General Roberts,the Occupation of Kabul,and the Problems of Transition,1879–1880

I. Authority in Military Occupation

In the event of rebellion or insurrection, the objective of the British military forces in the late nineteenth century was to use force to bring about the re-establishment of government control by either the colonial authorities or a local surrogate leadership. ⁵ The military task was to fulfil this mission by breaking the will of the insurgents, and throughout British contemporary writing the emphasis was on ‘decision’, looking for a decisive result that would avoid protracted warfare. This could be achieved by inflicting casualties, and, where insurgents remained elusive, they were to be induced into a show of force by attacking the fabric of their economy, which might include the destruction of property and livestock. The aim was not to antagonize the population but to draw fighters into a battle where the firepower and discipline of British forces and their trained auxiliaries could be brought to bear. Following the setbacks of the First Anglo-Afghan War (1838–42) and, indeed, the early stages of the Indian mutiny (1857–8), there was a great deal of emphasis on acting without hesitation, which would be interpreted by local populations, it was argued, as a sign of British weakness. ⁶ To bring about the decisive result, the exercise of bold leadership was thought to be the best method to ensure the continuation of military superiority and prestige, two components that would reinforce the idea of British governance. Roberts inculcated the idea throughout his career that protracted operations which favoured the insurgent were to be avoided.

Nevertheless, asserting British authority without resort to force was preferred by colonial authorities since this represented the cheapest means by which they could achieve their objectives. Armies alone were simply insufficient to guarantee security in the longer term. ⁷ The British knew that military force had its limits, and the risk of a temporary military setback had the potential to damage prestige profoundly and perhaps worsen an insurrection. ⁸ Lord Cromer noted:

There is truth in the saying … that the maintenance of the Empire depends on the sword; but so little does it depend on the sword alone that if once we have to draw the sword, not merely to suppress some local effervescence, but to overcome a general upheaval of subject races goaded to action either by deliberate oppression, which is highly improbable, or by unintentional misgovernment, which is far more conceivable, the sword will assuredly be powerless to defend us for long. ⁹

Considerations of cost frequently asserted themselves in colonial decision-making, and Afghanistan proved to be no exception. Money, in the form of subsidies, was used occasionally to augment the limited British military power as the cheaper alternative. ¹⁰ The chief problem was that cost also precluded the creation of larger civilian administration, and much work on the peripheries was still left to soldiers. ¹¹ The interim solution, to use political officers and advisers, some of them actually soldiers on secondment, risked a conflict of interest between military and political objectives. ¹²

If the strategic ends of pacification in military occupations were to restore order and authority, and the ways included a willingness to use escalating degrees of force, there was nevertheless a consciousness of the limited nature of the means available. Britain could deploy only a relatively small number of troops into Asia, compared with other continental powers, and this compelled the British army to raise local auxiliary forces, such as the three presidency components of the Indian Army, and a variety of semi-regular and irregular units. The governing military doctrine for the use of these formations in Afghanistan was to maintain a concentration of force with a mobile reserve that could reinforce any threatened garrison, line, or detachment. There was an emphasis on retaining the initiative by staying mobile, adopting an offensive posture and maximizing firepower. However, the campaigns in Afghanistan put considerable strain on logistics and transport. Protection of lines of communication not only absorbed much-needed manpower but also required time-consuming operations to clear routes, to relieve beleaguered garrisons, and to prevent or punish raiding. Most important of all was the weakness of political direction. Relations between the Indian Army and the civilian administration before the war in Afghanistan had been bad, and the army was to find itself confronted by the predictable and perhaps avoidable situation of facing popular unrest with no Afghan leadership to hand over to. ¹³ Nevertheless, the extended nature of the empire and limited means of communication had set a long precedent for independent action by both military and political leaders. ¹⁴ Using one’s initiative and acting within the broad outlines of a policy or ideology were the norm. ¹⁵

The alternative to the use of military power was the use of local rulers and chiefs to govern under British suzerainty. This had the advantage that existing structures, hierarchies, and authority could simply be incorporated under the British flag. Subordinate leaders could also be dismissed and replaced if they failed to remain compliant, and the British could potentially exploit rivalries between sectarian or ethnic communities if this was deemed necessary. ¹⁶ The British imperial project frequently tried to bring together quite distinct entities in order to create economically and administratively viable groupings in the form of federations or states, like the Indian Empire or the Union of South Africa. Viceroy Lord Lytton (1876–80) was an enthusiast for cultivating and bringing together the princes of India, and his imperial assemblage of 1877 harnessed together Indian and medieval British motifs to emphasize a feudal relationship of service and protection. ¹⁷ Just one year later Lytton was affronted by the attitude of the Afghan amir, Sher Ali, who appeared to profess allegiance to the British, but communicated with the Russians and entertained Colonel Stolietov, one of their envoys, at the height of the Eastern Question confrontation in 1878. ¹⁸ Afghanistan appeared to be drifting into a Russian sphere. It is ironic that Lytton had actually hoped that Sher Ali would be as compliant as other South Asian princes. ¹⁹ Believing Sher Ali now to be unreliable, and an ultimatum having failed to elicit the response expected, Lytton ordered the invasion of Afghanistan. ²⁰

II. The Occupation of Afghanistan

In Afghanistan the British had achieved their initial strategic objectives by 1879 in that they had defeated the Afghan forces, taken the capital, and denied the territory to their Russian rivals. Roberts expected to face Russian troops at some point either during the campaign or after, while Lytton contemplated breaking up Afghanistan into more easily controlled parts, each of which would resist Russia more readily. ²¹ Critics noted that if the strategic objective was to maintain a strong and independent Afghanistan as a buffer zone against Russian influence, then defeating and potentially dismembering the country was illogical. Finding suitable Afghan leaders was also a problem. Amir Sher Ali had died of natural causes in Mazar-i Sharif in December 1878 while on the run. ²² His successor, Yakub Khan, had spent the previous months being held under house arrest by his father and also initially opposed the British presence. ²³ Lord Lytton looked instead to Wali Mohammed, the half-brother of the late amir, to take over in Kabul. ²⁴ Faced with the prospect of losing control of the whole of Afghanistan to a rival, British-approved candidate, Yakub Khan offered peace negotiations.

Major Louis Cavagnari, the senior British political representative, responded positively in official communications and even offered condolences on the death of Yakub Khan’s father, but the sentiments were insincere. ²⁵ He dictated the terms with some impatience. Cavagnari wrote to the viceroy:

I found the whole lot to be pretty much of the ordinary Afghan stamp, and their avarice and suspicion were their leading qualities. Their arguments were so feeble and far from the point that I at once made up my mind to deal with the case as if it concerned an ordinary affair connected with border Pathan tribes. ²⁶

The terms of the subsequent Treaty of Gandamak were stern. Afghanistan was to accept a British resident with a military escort and would place its strategic passes under direct British control, while Afghanistan’s foreign policy was to be conducted by the Government of India. ²⁷ The loss of territory was designed to clarify the defeat of Afghanistan as much as secure strategic routes into India. ²⁸ The Treaty of Gandamak was also a piece of political and diplomatic theatre. Selecting the site of the last stand of the British brigade destroyed in 1842 for the signing of the terms of Afghan surrender reasserted a moral as well as a political authority. Interestingly, the flamboyant Russian Major General L.N. Skobelev believed that the British had made a mistake in allowing these political considerations to become more important than military ones. ²⁹ Skobelev’s own approach to the Central Asians, as at Geok Tepe (December 1880 – January 1881), was to inflict widespread destruction and significant casualties on combatants and non-combatants alike.

Having established a permanent British resident, the military columns commenced their withdrawal. While a British garrison was retained at Kandahar, the withdrawal was completed when the residency at Kabul was overrun, the chief British representative was killed, and the defenders were wiped out. ³⁰ Roberts made a forced march back to Kabul and reoccupied the city after two successful conventional battles. For him the moral effect of exacting vengeance rapidly was crucial. He was eager to demonstrate Britain’s military power to overawe the Afghans and prevent further resistance. Yet, for Roberts, it was also a personal conviction that the lesson of the Indian mutiny was to act swiftly and with resolution. ³¹

The initial reaction to Roberts’s arrival near Kabul was hostile. Within the city and the region there were several undefeated Afghan formations and there seemed a possibility that the British would have to engage in street fighting to gain control of the capital. However, the prospect of being encircled convinced the Afghan lashkars to abandon the city on 9 October 1879. Just three days later Roberts orchestrated a full dress parade through Kabul and occupied the Bala Hissar fortress to demonstrate British military strength. ³²

Roberts was forced to impose military rule because there was no political alternative. When Roberts had approached the capital, Amir Yakub Khan had first availed himself of British protection and then later abdicated with the famous remark that he would rather be ‘a grass cutter in India’ than attempt to reign in Afghanistan. ³³ General Stewart believed that, if the amir could not protect a British envoy within a stone’s throw of the Bala Hissar or even ensure his own safety, he was ‘of no real use as an ally’. ³⁴ The Foreign Department of the Government of India therefore ordered Roberts to ‘assume and exercise supreme administrative authority’. ³⁵ Lytton told him: ‘Any present [Afghan] administration should be provisional [and] only entirely subordinate to your supreme authority and removable at will.’ ³⁶ Lytton was prepared to admit that Roberts’s ‘political position at Kabul is undoubtedly a delicate one and the character and extent of your political authority ought, in my opinion, to be properly defined and proclaimed with the least possible delay’. ³⁷ Roberts thus assumed political as well as military authority in Kabul, but from the outset this was an appointment with significant problems. ³⁸

While the British and Indian governments debated the relative merits of partitioning the country and which Afghan leaders might be appropriate to reign, the Afghan factions in the north and east mobilized to take action against the occupation forces. ³⁹ Roberts managed to secure a base for future operations in the enlarged Sherpur Cantonment, a barrack complex just outside the capital. He also established an intermediate post at Buthak and sent a brigade to open communications with the Khyber Line Force and clear the entire route of fighters in November. However, he could command no more than 8000 men in and around Kabul, and so it was imperative for him to demonstrate British power in ways that would impress on the Afghans the futility of resistance.

The primary political objective was retribution against those that had perpetrated the attack on the British residency. Lytton authorized such action, but there was already general acceptance that imperial powers had the authority to impose their own legal norms on less developed polities. Roberts needed little encouragement. Indeed he clearly felt personally affected by the death of Major Cavagnari and his command: although cut off, without hope of relief, they had perished while fighting to the last man. ⁴⁰ Feelings ran high because the defenders of the residency had been dishonoured in death, despite their gallantry; their bodies had been dismembered and dumped on a rubbish tip. Roberts was convinced that punishment was required in order to satisfy the prestige of the British Empire. Attacks against residencies could not go unpunished lest such action occur across the globe, although critics pointed out that Cavagnari led an armed military mission, which had been installed by force of arms, and therefore was not a normal embassy. ⁴¹ Roberts issued a proclamation, with Lytton’s authority, that announced the establishment of martial law with military tribunals to prosecute the ringleaders and participants in the attack on the British residency. ⁴² Lytton had told Roberts that, while ‘it would probably be necessary to inflict death only in execution of the verdict of some sort of judicial authority … any such authority should be of the roughest and readiest kind such as a drumhead Court Martial’. ⁴³ He emphasized that it would not be ‘justice in the ordinary sense, but retribution’, and he urged Roberts: ‘Your object should be to strike terror, and strike it swiftly and deeply; but to avoid a “Reign of Terror”.’ ⁴⁴

Lytton believed that, since there had been a popular insurrection, all Kabulis were guilty, but Roberts prior to his advance on Kabul had initially tried to differentiate between participants and bystanders. He had urged civilians to avoid impeding the progress of the British forces and the amir, promising: ‘The British Government desires to treat all classes with justice, and to respect their religion, feelings and customs, while exacting full retribution from offenders. Every effort will therefore be made to prevent the innocent suffering with the guilty.’ ⁴⁵ However, the fact remained that the British army had had to fight its way to the capital, and that clearly many Kabulis had participated in that resistance. In a subsequent proclamation Roberts clarified that more Afghans were responsible:

The force under my command … has been pertinaciously opposed, and the inhabitants of the city have taken a conspicuous part in the opposition offered. They have, therefore, become rebels against the Amir, and have added to the guilt already incurred by them in abetting the murder of the British Envoy and his companions. ⁴⁶

Attempts to appeal to legal and moral authority can be traced throughout the proclamation. Despite the ‘indelible disgrace’ of the murder of the British envoy, the ‘Great British Government’, announced Roberts, ‘is ever desirous to temper justice with mercy’. He assured the Kabulis he would not lay waste to the city even though that would have been a ‘just and fitting reward’. He stated that to ‘provide for the restoration and maintenance of order’ martial law would be imposed up to 10 miles from the city. Moreover, he added:

With the consent of the Amir, a military Governor of Cabul will be appointed to administer justice and to punish with a strong hand all evil-doers. The inhabitants of Cabul and of the neighbouring villages are hereby warned to submit to his authority.

Disarmament was imposed, and the death penalty was threatened for non-compliance, although Afghans were compensated for handing in weapons. The reaction to the proclam-ation and to the presence of the British was described by one eyewitness as ‘sullen’. ⁴⁷

The first step was to establish a commission of inquiry to identify those to be charged. The military tribunals would try the accused and carry out their sentences. However, there was great difficulty in identifying the individuals responsible, and Lytton’s private advice to Roberts, that many Kabulis, indeed ‘every civilian, be he priest, or layman, mullah or peasant’, were guilty by association, made the atmosphere highly charged. ⁴⁸ Secret meetings were held in order to acquire evidence and cross-examine witness reports. Nominal rolls of the Herati regiments were located and used to identify their personnel. Nearby villages were cordoned and there were round-ups of suspects. Local leaders were found and interviewed to extract eyewitness statements. These witnesses were assured they would not have to stand in court, so the ‘evidence’ was given in secret. As military governor, Roberts had to approve every sentence handed down.

The number of executions carried out by the tribunal is contested, but the manner of death, by hanging, angered locals. The accusation that they had rebelled illegally against the will of the amir was lost on the population of Afghanistan. Their view was that the British had executed those who had resisted the British invasion. Indeed, not all the executions were carried out under the jurisdiction of the tribunal. ⁴⁹ Five men were condemned to death for attacking the post at Shutagarden, and six others were shot by firing squad by the 14th Bengal Lancers for being armed in the city contrary to the martial law imposed by the occupation forces. ⁵⁰ General Charles MacGregor felt that the measures were half-hearted and would generate more resistance. He wrote: ‘We are thoroughly hated and not enough feared … We have been cruel yet we have not made them quite acknowledge our supremacy; and they have not yet had time to appreciate our justice.’ ⁵¹

Roberts found it impossible to establish a clear link between Amir Yakub Khan and the attack on the residency. ⁵² He was nevertheless convinced that the amir had been behind the unrest. ⁵³ It is alleged the amir had asked the city’s leading figures whether they all ‘intended to join in a ghaza to fight the British’. ⁵⁴ Yet the amir seemed to be making an enquiry rather than issuing an executive order. Yakub Khan was probably only guilty of remaining passive at the time of the attack on the residency. There were several factors that had, in fact, given rise to unrest, including Herati soldiers’ resentment at the apparent Kabuli tolerance of foreigners in the city, and the Kabuli regiments’ defeat at the hands of the British. There was a cholera outbreak which had particularly affected the Herati troops, but anger had turned to violence when the soldiers were not issued with pay arrears. ⁵⁵ The rumour that the British had money for the taking in the residency fuelled the protests. Punishment of the Herati troops for mutiny was a useful means to assert British legal authority, but the absence of any Afghan intermediaries or executives of the Afghan legal system not only subverted the British claim to be acting on behalf of the Afghan sovereign, but also undermined the authority of the amir himself, since it appeared he was incapable of administering justice to his own people. Brian Robson describes Roberts’s use of the amir’s authority to impose martial law as ‘dubious and dishonest’. ⁵⁶

Roberts was eager to have Yakub Khan ‘quodded’, that is, detained, but he was advised against it. ⁵⁷ Yet, the amir had been under virtual house arrest with a guard on his tent throughout his stay with Roberts’s force. When Yakub Khan tendered his abdication, on the eve of Roberts’s parade through Kabul, it was not in British interests that he did so. Roberts initially refused to accept Yakub Khan’s decision, but Lytton authorized it, and ordered that the amir be escorted to India, although not all of his council were in favour. There were few other candidates for amir, though Lytton still favoured Wali Mohammed, Yakub Khan’s uncle and a son of Dost Mohammed. ⁵⁸ Roberts thought him unsuitable. ⁵⁹ However, with Yakub Khan gone, it was more difficult to justify the military tribunals as exercising legal authority through the Afghan monarchy. The Cabinet and Lytton himself were uncertain whether partition of the country into smaller and more manageable entities was more pragmatic and would give a more realistic degree of political authority to the three rival city-states of Kabul, Kandahar, and Herat. ⁶⁰ It was therefore 18 days before Roberts issued a proclamation announcing Yakub Khan’s abdication. Roberts stated:

The British government, in consultation with the principal sirdars, tribal chiefs and others representing the interests and wishes of the various provinces and cities, will declare its will as to the future permanent arrangements to be made for the good government of the people. ⁶¹

Roberts was now the sole executive authority in Kabul, and, by extension, in Afghanistan.

Lytton had persuaded the British government of his view that: ‘We should instantly take possession of the authority which falls from the hand of the Amir into our own and promptly, though provisionally, enforce that authority, so far as our practical power of enforcing extends, in every direction.’ He envisaged either the ‘permanent retention of that authority’ or its transfer, ‘with very careful and copious restrictions, to some form of native government’. ⁶²

III. The Afghan Challenge to British Authority

Kabulis resented the fact that they lived under martial law because of Herati actions, but the announcement that they were under direct British authority, and the summary manner of British retribution in the tribunals, regardless of any miscarriages of justice, created a temporary union of resistance: Kohistani Tajik-Afghans, Ghilzai Pashtuns, Kabulis, Logaris, and Wardakis shared a desire to drive the British out. Roberts was aware that popular resistance was stirring. ⁶³ He explained it was due to ‘vivid fanaticism’ which ‘has ever formed a prominent trait in Afghan character’. ⁶⁴ He noted that his amnesty for those that had fought the British in the past and the opening of a medical clinic had enjoyed some success among Kabulis and were successful in their design to pacify the population, but he was not prepared to admit that it was coercion and the occupation itself that were generating resistance. His decision to opt for greater leniency and the conclusion of the commission’s work in November were perhaps due more to the mounting criticism of his actions back in Britain than in Afghanistan. For Roberts the lack of ‘reason’ among Afghans meant condemnation could actually justify British moral authority. Surgeon Major Henry Bellew, a political officer among the Yusufzai who had taken part in a diplomatic mission to Afghanistan in 1857, was not untypical in his criticism of Afghans:

The most notable traits in their character are unbounded superstition, pride, cupidity and a most vengeful spirit … They despise all other races … They glory in being robbers, admit they are avaricious, and cannot deny the reputation they have acquired for faithlessness. ⁶⁵

In terms of the strategic situation, it is interesting to note that, at this point, the priorities for Roberts were shifting. Having invaded Afghanistan to enforce a permanent diplomatic presence against Russian influence, the focus since the hasty autumn occupation was the pacification of internal unrest. The British government was still uncertain whether to proceed with partition, and some diplomatic overtures had begun with Persia to act as the potential power over Herat in return for its support of the Turcomans against the Russians in Central Asia. ⁶⁶ There seemed little chance of restoring Yakub Khan, but there was no will to annex Afghanistan outright either, so Roberts’s political position was left unresolved.

Locally, Roberts had attempted to reinforce his authority through compliant elders and leaders, but this policy too had run into difficulties. The need for supplies for the occupation forces meant extending the area under his jurisdiction, and thus the appointment of local Afghan ‘governors’ with political and legal authority for collecting food and forage but also revenue. Roberts was effectively forced to extend the authority of his military governance by his logistical problems. The Afghan response was slow but there was increasing passive resistance. Rumours began to circulate that Mullah Mushk-i Alam of Ghazni was rousing people to fight against the infidels. In Nirikh, the centre of the resistance, fighters began to assemble around a local leader, Bahadur Khan. The governor appointed by the British, Mohammed Hussein Khan, was subsequently murdered by Bahadur Khan’s followers in revenge. Further north, Mir Bacha, a Kohistani leader, drove out another British-appointed governor, with popular backing.

Roberts decided to crush the gathering insurrection by contemplating a move against Ghazni, but the limited manpower and transport available ruled out this option. Instead, a concentric manoeuvre against the gathering clans in Wardak and a series of conventional battles would, it was thought, secure Kabul and deter any further resistance from either Kohistan or Ghazni. To mask his movements, Roberts ordered a review of the entire force on 7 December 1879, which also served as a visual demonstration to the Afghans of the military power at his disposal. He lacked any reserves, except those strung out on his lines of communication, but he was counting on the continued divisions among Afghans to even the odds against him. Even news of the approach of large numbers against Kabul was greeted with enthusiasm by his headquarters because a concentration of fighters would actually give the British forces the chance to bring the Afghans to battle and destroy them. However, over the following days, the British brigades were each outflanked and forced to withdraw by swelling numbers of lashkars who approached Kabul from every direction. Roberts was eager to regain the initiative and seize back the hills around Kabul, which he did on 12 December, but he lacked sufficient troops to hold the ground and thus he concentrated his force at Sherpur two days later, abandoning Kabul and all the high features that surrounded it.

Lytton had expected a rising since 9 December but thought it would materialize in the spring, and the cause of it, he claimed, carefully distancing himself from the original idea, was Roberts’s military tribunals. ⁶⁷ The British press anticipated disaster at Kabul, but Lytton was more confident of success and was critical of the ‘paroxysm of panic’ which seized London. ⁶⁸ His view was vindicated. News of a relief march by Brigadier General Charles Gough provoked the Afghans into a premature assault on 23 December on Sherpur, where their attacks were beaten off with heavy losses. The Afghans had committed a catastrophic error in believing they could overwhelm the British in a conventional action, particularly when Roberts was in a strong defensive position. Roberts had obtained his decisive battle.

In the aftermath of the failed Afghan attack, martial law was reimposed in Kabul and more defences were constructed at Sherpur and in the surrounding hills. Compounds and houses close to the cantonment were destroyed and the inhabitants driven out. The commission of inquiry was re-established on 26 December, this time to punish prisoners who had been under arms contrary to martial law. Afghan leaders of the most recent phase of the revolt were charged, and those accused of the murders of Roberts’s appointed Afghan governors were also called by the commission. The tribunals began carrying out sentences immediately. Not all the executions were by hanging, as eyewitnesses reported that firing squads had also been used. One telegram admitted: ‘All the villagers round Kabul quite hostile to us. No quarter is given to any one found firing upon us, and prisoners taken in fight are shot.’ ⁶⁹

There was again a wave of criticism in Britain, with condemnations offered by the Peace Society, Liberal MPs, and theologians, including the bishop of Oxford. Frederic Harrison, journalist for the Fortnightly Review, made fierce criticisms of Kabul’s martial law and the tribunals, while leading liberals, including Lord Ripon and Joseph Chamberlain, wrote to the prime minister demanding an end to the war in Afghanistan. ⁷⁰ Harrison asked: ‘By what right, in public law or in moral justice, do we now affect to treat the conquered people as rebels, and hang the generals and the priests who led them to defend their country?’ ⁷¹ William Ewart Gladstone made a particularly memorable speech during his Midlothian election campaign tour on 27 November 1879 at West Calder, calling for respect for the rights of the Afghan: ‘Remember the rights of the savage as we call him. Remember that … the sanctity of human life in the hill villages of Afghanistan among the winter snows, is as inviolable in the eye of Almighty God as can be your own.’ ⁷²

In a robust defence, Roberts claimed his actions had preserved peace and order, curbed the natural fanaticism of the Afghans, and protected his own troops. ⁷³ Lytton had advised Roberts not to justify the tribunals by claiming that the executions had been carried out against those who had rebelled against the amir. The report was in part drafted by Lepel Henry Griffin, the former secretary to the governor of the Punjab and Roberts’s new political adviser sent up from India, and the emphasis was on ‘force protection’, that is, the safeguarding of outnumbered British troops during an occupation on extended and vulnerable lines of communication. ⁷⁴ Along those Khyber lines of communication, which had been under constant harassment throughout the autumn of 1879, the Ghilzais had ceased their attacks when the assault at Sherpur failed in December. At Kabul, resistance had been suppressed completely. Roberts also benefited from the political turmoil in Britain, as parliament was dissolved just after the report arrived in London. Few stopped to consider either the alleged military necessity of inflicting a stern reprisal to deter further resistance, or the unusual position that Roberts had been placed in by his political masters. Robson argues that Roberts acted in accordance with standard practices of the day, in terms of collective and retributive justice, and was influenced by the manner of the residency defenders’ deaths and memories of the mutiny. ⁷⁵ He was also acting under orders. ⁷⁶ However, Robson suggests there is a ‘strong suspicion’ that many were hanged simply on the basis of their having opposed the British. Nevertheless, the verdict must surely be that Roberts was being asked to exercise political authority in the absence of civilian government, to implement a draconian legal authority, and to satisfy the demand for moral retribution only partly on his own initiative.

Despite Roberts’s success at Kabul, resistance was not entirely extinguished. Hensman wrote: ‘In Turkistan [hostility] could scarcely be called a revival, as it never died out. In that province were still organized regiments … whose sepoys had never suffered the disgrace of a defeat at our hands.’ ⁷⁷ Attempts to negotiate a settlement with Ghulam Haider, their commander, failed. Moreover, Ghazni was rapidly becoming the new centre of insurrection. It was said that at least 30 000 irregulars had assembled under Mushk-i Alam in the region. When General Sir Donald Stewart arrived to clear the Afghan resistance from Ghazni, the ‘shadow government’ had deserted the city, but there were cavalry patrols that reported Afghan forces nearby. However, the Afghans now avoided direct assaults and melted away when the British concentrated against them. It was only through subterfuge and political intrigue that the British anticipated difficulties. Hensman believed that ‘with all the cunning astuteness of Afghans, the tribal leaders will come in and try to outwit us, as they have always tried before’. ⁷⁸

IV. The Challenge to Military Authority

Three factors now began to intervene to make the continuation of the British military occupation unlikely. The first was the sheer cost of the campaign, which was seen as adding to delays in much-needed economic reforms in India. Further postponement added to the risk that unrest might take a violent form in the subcontinent. The second factor was the exhaustion of the troops, which was worsened by supply difficulties. The third was Lytton’s growing impatience at the absence of any viable surrogate leader to take over from the British military governance in Kabul. Lytton wanted to avoid the restoration of Yakub Khan on the basis he was still implicated in the death of the British resident, but it was likely that Yakub lacked the legitimacy in the eyes of the Afghan people to be acceptable to them anyway. By February 1880 Lytton therefore considered an evacuation of Afghanistan without establishing any successor authority at all. However, he charged Lepel Griffin to find a sovereign acceptable to the Afghans if he could, and to withdraw all British and Indian forces according to a timetable by the autumn. ⁷⁹ Griffin was authorized to give gifts, arms, and money to suitable candidates. However, at the same time, military power would be used in a spring campaign to reassert and extend control through Kohistan, along the lines of communication to the east of Kabul and south to Ghazni. Roberts agreed to the plans, but the perennial lack of transport and manpower reduced the operations to a single strike against Ghazni and clearance along the Khyber.

Roberts’s political authority was also being eroded. In January 1880 Abdur Rahman, a grandson of Amir Dost Mohammed and rival of Sher Ali’s line, returned to northern Afghanistan to raise an army among the old regulars, the clans of the north and his relations among the men of Badakshan. ⁸⁰ He also made an appeal to the Kohistanis, offering to lead them in a holy war against the occupation. ⁸¹ Roberts had been ordered to halt all punitive operations around Kabul to avoid the antagonism of the Afghan people, but it was still necessary to send out foraging parties, some of whom were attacked and had to defend themselves. ⁸² As supply demands increased, so local anger and frustration turned to resistance. Logar was particularly violent. Amid the ongoing insurgency, it was anticipated that when Sir Donald Stewart came up from Kandahar he would assume command of the military forces of Kabul while Lepel Griffin acted as the political authority of the viceroy. When Roberts queried the decision, the foreign secretary and the viceroy clarified that Lepel Griffin was the ‘Political Chief of Staff’ but that Roberts’s ‘political powers will be commensurate with your military authority and supreme over the area you command’. ⁸³ However, Stewart made it clear that he would not permit Griffin to have full political control. Roberts had not approved of Lepel Griffin before the war and argued that ‘while war lasts, the supreme authority should be vested in the military commander’, ⁸⁴ but he concurred with splitting the military and political responsibilities between Lepel Griffin and Sir Donald Stewart. He announced: ‘I am prepared to give up the political work altogether.’ ⁸⁵ He abolished his military governorship and Wali Mohammed became the de facto governor of Kabul. ⁸⁶

Prior to his arrival in Kabul, Lepel Griffin had expressed some misgivings about possible ‘friction’ with the army but in practice found there to be none with Roberts. He told Roberts: ‘The real objection to the severance of chief military and political functions is no doubt found in conflict of authority between high officials, each in his own sphere independent,’ adding generously that a division of powers would mean ‘I should lose the benefit of your intimate knowledge of affairs of the people and the country.’ ⁸⁷ Nevertheless, the prospect of having to be subordinated to Stewart provoked him to consider resignation. He wrote: ‘I could not accept the position of taking full political authority, as it would be contrary to the conditions on which I came here [as adviser]. Nor do I think it would be advantageous to the public service.’ Roberts accepted his own subordination to Sir Donald Stewart, his commander in India, but, while telling the viceroy his ‘only regret is that I had not the benefit of his [Griffin’s] advice and counsel at an earlier period, especially during the first weeks of our arrival at Kabul’, he resolved not to allow his political authority to be diluted again. ⁸⁸ He confessed to Lyall on the eve of his departure to relieve Kandahar that his acceptance of Griffin was for outward appearances only:

I use the expression ‘outwardly’ for practically after the arrival of Mr Griffin, I was not supreme in political affairs. His advent caused a division of authority which to my mind is most dangerous whilst an enemy’s country is in military occupation.

He continued: ‘I determined that nothing should ever induce me again to accept a military command on service unless I was politically supreme.’ ⁸⁹

Lyall sent an urgent reply, clarifying: ‘Your political instructions give you full latitude of action for the purpose of your military expedition, and the political officers in your camp are of course subordinate to you,’ and admonished Roberts with the comment: ‘I strongly recommend no further questioning on subject of instructions, and reliance on my support here, but do nothing, especially in South Afghanistan except upon military grounds for military objects.’ ⁹⁰ In a subsequent letter Lyall explained it was not possible to give instructions on the limits of Roberts’s political authority to cover every eventuality, but he felt:

the present arrangement is by far the best and simplest way of giving you complete authority to do anything [Lyall’s emphasis] you think fit on military grounds. The Government does not want any one to hold any important political powers – i.e. powers to act in politics without reference to the Government – in South Afghanistan just now. ⁹¹

Lyall had spelled out the limits of military authority in occupation, but its clarification had come at the very end of the campaign.

Roberts was clearly deflated by his new subordinate position. ⁹² To make matters worse, the Conservative government was defeated in April, and, although Lytton offered to stay on, partly to ensure no restoration of Yakub Khan, Lord Ripon was sent as his replacement anyway. ⁹³ In fact, despite considerable debate in London and Simla about the potential retention of Kandahar, Sibi, and Pishin as British garrisons, there was not a great change in policy between the two viceroys. ⁹⁴ The objective was to end the war, evacuate the forces, and keep the Russians out of Afghanistan.

The British response to Abdur Rahman, a ‘ram caught in a thicket’, was to negotiate with him. Griffin could offer him control of northern Afghanistan, including Kabul, as long as he agreed to stand by the terms of the Gandamak Treaty, which gave Britain control of Afghan foreign affairs in return for subsidies of arms and revenue. The Government of India added that it was ‘not disposed to hinder’ measures which he might take to obtain possession of the rest of the country. ⁹⁵ Stewart was concerned about the degree of popular backing Abdur Rahman was receiving and urged Griffin to break off negotiations, but Ripon insisted he continue, offering Abdur Rahman a free hand in his internal affairs, no British resident, and the protection of the British Empire against foreign foes. Griffin also put pressure on the Afghan leader by threatening to evacuate immediately if the deal was not accepted. ⁹⁶ Faced with an unsettled country, hostility from rival Ghilzais, and the loss of potential British support, Abdur Rahman knew he needed British cooperation, but he also had to ensure he had the backing of the other prominent resistance leaders, including Mohammed Jan and Mushk-i Alam. While making his ponderous moves towards the capital, Abdur Rahman secured the support he needed. The British government was eager to evacuate Afghanistan with some urgency, but there were still alternatives to Abdur Rahman, including members of Sher Ali’s family such as Ayub Khan, if the negotiations failed. Griffin patiently worked to get Abdur Rahman to accept the British terms. On 22 July 1880 Abdur Rahman agreed and was recognized as Amir of Kabul.

Stewart and Griffin nevertheless disagreed about what could be left behind for the new amir’s government, including stores and fortifications, largely because Stewart didn’t think Abdur Rahman would last long. Continuing violence between Afghans seemed to indicate that Afghanistan was descending into civil war. Initially the British believed that the dismemberment of Afghanistan was still a more satisfactory way of keeping control of the region, not least because this was, in essence, the de facto position. Kandahar had been quiet throughout the disturbances in the north, and was a useful counterweight to Herat and Kabul, while some even considered it a useful asset strategically in the defence of India. The governor, Sher Ali, a cousin of the late Amir Sher Ali, was widely regarded as a British puppet, but there seemed no reason to doubt his independent administrative abilities. He was to be given control of the entire south of the country, from Pishin in the east to Farah in the west, and he was to have his own army, with a British garrison on hand to support him. ⁹⁷

Problems arose where the British permitted their authority to be used by local Afghan leaders. The Wali of Kandahar asked the British to send a brigade to the Helmand River where Kandahari authority was weak and where the wali feared his troops might desert to the pretender to the Afghan throne, Ayub Khan. ⁹⁸ The problem was that the British had only 4800 men at Kandahar, and, although a small brigade could be sent, the British believed its role would be purely to bolster local Kandahari troops. ⁹⁹ As soon as they reached the Helmand, the wali’s troops defected. Unsupported, the British brigade was destroyed at Maiwand. ¹⁰⁰ Roberts was quick to condemn the civilian authorities for the disaster. ¹⁰¹ Yet it was he who was sent as commander of the relief expedition to Kandahar. ¹⁰² Roberts told Ayub Khan, who professed to ‘support the English’, to ‘submit unconditionally to the British government’. When unconditional surrender was not tendered, Roberts destroyed Ayub Khan’s army in September 1880 at the battle of Kandahar. ¹⁰³

Until the last moment, Roberts had hoped Britain would retain Kandahar, because although Afghanistan was a ‘wreck of her former self’ he believed ‘the warlike preparations carried on by Shere Ali in secret’ revealed the Afghans’ military potential, and Kandahar would be a useful springboard for operations against a Russian invasion. ¹⁰⁴ Indeed, it was not clear the new amir could hold Kandahar on his own. Abdur Rahman’s control of the city had not been assured until late 1881. ¹⁰⁵ Roberts also wanted to keep the Kurram Valley and the Khyber Pass in order to be within striking distance of Kabul. However, in 1880, after his victory at Kandahar, he felt that only the Khyber Line was viable and that, to hold the pass, local levies would be sufficient. ¹⁰⁶ Roberts concluded that ‘The less the Afghans see of us the less they will dislike us,’ and argued that since ‘we have nothing to fear from Afghanistan, the best thing to do is to leave it as much as possible to itself’. ¹⁰⁷ Instead he advocated more influence and control over the strategic parts of the border. This would allow troops to move rapidly into Afghanistan, to occupy the passes of the Hindu Kush and thus thwart a Russian annexation attempt. ¹⁰⁸

There would not be a solely military solution to the Second Anglo-Afghan War. Instead, the British and Indian governments put their faith in the authority of a surrogate Afghan leader, backing him with 5 lakhs of rupees, field guns, and 1000 rifles, figures later raised to 18 lakhs per annum with periodic injections of more arms. ¹⁰⁹ The civilian preference for local surrogate leaders, subsidies, and a diplomatic agreement, which, in fact, the Afghan amirs accepted until 1919, was reasserted.

V. Conclusion

The dilemmas of suppressing insurrection Roberts faced were not unusual in the era of imperialism, but his political position as the ‘King of Kabul’ was perhaps exceptional. Military officers expected to have ‘supreme authority’ in the conduct of operations, but Roberts found that his attempts to inflict punishments on rebellious Afghans using the sanction of his government and the terms of martial law still attracted a great deal of public criticism. Despite his intention to be systematic about the process, the actuality was little more than a drumhead court-martial with summary punishments. Most have been content to examine Roberts’s actions with a necessarily critical eye, but one might suggest that his justifications for the military commission and tribunals raise questions about the relative authority of a commander and his political masters in occupations faced by insurgency. Roberts felt he had legal and political authority to act decisively and independently, and he believed the military situation demanded a ‘firm hand’. In his autobiography he cited the example of Lord Napier of Magdala to infer how commanders should insist on a free hand:

Long experience had taught Lord Napier the wisdom of having only one head in time of war, and he impressed upon the Government his opinion that the civil officers, while acting as advisers and as the channels of communication with the tribes, should be subordinate to the control of the two Commanders [under his command], who, having been put in possession of the views and wishes of the Government, should be held responsible for carrying them out loyally so far as circumstances and the safety of the force would permit. ¹¹⁰

Roberts also believed he had a moral authority to exact punishments on selected miscreants who had broken a diplomatic protocol by attacking an embassy and committed a crime by murdering, dismembering, and defiling officers on a political duty. Furthermore, he believed Afghans had to see and feel the power of British arms, government, and justice to know that they were now subordinate and that resistance was hopeless. He welcomed the opportunities he had to inflict decisive military defeats on the Afghans in December 1879 at Kabul and September 1880 at Kandahar. He relied on such events to dampen Afghan enthusiasm for the protracted warfare many of them preferred. The relief of the British force at Kandahar, which was greeted with considerable enthusiasm in Britain, also had the convenient effect of sweeping away criticism of his tribunals in Kabul. ¹¹¹

Roberts had shown flexibility in meeting changing military and political situations. He showed a remarkable acumen for handling journalists, and removed the ones who were critical of his decisions. He was ruthless in his martial law measures but believed firmly that he was upholding principles of political, military, and moral authority. The latter was perhaps one of the most important to Roberts, for he not only believed in the moral force of the authority of leaders, but he also favoured moral conduct in his army. In Kabul he had even issued an order that no soldier was to conduct inappropriate liaisons with Afghan women, lest it lead to unrest. The practices established in Afghanistan continued throughout the rest of the century across the North-West Frontier, justified by soldiers on the basis that they were ‘endeavouring to extend our influence over, and establish law and order on, that part of the Border where anarchy, murder and robbery up to the present time have reigned supreme’. ¹¹²

Nevertheless, Roberts’s critics argued that he had implemented a regime of terrorism, and one that set a ruthless and immoral precedent. Harrison wrote in the Fortnightly Review:

What we complain of, is the slaughtering unarmed prisoners as part of a political plan, to overawe a civil population, when all resistance is over. It is the making of a soldier, reeking from battle, the supreme master of an entire nation, and the sole judge of a vast political problem. And it is the handing over of this political problem, the complex civil issues of rebellion and order, to the unbridled will of a victorious general, whilst placing in his hands the weapons assumed by a Pasha in reducing a Christian province. Nothing can ever make this right, or wise, or safe. ¹¹³

Subsequently lawyers struggled with the respective rights of individuals and the legality of actions by the military authorities in cases of insurrection and reprisals. When it was suggested in 1901 that military commanders ought to be held to account for their actions, the incumbent lord chancellor, Lord Halsbury, who in 1866 had defended Governor Edward Eyre after the implementation of martial law in Jamaica, seized the opportunity to modernize doctrine that had been drawn up in 1628. Halsbury said that civilian courts possessed no authority to judge acts of the military when a state of war existed, no matter that some civilian courts still operated in the usual fashion. ¹¹⁴ The Manual of Military Law of 1914 clarified some of the ambiguities thrown up by the South African War, but even in the 1960s and 1970s there was still disagreement about the parameters of martial law: the eminent Oxford legal specialist Professor Heuston maintained that military authorities ought to have cases reopened in retrospect, but Wade and Bradley argued that military personnel could not be held to account for actions they took under martial law conditions at the time. ¹¹⁵ Part of the objection to Roberts’s actions was the uncomfortable realization that martial law was being implemented in the empire long after it was discredited in the United Kingdom.

Historians of the Second Afghan War have shown that Roberts’s chief military weaknesses were his lack of manpower and his vulnerable, over-extended transport and logistics chain. With direct personal experience of the Indian mutiny, he knew the challenges of countering popular insurrection. He utilized techniques in Afghanistan that had emerged from frontier operations and insisted on the concentration of force to strike decisively, rather than dissipating his strength into small garrisons over a large area. He aimed to act with resolution and firmly, especially against rebels. Roberts knew that military forces were not suited to protracted ‘police actions’, a view reinforced by the desultory character of operations he faced in Upper Burma in the 1880s and by the difficulties he confronted later in South Africa at the onset of a widespread insurgency in 1900. Nevertheless, for Roberts, there were other considerations. His independent command in Kabul, while welcome, also made him vulnerable. He accepted the political advice he was given, but he was unimpressed with critical journalists, the liberal political conscience in Britain, or any suggestion that his superiors lacked confidence in him. He was also himself a ‘political animal’ in that he cultivated relations with the Queen, members of the Cabinet, leading MPs, and opinion formers. He went to some lengths to explain his actions, rebuff the claims of his critics, and to lobby for supreme authority and freedom of action.

Roberts was deprived of the ability to manage the political aspects of transition, although his military operations were still crucial to it. The British army, despite suffering a significant setback at Maiwand, ensured that it could remain mobile, retain the initiative, and strike successfully across the east and south of Afghanistan. This reputation clearly had an impact on Ayub Khan but also on Abdur Rahman, a ruler who spent the rest of his career trying to avoid another British military expedition into his country. Military success enabled the British to make a political transition viable.

Roberts’s conduct in Afghanistan raises interesting questions about the degree of licence officers could possess in the exercise of political, legal, and moral authority when faced with widespread popular insurrection. The dilemmas were not entirely resolved then, despite some later clarification of principles and doctrine, or by subsequent experience. Many of the problems of military authority returned during the South African War, and Roberts was to face considerable pressure from Alfred, Lord Milner, then governor of Cape Colony, for delays in establishing security and insisting on military control. He was also subjected to strong criticism from the Liberal Party in Britain for his punitive measures against Afrikaner families living in close proximity to railways attacked by insurgents. ¹¹⁶ The fundamental questions remained concerned with the extent that military forces were suitable, politically, legally, and morally, for the work of countering insurgency, and the degree of authority they and their commanders should possess.