There are two big challenges for American foreign policy: China and failed states. For the first, everyone agrees that the United States must do something but it is not clear what that something is. The answer depends on an assessment of the future trajectory of Chinese economic and political development.
If China remains an autocracy will its economic growth stall out, reverse itself, or continue. Alternatively China might transition into a rich, well-performing consolidated democracy with a large middle class. If China becomes a rich consolidated democracy, then the policy path for the United States, integrating China into the existing liberal world order, is obvious. If China remains an autocracy then the choice among integration, balancing, and hedging is less obvious. Nevertheless, the policy options available to China, and their relationship to China’s future political and economic trajectory, are well understood.
This is not the case for failed and failing states. There is no agreement on whether or not these polities pose a threat for the United States. And, if they do pose a threat what American policy ought to be. The core issue with regard to the threat to American national security is transnational terrorism. Other bads that could originate in ungoverned spaces, such as disease and criminality, are most effectively dealt with by changing domestic policies and increasing resilience in the United States itself, not be attempting to alter conditions in failed states.
There is, however, no consensus on the extent to which transnational terrorism originating in failed and badly governed states requires a foreign policy response. The minimalist position would be something like the following: Terrorist activities can originate from ungoverned spaces but they can also originate within the United States or other rich, industrialized, democracies. The damage that terrorists might inflict is limited. 9/11 was a horrific, but exceptional, event. The Boston Marathon bombings or the attack on the Charlie Hebdo offices are more typical. A few people might be killed but not a few thousand. The most effective way to address such threats is through intelligence and policing. Intelligence and policing can address domestic threats not just transnational ones.
The alternative to the minimalist position views that threat as more significant and does not limit the means for addressing it to domestic measures. Transnational terrorism coupled with the availability of weapons of mass destruction poses an existential threat to the security of the United States and the contemporary international order.
An existential threat is one that, if realized, could fundamentally change both domestic and international authority structures and policies. Weapons of mass destruction could be nuclear or biological. Biological weapons will become more readily available. Nuclear weapons are, obviously, much harder to obtain but a transnational terrorist organization, most likely one motivated by a jihadist ideology, could secure such weapons from a state program. Pakistan would be the obvious candidate.
A transnational terrorist organization could be self-generating in the United States or some other advanced democracy. It is, however, easier for such an organization to develop in ungoverned spaces in failed and badly governed states. Policing in ungoverned spaces will be non-existent. Intelligence gathering will be limited. Financial support for TN terrorist groups would be more difficult to monitor.
While ungoverned spaces might be more attractive for TN terrorist networks dedicated to procuring weapons of mass destruction than well governed (or at least effectively governed) spaces, it is still impossible to assign a probability estimate to a mass casualty attack actually occurring. Such an attack on a major city in the United States would be a black swan, and event that is highly consequential, without historical precedent, and impossible to predict. Black swans exist in a world of uncertainty rather than risk, we not only do not know whether an event will occur, we do even know what the underlying probability of the event is.
If there were a nuclear or biological attack on a major city the number of deaths would be in the thousands or tens of thousands, and the existing rules of the international order would go out the window. The concept of sovereignty would be junked. There would be no presumption that policing, arrests, interrogations, judgments, and imprisonments would be limited by national boundaries. Military strikes, manned or unmanned, would be conducted without permission from the government that nominally controlled the territory associated with a mass casualty attack. Colonialism, some form of direct control by more powerful states would be re-established. Defense budgets would dramatically expand. Civil rights in democracies would be curtailed. State surveillance would become more intrusive and pervasive. This would be a much less attractive world in which to live.
A WMD attack from an ungoverned space in a weak or failing state might or might not happen. Better intelligence and domestic policing are the first best defense against such an attack. Internationally the goal of American policy should be to improve state authority structures in countries with ungoverned spaces. The grandest policy option, the one pursued by the Bush administration, was to democratize these states. This is an impossibly ambitious objective, and one that as the last decade has demonstrated is extremely expensive. A more realistic policy option would be to aim for good enough governance in failed and badly governed states. Good enough governance means ensuring that a state is capable of keeping order within its own boundaries—at least enough order to contain transnational terrorists.
The provision of this order may sometimes be arbitrary and brutal. Maintaining order in some countries might require an American military whose primary mission would be to degrade transnational terrorist entities and perhaps intervene to maintain a balance of power among local strongmen. Where ethnic conflicts have eroded trust, we should encourage decentralization. Ideally, good enough governance would include providing some public services such as healthcare and primary education that would not threaten the local elite's ability to extract resources and stay in power. Some degree of economic growth might be possible provided we recognize that these rulers always require their cut of the profits.
Good enough governance is a far more modest goal than putting countries on the path to consolidated democracy. It would also be far less expensive. It would not eliminate the possibility of a mass casualty attack by a transnational terrorist group possessing nuclear or biological weapons but it would reduce the risk, even though there is no way of knowing the magnitude of such a reduction. It would make it more likely that American (or European, or Japanese, or Indian, or Chinese) deaths could be avoided, and it would preserve an international order in which the presumption is that recognized governments are responsible for policing their own territory.