Agency

The big idea: Your arm goes up. Two very different stories could be true. You raised it — to answer a question. Or it flew up — someone bumped it, or it twitched. Same movement; completely different meaning. That gap is the idea of agency.

An agent is a being that does things, on purpose, for reasons. When you act, there's a 'why' that's yours: you wanted to, you decided to. When something merely happens to you, there's a cause but no reason of your own.

The heart of agency is a special kind of 'because' — acting because of a reason you hold, not just because of a cause that pushes you.

Two kinds of 'because': Compare:

• 'She flinched because a spark hit her hand.' → a cause pushed her; a reflex, not an action.

• 'She moved her hand because she wanted to catch the ball.' → a reason she holds guided her; that's an action, an exercise of agency.

An agent is a being whose behaviour can be explained by reasons it holds — goals, wants, values — not only by causes acting on it.

Checkpoint — agency: In one line: an agent acts for its own reasons; a mere event has a cause but no reason of yours. Doing ≠ happening.

Agency is the third piece of the personhood picture — and it leans on the first two.

Go further — higher-level insight: A self-driving car 'chooses' to brake — is that agency? Many philosophers say no: it follows programmed causes, but the reasons aren't ITS reasons, held and understood by it. Whether that line really holds becomes the big question when we ask if an AI could be a person. Keep it in reserve.