I’ve just been doing a final read of some parts of David Kilcullen’s The Accidental Guerrilla, specifically his contagion model, before returning it to the library.
The Accidental Guerrilla syndrome is presented as a cycle however the cyclic nature of the model is not from Rejection back to Infection but is more likely as cycles between Contagion, Intervention and Rejection. Any spread of the insurgency within the host nation or theatre is more a factor of expanding Contagion than more Infection, although both may run alongside as other factions, groups and movements exploit the destabilised environment.
Please not that as much as possible, I am seeking to avoid use of the terms ‘insurgency’ and ‘counter-insurgency’ in the broader context in which they are too frequently employed today. In the absence of a suitable broader and manageable term that covers destablising activities, of which ‘insurgency’ is a subset, I am using ‘issue’ and ‘problem’. The lack of a suitable noun is one of the weaknesses in the adoption of Countering Irregular Activity (CIA) as a descriptor the the contemporary environment.
A medically-based analogy has considerable relevance for CIA and there is no arguing the with ‘accidentalness’ of many of the players in irregular campaigns. However, Kilcullen’s contagion model is flawed in a number of areas, most notably in its over-specific focus on Al Qaeda.
In focusing on AQ and its associate organisations, the models is of limited utility in considering other forms of irregular activity. AQ is but one of the current or potential irregular threats and the risk is that in concentrating too much on the current known evil, we will miss the rest of the forest. This is supported by two of the case studies in The Accidental Guerrilla, Pattani and East Timor, which do not conform to contagion model and are essentially ‘home-grown’ problems. More so, there has never been so much as a whimper from the intelligence community or any other source of Al-Qaeda/AGIM/JI involvement in East Timor.
‘Infection’ does not necessarily occur in “…remote, ungoverned or conflict-affected areas…” That may be the case in some places like Afghanistan but it does not hold true in Iraq, Vietnam, or Europe where the breeding ground for irregular activity has been the urban areas, ranging from universities to slums, from upper to lowest classes.
‘Infection’ also implies an external catalyst or driver for the problem which is not always the case, although it might be argued that global media coverage of insurgencies and other irregular activities promotes imitation. Rupert Smith’s ‘franchisers of terror’ definitely exist but I think that it is misleading to present external agents as key to the development of a problem. As detailed in The Accidental Guerrilla, movement leaders in Pattani have deliberately opted out of any alignment or support from AQ and its ilk. As an aside, this may an indication that at least some Islamic movements perceive association with AQ as counter-productive.
The ‘safe haven‘ implies a geographic sanctuary from which the insurgency spreads but insurgency ink blots tend to spread from those environments where the physical and ideological elements of the ’cause’ can take root and be nurtured. Geographic safe havens may be more for the safekeeping of key personalities e.g. the Ayatollah Khoumeini in France, than as a direct linear progression in the development of the insurgency. AQ’s alleged sanctuaries in Pakistan, for instance, have only really assumed prominence after US/NATO operations in Afghanistan rendered Afghan bases untenable.
The influence of ‘outside forces‘ is not a necessary element in the Intervention phase; in fact, outside forces, whether organic to the host nation or as part of external support, may not become involved until well into the Rejection phase. The initial intervention will more likely be through the internal national power of the host nation government and, while it may include military, police or other security forces, it may equally be built upon political and economic initiatives – any or all of which may have unintended consequences and contribute to escalation of the problem.
The population that reacts and thus contributes to the escalation of the problem, or the weakening of the host nation government (two different effects that contribute towards the success of the dissenting movement). might not necessarily be that of the host nation:
- The global outrage expressed at perceived US outrages during the battles for Fallujah and other US application of force during and after the warfighting phase on OEF is an example of this.
- The classic irony-laden example is that of the US response to the Tet Offensive in 1968 where the Viet Cong and NVA were decisively defeated but which led to a widespread public perception that the war was lost.
- Had the Soviets adopted a less brutal philosophy in Afghanistan in the early 1980s, it is possible that the Charlie Wilsons, Bob Browns and Osama Bin Ladens of this world would have lacked the platform upon which to build growing tangible (Stingers) support to the mujihadeen.
- International condemnation and sanctions against Rhodesia and South Africa greatly weakened the security forces of both nations, reduced the options available to them to suppress internal dissent, and led directly to an insurgent victory in Rhodesia in 1980, and a greatly weakened South Africa today.
A more applicable medical analogy for the CIA environment than that of an infection as used in the Accidental Guerrilla syndrome might be that of cancer. The catalyst or carcinogen that causes the cancer may be an external element or one drawn from an internal issue. As in COIN and CIA, there are few hard and fast rules defining what causes cancer of one sort or another, nor how it chooses its victims. Elements that we may have previously thought to be benign may be redefined as high risk, usually a lesson learned the hard way. Similarly, growths may lie dormant for years, considered benign or so far below the radar that they are not noticed, and explode across the host without warning.
Unlike a germ-based infection, cancers can rarely be vaccinated against and even for those that can, there is a statistical probability of harm for a percentage of those vaccinated. Antibiotic treatments against infection are usually non-intrusive, and have limited or no side effects although over time some germs may develop partial or full immunity against particular treatment regimes. The treatment for cancer is usually traumatic and causes harm to the host – a case of determining the lesser of a number of evils. Like insurgencies, cancer is not inherently contagious – the contagious or exportable aspect of both lies in the exploitation of the causes or carcinogens.
Untreated, cancer invariably causes the host to eventually succumb; treated, the host may still succumb if the cancer is too widespread, if the treatment has unintended adverse consequence, if the treatment is simply ineffective, or if the treatment regime lies beyond the available resources to fund it. As many doctors are aware, there are no golden rules that what works for one cancer patient will work for another – each case must be assessed and treated on its merits.
The Accidental Guerrilla model lacks a Recovery option; as writ in the book, it is a vicious downward spiral with an implied statement that intervention only worsens the situation. Somewhere along the path, there needs to be an off-ramp that breaks the cycle and leads to a recovered state. This state is unlikely to be a full recovery and even less likely to be a dramatic improvement upon the pre-problem situation. It is more likely that the host will take many years to recover and may not fully recover ever:
- The long-term effects of the US War between the States remain issues for some even today 150 years on.
- Similarly, New Zealand is still resolving the issues arising from the New Zealand Wars that were ‘won’ from a military perspective by 1873.
- As a result of friction (irregular or destabilising activity) between Muslim and non-Muslim populations, Pakistan and Bangladesh were split off from India. The ongoing antipathy between India and Pakistan still occasionally erupts into violence, as it did in 1999, and arguably, in 2008 in the Mumbai terrorist attacks.
- The insurgents in the Malayan Emergency that officially ended in 1960 were primarily ethnic Chinese and the relationships between Malays and Chinese continued to decline int eh early 1960s. As a result, the nation of Singapore was created as a home for the Chinese, leaving Malaysia primarily for Muslim Malays. The relationship between both nations continues to have its ups and downs and may come to a head when Singapore hits the lebensraum wall in the next decade or so.
- 35 years after Vietnam was reunified, it has still to reclaim its position as ‘the rice bowl of the world’ or fully resolve long-standing issues between North and South arising from the 1954-75 campaign.
- The Allies in WW2 sought not merely a return to the pre-war status quo but decisive change in Germany and Japan to prevent, not merely deter, future aggression.
- Likewise, the objectives of OIF in 2003 were clearly to force a regime change in Iraq.
- While British objectives in the 1982 Falklands War were a return to prewar understandings, it was well understood at the time that the party that lost the war would undergo a domestic change of regime.
One of the truisms of post-WW2 COIN has been that the object of any COIN campaign revolves around securing and maintaining the legitimacy of the host nation government. This has been a less successful process, often because the government in question has left much to be desired in terms of its culture and ethos:
- Early US attempts to bolster the Thieu regime in South Vietnam did generate David Kilcullen’s Rejection reaction and had the Ky government not been able to govern with a strong hand, it is likely that the Vietnam War might never have got to the 1965 surge.
- Currently US/NATO forces struggle to promote the legitimacy of Karzai’s government in Afghanistan, partially due to its lack of ownership amongst the Afghan people, partially because the notion of strong central government has never really been accepted in Afghanistan. In attempting to promote legitimacy, ISAF is actually trying to compel change inappropriately.
- Iraq is an example of a successful campaign where the legitimacy of the government is both a primary and achievable objective. In fact, Iraq in the first decade of the 21st Century is an interesting blend of all three outcomes of compelled change in toppling Saddam’s regime, promoting the legitimacy of a democraftic government, and restoring a state of normalcy (less one dictator and corrupt regime) for the people of Iraq.
- Thus, for Afghanistan, this may be a return to the loose confederation of tribes, possibly even under a degree of Taliban control, that existed in the mid-90s – with the strong and enforced proviso that sponsorship or harbouring of jihadist or similar takfir elements will not be tolerated.
- In East Timor, this was a return to the pre-referendum stability of 1999, albeit without Indonesia control, nothing that this change was mandated by the referendum and was not a result or objective of the INTERFET or UNTAET missions.
- Much as many were sympathetic to Bougainvillean aspirations of autonomy in the late 1990s and early part of the last decade, the objective of the BEL ISI mission in Bougainville 1997-2002 was a return to pre-rebellion normalcy under Papua New Guinea governance. While Bougainville may still achieve self-determination, that will be as a result of other processes and not of the intervention.
In developing objective prior to an intervention and reviewing them periodically during an intervention – what are we here to do? what are our freedoms and constraints? has the situation changed since the last time we thought about this? – commanders and planners at all levels must be considering which of these outcomes they are there to achieve:
To compel change in structures and governance?
To promote the legitimacy of the current government?
To restore what was ‘normal’?
Normalcy (PDF Version)