ABSTRACT
Many existing accounts of agent autonomy hold that individuals exercise self-determination when they act upon a will that is their own, upon desires that they have endorsed. The notion that autonomous action flows from desires that an agent has identified with or embraced via higher-order reflection is intuitively very compelling. However, three potential objections that face such accounts are the “arbitrariness problem,” the “regress problem,” and the “manipulation problem.” These critical worries question what makes acts of identification someone’s own, rather than being imposed from the outside or the expression of an adaptive preference. Although human action is continually influenced by social factors, surely this does not prevent human agency from ever being autonomous. An adequate account needs to acknowledge how autonomous agency is fundamentally relational and also distinguish between coercive and non-coercive social influences. In addition, it should emphasize the role played by a wide range of agentic skills; these include not just self-reflective deliberation, but also bodily experience, emotion, imagination, and perspective-taking. An adequate account also should acknowledge that autonomous agency is diachronic, unfolds over time, and comes in degrees. Lastly, this account should be grounded in naturalistic considerations and engage directly with philosophy of mind.
