Training and Tactics
`To lead an untrained people to war is to throw them away.’
Confucius, Analects, XIII (c.500 BC)
IN EARLY April, 1989, The Times carried a short article describing the trial of two alleged Pakistani spies in Kabul. One was said to be an Army intelligence sergeant, the other a Special Branch corporal. Both had been captured in Kandahar. They had supposedly confessed to their espionage or sabotage activities under torture, although the report indicated that their confessions were unconvincing and contradictory. Nevertheless, they received 18- and 16-year jail sentences respectively. Such a sentence in the infamous Pol-i-Charki prison outside Kabul would be a living nightmare; for many an execution would be preferable. The Pakistan Embassy had, inevitably, disowned them, while our foreign minister described the affair as a ‘propaganda stunt’.
I have no way of knowing whether the charges were true or false, but I know for certain that we at ISI were sending Pakistani military personnel
into Afghanistan from 1981 through to 1986. I know, because it was part
of my job to select the individuals, and brief them as to their tasks. It is quite likely that these highly secret activities were resumed after I had left the Army. I must make it clear, however, that the men we sent into Afghanistan were not spies, they were soldiers from the Pakistan Army, serving with the Afghan Bureau of ISI. Their mission was to accompany Mujahideen on special operations, they acted as advisers, assisting the Commander in carrying out his task. This assignment could range from blowing up an oil pipeline or mounting a rocket attack on an airfield to laying an ambush. During my time there were usually two Pakistani teams in Afghanistan at the same time throughout the period May to October. Depending on the distance, a team could remain in the field from one to three months. No team ever knew the other was operating. They were at their peak in 1984, when no less than eleven such teams operated, seven against Kabul, two against Bagram airfield and two around Jalalabad.
All these Pakistanis were volunteers from my staff at ISI. Officers and

NCOs were posted to ISI from all branches of the Pakistan Army and General Akhtar sent them to the various Directorates, reserving the best for the Afghan Bureau. They came to me for a 2-3 year tour, and I decided whether they would work on training, operations or logistics. I would always ask if anyone was willing to go inside Afghanistan, and from those who agreed I would carefully select those most suitable for special missions.
Normally a team would consist of an officer (usually a major), a JCO and an NCO, one of whom had to be a Pushtun speaker. I would have to make it absolutely clear to each individual the risks he would be taking. Under no circumstances must he allow himself to be captured, as this would expose the Pakistan government’s clandestine support for the Jehad. Of course we would deny everything, disown them, but they would certainly be subjected to the most vile and prolonged torture. As every man has his breaking point, eventually some information detrimental to our operations would be extracted, with the likelihood of a show trial and much publicity and propaganda. Nobody was encouraged to kill himself to avoid capture, no suicide tablets were issued, as to take one’s own life is forbidden to Muslims. It was repeatedly stressed that they were to escape from tight corners, or as a last resort to die fighting. If this occurred the Mujahideen with him had to do their utmost to retrieve the body. Similarly, if a Pakistani was wounded he had to be got out – somehow.
All my men going into Afghanistan had plenty of time to prepare themselves and the Mujahideen they would be accompanying. Once a mission had been decided, and a Commander selected, then the team would be responsible for the training of that Commander and his Mujahideen, although they never knew their instructors would be going with them until the end of the course. By this time the trainers had grown beards, were dressed as Mujahideen, so that they were indistinguishable from their guerrilla companions.
These officers and NCOs had to live and fight as the Mujahideen, enduring the same privations and hardships. There was none of the military back-up support to feed them or evacuate them if wounded. They became akin to special force advisers. Their duties included giving guidance on all aspects of military operations or duties to the Commander, training the Mujahideen in their operational bases, assisting with defensive measures for the bases, helping the Commander to plan and carry out his special tasks and, if necessary, fighting. Additionally, I relied on them for information as to what was happening in the field. They were a vital part of my intelligence organization, not only on enemy activities, but on the performance of the Mujahideen and their Commanders. I could rely on these men not to exaggerate, not to ignore Mujahideen weaknesses or gaps in their training. The information they brought back was invaluable in planning fresh operations, selecting suitable Commanders, or devising future training

programmes. But I had to wait until they returned to Pakistan before any debriefmg, as none of these teams ever carried long-range radio sets for fear of enemy interception.
I must admit that when I first took over I was not in favour of Pakistanis being involved in actual operations in Afghanistan. I felt the risk of capture was too high and that, should it happen, the damage it would do to Pakistan, and therefore to the Jehad, outweighed the tactical advantages. I recall several heated discussions with General Akhtar on this, but I was overruled. I had to accept that it was a part of my duty to organize these teams, so I resolved to do so to the utmost of my ability; in fact I set about increasing the number. During all those six years from 1981-86 they performed admirably, nothing ever went seriously wrong, and nobody was captured or killed. These men were a great credit to the Pakistan Army. Although they each received an award, roughly equivalent to the US Silver Star or British Military Cross, for their professionalism and daring, this is the first time their contribution has been made public.
Let me demolish a myth that has been built up by Soviet propaganda and many journalists. Up to the Soviet withdrawal from Afghanistan in early 1989, no American or Chinese instructor was ever involved in giving training on any kind of weapon or equipment to the Mujahideen. Even with the heavier and more sophisticated weapon systems such as the Oerlikon anti-aircraft guns, and later the Stinger SAM, it was always our Pakistani teams who trained the Mujahideen. This was a deliberate, carefully considered policy that we steadfastly refused to change despite mounting pressure from the CIA, and later from the US Defense Department, to allow them to take it over. From the start the Americans wanted to be directly involved with the distribution of the weapons, the operational planning of operations and the training of the guerrillas. From the start, until the last Soviet soldier quit the country, we successfully resisted.
We did so because the Parties were strongly opposed to direct dealings with Americans. They knew that such activities would be widely propagated, thus reinforcing Soviet and KHAD propaganda that the war was not a Jehad but merely part of the global capitalist-communist struggle. We also had every confidence in our Pakistani instructors. This was fully justified by events on the battlefield.
I well remember a visit by Mr Casey to some of our training camps in 1986. In all, three camps were visited, and the CIA delegation was most curious in its questioning of trainees. One senior US official, who spoke Pushtun, repeatedly asked individual Mujahideen at random how long had they been on the course. Had they ever been in the Afghan Army? Had they ever fired these weapons before in Afghanistan? The truth was that they had all been under training for eight days, and yet they were firing heavy machine

guns, mortars, RPG-7s and recoilless rifles with the confidence and accuracy of experienced soldiers. Casey was most impressed. At dinner with President Zia that night he expressed his admiration for the high standards achieved in such a short time. Within a month Zia came to see for himself. He too was amazed by what he saw, and as good as accused me of selecting the best shots to demonstrate the weapons. I told him he could select anybody to fire, as what he had seen was the average standard. He declined to do so, but at the end of the day he remarked, ‘I only wish our Army had half this standard of shooting.’ We had no need of American instructors.
The US did, however, have a role in training our Army instructors. In the case of new weapons, particularly anti-aircraft weapons, that were not on issue to the Pakistan Army, American trainers ran courses for our instructors; they then trained the Mujahideen.
Soon after taking over, General Akhtar and I discussed the importance of improving training at great length. By the end of 1983 only 3,000 Mujahideen had received any formal training at the two camps that had been established in Pakistan. We agreed that this was totally inadequate and set a target of 1,000 trainees a month completing courses. It was a target many thought impossible to hit.
As experienced soldiers, General Akhtar and I both understood that without proper training we would indeed be throwing away the Mujahideen. As the variety and quantity of arms grew, so did the demand for training, but with the guerrilla campaign in Afghanistan there was more to it than that. In a war over which we had no direct or formal control, training, like the supply of weapons, was a key to our being able to influence what went on on the battlefield. If we issued weapons such as MBRLs, or demolition charges, from a Party’s allocation to specific Commanders for a particular type of operation in their area, it followed that it was these Commanders, and their men, who must have priority with the training. We had worked out our overall strategy, we had selected sensitive targets for attack, we were providing the arms, so we had also to provide the training to get the tasks carried out.
Commanders, being Afghans, seldom missed the chance of enhancing their own prestige by fair means or foul. We exploited this by offering training and weapons to those who undertook specific operations in their area. If they succeeded they got more training and bigger and better weapons, thus boosting their status as Commanders. Our policy was as simple as that. As we were never able to issue orders direct to our forces in the field, this manipulating of the supply of arms, and training in their use, was the only effective way of getting an operational strategy implemented.
It was fundamental to our system that training should be mission-orientated. By this I mean that if we wanted the oil pipeline destroyed the course would be solely concerned with demolitions suitable to

this end. The Commander would receive instruction on the tactics of where best to place the charges, of how to approach the pipeline, how to distract or cover nearby enemy posts, where to lay mines to catch any repair parties, and on the likely Soviet reaction. His men concentrated more on the actual use of explosives and methods of detonation. At the end of the course they would lea*, their missions having been discussed in detail with the course officer, and the explosive charges made up ready for use, but without the detonators in place.
In order to increase our capacity our courses were all ‘hands on’ practical ones, with little theory or peacetime drills. From day one the trainees started handling the weapon and quickly progressed to live firing. We cut the length of the courses, but increased the daily training time. Courses never stopped for holidays as we made use of all 365 days in the year. For the students this was no great burden, but for the instructors the strain was immense, and we had to plan their rest periods with care. We also started running courses for Mujahideen instructors, who had been selected during their attendance at other courses. These men would go back to their bases in Afghanistan to organize courses there. Often we would send a Pakistani Mobile Training Team (MTT) in to help them establish themselves at locations that we had agreed with the Parties. We would also provide a syllabus and training aids. Once these local Mujahideen trainers had gained experience the MTT would merely pay periodic visits to offer advice.
At the end of 1983 we were operating two camps in Pakistan, each with a capacity of 200 trainees. By mid-1984 we were putting over 1000 a month through the system, and by 1987 we had seven camps operating simultaneously — four near Peshawar and three around Quetta. This crash programme necessitated more staff and more money, both of which General Akhtar quickly provided, so the resultant statistics were startling. In 1984 20,000 Mujahideen benefited from our efforts, with 17,700 completing courses in 1985 and 19,400 in 1986. It is no exaggeration to say that by the time I left ISI in late 1987 at least 80,000 Mujahideen had received training in Pakistan over a four-year period, and many thousands more had done so in Afghanistan. I salute my staff; they have never been called upon to work so hard before or since.
Setting up a training camp was never a simple matter of our commandeering an Army camp or using the military’s firing ranges. Like all our activities, complete secrecy was the name of the game. Nobody outside the Afghan Bureau was to know what we were doing. The public, the politicians, enemy agents, the Pakistan Army and Soviet spy satellites had to be kept in complete ignorance of the whereabouts of each camp. This necessitated our finding our own sites well away from prying eyes. It was easier said than done.
The camps had to be within a night’s drive from Peshawar or Quetta, as

all trainees were brought by truck during darkness so they would have no inkling of their location. They had to be administratively convenient and self-sufficient with water. We could not locate them near any Army base or exercise area, nor could we use places to which the public had easy access. Wherever they were sited they had to be concealed from the ground and the air. The latter caused the most difficulty, which we overcame by camouflage and strict track discipline. By this I mean we only used existing roads or paths to, from, or around the camp. Nothing shows up so clearly on an aerial photograph as fresh tracks.
Probably our greatest problem was in finding suitable sites at which we could fire all our weapons. It was not just small arms. We had to fire extensively, on a day-to-day basis, mortars, machine guns, rocket launchers, anti-aircraft guns and SAMs. To anybody within earshot it would sound like a major battle being fought every day, and often at night, with rockets and tracer rounds arching across the sky.
Theoretically we were bound by peacetime safety regulations, but if we had followed them, 90 per cent of our live firing would have been forbidden. We fired and prayed that nobody would fmd out and that there would be no accident. Allah and General Akhtar were kind, as we got away with it —but only just. Once, when General Akhtar was with us watching the firing of Blowpipe and SA-7 SAMs against illumination mortar bombs floating down on their parachutes, I had to order cease fire when some PAF planes roared overhead. General Akhtar demanded to know why the Air Force was overflying the area. Had they not been warned? He was most upset when I explained the situation, and only with reluctance did he allow the shooting to continue. Within a few minutes more aircraft flew over and we had to abandon the firing. It took a long time to convince General Akhtar that we had no other option. Amazingly we never did have an accident.
We did, however, frequently have to move the camp because we thought its location had been compromised. If civilians wandered into the area we invented some story about an Army exercise with soldiers dressed as Mujahideen. We were gone by the next morning. These hasty moves were most disruptive and we had to have an alternative location already earmarked. Fortunately, the physical effort of dismantling tents was not great, so we could vacate a site quickly.
Another precaution we took was that, until late 1985, none of our training camps had any means of communicating with us. The use of a telephone was obviously totally insecure, and I was fearful of Soviet radio interception being able to pick up transmissions and pinpoint the sites. By the end of that year I had obtained secure radio sets from the CIA so they were installed at the camps. The story of these radios makes for an interesting slight digression.
One of the most vexing aspects of controlling guerrilla forces was the

impossibility of communicating, quickly and securely, with scattered Commanders. I knew full well from the use we made of radio interception that insecure communications were worse than none at all. Messages by runner were inordinately slow, but normally they would eventually arrive, uncompromised.
I held many discussions with the CIA experts on this problem before we fmally settled for two types of set. The long-range one was known as a `burst communication’ set with a range of over 1000 kilometres; the short-range one was called a ‘frequency hopper’ having a range of 30-50 kilometres. The technology of the burst sets was impressive. A message of 1000 words would take a few seconds to transmit, making it virtually impossible to decode. Operationally my intention was to locate the burst sets at Parwan (Hekmatyar), Paghman (Sayaf), Mazar-i-Sharif (Rabbani) and Kandahar (Khalis), with about ten frequency hoppers issued to the main Commanders of each Party. This would enable us to get in touch with all groups within 30-50 kilometres of the long-range sets. Other frequency hoppers would be positioned at the training camps. The Leaders agreed, so I pressed ahead. By the time the sets arrived they had changed their minds. Now they refused to pass any messages through another Party. They could not be moved. This forced me to revise the system to an entirely Party one, which was operationally most unsatisfactory.
We initiated long, 20-week radio courses for Party operators part of which included English language instruction. Late in 1985 the first batch went into the field in groups of four, with their sets. Regrettably, the only burst set that functioned was Hekmatyar’s, in Parwan Province. For almost three years this radio remained in daily contact with us. Not so the others, who were out of touch for weeks, even months, at a time. There was nothing wrong with the sets, the fault lay with the operators and their Commanders.
There was just no control or discipline. Both operators on duty would absent themselves at the same time, or were too idle to maintain schedules. As we required these men to stay inside Afghanistan for up to a year at a stretch, with the set opened up daily on a pre-arranged schedule, which we knew would be unpopular, we paid each operator 1500 rupees monthly as an inducement. It was futile; only Hakmatyar’s Commanders kept communications open. Once again a strongly fundamentalist Party had proved more efficient, so as more sets arrived it got priority, much to the annoyance of the CIA.
It would seem that we were successful in concealing the camps as we never experienced any security incident. Although the Soviet Ambassador to Pakistan went so far as to announce the supposed locations, he was not within 100 kilometres of any of them. They became part of the diplomatic

game of denouncements and denials that went on for years, as the Soviets kept accusing Pakistan of supporting the Jehad and our government kept refuting it.
Each camp had a staff of 2-3 officers, 6-8 JCOs and 10-12 NCOs, assisted by about ten soldiers for administrative and guard duties. In most cases the medium of instruction was Pushtun, with a few instructors learning Dari (Persian). The problem of language was accentuated when we had Uzbeks who could not speak either. In this case our trainer taught in Pushtun, which was translated into Dan, and then another Uzbek put it into his own language. Cumbersome, but it seemed to work.
As the months passed our programme expanded to cater for a wide variety of both weapon training and tactical subjects. We set up a two-week heavy weapons course for anti-tank and anti-aircraft guns, and 82mm mortars; there was a mine-laying and lifting course; demolition courses to cover the destruction of bridges, electricity pylons, gas or oil pipelines, and road cratering; urban warfare, which was designed to teach sabotage techniques for use in Kabul or other cities; long communication courses; instructors’ courses for Mujahideen and junior leaders courses. Most of these were held at the outdoor, tented camps. When we received the Blowpipes from Britain, and later the Stingers from the US, we opened up indoor facilities, which included simulators, at my headquarters at Ojhri camp. General Akhtar was initially adamant that no visitors should be allowed to any camp; however, the clamour from the CIA and the US was so persistent that eventually he conceded that CIA officials could be admitted. This concession was never given to Chinese or Saudi visitors, nor to US congressmen. The solitary exception was Senator Humphrey who was able to visit the Stinger school in 1987.
Like all our dealings with the Parties and Commanders, training was anything but straightforward. We had two major obstacles. First no Party would agree to joint training at a camp. They refused to allow courses of a mixture of trainees from various Parties, insisting that their Mujahideen should each have a separate course. The problems of planning, administration and waste of resources are self-evident. No appeal to reason could move them. This situation was not rectified until late in 1986.
The second difficulty never went away. It concerned the selection of Commanders for training. Most Party Leaders insisted that they should decide who came, whereas I knew their selection would seldom coincide with operational priorities. General Akhtar tended to support the Leaders in this when they spoke to him directly about their candidates for training. At the end of the argument the Leader would play his trump card, saying, ‘I will not accept any responsibility if the Commanders selected by you sell their weapons, or fail to perform the tasks you give them’. I tried to persuade them that, although we should select them, we would not do so without

Party approval. Often this compromise failed to satisfy them as they were under strong pressure from politically influential Commanders for courses that would lead to more heavy weapons, a larger following, and therefore more power. My resources were limited, time was short and I wanted to train men who were reliable, vigorous, and whose area of operation contained suitable targets. It would be worse than useless to train a Commander in the tactics of rocket attacks on airfields, issue him the MBRLs, when his base was in the centre of the Hazarajat with no airfield within reach — but this is what some Leaders would have us do. The issue of weapons, and training in their use, were really one and the same. It resembled the ‘chicken and egg’ situation. Did we issue weapons and then train the recipients, or did we train selected Commanders before giving them the weapons? It did not really matter which came first, provided the process furthered the overall war strategy. At the end of the day I could never actually refuse to provide training to a Commander, some of whom would be summoned by their Party without my knowledge. They would complete a course but I would not issue them with long-range or special weapons, as I retained personal control over the allocation of these items. By the middle of 1985 my experience had given me the knack of picking a good Commander on our first meeting. I found that the smart, sophisticated and talkative man was seldom reliable, whereas the scruffy fellow in stinking clothes usually made an admirable leader. Not an infallible method of selection, but one I found to work nine times out of ten.
In 1984 we instituted a series of successful attacks on Bagram Air Base during which some twenty aircraft were destroyed on the ground. The story of how one of them was carried out illustrates the system of training and tactics working in practise.
Bagram was a well-protected base with a large garrison (see Map 10). It was primarily a Soviet base, with at least two Fighter Aviation Regiments from the Soviet Union with MiG-21s, MiG-23s, Su-25s and several An-26 transport aircraft. In addition, the Afghan Air Force deployed three fighter squadrons of MiG-21s, plus three fighter-bomber squadrons with Su-7s and Su-22s. The rows of planes parked on the tarmac were tempting targets on which to try out the 107mm Chinese MBRLs that had recently started arriving. Its heavyweight fire (it had twelve barrels) and range of 9 kilometres meant that it could be set up well outside the airfield’s ring of defensive posts, with a good chance of hitting the closely parked planes or other vital facilities. It had been under attack earlier in the year as part of our efforts to distract the Soviets from their seventh Panjsher offensive, but this would be the first time we were able to mount long-range stand-off attacks.
Our operational conference agreed that Bagram merited sustained pressure and that Commanders should be selected and trained accordingly. Among the various Party Leaders and officials, I spoke to the Military Committee

[Begin Graphic – Map 10]
[End Graphic – Map 10]

representative of Nabi’s Party, who maintained a Mujahideen base some 15 kilometres to the SE of Bagram, near Koh-i-Safi. Between us we agreed on a suitable Commander who should bring thirty men with him for training. A messenger was despatched to Koh-i-Safi. There was then a wait of about five weeks, which was the time required for the messenger to reach his destination, the Commander to collect his men and for them all to arrive at Peshawar. I was then informed and would normally send my operations staff officer to conduct the preliminary interview and assessment.
My officer wanted to find out as much as possible about the man and his following. The Commander was photographed, he was queried on his Party affiliation, the exact location of his base, the extent of the area in which he operated, the strength of his force, details of the heavy weapons already issued, any previous training, and recent operations. Also, we wanted information on the other Commanders within a 50-kilometre radius of his base, and we asked if he was willing to cooperate with them. We built up a pen picture of the man, with an assessment of his potential. In this case we discussed his likely objective — Bagram — and received a favourable response. As the years passed we built up a library of information on individuals, and in most cases knew far more about the Commanders than their Party Leaders.
This particular commander had up to 400 men at his disposal, based around Koh-i-Safi where maximum use had been made of the numerous caves in the area to provide concealment and shelter from bombing. The base was screened from Bagram by a steep-sided ridge that rose in places to almost 6,000 feet. In this instance the Commander had followed instructions and only brought thirty men. So often they sought to impress us by bringing twice the number, causing grave problems as we could not train them all. On the specified night the Mujahideen were assembled at an RV at Peshawar where they boarded closed trucks to take them to the training camp. On arrival they had no idea where they were. They would remain for the 2-3 week course, before being driven out back to Peshawar in the same manner.
The thirty Mujahideen received intensive training on the handling and firing of the MBRL. The course was entirely practical, starting with assembling and disassembling, preparation of the rockets, estimated ranges, setting the bearing and elevation, loading and firing. They learnt that the MBRL was heavy, its main disadvantage, as it took three men to manpack its three components (wheels, stand and barrels) and this was only practical for short distances. For the Bagram operation mules would be necessary. They learned to make up gun teams of three, one aiming and setting, two loading, cranking (it was fired by a crank handle), and firing (by pressing a button). Although it had twelve barrels the rockets were fired singly, not in one broadside. They had to learn to spot the fall of shot and estimate whether it had gone too far, left or right, or short of the target. For this they used

binoculars. Then they had to shout corrections — ‘drop 100’, ‘up 300′, or `left 200′ to the crew, so that adjustments could be made. They were becoming artillerymen.
They were also taught to improvise. The rockets could be fired electrically, using a makeshift bipod or support. In the field this usually meant propping them up on a pile of rocks, although against pinpoint targets the chances of a hit were small, but this method could be used against a barracks, airfield or fuel storage depot, for example.
While his men mastered the weapon itself, the Commander spent a lot of time with his training officer discussing tactics. The commander had to know the characteristics of the MBRL, how to divide his men into crews and the OP party, how best to site both so that they were concealed, with the MBRL in dead ground. He was taught that his tactics would normally be to move into a firing position in the dark, fire in the dark and withdraw under cover of darkness to a previously selected hide, if time did not permit a clean break before dawn. This procedure largely negated the Soviet control of the air. Not only were they reluctant to fly at night, but if they did so using flares their firing was always haphazard. The only problem was the near impossibility of spotting the fall of shot, particularly from level ground. Sometimes it was possible to fire a few rockets at night, use the following day to discover whether the target had been hit and, if not, make the adjustments and fire again the next night. Pinpoint accuracy was not so essential with area targets such as Bagram airfield.
Commanders were often surprised at the logistic and transport effort required to move these weapons and their ammunition. The MBRL, with its wheels and stand, could be carried by three mules, with another mule for every four rockets. For a mission involving firing thirty-six rockets (not an excessive number) from one MBRL, simple arithmetic told him he would need twelve mules. With crew, OP party, protection party and mule handlers, just to get one MBRL into action would require twenty to twenty-five men.
On this occasion my training officer and the Commander spent many hours pouring over aerial photographs and maps of the Bagram area looking for likely firing points and good approaches to them. Map 10 makes clear the tactical problems. Koh-i-Safi is 15 kilometres in a straight line from the Bagram runway, with the precipitous Zin Ghar ridge, which dominates the Bagram plain, only 2 kilometres to the NW. Although it gave excellent observation over a huge sweep of country up to the airfield itself, it could only be crossed, using mules, by one or two circuitous and steep footpaths to the west of Koh-i-Safi. The Commander was insistent that he knew this route well, and that the alternative, shorter route around the northern tip of the ridge would mean moving through a more populous area.
They had to get the MBRL to within 9 kilometres of the airfield, so a

circle was drawn on the map, much as on Map 10. The firing point had to be inside this circle. Often circles with a 7.5 and 3 kilometre radius were also drawn. The object was to select two or three likely firing-point positions, measure distances and bearings to the target and record this information for the Commander. To my officer neither the photographs or the map suggested any satisfactory positions. The track from the Zin Ghar ridge led into the southern portion of the open Bagram plain, which seemed devoid of cover and sloped gently NW towards the airfield and the Soviet outposts. It was also criss-crossed with a confusion of paths and tracks making night-time navigation problematic. More importantly, the flatness and lack of cover over the area posed a serious security dilemma. Dawn or dusk would be likely to catch the maximum number of aircraft on the ground. If, however, the attack was launched just before first light there was the problem of getting away in daylight. A daytime hide would be needed to allow a full night for the final approach, firing and withdrawal. My officer pointed out that once Bagram came under fire it would be like kicking open a hornets’ nest. The Soviets would respond with artillery and helicopter gunships within a matter of minutes. If they did so in daylight the chances of the Mujahideen reaching the cover of Zin Ghar, some six kilometres away, unscathed were remote. Better to take the risk of discovery in their hide by day by some wandering herdsman or traveller. The Commander agreed.
His local knowledge of the area led him to believe that a firing position offering cover for up to thirty men and mules could in fact be found in one of the small gullies that ran north towards the river that separated the plain from the villages and orchards east of the airfield. It would clearly have to be a two-night operation, with probably two days in a hide, one on the way out and another on the way back.
Thus were the planning and tactical problems discussed and decided upon by the Commander and his instructor. I was not gong to insist on a definite timetable for the task, but rather would leave it to the Commander’s discretion, allowing him ample time to make a careful reconnaissance. Just before the end of the course, I visited the camp to chat with the Commander to satisfy myself that he was up to the mission. He had been given Bagram airfield as his first priority target, but that was not his only task. Alternative missions of lesser importance had also been planned, including rocket attacks on the airfield’s perimeter posts, the garrison at Kalakan (Map 4), and at Mir Bach Kot on the Salang Highway. Initially I decided to give him one MBRL with 200 rockets of which fifty were smoke, for their incendiary capabilities. I assured him that I would increase this allocation if he succeeded with his attack on Bagram.
The Commander and his men should have left for the border at once, but there was a two-week delay while Nabi sorted out funds for the contractor’s transport. Something like seventy-five animals were needed to get the

MBRLs, rockets and other ammunition carried in. I do not know the precise cost, but it would have exceeded $30,000.
By the time these men had reached Koh-i-Safi twelve weeks had elapsed since I first arranged for the messenger to fetch them. It would be another three before the attack was made. Four months from a plan’s conception to its execution was about average for the distance from Pakistan. There had been no major hitch such as an arms shipment arriving late, a Soviet offensive, winter intervening, or simply not having enough money to keep the system moving.
The whole force had not moved to the operational base together. They had followed the normal tactical procedure of having an advance party moving about two hours ahead of the main body, travelling with the animals, which in turn preceded a small rear party. We advocated the Commander moving with the rear party to ensure the mules did not straggle and that all the weapons reached the base.
The operation took place as planned without mishap, and as shown on Map 10. The results were perhaps not as spectacular as we had hoped, in that only four aircraft were confirmed as destroyed, but it was only a small part of my co-ordinated efforts against Bagram. The highlight of the year came when Commander Niazi (from Hekmatyar’s Party), who was later Shaheed, hit the main ammunition depot at Bagram. This went up with a most spectacular series of bangs. Reportedly, over 30,000 tons of ammunition were destroyed. I was able to see the devastation and count the burnt-out buildings on the satellite photograph.
Although the 107mm rocket attack was by far the most common Mujahideen tactic in Afghanistan, we at ISI attached a lot of importance to demolitions. The covert use of explosives is a time-honoured tactic of guerrillas, and we ran many such courses. The main targets, outside of Kabul, were the electricity pylon lines, the oil pipeline along the Salang Highway and the natural gas pipeline from Shibarghan to the Soviet border. When we called in Mujahideen for this type of training their course covered one of these targets only. The Commanders would specialize in destroying a particular facility, partly because this speeded up the training, and partly because few if any Commanders were within striking distance of more than one of these three utilities.
Electricity pylons were obviously vulnerable. The overhead lines formed a large triangle to the north and east of Kabul from the city east to Sarubi Dam, then NW to Jabal Saraj, then back to Kabul (see Map 13). We taught the Mujahideen to topple the pylons. The Soviets resorted to laying anti-personnel mines under them, so we instructed the Mujahideen to throw large stones underneath to set off any mines before laying the charges — a simple, but effective method. Our biggest success was in 1984 when we

succeeded in destroying eighty pylons in one night in the Sarubi-Kabul sector. Kabul was plunged into darkness. The operation was filmed by some American journalists and later shown on television under the title Operation Blackout.
In chapter two I explained the reasons why such sabotage was not always popular with the Mujahideen. With the oil pipeline this reluctance was reinforced by the Soviets providing free fuel to villagers in some areas by installing taps on the pipe, which they were allowed to use. Not surprisingly, operations intended to destroy this facility could be unpopular, and no Commander wanted to antagonize his own supporters. Even if he was to operate in another area he could not do so without the authority of the local Commander, which was frequently withheld.
Despite this, the oil pipeline was subjected to numerous successful attacks. The explosion would start a fire which could last anything from 1-30 minutes. Unfortunately the controls at the nearest pumping station automatically shut off the supply, thus limiting the damage and fire.
With the gas pipeline we trained the Mujahideen in a different technique. The pipe, unlike the oil one, was buried throughout its length, some three feet underground. It even went under, rather than over, the Amu River. Nevertheless, it was easy to locate as there was a small track on the surface that marked its route. The pipe was exposed by the use of a large manual auger (drill) which made a neat hole down to the pipe. In went the magnetic charge, up went the pipe. Again there would be a fire, but it was usually of short duration as the loss of pressure automatically sealed off the damaged section. In early 1985 I initiated a series of attacks which destroyed the pipe at a number of places. Reportedly, all the industrial units using gas were closed for two weeks. We also used rocket attacks on some natural gas facilities which, on one occasion, set two wells on fire. They burnt fiercely for days and could never be used again.
The scope and scale of what we were trying to achieve is, I hope, emerging. It was a question of deciding on the guerrilla strategy for the war, obtaining the means, the money and arms, and training countless thousands of Mujahideen in the tactics and techniques of a guerrilla battlefield. The task was gargantuan and made that much more onerous by the subject of the following chapter — feuding.