The Kantian ethical formulation, "Always act according to that maxim whose universality as a law you can at the same time will," seems rather vague. Where, exactly, do we decide how to formulate one's actions?
Suppose a poor man steals medicine he otherwise can't afford, in order to save his child. Which maxim's universality as a law, exactly, should he be willing? If we examine the maxim as being "It is permissible to steal", then we clearly have a universal law that shouldn't be willed, but if we examine the maxim as being "It is permissible to steal in order to save lives, when there are no other options", then we have a maxim that might hold up under more scrutiny.
For any given action, there are multiple possible maxims that are more or less reprehensible, each of which can justify the action in question. So how does Kantian ethics treat situations like this? Do we boil it down to the most semantically/logically simple maxims? Do we go instead for the most nuanced maxims?
That’s an interesting question about how to interpret and apply the universal law test of Kant’s categorical imperative. That test is supposed to apply to our maxims, and our maxims are our subjective ends or goals. So presumably an agent’s maxim is a psychological fact about an agent having to do with her aims. So I don’t think we always choose the most coarse-grained maxim – for instance, stealing – or always the most fine-grained maxim – stealing $10 to feed a blonde six-year old child on a Wednesday because of a lack of other options. It depends on what things the agent -- for instance, the child’s mother – is aiming at and what does guide her actions, and this may depend on the answers to counterfactual questions. Would she steal the money in any case? Would she steal food if this were available? If there were other, morally innocent ways of feeding her child (e.g. going to a soup kitchen or picking berries), would she do so? If the mother would have stolen even if her child was not hungry,...