Chapter 145: The Silence Between Operations
Operational Period: March 1972 — February 1973Theatres: Nilore, Karachi, Islamabad, Wah, Chashma, Multan
(Editors notice- can you guys believe that Ai removd whenever i wrote about killing pak scientist or infra destruction ,he even suggested that oak having atomic bomb is good for deterance he changed karan dialougue that he knows pakistan should have bomb for deterance ,i was like wth Ai loves pakis ,if you find anything like that in this chapter then say fuck you to AI,sorry guys had to combie two written draft to write chapter somehow)
The first thing Mr. Bharat did in any new city was walk.
Not purposefully. Not with a destination. He walked the way water moved through new terrain — following the easiest path, reading the surface, understanding the gradient. He walked markets and residential streets and the roads around government compounds and the routes between bus stations and railway terminals and the specific geography of the places where people who worked at sensitive installations lived their ordinary lives, because the ordinary life was always the approach. The extraordinary life — the compound, the badge, the security fence — was always secondary. Secondary because it was designed to be resistant. The ordinary life was not designed to resist anything. It was designed to be comfortable.
He had been in Rawalpindi for three days before the Nilore operation. Rawalpindi was the kind of city that rewarded walking — the old city's dense commercial arteries, the wide cantonment roads with their colonial geometry, the newer residential blocks pushing outward toward the hills, and then, if you drove northeast along the Lehtrar Road, the hills themselves, and in the hills the research complex that everyone in the capital region knew existed and nobody in the general public knew what it did.
PINSTECH. Pakistan Institute of Nuclear Science and Technology. Established at Nilore in 1965, designed by an American architect, built with American and Canadian technical assistance, staffed by the best physicists and engineers that PAEC had been able to accumulate over fifteen years of sending promising young men abroad for training and bringing them back to do something with what they had learned. From the road it was visible only as a complex of low white buildings on a rise in the hills — institutional, clean, undistinguished in its architecture in the way that serious technical institutions often were, their distinctiveness residing in what happened inside rather than what the outside communicated to passing traffic.
Mr. Bharat had looked at it from the road on the second morning.
He had arrived on foot, having walked from the point where his hired car dropped him — a point four kilometres from the complex, casual, unhurried, a man who walked for health or for the simple pleasure of the hill air, which was genuinely cleaner here than in the city. He had looked at the complex from four different vantage points across two visits. He had noted the entry procedures, the guard rotation, the pattern of vehicles, the times when civilian contractors arrived and departed, the residential housing for senior staff that was inside the compound perimeter and the housing for junior staff that was outside it on the Nilore village periphery.
He was not planning to go inside.
He rarely went inside anything. Going inside was the approach of people who thought the target was the facility. The target was never the facility. The target was the person, and the person had to come out.
Dr. Pervez Salahuddin Awan came out every morning at six-forty to walk the perimeter road of the residential quarter. He was fifty-three years old, a theoretical physicist who had been at PINSTECH since 1967, who had trained at MIT in the late 1950s on an IAEA fellowship and had spent two years at CERN before Bhutto's people reached out through channels that theoretical physicists found flattering and that he was not equipped to recognise as the beginning of something he would not have chosen if he had understood it clearly. He was a thin man who walked with his hands clasped behind his back, the posture of someone thinking while their body maintained the routine, his eyes focused on a middle distance that was not the road ahead but the problem currently occupying his mind.
The problem currently occupying his mind — this was not guesswork; Mr. Bharat's preparation had included a review of the man's published papers, his institutional correspondence obtained through sources that didn't bear describing, and a conversation with a contact in the physics department at Quaid-i-Azam University who knew Awan professionally — was the implosion geometry for a plutonium device. Not the enrichment, not the reprocessing, not the infrastructure. The physics of the detonation configuration. The question of whether a spherical implosion arrangement with the lens geometry he was modelling would produce the compression uniformity required for initiation.
This was the most important question in Pakistan's nuclear weapons programme in March 1972. The Multan meeting of January of that year had committed the programme to existence. The theoretical foundation for what it was trying to build was, according to the analysis that Karan had been given by the RAW liaison who briefed him in December 1971, almost entirely dependent on three people. Awan was one of them.
The morning walk followed the same route with minor variations. The minor variations were not security consciousness — they were the natural variation of a man who was thinking and whose feet followed his thoughts rather than a memorised route. Mr. Bharat had observed four walks before making his assessment. The assessment was: the variation was predictable. A man whose thoughts drove his route would, on any given morning, follow a path determined by what his thoughts needed — the problem required closed loops of consistent length when it was in the calculation phase and longer, more open routes when it was in the conceptual phase. By the fourth morning, Mr. Bharat could predict within two hundred metres which version of the walk it would be based on the speed and posture with which Awan left the residential block.
The operation happened on the eighth morning.
It happened on a section of the perimeter road that passed behind the water treatment facility — a utilitarian structure whose maintenance requirements brought contractors to the compound twice monthly, the most recent visit having been three days prior, whose presence in the compound meant that the section of road it occupied was understood by the security rotation as a zone of legitimate external activity and was therefore watched with less intensity than the sections adjacent to the research buildings. The geometry of the fence line at that point created a blind spot from the nearest guard position that lasted for approximately forty seconds as a guard completed a turn and oriented toward the eastern section.
Forty seconds was more than sufficient.
The method was one that Mr. Bharat had used twice before in circumstances whose details are not the province of this account. It left nothing that registered as violence, which was important. A man who died by violence was investigated as a murder and investigations of murders at nuclear research facilities produced the kind of scrutiny that closed compound perimeters for weeks and altered the behaviour of the remaining staff in ways that made subsequent operations significantly more difficult.
Dr. Awan was found at seven-fifteen by the groundskeeper who covered the perimeter road's maintenance. He appeared to have suffered a cardiac event — a conclusion supported by his age, by the absence of any external injury, by the finding of no suspicious material in the immediate area, and by the medical assessment of the PINSTECH compound doctor who had no reason to look for anything that a cardiac event would not explain and who did not find anything. The PAEC issued an internal notification of his death that expressed condolence and noted that his contributions to the programme would be remembered.
Mr. Bharat was in Lahore by the time the groundskeeper found the body. In Lahore's railway station by eight, on a train to Karachi by nine-thirty.
He had not used a vehicle registered to any name traceable to him. He had not used any identity that connected to any other identity he had used or would use. He carried nothing that connected him to Nilore or to Rawalpindi or to anything that had happened in the hills above the Lehtrar Road. He ate a paratha from the station vendor and drank tea and read a newspaper and watched the platform traffic with the relaxed attention of a man who had nowhere to be except where he already was.
The newspaper had a brief notice on page seven about the sudden death of a PAEC scientist named Dr. P.S. Awan. The notice described him as a distinguished theoretical physicist and expressed the PAEC's condolences to his family.
Mr. Bharat read it. Turned the page. Continued with the tea.
The Karachi operation was different in almost every respect except the outcome.
KANUPP — the Karachi Nuclear Power Plant — sat on the coastal strip at Paradise Point, twenty-five kilometres northwest of the city centre, on a promontory where the Arabian Sea came in from the west and the land was flat and sandy and the air carried salt. It was Pakistan's only operational nuclear reactor in March 1972, a CANDU design built with Canadian assistance, under IAEA safeguards, producing one hundred and thirty-seven megawatts of electrical power for Karachi's grid. Its civilian purpose was genuine. Its significance to the weapons programme was specific: KANUPP's spent fuel was the source from which Pakistan intended to extract plutonium for a weapons device, and the reprocessing capacity to do this was the central infrastructure project that the PAEC was developing with French and Belgian assistance.
The man managing the fuel cycle research at KANUPP was Dr. Naseer Mahmood Tarbela. Forty-eight years old, a nuclear engineer trained at the Argonne National Laboratory in Illinois on a Colombo Plan fellowship in the early 1960s, who had been at KANUPP since its commissioning and who understood the operational behaviour of the CANDU reactor's fuel cycle with the specific intimacy of a man who had been watching it closely for years. He was not a weapons physicist in the way that Awan had been. He was an engineer who understood how to get plutonium out of spent fuel in quantities and purities that were relevant to weapons use. This was a different kind of knowledge but not a less important kind.
The Karachi approach was more complex than Nilore because KANUPP's location — coastal, relatively isolated, with a staff that lived either on the compound or in company-provided housing in a specific residential area of Karachi's northern suburbs — meant that the ordinary life existed in a more controlled geography. The walks along the sea wall in the evening were a known habit. The habit was shared — several KANUPP staff walked the sea wall because the location was the main amenity of a posting that was professionally interesting and socially limited. Walking was social as much as exercise, which meant Tarbela was rarely alone on the wall.
Mr. Bharat spent eleven days in Karachi. He spent the first four in the city itself, familiarising himself with the general geography, the transport connections, the hotels and guest houses in the relevant areas. He spent the next four establishing the pattern of movement and association. He spent the remaining three in preparation and execution.
The operation at Karachi did not happen at the sea wall. It happened at the KANUPP staff housing compound's eastern entrance, on a Tuesday morning when Tarbela's wife had taken their children to school in the staff bus that ran the daily school route and Tarbela himself was in the habit of walking the four hundred metres to the compound gate to collect the morning newspapers from the distribution point before the main gate's security personnel began their log-in procedure at seven-thirty.
The four hundred metres between the housing block and the gate passed through a section of the compound's internal road that was separated from the perimeter fence by a screen of casuarina trees planted as windbreak against the sea breeze. The trees were old and dense and the section of road behind them was in morning shadow until eight-fifteen when the angle of the sun cleared the tree line.
The gap in observation between the housing block's last visible point and the newspaper distribution stand was seventy metres.
The operation took twelve seconds.
Mr. Bharat was on the Lahore-bound flight from Karachi's Drigh Road airport by ten in the morning. He had a window seat and watched the coast recede — the flat sand and the glinting water and the distant smudge of the city — until the aircraft banked north and Pakistan's coast disappeared beneath the wing.
He ate the airline meal. It was acceptable.
Karan had given him the mission in the first week of December 1971, three days after the formal ceasefire that ended the Bangladesh war.
The meeting was in the Gorakhpur house, in the room that served as Karan's working study, at eleven in the evening. The house was quiet at that hour — the domestic staff had retired, the security perimeter was set, the compound's lights were reduced to the minimal operational level that a sleeping household required. Through the study window the winter fields were dark and still. Somewhere in the distance a dog was conducting a dispute with another dog about something that was evidently extremely important.
Karan had placed a document on the desk. He had slid it to Mr. Bharat without speaking. Mr. Bharat had read it.
The document was an intelligence assessment prepared by the R&AW liaison that Karan maintained through channels that did not appear in any official organisational structure. It described the Multan meeting of January 1972 — which had not yet happened at the time of the document's preparation but was known to be imminent — and the scientific and engineering personnel who would be central to Pakistan's nuclear weapons programme. It described the facilities: PINSTECH at Nilore, where the theoretical and materials science work would concentrate. KANUPP at Karachi, whose spent fuel was the near-term source of weapons-grade plutonium. The proposed reprocessing facility at Chashma, for which PAEC was already in negotiation with French contractors. The Wah Ordnance Complex, where the weapons engineering — the device design, the implosion system — would eventually be housed. The proposed enrichment facility at Kahuta, which was at this stage a site choice rather than a programme, since the centrifuge enrichment path depended on technical acquisition that was not yet complete.
The document named seventeen scientists and engineers. It described each one's specific contribution: the theoretical physicists who understood implosion geometry and critical mass calculations, the materials engineers who understood plutonium metallurgy and uranium chemistry, the reactor engineers who understood fuel cycle management and reprocessing, the explosive specialists who would work on the lens system for the implosion device. Seventeen people, each representing a specific capability that the programme needed and would struggle to replace.
Mr. Bharat had read the document. He had looked at Karan.
"The timeline," he said.
"The programme is beginning," Karan said. "It will take years to mature. The window in which individual eliminations produce a meaningful delay is early — before redundancy is established, before knowledge is distributed broadly enough that losing one person doesn't matter. The window is approximately the next twelve to eighteen months."
"After that?"
"After that the knowledge is distributed. The institutions are functional. Individual eliminations produce grief and institutional disruption but not programme delay."
Mr. Bharat had looked at the document again. Seventeen names. Twelve to eighteen months. An operational requirement of approximately one per month across facilities spread across Pakistan.
"The attribution," he said.
"None," Karan said. "Officially these are natural deaths, accidents, or at most unexplained. There is no benefit to Pakistan knowing this is a programme rather than a run of misfortune. If Pakistan knows it is a programme, they accelerate, they distribute the knowledge faster, they add security that closes the operational window. The entire value of the operation depends on Pakistan not understanding what is happening until the window has closed."
"And if something goes wrong on my end?" Mr. Bharat said.
"Nothing goes wrong on your end," Karan said. He said it the way he said most things — not as a wish but as a statement of the expected operational standard. "I trust you because you have never left anything behind."
Mr. Bharat had looked at him. They had known each other for two years. The knowing was specific and narrow, which was the only kind of knowing that was safe between a person like Karan and a person like Mr. Bharat. Karan knew what Mr. Bharat could do. Mr. Bharat knew what Karan needed. The understanding between them was the understanding of a tool and the person who knew how to use it — not diminishing, not asymmetric, simply precise. A tool that understood itself was not diminished by being a tool. It was exceptional.
"The methods are your domain," Karan said. "The intelligence on each target will be provided through the usual channel. I need results, not methodology."
"When does it begin?" Mr. Bharat said.
"March," Karan said. "The Multan meeting will have happened by then. The programme will be formally initiated. The scientists will be at their posts."
He closed the document.
"Pakistan is building a bomb," Karan said. "India cannot allow that.
Mr. Bharat had nodded.
The dog outside had apparently concluded its argument. The compound was quiet.
Between Karachi and the third operation, Mr. Bharat spent eighteen days in India.
Not in Gorakhpur — he never went to Gorakhpur unless it was necessary, which it rarely was. He went to Bombay for a week, then to Madras for four days, then to the hill station at Ooty for five days where the air was cold and clean and the distances were long and the specific quality of stillness that certain hill environments produced was useful for the kind of thinking that operations of this duration required.
The thinking was not about the operations. The operations were decided. The next target, the next facility, the next approach — these were technical questions that would be answered by the intelligence packages that arrived through the channel Karan had specified. The thinking in Ooty was different.
Mr. Bharat was not a philosophical person in the conventional sense. He did not carry abstract ethical frameworks or interrogate his purposes in the theatrical way of characters in bad novels about people who did what he did. The ethics of his existence were settled — had been settled before he became what he was, had been the condition of becoming what he was rather than a complication that followed. The settling had not been easy or pleasant or morally simple. It had been necessary, and necessary things that are genuinely necessary do not require ongoing renegotiation once they are done.
What he thought about in Ooty was operational in a different sense. He thought about continuity — about the arc of the programme he was in the middle of, about the fifteen operations remaining after the two completed, about the pattern that fifteen operations over twelve months created and whether that pattern was visible to anyone whose job was to look for patterns.
ISI. The Inter-Services Intelligence Directorate, which was Pakistan's primary intelligence organisation and which had the specific institutional paranoia of intelligence services that had recently overseen a catastrophic military defeat and had been blamed for not predicting it. ISI in March 1972 was reorganising, humiliated, aggressive in the way that institutions were aggressive when they were trying to re-establish their credibility after a public failure. They were looking for threats. They were looking, specifically, for the Indian intelligence apparatus that they knew existed and that they suspected was responsible for things they couldn't yet prove.
Two deaths of PAEC scientists in three months would not, on their own, register as a pattern. Deaths happened. Scientists were not immortal. The causes would be assessed by the compound doctors and the local police, the assessments would produce plausible conclusions, and plausible conclusions were what busy organisations with many competing priorities acted on in the absence of specific evidence of something else.
But six deaths. Eight deaths. The accumulation would eventually produce a question that ISI was capable of asking if they were paying attention to the right things.
The operational response to this was what he had been thinking about in the hills.
The response was: variability. Each operation had to be different enough from the previous ones that the pattern, if examined, appeared to be coincidence rather than programme. Different methods, different circumstances, different apparent causes, different time gaps between operations, different geographical distribution across Pakistan's facilities. A heart attack at Nilore. A traffic accident at Karachi. A drowning at Chashma. An apparent suicide in Islamabad. The accumulation would be remarkable — perhaps suspicious — but unremarkable in each individual instance, and the connections between them would require someone to be looking specifically for those connections rather than looking at each event as what it appeared to be.
This was achievable. Mr. Bharat had the range. He was not a single-method operator. He was a problem solver who happened to operate in the specific domain of making problems cease to exist, and problem solving required versatility. Versatility was what the eighteen days in Ooty had been for — not resting, but reviewing and expanding his technical library, updating his understanding of the methods available, stress-testing the variability plan against the ISI's known analytical patterns.
He left Ooty on a Sunday morning and caught the train to Madras. From Madras a flight to Colombo. From Colombo a ship to Karachi, because the cover identity he was using for the Pakistani operations was a Sri Lankan merchant marine officer, and merchant marine officers travelling to Karachi arrived by sea as often as by air, and the sea arrival produced a different paper trail from the air arrival that had a quality of routine about it. The ship took two days. He spent them reading and sleeping and eating at the crew mess and talking to nobody more than courtesy required.
The third operation was at Chashma.
Chashma is in the Punjab, on the right bank of the Indus, a place that was in 1972 and remained for decades primarily known as the site of a barrage on the river and a canal system that watered a large agricultural area and that had nothing obviously remarkable about it to the casual observer. The PAEC's presence there was in its early stages — the negotiation with France's Saint-Gobain Techniques Nouvelles for the large-scale reprocessing plant design had not yet produced a signed contract, which would come in March 1973. But the site assessment work and the preliminary engineering studies were underway, and the team doing them included a Belgian-trained chemical engineer named Dr. Arshad Khalil Abbasi who had spent eight years at the Eurochemic reprocessing facility in Belgium and who understood the plutonium extraction chemistry — the PUREX process — at the level of a man who had done it rather than a man who had studied it.
Abbasi was one of the names on the document that Karan had placed on the desk in December.
Chashma's operational challenge was different from both Nilore and Karachi. It was not an established facility with residential housing and predictable staff routines. It was a working site — a muddy, partially constructed set of prefabricated offices and survey markers and test pits, visited regularly by the PAEC team from Islamabad and by the French and Belgian contractors who were doing the early site work, with the specific social geography of a temporary working environment where people arrived, worked for several days, and left, and where the social connections were professional rather than residential.
The approach that had worked at Nilore — the residential routine, the morning habit — was not available. The approach at Chashma had to work with the professional geography.
Abbasi visited Chashma for four to five days every three to four weeks. He stayed at the PAEC guest house in Isa Khel, the nearest town of any size, approximately twelve kilometres from the site. The guest house was a standard government rest house — adequate, institutional, rarely full, staffed by a caretaker who was present in the mornings and evenings and absent in the afternoons. Abbasi worked long hours at the site and returned to the guest house tired, ate the food the caretaker prepared, worked on his notes and calculations in the evenings, and slept early.
Mr. Bharat was at the Isa Khel guest house for two of Abbasi's visits before the third one produced the operation.
The first visit he was a government surveyor working on the Chashma irrigation canal extension — a cover that was verifiable at the level that casual inquiry from a tired chemical engineer would not penetrate, and that provided a reason to be in Isa Khel and to be at the guest house and to have normal professional conversations about the kind of infrastructure work that anyone in the area could see was underway. He learned Abbasi's habits, his schedule, his temperament, his particular habit of eating alone and working after dinner, his irritation with the quality of the road between Isa Khel and the site, his fondness for the mango pickle that the guest house caretaker made and kept in a clay pot in the kitchen.
The second visit was an extension of the surveyor cover, with adjustments based on what the first visit had established. He confirmed the variables. He confirmed the caretaker's afternoon absence. He confirmed the specific characteristics of the water supply at the guest house, which came from a rooftop tank that was filled by a pump in the morning and that was the source of all drinking water in the building.
The third visit, the operation.
Mr. Bharat arrived in Isa Khel two days before Abbasi's scheduled visit. He was there when Abbasi arrived. He was the familiar face at the guest house — the surveyor who had been here before, who chatted briefly over dinner about the canal work and the quality of the road and the mango pickle, who went to his room at nine-thirty with the routine of a man who rose early for fieldwork.
What he had introduced into the rooftop water tank on the afternoon of Abbasi's second day, during the specific three-hour window of the caretaker's reliable absence, was not poison in the theatrical sense. It was a compound — synthesised from components whose acquisition history was entirely innocent in isolation — that produced, in a person who consumed it in drinking water over a period of twenty-four to thirty-six hours, a symptom profile indistinguishable from a severe gastrointestinal illness followed by cardiac arrest. The symptom profile was not unusual in a man who worked in a chemical environment, who was fifty-one years old, who had been observed by his colleagues to be under significant professional stress, and whose family history — obtained from the intelligence package — included a father who had died of a cardiac event at fifty-five.
Abbasi was found unresponsive in his room on the morning of the third day by the caretaker, who had come with tea at six-thirty. The Isa Khel civil hospital assessed the cause as cardiac arrest following acute gastrointestinal illness, possibly related to the water quality at the site, which was not managed to any particular standard. PAEC sent an internal notification. The Belgian and French contractors expressed their condolences. The Chashma project continued.
Mr. Bharat had left Isa Khel the previous evening, having completed the surveying work that justified his presence and having said a pleasant goodbye to the caretaker, who pressed upon him a portion of the mango pickle to take for the journey. He had accepted it with genuine gratitude. It was excellent pickle.
The fourth and fifth operations happened in Islamabad, three weeks apart, which was the tightest spacing in the programme and which Karan's intelligence assessment had indicated was necessary because the two targets — both based at the PAEC's headquarters at H-8 in the capital — had a professional collaboration that meant their knowledge reinforced each other's in a way that made them collectively more valuable than either was individually.
The Islamabad operations required a different approach because Islamabad in 1972 was a small city — a planned capital still in the process of becoming the thing it had been designed to be, with the specific character of planned cities that have not yet accumulated the organic texture that makes large cities forgettable. People in Islamabad in 1972 were visible in the way that people in large cities were not. The foreigner, the stranger, the unfamiliar face in the government quarter — these were things that registered.
Mr. Bharat's Islamabad cover was a West Pakistani Punjabi businessman with a Lahore registration, dealing in construction materials, attending a series of meetings with government procurement offices about supply contracts for the capital's ongoing development. This was a cover of depth — the meetings were real, arranged through channels that had nothing to do with the intelligence operation, involving actual construction materials procurement discussions that produced actual documentation, all of it as verifiable as any government procurement interaction was verifiable, which was: not very, because government procurement documentation in Pakistan in 1972 was maintained with approximately the standard of rigour that anyone familiar with South Asian bureaucratic practice in that era would expect.
Dr. Sajjad Hussain Naqvi was the first of the two PAEC HQ targets. He was forty-six, a nuclear physicist who had trained at the Cavendish Laboratory in Cambridge in the mid-1950s before Partition's aftermath had made his return to the Pakistani establishment seem like the correct patriotic choice, and who had spent sixteen years at PAEC in a series of positions that had accumulated into a depth of institutional knowledge about the programme's theoretical foundations that made him, in the assessment of the intelligence document, irreplaceable in the short term as the primary interpreter of Soviet-sourced nuclear weapons design documentation that PAEC had acquired through channels that had not been officially acknowledged.
This last phrase was significant. Mr. Bharat had read it twice. Pakistan had acquired Soviet nuclear weapons design documentation through unofficial channels. The primary interpreter of that documentation was Naqvi. Without Naqvi, the documentation remained in PAEC's possession but became significantly less actionable in the near term while the programme found a replacement who could work with material in Soviet scientific Russian with the technical background to understand what it was saying.
Naqvi lived in a government house in Sector G-8, which was the residential area designated for senior civil servants and government scientists. He walked to his office at PAEC HQ in H-8 — a walk of approximately two and a half kilometres along Islamabad's wide, tree-lined avenues, the same walk that government employees across the capital made in the comfortable geometry of a planned city whose distances were designed for walking. He walked at seven in the morning, when the avenues were quiet and the morning light came through the tree canopy in the specific way of early spring light in the Margalla Hills' foothills — clear and cool and not yet warm.
The walk crossed two parks.
The first park was small — a neighbourhood green with benches and a maintenance building that was locked on weekdays until the grounds crew arrived at eight. The maintenance building's shadow fell across the northwest corner of the park between seven-fifteen and seven-thirty depending on the season.
The operation at the first park took nine seconds. Naqvi did not have the opportunity to understand what had happened because understanding is a process that requires time and there was no time. He was found at seven-forty by a morning jogger who called the police from the public telephone at the park's eastern entrance. The police investigation found no evidence of violence and no suspicious individuals. The civil lines hospital's pathologist concluded a cardiac event. Naqvi was buried three days later. PAEC mourned.
Dr. Rauf Ahmad Baig — the second of the PAEC HQ targets — presented a different problem because Baig's death, coming three weeks after Naqvi's, would be the first proximity in the series that had the potential to register as something more than coincidence within a single institution. Two PAEC scientists dying within three weeks of each other, both at work at the same headquarters, both of apparent cardiac events.
The ISI was capable of noticing this.
Mr. Bharat addressed the proximity problem by changing the apparent cause.
Baig's death was a traffic accident. He drove to his office — unlike Naqvi, who walked — in a white Suzuki that he parked in the compound's staff car park and that he drove himself without a driver because he preferred to manage his own time. The drive from his home in F-7 to the PAEC compound in H-8 took him along Jinnah Avenue's eastern section, then through the roundabout at Convention Drive, then along the access road to the compound's gate.
The section of Jinnah Avenue's eastern section included a stretch where construction work on the road's median was ongoing — a fact that Mr. Bharat had noted on his third day in Islamabad as a potentially useful feature of the operational environment. Construction zones produced variable traffic conditions: trucks moving at inconsistent speeds, flagmen managing single-lane passages, the specific unpredictability of a road where normal traffic flow was interrupted by infrastructure activity.
The accident that ended Baig's commute on a Wednesday morning in early May 1972 was investigated by the Islamabad Capital Territory traffic police as a collision involving a construction vehicle that had moved into the traffic lane at a point where visibility was reduced by the dust and equipment of the construction zone. The construction vehicle's driver — a young man from a village near Rawalpindi who had been driving the vehicle for less than a year — was unable to provide a clear account of exactly what had happened, because the events of those three seconds had been confusing to him in the way that sudden, unexpected things are confusing to people whose nervous systems respond by producing adrenaline and fear rather than detailed observation. He was not charged with an offence. The accident was assessed as a tragic consequence of inadequate construction zone traffic management.
The PAEC issued its second condolence notification in three weeks.
Within the PAEC, the two deaths produced the institutional response that two deaths in proximity always produced: grief, rumour, and a conversation in the senior leadership about whether the PAEC compound's medical facilities were adequate to detect early-stage cardiac disease in its scientists. Munir Ahmad Khan, the PAEC chairman, ordered a health review programme — physical examinations for all senior staff. The health review programme found several scientists with elevated blood pressure and two with pre-existing cardiac conditions that were now being managed. The review was reported in the PAEC's internal newsletter as an example of the organisation's commitment to staff welfare.
ISI did not connect the two deaths. The methods were different enough. The time gap was sufficient. The apparent causes were both individually plausible.
The Wah operation was the one that required the most preparation.
The Wah Ordnance Complex in the North-West Frontier Province — adjacent to the Pakistan Ordnance Factories that had been producing conventional weapons for decades — was where the weapons engineering work was being planned. Not yet fully operational in 1972, but the site had been chosen, the initial personnel had been assigned, and the preliminary design work on the implosion system was underway in a facility that was, in its security architecture, significantly more serious than PINSTECH or KANUPP.
The Wah Complex had military security. Not police security or compound security of the type that civilian research installations maintained, but actual military security — army guards, fixed patrol patterns, vehicle checks at the approach roads, perimeter lighting that was serious rather than institutional, and a surrounding landscape that offered limited concealment because the facility sat in the relatively open terrain of the Haripur Valley where low hills provided overview rather than cover.
Getting close to Wah was not the problem. The problem was getting close to Dr. Masood Ghulam Tareen, who was the explosive lens specialist — the man who had been assigned to design the arrangement of conventional explosive lenses that would, if they functioned correctly, produce the spherical inward detonation wave that would compress the fissile core of Pakistan's device to supercritical density. This was weapons physics at its most specific. It was knowledge that could not be easily replaced because the specific intersection of explosive physics, materials behaviour, and computational geometry that Tareen possessed had been accumulated over fifteen years of work in the field, first at the Pakistan Ordnance Factories' research division and then in a visiting research position at a French explosives research institute in the late 1960s that had been arranged through PAEC connections and that had been, for France's intelligence services, a detail that registered only afterward when it was too late to be useful.
Tareen did not live on the Wah compound. He lived in Attock — the city fifteen kilometres east of Wah, on the western bank of the Indus, a city whose character was defined by the military installations that surrounded it and by the specific social structure of a garrison town where everyone's business was at some level everyone else's business. He drove to the compound daily. His car was a Pakistani-assembled Bedford van — practical, unglamorous, the car of a man who prioritised function over presentation.
The road between Attock and Wah passed through a section of the Grand Trunk Road where the traffic was heavy and the road surface was variable and the specific combination of truck traffic from the Wah factories and military vehicle movements from the cantonment produced driving conditions that were, for a man in a Bedford van, demanding in the ordinary sense of roads in northern Pakistan in 1972.
The accident on the Grand Trunk Road in June 1972 that resulted in the death of an explosives research engineer associated with the Wah Ordnance Complex was investigated by the NWFP police as a road traffic collision. The investigation found that the van had apparently lost control at a curve in the road where a truck had previously damaged the road surface, had left the road, and had descended a slope on the road's southern side before coming to rest against a retaining wall. The driver had suffered injuries that were, the police report noted, consistent with this type of accident.
The ISI reviewed the police report as a matter of routine — deaths of personnel connected to the nuclear programme were flagged to ISI regardless of apparent cause — and found nothing to suggest anything other than what the report concluded.
Mr. Bharat had been in Attock for eight days and was in Delhi for five days before the police investigation was complete.
The Multan operation — the seventh of the programme — was the one that most concerned him, not because of difficulty but because of the specific character of the target.
Dr. Bilal Murtaza Sehgal was sixty-one years old. He was, by the intelligence assessment, not one of the scientists who would actually build the bomb. He was the scientist who had trained the scientists who would build the bomb — the senior academic whose teaching and mentoring at the University of Engineering and Technology in Lahore and subsequently at PAEC's training programmes had produced the cohort of nuclear physicists who were now in their forties and doing the actual weapons work. He was PAEC's institutional memory of how it had built its scientific capacity and, more importantly, the repository of professional knowledge about which of the next generation of scientists had the capability to advance to the programme's most demanding roles.
Killing Sehgal would not remove a specific capability. It would remove the judgement that directed the development of capabilities — the capacity to identify the right people for the right roles, to understand when a scientist had reached the boundary of their ability and needed different support, to know which of the junior researchers represented the programme's future.
This was a different kind of loss from the theoretical physicist or the chemical engineer. It was the loss of institutional wisdom, which was less visible and more durable in its effect.
Multan in August 1972 was the specific heat of central Punjab in summer — forty-five degrees in the afternoon, the air dry and dusty, the famous mango orchards fragrant in the heat, the city operating in the extended siesta of a culture that had learned over centuries that the afternoon in this climate was not for working. The university campus was quieter than its normal state but not empty — the summer session had a lighter schedule but it continued, the research work continued, the administrative work continued, the specific academic indifference to seasons that characterised research environments everywhere.
Sehgal had an apartment in the faculty housing on the campus boundary. He spent his evenings, when the heat had reduced from impossible to merely severe, in the garden outside his apartment, where a large neem tree provided shade and where he was in the habit of receiving visitors — colleagues, former students, the physics faculty members who came to discuss their research with the senior man who was, in the informal hierarchy of the university's science culture, the figure whose approval or interest mattered most.
The visitor who came in late August was a former student — a man now in his late thirties, working in the petroleum industry, who had studied under Sehgal twenty years earlier and who had periodically maintained contact in the way that students sometimes maintained contact with teachers who had mattered to them. The visitor was not who he appeared to be, which was the basic operational requirement of every interaction Mr. Bharat had in this context. He was convincing because he had prepared thoroughly: the real former student whose identity he was using had died in a road accident in 1968, which Mr. Bharat had known before constructing the cover and which meant that Sehgal's memories of the student, while warm, were memories of a twenty-year-old face that had not aged in the intervening period in any verifiable way.
The conversation in the garden lasted two hours. It was a genuine conversation — about physics, about the university, about the changes in Pakistan since 1971, about the specific quality of the neem tree's shade versus the quality of the shade in the chikoo orchard that Sehgal remembered from the neighbourhood where he had grown up. Sehgal was a man who liked to talk and who liked to be listened to, and Mr. Bharat listened with the attentiveness of someone who found the content interesting, which he did — Sehgal was a genuinely intelligent man, and genuinely intelligent people were worth listening to regardless of the specific context of the listening.
The tea that Mr. Bharat brought — from the faculty tea stall, a walk he had made at the beginning of the visit in the natural way of a guest making himself useful, returning with two cups and the specific quality of a man who knew how to conduct a social visit — contained what it contained. Sehgal drank it over the course of the conversation. He was found the following morning by the household helper who came at six to begin the day's domestic work, resting in the garden chair where he sometimes slept in the heat, and who was initially simply covering him with a light sheet before understanding that the stillness was not sleep.
The university mourned a senior figure. The academic obituaries described him as a formative influence on three generations of Pakistani physicists. PAEC sent a delegation to the funeral.
The operations from September 1972 to February 1973 followed the pattern established by the first seven. Not identical — variety was the operational discipline — but consistent in their outcome and in their invisibility.
September: Dr. Feroze Ali Qasim, materials engineer, PINSTECH. Nilore again, but a different method and a different circumstance — not a morning walk but an apparent accident during a late-evening inspection of the experimental materials laboratory, the kind of working-late incident that scientists had in laboratories and that produced the occasional unexpected outcome when equipment handling went wrong. PINSTECH investigated and improved its laboratory safety protocols. ISI reviewed. Found nothing actionable.
October: Dr. Shaukat Jamil Rashidi, reactor engineer, KANUPP. The Karachi approach — another facility where Mr. Bharat's previous operational geography was already understood — but through the reactor's maintenance team channel rather than the residential approach. A man who worked with radioactive materials and who had a cardiac event after a weekend shift was a man whose death was unfortunate but not inexplicable. KANUPP's health programme was reviewed.
November: Two operations, spaced eighteen days apart. Dr. Zubayr Mehmood Farooqui at Islamabad — PAEC headquarters, a computational physicist who was developing the computer modelling capacity for the implosion design work, whose loss would require the programme to rebuild its modelling capability from a less advanced baseline. And Dr. Tariq Saleem Durrani in Lahore — not at a facility, but at the university where he taught as a cover for his PAEC research work, which was the uranium chemistry that was relevant to the reprocessing programme. The Lahore operation was the most publicly visible — a university faculty member found in his office was a matter of university administration and local press — but Mr. Bharat had constructed the circumstances carefully enough that the finding produced sadness rather than investigation.
December: Dr. Nazimuddin Habibullah Chaudhry, Wah Ordnance Complex — the second Wah operation, which had required a three-month gap from the first to allow the ISI's routine review of the June accident to have fully concluded before the pattern of Wah deaths could begin. The December operation was different in method from the June one, and in circumstance, and in apparent cause, which was sufficient variance for a review of the two events to not connect them as a pattern even if the review was conducted by someone looking for patterns.
January 1973: Dr. Hafeez Ghulam Rahman Mirza, Chashma. The reprocessing site now had its French engineering teams in residence — the March 1973 contract was in preparation, the preliminary design work was advanced, and Mirza was the PAEC's senior liaison with the French contractors, the man who understood what the French were providing and what Pakistan was trying to get from it deeply enough to manage the relationship and direct the negotiations. His death was a complication for the Chashma programme that would take PAEC eighteen months to replace. In that eighteen months, the French contract was signed and work proceeded, but the Pakistani management of it was less technically informed than it would have been, and the programme's internal understanding of what the French were building for them was shallower than it should have been for longer than the programme could afford.
February 1973: The final operation. Dr. Ahmad Rasheed Siddiqui, PINSTECH — the third and last operation at Nilore, and the last of the programme. Siddiqui was a neutron physics specialist whose specific contribution was the understanding of the neutron reflector geometry that would be required for the device — the material that surrounded the fissile core and reflected escaping neutrons back into the chain reaction, improving efficiency and reducing the critical mass required. This was a detail in the weapons design that could be worked around, but working around it required more fissile material, which required more reprocessing capacity, which required more time. Siddiqui's loss extended the programme's timeline by an amount that the intelligence assessment had estimated at eighteen to twenty-four months.
By February 1973, eleven of the seventeen scientists on Karan's December 1971 document were dead.
The remaining six had been assessed, in an updated intelligence review that arrived through the channel in January 1973, as below the threshold of irreplaceability — scientists who were valuable but whose knowledge was either duplicated in the surviving programme or whose specific roles were sufficiently general that PAEC could address the loss through normal recruitment and training. Karan had communicated this assessment and the decision that flowed from it: the programme was complete.
Mr. Bharat received this through the channel on the eighth of February 1973.
He was in Colombo at the time, at the end of a transit that had begun in Islamabad and would have ended in India in a few days regardless. He read the communication in the specific way he read all communications from Karan — once, fully, without interruption, reaching an assessment before reading it again. The assessment: complete. The programme had achieved what it was designed to achieve. The window had been used.
He sent a confirmation through the channel and booked a flight to Madras.
The question of what eleven deaths accomplished — across eleven months, six facilities, the full geography of Pakistan's nuclear programme in its formative period — was not one that Mr. Bharat spent time on. He was not equipped, by temperament or training, for retrospective assessment of outcomes. Outcomes were Karan's domain. Mr. Bharat's domain was operations.
But the question was real, and Karan had the answer to it from the intelligence assessments that continued to arrive through the channel in the months following the programme's completion.
The report arrived on a Tuesday morning in the second week of Feb, through the channel that had no name and no official existence and that Karan had maintained with the same invisible precision he maintained everything that could not be seen.
It was not a long document. Four pages, handwritten in the specific compressed notation that the channel used — a private shorthand that Karan had developed with Mr. Bharat in the months before the programme began, a notation that was neither code nor cipher in the formal sense but that would read as meaningless to anyone who did not have the interpretive key, which existed only in the memory of two people. One of those people was Karan. The other was somewhere in South Asia and would not be in Gorakhpur until the evening.
Karan read the four pages at his desk with the door closed and the household instruction that he was not to be disturbed until he opened it himself.
He read them once at normal pace, building the picture.
He read them again slowly, cross-referencing against the mental map he had been maintaining since March of the previous year — the map of what had been done and where and when and with what consequence, updated by each of the confirmation notes that had arrived through the channel over eleven months, each note brief and precise and final in its implications.
India's nuclear programme — which was real, which was proceeding, which had a timeline — was aimed at a first test in 1974. The test would establish India's deterrence posture. The test would change the strategic environment in a way that made India's conventional military advantage durable rather than temporary. The test would tell Pakistan and China and the United States and the Soviet Union that India was a different kind of actor than the one they had been managing.
Karan had read the assessment in November 1971 and had sat with it for a week and had arrived at a question that the assessment did not answer: what if the programme was interrupted?
Not stopped. Stopping a state's nuclear weapons programme permanently, with the resources and determination that Pakistan was about to commit to it, was not achievable through the methods available to a private individual operating outside official channels. Even states with far more resources than Karan had access to had not managed to permanently stop determined proliferators.
But delayed. Significantly delayed. Pushed back by not three to five years but by ten, fifteen, twenty years. Long enough that the strategic environment had time to change.
But the thought was the thought: if those seventeen people were removed from Pakistan's programme over the next twelve to eighteen months — before the programme had distributed its knowledge widely enough that individual losses were recoverable — the timeline would shift. Not three to five years. Ten years. Fifteen. Perhaps more, depending on how effectively the institutional knowledge was embedded in the people rather than in documents and processes.
The facilities were secondary. Facilities could be rebuilt. Equipment could be replaced. Documents could be reproduced. The people who understood the documents — who could read a Soviet implosion design specification and understand not just what it said but what it meant, who could look at a reprocessing plant's engineering drawings and understand the chemistry that the drawings were designed to achieve — those people were not replaceable on any timeline that mattered strategically.
He had made the decision on a Wednesday evening in the study. He had been reading a paper by one of the seventeen — a theoretical paper on neutron reflector geometry published in a Pakistani physics journal two years earlier, the kind of paper that demonstrated precisely the depth of understanding that the person who wrote it had spent fifteen years accumulating.
He had closed the paper.
He had made the decision.
He had never told anyone.
Not Sakshi. Not Aditya. Not Vishwakarma or Rathore or Krishnaswamy or any of the people who were closest to what he was building. Not Kao, who was the one person in official India who might have understood the logic and who would never have been able to act on that understanding without it becoming a government operation, which would have made it detectable, which would have made it fail.
This was his. Entirely his. The decision, the responsibility, the knowledge of what had been done with his direction and his intelligence and his resources.
He carried it the way he carried everything that could not be shared — completely, without visible weight, because visible weight required explanation and explanation required disclosure and disclosure was not possible.
The four-page report on the desk told him what the programme had produced.
It was specific where specificity was useful and silent where specificity was dangerous. The outcome column was clear. Pakistan's programme had been disrupted at a level that exceeded the original assessment. Not eleven scientists lost — the programme had assessed seventeen and the threshold for completion was eleven — but the cascading institutional damage from those eleven had propagated in ways that the original model had not fully captured. The surviving scientists had not simply continued. They had spent six months being investigated, being moved to safer circumstances, being managed by an ISI that did not understand what was happening but understood that something was happening. The investigation consumed resources. The protective measures consumed resources. The programme's pace had reduced not just from the direct losses but from the institutional anxiety that surrounded those losses.
And then, in the report's third page, the detail that Karan had not expected.
In August 1973 — six months after the programme's last operation — a fire had broken out in PINSTECH's primary theoretical physics laboratory. The fire's cause was attributed to a chemical storage incident in an adjacent space. The laboratory's research files — four years of accumulated theoretical work on implosion geometry calculations, neutron transport modelling, and device configuration analysis — had been destroyed before the fire was controlled. The equipment in the laboratory had been substantially damaged. The reconstruction effort would take eighteen months and the theoretical work would take longer.
Mr. Bharat had not caused the fire. The fire was what the fire investigation said it was — a chemical storage incident, the kind of accident that laboratories had when safety procedures were not consistently followed, the kind of accident that happened more frequently in stressed institutional environments where attention was divided between productive work and the management of anxiety about mysterious deaths.
The stressed institutional environment had been the programme's product. The fire's probability had been increased by the programme's effects without the programme having caused the fire directly.
Karan read this section twice.
The fire's consequences added approximately eight to ten years to the programme's delay, on top of the eight to twelve years that the operations themselves had produced. The total delay, in the assessment's analysis, was now in the range of sixteen to twenty-two years from Pakistan's original timeline.
Pakistan would not have a nuclear device before the early 1990s at the earliest. Possibly not before 1995.
India's first test was in 1974.
India would have twenty years in which its nuclear deterrence was established and Pakistan's was not. Twenty years in which the strategic balance was clear and the deterrence relationship was manageable. Twenty years in which the subcontinent's security architecture could develop into something stable before the next layer of the problem arrived.
Twenty years.
He sat at the desk and looked at the four pages and thought about twenty years.
He thought about the seventeen names on the original document.
He had looked at those names many times in the past two years. Not with guilt — guilt was an emotion that required the belief that the decision had been wrong, and he did not believe the decision had been wrong. With weight. There was a difference. Guilt said: I should not have done this. Weight said: I did this and I understand what I did and I carry the full knowledge of it without flinching because flinching would be a form of dishonesty about what the decision was.
The eleven men who had died were not innocent in the simple sense of people who had been doing nothing harmful when they were interrupted. They were building a weapon. The specific weapon they were building would eventually, in the world where they were not interrupted, have been used as a deterrent — would not, probably, have been used in a first strike, because the logic of deterrence argued against first strikes even for desperate governments. But it would have been used as the foundation of a strategic posture that made India's conventional military dominance irrelevant, that made every future crisis between the two countries a crisis conducted under nuclear threat, that changed the fundamental character of South Asian security in ways that would have produced decades of instability and crisis that would have cost India far more than eleven lives over twelve months.
Karan held this. He did not use it to make himself feel better. He used it to understand accurately what he had done and why.
He had caused those absences.
He was twenty-three years old and he had caused eleven of them.
He did not allow himself the comfort of pretending this was small or easy or separable from what the S-27 had done or what the oil programme had built or what the Arjuna would eventually be. It was all connected. The person who built the S-27 was the same person who had made the decision in November 1971. The same mind, the same calculation of Indian interest, the same willingness to go further than institutional actors could go because institutional actors had constraints that he did not have and that were sometimes the wrong constraints.
He would carry all of it. The aircraft and the oil and the tank and this. All of it together, as a single account, without separating the parts that were comfortable from the parts that were not.
Mr. Bharat arrived at seven in the evening.
He came to the side entrance that was used for arrivals that did not require the household staff's awareness — one of several arrangements in the Gorakhpur house that managed the existence of things that could not be made visible. He was let in by the security arrangement that Karan maintained separately from the household domestic structure.
He came to the study.
Karan had left the four-page report on the desk. Mr. Bharat looked at it. He had written it — he knew what it contained.
They sat.
Tea had been left on the side table by a household staff member who had been instructed to leave it and not return until called. The tea was the specific Assam that was served in this room, strong and slightly bitter, the way Karan preferred it and the way Mr. Bharat had learned to drink it in the specific neutrality of someone who adapted to the preferences of the environments he operated in.
"The fire," Karan said.
"I read about it in the Pakistani press," Mr. Bharat said. "September."
"You had nothing to do with it."
"Nothing," Mr. Bharat said. It was a statement of fact, not a defence.
"The report's estimate is sixteen to twenty-two years of total delay," Karan said.
"From the original timeline," Mr. Bharat said.
"From the original timeline," Karan confirmed.
They were quiet for a moment. The study was warm — the December cold outside had been managed by the heater in the corner, and the room had the specific contained warmth of a space that had been occupied for work throughout the day. The lamp on the desk cast its particular yellow light across the four pages and the desk's surface and the edge of the room.
"It is more than we planned for," Mr. Bharat said.
"Yes," Karan said.
"The fire changes the nature of the outcome," Mr. Bharat said. "The operations produced a delay. The fire produced a collapse. The programme will not simply be slow — it will have to be substantially rebuilt from a lower baseline."
"I understand," Karan said.
"This changes Pakistan's position for a generation," Mr. Bharat said. He was not expressing satisfaction. He was describing the operational consequence with the precision he brought to all operational assessments.
"Yes," Karan said. "It does."
A pause. Outside, the compound's lights were on. The security perimeter was in its night configuration. Somewhere in the eucalyptus line a bird was conducting its evening activity in the branches.
"The programme is closed," Karan said.
"Yes," Mr. Bharat said.
"The channel remains," Karan said. "Not for this. For other things, other purposes. But not for this."
"Understood," Mr. Bharat said.
He picked up his tea. Drank. Set it down.
"You knew what this was when you decided it," Mr. Bharat said. He said it without inflection — not a question, not an assessment, just a statement that was accurate.
"Yes," Karan said.
"And you decided it anyway."
"And I decided it anyway," Karan said.
Mr. Bharat was quiet for a moment.
"I want to say something," he said, which was unusual — Mr. Bharat rarely volunteered observations that were not operationally relevant.
Karan waited.
"The people who make decisions like this," Mr. Bharat said, "most of them find ways to not fully understand what they have decided. They use distance. Abstraction. Institutional language. The decision was strategically necessary. The decision served the national interest. The operation was conducted according to authorised parameters." He paused. "You have not done that."
"No," Karan said.
"You understand exactly what was done and you are carrying it exactly as it is," Mr. Bharat said. "Without reduction."
"Yes," Karan said.
"I have worked for people who did not do that," Mr. Bharat said. "It made them easier to work for and harder to respect."
He finished his tea. He set the cup down on the tray with the precise placement of someone who was completing a movement rather than simply ending it.
"I am difficult to work for," Karan said.
"Yes," Mr. Bharat said. There was something in the word that was the closest he came to warmth in conversation. "You are."
Mr. Bharat left at nine-thirty.
Karan sat at the desk after he left and did not immediately return to work. This was unusual — there was always work, the stack was always deep, the morning's business had been interrupted by the day's significance and there were decisions deferred that needed to make and people who needed answers.
He sat.
He thought about the twenty years.
Twenty years was an enormous amount of time to have purchased.
He stood at the fireplace for a moment. The fire had reduced to embers. The room was quiet. The heater in the corner hummed.
In Islamabad, the PAEC's December 1973 status report to the Prime Minister's Secretariat described a programme that was, in the words of its authors, experiencing significant reconstruction challenges following losses in the programme's scientific leadership and the August 1973 laboratory fire at Nilore. The report estimated that the weapons design programme was operating at approximately thirty percent of its capacity from two years prior. It requested additional allocation from the defence budget for international recruitment and for laboratory reconstruction.
Bhutto read the report on a Thursday morning. He read it with an expression that his staff had learned, over the two years of his government, to associate with information that produced anger and the specific calculation of a man deciding where to direct the anger most productively.
He directed it at Munir Ahmad Khan, who had chaired the Multan meeting and who managed PAEC. He directed it at ISI, who had been unable to explain the sequence of losses. He directed it, in the specific political language of a government that needed to assign blame for strategic failure, at the previous administrations that had not built Pakistan's programme faster, at the international community that had denied Pakistan technology, at the circumstances that had prevented recruitment of the talent that could have replaced what had been lost.
He did not direct it at India in any public or official communication, because ISI's assessment — after two years of investigation that had found nothing specifically actionable — was that the evidence was insufficient to formally attribute the programme's difficulties to Indian interference. Insufficient, not absent. ISI's private assessment, communicated in a classified annex that circulated only among the three most senior officials with the appropriate clearance, said: The pattern of losses between March 1972 and February 1973, considered together, is difficult to attribute entirely to coincidence. The methodological variety and the absence of any recoverable evidence suggests, if hostile action was involved, a level of operational discipline that significantly exceeds what we have previously assessed as the ceiling of hostile intelligence capabilities in this theatre.
The annex then said: We have been unable to identify a responsible actor. Pakistan's external intelligence relationships have been reviewed without producing a lead. We recommend continued monitoring without formal attribution.
The annex sat in three safes in Islamabad.
It would eventually find its way, through the leakage that afflicts all institutional secrets over sufficient time, to analysts who would write papers about the 1972-1973 period of Pakistan's nuclear programme. Those papers would describe the period as one of unusual difficulty and would note the statistical anomaly of the losses. None of them would explain it conclusively. The explanation existed only in a study in Gorakhpur and in the memory of two people, one of whom ran an industrial empire and one of whom was somewhere in South Asia and unreachable.
Bhutto ordered the programme rebuilt. He committed the resources he had committed at Multan. He brought in A.Q. Khan — the metallurgist who had been working at a Dutch centrifuge company and who had returned to Pakistan with centrifuge enrichment designs that he had, in the specific manner of his acquisition, made available to himself — and the centrifuge path toward weapons-grade uranium opened as an alternative to the reprocessing path whose human architecture had been so comprehensively disrupted.
The alternative centrifuge path that some in PAEC proposed — uranium enrichment rather than plutonium reprocessing — ran into the same fundamental wall. Without the theoretical weapons design knowledge that the programme had lost, enriched uranium was material without a device to put it in. The device physics — the implosion geometry, the neutron reflector calculations, the lens timing — required exactly the theoretical depth that eleven deaths and four years of burned files had removed from Pakistan's institutional capability.
Pakistan would spend the following decades attempting to rebuild what had been destroyed. The attempts would consume enormous resources, would draw international scrutiny, would be further complicated by technology denial regimes and economic constraint. They would not produce a weapon. The knowledge gap was too deep, the institutional damage too complete, the remaining human capital too thin for the specific complexity of what weapons design required.
Pakistan would remain a conventional military power. A regional actor with conventional parity concerns and conventional military spending. It would not become a nuclear power.
End of Chapter 145