The Emergence of Free, Intentional Control: Reply to Haselager
Pirn Haselager (Chapter 7, this volume) explores the question of how we should think about free will and more generally human intention in action in light of tentative and slowly developing frameworks in neuroscience and embodied cognition. He starts with a note of sympathy for Reidian ‘common sense’ epistemology, which contends that we are rationally entitled without independent evidence or argument to certain general beliefs about the world and the cognitive tools that we bring to our encounter with it. We have to start somewhere in coming to learn things about the world and ourselves. The reasonable starting points are, roughly, those that we find ourselves accepting instinctively and that appear to be essential to our ability to engage in critical inquiry and action. Among these instinctive beliefs is our self-conception as purposive agents: confronted by more than one option that we have reason to pursue, and being uncertain about what to do, we deliberate and choose. Choice is a kind of uncertainty-resolving intention to act that—so it seems—initiates and (in complex cases) guides the completion of the action. It seems to us, then, that we are effective, purpose-driven choosers. But further, it seems that our choosing is free: having chosen one option, it was open to us to have chosen differently. Nothing necessitated that particular option, and partly for that reason, we are responsible for having chosen as we did.
Haselager allows that this description captures our instinctive conception and experience of deliberate agency and that it is prima facie reasonable to accept it. However, he adds that epistemic starting points may be overturned by empirical evidence, and he quickly proceeds to suggest that our common sense conception of agency is in need of significant revision. Why, exactly?
It is not that our well-confirmed sciences pertaining to the human mind demonstrate its falsity. Haselager (correctly) acknowledges that they are too immature to have done that. Instead, he thinks, there is a broad metaphysics of nature that is more promising than any alternative. This framework is ‘scientific materialism’, which sees all of nature as constituted by‘physical-law abiding interactions of material elements’ (Haselager, Chapter 7, 3). High-level organized phenomena are one and all determined from the bottom up, even though scientific practice necessarily starts from our macroscopic perceptual level and tries to burrow down. Furthermore, in thinking about human choice and action, Haselager has ‘not seen’ any explanation of how macroscopically significant indeterminism might be anything other than mere ‘chance’ (Haselager, Chapter 7, 3), and so he struggles to understand “what ‘could have done otherwise’ might mean, or why it is considered to be so valuable” (Haselager, Chapter 7, 9). So in theorizing about human freedom and moral responsibility, we should discard this naive but possibly incoherent notion of an ‘open future’ that we partly determine and instead focus on the undoubted flexibility of our responses to stimuli over time, that is, on the fact that we learn from experience and often adapt our responses to similar future circumstances.
It should be very puzzling that we are naturally disposed to think of our own agency in profoundly confused ways. Indeed, I do not think that we are. Causal indeterminism is a merely negative condition on our freedom. It is the precondition on something positive: the exercise of a two- (or multi-) way power of choice that is guided but not determined by reasons we have (from our subjective perspective, however misguided) to act in particular ways. We humans are psychologically complex creatures, and we often enough have competing motivations that point toward different courses of action. These motivations are causal as well as rational influences, and so they make us more or less likely to choose one way rather than the other. Even so, we must choose. In choosing one way (and in experiencing the consequences), we influence incrementally the probabilities pertaining to future choices; over time, we typically come to have fairly settled dispositions concerning certain types of goals and outcomes.
I take the previous paragraph to be consistent with a reflective common sense perspective. Contra Haselager, I think it is readily understandable from a broadly scientific perspective, although it requires abandoning the ontologically reductionist straitjacket that he builds into ‘scientific materialism’ and seeing greater complexity in the relationship of the differently leveled structures studied by our successful sciences. First, it should be noted that probabilistic causal influence is now a firmly-established part of science. Sometimes, of course, this merely reflects relative frequencies of common variables over ensembles of processes that are similar in coarse-grained ways but whose fine-grained conditions on the underlying driving mechanisms differ. But we can also understand probabilities as reflecting stochastic propensities—individual-case, weighted tendencies of causal agents toward possible outcomes. Certain quantum phenomena such as radioactive decay are naturally understood in this way. Whatever the facts turn out to be, we can certainly conceive choice in a similar way. Haselager says that if actual, this would be mere chance and so of no help in understanding how we might control our actions in a way that grounds moral responsibility. However, we have to ask what ‘chance’ means, something that he nowhere defines. I suggest that our ordinary notion of a chance occurrence is of one that is causally undetermined and not intentionally controlled. By that definition, a motivated forming of an intention to act is hardly something that comes about by chance. Perhaps Haselager is supposing that chance occurrence may be defined more simply as anything that is causally undetermined for the reason that control over an outcome is necessarily exercised antecedently. So any causal conditions that fail to determine an outcome thereby fail to (fully) control it. But choice is not a mere outcome; it is an exercise of control. It is something we do, not something that happens to us. Or, again, so it naturally seems. So when Haselager asks, prior to the consideration of scientific evidence, what could have done otherwise ‘means’, and why it might be valuable, the common sense answer is quite simple: it means that we exercised a power to settle which course of action we would follow among a plurality of alternatives, and it is valuable because it constitutes a form of limited autonomy. We are not merely complex and changing conduits of causal influences from the past, flexible decisionmakers over time, but always pre-determined in how we decide on any occasion. We are fundamental difference makers—in the necessarily constrained manner of conditioned creatures rather than in godlike, ex nihilo fashion—yet still freely and creatively responding at a time.
Let us turn from Haselager’s conceptual misgivings to the more empirical question of whether the totality of present evidence from a range of sciences supports a reductionist conception of physical reality as wholly fixed from below by basic and impersonal physical forces. On this vision, all the structures and activity of chemistry, biology, neuroscience and psychology, sociology, and ecology are but the outworkings of a 2 lst-century version of Democritus’s atoms in the void. The descriptive concepts and principles of these restricted-domain sciences are not derivable from physical theory, Haslager allows, but the structural and dynamical realities they describe are wholly derivative: they are as they are in all respects because the basic physical realities are as they are. Physics is a complete and autonomous domain, determining but not determined by the structured realities. There is much to say about this reductionist vision, but here I can only comment briefly. This vision dominated scientific thinking in the first part of the 20th century between the breakthroughs of quantum mechanical explanations for chemical structure and the uncovering of the biochemical basis of life. However, developments in the second half of the last century strongly challenged it, as the power of complementary, top-down powers of organized structures to constrain and channel the behavior of their elements became clear in a variety of domains. Robert Laughlin (solid-state physics), Denis Noble (systems biology), and George Ellis (cosmology) are among the prominent voices articulating an alternative vision on which complex systems of many kinds exhibit reciprocal and inter-dependent bottom-up and top-down behavior in their evolving trajectories over time.1 Although neither vision is conclusively demonstrable, and (given measurement and computational constraints) likely never will be, the ‘emergentist’ vision is arguably a more natural fit with our evidential reality: profound successes in fundamental physics in describing artificially isolated interactions of small numbers of particles; equal success in delineating and confirming top-down explanations in many complex domains; and little prospect of ever being able to directly apply quantum theory to the entirety of large-scale composed systems involving billions or trillions of interacting particles. Certainly the emergentist vision is a more natural fit with available evidence if one gives any significant credence to our common sense perspective. Our encounter with the world begins with our conscious experiences, thoughts, emotions, and other intentional states. All of these present themselves to us as having distinctive intrinsic character that is quite unlike the character of our nonconscious environment and as continuously affecting our behavior and other, subsequent conscious states. To take them at face value is to implicitly endorse an ontologically emergentist understanding of their place alongside the subconscious neural machinery that sustains and modifies them in turn. To endorse, alternatively, a reductionist thesis regarding conscious states is to implicitly endorse a kind of illusionism, according to which our pervasive impression of directly apprehending the intrinsic character of conscious states is simply an illusion for future cognitive science to explain. (On such a view, conscious states are neural, which fact is not given to us in conscious experience. Our awareness of such states and ability reliably to compare them with previous such states is wholly constituted by subconscious mechanisms, as is the pervasive illusion of our grasping a distinctive qualitative and subjective character.)
A serious problem for the scientist who would endorse the reductionist’s rejection of our common sense perspective on conscious experience and agency is that this perspective is quite fundamental to the epistemology of science itself. Scientific theories, models, and results are themselves the products of scientific activity: of human persons acting in certain coordinated, purposive ways and communicating their activities and results to one another. The nature of this activity is not part of the content of such theories as the theories are not (in most cases) about such activity. However, we who embrace the theories and conclusions drawn in any scientific domain must tacitly accept that the theorists and experimenters were guided by purposes known to themselves and successfully carried out those purposes by intending a variety of specific behaviors thought to advance those purposes, where again they were aware of those intentions. Absent reasonable belief in their reliably self-aware agency, our confidence in what took place and what it resulted in would be undermined: we would have words on a page or a screen, but we would not be in a position to assess what they were indicative of. Or, this would be our predicament absent some alternative account that insured the reliability of the products, despite the unwitting behavior of the scientists yielding them. No such alternative account readily suggests itself, and any candidate that might be presented for our consideration would require us to trust the one providing it, raising the same problem all over again. Haselager seems dimly aware of the problem here when he likens the revisionist perspective that he envisions cognitive neuroscience as possibly producing to “a snake biting its own tail” (Haselager, Chapter 7, 2). But it would more accurately be characterized as the paradox of a snake swallowing its own tail. Or, to switch the metaphor, a scientist who comes to reject our fundamental self-understanding as agents who are regularly consciously aware of efficaciously intending actions is sawing off the branch on which he sits.
Now, affirming that we form conscious intentions that are efficacious in initiating and guiding action is not to affirm that we ever do so freely. But the sciences of brain and behavior do not (yet) pose serious challenges to our freedom. At any rate, no such challenge endures from the one set of studies that Haselager discusses, viz, those pioneered by Benjamin Libet and refined by many others on the timing of conscious intention or will. Haselager appears to agree up to a point. However, it is unclear how widely he would apply the conjecture of Schurger, Sitt, and Dehaene that the timing of action initiation may result from the passing of a threshold in random neural activity.2 This conjecture has some intuitive as well as experimental plausibility in Libet-style experimental scenarios where agents are asked to ‘spontaneously’ make arbitrary decisions, but no good basis has been given to generalize this to the seemingly very different, nonarbitrary, and deliberate choices that are better exemplars of our common sense understanding of free will.
Haselager closes his discussion by reciting in compact form the familiar scientific ‘creation myth’ that is the credo of many would-be deflaters of the human pretension to which prescientific people are supposedly naturally inclined:
The scientific picture, incomplete as it may be, indicates that we are bipedal great apes, possessing linguistic skills often used for confabulating societallv acceptable reasons for actions that are generally driven by subconscious processes, on a small planet circling around an insignificant sun in a remote area of the universe. The revisions of intentions and free will examined in this chapter may undermine other aspects of a cherished human self-perspective.
(Haselager, Chapter 7, 9)
What is true within this myth is that an important aspect of the advance of scientific understanding, especially in physics, has come from learning to prescind our view of the cosmos from our perspectival limitations with respect to time, space, and history. However, the resulting advance in knowledge concerning the cosmos and its (and our) history does not overturn our basic self-understanding or speak one way or another concerning our fundamental significance. An alternative framing of the story of the appearance of human beings on this planer that is at least as consonant with the established facts and well-confirmed theories in 2020 as Haselager’s is this:
Sundry basic features at the origin of our cosmos were exquisitely fine- tuned for the building blocks necessary for the subsequent appearance of life on our planet and its slow evolution toward greater complexity. Along one branch of that fecund biological bush there emerged a variety of complex creatures unified by the striking new feature of conscious awareness and culminating in ourselves: purposive, flexibly intelligent, self- and other-aware creatures. Our nature contains strata of our many precursors, such that we are curious hybrids governed by a patchwork of mammalian machinery and drives interacting with our distinctively personal capacities and attitudes. Curious but also wondrous, we are planet- bound, bipedal great apes by heritage yet, astonishingly, also capable of contemplating, probing, and coming to understand the entirety of the vast cosmos whose elemental traces our bodies bear—and capable, too, of contemplating and debating the possible purposes that may govern our remarkable species.3
- 1. See Laughlin (2005), Noble (2006), and Ellis (2016), as well as the multiau- thored volume edited by Murphy, Ellis, and O’Connor (2009), and the special journal issue edited by Ellis, Noble, and O’Connor (2012). For systematic analysis of the concept of emergence and its possible scientific applications, see O’Connor (2020).
- 2. Schurger, Sitt, and Dehaene (2012).
- 3. This publication was made possible through the support of a joint grant from the John Templeton Foundation and the Fetzer Institute. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of the John Templeton Foundation or the Fetzer Institute.
Ellis, George F. R. 2016. How Can Physics Underlie the Mind? Top-Down Causation in the Human Context. Berlin: Springer-Verlag.
Ellis, George F. R., Denis Noble, and Timothy O’Connor, eds. 2012. “Top-Down Causation: An Integrating Theme within and across the Sciences?” Themed Issue of Interface Focus: A Journal of the Royal Society 2 (February): 1—140.
Faughlin, Robert. 2005. A Different Universe: Reinventing Physics from the Bottom Down. New York: Basic Books.
Murphy, Nancey, George Ellis, and Timothy O’Connor, eds. 2009. Downward Causation and the Neurobiology of Free Will. New York: Springer Publications.
Noble, Denis. 2006. The Music of Life. Oxford: Oxford University Press.
O’Connor, T. 2020. “Emergent Properties.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta.
Schurger, Aaron, Jacobo Sitt, and Stanislas Dehaene. 2012. “An Accumulator Model for Spontaneous Neural Activity Prior to Self-Initiated Movement.” Proceedings of the National Academy of Sciences 109 (42): E2904-2913. https://doi: 10.1073/pnas. 1210467109.
9 Psychological Challenges
to Common Sense Philosophy