Many think it is plausible that agents enjoy freedom and responsibility with respect to their actions in virtue of being reasons-responsive. Extant accounts spell out reasons-responsiveness (RR) as a general modal property. The agent is responsive to reasons for and against ϕ-ing, according to this idea, if they ϕ in accordance with the balance of reasons in a suitable proportion of possible situations. This paper argues that freedom and responsibility are not grounded in such modal properties on the basis of a phenomenon I call ‘rational blind spots’. Agents have highly specific local blockages (or openings) that prevent them from seeing or reacting for a particular type of reason. When these blind spots are triggered, agents fail to be responsive to reasons in the sense relevant to freedom and responsibility. Thus, we judge that they are not free and responsible. But bind spots don’t remove the agent’s possession of general modal RR properties. Thus, extant accounts of reasons-responsiveness render the incorrect result that the agents in these cases remain responsible because they remain responsive to reasons in a general sense.

许多人认为，代理人因其对原因的反应能力而对他们的行为享有自由和责任是合理的。现存的帐户将原因响应性（RR）作为一般模态属性。根据这个想法，如果他们在适当比例的可能情况下符合理由的平衡，那么代理会对支持和反对 ϕ-ing 的原因做出响应。本文基于一种我称之为“理性盲点”的现象认为，自由和责任并非基于这种模态属性。代理具有高度特定的局部阻塞（或开口），以防止他们因特定类型的原因而看到或做出反应。当这些盲点被触发时，代理人无法对与自由和责任相关的理由做出反应。因此，我们判断他们不是自由和负责任的。但是绑定点并没有消除智能体对一般模态 RR 属性的占有。因此，现存的对原因响应性的解释给出了错误的结果，即在这些情况下，代理人仍然负责，因为他们仍然对一般意义上的原因做出响应。