Originally posted December 2009 

Notions of “security” can greatly vary and even be understood in a contrary way. In the case of Afghanistan, international observers heavily stressed the lack of physical security, circumscribing it with the term “warlordism.” This labelling was the expression of a modern, state-centric understanding of physical security, which generally assumes that the state institutions hold the monopoly on violence. Contrary to this blueprint, individual actors — so-called “warlords” — were identified as the ones who control de facto the means of violence. Between 2002 and 2006, virtually no influential political figure in Afghanistan could elude this label, which subsequently became the category for all actors spoiling or even casting doubts on the international agenda of the Afghan peace process.

In our opinion, “warlordism” and its connotations are not sufficient to characterize the structures of violence in Afghanistan. While the existence of “warlords” in Afghanistan is undeniable, the manifold forms of individual leadership as well as the local differences regarding security arrangements are significant; they could be positioned on a linear axis, with “warlords” on the one side and the modern state on the other.

Peripherization of the Center

“Warlordism” often is regarded as a local phenomenon characterized by a power struggle between the center and the periphery. The interactions between the center and the periphery are quite complicated, as each side endeavours to influence the other. However, in recent years, the “Afghan center” has not been in the position to strengthen its power in the periphery, while the periphery has gained the ability to impose its interests on the center. We can refer to this development as the “peripherization of the center” in Afghanistan.

The Afghan state never developed beyond an embryonic stage. The successive conflicts that have plagued Afghanistan since 1979 destroyed the remaining state structures. Therefore, the Afghan government which was established in December 2001 possessed neither a well-founded authority nor legitimacy in the eyes of the people. The twin objectives of the Afghan government were to re-establish a state-owned monopoly of violence and to dismantle local militias. The international community supported the Afghan government in pursuing these goals. However, their two main initiatives — the Disarmament, Demobilization, and Reintegration (DDR) process (2003-2005) as well as the subsequent Disbandment of Illegal Armed Groups (DIAG) process (begun in 2005 and planned to be completed in 2011) — have fallen short. Although they were able to collect a number of weapons, they were not able to disband the clientelistic structures between the commanders and their militiamen. In most cases, influential commanders were able to preserve their power by taking official positions or by transferring their militias to regular units of the army and the police. Accordingly, the security sector of the state to a large extent consists of commanders and their militias.

In addition, President Hamid Karzai’s strategy to restrict the power of commanders, who received a position in the state apparatus, was only moderately successful. Karzai decided to rotate governors, ministers, and police chiefs from one position to another in order to prevent them from establishing autonomous power bases. Initially, this strategy yielded some positive results, as in 2004, when Karzai removed Ismail Khan from his post as Governor of Herat and appointed him as Minister of Energy and Water in Kabul. Since then, however, this policy has encountered increasingly stiff resistance not only from the persons concerned, but also from the local elites. In many cases, local elites have prevented the government from taking action by mobilizing their clients in order to demonstrate that the enforcement of the state decision would be destabilizing and increase the likelihood of violence.

Furthermore, the political center in Kabul is strongly influenced by local politics. Local elites endeavour to maintain close relationships with executive branch officeholders or members of the Parliament. It is often the case that local elites are connected to political decision-makers in Kabul through family ties. The ability of local elites to influence decision-making processes in the capital, in turn, has an impact on local politics. Thus, local resistance constrains the central government’s efforts to implement its decisions beyond the capital, while the influence of local elites on the central government’s decision-making processes impinges on the latter’s sphere of activities. This highlights why the localization of power is so tremendously significant for understanding Afghan politics.

Local Security Architecture

Due to the absence of a reliable state, many Afghans regard the phenomenon of “warlordism” as a system of political life that is better than an unpredictable future. However, a certain socio-political order can only be sustained as long as it is regarded as legitimate or as unchangeable. This is why power-holders need legitimacy to endow their power with authority. Socio-political order under such conditions follows its own rules, and in many cases overlaps with modern concepts of statehood. Local power-holders resist state penetration, and aim instead to instrumentalize state resources for their own interests. Yet, at the same time, they are partly able to provide security and certain welfare functions to their constituencies, thereby gaining legitimacy. Moreover, local elites are strongly embedded in societal contexts. This “embedding” limits their scope of action and produces expectations within their constituencies.

As our research on Kandahar, Kunduz, and Paktia showed, one can even find a variety of different security architectures within a province, often diverging from valley to valley and from village to village. Our fundamental finding is that a contextualization is the key to understanding local security architectures.

First, social structures play an eminently important role and have to be always regarded in the very specific local context: The different social structures in Paktia and Kandahar make clear that a characterization as “tribal Pashtuns” as such is too superficial to say anything meaningful about the tribal impact on the security architecture. Moreover, the history of a given region must be taken into consideration. Due to the colonization process of the 20th century, the population of Kunduz, for example, is shaped by a high degree of heterogeneity, and also by a rift between the Pashtun latecomers on the one side and the indigenous inhabitants on the other. The lack of a common ground of values and rules between these groupings contributed to the fragmentation of “warlordism” in Kunduz. Paktia provides the opposite example: With a tradition of tribal culture accepted by the people at large, strong tribal institutions averted “warlordism.”

Second, local economies have an impact on the security architecture. In regions such as Kandahar, which rely heavily on drug cultivation and the drug trade, one can witness the establishment of strong warlord structures. Apparently, the financial resources connected with the drug economy contribute to the strengthening of hierarchical structures. This argument is supported by the example of Kunduz, where a strong clan succeeded in establishing itself in the district of Imam Sahib, which is strategically important for the trans-boundary drug trade to Tajikistan, while the district of Khanabad, which is not benefiting from the drug economy, faces a fragmentation of the control of power and violence.

Third, the presence of the state has a significant impact on the security architecture. In general, the state seeks to control the security sector and to establish a monopoly on violence. One could imagine that in places where the concept of state is more widely accepted, the dominance of the “warlords” is easier to break. But contrary to this, as the examples of Kandahar und Kunduz reveal, “warlordism” is very distinct in exactly those regions where the state (i.e., the notion of the state held by local elites) is regarded as important. “Warlords” often perceive the state as a desirable resource to control and to have access to. Thus, in the case of Afghanistan, it seems that “warlordism” is a concomitant phenomenon of the state-building process, rather than being diametrically opposed to it. In contrast, the egalitarian tribal structures in Paktia, where the state is hardly recognized as such, prevent the consolidation of “warlordism.”

Lastly, it is important to assess the role of the international community. The presence of international actors undoubtedly led to the disappearance of weapons being displayed in public — “warlords” and militias have been forced to keep a low profile. This trend is particularly discernible in those Afghan provinces where the international community has contributed substantial funds for reconstruction (e.g. Kabul, Herat). For many “warlords”, a share of these financial resources constitutes a vital economic incentive. Yet the international presence does not always have a taming influence on the violence structures. Ultimately, it was the establishment and equipping of Afghan “warlords” and their militias by the US Army in the War on Terrorism which caused the temporary emergence of “warlordism” and continues to determine the security structures in Kandahar to this day.


This contribution intended to stress the complexity and locally very heterogeneous security structures in Afghanistan. Even though a broad definition of the term “warlord” can be applied to many actors of physical violence in Afghanistan, it fails to take into account the vast variety of local security architectures. While the state and international actors have a direct influence on the security architecture, local social and economic conditions primarily shape the security mechanisms. Moreover, the concentration of power at the local level is so strong, that even the core institutions of the state are under siege by local interests.

Finally, it is important to note that the international media increasingly has resorted to using the term “Taliban” in place of “warlord” to denote the highly dynamic political structures in Afghanistan. Meanwhile, many analysts again have fallen into the trap of redefining the highly differentiated political landscape in Afghanistan again along a bipolar axis: The recent trend is to construct the isomorphism of the categories Taliban, Pashtuns, insurgents, and drug dealers to make clear who the enemy is. However, this tendency expresses much more the concerns of the interventionists than it reflects the highly differentiated local realities of Afghanistan.

The Middle East Institute (MEI) is an independent, non-partisan, non-for-profit, educational organization. It does not engage in advocacy and its scholars’ opinions are their own. MEI welcomes financial donations, but retains sole editorial control over its work and its publications reflect only the authors’ views. For a listing of MEI donors, please click here.