For decades, the specter of nuclear terrorism has haunted the halls of Washington policy-making and the imagination of Western popular culture. While this was true even before September 11, 2001, in the days since, a consensus has formed from which few dare deviate: Terrorist organizations are looking to acquire nuclear weapons, and if they are successful, they will use them in an attack as soon as possible. In a recent paper, we referred to this principle as the “acquisition-use assumption.” The underlying consensus around this is now virtually sacrosanct. It orients US strategic posture around the proposition that, as President Obama expressed it, nuclear terrorism is “one of the greatest threats to global security,” and “there is no doubt that if these madmen ever got their hands on a nuclear bomb or nuclear material, they most certainly would use it to kill as many innocent people as possible.”
The idea that terrorist organizations see value in trying to obtain nuclear weapons is fairly well-documented, and we don’t intend to question that here. But the assumption that terrorists only want those weapons in order to immediately attack Western capitals and local rivals deserves more scrutiny. In our work, we ask: How would a nonstate actor behave if it acquired a nuclear weapon? Is the working assumption of the policy and academic community that “acquisition equals use” a good one? We think this assumption deserves to be questioned because, if nothing else, holding it requires ignoring the long and rich history of deterrence and nuclear strategy, nearly all of which operated without a hostile detonation. And getting to the bottom of the matter will have important consequences. Assuming that use is a foregone conclusion holds world leaders hostage to only a single, worst-case version of a threat, while jettisoning the assumption enables planning for more tailored responses for a range of outcomes.
To be sure, understanding and predicting how a terrorist organization would behave if it had nuclear weapons is difficult and requires new ways of thinking about strategy. But the study and history of these organizations indicates that terrorists do behave strategically, and proceeding directly to a nuclear attack could be a strategically disadvantageous decision—and therefore less probable than many might think.
Nuclear weapons and terrorist strategy. While there are any number of far more likely scenarios for nuclear terrorism broadly understood, we focus only on groups with a working nuclear device, not a radiological dispersal device or the ability to attack a nuclear reactor. The threat posed by an operational device is fundamentally different, not least because possession would radically change the nature of the organization as a strategic, warfighting group.
A large body of work in terrorism studies teaches us that terrorist groups do behave strategically. Communications within Al Qaeda, Princeton Near Eastern Studies expert Michael Doran has written, have shown that the group behaves “almost exclusively according to the principle of realpolitik,” and is “virtually compel[led]” to do so by the “central doctrines of Islamic extremism” itself. While it may not appear so based on terrorists’ tactics, most groups have all the hallmarks of strategic decision-making, command and control, and sensitivity to costs.
This is all the more true for the hypothetical that concerns us: Only a large, well-organized, and heavily funded group would be able to attain operational nuclear capability. Regardless of what one thinks about the debates regarding terrorist organizations and their ability to acquire these weapons—either by theft or gift—acquisition and maintenance is going to be resource-intensive and difficult.
So, it stands to reason that if a group is able to cross this threshold and obtain a weapon, it will weigh all of its options seriously. And these options are not limited to a binary choice of attack or hold. In reality there are a slate of options for how to “use” a nuclear weapon, and we can divide these into five categories, only one of which is detonation.
For the other four options, there are two critical variables: whether to go public with the new capability or to remain covert, and whether to publicly communicate how they intend to use it as a strategic bargaining chip or leave their red lines unstated. While we isolate these for clarity’s sake, in practice strategic options interact with each other in important ways.
The first option—one we see in movies and television—is to reveal the new capability and publicly state under what conditions it would be used. In essence, this is nuclear “blackmail”: The organization declares that it has a weapon and will use it unless certain conditions are met.
Alternatively, the organization could make its new capability public, but not lay out terms for its use. This is the second option, which we term “opacity.” Pursuing this strategy has the virtue of instilling fear in the organization’s enemies without committing the organization to a definite strategic path.
While it may seem counterintuitive, it is equally possible for an organization to keep its new capability a secret while continuing its campaign of violence. So, as a third option, an organization could pursue a strategy of “latency” by keeping their bomb a secret, yet simultaneously making statements and threats about what it would do if it had one. Such an approach has been the core of Israel’s strategic posture for years.
Finally, given the high uncertainty that the first nuclear-armed nonstate actor in the world would face, the organization might just initially opt for “dormancy,” keeping the bomb “in the basement” until it decided that strategic conditions were ripe for global revelation.
Each option has its advantages and disadvantages depending on the context, but all of them entail strategic opportunity costs for the organization. For instance, once a group goes public as nuclear (and is presumably verified), it becomes nearly impossible to meaningfully return to a strategy of either latency or dormancy. Similarly, once conditions for use are articulated, going back on them would erode the credibility of the organization as a strategic bargaining partner, something of constant concern for these organizations. With states, we may fear a “use it or lose it” situation, but for an organization looking to get the most out of a limited, non-iterable capability, the reality is almost the opposite. The organization faces the traps of an escalation ladder: Each potential benefit it might accrue by adopting a more aggressive posture (declaring their weapons, making demands on their opponents) comes at a parallel cost to the group’s bargaining and strategic flexibility.
If these kinds of questions appear to be thoroughly familiar terrain for the history of nuclear strategy, that’s (part of) the point. Even more than for states, they take on an acute, existential significance for a terrorist organization. Because these organizations are engaged in asymmetric warfare against states that usually wield orders of magnitude more military might, terrorist organizations are forced to be even more exacting and calculating in their strategic decisions.
Terrorists might lose more than they gain by detonating a nuclear bomb. While terrorist organizations vary widely in their internal organization and structure, almost all are highly sensitive to benefits and costs, both external and internal. By examining these, it will become clear that terrorists might have more to lose than gain by proceeding directly to an attack. Doing so might alienate their supporters, cause dissent among the ranks, and give away a bargaining chip without getting anything in return.
Externally, terrorist groups rely on the support of the society in which they are embedded. As a result, they are deeply intertwined with the local population. Hezbollah is a paradigmatic example: In addition to possessing a quasi-independent political arm, it operates an extensive network of schools, hospitals, and other social services. These extended administrative networks provide a source of recruits, public support, and crucial cover for their financiers. They also engender a risk—what they have put so much effort into building can ultimately be undone by poor strategic decision-making.
Beyond the inevitable military response, committing a nuclear attack would represent an existential threat to an organization of this form. The unprecedented death toll of transgressing the nuclear taboo would have a predictably devastating effect on the support networks on which these organizations depend. Afterwards, maintaining the kind of large-scale external finance necessary to sustain them would become virtually impossible, particularly in the context of predictably heightened international law enforcement. In short, the organization would face radical alienation from its public base, as well as, where it exists, its state sponsorship.
From the outside looking in, it’s tempting to presume that the kind of public and even the kind of fighters who support Al Qaeda or the Islamic State simply don’t care about the level of violence those groups perpetrate. Empirically, however, this is simply untrue. Sustained analysis of the history of terrorist campaigns—even among those organizations willing to commit large-scale attacks—evinces a delicate balancing act between highly symbolic violence and concerns with stepping over invisible lines.
Equally, detonating a nuclear weapon in an attack would create intense strains on the internal dynamics of the organization itself. The sheer magnitude of the decision to proceed with a nuclear attack takes previously available opportunities off the table—returning to small-scale conventional attacks would appear to be a weakening of the organization’s position post–nuclear attack. The opportunity costs presented by the weapon have a significant potential to splinter organizations already concerned about command and control. As the last 20 years have demonstrated, terrorist organizations factionalize and splinter over goals and tactics under even conventional pressure. Post-attack, that pressure would increase exponentially and potentially reorder the international environment in ways that risk the organization’s continued existence. This reality is only magnified by the fact that most terrorist organizations exist in a political landscape in which they are not the only show in town; both the Taliban and Al Qaeda in Afghanistan, for example, face a constant competition with the Islamic State over recruits, bases of operation, and resources.
The result is a paradoxical effect on the calculations of terrorist leaders: Obtaining new capacity (even well shy of nuclear weapons) creates powerful incentives toward organizational centralization to prevent unauthorized activity, but such centralization could sideline or alienate certain factions, making it more difficult to hold the organization together. A nuclear attack, then, is the worst possible option for organizational leadership from the perspective of internal politics. This is because it risks setting in motion a series of events that could unravel the organization as a whole at precisely the moment when it needs unity to survive what will (likely) be an overwhelming reaction by the target state and its allies.
Finally, it must be remembered that what victory and defeat mean for the archetypal organization engaged in a terrorist campaign is significantly different than in interstate warfare. In conventional war, surrender is followed by capitulation to the opposition’s demands, even if exact terms must be worked out at the peace table. In conflicts between terrorist organizations and states (as well as other nonstate actors), determining what exactly constitutes “victory” or “capitulation” for either side is terminally ambiguous. Even a complete victory for the terrorist group on its own terms, such as the withdrawal of state forces from a region, will likely still be followed by reprisals—such as airstrikes or financial sanctions against the group. As a result, terrorists are inevitably circumscribed in what “success” could look like. Even with a nuclear attack, they could neither threaten the extinction of their opponent (as under the doctrine of mutually assured destruction, or MAD), nor threaten to further escalate costs, having already jumped straight to the top of the escalation ladder. Having played the entirety of its hand to bloody effect, a post–nuclear attack terrorist organization would face a hardened opposition and the prospect of a massive increase in costs, with little in hand to match.
Negotiating with nuclear-armed terrorists. If the processes we describe are accurate, then presumption should lie squarely on the side of skepticism whenever the threat of a nuclear attack is raised. But our analysis also highlights which conditions make an attack most likely and which make it increasingly difficult, revealing an entire palette of pressure points and leverage that states can use to further reduce the risk of an attack.
For example, a state might use both carrots and sticks to maximize the amount of pressure on key axes like organizational cohesion and local public support. The multiple audiences of a terrorist organization do not belong to them alone: Those audiences are themselves at the intersection of multiple lines of influence that can be pushed and pulled, if not always directly by the states themselves. The skillful use of threats and dangled political incentives can be highly effective in further deterring a nuclear attack. Intelligence is critical in this arena as well, particularly on the financial side. Financial knowledge allows potential target states to know how to threaten existential costs on an organization that might otherwise believe itself relatively impervious to such a threat.
In sum, we need to look at the strategic and organizational dynamics in play within terrorist groups in a clear-eyed way, without resorting to simple short-hand heuristics like the acquisition-use assumption. States have somehow muddled through the initial stages of the nuclear revolution with a set of reasonably intelligible and predictable strategies to match their capacities. There is no good reason to presume that a nonstate actor wouldn’t do the same.