Legitimacy as a Mere Moral Power? A Response to Applbaum

In a recent article, Arthur Applbaum contributes a new view—legitimacy as a moral power—to the debate over the concept of political legitimacy.^[1] Applbaum argues that his account is a genuinely alternative conceptual possibility because it neither reduces to the view of legitimacy as a liberty to coerce, nor collapses into the view of legitimacy as a claim to have the law obeyed.^[2] (For the sake of simplicity, let me follow Applbaum in calling them the power-liability view, the liberty-to-coerce view and the legitimacy-entails-duty view, respectively.^[3]) Moreover, he contends that the power-liability view gets the upper hand over two other views, for it captures the particularity of the normative relations entailed in legitimacy^[4] (where the liberty-to-coerce view fails^[5]) and yet leaves enough space for justified civil disobedience (where the legitimacy-entails-duty view allegedly fails). He then comes to the conclusion that “moral power, and only moral power, is at the core of the concept—what legitimacy is.”^[6]

While acknowledging that the power-liability relation is an indispensable element in conceptualizing political legitimacy and that the power-liability view is indeed a distinct conception of legitimacy, I remain unconvinced of Applbaum’s critiques of the legitimacy-entails-duty view and his claim that “only moral power, is at the core of the concept—what legitimacy is,” and therefore hope to challenge his position on the concept of legitimacy in this article. I first lay out Applbaum’s main argument, with the aim of showing how Applbaum utilizes the distinction between the concept and the normative conception of legitimacy to reject competing views of legitimacy for being normative conceptions. The second section of this article shows that there is only one plausible way for Applbaum to explain what a normative conception of legitimacy is and why such a conception must be rejected: Applbaum could argue that competing conceptions of legitimacy are normative and hence unacceptable because they mistakenly incorporate substantive moral argument into the conceptual analysis.^[7] But to show why substantive moral argument is involved in competing conceptions of legitimacy, Applbaum relies on the claim that moral duties are conclusory rather than prima facie. In the third section, granting Applbaum’s account and rejection of normative conceptions of legitimacy, I examine whether Applbaum’s power-liability views can stay clear of substantive moral argument. I will show that the power-liability view refers to more than a mere power and implicitly incorporates substantive moral argument; it would therefore also be an unacceptable normative conception of legitimacy. The fourth section then undermines Applbaum’s account of normative conceptions of legitimacy by showing that moral duties are better understood as prima facie, and thus, as can be demonstrated, there is no substantive moral argument involved in competing conceptions of legitimacy. Corresponding to Applaum’s original article, I end the article here with an argument that shows why Applbaum is also mistaken in rejecting the legitimacy-entails-immunity view.^[8]

I. Applbaum’s main argument

Following Rawls, Applbaum begins his article with a distinction between concept and conception. A concept “marks off the arena of a disputed idea that is filled, through normative argument, by conflicting conceptions with different content, criteria, or conditions.”^[9] With regard to legitimacy, Applbaum thus states:

[T]he concept of legitimacy is about the right to rule. Various conceptions of legitimacy differ on what the necessary and sufficient conditions are for A to have the right to rule B in some context C … But neither having a right nor ruling are transparent notions. Is the right to rule a claim-right that entails a moral duty to obey, or a mere liberty? Is the right of a legitimate ruler conclusory, or merely presumptive?^[10]

Applbaum warns that “we risk begging the normative question” and “assuming one’s conclusion” if in specifying a concept we make it too restrictive and narrow (for example by importing a particular conception into the concept).^[11] In Applbaum’s view, however, this is precisely the mistake committed by the legitimacy-entails-duty view, which holds that a state is legitimate only if its subjects have a moral duty to obey the law.

Applbaum first explores Simmons’s version of the legitimacy-entails-duty view. He points out that Simmons implicitly and mistakenly ties the duty to obey to consent, yet “the consent of the governed is not itself a conceptual requirement of legitimacy.”^[12] Therefore, Simmons’s view should be rejected because it is too restrictive; it amounts to “a normative conception of legitimacy.” ^[13]

With regard to Raz’s version of the legitimacy-entails-duty view, Applbaum agrees with Raz that “the exercise of legitimate authority by an actor entails some change in the normative situation or status of another.”^[14] Applbaum, however, argues that the normative change in question is not necessarily the generation of an obligation.^[15] In fact, he argues that the legitimacy-entails-duty view is conceptually unacceptable since it rules out the possibility of justified civil disobedience. “If the authority is legitimate, disobedience is not justified.”^[16] But, of course, whether civil disobedience is justified should not be settled by conceptual analysis, but by substantive moral argument.

Similarly, in refuting Copp’s competing view of legitimacy,^[17] which says that “political legitimacy confers a bundle of various Hohfeldian advantages,” Applbaum argues that “Copp does not make the sharp distinction between concept and conception that I am invoking.”^[18] While “reasonable normative conceptions of legitimacy would include most if not all elements of such a bundle,” it is “moral power, and only moral power, [that] is at the core of the concept.”^[19]

According to Applbaum’s power-liability view, “legitimacy is a kind of moral power, the power to create and enforce nonmoral (or perhaps I should say not yet moral) prescriptions and social facts.”^[20] And “when A has a moral power with respect to B … then B must have a correlative moral liability,” which means that “B’s moral claims against some other subject C or against the legitimate A can be changed by A.”^[21] Applbaum further explicates that:

By A’s decree, C might gain a moral privilege not previously held to act against B’s interests … and B may be subject to morally justified enforcement of its institutional duties to A or to C. But a moral liability is not a moral duty, and an institutional duty is not a moral duty.^[22]

According to Applbaum, whether such an institutional duty is also a moral duty is “a conceptually open question to be settled by substantive moral argument in light of morally relevant empirical circumstances.”^[23]

The power-liability view so far characterizes the internal aspect of legitimacy, that is, the relations between a legitimate state and its subjects. According to Applbaum, however, the power-liability view also challenges the legitimacy-entails-immunity view (which is concerned with the external dimension of legitimacy), for “there is no conceptual route from having legitimacy—having moral power—to having moral immunity. … These are all connections will have to be established by moral argument, not conceptual analysis.”^[24]

II. Making sense of normative conceptions of legitimacy: substantive moral argument and conclusory duties

As my emphases in the last section indicate, Applbaum utilizes the distinction between the concept and the conception of legitimacy to reject other competing views of legitimacy for being normative conceptions. Despite its importance, Applbaum does not elaborate on the distinction between the concept and the conception of legitimacy. He may well believe that the distinction is fairly clear. But it is not.

First of all, for the distinction between concept and conception in general, it is unclear why conflicting conceptions fill in the arena marked off by the concept “through normative argument.” For example, it is certainly not true that various conceptions of fish fill out the details of the concept of fish “through normative argument.”

Applbaum would surely reply that he only intends to discuss the distinction between normative (or moral) concepts and conceptions, such as the distinction between the concept and the conception of legitimacy. I have no objection to this clarification. ^[25] But since legitimacy is a normative concept, conceptions of legitimacy are necessarily normative. However, if this is so, why then reject competing views of legitimacy for being normative conceptions? What is wrong with being a normative conception of legitimacy?

At this point, Applbaum might reply that “normative argument” refers to “substantive moral argument in light of morally relevant empirical circumstances.”^[26] Similarly, “normative conceptions” of legitimacy refers to conceptions of legitimacy that incorporate substantive moral argument. To reach a final and determinate moral conclusion or evaluation, the reply continues, we need to resort to substantive moral argument, but at the conceptual level, even for normative concepts like legitimacy, no substantive moral argument is needed. Therefore, conceptions of legitimacy that incorporate substantive moral argument can be rejected for “assuming the conclusion.” For example, as Applbaum attempts to show, Raz’s legitimacy-entails-duty view should be rejected because already at the conceptual level it mistakenly assumes the conclusion that there could be no justified civil disobedience in a legitimate state.

I indeed think that the above clarification dissolves some confusions. And for the sake of clarity, let us restate Applbaum’s claim that normative conceptions of legitimacy should be rejected as follows: various conceptions of legitimacy, like those held by Simmons, Raz and Copp, should be rejected if they mistakenly assume moral conclusions by incorporating substantive moral argument.

But why does Applbaum think that those conceptions of legitimacy incorporate substantive moral argument? Recall how Applbaum distinguishes the concept of legitimacy from its conceptions: the concept of legitimacy is about the right to rule; various conceptions differ not only on the notion of the right to rule, but also with regard to reasons that justify or ground the right to rule.^[27] According to Applbaum then, a conception of legitimacy contains two elements: the specification of the concept of legitimacy and the justification of legitimacy as specified. Since one has to engage in substantive moral argument to reach the conclusion that A’s legitimacy over B is justified, Applbaum could argue that a conception of legitimacy necessarily incorporates substantive moral argument. Remember also that this is precisely the way that Applbaum rejects Simmons’s version of the legitimacy-entails-duty view, which he accuses of building “the consent of the governed” into the “conceptual requirement of legitimacy.”

This line of argument might seem plausible at first sight, but the problem is that Applbaum ignores the fact that he and Simmons simply do not use the word “conception” in the same way. Applbaum has a broad concept of “conception,” as it were, while Simmons has a narrow concept of “conception.” In other words, Simmons means by “conception” what Applbaum means by “concept”: for Simmons, a conception of legitimacy merely refers to the specification of the concept (sensu Applbaum) of legitimacy, without also including the justification of legitimacy. For instance, in his essay Justification and Legitimacy, Simmons says that:

[I]t is this conception of “legitimacy” that I will hereafter have in mind when I use that term. A state’s (or government’s) legitimacy is the complex moral right it possesses to be the exclusive imposer of binding duties on its subjects … Accordingly, state legitimacy is the logical correlate of various obligations, including subjects’ political obligations.^[28]

Clearly, Simmons does not build the justification of legitimacy into his conception of legitimacy. Rather than tying the consent of the governed to the legitimacy-entails-duty view, Simmons draws a sharp line between the conceptual analysis and the justification of legitimacy. With regard to the former, he defends the legitimacy-entails-duty view (or conception) against the liberty-to-coerce view (or conception). Only at the stage of justifying legitimacy, where substantive moral argument is involved, does Simmons contend that consent is the necessary condition for legitimacy.^[29] Therefore, Applbaum merely attacks a straw man by accusing Simmons of “mistakenly” invoking substantive moral argument in the conceptual analysis.^[30] Similarly, since Copp also separates the justification of legitimacy from the conception of legitimacy,^[31] Applbaum is begging the question (as to why Copp’s conception of legitimacy incorporates substantive moral argument) when he criticizes Copp’s view for “mistakenly” invoking the substantive moral argument in the conceptual analysis.^[32]

It remains possible, however, that for the sake of argument Applbaum could agree with Simmons and Copp that a conception of legitimacy refers only to the specification of what legitimacy is, yet still argue that competing conceptions of legitimacy necessarily incorporate substantive moral argument. Applbaum actually suggests this line of argument in his critique of Raz’s legitimacy-entails-duty view. Remember that Applbaum does not accuse Raz of implicitly injecting possible grounds of legitimacy into the conception of legitimacy. On the contrary, Applbaum comments that Raz “has made the strongest case for a conceptual connection between legitimate authority and obligation.”^[33] Acknowledging that Raz has strictly done conceptual analysis, Applbaum seems to accept that the legitimacy-entails-duty view is simply a conception of legitimacy specifying what legitimacy is. But if this is so, how could Applaum also claim that the legitimacy-entails-duty view still incorporates substantive moral argument and thus should be rejected? The answer, according to Applbaum, lies in the dispositive or conclusory nature of moral duties. “Grounds for duties can be overridden and outweighed,” Applbaum asserts, “[but] duties are conclusory.”^[34] That is to say, substantive moral argument will show that either there exists an unoverridable duty, or the alleged duty simply dissolves; there is no such thing as a prima facie duty. That is why Applbaum could object that the legitimacy-entails-duty view, even if it is merely intended to specify the concept of legitimacy, still mistakenly incorporates substantive moral argument and rules out the possibility of justified civil disobedience. Similarly, Copp’s view of legitimacy, since it also entails the duty to obey,^[35] is equally unacceptable.

To summarize, I have argued that (1) for Applbaum’s distinction between the concept and the normative conception of legitimacy to work, Applbaum must explain to us why various conceptions of legitimacy he attempts to reject incorporate substantive moral argument; and that (2) to provide such an explanation, Applbaum would do better to accept that a conception of legitimacy is merely a specification of what legitimacy is (without also including the justification of legitimacy),^[36] although the success of the explanation still rides on the claim that moral duties are of conclusory (moral )force.

III. But the power-liability view is not free of substantive moral argument

In contrast to those conceptions of legitimacy held by Simmons and others, Applbaum’s power-liability view is alleged to contain only a mere moral power, without entailing the duty to obey on the part of the subject or “the immunity from outside intervention.” Therefore, the presumed merit of the power-liability view is that it is a conception of legitimacy free of substantive moral argument.

Let us take a close look at the power-liability view and examine whether the presumed merit of the power-liability view exists. First, remember that according to Applbaum, “legitimacy is a kind of moral power, the power to create and enforce nonmoral … prescriptions and social facts.”^[37] Anyone familiar with Hohfeldian terminology, however, will get confused by the phrase “the power to create and enforce nonmoral … prescriptions.” According to the Hohfeldian analysis of rights that Applbaum also endorses, there are four distinct moral advantages: privilege or liberty, claim, power and immunity. Moreover, not only are these four moral advantages different from each other, they do not entail each other. Thus, a moral power of the state is a second-order advantage to change first-order moral relationships between the state and its subjects via the creation of the law or “not yet moral prescriptions”; specifically, a moral power is the state’s ability to either make the law or “nonmoral prescriptions” morally binding (so that the subject has a duty to obey), or make the enforcement of law permissible (so that the state has a liberty to coerce). The state’s moral power to change the moral status of its subjects via the creation of the law, therefore, should not be conflated with the state’s liberty to create the law. The state may have a liberty to create a law, yet without having the power to make it morally binding or make the enforcement morally permissible. Conversely, the state may have the power to impose certain obligations on citizens by creating a law, and by creating it the state would exercise that power – yet, this does not imply that the state also has the liberty to create the law in question. After all, the state might well be under a duty not to exercise its power here.^[38] In addition, we should also distinguish a liberty to create the law from a liberty to enforce the law. Thus, given the Hohfeldian terminology, “the power to create and enforce nonmoral … prescriptions” is a conceptual impossibility.

We find more misinterpretations and misapplications of Hohfeldian advantages in other places of Applbaum’s article. In one place, he says:

Having the power to change these moral statuses is moral power. Although all such moral powers can also be said to be moral privileges to exercise legal powers—hence the temptation toward reduction—not all moral privileges to exercise legal powers are moral powers, because the exercise of legal power does not always affect moral rights and privileges.^[39]

It is true that “the exercise of legal power does not always affect moral rights and privileges,” but this does not mean that when the exercise of legal powers (i.e., the creation of a law) does affect moral rights and privileges (of the subjects), the moral privileges to exercise them are, therefore, moral powers, nor can we say that a moral power to change the moral status of the subject by issuing a law entails a liberty to issue the law. Remember, a power and a liberty are conceptually separate; they do not entail each other.

“This is true in general,” Applbaum may reply, “but for the state, how can it have a power to change the moral status of the subject by issuing a law, yet without having a liberty to issue the law?” Well, it actually can. Consider a legitimate state where the moral value of democracy (i.e., democracy as an embodiment of equal respect among citizens) makes the law morally binding.^[40] Even if the democratic legislative body passes a mildly unjust law (say, it imposes slightly higher taxes on cat possession than on dog possession), here the state still exercises a moral power to impose a duty to obey on its subjects, thanks to the moral weight of democracy. But since the law is unjust, the state (or the democratic legislative body) does not have a moral liberty to create it (all else being equal, that is, barring a necessity justification to issue an unjust law), for the state has a duty not to do so.^[41] That being said, I am not denying that a legitimate state has both a power and a liberty to create reasonably just or morally innocent laws, and that a legitimate state has neither a power nor a liberty to create grossly unjust laws.

In another place, Applbaum says:

Even simple moral permissions come along with a duty not to interfere with the exercise of the permission in some way or another—not by conceptual necessity, but by a sort of practical necessity if the permission is not to be futile. … So, while it is true that on the power-liability view, the exercise of legitimate authority by A over B imposes on B not only liability for enforcement and punishment, but also a duty not to take certain measures of resistance to enforcement and punishment, this does not collapse the power-liability view into the right-duty view of legitimacy.^[42]

Note first that just like the aforementioned error of saying “the power to create and enforce” legal duties, the phrase “liability for enforcement and punishment” makes no sense. The correct way to formulate the power-liability view is to say that the exercise of legitimate authority by A over B strips B of the claim against A not to be enforced (hence, A gains a liberty to enforce).^[43] Moreover, Applaum is also flatly wrong in claiming that simple moral permissions, out of practical necessity, contain a duty of noninterference. On the one hand, Applbaum urges us to stick to conceptual analysis; on the other hand, he readily abandons the conceptual analysis whenever it seemingly does not work, and embraces “practical necessity” in a question-begging way. Is there any workable conceptual analysis that also accounts for practical necessity? Contrary to Applbaum’s view, there is; and it is a simple one. A liberty (or permission) together with a claim of noninterference that protects the liberty (so that the subject has “a duty not to take certain measures of resistance to enforcement and punishment”) will be enough to dissolve the practical concern. ^[44]

Let me pause to review all of Applbaum’s confusions identified in this section. 1. He confuses a power to change the moral status of the subject by issuing the law with a liberty to issue the law. 2. He confuses a liberty to issue the law with a liberty to enforce the law. 3. He confuses a liberty to enforce the law with a claim of noninterference in the enforcement of the law. ^[45] With all these confusions, Applbaum manages to squeeze other Hohfeldian advantages into a mere power so that the power-liability view is alleged to be free of substantive moral argument. But once unpacked, the power-liability view actually refers to: a liberty to issue the law, a power to change the moral status of the subject by issuing the law, a liberty to enforce the law and a claim of noninterference in the enforcement of the law.

Let me now show why Applbaum’s “power-liability view” (which, as we just saw, actually relies on Hohfeldian concepts beyond power and liability), in light of his own account of normative conceptions of legitimacy, is not free of substantive moral argument either and should therefore also be rejected. Remember first that on Applbaum’s view, since moral duties are conclusory and can only be established by substantive moral argument, any conception of legitimacy that involves a duty to obey necessarily incorporates substantive moral argument. But a duty to obey is just one instance of numerous conclusory moral duties. Strictly speaking, therefore, any conception of legitimacy that involves a duty (whatever it is) necessarily incorporates substantive moral argument. Now, according to Applbaum’s actual (as opposed to official) “power-liability view”, a legitimate state has a claim of noninterference in the enforcement of the law. But since the claim of noninterference correlates with the moral duty of noninterference on the part of the subject, it has to be established by substantive moral argument. Much worse for Applbaum, even the legitimate state’s liberty to enforce the law has to be established by substantive moral argument. This is because: (1) to establish a liberty to coerce is to show that the state has no duty not to coerce, yet (2) according to Applbaum, substantive moral argument must be invoked to show either a conclusory duty or no duty at all, therefore (3) substantive moral argument must also be invoked to establish a liberty to coerce. Similarly, to establish whether the state has a liberty to issue the law, that is, whether the state has no duty not to issue the law, we have to engage in substantive moral argument in light of the merits and demerits of the law in question.

While the above objection severely undermines Applbaum’s central claim that the power-liability view gains the upper hand over other competing views of legitimacy, its limitation has to be acknowledged. The objection would also refute the legitimacy-entails-duty view and Copp’s view based on Applbaum’s account of normative conceptions of legitimacy. But I believe that Applbaum’s account of normative conceptions of legitimacy is mistaken, that is, the legitimacy-entails-duty view and Copp’s view incorporate no substantive moral argument. This is what I am going to argue in the next section.

IV. Undermining Applbaum’s account of normative conceptions of legitimacy

As I have indicated in the second section, Applbaum’s rejection of normative conceptions of legitimacy is based on his claim that moral duties are of conclusory force and that there is no such thing as a prima facie duty. Thus, one can undermine Applbaum’s rejection of competing conceptions of legitimacy by showing that it does make sense to conceive moral duties as prima facie.

Applbaum provides two reasons for this view of conclusory duties. First, Applbaum believes that the view is supported by Raz’s preemption thesis in his account of authority, according to which “legitimate authority can be restricted in scope … but within its scope, the directive of a legitimate authority is an exclusionary reason that preempts the primary reasons that count for and against the directive.”^[46] Second, although the phenomenon of “apparent conflicts of moral duty” is often invoked to support the view of prima facie duties, the view of conclusory duties has no difficulty accounting for the phenomenon: “Grounds for duties can be overridden and outweighed, but a specified duty, if it is a duty, is not overrideable or outweighable.”^[47] Moreover, the view of prima facie duties misleads us into “a false temporal story … under which unforeseen circumstances come up to override,” yet the view of conclusory duties presents “the correct way to think about these circumstances is that there are exceptions built in from the start.”^[48]

Nevertheless, neither of the two reasons Applbaum provides survives further scrutiny. The first reason is based on Applbaum’s misinterpretation of Raz’s preemption thesis. In explaining the preemption thesis, Raz clarifies that “[i]t is not that the arbitrator's word is an absolute reason which has to be obeyed come what may. It can be challenged and justifiably refused in certain circumstances,” such as in the situation “if new evidence of great importance unexpectedly turns up.”^[49] In a more recent article, again, Raz explicitly says that “[a]uthoritative directives are not always conclusive reasons for the conduct they require. They can be defeated by conflicting reasons, or by conflicting directives. The reasons that can defeat them are those they do not exclude.”^[50]

Now Applbaum might reply that on Raz’s account of authority, the allegedly defeated authority is actually illegitimate in the first place. The reply might go like this: non-excluded reasons set the limits of the authority’s jurisdiction; if the authority “oversteps, and claims to preempt reasons over which it has no jurisdiction,” it is defeated by non-excluded reasons; since the authority “failed to be legitimate” in the first place, it “failed to generate a dispositive duty.” ^[51]

This reply, however, is a mistaken in overlooking two types of non-excluded reasons entailed in Raz’s account of authority. If non-excluded reasons only refer to reasons that the authority does not claim to preempt, then it is true that these reasons set the limits of the authority’s jurisdiction. For example, the medical authority of Dr. M over the nurse N is based only on medical reasons. Thus, other reasons, like tennis skills, are non-excluded reasons that limit Dr. M’s authority over N. But non-excluded reasons also include reasons that the authority would claim to preempt but actually does not preempt. To see this point, suppose that based on the previous diagnosis Dr. M orders N to inject some medicines into the patient P. When N enters the ward, he finds that P has some new symptoms that contradict the previous diagnosis. Being a responsive nurse, N reports this situation to Dr. M rather than going on with the injection. This is so because, as Raz put it, “new evidence of great importance unexpectedly turns up” and reasons entailed in the unexpected evidence are not preempted by the original directive of Dr. M. To be sure, these reasons actually stop the original directive from being authoritative. But it should be noted that the fact that this particular directive is not authoritative anymore does not mean that Dr. M is, therefore, no longer a legitimate authority over N. For Raz, Dr. M is still a legitimate authority over N, for the conditions set out in his service conception of authority are still met,^[52] but the legitimate authority of M need not generate a conclusory duty for N to obey.

In fact, Raz himself argues against the view of conclusory duties. In explaining the force of duties (or obligations, in his terminology), Raz resists the temptation “to assign to them absolute or at least great weight.”^[53] Instead, he argues that duties have preemptive force, for they are exclusionary reasons, that is, reasons “for not acting on certain conflicting reasons.”^[54] But, again, a duty can still be outweighed or overridden by some other conflicting reasons that are not preempted by the duty in the first place. For Raz then, an overrideable duty with preemptive force is not an oxymoron, but a constitutive element of his service conception of authority, which says that the function of authority is to “serve the governed,” that is, to help the governed more likely to comply with independent reasons in practice.^[55] But for authority to function properly, authoritative directives need only to carry preemptive (but not absolute) force in guiding our practice.

Supposedly, this is where Applbaum’s second reason enters. Applbaum could argue that the moral reasoning entailed in Raz’s account of authority and duty is mistaken, for the alleged unforeseen yet conflicting reasons are “built in from the start.” In Applbaum’s view, given any specific situation S with the option of doing X or not doing X, a rational agent A should directly engage in balancing all relevant moral reasons with their different weights before A realizes whether she has a duty (with conclusory force) to do X or not.

Suppose, for argument’s sake, that A does moral reasoning precisely in the way Applbaum suggests. Then, if the moral calculation tells A that in S the moral reasons in favor of doing X override the moral reasons against doing X, then A will directly realize that she should do X, without bothering to consider whether she has a duty to do X. Of course, Appblaum could insist that there is nothing wrong with saying that in S, A has a conclusory duty to do X. But since A does not need the idea of moral duty in guiding her action, Applaum’s view of conclusory duties actually renders the idea of moral duty irrelevant in moral practice.

I admit that this is not a decisive objection against the view of conclusory duties. (The dispute between the view of conclusory duties and prima facie duties is protracted, and both seem equally capable of accounting for the phenomenon of a conflict of duties). But nor does Applbaum make a decisive objection against the view of prima facie duties. His accusation that the view of prima facie duties is built into “a false temporal story” is not well-grounded, for as my objection shows, without “a temporal story,” it is not clear why the idea of moral duty has any relevance in our moral practice.

Thus I will simply follow Raz in contending that it makes much sense to conceive moral duties as prima facie as long as they could play a mediating role by simplifying our moral calculation and making us more likely to comply with moral reasons. For moral duties to work in this way, however, they need not carry absolute force (on the contrary, as I have just shown, the view of conclusory duties fails to play this role), and prima facie duties with preemptive force suffice for this purpose.^[56] A good example of prima facie duties with preemptive force is a deontological reason that constitutes a moral duty. When there is a deontological reason to do X, we often need not^[57] second guess whether common negative consequences of doing X could be counted against the deontological reason; it simply preempts. But, of course, this deontological reason could still be overridden if the cost of doing X is fairly high.

Once we adopt the view of prima facie moral duties, we find that the legitimacy-entails-duty view does not incorporate any substantive moral argument, for it does not yet reach the final and determinate conclusion that civil disobedience is implausible under any circumstance.

V. Legitimacy entails immunity

Finally, let us examine Applbaum’s argument as to why the legitimacy-entails-immunity view necessarily incorporates substantive moral argument. To quote from Applbaum:

What about the reigning orthodoxy in international law, that a legitimate authority has immunity from outside intervention? Again, if Hohfeld’s scheme holds up, having a moral power is not the same as, and does not entail, having a moral immunity. … Just because B lacks moral immunity from A, however, A does not have moral immunity from the interference of some third party C. There is no conceptual route from having legitimacy—having moral power—to having moral immunity. … These all are connections that will have to be established by moral argument, not conceptual analysis.^[58]

But Applbaum’s argument cannot stand, due to the following two reasons. First, although it is true that “[j]ust because B lacks moral immunity from A, however, A does not have moral immunity from the interference of some third party C,” it does not follow that A’s moral immunity against C must be established by substantive moral argument. Here we must pay attention to the normative status of C with respect to A. The third party C can be subject to A, equal to A, or superior to A.^[59] If C, like B, is subject to A, then apparently A has moral immunity from C. If C is equal to A, A also has moral immunity from C, because it is an analytical truth that being normatively equal, one party is not able to change the normative status of the other party. Only if C is superior to A does A then lack moral immunity from C, just as B lacks moral immunity from A. But in the case of a legitimate state (A), there is no political entity that is superior to the state, that is to say, the third party C is either subject to A, or independent of A (hence equal to A). In both situations, the legitimate state A has moral immunity from C. Therefore, contrary to Applbaum’s claim, there is a “conceptual route from having legitimacy—having moral power—to having moral immunity.”

Second, note that when Applbaum refers to a state’s immunity against outside intervention he does not only mean an immunity in the Hohfeldian sense but also a claim against outsiders not to interfere with the state’s law enforcement. Were the view of conclusory duties still intact, this combination could still lead Applbaum to conclude that substantive moral argument is involved in the legitimacy-entails-immunity view, since the state’s claim of noninterference would place a conclusory duty on outsiders. Unfortunately for Applbaum though, it is a mistake, in the first place, to confuse an immunity as a second-order Hohfeldian advantage with a claim as a first-order Hohfeldian advantage.^[60] In the second place, even if the combination were correct, according to the view of prima facie duties I have just defended no substantive argument is involved in the state’s claim of noninterference against outsiders.^[61]