State responsibility and counterterrorism

State responsibility and counterterrorism

Isaac Taylor*

Goethe University Frankfurt, Frankfurt am Main, Germany


It is widely thought that the international community, taken as a whole, is required to take action to prevent terrorism. Yet, what each state is required to do in this project is unclear and contested. This article examines a number of bases on which we might assign responsibilities to conduct counterterrorist operations to states. I argue that the ways in which other sorts of responsibilities have been assigned to states by political philosophers will face significant limitations when used to assign the necessary costs of preventing terrorism. I go on to suggest that appealing to the principle of fairness—which assigns obligations on the basis of benefits received from cooperative endeavours—may be used to make up the shortfall, despite this principle having received relatively little attention in existing normative accounts of states’ responsibilities.

Keywords: terrorism; remedial responsibility; capacity; effectiveness; liability; principle of fairness

Citation: Ethics & Global Politics, Vol. 9, 2016,

Copyright: ©2016 I. Taylor. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License, allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.

Published: 14 December 2016

*Correspondence to: Isaac Taylor, Forschungskolleg Humanwissenschaften, Goethe University Frankfurt, Am Wingertsberg 4, 61348 Bad Homburg vor der Höhe, Frankfurt am Main, Germany. Email:


There is a widespread norm against the use of terrorism. While some authors have suggested that terrorism—which is usually understood to involve the use of violence against innocent civilian targets with the aim of intimidating a wider population—may be justified in some instances, it is generally accepted that terrorist strategies are morally prohibited in at least most cases. This is true even if the goals for which terrorists are aiming for are themselves morally praiseworthy: one of the central tenets of just war theory, for instance, forbids the intentional targeting of non-combatants, even in pursuit of a just cause. Because of its prima facie moral impermissibility, terrorism is usually thought to require a response that will eliminate terrorist threats (or at least reduce them to an acceptable level) on the part of state actors, as well as others.

This is true of both state and non-state terrorism. Despite some legal definitions of terrorism ruling out the possibility of a state practicing terrorism by caveat, it is clear that governments can commit the same moral wrong that non-state terrorists do. Indeed, the ‘reign of terror’ following the French Revolution, in which the term ‘terrorism’ is thought to have originated, was practised by a government. In recent years, though, the question of what should be done when faced with non-state terrorist groups like al-Qaeda and ISIS has gained particular urgency, and it is this that will be the subject of this article. The exact response that is appropriate is the subject of ongoing controversy among political actors.

One source of controversy about how to respond to non-state terrorism concerns the sorts of actions that should be taken when attempting to prevent attacks and degrade terrorist organisations. At the time of writing, there is fierce disagreement about the correct response to ISIS. While we are arguably seeing an increased appetite for the use of ground troops by Western states, some politicians maintain that the West should limit itself to airstrikes, and other claim that foreign countries should not get involved at all.

My focus here, however, will be on a second source of controversy. Assuming that some actions need to be taken against non-state terrorist groups, and assuming that these actions involve costs for those carrying them out, who in particular should take up those costs and in what proportion? This question, too, has been a sticking point in debates about the appropriate response to ISIS. During a speech in January 2016, US Secretary of Defence Ash Carter called on partners in the US-led coalition to pick up the slack in counterterrorist operations: ‘there can be no free riders’, he said.1

The need to overcome these disagreements is crucial. In the absence of a widely accepted account of who should do what, there is a risk that important tasks will be left undone as each state waits for the others to pick up the costs. This of course provides one reason (among many) to create an institution that would coordinate the counterterrorist efforts of different states. Such an institution could set out an authoritative assignment of obligations to states. However, even if such an institution were to be set up, it would still need to come to a decision on who should do what. Ensuring a fair distribution of the costs of counterterrorism, thus, also points towards the need to find an account of what each state is required to do.

My aim, then, is to investigate how we should go about making this decision. On what basis or bases should we assign states responsibilities to conduct counterterrorism?2 While this has not been explicitly addressed by political philosophers in any detail, elsewhere in the literature on global justice we find a set of principles for assigning responsibilities to avert other harms—from humanitarian crises to the effects of climate change. It might be thought that the conclusions that we reach in these other areas can be straightforwardly applied to the harms of non-state terrorism; we might use the general principles that assign responsibilities to conduct humanitarian intervention or prevent dangerous climate change to assign counterterrorist responsibilities to states. In what follows, I will examine to what extent these principles can be used for this purpose. While they face greater limitations in assigning counterterrorist responsibilities than other sorts of responsibilities (including responsibilities to prevent state terrorism), I will go on to suggest that we may be able to assign responsibilities using principles not found in the global justice literature. In particular, I suggest that responsibilities here can be understood in part as reciprocal obligations incurred as a result of benefitting from existing counterterrorist regimes and assigned to states accordingly.

Some may view the project of assigning counterterrorist responsibilities to states as unnecessary and dangerous. Many states are quick to act—and, some would say, too quick—in attempts to combat terrorism without any external inducement. If anything, it might be thought, we should be curbing states’ actions here, not encouraging further ventures by loading responsibilities on them.

But assigning responsibilities to conduct counterterrorist operations to states can have a restraining effect as well. The reason why most states are currently motivated to conduct counterterrorist operations might be primarily that of self-interest. Conducting operations that aim at protecting one’s own citizens from attack leads to either the imposition of high costs on foreign populations—the use of drone strikes against terrorist leaders that create significant levels of collateral damage to civilian bystanders may be brought up in this context—or operations that simply shift terrorist threats away from their own populations to others without addressing the cause of the threat—proposals in a number of countries to annul the citizenship of ‘home-grown’ terrorists potentially has this result.

Viewing counterterrorism as a responsibility as well as a pragmatic policy, though, may lead to us viewing these sorts of actions in a different light. In particular, if conducting counterterrorism is what states owe to others as a matter of obligation as well as what they should do to protect themselves, efforts that achieve one nation’s security while rendering others insecure will be ruled out. An account of states counterterrorist responsibilities, such as the one I will propose below, can be welcomed by those who think some existing practices need to be scaled back.


Imagine that an individual is threatened with a harm, and there are a number of others who can potentially avert this harm at a cost to themselves. Perhaps, a child is drowning in the ocean, and a number of people are standing on the shore. While each stands a good chance of succeeding in a rescue attempt, they would incur a risk to their lives by doing so. If we think that someone should rescue the child, though, we need to know how to pick out one rescuer of the group. How should the responsibility to save the child be assigned?3

One way of approaching the question is to appeal to what I will call ‘non-relational bases’. Advocates of this approach look at certain characteristics of each potential obligation bearer apart from the relationships that they are in and assign responsibilities to agents who have these characteristics to the greatest extent. They might say that the most effective obligation bearer should do this: the strongest swimmer should undertake the rescue, perhaps. Or they might say that the agent with the greatest ability to carry out the rescue should do so: the strongest swimmer might be afraid of the water, and so someone who will take the rescue attempt in their stride may be a better candidate. Perhaps, a fully satisfactory decision will take both factors into account when picking out a rescuer.4

This approach has been used by a number of political theorists attempting to assign urgent responsibilities to states. Many problems threaten individuals with harm, and a number of states may be in a position to avert these. Some think that the effectiveness of states in averting these risks or the ability of states to incur the costs of doing so is important to consider when assigning responsibilities. James Pattison, for example, argues that the agent who can most effectively undertake humanitarian intervention should (usually) be the one to do so because of this.5 Others call for an assignment of responsibilities based on ability. Simon Caney claims that a state’s ability to pick up the costs of preventing or mitigating dangerous climate change (e.g. because of its greater wealth) should at least in part ground an obligation to do so,6 and Matthew Gibney argues for distributing refugees among states primarily based on each states’ integrative capacity.7

The approach might also be used to assign some counterterrorist responsibilities to states. We might be able to identify a state that will be more effective than others at conducting a particular operation—due to its superior military resources or intelligence services, for instance, or because it is located close to terrorist hotspots—and assign the obligation to it. Or we might assign the obligation to the state that has the greatest ability to take on the burdens of a particular operation—perhaps because of its greater wealth, or because it faces lower opportunity costs when it diverts resources towards counterterrorism.

Features such as effectiveness and ability should undoubtedly affect our views on the content of agents’ responsibilities. We do not want to assign responsibilities to avert harm to agents that will be completely ineffective at doing so, nor do we want to burden them with responsibilities that they will find it difficult to carry out, other things being equal. However, there are two shortcomings with this approach when it is used to assign counterterrorist responsibilities that should lead us to look to supplement it with other considerations.

The first shortcoming is one that applies to attempts to assign any responsibility on non-relational bases. This is that it might be unfair on those saddled with obligations. Why should the agent who happens to be most effective at averting harms be asked to take on costs every time the need arises?8 And why should agents who have acted prudently be asked to make sacrifices just because their prudence now means that they are most able to take on costs?9 If we were able to spread the costs among a number of different agents in proportion to their ability to contribute, then the unfairness might be less stark. But this will not always be possible if the costs are ‘bulky’ and cannot be easily divided (perhaps some operations are best done by single states acting alone). Many think that there are limits to how much an agent can be asked to take on in assisting others with whom they are not in a relevant relationship.10

There is an additional shortcoming that arises if the approach is used to assign counterterrorist responsibilities in particular. Suppose we want to assign responsibilities for a particular counterterrorist operation on the basis of effectiveness, but there might be a number of states that will be equally effective in carrying it out. Sometimes all an effective counterterrorist strategy needs is an injection of resources, and many states might have enough to do this. Appeal to effectiveness works best in cases where different states clearly vary in effectiveness. Pattison’s approach of assigning the costs of humanitarian intervention on the basis of effectiveness works well because he limits his concern to the question of who should actually conduct a humanitarian intervention, and states’ effectiveness at this task is almost certain to vary; additional tasks, such funding the intervention might properly be distributed in another manner.11 Because successful counterterrorism is likely to involve at least some tasks in which a number of states possess equal effectiveness, appealing to effectiveness to assign responsibilities here is likely to be more indeterminate.

Turning to ability will remove indeterminacy of this sort but generates additional indeterminacy in other areas. Suppose that a state that takes on a particular counterterrorist operation will have to put some of their citizens at greater risk in doing so—perhaps terrorists are likely seek reprisals against the state that is carrying this operation out. How are we to measure the ability of each state to take on this risk? By the psychological strength of the population to live with the risk? By the ability of their economy to recover if there is an attack on their soil? Both options seem to ignore the significance of the cost that a state is being asked to take on. In Caney’s discussion of responsibilities to prevent climate change and Gibney’s account of who should take in refugees, the costs involved are fairly clear: reducing greenhouse gas emissions and supporting newly arrived refugees, respectively, both of which we can usually come to straightforward conclusions about which states have the greatest ability to take on these costs. The many different sorts of costs that might be involved in conducting counterterrorism mean that judgements of relative ability are more difficult to make.

Grounding counterterrorist responsibilities on non-relational bases then, although appealing in some cases, has significant shortcomings if it is the sole approach to assign costs to states. An acceptable approach to allocating all the tasks that need to be undertaken here, it seems, must incorporate other bases at least to some extent.

Relationships to harms

In addition to non-relational bases like effectiveness or capacity, theorists discussing what states should do about harms across the globe also often appeal to certain sorts of relationships to ground responsibilities. Taking a cue from many jurisdictions’ tort law, these theorists argue that, just as an individual’s responsibility for harms is often assigned on the basis of that individual being somehow connected to the causal processes that caused those harms, states should be held responsible for harms that if they are connected to harms in similar ways. There are three main connections that are appealed to.

One way in which an agent might be taken to be responsible for a harm is through having culpably caused it. In tort law, an individual who acts in a blameworthy manner and causes a harm is often held accountable for remedying that harm. People who cause damage while driving negligently might be required to pay for that damage, for instance. In a similar way, many theorists think that the obligation to conduct humanitarian intervention, when necessary, should be assigned on the basis of culpability of a state.12 Cécile Fabre, in this vein, argues that if Belgium and France were both at least partially to blame for the Rwandan genocide in 1994 (because of the former’s role in fostering an environment of ethnic division and hatred there when it was a colonial power and the latter’s military and strategic support for the Hutu regime), then it is these states that should have been held primarily responsible to intervene and prevent the genocide.13 While their states were the agents that caused harm here, many theorists believe that citizens can be held responsible for the outcomes of their states’ actions so long as certain conditions are met that allow us, in some sense, to ascribe those actions to the citizenry as a whole.14

In many legal systems, one need not be blameable in order to be held responsible for a harm. Regimes of strict liability impose obligations of redress to individuals when they merely cause a harm through undertaking a permissible activity. A careful driver who nonetheless causes an accident might be held liable for the costs, for example. More generally, many have a moral intuition that blamelessly causing a harm can be sufficient to ground an obligation to rectify it: David Miller gives the example of a pedestrian who accidently trips and knocks someone else over. It seems right that this pedestrian should be the one to help the latter up, even though he or she was not blameable for the collision.15Mere causality has also been appealed to in order to assign responsibility for global harms, such as humanitarian crises. Erin Kelly, who is sceptical of attempts to hold citizens blameable for their states causing these crises (after all, it is often political elites in those states acting with questionable democratic credentials that is the cause), nonetheless thinks that those citizens might be under an obligation to pick up the costs of humanitarian intervention because of their causal role in producing the crisis. They support their states in a number of ways and so make the political elites’ actions possible.16

Causing a harm, whether one is blameable for this or not, is not the only way in which systems of tort law ascribe responsibilities to individuals. Principles of unjust enrichment ascribe these responsibilities to individuals on the basis of their having benefitted from injustices that harms others. Although they did not act wrongly, the mere fact that someone gains from a wrongful act is thought to render their gains ‘tainted’ and generate an obligation incumbent on them to compensate those who have lost out from the injustice in question (or perhaps even injustice in general) with the gains.17 Something like this idea has been used to argue for wealthy industrialised countries bearing the costs of preventing dangerous climate change18 or paying reparations for slavery.19 While the current citizens of these countries may not have acted wrongly—it was their ancestors who over polluted and supported the slave trade—, they nonetheless may have benefitted from these practices financially, and so, it is argued, they must use some of their gains to repair the damage that those practices caused.

There are, then, three main ways in which states are thought to come under responsibilities on the basis of their relationships with others: through culpably causing harms, through innocently causing harms and through benefitting from wrongdoing that causes harms. Can any of these relationships between agents and harms ground the counterterrorist responsibilities of states?

Let us begin with culpability. There are a number of ways in which states might be said to causally contribute towards terrorism in a blameworthy manner. In particular, there are three major sorts of wrongful actions taken by existing states that have been said to contribute towards terrorism:

  1. direct funding of terrorist organisations that provides them with the resources to carry out attacks,
  2. ill-judged foreign policy that creates both the institutional vacuums and sense of grievance that allow terrorist organisations to thrive,20 and
  3. international economic policy, supported by many states through international institutions like the IMF and the World Bank, that leads to widespread poverty and thus creates an effective breeding ground for terrorism.21

Some commentators have argued that the states that cause terrorism in this way should pick up the costs of preventing it. For example, it is often said that, because the United States’ invasion of Iraq in 2003 (a war that is taken to be unjust) led to the rise of ISIS, the United States should take on the lion’s share of the costs of degrading ISIS now. But just how far can basing counterterrorist responsibilities on these sorts of causal links go?

There are two pragmatic obstacles with this strategy, which, although do not tell against it decisively, are worth noting here. The first is that demonstrating states’ culpable actions cause terrorism in this way will often be difficult. State sponsorship, in particular, is hard to prove; indeed, one of the main reasons why states often resort to these measures to promote their goals is that they retain plausible deniability. The same is not true when states themselves practice terrorism, and so this shortcoming relates specifically to non-state terrorism.

While it may be easier to prove that states have engaged in unjust foreign or economic policies, it will sometimes be difficult to take the further step and link these actions to terrorism. While a link is often posited between poverty and terrorism, for example, many of those who join terrorist organisations can be described as bourgeoisie. This was true of the Russian anarchist organisations that were active in the late nineteenth century as of more recent examples such as Aum Shinrikyo, who were responsible for the sarin attack on the Tokyo subway in 1995.22 It is also the case that modern Islamist terrorist organisations such as al-Qaeda and ISIS draw significant members from middle-class backgrounds: a 2006 study found that over half of those who perpetrated the most significant Islamist terrorist attacks since the early 1990s had college degrees.23 Thus, even if we can show that states contribute to global poverty (and perhaps hold them responsible for reducing poverty in developing countries, as some theorists want to24), whether their actions also started a chain reaction that fostered terrorism is much more questionable. A more reliable trigger for violent extremism is actually frustrated expectations: members of the rising middle classes fail to do as well in socio-economic terms than they believed they would.25 While it may be the case that the same mechanisms that cause poverty have some role to play in developing these disappointed expectations, more complex arguments are needed to establish the link.

The second obstacle is that assigning counterterrorist responsibilities to states that culpably cause terrorism is that this will often result in important tasks not being carried out. States implementing neoliberal economic policies or sponsoring terrorism will deny that they are doing anything wrong, and so are unlikely to be moved by claims that they should pick up the costs of preventing terrorism that they had a role in fostering. Even in an international institution assigned with distributing responsibilities to states, fundamental disagreements about the permissibility of different actions are likely to block at least some assignments of responsibility based on culpability. Of course, this is not to say that we should take state leaders’ word for it when they say that they are doing nothing wrong; we can and should criticise wrongful policies when we see them. And if we are able to load the costs of counterterrorism on them against their will, this might be legitimate. But in cases where we require states to take the initiative in order to prevent terrorism, asking states to do so on the basis of their wrongful actions is often unlikely to help.

Before I offer more principled objections to this strategy, we can consider whether the second sort of basis discussed above—that of strict liability or responsibility based on mere causality of harms—can fare any better as a way of assigning counterterrorist responsibilities. The following sorts of action, which it has been claimed also cause terrorism, might be of this sort:

  1. Inadvertently advertising an image of prosperity that is out of the reach of many individuals across the world can lead to some being drawn to terrorism out of frustration.26
  2. Western states promoting liberal values, such as democracy and gender equality, makes some religious fundamentalists feel isolated, leading them to turn to terrorism.27 It also makes terrorism pay: it is generally thought that democracies will be less willing to suffer losses from terrorism and so will capitulate to terrorists demands faster.28

Can responsibilities to prevent terrorism thus be assigned to states who are involved in these causal relationships?

Miller himself noted that mere causality may not be able to play ‘more than a minor role’ in the assignment of responsibility.29 And when the costs become high, this basis might fall out of the picture. It appears plausible to say that I should help up a pedestrian who I have accidently knocked over, but to go any further than this and say, for instance, that I should foot his/her hospital bills looks more contestable. But defenders of strict liability principles might be willing to say exactly that and thus argue that assigning responsibilities to states on the basis of mere causality will often be the correct thing to do. Even if this is correct, there are issues with this approach when it is used to assign counterterrorist responsibilities specifically.

In particular, there are two more general problems with assigning counterterrorist responsibilities to states that cause terrorism, whether through morally wrong actions (such as 1–3) or morally permissible actions (such as 4 and 5). The first is that the actions in question do not automatically cause terrorism. Rather, for terrorism to result, intervening agency needs to be present. It is not simply the states acting in the ways described in cases 1–5 that causes terrorism but also a set of other agents (namely, the terrorists themselves) choosing to react in unjustified ways that harm individuals. This fact might remove (or at least diminish) states’ responsibilities.

To see the force of this point, we can alter the details of Miller’s example. Suppose now that whenever I stumble in the street, an enemy of mine will push over a random pedestrian. Do I bear any special responsibility to help these pedestrians back to their feet merely on the basis of my causal role in their being pushed over? It appears not. Similarly, states might not bear any special obligation to avert terrorism simply because they played a causal role in those harms coming about.

This will not always be the case. In example 1, the sponsoring state can plausibly be described as an accomplice to terrorism and might on this basis be asked to take on the costs of preventing the harms that result. In examples 4 and 5, though, no such description is appropriate. These cases are more like the modified example of the pedestrian set out in the previous paragraph, and thus, no special responsibility appears to attach to the states in question here.

Cases 2 and 3 are less clear cut. Whether these states can properly be held responsible as accomplices will depend on how we understand complicity, and I will not discuss this issue here. Nonetheless, if (as I believe) cases 4 and 5 are representative of a significant number of causal processes that cause terrorism, there will be many instances in which no counterterrorist responsibilities can legitimately be assigned on the basis of culpability or causality. This is not true with respect to other global harms. Often, there is no intervening agency when states cause these harms (as when states cause climate change by emitting greenhouse gasses), and even when intervening agency does play a role, it will often still be appropriate to take states to be accomplices or at least to have acted in ways that properly leave them liable to costs (as in the case of France and Belgium’s role in causing the Rwandan genocide, which Fabre discusses).

A second problem with assigning counterterrorist responsibilities on these bases is the following. In many cases, there is no state that can be said to have caused terrorist harms—either in a blameworthy or innocent way. This is particularly true when we consider how terrorism has evolved over the last few decades. In an important study, Peter Neumann argues that we have seen a significant shift in the nature of terrorism beginning in the 1990s. Among other things, terrorists’ aims have changed. While ‘old’ terrorist groups were often motivated by political ideologies (like Marxism) and nationalism, in ‘new’ groups we find political goals being replaced by religious ones.30 Even the IRA, who drew their support from a largely Catholic constituency, did not make religion a central part of their guiding ideology,31 but the likes of al-Qaeda and ISIS give religion a more central place in their worldview. Attacks against civilians are often justified through questionable readings of the Quran, and these groups sometimes fight for the implementation of a stricter interpretation of Sharia law.

If it is this that partly explains why individuals turn to terrorism, then states’ roles in causing terrorism will be significantly diminished: there is little they could have done or avoided doing that would ensure that terrorist groups did not thrive. While the ongoing British presence in Northern Ireland clearly played a large causal role in the rise of republican terrorism, it is more difficult to point to any actions of states that play a comparably large role in causing much modern Islamist terrorism. Consequently, the degree to which we can assign counterterrorist responsibilities on the basis of states having caused terrorism will be limited, at least in recent times.

As we saw above, the idea of unjust enrichment is also used to assign responsibilities to states. And some states do sometimes benefit from the unjust actions of terrorists. Consider this final case:

  1. When terrorists attack one country, it faces a number of economic costs, such as reduced tourism. Tourists may go to other countries instead, thus providing them with additional revenue.

These sorts of effects might happen as a result of much terrorist activity. Following the shooting of 38 tourists in Tunisia by terrorists in 2015, for example, tourism minister Selma Elloumi Rekik estimated that the country would lose more than US$500 million as a result of decreased demand for tourism.32 If neighbouring countries increased their tourism revenue as a result, it might be argued that they are under an obligation to give up their benefits to prevent future terrorist attacks.

One problem with this thought is that theorists who accept unjust enrichment as a ground of obligations generally hold that it only applies to cases where the benefits are the ‘automatic consequences’ of the harm in question.33 There is a significant difference between unjust gains that are derived ‘in the course’ of wrongdoing, rather than ‘as a consequence of it’, and the principle of unjust enrichment loses its intuitive force when benefits are of the latter, more causally distant, variety.34 Now the benefits of increased tourism revenue are not entailed by the wrongdoing itself; they only come about because tourists decide to go elsewhere after terrorist attacks and are thus a consequence of wrongdoing. Neighbouring countries are unlikely to benefit merely from civilians being killed. Of course, if terrorists fight for political concessions that will be advantageous for other states (perhaps the latter fund them to carry out the attacks), they will benefit in this way as well. But, again, these do not look like automatic effects of terrorism. If responsibilities could be assigned in such a case, it would have to be because of the sponsorship itself rather than the benefit received. This assignment would thus be based on causality rather than unjust enrichment (and thus face the problems with assigning responsibility on the basis of causality that I discussed above).

A further problem is that these sorts of situations are probably not going to be zero-sum games. Terrorism may make everyone worse off globally. Following the 9/11 attacks, for example, consumption fell across the world, not just in the United States, and many different countries’ tourist industries suffered.35 If nobody benefits from terrorism as much as the costs of stopping it, appealing to unjust enrichment will be of little help in assigning those costs. This problem does not attach to the other cases in which theorists have appealed to unjust enrichment as a basis for responsibility. Present-day economic inequalities that have resulted from slavery, for instance, can usually be fully corrected for by those who are on the winning end of these inequalities contributing their undeserved gains.

Assigning counterterrorist responsibilities to states that have either culpably caused, innocently caused or profited from terrorism, then, can only go so far. Not only does the exact nature of the relevant relationships between state actions and terrorism render the assignment of responsibilities questionable here but also with respect to much terrorism, no such relationships exist in the first place.


In this section, I will consider another basis on which counterterrorist responsibilities might be allocated: benefit from the provision of goods by others. This basis has been thought capable of grounding political obligations of citizens within a state, but it has been used far less to assign obligations to states. I will argue, however, that the nature of modern terrorism means that this is a promising way in which to assign some counterterrorist responsibilities.

It is not only through tort law that individuals are thought to gain responsibilities to rectify or prevent harms. Many believe that citizens of a state can be under obligations to help each other avoid harm through obeying the law, paying taxes to pay for police and national defence, and so on. There are a number of competing accounts purporting to explain why individuals have these obligations, but one of the most influential appeals to the principle of fairness. This principle, roughly speaking, says that when one benefits from others’ contributions to cooperative endeavours, one is under an obligation to make a contribution as well. Or, in H.L.A. Hart’s seminal formulation:

when a number of persons conduct any joint enterprise according to rules that restrict their liberty, those who have submitted to these restrictions when required have a right to a similar submission from those who have benefitted by their submission.36

This implies, for example, that when individuals’ co-citizens pay taxes towards policing services, they may be under an obligation to pay taxes too when they benefit from these services. This is a rather different principle than those found in tort law; obligations arise out of beneficial cooperative endeavours rather than one’s harmful actions (or the benefitting from others’ harmful actions).

The principle of fairness has not been universally accepted. Robert Nozick has presented a number of powerful counterexamples to the idea that one can come to be under obligations merely by benefitting from the efforts of others. It is implausible to suggest, says Nozick, than an individual who thrusts a book into your hands is owed payment for benefitting you.37 A commitment to individual liberty leads us to think that we should leave it up to the prospective recipient of the book about whether or not to pay for it. Yet, the principle of fairness, as presented above, implies obligations to pay up here and is thus, for Nozick, implausible.

Those who wish to defend the principle of fairness as a way of justifying political obligations have responded by placing a number of limits on its application. I will not discuss whether they succeed here. But we can note that the goods, receipt of which most plausibly generates obligations to contribute towards meeting two conditions: they are goods that are difficult to exclude people from and which are of such importance that we can presume that people benefit from them, and thus seek them out were they not to receive them passively.38

A book is not this sort of good. The bookseller in Nozick’s example could have quite easily kept the book to herself and thus excluded the potential recipient from the benefits. And we cannot presume that the recipient would have sought out the book were he not to receive it; he might have spent his money on another book, or something altogether different. A commitment to individual liberty implies that we should allow the prospective recipient to decide for herself or herself whether or not to pay for the book, and it is quite possible in this case.

When a good meets these two conditions, however, limiting the liberty of individuals and requiring that they pay for it—as the principle of fairness does—looks less troubling than in the case of trivial and excludable goods. While leaving the choice about whether to pay for an excludable good like a book up to the potential recipients does not come with any significant costs, leaving the decision about whether to contribute towards public goods is likely to have consequences that are more problematic. Because individuals will receive these goods whether or not they contribute, they will have an incentive not to do so, and to ‘free ride’ on the efforts of others. This in itself might be viewed as unfair, but if enough people act on this incentive, the good will not be supplied at all, leavening everyone less well off than they could be. Pointing to the possibility of both sorts of costs arising when contributions to non-excludable goods are voluntary is the first step in the argument for holding individuals under an obligation to pay for them.

But this cannot be the end of the story. Individuals might claim that they do not care for the non-excludable goods that they are being asked to pay for, or that the cost that they are being asked to pay is too high. When non-excludable goods also meet the second condition identified above—being of significant importance—their doing so lacks plausibility. Because national defence, for example, plays such an important part in most peoples’ lives, very few, if any, people can argue that they would rather do without it. It thus looks less problematic to require people to pay for this, because, were they given the option, they would seek out this good and take on the associated cost.

The principle of fairness is most often invoked to justify obligations to contribute towards goods provided by a state for the benefit of its citizens (like national defence); far less has been used to justify obligations of states to contribute towards goods provided by the international community.39 It is my claim here, though, that responsibilities to conduct counterterrorist operations can be assigned to states through this principle too. States (and individuals within them) benefit from many existing international counterterrorist practices just as individuals within a state usually benefit from national defence. And, as I argue below, these benefits meet the two conditions that are thought necessary to ground genuine obligations.

The first condition was that the benefits be non-excludable. While some measures that states might take to protect their citizens from terrorism (like closing borders, for example) do not produce benefits that spill across borders, it will be my claim here that many more actions that states take in order to ensure the safety of their citizenry against terrorist attacks unavoidably benefit individuals the world over.

In earlier years, terrorism tended to be limited to within national borders. Terrorists’ grievances were generally with specific national governments, and they operated largely within those governments’ territories. Another aspect of the ‘new’ terrorism that Neumann discusses, though, is the expansion of operations beyond a single geographic territory.40 Al-Qaeda’s internal structure was highly decentralised, and it had ‘franchises’ in many different countries.41 Furthermore, its attacks were widely dispersed; its target was not a particular government or state apparatus, but rather a large and nebulous collection of states.42

ISIS may appear to represent a return to the more territorially limited sort of terrorism. Its immediate goal is the maintenance and gradual expansion of its so-called caliphate. We should not assume, though, that this means that the harms its members cause will be limited to their immediate environment. This is an organisation whose motivating ideology permits only temporary peace treaties and calls for war to be waged at least once a year (and recent attacks in Europe suggest that this is not simply idle rhetoric).43 And even when a particular state is not directly targeted, there may be knock-on effects that they feel from international in security more generally. The effects that terrorist attacks in one country can have on a globalised economy and society are beginning to be felt.

What is the significance of this trend? When states conducted counterterrorist operations against the organisations of ‘old’ terrorism, they generally only benefitted the population of a particular state. With ‘new’ terrorism, however, they often benefit individuals in many different states, because the harms that are being prevented and the risks that are being reduced would otherwise fall on diverse populations. Counterterrorism has thus, to some extent at least, become a ‘global public good’ in that, through supplying it, we cannot easily exclude individuals in different countries from the benefits. It meets the requirement that the benefits be non-excludable.

What about the second condition that the benefits produced be sufficiently important to individuals’ well-being? The benefits provided through counterterrorism certainly look fairly crucial; without the security that counterterrorism brings, people around the world may be in something approaching the state of insecurity that they would be in if they resided in a country without national defence. We can thus presume that people benefit from these goods, and that they would seek them out were they to be excluded from them. This condition is thus met too, and so responsibilities to conduct counterterrorist operations can be assigned to states based on the benefits they receive from current efforts being undertaken by others.

As I said above, the principle of fairness has used to assign obligations to states far less than it has obligations to citizens within states. One reason for this might be because it is thought that there are relevant differences between the way in which individuals receive benefits from their state and the way in which they receive benefits from foreign states, which warrants ascribing obligations in the former case, but not in the latter case. I will discuss three potential disanalogies of this sort below and argue that they are not capable of undermining my argument for assigning obligations to states via the principle.

First, it might be thought that, while individuals have no other way of receiving the important benefits provided by their own state, states have alternative ways of getting the benefits that other states provide them through conducting counterterrorist operations. In particular, they could implement a number of domestic policies—including, for instance, tighter border controls, the exclusion of migrants and widespread surveillance within their own territory—that would reduce their risk of exposure to terrorism to an acceptable level without the help of others. George Klosko has suggested that when agents can demonstrate that they can attain benefits in alternative ways to public provision, they are under no obligation to continue contributing to that public provision.44 We would not think that an individual has violated any obligation if he or she refuses to participate in a communal project to bring clean water to his or her neighbourhood if he or she prefers to dig a well for himself or herself, for example.45 It is perhaps this thought that motivates Klosko when he denies that obligations based on reciprocity can extend across borders, on the basis that ‘while the public goods noted in regard to the intrastate realm are necessary for acceptable lives, such interstate public goods are not similarly necessary’.46 Because states can guarantee acceptable lives for their citizens without help, it is only to domestic public goods that individuals owe a contribution.

Looking at what is necessary today to reduce terrorist threats should lead us to reject the empirical assertion in this argument. Because ‘new’ terrorist organisations operate on a more transnational scale than their predecessors, the ability of states to combat them alone has diminished. ‘There is nothing new … in calling for more international cooperation’, says Peter Neumann, ‘but the call has become more urgent with the rise of transnational terrorism … the era of new terrorism requires more—as well as more demanding—interaction with foreign countries’.47 There is now no alternative way to receive the benefits of security without the help of foreign states, and so when a population receives those benefits, the principle of fairness generates responsibilities.

The second potential disanalogy is the following. When many states provide domestic public goods, the benefits and burdens of these goods are often shared roughly fairly. Adjusting tax systems ensures that each individual pays their fair share towards the services that their government supplies. The global public good of counterterrorism, by contrast, is often supplied with grossly unjust distributions of costs. Many counterterrorist operations, as I said earlier, impose high costs on certain individuals. For example, the practice of drone strikes has been thought to place unacceptable risks on foreign civilians in order to secure the safety of those in the country conducting them, and state surveillance practices threaten the civil liberties of some ethnic and religious groups more than others.48 Many theorists hold that it is only in just schemes of cooperation that the principle of fairness can ground operations; it is inapplicable in schemes that distribute benefits and burdens unfairly.49 Does this mean that the principle cannot ground benefits on the basis of benefit from counterterrorist operations?

I think that the requirement of schemes to be justly organised for obligations to be created is too strong. As A. John Simmons points out, although an unjust distribution of benefits and burdens in a cooperative scheme might plausibly nullify any obligation on the part of those on the losing end, it is less clear whether those who are part of such schemes yet are not treated unjustly lose all responsibilities to contribute.50 There may be good reasons in particular cases why they should not contribute of course: If their contribution will simply exacerbate the existing injustice, then they should refrain. But if they can make a contribution that alleviates the injustice—by focusing their efforts on reducing the burdens for those unjustly treated, for instance—then the fact that they have benefitted from the scheme seems to provide a good reason for requiring them to do so. Even if existing counterterrorist operations generate unfair distributions of harm, then, the principle of fairness can still explain why some states have obligations to take on more responsibilities.

The final potential disanalogy relates to the nature of costs that are assigned. In discussions of the principle of fairness in the domestic context, the costs are usually taken to be financial. It is the payment of taxes towards public goods that is being argued for as an obligation. While financial costs need to be taken up in the global context too—some material resources need to be injected for an effective international counterterrorist strategy—this does not exhaust the sorts of responsibilities that states take up. At some point, it might be thought, military action needs to be taken against terrorist groups like ISIS, and in addition to financial costs that would be involved in this (paying for weapons, transportation, and so on), other sorts of costs need to be met. In particular, some soldiers will have to undergo a risk of harm, and some may even be killed. And civilians of a country which engages in counterterrorist operations may find themselves targeted, and thus at a greater risk of being harmed. Can the principle of fairness ground responsibilities to take up these sorts of costs?

There seems to be little that is objectionable with increasing general security of a population at the cost of each of them taking on a smaller and distinct security risk. If the degree of security that is obtained is important enough, we may be able to presume that individuals would be willing to take on the costs, and so the argument for obligations for taking on the costs that the principle of fairness rests on seems to apply here too. Issues arise, however, when costs to security are ‘bulky’ and cannot be evenly spread over individuals. A small group of soldiers may have to take on the high risk of being harmed to provide for the security of others. But, in my view, so long as these risks are assigned in a fair way among the population who have the obligations to take them up, there is nothing objectionable about this. It is always the case that states require some individuals to take up the risks associated with service in the armed forces, and as long as those individuals are chosen through fair procedures, this looks acceptable. What procedures should be used is of course open to dispute—we could fill the armed forces through conscription or with volunteers, for instance—but the point here is that there are plausible ways in which indivisible costs can be fairly distributed among a population.

In this section, I have argued that responsibilities to conduct counterterrorist operations can be understood, in part, as reciprocal obligations, and assigned to states on the basis of their having benefitted from existing international counterterrorist efforts. Ash Carter, who I cited at the beginning of this article, called those who were not taking up the slack in international efforts against ISIS ‘free riders’, and we can now see that this language is appropriate: those who do not do their fair share of counterterrorism can be criticised as failing to take on some costs of a collective project while taking a ‘free ride’ on the efforts of others.

What of Carter’s more substantive claim, that the coalition partners are free riders, and should be doing more? Assessing this requires further analysis that goes beyond the scope of this article. For one thing, we would need to know whether those states he had in mind received a fair share of the benefits of existing operations; if some do not, they—and perhaps others—incur no responsibilities to reciprocate. Furthermore, nothing in what I have said tells us anything about the extent of the obligations that attach to states who benefit from counterterrorism. We would need an independent account of how costs should be distributed within a mutually beneficial system of cooperation.51 A ‘progressive’ distribution of responsibilities among all those who benefit—whereby those with greater wealth (like the United States) should pay more—might be appropriate here, and so there might be nothing particularly problematic with the United States taking on the lion’s share.


I have been surveying some possible bases on which counterterrorist responsibilities can be assigned to states. While existing ways of assigning responsibilities elsewhere in the global justice literature might provide us with an acceptable way of assigning some responsibilities in this case, they also have significant limitations. However, other strategies may be available here. In particular, I have suggested that responsibilities might be justified as reciprocal obligations that are grounded on benefits received from others currently undertaking counterterrorist operations. Ultimately, I have suggested that a plurality of principles needs to be appealed to in order to ensure that important tasks are assigned to states. While the list of bases that I have provided may not be exhaustive, I believe that it will allow us to assign most of the responsibilities that need to be taken up.

I have said little about how we should choose among potential bases when assigning a particular task, and there is no straightforward answer to this that can be presented neatly here.52 However, as some bases face shortcomings with respect to certain sorts of task, others will look more promising. For example, we saw earlier that non-relational bases face particularly strong objections when the tasks being assigned involve significant costs that must be taken on by a limited number of agents: it appears questionable that one group of soldiers should take on risks to life and limb simply because their state is the most effective militarily or has the greatest ability to take on these costs (however that is understood). Relational bases such as liability or receipt of benefits might be more appropriate to appeal to in this case. When we are considering tasks that simply involve contributing money, in contrast, appealing to capacity (understood as ability to pay) might look more apt.53 What bases are relevant will depend on the case at hand?


I would like to thank the audiences at a conference on ‘Collective Responsibility for the Future’ at University College Dublin and a Justitia Amplificata Workshop at Goethe University Frankfurt, as well as Nathan Adams, Fredrik Hjorthen, and two anonymous reviewers for comments on earlier drafts. Work on this article was conducted while I held a Postdoctoral Fellowship at Goethe University Frankfurt, funded by the Centre for Advanced Studies ‘Justitia Amplificata’ and the Kassel Foundation.


1. Cheryl Pellerin, ‘Carter: Lasting Defeat of ISIL Depends on Coalition Strength’, Department of Defence News, January 13, 2016, (accessed November 10, 2016).

2. Unless otherwise stated, in what follows I will understand responsibility to mean the moral requirement to put a bad situation right. David Miller calls this concept ‘remedial responsibility’ to distinguish it from other meanings of the term ‘responsibility’. See David Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2007), 84.

3. This example and potential solutions that follow are taken from ibid., 104

4. On the relative importance of both sorts of consideration, see Simon Caney, ‘Two Kinds of Climate Justice: Avoiding Harms and Sharing Burdens’, Journal of Political Philosophy 22 (2014): 125–49.

5. James Pattison, Humanitarian Intervention and the Responsibility to Protect: Who Should Intervene (Oxford: Oxford University Press, 2010). See also Kok-Chor Tan, ‘The Duty to Protect’, in Humanitarian Intervention, ed. Terry Nardin and Melissa Williams (New York: New York University Press, 2006), 84–116, at 99–101.

6. Simon Caney, ‘Climate Change and the Duties of the Advantaged’, Critical Review of International Social and Political Philosophy 13 (2010): 203–28, at 213–18.

7. Matthew J. Gibney, ‘Refugees and Justice between States’, European Journal of Political Theory 14 (2015): 448–63.

8. For this reason, Pattison suggests that the most effective states may be relieved from their responsibility to conduct humanitarian intervention when they have already taken on significant costs. See Pattison, Humanitarian Intervention and the Responsibility to Protect, 198.

9. David Miller, ‘Distributing Responsibilities’, Journal of Political Philosophy 9 (2001): 453–71, at 461.

10. Rawls, for example, held that individuals are under a ‘natural duty’ to ‘assist in the establishment of just arrangements when they do not exist, at least when this can be done with little cost to ourselves’. See John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 334.

11. Pattison, Humanitarian Intervention and the Responsibility to Protect, 197–8.

12. Cécile Fabre, Cosmopolitan War (Oxford: Oxford University Press, 2012), 189; David Miller, ‘Review of “Humanitarian Intervention and the Responsibility to Protect: Who Should Intervene?” by James Pattison’, Global Discourse 2 (2011), 205–8, at 207–8.

13. Fabre, Cosmopolitan War, 190.

14. For example, Miller, National Responsibility and Global Justice, 111–34.

15. Ibid., 102.

16. Erin Kelly, ‘The Burdens of Collective Liability’, in Ethics and Foreign Intervention, ed. Deen K. Chatterjee and Don E. Scheid (Cambridge: Cambridge University Press, 2003), 118–39.

17. Daniel Butt, ‘On Benefitting from Injustice’, Canadian Journal of Philosophy 37 (2007): 129–52; and Robert E. Goodin, ‘Disgorging the Fruits of Historical Wrongdoing’, American Political Science Review 107 (2013): 478–91.

18. Caney, ‘Climate Change and the Duties of the Advantaged’, 217–18.

19. Daniel Butt, Rectifying International Injustice: Principles of Compensation and Restitution between Nations (Oxford: Oxford University Press, 2009), 5–6.

20. Studies on terrorists often find a desire for revenge against a perceived unjust aggressor among the motivating factors. See Louise Richardson, What Terrorists Want: Understanding the Enemy, Containing the Threat (New York: Random House, 2006), 88–94.

21. The link between poverty and terrorism is discussed in ibid., 57. On the role of states in causing global poverty, see Thomas Pogge, World Poverty and Human Rights, 2nd ed. (Cambridge: Polity, 2008).

22. Peter Bergen, United States of Jihad: Investigating America’s Homegrown Terrorists (New York: Crown, 2016), 49.

23. Peter Bergen and Swati Pandey, ‘The Madrassa Scapegoat’, The Washington Quarterly 29 (2006): 117–25, at 123.

24. For example, Pogge, World Poverty and Human Rights, 87–123.

25. Diego Gambetta and Steffan Hertog, Engineers of Jihad: The Curious Connection between Violent Extremism and Education (Princeton, NJ: Princeton University Press, 2016), 34–59.

26. This causal mechanism is suggested in Monte Palmer and Princess Palmer, Islamic Extremism: Causes, Diversity, & Challenges (Lanham: Rowman & Littlefield, 2008), 18.

27. The view of modern Islamist terrorism as a reaction against liberal society is found in Paul Berman, Terror and Liberalism (New York: W.W. Norton, 2004).

28. Robert Anthony Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), 94.

29. Miller, ‘Distributing Responsibilities’, 458. Miller thinks that more (remedial) responsibilities may be assigned on the stronger relationship of ‘outcome responsibility’, in which an agent causes a harm to another through genuine agency, as opposed to uncontrolled action (such as tripping over in the street). See Miller, National Responsibility and Global Justice, 100–1. What I say here with respect to mere causality will apply mutatis mutandis to outcome responsibility.

30. Peter R. Neumann, Old and New Terrorism: Late Modernity, Globalization and the Transformation of Political Violence (Cambridge: Polity Press, 2009), 21–5.

31. Ibid., 33–5.

32. Tunisia declares state of emergency after Islamic state-linked beach attack that killed 38 tourists’, ABC News, July 5, 2015, (accessed November 10, 2016).

33. Butt, Rectifying Historical Injustice, 127.

34. Goodin, ‘Disgorging the Fruits of Historical Wrongdoing’, 486.

35. International Monetary Fund, ‘World Economic Outlook: The Global Economy After September 11’, December 16–17, 2001, (accessed November 10, 2016).

36. Herbert L.A. Hart, ‘Are there Any Natural Rights?’ The Philosophical Review 64 (1955): 175–91, at 185.

37. Robert Nozick, Anarchy, State, and Utopia (Oxford: Blackwell, 1974), 95.

38. Richard J. Arneson, ‘The Principle of Fairness and Free-Rider Problems’, Ethics 92 (1982): 616–33, at 618–19; and George Klosko, The Principle of Fairness and Political Obligation (Lanham: Rowman & Littlefield, 1992), 35–9.

39. For an important recent exception, see Chris Armstrong, ‘Fairness, Free-Riding and Rainforest Protection’, Political Theory 44 (2016): 106–30. The view that moral (as opposed to political) obligations can arise between states on fairness grounds has been put forward by David R. Mapel, ‘Fairness, Political Obligation, and Benefits Across Borders’, Polity 37 (2005): 426–42.

40. Neumann, Old and New Terrorism, 39–41.

41. Peter L. Bergen, Holy War Inc.: Inside the Secret World of Osama bin Laden (New York: Touchstone, 2002), ch. 10.

42. Joseph M. Schwartz, ‘Misreading Islamist Terrorism: The “War against Terrorism” and Just War Theory’, Metaphilosophy 35 (2004): 273–302, at 290.

43. Graeme Wood, ‘What ISIS Really Wants’, The Atlantic, March 2015, (accessed November 10, 2016).

44. George Klosko, Political Obligations (Oxford: Oxford University Press, 2005), 61–70.

45. A. John Simmons, Justification and Legitimacy: Essays on Rights and Obligations (Cambridge: Cambridge University Press, 2001), 34.

46. George Klosko, ‘Cosmopolitanism, Political Obligation, and the Welfare State’, Political Theory 37 (2009), 243–265, at 253.

47. Neumann, Old and New Terrorism, 155–6.

48. Whether or not these practices could be reformed to ensure a more acceptable distribution of costs is an issue that I will not pursue here. However, I suspect that the tendency of some sorts of practices to impose unfairly high costs on some individuals will render them always impermissible on this approach.

49. Arneson, ‘The Principle of Fairness and Free-Rider Problems’, 622; Klosko, The Principle of Fairness and Political Obligation, 39; and Rawls, ‘Legal Obligation and the Duty of Fair Play’, 122.

50. A. John Simmons, Moral Principles and Political Obligations (Princeton, NJ: Princeton University Press, 1979), 111–12.

51. For an examination of principles that might be used for distributing the costs and benefits of public good supply, see David Miller and Isaac Taylor, ‘Distributive Justice and Public Goods’, The Oxford Handbook of Distributive Justice, ed. Serena Olsaretti (Oxford: Oxford University Press, forthcoming).

52. David Miller suggests that, when adjudicating among different bases of responsibility, ‘[w]e have to rely on our intuitions about the relative importance of different sources of connection’. See Miller, National Responsibility and Global Justice, 107.

53. For the more general view that the significance of non-relational bases will diminish as the costs become higher, see Stephanie Collins, ‘Distributing States’ Responsibilities’, Journal of Political Philosophy 24 (2016): 344–66, at 347.

About The Author

Isaac Taylor
Goethe University Frankfurt

Postdoctoral Fellow at the Centre for Advanced Studies "Justitia Amplificata", Goethe University Frankfurt.

Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Related Content