On 7 August, in Makkah, Recep Tayyip Erdoğan, Mohammed bin Salman and Shehbaz Sharif signed an agreement stipulating that an armed attack against any one of the three states shall be regarded as an attack against them all. The clause is Article 5 in substance if not in name, and it binds together the Muslim world’s only nuclear-armed state, NATO’s second-largest army and the world’s largest oil exporter. Ishaq Dar has said the arrangement remains open to others.
Within hours the commentary settled into its expected grooves: what it means for Iran, what it means for India, whether Washington is pleased, whether the commitments are real. Reasonable questions. None of them the important one, I must say.
The important question is what Pakistan brings to that table over the next twenty years, and whether the thing it brings is manpower or industry. Those are different futures. The choice between them is being made now, largely by default, by an apparatus that has never been asked to make it.
The received wisdom is wrong
Let us begin with the current defence production ecosystem, because the belief that Pakistan cannot find enough engineers is the first thing that has to go. Count them. NESCOM was assessed at roughly sixteen thousand staff, a figure nearly two decades stale and certainly at least twice that now, carrying the National Defence Complex at Fateh Jang, the Air Weapons Complex at Kamra, the Maritime Technologies Complex, NECOP, and some more. KRL runs several thousand more. SUPARCO, DESTO and the Directorate-General Research and Development Establishment (DGRDE) add a few thousand between them. The Atomic Energy Commission is reported above a hundred and twenty thousand people, but the overwhelming bulk of that is power generation, oncology and agricultural isotope work; the strategic slice is perhaps ten thousand.
Pakistan Ordnance Factories (POF) states on its own site a workforce of over twenty-seven thousand across fourteen factories at Wah. Pakistan Aeronautical Complex (PAC) runs four rebuild and manufacturing factories plus AvRID. Heavy Industries Taxila (HIT) carries roughly five thousand two hundred. Karachi Shipyard and the Naval Dockyard, several thousand. NRTC at Haripur, a few thousand more.
The base runs to something near a hundred thousand people. Compare like with like, which is the only comparison worth making: Israel’s military-industrial workforce approaches a hundred thousand as well, with the first tier of IAI, Rafael, Elbit, Israel Shipyards, Tomer, Aeronautics and the Ministry of Defence’s Tank and APC Directorate accounting for roughly forty-five per cent of it. Türkiye’s is larger still. SaSaD recorded 67,239 in 2018; Ankara province alone now carries between sixty-five and seventy-two thousand, which its own recruiters put at forty-two per cent of the national total. On raw headcount Pakistan is behind neither of them.
What proportion are engineers is harder, because nobody in Pakistan publishes it. ASELSAN’s investor presentation states the figure without hesitation: 5,824 engineers of 9,594 staff, 61%, with 29% technicians beneath them. TUSAŞ published seventeen thousand employees comprising seven thousand engineers and six thousand technicians. Roketsan discloses two thousand of five thousand in research and development. No Pakistani defence entity has ever published the split. POF, PAC, NESCOM, HIT; none.
The silence is the first piece of evidence. An organisation that measures itself by design density publishes design density. An organisation that measures itself by units delivered to a single customer publishes units delivered to a single customer. This is definitely not the secrecy but the fact that the ratio was never a management variable.
Estimate it anyway, but refuse to apply one number across a base that is not homogeneous. A filling factory at Wah and a seeker laboratory at Kamra are not the same kind of place. The result lands near 17,700 engineers on a base of 97,000 – a blended density of about 18%.
Seventeen thousand. Roughly what Rafael, Elbit and Israel Aerospace Industries carry between them, and roughly what ASELSAN, TUSAŞ and Roketsan carry between them; except those are three firms apiece and this is a country. Nation against nation the ranking reverses. Türkiye fields thirty to fifty thousand, Israel comparable or more, on estates no larger than Pakistan’s own.
So the deficit is not in the pipeline. Eighteen per cent against a third, across the same hundred thousand people. Pakistan produces engineers in numbers nobody disputes and at a cost nobody can match. It put fewer of them in the rooms where things are designed.
The ratio, and why it is not an indictment
Set the workforce against what the workforce sells. Pakistan’s officially reported defence exports stood at $416M in 2023, against $210M in 2019. Türkiye exported $7.15B in 2024. Israel exported $14.7B in the same year.
Per head of defence-industrial workforce, measured at the same scope for all three: roughly $4,300 in Pakistan, between $42,000 and $71,500 in Türkiye depending on where its sector headcount is actually pinned, and about $147,000 in Israel. Ten to seventeen times, and thirty-four times.
Now normalise away the density gap, because that is the number that isolates what is actually broken. Per engineer, Pakistan converts about $23,500 of export revenue a year. Türkiye converts roughly $200,000. Israel roughly $420,000. Pakistan holds a half to a third of their engineers and extracts a ninth to an eighteenth of what each of those engineers extracts. The headcount deficit accounts for a fraction of the gap. The remainder is a fact about what the engineers are sitting inside.
The reflex is to read that as failure. The reflex is wrong, and getting it wrong is how most commentary on this subject discredits itself in front of the people who would have to act on it.
Pakistan’s defence establishment was never given an export mandate. It was given four, they were explicit, and they were set by a state that correctly believed it faced an existential problem.
The first was engineering capacity development: build, from essentially nothing, a cadre capable of designing weapons systems. The second was indigenisation: reduce platform dependence to the point where a supplier’s withdrawal is survivable. The third was sanctions-bypass: construct supply chains and design methods that assume no external input can be relied upon. The fourth was strategic independence sufficient for deterrence and defence.
All four were met, and this deserves stating without hedging, because it is among the more remarkable industrial achievements of the late twentieth century and Pakistanis routinely underweight it themselves.
Under embargo, without a patron, with GDP per capita in the low hundreds of dollars for most of the relevant period, Pakistan assembled a nuclear weapons enterprise, a solid-propellant ballistic family running from Hatf through Shaheen to Ababeel, an indigenous cruise line in Babur and Ra’ad, an operating fighter production line at Kamra and submarine construction at Karachi. Israel built its enterprise with French assistance at Dimona and American assistance at Soreq, and buys its Dolphin hulls from ThyssenKrupp to this day, on top of the favouritism and lobbying the global Zionist cult commands everywhere from Germany to the United States. Türkiye has no nuclear enterprise and KAAN has not entered service. Neither did what Pakistan did, and neither did it under the conditions Pakistan did it under.
$416M is therefore not evidence of incapacity. It is evidence that nobody asked. An institution optimised for autarky under blockade will not spontaneously develop an export function, because export requires exactly the outward-facing exposure that autarky was designed to eliminate.
The fourteen factories at Wah are the correct industrial shape for the mandate that was given. They are the wrong shape for the mandate now required, and the distance between those two sentences is the entire argument.
The gap May 2025 exposed
May 2025 revealed two separate failures of coupling, and only one has been addressed correctly so far.
The first was the gap between strategic capacity and operational-tactical capacity. Pakistan held a deep, survivable strategic arsenal and a comparatively thin conventional fires layer beneath it, which meant escalation options bunched at the top of the ladder rather than distributing across it. The Army’s answer was to raise the Army Rocket Force. That is the correct answer, arrived at quickly since it was within the immediate ambit of army, and it deserves to be recognised as such: it decouples conventional deep fires from the strategic tier and gives the theatre commander rungs he did not previously have.
The second failure has not been addressed, and it is the harder one, because it is not a weapons problem at all, and far less tractable than the decision cycle that produced the Army Rocket Force.
Pakistan cannot mass-produce. Not cannot design, which it demonstrably can; cannot produce at volume, at speed, from a standing start, in wartime. The demand signal from May 2025 and from Ukraine before it points in exactly one direction: thousands of attritable airframes, loitering munitions, first-person-view drones, cheap interceptors, low-cost cruise bodies. Quantities in the tens of thousands, at a rate of thousands per month, rather than in the tens. Pakistan’s dependence on Turkish supply for precisely this class of system is documented without needing to rely on anybody’s wartime rumour mill: SIPRI records that ten per cent of Türkiye’s total arms exports between 2020 and 2024 went to Pakistan, Asisguard Songar systems featured in the May exchanges, NASTP developed Yiha-3 with Baykar, and Kemankeş came from Baykar. Joint design has not produced joint volume. Design partnership without production capacity is a supply line without an industrial base – the distinction is very important.
Here is what makes this maddening. Pakistan is a country of tremendous production capacity. It is one of the world’s larger textile manufacturers, runs a substantial fertiliser and chemical industry, feeds a population of two hundred and forty million from its own agriculture, and operates automotive assembly, machine tooling, precision casting and electronics assembly at commercial scale. The constraint is not industrial competence. It is that none of that capacity has ever been connected to defence demand, because the defence sector was built to be self-contained by design, and self-containment was the correct design for the first mandate.
The American precedent is the one worth holding in mind, and it is worth stating precisely rather than gesturing at. The United States did not answer 1941 by inventing a new bomber. It answered by converting Detroit. Ford built B-24s at Willow Run at a rate the aircraft industry could not have approached; General Motors, Packard and Chrysler turned out engines, airframes and armour on lines that had been producing sedans eighteen months earlier. It was the institutional machinery, rather than the creative design, that let a car company become an aircraft company on a war footing and back again afterwards: standby tooling, surge contracts, cost-plus arrangements that made conversion survivable for the firm, and above all a state that trusted private industry enough to hand it the work.
Pakistan has none of that machinery. It has no surge doctrine, no standby tooling contracts, no dual-use line certification, no mobilisation register of qualified private manufacturers, and no procurement culture that would permit a textile group or an auto-parts group to be handed a munitions order and be believed capable of filling it.
And the defence sector will not build that machinery, because it is not a defence problem. It is an industrial mobilisation problem, and industrial mobilisation belongs to the state rather than to the services. POF cannot conscript Faisalabad. Neither can GHQ.
The tier nobody counts
The second thing May 2025 should have made unavoidable is that the private sector Pakistan needs for this already partly exists, and is essentially invisible to policy.
Not conglomerates. Small design houses, engineering firms, specialist manufacturers, most of them founded in the last decade, most of them at IDEAS and PIMEC every cycle, most of them absent from any serious discussion of national industrial capacity.
SoluNox in Karachi, registered in 2017, built the first genuine private-sector aerospace and maritime design house in the country, works on fourth-generation launching systems for high-mass high-speed target drones, developed the MANJNEEQ MR1-1 launcher jointly with PAC Kamra, has partnered with SIAT’s missiles division, and has been working a locally designed micro-turbine engine for unmanned systems along with hydrogen fuel-cell propulsion. Integrated Dynamics runs a ninety-thousand-square-foot facility in Karachi, exports unmanned systems and provides turnkey UAV commissioning abroad. CARE designed and delivered the Solid State Auto Loader (SSAL) for the Al-Khalid MBT, and carries substantial assignments still. East West Infiniti, SATUMA, Micro Electronics International sit alongside them. Newer names have surfaced in the last two years: Woot-Tech testing rocket-assisted take-off motors, and derivative corporate vehicles like Stingray Technologies and Qaswa Industries spun out of state enterprises. NICAT, the aerospace incubator launched in August 2023 through Ignite and a NETSOL-led consortium with NASTP, Air University and PAC, exists precisely to grow this layer.
Count the Tier 1 and Tier 2 private firms across a single IDEAS or PIMEC cycle and the number is somewhere near two dozen. Not three and a half thousand, which is Türkiye’s supply base, but not the handful that most analysis assumes either. It is a real tier, and it is the correct starting point for a mobilisation register.
The single most instructive item on that list is Alsons, an automotive-components group, now producing indigenous piston engines for unmanned aircraft. That is the Willow Run mechanism starting spontaneously, from the bottom, without policy, without a surge contract, without anybody in Rawalpindi having decided that auto-parts manufacturers are a strategic reserve. If it can happen by accident, it can be made to happen deliberately, and at a hundred times the scale.
The obstacle is not capability. It is the procurement relationship. The complaint one hears repeatedly from founders in this space is that state organisations approach private firms to extract transfer of technology rather than to buy product, and then ask why the organisation could not have done it in-house. That pressure runs downward onto the engineering base these firms depend on, much of it ex-service, which they are then unable to compensate properly. That instinct is not corruption or malice. It is the first mandate still running: an apparatus trained for thirty years to internalise every capability because externalising it was a vulnerability, applying that reflex to domestic firms who are not a vulnerability at all.
What the Air Force and the Navy already proved
There are two places where the second mandate has effectively been executed. Both are worth studying rather than praising, because they are working templates and neither was ever converted into policy.
The first and older is the JF-17. Co-developed with Chengdu from 1999, assembled at the Aircraft Manufacturing Factory at Kamra, and deepened progressively across blocks so that Pakistani content in airframe, avionics and weapons integration rose with each iteration rather than plateauing at licence assembly. The Air Weapons Complex built the munitions ecosystem around it, and NASTP now carries the design and integration work downstream of it. Critically, the JF-17 is the only Pakistani platform that has reached third-party customers at scale: Myanmar, Nigeria, and then Azerbaijan with Turkish help, the Azerbaijani contract for forty Block III airframes at $4.6B being the largest defence export agreement in Pakistan’s history. The traffic even runs both ways in the case of the latter, with all fifty-two MFI-395 Super Mushshak trainers delivered to the Turkish Air Force by December 2025. Whatever else is true of Pakistan’s industrial base, an operating fighter line that exports is a thing Türkiye does not yet have and Israel abandoned.
The second is more recent and goes further on the question that matters most. The Navy’s MILGEM programme with ASFAT, contracted in September 2018 at roughly $1.5B, built four Babur-class corvettes with two hulls laid down at Istanbul Naval Shipyard and two at KS&EW. That much is ordinary offset practice. What followed is not. ASFAT’s design office and the Navy’s Platform Design Wing jointly designed the Jinnah-class frigate, designated AS3400, with intellectual property rights shared equally between the two parties, and the construction contract for the lead ship was signed with KS&EW on 3 November 2025 during PIMEC. As ASFAT itself has stated, this was the first occasion on which Türkiye transferred the design know-how of a main combatant warship to any foreign country.
Read both carefully, because together they are the whole argument in miniature. In neither case did Pakistan buy platforms. It bought the capacity to build platforms, and in the naval case secured co-ownership of the resulting design outright. Six to eight Jinnah-class ships are planned. What made the absorption possible in each case was a cadre capable of receiving what was transferred: the PAF’s decades-old engineering establishment at Kamra in the first instance, the Navy’s decades-old vendor and supplier development track carried by ex-mariners and naval architects in the second. Transfer of technology to an organisation with nobody able to receive it produces a warehouse, not a capability.
There is a third case, and it is the one that settles the argument, because it is the largest industrial effort of the three and it went the other way. Heavy Industries Taxila built real capacity and a great deal of it: over a thousand Al-Khalid delivered, roughly 320 Type-59s rebuilt as Al-Zarrar through more than fifty modifications from January 2003, and alongside them Talha, Saad, Al-Hamza, the Viper IFV, an indigenous 155mm towed gun, an indigenous active protection system and a remote weapon station. In sheer physical output Taxila may well has produced more than Karachi Shipyard ever has. Nobody serious should call that nothing.
What Taxila never acquired was the design. Al-Khalid is NORINCO’s Type 90-IIM. Pakistan selected a powerpack across four competing prototypes, adapted the vehicle to its own terrain and heat, and localised the manufacture; it did not own what it was building. Set that beside the Jinnah-class, where the intellectual property is co-held outright, and the contrast is not one of effort or competence but of what was asked for at the table.
The powerpack carries the whole lesson in a single component. The Ukrainian KMDB 6TD-2 was never indigenised across three decades of production: three hundred and fifteen engines contracted from the Malyshev Plant in May 2002, a further $600M memorandum with Ukrspecexport in November 2016 covering two hundred more, and then Crimea in 2014 and the full war in 2022. The mobility of Pakistan’s armoured corps was made hostage to Kharkiv, and it is precisely the loitering-munition dependency of May 2025, arriving thirty years earlier, in the Army’s flagship programme, with nobody drawing the lesson forward.
Two further facts make this tale quite serious. HIT told the Senate in December 2017 that budgetary constraints had reduced Al-Khalid output to roughly eighteen tanks a year, which is the mass-production problem with a number attached to the Army’s own industrial crown jewel. And the export sales, 44 to Bangladesh and 22 to Sri Lanka, went out as MBT-2000, NORINCO’s export designation. China marketed Pakistan’s tank.
This whole trajectory is the plain verdict. Al-Khalid 2 was conceived in the mid-2000s and was still being discussed at IDEAS 2016; it has never been fielded. What arrived instead was Haider, 176 VT-4s bought from NORINCO for $859M around 2019, with licence co-production sought afterwards rather than secured first. Across thirty years the Army’s armour programme travelled from co-development to off-the-shelf purchase, while the Navy’s travelled from purchase to co-owned design. Two services, the same state, opposite directions on the only axis that compounds.
That is what a second-mandate transaction looks like. Not procurement, not licence assembly, but the acquisition of design authority with export-relevant intellectual property attached. And yet neither instance produced a repeatable mechanism. The JF-17 took a quarter of a century to yield one large export, and it did so through a bilateral relationship rather than through an institution built to generate such deals. The Jinnah-class design authority has not been pointed at a third-party customer at all, though the Gulf and East African navies buying inexpensive patrol and escort tonnage are sitting in plain sight. And Taxila, which out-built both, ended the period owning less of what it makes than when it started. Two proofs of concept across five decades, both achieved on the initiative of small professional cadres, and no policy anywhere in the state that requires a third.
The nemesis of creativity, correctly stated
Toynbee’s term for the characteristic failure of successful civilisations was the nemesis of creativity, and it is usually quoted badly. The failure he described is not incompetence. It is idolisation of an ephemeral technique: a civilisation meets a mortal challenge, devises a response that works completely, and thereafter treats the specific institutional form of that response as permanent rather than as one answer to one question. The creative minority does not become stupid. It becomes custodial, administering with great skill the arrangement that solved the last problem.
Spengler drew the line more sharply. A culture creates forms; a civilisation administers the forms a culture created. The transition is not marked by collapse but by competence. Everything continues to function. Budgets disburse, inspections complete, factories man, tenders let, delegations arrive and are received. What ceases is the generation of anything new, and the ceasing is invisible from inside, because the metrics were never designed to measure novelty.
The United States is the demonstration case, and it is more instructive than any comparison with Türkiye or Israel, because America did nothing obviously wrong and it happened anyway. Between the early 1990s and the early 2000s the number of aerospace and defence prime contractors fell from fifty-one to five. Tactical missile suppliers went from thirteen to three, fixed-wing aircraft from eight to three, satellites from eight to four, surface ships from eight to two, expendable launch vehicles from six to two. 90% of American missiles now come from three sources. Missile primes fell from thirty to seven, and in the hypersonic weapons sector the Pentagon’s own 2022 competition review counted exactly one. Every individual merger was defensible on its own terms; budgets were falling, capacity had to be preserved, and the state actively encouraged the consolidation. The aggregate was an industrial system optimised to administer programmes of record with great precision and progressively less able to originate anything. No amount of Anduril or Palantir, with their dozen or so neo-prime contenders, is currently projected to plug that gap. As one senior official put it, the Pentagon went from an industry it judged too large to sustain to one now too small to surge.
Consider what that culture had been. The firms and design bureaux that produced the SR-71 in the 1960s and the B-2 in the 1980s, that commercialised the ring laser gyro and then the MEMS inertial sensor on which every subsequent guided weapon in the world depends, are the direct institutional ancestors of the five that remain. And they are now demonstrably behind on the things that matter most. China has flown two sixth-generation prototypes in public since December 2024; the F-47 was awarded in March 2025 and is not expected to fly before 2028. Iran, working with a fraction of the budget, set the global price point for one-way attack munitions with the Shahed, and Washington’s response was Replicator, an initiative explicitly created to close a gap that should never have opened. Russia and China fielded hypersonic glide vehicles first. None of this reflects a loss of engineering talent, of which America has more than anyone. It reflects an industrial architecture built to sustain the response to one challenge, never re-tasked when the challenge changed, and staffed throughout by people executing their mandates faithfully.
Applied here the point is precise and it is not an accusation. The mandate of 1974 was set by the state, not chosen by the institutions. The institutions executed it completely, over five decades. What has not happened is the second act: the state has not returned to reset the mandate now that the first is discharged.
The evidence sits in the shape of the estate. The National Defence Complex is a genuine design house reporting through the Strategic Plans Division to a single captive customer, which is exactly correct for a deterrence mandate and structurally incapable of a subsystem export business. GIDS is a marketing conglomerate rather than a manufacturer, which is what an export arm looks like when export is bolted onto production as an afterthought rather than designed in. IDEAS 2024 drew over five hundred and fifty exhibitors and produced eighty-two MoUs, of which very few became hard contracts, because a memorandum is what you generate when you have product but no institution to convert intent into delivery.
And the services diverged, which is what happens when re-tasking is left to individual initiative rather than directed.
The Air Force re-tasked itself. CENTAIC in 2020, NASTP inaugurated at Nur Khan in August 2023, campuses at Rawalpindi, Lahore, Kharian and Karachi, the Silicon site carrying design work in simulation, space, wireless, cyber and AI, with NICAT incubating beneath it. Turkish Aerospace Industries and Baykar both rent space inside it.
The Navy chose a direction and, in the Jinnah-class, proved it works. The risk is that it remains personality-dependent and generationally shallow, and that the pool dissipates within a decade because nothing institutional synthesises it.
The Army consolidated paperwork. Taxila kept building, and kept building somebody else’s designs. In May 2020 the Ministry of Defence Production merged the Military Vehicles Research and Development Establishment (MVRDE) of 1972, the Armament Research and Development Establishment (ARDE) of 1974 and the Institute of Optronics of 1984 into a single directorate. Three establishments whose youngest component predates the fall of the Berlin Wall, given a new letterhead. The Army’s own electronics estate consists of the Inspectorate of Electronics and Instruments at Chaklala and 509 EME Workshop at Gujranwala: conformance and sustainment, with no design mandate anywhere in the chain. The institution the Army did inaugurate at Chaklala, in 2018, was the Army Institute of Military History. The Rocket Force is the exception and it is a real one, but it is a force structure decision rather than an industrial one, and it will need somebody to build the magazines it is going to empty.
What the second mandate would have to contain
Convert a deterrence-optimised industrial base into an export-capable and surge-capable one without degrading the deterrent, fast enough to enter the Makkah architecture as an industrial partner rather than a manpower supplier. That is the whole of it, and every component has been done by somebody else already.
Separate design from production in the corporate form, because seventeen thousand engineers currently sit buried inside a hundred-thousand-person estate whose dominant activity is metal and ordnance. Give the design entities a profit and loss account and a mandate to sell subsystems rather than only platforms; Roketsan’s business is seekers and motors sold into other people’s integrations, and that is where the volume lives. Generalise the Jinnah-class transaction: acquire design authority with shared intellectual property, not licences.
Build the mobilisation architecture that does not exist. A register of qualified private manufacturers with dual-use lines. Standby tooling contracts that pay a firm to hold conversion capacity it is not currently using. Surge clauses in peacetime procurement. Certification pathways that let a fertiliser plant, a textile group or an auto-parts manufacturer be a munitions supplier inside ninety days rather than never. Alsons is already halfway there without being asked.
And treat the two dozen private design houses as a strategic asset rather than as vendors of last resort. Buy their product instead of extracting their technology. That single change of posture is worth more than any subsidy.
One caution against reading Ankara as a finished model to be copied. Türkiye is now losing the advantage Pakistan still holds. Real wages for senior Turkish defence engineers fell fifteen to twenty per cent through 2024, time-to-fill for senior roles stretched from fifty-four days to eighty-nine, and a 2024 survey put forty-two per cent of ASELSAN engineers actively seeking other work. Cheap, deep engineering labour is a wasting asset. Pakistan has held it for fifty years and converted almost none of it. The window is narrower now than when Ankara started.
The thing that has to be said plainly
The establishment did what it was asked. Every task set in the 1970s was completed, under conditions no comparator faced, and the deterrent that resulted is why Pakistan sat at the table in Makkah rather than appearing as an item on somebody else’s agenda. Any argument that opens by treating that record as failure will be dismissed by the people who would have to act on it, and will deserve to be.
But a mandate discharged is not a mandate renewed. 17,000 engineers sit inside an architecture designed to survive a blockade and never redesigned for a market, converting $23,500 apiece where Ankara converts two hundred thousand and Tel Aviv four hundred. A hundred million people’s worth of industrial capacity sits outside the defence perimeter with no mechanism to bring it in when the shooting starts. Two dozen private design houses exist and are treated as a source of technology to be extracted rather than an industrial base to be capitalised. And a trilateral security order has just been signed in which the other two parties bring capital and industry respectively.
The question is no longer whether Pakistan can defend itself. That was answered decisively, and the answer is the reason the agreement exists. The question is whether it enters this order as the industrial leg or the manpower leg, and stays there for a generation.
The National Command Authority set the first mandate. Somebody has to set the second, and Makkah is the last convenient moment to do it.