Afghanistan’s Taliban – Legitimate Jihadists or Coercive Extremists?

ABSTRACT The military intervention in Afghanistan in 2001 was portrayed as a fight to oust the extremist Taliban. But the Taliban have long been regaining influence, with the military victory of the Afghan government and its foreign allies now seeming less likely than ever. In light of these developments, this article investigates what the affected people – rather than the foreign interveners – think about the Taliban, and whether they perceive them as coercive or legitimate. Building on a conceptual understanding of legitimacy that has been adjusted to the dynamics of conflict-torn spaces, the article suggests that people judge the Taliban on the basis of how their day-to-day behaviour is perceived. While the Taliban are a coercive threat in urban centres and other areas where they launch attacks, they nonetheless manage to construct legitimacy in some of the places which they control or can access easily. A major source of their legitimacy in these areas is the way in which they provide services – such as conflict resolution – which some people consider to be faster and fairer than the state’s practices.


Introduction
The Taliban are usually depicted as ideological fightersreligious extremists who want to introduce harsh rules in Afghanistan, including the prohibition of music and the suppression of women. Their mode of governance stands in stark contrast to Western ideals, with the fall of the Taliban government in 2001 being portrayed as a victory against terrorism and human rights abuses. However, the influence of the Taliban and other armed opposition groups is steadily growing again throughout Afghanistan. They describe themselves as 'jihadists' that fight against the government and its foreign supporters. According to a report for the United States (US) Congress, not even 60% of the country's districts were under Afghan government control or influence in 2017 (SIGAR 2017, 87). At the same time the US is welcoming direct peace talks with the Taliban (Tolo News, February 24, 2016). This development raises the question as to what the affected peoplerather than the foreign intervenersthink about the Taliban.
In response, this article sets out to investigate the reasons why people in Afghanistan view the Taliban as legitimate or illegitimate. Such an analysis can help to move away from peacebuilding and statebuilding efforts, which rest on ideological assumptions of what ought to be built, to policies grounded in an understanding of what the affected people think. The article also helps to deal with the difficulty of conceptualizing legitimacy in conflict-torn spaces. To enable an analysis of legitimacy, a framework is developed herein that builds on the traditional understanding of authority and legitimacy, but which considers the characteristics of contexts with a low degree of monopolization of force. Three analytical ideal types are proposed that can help to explain obedience to social control: coercion, which achieves obedience through force and threats; instrumental legitimacy, which 'buys' obedience by responding to needs; and the traditional more substantive understanding of legitimacy, which is underpinned by shared values and a belief in rightfulness.
This framework is applied to an empirical analysis of the local perceptions of authority in Afghanistan. Insights into these perceptions were gained through an extensive number of interviews, conducted in different parts of Afghanistan in 2014 and 2015. 1 This paper draws on those interviews that were collected in the eastern province Nangarhar. It uses selected interviews to illustrate the findings of the wider analysis of perceptions in the province (Table A1 in the Appendix). At the time of these interviews, some districts of Nangarhar were under government rule and others were opposition-controlled. The interviews present a snapshot of the situationand its perceptionat a certain time, but more importantly they indicate that people judge the Taliban on the basis of what they actually do, and consider them to be legitimate or illegitimate depending on their day-to-day behaviour. In the government-controlled areas, the behaviour of the Taliban is perceived to be threatening and illegitimate by many, if not most. The main concern of people in these areas is not religious extremism but the attacks the Taliban launch and the instability they cause. But in territories under Taliban controlas well as in some territories that are government-controlled which the Taliban can accesssome people consider the Taliban to be legitimate. The findings suggest that this legitimacy is often not underpinned by 'big' ideological concerns, whether traditional or religious, but by the perception of being treated better by the Taliban than by the government. For instance, the Taliban successfully construct legitimacy by providing conflict resolution in a way that is perceived to be accessible, fast and fair. In the massively corrupt and volatile political order of Afghanistan, people appear to be longing mainly to be treated with respect and as equals, caring less about which code of law is actually applied. In some cases, the Taliban seem to be responding to these local demands better than the government.

Authority and legitimacy in conflict-torn spaces
The literature on legitimacy usually deals with political orders that are characterized by a high degree of monopolization of force. In such static settings, researchers can conceptually distinguish the institutionalized 'state' from the 'people' who bestow legitimacy. In this way, one can drill into the rational-legal source of legitimacy and investigate, for instance, how different policies affect a state's legitimacy. However, in a conflict-torn space there is, by definition, a contestation of at least two authorities, if not more. This conflict does not take place within the institutionalized structures of the state but outside of them. Usually the state is part of the conflict, fighting with armed opposition groups, whether they are called rebels, militias, insurgents or terrorists. In such a dynamic setting, the analysis of legitimacy becomes more complex due to the multiple referent objects, multiple audiences and multiple potential sources of legitimacy (von Billerbeck and Gippert 2017;Weigand 2015). The empirical analysis of a complex concept like legitimacy in such a dynamic setting requires a well-defined framework.
It is suggested that in order to conceptualize the political order of conflict-torn spacesthe context in which legitimacy 'happens'the static understanding of the Weberian ideal-typical state should be left behind. Adopting such blinkers narrows the focus down to the degree of monopolization of force in a defined territory. As a consequence, conflict-torn spaces are easily dismissed as 'failed' or 'fragile' states, a category based more on the absence rather than the existence of characteristics, and hence devoid of much analytical leverage. In order to focus on what is rather than what is not, herein the political order is considered to be dynamic and in a constant process of transformation. In line with scholars such as Migdal and Schlichte (2005), political order is viewed in the present study as a constantly changing arena of competition involving multiple authorities. As Agnew points out, 'political authority is not restricted to states and … is thereby not necessarily exclusively territorial ' (2005, 441). Hence, choosing authority as the unit of analysis allows more flexibility in dealing with the globalized world order (see Kaldor 2009). Nonetheless, for this view on political order Weber's work is still considered to be fundamental. The understanding of authority which informs the present study is influenced by the German term Herrschaft, suggesting an analysis of governance beyond government. Weber defines the term as 'the chance of a specific (or: of all) command(s) being obeyed by a specifiable group of people ' (1980, 122). Building on Weber, the expression authority is used to describe social control both as a relationship of command and obedience 2 and, accordingly, the (commanding) actor or entity to whose social control a group of people obeys.
The question driving this research is why people accept or even support an authority and obey. Conceptually speaking there are two options: either people want to obey or they have to obey. The voluntary obedience is based on legitimacy, whereas involuntary obedience is based on coercion. If authority rests on coercion then people are threatened or forced to obey in a violent or non-violent way. 3 The definition of legitimacy as voluntary obedience to social control is in line with traditional empirical definitions of legitimacy (e.g. Levi et al. 2009)however, its referent object is not necessarily the state or government, but can be any authority. Further developing the conceptual understanding of legitimacy, it is suggested that a distinction should be made between two different kinds of theoretical reasons for voluntary obedience, which reflect two fundamentally different ways of looking at legitimacy. For this purpose, the rational assessment of usefulness of authority is termed instrumental legitimacy, describing the extent to which an authority responds to needs. People may voluntarily obey authority simply because it is beneficial, or else because they have no alternative. But instrumental legitimacy is of a short-term nature, and only lasts as long as people benefit, or hope to do so in future. Instrumental definitions of legitimacy are implicit in policy literature through suggestions aiming at enhancing legitimacy through service delivery or improved performance. Substantive legitimacy meanwhile is a more abstract normative judgementa belief in rightfulness which is underpinned by shared values. Substantive definitions of legitimacy are dominant in the social science literature and go beyond the simple assessment of advantages and disadvantages, as they are centred around beliefs. If a person believes that an entity has the right to exercise social control, he or she may also accept personal disadvantages. Hence, according to Weber, belief in the legitimacy of an authority [Legitimitätsglauben] is necessary to achieve long-term voluntary obedience (1980,122).
The most complex of these ideal types is substantive legitimacy. While coercion rests on force and threats, and instrumental legitimacy rests on the provision of goods and services, there are various sources of substantive legitimacy which can be categorized in different ways. Weber famously distinguishes between rational-legal, charismatic and traditional legitimacy, underpinning the belief in the right to exercise social control (2009,(78)(79). Most of the contemporary literature on sources of legitimacy focuses more specifically on legitimacy within rational-legal systems that require a high degree of monopolization of force. For instance, Scharpf (1997) describes input and output legitimacy as two dimensions of a democratic system, where output goes beyond simple service delivery and is linked to the input of the people through representative institutions that ensure accountability. And the psychology literature, looking particularly at democratic policing, suggests considering the procedures of how police officers interact with the public (e.g. Tyler 2004Tyler , 2006Jackson et al. 2013;Mazerolle et al. 2013). Empirical research can help improve the understanding of the extent to which these concepts can also be used to explain voluntary obedience in a conflict-torn space with a low degree of monopolization of force. 4 In a nutshell, this article proposes three analytical ideal types which explain why people may obey social control. Coercion forces people to obey. If authority rests on the traditional understanding of legitimacy, or what is called herein substantive legitimacy, people obey because they believe in the rightfulness of authority. This belief may, among other potential sources, result from traditions or rational-legal structures. Instrumental legitimacy, underpinned by usefulness, fills the conceptual gap between these two explanations for obedience. This is obviously a purely conceptual exercise. The categories are interconnected, and in almost every empirical case different sources of authority will play a role. The analytical distinction does not necessarily contradict any conclusion on the relationship between the concepts. While normative definitions set the criteria with which to assess authority and legitimacy, these empirical concepts rest on Weber's general approach of making research as independent as possible from the researcher's own views and values (Beetham 1991). Hence, the framework allows researchers to work more inductively and investigate the perceptions of different groups of people to explain why they obey social control. The framework can be applied to analyse the sources of authority and legitimacy, independent of the political system in which the authorities operate and the degree of monopolization of force involved.
On this basis, this framework is used to analyse the perceptions of people who are affected by the Taliban in Afghanistan and then draw conclusions about the basis of their authority and legitimacy. This in turn can also help to create a better theoretical understanding of the mechanisms that explain obedience to authority, especially with regard to sources of substantive legitimacy in conflict-torn spaces.

The Taliban in Afghanistan
The Taliban's history and claim of legitimacy The 'armed opposition' or 'insurgency' groups in Afghanistan today are commonly associated with the label 'Taliban'. Indeed, after successfully turning an insurgency into a government in 1996 and being toppled again in 2001, the Taliban have returned to insurgency strategies to subvert the current government and its foreign allies. However, armed opposition today is a complex phenomenon, consisting of various groups and factions that change alliances fairly readily.
The history of the Taliban is closely linked to the 'jihad' against the Soviet occupation of Afghanistan and the subsequent civil war. 5 In response to the occupation, multiple Mujaheddin groups in the country took up arms to fight the Soviets, often with the support of Western countries that were channelling money and weapons through the Pakistani intelligence agency, the Inter-Services Intelligence (ISI) (Coll 2004). After the Soviet withdrawal from Afghanistan in 1989 and the fall of the pro-Soviet Najibullah government that held Kabul until 1992, the Mujaheddin groups turned against each other. Afghanistan quickly disintegrated into multiple regions controlled by different strongmen, fighting each other with changing alliances. At that time, the southern province of Kandahar was also divided. As foreign funding dried out in the province, competing commanders began to mistreat the population and extract money; the highways were littered with checkpoints of various groups who put chains across the road and demanded tolls (Strick van Linschoten and Kuehn 2012, 113-117;Rashid 2001, 22). A number of Mujaheddinmany of whom had stopped fighting since the fall of the Najibullah government and returned to study at madrassas [religious schools] in Quetta and Kandaharbecame increasingly concerned and disillusioned (Rashid 2001). In response, they formed a group around their leader Mullah Omar to 'restore peace, disarm the population, enforce Sharia law and defend the integrity and Islamic character of Afghanistan' (Rashid 2001, 22). They 'distanced themselves from the party politics of the Mujaheddin and signalled that they were a movement for cleansing society rather than a party trying to grab power' (23), calling themselves 'Taliban'the 'students' or those 'seeking knowledge'. However, this narrative of the history of the Taliban is disputed. Barfield (2010, 257) points to an alternative story of Pakistan shifting its support away from the strongman Hekmatyar and instead helping to form the Taliban to fight the new government in Kabul. In any case, the Taliban were successful in gaining local legitimacy quickly, as they portrayed themselves as a 'Robin Hood' taking a stand against Kandahari warlords (Rashid 2001, 25). Their popularity grew steadily, with more and more people joining the Taliban. By late 1994, the Taliban had taken control of Kandahar, and on 26 September 1996 they took Kabul (Barfield 2010, 258-260;Strick van Linschoten and Kuehn 2012, 144).
While the Taliban expanded their influence further to the north, they introduced a new political order for the territory they controlled. They called this territory the Islamic Emirate of Afghanistan, which was governed by Mullah Omar, who had been proclaimed Amir ul-Mumineen [Commander of the Faithful], and a small shura [council] of selected people. While they claimed to be implementing Islam, their way of de facto governing included many elements of the Pashtun's cultural code, the Pashtunwali. Building on this mix of values and ideas they enforced strict rules, including the prohibition of music and the exclusion of women from public life, which stood in sharp contrast to the lifestyle of the population in urban places like Kabul and Mazar-e Sharif (Barfield 2010, 261-263). They taxed the people, asking everyone to pay zakata form of income taxand making farmers pay ushr [Islamic tithe], 10% of their produce (Nojumi 2002, 155).
Meanwhile, a number of strongmen and their militias tried to defend the north of the country against the Taliban, banding together as the United Islamic Front for the Salvation of Afghanistan, which is also known as the Northern Alliance. When the US and the United Kingdom (UK) launched so-called Operation Enduring Freedom in October 2001, their air support enabled the Northern Alliance to quickly advance and capture Kabul in November of the same year. The Taliban first retreated to Kandahar before going underground, with some of their commanders moving to Pakistan. They were excluded from the negotiations on the future of Afghanistan in Bonn in December 2001, in whichapart from the international communitymainly strongmen of the Northern Alliance participated. These negotiations resulted in the establishment of the International Security Assistance Force (ISAF). While the Taliban were no longer of much relevance for the international community, they reorganized in the Afghan-Pakistani border area and began fighting again as an insurgency. Today the Taliban, who still call themselves the 'Islamic Emirate', claim legitimacy by portraying themselves as 'jihadists', fighting against the 'occupying' US forces and the 'infidel' government forces. 6 This rhetoric underpins most of the Taliban's public statements on their websites al-emarah and shahamat, 7 which are used to justify attacks and announce the beginning of the annual 'fighting season' in spring (D'Souza 2016, 24). For instance, in April 2016 the Taliban declared: The Islamic Emirate's armed Jihad against the American invasion has completed fourteen years and is now in its fifteenth year. Jihad against the aggressive and usurping infidel army is a holy obligation upon our necks and our only recourse for reestablishing an Islamic system and regaining our independence (Shahamat English, April 12, 2016). 8 Indeed, despite the ongoing international support, it is estimated that, in 2017, only 59.7% of Afghanistan's districts were under the control or influence of the Afghan government, with the insurgents controlling 11.1% and the remaining 29.2% of the districts being contested (SIGAR 2017, 88). 9 The fighting over the major city of Kunduz in the north of Afghanistan in 2015/16 illustrates that there is a chance of insurgency control continuing to expand, even into urban areas. However, while the numbers look impressive, the Taliban are also considered to be responsible for a large amount of civilian casualties (UNAMA 2017). Furthermore, the Taliban today are a fragmented movement, as illustrated by the complexity of the insurgency in Loya Pakita: It is composed of four different strands. There are two networks led by the Haqqani and the Mansur families respectively. Besides them, there are Taliban groups acting independently from these two networks, led directly by the Taliban Rahbari or Ali Shura (Leadership or Supreme Council) or by individual influential commanders in Quetta. (Ruttig 2009, 59) But the exact relationship between the various groups remains unclear. Ruttig concludes: 'The Haqqani and Mansur networks are clearly part of the Taliban universe … but their modus operandi is that of semi-independent warlords who have joined the rather heterogeneous insurgency movement for reasons of expediency ' (2009, 88). 10 Probably the best-known institution of the Taliban is the Quetta Shura. Mullah Omar used to be its head, with Mohammad Mansur as his deputy. In addition, there is the socalled Peshawar Shura, which has allegedly gained importance since Pakistan's ISI started shifting funds away from the Quetta Shura after it began opening up for negotiations and trying to open an office in Qatar (Martin 2014, 208). The fragmentation of the Taliban appears to have continued since Mullah Omar was revealed to have died in 2013. In July 2015, Mohammad Mansur was announced as Mullah Omar's successor (Giustozzi and Mangal 2015). However, members of the Rahbari Shura publicly complained about this decision, with one group led by Mohammad Rasoulwho fought the Soviets together with Mullah Omaropenly declaring war. Meanwhile the head of the Haqqani Network, Jalaludin Haqqani, was appointed as a deputy of Mansur (Osman 2015). Mansur was subsequently killed in a US drone strike in Pakistan in May 2016 and replaced by Mawlawi Haibatullah Akhundzada. However, according to Farrell and Semple (2017), Akhundzada is considered to be a weak leader by many within the Taliban, contributing to an ongoing fragmentation of Afghanistan's armed opposition. This fragmentation on the macro level is complemented by similar processes on the micro level. For instance, Smith describes the dynamics of the insurgency in Kandahar, explaining that 'there is no evidence that the formal structure of the insurgency has any real importance ' (2009, 193). He points out that the Taliban in Kandahar are a much more dynamic and fluid 'entity', consisting of different units (see also Jackson and Giustozzi 2012). 11 The Taliban's public legitimacy While the Taliban portray themselves and claim legitimacy on the basis of being 'jihadists' and wanting to 're-establish an Islamic system', it remains unclear how their legitimacy is perceived by Afghans. The importance of gaining a better understanding of what underpins legitimacy in the Afghan context has been widely acknowledged. Scholars working on peacebuilding, statebuilding, development and the international intervention in Afghanistan have emphasized that a major reason for failure is one of legitimacy (e.g. Goodhand and Sedra 2013;Kühn 2008;Roy 2004). While the international approach in Afghanistan focused on gaining legitimacy in the intervener's countries, local perceptions of what is legitimateor notwere often ignored (Rubin 2006, 184-185) or misconstrued (Coleman 2017).
Building on this realization, most of the literature on legitimacy in Afghanistan looks at the state and the government. More quantitative, survey-based studies focus almost exclusively on service delivery, indicating an instrumental understanding of legitimacy. For instance, Sabarre et al. (2013, 17) conclude that 'perceptions of security are key indicators of legitimacy scores'. Scholars with a more substantive understanding of legitimacy often claim that local customs are crucial, pointing at what Weber would call traditional legitimacy. For example, Libel emphasizes the importance of the Pashtunwali over religion for the Pashtuns in the south and east of Afghanistan: Pushtuns have tended to accept any government and its civic laws as long as the government is controlled by Pushtuns and follows the basics of the Pashtunwali, and as long as its governmental laws and decisions are sanctioned by Pushtun jirgas. (Liebl 2007, 507) The notion that legitimacy in Afghanistan is linked to traditions has also been adopted in the policy world. A USAID study claims: 'Legitimacy begins with empowering the local level with traditional decision making processes and from there, slowly establishing links with the Weberian rational-legal institutions of the state can ensure accountability' (Melton 2015, 4).
Conversely, other scholars point out that there are more general requirements which the state needs to fulfil to be considered legitimate. Barfield and Nojumi argue that historically, the government delegated authority to non-state actors, acting as mediators for grievances on the local level; according to their analysis, '[t]his system was highly functional and grounded in local perceptions of fairness and trust. It crossed ethnic, linguistic and tribal boundaries with ease because it was in the interest of all parties to cooperate' (2010,42). Similarly, Roy argues that the 'Afghan identity is based on a common political culture ' (2004, 173). He suggests three criteria that the state in Afghanistan needs to fulfil to be considered legitimate: building on the concept of Afghanistan being an independent Muslim territory, acting as a mediating broker between competing groups and providing basic services.
The Taliban's legitimacy post 2001 on the other hand has not been investigated much. Perception data indicates that public support for armed opposition groups in Afghanistan has been decreasing since 2009. According to the Asia Foundation (2016, 7), only 16.7% of the people in areas their researchers can access have sympathy for armed opposition groups today. But this does not explain why people support the Taliban or consider them to be legitimate. Some of the sources of legitimacy that are discussed in the context of the Afghan state may also matter for the Talibanin particular, 'tradition' in the form of the Pashtunwali could still play a key role. Liebl (2007) suggests that the people's expectations in the Pashtun areas of Afghanistan regarding governance rests on the Pashtunwali. At the same time, the Pashtunwali influences how the Taliban govern, matching local expectations. But service delivery may also provide the Taliban with some legitimacy. Giustozzi (2012) offers some helpful additional insights. On the basis of interviews with members of the Taliban, he concludes that the provision of justice is central. When the Taliban manage to establish a local monopoly of force, Giustozzi argues, they can construct legitimacy by providing justice: 'by and large, the Taliban seem to have greatly benefited from their ability to mediate disputes among communities. It could be argued that such ability is a major source of legitimacy for the Taliban' (2012,73). Building on these ideas, this paper further contextualizes the perception data and compares the Taliban's claims to legitimacy with the reasons why people perceive them to be legitimateor not.

Local perceptions of the Taliban
To gain a better understanding of the Taliban's authority and sources of legitimacy, the ways in which they are perceived by members of the public are examined. To do justice to the 'fluid' nature of the Taliban, with characteristics varying spatially and changing over time, the analysis focuses on Afghanistan's Eastern province Nangarhar in 2014/ 15 ( Figure 1). This geographical focus complements the existing literature on the Taliban that concentrates mainly on the south of Afghanistan.
Travelling from Kabul, it only is a three hour ride by car to the city of Jalalabad, the capital of Nangarhar Province. The road follows Kabul River and twists down 1300 metres through the mountains. In the mountains one can see patches of new asphalt on the road frequently, serving as reminders of the fuel trucks that have caught fire after being shot at from the mountains. Surobi District, which is one of the eastern districts of Kabul Province, is well known for frequent attacks on the Afghan National Security Forces (ANSF). Leaving the mountains of Kabul Province behind, the road reaches the plains of eastern Afghanistan, first Laghman Provincewhere many people take a break for a photo with an old Soviet tank lying next to the roadand finally Jalalabad in Nangarhar Province. The terrain is much flatter and the temperature is much higher here than in Kabul. In this part of the country it hardly ever rains, and the heat is often unbearable in summer, but because of the pleasant temperatures in winter, Jalalabad was a popular seasonal home for many Afghan kings over the centuries. Today, with about 200,000 inhabitants, Jalalabad is one of the biggest cities in Afghanistan. From here it is barely 80 km to the Pakistani border. Indeed, the main currency used in this part of Afghanistan is the Pakistani rupee. Like most other parts of Afghanistan, the predominantly Pashtun province of Nangarhar was controlled by the Taliban from 1996 to 2001. Today governance in Nangarhar is dominated by a number of strongmen, who compete for influence (Jackson 2014). In the period during which the interviews were conducted, the Taliban had gained full control of many parts of the province again. However, the provincial capital Jalalabad remained under government control, with some districts close to the city being 'grey' areas, technically controlled by the government but also influenced by the Taliban. But even though the visible presence of the Taliban in Jalalabad was low, there was an atmosphere of fear, driven by suicide attacks, kidnappings and robberies.
The interviews were conducted in the format of natural conversation, triggered by the use of open-ended questions. As legitimacy is an abstract social phenomenon, which may be understood differently by different people, it was not used during the interviews. Instead, the discussion was on the Taliban more generally, using two thematic examples of exercising social control to investigate perceptions that can be related to legitimacy: security provision and conflict resolution. Of particular interest for the purposes of this research are the interviewees' explanations of strong positive or negative views regarding the Taliban, the reasons why they thought that the Taliban or other authorities should be responsible for security provision and conflict resolution, and their views as to who should govern Afghanistan in the future. Interviewees were selected in such a way that they covered a wide range of characteristics with regard to age, sex, income, social position and district of residence. The research does not, however, build on a 'representative sample' in a positivist sense; being about the mechanisms of legitimacy, it does not aim at drawing general conclusions on the extent of the Taliban's legitimacy in Nangarhar, let alone Afghanistan. The details of the interviews this article draws on are given in the Appendix. 12 The selection of interviewees also depended heavily on pragmatic considerations, particularly accesshence, certain groups of people, particularly women and people living in areas controlled by the Taliban, are underrepresented. In addition, certain ideologies or views might be underrepresented, for instance because people who are critical of foreigners may be less likely to participate in a research project. It also turns out that the number of people interviewed who supported the Taliban is much lower compared to those who perceived the Taliban as a threat. As a consequence, it was necessary to work with a comparatively small number of interviews to understand what legitimizes the Taliban, while there are many more interviews which shed light on what delegitimizes them. However, understanding what delegitimizes the Taliban also helps to draw conclusions on the sources of legitimacy.

Jalalabadthe Taliban as a security threat
Even though the Afghan state was in control of Jalalabad at the time of this research, the Taliban frequently arose as a topic of conversation in the interviews, as both a historical and a contemporary phenomenon. Looking back to the time when the Taliban were in power, most interviewees had very positive memories and stated their satisfaction, particularly in terms of security. Comparing the past with the present, the mechanic Rahmanullah stated: 'I think the security situation is worse than 15 years ago. During the Taliban regime there were no kidnappings or other crimes in our province' (N31). Similarly, Khyber, a middle-aged man who was unemployed at the time of the interview, concluded: During the Taliban regime there were no robberies, kidnappings or suicide attacks. During the first term of Karzai, the situation was good too. But now it has changed completely. The number of suicide attacks and kidnappings have increased, and there is a high level of corruption in the public administration. (N30) Views like those of Rahmanullah and Khyber were widely shared, and not limited to Jalalabad. Also, in the more rural districts of Nangarhar, people often had good memories of the time when the Taliban were in power (e.g. N34, N35, N36). Nonetheless, some people, less frequently, also acknowledged that not everything was good during the Taliban's regime. For instance, the farmer Taher explained that under the Taliban the security situation was better but that his economic situation was worse (N28).
Today the phenomenon of the 'Taliban' is considered to be more complex. There are a number of armed 'opposition' groups with different interests and alliances, sometimes linked to strongmen or to the government, sometimes claiming to be 'Taliban' and sometimes only being labelled as 'Taliban' by others. The armed groups that are mentioned most frequently in the interviews are Hekmatyar's Hezb-e Islami 13 and the forces controlled by the Quetta Taliban Shura, also referred to as the Islamic Emirate, as well as the militias of the various strongmen. A local civil society activist explained the structure of the armed groups in Nangarhar as follows: There are two main opposition groups here. It's the Islamic Party [Hezb-e Islami] and the Taliban. The Islamic Party usually only attacks foreigners, not the ANA [Afghan National Army] or the ANP [Afghan National Police]. The Taliban attack foreign and government forces. The Taliban also collect ushr and zakat, which the Islamic Party isn't doing. The Taliban are much stronger and consist of different factions. The government is linked to these groups as it has some of its people within the Taliban. In addition, strongmenwhich often are part of the government toohire militias to achieve their personal interests. Every strongman, every political party and every group has its own armed force. Most people just call all of these armed groups Taliban. (N18) In 2015, a new opposition actor suddenly entered the ring, the so-called Islamic State Khorasan (IS-K), which people in Afghanistan widely referred to as Daesh. While in late 2014 not a single interviewee mentioned IS-K, in 2015 almost every interviewee was concerned about its growing influence in the province. At the time of the interviews, reports on the Islamic State in Afghanistan had been limited to Nangarhar. But soon after the Taliban declared 'jihad' against the group (Khaama Press, April 20, 2015; Osman 2016). 14 Most of the people interviewed in Jalalabad had strong negative views on the Taliban as being responsible for multiple attacks that had killed a large number of civilians in the city. The positive memories from the past rarely translated to the present perception. For example, the university student Gulagha proclaimed: 'Taliban means "the ones who seek knowledge". But now it is nothing more than the name of an illegally armed group, which is destroying the country with bombs and suicide attacks' (N10). Similarly Rohullah, also a student, stated: 'Taliban for me means murderers. Especially city people hate them' (N08). Very negative perceptions were particularly prominent in the second round of interviews in the spring of 2015, reinforced by an attack on Kabul Bank in Jalalabad killing and injuring dozens of people around the time of the interviews. But Rohullah, like many of the interviewees, also blamed the Taliban for crime in the city. Kidnappings by unknown groups were a regular occurrence during this period, and many people attributed them to the Taliban even though they appeared to be carried out only for income generation via ransom. Rohullah summarized: 'Jalalabad is not secure at the moment. On a daily basis we witness bombings and the crime rate is going up' (N08). Although people in Jalalabad blamed the Taliban for all kinds of criminal activities, they still perceived them as a political actor (e.g. N27, N29)albeit one that is not necessarily fighting for ideological reasons, but as a tool of foreign intervention driven mainly by Pakistan to consciously cause insecurity in Afghanistan: 'In case the Taliban managed to stop the infiltration of their organization through foreigners they might be accepted. But our nasty neighbour Pakistan causes insecurity through the Taliban' (N07). Some people in Jalalabad had also experienced recent personal encounters with the Taliban. Based on these encounters, people in Jalalabad did not necessarily perceive the Taliban to be an armed group linked only to Pakistan, but also saw connections to the Afghan state. For instance, the civil society activist told me: My son was kidnapped and I was asked to provide them [the kidnappers] with 10 AK47s. … When I was walking out of the police station after filing my case I got a call from the Taliban telling me that the police couldn't help me. About six months ago I was attacked in front of a police station. And after filing my case they arrested a number of police officers. They called me to recognize their faces, but I was too afraid to do so. In fact, the police commanders cooperate closely with the illegally armed groups. Whenever these people notice that their interests are in danger they will take advantage of their links with the insurgents and call on them for help. (N18) While the civil society activist was particularly open about his experiences, other interviewees also complained about the links they see between the police and the Taliban (e.g. N08).
As far as the people interviewed in Jalalabad were concerned, the Taliban did not have much authority, as the people did not obey themeither voluntarily or because of coercion. They saw the Taliban mainly as a threat from the outside and a driver of insecurity, an agent of Pakistan that is responsible for numerous attacks. But people also blamed the Taliban for other crimes, often perceiving them as partners rather than opponents of criminal government authorities. Even though many interviewees had positive memories of the Taliban regime, these memories did not translate into an overall positive perception of the Taliban, as their perceptions rested only on the present activities of the Taliban. The violent behaviour had delegitimized the Taliban in the eyes of the people, while the Taliban's attacks in Jalalabad also illustrated that the Afghan state was incapable of providing security for its citizenseven in major urban areas. 15 Hence, people had become dissatisfied with the Afghan state, making the attacks a successful delegitimization strategy for the Taliban.

Behsod and Surkh Rodthe Taliban as a sporadic phenomenon
In the other parts of Nangarhar province in which research was conductedthe districts of Surkh Rod and Behsod (see Figure 1)people shared an overwhelmingly negative perception of the Taliban.
In Behsod, which is a more rural district that is in close proximity to Jalalabad, almost all the interviewees saw the Taliban as the main source of insecurity. However, while some were concerned that 'security [was] getting worse day by day' (N34; see also N36, N40, N47, N51), most people were confident about the security situation (e.g. N35, N38, N39, N45, N48, N49). To explain this variation, many people pointed out that the Taliban were not active in all parts of the mainly government-controlled district. However, the Taliban did infrequently show up in some areas. For instance, the farmer Haiatullah reported: 'There are no Taliban in our village. But they sometimes show up in Samarkhel village close by' (N44). In contrast to Jalalabad, the experience of insecurity in these areas was based less on attacks and more on the Taliban's attempts to extract money. Like other interviewees, the non-governmental organization (NGO) employee Subhanullah reported: 'Every now and then the Taliban come from Khogyani and Chaparhar districts and ask us for tithes. And when they come we feel insecure' (N40). But despite the limited presence of the Taliban in the district and their negative perception, they managed to recruit successfully in Beshod. An interviewee explained: 'Many people from our district go and join the Taliban because of poverty and unemployment' (N50).
Only one of the interviewees in Behsod had an openly positive perception of the Taliban. The village elder Wais explained that he liked the Taliban because they were helping to solve conflicts on the community level: Here in our village two brothers had a conflict on land rights. They went to the government to solve the conflict but nothing happened for a month. Then one brother went to the Taliban instead. They solved the conflict very quickly. And the result was acceptable for both brothers. (N34) As he was unsatisfied with the government's conflict resolution procedure and the growing insecurity, he expressed hope that in the future the Taliban would be the only provider of security and conflict resolution. He also explained that the Taliban offered 'mobile courts', making access much easier than the government. In the case of a conflict, people could request that they come by motorbike to help. Despite his preference for the Taliban over the government, Wais criticized, like everybody else, the interference of neighbouring countries: 'I don't like Iran and Pakistan because they cause insecurity in our country' (N34). But in contrast to most of the other interviewees, he did not see a connection between the neighbouring countries and the Taliban. Furthermore, he did not consider himself to be a Talib, and carefully distinguished between 'the Taliban' and 'the people in the village' in the interview. He complained that foreign and government forces sometimes accuse people from his village of being members of the Taliban: 'A month ago international forces and ANA soldiers came and accused our Mullah [of being] a Talib. They looted his house and burned his motorcycle even though he is innocent' (N34). According to Wais this reinforced the perception in his village that the government was a threat while only the Taliban could provide security.
The third area of Nangarhar in which this research was conducted is the district of Surkh Rod. Surkh Rod, like Behsod, is rural but still close to the urban Jalalabadand like in Behsod, the interviewees there had mixed perceptions of the security situation, but widely agreed that the Taliban were the main threat (e.g. N01, N02, N03, N04, N05, N21, N23). However, according to the teacher Nader, the Taliban were only present and a concern in the Kakrak area of Surkh Rod at the time of the interviews (N16). Like in the other districts, the label 'Taliban' was used to describe all kinds of armed groups. For instance, Mohammad, a member of a local council reported: 'In general, security is enforced and there is no threat from insurgents. But there are criminal activities. … The Taliban use weapons to kidnap, rob, ambush and threaten people' (N05). The head of a civil society organization further explained that many people and groups describe themselves as Taliban: 'A great number of people are only fighting for their personal interests. But they do so using the Taliban label' (N02). As in Behsod, many people were reported as joining 'the Taliban'although it remains unclear which group exactlydue to the pressures of poverty induced by unemployment (e.g. N16). Many interviewees concluded that the violence in the country was not a consequence of opposing ideologies, but instead only a matter of economic interests.
In Surkh Rod and Beshod, it was the state authorities who dominated the political orderand, like in Jalalabad, many people perceived the Taliban as a threat. In those parts of Behsod that were accessed frequently by the Taliban, this perception did not only rest on attacks and criminality in general, but, more specifically, also on the experience of the Taliban demanding obedience to their rules and attempting to extract money from people. Here the Taliban were an actual authority for people, albeit one founded on coercion rather than legitimacy. However, despite their negative reputation in both districts, some people decided to join the Taliban in order to make a living. And not everybody saw the Taliban as a threat, as the case of Wais indicates. While his support for the Taliban was an exception, helike the other intervieweesformed his opinion on the basis of his perception of the behaviour of the state in comparison to the Taliban on a day-to-day basis. His support for the Taliban was based on the perception of the Afghan state not only being incapable of providing security and conflict resolution, but even behaving in a threatening way, while he felt that the converse was the case with the Taliban. He therefore voluntarily chose the Taliban for conflict resolution, indicating an authority resting not only on force but also on legitimacy. The Taliban's legitimacy in this case was based on very rational considerations: the experience of procedures he considered to be fast and fair. None of the interviewees explained support or rejection of the Taliban with more ideological considerations, such as a preference for a certain code of law (e.g. state law or Islamic law) or a preference for a certain defined procedure of how an authority should be gaining power (e.g. democracy or theocracy). Indeed, the interviews illustrate that pragmatic decision-making seems to play a key role, with people joining the Taliban to escape poverty and unemployment.
Khogyani and Sherzadthe Taliban as a provider of fast and fair conflict resolution A number of interviewees in Jalalabad, Behsod and Surkh Rod turned out to be from other more distant districts, such as Khogyani, Chaparhard and Sherzad (see Figure 1). Their views offer valuable additional insights, as these districts were often described as the Taliban's 'bases' for driving insecurity in Nangarhar in the other interviews, but were difficult to access due to safety considerations. And indeed, many of these districts appeared to be at least partly controlled by armed opposition groups at the time of the interviews.
The shopkeeper Obid (N13) and the engineer Khalid (N14) together described the situation in Khogyani District in the autumn of 2014. According to them, the security situation had become worse since 2010: 'Five years ago at least the main roads were safe. Today not even these roads are safe any more, as the number of bombings and kidnappings has increased dramatically' (N13 and N14). They assumed that the reason for this was the collaboration between government forces and other armed groups: 'The security is both enforced and sabotaged by the government, as the government cooperates with insurgency groups' (N13 and N14). And, according to them, the political landscape of insurgency groups was complex: There are … the Islamic Political Party [Hezb-e Islami], the Islamic Emirate, Mahaz, a new group called Karwan Fidaye as well as ordinary criminals. Karwan Fidaye is particularly active in Khogyani. Once they get you, escaping from them is very challenging. The members of this group cover their faces with black masks and people say that most of them are from Punjab in Pakistan. This group fights both the Taliban and the government. (N13 and N14) According to the two interviewees, all of these groups would make attempts to extract money: In rural areas they come several times per year and ask landowners to pay tithes. In cities they target rich people, by calling them or sending them threatening letters, asking for money. Two people who introduced themselves as members of the Mahaz Party called me recently, threatened me and asked for money. (N13 and N14) But the interviewees also felt threatened by the government: My father in law spent US$90,000-100,000 on building a house in the Ahmad Khail area of Khogyani. Now the head of the Afghan Local Police, Malik N. has taken the house. He refuses to leave the house, and doesn't even pay rent. The Taliban started to attack him and as a result the house got partly destroyed. A suicide bomber even blew himself up inside the house. (N13 and N14) They were not able to solve the conflict in the formal system: When we wanted to complain about our house being stolen at the district police department, they didn't even let us inside because the head of the district police is also the head of the village and a friend of Malik N. (N13 and N14) Given these circumstances, these two interviewees saw Taliban courts as the only feasible alternative: 'The next time I am involved in a conflict I will go to the Taliban first. They solve conflicts quickly. The government isn't solving my problems, it's making them bigger' (N13 and N14).
Another interviewee, Wasiullah, a young man from Sherzad District, was even more convinced by the Taliban: Sherzad district is completely controlled by the Taliban … I think it is much better in Sherzad than in Jalalabad. Because in Jalalabad there are two governments in one city, in Sherzad it is only the Taliban. And all people in Sherzad are happy with the Taliban. (N33) According to Wasiullah, the reason for this was not only the good security but also the fast and non-corrupt conflict resolution: When a person has a problem he goes to the Taliban. The Taliban then refer the conflict to their Hoquqe Department and courts. Their conflict resolution procedure is much simpler than the formal justice mechanisms. The decisions are made on the basis of Islamic law, and they are fast. They sometimes ask for a small bribe to cover the expenses of their motorcycle. But apart from that there are no costs for the involved parties. (N33) When asked about the use of force by the Taliban, Wasiullah responded: 'Yes, they use force if necessary. But they are not corrupt. So people prefer the Taliban's conflict resolution' (N33). He expressed his hope that the Taliban would be governing the country again soon.
It can be easily doubted whether, as Wasiullah claims, all of the people in Sherzad were happy with the Talibanbut it has to be acknowledged that some people in Sherzad District apparently did prefer the Taliban to other authorities, and that the Taliban have legitimacy because they do some things better than the government. And again, it is their day-to-day behaviourthe perceived speed and comparatively lower level of corruptionthat was a major source of their legitimacy. The interviews with people from Khogyani District further support this narrative. The interviewees complained about the various insurgency groups collecting tithes. Nonetheless, they preferred the Taliban courts to the government ones for conflict resolution. This preference rested on their personal negative experiences of theft and unfair treatment by people associated with the government, combined with the perception of comparatively fast and fair procedures at the Taliban courts.

Conclusions
Given the complexity of the situation in Afghanistan as outlined throughout this article, it is difficult to define who the Taliban actually are, and how to categorize and distinguish them from other armed groups. Furthermore, a very prominent perception is that some of the armed 'opposition' groups are closely linked to criminal government authorities. Nonetheless, all of the people interviewed did have strong views on 'the' Taliban. These views were predominantly negative, which corresponds with survey data (Asia Foundation 2016). However, broader conclusions on the extent of the Taliban's legitimacy should not be made on the basis of this research, as the sample size is very small and is not aimed at being representative. Whom people consider to be a legitimate authority can vary from village to village, from person to person and, over time. Still, the reasoning that underpins these different perceptions is very similar, indicating that the mechanisms that legitimize or delegitimize authorities are of a more general nature and remain relevant regardless of how the situation in Afghanistan develops.
Overall, people appear to assess the Taliban pragmatically, predominantly based on the day-to-day experience of their behaviour rather than on their history or ideology. Further empirical research is necessary for a more complete picture which more closely examines Taliban-controlled areas, considers other provinces in Afghanistan and investigates the views of a larger number of peopleincluding the self-perception of Talibs. However, two conflicting images of the Taliban emerged from Nangarhar. While people in most parts of the government-controlled territories looked at them through the lens of 'insecurity', some people in the Taliban-controlled territories viewed the Taliban more favourably, focusing on their effective role in 'conflict resolution'. These views are closely aligned with the two different roles the Taliban play in Afghanistan, governing some parts or offering services that are associated with governing authorities while fighting in others.
On the one hand, in the government-controlled areas, most of the interviewees perceived the Taliban as a coercive authority or simply as a threat to their security, a group acting in the interest of Pakistan. This view stands in stark contrast to the picture of the Taliban as 'jihadists', which they use to claim legitimacy. However, it is not religious extremism per se that people were concerned about; the negative public perception rested on attacks and other criminal activities in the city for which the Taliban were blamed. While some had positive memories of the Taliban regime, these thoughts did not translate into a positive perception today. In sum, it appears that the Taliban were not even trying to construct legitimacy, but were focusing on undermining the legitimacy of the state by illustrating that it could not protect its citizens. Nevertheless, in spite of the widespread rejection of the Taliban, some people from these areas decided to join the group in order to make a living. This choice to not only accept but become active with the Taliban can be read as legitimacy. However, this legitimacy appears to be purely instrumental, based not on beliefs but instead arising out of economic need.
On the other hand, some residents from Taliban-controlled areas and territories that were technically government-controlled but accessible to the Taliban had more positive opinions. These people described the Taliban as their preferred authority, or viewed their relationship with them as a voluntary onewhich likewise indicates legitimacy. But again, it was not the Taliban's 'jihad' that people focused on; in terms of the Taliban's role in security provision, their legitimacy appears to have been instrumental. People viewed the Taliban as the best security provider for very pragmatic reasons, not because of values and beliefs. For instance, they considered armed opposition groups to be threatening but thought that the state was even worse, or they simply preferred having a monopoly of force over ongoing violent competition.
But the Taliban's role in security provision was marginal for most, who instead focused on their positive role in conflict resolution, confirming Giustozzi's (2012) conviction that justice plays a key role in the Taliban's legitimacy. Supplementing Giustozzi's theory, these findings indicate that the Taliban apply this strategy not only in territories in which they have a local monopoly of force, but also in some accessible territories that are government-controlledthereby attracting people for conflict resolution and building legitimacy. This is noteworthy, since in the absence of a local monopoly of force, the Taliban's ability to enforce decisions is limited and requires more popular acceptance. Some people choose the Taliban because they think that the group responds better than the state to their need for fast and cheap conflict resolution, and accessibility plays an important role as well: for some, the Taliban courts are closer than the state ones, and in some instances its 'mobile courts' even come to the village. As such assessments are based on usefulness and personal advantage, they again illustrate that the legitimacy of the Taliban is very instrumental. This prevalence of instrumental legitimacy shows that many people simply think that the Taliban are the best available choice, or at least the lesser evil. Importantly, this makes their authority vulnerable. Supporters may easily be convinced to follow competing authorities, such as the state, if they were accessible and offered services of a similar speed and standard.
However, there is also a substantive dimension to the Taliban's legitimacy in this context, as the interviewees also preferred the Taliban because of the fairness and predictability of their conflict resolution procedures. The repeated description of the Taliban as not being corrupt, or at least being less corrupt than the state, indicates that people do not only choose the Taliban for conflict resolution because they assume that they will have the best chances of winning their case or securing the best outcome, but also because the procedures correspond with a shared belief of what is right. This shows that the perceived fairness of proceduresa source of legitimacy emphasized by scholars like Tyler (2004Tyler ( , 2006 in the context of political orders with a high degree of monopolization of force, as well as by Barfield and Nojumi (2010) in the context of Afghanistan's historycan play a central role in the construction of legitimacy in conflict-torn spaces today. More ethnographic research is necessary to explore the values that underpin the notions of 'corruption' and 'fairness' in Afghanistanbut the present research indicates that these values, most notably equality, are very general ones that are not necessarily linked to Islam, to the Pashtunwali or to other culturally-specific ideals. People seem to care less what kind of law is appliedwhether it is state law, Islamic law or 'traditional' lawthan about how the law is applied. This finding illustrates that the Taliban's second claimed source of legitimacythe 're-establishment of an Islamic system'does not matter much for the people. What matters much more is the perception of the Taliban as being less corrupt than the state.
Put differently, the people's immediate concern is having any rule of lawregardless of its ideological sourcesto counter the perceived high level of corruption and arbitrariness. It is only when the procedures are clearly defined and implemented accordingly that there is a certain degree of predictability for people; and only when people have the feeling that everybody is treated the same way, regardless of money or influence, are the procedures of an authority considered to be fair. By living up to these very basic expectations and making people feel like equal subjects to their authority, not perfectly but better than the state, the Talibaneven if only perceived as the lesser of two evilshave already constructed some substantive legitimacy. This means that if the Afghan state is to win the fight against the Taliban, it is not sufficient to defeat them on the battlefield; the state also has to construct legitimacy more successfully than the Talibanand, without having to compromise on human rights or democratic values, it can do so by taking some cues from its enemy.
Notes 5. For a detailed history of the Taliban, see e.g. Barfield (2010), Cramer and Goodhand (2002), Edwards (2002), Rashid (2001), and Strick van Linschoten and Kuehn (2012). 6. For a discussion of the Taliban's ideology, see Gopal and Strick van Linschoten (2017). 7. The website shahmat was offline at the time of publication. 8. For a discussion of the Taliban's views on a future state, see Osman and Gopal (2016). 9. In 2014/15, the period during which the interviews were conducted, the number of districts controlled or influenced by the Taliban was still considerably lower, but was growing steadily (see e.g. Smith 2014). 10. For a recent analysis of the relationship between the Taliban and the Haqqani Network, see Joscelyn and Roggio (2017). 11. Jackson and Giustozzi's (2012) work on humanitarian access to areas under Taliban influence further illustrates how policies of the Taliban leadership are not necessarily implemented on the local level. Therefore, 'rules are fluid and vary depending on who is in charge' (Jackson and Giustozzi 2012, iii). 12. For their protection, the names of all interviewees were replaced with pseudonyms. 13. Hezb-e Islami was one of the biggest Mujahedin groups during the Soviet occupation, established and led by Gulbudin Hekmatyar. From 2001 to 2016, and in the period during which the interviews were conducted in Afghanistan, Hezb-e Islami consisted of a political wing, a political party with numerous MPs in the Afghan parliament, and a military wing that was fighting the Afghan state as an insurgency group under the leadership of Hekmatyar. After signing a peace deal, Hekmatyar returned to Afghanistan in May 2017 (see Rasmussen 2017). 14. For a summary of the history of the Islamic State in Nangarhar, see Osman (2016). 15. For a discussion of 'spectacular violence' in Kabul, see Esser (2014).