Does "Ought" entail "Can"?

[Theftz22]

It is a classic Kantian moral principle that if it is the case that x ought to be done, then it is the case that x can be done.

Intuitively, it seems odd to say that we could have a moral obligation to do something that that cannot be done. But on second look, you might think that moral principles are things that ought to be followed regardless of whether or not you actually can. A principle doesn't cease to be true because it cannot be fulfilled. I don't have any strong inclination either way in terms of the intuitiveness of the Kantian claim.

Can the Kantian claim be grounded in various ethical systems? Ironically, utilitarianism is perhaps the easiest ethical system on which to ground the Kantian claim. Consider the following argument:

1. If utilitarianism is true, x ought to be done if and only if x maximizes utility.
2. If x cannot be done, x does not maximize utility.

Given these two premises, it is derivable that if utilitarianism is true, then if x ought to be done, then x can be done. Here's the logic:

1. If p, then q iff r
2. If ~s, then ~r
3. If p, then (if ~r, then ~q) (from 1 by the def of biconditionals)
4. If p, then (if ~s, then ~q) (2+3 by transitivity)
5. If p, then (if q, then s) (contrapositive of 4)

Thus the final conclusion is, if utilitarianism is truen, then if x ought to be done, x can be done.

Looking at the 2 premises, each is obviously true. Premise 1 is just the definition of utilitarianism. And 2 is obvious because in order for x to maximize utility, x must be actualized, otherwise no maximization is actualized. But if x can't be done, then x can't be actualized. To put the argument more colloquially, an action won't make the world a better place unless it's performed. So if it can't be performed, it can't make the world a better place. But on utilitarianism, an action ought to be done if and only if it makes the world a better place. So on utilitarianism, if an action ought to be done, it must be the place that it can be done, or else it couldn't make the world a better place.

It is ironic that it is easiest to ground the kantian principle on a consequentialism that is radically opposed to the deontology of kant. Another ethical system on which the kantian claim is easily grounded is divine command theory. Here's the argument:

1. If divine command theory is true, x ought to be done if and only if x is commanded by god.
2. If x cannot be done, god would not command x.

The logical form of the derivation is then identical to the previous one, so I will omit it. Again, premise 1 is just the definition of divine command theory. Premise 2 is need of further argument however. One such argument that might be given in favor of premise 2 is that god holds us accountable for not obeying his commands. But it would be evil for god to hold us accountable for not obeying commands that couldn't be obeyed. Thus, given that god is omnibenevolent, it follows that god wouldn't make commands that couldn't be obeyed. Another argument that could be given is that god would have no reason to issue a command that could not be obeyed. God, being omniscient, foreknows for any given command whether it can be obeyed. But then what would be the purpose of issuing such a command, knowing it cannot be obeyed? Thus, given that god acts only with sufficient reason to act, god would not issue commands that could not be obeyed.

I cannot see any other systems on which the kantian claim is easily grounded. But, can we ground the claim appealing to more general moral principles that are accepted on every ethical system? I think we can, and here's the proof:

1. If x ought to be done, agents are morally accountable for failing to do x.
2. If x can't be done, agents are not morally accountable for failing to do x.

These propositions entail that ought entails can:

1. If p, then q
2. If r, then ~q
3. If q, then ~r (contrapositive of 2)
4. If p, then ~r (1+3 by transitivity)

The conclusion is that if x ought to be done, then x can be done. Both premises are eminently plausible. As for 1, I can't make sense of the notion of "ought", if not in the sense that you would be morally accountable for not doing it. That is just what one means by "moral duty". As for 2, if x can't be done, then one is not accountable in any sense for failing to do x, because it is not in your causal capacity to do x, and therefore you are not responsible or accountable for failing to do x. And therefore one is certainly not morally accountable for failing to do x.

I think then that there are good general moral considerations for believing that ought entails can. But, when examining specific ethical systems, it is hard to ground that claim except on utilitarianism or divine command theory. This may be thought of as constituting evidence against ethical systems other than those two. Further, the claim could be furnished in an interesting kantian moral argument for the existence of god. For it could be argued that we ought to bring about the greatest good. But, if god does not exist, then we cannot bring about the greatest good. Conjoined with the claim that ought entails can, this would imply the existence of god.

All food for thought.

 

[rvkevin]

But, can we ground the claim appealing to more general moral principles that are accepted on every ethical system? I think we can, and here's the proof:

1. If x ought to be done, agents are morally accountable for failing to do x.
2. If x can't be done, agents are not morally accountable for failing to do x.

What do you mean by being morally accountable? From a utilitarian perspective, it seems almost seems like coming up for another term for ought. However, on the divine command theory perspective, it could be deemed to be the cosmic judge who meets out punishment for failing to live up to moral duties as, IIRC, Craig mentioned in his debate with Shelly Kagan. In other words, the cosmic judge makes you accountable for your actions. On this meaning, only religious forms of morality would have duties and accountability, but then I would simply dismiss the existence of those; hence my inquiry into what you mean by a form of moral accountability that is generalize to all ethical systems.

Edit: Background, agree with utilitarianism and agree with the sentiment that ought implies can.

 

[ElvenKing]

No, an ought is does not imply a can(unless an argument is making the case that the "ought" is only what can be done). The argument presented in the OP falsely equates the "if" in the definition of ethical truth with the ability for ethical action to be achieved.

That x cannot be done does not change what would be so IF x were done. The fact that x cannot be done, and so maximal utility cannot be achieved, does not change that IF x were done maximal utility would be achieved.

 

[rvkevin]

That x cannot be done does not change what would be so IF x were done. The fact that x cannot be done, and so maximal utility cannot be achieved, does not change that IF x were done maximal utility would be achieved.

This is irrelevant. If a ship were to capsize, the best advice to the crew is not that they should walk on water. It doesn't matter that if they were able to do it that it would be good advice. They shouldn't try because they can't. The best course of action would be to put on a life suit because they can actually do that. They should put on a life suit, they shouldn't try to walk on water. Should is inherently tied to the outcomes. You should do something that if it brings you to an outcome that you desire. If you can't reach such an outcome by that means, then it becomes pointless to say that they should do it.

 

[ElvenKing]

This is irrelevant. If a ship were to capsize, the best advice to the crew is not that they should walk on water. It doesn't matter that if they were able to do it that it would be good advice. They shouldn't try because they can't. The best course of action would be to put on a life suit because they can actually do that. They should put on a life suit, they shouldn't try to walk on water. Should is inherently tied to the outcomes. You should do something that if it brings you to an outcome that you desire. If you can't reach such an outcome by that means, then it becomes pointless to say that they should do it.

It is relevant to understanding the nature of logic in reference to systems of ethics. Should is only inherently tied to outcomes when it is actually tied to the capacity of that outcome occurring. The presentation of the argument in the OP gives a false impression that the truth of an "if" ceases to be in a situation where a state cannot be achieved and consequently gives a misleading impression of what is actually possible under Utilitarianism. If, for example, the action which maximised utility was believing in a system of ethics which was impossible to achieve, the argument given in the OP would close off the capacity to comprehend this because it would present that a notion of unachievable ethics is incoherent. Of course, such a situation actually involves having a achievable state(believing in the unachievable ethics), so the sentiment of the OP's argument is correct in reference to Utilitarianism, but it form is misleading to the point where people could be lead away from the state which grants them maximal utility in the hypothetical example.

All ethical systems which have their positions defined by an occurring event by definition are tied to what can happen. Appropriate action under Utilitarianism is denied by what causes the majority to be happy. Divine Command Theory requires it be true that a deity can make a particular command(the OP's assertion that it must be because a deity would not command an impossible action is actually mistaken. This would only be the case if the deity was restricted to making commands humans were capable of fulfilling. They do get close to identifying this, but they end up pulling back and just presenting the notion that a deity could give an impossible command and punish someone for it as "evil" without fully identifying why this must be so) However, this is not a necessary part of any coherent system of ethics and that it applies to both Utilitarianism and Divine Command Theory is not evidence against other systems of ethics at all.

 

[rvkevin]

The presentation of the argument in the OP gives a false impression that the truth of an "if" ceases to be in a situation where a state cannot be achieved and consequently gives a misleading impression of what is actually possible under Utilitarianism.

I don't see how it is misleading at all.

1. If your ship is sinking, you wish to live, and you have access to a life suit, then you should put it on.
2. If your ship is sinking, you wish to live, and you can walk on water, then you should walk on water.

The condition for 1 is true for the hypothetical and as a result, the consequence is true. However, the condition for 2 is false, in all cases, and as a result, the consequence is also false, in all cases. While the conditional itself is true (you won't find a case where the ship is sinking, the people want to live, the people can walk on water, and they shouldn't walk on water), the consequent is false, which is the 'ought' part of the equation. For every true conditional ought proposition, whenever you find a true condition, you will find a true consequent, and vice avers because ought implies can.