Afghanistan Analysts Network – English

War and Peace

‘Ghosts of the Past’: New Special Report on Local Force Mobilisation in Afghanistan

Kate Clark 11 min

Today, AAN and the Global Public Policy Institute (GPPi) co-publish a new special report, ‘Ghosts of the Past: Lessons from Local Force Mobilisation in Afghanistan and Prospects for the Future’. The result of a three-year research project, the report considers why governments and their foreign partners have kept mobilising local forces in Afghanistan. It considers the potential benefits – greater local intelligence and motivation to defend one’s own land – and risks – of ending up with unaccountable and abusive militias. One of the co-authors, Kate Clark, here introduces the main findings of the report. She also says the need to understand local mobilisation is more urgent than ever. In a time of uncertainty for Afghanistan – over United States military and financial support support, intra-Afghan peace talks and the intensifying conflict – the Afghan Local Police is about to be dissolved with barely any plan as to what to do with those about to be unemployed. There are also proposals both to expand the Afghan National Army Territorial Force and ‘re-integrate’ Taleban into it, should there be a peace deal.

The ALP has been a lightning rod for attention, with an equal share of critics and proponents. Critics note the ALP’s long record of abuse and that many units are actually re-hatted militias. Proponents point to ALP units like the one pictured above in Helmand’s Nad Ali District. They defended their communities, holding the line against the Taleban, who were in villages only a few hundred metres away. One month after this picture was taken, the Taleban surged toward Helmand’s provincial capital, Lashkargah and this ALP unit was forced to withdraw, along with an ANA company that had been stationed in its village. Photo: Andrew Quilty, 2016.

Local defence forces are present in almost all of Afghanistan’s 34 provinces. All are supposed to be mobilised from and with the support of local communities, with the aim of defending people and land against the Taleban and/or the Islamic State in Khorasan Province (ISKP). Currently, three different forces operate, with some districts hosting all of them.(1) They are the Afghan Local Police (ALP), Afghan National Army Territorial Force (ANA-TF) and Popular Uprising Forces. They come from a long line of local forces, dating back to the mobilisation of mujahedin and subsequently pro-government ‘tribal militias’ (kandak-e qawm) in the 1980s. Post-2001, there have been fresh waves of mobilisation in the face of the Taleban insurgency.


The research for our new report, ‘Ghosts of the Past: Lessons from Local Force Mobilisation in Afghanistan and Prospects for the Future’, started from an apparent paradox, summed up in UNAMA reporting on the ALP (this and all other sources can be found in the special report). As part of its work on the protection of civilians during the conflict, UNAMA has documented abuses by some ALP units. Yet it has also reported that many communities said the ALP improved their security. We wanted to understand the reasons why some local defence forces were successful, protecting people and land, while others abuse the population, are co-opted by factional, ethnic or criminal interests and even behave so badly they draw support to the insurgency. The importance of this question was confirmed as important during the research by our work on how the Taleban view local forces as ‘enemy number 1’; they are especially hostile towards the ALP and Uprising Forces, seeing them as a greater threat than regular Afghan or even international forces. The continuing significance of local forces was also underlined by another dynamic, the tendency for local armed groups to be repeatedly re-mobilised – or ‘re-hatted’ – as ALP, or Afghan National Police (ANP), Taleban or private security companies.

In answering the question of what is likely to make a local force successful, protecting local people and not abusing them, one of our conclusions is that getting the genuine backing of local people for the force is fundamental. All too often, however, despite the Afghan authorities and their international backers paying lip service to the idea of a community defence force, local people’s wishes have been ignored. In the case of the ALP, that led to many units being established where people did not want them, including in places where they were bound to be co-opted by power brokers, factions and/or criminal interests or to aggravate local conflict. Or, the rush to get ‘boots on the ground’ as quickly as possible lead to safeguards being passed over, again with ruinous consequences. Where guidelines are ignored, our research suggested, local forces typically fail to protect local people.

Another conclusion is that local forces, while they can work well in protecting local people, do not work everywhere in Afghanistan. Often, it is local dynamics that determine whether or not a ALP unit is likely to work well. Where local strongmen with pre-existing militias dominate, particularly where they are connected to factional networks (for example in Takhar, Kunduz and Baghlan provinces), ALP have performed poorly. Establishing the ALP has also proven risky in areas with a strong history of multi-ethnic or intra-tribal division, because where units are mobilised from one side over another, they may deepen or exacerbate divisions. Access to natural and/or illicit resources is another risk factor. The ALP is more likely to succeed in places where local communities are organised, representative and actively engaged in establishing the force (for example Yahyakhel in Paktika, latterly in Shajoy district of Zabul and Kunar province).

Yet, the report also found that repeated cycles of mobilising local forces has itself contributed to the larger degradation of community structures and the intensification of conflict. This can be seen in how few places still have strong, organised, representative community structures (as in Yahyakhel and Kunar). Instead, in many places, commanders dominate and ethnic, tribal or factional conflict is entrenched; establishing new local forces in these places is likely to worsen conflict and result in greater harm to civilians.

The research also suggested that local force mobilisation can affect levels of violence. In Yahyakhel district in Paktika, for example, the ALP had such widespread popular backing and gained such control over the district, that insurgents had difficulty operating and levels of violence fell. In other districts, mobilising local men to fight other local men led to particularly nasty conflict, with both sides breaching the customary norms of Afghan warfare (seen, for example, in Andar and Muqur districts in Ghazni and Arghandab in Kandahar).

Finally, we considered whether the lessons learned from previous local force mobilisations have resulted in better safeguards for Afghanistan’s newest force, the ANA-TF (a separate paper on the ANA-TF will come out later this month with greater detail on this question).

Afghanistan’s local defence forces

Afghanistan’s three local defence forces have different institutional backers and funders. In 2020, all three face uncertain fates.

The Afghan Local Police (ALP), established in 2010, currently has 18,000 personnel (2) and is spread across 31 of Afghanistan’s 34 provinces. This force grew out of US Special Operations Forces’ projects to mobilise village-based forces against the insurgency from 2009 onwards. The ALP got grudging permission in 2010 from then President Karzai to become a national force within the Ministry of Interior and under Afghan National Police (ANP) command. It is funded solely by the United States – other international donors did not want to fund a ‘militia programme’ – and the Afghan government. It is due to see American funding dry up on 30 September 2020.

Afghan National Army Territorial Force (ANA-TF) has about 10,000 personnel mobilised in companies in 32 provinces and was established in 2018. The decision to place the ANA-TF within the Ministry of Defence and under Afghan National Army (ANA) command was intended to ensure there was better institutional control of the force compared to the ALP (given their better reputation for discipline than the ANP and Ministry of Interior). The ANA-TF is funded by NATO. There are plans, currently on hold, to expand it further, and at the same time from the US Department of Defence, the idea of using the force as a ‘reintegration vehicle’ for Taleban in the wake of any peace deal.

Popular Uprising Forces (wulusi patsun in Pashto; khezesh-e mardomi in Persian) are a more ad hoc set of local, counter-insurgency forces that are supported by the NDS. They appear to have proliferated under the National Unity Government. There is little publicly available information about their number, cost, weaponry, training, locations, or how commanders and locations are chosen. They are under no known formal mechanism of accountability and, as UNAMA has pointed out, they “have no legal basis under the laws of Afghanistan.” The NDS’ main sponsor is the CIA, which is itself outside the scrutiny of the main US watchdog, the Special Inspector General for Afghanistan Reconstruction, or SIGAR. It is assumed that some CIA funds go to the NDS-backed Uprising Forces.

Afghanistan’s local defence forces in an uncertain time

What happens to all three forces this year will depend on various factors: funding, international support or lack of it, Afghan government actions and the state of the conflict. Various scenarios for Afghanistan are possible – from a peace deal to even greater violence and instability – which makes understanding the dynamics of local force mobilisation important.

US and other international support, both aid and military, to the Kabul government has become far less dependable than it once was. The 29 February deal between the United States and Taleban has led, already, to the withdrawal of several thousand US troops (from about 12-13,000 to 8,600), with the remainder, according to the agreement, withdrawing after the Taleban provide some rather vague guarantees on terrorism and not posing a threat to the interests of the US and its allies (see AAN analysis here). Possibly pre-empting that, President Donald Trump has been discussing withdrawing US forces early, even before the November 2020 election “It is time,” he tweeted on 27 May, “for [the Afghans] to police their own Country. Bring our soldiers back home but closely watch what is going on and strike with a thunder like never before, if necessary!”

The US Special Inspector on Afghan Reconstruction (SIGAR), John Sopko, has also fired warning shots across the government’s bow concerning international aid: “After 19 years of war, foreign governments, including the United States, are growing weary of paying Afghanistan’s bills.” Sopko’s remarks came a few days after a damning report by UNAMA on the government’s efforts (or lack of them) to tackle corruption (see our analysis here). Sopko warned that the patience of members of Congress and their voters:

…is not endless, even if it seems the war is. Time is running out. The Afghan government must finally get serious about addressing the problem of corruption if it is ever to bring lasting peace to its people.

Given that the Afghan government is dependent on international aid for 75 percent of its budget, threats to reduce it are a serious matter. Moreover, by contrast, Najibullah’s government relied on support from Moscow for just 26 per cent of its expenditure, but when that was withdrawn in 1992, it led to state collapse and a new and bitter phase of the civil war (for more detail on this, see this recent special report). Yet, as we have also reported in the face of these threats to international support, Afghanistan’s political elite spent months negotiating a power-sharing deal. Even now, the slow pace of appointment-making suggests they have no idea of the urgency of the situation.

Meanwhile, the spring and early summer have unfolded with brutality and violence. The Taleban refused a ceasefire in the face of the Covid-19 pandemic, seeing it as a government ruse to weaken them. Tragically, it seems the conflict is the only facet of life not to have slowed down in the face of Covid-19. UNAMA has reported that twelve hundred civilians had been killed and injured in the conflict in the first three months of 2020 and that April 2020 saw more casualties than in April 2019 (for sourcing, see this recent AAN report). Apart from a brief reprieve over Eid, the violence has been unremitting.

As to possible peace, talks between the government and the Taleban could begin soon, with the freeing of almost all of the 5,000 government-held Taleban prisoners specified for release in the US-Taleban agreement. Again, however, there is uncertainty: Are the Taleban intent on negotiating a peace deal with the government, or do they just want to get their prisoners freed and the US military off the battlefield before carrying on the war against the ANSF? It is also not clear if the government actually wants to negotiate; so far, it has proven unenthusiastic about the US-brokered talks.

Dissolving the ALP

In the midst of all these uncertainties, the US is going ahead with its decision to stop funding the ALP on 30 September 2020. The US, and presumably also the Kabul government, is fully aware of the dangers the dissolution of the ALP creates. A 30 March 2020 report to Congress on the US combat mission in Afghanistan, Freedom’s Sentinel, by the United States’ Lead Inspector General (3) repeats it concerns as to:

… whether well-armed but newly unemployed ALP members would join the ranks of violent extremist groups or local power brokers, who have previously used ALP units as their own private militias.

Not mentioned in the report is that, in some districts, the ALP is an important element in defending people and territory against the insurgency. This means that whatever the nature of the ALP locally, the decision to demobilise the force when so many rural districts are under intense pressure from the Taleban, is highly questionable. That is especially the case, given the lack of preparation.

Moreover, less than three months before funding is cut, plans for what to do with demobilised local policemen are still vague and tentative. This is despite the US decision having been made at least as far back as July 2019. On preparations, the Lead Inspector General report quotes the NATO Special Operations Component Command–Afghanistan (NSOCC-A):

To mitigate potential security risks, the Afghan government has tentatively scheduled a plan for post-dissolution employment options for ALP members and for recovering ALP weapons and equipment. NSOCC-A reported that in order to prevent the creation of future insurgents, it is working with the Afghan Ministry of Interior Affairs (MoI), the Ministry of Defense (MoD), and the Office of the National Security Council to identify and encourage recruiting of ALP members into the Afghan National Army (ANA) and ANA-Territorial Force (ANA-TF), and the Afghan National Police.

The plan appears to be:

  • Provincial headquarters collect weapons and equipment
  • The MoI dissolves ALP units by district, “according to their assessed effectiveness and an estimated level of risk, and will include severance pay, depending on final MoD and MoI input, as well as CSTC-A [Combined Security Transition Command – Afghanistan, ie the US military and its allies] approval”
  • Attempts will be made to recruit ALP members to other branches of the ANSF. If that fails, the MoI has tasked “provincial governors with finding civil employment for ALP members ineligible for recruitment.”

The report describes the implementation of this strategy as “challenging,” given that the ALP leadership has stated they have “limited ability to carry [it] out and there is a lack of coordination with civilian public and private sector organizations that could help to find employment opportunities for former members of the ALP.” It stressed the “very high” unemployment in Afghanistan and unclear job opportunities for former ALP men.

Meanwhile, there is a proposal not just for finding places for former local policemen in the ANA-TF, but also using it as a ‘reintegration vehicle’ for reconciled Taleban, should there be a peace deal. That this is a poor idea is evidenced in the special report: a fundamental lesson from our ALP case studies is that local forces are most likely to be harmful when they are set up as a result of political pressure or perceived security exigencies, rather than the needs and wants of the local community. Given that previous DDR programmes have all resulted in rampant corruption, a marginalisation of civilian wishes and often a re-hatting of militias, the authors have little faith that the Department of Defence proposal to use the ANA-TF to deal with reconciled Taleban would fare any better.

Whatever happens in the coming year, the role of local forces will be significant. How they behave on the ground will also be affected by the wider picture and whether the conflict tends towards sustained peace or worsening violence. For local people, the question of whether a local defence force is firmly within the institutional control of the Afghan state or loyal to a power-broker or strongman, whether it has been set up with their blessing and support or against their wishes, and whether it has so far fought or collaborated with the Taleban could be become very important. It will also be essential to better understand local force mobilisation in Afghanistan – as this special report attempts to do.

Edited by Rachel Reid

(1) There are also unrecognised pro-government militias which get occasional support from the government and may fight on the government side against the Taleban. There are also ‘campaign forces’, local armed groups which answer to a foreign command, typically US Special Operations Forces or CIA. Both are mentioned in the special report, but were not part of the research brief.

(2) The ALP’s tashkil, the approved size and structure of the force, was set at 30,000 in 2012, a size it approached, but never reached. Reform of the pay-roll brought the number of local police, whose identities the US military could confirm and who therefore were on the pay-roll down to about 22,000 in 2017 (spending fell by about eight per cent between the last quarter of 2016 and the first quarter of 2017). Some of those who had been paid would have been ‘ghosts’, on the payroll but not existing; others could not be registered because of insecurity. Ministry of Interior figures for ALP therefore tend to be higher than US ones. The report to Congress by the Lead Inspector General referred to later in this piece, gives a number of 18,000 ALP for the first quarter of 2020.

(3) The Lead Inspector General (Lead IG) is a joint agency initiative involving the inspector generals of the DoD, Department of State and the Agency for International Development (USAID). It provides quarterly reports to the Congress on “active overseas contingency [ie military] operations.” See this State Department webpage for more detail.  


ALP ANA ANA-TF conflict Ministry of Defence uprising forces US military US-Afghan relations