Land, People and History


Introduction: Understanding Pakistan

Eppur si muove

(And yet it moves)

(Galileo Galilei)

There have been times during the writing of this book when it seemed that it would have to be titled ‘Requiem for a Country’. At the time of writing, the pressures on Pakistan from without and within are unprecedented even in its troubled history. Yet such despair would be premature. Tariq Ali wrote Can Pakistan Survive? The Death of a State in 1983, a generation ago. That’s quite a long deathbed scene by any standards.1

It is possible that the terrible floods of the summer of 2010 have fundamentally changed and weakened the Pakistani system described in this book. This, however, will not be clear for a long while – and in the meantime it is worth remembering the extraordinary resilience that South Asian rural societies have often shown in the face of natural disaster, from which they have repeatedly emerged with structures of local authority and political culture essentially unchanged.

What is certainly true is that if floods and other ecological disasters on this scale become regular events as a result of climate change, then Pakistan will be destroyed as a state and an organised society – but so too will many other countries around the world. Indeed, such a development would reduce present concerns about Pakistan to relative insignificance. In the meantime, however, the floods have obviously damaged Pakistan’s national infrastructure, and retarded still further the country’s already faltering economic progress.

This book is intended to describe and analyse both Pakistan’s internal problems and the sources of Pakistan’s internal resilience. In consequence, it of course deals extensively with the threat from the Pakistani Taleban and their allies, the roots of their support, and the relationship of this support to the war in Afghanistan. It also examines the policies of the Pakistani security establishment towards Afghanistan and India, since these have had very important effects on domestic developments in Pakistan. It is not meant, however, to be a study of Pakistan’s international position, though the conclusions contain some recommendations for Western policy.

Trying to understand Pakistan’s internal structures and dynamics is complicated; for if there is one phrase which defines many aspects of Pakistan and is the central theme of this book, it is ‘Janus-faced’: in other words, many of the same features of Pakistan’s state and government which are responsible for holding Islamist extremism in check are at one and the same time responsible for holding back Pakistan’s social, economic and political development.

Pakistan is divided, disorganized, economically backward, corrupt, violent, unjust, often savagely oppressive towards the poor and women, and home to extremely dangerous forms of extremism and terrorism – ‘and yet it moves’, and is in many ways surprisingly tough and resilient as a state and a society. It is also not quite as unequal as it looks from outside.

Pakistan contains islands of successful modernity, and of excellent administration – not that many, but enough to help keep the country trundling along: a few impressive modern industries; some fine motorways; a university in Lahore, parts of which are the best of their kind in South Asia; a powerful, well-trained and well-disciplined army; and in every generation, a number of efficient, honest and devoted public servants. The military and police commanders of the fight against the Taleban in the Pathan areas whom I met in Peshawar and Rawalpindi in 2008 – 9 struck me as highly able and patriotic men by any standards in the world.

The National Finance Commission Award of 2010, which rebalanced state revenues in favour of the poorer provinces, was a reasonable if belated agreement. It demonstrated that Pakistani democracy, the Pakistani political process and Pakistani federalism retain a measure of vitality, flexibility and the ability to compromise. None of these things is characteristic of truly failed or failing states like Somalia, Afghanistan or the Congo.

That doesn’t mean that Pakistan always smells nice (though sometimes it does); and indeed, some of the toughest creepers holding the rotten tree of the Pakistani system together are at one and the same time parasites on that tree, and sometimes smell bad even by their own standards. Nonetheless, tough they are; and unless the USA, India, or both together invade Pakistan and thereby precipitate its disintegration, the likelihood is that the country will hold together, and that if it eventually collapses, it will be not Islamist extremism but climate change – an especially grim threat in the whole of South Asia – that finishes it off.

Support for extremist and terrorist groups is scattered throughout Pakistani society, but as of 2010 mass support for Islamist rebellion against the Pakistani state is so far present only in the Pathan areas, and in only some of them – in other words, less than 5 per cent of the population. That is not remotely enough to revolutionize Pakistan as a whole. During their rule over the region, the British faced repeated revolts in the Pathan areas, without seriously fearing that this would lead to rebellion elsewhere in their Indian empire.

Any Pakistani national revolution would have to gain not just mass but majority support in Pakistan’s two great urban centres, Lahore and Karachi; and as the chapters on Punjab and Sindh will make clear, this is unlikely for the foreseeable future – though not necessarily for ever, especially if ecological crisis floods the cities with masses of starving peasants.

When terrorist groups attack India, or Western forces in Afghanistan, their actions enjoy a degree of instinctive, gut sympathy from a majority of Pakistanis – not because of Islamist extremism, but because of Muslim nationalism and bitter hostility to the US role in the Muslim world in general and Pakistan’s region in particular. Support for a civil war and revolution in Pakistan itself that would turn Pakistan into a revolutionary Islamic state is, however, a very different matter from sympathizing with attacks on the US and India. That would mean Pakistanis killing Pakistanis on a massive scale, and by and large they don’t want to. They may well want to kill some set of immediate rivals, but that’s another matter.

It is important not to be misled by the spread of terrorism in Pakistan in 2009 – 10. In many ways, terrorism by the Pakistani Taleban is a sign not of strength but of weakness. If you want to overthrow and capture a state, you need either a mass movement on city streets that seizes institutions, or a guerrilla movement in the countryside that seizes territory, or a revolt of the junior ranks of the military, or some combination of all three. No movement relying chiefly on terrorism has ever overthrown a state. The Pakistani Taleban looked truly menacing when it took over most of the Federally Administered Tribal Areas (FATA), followed by the districts of Swat and Buner. When it blows up ordinary people in bazaars and mosques, it merely looks foul.

Pakistan is thus probably still far from the situation of Iran in the late 1970s or Russia in 1917. Apart from anything else, the army is a united and disciplined institution, and as long as that remains the case, it will be strong enough to defeat open revolt – as it proved by defeating the Taleban in Swat and south Waziristan in 2009. Unlike in Africa and elsewhere, military coups in Pakistan have always been carried out by the army as a whole, on the orders of its chief of staff and commanding generals – never by junior officers. As my chapter on the military will describe, there are very deep reasons for this in terms of material advantage as well as military culture.

The only thing that can destroy this discipline and unity is if enough Pakistani soldiers are faced with moral and emotional pressures powerful enough to crack their discipline, and that would mean very powerful pressures indeed. In fact, they would have to be put in a position where their duty to defend Pakistan and their conscience and honour as Muslims clashed directly with their obedience to their commanders.

As far as I can see, the only thing that could bring that about as far as the army as a whole is concerned (rather than just some of its Pathan elements) is if the US were to invade part of Pakistan, and the army command failed to give orders to resist this. Already, the perceived subservience of the Pakistani state to Washington’s demands has caused severe problems of morale in the armed forces. I have been told by soldiers of all ranks that faced with open incursions on the ground by US troops, parts of the Pakistani army would mutiny in order to fight the invaders. With the army splintered and radicalized, Islamist upheaval and the collapse of the state would indeed be all too likely – but even then, the result would be rebellions leading to civil war, not, as in Iran, to a national revolution that would be successful in taking over the whole country.

I hope through this book to strengthen the argument that however great the provocation, the US must not contribute to the destruction of Pakistan – even though, as this book will make clear, neither the Pakistani army, nor the Pakistani state, nor the great majority of Pakistanis will ever give more than very qualified help to the US campaign against the Afghan Taleban, since Pakistanis of every rank and class see these in a quite different light from Al Qaeda and the Pakistani Taleban.


Pakistan is quite simply far more important to the region, the West and the world than is Afghanistan: a statement which is a matter not of sentiment but of mathematics. With more than 180 million people, Pakistan has nearly six times the population of Afghanistan (or Iraq), twice the population of Iran, and almost two-thirds the population of the entire Arab world put together. Pakistan has a large diaspora in Britain (and therefore in the EU), some of whom have joined the Islamist extremists and carried out terrorist attacks against Britain.

The help of the Pakistani intelligence services to Britain has been absolutely vital to identifying the links of these potential terrorists to groups in Pakistan, and to preventing more attacks on Britain, the USA and Europe. Pakistan therefore has been only a partial ally in the ‘war on terror’ – but still a vital and irreplaceable one. For we need to remember that in the end it is only legitimate Muslim governments and security services that can control terrorist plots on their soil. Western pressure may be necessary to push them in the right direction, but we need to be careful that this pressure does not become so overwhelming that it undermines or even destroys those governments, by humiliating them in the eyes of their own people.

Finally, Pakistan possesses nuclear weapons and one of the most powerful armies in Asia. This means that the option of the US attacking Pakistan with ground forces in order to force it to put pressure on the Afghan Taleban simply does not exist – as both the Pentagon and the Pakistani military have long understood. Deeply unsatisfying though this has been for the West, the only means of influencing Pakistan has been through economic incentives and the threat of their withdrawal. Economic sanctions are not really a credible threat, because the economic collapse of Pakistan would play straight into the hands of the Taleban and Al Qaeda.

Pakistan’s relationship with India has been central to Pakistan’s behaviour since 9/11 – as of course it has been ever since partition and independence in 1947. Fear of India has both encouraged and limited Pakistani help to the US in Afghanistan. This fear is exaggerated, but not irrational, and neither are most of the policies which result from it. On the one hand, fear of a US – Indian alliance against Pakistan seems to have been a genuine factor in Musharraf’s decision to help the US after 9/11, and was certainly used by him to convince the military and – initially – many ordinary Pakistanis of the necessity of this help. On the other hand, fear of India has been both a reason and an excuse for Pakistan not to redeploy more troops from the eastern border with India to fight against the Taleban in the west.

Lastly, the Pakistani establishment long cherished the hope that it could use Pakistani help against the Taleban to bargain for US pressure on India to reach a settlement with Pakistan over Kashmir. This hope has faded with the refusal (compounded of unwillingness and inability) of both the Bush and the Obama administrations to play such a role; but this refusal, and America’s ‘tilt towards India’, have added greatly to longstanding Pakistani feelings of betrayal by the US.

Pakistan’s help to the West against the Afghan Taleban would, however, have been limited in any case both by strategic calculation and mass sentiment. In terms of mass sentiment, the overwhelming majority of Pakistanis – including the communities from which most Pakistani soldiers are drawn – see the Afghan Taleban as engaged in a legitimate war of resistance against foreign occupation, analogous to the Mujahidin war against Soviet occupation in the 1980s.

In terms of strategy, the Pakistani establishment’s approach to Afghanistan has long been driven by a mixture of fear and ambition. The fear is above all of Afghanistan, under the rule of the non-Pashtun nationalities, becoming an Indian client state, leading to India’s strategic encirclement of Pakistan. This fear has been increased by a well-founded belief that India is supporting Pakistan’s Baloch nationalist rebels via Afghanistan, and by what seems by contrast to be a purely paranoid conviction that India is also supporting the Pakistani Taleban.

The greater part of the Pakistani establishment therefore believes that it needs to maintain close relations with the Afghan Taleban, since they are Pakistan’s only potential allies in Afghanistan. In recent years, belief in the need for a relationship with the Taleban has been strengthened by the growing conviction that the West is going to fail in Afghanistan, and will eventually withdraw, leaving anarchy and civil war behind – just as occurred after the Soviet withdrawal and the fall of the Communist regime from 1989 – 92. In the resulting civil war, it is believed, every regional state will have its own allies – and so must Pakistan.

Incidentally, it is worth pointing out that even entirely secular members of the Pakistani establishment do not see the Afghan Taleban as morally worse than the Taleban’s old enemies in the Afghan Northern Alliance leaders, with whom the West has in effect been allied since 2001. Their atrocities and rapes in the 1990s helped cement Pathan support for the Taleban. They massacred Taleban prisoners and looted Western aid after the overthrow of the Taleban in 2001, and their role in the heroin trade has helped destroy any hope of the West curtailing that trade since 9/11.

Equally, it is important to note that in the great majority of cases, both in the elites and in the mass of the population, this sympathy or support for the Afghan Taleban does not imply ideological approval, or any desire that Pakistan should experience a Taleban-style revolution – any more than support for the Mujahidin in the 1980s implied much liking for them.

Hence, too, the great difference in Pakistani attitudes to the Afghan and to the Pakistani Taleban. There was never a chance that the Pakistani establishment and army were going to let the Pakistani Taleban conquer Pakistan. The long delay in fighting them seriously was because they were not generally regarded as a serious threat to Pakistan, but were seen as a local Pathan rebellion which could be contained by a mixture of force and negotiation; because many ordinary Pakistanis (including soldiers) saw them as misguided but nonetheless decent people dedicated to helping the good jihad in Afghanistan; because there was deep opposition to the state engaging in a Pakistani civil war for the sake of what were seen to be American interests – especially among all sections of Pakistan’s Pathan population; and, finally, because the Pakistani military and its intelligence services were deeply entwined with jihadi groups which they had sponsored to fight against India in Kashmir, and which were in turn entwined with the Pakistani Taleban.

As soon as the Pakistani Taleban were seen by the establishment to be a really serious threat to the central Pakistani state, in the spring of 2009, the army, with the backing of the PPP-led government and much of the establishment in general, took strong action to drive them back. The army’s victories over the Pakistani Taleban in Swat and South Waziristan have settled the question of whether Pakistan will survive the Pakistani Taleban’s assault (barring once again an attack by the US or full-scale war with India). They have, however, settled nothing when it comes to the question of the army’s willingness to fight hard against the Afghan Taleban for the sake of a Western victory in Afghanistan.


Failing a catastrophic overspill of the war in Afghanistan, Pakistan will therefore probably survive as a state. The destruction of united Pakistan and the separation of Bangladesh in 1971 are often cited as possible precedents for the future disintegration of today’s Pakistan; but this is quite wrong. No freak of history like united Pakistan, its two ethnically and culturally very different wings separated by 1,000 miles of hostile India, could possibly have lasted for long, quite apart from the immense cultural and linguistic differences between the two halves. The tragedy is not that it failed, but that a situation made for a civilized divorce should instead have ended in horrible bloodshed.

West Pakistan by contrast is far more of a natural unity in every way, with a degree of common history and ethnic intertwining stretching back long before British rule. Pakistan in its present shape has already survived considerably longer without Bangladesh (thirty-eight years) than the original united Pakistan managed (twenty-four years).

It is true that ‘Pakistan’ as a name is a wholly artificial construct, invented by Rehmat Ali, an Indian Muslim student in Britain in 1933, to describe a future Muslim state in the north-west of the then British empire of India embracing Punjabis, Pathans, Kashmiris, Sindhis and the peoples of Balochistan, different parts of which names make up the word Pakistan. ‘Pak’ in turn means ‘pure’ in Urdu, and so Pakistan was to be ‘The Land of the Pure’.

In the imagination of the coiners of this name, there was no thought of including Muslim East Bengal in this state, so Bengalis had no part in the name; another sign of how completely improbable and impractical was the attempt in 1947 to create a viable state out of two pieces 1,000 miles apart. Certainly most of the Punjabis and Pathans who dominate West Pakistan never really thought of the East Bengalis as fellow countrymen or even true Muslims, shared much British racial contempt for them, and contrasted their alleged passivity with the supposedly virile qualities of the ethnicities dubbed by the British as ‘martial’, the Punjabis and Pathans.

The official language of Pakistan is native to neither of its old halves. Urdu – related to ‘Horde’, from the Turkic-Persian word for a military camp – started as the military dialect of the Muslim armies of the Indian subcontinent in the Middle Ages, a mixture of local Hindustani with Persian and Turkic words. It was never spoken by Muslims in Bengal – but then it has never been spoken by most of the people of what is now Pakistan either. It was the language of Muslims in the heartland of the old Mughal empire, centred on the cities of Delhi, Agra, Lucknow, Bhopal and Hyderabad, deep in what is now India. Urdu is the official language of Pakistan, the language of the state education system, of the national newspapers, and of the film industry; but the only people who speak it at home are the Mohajirs, people who migrated from India after partition in 1947, and who make up only 7 per cent of Pakistan’s population.

However, what is now Pakistan is not nearly such an artificial construct as the old Pakistan of 1947 – 71. It has a geographical unity which in some respects is thousands of years old, being basically the valley of the River Indus plus neighbouring mountains, deserts and swamps. To a much greater extent than most post-colonial states, Pakistan therefore has a core geographical unity and logic.2 Moreover, most of Pakistan’s different ethnicities have lived alongside each other for millennia, have been Muslim for hundreds of years, and have often been ruled by the same Muslim dynasties.

Regional identity may be growing in political importance, with the 2008 elections showing a lower vote for the PPP in Punjab, and a lower vote for the Punjab-based Muslim League in other provinces. All the same, with Pakistanis, there is usually a wheel within a wheel, an identity within an identity, which in turn overlaps with another identity. The only exceptions, the people with a single identity, are some of the Islamists, and some of the soldiers – but by no means all of either. Or as Ali Hassan, a young Lahori executive with a Norwegian company, said to me:

If I were to jump on a box and preach revolution, with the best programme in the world, you know what would happen? First, people from all the other provinces would say that we can’t follow him, he’s a Punjabi. Then most of the Punjabis would say, we can’t follow him, he’s a Jat. Then the Jats would say, we can’t follow him, he’s from such-and-such a biradiri. Even in my own village, half the people would say something like, I can’t follow him, his grandfather beat my uncle in a fight over land. If you preach Islamic revolution, most Pakistanis won’t follow you because they practise different kinds of Islam and worship different saints. So you see we Pakistanis can’t unite behind a revolution because we can’t unite behind anything.3

Or, in the saying common in Pakistan as across the Greater Middle East: ‘I against my brother, I and my brother against our cousins, and our family against our biradiri and our biradiri against other biradiri.’ Occasionally they end, ‘and Pakistan against the world’ – but not often.

Not surprisingly then, until recently at least, every attempt to unite large numbers of Pakistanis behind a religious, an ethnic or a political cause ended in the groups concerned being transformed by the everpresent tendency to political kinship and its incestuous sister, the hunt for state patronage. This wooed them away from radicalism to participation in the Pakistani political system, which revolves around patronage – something that is true under both military and civilian governments in Pakistan.


Indeed, a central theme of this book is that the difference between civilian and military regimes in Pakistan is far less than both Western and Pakistani analysts have suggested. A fundamental political fact about Pakistan is that the state, whoever claims to lead it, is weak, and society in its various forms is immensely strong. Anyone or any group with the slightest power in society uses it among other things to plunder the state for patronage and favours, and to turn to their advantage the workings of the law and the bureaucracy. Hence the astonishing fact that barely 1 per cent of the population pays income tax, and the wealthiest landowners in the country pay no direct taxes at all. As a state auditor in Peshawar said to me with a demoralized giggle, ‘If anyone took taxes seriously, I’d have the most difficult job in the world, but as it is I have the easiest.’

The weakness of the state goes far beyond a dependence on patronage for the survival of governments. To an extent most Westerners would find hard to grasp, the lack of state services means that much of the time, the state as such – as an agent with its own independent will – does not necessarily affect many people’s lives very much, either in terms of benefits or oppressions. The presence of policemen, judges and officials may make it look as if the state is present, but much of the time these people are actually working – and sometimes killing – on their own account, or at the behest of whoever has the most power, influence and money at a certain point, in a certain place.

The nineteenth-century British colonial official Sir Thomas Metcalf described the traditional villages of northern India as ‘little republics’, administering their own justice, deciding their own affairs, and paying only what tribute to the ‘state’ could be extracted from them by force. This independence has been very greatly reduced over the years, but compared to any Western society a good deal of it still exists in many areas, if not specifically in the village, then in local society generally. Society is strong above all in the form of the kinship networks which are by far the most important foci of most people’s loyalty.

For those readers who are really interested and have a few brain cells to lose, a brief description of the horribly complex subject of kinship terms and groups in Pakistan is appended to the end of this introduction. Suffice it to say here that the language of kinship – even among people who are not in fact related – permeates most of Pakistan as it does most of South Asia, whether it is a matter of affection, responsibility, asking for favours or asking for protection. The most wonderful expression of this, which perfectly sums up India’s mixture of kinship, democracy and hierarchy, is the term with which you may wish to address a relatively menial person in northern India who happens to be in a position to help or harm you (like a bus-conductor): Bhai-sahib, or ‘Brother-Lord’.

Kinship is central to the weakness of the Pakistani state, but also to its stability, above all because of its relationship with class. Because the Pakistani political elites, especially in the countryside, rely for their strength not just on wealth but on their leadership of clans or kinship networks, kinship plays a vital part in maintaining the dominance of the ‘feudal’ elites and many of the urban bosses.

By helping to enforce on the elites a certain degree of responsibility for their followers, and circulating patronage downwards, kinship also plays a role in softening – to a limited extent – class domination. Kinship is therefore partially responsible for Pakistan’s surprisingly low rating of social inequality according to the Gini Co-efficient, which I will discuss further in Chapter 6.4 In both these ways, kinship is a critical anti-revolutionary force, whether the revolution is of a socialist or Islamist variety.

The importance of kinship is rooted in a sense (which runs along a spectrum from very strong to very weak depending on circumstance and degree of kinship) of collective solidarity for interest and defence. This interest involves not just the pursuit of concrete advantage, but is also inextricably bound up with powerful feelings of collective honour or prestige (izzat) and shame; and, indeed, a kinship group which is seen as ‘dishonoured’ will find its interests decline in every other way. A sense of collective honour among kin is thus reflected most dramatically in preventing or punishing any illicit sexual behaviour by the kinship group’s women, but also in working to advance the political and economic power and public status of the group.

This is a cultural system so strong that it can persuade a father to kill a much-loved daughter, not even for having an affair or becoming pregnant, but for marrying outside her kinship group without permission. You don’t get stronger than that. As Alison Shaw and others have noted, the immense strength and flexibility of the kinship system in Pakistan (and most of India too) are shown, among other things, by the way in which it has survived more than half a century of transplantation to the very different climes of Britain. Shaw writes:

Families in Oxford are therefore best seen as outposts of families in Pakistan whose members have been dispersed by labour migration ... [In Britain] a distinctive pattern of living near close kin has emerged, echoing that of earlier migrations within the Indian subcontinent.5

Defence of the honour and the interests of the kinship group usually outweighs loyalty to a party, to the state, or to any code of professional ethics, not only for ordinary Pakistanis, but for most politicians and officials. It is important to understand therefore that much Pakistani corruption is the result not of a lack of values (as it is usually seen in the West) but of the positive and ancient value of loyalty to family and clan.

Since the kinship group is the most important force in society, the power of kinship is inevitably reflected in the political system. Just as in much of the rest of South Asia, a majority of Pakistan’s political parties are dynastic. The PPP is the party of the Bhutto family; the PML(N) is that of the Sharif family; and the Awami National Party (ANP) in the Frontier is the party of the Wali Khan family.

The local political groupings which are the building blocks of these parties are themselves based on local dynasties. Hence the phenomenon of a woman such as Benazir Bhutto rising to the top of the political system in an extremely conservative male-dominated society. This was power by inheritance, and says not much more about ordinary women’s rights in modern Pakistani society than the inheritance of the throne by Queens Mary and Elizabeth from their father said about ordinary women’s rights in sixteenth-century English society.

The only institution which has succeeded to some extent in resisting this in the name of state loyalty and professional meritocracy is the army – and you could say that it has managed this in part only through turning itself into a kind of giant clan, serving its members’ collective interests at the expense of the state and society, and underpinned to some extent by ancient local kinship groups among the north-western Punjabis and Pathans.

If the importance of kinship links has survived transplantation to the cities of Britain, it is not surprising that it has survived migration to the cities of Pakistan, especially because in both cases (the British through marriage with people from home villages in Pakistan) the urban populations are continually being swelled and replenished by new migration from the countryside. The continued importance of kinship is a key reason why Pakistan’s tremendous rate of urbanization in recent years has not yet led to radical changes in political culture, except – for reasons I will explain – in Karachi.

Largely because of the strength of kinship loyalty, Pakistani society is probably strong enough to prevent any attempt to change it radically through Islamist revolution, which is all to the good; but this is only the other face of something which is not so good, which is society’s ability to frustrate even the best-designed and best-intentioned attempts at reform and positive development. Key factors in this regard are the gentry in the countryside and the intertwined clans of business, political and criminal bosses in the towns, all of them maintaining continuity over the years through intermarriage, often within the extended family, almost always (except among the highest elites) within the wider kinship group.

Marriage with members of the same kinship group, and when possible of the same extended family, is explicitly intended to maintain the strength, solidarity and reliability of these groups against dilution by outsiders. Shaw writes of the Pakistanis of Oxford that in the year 2000, almost fifty years after they first started arriving in Britain, there had been barely any increase in the proportion of marriages with non-kin, and that over the previous fifteen years 59 per cent – 59 per cent! – of marriages had been with first cousins; and the proportion in strongly Pakistani cities such as Bradford is even higher:

Greater wealth was perceived not solely in terms of individual social mobility, although it provides opportunities for this, but in terms of raising the status of a group of kin in relation to their wider biradiri and neighbours in Pakistan ... Status derives not only from wealth, mainly in terms of property and business, but also from respectability (primarily expressed by an ashraf [‘noble’] lifestyle). One element of being considered a man worthy of respect derives from having a reputation as being someone who honours his obligations to kin. Cousin marriage is one of the most important expressions of this obligation. The majority of east Oxford families have not achieved social mobility and status though massive accumulation of property and business. For them, the marriages of their children to the children of siblings in Pakistan is an important symbol of honour and respectability, a public statement that even families separated by continents recognize their mutual obligations.6

It is above all thanks to locally dominant kinship groups that over the years, beneath the froth and spray of Pakistani politics, the underlying currents of Pakistani political life have until recently been so remarkably stable. Civilian governments have come and gone with bewildering rapidity, whether overthrown by military coups or stranded by the constantly shifting allegiances of their political supporters. Yet the same people have gone on running these parties, and leading the same people or kinds of people at local level. The same has been true under military governments. None of these changes of government or regime has produced any real change to the deeper structures of Pakistani politics, because these are rooted in groups and allegiances which so far have changed with glacial slowness.


In the countryside, and to a great extent in the towns as well, the most powerful elements are not individuals – though they are led by individuals – or even families in the Western sense. From this point of view, as from many others, the description of the rural landowners as ‘feudals’ is false, in so far as it suggests any close comparison with their medieval European equivalents. If this were so, the Pakistani ‘feudals’ would long since have been swept away by pressure from below and reform from above.

Each individual ‘feudal’ is quite a small landowner by traditional European standards (500 acres is a big estate in Pakistan), and most of his (or sometimes her) wealth may well now come from urban property. In Punjab at least, the really big landowners lost most of their land in the land reforms of Ayub in the late 1950s and Bhutto in the early 1970s.

In fact, I thought of arguing that there is in fact no such thing as a feudal in Pakistan; but then I remembered wild-boar hunting with the noble landowners of Sindh – a remarkably ‘feudal’ experience, described in Chapter 8. So what I’ll say instead is that while there certainly are ‘feudals’ in Pakistan in the loose Pakistani sense of that term, there are no feudals in the European historical sense.

A great many leading ‘feudal’ families, especially in northern Punjab, are not old landowning families at all, but emerged quite recently, often on the basis of urban property or successful corruption when in state service. However, they adopt the same manners and behave politically in the same ways as the old families. Above all, they appeal for support in their own kinship groups, and do so on the basis of the same old politics of patronage and protection.

Indeed, the most powerful remaining ‘feudals’ in Pakistan owe their power not to the extent of their personal landholdings but to the fact that they are the chiefs of large tribes; and the importance of leadership roles in kinship groups extends down to the lesser gentry as well. It is the way in which individual landowners are embedded in landowning clans (like the Gujjars of Attock, the subject of a classic study by Stephen Lyon) which gives them their tremendous strength and resilience, and allows them to go on controlling the politics of the countryside, and dominating those of the country as a whole.7

If the political power of the kinship group in Pakistan depended only on the distribution of patronage, then this power might well have declined over time, given that patronage will always be limited; but it is also rooted in the oldest of social compulsions: collective defence. As one landowner-politician in Sindh told me, in words which were echoed by many other people and provide the title for this book,

This is a hard country. You need family or tribal links to protect you, so that there are people who will stick with you and sacrifice for you whatever happens. That way you will not be abandoned even when out of government. The tribal people gives even ordinary tribesmen some strength and protection against attack, whether by dacoits, the police, the courts – your tribesmen will get you out of jail, lie for you to the court, avenge you if necessary.

Since British days, outside the Baloch and Pathan areas this has rarely been a matter of the whole clan taking up arms against a rival clan. Rather, in a violent society in which none of the institutions of the state can be relied on to act in accordance with their formal rules, close relations with kinsfolk are essential for help against rivals, against the predatory and violent police, in the courts, in politics, and in the extraction of political patronage – all areas of activity which overlap and depend on each other.

A combination of the weakness of the state and the power of kinship is one critical reason why urbanization has had a much smaller impact on political patterns and structures than one might otherwise have expected. Rather than a new urban population emerging, what we have mainly seen so far is huge numbers of peasants going to live in the cities while remaining culturally peasants. They remain deeply attached to their kinship groups, and they still need their kinship groups to help them for many of the same reasons they needed them in the countryside. Underlying all this is the fact that so much of the urban population remains semi-employed or informally employed, rather than moving into modern sectors of the economy – because these usually do not exist.

And of course while the power of kinship is necessary to defend against the predatory state, it is also one of the key factors in making the state predatory, as kinship groups use the state to achieve their goals of power, wealth and triumph over other kinship groups. As one informed description of the state legal system has it:

Below the level of the High Courts all is corruption. Neither the facts nor the law in the case have real bearing on the outcome. It all depends on who you know, who has influence and where you put your money.8

So the ancient Pakistani kinship groups and the modern Pakistani state dance along together down the years, trapped in a marriage that ought to be antagonistic, but has in fact become essential to the nature of each party.

The problem about all this is that while in one way the power of kinship, underpinning the rule of the elites, has so far maintained the basic stability and even the existence of Pakistan, in other ways the plundering of state resources for patronage which this politics breeds has been extremely bad for the development of the country.

It is striking that the most economically and socially dynamic sections of the Pakistani population are those which have to a greater or lesser extent been shaken loose from their traditional cultural patterns and kinship allegiances by mass migration. This is most obviously true of the Mohajirs of Karachi, who emigrated from India after 1947; but even more important are the Punjabis who fled from east Punjab in the dreadful summer of 1947, and now form the backbone of the Punjabi economy. A much smaller, but in some ways even more striking, group are the Hazaras of Balochistan who fled from Afghanistan in the late nineteenth century, and whose educated middle-class society forms a remarkable contrast to the stagnation that surrounds them.


The original title for this book was ‘How Pakistan Works’, and one of its core goals is to show that, contrary to much instinctive belief in the West, it has actually worked according to its own imperfect but functional patterns. One of the minor curses of writing on world affairs over the past few years has been the proliferating use of the term ‘failed state’. Coined originally for genuinely failed and failing states in sub-Saharan Africa, this term has since been thrown around with wild abandon to describe a great range of states around the world, pretty much in accordance with the writer’s prejudices or the need of his or her publication for a sensational headline.

In this respect, it is instructive to place Pakistan in the context of the rest of South Asia. All of the states of this region have faced insurgencies over the past generation, which in two cases (Afghanistan and Nepal) have actually overthrown the existing state. Sri Lanka and Burma have both faced rebellions which have lasted longer, covered proportionally far more territory, and caused proportionally far more casualties than has been the case with the Taleban revolt in Pakistan.

India, the great power of the region, is a stable democracy compared to its neighbours; yet India too has faced repeated rebellions in different parts of its territory, some of them lasting for generations. One of these, the Naxalite Maoist insurgency, affects a third of India’s districts, and effectively controls huge areas of the Indian countryside – a far greater proportion of India than the proportion of Pakistan ever controlled by the Pakistani Taleban. The Indian Prime Minister, Manmohan Singh, in September 2009 described this insurgency as the biggest threat facing India, and said that to date India had been losing the struggle. A recent book by an Indian journalist describes the greater part of the countryside in several Indian states as effectively ‘ungoverned’.9

This is not to argue that India is in any danger of breaking up or collapsing. Rather, one should recognize that states in South Asia have not traditionally exercised direct control over much or even most of their territory, and have always faced continual armed resistance somewhere or other. As in medieval Europe, for most of South Asian history government was mostly indirect, and implemented not by state officials but through local barons or tribal chieftains – who often revolted against the king, emperor or sultan if they felt that he was not treating them with sufficient respect and generosity. The world of the Pakistani landowners of today would in some ways have been immediately recognizable to their fifteenth-century English equivalents.

The British introduced a modern state system which all the present countries of the region have inherited. Yet British rule too was to a great extent indirect. Two-fifths of the territory of British India was in fact ruled by autonomous princes who owed allegiance to the king-emperor (or queen-empress) but governed their own states under British tutelage. Even in the areas which came directly under the British Raj, British rule could not have long maintained itself without the constant help of the local landed aristocrats and chieftains, who in consequence often had pretty much of a free hand when it came to their treatment of their own tenants and labourers. As in parts of Pakistan and India today, these local princelings also sheltered and sponsored bandit groups (dacoits) to help in their constant feuds with their neighbours.

When compared to Canada or France, Pakistan inevitably fails. When compared to India, Bangladesh, Afghanistan, Nepal and Sri Lanka, things therefore do not look so terrible. In fact, a good many key features of Pakistan are common to the subcontinent as a whole, from parties led by hereditary dynasties through the savagery of the police and the corruption of officialdom to the everyday violence and latent anarchy of parts of the countryside.

Pakistan is in fact a great deal more like India – or India like Pakistan – than either country would wish to admit. If Pakistan were an Indian state, then in terms of development, order and per capita income it would find itself somewhere in the middle, considerably below Karnataka but considerably above Bihar. Or to put it another way, if India was only the ‘cow-belt’ of Hindi-speaking north India, it probably wouldn’t be a democracy or a growing economic power either, but some form of impoverished Hindu-nationalist dictatorship, riven by local conflicts.

In order to understand how Pakistan works, it is necessary to draw heavily on the field of anthropology; for one of the things that has thoroughly befuddled not just much Western reporting and analysis of Pakistan, but the accounts of Pakistanis themselves, is that very few of the words we commonly use in describing the Pakistani state and political system mean quite what we think they mean, and often they mean something quite different.

This is true whether one speaks of democracy, the law, the judicial system, the police, elections, political parties or even human rights. In fact, one reason why the army is by far the strongest institution in Pakistan is that it is the only one in which its real internal content, behaviour, rules and culture match more or less its official, outward form. Or, to put it another way, it is the only Pakistani institution which actually works as it is officially meant to – which means that it repeatedly does something that it is not meant to, which is seize power from its weaker and more confused sister institutions.

Parts of Pakistan have been the subject of one of the most distinguished bodies of anthropological literature in the entire discipline; yet with the partial exception of works on the Pathans, almost none of this has made its way into the Western discussion of political and security issues in Pakistan today, let alone the Western media. Critically important works like those of Muhammad Azam Chaudhary and Stephen Lyon on Punjab are known only to fellow anthropologists.10

Incidentally, this is why in this book I have chosen to describe as Pathans the ethnicity known to themselves (according to dialect) as Pashtuns or Pakhtuns. It was under the name of Pathans (the Hindustani name for them, adopted by the British) that this people is described and analysed in the great historical and anthropological works of Olaf Caroe, Fredrik Barth, Akbar S. Ahmed and others; and it was also by this name that this people was known for more than a century to their British military adversaries. The name Pathan recalls this great scholarly tradition as well as the glorious military history of resistance to British conquest, both of which are crucial to understanding developments among the Pathans of the present age.

When it comes to other parts of Pakistani society, the lack of detailed sociological research means that analysts are groping in the dark, and drawing conclusions largely based on anecdotal evidence or their own prejudices. It is striking – and depressing – that more than eight years after 9/11, by far the best US expert on the vitally important subject of Islamist politics in the North West Frontier Province (NWFP)11 is a young graduate student, Joshua White – one key reason being that he has actually lived in the NWFP.12 This lack of basic knowledge applies for example to the critical area of urbanization and its effects – or lack of them – on religious, cultural and political patterns.

According to standard theories concerning urbanization in the Muslim world, the colossal movement of Pakistanis to the cities over the past generations should have led to fundamental cultural changes, reducing the power of the old political clans and traditional forms of Islam, and strengthening modern and radical forms of Islam and modern mass parties. But is this really happening? My own impressions would tend to suggest that things are much more complicated, for reasons that will be discussed in this book. But they are only impressions. Systematic studies of these questions have not been carried out for almost a generation.

As a result of this lack of basic information, too often in Western analysis, when local forms differ from the supposed Western ‘norm’ they are not examined, but are treated as temporary aberrations, diseases to be cured or tumours to be cut out of the otherwise healthy patient’s system. In fact, these ‘diseases’ are the system, and can only be ‘cured’ by a revolutionary change in the system.

The only forces in Pakistan that are offering such a change are the radical Islamists, and their cure would almost certainly finish the patient off altogether. Failing this, if Pakistan is to follow Western models of progress, it will have to do so slowly, incrementally and above all organically, in accordance with its own nature and not Western precepts.


In the course of Pakistan’s sixty-year history, there have been several different attempts radically to change Pakistan, by one civilian and three military regimes. Generals Ayub Khan and Pervez Musharraf, military rulers in 1958 – 69 and 1999 – 2008 respectively, both took as their model Mustafa Kemal ‘Atatürk’, the great secular modernizing nationalist and founder of the Turkish republic. General Zia-ul-Haq, military ruler from 1977 to 1988, took a very different course, trying to unite and develop Pakistan through enforced adherence to a stricter and more puritanical form of Islam mixed with Pakistani nationalism. Zulfikar Ali Bhutto, founder of the Pakistani People’s Party and civilian ruler of Pakistan in the 1970s, for his part tried to rally the Pakistani masses behind him with a programme of anti-elitist economic populism, also mixed with Pakistani nationalism.

And they all failed. Every single one of them found their regimes ingested by the elites they had hoped to displace, and engaged in the same patronage politics as the regimes that they had overthrown. None was able to found a new mass party staffed by professional politicians and ideologically committed activists rather than local ‘feudals’ and urban bosses and their followers. Indeed, with the exception of Bhutto none tried seriously to do so, and after a short while Bhutto’s PPP too had ceased to be the radical party of its early years and had become dependent on the same old local clans and local patronage.

The military governments which took power promising to sweep away the political elites and their corruption also found themselves governing through them, partly because no military regime has been strong enough to govern for long without parliament – and parliament is drawn from the same old political elites, and reflects the society which the military regimes wish in principle to change. Western demands that such regimes simultaneously reform the country and restore ‘democracy’ are therefore in some ways an exercise in comprehensively missing the point.

To have changed all this, and created a radical national movement for change like that of Ataturk, would have required two things: firstly a strong Pakistani nationalism akin to modern Turkish nationalism – something that ethnically divided Pakistan does not have and cannot create; and, secondly, a capacity for ruthlessness to equal that of Ataturk and his followers in suppressing ethnic, tribal and religious opposition. For the pleasant Western story of Turkey’s ascent to its fragile democracy of today ignores both the length of time this took and the hecatomb of corpses on which the modern Turkish state was originally built.

With the exception of the dreadful atrocities perpetrated in East Bengal in 1971 – committed against people whom the Punjabi and Pathan soldiery regarded as alien, inferior and Hindu-influenced – the Pakistani state has not been able to commit abuses on a really massive scale against its own people, either because, in the case of Punjab and the NWFP, its soldiers were not willing to kill their own people, or, in Sindh and even Balochistan, because it always in the end had to make compromises with the local elites.

One of the most striking things about Pakistan’s military dictatorships is in fact how mild they have been by the historical standards of such dictatorships, when it comes to suppressing dissent and criticism among the elites. Only one prime minister (Zulfikar Ali Bhutto) and a tiny handful of politicians have ever been executed in Pakistan – far fewer than have been killed in feuds with each other. Few senior politicians have been tortured.

Of course, the poor are a different matter, but, as noted, they get beaten up by the police whoever is in power. Perhaps the single most important social distinction in Pakistan is that between what Graham Greene called the ‘torturable classes’ and the ‘untorturable’ ones.13 This view has support from a surprising source. As General Musharraf writes in his memoirs, ‘Whatever the law, civil or military, the poor are always the victims of oppression.’14

Nor indeed has the Pakistani state ever faced rebellion in West Pakistan on a scale that would have provoked massacre in response – though that could be changed by the Taleban insurgency in the Pathan areas which began in 2004 and gathered strength in 2008 and 2009. India has faced much more serious rebellions, and has engaged in much largerscale repression in response.

But in India, as in Pakistan, the state is not responsible for most human rights abuses. This is something that human rights groups in particular find hard to grasp, since they stem from a modern Western experience in which oppression came chiefly from over-mighty states. In Pakistan, however, as in India, the vast majority of human rights abuses come not from state strength, but from state weakness. Even when they are committed by state policemen, they are not on the orders of the government, but are the result of individual policemen or groups of police preying on the population as their ancestors did for centuries. Take the police chief in the interior of Sindh, who told me, ‘I try to stop my boys raping women and torturing people to death. Beyond that, you have to be realistic. Anyway, we need to raise more money from the people just to do our job half-way properly.’

No one can seriously imagine that when police rape a woman or torture a suspected criminal in their custody, that this is the will of Prime Minister Manmohan Singh of India or President Zardari of Pakistan. They can be accused of not doing enough to stop such abuses – but their ability to do so is very limited. So Pakistan – and indeed South Asia and much of Latin America – demonstrates the frequent irrelevance of democracy even in an area where we instinctively think that it makes all the difference, namely human rights. The overwhelming majority of human rights abuses in Pakistan stem from a mixture of freelance brutality and exploitation by policemen, working either for themselves or for local elites; actions by local landlords and bosses; and punishments by local communities of real or perceived infringements of their moral code.

The murder of women in ‘honour killings’, the giving of young girls in marriage as compensation in the settlement of clan feuds, the dreadful cases of gang rape as a punishment which have taken place in recent years in southern Punjab, Sindh and Balochistan are the work of families, or local clans and their collective leaderships, not of the state. Atrocities by local dominant clans and the police are also entirely characteristic of much of neighbouring democratic India. As regards the police, this is starkly revealed in a report by Human Rights Watch of August 2009.15

Not for nothing was an old Hindustani popular term for banditry ‘padshahi kam’, the imperial trade. This indicated not only that ordinary people could usually see no difference between bandits and soldiers, but that they often changed places. Unpaid soldiers became bandits; successful bandits became soldiers of conquering armies, and their leaders became kings.

According to standard Western models, and to the Pakistani constitution that derives from them, authority stems from the sovereign people through elections, and then spreads downwards from the government through hierarchical structures, which transmit orders from above, from superior to inferior officials, in accordance with laws made by parliament or at least by some formal authority.

In Pakistan, only the armed forces work even in the second half of this way. For the rest of the state, the law, the judiciary and the police, authority is a matter of constant negotiation, with violence or the threat of it very often one of the cards that can be played on either side. The negotiated nature of the Pakistani state was summed up for me in a grim anecdote from a retired general who in the 1990s was responsible for commanding anti-dacoit operations in Sindh.

A subordinate had run a dacoit gang to earth on the estate of a parliamentarian from the then ruling party, and wanted to send troops in to get them – which would have led to furious protests from the governments in Islamabad and Karachi, and most probably the immediate release of the men arrested. His commander overruled him, and instead invited himself to lunch with the landowner concerned. At the end of a convivial meal, he passed his host a note and said that he’d be personally obliged for his help. The next day, four of the dacoits were handed over to the army with a message from the landowner-politician saying that the general could shoot two of them, but could he please charge the other two before the courts.

‘Any two?’ I asked, somewhat faintly.

No, he said which two we could shoot. Probably they had offended him in some way, or they were not from his tribe. As to the other two, he knew perfectly well that his influence meant that the courts would never convict them, and they would be released after a few months. The courts are useless when it comes to criminals in this country if the criminals have any connections – they are bribed, or scared, or both. That is why if you really want to deal with a miscreant, the only way is to kill him out of hand. This is a hard country, and this is the way things are here, sadly.

This negotiated nature of the state also applies to the workings of democracy. For democracy is representative not only of the people, but of all those classes, groups and institutions through which the popular will is refracted until it eventually finds some kind of distorted reflection in elected institutions. In other words, democracy usually reflects not so much ‘the people’ or ‘the electorate’ as the distribution of social, economic, cultural and political power within a given society. The nature of Pakistani society, and the weakness of real democratic development, are shown among other things by the lack of real, modern, mass political parties, with their own cadres of party workers.

A while spent pondering on these themes should bring out why so much Western analysis of Pakistan misses the mark, because it expects institutions with names like ‘the law’ and ‘the police’ to work as they are meant to work in the West, according to rules rather than negotiation. Similarly, Western language about ‘corruption’ in Pakistan suggests that it can and should be cut out of the political system; but in so far as the political system runs on patronage and kinship, and corruption is intertwined with patronage and kinship, to cut it out would mean gutting Pakistan’s society like a fish.

This of course is precisely what the Islamist revolutionaries would like to do. The modern Islamist political groups are trying to replace the clan and patronage politics of the ‘feudal’ landowners and urban bosses with their own version of modern mass politics, so far with only very limited success. With the partial exception of the Jamaat Islami, the Islamist political parties have themselves been swallowed up by the patronage system. As for the Pakistani Taleban (the Tehriq-e-Taleban Pakistan, or TTP), they are so far a primitive collection of guerrilla and terrorist groups, which would be completely at sea if they found themselves responsible for Peshawar, let alone Lahore or Karachi.

Of course, they do draw a great deal of their strength from the glaring inequities and oppressions of the Pakistani system, and above all the justice system. It is true, as I have said, that ordinary Pakistanis are themselves part of endless conspiracies to pervert the course of justice – but it is also true that they feel they have no choice, given the nature of the justice system. The state’s law is felt by many ordinary people not just to be rigged in favour of the rich, and hopelessly slow, corrupt and inefficient, but also to be alien – alien to local tradition, alien to Islam, the creation of alien Christian rulers, and conducted by the elites for their own benefit.

So, as this book will describe, when ordinary people speak of their reverence for the Shariah, and their respect for the Taleban when they introduce the Shariah, this should not necessarily be taken as active support for the Taleban’s complete programme. Rather it is a mixture of a reverence for the Shariah as part of the word of God, dictated to the last Prophet, with a vague yearning for a justice system that might be cruder than that of the state, but would also be quicker, less biased in favour of the elites, and conducted before the eyes of the people, in their own language. Mixed in with this is a great deal of somewhat veiled anti-elitist feeling, which in the eyes of parts of the Pathan tribal populations helps fuel mass acceptance of Taleban attacks on the local maliks and khans, or tribal bosses and local landowners.

But then, the Pathans have always been the most culturally egalitarian people of Pakistan. Among the masses elsewhere, the progress of the Islamists has so far generally been very limited when it comes to gaining active mass support. One key reason for their failure to date is the deeply conservative nature of much of Pakistani society; for – quite contrary to most Western perceptions – Islamist mobilization often thrives not on backwardness, but on partially achieved modernity. Thus, to judge by all the economic evidence about poverty and landownership, radical Islamist groups preaching land reform ought to be flourishing in the Pakistani countryside.

In fact, the only areas where they have had any significant success (outside the Pathan territories) are where a sectarian (Sunni versus Shia) or tribal element comes into play. This is partly because of clan solidarity, but also for the simple reason that the only people who could lead such a radical Islamist movement in the countryside are the local mullahs, and they are in effect chosen by the local ‘feudal’ landowners – who do not exactly favour radicalism of any kind, least of all involving land reform.

In the cities, things are freer, but even there most attempts at political mobilization from below are stifled by the grip of the political bosses and the kinship groups they lead, as well as by the politically apathetic condition of society, and by divisions along religious lines. In other words, while there is certainly a great deal of economic, social discontent in the Pakistani population, being discontented is not at all the same thing as being able to do something about it. As of 2009, the perennial discontent of the urban masses in most of Pakistan continues to express itself not in terms of political mobilization behind new mass movements, but sporadic and pointless riots and destruction of property – including most notably the buses in which the rioters themselves have to travel every day.

According to the standard Western version, by which the Western way is the only way to modernity, the key ideological struggle in Pakistan is between Westernized modernity (including democracy, the rule of law, and so on) and Islamic conservatism. A more accurate way of looking at it would be to see much of Pakistan as a highly conservative, archaic, even sometimes quite inert and somnolent mass of different societies, with two modernizing impulses fighting to wake it up.

The Western modernizers have on their side the prestige and success of the Western model in the world in general, and the legacy of British rule, including a vague belief in democracy – but are crippled both by the conservative nature of Pakistani society and by growing popular hatred for the US and its Western allies.

The Islamist modernizers can draw on a much more ancient and deeply rooted tradition, that of Islam – but are crippled by the conservative nature of Pakistani society, by Pakistan’s extreme fissiparousness, by the failure of their programme elsewhere in the Muslim world, and by the fact that the vast majority of the Pakistani elites reject their model, for cultural as well as class reasons. Both Westernizers and Islamists see the battle between them as apocalyptic, and ending with the triumph of good or evil. Yet there is a fair chance that Pakistan will in effect shrug both of them off, roll over, and go back to sleep.


Pakistan cannot however afford to do so, because time is not on Pakistan’s side. In the long run, the most important thing about the people of Pakistan is not who they are or what kind of religion they follow, but that whoever they are, there are too many of them for the land in which they find themselves – and more of them all the time. In 2010, the population was estimated at between 180 and 200 million, making Pakistan the sixth largest country on earth in terms of population. It had risen from 131 million at the census of 1998, 34 million at the census of 1951 (four years after independence) and only 19 million at the British census of 1911 – and is today ten times what it was 100 years ago. Officially, population growth now stands at 2.2 per cent a year – which would seem to be a serious underestimate.

Pakistan’s inability to bring this rate down more quickly reflects state weakness, social conservatism, lack of education (above all among women) and the ability of the religious parties to play on popular prejudices. Since Ayub Khan in the late 1950s, no Pakistani government has dared to promote family planning seriously, and the reduction that has occurred has happened because of socio-economic change and urbanization, not through state action.

The huge youth bulge making its way through the Pakistani population means that this population will continue to grow steeply for a long time to come (in 2008, 42 per cent of the population was estimated as under the age of fourteen). If present trends continue, then by the middle of the twenty-first century, according to World Bank projections, Pakistan may have as many as 335 million people.16

This is far too many people for Pakistan’s available water resources to support, unless the efficiency of water use can be radically improved. If the old Indian economy used to be described as ‘a gamble on the monsoon’, then the entire Pakistani state can be described as ‘a gamble on the Indus’ – and climate change means that over the next century this may be a gamble against increasingly long odds. The capricious power of water in this area is demonstrated by the remains of numerous cities – starting with those of the Indus Valley civilization 4,000 years ago – that have been either abandoned because rivers have changed their course, or been washed away by floods, as so many towns and villages were by the great floods of 2010.

At an average of 240 mm of rainfall per year, Pakistan is one of the most naturally arid of the world’s heavily populated states. Without the Indus river system and the canals flowing from it, most, even of Punjab, would be semi-desert and scrub-forest (called ‘jungle’ in Pakistan) – as it was before the British began their great irrigation projects.

This is apparent if you fly over the country. Once the five great rivers of Punjab and the Kabul river flowing from Afghanistan have paid their tribute to the Indus, the vast majority of cultivated land in the southern end of Punjab and the whole of Sindh is only what can be irrigated from the Indus. Beyond these lands, all is brown, yellow and grey, dotted with the occasional oasis provided by natural springs or more often tube-wells. Only 24 per cent of Pakistan’s land area is cultivated – the great majority through man-made irrigation systems. The rest is pastoral land, or uninhabited: desert, semi-desert, and mountain.

Chronic over-use, however, means that many of the natural springs have dried up, and the water table is dropping so rapidly in many areas that the tube-wells will also eventually follow them into extinction. That will leave the Indus once again; and in the furore surrounding the debunking of the exaggerated claim that the glaciers feeding the Indus will disappear by 2035, it has been forgotten that they are nonetheless melting; and if they disappear a century or two later, the effects on Pakistan will be equally dire, if no serious action is taken in the meantime radically to improve Pakistan’s conservation and efficient use of water.

If the floods of 2010 are a harbinger of a long-term pattern of increased monsoon rains, this on the other hand would potentially be of great benefit to Pakistan – but only potentially, because to harness them for agriculture requires both a vastly improved storage and distribution infrastructure, and radical measures to stop deforestation in the mountains and to replant deforested areas. Otherwise, increased rainfall will risk more catastrophes like that of 2010, with the water rushing off the deforested hillsides in the north to swamp first the valleys and then the plains. It should be added, though, that an absolutely essential part of existing infrastructure did work during the floods: the great barrages along the Indus and its tributaries. If these had broken, several of Pakistan’s greatest cities would have been inundated, and the death toll would have been vastly higher than the 1,900 who lost their lives.

This dependence on the Indus is the greatest source of long-term danger to Pakistan. Over the next century, the possible long-term combination of climate change, acute water shortages, poor water infrastructure and steep population growth has the potential to wreck Pakistan as an organized state and society. Long-term international aid projects in Pakistan should be devoted above all to reducing this mortal threat, by promoting reforestation, repairing irrigation systems and even more importantly improving the efficiency of water use. Human beings can survive for centuries without democracy, and even without much security. They cannot live for more than three days without water.

The extent of the water crisis that is already occurring will be described in the chapters on Sindh and Balochistan. As two of the authors of the World Bank’s very worrying 2004 report on Pakistan’s water situation write:

The facts are stark. Pakistan is already one of the most water-stressed countries in the world, a situation that is going to degrade into outright water scarcity due to high population growth. There is no feasible intervention which would enable Pakistan to mobilize appreciably more water than it now uses ...

There are no additional water resources to be exploited and agricultural water use must decline to enable adequate flows into the degrading Indus River Delta. Pakistan’s dependence on a single river system makes its water economy highly risky ...

Groundwater is now being overexploited in many areas, and its quality is declining ... There is little evidence that government (or donors, including the World Bank) have re-engineered their capacity and funding to deal with this great challenge. And here delay is fatal, because the longer it takes to develop such actions, the greater will become the depth [beneath the earth] of the water table.17

According to a 2009 study by the Woodrow Wilson Center drawing on a range of different works, by 2025 population growth is likely to mean that Pakistan’s annual water demand rises to 338 billion cubic metres (bcm) – while, unless radical action is taken, Pakistan’s water availability will be around the same as at present, at 236 bcm. The resulting shortfall of 100 bcm would be two-thirds of the entire present flow of the Indus.18

And this frightening situation would have come about even before the potential effects of climate change begin to kick in. These effects could be to turn stress into catastrophe by the end of the twenty-first century. Well before Pakistan reaches this point, however, it is likely that conflict over access to the shrinking Indus will have raised tensions between Pakistan’s provinces to levels which will be incompatible with the country’s survival.

If anyone thinks that the condition of Pakistan will be of little consequence to the rest of the world in the long run, they should remember that a hundred years from now, if it survives that long, Pakistan will still possess nuclear weapons, one of the biggest armies in the world, one of the biggest populations in the world and one of the biggest diasporas in the world, especially in Britain. Islamist radicalism, which has already existed for hundreds of years, will also still be present, even if it has been considerably reduced by the West’s withdrawal from Afghanistan.

All of this will still mean that of all the countries in the world that are acutely threatened by climate change, Pakistan will be one of the most important. Moreover, what happens to Pakistan will have a crucial effect on the rest of South Asia, where around one-fifth of the world’s entire population live and will live. Those Indians who would be tempted to rejoice in Pakistan’s fall should therefore consider that it would almost certainly drag India down with it.

The World Bank has valuable programmes in place, on which much more could be built. For example, at present, Pakistan harvests a good deal less of its rainfall than neighbouring areas of India, and only a fraction of China’s harvest per cubic metre of rainfall. Intensive effort is needed to improve this performance – something which does not require expensive high-tech solutions, but rather a mixture of spadework and better, more innovative management.

Concentrating development aid on the improvement of Pakistan’s water infrastructure has the added advantage that such improvement is highly labour-intensive, and provides a range of jobs, from masses of ordinary workers with spades to highly trained engineers. This means that benefits from international aid will flow immediately to many ordinary people and be immediately apparent to them. By contrast, most US aid in recent decades, though often very useful to the economy as a whole, has not been visible to ordinary people and therefore has had no political effects in terms of attitudes to the US.


There would be no point in talking about any of this if Pakistan were in fact the hopeless economic and administrative basket-case that it is so often made out to be. This is not the situation, however. Pakistan has never followed the ‘Asian tigers’ in radically successful modern development, and shows no signs of doing so, but for most of the time since 1947 its rates of growth have been substantially higher than India’s. Pakistan would be a far more developed and prosperous state today but for the economic disaster of Zulfikar Ali Bhutto’s nationalization programmes of the 1970s, which led to a steep drop in growth.

After another period of economic stagnation in the 1990s (worsened by US sanctions imposed as a punishment for Pakistan’s nuclear programme), under the Musharraf administration from 1999 to 2008 economic growth returned to a rate of between 6.6 and 9 per cent a year, before dropping again as a result of the global economic recession.

Certainly, most people in Pakistan are poor; but all the same, as a result of this economic growth, together with the effects of Islamic charity and the circulation of state patronage to the kinfolk and followers of successful politicians, Pakistan lacks the huge concentrations of absolute poverty to be found in India’s cities and countryside. In fact, the absolutely poor and defenceless people in Pakistan are often the same as in India – the descendants of the old ‘untouchable’ castes, who – seeing nothing for them in India – remained in Pakistan at partition but have never escaped their traditional poverty and marginalization. There are however far fewer of them in Pakistan.

From this point of view as so many others, Pakistan has a rather medieval look. The state is very bad at providing modern services such as clean water, medicine, public transport and education, because it is too weak either to force much of the population to pay taxes or to control corruption on the part of its own officials. In part as a result of the lack of education, ordinary people are also very bad at organizing themselves to demand or create such services. Certain groups are outside the system altogether, and have no access to protection, patronage or charity. On the other hand, the system does ensure that the great majority of the population does at least have enough to eat.

And where the state decides that a particular development project is of great national importance, it can in fact partially isolate it from the corruption of the rest of the system and ensure that it is built successfully. This was true of the vast extension of dams and irrigation in the 1950s, and in recent years the construction of the port of Gwadar and the fine motorways linking the great cities of northern Pakistan.

Pakistan’s GDP as of 2009 stood at $167 billion, making it the 48th largest economy in the world (27th if adjusted for purchasing power). Despite the image of Pakistan as an overwhelmingly rural society, and the dominance of political, social and cultural patterns drawn from the countryside, agriculture as of 2009 accounted for only about 20 per cent of GDP. The ‘service sector’ accounted for 53 per cent (most of it in informal, very small-scale businesses and transport), with industry at 26 per cent. However, around 60 per cent of the population continued to live in the countryside, helping to explain the continued power of the rural elites. Most of Pakistani industry is made up of textiles and food processing. In 2007 – 8 Pakistani exports stood at $18 billion, the majority of them textiles.

Pakistan also contains certain islands of high technology – above all the nuclear industry, which (whatever you may think about its strategic implications) is a very remarkable achievement for a country with Pakistan’s economic profile, and shows what the Pakistani state can achieve if it really sets its mind to it, and can mobilize enough educated, honest and committed people.

It is miserably clear, however, that – as with the other South Asian countries – the greater part of the Pakistani economy has not made the breakthrough to modern development and seems nowhere near doing so. As of 2009, GDP per head stood at a mere $1,250 (before adjustment for purchasing parity). Between 1960 and 2005, per capita income as a proportion of that of the USA actually fell from 3.37 to 1.71 per cent. Some 23 per cent of the population live below the poverty line. Underlying this lack of development is a literacy rate which in 2010 stood at only 55.9 per cent, above all because of the complete absence of education for women in much of the countryside.


If the West and China want to help improve this picture, they need to develop an approach to Pakistan which recognizes the supreme importance of the country but is based on a real understanding of it, and not on fantasy, whether of the paranoid or optimistic variety. This book is an attempt to strengthen such understanding. It is based on travels to Pakistan dating back to my time there as a journalist for The Times (London) in the late 1980s, and on five research trips in 2007 – 9 lasting a total of six months, during which I visited all Pakistan’s provinces and major cities.

It should be said that, with the exception of my stays in some of the Pathan areas, at no point during my visits did I feel under any direct physical threat, except from the execrable local driving – and if you were going to be too affected by that you’d have to avoid visiting about half the world. Moreover, the Pathan areas are only a small proportion of Pakistan as a whole. It is worth stressing this, because one reason why Pakistan is so little known and so badly misinterpreted in the West is that so many analysts and commentators are too afraid to go there, or, if they go, to travel outside Islamabad. This reluctance to visit Pakistan is true even in Britain, which is organically linked to Pakistan by the large Pakistani diaspora; and it is largely unjustified – not to use a stronger word for this behaviour.

Of course, an unescorted visit to the tribal areas, or a prolonged stay in Peshawar in unprotected accommodation, might very well prove fatal; but it is entirely possible to live for months on end in a dozen other different Pakistani cities, and most of the Pakistani countryside, without in my view running any very serious risk. Researchers, analysts and officials dealing with aid to Pakistan need to do this if they are to do their jobs properly.19

My own recent researches in Pakistan have been not nearly as long as I would have wished, owing to teaching commitments and a short deadline for this book; but they did give me the chance to talk to hundreds of ordinary Pakistanis from every walk of life, most of whom had never been asked for their opinion before by any Pakistani or Western observer or organization. The views of these voiceless masses form the heart of this book, and I have tried to reflect them as faithfully as I can and draw from them an understanding of Pakistani reality.

This has not always been easy. I remember a conversation with an elderly leader of the Awami National Party in Peshawar in 2008. In his garden were a pair of strikingly graceful long-legged grey birds with crests. I asked him what they were. ‘Flamingos,’ he replied.

‘Um, I don’t think so, sir ...’

‘No, you are right of course, they are not flamingos. They migrate to Russia in summer, sensible birds, and then come back here again to lay their eggs. We hoped one of them was female and would lay eggs, but it turns out they are both males. So they fight all the time or stand at opposite ends of the garden sulking.’

‘Like Pashtun politicians, sir, perhaps?’

With a deep laugh: ‘Oh yes, yes indeed, unfortunately! But what are they called now in English? We call them koonj.’ Turning to his staff, he asked, ‘What are koonj in English?’ And with absolute, unquestionable conviction, one of them replied:


So if in the course of this book I have sometimes mistaken flamingos for swans, or indeed pristine Pakistani political swans for ugly ducklings, I can only plead in self-defence Pakistani society’s ability to generate within an astonishingly short space of time several mutually incompatible versions of a given event or fact, often linked to conspiracy theories which pass through the baroque to the rococo.

Conflicted Pakistani relations with reality notwithstanding, I am deeply attached to Pakistan, which has provided me with some of the best moments of my life; and Pakistan’s people have treated me with immense kindness and hospitality. Pakistan is one of the most fascinating countries of my acquaintance, a place that cries out for the combined talents of a novelist, an anthropologist and a painter. But after twenty years of intermittently covering Pakistan, this is a clear-sighted affection. I have good friends among the Pakistani elite, but I also take much of what they say with several pinches of salt – even, or especially, when their statements seem to correspond to Western liberal ideology, and please Western journalists and officials.

The problem goes deeper than this, however. Western-style democracy has become so associated over the past generation with human rights, wealth, progress and stability, that to accept that a country cannot at present generate stable and successful forms of it is an admission so grating that both Westerners and educated Pakistanis naturally shy away from it; Westerners because it seems insulting and patronizing, Pakistanis because it seems humiliating. Both, therefore, rather than examining the reality of Pakistan’s social, economic and cultural power structures, have a tendency to reach for simple explanations concerning the wicked behaviour of malignant forces, whether the generals, the ‘feudals’, the mullahs or the Americans.

Seen from a long historical perspective, things look rather different. Modern democracy is a quite recent Western innovation. In the past, European societies were in many ways close to that of Pakistan today – and indeed, modern Europe has generated far more dreadful atrocities than anything Islam or South Asia has yet achieved. In the future – who can say? The virtues of courage, honour and hospitality, in which Pakistanis excel, do after all have their permanent worth.

It may also be worthwhile in this context to recall the words which the science fiction novelist John Wyndham put into the mouth of a professor asked to give advice on future careers in a world threatened by radical climate change. He had two recommendations: ‘Find a hilltop, and fortify it’; and ‘Enlist in a regiment with a famous name. There’ll be a use for you.’21 Pakistan has plenty of famous regiments, and local chieftains have been fortifying hilltops for thousands of years. They may yet cope better with the future than the successful elites of today’s world.


That kinship is of critical importance in Pakistan is something on which all the academic experts agree – which is nice, because they tend to agree on nothing else about the subject. For me, the definitive word was said 100 years ago by the great British civil servant and ethnographer Sir Denzil Ibbetson. After an official career lasting decades in the Punjab, he wrote:

An old agnostic is said to have summed up his philosophy in the following words: ‘The only thing I know is that I know nothing, and I am not quite sure that I know that.’ His words express very exactly my own feelings regarding caste in the Panjab. My experience is that it is almost impossible to make any statement whatever regarding any one of the castes we have to deal with, absolutely true though it may be for one part of the province, which shall not presently be contradicted with equal truth as regards the same people in some other district.22

Thus the English term ‘tribe’ can be translated into several different local words, which overlap with other English meanings. Qaum can mean tribe, people, ethnicity or even nation. Zat is related to the Indian word jati, for a ‘sub-caste’ in Hinduism. ‘Tribe’ can also mean several very different things in different parts of Pakistan. Among the Baloch tribes (not just in Balochistan but in Sindh and southern Punjab as well), a tribe means something like the old clans of Scotland, a tightly knit group under an autocratic chieftain.

Among the Pathans, however, while tribal membership is a tremendously important marker of identity and status, the tribes are divided into endless rivalrous sub-tribes, for which new leading men emerge in every generation. Meanwhile in Punjab, the Rajputs, Jats, Gujjars and others were presumably tightly knit nomadic tribes in the distant past, but long ago spread out and intermingled territorially across the whole of what is now northern India and Pakistan (the Gujjars have given their name to a state in India as well as a city in Pakistan, among many other places).

Originally assimilated to the Hindu caste system as kshatriyas (the warrior caste) thanks to their conquests, many Jats, Gujjars and Rajputs later converted to Islam or Sikhism. Within Pakistan, they have no collective overall political identity at all, but have a certain community of sentiment, including a strong sense of superiority to everyone else, and a strong preference for marrying each other. An appeal to fellow Rajput or Jat feeling may gain some limited help when all else fails. More important in terms of loyalty and mobilization is the local sub-clan, as in Chauhan Rajput, Alpial Rajput and so on. The Sayyids and Qureishis are groups peculiar to Islam, being (ostensibly) descendants of the Prophet and his clan, and therefore of Arabic origin. Yet their role and status in South Asian Muslim society has certain limited affinities to that of the Brahmins in South Asian Hindu society.

Meanwhile, other kinship groups are descended straight from the lower castes of the Hindu system. These include the kammi artisan and service groups of the towns and villages; and below them, the old untouchables and tribals, who are so far down the system that no one even bothers much if they are Muslim, Hindu or – what most really are – pre-Hindu animist. As in India, these last are the most vulnerable groups in Pakistani society, liable to be preyed on economically and sexually by local dominant lineages and by the police.

As to effective political roles within kinship groups and in wider politics, this spreads outwards from the khandan – denoting both the immediate family and the extended family (often a joint family, in which several brothers and their families live together under one roof) – to that hellish concept, invented for the confusion of mankind: biradiri (related to the Indo-European root for ‘brother’), which is supposed to denote the descendants of one common male ancestor. To judge by my interviews, however, biradiri can be used to mean almost any kind of wider hereditary kinship link depending on context. In this book, I have therefore used ‘kinship group’ or ‘kinship network’ rather than biradiri when speaking of such wider groups.

If you find an error please notify us in the comments. Thank you!