and still bear the mark of that origin. Indeed, in a thoroughly secular age, the main challenge of Flanagan's book is really directed at people who trace their political ideology to the Enlightenment's classical liberal tradition.
Fragments of the manifest image show up in the most unlikely places. The aggressively atheistic Ayn Rand grounded her theory of enlightened self-interest on the premise that there's something uniquely rational about making decisions on the basis of one's interests over the course of an entire lifetime. That position becomes more difficult to sustain when we cannot appeal to a transcendent permanent "self," a subject of experience who remains "the same person" from childhood through old age. We can't identify that self with the contents of the mind or any specific personality traits, since those are always subject to change. But something more general, such as formal or structural features of the mind, would leave nothing to distinguish one person's self or soul from another's.
Flanagan endorses the view that there is no "further fact" of personal identity. When you strip away all of a person's experiences, memories, thoughts, and dispositions, there is no "self" left over who has all these things: The person just is these things. Yet Flanagan underplays the consequences of adopting this view. Among other things, it means that you may not have any reason either to care about the welfare of your future self or to feel bound by contracts or agreements made by your past self. Our inclination now is to create those reasons, partly out of a desire to situate ourselves in some coherent life narrative, but probably also because the idea of the persisting self or soul continues to influence the way we conceive ourselves. Whether that conception and the reasons it generates will outlast the manifest image remains to be seen.
The human capacity for free choice is another cornerstone of liberal thought that seems threatened by a thoroughly naturalized conception of persons. Real choices are supposed to be undetermined by what came before. When I make a genuinely free decision, no set of antecedent causes predetermines what I must do. But an exercise of free will is also supposed to be something that the agent does, not merely something that happens. It would not count as an act of free will if some nondeterministic quantum fluctuation in my brain caused me to do good rather than evil. These two conditions -- indeterminacy and authorship -- together define free will as traditionally conceived; but as Flanagan observes, they are mutually incompatible. To the extent that my actions are undetermined -- that I could have turned right just as easily as left -- they are not bound to any of my own past mental states. To the extent that my own experience and reasoning do explain my actions, those actions are determined and, therefore, not "free" in the radical sense.
Free will may be conceptually incoherent, but some believe that its denial is equally self-undermining. For instance, libertarian author Sheldon Richman has written that to "announce that one cannot freely choose is to announce that one is not really speaking but only making random noises by electrochemical necessity." If we don't believe in free will, the argument runs, then we cannot trust the reasoning that led us to that conclusion, since it is only the upshot of a physical process that would inexorably give rise to the belief whether it were true or not. To see why this is wrong, consider computers, which nobody thinks are endowed with free will. Not only are they capable of generating true conclusions by acting on perfectly deterministic rules, but the determinism is a crucial part of our confidence in their accuracy. Would you be more inclined to trust the calculations of a math program that could "freely choose" to spit out any answer, in defiance of the rigid laws of arithmetic? When we look closely, it becomes clear that free choice could play no important role in conveying meaning or inferring truth.
Resistance to these ideas, Flanagan believes, is largely a function of the discomfort they inspire. People cling to the manifest image not because it is defensible but because we fear that our lives lose meaning without it. The Problem of the Soul aspires to provide therapy along with argument, to convince us that the truth is nothing to fear: There isn't a Santa Claus, but it turns out you get presents anyway.
There is an unmistakably Wittgensteinian flavor to this approach, an attempt to dissolve the apparent problems generated when "language goes on holiday." We have no problem talking about "choice" or the "self" in everyday contexts, but when we attempt to construct philosophical theories about "free will" and "personal identity," we go astray. When our map proves to be badly drawn, we mistakenly think this means the territory doesn't exist. We are in the position of medieval astronomers who fear that if the Copernican theory of the solar system is true, "the earth" as we have conceived it does not exist. Yet the ground is still there beneath our feet.
Even so, Flanagan cannot quite claim, as Ludwig Wittgenstein did, that his analysis "leaves everything as it is." On the contrary, Flanagan expects that a more naturalized conception of persons will "dislodge some of our individualistic ways of thinking and acting." One sense in which Flanagan intends this is quite conducive to liberal values. Just as for Socrates the beginning of wisdom consisted in acknowledging one's ignorance, recognizing that our behavior is often automatic and socially conditioned may help us to be, if not more "free," then at least more reflective in our actions.
Often, though, the recognition that we are not radically free in some metaphysical sense is taken to undermine classical liberal views about justice. Perhaps the case for retributive punishment -- punishment for its own sake -- is weakened, but it would surely be a mistake to conclude that only radical freedom would make it appropriate to hold people responsible for their actions.
F.A. Hayek, who called free will a "phantom problem," turned that view on its head, observing that if we truly had free will, there would be no point in holding people responsible. What are praise and blame, after all, but attempts to causally shape the behavior of others? One might add, how could we simultaneously deny responsibility to others while acknowledging that we would deserve blame if we punished them inappropriately?
Similarly, critics of liberalism -- and some liberals as well -- believe that disparities of wealth and income are justified only if the well off "deserve" what they have in some deep sense. But as the late philosopher Robert Nozick observed, there are many things to which we are entitled, even though they are not deserved "all the way down." Being born with two working eyes is an accident of fate, not something the sighted have done anything to "deserve."
It does not follow that our eyes are up for grabs, subject to political reallocation. Our decisions -- our capacities and the uses we make of them -- are as much a constitutive part of us as our bodies. Respect for embodied persons still requires deference to our "unfree" choices and their consequences.
The account of morality given in The Problem of the Soul is emotivist: "Good" and "bad" are expressions of approval and disapproval, and saying that "murder is wrong" is tantamount to exclaiming "boo, murder!" It is here that Flanagan's claim to leave the core of what we care about intact is least plausible, but he does make a good case that the demand for some more transcendent basis for ethics is misplaced. What we seem to want in ethics is what Nozick would have called a "coercive" argument -- one so powerful that it must be accepted, even by a full-blown moral skeptic. But this is too much to ask, and it is more than we ask in the epistemic realm, where our situation with respect to a radical skeptic is perfectly analogous. There is nothing we can say to someone who rejects the axioms and inference rules of deductive logic without begging the question. Similarly, there is no deductive proof that induction -- inference from past evidence to future occurrences -- is valid. There's something obviously circular about asserting that we're entitled to inductive inferences on the grounds that they've worked pretty well for us so far.
If this seems comforting, it is probably because truly radical skeptics about either morality or logic are in short supply. But we do regularly face the problem of moral disagreement. Evolution gives us reason to expect that the most basic moral sentiments will be as universal as the ability to appreciate logic, but this still leaves the tricky moral details radically unsettled. Yes, we can all agree with Flanagan that murder and rape are bad, but where does that leave us with respect to the more subtle questions that are the focus of real debate?
Saying, as Flanagan does, that ethics should concern itself with "human flourishing" does not tell us who should flourish and to what degree. Should we be especially concerned with the worst off in society, or is it only the sum total of benefits and burdens we face, rather than their distribution across persons, that matters? How do we weigh different kinds of flourishing? Perhaps Flanagan is right that there will be no one "correct" way to answer these questions, but this means that they may ultimately have to be decided by nothing more exalted than the relative power of those who disagree.
Nevertheless, the fear that a naturalistic worldview will undermine all morality and meaning is exaggerated. Finding out that you are not the kind of metaphysical entity you might have thought you were doesn't make pain hurt any less, strawberries taste any worse, or love any less worth pursuing. It is not yet clear whether the more fleshed-out view we may ultimately develop will, as Flanagan believes, prove as psychologically satisfying as the one we leave behind. But if naturalism leaves us with a new set of problems in place of our old and comforting myths, it also leaves us with a new opportunity to create meaning. That, too, is a kind of freedom.