The 1947-48 Kashmir war. Forgotten lessons
The 1947-48 India-Pak war, or the first Kashmir war, was the longest India-Pakistan war. It is also the least studied. Having read several accounts from both sides, I believe our (India’s) view of the war emphasises some remarkable achievements that prevented the loss of Kashmir and key towns in it e.g The airlift that saved Srinagar (and Kashmir), along with the breaking of the siege of Poonch and Leh.
While it is right to celebrate our achievements and the bravery of the soldiers
of both the India army and state forces, we also need to understand where things
went wrong – I believe they did, at the level of political objectives and
planning. Those lessons may be more valuable for us.
As someone with an amateur interest in geopolitics and military history, I believe there were larger lessons of this conflict and the mistakes made – some of which were repeated in India’s later wars.
Before
commenting on mistakes and lessons to be learnt, a summary of what went right
is important, because it would indicate that the margin of victory was slender
– indeed, the best book on military operations in the Kashmir, by Lt Gen (then
Brig.) `Bogey’ Sen, war is titled
`Slender was the thread’.
I believe
the following were our best achievements of the war:
1.The airlift, put together at very short notice, despite a lot of constraints,
that airlifted the men that saved Srinagar and then threw the raiders
back.
2. Preventing the capture of Poonch and sustaining it by air, until the siege
was broken
(Operation Easy).
3. Breaking the siege of Leh and recapturing Zojila pass and Kargil, reconnecting it with
Srinagar.
These were
won with strong leadership at lower levels (battalion and below) with bravery
displayed by individual soldiers and pilots. Some of the best known Indian
formations involved in the war:
1 Sikh, 4 Kumaon, 1 Kumaon (Para), 1,2,3 & 4 Rajput, 6 Raj Rif,
Gurkha Rifles battalions – 1/5 GR, 2/4 GR & 3/9 GR along with the State
forces (8 JAK rifles) also fought in WW2, so they had an experienced cadre of
JCOs and junior officers. In contrast, the senior army leadership were either
British, with a limited, or biased interest in the war, or less experienced
Indian officers.
The mistakes we made were at the higher levels and less spoken of.
Contingency planning: Planning is something all armies do. The German army probably has a contingency plan to invade Austria. After WW2, the Indian army had an experienced General staff, used to planning large operations and contingency planning. At the time of independence, it should have been clear that the biggest threat to the country would be in Kashmir. On 15th Aug 1947, the Maharaja of Kashmir was undecided about acceding to either India or Pakistan. The leadership of the main political party, the National conference was in jail. The NC led by Sheikh Abdullah favoured India. The other political party, the Muslim conference, wanted Pakistan. There were serious communal riots in the State.
In this context, anything that the Maharaja of J&K decided would probably
have led to the Indian army being required to intervene. Consider these
scenarios:
2.The Maharaja does not decide. Pakistan imposes a blockade (as it happened). There is a civil war between the cadres of the National conference and Muslim conference and a communal war. The Indian army has to intervene to protect minorities. Pakistan does the same. The State has a de-facto partitioning between India and Pak. The area of influence depends on what territory the Indian army can physically control.
3.The Maharaja chooses India. Pakistan intervenes. This is what happened.
In all cases, newly independent India should have planned for the possibility of intervening in Kashmir. Even if the Maharaja had not decided, there could have been staff consultations between the officers of the Kashmir and the Indian armies. This was not done.
Pak planning: Pakistani planning for intervention in Kashmir began before 20th Aug 1947. We know this because the plan – Operation Gulmarg, accidently fell into the hands of Major Kalkat, the brigade Major of a Pakistani brigade, on 20th Aug – the communication was intended for the British commanding officer. Major Kalkat, who was supposed to join the Indian army, was kept under virtual house arrest, but managed to escape and made it to India on 18th Oct. His account was not believed by Brig Kulwant Singh (who was to command a division in Kashmir later) and the DGMO Col. Thapar. The same person who as Army chief in 1962 did not believe that the Chinese were about to attack.
The Pak plan was simple. One group of tribal militia would head for Srinagar, on trucks, via the road
from Muzaffarabad. All that stood in their way was a battalion of 8 JAK - half the soldiers, who were Muslim would defect, leaving just two companies of Dogra soldiers to face 6000 fighters. In theory this force could be in Srinagar within a day.
The other force, of about 10,000 men, would attack and pin down the remaining formations of the Kashmir state forces, in the Bhimber, Poonch, Rajouri, Mirpur sectors.
There was no similar contingency planning from our side.
There was also local intelligence available with J&K state forces about the
movement of tribal militia into Kashmir and raids on border villages, but this
information did not reach the Indian army, as we did not even have the
equivalent of a military attache in the state, let alone staff consultations
between the Indian army and Kashmir.
The Maharaja would probably understand that in the Jammu region, it would be the Indian army and
not his army, who would be better placed to protect his Hindu and Sikh subjects and would have agreed to consultations between the two armies.
Why we should have acted earlier.
On 9 Sept, Pakistan started an economic blockade of J&K, to pressurise the Maharaja to accede to Pakistan. The Maharaja asked India for help in sending essential commodities, without any discussion on how it was going to happen. Instead of a lengthy exchange of letters (the telegraph lines went through Pakistan and were cut off), India could have sent a team to Kashmir to take decisions on the spot, including a bridge connecting India with Kashmir and improving the airstrip.
The communal situation in Jammu had gone out of control by Sept. The J&K state forces were seen as both facilitating attacks on Muslims and unable to protect people. There was a rebellion against the Maharaja in districts bordering Pakistan. There was every chance of a humanitarian catastrophe, similar to what Punjab was going through. In that context, India should have liaised with the Maharaja to discuss stationing of troops in Jammu for refugee protection, or near the Punjab-Jammu border with pre-positioned supplies.
The administration of Kashmir had broken down. Sheikh Abdulla and the key members of the national conference - the only organization with widespread grassroot support, were jailed. Pro Pak Prime Minister Kak was dismissed on 11 Aug. His successor Mehr Chand Mahajan was appointed only in Sept, but took over on 15 Oct (replacing Maj. Janak Singh as the interim PM). The Kashmir army chief, Maj. Gen Scott resigned in Sept and was replaced by Rajinder Singh Jamwal on Sept 25.
This team, which was now pro India, should have been in close but informal contact with the Indian govt, at least on matters like transport links and maintaining law and order. The new Kashmir PM Mehr Chand Mahajan, knew better than most, the importance of the road link from India (Gurdaspur) to Kashmir, because he was part of the Radcliffe commission which demarcated the Indo-Pak border.
A bridge and an airfield The main factor determining the pace of Indian intervention in Kashmir was geography. The road links connecting Kashmir with British India ran through what became Pakistan on 15th August 1947. Fortunately for India, the boundary award gave Gurdaspur to India, which provided India’s only link to Jammu & Kashmir via Pathankot. Even so, it would be clear even to an amateur strategist that Pakistan would be able to send forces into Kashmir much faster than India, due to the location of both its cantonments and the road network.
To make matters worse, the road route from India (Gurdaspur to Jammu), went
across the river Ravi, over which there was no bridge capable of carrying
vehicles. While Indian army engineers (15 Field company) managed to construct a
pontoon bridge between Oct 27 and Nov 2, to transport troops and armoured cars
to Kashmir, the bridge would have been completed earlier, if it was started
earlier and not waited till the Maharaja signed the instrument of accession.
Even if not for possible military movements, the bridge would have permitted
refugees to move from Kashmir to India. Contingency planning would have
identified the lack of a bridge to be a bottleneck, preventing rapid deployment
of Indian forces to Kashmir.
When the war started, we had our most mobile infantry formation - 43rd Lorried infantry brigade
in East Punjab, looking after refugee protection. Although its had given up some of it trucks and two of its three Gurkha battalions had opted to be part of the British army, it had 2/8 GR available with enough trucks to move them and their supplies. If we had a bridge across the Ravi capable of carrying 3 Ton trucks and prioritised fighting the battle over refugee protection, 2/8 GR could have made a big difference to the battle of Shalateng - along with the rest of 4 Kumaon and 1 Punjab who, in my view, we had the means to transport by air, before 7 Nov 1947.
2/8 GR eventually was inducted in in the Ladakh sector.
Maj. Gen Akbar Khan who planned Operation Gulmarg was of the view that we could have taken Muzaffarabad in Nov.
There was a similar problem in the Jammu region, where two under strength battalions of 50 Para brigade could not achieve their objectives, because they were deployed piecemeal and not fast enough, compared to what might have been achieved, if the brigade was deployed in advance, in the Pathankot region and there was a bridge across the Ravi, between Pathankot and Jammu.
I believe we also made a major mistake in not attacking bridges that Pakistani forces were dependent on for their supplies – we had complete air superiority to do so.
Pakistan’s most critical bridge was the Kohala bridge over the Jhelum, on Pakistan’s MSR (main supply route – Rawalpindi-Murree-Muzaffarabad-Baramula / Uri. The IAF was not permitted to attack bridges by Air Marshal Elmhirst – the view of senior British officers on both sides, was not to escalate the fighting, but restrict the IAF (RIAF as it was called then) to transport an close air support (which also operated with a lot of restrictions). IAF pilots who `accidently' dropped bombs near the bridge were prevented from doing it again after Pakistan protested !
Nehru’s concern was that we might be blamed for a humanitarian crisis, if
refugees could not cross the bridge. The IAF may well have been unable to
destroy the bridges, as we had no specialised bombers, but that is moot. We did
not attempt it. Wars are won by fighting as unfairly as possible and since we
had air superiority, it should have been fully exploited.
It was a similar situation in 1962, when the political leadership did not want
to use the air force in an offensive role. It is debatable if that would have made
any difference to the outcome, but again, it was not attempted.
In the absence of a road link with India (which did not exist till the boundary
award on 15th Aug) the only link that Kashmir had with newly
independent India was by air, with the only rough airfield near Srinagar. While
a lot is made of our ability to improvise and conduct an airlift at Srinagar,
from 27th Oct, our task would have been made easier if we had a
survey of the airfield before the war – which the Maharaja would probably have
been happy to help with, since that might be the only escape route from
Srinagar for himself and other Hindu and Sikh families – thousands were indeed
evacuated by air.
The brigade defending Srinagar had only one map of the State in its HQ. During the battle of Badgam the battalion commander was given the relevant square, but out of the original map, for his reference.
The airlifts: The Indian air force has rightly been commended for operating two airlifts – Srinagar and later Poonch, under very adverse circumstances, as well as feats like landing in Leh - with senior pilots invariably leading form the front. Some of the problems included the availability of only 1 transport squadron in India - but with the spares in Pakistan. No proper airstrips, inadequate airstrip length, aircraft operating close to their maximum altitude and rudimentary air traffic control at both airfields – Poonch had added complications of a difficult approach before landing and the airstrip itself being within artillery range of the enemy and frequently shelled.
When the
Srinagar airlift started on 27th Oct, the IAFs no 12 squadron
(Dakotas) had only 10 aircraft, of which no more than 7 were operational*. 6
civilian aircraft were also used and together they flew 28 sorties on day 1 of
the airlift. While the highest number of transport sorties per day was 35
during the Srinagar airlift, we know that a max of 704 sorties were
flown in 22 days, for an average of 32 a day.
If 704 includes fighter
sorties from the two aircraft stationed at Srinagar, the average drops.
At 32 sorties per day and a minimum of 2 sorties per aircraft per day (some did
3), a maximum of 16 military and civil aircraft were used.
* No 12 squadron strength was increased to 16, from 1948, after the transfer of 3 civilian aircraft
and other Dakotas in the air force inventory.
While the use of civilian aircraft and the dedication of their pilots was commended, at the time of Independence, India’s various airlines had a total of 74 Dakota aircraft (115 in all). Only 33 of these aircraft were used in the war and most likely no more than 10 on any day. Fighter and transport sorties together averaged just 420 per month (i.e just 7 aircraft flying 2 sorties per day)
It is true that partition had disrupted civil aviation and availability of
aircraft and crew was probably not more than 50% That said, if the airlift was
implemented as part of a well thought out plan and aircraft requisitioned on a
truly war footing, rather than ad-hoc planning after the Pakistan
invasion had started. We might have been able to requisition more civilian
aircraft from the outset, particularly in the critical period from 27 Oct-5
Nov 1947, when we used no more than 10 (probably just 6) civil aircraft
out of 34 possibly available (50% of the total strength of 74 civil Dakotas).
We lost just 2 Dakotas in 15 months (one crash and one hit by artillery fire on Poonch airstrip), so my contention is they were not pushed to the limit.
As an example, of flights that might have been diverted to the war effort, Deccan
Airways, (owned by Hyderabad state and
the Tatas), operated 4 refugee flights per day till Feb 1948.
Air India had a daily Bombay-Delhi flight throughout Nov 1947 - it's Vickers Viking aircraft had 35 seats compared to 15 for the Dakota.
Under Operation India, we used 21 aircraft, mostly from British airlines to fly 35,000 refugees between India and Pak in Oct and Nov 1947. The services of some of these aircraft and pilots could have been used in the Srinagar airlift, once the airstrip was safe. This did not require British approval, as it would have been an agreement with a private company. The pilots and these aircraft that were used in WW2, were not in demand after the war and it is possible a couple of additional aircraft could have been requisitioned. These could have been for flying in medical and civil reconstruction teams (army doctors and engineers) and carrying out refugees.
All aircraft flying into Srinagar and Poonch carried out refugees, so we would not have carried out
fewer refugees, if `refugee only' flights were diverted for a war effort.
If we had 3
more civil aircraft flying 2 sorties per day into Srinagar for those 10 days,
they could have flown in an additional 900 men. It would have meant 4 Kumaon
would be at full strength, instead of only 2 companies. That may have meant that Maj Somnath Sharma’s (PVC) company of 4
Kumaon would not have been isolated and an attack on it would have been
repulsed with a more adverse casualty ratio for the attackers. It would also
have meant that 1 Sikh would be at full strength (1 company was left behind in
Gurgaon) as well as 1 Punjab (which was understrength). The presence of 4 extra
companies and a squadron of armoured
cars, instead of a troop, may have meant that the enemy, tactically encircled
at the battle of Shalateng, could have been annihilated instead retreating -
albeit in a disorganised manner.
Another way to look at the battle of Shalateng, is that we were lucky. Pakistan had a squadron of armoured cars (11 Cav) at Rawalpindi, a day from Srinagar by road. On 30th Oct, it was proposed to use them, with Pak army men willing to `unofficially’ operate them. This idea was vetoed by both the army leadership (which was British) and the political leadership, who did not want a full fledged war with India. If that armoured squadron was deployed and there was better leadership of the tribal forces at Shalateng (the few Pakistani army and defected state forces officers being unable to influence them) they might have broken through to Srinagar.
In the Poonch airlift – which similarly saved the city, the airstrip was improved and lengthened by civilian labour. This was not done in Srinagar. The Dakota was capable of operating at night, but this was not done in Srinagar – IAF specialists landing in the initial sorties could have facilitated night operations.
Interestingly, the Pakistani account of the war suggested we used a lot more Dakotas than we did, because it was logical for them to make that assumption.
In the siege of Demyansk, on the Eastern front in 1942, the Luftwaffe flew in far worse conditions than Srinagar (heavy snow, low visibility, extreme cold, rough airstrips and attacks from Soviet aircraft), with an average of 150 sorties a day on a Srinagar sized airstrip (and one smaller one) supplying 90,000 soldiers.
A siege India failed to lift was at Skardu, which was gallantly defended, by Col Thapa of 6 JAK rifles, with just 130 men. While its distance from the front prevented reinforcements reaching Skardu, there was no such excuse for the fall of Mirpur. The Mirpur garrison of 600 men (JAK state forces) pulled out without a fight on 26th Nov (while 130 men at Skardu resisted for 6 months). On 19th Nov, India's 50 Para brigade, deployed for the relief of besieged towns in Jammu, was a day away from Mirpur. However, 50 Para was ordered to capture Kotli, instead of Mirpur, though the brigade commander Brig. Paranjape felt Mirpur was a more feasible objective.
India's 50 Para brigade, started its movement from Akhnoor only on 16th Nov. They would have done it faster, if they were deployed close to Jammu earlier (if contingency planning and field intel anticipated communal trouble in Jammu) and if a bridge across the Ravi was available earlier. 50 Para comprised only two under strength battalions of 1 Punjab and 3 Rajput (combined strength of around 750 men) and did not have the ability to take both Kotli and Mirpur in the time available. Kotli was relieved on 26th Nov, the day Mirpur fell. The other units of 50 Para - 1 Kumaon and 7 Cavalry were diverted for the
defence of Srinagar.
Fighter aircraft under utilized.
Of the average of 420 sorties per month flown by the IAF during the war, 107 (1500 in 14 months) were flown by fighters. We had 6 fighter squadrons but mostly used two - Nos 1 & 7 squadrons. Later, missions were flown by no 8 & 10 squadrons. While one single transport squadron (and a squadron equivalent of civilian aircraft) flew 310 sorties a month, six fighter squadrons between them, with the same level of aircraft availability, flew 107 sorties a month. These 107 sorties a month included reconnaissance flights.
We had bases available in Jammu, Ambala, Pathankot and Srinagar. It appears fighters were under utilized, particularly in the early stages of the war, against tribal militia, when more sustained strikes against their vehicles and troop concentrations would have made a difference - all accounts suggest that even attacks by a couple of fighters made an impact. Our most significant victory over regular Pakistani forces - the battles of Zojila - Dras in Oct-Nov 1948, the IAF had a higher concentration of ground attack (Tempest) aircraft, showing that it was possible to have multiple strike package of 4 aircraft each, rather than single missions with one pair of aircraft.
These are accounts from two of the fighter pilots.
https://www.bharat-rakshak.com/iaf/history/1948war/8sqn-blake/
https://www.bharat-rakshak.com/iaf/history/1948war/donmich/
Missed
opportunities: Every
published account of the fighting suggests that Pakistan lost an opportunity to
capture Srinagar when its tribal Lashkars stopped at Baramulla to pillage the
town. On our side, we lost an opportunity to capture Muzaffarabad (or the town
of Domel before it) in the time between the recapture if Baramulla and Uri and
before winter closed the roads and the Pak army then got fully involved in the
fighting – Pakistan’s 101 brigade was officially deployed in Muzaffarabad area
in May 1948 (unofficially before that).
After the recapture of Uri on 13 Nov, the brigade commander’s view was that his (161) brigade should have continued its advance towards Domel and Muzaffarabad. Instead, a battalion (1 Kumaon) was diverted to break the siege of Poonch. The effort failed – inadequate recon meant a bridge was destroyed by J&K state forces, mistaking the Kumaon battalion for Pakistanis, after which 1 Kumaon was ambushed. The battalion finally entered the town, but had to remain there and to defend it from the besiegers. While the political leadership felt the relief of Poonch took precedence over trying to recapture Muzaffarabad, the commanders on the ground felt there was an opportunity to recapture Domel and Muzaffarabad. Sardar Patel agreed with that view, but the majority view of the army leadership and Nehra prevailed, rather than that of the commander on the ground.
Brig Sen (later Lt Gen), commanding 161 brigade felt that the capture of Domel
would have reduced pressure on Poonch anyway. Sardar Patel’s view was that the
garrison at Poonch could hold out unassisted for a few weeks more – Pakistan did
not attempt any offensive until Dec (capturing Jhangar).
In any case, the southern arm of the offensive to break the siege on Poonch did
not make progress, so
the weaker Northern arm – with only 1 Kumaon detached from 161 brigade would
not have succeeded.
At the time of the ceasefire on 1 Jan 1949, Indian positions facing
Domel /Muzaffarabad were where they had been on 13 Nov 1947.
Once the Pakistan
army was fully involved, the ground forces of each side in Kashmir were roughly
of equal strength, making further capture of territory difficult. India deployed 50 infantry battalions (incl. Indian state forces) and 14
battalions of the J&K and East Punjab militia.
Pakistan deployed 23 infantry battalions and 40 battalions of
Azad Kashmir forces – overall 64 Indian vs 63 Pakistani battalions.
The Butcher papers (Gen Sir Roy Bucher’s letters to Nehru) point out that by
Nov 1948, the army was tired and suffering from a logistics strain, making further
capture of territory infeasible.
Towards the end of the war, India had two major successes – the breaking of the siege around Poonch (operation Easy) and the breaking of the siege around Leh, which also opened the Srinagar -Leh road, recapturing Zoji la pass (with the innovative use of tanks) and Kargil. While both these operations have been publicised, with innovative use of tanks and good close air support, we lost Haji Pir pass in Pakistan’s last major attack of the war, after capturing it earlier.
Officially, part of the reason was the vacating of two posts (Led Gali and Pir
Kanthi – taken at some cost in summer) for the winter – shades of Kargil in
1999. Lt Gen. Sen (then Brig and CO of 161 Brigade) in chapter 21 of his book
`advantage thrown away’, covers it in more detail.
We could not attempt to recapture it, as a ceasefire was declared on 1st
Jan, without letting the army plan for a better ground situation before the
ceasefire, including the recapture of Haji Pir
In the 1965
war, we recaptured Haji Pir and gave it up in the Tashkent agreement
Publicity: Nehru himself wrote (in his papers in Dec 1947) that the Pakistani newspapers treat the war as a war of national liberation, with highly embellished stories of victories and valour. He note the effort of Dawn correspondent Desmond Young. In contrast, Nehru stated:
`Our publicity is feeble in the extreme. There is very little about Kashmir in the newspapers or radio and what is there makes for dull reading. I am told that many acts of great personal bravery have taken place. In the recent Uri encounter once of the VCs of the Indian army was killed. According to reports, he fought with great gallantry. Something should have been said about it, but not even a mention of his death has been made. There are numerous instances of this kind'.
He goes on: `Our whole outlook has been defensive and apologetic, as if we are ashamed of what we are doing..... Similarly `wounded soldiers from the Kashmir front who have been sent to Delhi feel they are ignored...'
What Pakistan learnt: Operation Gulmarg of 1947, became operation Gibraltar in 1965, where militants with the aid of the Pak army, tried to seize Kashmir and was conceptually repeated in Kargil in 99. Like in 1947, we were taken by surprise.
The Indian army did not use Subhash Bose's INA veterans, as they were not re-absorbed into the Indian army. Pakistan used them in the training and leading of units of the Azad Kashmir militia.
Maj Gen. Akbar Khan who planned operation Gulmarg, was frustrated at the lack of support from the political leadership - in not directly inducting the Pak army into the fighting in Nov 47. He attempted coup and was jailed. That began Pakistan's history of generals believing they knew best how to run the country.
Trivia:
No Pakistani
officer above the rank of Capt. died in the war. whereas the Indian army's very first casualty was the CO of 1 Sikh, Lt Col. Ranjit Rai (previous casualties were J&K State forces).
Pilot officer GB Cabral, crashed his Tempest aircraft twice. Surviving the first accidental crash, he
was shot down the second time, walked across enemy lines and continued in active duty.
The Pathan tribal militia that invaded Kashmir, included an American - Russell Haight.
Online reads:
Maj Humayun Amin's detailed analysis of the war from the Pak side.
The war of lost opportunities - Maj. Amin
Paper from CLAWS analysing operations during the war.
CLAWS paper on operations during the war
Official history of the war. MOD.
https://ia601508.us.archive.org/27/items/in.ernet.dli.2015.116302/2015.116302.Operations-In-Jammu-Amp-Kashmir-1947-48_text.pdf
The best book on the war, in my opinion is Lt Gen. LP Sen's `slender was the thread'. I reviewed it in my blog post on the `best of Indian military writing'.
https://rpdeans.blogspot.com/2025/07/the-best-of-indian-military-writing.html
The Srinagar airlift
The airlift that saved Kashmir
Notes from Nehru's archive:
https://nehruarchive.in/documents/a-note-on-kashmir-19-december-1947-7y3nx4
No 12 squadron's airlift to save Poonch.
No 12 sqdn in Poonch
Comments
Post a Comment