Loading field map
When you raise your arm, what makes that movement an action rather than a mere bodily event? A twitch is something that happens to you; a wave is something you do. The philosophy of action asks what this difference consists in, and the answer has proved surprisingly elusive. Is an action an event caused by a mental state—a desire, a belief, a volition? Or is it something that can only be understood by grasping the agent's reasons, which may not be reducible to causes at all? Does agency require a self that is more than a bundle of mental events? And where does the mind that initiates action end—at the skull, or in the tools and environment we use? These questions have generated a rich landscape of competing frameworks, each offering a different picture of what it is to act.
The first systematic frameworks of the twentieth century approached action from very different directions. Social Action Theory, developed by Max Weber in the 1920s, treated action as behavior to which the agent attaches subjective meaning—meaning that is oriented toward the behavior of others. Weber's framework was sociological: it aimed to make action intelligible by placing it within a web of shared norms, values, and expectations. Action, on this view, is not a private mental event but a public, socially embedded phenomenon.
Around the same time, a very different tradition was taking shape in continental philosophy. Phenomenology of Action, rooted in the work of Maurice Merleau-Ponty in the 1940s, focused on the lived, bodily experience of acting. Merleau-Ponty argued that much of our action is not preceded by a conscious decision or a mental representation. When a skilled pianist plays a phrase, her fingers move not because she forms an intention and then executes it, but because her body already 'knows' how to respond to the music. This pre-reflective, embodied skill is the primary mode of agency, on this view. Where Weber looked outward to social meaning, the phenomenologists looked inward—but not to a detached mind; they looked to the body as the subject of action.
In the 1960s, analytic philosophy took up the question of action with a new focus on causation. Volitional Theory proposed that an action is a bodily movement caused by a special mental event called a 'volition'—an act of will. This seemed to capture the intuitive idea that actions are willed, but critics objected that volitions themselves looked suspiciously like little actions, threatening an infinite regress.
Donald Davidson's Causal Theory of Action, introduced in his landmark 1963 paper 'Actions, Reasons, and Causes', offered a more parsimonious alternative. Davidson argued that a reason for acting—a combination of a belief and a desire—is itself a cause of the action. When you wave to a friend, your desire to greet her and your belief that waving will do so cause your arm to rise. The action is an event caused by mental states, and the causal relation is what distinguishes it from a mere bodily movement. This 'event-causal' model became the dominant framework in analytic philosophy for decades.
But the Causal Theory faced a persistent challenge: if an action is just an event caused by prior mental events, where does the agent go? The Agent-Causal Theory, developed by Roderick Chisholm and others in the 1970s, insisted that the cause of an action must be the agent herself, not merely a mental event within her. On this view, when you wave, the cause is you—a substance, not an event. Agent-causation is irreducible to event-causation; it is a distinctive kind of causation that only persons can exercise. This framework preserves the intuition that agents are genuine originators of their actions, but critics have worried that it introduces a mysterious, non-natural form of causation.
The disagreement between the Causal Theory and Agent-Causal Theory is not just about what causes action; it is about what kind of metaphysics agency requires. The Causal Theory fits comfortably within a naturalistic worldview where all events are caused by prior events. The Agent-Causal Theory insists that agency requires a break in that chain—a self that is not just another link.
A related debate concerns the nature of the reasons that explain action. Internalism about Reasons, championed by Bernard Williams in the late 1970s, holds that a reason for action must be capable of motivating the agent, given her existing motivational set—her desires, commitments, and values. If you have no desire to learn chess, the fact that chess is intellectually rewarding is not a reason for you to play. Reasons are relative to an agent's psychology.
Externalism about Reasons, defended by Thomas Nagel and others, rejects this constraint. Externalists argue that some reasons apply to agents regardless of their desires. The fact that an action would be morally required, or objectively good, gives the agent a reason to perform it, even if she does not care about morality or goodness. For externalists, reasons are facts about the world, not merely facts about the agent's psychology.
This debate connects directly to the earlier causal theories. Davidson's Causal Theory, which identifies reasons with belief-desire pairs, naturally aligns with internalism: a reason must be a mental state that can cause action. But externalism challenges this picture by insisting that some reasons are not internal to the agent's psychology at all. How could such a reason cause action? Externalists must either offer a different account of action explanation or argue that reasons can motivate without being causes.
Not everyone accepted that action explanation must be causal. Two frameworks from the 1980s offered alternatives.
Interpretivist Theory of Action Explanation, associated with Daniel Dennett, holds that we explain actions by adopting an 'intentional stance'—treating the agent as a rational being with beliefs and desires. Explanation is a pragmatic, interpretive activity, not a causal hypothesis. When we say someone opened the window because she wanted fresh air, we are not identifying a cause; we are making her behavior intelligible by fitting it into a rational pattern. Interpretivism is stance-based: it works when it helps us predict and understand behavior, regardless of whether the agent really has internal mental states with causal powers.
Teleological Theory of Action Explanation, by contrast, explains actions by their goals or purposes, without reducing those goals to efficient causes. On this view, the reason an agent acts is the end toward which the action is directed, and this end-directedness is not the same as being caused by a prior mental state. Teleological explanation is non-causal: it makes action intelligible by showing what it aims at, not what produced it.
Interpretivism and the Teleological Theory share a skepticism about the causal model, but they differ in their positive proposals. Interpretivism is pragmatic and stance-based: it is about how we interpret agents. The Teleological Theory is metaphysical: it claims that action is in fact goal-directed in a way that resists causal reduction. Both remain minority positions, but they keep alive the question of whether the causal paradigm has a monopoly on action explanation.
The most recent major framework, Extended Mind and Embodied Cognition, emerged in the late 1990s and has reshaped the philosophy of action. Drawing on insights from cognitive science, this framework argues that mental processes—including the processes that generate action—are not confined to the brain or even the body. When you use a notebook to remember a phone number, the notebook is part of your memory system. When a skilled carpenter uses a hammer, the hammer is part of her action system. The boundaries of the mind are not fixed by the skull.
This framework revives and extends the Phenomenology of Action's emphasis on embodied skill and the environment. Merleau-Ponty's insight that the body 'knows' how to act without explicit representation is now supported by empirical work on motor control and ecological psychology. But Extended Mind goes further: it claims that the tools and artifacts we use can become literal parts of our cognitive and agential systems, not just instruments of a pre-existing mind.
This expansion challenges the internalist assumptions of the Causal Theory and Internalism about Reasons. If the mind extends into the environment, then the causes of action may include features of the world that are not inside the agent's head. The debate over where agency begins and ends is now a live empirical and philosophical question.
Today, no single framework commands universal assent. The Causal Theory of Action remains the default position in much of analytic philosophy, but it has been refined to address objections about deviant causal chains and the role of the agent. Agent-Causal Theory persists as a minority view, kept alive by the intuition that event-causation cannot capture genuine agency. Internalism and Externalism about reasons remain in live disagreement, with each side producing sophisticated arguments about motivation, normativity, and practical reason.
What the leading frameworks agree on is that action is not a brute physical event: it involves mentality, normativity, and the agent's perspective. They disagree on whether that mentality is best understood causally, teleologically, or interpretively, and on whether the agent is a substance, a bundle of mental states, or an embodied system extended into the world. The philosophy of action today is a field of pluralism, where different frameworks capture different aspects of agency—causation, meaning, embodiment, and social context—and no single framework has yet absorbed the others.
What unites the field is the conviction that understanding action is essential to understanding persons. We are, above all, the beings who act. Getting clear on what that means remains one of philosophy's most urgent tasks.