A resounding victory won at small cost solves many military problems. The war against the Mahrattas was going astonishingly well in Wellesley’s immediate theatre of operations. After the battle on 29 November 1803 both British armies were in fine physical and mental condition. The enemy was still numerous, but Mahratta morale was low.
Elsewhere in India there was more good news. On 1 November General Lake won decisively at Laswaree in a battle comparable to Assaye in viciousness and in what it achieved. Perron was utterly defeated. As we have seen, Scindia had lost in Guzerat also; Baroch had fallen. Colonel John Murray of the King’s 84th Foot had been engaged in a local conflict on behalf of the Gaikwar against a rival for the throne. Murray did better than expected and was able to consider an advance on Scindia’s capital at Ougein. Lieutenant-Colonel G. W. R. Harcourt of the King’s 12th Foot had conquered Berar’s entire province of Cuttack with a force of no more than 3,000. This campaign was complicated by weather and terrain, but it did not involve a great deal of fighting. For the first time British controlled territory extended along the coast from Madras to Calcutta. The Mahratta Confederacy no longer touched on the Bay of Bengal.
The two British armies that had won at Argaum moved by easy stages carrying their wounded towards Ellichpoor. On 3 December Wellesley camped only fifteen miles south of Gawilghur and reported to Stuart in Madras that from the plain ‘It does not appear to be as strong as many hill forts in Mysore taken by our troops.’ Even nowadays the resemblance to Nundydroog is remarkable when Gawilghur is observed from the south, though Gawilghur is much larger. The fortress town was thought to contain Berar’s treasure and some of his family. He also used it as a kind of fortified hot-weather retreat. But British armies in India usually had less trouble taking hill forts of all types than one would expect from looking at the places.
A hospital was established at Ellichpoor. Wellesley then apparently approached to within two miles of Gawilghur from the south south-east during a personal reconnaissance. The principal problem in attacking the fortress was one of getting close to it.
The approaches from the south consisted of two ‘roads’ leading to the fortress from the valley below. The easterly approach was so difficult that it would not even accommodate bullocks. It is still in use, but one must climb, not walk. The westerly road was narrow and steep, but moderately loaded carriage bullocks could go up. It was scarped on both sides at the top, however, and had the final disadvantage of passing for half a mile within point-blank range of the guns on the west wall of the Inner Fort.
If the fortress-town had been as inaccessible on all sides as it was from the south, the place might have been impregnable. Unfortunately for Berar’s garrison, the two rocky hills on which the fortress had been built were connected on the north by a narrow tongue of land to a whole low range of flattened mountains of similar elevation (about 3,600 feet) which extended east and west for many miles. Since the place could not be taken from the south, it had to be besieged from the north. At Ellichpoor, which was part of Hyderabad and in possession of the Nizam’s killadar, Wellesley learnt that the main approach to Gawilghur was from the north and lay along the narrow tongue of land.
The information available at Ellichpoor was meagre perhaps because Gawilghur, though only thirteen miles away as the crow flies, was in another country and much further by the only practical route which led through hilly jungle – a glorified path not wide enough for any wheeled vehicle. On the other hand, the jungle was not as impenetrable as that in India ‘below the ghauts’. Madras pioneers with strong working parties from the infantry should be able to help the bullocks and elephants pull and push the artillery up the hills and then west along the more or less flat crests. The total distance was about twenty miles.
Wellesley began his operations from Ellichpoor on 6 December 1803. He sent Chalmers and the 1/2 Madras from his own army to clear Deogaum and the valley four miles south of Gawilghur. From Stevenson’s army he sent Captain Alexander Maitland with the 1/6 Madras and two companies of the King’s 94th ‘to seize the fortified village of Damergaum which covers the entrance to the mountains’. Both these detachments succeeded in their missions, although Mahratta strength in the area was considerable. By this time Gawilghur was known to contain not only its normal garrison, but most of the survivors of Manoo Bappoo’s regular infantry. Enemy patrols were active in the foothills, so ample guards would have to be left at Ellichpoor for the hospital.
The two British armies moved out of Ellichpoor at sunrise on the 7th. Wellesley advanced only as far as the village of Deogaum, nine miles from his starting place and in a direct line between Ellichpoor and Gawilghur. A standard camp was established near the village. Stevenson’s army, temporarily reinforced by two of Wellesley’s iron 12-pounders and artillery and engineer personnel, had a much more difficult assignment. The troops began to climb the Gawilghur hills at Damergaum and continued into rugged country. They had to cut out trees and build roads with earth and rock, at one point filling in a chasm to save miles of additional road. After four days of exhausting work Stevenson’s army complete with its battering artillery and ammunition reached the village of Lobada on the ridge level with Gawilghur.
From this side the fortress was not so awe-inspiring. Although it was built on the summits of two hills with deep and precipitous slopes almost all round, a corridor about 400 yards broad led from the hills to the northern wall. The tongue of land was not open to the wall; two-thirds of it was protected by a tank or artificial lake nearly full of water. There remained, however, a ribbon of meadow about 120 yards wide which led up to the double northern wall with an extremely complicated entrance system.
Wellesley was at Lobada on the evening of 10 December because Stevenson’s health had not improved. For the next five days he was to divide his time almost evenly between the two armies. This involved a ride of just over twelve miles from Deogaum to Lobada over the rough new road, but he could probably cover the distance in about an hour and a half.
During the night of the nth a breaching battery was begun on the crest of a small rise overlooking the tank only 250 yards from the outermost wall. Fire was opened on the morning of the 12th from two 18-pounders and three iron 12-pounders. There was an enfilading battery of less powerful pieces – two such batteries later on – set further back and to the east to keep enemy personnel from repairing the walls or retrenching the breach.
The weather of India is hard on masonry. The stone used originally at Gawilghur was probably a by-product of scarping the hilltops on which the place was built, and was not good building stone. As at Ahmednuggur, the old solid-masonry walls appeared stronger than they actually were; 12-pounder and 18-pounder shot travelling at more than 1,200 feet per second caused extreme damage after a few hours. Almost every round brought down chunks of masonry. There were to be three breaches in all, a wide one in the lower wall and two in the upper structure.
The fortifications of Gawilghur still are quite complicated. They were built to fit the terrain rather than according to any regular plan. Gawilghur had, therefore, a weakness common to all fortresses of India design; it had little or no means of delivering flanking fire. The outer defences extended for more than six miles and varied in strength in accordance with the designers’ estimates of the inaccessibility to an enemy. In March 1968, for instance, I found one stretch of nearly a mile on the north-east side of the Inner Fort where I could see no trace of any fortifications. In 1803 there might have been a palisade or a trench of some sort, but nothing substantial since the slopes below were unclimbable from a military point of view. Such gaps in the fortifications did not constitute a physical weakness and did not contribute to the ultimate fall of Gawilghur. But they may have undermined the morale of the defenders.
The enemy had more than fifty pieces of artillery on the walls and in cavaliers, concentrated where targets were likely to appear. Some of the guns were large. One is still there, an enormous wrought-iron gun mounted on a small mamelon on the north side of the Inner Fort which could fire through nearly 180 degrees at any target that appeared on, or south of, the crest to the north. Another similar piece was mounted so as to fire into the valley to the south; its balls were said to carry for a distance of several miles, but accuracy would have been poor and a single plunging ball would have been most ineffective.
Stevenson’s army had only an imperfect knowledge of Gawilghur’s internal design. No accurate plan or sketch was available. The British did not understand the communications between the smaller, but slightly higher, Outer Fort on the north-western hill and the Inner Fort on the larger south-eastern hill which was unapproachable except by way of the Outer Fort. The two flat peaks were separated by an irregular ravine up to 300 feet deep, but it could not be clearly seen from any accessible point to the north. In addition to fortifications the Inner Fort contained a number of tanks and many solid buildings. It was at that time a considerable town.
In December 1803 Gawilghur had a garrison of 2,000–4,000 men under a Punjabi killadar, whose name may have been Beny Singh, and a civilian population of about 15,000–30,000. After Argaum Manoo Bappoo and some 4,000–6,000 of his regular infantry came in, accompanied undoubtedly by camp followers who in Mahratta armies were often semi-armed.1 Gawilghur was naturally strong, well fortified by Indian standards and amply garrisoned. Weapons, ammunition and military equipment were plentiful. The tanks were still reasonably full in spite of the poor monsoon and there was plenty of grain.
Throughout history, however, sieges have depended more on skill and morale than on walls and weapons. Wellesley’s engineers were well trained and veterans of similar operations in India. They had skilled pioneers to do their work. The gunners knew how to hit where their shot would be most effective and how to maintain their pieces in action efficiently. The assault would be led by active and courageous British officers who were exceptionally capable with their personal weapons.
By contrast, the Mahratta leaders had little knowledge or skill in the defence of their fortress. They did not try to prevent Stevenson’s army from approaching the ‘isthmus’ which was the only effective breaching ground. They made no effort to protect the wall with an earthen glacis or any form of outwork. They did not fire during the night of the nth at the place where the main battery had to be located.
We should look briefly at what had occurred south of Gawilghur. Between 6 and 8 December Wellesley had driven in all the Mahratta pickets, but he did not endeavour to invest the enormous fortress. He kept the bulk of his troops in camp at Deogaum four miles away, though he had a forward concentration post in the small village of Baury at the junction of ‘roads’ from the south and the north-west gates. British patrols pushed north on both tracks to within a musket shot of the walls. The difficult eastern route to the south gate of the Inner Fort was the only one that could possibly be used to get artillery within range of the fortifications. The other road was better in that it could accommodate draft animals, but was commanded by fire from the guns on the walls of the Inner Fort.
EIC engineers had much experience in moving guns over impossible terrain mainly by manpower. They attempted to get Wellesley’s two remaining iron 12-pounders up the eastern route on the night of the nth. The Indian and, perhaps, European pioneers were reinforced by working parties of muscular Scots from the 74th and 78th regiments. If the task was humanly possible, these men would accomplish it.
Early in the evening engineers, pioneers, artillerymen and working parties began their efforts. We should remember that they had to do their work without artificial illumination, and there was no possibility of dragging the pieces up complete with their carriages – the wheels would not roll over the small steep cliffs. Stripping the carriages was no problem, but each gun, in modern terminology the tube only, became a nine-foot fiend weighing 4,100 pounds (32 cwt) able to crush men with the smallest slip or roll. There was no way to secure tackle above, not even room for a team to pull from a distance. Elephants, which normally were used in all difficult gun movements, could not negotiate the terrain.
The job simply could not be done; the route was too steep and too uneven. After ten hours’ labour, the men buried the pieces under debris and retired as dawn was coming. On the night of the 12th they did manage to get forward two brass 12-pounders and two 5·5-inch howitzers, much lighter pieces, which they mounted in a battery within 400 yards of the south gateway, but about 450 feet below it. The brass 12-pounders had to fire at an elevation of almost thirty degrees and did no serious damage. Their shot are said to have rebounded back to the guns themselves and perhaps into the valley below. The battery was more like a sheepfold than a normal emplacement.
Stevenson’s battering pieces did far better on the northern side. They opened on the 12th, and by the morning of the 14th the breaches were thought practical: an armed man could climb into the fortress. Wellesley had a close look with a telescope and decided on an assault the next day. Stevenson was in no better health, so Wellesley continued to direct both armies, giving verbal orders and discussing all pertinent details with Stevenson’s corps commanders. He confirmed his instructions in writing later that day from his camp at Deogaum.
Wellesley knew from his own inspection that Stevenson’s breaches into Gawilghur were moderately difficult; if the garrison worked hard at repairing the defences on the night of the 14th, they might be un-negotiable the next morning. A skilful fortress commander would surely do this and perhaps place mines and other obstacles in the way of the assault. To discourage such measures, however, a large gun loaded with grape was discharged at the breaches every twenty minutes throughout the night.
A dawn assault had some advantage, but not enough to outweigh a few hours’ additional battering if it should be found necessary. Wellesley also wanted the enemy inside Gawilghur to see his two powerful British forces approaching from the south. The assault was set for 10 a.m.
Wellesley’s attack from the south had no hope of taking the place, but some of the Mahrattas inside had surely heard of the British escalade of the pettah at Ahmednuggur. Wellesley was still relying on audacity. If the Mahrattas had fought skilfully and courageously, Gawilghur could hardly have been taken at all, at least not on the 15th. But the image of British invincibility was already established. Even Hindoos who did not place such a high value on their lives as Europeans could fight effectively only in an atmosphere of some hope.
Stevenson’s attack through the northern breaches was to be led by Lieutenant-Colonel William Kenny of the 1/11 Madras with the grenadier company and two battalion companies of the King’s 94th and the flank companies of three EIC battalions – his own, the 2/11 and the 2/7 Madras. There were also small units of pioneers and artillery, making a total of about 1,000 men in all.
The force that would make the second assault through the breaches if the first should fail, or would follow into the Outer Fort if it succeeded, was commanded by Lieutenant-Colonel Peter C. Desse of the 2/2 Madras; he had the light and two more battalion companies of the King’s 94th with the flank companies of the other three EIC battalions of Stevenson’s army – the 2/2, the 1/6, and the 2/9 Madras. Desse also had pioneers and artillery for a total of about 1,000 fighting men.
Behind these two forces Major James Campbell of the King’s 94th led the other four battalion companies of his corps, backed up by the battalion companies of the 2/7, the 1/11, and the 2/11 Madras under EIC Lieutenant-Colonel John Haliburton, who was senior to all officers except Stevenson in the army. We will hear more of Haliburton.
The assaults were to be pressed home regardless of cost; a total of about 4,600 first-quality fighting men were assembled in the four assault commands. Only EIC Lieutenant-Colonel H. Maclean with the other three EIC battalions less their flank companies was held in camp as a reserve.