Introduction
© David Weissman, CC BY 4.0 https://doi.org/10.11647/OBP.0197.06
Agency implies purpose, action, and autonomy. We are inventive, effective, and self-appraising. But there is an ambiguity. Does agency signify reality-testing in the near-world where the existence and character of other people and things are independent of our thinking? Or is agency the activity of a mind inspecting sensory data, ideas, and itself? These are alternative foundational claims: one material and existential; the other subjective and epistemological. Aristotle described agency in detail, but his formulation is obscured by the Cartesian tradition that elides thought with being. It supposes that we know reality when mind turns on itself, though agency is apparent to anyone taking a walk or calling a friend.
The difference between these views is suppressed rather than resolved when reality-testing is construed as the mere look of the data inspected: we only seem to be agents afoot in a world we haven’t made. These are sample passages from Heidegger’s Being and Time:
[S]ciences have the manner of Being which this entity—man himself—possesses. This entity we denote by the term “Dasein.”1
Ontology and phenomenology are not two distinct philosophical disciplines among others. These terms characterize philosophy itself with regard to its object and its way of treating that object. Philosophy is universal phenomenological ontology, and takes its departure from the hermeneutic of Dasein, which, as an analytic of existence, has made fast the guiding-line for all philosophical inquiry at the point where it arises and to which it returns.2
Heidegger is rightly credited with bringing Hegel to ground—universals cashed out in particularity—yet his ontology is speculative in a traditional way. For Dasein is an intuiting ego: everything real because intelligible is located in the minds discerning it. This is phenomenology as it claims transcendental competence: minds discern reality’s essential structure and details by inspecting experience and themselves.
Much of philosophy is comfortable with this bias and its deference to Plato’s cave allegory: he alleged that minds can know the ordering principles of the cosmos because the quest for reality’s Forms is intuition’s trajectory.3 Descartes inverted Plato’s focus: distilling minds are self-discovering. Deflecting the skepticism of his first Meditation, Descartes discovered the theater where everything is set before our inspecting minds, all of it informed by mind’s clear and distinct ideas. Nor is anything else thinkable. For what can we know or imagine when nothing is either conceivable or existent if not inspectable? Esse est percipi was Descartes’ idea before Berkeley used the phrase. It embodies two claims: that evidence of existence is a necessary condition for existence, and that evidence isn’t secured if a mind isn’t aware of itself (thereby confirming its own existence) while inspecting things perceived. For “I am, I exist, that is certain. But how often? Just when I think,” 4 implying both awareness of something and mind’s awareness of its own existence while inspecting that thing.
Put aside Descartes’ gesture when he introduced a god as the deus ex machina justifying moral certainty that the external world is largely as it seems. For all states of affairs lying beyond the arc of consciousness earn Kant’s designation as “negative noumena.”5 He reduced agency to the dimensions—the single souls—of Leibnizian monads;6 he fortified the self-inspecting cogito, re-describing it as the agent responsible for schematizing experience in ways that satisfy the maker’s values and interests. But this is odd: purpose succeeds because we’re organized, stubborn, and clever, not because we create surrogate realities while staring at ourselves.
Why make these antique allusions? Because the subjectivist tradition is second nature to thinkers fearful that invoking agency’s stance in the natural world would make them seem naive. Aristotle was confident that thinking, making, and doing engage human agents with other people and things. His assumptions—ontological realism, truth as correspondence, and a representational theory of perception—are known but discounted by skeptics who defend Cartesian subjectivism and Kantian idealism. Principal ideas of knowledge, perception, and agency fail to assimilate either his realism or the Darwinian implications of minds that evolve while accommodating their material circumstances.
Our subjectivist tradition construes us as passengers on a train: seeing the countryside through the windows of the observation car, we imagine saving the land while never leaving the train. What does it matter that philosophy distorts self-understanding by reducing agency to Cartesian self-reflection or Kantian world-making? Do we believe with Quine that “the quality of myth is relative,” so reality has whatever character satisfies our “various interests and purposes”?7 This is fanciful; it doesn’t survive an accurate reading of agency, frustration, or error. These chapters face the other way: they mate Aristotle with Darwin, Peirce, and Freud. I assume that every living creature is situated in the material world, and that each has multiple effects on things or events that are otherwise independent of human agents and one another. Agency’s profile emerges as we observe others and ourselves, then augment observation with inferences that identify the likely intra-psychic, material, and social causes for our actions and effects. We rebalance this schism—action or inspection—by controlling action with appraisal. Storms have effects they don’t control. We alter our effects by revising our aims or their means.
Yet agency is elusive. This is true superficially because we often qualify subject terms with active verbs—“She tripped”—though it was a loose rug, nothing careless, that caused her fall. Donald Davidson made this point, though there are, he agreed, many actions rightly attributed to the agents causing them.8 The more abiding obscurity is mental and material. Agency is too close and self-entangled for easy comprehension: too many of its effects aren’t perceived, inferred, or foreseen. It once seemed that mystery dissolves because autonomy and free will are apparent; one could reliably say that mind is whatever self-inspection shows it to be. But nothing has changed more radically than this self-conception. Having analogized machine intelligence to human ways of thinking, we infer that consciousness and its qualifications are emergent effects of a body’s nervous system. We fear that autonomy may fall to socialization and machine management: we sing any tune we please, though the tunes we know are usually familiar to our neighbors. For almost everything we do comes from an inventory accessible to all a society’s members. Tastes change, choices alter, though changes are usually those we accept, not those we provoke. Autonomy is restricted to small initiatives—coffee or tea, white shirt or blue—in contexts requiring choices that individuals don’t control.
Successful practice is evidence that our brains effectively process information about the ambient world and our responses to it, but there is no direct perception of brain’s complexity and no comprehensive understanding of the gap between our powers of self-inspection and the neuro-mechanical processes of self-regulation. What’s to be made of aims pursued steadily from provocation and desire through planning, initiative, frustration, and revision? Some thinkers tease subtlety out of introspective data; others reduce intention to the persistence of our behavior. Or, like Ryle, Wittgenstein, and Anscombe, they resort to a different sort of data. Ignoring the evidence of self-inspection, they parse the grammar of the words used when speaking of one’s intentions. Yet different uses and different ways of construing them require verification. How is that achieved? We test alternate linguistic implications against experience and behavior, while supposing that choice and action are impelled by intentions and volitions known several ways: by self-inspection, by observing other people’s behavior, and—in a preliminary way—by brain scans that trace their neural course. Grammar is suggestive, but never more than provocative because its rules lack the diversity and nuance of aims that are variously pursued and expressed. Nor is this surprising: anger and fear are persistent motivators; they are often disguised or suppressed, though never reduced to the grammar of anger or fear.
This essay embodies a tradition that runs from Jonathan Edwards and Ralph Waldo Emerson to C. S. Peirce, William James, and John Dewey. Many agents—storms and trucks, for example—are impelled by other things; humans (and some other animals) are self-moving. Philosophy’s American roots are dominated by ideas of agency, initiative, and self-control. Regulation wasn’t urgent on the frontier, where people had few neighbors; it became an issue as America filled with immigrants. How would an open society constrain its citizens while liberating them from rigid practices and traditions? Seek opportunities, take risks, though effects may be consequential, so deliberate about aims and means. Purpose is often steady, but don’t expect that it will always be present, like blinking lights, to inspecting minds. Is my trajectory fixed when driving in fog and rain? Wouldn’t that detract from the care required to do it safely?
This tradition emphasizes autonomy, while acknowledging that freedom carries responsibility for controlling one’s actions and appraising one’s effects. Hard determinism objects that this self-critical response is illusory: it says that every current action is a moment in a trajectory initiated at time’s origin, an arc we don’t control. Chapter Two argues that hard determinism is false: many things we do are habitual; many habits have a lineage that includes DNA and its evolutionary survival in circumstances long superseded. Yet agents—material though we be—often address situations for which history has prepared no specific response.9 Middle-aged, male, and childless, I hold the infant left with me while her mother runs an errand. The baby pauses, considers, then twists and screams; she isn’t happy. I hold her tighter, then loosely; nothing works until I blow across her ear. The screaming stops; she relaxes in my arms. Where was this in my history? Where is hard determinism in any situation where an agent has choices but no established responses to circumstances that are novel or surprising? Evolution is deterministic: it explains the generation of speech and cooperation, though not their expression or effects in circumstances for which we have no preparation. For every situation invites free choice because uncertainty and initiative trump the causal tide: canny and skilled, we respond to situations for which there is no history.
Validating free will—choice when history isn’t determining—is critical because moral responsibility requires it, and because having it is a condition for moral identity. Rather than being programmed for action, I declare myself to others by appraising and regulating my effects on them. This isn’t news: we don’t convince others or ourselves that we avoid responsibility because we locate our choices within an ancient lineage. Arguments for or against free will won’t change the fact that we do or don’t have it; but they are consequential because of their real or potential effects on our attitudes. Hard determinism is demoralizing: don’t imagine that any difference you make is one you’ve freely chosen, given that each of us is a vehicle for effects long ago determined. There must be compelling logic and evidence before we concede that our apparent control of judgment, choice, and behavior is an illusion.
These issues have a contentious history. My reading of Aristotle locates him within the pragmatic naturalist tradition because of his belief that mind is the activity of a body having a certain complexity, and because he invariably emphasized self-regulation and one’s responsibility for action’s effects.10 Descartes is his contrary: mind is self-sufficient; it connects to body by way of the pineal gland, though ablating the gland wouldn’t stop mind from making choices as it thinks, remembers, or imagines. Yet mind’s choices wouldn’t have effects beyond itself, given Descartes’ reasons for doubting that there is an ambient world. Should we settle, as with Kant’s “transcendental object,”11 for an “as if” reality, one having no consequence beyond our inclination to assume it? Freedom to choose when thinking is, he said, all the freedom we have, but is that enough freedom to conduct the business of everyday life?
Subjectivism collapses because no one has identified these putative agents—Descartes’ appraising cogito or Kant’s schematizing ego—and because material systems are able to perform as dualism alleges they cannot. Consciousness and its qualitative data resist physiological explanation, though this obstacle is somewhat reduced by engineers who build systems acting in ways that would be described as conscious if performed by minds. Challenge passes to the Aristotelian side: what survives of choice when all its conditions are the capacities of material systems? What are autonomy, choice, and moral identity in the context of the body and brain? Answers seem apparent when observing people who go about their ordinary tasks: weighing likely outcomes, they estimate costs and benefits before deciding what to do. Some agents identify with a role having specific tasks and duties; they perform as the role prescribes. Others innovate. Either may interrupt standard or experimental practice because of regretted effects; we plausibly say that both could have chosen to act differently. Determinists agree that self-inspection seems to confirm these powers, but nature, they say, is a closed book. We aren’t autonomous, and don’t have free will; history is the unspooling of causes and effects whose trajectory was decided at the start; whenever, whatever its character. Is this judgment supported by empirical and logical evidence or is it an a priori dogma? These chapters are an appraisal: I conclude that hard determinism is multiply flawed with very little supporting argument or evidence.
Why does it matter that this version of the determinist argument is false? Because there are critics so skeptical of libertarian excesses that they gloat when saying that freedom is a delusion. Their posture squares comfortably with discoveries of several kinds: that mind is a physical system performing in fundamental ways as computers do; and that thought, perception, and memory are material processes. Yet biology doesn’t confirm that every contemporary choice is the current expression of a trajectory started eons ago. Someone believing that reasons make no difference when making hard choices would throw up his hands or use a Ouija board when deciding: let history choose. For reasons that seem to justify a choice have a cosmetic but no causal effect if all was determined at the beginning of time. I propose an alternative that eschews both unconditioned freedom and the determinism which explains that every event—past or future—was decided by natural laws and the conditions prevailing at time’s inception. Soft determinism is the hypothesis that every event has causes sufficient to determine its character and existence, many that are past but some that express the interests or values of agents responding to current circumstances. For it often happens that actions are provoked by situations having no antecedent in an agent’s history. Laplace assumed that causal energy is continuously transformed when sustained in accord with laws that have operated since the beginning of time. He would have us explain an agent’s response by citing those laws and original conditions. I argue that this explanation is incomplete: it ignores the myriad times when causes are insufficient to provoke an effect until completed by the responses of agents engaged by provocative situations.
Are there considerations—evidence and arguments—that defeat hard determinism? My solution has four elements:
i. Agreeing that every property and event has causes sufficient to determine its existence and character, it affirms that choices underdetermined by their antecedents alter the trajectories of causal histories. This happens when choice or action ensues because one or more features in a current situation provoke an unprogrammed response.
ii. We have this power because of three factors: a. Agents in one causal lineage often address new or surprising situations having constituents in lineages independent of their own. Having no fixed reaction (or overriding a fixed response), agents experiment; they test and revise their actions until effects satisfy their needs. b. Emergence has sufficient causal conditions (life emerges from assembled molecules and cells), but its effects are liberating. No longer restricted to the powers of their lower-order constituents (organs, cells, and molecules), agents have a repertoire of powers and skills that redirect the causal trajectories of both themselves and the things with which they interact. Or the causal tide is interrupted when inherited responses are inhibited. c. Will is two things: life-force, and a power that energizes and enacts particular aims. Both emerge when evolution produces living things shielded from their environments by permeable surfaces (cell walls, for example) that resist intrusions while supplying access to nutrients and partners. Every such agent is an evolutionary experiment: how will it adapt to circumstances when its external buffer enables the development of capacities for using circumstances to its advantage? Millenia pass as these functions evolve, and others emerge. In humans, emergent powers include deliberation, judgment, will, and choice. Is a situation new or surprising? Free will is the activation of powers that evolution has enabled: powers for reality-testing, coping, and restraint. Creatures less endowed, react; we consider our options. The full history of will’s causal lineage goes back very far: all emergents do. Its activation as free will is the effect of current situations: responses are inhibited until circumstances are appraised, options are considered, choices are made: I see and like it, but don’t buy it.
Calling will “free” doesn’t imply that the act is unconditioned; it signifies that willing is free from external constraints, when its sufficient conditions are local to the psychic space where decisions are made. An observing god might predict will’s evolution with near or perfect accuracy; its guesses about agents’ responses to particular states of affairs—addressing situations for which there are no prepared responses—would be no better than likely.
iii. Agency is an essential topic in the thinking of pragmatists for whom thoughtful action is the key to understanding our relations to one another and our circumstances. Yet pragmatism too often reduces to the urgencies of practical life or the values and planning that make action effective. We emphasize actions in the public world; we neglect interiority, though one is altered by a book, a play, or the horror of an accident. Active in the near-world of things and opportunities, active in relations to one another, we too often ignore the resonance of taste, skill, and deliberation. Their cultivation is decisive for aesthetic and moral judgment. Autonomy is incomplete if sensibility is unformed.
iv. Equilibrium implies a vague but useful measure of agency’s principal aims. Signifying moral, aesthetic, and emotional balance in the relations of agents to those they engage and within agents themselves, it expresses a psychic and moral ideal: in control of ourselves, we are also responsible for our effects on other people and things. Many factors explain the want of balance, but philosophic ideas—hard determinism, for example—are a symptom: why learn to modulate our responses if all was settled long ago? We reduce the severity of this effect by dispelling several persuasions. They include the bellicose reading of Nietzsche’s persuasion that power and action are our raison d’être; Descartes’ belief that intellect is self-sufficient; the Hegelian, Heideggerian aversion to privacy, and Plato’s disdain for emotion. We repair these excesses by binding the extremes. Women were once excused from being publicly effective, men from educating sensibility. Equilibrium implies the convergence of these contraries. Like health, it signifies a condition to which we aspire, one whose absence may explain both our displacements (money, status, and possessions) and our addictions (alcohol and dope). Agency is these two things: control of one’s circumstances; cultivation, appraisal, and control of oneself.
Free will is conceptually enabling: it clarifies the requirements for socialization, autonomy, and moral identity. Yet autonomy is problematic: a factional or atomized society is perpetually barren and thwarted. Its individuals can’t organize for common or complementary aims; they lack the milieu required if art or ideas are to sensitize taste and talents. This essay construes socialization as the process of folding autonomous agents into the fabric of meanings, roles, and rules responsible for order, productivity, and mutual understanding. The conditions for free will—self-sufficiency, resistance to intrusions, and inhibition—make us singular, but vulnerable because inaccessible or unintelligible to others. We risk being unable to sustain ourselves because unable to acquire habits and sensibilities that would make us interesting or useful to other people. Separate but equal is a political problem; separate but accessible and mutually responsive is a psychological and moral virtue.
Moral identity is the achievement of socialized autonomies. We participate in an earthly version of Kant’s kingdom of ends while facing two ways: enjoying our separate talents, tastes, and opportunities, we are also responsible for duties to other people, and to the systems—families, friendships, businesses, and states—in which we participate. Autonomy is crystallized discipline; it makes us resourceful and reliable. The chapters that follow elaborate these themes by addressing agency’s constitutive variables: free will, socialization, autonomy, and moral identity. They argue that action is constrained by circumstances, rules, roles, and affinities, but free within those limits. The degree of one’s freedom varies: slaves have little or none; the wealthy residents of Paris or New York have a lot. But everyone has projects, beliefs, and inclinations. Agency implies autonomy; but autonomy wants power, opportunity, partners, and a voice.
1 Martin Heidegger, Being and Time, trans. John Macquarrie and Edward Robinson (New York: Harper and Row, 1962), p. 32.
2 Ibid., p. 62.
3 Plato, Republic, Collected Dialogues, eds. Edith Hamilton and Huntington (New York: Pantheon, 1961), 514a-517e, pp. 747–49.
4 René Descartes, Meditations on First Philosophy, in: Discourse on the Method and Meditations on First Philosophy, ed. David Weissman (New Haven: Yale University Press, 1996), p. 65.
5 Immanuel Kant, Critique of Pure Reason, trans. Norman Kemp Smith (New York: St. Martin’s, 1965), p. 270.
6 G. W. V. Leibniz, Monadology and Other Philosophical Essays, trans. Paul Schrecker and Anne Martin Schrecker (Indianapolis: Bobbs-Merrill, 1965), paras. 1–19, pp. 148–51.
7 Willard Van Orman Quine, From a Logical Point of View (Cambridge: Harvard University Press, 1980), p. 19.
8 Donald Davidson, “Agency,” in Essays on Actions and Events (Oxford: Clarendon Press, 2002), pp. 43–62.
9 An earlier version of Chapter Two was published as “Autonomy and Free Will,” Metaphilosophy 49 (2018), pp. 609–45.
10 Aristotle, The Basic Works of Aristotle, ed. Richard McKeon (New York: Random House, 1941), 431a1–435a10, pp. 593–602.
11 Kant, Critique of Pure Reason, p. 137.