Is Objective Consequentialism Compatible with the Principle that “Ought” Implies “Can”?

Abstract

Some philosophers hold that objective consequentialism is false because it is incompatible with the principle that “ought” implies “can”. Roughly speaking, objective consequentialism is the doctrine that you always ought to do what will in fact have the best consequences. According to the principle that “ought” implies “can”, you have a moral obligation to do something only if you can do that thing. Frances Howard-Snyder has used an innovative thought experiment to argue that sometimes you cannot do what will in fact have the best consequences because you do not know what will in fact have the best consequences. Erik Carlson has raised two objections against Howard-Snyder’s argument. This paper examines Howard-Snyder’s argument as well as Carlson’s objections, arguing that Carlson’s objections do not go through but Howard-Snyder’s argument fails nonetheless. Moreover, this paper attempts to show that objective consequentialism and other objectivist moral theories are compatible with the principle that “ought” implies “can”. Finally, this paper analyses a special kind of inability: ignorance-induced inability.

Some philosophers hold that objective consequentialism is false because it is incompatible with Ought Implies Can. Roughly speaking, objective consequentialism is the doctrine that you always ought to do what will in fact have the best consequences. According to Ought Implies Can, you have a moral obligation to do something only if you can do that thing. Frances Howard-Snyder has used an innovative thought experiment to argue that sometimes you cannot do what will in fact have the best consequences because you do not know what will in fact have the best consequences.

Erik Carlson has raised two objections against Howard-Snyder’s argument. This paper sets out to scrutinize Howard-Snyder’s argument as well as Carlson’s objections, arguing that Carlson’s objections do not go through but Howard-Snyder’s argument fails nonetheless. In the course of examining the Howard-Snyder/Carlson debate, the paper attempts to shed light on the concept of ignorance-induced inability and to show why objective consequentialism and other established moral theories are compatible with Ought Implies Can, as far as ignorance-induced inability is concerned. Finally, I hope that this paper has to share interesting thoughts on abilities, intellectualism, and by-relations between actions.

Howard-Snyder’s Argument

The worry that objective consequentialism could be incompatible with Ought Implies Can has been known for a long time and recently been reinvestigated.¹ At the outset of her article “The Rejection of Objective Consequentialism”, Howard-Snyder provides the following definition of objective consequentialism:

Objective consequentialism is the view that the agent ought to perform an action of those available to her which will have consequences at least as good as the consequences of any other action available to her.²

In the remainder of her article, Howard-Snyder paraphrases objective consequentialism as the position according to which agents ought to produce the best consequences.³

I think that Howard-Snyder’s own definition of objective consequentialism suggests that objective consequentialism is compatible with Ought Implies Can. Since objective consequentialism judges only available acts to be right or wrong, it has built-in the compatibility with Ought Implies Can. Objective consequentialism neither requires the non-swimmer to swim nor anyone to travel back in time, irrespectively of how good the consequences would be.

Nonetheless, Howard-Snyder thinks that she can demonstrate the incompatibility of objective consequentialism and Ought Implies Can by means of the Karpov case:

Imagine that something important (say, your career) hangs on my winning a chess championship by defeating Karpov. Suppose I try to do so and fail.⁴

The crucial feature of the case is that the chess layperson – let us call her Frances – could not have beaten Karpov and therefore could not have produced the best consequences.

Howard-Snyder has put forward the Karpov case as an analogy.⁵ The thought is that in ordinary choice situations, which series of choices will produce the best consequences is an even more complicated question than which series of moves will lead to victory over Karpov.

Let us assume that not only good but the best consequences would have obtained if, and only if, Frances had beaten Karpov. And let us treat the Karpov case not as an analogy but as a candidate situation in which the agent cannot do what she ought to do according to objective consequentialism.

Howard-Snyder’s argument can be summarized as follows:

• (1) According to objective consequentialism, you ought to produce the best consequences in all possible situations.⁶

• (2) In some possible situations, you cannot produce the best consequences.

• (3) Ought Implies Can: Necessarily, a person is morally obligated to φ only if the person can φ.

• (4) Therefore, in some possible situations it is not true that you ought to produce the best consequences.

• (5) Therefore, objective consequentialism is false.⁷

Carlson’s first objection pertains to premise (2). Carlson’s second objection and my own critique concern premise (1).

Ignorance-Induced Inability

The argument’s central idea, I take it, is that ignorance can lead to inability or, more precisely, that ignorance about the consequences of your options can make you unable to produce the best consequences. How could this idea work? Human agents often do not know the consequences of their options. But how can Howard-Snyder get from this triviality to the conclusion that there are situations in which you cannot produce the best consequences? Two assumptions are needed.

The first assumption is that your ignorance about the consequences of actions renders you ignorant about certain by-relations – such as those by-relations illustrated in Fig. 1 for the Karpov case. If you do not know the consequences of your options, then you do not know by the performance of which act you would produce the best consequences. For instance, Frances does not know the consequences of the available chess moves. Let “e2-e4, etc.” designate the sequence of chess moves Frances has to play in order to beat Karpov. Frances does not know that it is by playing e2-e4, etc., that she would produce the best consequences. (To put it differently: Frances does not know that she has to play e2-e4, etc., in order to produce the best consequences; she does not know that her playing e2-e4, etc., is a way for her to produce the best consequences.) The general point, again, is that if you do not know the consequences of your options, then you do not know that certain by-relations hold.

Fig. 1

The structure of the Karpov case

The second assumption is that ignorance about by-relations leads to inability to do what ranges at the higher levels of a chain of by-relations. See Fig. 2. Frances can perform the relevant basic actions⁸, that is, she can make the bodily movements that constitute her playing e2-e4, etc. And she can play e2-e4, etc. But since she does not know that by playing e2-e4, etc., she would be beating Karpov, she cannot beat Karpov. And in turn she cannot produce the best consequences.

Fig. 2

Frances’ “cans” and “cannots”

How can Howard-Snyder argue for the move from ignorance about by-relations to inability? Two ideas suggest themselves. One idea is to apply the well-known theories of ability to the Karpov case and show that these theories entail that Frances cannot produce the best consequences. Another idea is to make use of intellectualism. Let us start with the second idea.

Intellectualism is a view that connects knowledge-how to knowledge-that. As Jason Stanley explains:

According to Intellectualism about knowing how, knowing how to do something is equivalent to knowing in what way one could do it. So, for example, you know how to ride a bicycle if and only if you know in what way you could ride a bicycle. But you know in what way you could ride a bicycle if and only if you possess some propositional knowledge, viz. knowing, of a certain way w which is a way in which you could ride a bicycle, that w is a way in which you could ride a bicycle.⁹

If intellectualism is true, it follows that since Frances does not know that she has to play e2-e4, etc., in order to produce the best consequences, Frances does not know how to produce the best consequences. How does this support Howard-Snyder’s argument? If intellectualism is true, all Howard-Snyder has to show is that Frances’ lack of knowledge-how to produce the best consequences leads to Frances’ inability to produce the best consequences. This does not seem too difficult because talk of knowledge-how and talk of inability are intimately connected in everyday discourse.

Intellectualism, however, is far from uncontroversial. Many philosophers, influenced by Gilbert Ryle, believe that knowledge-how cannot be a species of knowledge-that because knowledge-how is an ability or a complex of dispositions, whereas knowledge-that is a relation between a thinker and a true proposition.¹⁰

Jason Stanley and Timothy Williamson have recently defended intellectualism on broadly linguistic grounds. They offer an influential critique of Ryle’s main argument as well as of other arguments for non-intellectualism.¹¹ However, Stanley’s and Williamsons’ account is still hotly disputed.¹²

Intellectualism has some intuitive appeal. Often you perform an action by performing another action and it seems true that you know how to perform the former action if and only if you know that you would perform it by performing the latter action. However, intellectualism is not very plausible when it comes to the performance of basic actions. Since you do not perform a basic action by performing another action, there is no by-relation you could have knowledge of in order to know how to perform a basic action. But since you can perform basic actions, it seems plausible to say that you know how to perform them.

It would go beyond the scope of this paper to discuss intellectualism.¹³ It seems clear that if intellectualism were true, or even true only with respect to non-basic actions, and I guess it might well turn out to be, then this would lend support to Howard-Snyder’s argument.

Let us then turn to the second way to connect ignorance about by-relations to inability. Here we can start with Howard-Snyder’s own ability-theoretical remarks. Howard-Snyder argues that an appropriate account of ability should make reference to why we are interested in it.¹⁴ According to her, we want to know how to understand the “can” implied by the “ought” in Ought Implies Can. Howard-Snyder connects “ought” to moral responsibility:

To say that the agent ought to ϕ is to make the agent morally responsible for ϕ-ing. To say that she ought to have ϕ-ed is to make her morally responsible for not ϕ-ing.¹⁵

Howard-Snyder goes on to argue that the attribution of responsibility makes sense only if the agent’s φ-ing is up to her, and that it is up to her only if her choosing to φ makes it at least very likely that she will φ. Hence, the “ought” in Ought Implies Can suggests that we should understand the “can” in such a way that an agent’s ability to perform an action consists in her choice’s making it very likely that she will perform the action.

Howard-Snyder seems to have in mind a variant or a derivative of the so-called conditional analysis of ability, one of the three major accounts of ability.¹⁶ According to the conditional analysis, an agent can perform an action if and only if the agent would perform the action if she tried (or chose, willed, …) to perform the action. According to Howard-Snyder’s account, the truth of the conditional, though not sufficient, is necessary for an agent’s ability to perform an action. Howard-Snyder thus seems to endorse the following principle:

Howard-Snyder’s necessary condition for ability An agent can φ only if it would be very likely that the agent φ-ed if she tried (chose, etc.) to φ.¹⁷

How does this principle support the move from ignorance about by-relations to inability? The answer is, of course, that since Frances does not know that it is by e2-e3, etc., that she would produce the best consequences and since there are countless series of chess moves Frances could play, Frances’ attempt to produce the best consequences would not make it very likely that she will produce the best consequences.¹⁸ This seems to be true.

Against Howard-Snyder’s necessary condition for ability, one could object that in order for an agent to be able to φ, it is not always necessary that the agent’s choosing to φ would make the agent’s φ-ing at least likely. For actually performing an action may be a sufficient condition for having the ability to perform that action even if it is not true that an agent’s choosing to φ would not make her agent’s φ-ing at least likely. J. L. Austin has famously put forward the example of a golfer who sinks a difficult putt. Austin argued that “it follows merely from the premise that he does it, that he has the ability to do it, according to ordinary English”.¹⁹

Howard-Snyder argues that actually performing an action is not sufficient for having the ability to perform the action.²⁰ I will leave it open whether her argument is successful. Although the conditional analysis seems at least to be a good rule of thumb for ascribing abilities or inabilities, respectively, there are more objections to it than the one just considered. However, insofar as other accounts of abilities also lead to the conclusion that Frances cannot produce the best consequences, Howard-Snyder’s necessary condition for ability is dispensable. And as I will argue in the following section, other accounts of ability indeed must lead to the conclusion that Frances cannot produce the best consequences in order to be convincing. At this point, it is enough to point out that, since Frances does not actually beat Karpov but tries to do so and fails, it does not follow from “ordinary English” (as Austin put it) that Frances can beat Karpov.

Carlson’s First Objection

In developing his objections to Howard-Snyder’s argument, Erik Carlson discusses a situation involving a safe:

We might compare the task of producing the best consequences with that of opening a safe to which you do not know the right combination. Suppose that the right combination is 448-961-5237. Because you can dial any ten-digit combination, you can dial this one. But since you do not know that this is the right combination […] you cannot open the safe.²¹

Let us assume that the person who does not know the right combination – Erik – had only one chance to dial 448-961-5237 in order to produce the best consequences. Erik thus did not have time to find out what the right combination is. Erik tried to open the safe and failed. Provided this modification, the case of the safe is as supportive of Howard-Snyder’s argument as the Karpov case.

Carlson’s first objection says that Erik could have opened the safe – just as Frances could have beaten Karpov and, in general, you can produce the best consequences.²² But why should one believe that Erik could have opened the safe? The answer, according to Carlson, is that he could have dialled 448-961-5237 and that, by doing so, he would have opened the safe. At this point, Carlson introduces two views on act-individuation and two corresponding derivatives of the conditional analysis:

If you were to try to dial 448-961-5237 you would succeed in doing this. On the other hand, if you were to try to produce the best consequences you would fail. Assume that we whish to claim that you can, nevertheless, produce the best consequences. If producing the best consequences and dialling 448-961-5237 are one and the same action, this claim is compatible with the following condition: A person can perform an action a only if there is some description of a, such that if he were to try to perform a under this description he would succeed. If, on the other hand, producing the best consequences and dialling 448-961-5237 are different actions, the claim that you can produce the best consequences is still compatible with this condition: A person can perform an action a only if there is some action b, such that (i) he would do a by doing b, and (ii) if he were to try to perform b he would succeed.²³

The challenge for Howard-Snyder is to point out why her necessary condition for ability, according to which an agent’s trying to produce the best consequences makes it at least very likely that the agent will produce the best consequences, is preferable to Carlson’s alternatives.

Against Carlson’s first objection, Howard-Snyder argues that, although there is a sense of “can” in which it is true that Frances could have beaten Karpov, Frances could not have beaten Karpov in the sense required by Ought Implies Can in order for her to have had an obligation to beat Karpov. In support of this claim, Howard-Snyder points out that in everyday discourse we would assume that Frances could not have beaten Karpov.²⁴ Howard-Snyder also alludes to the connection between Ought Implies Can and moral responsibility, which we encountered in the previous section.

Howard-Snyder is right: There are strong common-sense intuitions to the effect that Frances could not have beaten Karpov in a morally relevant sense and that, in consequence, it would be unfair to blame her for failing to do so. Hence, common-sense intuitions count against Carlson’s first objection.

Howard-Snyder’s argument can be strengthened in two respects. First, Eric Wiland has added further examples of presumably ignorance-induced inabilities: you cannot write an article that explains how to cure AIDS, you cannot write the Great American Novel, you cannot write the Islamic equivalent of Locke’s Letters on Toleration.²⁵ Let us focus on your inability to write an article that explains how to cure AIDS. Given the astronomically good consequences your writing an article that explains how to cure AIDS would have, it seems that if you could write the article, you would be obligated to do so not only according to consequentialism but also according to common-sense morality. This allows for the following argument:

• (6) If you could write an article that explains how to cure AIDS, then common-sense morality would require you to write it.

• (7) But you are not required to write the article according to common-sense morality.

• (8) Therefore, you cannot write it.

Here is a further reason to accept Howard-Snyder’s reply to Carlson. Howard-Snyder has not explicitly addressed Carlson’s challenge to show what is wrong with his alternatives to her necessary condition for ability. But this can easily be fixed: Any persuasive theory of ability has to do justice to our ordinary judgments about abilities and hence has to obey certain extensional constraints: It must not ascribe too many or too few abilities to agents.²⁶ A theory of ability according to which you can write an article that explains how to cure AIDS ascribes too many abilities, as from the perspective of everyday discourse, and thus violates an extensional constraint.

In sum, Carlson’s first objection fails. There are strong common-sense intuitions to the effect that Frances could not have beaten Karpov. A persuasive theory of ability has to take these intuitions into account. It follows that you cannot produce the best consequences.

Carlson’s Second Objection

Howard-Snyder discusses the following objection: Objective consequentialists “tell us to perform that action, of those we can perform, which will have the best consequences”.²⁷ Frances can play e2-e4, etc. By implying that Frances ought to play e2-e4, etc., objective consequentialism does not require anything impossible from Frances. This is, of course, the worry I mentioned already when presenting Howard-Snyder’s argument.

Howard-Snyder has a twofold reply to the objection. First, she claims that objective consequentialism by definition does not only require Frances to play e2-e4, etc., but also to do what has the best consequences. And Frances cannot do the latter. Second, if we modify objective consequentialism such that it does not anymore require Frances to do what has the best consequences but only to play e2-e4, etc., then objective consequentialism becomes a very undesirable moral theory.

Why does Howard-Snyder think that we cannot do what objective consequentialism, by definition, requires us to do?

Here is an instruction which objective consequentialism issues and which I cannot obey: “perform that action, of those you can perform, which will have the best consequences”. I cannot perform this action for the same reason I cannot beat Karpov. But […] it is true by definition that objective consequentialism requires this of me.²⁸

The details are important: According to Howard-Snyder, Frances can play e2-e4, etc., and her playing e2-e4, etc., would have the best consequences. But Frances cannot obey the instruction to do what would have the best consequences. For Frances does not know that playing e2-e4, etc., is the action with the best consequences.

Carlson has replied to the claim that Frances cannot do what has the best consequences. He argues that a distinction between properties of acts and act-types is crucial for rebutting Howard-Snyder’s argument:

It might be thought, however, that the objective consequentialist involves herself in a contradiction, by denying that you ought to produce the best consequences. Is she not, after all, committed to precisely the claim that you ought to produce the best consequences? The answer is yes and no. The question whether you ought to produce the best consequences is ambiguous. On one interpretation, it is the question whether there is some action that you ought to perform, which has the property of producing the best consequences available. The answer to this question is, “Yes”, according to objective consequentialism. Your dialing 448-961-5237 is such an action. On the other interpretation, the question is whether you ought to perform an action of the act-type “producing the best consequences”. The objective consequentialist’s answer to this question may well be no, since it may be the case that you cannot perform an action of this type.²⁹

If Frances cannot produce the best consequences, then, so the suggestion goes, objective consequentialism does not require her to perform an act of the act-type “producing the best consequences”. Objective consequentialism only requires Frances to produce an act that has the property of “having the best consequences”. Since Frances can play e2-e4, etc., and Frances’ playing e2-e4, etc., has the property of having the best consequences, Frances can perform the act that has the best consequences. Hence, objective consequentialism is compatible with Ought Implies Can. Carlson’s suggestion is illustrated in Fig. 3.³⁰

Fig. 3

Carlson’s suggestion

I think that Carlson is right in that, according to objective consequentialism, Frances has a moral duty to play e2-e4, etc. However, I think that Carlson’s distinction between act-types and act properties is not sufficient for rebutting Howard-Snyder’s argument. First, Carlson fails to make clear what exactly it is that he has in mind. Is his view that according to objective consequentialism Frances has a moral duty to (i) perform an act that has the property of having the best consequences or (ii) play e2-e4, etc.? Howard-Snyder has replied to both possible interpretations.³¹ In case of (i), Howard-Snyder’s argument still goes through. For Frances does not know what she has to do in order to do what has the property of having the best consequences.³² In case of (ii), objective consequentialism delivers an incredibly huge list of moral duties that apparently have nothing in common. Frances has the moral duty to play e2-e4, etc., I may have the moral duty to check my emails, you may have a moral duty to make two steps to the left, and so on. Since this list lacks coherence, objective consequentialism becomes an undesirable doctrine.

Carlson’s Second Objection Revisited

I think that Carlson should have chosen answer (ii): The only moral duty objective consequentialism implies in Frances’ case is to play e2-e4, etc., cf. Figure 4. Does this mean that objective consequentialism becomes an undesirable “huge-list” theory? I submit that objective consequentialists can stick to the established definition of objective consequentialism and nonetheless insist that Frances’ only moral duty is to play e2-e4, etc. How is this possible?

Fig. 4

The solution

The general definition of objective consequentialism, according to which an act is obligatory if and only if it has better consequences than any alternative, does not state a moral duty. Rather, it is a statement of necessary and sufficient conditions for an act’s being obligatory. It is an act’s having better consequences than its alternatives which makes the act obligatory. The point of the general definition of objective consequentialism is to explain what makes an act right or obligatory by stating the right-making or ought-making features. It is the point of objective consequentialism qua moral doctrine to provide explanations of this sort. Thus, objective consequentialism explains the rightness of Frances’ playing e2-e4, etc., by the fact that this action has the property of having the best consequences.

Objective consequentialists can have their cake and eat it, too: they can endorse a general formula, which states the conditions under which acts are right, wrong, or obligatory, respectively, and at the same time deny that their theory issues any general instructions that agents would not be able to follow. Due to the general formula, the huge list of instructions issued by objective consequentialism does not lack coherence at all. At the same time, it does not make sense for Frances to complain that she did not know how to obey the instruction “Do what has the best consequences”, for objective consequentialism, properly understood, does not say that Frances has a moral obligation to do what has the property of having the best consequences. Of course, it does make sense for Frances to complain that she did not know which action she ought to perform according to objective consequentialism. But this is a different issue. For as we have seen in the “Ignorance-Induced Inability” section Howard-Snyder’s argument is based on the idea to derive inability from lack of knowledge-how, not from lack of knowledge-which.³³

A defender of Howard-Snyder’s argument could try to object, “Since Frances did not know that objective consequentialism required her to play e2-e4, etc., she did not know how to do what objective consequentialism required her to do. Therefore, Frances was unable to do what objective consequentialism required her to do”. But the reply is the same as before: according to objective consequentialism, you do not have a moral obligation “to do what objective consequentialism requires”. In most situations, we do not know which act we have a moral obligation to perform according to objective consequentialism. But we know that it is an act with consequences at least as good as those of any of its alternatives.

Here is a related objection.³⁴ It seems to be trivially true that you ought to do what you ought to do. However, on the view defended here, apparently, in cases like that of Karpov agents cannot do what they ought to do: while Frances can, and ought to, play e2-e4, etc., she cannot do what she ought to do – despite the fact that by playing e2-e4, etc., she would be doing what she ought to do. Now, if a theory implies that you cannot do what you ought to do, the theory seems to be incompatible with Ought Implies Can after all.

The objection focuses on this claim:

• (9) Necessarily, you can do what you ought to do.

Does objective consequentialism, in the version I suggest, contradict (9)? It depends. Let me start by repeating the distinction between a coarse-grained and a fine-grained account of act individuation, which has been introduced in the section entitled “Carlson’s First Objection”. On the coarse-grained account Frances’ playing e2-e4, etc., and her doing what she ought to do are different descriptions of one and the same act – just like her playing e2-e4, etc., and her beating Karpov. The proper object of obligation on this view of act individuation are not acts per se but acts under certain descriptions: not Frances’ beating Karpov – under this description of her act – but her playing e2-e4, etc., − under this description of her act. On the fine-grained view, in contrast, Frances’ playing e2-e4, etc., and her doing what she ought to do are not just different descriptions of the same act but different acts. Now, if (9) is meant to have the same truth conditions as Ought Implies Can (the doctrine that, necessarily, a person ought to φ only if the person can φ), then the objective consequentialist can accept (9). If, however, (9) is meant to be identical, on a coarse-grained view of act individuation, with

• (10) Necessarily, if you ought to perform an act, then you can do what you ought to do (under this description of your act),

or, on a fine-grained view,

• (11) Necessarily, if you ought to perform an act, then you can perform the act of doing what you ought to do,

then, indeed, I think that the objective consequentialist should reject (9). At any rate, this paper attempts to show that objective consequentialism is compatible with Ought Implies Can and can be silent on (10) and (11).

But maybe this is too quick. Cannot something like (10) or (11) be derived from the triviality that you ought to do what you ought to do in conjunction with Ought Implies Can? The challenge for the objective consequentialist is to show that

• (12) You ought to do what you ought to do,

is not trivially true on a reading that, when combined with Ought Implies Can, leads to (10) or (11).

Let us start by disambiguating the claim that you ought to do what you ought to do. (12) might be understood as follows:

• (13) If you ought to perform an act, then you ought to perform this act.

(13) is indeed trivially true. (13) implies, for instance, that, if Frances ought to play e2-e4, etc., then Frances ought to play e2-e4, etc. However, neither (10) nor (11) can be derived from (13) in conjunction with Ought Implies Can. (13) is compatible with objective consequentialism.

Let us now come to the problematic interpretations of (12). Here is one interpretation, based on a coarse-grained view of act individuation, of the claim that you ought to do what you ought to do that is indeed incompatible with objective consequentialism in the version defended in this paper:

• (14) If you ought to perform an act, then you ought to perform this act under the description “act which I ought to perform”.

(14) is not trivially true. Objective consequentialists typically hold that the desire to produce good consequences should not always be the motive of your actions. From there it is only a small step to the claim that you should not always perform obligatory acts under this description. As long as the defender of Howard-Snyder’s argument does not offer an argument in support of (14), the objective consequentialist can dismiss (14) as question-begging.

On the fine-grained view of act individuation, a natural understanding of (12) – apart from (13), which turned out to be unproblematic – is this:

• (15) If you ought to perform an act, then you ought to perform the act of doing what you ought to do.

Of course, (15) in conjunction with Ought Implies Can is incompatible with the version of objective consequentialism suggested here. However, (15) isn’t trivially true either – for the same reasons that apply to (14).

The bottom line is that the objective consequentialist can reject the claim that, necessarily, you can do what you ought to do, insofar as this claim contradicts objective consequentialism. At least, this is true if one tries to justify the claim on the basis of the alleged triviality that you ought to do what you ought to do in conjunction with Ought Implies Can.

Conclusion

This paper has argued that Frances Howard-Snyder’s argument for the incompatibility of Ought Implies Can with objective consequentialism fails. The argument fails because, although objective consequentialism tells you qua moral doctrine that an act is obligatory if and only if its consequences are better than the consequences of any alternative, objective consequentialism does not say that it is obligatory for you to do whatever has better consequences than any alternative. The former claim is a statement about necessary and sufficient conditions for an act being obligatory, not the expression of an obligation.

We have also seen a general reason to think that objective consequentialism is compatible with Ought Implies Can. Objective consequentialism ranges only over acts that you can perform (or omit) when assigning rightness or wrongness. I conclude that objective consequentialism is compatible with Ought Implies Can.

Finally, notice that as far as ignorance-induced inability is concerned all that has been said in this article about the compatibility of objective consequentialism with Ought Implies Can generalizes to non-consequentialist moral theories. Take a rights-based moral theory according to which you ought not to do what in fact violates someone’s rights. Suppose you open a door and thereby hurt someone, while you believed, in accordance with the evidence available to you, that you would not hurt anyone by opening the door. A rights-based theory may say, without contradicting Ought Implies Can, that you had a moral duty not to open the door because your opening the door has the feature of violating someone’s right to physical integrity. The theory must not say, though, that you had a moral duty not to violate the victim’s physical integrity; for this claim would be incompatible with Ought Implies Can.