With his volume The Problem of Political Authority Michael Huemer has made an important contribution to the debate on political obligation and political authority. The first part of the volume is devoted to considering a number of classic theories that attempt to justify political authority by grounding it in social contract, democracy, consequentialism and fairness. All these theories are found wanting by Huemer, who therefore concludes that political authority lacks justification and that there is no duty to obey the law.[1] The second part is devoted to elaborating an alternative way of organizing social interaction: one that revolves around a system of private agencies and arbitrators which, according to Huemer, could take over all of the most important functions normally monopolized by the state. This alternative arrangement would have two advantages: first, it would be grounded in genuine contractual obligations, which is rarely the case with states; second, it would be subject to genuine market competition, which is likely to lead to better services at a lower cost.

Although there is much I disagree with in the book, its value can hardly be overstated.  Both parts are a model of clarity and contain a wealth of arguments to which I cannot begin to do justice here. Most importantly, Huemer never takes a cheap shot.  He always presents the positions he intends to criticize in their strongest possible form, and rejects them only once he has considered several strategies that might be employed to rescue them from his objections.

Here I concentrate on one of the main arguments that Huemer develops in the first part of the volume, one that goes at the heart of his challenge to political authority. The argument concerns the alleged right of political authorities to coercively impose whichever laws they decide to pass,[2] independently of whether these laws are good or bad (what Huemer calls content-independence; see p. 12). How can this ever be the case? Consider a situation in which we are to decide whether to φ or not. Suppose that φ-ing would be unjust and thus, before the state issues any directives, we are under an obligation not to φ. Imagine now two possible scenarios. First scenario: the state requires that we φ. How can that make a difference? Whatever justification for political authority we might be tempted to subscribe to, that will not change the fact that φ-ing is unjust. The fact that we consent to the authority, for example, cannot change that (if I consent to kill your cat, that doesn’t change the fact that killing your cat is wrong and I ought not to do it); nor can the fact that, say, we all voted in favour of φ-ing (if most people voted to kill your cat, that wouldn’t change the fact that it would be wrong to do so). If φ-ing was unjust before we were required to do φ by the authority, how can it become permissible simply because the authority issues a directive to φ?

Consider now the scenario in which the state requires that we do not φ. Certainly in this case we should not φ. But what has this to do with the directive we have received from the authority, Huemer asks? After all we were under an obligation not to φ even before the authority issued its directive. The fact that the state requires that I do not kill your cat does not seem to make any normative difference with respect to what I ought to do, since I ought not to kill your cat independently of what the state tells me to do. To be sure, the state can coerce me to prevent me from killing your cat. It can also punish me for doing so and extract compensation from me. But so can anyone else, Huemer argues, as long the intervening party acts according to what justice requires. (There will be limits to what can be done to me in order to prevent me from killing your cat or in order to punish me if I do.) Thus, it looks as if I should φ or refrain from φ-ing simply by looking at whether φ-ing is right or wrong. Most importantly, there is nothing the state can do to coerce me to φ or refraining from φ-ing that cannot also be done to me by anyone else.

II.

The main problem I see with this line of argument is that it presupposes that there is always an independently correct course of action that agents can decide to take while acting “on their own”, i.e. outside any institutional arrangement. According to this view, we have a number of natural duties that we can discharge in what we might call the state of nature (i.e. in a pre-institutional condition) simply by refraining from violating others’ rights and by aiding others in need (at least when we can aid them at a reasonable cost to ourselves). It is because we can discharge these duties by acting on our own, that the existence of authorities or political institutions can make no difference, according to Heumer, as to what we should do. An example that he resorts to in discussion is, I think, revealing. He asks us to consider a lifeboat which is taking on water. The water needs to be bailed out if the passengers are to prevent the boat from sinking. Bob, who has been designated by some as the person in charge to devise a solution, starts issuing directives to all the passengers. Surely here we need not obey anything Bob orders us to do, Huemer argues. What we need to do is grab a bucket and start bailing water.

I think that this way of looking at the issue is problematic for a number of reasons. To begin with, fulfilling our natural duties and respecting others’ rights often requires that we do not act on our own, but rather that we coordinate with others. To illustrate this point, consider a slightly different version of the boat case. Suppose that lifeboat is not taking on water. Rather the problem is that we need to move the move the boat to the land (none of us can swim). In this case, if each passenger starts rowing as hard as she can, we will not be able to reach the land. Successfully reaching the land requires that we coordinate in a number of ways: we need to decide which rowing technique to adopt, where we should try to direct the boat, how to organize turns in taking rests, etc. In other words, we need to set up some sort of cooperative scheme. Only if we act together in this way will we be able to reach the land. Acting on our own is not an option.[3]

Here it might be objected that I’m running together two different things: the need to act together and the need to set up some sort of cooperative scheme in which someone like Bob issues authoritative directives to the other members of the scheme. Even if I am right that we cannot discharge our natural duties by acting on our own, but only by coordinating with others, why is there a need to have an authority? If what we ought to do is point in a certain direction, row in a certain way, rest at certain intervals etc., what need is there for an authority? Why cannot we just contribute to whatever needs to be done without Bob telling us what to do? The problem is that typically in this sort of case there is more than one way in which we can coordinate, and people tend to disagree about how best to do so. Sometimes the disagreement will be simply due to cognitive limitations affecting some of us, but other times the problem will be more serious. Sometimes, we will reasonably disagree in virtue of the fact that we hold irreconcilable and yet reasonable moral beliefs. For these reasons, even assuming that a) we hold sound moral beliefs about what our natural duties are, and b) we aim to always act conscientiously on these moral beliefs, we will fail to discharge these duties unless we enter some sort of authoritative cooperative scheme. Authority is required in order to determine which of the many possible ways in which we can structure our interactions should be adopted, and to make sure that everyone complies with it.

States are in important ways analogous to the rowing scheme in my example. They are necessary for us to be able to fulfil our duties of justice and respect each other’s rights, because in order to do so we often need to coordinate our efforts. And despite acting with the best intentions, we will often reasonably disagree over what the best way to coordinate our efforts is, i.e. over which rules and principles should guide our attempts to act together according to justice. [4]

In fact, the problem goes deeper. The point is not only that people can reasonably disagree about what justice requires (which would be presupposing that there is a specific course of action required by justice, although we reasonably disagree about what this course of action is). Rather, the problem is that to some extent, justice in the state of nature is indeterminate. Not all natural duties of justice are as straightforward as “don’t kill cats for fun”. In some cases at least, it looks as if there is no pre-institutionally correct answer to the question of what justice requires until an authoritative decision has been made within a certain cooperative scheme. Think, for example, about the different sets of rules that could be adopted to effectively protect our interest in having private property (including rules whose function is to regulate market transactions, taxation or inheritance). Accepting that individuals have a natural right to property does not tell us much about which of these sets of rules we should adopt.[5] One of the functions performed by political authorities is to provide an answer to this question. As Joseph Raz puts it, ‘‘mediation through law serves the role of concretizing moral principles—that is, of giving them the concrete content they must have in order for people to be able to follow them’’.[6]

To be sure I am not saying that there are no duties that we can discharge by acting on our own in the state of nature. Our duty not to torture innocents and our duty not to kill cats for fun are good examples. My claim is rather that not all our natural duties are like this. If so, we should reject the assumption that we can discharge all of our natural duties of justice while acting on our own in a pre-institutional environment. The only way in which we can discharge some of our most important duties of justice is to leave the state of nature and enter a “civic condition”, in which we are subject to some sort of political authority. This is what justifies state legitimacy and political obligation.[7]

III.

But even if I am right about all this, have I vindicated the right of political authorities to coercively impose whichever laws they decide to pass, independently of whether these laws are good or bad? This might seem still doubtful for reasons that Huemer articulates in his volume. What if Bob in addition to ordering that we row in a certain direction, by using a certain technique etc., also orders that we flagellate ourselves to prove to him our loyalty, that we beat up children on the boat for his entertainment, that we dance for him or that we pay $50 to Sally who helped him to get elected?[8] Do I really want to say that Bob has a right to coercively enforce these commands? This view seems implausible. But isn’t rejecting this view tantamount to denying political authority? After all, one of the marks of authority –indeed, perhaps the most important mark of authority– is content-independence, i.e. the idea that authorities are entitled to enforce their directives independently of whether they are sound or inadequate.

To answer these questions we need to distinguish different ways in which the orders issued by a certain authority can be inadequate. Huemer tends to focus on cases in which a directive is inadequate because it is useless or unjust, but a directive can also be inadequate simply because it falls outside the scope of the authority. Indeed, this is the first thing to consider in deciding which obligations the authority is permitted to enforce, before we even start inquiring as to whether the directive is useless or unjust. My boss, for example, lacks the authority to decide which music I should listen to in my spare time or whom I should date. This has nothing to do with the fact that following his directives about which music I should listen to or whom I should date would be good or bad. We can imagine that my life would in fact go better if I were to follow his directives. The point is that even if this was the case, his directives would not be authoritative for the simple reason that they fall outside the scope of what my boss can demand of me. For the same reasons, Bob cannot successfully issue any authoritative directive that requires the other passengers to dance for him or to give Sally $50. Both of these actions fall outside the scope of his authority, which is primarily determined by the nature of the task that Bob has been appointed to complete (take the boat to the shore).

The scope of the authority is also delimited by obvious requirements of natural justice. For the same reason why we cannot acquire a duty to kill an innocent or beat up someone for fun (against his will) by promising or signing a contract, we cannot acquire a duty to kill an innocent or beat up someone for fun (against his will) as a consequence of the fact that we are so commanded by the authority. This is why Bob cannot successfully issue an authoritative directive that requires the other passengers to beat up children for his entertainment on the boat (p. 161).

But what about directives that do fall within the scope of the authority, but are wrong or useless? What if, for example, Bob chooses a route that is in fact longer than other routes available, thereby making us row for longer than it is necessary? (This is unjust, Huemer would argue, in that we are only under an obligation to row as much as it is necessary to take the boat to the land. Any extra rowing constitutes an unjustifiable interference with our liberty.) Or what if Bob comes to the mistaken conclusion that sitting in a particular way on the boat will make us go faster? My view is that in these cases Bob will retain his authority, despite the fact that he is requiring us to do unnecessary extra-work or to do something completely useless. Contra Huemer, I believe that authorities must be “accorded some leeway in the form of a content-independent entitlement to make rules as long as its rules are not too unreasonable” (p. 175. I suspect that the order to flagellate ourselves might be one that is too unreasonable). Authorities perform the valuable functions that justify their existence (described in section II above) despite the fact that they are not perfect. No authority is infallible. This is why infallibility should not be adopted as the standard of their justification.

Huemer tries to resist this conclusion by pointing at the fact that we do not seem to adopt the same view in relation to private agents. We do not expect private corporations to be perfect and yet we do not ascribe to them “a moral entitlement to periodically perform unjust or wrongful actions just as long as they are not too unreasonable. We recognize that a large corporation will sometimes do wrong, but we do not acquiesce in those wrongs. We condemn them when they happen and demand that the corporation make amends. In the same way, we should not acquiesce in wrongdoing by the state, however predictable it may be; we should condemn it when it happens and demand that the state make amends” (p. 175). I think Huemer is right about all this. We should not acquiesce in wrongdoing by the state, at least when they require that we do something profoundly unjust (i.e. something that goes beyond the margin of error discussed in the previous paragraph). When they do, we should certainly condemn them, and demand that they make amends. But this is all compatible with acknowledging that states do have political authority. After all, no matter how we understand the duties imposed by political authority, these duties will only be prima-facie, which means that they can be overridden by other moral obligations we have. This explains why we should not acquiesce in wrongdoing by the state. Although we have some reason to follow the directives issued by the state, we will have stronger reasons to disregard them whenever following them would require us to do something profoundly unjust.

Similarly, accepting the claim that states have a right to enforce directives that turn out to be unjust, at least when they are acting in good faith (as Bob when he chooses a longer route), does not prevent us from accepting the claim that we have the right to condemn state representatives and demand that they make amends when they are responsible for particularly serious injustices. One problem here is the tendency in the literature to run together the discussion of the legitimacy of states and the discussion of the legitimacy of governments. It seems to me that the two should be kept separate. A state might retain its legitimacy even when a particularly corrupt or inefficient government loses its own legitimacy. In that case, the government should be replaced and called to answer for its wrongs, without necessarily compromising the legitimacy of states.

Here it might be objected that while in principle the right to disobey, the right to condemn unjust and inefficient governments, and the right to require them to make amends can be preserved in the way I suggest, in practice we will rarely be in a position to exercise such rights. I agree. I also agree that many existing states are in fact so unjust that they cannot be considered legitimate in light of the justification I have outlined above. My disagreement with Huemer, however, is not so much about the legitimacy of existing states, but rather about the question of whether in principle states can enjoy legitimacy. Huemer argues that they cannot because no state can issue directives that are genuinely content-independent. I have tried to cast some doubts on this view.