Select Committee on Foreign Affairs Appendices to the Minutes of Evidence


Memorandum submitted by Professor Adam Roberts, Balliol College, Oxford


  1.  My job is Montague Burton Professor of International Relations at Oxford University. I have written extensively on the laws of war, on strategic issues, on terrorism, and on wars and military occupations in the Middle East. My principal publication relevant to this memorandum is (co-edited with Richard Guelff), Documents on the Laws of War, 3rd edition, Oxford University Press, 2000. In this memo I refer briefly to two newspaper articles that I have written during this crisis, in order to indicate how the particular issues addressed here were seen as events unfolded.

  2.  This memorandum is a short examination of one limited and specialised aspect of ongoing events: the application of the laws of war, with particular reference to the hostilities in Afghanistan. An issue raised in the memo is the impact of laws of war issues on the maintenance of the International Coalition of states engaged in or supporting the war against terrorism. The main question addressed is the extent to which the UK government generally, and the FCO and Defence Ministry in particular, have worked to ensure observance of basic standards by the Coalition in this unusually complex war.

  3.  The International Coalition of states engaged in the overall campaign against terrorism has an unusual character. Its membership does not appear to be precisely defined, and the contributions of many of its members to the actual conduct of military operations in Afghanistan (as distinct from the many other areas of activity involved in this campaign) has been limited. None the less, the maintenance of the International Coalition is important for the campaign against terrorism, for a variety of military and political reasons which need no elaboration here.

  4.  For centuries the laws of war (today sometimes called "international humanitarian law applicable in armed conflict") have had an important function in the policies and practices of international coalitions engaged in military operations. Although in strict legal terms there is no direct connection between the law relating to the resort to force (jus ad bellum) and the law governing the use of force in war (jus in bello), there are certain ways in which these two bodies of law interact in practice. A perception that the states involved in a coalition are observing basic international standards may contribute to public support for military operations within the member states; support (or at least tacit consent) from other states for coalition operations; and avoidance of disputes within and between coalition member states. In short, there can be strong prudential considerations (not necessarily dependent on reciprocity in observance of the law by all the parties to a war) which militate in favour of observing the laws of war. The 1991 Gulf War and the 1999 Kosovo War confirmed the utility of having common standards in certain basic matters: treatment of captured personnel; protection of civilians; humanitarian assistance; policy towards neutral countries, individuals and trade; and prosecution of alleged war criminals.

  5.  In wars in Afghanistan over the centuries, observance of the laws of war has been limited. Some local customs in certain matters are consistent with international legal provisions. Other customs are different from what is envisaged in the conventions, but not necessarily a violation of them: an example of this is the practice of soldiers of the defeated side not to be taken prisoner, but to join the forces of their adversary. However, taken overall, there has not been a high level of compliance. Guerrilla warfare, which was already endemic in Afghanistan in the nineteenth century, notoriously makes it difficult to observe the distinction between soldier and civilian which is at the heart of the laws of war. To the extent that wars in Afghanistan have been and remain civil wars, fewer rules are formally applicable to them than to inter-state wars.

  6.  On the technical legal question of which of the main laws of war treaties are formally binding on the belligerents, Afghanistan is a party to the 1925 Geneva Protocol on gas and bacteriological warfare, to the 1948 Genocide Convention, and to the four 1949 Geneva Conventions. The main members of the Coalition are also parties to these agreements. In addition, the 1907 Hague Land War Convention applies because of its customary law status. Although some of the belligerents are parties to this or that additional agreement, the above-named agreements provide the basic treaty framework for considering the application of the law in this particular armed conflict. In the case of non-international armed conflict (i.e. civil war), which is the character of certain aspects of this war, some but not all of the provisions of these agreements apply. The 1907 Hague Land War Convention's Article 2 indicates that it only applies between states. The 1925 Geneva Protocol is not formally applicable. The 1948 Genocide Convention is considered to apply to non-international as well as international armed conflict. In the 1949 Geneva Conventions, common article 3 lists certain minimum provisions to be applied in non-international armed conflict.

  7.  The wars in Afghanistan in the past two or more decades, like many wars in many countries in the period since 1945, have had the character of an internationalised civil war. That is to say, they have in varying degrees involved elements of an international war (to which the main body of the laws of war applies) and elements of a civil war (in respect of which there are fewer agreed international rules). The presence of significant numbers of foreign fighters on the Taliban side, especially as they appear to have little respect for the laws of war, further complicates both the legal classification of the war and the application of certain minimum rules.

  8.  The US armed forces have long kept open the possibility of observing the rules governing international armed conflicts even in situations which may differ in certain respects from the classical model of an inter-state war. A simple principle codified in the Standing Rules of Engagement issued by the US Joint Chiefs of Staff on 1 October 1994 spells this out: "US forces will always comply with the Law of Armed Conflict. However, not all situations involving the use of force are armed conflicts under international law . . . Law of Armed Conflict principles may nevertheless be applied as a matter of national policy."

  9.  In respect of the war, principally between the Taliban and the Northern Alliance forces, which has been going on in Afghanistan for many years, there have been authoritative pronouncements calling on the parties to the conflict in Afghanistan to comply with their obligations under international humanitarian law. In particular, UN Security Council Resolution 1193 of 28 August 1998, passed unanimously, reaffirmed inter alia 'that all parties to the conflict are bound to comply with their obligations under international humanitarian law and in particular the Geneva Conventions of 12 August 1949, and that persons who commit or order the commission of grave breaches of the Conventions are individually responsible in respect of such breaches.' The reference to grave breaches would appear to suggest that the Security Council viewed all the rules of the 1949 Geneva Conventions as applicable, and not just common Article 3 which deals with civil war.

  10.  In the period between the attacks of 11 September and the start of the US bombing campaign in Afghanistan on 7 October, when the International Coalition was being formed, it was not difficult to foresee that implementation of the laws of war would be a difficult problem in the military operations then in preparation. The number of forces involved in the war in Afghanistan, many of which have a strong regional basis, and the lack of clear structures of authority, decision-making and military discipline within them, were all likely to be complicating factors so far as implementation of the laws of war was concerned. Some of the forces involved, especially Al Qaida, had committed and/or approved numerous acts in foreign countries which could only be called crimes against humanity. In the USA, there was evidence of a school of thought that in a war against brutal terrorists, certain normal restraints and safeguards might not apply. In an article ("Apply the Law of War in an Anti-Terror War, Too"), published in the International Herald Tribune on 4 October 2001, I drew attention to certain obvious issues: the need to conduct Coalition operations discriminately; the likelihood of adversary forces executing Coalition prisoners; and the possibility that some captured enemy personnel might not qualify for prisoner-of-war status, though there was still a requirement that they be treated humanely. Other practical issues that required careful thought and planning included: maintenance of order (and avoidance of looting and revenge killings) in liberated towns; how to bring pressure on Northern Alliance forces to observe basic norms; and assistance for humanitarian relief operations. A particularly difficult problem was determining where overall responsibility would lie in such matters as treatment of enemy prisoners, granted that the USA and the Northern Alliance forces might both be involved in the events leading to their surrender. I assume that in the four weeks after 11 September considerable thought was given by US and UK governments to problems such as these, but have no direct information on this.

  11.  Since the direct US and Coalition military involvement began on 7 October 2001, many of the problems indicated in the preceding paragraph have in fact arisen. They have been heavily publicized because of the remarkable role of the press in this war, operating close to (and even in some cases in front of) the front lines, and sending back reports and high-quality pictures as events unfold. While the discriminate character of much Coalition bombing has been a generally positive factor, Coalition actions in respect of some of these issues have in my judgement probably had a negative effect on international support for Coalition operations. Two principal causes for concern have related to the bombing of certain locations in Afghanistan, and the treatment of prisoners, each of which is discussed separately below. A third issue discussed briefly is assistance for humanitarian relief.

  12.  There has been much international concern over aspects of the US/Coalition bombing in Afghanistan. There have been certain bombing attacks causing significant numbers of apparently civilian casualties, and also the two attacks which hit an ICRC warehouse in Kabul. There have also been reported killings of friendly forces. The most serious reports of attacks affecting civilians relate to reported bombings on 1 and 2 December of Kama Ado and neighbouring villages in eastern Afghanistan, not far from the cave complex at Tora Bora: while reports indicate that over a hundred villagers may have died, the Pentagon has denied that such attacks took place. In legal terms, it is wrong to assume that civilian deaths per se constitute a violation. There are strong reasons to believe US statements that civilian deaths in the above types of episode were unintended. Some may well have resulted from errors of various kinds, and some may have been unavoidable or unlucky collateral damage. Much of the bombing to date has been discriminate. However, events of the type indicated, even if they do not constitute violations of the law, do have the potential to harm the Coalition cause.

  13.  There has also been much concern over the difficult issue of treatment of prisoners. Here the main focus has been on an exceptional event: the killing of a large number of Taliban prisoners following the revolt at Qala-e-Jhangi Fort near Mazar-e-Sherif in the period 25 November 1 December 2001. Already before the prisoners were taken at Kunduz, it was evident that the surrender and imprisonment of the foreign forces was likely to prove extremely difficult. In an article ("Crisis at Kunduz") published in The Guardian on 24 November I wrote: "The US and other governments need to take a much clearer and tougher stance towards their Afghan allies, requiring them to establish a prisoner-of-war camp, not to institute a bloodbath." I was concerned about a number of statements made by US Secretary of Defense Donald Rumsfeld at press conferences, including his press conference on 19 November, and in particular at the apparent failure to make clear that if the fighters in Kunduz surrendered, they would be treated humanely. This is not the place to go into detail about the course of events during the rebellion at the fort which started on 25 November. The situation was extremely difficult, especially as the Taliban prisoners continued to pose an extreme threat to their captors; and by not having laid down their arms they failed to meet the requirements for prisoner-of-war status. The precise chain of events leading to the revolt has yet to be established, but the causes appear to include the following heady mix: (a) these were particularly fanatical soldiers, for whom the whole concept of surrender would be anathema; (b) the arrangements for receiving, holding and processing the prisoners appear to have been ad hoc and casual; (c) not all the prisoners had surrendered all their weapons; (d) the prisoners were held in a place where there was a large store of weapons, to which they gained access; (e) some reports suggest that the prisoners feared that they were about to be killed, so had nothing to lose by revolt; and (f) the presence of foreigners at the fort, including journalists and CIA men, may have added to the prisoners' fear and anger.

  14.  In public discussion in the UK there has been a tendency to focus on the events at the fort, including the question of whether too much force was used once the rebellion there had begun. There are issues there than need to be explored, but if the situation was as desperate and threatening as some have described it, the use of extreme measures is not surprising. The public discussion should focus a bit less on blame once this had begun, and a bit more on what the right overall policy would be for receiving and dealing with an exceptionally difficult class of prisoners. The most urgent issue raised by the events at the fort is whether the USA and the Northern Alliance have a clear overall policy for treating prisoners, and for handling the particularly difficult category of foreign fighters. It may have been failures to think the issue through before the prisoners ever got to the fort that was the real cause of the disaster.

  15.  It is a difficult question whether the US and other members of the International Coalition have influence over the Northern Alliance's actions in such basic matters as protection of prisoners. Even though this is primarily a Northern Alliance responsibility, the Coalition is inevitably involved in the matter. In his Pentagon press briefing on 30 November, Donald Rumsfeld indicated in general terms (not in connection with the prisoner question) that the US does have influence with the forces with which it operates in Afghanistan: "We have a relationship with all of those elements on the ground. We have provided them food. We've provided them ammunition. We've provided air support. We've provided winter clothing. We've worked with them closely. We have troops embedded in their forces and have been assisting with overhead targeting and resupply of ammunition. It's a relationship." This contrasts with a statement of Prime Minister Tony Blair, who was asked on 13 November, again in general terms, "What sanctions do we have over the Northern Alliance?" He replied simply "None." (The above texts are respectively from the US DoD and UK 10 Downing Street websites.)

  16.  An important question not explored in detail here is the obligations of the various parties to assist in and protect humanitarian relief operations. The main relevant provisions are those in Part II (i.e. Articles 13-26) of 1949 Geneva Convention IV on civilians. Now that many parts of Afghanistan are no longer under Taliban control, there should be scope for making appropriate security arrangements to reduce the risk of aid being hampered by banditry or internecine fighting. The whole issue is urgent now that winter has arrived.

  17.  In view of the likelihood that further issues relating to the laws of war will arise in the continuing war in Afghanistan, not to mention in any other countries which may become involved in the war against terrorism, there is a need for greater clarity about observance of basic rules of the laws of war generally. In particular there is a need for better preparation in practical matters such as the taking and holding of prisoners. To judge by events since the fall of Kunduz, there is a risk that laws of war issues will become contentious in relations between Coalition members. It is the task of diplomacy, including British diplomacy, to try to ensure that there is a consistent and defensible approach to these tangled and difficult issues. It will be well worth exploring what the FCO and Ministry of Defence have done, and plan to do, on them, especially granted the significance of these issues for the maintenance and effective operation of the Coalition.

Professor Adam Roberts

Balliol College


4 December 2001[1].

1   [On 16 February 2002 Prof. Roberts supplied to the Committee an advance copy of his article on "Counter-terrorism, Armed Force and the Laws of War" which appeared in Survival, quarterly journal of the International Institute for Strategic Studies, London, vol. 44, no. 1, Spring 2002, pp. 7-32. This supplemented and updated paragraphs 5 to 16 of the memorandum.] Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 20 June 2002