House of Commons - Foreign Affairs Committee

Global Security: Afghanistan and Pakistan - Foreign Affairs Committee Contents

Submission by Adam Roberts

1. The central question that is explored in this submission is: what are the implications of wars in Afghanistan for international security—not only in the region but also more generally? The central question can be approached by looking first into four related questions about wars in Afghanistan and their influence on international security.

— How should the almost continuous wars in Afghanistan since 1989 be characterized, and what have been the effects of their Pakistan dimension?

— What have been the roles of the United Nations in the long-running Afghan crisis, including in its post 2001 post-conflict peace-building role and in assisting the return of refugees?

— In the war since 2001, what problems have there been in fitting Western military doctrines, practices and institutions to Afghan realities? What has been the role of air power? How has NATO performed in this unanticipated commitment? Are counter-insurgency doctrines fit for purpose in Afghanistan? And how can progress be judged?

2. The exploration of the fourth question, which forms the main part of this survey, leads on to the concluding discussion of the actual and possible future effects of the war on international security, including on the United States, the United Nations and NATO. Some policy choices are briefly summarized. They involve a difficult underlying issue: whether to go with the grain of Afghan society, with all the compromises that would be involved; or to continue with a modernising and centralising project which is alien to Afghan traditions, and important aspects of which are increasingly in trouble?

3. I am President-elect of the British Academy and Senior Research Fellow at the Centre for International Studies in Oxford University's Department of Politics and International Relations. I am also an Emeritus Fellow of Balliol College, Oxford. I was the Montague Burton Professor of International Relations at Oxford University from 1986 to the end of 2007. I have written extensively about international security, war, and international law and organization.

4. This paper is a product of research conducted under the auspices of the Oxford University Leverhulme Programme on the Changing Character of War. It is based on a presentation at the US Naval War College International Law Department experts' workshop on "The War in Afghanistan", 25-27 June 2008, and a version will appear in due course in Michael N Schmitt (ed), The War in Afghanistan: A Legal Analysis (Newport, Rhode Island: US Naval War College International Law Studies vol 85). A shortened version of this paper is "Doctrine and Reality in Afghanistan", Survival, London, vol 51, no 1, February to March 2009. I am grateful to Alex Alderson, Jeremy Allouche, Rob Johnson, John Nagl, Hew Strachan, Astri Suhrke and Susan Woodward for their comments on earlier drafts of this paper. The usual disclaimer, that all responsibility for error is mine and mine alone, applies with particular force in this case.

A. LESSONS FROM AFGHAN WARS UP TO 1989

The 19th Century and after

5. Many modern wars, including that in Afghanistan, fit quite well the general description of colonial conflicts offered by Major C E Callwell of the Royal Artillery in 1899 in his justly famous manual Small Wars. Callwell himself had served during the closing stages of the Second Anglo-Afghan War, when he marched through the Khyber Pass to join the Kabul field force.[16] It was on the basis of experience that he wrote two decades later:

6. In the 19th century the British army was involved in two major campaigns in Afghanistan, in 1839-42 and 1878-80. The first, fought ostensibly to assist a weak ruler and to provide a friendly buffer state on India's north-west border, was a hubristic enterprise that was marked by disaster—the wiping out of a reduced garrison as it struggled back to the Khyber Pass.[18] The second war, which was fought to counter-balance Russian influence in Afghanistan, provided evidence that apparent success in Afghanistan can be quickly followed by uprisings and setbacks. The British, having defeated the Afghan state, had no political solution except to appoint a suitable "warlord" as head of state. What did Callwell have to say specifically about the type of war that had been encountered in Afghanistan and elsewhere in the late nineteenth century? His words are as pertinent today as when they were penned over a century ago:

With the capture of the capital any approach to organized resistance, under the direct control of the head of the State, will almost always cease; but it does not by any means follow that the conflict is at an end. ... [T]he French experiences in Algeria, and the British experiences in Afghanistan, show that these irregular, protracted, indefinite operations offer often far greater difficulties to the regular armies than the attainment of their original military objective.[19]

7. The wars in Afghanistan in the 19th century have been the foundation for a view of the country and its peoples—especially the latter—as unusually resistant to any kind of foreign influence or control, actual or perceived. David Loyn, the veteran BBC reporter on Afghanistan who has charted these previous conflicts, argues that mistakes are being repeated today because of a neglect of the study of history. He charges that the US and Britain have failed to understand the extent of resistance in Afghanistan to anything that looks like foreign control. It follows, states Loyn, that it is necessary for outsiders to accept a very limited role, and to negotiate with the Taliban.[20] This is one important perspective on wars in Afghanistan. However, it should not be taken to imply that there is uniform hostility to all foreign influence.

8. Twentieth-century Afghan history was characterized not only by wars against foreigners, such as the Third Anglo-Afghan War of May 1919, but also by civil wars, assassinations and coups, as in the conflict of 1928-31 and the seizures of power by Daud Khan in 1953 and 1973. Throughout the twentieth century, there was a continuous interplay between the development of constitutional government and the continuation of political violence. The role of the Pashtun peoples in Afghanistan was one of many bones of contention. The political culture of Afghanistan was characterized by state weakness and general instability.

The Soviet War in Afghanistan, 1979-89

9. The war in 1979-89 between the Soviet-backed government of Afghanistan and its mujahidin adversaries had major effects on international politics. In particular, the war had a vast impact in the Soviet Union. It accentuated the Soviet Union's sense of imperial overstretch; contributed to a decline of faith in the use of force to maintain the empire; and accentuated doubts about a central purpose of Soviet foreign policy—the maintenance of a network of dependent, demanding and hardly popular socialist regimes in an assortment of countries around the world. It formed part of the background to the role of civil resistance movements in central and eastern Europe pursuing their struggles by non-violent means to a successful outcome in 1989. In short, the Afghan war contributed to the collapse of the Soviet empire. This very fact is not only proof of the fateful consequences that may flow from war in Afghanistan, but is also one driver of the present war. Osama Bin Laden has made no secret of his belief that, having helped to destroy the Soviet Union, he aims to do the same for the US. One down, one to go! This was not the only case of post-Cold War hubris—there were also many variants of this condition elsewhere, including in the British and American governments—but it was a notably severe one. Bin Laden's interpretation of events led him to 9/11 and engulfed Afghanistan in continuing war.

10. There were other ways in which the Soviet-Afghan War led to subsequent wars. The channelling of much international aid to mujahidin groups through Pakistan reinforced the fateful link between events in Pakistan and those in Afghanistan. The power of non-state groups and regional military chiefs, and their tendency to rely on threats and uses of force not controlled by any state, became more deeply engrained than before in both Afghanistan and the frontier areas of Pakistan. The religious element in Afghan politics—which was particularly prominent in the struggle against Soviet influence, and was encouraged by the outside powers that provided much-needed finance and weapons for the mujahidin—did not disappear with the departure of Soviet forces in 1989. Indeed, within a few years religious warriors trained in the hard school of combat against Soviet forces in Afghanistan were to turn up in a wide range of other locations, including in the former Yugoslavia.

11. These legacies of the war against Soviet control remain most important in Afghanistan itself. The problem of non-state violence, regional rivalries, and the religious element in politics are not new to Afghanistan, but they were reinforced. Long-held suspicions towards certain types of foreign presence remained prominent.

B. THE WARS IN AFGHANISTAN SINCE 1989

12. The current multi-faceted and complex situation in Afghanistan is best understood as the continuation of a protracted war over the country's future which began many years before 2001. Understanding its character is important not only for developing military and political policy in the country, but also for understanding its likely impact on international security generally. There are fundamental differences of understanding about its nature.

13. Whether viewed as a war or a stabilization mission, there is a tendency to present the situation as a conflict between an essentially progressive cause represented by the Karzai government in Kabul on the one side, and two reactionary Islamist forces on the other: the Taliban and al-Qaeda. This view may be too simple in its views both of the Afghan government and of its opponents. Most strikingly, it tends to overstate the effectiveness of the Afghan government. It also understates the importance of ethnic/linguistic divisions within Afghanistan, where the largest ethnic group, the Pashtuns, constitute over 40% of the population. Elements of Afghan and Pashtun nationalism play a significant part in the resistance to the Afghan government and its foreign backers. A review of the 20 years' crisis in Afghanistan since the Soviet withdrawal, and of the place of Pakistan in that crisis, is necessary for an understanding of the nature of this war.

The crisis since 1989

14. Following the withdrawal of the last Soviet forces from Afghanistan in January 1989, an internal crisis and civil war erupted. Indeed, the civil war can be traced back further, and can be said to have begun in about 1978.[21] It has never really ended. Throughout the two decades since 1989 there have been continuing regional rivalries, involvement of outside powers in support of particular factions, and ongoing conflict between modernizers and Islamists. There have been two moments when the conflict was viewed by some as having ended—after the Taliban victory in September 1996 and after the Northern Alliance victory in December 2001.[22] However, on both occasions the conflict continued in new forms.

15. This first phase of Afghanistan's long-running war following the departure of Soviet forces was only partially concluded on 26 September 1996 when Kabul fell to the Taliban, which established a theocratic style of government throughout the areas under their control and in 1997 renamed the country "Islamic Emirate of Afghanistan". Then and thereafter the Northern Alliance continued to control an area of northern Afghanistan and to challenge Taliban rule.

16. From 7 October 2001 onwards, following the al-Qaeda attacks in the US of 11 September, direct US and coalition military intervention in Afghanistan changed the character of this continuing war. Of course it did not transform the situation completely: resistant to change as ever, rival warlords sought to maintain their fiefdoms against intervention unless it could offer more by extending the chance of collaboration. However, there was now an undeniably international war inside Afghanistan. There was not much doubt that this was, for a few months, an international war in the sense of a war between sovereign states—the US-led coalition v the Taliban government of Afghanistan. In November to December 2001 the US-led intervention, and the military campaign of the Northern Alliance, toppled the Taliban regime, which had been supported by al-Qaeda. This military action was widely though not universally viewed as a justifiable response to the Taliban for having allowed Afghan territory to be used for preparing attacks on the US, and additionally had the effect of freeing Afghanistan from an unpopular regime. Initially there was much popular support in Kabul and elsewhere for the incoming forces of the International Security Assistance Force (ISAF), but this situation was to change.

17. The international war of October to December 2001 had been superimposed on two other more enduring conflicts: the non-international armed conflict of the Taliban v Northern Alliance, and the US-led struggle against al-Qaeda terrorists. Both of these "other conflicts" continued. The war against al-Qaeda and related terrorists, who were now based in Pakistan as well as Afghanistan, carried on without interruption. In addition, there was growing resistance in southern Afghanistan to the new regime. This insurgency began relatively slowly, so that its seriousness was not recognized for some time.

18. How should this resistance be characterized? It is commonly labelled as the Taliban insurgency—a description which may conceal the possibilities that the sources of support for the insurgency have been more numerous than the label "Taliban" suggests, or that the ideology of the Taliban may have evolved. The insurgent movement has drawn on elements of both Afghan and Pashtun nationalism; it has operated alongside traditional forms of social organization and systems of justice; its recruiting has been facilitated by Afghanistan's high levels of unemployment and by the fact that it is able to pay its soldiers; and its willingness to support poppy cultivation not only increases its acceptance in certain provinces but also exposes the incoherence of the policies of the various NATO countries on this issue.[23] None of this is to suggest that all those forces labelled Taliban should be seen simply as heroic patriots or as Pashtun traditionalists. Ahmed Rashid has written:

19. Into this ongoing conflict a new element was added from 2005 onwards: the involvement in combat activities of contingents of the NATO-led ISAF. The original authorization of ISAF in 2001 had been "to assist the Afghan Interim Authority in the maintenance of security in Kabul and its surrounding areas, so that the Afghan Interim Authority as well as the personnel of the United Nations can operate in a secure environment".[25] Initially in January 2002 the UK took the lead in organizing ISAF, followed at six-monthly intervals by other "lead states" until NATO as such took over in August 2003. ISAF's remit gradually extended across Afghanistan, and in some provinces came to involve direct combat.[26] By 2006 ISAF comprised troops from 32 countries. Those deployed in the southern provinces of Afghanistan became increasingly geared to a counterinsurgency campaign. This campaign had not been part of ISAF's original role: the transition to it, involving a gradual stretching of the initial mandate, resulted in some unavoidably uneven burden-sharing between NATO member states. Thus NATO had put itself in the unenviable position of staking its impressive reputation on the outcome of a distant and little-understood war in a country well known to be a graveyard for foreign military adventures.

20. The outsiders—military and civilian—involved in Afghanistan since 2001 have generally had short-term tours of duty. This has serious consequences. Few of them have learned the relevant languages, and there is remarkably limited institutional memory, especially as regards knowledge of local communities and political traditions.

21. One special feature of the ongoing war in Afghanistan that distinguishes it from certain other post-Cold War US involvements has been that the US-led forces had at the start significant allies within the country: originally the Northern Alliance, then the Government of Afghanistan. This made the Afghan involvement different from some of the other conflicts in which the US has been involved, including Iraq in the first years of the US-led presence and Somalia over a much longer period, in neither of which were there strong local forces in place with which to work.

22. However, this apparently favourable situation had inherent limitations and was vulnerable to change. Even after its capture of Kabul in December 2001, the Northern Alliance, which at the best of times was an unstable coalition, never controlled all of Afghanistan. The Afghan authorities conspicuously lacked the bureaucratic back-up that provides the essential underpinning of most governments around the world. The Pashtuns generally resented the Northern Alliance's US-assisted victory; and when, around 2003-04, the Pashtuns came back strongly in the government (thanks to the new constitution and law on political parties), Afghan opinion critical of the US found a voice. Indeed, the boot was now on the other foot, with minorities complaining of Pashtun nationalism and structural exclusion. In short, the social foundations of the foreign presence in Afghanistan proved to be weaker than they had first seemed in 2001-02.

23. In legal terms, there has been a tendency to focus attention on the question of whether particular aspects and phases of the ongoing war in Afghanistan should be characterised as "international", "non-international" or something else. The main problem with debates on this topic is that the passion for pigeon-holing risks obstructing understanding of a complex reality. Actually the wars in Afghanistan have been all of these things. If one were forced to apply a single label to all their aspects, it would probably be "internationalized civil war"—an under-explored but important category of wars. Yet whichever of these terms is adopted has only limited relevance to, or effect on, policy-making. Although technically it is true that more rules apply to international war than to non-international armed conflict, in this case most of the powers involved in the war do at some level recognize the need for restraint in the conduct of the war—a matter discussed further below.

The Pakistan factor

24. Afghanistan's neighbours—including China, Iran, Tajikistan, Turkmenistan and Uzbekistan—all have legitimate interests in the country and its long-running conflicts. Many other states, including India and Russia, also have legitimate interests in whether Afghanistan can manage to stay together, make progress in development, and attract refugees back. Of all the relationships with other states, that with Pakistan is the most complex, and has contributed most to Afghanistan's ongoing divisions.

25. All borders are artificial constructs created in peoples' minds. Thus in itself it is hardly a remarkable statement to say that the border between Afghanistan and Pakistan—the Durand Line imposed by the British on a reluctant Afghan government in 1893—is artificial. What is significant about this border is that Pashtuns on either side of the line view it as artificial. This does not mean that they are committed to a definite idea of a new state of "Pashtunistan", separate from both Pakistan and Afghanistan. Rather it means that conflicts on either side of the line immediately acquire a cross-border and therefore an international dimension. What creates an issue, both for governments and peoples, is its chronic porousness, the existence of linked conflicts on both sides of it, the strength of the bonds of common identity and experience that link Pashtuns in Afghanistan and Pakistan, and the inherent weakness of both of these states. It is too simple to say that the frontier areas of both states are ungovernable: they have their own systems of authority, which leave little room for control by the state.

26. Pakistan's Federally Administered Tribal Areas (FATA), which run along the border with Afghanistan, remain almost completely outside the control of the Pakistan government, and have provided fertile ground for the exercise of dominance by the Taliban and al-Qaeda. They are a legacy of empire. The British had also practised containment, occasional chastisement and periodic negotiation; and resistance meant that a final occupation was simply too expensive to justify in imperial terms. One remarkable feature of this situation is that successive Pakistan governments have had no counter-insurgency policy in these areas. Occasional sweeps and demonstrations of firepower are in no way a substitute for a serious policy aimed at gaining a degree of consent from the population or the power-brokers. The US has not used the power that ought to come with its generous support for Pakistan to persuade it to adopt a strategy in these areas. The FATA constitutes a haven for terrorists that is in some respects comparable to the one that existed in Afghanistan before 2001.

27. Overlapping with all this, and compounding the problem of relations between the two countries, is the fact that opinion in Pakistan generally on matters relating to the use of force has never favoured the US vision of the "war on terror". A BBC World Service Poll in 23 countries, published in September 2008, when asking respondents to indicate their feelings regarding al-Qaeda, found high levels of support for it in Pakistan. This was combined with a mere 17% of Pakistanis stating that they had negative views of al-Qaeda—the lowest proportion of respondents in any of the countries polled.[27] However, this may reflect more a desire to take an anti-US position than an acceptance of terrorist bombings. Indeed, in four weeks in Autumn 2008 an anti-terror petition in Pakistan—"This is Not Us"—attracted almost 63 million signatures in what is possibly the biggest such lobby effort anywhere in the world.[28]

28. The Pakistan connection has deeply affected events in Afghanistan in all the wars there since the Soviet intervention in 1979. Throughout, Pakistan's Interservices Intelligence (ISI) has had a major, and not always controlled, role. In the 1980s Pakistan, with massive Western support, provided crucial assistance for the anti-Soviet rebels in Afghanistan. Then from 1994 onwards there was extensive Pakistani official support for the Taliban movement in Afghanistan.[29]

29. In the ongoing war in Afghanistan a number of consequences in the security field have flowed from the Pakistan connection. The first is that, since Pashtuns on either side of the border are more likely than most others to view the Western military presence in Afghanistan as illegitimate, there is inevitably a trans-border hinterland for the insurgency. Second, since Pashtuns play a large part in the Pakistan Army, and in the Frontier Corps which comes under the Ministry of Interior, there are built-in difficulties in Pakistan government attempts to impose the capital's rule by force on the various Pashtun-inhabited areas.[30] As a consequence of these two factors, the insurgency in southern Afghanistan is likely for the foreseeable future to have safe base areas inside Pakistan. In sum, like so many border regions in the world, the Pakistan-Afghanistan border presents excellent opportunities for the organisation and continuation of insurgency.

30. This creates the third consequence of the Pakistan connection: the strong pressure on US military leaders to take the war unilaterally into the territory of Pakistan. US policy towards Pakistan notoriously lacks strategic coherence.[31] The fact that the US considers the Pakistani authorities unreliable, with certain elements willing to pass on intelligence to the US's enemies, means that the US military role on the territory of Pakistan cannot be based on close military cooperation. As a result, US military action in Pakistan is bound to be perceived as an infringement of Pakistan's sovereignty. The US killings of Pakistani soldiers in several such incidents, and the strong reactions to this in Pakistan, confirmed the chaotic and inflammatory character of the situation.[32] George Bush's presidential order of July 2008, authorizing US strikes in Pakistan without seeking the approval of the Pakistan government, while an understandable reaction to a troubling situation on the border, risks further destabilizing a country that is a crucial if deeply flawed ally.[33]

C. THE MANY ROLES OF THE UNITED NATIONS IN AFGHANISTAN SINCE 1979

31. The United Nations has a long history of involvement in the conflicts in Afghanistan—and such a continuing commitment there that failure would impact on the UN's already tarnished reputation. There have been three main phases of UN involvement: during the Soviet War 1979-89, in the largely civil war of 1990-2001, and in the war since 2001 that continues today.

UN roles during the Soviet War in Afghanistan, 1979-89

32. During the Soviet war (1979-89) the main action was not in the Security Council: there the Soviet Union could veto any direct UN involvement in the conflict, so Council referred the matter to the General Assembly under the UN's "Uniting for Peace" procedure.[34] From then on the conflict was mainly handled in the General Assembly and in the office of the Secretary-General. In January 1980 the General Assembly called for "the immediate, unconditional and total withdrawal of the foreign troops from Afghanistan".[35] Subsequently, under the auspices of the Secretary-General, the UN initiated a "good offices" function to assist negotiations involving the Afghan and Soviet governments on the one hand, and Pakistan on the other. This led eventually to the April 1988 Geneva Accords on Afghanistan, which were a crucial landmark in the ending of the Cold War.[36] Later in 1988 the UN Good Offices Mission in Afghanistan and Pakistan (UNGOMAP) was established.[37] This was the first UN peacekeeping mission since the establishment of UNIFIL in Lebanon in March 1978—evidence of the key part played by Afghan events in the post-Cold War re-emergence of the UN.

33. At the same time, the process of ending the Soviet involvement posed a classic dilemma for the UN. The internal conflict presented the delicate question of the extent to which the UN, as an organization of governments, could be seen to negotiate with rebel forces that were battling it out throughout the country. As Secretary-General Javier Pérez de Cuéllar put it in 1988, it would be "against our philosophy to be in touch with the enemies of governments".[38] Yet that is exactly what the UN started to do in the following year, in the attempt to facilitate a comprehensive political settlement and to set up a broad-based government. In presenting the UN with this dilemma, the war in Afghanistan was truly characteristic of the post-Cold War era. The UN's limited success in persuading the parties to a largely internal conflict to agree a peace settlement would also be a harbinger of things to come.

UN roles in the continuing civil war, 1990-2001

34. The continuing civil war following the Soviet departure presented a difficult challenge for the UN. By March 1990 UNGOMAP, having completed its key mission of observing the Soviet withdrawal, was wound up. Yet there was a chaotic situation on which the Security Council, the General Assembly, and the Special Representative of the Secretary-General had remarkably little capacity to influence events. The General Assembly established the UN Special Mission to Afghanistan (UNSMA) in 1993, in the distant hope of facilitating national rapprochement and reconstruction.[39] The post of Special Representatives to Afghanistan, who headed the mission, was held successively by two of the ablest and most experienced UN trouble-shooters, Lakhdar Brahimi and Francesc Vendrell. However, they could achieve little in UNSMA's lifetime, which ended in 2001-02.

35. At the same time the Security Council gradually became more active over Afghanistan. One month after the Taliban came to power in September 1996 the Council passed a resolution which staked out a number of critically important positions. As well as stating its unsurprising conviction that "the United Nations, as a universally recognized and impartial intermediary, must continue to play the central role in international efforts towards a peaceful resolution of the Afghan conflict", it called for an immediate end to all hostilities, denounced the discrimination against girls and women, and called for an end to the practices that had made the country a fertile ground for drug-trafficking and terrorism.[40] Then in August 1998, following an upsurge in the fighting between the Taliban and the Northern Alliance, the Security Council passed a further resolution, again setting out some useful principles. It noted that there was "a serious and growing threat to regional and international peace and security, as well as extensive human suffering, further destruction, refugee flow and other forcible displacement of large numbers of people"; it expressed concern at "the increasing ethnic nature of the conflict"; it deplored the fact that, despite numerous UN pleas, there was continuing foreign interference; condemned the attacks on UN personnel in the Taliban-held areas; condemned the Taliban's capture of the Iranian Consulate-General in Mazar-e-Sharif; reaffirmed that "all parties to the conflict are bound to comply with their obligations under international humanitarian law"; and demanded the Afghan factions "to refrain from harbouring and training terrorists and their organizations and to halt illegal drug activities".[41] In October 1999, it imposed sanctions on the Taliban regime in Afghanistan:[42] arguably this decision undermined whatever was left of the UN's good offices mission. The Council may have been ineffective in the 1990s civil war in Afghanistan, but it was certainly not asleep. Some of the positions that it had staked out would be important for the future, in that they provided a basis for subsequent tough action against the Taliban, and for serious efforts to rebuild the Afghan state.

UN roles in the war since September 2001

36. The attacks on the USA on 11 September 2001 were a clear indication of the connection between Afghanistan and international security. In 1996 and 1998 the Council had warned of the terrorist danger in Afghanistan. Now it was to have a more prominent role, giving implicit authorization to the US-led use of force, and becoming deeply involved in the subsequent reconstruction of Afghanistan.

37. The most significant acts of the Council after 9/11 took the form of two resolutions which had profound implications for the management of international security issues. The first, Resolution 1368 passed the day after the attacks, by recognizing "the inherent right of individual or collective self-defence in accordance with the Charter" implicitly accepted the proposition that it could be lawful for a state to take action against another state if the latter failed to stop terrorist attacks being launched from its territory. The same resolution called on all states to bring the perpetrators to justice, and to co-operate to prevent and suppress terrorist acts.[43]

38. In this resolution the Council accepted that a right of self-defence could apply to a state when it was attacked by a non-state entity. To those who believe that action against terrorists should be confined to police methods, this was controversial. However, the resolution was passed in the specific and hopefully unique circumstances of 9/11, when the Taliban regime was refusing to take any action against the terrorists in their midst. The resolution does not mean that there is or should be general Council approval of responding to terrorist attacks by cross-border military actions, or that such action should generally be viewed as lawful. The history of such responses is dismal, as evidenced for example by the Hapsburg attempt to wipe out the terrorist "hornets' nest" in Serbia in 1914, and the various Israeli counter-terrorist operations in Lebanon in the past 30 years. The effectiveness of the military campaign in Afghanistan in late 2001 is an exception to the proposition that it is unwise to attack states from which terror originates. While that remains strong, it is bound to face severe challenges if state-sponsored or state-tolerated terrorism continues to be a major feature of international politics.

39. The second key resolution passed by the Council in September 2001, Resolution 1373, recognized "the need for States to complement international cooperation by taking additional measures to prevent and suppress, in their territories through all lawful means, the financing and preparation of any acts of terrorism". It then indicated the remarkable extent of such measures, and the key role of the Council in overseeing them. It used strong language—the Council "decides that all states shall" take action, rather than merely calling on them to do so.[44] The General Assembly—often wary of any increase in the Security Council's powers—was duly nervous but did not go against the Council's approach.[45] It remains possible that in the long run the greatest effect of Afghanistan on international security will be that it compelled the Council to take on a more intrusive role in relation to states than had ever previously been contemplated.

40. Yet the actual role of the Council in the events following the 9/11 attack was limited. True, its resolutions and other actions were important for the international legitimacy of the US-led military action in Afghanistan and for the attempts to build up a post-Taliban system of government there.[46] However, there was no way in which the Council could have been centrally involved in mustering and commanding the military coalition that resulted in the closing of the al-Qaeda bases in Afghanistan and the removal of the Taliban from power in Kabul. The most striking feature of the Council's role in the hostilities of late 2001 is its limited character.

41. Following the installation of the Karzai government in Kabul on 22 December 2001, the two main tasks facing the new government and its outside backers were perceived to be reconstruction, and the provision of security. The UN was widely seen—even by the US administration—as being pivotal in tackling these tasks. The key statement of this period, which did much to define the role not just of the UN but of the international community generally, was made by Lakhdar Brahimi, Special Representative of the Secretary-General for Afghanistan. In discussing the planned UN Assistance Mission in Afghanistan (UNAMA), he famously said:

42. This immediately raises the question of whether a light footprint is indeed possible in a country with such a limited, and distrusted, a state structure as that of Afghanistan. The concept was inevitably buffeted by events and modified to the point where some did not recognise it. Within a year or two, a reviving insurgency, and major military operations on Afghan territory by the US and NATO, created the dual risks that the footprint would be perceived as heavy, and that UNAMA would be seen as powerless to implement important parts of its mandate. It was not the only part of the UN system that faced the problem of appearing to be partial, or powerless, or both. As Gilles Dorronsoro has pointed out in a critical survey of the UN Security Council's roles in Afghanistan up to the end of 2006, "the direct involvement by Permanent Members of the Security Council in a counter-insurgency war has resulted in the Council being silent on specific violations of international humanitarian law".[48]

43. In the years since 2002 in which it has operated in Afghanistan, UNAMA has sought to assist political and economic transition and the rule of law. The report of its activities up to March 2008 presented a sobering picture:

D. FITTING MILITARY DOCTRINE AND PRACTICE TO AFGHAN REALITIES

44. The limitations of military doctrines and practice are often exposed, not by arguments, but by events. Thus it was mainly events in Iraq and Afghanistan that exposed the inadequacies of the so-called "revolution in military affairs"—an idea that was popular in the US from the mid-1990s until at least 2003.[50] Afghanistan was always likely to be a difficult theatre of operations for outside military forces. Seeing this (and perhaps also because he did not want an ongoing distraction from the future invasion of Iraq, for which he was already lobbying) Paul Wolfowitz said in November 2001:

45. Many problems have been encountered in implementing and adapting military doctrine and practice in face of Afghan realities. Three issues considered here are the role of air power, the complexities of operating in an alliance framework, and the appropriateness or otherwise of counterinsurgency (COIN) doctrine. The first two are touched on here briefly: more attention is paid to the third.

Air power in Afghanistan

46. Ever since October 2001 air power (which mainly means US air power) has played an important part in military operations in Afghanistan. The apparent success of the use of air power in October-December 2001 was deceptive: a major factor in the Taliban's defeat was the advance of ground forces—those of the Northern Alliance. Since then, the role of air power in the Afghan conflict has been a subject of contestation, principally between the Army and Marines on the one hand, and the USAF on the other. A key issue has been whether air power is a major instrument in its own right, or is mainly useful in supporting ground forces. Self-evidently, the US and NATO ground forces in Afghanistan, widely dispersed and few in number, frequently need air power in support of their ground operations. Indeed, tactical air support has been vital to any success they have had, and has often saved the small numbers of ISAF forces from being overwhelmed. In military terms, a "light footprint" on the ground inevitably means a heavy air presence.

47. Those planning coalition military operations in Afghanistan have shown awareness of the dangers of reliance on air power—especially of the adverse consequences of killing civilians. On occasion they have even claimed to have set an aim of no civilian casualties.[52] While this aim actually goes further than the strict requirements of existing law applicable in an international armed conflict, in practice it has not been achieved. Part of the difficulty is that the very definition of civilian is problematic in a war such as that in Afghanistan. In addition, many other factors have prevented realization of the aim of no civilian casualties: shortage of ground forces, different approaches of individual commanders, poor intelligence, the heat of battle, weapons malfunction, the co-location of military targets and civilians, and the frayed relationship between ground and air forces operating in Afghanistan.[53] A Human Rights Watch report in September 2008 summarized the situation thus:

In the past three years, the armed conflict in Afghanistan has intensified, with daily fighting between the Taliban and other anti-government insurgents against Afghan government forces and its international military supporters. The US, which operates in Afghanistan through its counter-insurgency forces in Operation Enduring Freedom (OEF) and as part of the NATO-led International Security Assistance Force (ISAF), has increasingly relied on airpower in counter-insurgency and counter-terrorism operations. The combination of light ground forces and overwhelming airpower has become the dominant doctrine of war for the US in Afghanistan. The result has been large numbers of civilian casualties, controversy over the continued use of airpower in Afghanistan, and intense criticism of US and NATO forces by Afghan political leaders and the general public.

48. That last figure needed to be increased when it was revealed in October 2008 33 civilians had been killed in a single US airstrike on 22 August. Such incidents do serious damage to the coalition cause. Largely as a result of the long history of such incidents, there has been a strong anti-coalition reaction. Already in 2006 the Afghan parliament had demonstrated its concern about coalition military actions, and such expressions of concern have subsequently become more frequent. Meanwhile, President Hamid Karzai, whose authority has been diminishing, has made a number of criticisms of the coalition forces, calling for an end to civilian casualties, and even stating that he wanted US forces to stop arresting suspected Taliban members and their supporters.[55]

The NATO Framework

49. From 2001 onwards the US has operated in Afghanistan with coalition partners and, especially since August 2003, with the formal involvement of NATO. Indeed, in Afghanistan NATO is involved in ground combat operations for the first time in its history, far from its normal area of responsibility and against a threat very different from the one it had been created to face. The NATO involvement in Afghanistan is widely, but perhaps not wisely, viewed as "a test of the alliance's political will and military capabilities".[56] It is an exceptionally hard test. Indeed, the implication that the future of the alliance hangs on this test is reminiscent of earlier views that US credibility was on the line in Vietnam.

50. NATO's involvement in Afghanistan is in sharp contrast to its conduct during the Cold War. In that period it repeatedly and studiously avoided involvement in colonial conflicts—the French wars in Indochina and Algeria, the Portuguese Wars in Africa, the British in Malaya, the Dutch in Indonesia and so on. Its individual members were involved in these, but the alliance was not. NATO also avoided involvement in post-colonial conflicts or—as in Cyprus, limited itself to an essentially diplomatic role. Now in Afghanistan, which has all the hallmark features of post-colonial states undergoing conflict—especially the lack of legitimacy of the constitutional system, government and frontiers—NATO became engaged, all with little public debate.

51. The NATO role in Afghanistan began in a problematic way, and so it has continued. On 12 September 2001, the day after the 9/11 attacks, the NATO Council stated: "If it is determined that this attack was directed from abroad against the United States, it shall be regarded as an action covered by Article 5 of the Washington Treaty, which states that an armed attack against one or more of the Allies in Europe or North America shall be considered an attack against them all".[57] When the US gave this offer the brush-off, preferring to have a "coalition a" la carte" in which there would be no institutional challenge to its leadership, there was disappointment and irritation in Europe. The war in Afghanistan in October to December 2001, while it was effectively conducted under US leadership, was also one chapter in the story of the declining size of US-led wartime coalitions.

52. However, NATO rapidly came back into the picture, not least because the US came to recognize the need for long-term assistance in managing societies that had been freed from oppressive regimes by US uses of force. NATO has been directly involved in Afghanistan at least since 9 August 2003, when it took formal control of the International Security Assistance Force, which had originally been established under UK leadership in January 2002. It was in the autumn of 2003 that an upsurge of violence began that was part of a deteriorating security situation.[58] Since 2006 ISAF has undertaken an expanded range of responsibilities in Afghanistan, involving combat as well as peacekeeping, in an expanded area which includes provinces in which conflict is ongoing.

53. ISAF's notably broad UN Security Council mandate involves it in a wide range of activities, including military and police training. Many of its activities are carried out through Provincial Reconstruction Teams (PRT)—civilian-military units of varying sizes designed to extend the authority of the central government, provide security, and undertake infrastructure projects. There are 26 PRTs in 26 of the country's 34 provinces. Operating under different lead states, with 12 of the 26 led by the US, the resources and tasks of the PRTs have varied greatly.

54. Not surprisingly, there have been controversies about numerous aspects of the overall ISAF mission. Four key problems concern the coherence or otherwise of the different members of ISAF; the problematic command and control arrangements; differences over detainee treatment; and the difficulty of raising forces.

55. The lack of coherence of the approaches taken by different foreign forces in ISAF and their governments at home is evident. Different contributing states have different visions of ISAF's role. The most obvious difference is that the US, UK and Canada tend to see it, albeit with some variations within each of these countries, as encompassing a counterinsurgency operation, while Germany and some others see it more through the lens of a stabilization mission. These positions are not polar opposites, and each may have validity in different provinces of Afghanistan, but the clash of perspective on this issue does not assist cooperation of forces in difficult operations. Daniel Marston has gone so far as to conclude: "As of 2007, the main problem impeding coalition forces' successful application of counterinsurgency was decentralization of responsibility".[59]

56. The complexity of the command and control arrangements in Afghanistan is greater than that in past counter-insurgency campaigns. Debates about this have inevitably reflected the US desire that more contingents in ISAF should become directly involved in combat operations, and the concern of some contributors that this should not happen. Although ISAF is now under a US commander, and the continuous rotation of senior posts is ceasing, the arrangements for coordinating the work of these three distinct forces continue to pose problems.[60]

57. The important, and scandal-ridden, matter of treatment of detainees is another issue on which there are differences of approach. Anxious not to be associated with shocking US statements and practices in this matter, and insufficiently staffed and equipped to hold on to the prisoners they capture, other NATO members have drawn up separate agreements with the Afghan authorities, embodying a variety of different approaches to how they should be treated once in Afghan hands. There are serious concerns that some detainees handed over to the Afghan authorities on this basis have been maltreated.[61]

58. The provision of forces in the numbers required for ISAF has been a highly contentious matter within NATO states. The coalition of forces acting in support of the Afghan government consists of three basic elements. The first is the Afghan National Army which has a modest manpower level of about 57,000—a fact which has led to US accusations that the Afghan government has been slow in building up its army. The second is ISAF, which now comprises some 51,350 troops from forty NATO and non-NATO countries. Much the largest contingents are those of the US, with 19,950 troops, and the UK, with 8,745. The third basic element is the force of well over 10,000 troops (almost all of them American) who are part of the US Operation Enduring Freedom, which focuses particularly on the counter-terrorist mission in Afghanistan.[62] Granted the scale of the problems in Afghanistan, all these numbers are widely seen as low, yet in many NATO member states there is a reluctance to increase the commitment. Opinion polls in five NATO member states with a high level of involvement in Afghanistan show the public to be highly sceptical about it.[63] An increase in such numbers risks running into opposition in many NATO states, and also further antagonizing Afghan opinion. If counterinsurgency theory is a guide, a massive increase in such numbers would seem to be called for. So how reliable a guide is the writing on counterinsurgency?

Counterinsurgency doctrines and practice

59. Contrary to myth, counterinsurgency campaigns can sometimes be effective. Doctrines and practices of counterinsurgency—the best of which draw on a wide and varied range of practice—have a long history.[64] The revival of counterinsurgency doctrine in the past few years has been driven primarily by events in Iraq, but also, if to a lesser degree, by the development of the insurgency in Afghanistan. This revival of COIN is hardly surprising. The response of adversaries to the extraordinary pattern of US dominance on the battlefield was always going to be one of unconventional warfare, including the methods of the guerrilla and the terrorist; and in turn the natural US counter-response was to revive the most obviously appropriate available body of military doctrine.

60. The key document of the US revival of COIN doctrine is the US Army Field Manual 3-24.[65] It is very much an Army and Marine Corps manual: the USAF refused to collaborate in the exercise. Improbably for a military-doctrinal document, it has been in demand in the US. It has been heavily accessed and downloaded on the web, and is also available as a published book from a major university press.[66] Although it has some flaws, explored further below, it is a significant contribution to COIN literature.

61. By contrast, the UK has not produced any major new manual. This is partly because, much more than their US counterparts, the British had extant doctrine.[67] It is also because there was some opposition to COIN doctrine on the grounds that it would result in the same hammer being used on every problem. As a result there has not yet been a UK equivalent of FM 3-24. The Ministry of Defence's short (23 pages) Joint Discussion Note of January 2006, on The Comprehensive Approach, is a more general survey intended to be relevant to a wide range of operations: the word "counterinsurgency" does not appear in it.[68] It was followed in 2007 by a paper on Countering Irregular Activity.[69] This document, which has not gone into general public circulation and has not been greeted with enthusiasm in the Army, "seeks to instruct military personnel about counter-insurgency as a whole and about associated threats, and emphasizes the need for military activity to be part of a comprehensive approach involving all instruments of power".[70] This summary, by Sir John Kiszely, until 2008 Director of the Defence Academy of the UK, is immediately followed by a down-to-earth reminder that "every insurgency is sui generis, making generalizations problematic".[71] This important point has been emphasized by military professionals on both sides of the Atlantic.

62. The "comprehensive approach", central to both the US and UK doctrines, essentially means the application of all aspects of the power of the state within the territory of which the insurgency is being fought. The apparent assumption that there is a state with real power is the key weakness of the approach, especially as it applies to Afghanistan. Before exploring this in more detail, it may be useful to glance at the problematic nature of assumptions about the political realm in the counterinsurgency doctrines inherited from past eras.

63. The US manual revives and updates doctrines that were developed in the Cold War years in response to anti-colonial insurrections (some of them involving leadership by local communist parties). It relies especially heavily on two sources from that era.[72] The first is David Galula's Counterinsurgency Warfare—one of the better writings of the French thinkers on guerre révolutionnaire.[73] The second is Sir Robert Thompson's Defeating Communist Insurgency.[74] Both works had placed emphasis on protecting populations as distinct from killing adversaries—a crucial distinction which implies a need for high force levels.

64. According to the Introduction, FM 3-24 aspires to "help prepare Army and Marine Corps leaders to conduct COIN operations anywhere in the world".[75] This might seem to imply a universalist approach, but the authors emphasise that each insurgency is different. The foreword by Generals Petraeus and Amos is emphatic on this point: "You cannot fight former Saddamists and Islamic extremists the same way you would have fought the Viet Cong, Moros, or Tupamaros; the application of principles and fundamentals to deal with each varies considerably".[76] FM 3-24 is also emphatic on the importance of constantly learning and adapting in response to the intricate environment of COIN operations—a point which strongly reflects British experience.[77]

65. Past exponents of COIN doctrine have generally placed heavy emphasis on achieving force ratios of about 20 to 25 counterinsurgents for every 1000 residents in an area of operations. Noting this, the manual states: "Twenty counterinsurgents per 1,000 residents is often considered the minimum troop density required for effective COIN operations; however as with any fixed ratio, such calculations remain very dependent upon the situation".[78] This emphasis on force ratios is controversial. In any case, in Afghanistan there appears little chance of achieving such numbers. If the entire country with its 31 million inhabitants were to be viewed as the area of operations, a staggering 775,000 counterinsurgents would be needed. Even if the area of operations is defined narrowly, and even allowing for the fact that not all have to be NATO troops, the prospects of getting close to the force ratio indicated must be low.

66. A flaw in some, but not all, past counterinsurgency doctrine has been a lack of sensitivity to context and, in some cases, an ahistorical character. Some specialists in counterinsurgency have seen their subject more as a struggle of light versus darkness than as a recurrent theme of history or an outgrowth of the problems of a society. Examples of such an ahistorical approach to the subject can be found in the French group of theorists writing in the 1950s and early 1960s about guerre révolutionnaire. Some of these theorists denied the complexities—especially the mixture of material, moral and ideological factors—that are keys to understanding why and how guerrilla and terrorist movements come into existence. Colonel Lacheroy, a leading figure in this group and head of the French Army's Service d'Action Psychologique, famously stated: "In the beginning there is nothing".[79] Terrorism was seen as having been introduced deliberately into a peaceful society by an omnipresent outside force—namely international communism. It is a demonological vision of a cosmic struggle in which the actual history of particular countries and ways of thinking has little or no place.

67. A related fault in some counterinsurgency writing was the tendency to distil general rules of counterinsurgency from particular struggles and then seek to apply them in radically different circumstances. The campaign in Malaya in the 1950s, because it was successful in ending a communist-led insurgency, was often upheld as a model, and is described favourably in the US Field Manual.[80] Certain lessons drawn partly from Malaya were subsequently applied by the British in Borneo and Oman with some effect. However, successes such as that in Malaya can be great deceivers. Attempts were made to apply the lessons of Malaya in South Vietnam in the 1960s.[81] These largely failed. The main reason for failure in South Vietnam was that conditions in Vietnam were utterly different from those in Malaya. In Malaya the insurgency had mainly involved the ethnic Chinese minority, and had never managed to present itself convincingly as representing the totality of the inhabitants of Malaya. The insurgency was weakened by the facts that the Chinese minority was distinguishable from other segments of society; Malaya had no common frontier with a communist state, so infiltration was difficult; and the British granting of independence to Malaya undermined the anti-colonial credentials of the insurgents. In South Vietnam, by contrast, the communist insurgents had strong nationalist credentials, having fought for independence rather than merely having power handed to them by a departing colonial power.[82] At the heart of the US tragedy in Vietnam was a failure to recognise the unique circumstance of the case—that in Vietnam, more than any other country in South-East Asia, communism and nationalism were inextricably intertwined.

68. One lesson that could have been drawn from the Malayan case is that it is sometimes necessary to withdraw to win. FM 3-24 places much emphasis on the fact that the US withdrew from Vietnam in 1973 only to see Saigon fall to North Vietnamese forces in 1975.[83] It does not note a contrary case: it was the UK promise to withdraw completely—a promise that was followed by the Federation of Malaya's independence in 1957—that contributed to the defeat of the insurgency in Malaya.[84] The value of such promises needs to be taken into account in contemporary COIN efforts and indeed COIN theory. This is especially so, as the idea that the US intended to stay indefinitely in Iraq and Afghanistan, and to build networks of bases there, had a corrosive effect in both countries and more generally. The decision of the Iraqi cabinet on 16 November 2008 that all US forces will withdraw from Iraq by 2011 is evidence that a guarantee of withdrawal is seen as a necessary condition (and not simply a natural consequence) of ending an acute phase of insurgency.

69. One weakness in the US manual, likely to be remedied in any future revisions, is the lack of serious coverage of systems of justice—especially those employed by the insurgents themselves. The references to judicial systems in FM 3-24 are brief and anodyne, almost entirely ignoring the challenge posed by insurgents in this area.[85] Insurgencies commonly use their own judicial procedures to reinforce their claims to be able to preserve an existing social order or create a better one. The Taliban have always placed emphasis on provision of a system of Islamic justice.[86] In the current conflict, taking advantage of the fact that the governmental legal system is weak and corrupt, they have done this effectively in parts of Afghanistan.

70. This leads to a more general criticism. In addressing the problem of undermining and weakening insurgencies, both traditional COIN theory, and its revived versions in the 21st century, place emphasis on the role of state institutions: political structures, the administrative bureaucracy, the police, the courts and the armed forces. The institutions are often taken for granted, and assumed to be strong. Indeed, the current British COIN doctrine stemmed from a project started in 1995 to capture the lessons and doctrine from Northern Ireland. A common criticism of much COIN practice is that it was enthusiastically pursued by over-powerful and thuggish states, especially in Latin America.[87]

71. Today, COIN theories risk being out of joint with the realities of assisting the so-called "failed states" and "transitional administrations" of the 21st century. These problems are not new—one of the problems that undermined US COIN efforts in Vietnam was the artificiality and weakness of the coup-prone state of South Vietnam. Yet the central fact must be faced that in the two test-beds of the new COIN doctrines of recent years, Iraq and Afghanistan, state institutions have been notoriously weak—in Iraq temporarily, and in Afghanistan chronically. Indeed, in post-colonial states generally, where insurgencies are by no means uncommon, indigenous state systems tend to be fragile and/or contested. The role of the state in people's lives, and in their consciousness, may be thoroughly peripheral or even negative.[88] So when the US manual speaks of "a comprehensive strategy employing all instruments of national power" and stresses that all efforts focus on "supporting the local populace and HN [Host Nation] government",[89] it is necessary to remind ourselves that support for government is not exactly a natural default position for inhabitants of countries with such tragic histories as Iraq and Afghanistan. On the other hand, General Petraeus worked on the manual after completing two tours of duty in Iraq, with an eye to applying it there, and then did so to some effect when he was Commander of the Multinational Force—Iraq. In 2008 the Iraqi government is looking stronger than in the first years after the invasion. The fact that a government is weak in face of insurgency does not mean that it is necessarily fated to remain so.

72. Of the many critiques of the US revival of COIN doctrine, one of the most searching is an American Political Science Association review symposium published in June 2008.[90] Stephen Biddle of the US Council on Foreign Relations queried the manual's fundamental assumption when he stated that

73. Biddle also pointed out that the US manual has little to say about the comparative merits of waging COIN with large conventional forces as against small commando detachments, on the relative utility of air power in COIN, and on the willingness of democracies to support COIN over a long period. Further, the manual does not fit particularly well the realities of Iraq, where the insurgencies are far more regional and localised in character, and more fickle in their loyalties, than were many of the communist and anti-colonial insurgencies of earlier eras. As Biddle points out, the negotiation of local ceasefires between insurgents and US commanders has been of key importance in Iraq.[92] Such webs of local ceasefires, valuable despite their fragility, do not come from counterinsurgency doctrine. These criticisms are another way of saying what General Petraeus knows: that all doctrine is interim, and some parts are more interim than others.

74. The need to adapt doctrine, so evident in Iraq, applies even more strongly to Afghanistan, a subject about which the US manual says remarkably little.[93] The key issue is whether the revival of counterinsurgency doctrine really offers a useful guide in a situation where there are some distinct elements in the insurgencies, where negotiation with some of the insurgents may have a role, and where the state does not command the same loyalty or obedience that more local forces may enjoy.

75. After a difficult year in 2008, the US and Afghan governments began to place increased emphasis on local social structures. The US Ambassador to Afghanistan said at the end of the year that there was agreement to move forward with two programmes: first, the community outreach programme, "designed to create community shuras" (local councils); and second, the community guard program, which is "meant to strengthen local communities and local tribes in their ability to protect what they consider to be their traditional homes".[94] While neither programme was well defined, the move in this direction was evidence of willingness to rely on a less state-based approach than hitherto.

Judging progress in the war in Afghanistan

76. Judging progress in counterinsurgency wars is by nature a contentious task, and involves difficult questions about the appropriate methodologies. Sometimes unorthodox methods of analysis yield the most valuable answers. The war in French Indochina from 1946 to 1954 provided a classic case. When a French doctoral student, Bernard Fall (1926-67), went to Vietnam in 1953, the French authorities claimed that the war was going well, and showed maps and statistics indicating that they controlled a large proportion of the territory. But he soon realised that French claims about the amount of territory they controlled were exaggerated—or at least lacked real meaning as far as the conduct of government was concerned. He reached this conclusion both by visiting Vietminh-held areas, and by inspecting tax records in supposedly government-held areas: these latter showed a dramatic collapse in the payment of taxes, and thus indicated a lack of actual government control.[95] In Afghanistan, the long-standing lack of a tax collection system continues today. As Astri Suhrke has shown, taxation constitutes a uniquely small proportion—in 2005 it was only 8%—of all estimated income in the national budget.[96]

77. By one key measure serious progress may appear to be being made in the Afghan war. The numbers of refugee returns to Afghanistan since the fall of the Taliban regime at the end of 2001 are one possible indicator of a degree of progress. According to UNHCR, which played a key part in the process, between 1 January 2002 and 31 December 2007 a total of 4,997,455 refugees returned to Afghanistan, as follows:

2002 2003 2004 2005 2006 2007

1,957,958 645,864 879,780 752,084 387,917 373,852

78. This is the largest refugee return in the world in a generation. It is striking that even in 2006 and 2007—years of considerable conflict in parts of Afghanistan, the returns continued, if at a reduced rate. In the whole period 2002-07, the overwhelming majority of refugees have been in two countries: Iran, from which 1.6 million returned, and Pakistan, from which 3.3 million returned.[97] Impressive as the figures of this return are, four major qualifications have to be made:

— First, they have to be understood against the backdrop of the sheer numbers of Afghan refugees: at the end of 2007 Afghanistan was still the leading country of origin of refugees world-wide, with 3.1 million remaining outside the country. Thus in 2008, even after these returns, Afghan refugees constitute 27% of the entire global refugee population.

— Secondly, not all returns were fully voluntary. Within the countries of asylum there have been heavy pressures on these refugees to return, including the closing of some camps.

— Thirdly, the experience of many returning refugees has included lack of employment opportunities in Afghanistan, and in some cases involvement in property disputes. There has been mismanagement and corruption in the Afghan Ministry of Refugees and Returnees. Some returnees live in dire conditions in makeshift settlements. All this has created much disappointment, bitterness, and anti-government feelings.

79. Other developments confirm this sobering picture. The Afghan army remains relatively small, and highly dependent on outside support. As for the insurgent forces, they appear to have no shortage of recruits. Large numbers of fighters are able to cross into Afghanistan, mainly from Pakistan; and the Taliban can also employ many locals, especially in seasons when other work is in short supply. The fact that the estimated unemployment rate is 40% means that insurgents continue to have opportunities for recruitment. In Kabul and other cities, terrorist attacks, once rare, have become common. Serious observers have reported an atmosphere of disappointment and bitterness in Afghanistan in 2008.[99]

80. The UN Secretary-General's report of September 2008 summarizes the situation thus:

100

81. The Secretary-General's Report states bluntly that the number of security incidents rose to 983 in August 2008, the highest since the fall of the Taliban in 2001 and "represents a 44% increase compared with the same month in 2007". It also states: "While the main focus of the insurgency remains the southern and eastern parts of the country, where it has historically been strong, insurgent influence has intensified in areas that were previously relatively calm, including in the provinces closest to Kabul".[101] Overall the report is far from negative. It reports some successes in the campaign against poppy cultivation, and it strongly endorses the Afghanistan National Development Strategy, adopted at the Paris Conference in Support of Afghanistan, held on 12 June 2008. However, as an account of the state of progress in the war against the Taliban, it confirms the picture which has also been depicted by other sources. The latter include the sober report General David McKiernan, the top US Commander in Afghanistan, who, at the same time as seeking specific troop increases, has rejected simple notions, and indeed the terminology, of a military "surge";[102] and the US National Intelligence Estimate on Afghanistan, a draft version of which was leaked in October 2008, and which stated that the situation there was in a "downward spiral".[103] One grim statistic of the downward spiral is the casualty rate of IFOR and Operation Enduring Freedom forces in Afghanistan. Fatalities have increased each year from 57 in 2003 to 296 in 2008.[104]

82. As so often in counterinsurgency wars, the most useful assessments may be those of independent witnesses who, just as Bernard Fall did in French Indochina, have deep knowledge of a society and a healthy open-mindedness about the contribution that outside forces can make to security. Rory Stewart, who walked across Afghanistan in 2002, and later retired from the UK diplomatic service to run a charitable foundation in Kabul, is perhaps Fall's nearest equivalent today. He has argued that "we need less investment—but a greater focus on what we know how to do". He is specifically critical of increases in forces: "A troop increase is likely to inflame Afghan nationalism because Afghans are more anti-foreign than we acknowledge and the support for our presence in the insurgency areas is declining. The Taliban, which was a largely discredited and backward movement, gains support by portraying itself as fighting for Islam and Afghanistan against a foreign military occupation".[105]

E. CONCLUSIONS

83. Four kinds of conclusions follow. First, about the implications of Afghanistan for the UN; second, on the role of NATO; third, on international security generally; and finally, on the debate about policy choices that is emerging from the difficult experience of attempting to transform Afghanistan. These conclusions are based on the presumption that the present campaign in Afghanistan is unlikely to result in a clear victory for the Kabul government and its outside partners, because the sources of division within and around Afghanistan are just too deep, and the tendency to react against the presence of foreign forces too ingrained. The war could yet be lost, or, perhaps more likely, it could produce a stalemate or a long war of attrition with no clear outcome. The dissolution of Afghanistan into regional fiefdoms—already an accustomed part of life—could continue and even accelerate.

84. To some it may appear remarkable that Afghanistan has not reverted more completely to type as a society that rejects outside intrusion. Part of the explanation may be that this is not the only natural "default position" for Afghans: there have also been countless episodes in which Afghan leaders have sought, and profited from, alliances with outsiders. A second factor is the "light footprint" advocated by Brahimi: for all the limitations of this approach, and the many departures from it since it was enunciated in 2002 with specific reference to the UN Assistance Mission in Afghanistan (UNAMA), no one has convincingly suggested a better one. A third factor is that—notwithstanding the disastrous killings of civilians as a result of using air power—there has been a degree of restraint in the use of armed force: this has been important in at least slowing the pace of the process whereby the US and other outside forces come to be perceived as alien bodies in Afghanistan. The interesting phenomenon of application of certain parts of the law of armed conflict—namely the rules of targeting—as if this was an international war is part of this process.

The UN

85. A few conclusions on the UN's various roles in Afghanistan flow from this brief survey. First, the UN has some remarkable achievements to its credit in Afghanistan. It helped to negotiate the Soviet withdrawal from Afghanistan that was completed in 1989; ever since then it has remained engaged on the ground in Afghanistan; it gave a degree of authorization to the US-led effort to remove the Taliban regime in 2001; it has authorized ISAF and has provided a legitimate basis for its expanded roles throughout the country; it has been involved in the many subsequent efforts to help develop Afghanistan, not least by assisting in the various elections held there since 2001; it has assisted the largest refugee return to any country since the 1970s.

86. Second, despite these achievements, the UN's roles have been more limited than those of the US and its various partners, especially in matters relating to security. The fact that the UN's role in this crisis has been modest is not especially shocking. Neither the terms of the UN Charter nor the record of the Council justify the excessively high expectations that many have had in respect of the Council's roles. It was always a mistake to view the UN as aiming to provide a complete system of collective security even in the best of circumstances—and circumstances in and around Afghanistan are far from being favourable for international involvement.

87. Third, international legitimacy is never a substitute for local legitimacy. The Council's acceptance of regime change in Afghanistan was justified once the Taliban had refused to remove al-Qaeda, and did much to legitimise the aim of regime replacement, which could otherwise have seemed a narrowly neo-colonial US action. Yet there is a danger that such international conferrals of legitimacy can contribute to a failure to address the no less important question of securing legitimacy in the eyes of the audience that matters most: in this case, the peoples of Afghanistan and neighbouring countries.

NATO

88. The single most fateful impact of the current Afghan war on international security is the involvement of the NATO alliance in this distant, difficult and divisive conflict. It is truly remarkable that the reputation of the longest-lived military alliance in the world, comprised of states with fundamentally stable political systems, should have made itself vulnerable to the outcome of a war in the unpromising surroundings of Afghanistan. There is much nervousness about this among NATO's European members, and this may explain the reluctance of European leaders to make the kind of ringing statements that often accompany war. Knowing that the outcome of any adventure in Afghanistan is bound to be uncertain, they have wisely kept the level of rhetoric low.

89. There may be another reason for the reluctance of many leaders of European member states to make strong endorsements of their participation in the war in Afghanistan. Many of the claims that can be made in favour of the Afghan cause are also implicitly criticisms of the involvement in Iraq. From the start in 2001, the US-led involvement in Afghanistan and the subsequent involvement of ISAF have both had a strong basis of international legitimacy that was reflected in Security Council resolutions. In Afghanistan there was a real political and military force to support, in the shape of the Northern Alliance. In Afghanistan and Pakistan there were real havens for terrorists. In Afghanistan, up to five million refugees have returned since 2001. To speak about these matters too loudly might be to undermine the US position in Iraq, where the origins and course of the outside involvement have been different, and where the flow of refugees has been outwards. NATO leaders, anxious to put the recriminations of 2003 over Iraq behind them, may be nervous about highlighting the differences between Afghanistan and Iraq.

90. A major question, heavy with implications for international security, is: how are the setbacks experienced in Afghanistan to be explained, especially within NATO member states? The UN may be accustomed to failure, but NATO is not. So far, the tendency has been to blame Pakistan, the messy NATO command, the poor attention span of consecutive US governments, the unwillingness of NATO allies to contribute, the weakness of Karzai, the corruption of his government, the shortage of foreign money and troops—in other words, to blame almost everything except the nature of the project.

91. The various reasons that have been given cannot be lightly dismissed. For example, the lack of NATO unity in certain operational matters has been striking: the inability of member states to agree on a straightforward and defensible common set of standards for treating prisoners in the Afghan operations is symptomatic of deep divisions within the alliance. Political divisions have never been far from the surface, and will no doubt be projected into future explanations of what went wrong. Continental Europeans can convincingly blame the Americans and the British for having taken their eye off the ball in Afghanistan in 2002-03, foolishly thinking that the war there was virtually won and that they could afford to rush into a second adventure in Iraq. Americans can blame the Europeans for putting relatively few troops into ISAF, and being slow to back them up when the going got rough in 2006-08. A less blame-centred explanation might be that the reconstruction of Afghanistan, and the pursuit of counterinsurgency there, was always going to be an extremely difficult task; that there are limits to what outsiders should expect to achieve in the transformation of distant societies with cultures significantly different from our own; and that it never made sense to invest such effort in counterinsurgency in Afghanistan without having even the beginnings of a strategy for the neighbouring regions of Pakistan.

Impact on International Security

92. The problem of Afghanistan—including the complex interplay of international actors who have pursued their interests there—has had an impressive impact on international security issues in the past generation. It contributed to the end of the Cold War and indeed of the Soviet Union itself. The Taliban regime's failure to control al-Qaeda activities launched the US into the huge and seemingly endless "War on Terror", and also resulted in the UN Security Council claiming unprecedented powers to affect activities within states. The Afghan war has embroiled NATO in a largely civil war thousands of miles from its North Atlantic heartlands.

93. For the future, the greatest impact of Afghanistan on international security may turn out to be highly paradoxical. It is obvious that Afghanistan, along with Iraq, has called into question the idea that the US, in its supposed "unipolar moment", could change even the most difficult and divided societies by its confident use of armed force. But it is not only the ideas of the neoconservatives and their camp-followers that are in trouble. In many ways the involvement of NATO in Afghanistan was textbook liberal multilateralism: approved by the UN Security Council, involving troops from forty democracies, coupled with the UN Assistance Mission, and with admirable aims to assist the development and modernization of Afghanistan. The very ideas of rebuilding the world in our image, and of major Western states having an obligation to achieve these tasks in distant lands—whether by unilateral or multilateral approaches—may come to be viewed as optimistic. Or, to put it differently, and somewhat cryptically, Afghanistan may not have quite such a drastic effect on the American imperium as it had on the Soviet one in the years up to 1991; but it may nevertheless come to be seen as one important stage on the path in which international order became, certainly not uni-polar, and perhaps not even multi-polar, but more based on prudent interest than on illusions that Western ideas control the world. Afghanistan, like Somalia, may contribute to greater caution before engaging in interventionist projects aimed at reconstructing divided societies.

94. Despite all the difficulties encountered in Afghanistan since the fall of the Taliban in 2001, in the US presidential election campaign in 2008 both Barack Obama and John McCain promised to increase the US commitment to Afghanistan in 2009. There was little prospect either that the insurgency would subside or that the US would tiptoe out of the war. Furthermore, both candidates advocated continuing and even extending the practice of using US force against Taliban and al-Qaeda targets in Pakistan. The war's international dimension, and its significance for international security more generally, was set to continue.

The Debate on Policy Choices

95. The Obama administration's policy planning for Afghanistan is based on the sound presumption that the Afghan problem cannot be addressed in isolation. Although many countries have a potentially important role in any settlement in Afghanistan—especially Iran, with its large numbers of Afghan refugees and its major drug problem—Pakistan is at the core of this approach. Granted the indissoluble connection between Afghanistan and Pakistan, any policy in respect of the one has to be framed in light of its effects on the other. At times it may even be necessary to prioritise as between these two countries. The simple truth is that Pakistan is a far larger, more powerful and generally more important country than Afghanistan. If the price of saving Afghanistan were to be the destabilization of Pakistan, it would not be worth paying. A principal aim of the US in the region should have been, and indeed may have been, to avoid creating a situation in which that particular price has to be paid: yet at least once before, in the Soviet-Afghan War in the 1980s, something very like it happened.

96. The main conclusion of any consideration of the Pakistan factor in the ongoing conflict in Afghanistan has to be that the policy of the US and allies—to strengthen central government in both countries—has been operating in extremely difficult circumstances, has been pursued erratically, and has been largely unsuccessful. While it is not obvious what the alternatives might be—open acceptance of regional autonomy in both societies would have some merits—the general approach of backing non-Pashtuns in Pakistan and Afghanistan risks exacerbating the Pashtun problem in both countries. Three distinct causes—Pashtun, Taliban and al-Qaeda—have become dangerously conflated. It should be a first aim of Western policy to reverse this dangerous trend.

97. Because of the grim prospects of a stalemate, a war of attrition or worse in Afghanistan, and also because of the advent of new governments in Pakistan in 2008 and the US in 2009, there has been at least the beginning of consideration of alternative policies. Two stand out: each in its way addresses directly the growth of the insurgency, and each is based on a recognition that the Pakistan dimension of the problem has to be considered alongside the Afghan one. Both options take into account the central requirement of any approach—that it be geared to ensuring that neither Afghanistan nor Pakistan offer the kind of haven for organizing international terrorist actions that Afghanistan did under Taliban rule.

98. The first option centres on negotiation with Taliban and other Pashtun groups. The first question to be faced is whether, on either side of the border, there are sufficiently clear hierarchical organizational structures with which to negotiate. The second question is whether Afghan Taliban/Pashtun goals are framed more in terms of control of the Afghan state along the completely uncompromising lines followed by the Taliban in the years up to 2001, or in more limited terms. Whatever the answers, negotiation in some form with some of the insurgent groups and factions is inevitable. Indeed, in an informal manner some is already happening. Combining fighting with talking is quite common in insurgencies, not least because of their tendency to result in stalemate. Yet it is never easy, and is likely to be particularly difficult for those on both sides who have chosen to see the war in Afghanistan as a war of good against evil. It is also likely to be difficult if, as at present, the Taliban believe they are in a position of strength. A critical question to be explored in any talks is whether, as some evidence suggests, Taliban leaders have learned enough from their disasters since seizing Kabul in 1996, and in particular from their near-death experience in 2001, to be willing to operate in a different manner in today's Afghanistan.[106] The continuing commitment of the Taliban in Pakistan to destroying government schools, and its opposition to education for girls, does not inspire confidence. The scope and content of any agreement are matters of huge difficulty. Some agreements concluded by the Pakistan government in the past few years are widely seen as having given Taliban leaders a licence to continue supporting the insurgency in Afghanistan. This serves as a warning of the hazards of partial negotiation. Yet the pressures for negotiation are very strong, and a refusal to consider this course could have adverse effects in both countries.

99. In October 2008, after a two-week debate that was not always well attended, the Pakistan parliament passed unanimously a resolution widely interpreted as suggesting above all a shift to negotiation. Actually it was a complex package, in which the parliament united to condemn terrorism and at the same time was seen as "taking ownership" of policy to tackle it. The resolution said that regions on the Afghan border where militants flourish should be developed; and force used as a last resort. It opposed the cross-border strikes by US forces in Pakistan, but at the same time indicated a degree of support for US policy. It called for dialogue with extremist groups operating in the country, and hinted at a fundamental change in Pakistan's approach to the problem: "We need an urgent review of our national security strategy and revisiting the methodology of combating terrorism in order to restore peace and stability".[107] At the very least it provides one basis for the incoming US administration to recalibrate the US's largely burnt-out policies towards Pakistan.

100. The second option under discussion involves a fundamental rethinking of security strategy in both Afghanistan and Pakistan. On the Afghan side of the border it would call for some increase in ISAF or other outside forces, especially to speed up the pace of expansion of the Afghan Army, and thereby to provide back-up so that certain areas from which the Taliban have been expelled can thereafter be protected. It would also call for cooperation in security matters with local forces and councils, with all the hazards involved. One informed and persuasive critique of the approach to counter-insurgency used in Afghanistan since 2003 suggests that its emphasis on extending the reach of central government is precisely the wrong strategy: its authors, specialists in the region, argue instead for a rural security presence that has been largely lacking.[108] A security strategy based on local forces and councils would also call for expansion of aid and development programmes, especially in urgent matters such as food aid in areas threatened by famine; and for a serious effort to address the widespread corruption which makes a continuous mockery of Western attempts to bring reform and progress to Afghanistan. On the Pakistan side it would involve a protracted effort to develop a long-term policy—hitherto non-existent—for establishing some kind of government influence in the FATA, and for a joined-up policy for addressing the Taliban and al-Qaeda presence. On both sides of the border it would necessitate reining in the use of air power to reduce its inflammation of local opinion.

101. For reasons indicated in this survey, it is highly improbable that either of these options on its own could provide a substantial amelioration of a tangled and tragic situation. However, a combination of the two policies—both negotiating, and rethinking the security strategy—might just achieve some results. Such a dual approach has been supported in 2009 by John Nagl, one of the architects of the new US counterinsurgency doctrine. Advocating the adaptation of this doctrine in the special circumstances of Afghanistan, he has stated: "At the time, the doctrine the manual laid out was enormously controversial, both inside and outside the Pentagon. It remains so today. Its key tenets are simple, but radical: Focus on protecting civilians over killing the enemy. Assume greater risk. Use minimum, not maximum force". His advocacy of these principles is accompanied by emphasis on the importance of dealing with local forces as well as national governments both in Afghanistan and in Pakistan.[109]

102. An approach along such lines would need to include other elements as well, including a strong and credible commitment to leave as soon as a modicum of stability is achieved. Such a combination would need to be pursued in both Afghanistan and Pakistan. It could only work if a new US administration rejected the worst aspects of previous policies, and pursued the matter with more consistent attention than in the past. It would be likely to result in some unsatisfactory compromises, and might build on, rather than fundamentally change, the pattern of local loyalty and regional warlordism that is so rooted in Afghanistan. Yet if the war in Afghanistan is not to have even more fateful consequences for international order than those seen in the past three decades, it may be the direction in which events have to move.

23 January 2009

16 T R Moreman, "Callwell, Sir Charles Edward (1859-1928)", in H C G Matthew and Brian Harrison (eds), Oxford Dictionary of National Biography (Oxford: Oxford University Press, 2004). Back

17 Maj C E Callwell, Small Wars: Their Principles and Practice (first published 1896), rev edn (London: HMSO, 1899), p 5. Back

18 For an account of the retreat from Kabul in January 1842 see Saul David, Victoria's Wars: The Rise of Empire (London: Viking, 2006), pp 55-67. Back

19 Ibid, p 16. Back

20 David Loyn, Butcher & Bolt: Two Hundred Years of Foreign Engagement in Afghanistan (London: Hutchinson, 2008). Back

21 Rob Johnson, A Region in Turmoil: South Asian Conflicts since 1947 (London: Reaktion, 2005), p 166. Back

22 The Northern Alliance, more correctly called the United Islamic Front for the Salvation of Afghanistan, is a loose association of regional groups founded in 1996 to fight against Taliban control of Afghanistan. Back

23 On the Taliban's history of supporting opium production, which became the mainstay of their war economy in the late 1990s, see Ahmed Rashid, Taliban: The Story of the Afghan Warlords (London: Pan Books, 2001), pp 117-24. Back

24 Ahmed Rashid, Descent into Chaos: How the War against Islamic Extremism is Being Lost in Pakistan, Afghanistan and Central Asia (London: Allen Lane, 2008), p 401. Back

25 SC Res. 1386 of 20 December 2001, operative paragraph 1. Five months later SC 1413 of 23 May 2002 addressed ISAF's entitlement to use force more explicitly, authorizing "the Member States participating in the International Security Assistance Force to take all necessary measures to fulfil the mandate of the International Security Assistance Force". Back

26 SC Res. 1510 of 13 October 2003 expanded ISAF's sphere of operations to other parts of Afghanistan. By the end of 2006, now operating under NATO, it had responsibilities in virtually all Afghanistan. Back