Over the past few weeks, several blogs have spotlighted a recent article by scholars Keir Lieber and Daryl Press that analyzes the threat of a nuclear weapons state transferring nuclear weapons to a terrorist organization. The article – “Why States Won’t Give Nuclear Weapons to Terrorists” — asserts that a state has little incentive to deliberately transfer nuclear weapons to a terrorist group, because if that group were to carry out a nuclear attack with said weapons, neither party would remain anonymous, and retribution from the attacked state would undoubtedly ensue.
The article’s quantitative focus makes it something of a novelty in the field of nuclear weapons studies, and while it is worth reading in full (according to The Monkey Cage, MIT Press is making it freely available through mid-August), the authors’ use of statistical analysis merits a brief exploration.
Essentially, Lieber and Press argue that, once a nuclear terrorist attack could be attributed to a specific terrorist organization, connecting that group to a sponsor state would be relatively easy. The key challenge, then, would be making the link between the attack and its terrorist perpetrator. Drawing upon the Global Terrorism Database (GTD), a dataset compiled by the National Consortium for the Study of Terrorism and Responses to Terrorism (START), Lieber and Press found that the attribution rate of terrorist attacks is closely correlated to the number of fatalities caused by an attack – according to their analysis, the attribution rate for attacks causing 0-4 deaths was 40%, but for attacks causing more than 100 deaths (a threshold that a nuclear attack would surely clear), that number jumped to 73%. Moreover, for attacks that took place on the soil of the US and it allies, and killed more than 10 people, the attribution rate spiked even higher, to an imposing 97%. Clearly, then, the odds of an act of nuclear terrorism going unattributed are quite steep, particularly if the attack is carried out on the territory of the US or its allies.
In light of the quantitative evidence that they present, I can certainly get on board with the Lieber/Press assertion that a nuclear weapons state is highly unlikely to deliberately transfer a nuclear weapon to a terrorist organization. The chief problem with this article, however, is that it focuses on a fairly far-fetched nuclear terrorism scenario. Simply put, on a conceptual level, it is exceedingly difficult to imagine any sane government putting nuclear weapons in the hands of a terrorist group that, in the end, it has no direct authority over. The notion of a rational government willingly exposing itself to such a massive risk simply does not pass the smell-test. So, while Lieber and Press’s quantitative approach does lend the article a certain novelty, one can’t help but feel that the statistics presented merely confirm something that was fairly obvious to begin with.
For this reason, it would be unwise to take Lieber and Press’s findings and conclude that concerns about the threat of nuclear terrorism, as a whole, are alarmist (as Hegemonic Obsessions’s Matt Fay argues here, and Foreign Policy’s Stephen Walt seems to imply here).
Keep in mind, Lieber and Press focus specifically on the possibility of the “nuclear handoff” (to borrow Walt’s phrasing), and find that particular scenario to be implausible. However, this sort of government-authorized transfer is just one potential route to an act of nuclear terror, and probably not the most realistic one. I would argue that a much more viable threat is that of a terrorist group stealing a sufficient amount of fissile material, and then constructing the actual nuclear device themselves. Fissile material is far from a rare commodity – the size of the current global stockpile of highly-enriched uranium, for instance, sits at roughly 1.39 million kilograms, meaning that, as Lieber and Press put it, “the material needed for a single crude weapon could be found within the rounding error of the rounding error of global stocks [of HEU].” Of course, not all of this HEU would be equally accessible to an aspiring nuclear terrorist, but the nuclear security problems of certain states (Russia and Pakistan, in particular) are well–documented, and provide an avenue through which a terrorist group could potentially acquire the raw materials for a nuclear weapon. With those materials in hand, the construction of a simple nuclear device would not be out of reach for a group of technically proficient terrorists.
Ultimately, if an act of nuclear terrorism were to occur, a state would have to be involved to some degree, as producing fissile material lies well beyond the capabilities of even the most sophisticated terrorist organization. However, that “involvement” is very unlikely to come in the form of a government-authorized, deliberate transfer of nuclear weapons or materials – this is something that Lieber and Press’s study makes clear, and quite reasonably so. And yet, we must be careful not to broaden the implications of Lieber and Press’s study beyond the scope of the authors’ actual argument. They question the risk of a specific route to nuclear terrorism – the “nuclear handoff” – but the broader threat of nuclear terrorism remains.