Aaron Mannes cited by Reuters on Assassinating Terrorist Leaders

In the wake of the demise of Baitullah Mehsud, Reuters blog did a piece on whether or not this would have any impact on the Pakistani insurgency. They quoted my statistical analysis of the decapitation strategy (not cutting off heads – eliminating the leaders of terrorist groups). My findings were, in short, that, well, it depends. As it happens I am revisiting the paper now and the fact that it pops up so quickly when researching the issue only shows how thin research on the topic has been.

Broadly, I believe there are four categories of group in terms of their vulnerability to decapitation. Some groups are held together by a strong personality and may be vulnerable to decapitation strikes. Some groups are very robust, have extensive capabilities and deep benches of alternate leaders – these groups can and will respond to the loss of their leaders with revenge attacks. There are groups that will become more radicalized when they lose their leader. All three of these groups are the outliers. The most common groups will not necessarily be capable of carrying out a revenge strike, but will also not collapse at the loss of their leader. The open question then is whether or not the decapitation strike at least degrades the group’s capabilities.

Mehsud probably falls in this category. The Pashtun insurgency is deep and has multiple components, it isn’t going away quickly. However, it may be a step in the right direction – at the very least it is difficult to see how the Pashtun insurgency can become more radical than it is already.

To cite the classic maxim of academia, “More research is needed.”

Here’s the Reuters post:

Targeted killings inside Pakistan — are they working?
The death of Pakistani Taliban leader Baitullah Mehsud in a U.S. Predator strike last week – now considered a certainty by U.S. and Pakistani security officials – and subsequent reports of fighting among potential successors would seem to justify the strategy of taking out top insurgent leaders

The Taliban are looking in disarray and fighting among themselves to find a successor to Mehsud, the powerful leader of the Tehrik-e- Taliban Pakistan, the umbrella group of militant groups in the northwest, if Pakistani intelligence reports are any indication. Top Taliban commanders have since sought to deny any rift, but they certainly look more on the defensive than at any time in recent months.

So is decapitation or targeting the heads of militant groups, as a strategy to destroy these organisations, beginning to work in Pakistan ?

A considerable amount of research has gone into such a snake-head strategy, or the killing or capture of militant leaders, since Israel went down this road decades ago and the results are mixed.

Daniel Byman, Director of the Center for Peace and Security Studies at Georgetown University, says that while the U.S. strategy could tamp down the threat from al Qaeda, it can neither defeat the group nor remove it from its stronghold in Pakistan. In a piece for Foreign Affairs, Byman who previously studied the Israeli campaign of targeting enemy leaders, lays out the gains as well as the limits to such a strategy.

– A sustained campaign of targeted killings can disrupt a militant group tremendously, as slain leaders are replaced by less experienced and less skilled colleagues. This can lead the group to make operational and strategic mistakes, and over time, pose less of a danger. Moreover, constant killings can create command rivalries and confusion. Most important, the attacks force an enemy to concentrate on defense rather than offense.

And the limits as in Pakistan’s case are:

– The Predator strikes can force al Qaeda to watch its step in Pakistan, but it can still carry out some operations. Moreover, their local jihadi partners (such as Lashkar-e-Taiba) remain unaffected. So far, the strikes have been confined to tribal areas near the Afghan-Pakistani border, meaning that al Qaeda and the Taliban have been able to relocate parts of their apparatus further inside Pakistan, which may work to actually widen the zone of instability

– Although Israel achieved some success through its campaign of targeted killings during the second intifada in the early years of this decade, it was able to fully shut down Palestinian militancy only by reoccupying parts of the West Bank and building a massive security barrier between itself and much of the Palestinian territories — options that are not available to the United States in Pakistan, Byman notes.

Over the longer term the results of the decapitation strategy are even more mixed. Aaron Mannes, a researcher at the University of Maryland, says his study “in general found that the decapitation strategy appears to have little effect on the reduction of terrorist activity.”

In fact he found a distinction between groups that are ideologically driven or nationalist- separatist ones like the IRA and ETA – and religous groups such as al Qaeda or Hezbollah. While the ideological groups were forced to restrict acivity following a decapitation strike, the religious groups actually grew even more deadly. Hezbollah and Hamas are more reboust organisations, which is an important criterion for surviving the loss of a leader, his study found

Revenge also plays a key role in upsurge of violence following the loss of a leader. Another explanation might be the rise of the most violent elements within a religious militant group to the fore. “Based on this data, decapitation strikes are not a silver bullet against terrorist organisations. In the case of religious groups, they may even be counter-productive,” he says.

Ultimately, as Nighwatch intelligence here notes, there is no alternative but to destroy the sanctuaries in which militant groups operate. And it is hard to see that being done through these “bolts from the blue.”

[File photograph of Baitullah Mehsud at a news conference, and a village in South Waziristan cleared of fighters loyal to him]

Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *