In the article "Normativity, deliberation, and Queerness," (in A World Without Values: Essays on John Mackie's Moral Error Theory, (Richard Joyce and Simon Kirchin, eds.), 2010), David Copp considered and raised objections to five arguments for the Authoritative Reasons Proposal.
The Authoritative Reasons Proposal says this:
Authoritative Reasons Proposal: Necessarily, if a person has a moral obligation to do something, then (a) there is a moral reason for her to do it, and (b) if she is fully rational and believes she is obligated to do it or that there is a moral reason for her to do it, she takes this obligation or reason appropriately into account in deciding what to do.David Copp rejects the Authoritative Reasons Proposal. In doing so, he looked at five arguments that might be offered in its defense, and raised objections to each of them.
One of those arguments was the Argument from "Ought" implies "Can".
The first premise is that moral obligations are incumbent on all rational agents.... The second premise is the maxim that “ought implies can”.... An agent who, all things considered, is morally obligated to do something must be able to do it....But then, third, anyone who is subject to a moral obligation must be able to fulfill the obligation simply in virtue of exercising her rational capacities, irrespective of contingent features of her circumstances or of her desires or values and the like.... We can now draw the desired conclusion. For if an agent can comply with the demands of morality simply by exercising her capacity for rational deliberation and action, it must be that the reasons to act in accord with these demands would motivate her to act accordingly insofar as she were thinking rationally.I am going to object to the third premise in this argument. If an agent is under a moral obligation, this does not imply that the agent can perform the action irrespective of her desires or values and the like. This would require some sort of magical force that does not exist in the real world. The capacity to act independent of our desires and values simply does not exist. If morality requires such a capacity, this would not be an argument that shows that capacities of this type are real. This would be an argument that shows that morality itself is a fiction.
However, morality does not require this fictitious capacity. Instead, to say that an agent is under a moral obligation is to say that if the agent had the right desires or values, then the agent could (and would) perform the action. In the language of the compatibilists, "could have done otherwise" implies "would have done otherwise if she had wanted to."
Since an agent cannot, in fact, "comply with the demands of morality simply by exercising her capacity for rational deliberation" it does not follow that "the reasons to act in accord with these demands would motivate her to act accordingly insofar as she were thinking rationally."
It is actually the good desires that would motivate her to act accordingly - the good desires that she would have if the reasons (desires) she had matched up with the reasons (desires) she SHOULD have. The reasons (desires) that people generally have reason to cause her to have.
There is a place for adding in this account that the agent not only "would have acted otherwise if she had wanted to" but also that she "could have wanted to". However, in saying this we are only saying that the reasons under discussion are reasons that can be molded using rewards such as praise and punishments such as condemnation. It is to say that this is one of the ways in which society is capable of molding malleable brains.
David Copp takes these same points in the opposite direction and comes up with a similar problem.
If it were the case that our moral obligations depended on the reasons we had and not the reasons we should have, then it would be very easy to get out of many claims concerning our obligations.
If I am unloving, careless, and lazy, I will not take care of my children. I could not cease to be unloving, careless, and lazy by a simple exercise of my rational capacities. But I certainly cannot invoke the Maxim to argue on this basis that I have no duty to feed my children.A person who must "change psychologically in ways he could not bring about simply by exercising his rational capacities" (to exchange the reasons has with for reasons he should have) cannot escape morality by this fact alone.
In fact, there is something quite dangerous in the idea that our moral requirements are determined by what it is rational for us to do. It does make it easy to escape moral obligations whenever the case that what it is rational for a person to do is to act in ways that are harmful to others. If it is rational for an agent to kill a person he just raped (to prevent her from talking), and a person's moral obligations are limited to what it is rational for a person to do, then morality must command the agent to kill the person he just raped.
The way around this is to deny that morality has anything to do with what it is actually rational for an agent to do at any given moment, but rather with what it would be rational for the agent to do if he had the reasons to act that he should have (that people generally have reasons to cause him to have).