Khans & Warlords in Pashtun Society PART 1

Khans and Warlords: Political Alignment, Leadership and the State in Pashtun Society

- 1 -

PART I

Political Alignment, Leadership and the State in Pashtun Society

kit10 .info Bern, Switzerland

_________________________________________________________________ © José Oberson (kit10.info) 2002 / Institute for Ethnology, University of Berne

Khans and Warlords:

Political Alignment, Leadership

and the State in Pashtun Society

Anthropological Aspects

and the Warlordism Debate


- 2 -

Contents PART 1 ................................................................................................................................................... 4

POLITICAL ALIGNMENT, LEADERSHIP AND THE STATE IN PASHTUN SOCIETY....... 4

1 Introduction ........................................................................................................................................ 4

2 A macro-perspective: The wider environmental and historical setting ........................................ 9

2.1 The geographical setting ............................................................................................................. 11 2.2 Ethnonyms, identity and common history .................................................................................. 11 2.3 Tribal society and the state.......................................................................................................... 16 2.4 Socio-economic variability ......................................................................................................... 17 2.5 ‘Nang’ and ‘qalang’: two distinctive socio-economic categories ............................................... 19

3 Segmentary tribal organization and politics.................................................................................. 22

3.1 The Pashtun tribal system ........................................................................................................... 23 3.2 Tribal affiliation and political matters: Does it matter? .............................................................. 26 3.3 Contractual alliance and the two-bloc system: Conceptual explanations ................................... 29 3.4 Tribes and politics: Critical and conclusive remarks .................................................................. 32

4 Pashtunwali: The tribal law ............................................................................................................ 36

4.1 ‘Nang’, the concept of honour and shame .................................................................................. 37 4.2 ‘Badal’ and ‘tarboorwali’ (agnatic rivalry) ................................................................................. 40 4.3 ‘Melmastia’ ................................................................................................................................. 41 4.4 ‘Nanawatee’ ................................................................................................................................ 41 4.5 Jirga............................................................................................................................................. 42

5 Being Pashtun – oncepts of person and politics............................................................................. 43

5.1 ‘Turá’: the sword......................................................................................................................... 43 5.2 ‘Aql’: responsibility and the corrective of ‘turá’ ........................................................................ 44 5.3 Egalitarianism: ’har saray khan day’ – every man is a khan....................................................... 44 5.4 Pride and the primacy of emotional control................................................................................ 46 5.5 The Pashtun Code, concepts of the ideal Pashtun person and their impact on leadership.......... 47

6 Forms of political leadership and the state .................................................................................... 51

6.1 Leadership: Traditional political elite ......................................................................................... 52 6.2 Political leadership in nang society............................................................................................. 54 6.3 Political leadership in qalang society.......................................................................................... 55 6.4 Sources of authority .................................................................................................................... 57 6.5 Traditional religious elite ............................................................................................................ 59

7 The theory of person-centred politics ............................................................................................. 61

8 Critical remarks and contemporary contextualization................................................................. 65


- 3 -

PART 2 ................................................................................................................................................. 70

WARLORDISM AND THE PRIVATIZATION OF WAR ............................................................ 70

9. The theoretical framework of the warlordism debate ................................................................. 72

10. Markets of violence ....................................................................................................................... 78

10.1 Preconditions and development ................................................................................................ 80 10.1.1 The weak state ................................................................................................................... 80 10.1.2 Ideological beginnings and the gradual economization..................................................... 85

11. Types of warlordism...................................................................................................................... 92

11.1 Violence from below................................................................................................................. 92 11.2 Violence from above................................................................................................................. 92 11.3 Local warlordism ...................................................................................................................... 93 11.4 Frontier warlordism................................................................................................................... 93 11.5 Warlords of international crime ................................................................................................ 93 11.6 Client warlords.......................................................................................................................... 94 11.7 Partisans and liberation leaders................................................................................................. 94 11.8 Actors and commodities............................................................................................................ 95

12. Tribal fragmentation and warlordism: Somalia and Pashtun Afghanistan in comparison ... 98

12.1 Somalia ..................................................................................................................................... 98 12.2 The comparison with Pashtun Afghanistan............................................................................... 99

13. Conclusive remarks: Khans are warlords are khans? ............................................................. 100

References....................................................................................................................................... 105

Maps Map 1: Pashtun area in Afghanistan and Pakistan................................................................................10 Figures Figure 1: Pashtun putative genealogy……………................................................................................ 23 Figure 2: Segmentary opposition and the two-bloc system................................................................... 27 Figure 3: Auto-dynamics of the escalation in civil wars…................................................................... 86 Figure 4: Khans and warlords: Conditions and distinguishing features...............................................102 Tables Table 1: Nang and Qalang: Socio-economic categories....................................................................... 19 Table 2: Nang and Qalang: Self-perception and perception of the other............................................. 20


- 4 -

PART 1

POLITICAL ALIGNMENT, LEADERSHIP AND THE STATE IN PASHTUN SOCIETY

1 Introduction At the dawn of a new century, a human tragedy came about. A tragedy that provides sufficient impetus

to set off additional calamities in southern Central Asia. Furthermore, the disastrous event had the

power to promote an ordinary autumnal calendar date into a legendary lexical item with an epochal

meaning. In the aftermath of September 11 and the subsequent events, eloquent opinion from

commentators was mushrooming in the manner of plumes from laser-guided bombs in the wider

Hindukush region. Both at times missing their targets, notwithstanding their pretence of smartness.

The American led war against terror made the Afghan Pashtuns1 to become a focus of worldwide

public attention, whereas previously the Pashtuns had been largely unknown in the general public.

However, in the minds of some interested contemporaries these people still evoke nostalgic ideas of a

nomadic and heroic people abiding in black tents. Some commentators reapplied an alternative picture

of a warring and even cruel people, their culture allegedly harbouring an inherent predisposition to

violence. In doing so, some see in such cultural harshness the unmistakable face of Islam and attribute

to them a lower stage of civilization, if not denying them any measure of cultural sophistication

whatsoever. All this seemingly in defence of another extremism and under the trustful auspices –and

quotations– of former political leaders of the western world:

The Pathan [i.e. Pashtun] tribes are always engaged in private or public war. Every man is a warrior, a politician and a theologian. Every large house is a real feudal fortress....Every family cultivates its vendetta; every clan, its feud [...]. Nothing is ever forgotten and very few debts are left unpaid. (Winston Churchill cited in Crossroads [n.y.])

The Pashtuns are regularly referred to in terms of feuding, warring, constant retaliatory efforts,

factionalism and physical brutality apparently beyond any standards:

The warlike Pathans form one of the world's largest tribal societies [...] and defend their territory and honor against all invaders […and] have a passion for freedom and independence[…]. They are fearless guerrilla fighters […and] no one has ever managed to subdue or unite them. To safeguard his honor, or the honor of his family or clan, a Pathan will sacrifice everything, including his money and his life. He will return even the slightest insult with interest. According to a Pathan proverb, 'He is not a Pathan who does not give a blow for a pinch.' The Pathans are notorious for

1 This ethnic group living predominantly in Afghanistan and Pakistan is in the literature also denoted as Pukhtuns or Pathans, the latter mostly referring to north-western Pakistani Pashtuns. In Afghanistan they are frequently referred to as Afghans from other ethnic groups in the country since they are historically seen as the ethnic group representing the Afghan state (Rogg and Schuster 1992: p. 306) .


- 5 -

the family feuds, often the result of disputes over zar, zan or zamin – gold, women or land. (Crossroads sourcing to Shaw 1998: 394ff.)

Examples that display similar viciousness are countless. They usually report historical voices from the

period of British colonial wars during the 19th century. Indeed, Pashtuns appear to be a classic even in

the realm of scientific literature if addressing issues of conflict, feuding and warring (Orywal et al.

1996; Dupree 1984; Lindholm 1991). Who wonders then that contributions to reader’s letters pages

stand like a voice in the wilderness:

Your short history of Afghanistan’s Pashtun tribe might leave readers with the impression that its history is one of unbroken war, feuding and violence that reaches back as far as Alexander the Great […]. The public doesn’t know that an Islamic leader [i.e. Badshah Khan in the 1930s and ‘40s] took those same people to the pinnacle of their humanity, what Gandhi called the non-violence of the brave. (Reed 2001: 7)

Before the Peace Conference in German Königswinter, resulting in the Bonn agreement on December

6th in 2001 there was optimism over finding a political solution to the satisfaction of every involved

party “given the Afghan factionalism”(CDE 2002). Yet heavy clouds of scepticism overshadowed the

conference as it became apparent that there was little hope of uncovering an Afghan type of “magic

formula” in the face of political realism. Such realism has been well nourished by painful lessons of

the past. An Afghan past harking back to never-ending sequences of switching coalition groups and

so-called warlords: apparently fragile in their fabric, flexible in their orientation of alignment,

unpredictable in their dynamics and last but by no means least lethal in their consequences. During the

Mujahideen jostle for power in a fratricidal conflict between 1992 – 1994, roughly 45,000 Afghans

were killed (Matinuddin 1999: 10). In those clashes, Afghan factions with different ethnic or tribal

backgrounds were involved physically. They share, or more precisely claim nonetheless the same

religious legitimacy expressed in a strive for an Islamic state or emirate of Afghanistan. Almost a

dozen Islamic parties turned up during the Afghan Jihad against the Soviet occupation in the 1980s, all

of them claiming the exclusive representation of the true Islam. The engaged parties belonged to

different ethnic and tribal groups. Islam and its mundane implementation in policies appeared to be

instrumental. Consequently, the religiously orientated political factions were distrustful of each other

(ibid. 1999: 7). The impression of an evident propensity for a continual factionalism paired with

manifestly violent practice entices authors to label the phenomenon as an Afghan commonplace. Even

more, they associate it with a distinctive cultural milieu: “Unfortunately, the negative side of the

Afghan character comes to the fore when an injury or an insult has to be avenged, even if it be the

proverbial cutting off of the nose to spite the face” (ibid. 1999: 9). Here, the social anthropologist’s

resistance both to cultural reductionism and a certain arbitrariness in applying terms of ethnic

implication is stimulated. First, we have to ask whether in such intense expressions of factionalism and

violence the ethnic factor is of chief importance. If so, we can close the chapter and open a new one,


- 6 -

as this paper deals with contemplations on topics of political anthropology and is largely neglectful of

addressing questions concerning inter-ethnic cleavages. Second, we must examine whether a cultural

factor accounts for the assertions made above. If this is at least partly the case, similar occurrences

must consequently be found within ethnic boundaries. Third, a critical expansion of the focus should

not be ignored, since Afghanistan and its wider region is (and never ceased to be during the last

centuries) still the object of geopolitical interests of alien powers. All those influences should be taken

into account, since the incursion of foreign ideas, funds and weaponry has had an impact on the

balance of internal power assemblage and on the means supporting political disputes. Thus, the

promotion of new political actors in the arena may change the structure of political organization. This

contribution ignores to a large extent conflicts set off by inter-ethnic tensions. When the Afghan

interim government led by Hamid Karzai has been trying to mobilize the cohesive elements among

dozens of diverse factions throughout the country, the fears spread by political commentators and

analysts turn out to be accurate anticipation. During the first quarter of 2002, the flare-up of several

violent tensions between competing local Pashtun groups mainly in the Eastern part of the country, the

Pashtun showcase speaks for a constantly recurring truism: Pashtun society and its deep-seated

tendency for factionalism and flexible modes of political alignment. Hence, such observation appears

to demonstrate the assumed endemic nature of factionalism that resides in Pashtun society. This has

brought up the cultural factor of such seemingly persistent modes of collective political action within

ethnic boundaries. The focus is hence directed towards an examination of such reproductive patterns

within the cultural apparatus of Pashtun society. Relating those kinds of actions with a cultural pattern

implies its reproductive character. Therefore, it requires justifications of cultural boundaries. Is it

permitted to speak of the Pashtun society? The first part is an attempt to display important components

which give sound reasons for a cultural coherence of Pashtun society at least to a certain degree.

Cultural practice and its implicit values may reappear most likely on an upper level of collective action

and function as a starting point for the very goal of this paper: a) to examine the common basis of

traditional manifestations of political organization, political alignment and mobilization of followers;

b) to illustrate the flexible nature of alliances and networks; c) to exemplify the role of kinship for

political alliances on the basis of the case of Swat Pashtuns in Pakistan; d) to give evidence for a

distinctive set of cultural features that contribute to collective political action; e) to examine crucial

personal qualities that draw from the idealized Pashtun person in connecting them to political

leadership and f) to reveal the underlying relational, material and non-material factors that have an

impact on power exertion of political leaders. With reference to Barth’s methodological individualism,

the observations are constantly linked to an actor-centred theory that favours the individual decision

process based on free choice as an analytical focus. This paper tries to account for the cultural

variability of the large Pashtun tribal society. Finally, a comprehensive concept that associates a theory

of leader-centred politics with wider socio-economic circumstances as well as with a critical

evaluation of the theoretical framework of a Barthian transactionalism. Apart from cultural variability

and being critical towards a structural-functionalist approach, the cultural variables are embedded in


- 7 -

the wider spatial and temporal context. The external factors such as foreign interference in the context

of interest-driven geo-political strategies may have a considerable impact on domestic politics. We

might assume that selective foreign support for political groups in terms of money, logistics,

knowledge and weaponry together with the income from drug trafficking may affect heavily the

stability and functioning of domestic politics. Ultimately, under the sway of similar impacts of

modernization or in periods of lasting war and public violence, cultural elements are subject to change.

For instance, a new type of political leaders may supersede the traditional elite.

Currently, prominent political representatives are regularly referred to as warlords, while the same

articles suppose identical actors to be clan-chiefs. Uncritical and random labelling of political actors

does not help to disentangle the more refined and complex political realities in the region, nor does it

serve to modify the platitude of a martial Pashtun culture where the sword appears to replace any

corrective of social control and legal proceduralization. The second part of this paper therefore

addresses the academic debate over warlordism. Warlordism is considered as arising from a distinctive

political environment, in which a pattern of features may result in a violence-open area. The collapse

of a state or the partial dysfunctionality of a state may give rise to the lack of an effective monopoly of

violence of the state. Additionally, a violence-open area is regularly preceded by the deficiency of a

functional state administration as well as an ineffective system of law enforcement and governance.

Subsequently, the political vacuum is to be filled. Prominent political actors emerge. They secure their

power by the control and the use of force. The purchasable nature of violence is a typical characteristic

of emerging warlord systems. As a consequence, warlordism will be examined by applying theories of

the privatization of violence (cf. Elwert 1999; Eppler 2002; Bollig 2001 and others). Warlords secure

their allegiance through a combination of the control of force and a strongman position they had held

before the conflict broke out. Apart from the capability to control a privatized apparatus of violence,

the allegiance and the power of warlords largely draws either on a strong economic position as a result

of running a lucrative business or on a recent past in which they were incumbents in the military or

administrative organization of the state. Ahead of warlord systems, a variety of causes may be

responsible for the upsurge of a conflict. Besides mere economic intentions and power politics,

ideological, political, social or religious motives are often a starting point for violent conflicts.

However, in the course of the events political and ideological motives more and more serve as a an

instrument for mobilizing allegiance as well as for a justification for the existence of the violent

organization. But the organization may reach a stage of self-independization, where private violence is

primarily a means for securing the private business. A profitable business which is based on lucrative

commodities or illicit trade as well as on unlawful activities, such as kidnapping, traffic in human

beings, arms trade or extortion racket. Despite a certain degree of convergence, warlord-type conflicts

are far from being uniform. Rather, they differ in terms of causes, types of actors and structure of

political allegiance. For instance, both in Somalia and in the Pashtun-dominated parts in Afghanistan,

armed militias, non-state systems of law enforcement and social control have been based on tribal

regulations for a long time. Furthermore, the nation-state has rarely had the inclusive administrative


- 8 -

form, which is applicable to standards of the western states. By raising similar objections, would it still

be appropriate to speak of warlords when addressing local power holders in both countries? If we can

nonetheless observe some degree of “warlordization”, what characteristics give substantiation for the

political circumstances to fit into the category of warlordism? The subsuequent digression on

warlordism serves as to tackle related questions. After all, it becomes apparent that warlord-type

conflicts are much more diverse. This is also a consequence of unlike historical, political and cultural

backgrounds of the conflict. Moreover, underlying biased assumptions relating to the administrative

arrangement of a state do infrequently not meet a political and cultural reality, all the more, if we deal

with functional tribal structures. Political circumstances and change should not be disregarded, either.

Likewise, in the course of floods of war refugees, a substantial drain of population may possibly have

an effect on local politics. Most observations described have been derived either from former (Barth’s

studies) or current academic material based on political anthropology and political science. The largest

amount of contribution provide articles that have been drawn from ethnographic research in the

Pashtun tribal areas in north-western frontier region in Pakistan. Both the classical monographs by

Frederik Barth on political leadership among Swat Pashtuns as well as Akbar Ahmed’s articles on the

neighbouring tribal districts provide the basis. They are completed with alternative contributions by

other anthropological authorities in the field, among them Pierre Centlivres and Micheline Centlivres,

Louis Dupree, David Edwards, Bernt Glatzer, Charles Lindholm, Nazif Shahrani and Nancy Tapper. It

must be conceded that the material about Southern and Eastern Pashtuns in Afghanistan is particularly

scarce. With regard to anthropological literature, many smaller marginalized communities are

represented in quite a considerable number compared to the large number of Pashtun agriculturalists

e.g. in the south. As a consequence, it is a disputable question whether observations and analogous

interpretations are transferable and thus applicable to a larger society which is confronted with

alternative socio-economic realities, other political characteristics, a variable degree and mode of state

integration as well as with unlike historical experience. For those reasons, many questions that are

tackled here turn out to be an uncertain venture. Nevertheless, current actualities and a fascinating

world one immediately discovers after entering the Pashtun cultural universe, are worth enduring a

certain degree of tension resulting from apparent uncertainty and contradiction.


- 9 -

2 A macro-perspective: The wider environmental and historical setting

Before embarking on explanations that attempt to transcend traditional behaviours within a tribal

framework to political action on a state level, justifications that tackle questions of cultural boundaries

and identity are imperative. With 20 millions of members1 or probably more, the Pashtuns are a

relatively large ethnic group (Rogg und Schuster 1992: 306). Furthermore, they live within different

national boundaries neighbouring on a variety of other ethnic groups and thus cultural influences.

Pashtun population is widely extended over an ecologically diverse area, it displays a great range of

economic and social forms and is penetrated by, and hence encapsulated in, centralized administration

to various degrees. Although assuming Pashtuns to constitute “a large, highly self-aware ethnic group

[…] organized in a segmentary, replicating system without centralized institutions” (Barth 1981: 103).

Cultural patterns require some types of mechanism in order to generate and maintain social

reproduction buttressing an ethnic identity across residential and tribal groups. Can we take a similar

replicating system for granted, given the fact of limited communication between distant communities

so as to maintain and reinforce a shared body of knowledge and cultural practice that is powerful

enough to activate inclusive ethnic identity, if required? More precisely, can we then tacitly assume a

cultural coherence and homogeneity that reproduce a similar web of meanings and consequently

comparable patterns of political behaviour across Pashtun tribes? Questions like these elicit

challenging issues of identity and call for further preparatory grounding. Consequently, the lines to

follow serve as a condensed presentation of cultural traits in Pashtun society that are deemed being

widespread, relevant and persistent. They address cultural attributes which are a matter of recurrent

record in classical scientific literature of major reputation among scholars. The relevant features aim at

laying emphasis on those components which are considered to have a significant effect on issues of

political organization, leadership and public conflicts. Before entering the cultural universe, I shall

briefly give some indispensable background facts such as the geographical spreading of Pashtun tribal

groups as well as a cursory synopsis on Pashtun collective identity and history. The latter subjects,

although unsatisfying in its offered depth, deserve an entry due to the suggestion that a common

history of conquest and the far-reaching societal and political implications of alike events may as a

minimum form a common identity if challenged by non-Pashtun forces. However, the degree of

fragmentation of Pashtun society, a long history of conquest and subsequently the membership to an

ethnic group that ranks prominently above subjugated non-tribal minorities has a deep impact on the

way Pashtun tribal members see themselves and their position within blurred boundaries of society. In

this sense, history and the function of Pashtun dominance may serve as a counterbalance against inter-

tribal crumbling and disintegration. On the other hand, it nourishes ideas and convictions of

1 There is a current dispute about the number of Pashtuns spread predominantly in Afghanistan, Pakistan and India. Some reckon the number of tribal Pashtuns about 100 millions, whereas the number living in Afghanistan is alleged to exceed 20 millions. The political function of such popular manipulations of census should not be disregarded, either.


- 10 -

characteristic values and virtues, that centre around tribal membership, tribal law and institutions as

around a set of distinctive personal qualities that imply ideas of superiority and political pre-eminence.

Since those features are deeply rooted in Pashtun culture and hence largely seen as exclusively

Pashtun qualities by its members, they are likely to reinforce the consciousness of Pashtun ethnic

identity, despite a visible tendency for tribal fragmentation. Furthermore, today’s challenges to

Pashtun political dominance actualize the significance of similar questions. To sum up, the linkage of

research findings derived from a specific context to a generalized reference of Pashtun culture or

Pashtun society is a risky task and thus prone to over-simplification. Therefore, it is imperative to keep

in mind the high complexity and various contexts in which social groups who claim proudly to be

Pashtuns live and thrive. In spite of alike implications, there is evidence of particular cultural

universals within the complex Pashtun universe that are evidently able to cross international, tribal and

regional restrictions. In suggesting related “Pashtun universals”, I follow to a large extent David B.

Edwards, who draws similarities between Pashtun groups despite the international border from similar

linguistic forms and “[they] share the same genealogical and mythic charter, the same forms of

economic livelihood, religious beliefs, and ethos and understanding of life and death” (1998: 714). By

transcending a relevant set of evidence found within a restricted context, we may assume such cultural

traits to be applicable on a more abstract, higher and thus comparable level of analysis, too.


- 11 -

2.1 The geographical setting

Pashtun Area of Western Pakistan (settled/qalang & tribal/nang districts; Pakistan)

Swat Region in Pashtun settled/qalang-districts; Western Pakistan

MAP 1 PASHTUN AREA IN AFGHANISTAN AND PAKISTAN: Sketch line encloses the approximate extension of the territory predominantly inhabited by Pashtun population in Afghanistan and Pakistan. Sources: Centlivres and Centlivres-Demont (1988: 58) and Ahmed (1980 : 10).

2.2 Ethnonyms, identity and common history

To start with, Afghanistan is the epitome of the plasticity of ethnic terms (Centlivres and Centlivres-

Demont 1988: 33-38). On the one hand, this statement gives expression to the remarkable ethnic

diversity of the country. One the other hand, it demonstrates subtle social stratifications along ethnic

lines as a result of historical processes of territorial conquest, constellations of dominance and hence

ethnic marginalisation of other ethnic groups. In this context, the dialectical character of ethnic

identity becomes apparent in the background of a weak national identity, where “l’image de l’autre est

inséparable de l’image de soi, et dans le cas [afghan] qui nous occupe, la constitution de l’identité d’un

groupe ethnique donné se fait par référence aux autres groupes ethniques” (Centlivres and Centlivres-

Demont 1988: 34). For international public afghân refers to all citizens living in Afghanistan (or once

used to live there bearing in mind the enormous number of Afghan refugees). However, within

Afghan borders where the impact of ethnic variety is evident and the abstract idea of nationality is

largely unfamiliar, it is not quite so clear. Inside Afghanistan, being afghân means to belong to an

Afghan tribe, i.e. to a Pashtun tribe. Non-Pashtun inhabitants declare either to be tâjik, uzbek, balûc or

hâzâra. Local identity is even more pronounced among non-Pashtun groups. Thus, they are first and


- 12 -

foremost herâti or kâbuli. Of course, according to the level of reference and disassociation compared

to an opposite group, the family, the region, the language or religion is more compelling for identity.

Orywal (1986: 80 – 85) offers a systematic discussion based on a concentric model of identification.

What appears to be a linguist’s clowning around for semantic amusement has far-reaching implication.

The Pashtuns are seen as the ethnic group representing the state. As such, it is the dominant ethnic

group whose members traditionally hold a large amount of functions of national or local authorities.

This is to a large extent a result of the historical process of Pashtunization in the 19th century that led

to the constant spread of Pashtun groups towards the north and west of Afghanistan. Those

circumstances explain why non-Pashtun Afghan inhabitants claim to be afghân within national borders

only if they hold a position as a public servant. Other reasons for adopting a virtual Pashtun-like

identity are a marked nationalistic attitude of politicians and an instrument for ethnic camouflage by

marginalized groups. The latter case again discloses the value-loaded instrument of ethnicity and

hence the supremacy of Pashtun identity in Afghan socio-political life (Centlivres and Centlivres-

Demont 1988: 35). The inter-ethnic cleavage is a function of unequal allocation of political power.

This again mirrors the historical process of conquest and the spread of Pashtun society during the last

centuries, which has enabled the construction of a mainly Pashtun directed state in Afghanistan

(Rasuly 1993: 30-35). Another legacy favouring the Pashtun supremacy in the region as well as within

state-run administrative authorities was their integration into the power structure of the Persian empire

as soldiers in the standing army. political as well as military functions. As a consequence, some

Pashtun tribes in the South and the West might have become familiar with political structure that had

been holding sway over a large territory. But the prior incorporation into the Persian empire and the

extent of its impact on Pashtun social organization and lineage-structure remains uncertain (Rasuly

1993: 34). However, according to Rubin the Persian Safavid Shas have imposed particular elements of

the Turco-Mongolian tribal structure onto the Pashtuns living in their territories. Thus, the major

division of the Pashtun tribes in two large and occasionally powerful federations (Ghilzai and Durrani)

is a result of Persion colonial efforts. The Persian Safavids created (or rather recognized) the ruling

lineages among each tribal confederation by appointing a ruling chief (khan) in their duty (Rubin

2002: 29 referring to Elphinstone 1972 (1815) and Caroe 1958). In early 18th century, Persian rulers

were driven out. Under Ghilzai rule, a Pashtun tribal association, a large part of Persian territory

including Isfahan was subjugated and as a consequence the Safavid reign collapsed (Rasuly 1993:

36). However, the Ghilzai rule survived only a short time. If not considered as a Pashtun peculiarity,

this episode (and comparable ones) might be exemplary for an inherent dilemma of tribal societies: A

regular pattern including diametrically opposed and incompatible conflicting tendencies. On the one

hand, kin-based mergers can be seen as a potential to form political corporate groups along ramified

lines of kinship. On the other hand, the quick dissolution of the tribal merger while it splits into ever

smaller factions of its segmentary constituents. It reveals an apparent and recurrent ineffectiveness of

political fusion based on tribal kinship pattern from which Ibn Khaldun derived his cyclical theory of


- 13 -

the state (14th AD). That is, members of tribal societies regularly succeed in seizing the central power,

but fall short in persevering with the control owing to a lack of tribal cohesion (Glatzer1995: 51).

In the year 1747, an inter-tribal jirga in Kandahar proclaimed Ahmad Khan the first Pashtun Shah

(King) ruling over today’s southern and western Afghanistan. Ahmed Shah was a tribal member of the

Abdali tribal confederation (later called Durrani) which dominated the south-western part with

Kandahar as the political centre. A jirga is a traditional assembly with consultation character

composed of various local elders, political or religious representatives or any involved person of

respect and functions as a council meeting Tapper, Nancy 1991: 32; cf. chapter 4.5). The procedure of

the proclamation itself is deeply rooted in a widely respected tribal institution. At that time, it

demonstrates the state to be not much more than a structural transformation of a traditional tribal

system. Furthermore, the process of formation of an inter-tribal power-sharing agreement was based

on traditional political institutions. Therefore, the compliance with traditional tribal values was crucial

for reaching a political consensus and in order to avoid animosities. A comprehensive inter-tribal

legitimacy was required (a) to respect the liberal spirit (i.e. the “ethos of egalitarianism”; cf. chapter

5.3) of the Pashtuns. Accordingly, an overly powerful tribal leader was out of consideration; (b) a

Pashtun leader was dependent on the legitimation for his decisions and actions on the basis of the

jirga; (c) genealogical descent and charisma constituted two important features that enhance political

legitimacy of traditional Pashtun leadership (Rasuly 1993: 36). However, such mergers of political

factions should not obscure the fact of tribal fragmentations. Conversely, the claims of tribal interests

and the concerns of the central state were in a state of permanent conflict and often diametrically

opposed. Rulers could only maintain an enduring counterbalance if taking into account expectations of

tribal loyalty. The traditional Afghan state in those days and throughout the following centuries was

characterized by a precarious coexistence of state-like structures and tribal autonomy. In addition, the

ruling dynasty itself was shaken by internal conflicts due to succession. Moreover, the ruling party

tried to dominate the tribes and to incorporate them at least partially into its political system. In

contrast, the governing dynasty was dependent on the faithfulness of their followers. They secured

tribal loyalty by means of traditional instruments: marriage exchanges and clientelistic relationships of

various kinds were powerful in creating reliable social bonds that bore political functionality (Rasuly

1997: 22ff.). More than a century later, Abdurrahman Khan (1880-1901), a Durrani Pashtun with his

absolutist politics was responsible for the extension of the national administration and hence for the

curtailment of above-mentioned traditional instruments as well as for the influence of tribal

representatives hitherto indispensable as a tribal participation for a national rule. As a consequence,

the significance of jirga-based decisions on the state-level was eliminated to a large extent. Civil and

military administration expanded drastically (ibid.). Likewise, Abdurrahman’s nationalistic and

centralist politics restrained the autonomy and privileges of Islamic clergy. Those reforms weakened

the authority of the segmentary tribal system and consolidated central power and thus led to a

widening gap between state-authorities and the rest of a rural, decentralized society (ibid: 25). In this

era, the centralist politics attempted to break up the political structure of non-Pashtun groups, although


- 14 -

the political elite of Pashtun tribes successfully kept the most part of their power owing to a policy of

non-intervention in terms of local, tribal politics (ibid: 105). It was also during this period, when the

process of Pashtunization came to the fore. After the Pashtun conquest of the Afghan northern part,

many Pashtun nâqelin (resettled) groups were settled in the midst of non-Pashtun ethnic groups. The

Pashtunization served a couple of purposes: unruly Pashtun groups from the southeast were deported

with the intention of preventing further detrimental effects, such as dissidence, on political stability.

On the other hand, it extended the sphere of Pashtun influence. The latter had also the effect to abate

alternative tribal units since the dispersion of tribal segments averted the concentration of opposing

and power-competing forces, but it also supported the strive for a Pashtunization in a nationalistic

sense (Centlivres and Centlivres-Demont 1988: 232-233). The Afghan national version of

Pashtunization includes the will to equate the complex social and political makeup of the country with

a Pashtun administrative system. This comprises the endorsement of Pashtun life-style and values, the

political and ideological integration of fragmented Pashtun groups as well as the active promotion of

Pashtu as lingua franca. In a broader sense, Pashtunization attempts used to embrace the incorporation

of Pashtun tribal land in its totality, i.e. the integration of tribal areas set apart by the international

frontiers between Afghanistan and Pakistan into an own state: Pashtunistan. When Daoud Khan, head

of the Afghan government 1953-1963, had boosted again excessive Pashtunistan ambitions, a deep

crisis in foreign and economic policy was the consequence. Still today, the Pashtunistan issue appears

to be latent. Afghan Pashtun tribal leaders still cite Pashtunistan as an unresolved problem. Small

Pashtun parties on the Pakistan side of the border, such as the Pashtun National People's Party, call for

the creation of a Pashtun homeland. Besides, Pakistan is suspected of supporting Islamic extremism in

order to keep the Pashtunistan issue suppressed. In doing so, another collective identity than an ethnic

one is attracted. The support of groups aimed at strengthening Muslim community means to de-

emphasize state borders.

To sum up, the previous outline addressed questions of ethnic identity expressed in ethnic terms,

which bear implications of inter-ethnic relations. Not only in today’s Afghanistan, but also in the tribal

areas of north-western Pakistan, inter-ethnic relations reflect relations of mainly Pashtun domination

and subordination of non-Pashtun groups in the Pashtun heartland. This vertical cleavage is a result of

historical processes that included Pashtun favouritism within the Persian empire in the region and the

subsequent Pashtun politics of conquest and subjugation of non-Pashtun groups as well as a selective

settlement practice. In addition, the penetration of society by Pashtun-dominated public administration

as well as the consolidation of a tribal dynasty mainly by members of the Pashtun Durrani ensured

their tribal pre-eminence. As a reinforcing effect for Pashtun control, this process had far-reaching

practical consequences: social mobility of non-Pashtuns was complicated or precluded, since most

relevant administrative positions both on state and local level were held by Pashtun tribal members.

Select historical events serve as to display two counterbalancing dynamics in Pashtun social and

political fabric: (a) a political decision-making on an inter-tribal level based on traditional tribal

institutions. Its effectiveness as an integrative force enables a persistent continuation for Pashtun


- 15 -

domination, more often than not under the dynastic rule of Durrani tribes; (b) conversely, the

autonomy among the various tribal segments of Pashtun society is a paramount value. At all times the

leaders on a state or inter-tribal level had to consider carefully the need for tribal sovereignty and

political participation to ensure a stable allegiance. Such a quandary is a manifestation of traditional

tribal values with key significance in a tribal society. Centlivres and Centlivres-Demont emphasize the

importance of traditional values and institutions now and then:

Les homes de l’Afghanistan professent un ideal d’égalité et d’autonomie, souvent affirmé dans les institutions tribales: les assemblées qui réunissent les homes adultes, et dont la forme circulaire symbolise les droits égaux des participants, en sont un exemple; l’exigence d’autonomie, l’insistance sur la fierté et l’honneur de l’homme libre, sur les rapports entre pairs en sont un autre. Pourtant la société afghane apparaît à l’observateur comme fondamentalement inégale. […] Cette égalité suppose la liberté de prendre part aux conseils tribaux et locaux, ou jirga. […] C’est une jirga des tribus durrani qui, au XVIIIe siècle, a mis Ahmad Shah sur le trône et a été le point de départ de la dynastie qui est à l’origine de l’Etat afghan. Les Pachtouns ou Afghans étant dominants en Afghanistan, les collectivités pachtounes, du moins quand elles sont loin des grands centres, bénéficient d’une certaine autonomie interne ; les représentants de l’Etat ne se distinguent pas toujours des notables tribaux, même si la bureaucratie et les fonctionnaires tendent à s’imposer toujours plus comme instances d’arbitrage. (1988: 22-23)

An analysis of types and dynamics of political alliances and decisions of political actors must

consequently be sourced to the complex fabric of cultural meanings and practices of Pashtun society.

Even more, besides a shared set of cultural values and institutions a common history can also

contribute to latent cultural coherence, since it creates a collective identity by disassociating the

ethnically different. This coherence may be functional and manifest in collective actions in periods of

crisis or if common political achievements are threatened. In a practical sense, a positive connotation

of a particular ethnic membership helps to achieve individual advantages and consequently adds to a

persistence of cultural traits as a distinguished set of ethnic characteristics. Based on the assumption

that being a Pashtun and hence a member of a widely dictating collective, the adoption of attributes

attached to tribal affiliation is significant. One postulation in order to prove tribal –if only putative–

membership to benefit from collective primacy should consequently require an outstanding and visible

set of cultural characteristics that are publicly exposed in practice and discourse. Besides, structural

features such as patrilineal descent are overt cultural attributes that are put on display and referred to

explicitly in a concept of doing Pashto, if applied. As a basis, Pashtun emic self-ascription of ethnic

identity covers some crucial traits necessarily associated with being a Pashtun:

(a) Pashtun language,

(b) Islam,

(c) Patrilineal descent,

(d) Pashtunwali, the Pashtun tribal law based on

(e) Pashtun custom and values, a widely coherent and accepted set of cultural attributes (Barth 1981:

105).


- 16 -

As with most segmentary Islamic societies, religious belonging goes hand in hand with genealogies.

Descent, religious leaders and a distinctive body of customs are important components that contribute

to the dynamics of political affiliation. Accordingly, such inter-tribal similarities will be discussed in

full explanation after having drawn concisely a more discriminating picture including socio-economic

differences. Such variation may be functional for an analysis of political matters.

2.3 Tribal society and the state

In countries such as Afghanistan or western Pakistan with a large proportion of people with tribal

membership, the relationship between tribal society and the state is critical. In Afghanistan as well as

in Pakistan, the penetration of state administration such as law enforcement institutions is restricted in

some areas. Again, the issue of Pashtunistan represents a subject matter of some political delicacy for

governmental decision-makers. It has dominated the political agenda of political actors for more than a

decade around the seventies of the last century. The vision of uniting all Pashtun tribes under one

administrative umbrella exhibited the complexity of counteracting dynamics in Pashtun political

affairs. This displays a strong effect of group inclusion owing to Pashtun identity if faced with an

external threat. Ethnic cohesion among Pashtuns may increase if a curtailment of tribal autonomy is

feared. The fear of loss of autonomy turns out to be a Pashtun constant. Analogous tendencies were

mobilised against foreign intruders and world powers such as the British empire a century ago.

Similarly, it rallied support in order to oust the Soviet invaders. Afghan pro-soviet government under

Daoud Khan, who was a Cousin of former King Mohamed Zahir Shah and also responsible for

abolishing Pashtun monarchy led by Durrani-Muhamadzai, revived Afghanistan’s campaign for

Pashtunistan in 1976. Such ethnic nationalism prompted the concern of Pakistan that was in turn eager

to support any source of counterpressure against the revival of the Pashtunistan issue. Some Pashtun

parties on the Pakistan side called for the creation of a Pashtun homeland (Rubin 2002: 100). The

following sequence of reactions again revealed the flip side of the double-edged coin, on which

Pashtun ethnic politics is engraved. With the objective to keep the Pashtunistan issue suppressed,

Pakistan supported Islamist protagonists and welcomed Islamist exiles and consistently favoured the

Pashtuns among them. “Pakistan covertly organized and supported the attempted uprising by the

Islamists in 1975, which, together with pressure from the shah, had the desired effect of bringing

Daoud to the bargaining table over his support for Pashtun and Baluch separatism in Pakistan and

other regional issues”(ibid.). Accordingly, some prominent Pashtun politicians like Gulbuddin

Hikmatyar and others probably de-emphasized state borders with the purpose of uniting the Muslim

umma. This enabled them not only to disband the overriding significance of ethnic or tribal

membership, but also to expand their political influence beyond such collective restrictions. Islam

seems to increase plausibility of political endeavours and most likely strengthens the legitimacy of

leaders. Political actors regularly draw on religion that serves as an alternative instrument to

strengthen collective identity of the rallied supporters. In doing so, they apply a successful strategy by


- 17 -

circumventing inter-tribal loyalties and thus precluding an indispensable political and symbolic

resource that provides a component for fostering ethnic nationalism. For Nazif Shahrani, an Afghan

fieldwork-experienced social anthropologist, a similar perspective provides a partial explanation for

the rise of the Taliban during the 90ies of the last century. In his view, the religiously motivated

martial movement was systematically taking advantage of the manipulation of the Muslim identity of

believers for political objectives, “to re-define, and even to determine not only the basis for political

legitimacy and exercise of authority, but also to proclaim what ought to constitute Muslim religious

orthodoxy/orthopraxy” (Shahrani 2000). Hence, both the Pashtunistan issue as well as far-reaching

forms of religious extremism such as the Taliban phenomenon offer exemplary occurrences. They

illustrate convincingly the complex situation and multidirectional dynamics, engendered in a context

of societal fractures at the crossroads, where tribal, state, religious impacts and external strategic

interests collide. Comparable events that have been addressed previously suggest also alternative

explanatory models. The cohesive power of supra-ethnic identity, such as religion, displays its partial

political success, but it also evokes alternative compelling questions. Not to tackle them would be

neglectful of an obvious constant throughout Pashtun history of politics: The often quoted

fragmentation of Pashtun political constituents. Whether the imperfect success of religious politics and

the failure of Pashtun national politics are due to a balanced product of either their conceptual

weakness or strength, respectively, is an open question. But the contribution of disjointing forces that

result in an impression of deep-seated propensity for political and societal fragmentation may even

account more for political events than other factors. My suggestion will be to demonstrate that

political fragmentation is at least to a certain degree endemic in Pashtun political organization. It is

structurally sponsored and sustained by structural conditions, cultural tradition and a distinctive set of

features. The persistency of political fragmentation will subsequently also be associated with active

elements of Pashtun culture. Accordingly, traditional tribal institutions, customs and values need to be

be found to back up an appropriate social anthropological model that reveals systematic processes of

political action. Accounts for political fragmentation, but also for cohesive countertendencies, will be

given.

2.4 Socio-economic variability

Unlike one might imply, owing to the previous explanations of cultural and ethnic consistency, the

Pashtun society is far from being homogeneous. On the contrary, it comprises a wide variation of

socio-economic life styles. But does such a diversity of lifestyle impair significantly Pashtuns’ “self-

image as a characteristic and distinctive ethnic unit with unambiguous social and distributional

boundaries”, as Barth assumes (1981: 105)? In areas of East Afghanistan and in the Pakistan north-

western frontier region which are included in the particular examples of Swat and the neighbouring


- 18 -

tribal territories, a great range of socio-economic variety can be found depending on forms of

adaptation.

(1) Barren hills: mixed agriculture Villages of mixed agriculturalists can be found in ecologically less favoured conditions like barren

hills. Their social organization is based on egalitarian patrilineal descent segments with an acephalous

political form (Barth 1981: 104).

(2) Ecologically privileged regions: intensive agriculture In the broader valley regions and in favoured localities in the mountains, more intensive agriculture

based on artificial irrigation, dominates as economic practice. Pashtuns in these area are landowners or

owner-cultivators. In the south and the west of the focused area, a large part of the village population

consists of tenant Tajiks and other ethnic groups. The political organization is centred around the

segmentary order of Pashtun descent groups. Some are incorporated in bureaucratic administration.

Quasi-feudal systems as well as acephalous systems with less perceptible political hierarchy can be

found (ibid 1981: 104-105).

(3) Other sectors: administrators and craftsmen Mainly in the towns of Afghanistan and Pakistan, administrators, traders, craftsmen or labourers

dominate (ibid: 105). Provided that tribal membership is associated with estate property, most

Pashtuns are landowners within agriculturally used areas. However, as a result of land selling,

migration or other occasions, many Pashtuns have followed alternative options of livelihood.

(4) Pastoral nomadic life Pastoral nomadic life together with great autonomy and tribal political organization is important in

more secluded and harsher areas in the south of Pashtun mainland (ibid: 105). But also in the central

and northern parts of Afghanistan, some Pashtun groups live as nomads or subsist on a variant of

pastoral lifestyle like transhumance.

The economic base has great significance in the Pashtun-dominated society of the Swat region where

the land is owned by a few families. Its relevance on social life also account for political processes,

since it concerns the system of land allotment and the ownership of land. Though estate property

explains to a certain extent the material capital to be a means for building up political following, the

economic aspect is but one side of the explanatory framework for political practice. Nevertheless, the

significance of land, its property and allocation as well as the respective cultural connotation must be

included in a wider analysis of political procedures. Ahmed insists that the economic structure and

ecological constraints have great analytical meaning in order to understand political organization and

power (1976: 71). But being critical towards any kind of economic determinism and reductionist

explanatory models, other features than mere economic forms of production supposedly have an

impact on the cultural arrangement including the political structure. Thus, the cultural diversity


- 19 -

between different communities appears to cover a scope of magnitude rather than to represent different

unrelated cultural units (Barth 1981: 105). Even more, certain cultural traits from the realms of social

organization, customs, beliefs and values are selected to constitute not only a set of criteria for an

ascription to ethnic identity. They also contribute to the inventory that accounts for political legitimacy

and hence add to the regulatory frames of the formation of political alignment. For that reason, central

attributes as necessarily attributed to Pashtun culture and identity will be addressed hereafter. Apart

from economic and ecological factors, various authors assume non-material features such as

personalized qualities related to a concept of the idealized Pashtun character to be persistent even if

their functionality in a changing economic and social environment may justly be questioned. Besides,

the following cultural standards serve partly as a distinctive device to assess whether Pashto is done

generally. Cultural standards are based on ideal-type models. Even if not compatible with observable

realities at all events, they are a collective instrument to measure, for instance, the congruence of

political actions of political leaders with cultural ideals (Barth 1959: 82). When speaking of ideal-

types, the notion refers to basic concepts of the pertinent discussion. This term very often follows

Ahmed’s most simple definition as ideal-types to represent “structures based on frequency of

observable behaviour and a concept of a species type and not based on quantitative analysis of

statistical averages. Ideal-types help to explain the ‘ideal’ while examining the ‘actual’ and to

contemplate the ‘pure forms’ and not the ‘average’ “ (Ahmed 1976: 73f.). The importance of tribal

ideals and values as a resource for a type of non-material capital and their interconnections with the

economic system, economic contracts and house tenancy interdependencies provide altogether

explicatory components for political action as well as for political legitimacy. Related interconnections

of this type will be illustrated later.

2.5 ‘Nang’ and ‘qalang’: two distinctive socio-economic categories

As with other tribal groups and as insinuated earlier, geographical, political and social environment of

the varied Pashtun groups may differ significantly. For example, a distinction regarding the degree of

state inclusion is to be made for the purpose at hand. The degree of central government control affects

traditional repositories of political institutions and their functioning. Additionally, the socio-economic

foundation determines to a certain extent the importance of customary repertoires such as life-guiding

standards. First, the relational pattern of political leaders and followers may differ according to the

socio-economic environment and the degree of integration into the administrative system along with

the law enforcement system of the state. For instance, hierarchies may be more or less pronounced.

Second, core values or general distinctive personal qualities attached to political leadership may be

subject to change if general environmental circumstances alter. Therefore, the analysis of the diverse

forms of Pashtun society should not leave out the socio-economic milieu in which it is embedded nor

ignore the potential for change. A distinction between mountain and plain tribes is indispensable

seeing that it is important to the Pashtun self-conception (Rubin 2002: 28). To simplify, plain tribes

(qalang) are integrated into a system of administrative and political control of the state. As a


- 20 -

consequence, they pay taxes. Furthermore, the state’s monopoly of violence and a law enforcement

system is effective. Powerful Pashtun landlords dominate and manipulate patronage networks.

Conversely, hill tribes (nang) are relatively free of any non-tribal control from outside. Even within

tribal borders they are free of domination by others, leaving aside justifiable objections against this

assessment from a qualified female point of view:

“[Hill tribes] embrace, much more fully than can the qalang Pashtun, the ideal type of egalitarian

tribalism expressed in the Pashtun Code, Pashtunwali. In the tribal model, conformity to Pashtunwali

defines what it means to be ‘really’ Pashtun” (ibid.). A Pashtun proverb gets the two models

symbolically to the heart: “Honour, nang, ate up the mountains and taxes, qalang, ate up the plains”

(Ahmed 1975: 47). Honour refers to the great importance of revenge related to honour and shame as

well as to the prevalence of blood feuds among hill tribes. Subsequently, in order to give an example

of Pashtun variation in north-western Pakistan, two socio-economic organizations will be opposed.

The categories of nang and qalang will serve as differing categories. Specific attributes serve as

characterizations. Applying the categories elaborated by Ahmed, they are opposed so as to make clear

the differences in a very brief and simplified table (Ahmed 1976: 82f.). The referential terms nang (i.e.

honour) and qalang (i.e. rents and taxes) are symbols for the socio-economic categories and stand for

an emphasis on particular attributes or rather key concepts that stand for the whole set of defining

features of each respective system. Ahmed derives the category from sections of a related tribe, but

living under dissimilar conditions: The Tribal Area Mohmands (TAM approximating nang category)

and the Settled Area Mohmands (SAM approximating qalang category). On accounts of clarity, I will

use the terms qalang and nang, respectively.

The nang category: The “hills” or Tribal Areas of the six Tribal Agencies in the North-West Frontier Province. The regular criminal and judicial laws do not apply. Mainly along the international border.

The qalang category: The “plains” or Settled Disctricts administratively subdivided into three divisions of the Province.

-weak encapsulation: less administrative control of the state (“tribal area”) -outside larger (juxtaposed to) state system; illiterate; oral tradition

-encapsulated within larger systems; district boards; literate; written tradition -more administrative control of the state

-hill areas; largely unirrigated; scarce population dispersed in fort-like hamlets and nucleated settlements

-plain areas; largely irrigated; dense population in large villages and tendency to urbanization

-pastoral tribal economy -agricultural feudal economy -no rents and taxes -rents and taxes - egalitarian social organization; mainly ‘achieved’ status of elders (mashars)

-mainly ‘ascribed status of khans; hierarchical social organizational based on authochthonic population

-warriors participating in raids -warlords organizing battles -acephalous tribal society organized in segmentary descent groups

-autocephalous village organization under khan within larger acephalous society

emphasis on ‘nangwali’ (Code of honour) -emphasis on ‘tarboorwali’ (agnativ rivalry) -jirga represents interests of entire tribe (the vast majority of the population)

-jirga members from landowning Pashtun tribes only

TABLE 1: Nang and Qalang: Socio-economic categories. Source: Ahmed (1976: 82).


- 21 -

It is important to point to the artificiality of the chart. The binary character should not obscure the fact

that categories may prove to be arbitrary and contradict empirical knowledge in particular instances.

Cultural boundaries rarely enclose pure closed systems. The categories at times overlap and borrow

from each other. They are not residential groupings either, since in reality the same tribal group may

be found on both “sides”, occasionally. Finally, both socio-cultural systems are dynamic owing to

effects of migration, exchange of goods and ideas or economic transformation. After all, cultural traits

are subject to change. However, ecological, historical and socio-economic prerequisites account for

the sufficient functionality of the categories since the underlying factors affect cultural evolution. In

consequence, they shape the political structure to some extent. Therefore, one should constantly bear

in mind socio-economic differences, even if the applied categories may appear simplistic. In terms of

what constitutes and ideal-type Pashtun behaviour and person, the gap between the evaluation of the

self-image and the perception of the outsider adds to the confusion. With important features of Pashtun

practice and Pashtun personality such as revenge, cousin enmity, hospitality, bravery and politics, the

respective evaluations differ considerably. While the nang tribes see themselves as practicing revenge

(badal, tor; cf. chapter 4.2) uncompromisingly, they maintain that qalang tribes are lax in taking

revenge. They reproach qalang tribes even to compromise cases of tor, where the impaired chastity of

women is involved. This is considered as the most serious threat. The qalang on the other hand are

aware of their deviance from the ideal, but still talk in badal (revenge) idiom. But they claim to take

revenge in cases where tor is concerned. Qalang regard the rigid practice of nang as wild and unruly

while killing indiscriminately (Ahmed 1980: 329). With cousin enmity, hospitality and bravery, the

evaluations are comparable:

self-image and perception of the other…

nang about nang nang about qalang qalang about qalang qalang about nang

cousin enmity• involved in agnatic rivalry

too weak to practice it

a milder form, other expressions than killing

nang not motivated by nang, but by jealousy and boorish nature

hospitality• very hospitable too poor to be hospitable

hospitable, civilized, cultured

calculating in their hospitality

bravery• very brave submissive, cowardly

ascendants the bravest, but aware of limitations

admit bravery of nang, but seen as ill-directed, wild

TABLE 2: Nang and Qalang: Self-perception and perception of the other. Source: Ahmed (1980: 329).

Moreover, the categories are referred to from relevant authors who deal with Pashtun culture (e.g.

Ahmed, Barth). Hence, they will also serve as experienced analytic categories in this paper. In terms

of the integration into an administrative system controlled by a central government, some areas have

undergone important transformations which have a bearing for the political organizations. For

instance, in Swat in the 1920s, a central government was superimposed over a structure which


- 22 -

consisted of numerous and ephemeral small-scale leaders in every village. But in the last decades, the

government has weakened significantly. This allowed the old segmentary system to reassert itself

(Lindholm 1991).

3 Segmentary tribal organization and politics

Among Durrani Pashtuns, a saying goes: “Descent (nasab) makes a tribe (tayfa)” (Tapper1991: 45).

Under the effect of literature, one is tempted to add: a tribal belonging makes a Pashtun. Pashtuns are

often referred to as the tribal people and are likewise said to be the largest tribal grouping in the world

(recurrent stereotype and probably first formulated in Spain 1963: 17). This points to the pre-eminence

of tribal affiliation, since it has not only an enclosing effect against other ethnic groups, but it also

links Pashtuns together on the basis of a common apical ancestor. But given that the term tribe is very

often used popularly with an unsystematic elasticity, if not to say in an appalling randomness, the

meaning applied here needs a succinct clarification. In a sharp contrast to the popular use of tribe as a

fuzzy term which is used to referring to any small ethnic group and at times does not even refrain from

derogatory connotations, the term applied here is descriptive for a particular kind of social

organization. Tribes, as understood here, consist of an organizational principle based on segmentary

lineages. Essentially, they are characterized by (a) ramifying segmentation of tribal groups based on a

genealogical concept and therefore include (b) unilineal descent from a common eponymous ancestor

(Ahmed 1980: 82, following Leach 1960). Ahmed completes his definition with tribes to form an

egalitarian or an acephalous type of political organization. It is important to note that, for a tribal

individual, the affiliation to a sub-tribe, tribe or clan has not only organizational implications but

entails far-reaching repercussions. Basically, it confers rights of land and property. In particular with

the Pashtuns, it indicates a dignified social status considering the catalogue of exclusive qualities

attached to it emically: e.g. marked nobility, pride, bravery and honour. Within tribal worlds in

general, tribal membership is evidently a banner to carry in height with distinction:

Arab rulers would be deeply offended if their tribal background was questioned, Afghan dignitaries and intellectuals increasingly use their tribal names as a second name, a similar tendency is noticeable in Pakistan, e.g. Ghulam Ishaq Khan, a former President of Pakistan preferred to appear in tribal costume on official occasions (Glatzer 2002).

In addition, it may provide for a large network of primordial obligations on the basis of well-known

and structured genealogical ties. Generally, tribal membership may perform several functions. It

confers distinctive ethnic identity by its disassociative boundary against non-members on every

segment. Moreover, the structural connections and a sense of commonality between segments

(lineages, sub tribes) may ensure solidarity being activated in the occurrence of predicaments. In most

respects, alike solidarity is also contingent on the distance of kinship. According to Evans-Pritchard,

people within segmentary lineage organizations interact on the basis of genealogical closeness or


- 23 -

remoteness, respectively (Barnard and Spencer 1996: 152). Similarly, ancestors and genealogies serve

accordingly as a repository of political identity and as a resource for mobilizing adherents for political

action. Depending on the situation, the scope of kin-based inclusion varies. A famous Arab proverb

reflects this flexible system for political mobilization: “Myself against my brother; my brother and I

against my cousins; my cousins and I against the world” (ibid.). For the purpose of the present paper,

the relevance of segmentary descent groups for political matters must be examined further. Many

authors characterize tribe as a political unit which in turn suggests any descent group on a charter to be

a corporate group. If used by social anthropologists, the term corporate group commonly portrays a

collection of people recruited on a recognized kin-based principle applicable to groups that (a) share a

common identity with a single legal personality; (b) imply permanent existence; (c) include common

interests in property and common rules and norms (Boissevain 1974: 171; Barnard and Spencer 1998:

599). This definition of a descent group as a corporate body remains in contrast to an alternative

characterization of coalitions. Coalitions are generally defined as temporary alliances of parties for a

limited purpose and duration as well as a joint use of resources in an attempt to pursue a clearly

determined goal (Boissevain 1974: 171f.). An analysis of descent groups relative to their political

inference must therefore take into account three aspects: temporality (relative duration), goal-

orientation and instrumentality. As a consequence, an assessment of the political role of lineage

formation should not only include structural considerations, but must also comprise the individual

choice of political actors. Only the basis of individual choice is able to lay bare personal goals with

their inherent instrumental nature.

3.1 The Pashtun tribal system

Before setting out to consider pertinent aspects of segmentary systems for political activities, a few

words about Pashtun genealogy may be useful. All Pashtuns claim descent from their putative

ancestor, Qais bin Rashid. His real name may be disputed, but solely the belief of belonging to one

common prime ancestor, i.e. one huge kinship group on the top level is of major importance (Glatzer

2002). Qais was converted to Islam by the Prophet himself. He is said to have married the daughter of

the most renowned general of early Islam, Khalid bin Walid. From her, he had three sons: Sarbarn,

Bitan and Ghurgust. Karlanri, a fourth son, is said to have been adopted later into the family of the

third son (Ghurgust). All Pashtun tribes trace their origin to the offspring of Qais (ibid.). Two aspects

of considerable importance underlie this information about the genealogical origination of all Pashtun

tribes, whether historical true or fictitious: Pashtuns lack a collective hang-up attributable to a coercive

conversion to Islam. On the contrary, they were “enlightened” on the behalf of their genealogical apex

person. What’s more, the link to a famous general fighting in the name of Islam creates the mythology

of martial descent (Ahmed 1980: 129). This is consistent with the proverbial belligerency of Pashtuns

and adds to the characteristic and the self-image which advocates an ideal of the brave, martial,

intrepid and honourable man (turialáy). Before Qais, descent is traced back to Afghana, a descendant


- 24 -

of Saul, the first king of the Jews. This connection explains the name of today’s nation-state. Short

clarifications of terminology should include the term qawm (from the Arabic, meaning people, nation,

tribe, group). Qawm refers to human groupings of all kinds and analytical levels. Amid Durranis who

otherwise claim their superiority towards others, a qawm is ideally endogamous. The term therefore

defines all Durrani as equally eligible as affines (Tapper 1991: 47). Sub-tribe or sub-clan means khel

in Pashto, whereas larger tribal units (but not all of them) are referred to by adding the suffix –zay

(“son of”). Accordingly, Muhamadzai literally means “offspring of Muhamad”. Eastern tribes and

others may lack this typical tribal suffix. The Durrani and Ghilzai are seen as the two largest tribal

units and are therefore often referred to as a tribal confederation. Both the Durrani and the Ghilzai

were the founders of the preliminary state created in the 18th century and later.

Figure 1 shows an extensive genealogical charter of the tribal society of the Pashtuns.

Saul

Ghurghust

Armia

Bitan Sarban

Khalid bin Walid

Afghana

Qais

Karlanri

-Sharkbun (western) -Abdali (=Durrani) -Popalzai (Karzai) -Barakzai -Achakzai

-Kharshbun (eastern) -Ghoriah Khel: -Khalil -Daudzai -Mohmand -Chamkani -Khakey: -Mand: -Yusuf -Yusufzai Dir/Swat -Mandar -Yusufzai Mardan -Muhamadzai -Shinwari

-Ghilzai -Bhitanni -Ghalji -Lodhi -Niazi -Suri -Marwat

-Kakar -Gadun -Safi

-Utman -Orakzai -Afridi -Khattak -Wazir -Mahsud -Bangash

Figure 1:Pashtun putative genealogy. Based on Ahmed (1976: 7) and Ahmed (1980: 129).


- 25 -

Pashtun society is strongly male centred. With reference to the genealogical system, the term

patrilineal descent system is appropriate in order to reflect the emphasis on structural male dominance.

Barth does not hesitate to label Pashtun society patriarchal (1959: 22). As a rule, the senior male is the

formal representative of the domestic household in public. Pashtun males exert absolute dominance

not only in the public sphere. They also exercise extensive control in the domestic sphere. In the

family, the husband and father has all authority. He employs his pre-eminence in an attempt (a) to

control arbitrarily the social intercourse of the family members; (b) to cut his wife from her natal kin;

(c) to control all property; (d) to control all productive resources; (e) to use physical compulsion to

pull off his purposes; (f) to exercise his exclusive right in dissolving the domestic unit or in matters of

expulsion of family members (Barth 1959: 22f). In a more general sense, descent ideology is formally

relevant on the level of the ethnic group. There it determines marriage choice and political allegiance

(Tapper 1991: 46). Thus, the strict emphasis on male line in the genealogical charter is reflected and

reinforced in formal rights which in turn have a great impact on society. It should be clarified that

tribal membership does not manifestly call for sanguineal substantiation. Several genealogical myths

prove Pashtun pragmatism and a flexible handling of genealogical absorption. As a result, many

famous tribes, among them the tribal Ghilzai federation, are connected by adoption or by female links.

All the same, especially the Ghilzai case demonstrates the noble origin to be a crucial characteristic for

the acceptance of outsiders at least in former times (Glatzer 2002). In addition, outsiders who reside in

tribal areas for one generation or more may be accepted as tribal members provided that they honour

the tribal Code of behaviour (Glatzer 19982). The significance of genealogy is subject to change. One

indication for this might be that many contemporary Pashtuns are more likely to use residential

designations (e.g. Qandahari) than genealogical reference to denote groupings of tribes (Rubin 2002:

28). Nevertheless, one should take into account the aspect of security provided by the lineage

membership. Especially in times, when migratory displacement due to warfare events or other causes

that result in expulsion, one can hardly be “de-pashtunized”: “Nothing can strip […a Pashtun] of his

place on the tribal charter to which he is recruited by birth. The kinship chart provides him with a

permanent set of terms […] and preventing him from becoming anomic” (Ahmed 1980: 84). But an

inconsistency to this observation arises: a description of the Yusufzai segmentary system in Pakistani

lower Swat valley, where (descent) group membership is categorically connected with a title of estate

property, seems to challenge this view. Sons of a Yusufzai tribal membership who have lost their title

of the inherited land are no longer regarded as members of the group (Barth 1981: 60). Most likely,

similar discrepancies often ensue from disparate socio-economic backgrounds. Ahmed mainly resorts

to nang tribes in his accounts, whereas Barth for the most part portrays qalang societies in Swat.

Another aspect might demonstrate the importance of tribal membership. Since tribal descent and birth

is a principle of recruitment for Pashtunness, tribal affiliation confers on the member elite status. This

in turn underscores the prominence of tribal membership against non-Pashtun people. To be fair, one

should not fail to mention inequality between lineages. A certain favouritism towards seniority of a

lineage compared to a more recent one on the same comparative level of segment is manifest in


- 26 -

settlement arrangements. For instance, it may occasionally explain why some (senior) lineages have

access to the better lands along fertile river banks (Ahmed: 132).

3.2 Tribal affiliation and political matters: Does it matter?

For an analysis of political alignment, segmentary lineage systems need to be examined under the

aspect of common identity and interests. The essential question at hand is whether segmentary groups

operate as corporate groups. As maintained by Boissevain, corporate groups differ from coalitions in

various features. In the former, an individual identity is replaced by a group identity. In the latter, a

clear distinction of each partaking section remains and prevents individuals from adopting a group

identity. However, the maintenance in time may possibly set off a transition towards a corporate

nature of group cohesion by acquiring some of the relevant attributes. Apart from a collective identity

and temporal persistence, corporate groups are often marked by an individual commitment to an

uniform set of rights and obligations (Boissevain 1974: 172). However, I suggest that the most basic

distinction for the formulation of the imminent question relies on the degree of instrumentality. Thus,

two questions arise: Can assemblages of people be aligned only due to a common identity and a

preformed commonality? Conversely, are temporary groupings which are formed to achieve specific

political goals actually a result of an appeal to individual motives and self-interest? The appraisal of

whether tribes or tribal segments can be mobilized as complete corporate political groups or whether

they are no more than “alliances at events” is crucial. It is at the core of the theoretical discussion that

was initiated by a Barthian methodological individualism which in turn challenged the structural-

functionalist paradigm in the 1950ies: individual choice versus structural predisposition of the

individual lay at the heart of those armchair-row. Pashtun tribal associations in Afghanistan are seen to

serve more as a blueprint for political alliances, kinship being rather a convenient than a compulsory

way to build up alliances (Glatzer 2002). A clear distinction must be drawn between the merely

structural arrangement of units based on a unilineal descent system in a formal sense and the way

these units are made relevant for concrete corporate actions. This is a distinction in which many

descriptions of lineage systems are deficient (Barth: 1981: 55). Applying the idea of mechanical

solidarity, descent charters are generally composed of a hierarchy of homologous groups. Societies

based on mechanical solidarity are thus held together by their uniformity. A consequence of internal

tightness and coherence is a relatively strong collective consciousness (Barnard and Spencer 1996:

613). In a classical consideration of segmentary systems, solidarity derives from likeness. Aside from

certain restrictions, descent groups amalgamate with their respective political interests. However,

within Yusufzai Pashtun society in Swat for example, “descent charters do not unequivocally define

corporate units; […] the manner of recruitment of Pathan political groups cannot be understood

directly in terms of the descent system” (Barth 1981: 56). Why is this the case? For a transactionalist

approach maintained by Barth, political actors are not seen as single elements that function according

to unilineal descent principles based on their respective genealogical positions. Quite the reverse, the


- 27 -

implications of unilineal descent are related with their significance for individual choices. With the

Yusufzai case, variations from the classical description of segmentary systems are seen as a result of a

particular arrangement of dual opposing factions, which is a corollary of a specific structure of land-

allotment. In order to draw the comparison comprehensibly, it should be recalled that the alleged

principles underlie the classical rationale of segmentary systems (Barth 1981: 57fff. referring to Fortes

and Evans-Pritchard 1940: 284-296): (1) Unilineal descent defines a hierarchy of descent groups. The

remoteness of one’s common ancestor indicates all at once the scale of his group of descendants.

Situational opposition or alignment is flexible and rests on the nature and the magnitude of conflict.

As a result, a system of fission and fusion is inherent in an ever-changing texture of political

organization. Situational grouping for corporate political action runs along ascending line of unilineal

male agnates. Along those lines, a unilineal charter serves as a charter for the fusion of interests of the

kind enumerated above. (2) A progressive interlocking of segments along the ascending line of

agnates is seen as an intrinsic part of segmentary allying. That is, the fusion of interests initially

encompasses a merging of those more closely related by lineage bonds against those more distantly

related. (3) The equivalence of siblings creates a basic bond. Its ideal approaches a state which Barth

(ibid.) terms the merging of social personalities of siblings.

Technically, the Pashtun kinship system does not form a dramatic exception to those principles. A

major deviation might be assumed for the second postulation, though conflicts with the closest

collateral agnate are frequent in many segmentary societies. But in political contexts with Pashtuns,

Barth interprets the manner of such generalizations to lack reminiscence of African lineage

organizations, on which classical observations of segmentary systems are based (Barth 1980: 58). In

the context of Swat Pashtuns, the basic frame of unilineal descent is intertwined with political

alliances derived from individual choice and completed with a dual division. This is a result of an

inherent dilemma of a system of alignment giving political primacy to the bonds between brothers and

to the bonds between fathers and sons. A subsequent alliance between close agnate kins would unite

close collateral agnates as well. But in Pashtun society, the relationship between close collateral

agnates such as patrilineal cousins typically is one of profound enmity. The incompatibility of the total

of such requirements constitutes the dilemma for any Pashtun male ‘ego’. Individual strategies that

seek to solve it are seen as one constituent of Pashtun political dual organization in Swat. The

complements are a limited number of potential sources of conflict. Recurring matters of conflict

essentially concern issues related to land, such as the contentious use of irrigation facilities or a

controversy over the course of land borders, respectively. Alternative sources of conflict are about

violent events and issues of revenge involving honour and women. In councils (jirga), where most –

but not all– of the mentioned issues are debated openly and conflicts possibly settled unanimously or

arranged by informal lobbying, specific features of political organization become apparent. They

reflect the dilemma to a certain degree:


- 28 -

A hierarchy of councils is defined, corresponding in terms of their membership to a merging series of agnatic descent groups. But when members align against each other in debates, or any other form of opposition, they do not act in terms of such a merging series. In meeting of a council of a wide area, there is not the fusion of interests of smaller, related segments of a minor council vis-à-vis larger segments which one would expect in a lineage system […]. On the contrary, the opposition between small, closely related segments persists in the wider context, and these segments unite with similar small segments in a pattern of two-party opposition, not in a merging series of descent segments. (Barth: 1980: 63)

Obviously, the society of Swat Pasthuns is composed of close collateral lineages opposing each other

in case of conflict. But the more distantly related segments tend to form clusters of alliances. This

gives rise to a pattern that constantly creates a two-bloc system. Close collaterally neighbouring

groups or lineage groups that are integrated due to a common direct genealogical link do rarely merge

for political action. Rather small segments unite with alternative small segments, that lack immediate

proximity in terms of their genealogical position within a segmentary system. Inferred from this

specific system of segmentary opposition, a two-bloc system has constantly been developing while it

serves as a guideline for dual alliance or opposition (Barth 1980: 65fff.). If extended and reproduced

widely so as to cover a larger society, it generates a structure of political dual division into two

opposed blocs. In figure 2, the function of such a two-bloc system is illustrated on the basis of a

simplified unilineal segmentary structure.

Figure 2: Segmentary opposition and the two-bloc system. Source: Barth (1980: 67-68).

Explanation of figure 2: A and B are structurally opposed. a are the rivals of b, and c are the rivals of d. a may form an alliance either with c or d, while c may form an alliance with a or b, since either of the mentioned party represents a more distantly related collateral relative. Thus, the potential number of issues of conflict is limited. In case of meeting on a higher level (e.g. when A and B segments form a council as well as in a conflictive event), a and d contract to combine against their respective relatives b and c. The extended version amounts to a recurrent pattern of political dual division called dullah.

Consistent with the suggested person-centred politics in Swat and in other Pashtun-dominated regions,

each party of the dualistic structure is named after their local leader’s clan. With a web of alliances

A

d c a b

B

kinship relation

structural opposition

possible alliance


- 29 -

throughout the region based on the two-bloc system (dullah), powerful leaders in Swat are able to

name members of their party in 50 villages or more (Lindholm 1991: 172). Since descent groups in

Swat can be localized spatially, the collateral relatives are customarily neighbours, taken literally. As

said before, conflicts owing to land issues are the main source of inner conflict. For that reason,

collateral relatives such as ego’s patrilateral parallel cousin and his equivalent counterpart are

structurally prone to conflicts. First, because in a strictly patrilineal system they are opponents with

regard to the inheritance of estate property. In this respect, they are competitors. Second, as a result of

evident localization of descent groups, they are usually neighbours and therefore actors in an often

violent contest of land-rights or questioning the course of land borders or irrigation practices. Third,

the tension between agnate collaterals is manifest in kinship terminology. In ego’s generation, the

children of the father’s brother are distinguished by a particular term: tarbur. It has further the

connotation of rival or enemy. The problematic relationship between agnate collaterals is recognised

by Pashtuns. Though it is not regarded as ideal, since loyalty to one’s agnates is highly valued, it is

still regretted as a lamentable fact of life (Barth 1959: 109). Lindholm criticizes Barth’s suggestion to

give impermanent alliances priority over the principle of the unity of patrilineal segments, because “in

actual fact, party alliances are set aside in cases of blood revenge. No man would ever support the

murderer of his tarbur under any circumstances” (Lindholm 1991: 172).

We have just seen that the formal framework sponsored by the territorial system and its underlying

arrangement based on opposed descent groups creates a two-bloc system. Many smaller corporate

groups are aligned in two grand dispersed political divisions. Each bloc integrates members from

nearly every local community (ibid.: 104). But how are these groups held together? What is the

conventional basis on which politically operating groups function?

3.3 Contractual alliance and the two-bloc system: Conceptual explanations

In Swat, a series of alliances between local leaders (Pashtun khans and land-owners) lead to the fusion

into two great blocs across a comprehensive tribal area. The basic status of a mutual attachment is an

alliance between two political leaders based on a contractual agreement. The agreement assures each

joint partner of being supported in the defence of their several interests if opposed to outsiders. In the

absence of a given contract of political fusion, a leader however is not compelled to give political

support to any particular person or to any person at all. Thus, a leader may establish such an agreement

with any political actor, always keeping in mind the obligations of tacit loyalty to his father and

brothers. Conversely, leaders are not committed to oppose any particular person or group. Unlike

alternative political alliances such as agreements between landowner and tenant, fusion alliances are

composed of independent leaders and their equal partners. Being essentially equal towards each other,

“they are not engaged in any other joint co-operative effort, nor is one able to bring particular

sanctions to bear on the other. Their solidarity derives from the mutual strategic advantage they obtain

from the relationship” (Barth 1959: 105). Once a political constellation is short of such a strategic


- 30 -

advantage, the repudiation of and thus terminating of the relationship is implicit. In Swat, political

alliances are accordingly deemed political contracts by Frederik Barth. The described circumstances of

alike political relationships that differ in nature from any other relationship serve as the crucial point

of the argument: the very nature of free individual choice. According to Barth, free individual choice

is thus the vital principle on which similar political coalitions are based. Unlike membership owing to

common descent or territorial propinquity, political contractual promises for mutual support are hardly

embedded in imperative structural principles others than decisions of free choice (ibid.). But once

entered into a contractual agreement, mutual support is systematic and tacitly expected in political

conflicts, particularly if related to debates in public assemblies and to the use of force. In addition, if

an allied partner’s material rights or privileges are threatened, mutual assistance is anticipated.

Nonetheless, there are issues of conflicts of which the kind of alliance partnership at issue is not

susceptible: For instance, conflicts in defence of honour does never involve the alliance partner. An

issue involving blood revenge is a personal matter and a subsequent conflict is confined to individuals,

since “a man’s honour can only be defended by his own might” (ibid.: 106). On the previous accounts

and by suggesting Swat Pashtuns to be a functional segmentary organization, one might expect

constant processes of political fission and fusion. But this characteristic of any segmentary society is

less applicable to the alliances of Swat. Rivalries between individuals within an alliance is kept down.

An ally is a supporter in any situation irrespective of whom he opposes. To oppose to a person in one

situation and allied to him in another is scarcely conceivable. Therefore, we may attribute to such

contractual alliances some consistency, a considerable persistency as well as a qualitatively systematic

nature. Resuming the theoretical discussion over whether political alliances like the ones under debate

comply with the requirements of corporate groups or not, is a complex endeavour. To summarize,

incompatible interests between collateral agnates based on (1) a segmentary agnatic lineage system

and (2) a socio-economic arrangement built on genealogy-based allocation of land and estate property

foster the creation of a political two-bloc system. The main actors consist of Pashtun political leaders

and their followers, mainly non-Pashtun tenants. Smaller lineage segments ally with other small

segments based on a contractual agreement between their political leaders. Genealogical distance of

each segment appears to correlate positively with the leanings to enter into an alliance. Alliances

function as a strategy based on a rational decision which is derived from a series of free individual

choices. Being individual and free, the choice to join in an alliance is not structurally predetermined

either in terms of possible partners nor with regard to the mere constraint to be allied at all. But once

set up an alliance, it is subject to systematic assistance in case either partner’s interests are threatened.

In this respect, the assemblage of allied partners may be seen as corporate groups, if the definition

includes (a) a persistence through time; (b) groups acting as units in a co-ordinated fashion; (c) have

an internal distribution of authority and responsibility; (d) they exist unambiguously to protect a joint

estate such as interests and privileges of the constituting members (Barth 1959: 119). If we take into

account the important aspect of instrumentality in this context, we meet again the problematic nature

of terms which suggest dichotomy rather than a continuum. For the complex reality at hand, being


- 31 -

suspicious of applying simplistic labels is more than appropriate. Presuming that bloc alliances are

held together by common interests and calculated strategy so as to gain political significance, the term

mechanical solidarity is unlikely to be adequate. Bloc alliances lack a strong collective consciousness

provided by daily joint routine as well as an automatic, i.e. mechanical mode of mobilization. Barth

examines the Pashtun framework of unilineal descent and political organization by means of the

Theory of Games (derived from Neumann and Morgenstern: 1947) that favours focussing the attention

on the decision of the individual subject. In accordance with this theory, political groups are formed

through the strategic choices of persons. Hence, the solidarity of groups arises from the calculated

advantages which persons obtain –or anticipate in obtaining– from being a member of the group

(Barth 1981: 72-82). Alliances with Swat Pashtuns are explicit inasmuch as they require an overt

contractual agreement between political leaders. To a certain degree, they remain an instrumental

body, only mobilized in overt political action, if needed. However, the membership of a bloc surely

confers a certain degree of identity perceived by the members. Furthermore, it persists through a

certain time. Though political fission or factionalism might be assumed at least for every line in every

generation, when the estates of brothers passes on to parallel cousins, because brother’s sons become

cousins after succeeding their fathers. Barth points to important features that complicate

straightforward fissions of latent political units. First, son’s are active in political life before

establishing as independent landowners. During this stage, the contact with their father’s allies (and

later on their potential opponents) delays fission. There is some inhibition to abandon his established

friends for the prospect ally. Second, the leader of the bloc tries to exert his influence to hold the group

together. A large amount of time is invested in arranging settlements and compromises between

constituents of allies. Third, the mere necessity of belonging to a politically viable unit prevent an

immediate dissolution of political mergers. Thus, only events of overt conflict result in a political re-

arrangement. Tensions alone spare the “old world order” fission: “Thus persons whose structural

position implies rivalry may continue together in alliance for a considerable length of time. Swat

Pathans use the term mariz for persons who are local rivals but nevertheless belong to the same bloc”

(Barth 1959: 111). In accordance with that, the co-ordinated strategies develop in order to settle

disputes or to reach compromises with their structural opponents. Concessions in one dispute are given

in return for gains in another (ibid.: 111-126). Finally, some caution is advised in applying the political

framework of Swat Pashtuns to the tribal society as a whole. Given the socio-economic variability of

the entire Pashtun tribal society, alternative models of political organization must be assumed.

However, during the US-led war against the Taliban in the end of 2001, the attentive spectator was

becoming aware of the fact that even in major cities such as Kandahar, a type of two-party system had

obviously played a part when facing the handing over of control after the Taliban surrendered the

town. This points to the likely context-specificity of the system related with a quasi-feudal social

organization consisting of landowning Pashtun patrons in an ecologically advantageous environment.

It was shown that agnatic rivalry provides the structural foundation for a two-bloc system. The term

tarbur with its hostile connotations exists in largely all Pashtun language communities. Moreover, a


- 32 -

two-bloc system of named alliances (Gar and Samil) is characteristic to southern areas of the tribal

territories in today’s Pakistan and Afghanistan (ibid.). Therefore, we might altogether assume some

applicability of the political two-bloc system to a wider Pashtun community, given similar socio-

economic realities. Glatzer illustrates the functioning of a two-section division near Khost in East

Afghanistan named tor gund (white faction) and spin gund (black faction). It is rather latent than

overtly apparent, since it has practically become obsolete. But even there, in recent violent land

disputes between “former” members of different factions, an uninvolved party sided spontaneously

with their actual gund fellows (Glatzer: 2002). The case indicates a possible transience of similar

systems in the course of socio-economic and political transformation, but nonetheless unveils their

persistency and latent character as a part in the conscience collective of a community. At least in Swat

the superimposed dual party system has proved importance in showing a predictive power during

national elections in recent times. The Pushtun vote in Swat is likely to split in the traditional balanced

opposition reflecting the two-bloc division (Lindholm 1991: 171).

3.4 Tribes and politics: Critical and conclusive remarks

Altogether, tribal affiliation obviously does matter. It is less evident in the sense that larger descent

segments as corporate groups fuse for political interests. Rather, the strong tensions between

landowning agnatic collaterals regularly prevent descent groups from forming political units.

Conversely, they encourage to create an alternative system of contractual alliance partnerships which

serve as instrumental factions to secure protection from, or dominance over, his local agnatic rivals

(ibid.: 110). Despite all objections to factional tendencies, one should not overlook the weakness of a

segmentary system of the Swat Pashtun type. Alliances are instrumental. That is, they serve as to

achieve a political goal and are therefore primarily activated at times of conflict. In every new

generation, a new situation challenges the previous arrangement of coalitions due to changing

positions on the genealogical charter. Additionally, free choice derived from individual, goal-

achievement oriented strategy as the basis of a political joint venture is far from being a reliable basis

for durable political mergers. After all, the concentration of political actions on sole political leaders

fosters a person-centred politics, which is prone to personal capriciousness and possibly also to private

ambitions of the political leader. But there is another uncertainty factor which may obscure the

stability of the political fabric. The strive for individual as well as tribal egalitarianism prevents a

persistent coherence of political joint enterprises (Glatzer 2002). In addition, egalitarianism averts the

unstable and volatile leadership. Tribal divisions are structurally stable, tribal leadership is not. What

remains after having considered some aspects is the proverbial unpredictability of political life based

on tribal leadership (Glatzer 2002). Before extending the debate, let me emphasize an important point

about the accountability of tribal systems for political matters. In most lineage organizations, descent

groups fuse as corporate groups for political objectives. Even though opposition between collateral

agnates is to some extent inherent in any segmentary system, lower segments of descent groups merge


- 33 -

on an ever higher level of segmentation for political action. Extending this strict logic, ramifying

smaller segments fuse into larger ones. However, among Pashtun societies in Swat, small adjoining

descent segments rarely merge on a higher segmentary level for political action. Instead, they join

politically with other segments, collaterally more distantly related, in a system of two blocs. Thus,

opposition between close collateral segments is maintained in all circumstances (Barth 1981: 80). But

objections to this suggestion are appropriate, since Pashtun history is evocative of an alternative

implication of tribal associations. In 1722, after ousting the Persian Safavid rulers from Qandahar, a

tribal confederation of the Ghilzai succeeded conquering and usurping the throne of the Safavids in

Isfahan. A powerful khan of the Ghilzai tribe of the Hotakis led the resistance. However, the tribal-

based kingdom was shortlived and disintegrated in bloody battles of succession. A lack of any

organizational basis quickened the decay among other reasons for Ghilzai decline. They subsequently

succumbed to the Turkish conqueror under Nadir Shah Afshar. The tribal Abdalis (today’s Durrani)

offered him generous assistance without going away empty-handed, of course: The Abdalis amassed

wealth and military experience in the Turkish army (Rubin 2002: 45; Glatzer 2002). This short outline

of historical events serves as an exemplary analysis. It may be instructive in order to interpret some

vital characteristics of Pashtun politics on a more abstract tribal level: (1) Towards the top level of a

segmentary tribal union, an inclusive boundary appears to arise. It delimits a margin in opposition to

other tribal mergers. This tribal grouping conveys a type of tribal or ethnic identity (the Abdali; the

Ghilzai). (2) In order to be successful in tribal merging on a higher segmentary level for political

action, the existence of a powerful and integrating political leader seems to play a key part in this

process (cf. chapter 7). (3) Tribal identity overrides a comprehensive Pashtun ethnic identity. Each

tribal association manipulates actively or benefits from political events and relationships at the

expense of a tribal competitor. If advantages can be anticipated rationally, such unscrupulous

interfering is favoured while ideological ethnic considerations are suspended. The appeal to an all-

embracing Muslim identity within the religious community may be an exception. (4) In spite of an

initial success to rule as a tribal unit, a strong propensity for internal factionalism thwarts efforts to act

as a persistent corporate group for common political action. The frequently addressed phenomenon of

a cousin enmity between patrilateral cousins is a recurring theme in Pashtun politics. In 1973, when

Muhamad Zahir Shah, at that time sitting enthroned as the king of Afghanistan, had relished a

treatment stay abroad, Daoud Khan replaced him on the spot without asking for audience with the

king, beforehand. Daoud Khan was essentially Zahir’s cousin (Rubin 2002: 58). (5) Like in a zero-sum

game, the wrangle over power has obviously the effect of bringing about a triumphing party while the

parallel contestant’s power is on the wane. The competitor is usually outdone by manipulating exterior

groups of political actors, or rather players. A similar process can be observed when Pashtun groups in

today’s north-western Pakistan tried to use the rule of the British to play off their tribal counterparts

(Ahmed 1980: 72-78). A constant scheming and manipulation which subsequently produced

unpredictable political manoeuvres on almost each side of acting groups has also been observable in

the last decades amidst the geo-strategic game of super- or middling –powers in the wider region.


- 34 -

After the withdrawal of the Soviets, countless defections and re-arrangements in the political fabric of

the violent conflicts turned out to be an unpredictable but constant factor. Again, once having got rid

of an exterior enemy that united the factions under the umbrella of jihad, “the fragmentation of the

political and military structures of the resistance prevented the mujahidin from turning local victories

into a national one. For many, […] the personal obligation (farz-i ´ain) of jihad ended […]. As in more

normal times –but with more weapons– they engaged in struggles for local power” (Rubin 2002: 247).

Exemplary illustrations in a specific context only serve as a contributory factor of a possible function

of the Pashtun segmentary system for political actions. Being context-bounded as well as subject to

change, the explanatory significance of a segmentary system is an open question. Various authors

evoke an ongoing effect of de-tribalization (Rubin 2002: 29). Furthermore, the causal direction as well

as a potential over-accentuation of the tribal factor should be considered:

Ethnicity and tribalism are often held responsible for the Afghan disunity. Indeed practically each of the conflicting parties and groups […] show a certain emphasis towards one or another ethnic group. This, however, is no proof that ethnic and tribal divisions are the cause of political cleavages and violent conflicts. Every Afghan belongs to one of the ethnic groups and every Pashtun belongs to one of the tribes, thus a quarrel between two Afghans who incidentally do not belong to the same group or tribe may easily be misinterpreted as ethnically or tribally motivated . (Glatzer 2002)

A future debate ought to centre upon the often neglected aspect of forced migration with regard to

tribal systems and their political function. In a tribally organized ethnic group like the Pashtuns with

millions of war refugees, a question arises: Does migration support tribal bonds, or quite the reverse

do migratory processes result in severing kinship ties? Despite all the uncertainties regarding the

relevancy of the Pashtun tribal organization for political actions, the importance of the genealogical

basis should not be rejected, either. Rather, I suggest the tribal organization and the kin-based identity

to be analysed against the background of a goal-oriented calculation of political actors who weigh the

utility of tribal bonds against advantages and drawbacks of alternative options. Besides, the simple

convenience and practical aspect offered by already existing tribal or local ties should not be

underestimated. The former proposal follows Barth’s analysis adopts a perspective of utility

maximization and rational choice. It focuses on rational decisions of the individual actors based on

options of choice to solve his problem, rather than defining a set of formal structural positions. From

this perspective on political acting, one might be inclined to insinuate a large portion of

instrumentality: “Whether social action is based on tribal and ethnic criteria depends on opportunities

and tactics and may change quickly. E.g. the Pashtun party leader and warlord Gulbudin Hikmatyar

initially laid a stress in his public speeches on panislamism and the Muslim ummah. […] Later, during

his campaigns for recruitment in Pashtun areas he appealed to the ethnic and tribal solidarity of the

Pashtuns who should defend their identity and honour against the rest of the world” (Glatzer 2002).

As final points, let me summarize the most important key aspects of the structural basis of the

PAshtun Segmentary Tribal Organization (“PASTO”):


- 35 -

• IDENTITY: PASTO confers ethnic identity. Members are equipped with and guided by specific qualities attached

to membership of PASTO.

• EGALITARIANISM: PASTO is the basis of the ideal of egalitarianism.

• EXPEDIENCY: For practical matters, PASTO-links are most convenient as a basis for political alliances. It

enables leaders to form political entities.

• FACTIONALISM: The divisive character prevents PASTO from forming enduring political entities as well as

from performing any collective action as a tribe. Apart from structural features such as inherent egalitarian

predispositions, agnatic rivalry and enmity of cousins are largely responsible for tribal factionalism.

• THE LEADER: There are at least two constructive factors that appear to have the potential to increase (sub-)tribal

merger and to mitigate tribal parting: First, a capable political leader of high integrity and wide acceptance seems to

be a indispensable prerequisite so as to succeed in transforming PASTO in a political merger. Tribal khans are seen

to be the key actors in order to “tie the knot of the tribe” (Glatzer 2002; citing Anderson 1978). Second, external

threats such as invasions or impending dangers are able to combine efforts in an attempt to gather tribal segements

as politically functional groups. Religious leaders often emerge as powerful actors to take the lead in times of crisis.

Looking at it that way, we should be cautious not to overstress PASTO as the merging factor. More accurately, the

Muslim community substitutes for any other collective identity in the process of political mobilization.

Despite all indicators of tribal factionalism, Pashtuns educate, view, assess, act and react on common

grounds as well. Apart from shared historical experiences which have been introduced in chapter 2, a

fundamental principle of Pashtun views is represented by Pashtunwali and its inherent attributes to an

idealized Pashtun person. Pashtunwali is of paramount importance for the tribal daily life and the

political behaviour: It is often referred to as the tribal law. The succeeding sections will deal with

implications related to the tribal law, the idealized tribal man and the way these concepts interact with

political matters.


- 36 -

4 Pashtunwali: The tribal law

Being a “true” Pashtun requires being a good Pashtun. From an emic point of view, Pashtun is not

only considered as a label for ethnic or linguistic membership, but also as an expression of fulfilling a

set of stringent qualities. Consequently, being a good Pashtun implies living and acting in accordance

with the ideal of ethnic self-representation that is embodied in the concept of pashtunwali. The

concept encompasses the idealized core values of the Pashtun person and his behaviour, since it is the

epitome of Pashtun moral and juridical standards. Pashtunwali has been the customary law of the

Pashtuns for centuries. We can find evidence in descriptive sources already a couple of centuries ago

as “the manners and customs of the Afghan tribes” (Glatzer 1998a). Even where the term is unknown,

the identical substance of traditional norms, that hardly differ from those of their neighbours, is active.

‘Pashtunwali’ provides a definition for the ideal person (ibid.). Pashtun culture is described as being

immensely imbued with ‘the Code’. Through this pervasive nature it has a great plausibility for social

behaviour:

Although unwritten and precisely undefined it is the theme of song, proverb, metaphor and parable and never far from men’s minds; like most codes it is part-fiction and part-reality. (…) It is remarkable how native and foreign models coincide with regard to the main features of Pukhtunwali and its relevance and sanctity in society. It is equally remarkable how similar the ideological model is to the actual immediate model of empirically observed social behaviour and organization. (Ahmed 1980: 89)

The ambiguity of the term pashto as a reference to the language as well as to a distillate of an ideal

pattern of behaviour must be interpreted but as one expression since the Code is seen as the language

or idiom through which the Pashtun conveys his “pashtunness” (ibid: 91). Pashtunwali appears to

bridge the regular gap between etic models of outsiders and emic explanations. Accordingly, the

empirical reality of behaviour seems to act in far-reaching accordance with the ideological claims In

West Afghanistan, an equivalent code is called rawaj, instead. Rawaj equals to a set of principles for

aspired human qualities that is termed ghairatman, the “good Pashtun”. Contrary to southern and

eastern Pashtun area, the qualities of ghairatman do not enclose barely the ethnic Pashtuns. A “good

Tadjik” or a “good Aimaq” as an epitome of the “good Afghan Muslim” may also be marked by the

label ghairatman (Glatzer 1996: 111).

In practice, pashtunwali has many functions:

(a) It serves as a guide-line for behaviour as well as a model and an orientation for education.

(b) It provides an assessment criterion for conflict resolution.

(c) It is a selective tool that may exclude ethnic foreigners. Consequently, it endows the members

that follow the Code with an ethnic distinction. It is powerful in accentuating ethnic membership


- 37 -

and social boundaries. This quality may be activated in times of change or conflict since it

imparts comprehensive plausibility.

(d) It has catalytic power, since it holds up and reinforces a distinct set of values that stand for the

“good Pashtun”. Similarly, it closes off desirable qualities from sanctioned characteristics and

behaviours.

(e) It provides legal standards in territories where state-like institutions based on a framework of

legislation and a sound system of law enforcement have been partly absent for centuries.

It is important to see that in an ideal-type model of pashtunwali the actor is tacitly compelled to pursue

his objectives in accordance with the Code. That is, in a case of murder or robbery that is placed

outside the recognised laws of revenge, the actor will be severely dealt by society through the tribal

assembly. On the contrary, the negligence of normative behaviour such as adequate retaliation will be

condemned just as clear as not doing Pashto. “In his freedom he is slave to the Code” (Ahmed 1980:

94).

Before it is adequate to mention some critical objections in view of the praises that coat the

phenomenon of the Pashtun Code, a description of some central features is requisite. According to a

multitude of sources that cover the literature on the Pashtuns, the main values and institutions appear

to be nang, badal, melmastia, nanawatee and jirga.

4.1 ‘Nang’, the concept of honour and shame

The crucial term in pashtunwali includes the complex idea that centres around honour and shame,

dignity, courage and bravery. The pursuit of an honourable life is at the core of the idealized concept

that approximates how not only to speak, but to live Pashto. Nang is found as a central theme in poetry

and literature. It guides every-day action as well as social and military action: “I despite the man who

does not guide his life by honour. The very word ‘honour’ drives me mad”, as Khusal Khan Khattak

(1613-89) is quoted (Spain 1963: 63). Nang is still today so essential that pashtunwali is at times

freely equated to nangwali (Ahmed 1980: 91; Glatzer 1998a). The juridical function of the Code

becomes apparent when societies in qalang and nang situations are opposed. Apart from the

polysemous implication of the latter which gives some account of the importance of nang-qualities in

this context, the absence of any other law underpins the significance of a tribal law, in which nang is a

dominant component. Ahmed bases this observance on economical and ecological circumstances that

have a deep impact on the quality of personal prerequisites for political leaders: “Economic channels

of mobility and power are so obviously restricted in the ecological situation in which the tribe finds

itself that nang, as a symbol of prestige and social mobility, assumes a key aspect” (Ahmed 1976:

75fff). Nang as a pivotal concept and a complete codification of behaviour offers also an important

analytical instrument and a standard of reference. Its ideal-type gives insights into normative tribal

behaviour and socio-political understanding. In other words, the inclusive bearing of nang can be


- 38 -

sometimes applied to explain a “subordinate” phenomenon such as badal, since only the absence or

impaired nang of a person or family sets off actions based on the concept of badal. We will see later,

how ideas revolving around nang qualities play a key role in constituting a personal capital in order to

gain political authority. Political success of leaders can partly be attached to whether they succeed in

living up to the ideal of an honourable and dignified life-style. Similarly, even economic achievement

is linked to it. In particular, nang provides also explanatory aspects in order to understand the

normative behaviour towards women. A nangialáy is a man who meets with the requirements of nang.

In doing so, a person brings honour and fame to his kin-group. In contrast, “to be called benanga

(‘shameless’, ‘undignified’) is the worst possible insult in Pashtu and a deadly threat to the social

position of the insulted” (Glatzer 1998a). A culturally accepted and even encouraged way of regaining

one’s nang includes retaliatory actions, not ruling out killing the insulter. The insult of a person’s nang

adds up to the involvement of his kins. Some important characteristics of the concept become obvious

here: (a) the extension of the concept beyond the individual by engaging family ties (especially

women) as demonstrated by the concept of badal; (b) the instrumentality of nang that makes it to be a

major non-material resource to gain respect through highly valued “Pashtunness”. As a result, it

enables to acquire political and economic advantages; (c) the interconnections with alternative notions

within pashtunwali such as badal; (d) the complex and complementary character of nang, since it is

not only defined by what it constitutes, but above all by the negative and hence complementary

attributes that prevent the accomplishment of nang criteria.

This is why in this context two features of nang need a closer look: sharm and namus. The former is a

significant part of nang. Sharm is roughly equivalent to the English term ‘shame’ in the sense of noble

modesty. Again we can identify a strict ambivalent quality of the notion: its implementation into

concrete action or the failure by counteracting the ideal of the concept in every-day life, respectively.

For example, a boy may guzzle down a meal greedily. His father might possibly ask him: “sharm

nálare?!” (“don’t you have shame?”). Similarly, a man does not check her daughter’s flirt with the

neighbour’s son. In this situation, people would probably say: “sharm nálarî” (“he has no shame“).

The latter is an extremely serious insult and requires impressive and categorical action. A multitude of

incidents that concern sharm is related to the behaviour of females, their control from men and the

consequences for women’s family and kin-group. Less the honour of the woman itself but the honour

of her whole group is at stake (Glatzer 19981). The term namus belongs to the complex of nang, too. It

means privacy and the protection of its sanctity. It is essential to analyse it in terms of women’s moral

credit and its significance for consanguineal and affinal male kin in public discourse: “In the narrower

sense namus refers to the integrity, modesty and respectability of women and to the absolute duty of

men to protect them. In a wider sense namus means the female part of the family, of the clan, tribe and

of the Afghan [i.e. Pashtun] society; in the widest sense it is the Afghan home-land to be protected”

(ibid.).

The importance of women in the described concept of honour and shame reveals an explicit perception

of the role of women and their inherent characteristics. Among Durrani Pashtuns, the superiority of


- 39 -

men over women is expressed in a dichotomic mental chart of gender attributes: Whereas men are

associated with nobleness, authenticity and perfection, women embody imperfection, weakness,

polutedness and are seen as “imitations” (Tapper 1991: 52). Women are routinely linked to emotion

(nafs). By contrast, men have reason or responsibility (aql) (ibid: 209-211). Young women are

considered less to be able to act rationally. They are believed to have less self-control and to be more

inclined to sexual activity. As a consequence of this unambiguous gender-specific stereotyping,

women are an easy prey to any seducer coming along (Glatzer 1998a). By inference, the possible

threat of a violation of a woman’s namus is palpable. Since namus in a wider sense means the female

part of the whole family and affects severely the husband’s integrity, the honour and shame (nang) of

a wider kin-group is at stake. Therefore, “men feel obliged to fight for maintaining their namus, i.e.

first of all to keep the women of their families under tight control and to protect the women from their

own ‘weakness’ “ (ibid: 5). However, men are less worried about the actual behaviour of women.

Rather, the belief and perception of the public is concerned. Neighbour’s gossip is feared the most.

Gossip is difficult to control and may erode namus effectively. This in turn reinforces men’s attempts

to protect their women: “Better not let anybody see the women” (ibid: 5). Still, the reality appears to

be more complex. Glatzer was told by north-western Pashtuns in Afghanistan that the authentic

strength of a man need no austere protection management of women since no outsider would dare to

come too close and “only weaklings need to hide and lock-up their women” (ibid: 6). This may be

true. One is tempted to deduce positive correlations between the lack of power and the efforts to

protect women from public sphere. All the same, journalists nowadays still report on the difficulties to

get occasions where women would be allowed to talk to male interviewers (Sorg 2002fff). In a paper

that pretends to deal with conflict and political matters, one may be surprised at the relative

importance of gender subjects granted here. Violent conflicts like feuds can be most typically put

down to three causes: zin, zer and zamin. Zin stands for women. Zer literally means gold and refers to

any kind of portable property and is also utilized in making gifts or bribes. Zamin refers to land and

other immovable property or rights to land and water, respectively (Barth 1959: 73; Dupree 1984:

269). We can hence identify a trinity of causes for conflicts. My proposal would be even to interpret

these features as resources to create and maintain political influence by adding the whole set of

concepts that centre upon idealized qualities of person. Some of them have been addressed already. To

resume the topic, women are deeply involved in generating and maintaining men’s and family’s honor

by avoiding shame. Other sound arguments account for a comprehensive, though passive, role of

women: Firstly, the inviolability of women is nearly equated with the significance of land, as a

common saying goes: “The way to the women leads over the land [since he] who cannot protect the

integrity of his family cannot protect anything, anyone is free to snatch away from him what he wants,

his possessions, his land” (Glatzer 1998a: 6). Secondly, most authors agree that an impaired namus is

a frequent cause for violent conflicts (ibid). Thirdly, the concept of namus appears to set off extensive

political plausibility. Apart from religious arguments, mujahedin were engaged for the resistance

against the socialist regime (and subsequently against the Soviet invasion) by reasoning that the namus


- 40 -

of Afghanistan is threatened. This happened during the late seventies of the last century. In the course

of similar events, many people in the basically rural country defied the strict school enrolment policy

of girls. But the reason for the suspicion was not a categorical rejection of female education that was

being raised. Rather, the encroachment on the private sphere engenders rebuffs. This reflects the neat

separation between public and private spheres in traditional Pashtun society. To be precise, the

interference in matters where females are involved are intensely private (ibid: 6).

4.2 ‘Badal’ and ‘tarboorwali’ (agnatic rivalry)

Badal generally denotes forms of exchange. To make a distinction from the exchange of women in

badal marriages between close agnatic kins, the term badal being considered here is about exchange in

the sense of activities in retaliation for a morally sanctioned behaviour. The translation reflects a

notion that accounts for retaliatory concepts such as revenge, feud and vendetta: “He is not a Pukhtun

who does not give a blow for a pinch” (cf. Ahmed 1975: 57). Between cousins, rivalry is strong and

outspoken. Dispute express or question dominance and power. Tarbur do not fear another. Rather,

they challenge one another. Lindholm renders an account of a devastating feud ongoing during his

fieldwork in Swat, which illustrates the severe economic repercussions pattern of solidarity between

direct relatives in cases of tarbur conflict:

[It] began with a boy’s refusal to let his second cousin play soccer with him. This insult led to a fight which spread to the boys’ fathers. At the close of the fieldwork three men were dead and the fields of both families had either been sold for weapons or else left fallow as the remaining men sought to eliminate their rivals. (Lindholm 1991: 169)

As stated by Ahmed, badal provides the primary law of Pashtunwali. (1980: 90). Dupree simply states

it in the slogan “To avenge blood” (1984: 282). As with other features of pashtunwali, badal is

interconnected with complementary elements such as nang, sharm or melmastia. For instance, with

regard to the latter that encompasses the hospitality to guests, one must revenge and thus “to fight to

the death for a person who has taken refuge with me no matter what his lineage” (ibid: 282). Badal

(“the primary law”) can be extended to comprise tarboorwali and tor (Ahmed 1980: 91). Tarboorwali

revolves around the current agnatic rivalry usually between patrilateral parallel cousins of the same

generation. Tor approximates to the prior term of namus. It is used in cases where the chastity of

women is compromised. In such a case, both affected parties are said to be tor and its offence is

considered to be of the utmost seriousness. Ahmed stresses on the one hand its paramount significance

within Pashtun patterns of norms and on the other hand he points to the uncompromising nature of

those central features:

The very concept of nang is equated and reduced to tarboorwali and tor. It is in the pursuit of these two that the other principles of Pukthunwali may be tacitly suspended. Considerations of tarboorwali and tor override sanctions of custom, Code and even religion. Around these two features the boundaries are clear and unambiguous. There is no compromise in practice. (ibid: 91)


- 41 -

Furthermore, the binary assessment of a behaviour is strictly based on the compliance with the norm

and refers to the accomplishment of the act, but also to its omission. Either one has done Pashtu (killed

his tor daughter; restored his own nang by killing the responsible) or one has failed to do Pashto (not

killed his tor daughter; not taken revenge to restore his own impaired nang). Here the extension of the

Code to a more general evaluation of the ideal Pashtun person becomes apparent. In addition, actual

social behaviour –or at least its evaluation– and ideal behaviour seem to correspond in practice.

4.3 ‘Melmastia’

Considering the proverbial expression of a Pashtun elder that equates a Pashtun with “a friend of

guests” (“da melma dost”), it becomes clear that hospitality appears to be a key feature of behavioural

ideal towards invited company that seek shelter and protection. The custom of hospitality is imperative

for a Pashtun host. Any guest will be given the most sumptuous commensal service as possible, even

in the absence of a household’s wealth. The guest-host relationship is complex though given that e.g.

the Durrani Pashtuns are described to humble themselves before their guests. He accepts that the guest

has the right to demand and receive whatever thing. However, this apparent superiority of the guest is

deceptive, for receiving hospitality is also a submissive act since he honours his host and implicitly

acknowledges the host’s ability to satisfy his requests (Tapper 1991: 172). Additionally, entertaining

guests has wider social and political implications. Not only should a village or any associated

community share the burden of the guest’s expenses, an elder, a patriarch or senior lineage mashar

(elder) is responsible for the guest’s costs, but also for his security (Ahmed 1980: 90). Hospitality as a

special form of gift-giving implies solidarity and may express inequalities between guest and host. The

recipient is under an obligation of respect. The host proves his economic and political success through

the generous offering of his wealth (Barth 1959: 77). Being a host, on is obliged to assure the security

of the guest. What's more, the host proves and at times risks commonality with a guest. An unpleasant

guest from the point of view of the community may harm the integrity of the host. Currently, we can

observe that western journalists in the area of Kandahar, a former stronghold of the Taliban and a core

region of southern Pashtuns, are deeply distrusted. A shadow of suspicion and disappointment is cast

over Pashtun hosts of western reporters who are on occasion held for or equated with Americans (Sorg

2002; Espinosa 2002). The blunt refusal of Taliban’s political leaders to meet the American demands

to deliver some of their Arab “clients” must also be seen in the light of Pashtun custom.

4.4 ‘Nanawatee’

According to Ahmed, nanawatee is an extension of melmastia. It is deduced from the verb that has the

signification of “going in”. Whenever an enemy “goes in” (i.e. “comes in”) to offer peace, the act is

considered a supplication. This in turn is to be acknowledged tacitly by the receiving counterpart.


- 42 -

Nanawatee is usually substantiated by a ritualized slaying of goats or cows (Ahmed 1980: 90). The

obvious existence of an institutionalized act of exoneration from the social constraint to seek

retaliation is of great importance: “More recently, when the son and grandsons of [the elder] Shahzada

fired on Mazullah […], Shahzada swallowed his pride and asked for nanawatee rather than see one of

his offspring killed.” (ibid: 90) In view of the widely expressed assertion that Pashtun culture demands

imperatively acts of retaliation for physical and nang violations, such alternative features are helpful

mitigations. Nanawatee is additionally complemented by an attribute of the ideal person: The virtue

that corresponds with aql (reason), which is the counterpart of turá (the sword).

4.5 Jirga As a public assembly of elders, the jirga is composed of elder respected persons and informal political

leaders on local level. It deals with specific issues for which the members (jirgaeez) of the assembly

must find decisions that are binding on parties in conflict. The size of a local jirga may vary

depending on the scale of the issue under debate. The treated issues also are diverse in their

magnitude, with regard to the actors as well as to the topics involved: “The jirga regulates life through

decisions ranging from the location of a mosque to the settling of conflict within sub-sections, to

larger issues such as regulating foreign relations with other tribes and even conveying decisions of the

the tribe to government” (Ahmed 1980: 90–91). Both the Islamic law and the Pashto custom are the

guidelines on which decisions of jirga are based. With reference to the political life of Pashtun culture,

three particular aspects illustrate wider implications: First, the existence of a traditional political

institution such as the jirga shows again the institutionalized attempts to provide a legal or rather

legitimate alternative as a substitution to written or formal law, respectively. Second, the jirga is

regularly referred to an radically democratic institution. Third, it demonstrates an ideal of equality and

autonomy that give priority to sameness. It reflects the highly valued importance of the “free man” and

highlights the inter pares, though a primus is certainly far from being absent as the heading in a

relevant title strikingly implies: A somewhat hierarchical society with an ideal of egalitarianism

(Centlivres and Centlivres-Demont 1988: 22; cf. chapter 5.3).


- 43 -

5 Being Pashtun – Concepts of person and politics

Pashtunwali serves as a tribal law. Explicitly, it must provide a customary basis for decisions in cases

of conflict. It is not a written charter of unambiguous rules. Therefore one must be conversant with

behavioural features that are to be reinforced as well as proved not only by verdict, but also in the

course of daily routine and public discourse. Proverbs, self-ascription and public discourses are filled

with values, that are attached to the idealized Pashtun person (Ahmed 1980; Dupree 1984). This

idealized person is not only supported by public discourse and verdicts based on Pashtunwali. The

tribal Code itself is packed with ideal qualities of a Pashtun person, as we have seen. The compliance

with badal requires courage and a personal strength to execute an obligation. It calls for loyalty

towards kin members too. Melmastia requires generosity and openness. The ability to show an

adequate deference is needed as well. Nanawatee demands forgiveness and the ability to suppress the

need for retaliation. Jirga is based on deference towards a judgement that is not based on transparent

law or written theorems, but on debate and unanimity. More important: It is founded on the belief that

everyone is equal. The concepts of nang, sharm, namus are all about obscure notions such as dignity,

shame and honour. Those themes are recurring in scientific literature on Pashtun societies. This is why

I take their significance to a certain degree for granted. To complete the panorama of components that

encompass the ideal Pashtun person, two further dimensions are important to tackle questions of

violent conflicts: turá, aql and the ideal of egalitarianism.

5.1 ‘Turá’: the sword

According to Glatzer, nang and the concept of honour and dignity has two counteracting sides: Turá

and aql (1998: 6). Turá may be translated as “the sword”. As an allegorical reference, it encompasses

the ideal of aggressiveness as well as the readiness to fight until self-sacrifice. Turá is supposed to be a

dominant element of the ideal personality associated with young men. It validates hotheadedness and

rashness in situations where provocation is answered with impetuous acts of violence. Channelled by

the discipline of tribal militias (lashkar), turá is instrumental in achieving objectives. Therefore its

social acceptance is reassured. All the more, as a characteristic attached to the individual, it must be

constantly proved by courageous actions. Consequently, the turialáy, the one who embodies turá, is

respected. Turá does not only serve as to demonstrate his personal honour, but it also reinforces one’s

own equality and autonomy and hence that of one’s family. In displaying turá, “one has not to bow

down before any arbitrary power” (ibid: 7). In this respect, turá combines with the ideal of

individualism and egalitarianism. It must therefore be seen as an operative measure to prevent the

intimidating loss of one’s honour and independence. Besides, it supports a select behavioural trait that

is indispensable in meeting requirements of traditional institutions or rather events such as blood feuds

or comparable violent occurrences. Were Pashtuns successful warriors due to their cultural

prerequisites that underscore martial values or doe they emphasize bravery their because of their role


- 44 -

in military systems and the importance of violent public conflicts in their history? Regardless of the

causal direction, fact is: Pashtuns did not only play a crucial role in the standing army of the Persian

empire, but the image of the Pashtuns as courageous and heroic warriors is effective until today. With

a share of about 25%, they are proportionally overrepresented in the army of Pakistan. Despite their

relative small share of Pakistani national population (14%), the Pashtuns are the second largest group

in the army (Rasuly 1993: 34; Schaffer and Mehta 2002: 5). Of course, the setting of national and

regional politics may as well explain the role of Pashtuns in the national army, since Pakistan attempts

to undermine Pashtun nationalism by integrating Pakistani Pashtuns institutionally. However, a select

integration of Pashtuns as soldiers is supposed to reinforce their (self-) image as heroic warriors. In

doing so, an accentuation of martial characteristics might be assumed.

5.2 ‘Aql’: responsibility and the corrective of ‘turá’

Since turá might be harmful for social existence, if undirected, we find its counterpart. Virtually

organic, aql averts the destructive occurrence of a literal overkill of turá. Aql, originally an Arabic

term, may be translated as “reason” or depending on context as “responsibility”. The notion suggests

competence as well as predictions and decisions based on rationality, sensibility, common sense and

prudence. Furthermore, it also has connotations of a willingness to accept the consequences of action,

self-control and discipline (Tapper 1991: 209). Aql-guided decisions should definitely bear in mind the

benefit to one’s family and one’s wider social environment: a definition of social responsibility that

may include the ethnic group, the nation or the entire Muslim community (Glatzer 1998a). Young

men’s turá is not expected to be tempered by young men’s aql. Rather, it is supposed to be checked by

the aql of the elders, the “white beards” (spin gíri). Consequently boys are educated to obey the elders

(ibid: 7). In practice, aql as the respected feature of elders’ personality is the basis on which decisions

of jirga are founded. Aql is thus functional in jurisdiction and mediation in case of conflicts. The two

sides of nang are connected with different ages. But they imply a symbiotic aspect, too. For instance,

decisions that are swayed by aql depend at times on the persuasiveness of turá in lashkars to execute

the decisions of the egalitarian jirgas.

5.3 Egalitarianism: ’har saray khan day’ – every man is a khan

Throughout scientific literature, more or less implicit allusions are numerous, where the facts are

linked to an inherent idea of Pashtun strive for egalitarianism and autonomy: “Les hommes de

l’Afghanistan professent un idéal d’égalité et d’autonomie, souvent affirmé dans les institutions

tribales” (Centlivres and Centlivres-Demont 1988: 22). Tribal institutions such as jirga provide

exemplary evidence of the weight that is attached to sameness. The participants are numerous. They

must have legitimacy in terms of their political influence (not power) as well as their personal qualities

(aql) that are culturally accepted. Decisions based on the public assembly of jirga are best to be

unanimous. The equal rights of the assembly’s participants are symbolised in the spatial order that


- 45 -

favour a circular arrangement. The public disapproval of a deficiency of independence also of men in

Swat has been described by Lindholm: “All forms of dependence and of need are reviled as weakness,

since they show that a man cannot stand on his own against the world” (1988: 234). Historical

incidents that prove a martial defence of independence are further instances for the cultural

endorsement of egalitarianism. Among such events, the violent resistance to colonial subjugation

during the British occupation in the 19th century or the battles against the Russians account for a stern

reflex against the loss of freedom. Conversely, one must concede the ambiguous role of ephemeral

coalition building on the part of some Pashtun groups, if only to do harm to a rivalling Pashtun foe.

Furthermore, the insistence and cultural accentuation of values such as individual pride as well as the

honour of the “free man” provide for a passionate care of independence (ibid.). In a more traditional,

i.e. less encapsulated context, “democracy is almost total, it can come close to anarchy: “ ‘Every man

is a Khan’ (har saray khan day) ” (Ahmed 1980: 94). The idea of equality is most often connected with

the tribal organization of social life. According to Ahmed, an intense spirit of democracy belongs

consistently to the two fundamental principles (besides tarboorwali) that have an impact on Pashtun

social organization. A similar sense of democracy finds ratification in the tribal charter (ibid.: 5). The

confrontation of a traditional tribal setting with an imposed colonial administration demonstrated the

democratic nature of tribal realities. When British occupying forces in the 1930ies were trying to

bolster the growth of an administrative malik (‘petty chiefs’) class in the Tribal Areas, the very core of

tribal democracy was touched. In the administrative structure of the agency, conflicts were created

between mashars and kashars. Hitherto, petty chiefs in the Tribal Areas had more influence than

power in restricted realms of war and counselling, while their authority was more achieved than

ascribed. Yet, the somewhat artificial nomination of administrative maliks who gained a profit from

secret allowances and political privileges created resentment and endangered the political stability.

Even so, the mechanical end on a Durkheimian’s dimension scale of solidarity proved to prevail:

“Nevertheless, the Maliks … remained little more than glorified tourist chiefs. In the interior of the

Agency the weight of their word depended to a great extent on their personal influence. The Agency

remained a close system” (ibid.: 70f.). To sum up, the strive for autonomy appears to be a pervasive

and recurrent theme that guides Pashtun behaviour in conflictive situations. The ideal of equality

(“égalité”) is based on the standards of freedom and autonomy. The one is equal who does not depend

on anyone’s endowments neither as a means to survive nor to protect one’s honour. The status of

freedom is expressed in the capability to supply gifts generously. Quite the reverse, being in the

position of the recipient is far from being desired. Since he provides gifts and offers services, a patron-

like leader embodies the free person. Conversely, his clients do not. Wealth as a means to express

autonomy hence is a helpful if not essential equivalent to independence. Independence is a

supplementary quality to be accepted as a participant in jirga. The community considers the freedom

and autonomy of its members as a commodity to be protected and defended, if imperilled (Centlivres

and Centlivres-Demont 1988: 23). Seen this way, the threat to lose individual autonomy may be

regarded as a collateral damage of one’s nang. Despite an apparently sound picture of an equal


- 46 -

community, the unfamiliar observer perceives fundamental dissimilarity. The quintessence of vertical

stratification may not only be found in the existence of khans, mashars and powerful saints (religious

leaders). The definition of relationships between men and women is in a strong contrast to the ideal of

egalitarianism that guides relationship between males. Additionally, a deeply entrenched principle of

seniority marks the relationship between young men and older people (“white beards”) less in terms of

material matters but with regard to expressions of respect and deference. Besides, the economic

production is based on a particular arrangement of clientelistic networks which brings forth unequal

patron-client relationships (ibid.: 22fff). Finally, one should not refrain from taking into account

hierarchies between collectives. Such is not restricted to inter-ethnic relations (cf. chapter 2.2). Among

Pashtun tribes, a subjective perception of superiority towards other Pashtun tribes prevails. For

example, this collective self-image guides endogamous marriage patterns with Durrani Pashtuns.

Durrani consider themselves as superior to other ethnic or tribal groups, including Durrani women

who are rated even superior to both men and women of other non-Durrani groups (Tapper 1991: 52–

61). The ideal of egalitarianism needs obviously redefinition once crossing tribal borders. Again, we

must be careful in taking the ideal (i.e. being equal) for granted in every-day life which may reveal

divergent social realities. Therefore, the consideration of the empirical context is necessary in order to

make sound judgments upon political matters.

5.4 Pride and the primacy of emotional control

As stated by Lindholm as well as other relevant authors from the beginning of ethnographic

description until today, pride is a recurrent referential term to summarize a desired quality among

Pashtuns: “Every [Swati-Pashtun] […] is filled with the idea of his own dignity and importance…

Their pride appears in the seclusion of their own women, in the gravity of their manners, and in the

high terms they speak of themselves and their tribe, not allowing even the Dooraunees [i.e. the kingly

tribe of Durrani Pashtuns] to be their equals” (Elphinstone (1815) cited in Lindholm 1988: 235). I

assume this conception of dignity to approximate to the set of values attached to nang. But how are

these attitudes of dignity, pride, honour and autonomy expressed in public? If we expect them to have

an effect of collective inculcation, we must as well be able to find a fairly unanimous form among

individuals. The mentioned set of values appears to have a great impact on personal (and political)

success. Can we therefore expect to find people behaving in weakly –i.e. deviant– manners in public,

after all? What are then publicly visual indicators that comply with the rules of dignity? In contrast,

what are pointers that indicate deviant –i.e. weakly– attitudes and should therefore be avoided in

public? To answer those questions, another culturally viable constituent of Pashtun set of values,

relevant also for political actions, must be addressed: mechanisms of emotional control. As an

illustration for the issue, Lindholm made up a game with four Pashtun tribesmen during his fieldwork:


- 47 -

The game was for one person to pretend to feel an emotion and for the others to guess which emotion it was. My [Swati] Pukhtun friends were enthusiastic about this idea, but each emotion they portrayed looked exactly the same: impassive and stoic, with perhaps a slight trembling of the lips. Eventually I realized that the actor was stimulating a feeling inside himself, and then concealing it behind this habitual inexpressive mask. (Lindholm 1988: 232)

Emotions like anger, sadness, happiness, or any other emotions within human repertoire are to be

obscured in public. Even at public occasions, including funerals or celebrations, the expression of

emotions must be suppressed. Boisterous spirits as well as mourning waters are strictly shunned.

Emotional expressions must remain imperturbable. The primacy of emotional control is pervasive and

normative. The control of emotion is consequently a virtue to be impressed persistently on young male

children. Deviant behaviour is punished. The aura of control and self-sufficiency are deemed

appropriate aspects of the ideal Pashtun. Control and restraint are subjects of poetry and proverbs.

Ghani Khan, a contemporary Pashtun writer, still gives faint advices to hide the personal, the private,

hence the weak: “The eye of the dove is lovely, my son. But the sky is made for the hawk. So cover

your dovelike eyes [sic] And grow claws” (Lindholm 1988: 233). This allegory allows to assume that

the existence of a private face is recognised. According to Lindholm, it may be one of anger,

frustration and depression. But the public face is often in marked contrast. Lindholm gives as an

explanation the social circumstances, since in Swat society individual antagonism and rivalry abound

as a result of the classical acephalous patrilineal segmentary organization. Consequently, “men also

must continually present themselves as proud, courageous, and impassive. In this environment,

emotions of depression, fear, jealousy, tenderness or other forms of attachment, are viewed negatively

as displaying an inability to cope and to keep one’s autonomy” (ibid.: 234). Display of weakness

indicate vulnerability. Therefore, expressions of emotions in public are kept in check, since the

evidence of a lack of control allows omnipresent opponents an unwanted advantage (ibid.). In

addition, emotional control essentially is supplemental to the prevailing ideal concepts of person,

among them the dominant nang and aql. What is important if discussing non-material, “soft” qualities

of persons in connection with political leadership is the greater repercussions they have for political

leaders. As we will see, leaders and their actions are traditionally more than anyone else evaluated

whether they fulfil the requirements of pashtunwali. They are more than anyone else required to

comply with the demands of a pure form of the idealized Pashtun person.

5.5 The Pashtun Code, concepts of the ideal Pashtun person and their impact on leadership

The idealized concept of egalitarianism has been previously portrayed. Within the extended family,

disparities manifest along the lines of gender and age while the principles of primogeniture and

seniority are standards. Outside this core of kin group, the cardinal pattern of egalitarianism seems to

be predominant, at least as an emic norm, if not as an observable social fact at any time and place.

This pattern has an impact on the preferred type of leadership. Regardless of the socio-economic


- 48 -

setting, the political leadership of Pashtuns displays informal types of leadership. Traditionally, they

lack the administrative framework and the permanence of bureaucratic authority. As to political

leadership, the fully egalitarian and hence ever contestable position of the mashar (elder, petty chief)

in the nang society may differ from its counterpart in the qalang context to some degree. But even in

the qalang setting, the title of khan (apart from its use for courtesy) does not involve any formal office

(Barth 1977: 214). No questions of succession arise since the status of political leadership is by no

means ascribed but achieved invariably: “Leadership cannot be transmitted and is inherent in the

individual. Tribal society remains democratic to the end. The good and great qualities of a leader are

buried with him in his grave” (Ahmed 1980: 158). The achievement of chiefly status depends on a

variety of factors. Some of them will be addressed below. In a political system such as the one which

seems to prevail among the Pashtuns, the decisions of political actors are rather a result of a sequence

of individual decisions based on free choice and strategic calculations than of e.g. predetermined

alignment. Then, a question inevitably arises: If political actors are structurally unable to rely on

predetermined bonds of dependent adherents and allies: how is leadership created? What are potential

sources of authority that can be manipulated and serve as a means to attract followers so as to gain

political support? A selection of sources of authority will be given in the next chapter. But many

scholars see non-material cultural foundations such as the tribal law, promoted values and selected

features of the idealized Pashtun person as an influential basis from which political authority may

spring. Political leaders “build up ephemeral prestige and power because of exceptional personal

qualities of leadership […] but this is not hereditary and cannot be passed on to their sons” (Ahmed

1980: 141). Subsequently, Ahmed describes the behaviour of tribal leaders with the nang Mohmands,

a hill tribe fitting in the nang model. Together with the precedence to seniority, mashars as the living

senior agnate of a tribal section represent the direct unilineal descent from the tribal eponym. He is

therefore a symbol for and the living link with the ancestors (ibid.: 149). But unlike other resources of

authority, unilineal descent is hardly an influence-raising factor for a mashar. Certainly, descent may

be coupled with the holding of a fairly advantageous share of estate. An offspring may additionally

trace his descent to a direct agnate who distinguished himself with a particularly honourable life or

martial deeds during a jihad, for instance. “Historical capital” of this type might contribute to a better

start among competitors of potential chiefs. Therefore, a shift of the attention towards variables within

the bounds of a leader’s influence is appropriate. Being an elder and an example for the younger men,

a mashar is obliged to comply with the ideal of aql in order to be respected as a political leader:

Mohmands expected in formal questionnaires a leader to ensure peace among his sub-section with

justice and wisdom. Thus, loyalty, reliable caring and helpful assistance to his lineage sub-section or

group is another yardstick for political legitimacy. Furthermore, a leader ideally must display bravery.

Bravery, proven in reputable battles or in the name of jihad, increases his political eligibility.

Furthermore, a mashar symbolizes Pashtunwali, the tribal law and regulatory frame of Pashtun custom

and behaviour. Traditionally, the eligibility of any leader is measured by his compliance with the

norms of pashtunwali. Among the central terms which are conferred by pashtunwali, honour, shame,


- 49 -

counsel and the requirements of hospitality are the most prominent qualities related to plausible assets

of leaders. The requirement of hospitality is dependent on socio-economic conditions. According to

Ahmed’s interpretation based on his survey, hospitality had lower priority in nang areas. Conversely,

in areas which approximate to qalang conditions described by Barth, bounteousness of mashars had

greater importance, defining a mashar to be the one who “helped the poor” (ibid.: 145). Paradoxically,

for the latter, the mere virtue of hospitality came low in priority. For Ahmed, this is indicative for a

deviance from the ideal concept of pashtunwali. In his view, hospitality is reduced to sheer material

assistance in redistributive qalang contexts. It mirrors their subconscious desire for dependence in a

social and environmental setting where stronger stratification and administrative encapsulation are

effective. This may result in stronger leaders. Followers expect from a capable leader to be successful

in resolution of conflicts. In jirgas, a leaders talent is judged by his skills in settling conflicts, in

brokering discords in a wise manner, in short: in displaying in public aql and in taming the turá of the

young. Honour is another major concern for political pretenders. The leader’s ability is not only

measured by his wealth and his actions. Rather, his actions are judged in relation to the ideal. To

defend his honour in cases of insults or as a consequence of impaired namus, is an undisputed standard

for every Pashtun. The ability to defend the honour of his kin wives and to prevent his land and wealth

being threatened is the gauge of his honour. His impeccable honour, in turn, functions as a valuable

currency for his reputation as a capable leader. As a consequence, a leader should never fail to comply

with the code of honour, which may also require acts of lethal revenge, depending on the magnitude of

the insult. Only the show of superior force by the insulted leader does not impair his most effective

non-material currency as an indicator of political ability and success. In the materially prosperous

context of Swat, we might expect such non-material factors to have less importance. But also in Swat,

a chief must command respect by virtue: “Thus the intangible factor of prestige of reputation becomes

a major source of authority, an important means by which a political pretender rallies supporters. The

qualities are evaluated in terms of the polar opposites izat honour, and sharm – shame. The ideal

personality is virile and impetuous, […] always brave” (Barth 1959: 81ff.). Why do personal qualities

weigh so much also in conditions that prevail in Swat, where one would expect a slow disappearing of

the significance for political rallying? Pashtun males in Swat may break their contractual relations at

will with any party, at any time they wish to. As a rule contractual relationships imply political support

for the patron. On the other hand, the situation of a tenant as the contractual party is dependant on the

success of his patron and his ability to defend his property. Being free to terminate a contract, but

depending on the success of his patron, potential political pretenders are constantly evaluating the

anticipated success and the ability of the competing leaders (i.e. patrons). A chief may be wealthy in

inherited land but lose his land to rival chiefs. A poor man depends on the hospitality and assistance of

a chief for his sustenance. But they are free to join an alternative rivalling chief in case of expecting

better success with him (ibid.). For an evaluation, the latest state of currency of a leader’s honour,

prestige and reputation indicator may help him with his decision-making. Another structural aspect

may be supportive in perpetuating the system. Between competing leaders, none of them is


- 50 -

superordinate or subordinate by structure. In conjunction with an egalitarian ethos, this facilitates

shifts and changeability of distribution of political power. In summary, a selection of the classic

ethnographic research from Barth and Ahmed provides enough evidence to suggest non-material

factors such as values and personal qualities to have great significance for the functioning of political

leadership. Whether alike “soft” factors which affect political leadership are fundamentally subject to

change when exposed to modern or foreign influences remains speculative. An appropriate perspective

will be given later. Again, the paucity of contemporary and comprehensive literature on the relevant

topic makes an adequate conclusion more difficult. Furthermore, the importance of some features

appear to vary according to the underlying socio-economic setting.


- 51 -

6 Forms of political leadership and the state

As particularized earlier, the share of power in Afghanistan has mostly been based on Pashtun

domination throughout Afghan history. The situation is similar in Pakistani tribal territories where

Pashtuns make up the majority. Seeing that attempts of extensive pervasion of the Afghan central

authority on regional level has failed repeatedly, the political integration of Pashtun society has proved

to be a balancing act. In spite of intense integrative efforts during the reign of Amir Amanullah Khan

(1919–1928), the incorporation of local and regional tribal power on state level remained limited.

Major efforts to deprive traditional elites of power fell short and recurrently culminated in rebellions.

This is partly due to scarce economic and military resources of the state, but to a certain extent also a

result of an inherent proclivity for political independence in Pashtun society. In addition, a high degree

of legitimacy of Pashtun leaders on regional level contribute to the obduracy towards non-tribal

political determination. Thus, owing to historical experience, the state and its representatives have

been undeniably dependent on the cooperation and clemency of traditional Pashtun leadership. Since

then, the state has strictly avoided to infringe upon local politics and regional sovereignty. Claims of

state rights of sovereignty such as imposition of taxes, a rudimentary administration and jurisdiction

has been limited. The strategic policy was confined to a comprehensive local and regional

participation as well as to a non-violation policy that tolerated regional sovereignty. As a consequence,

the state did hardly interfere in the political actions of regional and local representatives. As a

counterpart, local leaders were essentially precluded from influencing state politics based on tacit

consent. The dichotomy between traditional political elite on regional and local level and political

representatives on state level is hence a determining shaping factor of Afghan politics (Rasuly 1997:

101ff.). The latter has been mostly backed by a traditional power structure of the Durrani Pashtun clan.

During more than two centuries, with only a few short periods of interruption, its political leaders

reigned the country with the mentioned federal restrictions on the basis of a monarchy-like claim of

power. One might deem this phenomenon to be more an Afghan specificity. But a look at the north-

western frontier region of Pakistan suggests an alternative interpretation towards a Pashtun special

case. In British India a century ago, the Pashtuns in the north-west frontier region in Swat (Pakistan)

were administratively acknowledged as a ‘tribal territory’, i.e. an area in which extensive local

autonomy was effective. The Swat interpretation of local autonomy included working out their own

political problems in any manner they chose. The Indian Government (in pre-Pakistan times) did not

interfere in political matters nor did they tolerate external interference from the dynastic centre of

Afghanistan (Barth 1959: 8f.). Here again the fear of colonial administration in India towards Pashtun

political amalgamation, that later anticipated Pakistan’s concern, becomes apparent. After an era of

decolonization and after violent series of turmoil that led to the foundation of the Islamic state of

Pakistan, this statutory autonomy has been modified, but actual conditions still approximate closely to

the ideal of local autonomy within borders of Pashtun dominated ‘tribal territory’ (ibid.). For the goal

of this paper, a short outline of the Pakistan policy in terms of the tribal areas of the north-western


- 52 -

frontier provinces must bring up (a) the invalidity of criminal or civil procedure codes of Pakistan; (b)

a wide-ranging exemption from payments of taxes or rents of any kind, (c) the existence of only a

loose establishment of a political administration whose main purpose is to ensure general tribal peace

so as to guarantee the functioning of infrastructural installations with great significance for the

Pakistan state; (d) the work of political parties or modern politics has been prevented by the original

tribal treaties (Ahmed 1980: 10f.). Still today, the Pakistani Pashtun tribesmen in the tribal nang area

are not subject to national laws in a large part of their own territories. This fact backs the concept of a

keen sense of independence to be a Pashtun peculiarity. Based on the previous evidence, my

argumentation will therefore promote the idea that political matters in the Pashtun-dominated areas

within state borders are strongly rooted in tribal politics. In addition, Pashtun tribal politics is deeply

marked by actions of political leaders. This suggestion has been repeatedly insinuated so far. A more

elaborated theoretical foundation will be given later. As a result, an understanding of Pashtun politics

and its effect even within state borders requires an analysis of political actors as well as their

significance, acting and legitimizing within tribal structures and values. Generally, the traditional

political elite can be subdivided into two large groups: on the one hand a political elite with secular

orientation that pursue mainly political aims and comprises local leaders presiding over their tribal

affiliation (lineage, clan); on the other hand a religious elite with great religious significance for the

people but at times aspiring political power, too (ibid.). Mohamed Omar, a mullah of the Pashtun

Hotak Ghilzai tribe near Kandahar and the political as well as religious leader of the Taliban (Rashid

2000: 65) is a popular example for the latter. This fact shows what a religious office-holder with minor

formal significance can achieve in a social context where Islam has great plausibility in times of

crises.

6.1 Leadership: Traditional political elite

The implication of the term ‘traditional’ as a descriptive notion to label political systems is

problematical. As described earlier, the problem is reminiscent of Ahmed’s objections towards a

generalizing notion of Pashtun social organization: “The absence of a general societal typology has led

authoritative writers to take nang values for granted while analysing empirically observed qalang

society” (1980: 116). Following his clarifications, I tried to unravel the maze of societal typologies

and their potential of conceptual aberration in two ways: First, since this paper also deals with the

possible effects of traditional political systems on contemporary political life within a framework of a

state, I give a deliberate preference to structural components and behavioural traits that are connected

with a situation of at least partial encapsulation (i.e. the Swat expample). In history, the process of

partial structural integration was constantly moderating cultural features of a pure ‘nang’ society

(Ahmed 1980: 92fff.). Second, central features of Pashtun culture such as the set of tribal laws and

values based on pashtunwali are persistent also in an encapsulated situation, even if only

idiomatically, or as an emically idealized concept. Furthermore, a qalang society wishes “to live up to


- 53 -

the nang ideal but finding it difficult to do so in an encapsulated situation” (ibid: 118). Ahmed has

substantiated in his thesis, that central features of pashtunwali can survive and yet be prominent

despite encapsulation, though a strict application of some features has been rejected or reinterpreted.

Still, the self-confession to comply with Pashtun values in qalang society proves at least their emic

significance. With reference to protagonists of political leadership, the discussion has a bearing

whenever an analysis of political organization is carried out: a high degree of centralization in an

environment which allows material accumulation and thus a source of power is confronted with an

acephalous tribal system with no central authority; the former reveals a considerable power of khans,

backed by coercive authority which is in turn nurtured by the accumulation and access to material

resources. The latter limits authority to the genealogical position of senior agnates. A political leader

has power in a qalang situation, while in nang societies the effect of political legitimation is influence.

Nevertheless, the assumption of the persistence of an invariable set of features that affects political life

of Pashtun society should allow to give an account of political leadership. After all, the model-

building in social science always remains an image of the reality and therefore hardly an uncritical

reflection of vivid social reality itself, since only defining a concept does not make it a social reality

(Boissevain 1974: 171). For that reason, description of social and political organization in a very

complex environment leaves space for alternative interpretations and empirical realities. In

Afghanistan as well as in Pakistan, multifaceted relations that had a wide range of impact blurred

social boundaries and political spheres. We find a great variety of political organizations that do not

warrant simplistic descriptions. Besides, the Pashtun region has been a steady hot spot rimmed with

playgrounds for international power politics.

Unlike other tribal societies, Pashtun politics puts barely forth positions of uncontested leadership.

Rasuly detects a higher degree of concurrence among potential leaders that strive for power and

influence in comparison with other ethnic groups in Afghanistan whose political structures were the

target of suppressive efforts on the part of the state (1997: 105fff.). This is a result of a non-

interference policy of Afghan central authorities towards Pashtun tribes. Henceforth, a partial

autonomy hampered a socio-political transformation of Pashtun tribal structure. For that reason, I

consider the social anthropological literature to have great plausibility in describing political practice,

since it conventionally allows for traditional political organization adequately. In a most general

assessment, Pashtun politics is a reflection of Pashtun society, its value orientation and social

organization. In principle, it is based on a strong patriarchal predisposition. Authority lies in the hands

of elder men. The ideal-type of a household consists of several generations and forms a residential

unity. More important for political issues is the function of the family with its wide genealogical

ramifications as a solidarity group (ibid: 107). The leader and hence political representative of any

sectional level is almost invariably the living senior agnate. On household level, the family elder

assumes this role. They are representatives and speakers of a group, may it be a household, a village or

a kin-group, e.g. a lineage. A respective local leader is termed mashar (or more formally málak) and

implies ‘petty chief’ or ‘headman’ (ibid; Ahmed 1980: 141ff.). Khan is another term and serves as a


- 54 -

title for political leaders further up the scale of power, compared to mashars. However, the title does

not legitimize a formalized position of authority in a hierarchical system. Rather, a khan’s claim to

authority over others is restricted to the leader’s willingness to lead (Barth 1959: 74). In the nang

context, the deep aversion to authority and titles implying political power comes to light in the distaste

for the label khan. In terms of political power, the discussion about leadership in social, political and

anthropological science can be demonstrated exemplary along the fuzzy lines of power in the Pashtun

context. The classic dichotomy of analytical categories differentiates between leadership on the one

end of the continuum and chiefdoms on the other end of the relevant measure. In short, the former

entails achieved status, temporariness, transient and situational power, personal qualities as well as

relational instability and brittleness of the political followers. The latter category implies ascribed

status, permanency, office and inherent conditions attached to it as well as a certain relational stability

of the adherents (Luque 1996: 142fff.). In the Pashtun context neither reality meets the requirements

of chiefdoms. Though some restrictions are appropriate, Pashtuns are typically a non-hierarchical

society based on an ethos of tribal egalitarianism. Traditionally, neither an ascribed status nor a

permanency in office can be asserted according to the literature. If adding variables of genealogical

sacredness and heredity of power, none of the described political systems may be pigeonholed under

the chiefdom category (Luque 1996: 146; Keesing 1981: 290). As a consequence, a manifestation of

power among Pashtuns complies largely with the conceptual prerequisites of leadership. The nang

Pashtuns reject emphatically any form of structural inequality, since “no ‘chiefly model’ has evolved

or is possible in the Tribal [i.e. nang] Areas. Elders yes, petty chiefs for temporary periods possibly,

but chiefs: no. By some fortune or traits of personality an individual may exceed another but none is

by right superordinate and none by structure subordinate, all are as good in the sight of man and by

their own lights” (Ahmed 1980: 146). But this is also true for Pashtun tribes under qalang conditions.

When a western researcher asked in a hujra (village guest-house), who was the khan and how he is

customarily elected, the gathering answered unanimously: “Every man is a khan” (ibid.). Generally, in

a traditional context, Pashtun descent, wealth and a set of personal qualities were imperative

prerequisites for the assumption of political leadership. In Afghanistan, other qualities turned out to be

more relevant and successful in the course of the war. Other types of actors superseded traditional

leaders with direct bonds to local politics and custom. Some effects of the change on the arena of

politics through the preceding decades will shortly be discussed below.

6.2 Political leadership in nang society

In accordance with their aversion to hierarchy and titles and their preference to equality, tribal nang

societies like the Mohmand tribes in the tribal territory in Pakistan dislike the formal reference of

malik for ‘petty-chiefs’ (Ahmed 1980: 142). Tribal Area Pashtuns or “Pathans of the hills" arrange

their socio-political life along egalitarian lines: “The ‘Malik’ is seldom more than their leader in war

and their agent in dealings with others; he possesses influence rather than power[!]; and the real


- 55 -

authority rests with the “jirgah”, a democratic council composed of all the “Maliks” “ (Ahmed 1976:

74). In nang society, deviant behaviour is more confronted with tribal and customary laws. They are

rooted in the set of codes that is provided by pashtunwali owing to the absence of any other law. The

picture that emerges from the socio-political illustration of Ahmed is one of acephalous and

segmentary tribal groups, where traditional tribal customs and codes have much more impact on

political life (ibid: 74f.). One might theorize that this is –apart from the situation of administrative

autonomy– also a consequence of a cultural premise of egalitarianism which is in turn rooted in the

socio-economic environment: “Tribal life is supported by primitive modes of production and equitable

forms of distribution. Equality is thus inherent […] and acts through circular causation with existing

socio-political factors to reinforce itself” (ibid.). The mere absence of state control, the lack of

juridical and administrative encapsulation as well as a scarce economic basis together with the strive

for equality between small kin units prevent the creation of powerful leaders. The formation of

powerful leadership requires an economic environment which allows accumulation of resources. In

accordance with an acephalous political system and more informal leadership, resources of power

differ from those exploited “in the plains” (qalang societies). Features of the idealized Pashtun person

are key symbols that stand for prestige and social mobility. In a poor environment, non-material

properties become a fundamental pool of resources in order to establish influence, not power. Nang,

for example, develops into a pivotal concept and a standard of reference in understanding socio-

political interaction and the formation of informal leadership: “In the face of severe economic poverty,

personal valour, marksmanship and skills in combat become the symbols that confer nang and

therefore status. For a different and older age-set wisdom and propriety achieve honour” (Ahmed

1976: 76). The listed set of personal traits, which embodies personal valour is partly identical to the

introduced concepts that centre upon the idealized Pashtun person. They are reminiscent of

characteristics such as nang, turá and aql.

6.3 Political leadership in qalang society

In Swat, male elders are the representatives of a Pashtun household. On a higher level of a segmentary

section, a dominant khan is as well the legitimate representative, i.e. the political leader of a sub-

lineage, of a lineage or of a clan, respectively. Pashtun households in Swat are based on their estate

property. Swat society is divided into social strata. Those qawms (social groups) are also referred to as

castes. They constitute patrilineal, hereditary, ranked occupational, endogamous groups. Membership

is determined by birth. The Pashtuns as the conquerors of the region in the 16th century belong as a

rule to the caste of landowning tribesmen (Barth 1959: 16fff.). Being neglectful in giving an all-

encompassing account of the social framework of organization in Swat, only a few words should

suffice to gain an overview of qalang political make-up. In the function of a landowner, the khan is the

patron of non-Pashtun tenants or landless Pashtuns as well as small Pashtun landowners. Consistent

with the interpretation of Barth, the dyadic relationship does not involve any asymmetrical inequality


- 56 -

that would necessarily engender dependence. But even Barth concedes the superiority of the person

who gives the contract. Therefore, the “employee”, i.e. contract-taker, is economically in a weaker

bargaining position. It does however not imply aspects of political dominance and submission. But

still, mutual relations are regulated by a series of individual contracts. Not only agricultural labourer as

tenants, but also professional specialists such as craftsmen are integrated within a contractual

relationship. Most of those contracts are thus by definition purely economic in character (Barth 1977:

212). What is central for the political realm –and there is the effective difference to classical patron-

client dyads– is that most of such contractual relationships have no lasting political implications. That

is, contracts can be suspended any time. By inference, the ones who establish contracts with a Pashtun

landowner, is at the same time regarded as political followers in an implicit agreement. The services of

the followers are bought by gifts and promises. In return, the service of a leader is expected to be

compensated by political support. If necessary support is refused, a leader is able to exercise control

by threatening to withdraw benefits. A follower can expect to obtain rewards in the form of

advantages through the success of his leader (Barth 1977: 214ff.). In this respect, he is dependant. But

since the follower may at least theoretically switch sides so as to support another –allegedly more

successful– leader, he is more dependant of the leader’s success than structurally of the leader itself.

As a consequence, followers seek the leaders from whom they anticipate to have the greatest success

which results in advantages for themselves. As a principle, this system is structurally very instable.

One might expect a considerable political “turnover” in number and frequency. As a result, “the

position of a leader is thus never secure; his following may swell or shrink almost without warning.

Since leaders are permanently in competition, the sources of their authority are most clearly exhibited

in situations of conflict” (ibid.). Leaders appear to be permanently in competition. They attempt to

expand their estate property, their wealth and field of political influence. The arena where personal

superiority and power can be exhibited most effectively is the showdown events of conflicts. A

specific deduction becomes tempting: Is it then mere speculation to suggest that the high tension of

public violence is also due to leader’s active search for the public arena where one’s capability,

influence and strength can be proven in conflictive competition? As a preventive means of losing

adherents, the public conflicts seen in the light of advancing power, may even be interpreted as a

legitimate structural part of the political organization in Swat. The two-bloc system in Swat may be a

mitigating factor that complicates defection from one political faction to another. Of course, some

restrictions regarding the assumed stability of a lineage-based two-bloc system are needed. Some

arguments have been presented in an previous chapter. A further variable prevents followers from

switching sides easily with the purpose of entering into an agreement with a leader who promises

better prospects. A critical reinterpretation of Barth’s analysis of Swat Pashtun evidence must stress

the unequal distribution of the control of resources as well as the means of production. From this point

of view, the distinguished landowning power-holders are maintaining their power within a system

based on exploitative economic relationships with dependant landless tenants (Asad 1972). That is, in

an environment where estate property is not only vital but also a scarce resource, the fear of loss may


- 57 -

hesitate a tenant before leaving the safe haven. With house tenancy contracts, practical reasons also

impede an easy “contractual defection”. Ultimately, not only a structural-functional approach which

overemphasizes systemic stability, but also the concept of individual free will and choice should be

balanced by an inclusive class analysis. Emphasizing individual strategies and motives in search of

maximizing advantages or power is likely to hide structural obstructions which preclude theoretical

free choice from action.

6.4 Sources of authority

In a situation of inherent structural instability of the political system. like in the case of Yusufzai

Pashtuns in Swat, the sources of influence become the centre of interest. Given that authority and

superiority can only be demonstrated through the sources of power, land, wealth, honour and martial

qualities become essential issues.

–Land: In Swat, estate property constitutes the basic assets. Compared to other property, land is

associated with the claim of high status. The only access for non-landowners to it is through

contractual agreements with landowners. For the leader, land is the most important source to gain

contractual partners. Since only through a considerable number of contractual relationships, a leader

can gain political supporters. The number of contractual and political adherents is in turn a public

yardstick for a leader’s success. In rallying support, we can possibly expect the mechanism of social

comparison to be also effective. Nevertheless, the landowner certainly holds the whip hand and exerts

his dominance, since “the tenant depends on the land for his livelihood, and the power of eviction

gives the landowner a hold on him” (Barth 1977: 215). Political influence is thus directly correlated

with the ownership of land. Increasing one’s land holdings enhances a leader’s chances to gain

political influence: “The competition between chiefs is thus largely for the control of land, and the

acquisition of land is an important move in a political ascent” (ibid.). How can property of land be

expanded? There are mainly four possibilities: (1) land can be inherited. Brothers share equal inherited

estates; (2) one may be rewarded with land for political loyalty; (3) purchase; (4) acquisition by

violence or threats. The last option is extensively used, since there is often a confusion owing to

conflicting claims, or because of disputes relating to the course of land borders. For instance, a

powerful landlord attempts to encroach on the land of his weaker neighbours. Smaller landowners may

be forced to transfer partial claims to his more influential neighbour in exchange for his support and

protection against another powerful leader who in turn tries to seize his entire estate (Barth 1977: 217).

–Wealth: Through the well-directed manipulation of his wealth, a chief attempts to strengthen

his position. He tries to assure the loyalty of his adherents and clients in several ways: by bribes,

payments, gifts and hospitality. The unilateral gift-giving does not explicitly entail an obligation for

the receiver to respond to the command of the giver. But it is an expression of difference in status.

–Hospitality: Hospitality is a direct function of the manipulation of a leader’s wealth. Commensality

generates solidarity. It is a concealed form of a distributive economic system. It may be expressed in


- 58 -

terms of food, material support or generous assistance on cost-intensive occasions such as weddings.

For political matters, hospitality appears to be the most important and least detrimental form of giving:

“Bribes and payments create relationships which render them onerous or hazardous. Gifts and

hospitality, on the other hand, are of prime importance in the building up of a political following”

(ibid.). The recipient is highly dependent on the continuation of the gifts he receives. Therefore,

hospitality is a potent measure in order to control others. A continuous flow of gifts fosters

dependence. As a consequence, it reinforces the bond of contractual relationships, which can be

mobilized for a leader’s political actions (Barth 1959: 79f.). In this sense, the Pashtun leader’s gift

practice is reminiscent of the institution of Northwestern American potlatch, where the public display

of lavishness is used to gain social prestige (Ahmed 1980).

–The men’s house: In connection with the importance of hospitality, the guest-house of the leader

offers to his adherents a space for public meeting and mutual exchange. It is also an opportunity to

gather his following so as to show the strength of support publicly as well as to demonstrate his

generosity.

–“Key or soft qualities”: As illustrated before, a capable leader must comply with admired personal

qualities of the idealized Pashtun person. He is required to meet with the conditions of pashtunwali.

Furthermore, he must constantly provide evidence for his superiority and personal abilities. That is, he

ought to protect his honour. In debates such councils (jirga), he must demonstrate his wisdom (aql).

As head and commander of local lashkars (tribal armed gathering for battles), a chief must also prove

martial qualities: a brave character and a determined willingness to defend his honour and property at

all costs. This way he keeps his contenders from the temptation to challenge him. As Lindholm has

demonstrated, emotional control is another attribute. As ambitious competitors are present

everywhere, the expression of weakness must always be concealed. The conditions and corollary of a

generous behaviour and its functionality in gathering support has been illustrated in detail by now.

In terms of importance, the share of single resources within the whole variety of possible “stakes” may

subject to change. As will be claimed in the conclusive chapter, the economic role of the whole range

of official or unofficial actors from several states has made the Afghanistan war an international

conflict. This may influence the type of leadership qualitatively as well as quantitatively with regards

to the type of material resource involved. A political economy of war may entail a pragmatic

economization as well as a militarization of a leader’s activity:

The pursuit of politics through both peaceful and violent means requires money. Just as in many parts of the world political power is a principal means to the pursuit of wealth, war too may create conditions for economic activity, though often of a predatory nature. Political leaders speak in public about their ideas and goals, but much of their daily activity is devoted to raising the resources to exercise power and reward supporters or themselves. How political leaders raise and distribute these resources often determines the outcome of their acts, as much as if not more than their stated goals and intentions. (Rubin 1999)


- 59 -

The warlordism debate may bring out more clearly the characteristics of change Afghanistan has

undergone for a couple of decades. Many scholars suggest that Pashtun leaders who base their

authority on features of traditional legitimacy have replaced by a modern type of leader. As a

consequence of an increasing economization of conflicts, a warlord-type elite dominated the political

landscape.

6.5 Traditional religious elite

Within the area taken under study, roughly four religious elites can be discerned: (1) the ulema with its

Islamic dignitary and scholars (Alim, Malauwi); (2) the pirs, i.e. the spiritual leaders of different Sufi

orders (tariqats); (3) the holy saints (the sayyeds with the Pashtuns); (4) the mullahs on the local level.

As a rule, mullahs do not have a paramount significance in politics. But particularly in Pashtun areas,

single charismatic mullahs played an important part over and over again in history. Apart from

isolated incidents, the political influence of mullahs do rarely extend beyond the local level of politics

and religious instruction (Rasuly 1997: 119ff.). Under the British colonial “rule”, proven political

authorities such as mashars or khans were superseded by religious leaders such as mullahs and

sayyeds. When signs of political trouble emerged, mullahs took the initiative. They became political

leaders highlighting religious themes (Ahmed 1980: 70). The sayyeds are of distinctive character,

since they claim special status which is derived from their holy descent of Pashtun ancestor. In some

areas they exert considerable political influence, while in Swat, their main source of authority is land.

Their role as mediators in conflicts has greater importance (Barth 1959: 92–103). In general, the role

of religious leaders must be viewed in the context of an Islamic society. In Islamic societies like in

Afghanistan, the influence of Islamic dignitaries the influence exceeds the religious sphere. Rather, in

educational as well as in juridical areas, religious representatives have immense influence (Rasuly

1997: 120f.). The main importance of religious dignitaries who become substantial political leaders in

certain contexts can be confined to three functions:

(1) They are mediators. In times of deepening political cleavages between single segments of society

or e.g. between the Pashtun central authority and local segmentary collectives, religious figures

mediate between the parties at odds (ibid.).

(2) In periods of crisis and threats, religious leaders have recurrently taken the lead. If Afghanistan or

Pashtun tribal homeland is threatened by foreign intruders (the British, the Russians), the importance

of the religious elite as political leaders rose significantly (ibid.).

(3) Furthermore, they are contenders of political leaders, too. But unlike other political representatives

such as Pashtun khans, their influence is not restricted to tribal or ethnic boundaries. Rather, they

attempt to appeal to the Islamic community in general in order to gain influence, regardless of tribal

identity. Even more, they are interested in breaking confined ethnic ties. This is instrumental for other

political players, too. For instance, in the last decades Pakistan supported deliberately religious


- 60 -

representatives only to prevent the Pashtun community from adopting a strategy of reunification (cf.

“The Pashtunistan issue”: chapter 2.2).

Not only the religious elite attempts to stress the shariat networks. Many secular political leaders

pursuing an extremist Islamic policy try to address both the pan-Islamic identity as well as the ethnic

identity; a double-cross strategy according to the intention and the audience. A recent example is the

Pashtun warlord Gulbudin Hikmatyar. For the Islamic clergy, traditional tribal affiliation had less

importance. Therefore, they tried to supersede the tribal divisions. In recent times, they stress bonds of

networks provided by Islamic parties. During the 1970s, young Islamist intellectuals, such as

Hikmatyar, attempted to root themselves in new political patterns, such as affiliation to an Islamist

political party. But in the south, young intellectuals had more difficulty to be successful. There, the

new leadership consisted of mullahs who gained legitimacy through their prominent religious position

(Roy 1989: 73).


- 61 -

7 The theory of person-centred politics

Shahrani (2000) sees person-centred politics as “the cornerstone of kin-based mode of Pashtun tribal

social and political organization [...] has been the defining attribute of Afghan politics since the

creation of Pushtun-dominated centralized polity” by assuming the Pashtun political leader to have a

pivotal role in political processes. Quoting Eric Wolf, Shahrani points to the Achilles heels of kin-

based politics in general, that a leader must attract following through judicious management of

alliances and redistributive action. But a leader in a tribal context reaches a natural limit to expand his

influence, as his scope of potential authority is contained within borders of the kinship order. To

surpass the limitations of kin-based politics a leader is forced to gain independent access to reliable

and renewable resources of his own. As illustrated above, such resources may be of material,

monetary, personal or ideological i.e. religious nature. In Afghanistan, the latter has been successfully

activated in order to overcome tribal constraints and kin-based political economy with considerable

success in the beginning, as waging jihads against non-Muslim invaders or the initial triumph of the

Taliban movement have shown. Another historical example is a form of internal colonialism imposed

on non-Pashtun communities gives evidence for this strategy. Also extensive foreign subsidies from

the wider region have brought a resource into play that allowed to expand power of politically

dominant groups around ambitious leaders. Still, as Shahrani speculates, the “effectiveness of these

strategies, however, has proved to be episodic and transient (ibid.)”. The tribal political culture is

anchored in person-centred politics. This is exactly, where Shahrani spots the source of the failure to

shift from such person-centred politics “to a broader, more inclusive, participatory national politics

based on the development of modern national institutions and ideologies” (ibid.). As a result,

fratricidal hostilities, violent confrontation and wars of succession on local as well as on national level

together with holy wars against foreign powers have devastated the whole country. It is not merely

speculative, that incessant incidents of internal conflicts have facilitated aggressive foreign

interventions. Foreign powers turned into actively involved “stake-holders” of the conflicts since they

were able to benefit from them by playing off parties against each other. Internal conflicts were also

constantly covered up with the rhetoric of religious or ideological justifications. Consequently, without

compromising his theorizing, Shahrani resumes the central function of leaders entangled in their

traditional setting. He denies any implication of ideological or institutional causes. Instead, the

conflicts “were fought for or against specific individuals, families or clans out of personal, but often

rapidly shifting, commoditicized loyalties” (ibid.). Another term, employed by Banfield (1958) within

a Mediterranean peasant setting describes best, what is insinuated here: an “amoral familism” that puts

an individual’s responsibility primarily towards his or her family (Barnard and Spencer 1996: 212).

Thus, if loyalty to the immediate kin-group as a determining ethos for political actors take precedence

over corporate obligations, we can expect a similar dynamics to be even much more effective and

significant in a tribal setting, where kin-based mobilisation for political matters has structural

advantages. It can exploit expanded primordial ties which are embedded in a cultural system of


- 62 -

retributive responsibilities. This is especially the case in remote areas that resemble to nang

conditions. Conversely, the tribal variant and thus expansion of the concept of “amoral familism”

employed for the Pashtun context is not fully compatible with Barth’s concept of free choice. In

contrary to the appearance among other tribal contexts, where “one has the impression, that political

allegiance is not a matter of individual choice. Each individual is born into a particular structural

position, and will accordingly give his political allegiance to a particular group or office-holder. In

Swat, persons find their place in the political order through a series of choices, many of which are

temporary or revocable” (Barth 1959: 2). Consequently, the application of an alternative notion of

amoral familism needs further conceptual clarification and extension. What segments does a similar

force of social cohesion as a means for political mobilization encompass? Is only a household with its

co-residents included, or is a similar concept to comprise a wider network and deeper genealogy of

kinship? If it only consists of the former, as Barth’s remarks insinuate, is it still adequate to speak of a

tribal society as a kin-based political unit? The resumption of the discussion that centres around the

core concept of leadership may help to interpret apparent inconsistencies. At least, the concept of

amoral familism matches methodically with Barth’s observations in its actor-centred logics of

methodological individualism. Both explanations are based on an actor’s tendency to maximize

material advantage of the core kin group, assuming that all others will do likewise. In addition, Barth’s

observations lay stress on the paramount significance of political leaders (1959: 71). This emphasis is

in accord with the introduced assessment of Shahrani. In the context of Swat Pashtuns, people were

unable to indicate any simple principle for the recruitment of corporate units for political purposes.

When they refer to such groups, phrases like “my descent group” or “my association” are absent both

in two-bloc systems and within associated groups indirectly linked by contractual ties with the same

leader. The activities of groups are strictly seen in terms of the actions of their leader. As a

consequence, conflicts have a personalized face : “the reference is always to ‘the party of so and so’.

[…] From Pathan descriptions of conflicts, one might think they were duels (Barth 1959: 71)”. A

historical extract of Pashtun history of invasion of the upper Swat region illustrates the strong focus on

a political leader in holding together a corporate group. There were four Khans, among them Taj

Mohammed Khan. They invited the Nawab of Dir, an adjacent region, to invade upper Swat. The other

party that invaded and later ruled Swat together with the former was Darmei Khan in collaboration

with the Badshah (later the founder of Swat state). The brother’s son of Taj Mohammed Khan rebelled

against his uncle once the Nawab had been driven out and thus weakened Taj Mohammed Khan.

Subsequently, Taj Mohammed Khan’s nephew switched side and joined the Badshah. The four old

Khans fled to Dir. After they returned temporarely to Swat, their rifles were confiscated, their movable

property taken. Unlike any of the four old Khans, their followers returned to Swat, while a large part

of them did not even joined the flight of their leaders in the first instance. The number of followers and

combatants amounted to thousands, so did the number of confiscated arms (Barth 1977: 213). This

incidents show some of the most crucial properties of a recurrent pattern with political conflicts, when

political groups for corporate actions are involved in violent disputes:


- 63 -

First, an extension of the principle of agnatic rivalry as an expression of tarboorwali. This principle

describes a latent sphere hostility regarding male agnates. It opposes collateral patrilineal relatives

against each other, especially parallel cousins jostling for power (cf. chapter 4.2). The ambiguous

relationship between agnatic rivalry reveals on the one hand the instrumental character of kin relations

as a means to achieve collectively political goals. On the other hand, it provides an opportunity to

surpass, outclass and finally eliminate his tarboor.

Second, the survival of a corporate group for concerted action is dependent on a leader. A corporate

group should by definition persist in perpetuity, independently of the life and death of individuals. The

leaving or the decease of the leader creates a crisis. The conflicts over succession to leadership often

disband a group (ibid.). The mobilisation, existence and purposeful action of a corporate group for

political action accordingly appears to function only in connection with a leader. Again, the principle

of person-centred politics is supported.

Third, in qalang societies such as Swat as well as in broad river valleys like the region around

Qandahar (tribal homeland of the Barakzais or the Popalzais), we can suggest a comparable ideal

pattern of political structure. Powerful leaders compete for scarce resources based on land. Land and

wealth enables them to enter into contractual agreements with tenants and other service providers,

mostly landless or petty Pashtun landowners or non-Pashtun clients of other ethnic identity. In a

concealed redistributive system, the leader provides in turn services, material assistance and

protection. Due to the asymmetry of the contractual relationship, dependence cannot be fully denied.

In return for his services, the Pashtun khan expects the follower to be supportive in his political

actions. Political actions include the active maintenance of a leader’s power, possibly the extension of

his influence through the expansion of land and accumulation of his wealth. His contenders are other

ambitious Pashtun khans. Since also followers, contractually committed, attempt to maximize their

advantages, they are constantly evaluating the influence and success of their leader and their

challengers. Theoretically, a switch of sides may improve his situations. The process of decision-

making is based on a principle of rational and free choice, on either side, the khan’s as well as his

adherent’s. As a consequence, khans might permanently feel forced to demonstrate their ability.

Appropriate opportunities of exhibition of strength and power is on the one hand the triumph in

conflicts over their competitors, on the other hand the successful settlement of conflicts between a

leader’s adherents in order to prevent fissions.

A critical application of Barth’s analysis of Swat Pashtuns together with the explanatory completion

by Shahrani’s theory of person-centred politics may serve as an appropriate instrument in order to

interpret striking phenomena of the political life among Pashtun tribal groups. Observable facts of

Pashtun political organization based on atomized, close-kin oriented and person-centred politics

comprise the following characteristics:


- 64 -

(1) Person-centred politics makes the political structure highly dependant on the goals and the

success of single leaders, such as tribal khans, religious elites or military commanders. The

existence of a corporate political group is subject to the success/failure of its leader.

(2) A tendency towards factionalism of political groups. The tribal structure may serve as an

explanation for factional tendencies, though not a sufficient reason.

(3) A structural instability manifest in frequent changes of sides of political actors. This is a

consequence of a leader-centred politics. Furthermore, flexible loyalties beyond ties of close

kins as well as a strong inclination to favour short-run advantages for one’s immediate family

reinforce the effects of shifting loyalties.

(4) Paternalistic politics encourages commoditization of loyalties. That is, a quasi redistributive

economic system fosters the creation of political loyalties as a means of economic

relationship.

(5) A strong need for egalitarianism, at least among Pashtun tribal members prevents tribal groups

from forming persistent and stable political mergers, since it would require hierarchical

political structures.

(6) A high amount and frequency of public conflicts often ending up in violent and lethal clashes.

Personal feuds, competition and enmities to be carried out publicly add to the endurance of

public violence.

(7) A segmentary tribal system of high complexity oscillating between two diametrical poles:

agnatic rivalry between patrilateral cousins and their followers on the one end and a strong

preference to kin groups for corporate actions against tribal outsiders at the other end.

(8) National and socio-political ideologies as well as moral principles serve as to protect personal,

familial, tribal or ethnic group honour. The preservation of self-interest is prevailing. “This

has resulted in serious discrepancies between public policy pronouncements of the contending

groups and their actual practices” (Shahrani 2000). Ideology is often instrumentalized for

political goals.


- 65 -

8 Critical remarks and contemporary contextualization

The optimistic assumptions or rather insinuations made in this paper resume an extrapolation of Barth.

Namely the daring assertion that salient core features of political organization is not only applicable on

lower levels of local politics, but also characterize an isomorphic organizational framework of wider

political alignments (Barth 1959: 104). I would also like to join in the moaning of some social

anthropologists that precious contributions of their field are largely ignored when it comes to an

explanation of striking contemporary phenomena, such as prominent conflicts. In the public and

academic discussion, political analysts, political scientists and military strategists occasionally armed

with geo-strategic glasses are dominant since they constantly appear to take the lead (cf. Edwards

1998: 714). Rather often one may gain the impression that at times a somewhat simplified perspective

takes the lead. It is habitually marked by the western ideal of a fairly uniform nation-state. But

particularly in Central Asia, nations are often fractioned patchworks of different ethnic groups which

are de facto held together either by autocratic regimes or, more rarely, by explicit policies that allow

quasi autonomous regions. The latter arrangement applies widely to Afghanistan and western Pakistan.

In assessing the Afghan state of affairs, one is inclined to disregard the long history of societal and

ethnic fractures that have been lasted for centuries. The conquest and partial dominance of one ethnic

group (the Pashtuns) as well as their role in a nation-state has avoided dreadful clashes to an even

larger extent. But this argument is only true, if it is completed by another two facts: First, the Pashtun

dominance in Afghanistan has mostly been concentrated on the southern and eastern part.

Furthermore, the central state of Afghanistan since it had been founded has been restricted to a limited

core centre (first Qandahar, later Kabul). Then, the Pashtuns themselves were highly fragmented

according to their tribal segmentation. Besides, the “Pashtun” central government has often been

comprised of a distinguished elite of Durrani tribal members. Other tribal or ethnic groups have been

incorporated by tacit approval of the government in exchange for far-reaching tribal and ethnic

autonomy. Second, external imperial threats have again and again merged together political forces

under the common umbrella of a unified Muslim Umma. Following Asad (1972) and Ahmed (1980),

an analysis of political frameworks and conflicts must definitely include a historical dimension. That

is, even the evidently structural features are subject to change. In addition, and against a structural-

functionalist approach, societies are far from being closed systems. And this is also my central

argument in embedding “traditional” cultural aspects in contemporary incidents: external influences

must unconditionally be considered, in spite of alleged persistence of cultural conventions. I tried to

give a wider focus on the distinguished role of the Pashtuns within frames of a nation-state. Their

prominent position is a result of a privileged treatment and administrative and military incorporation

into the colonial empire during the Persian occupation. Subsequently, the Pashtuns were constantly

taking advantage of their military skills and their organizational superiority through periods of

conquest towards north and east, but also during limited successes in repulsing the empires of the

Persian as well the Indian Mughals in the east. The settlers in the conquered territories had been


- 66 -

subjugated while a Pashtun landownership was established for the largest part. This is especially

verified for today’s north-western part of Pakistan, where the dominated non-Pashtun population was

either driven out or mostly adopted the role of dependant tenants in a caste-like system. In

Afghanistan, the dominance of the Pashtuns was culminating in the formation of a state, in which the

tribal confederation of the Durrani (formerly Abadali) played a key role in representing the central

state. But this state functioned to a large extent within a system that assured the tribal regions and their

leaders of extensive autonomy. This also included the traditional 1lashkars remaining untouched.

Thus, the state’s monopoly on the use of force has been persistently denied. It remained largely in the

hand of local tribal leaders. Together with their numerical strength, the historical experience of

Pashtun dominion may explain the claim for Pashtuns to assume a crucial role in governmental key

departments. Moreover, it explains to a certain degree inter-ethnic relationships in Afghanistan and

north-western Pakistan. In the framework of a wider context, I attempted to address selected key

features that are alleged to contribute to political organization of the Pashtun tribal society.

Pashtunwali, the tribal law, is of major relevance for almost all aspects of life. With regard to political

matters, a leader’s eligibility is also measured by his compliance with standards which are implied in

the tribal law. Directly linked to pashtunwali are specific characteristics of a concept that centres

around the idealized Pashtun person. It includes a distinctive set of personal qualities. Among the

most important characteristics are the publicly perceivable willingness to defend one’s honour that

revolves around an intact reputation of the closest female members of one’s kin. Besides, honour is

associated with consequent retaliatory actions if namus face threats to be insulted or violated. Other

personal qualities comprise the complementary pair of aql (wisdom) on the one and turá (martial

bravery) on the other pole. Emotional constraint is a consequence of those characteristics, since it

conceals weakness. Weakness is likely to be a seen as a gateway through which challenges of honour

may cross. As a standard, the ability of a political leader is measured by the degree of his personal

qualities. But the emphasis differs according to the role political leaders represent: “Khans, whose

support requires a reputation for self-assertiveness and ruthless defense of their interests, […] saints

will have established a reputation for moderation, piety, indifference to physical pleasure” (Edwards

1998: 715). The fulfilment of the mentioned collection of non-material requirements is especially

important for leaders against a background of a highly competitive environment in which political

leaders are competing for influence. The wide-spread, essential and very plausible ethos of

egalitarianism is congruent with the tribal system that stresses the value of individual equality and

fosters an aversion against symbols of inter-tribal disparity. Ideally, non-material characteristics are

seen as currencies that indicate ability of a leader. Apart from non-material dispositions, leadership is

mainly based on land and wealth. Land enables political leaders to contract tenants and professional

service-providers. Since the numerical increase of contractual clients expands equally the number of

1 Lashkars are tribal armed gatherings. They ensure tribal security and the binding decisions based on jirga arbitration being imposed or respective violations being penalized. Furthermore, they may function as tribal armed forces at the serve of the matters of a political leader.


- 67 -

political followers and his reputation as influential and successful leader, he attempts to attract as

much clients and political adherents as he can. Where land is scarce, income low and where unchecked

public violence abounds, clients are partially dependant not only on a chief’s hospitability and gift-

giving, but also on his assistance in conflicts. From the point of view of a leader, an implicit

redistributive system ties economically deprived clients for political matters. However, if we adopt

Barth’s model of actor-centred analysis of political behaviour, the position of a leader is unstable and

threatened at any time. Actors, leaders and clients alike, are free to choose their support based on

rational choice, that is constantly seeking opportunities in order to maximize their advantage. In

contractual relationships, they are theoretically free to switch sides in case an alternative leader

promises more advantage, since the success of a political leader reverts to his clients. Observable

material wealth coupled with a non-material set of personal qualities are indicative to him so as to

reach a decision. Accordingly, political leaders are in return relentless to demonstrate their skills,

success and superiority. This is best proved in conflictive events. Conflicts may therefore be seen as

functional in this game for influence. A critical remark must however be added: Barth’s model of free

choice as an underlying framework has been criticized repeatedly. Bearing in mind the social

stratification and increasing economic disparities, the objection of whether free choice can be

exercised under those conditions, is adequate.

Far from being homogeneous, the Pashtun tribal community is neither socio-economically uniform nor

are the above-made assumptions made above fully applicable to each standard of collective living.

According to the classical literature, two socio-economic categories arise. Mainly based on Akbar S.

Ahmed (1980), Pashtuns can be largely divideded into a nang category and a qalang category. The

former referring to a geographically remote, an ecologically disadvantaged, an economically deprived

and a socio-politically more traditional situation. They appear to be culturally more homogeneous as

well as marked by a lower degree of state encapsulation. Qalang matches up with the antithetical set

of socio-economic description, quite inclining to the other end of the continuum. In respect of political

organization, within nang context, the strive after egalitarianism is stronger, factions align more with

kin-based lineage groups. The concentration and accumulation of wealth is limited. Thus, political

power does not allow the same concentration as with qalang societies. Furthermore, a higher

homogeneity reduces inequality and results in a smaller number of potentially dependant clients.

Traditional institutions such as pashtunwali, and hence the practice of agnatic rivalry and prescriptive

revenge in specific cases seem to be of greater importance compared to qalang. For that reason,

caution is appropriate in generalizing findings derived from a defined cultural area. Nevertheless, one

might conjecture, that the often suggested factionalism and political instability is at least to a

considerable extent attributable to cultural practice that is inherent in the traditional political set-up. It

has been illustrated previously in fuller detail. This is but one misleading interpretation, if not

completed by counter-acting forces. As described in chapter 1.2, Pashtuns may also merge as a

corporate group for political action. Tribal merger on a higher segmentary level seems to be functional

as a reaction to threat from the outside. If challenged by other ethnic groups or by colonial endeavours,


- 68 -

if autonomy, independence or the namus of one’s country is at stake, the merger of tribal

confederations is purposeful. However, if the source of the immediate threat has disappeared, the

temporary and goal-oriented tribal merger is most likely to fracture. History provides several

evidences about this matter. In contrast to other regions of the world, the Pashtunistan issue provides

substantiation for the ineptitude in forming an administrative and political unit of Pashtuns. History

presents another lesson. Consistent with the assumption about actor’s decisions based on free choice,

calculation and objective-oriented rationale given above, Pashtun politics generally appears to be

pragmatic. We find such pragmatism in wider historical context. In the seventies, when the Marxist

Khalq party in Afghanistan tried to condemn the large landowners for their fortune while glorifying

peasants as heroic, this political rhetoric was largely rejected. Whatever stratified the social structure,

in a country where a large population is still living in rural areas, rhetoric that is ideological in nature

and denounces people’s leaders, is given no odds (Edwards 1998; 717). May pragmatism, opportunism

and leader-centred actions be added as basic constants to Pashtun political culture? It seems that above

all, Islam with its popular plausibility, is able to mobilize extensive political support in periods of

crisis. One may argue that most of the suggestions made here are based on ethnographic data that have

become obsolete if addressing contemporary events. This is certainly true to some extent, even if the

persistence of cultural traits are underestimated at times. But let me demonstrate my last argument on

the basis of current events. Warlords and commanders replaced largely traditional leaders such as

mashars and traditional tribal khans, whose significance has constantly been diminishing during the

last decades.

Status differences between warlords and commanders were reflected most dramatically in their vehicles […as well as in] the number of bodyguards […]. They simply followed the lead of their commander, and they told me that if at any point their patron decided to join a different political party they would do likewise without any hesitation. While the willingness of these men to switch parties might appear to be evidence of their loyalty to their leaders, it appeared to me more the result of necessity in that a strong commander provided security and opportunities that would otherwise be hard to come by. (ibid.: 722)

The new political elite that has been crystallized during the last violent decades has changed the

importance of traditional values and norms (Shahrani 1997: 162ff.). Less wealth and descent but

alternative qualities like military, organizational and logistic competencies enabled a leader to become

an eligible and legitimate leader. My suggestion is that such a development not only has its basis in a

conflictive environment but also draws on strategic considerations of supranational “big regional

players” in Central Asia. Neighbouring countries and hegemonies have been increasingly sponsoring

ambitious political groups with military abilities. Apart from drug trade as an important source of

income, large amounts of money and a large stock of weapons from abroad made it possible for the

new political elite such as warlords and commanders to expand their influence. They were able to

replace traditional leaders by having the demanded material and redefined non-material resources at

their disposal. We might conjecture that such a development is powerful enough to facilitate a

transformation towards a pronounced qalang type of society where the dependence on leaders is


- 69 -

increasing as well as correlative with the concentration of power. External influences on such a large

scale will likely to weaken traditional political organization and their representative elites. A new type

of leaders may have been arisen while redefining the basis of political eligibility. I suggest that the

available set of cultural standards is catalytic in this process. In this respect, Pashtun culture might be

supportive with its predisposition in favour of pragmatic, profit-maximizing decisions, person-centred

politics and structural factionalism as well as its approval of distinctive personal qualities that foster

martial, honour-related and retribution-seeking characteristics. An analysis of cultural change should

never ignore the fact that socio-economic and political systems are far from being closed ones.

Historical as well as external factors and actors are to be considered. An analysis that is able to address

ways of cultural entrenchment of new leaders against a background of periods of crisis is required.

How do tragic and lasting violent conflicts change the political arena and its cultural basis? Finally, an

analytical framework in an attempt to define whether the popular and ubiquitously used label

“warlord” is necessary. For instance: under what circumstances and at what stage is a leader turning

into a so called warlord? Does he belong to a distinctive category of leader in violent times when the

decay of regular political organization leaves a vacuum of control to be filled? In any case, the type of

a internationally involved political economy of war (and peace) in Afghanistan presumably finds its

corollary in a changing map of political arenas with altering forms of leadership. Related effects on a

tradition-guided pre-war society with its cultural conventionalities of leadership should not be

underrated. They are apparent in an increasing concentration of power which is maintained by large

amounts of raised physical capital as well as their adequate military resources in a violent

environment. In a de facto redistributive system, where asymmetrical relationships of dependency

intensify, the repercussions on national politics may result in an increasing regionalism and hence a

lack of national political cohesion. This may ensue from the creation of powerful local leaders who

aspire after the consolidation of their power against similarly potent adversaries. However, in exactly

what way the underlying cultural framework serves as a reinforcing matrix for this development in a

violent context, or whether it is a rather vanishing factor, or why it looks as if to be so vulnerable to

the effects previously introduced is an open question to be answered in further investigation that

requires greater empirical foundation.

Khans & Warlords in Pashtun Society PART 1

Documents