Wellington’s army did not commence its easterly advance until the last week of May 1813. A few days before the first unit of Wellington’s allied army began to move, Murray had written to the commander-in-chief expressing his concerns about his part of the wider plan:
Upon the subject of our expedition I have made some calculations, and the result is, that Suchet will be enabled to collect, in the neighbourhood of Tarragona, a force of about 24,000 men, in twelve days, after he knows of our embarkation. In that time, provided we have a fair wind and everything goes on well, the place may be taken. Our force ... in all 22,000 of which 3 or 4,000 will be employed at the siege. Allowing, therefore, that the force of the enemy is exaggerated (which I do not believe it to be) he will have a great superiority in any attempt he may make to relieve the place.1
Wellington had been notified that Suchet’s forces to Murray’s front had not been dealt quite the devastating blow the force commander had intimated in his post-Castalla despatch.2 Nevertheless, Murray’s letter was an early warning that he was uncomfortable with his mission and probably not up to the job. Murray became fixated with troop numbers and dispositions and this obsession clearly began to affect his judgement. Wellington received Murray’s letter as the allied advance commenced and he responded with some urgency; his words were designed to allay Murray’s concerns and inspire some confidence in the execution of his mission.
I think that in reckoning that Suchet will bring 24,000 men to raise the siege of Tarragona, leaving 14,000 in Valencia, you have overrated him, and have not allowed for the necessity of keeping respectable garrisons in all the places in Catalonia. If he is to weaken these garrisons, he cannot then be at Tarragona in twelve days. I think you reckon too much upon the Spaniards. But if it be true, as Copons and Cabranes agree, that the 14,000 disposable troops would keep the French in check in Catalonia, how can Suchet collect 24,000 men disposable in Catalonia, leaving 14,000 in Valencia?3
Murray’s estimates were based on three intelligence sources: firstly, from reconnaissance elements within his own force; secondly, from the Spanish commanders in Catalonia; and thirdly, from our old friend Mr Tupper who was receiving constant reports from merchants, businessmen and Spanish spies who were habitually interacting with the French.4 Murray, as we shall see, remained convinced throughout the siege that 24,000 to 25,000 Frenchmen were poised to fall on Tarragona from the north and south within days. Detailed reports he received in May indicated that Suchet’s force numbered 31,260 men in Valencia and Aragón, while Decaen’s force in Catalonia was 24,805.5 From this combined figure of 55,000 he estimates, not unreasonably, that slightly less than half would be used by Suchet and Decaen to attack his force. That, indeed, was the basis of his figure in his letter to Wellington. In fact, at this time Suchet’s field and garrison forces numbered nearly 23,000; Pannetier’s flying reaction force added another 3,000 and General Paris had a further 6,000 in garrisons in Aragón (see Appendix I). In Catalonia, General Decaen had 25,500 men. So Murray’s estimate of 55,000 is about right, however, a more experienced and perhaps talented general would have realized that releasing half that number for combat operations was unrealistic. Suchet would have to leave at least two-thirds of his force in Valencia to retain control of the region and keep his principal garrisons at Valencia, Sagunto, Peñiscola and Denía, leaving him a force of no more than 10,000 to move north. General Paris in Aragón could spare no man as he was assailed from all sides by El Empecinado and Duran; General Decaen in Catalonia was unlikely to be able to spare many more than 6,000 to 8,000 men without endangering his vast garrisons and losing control of the lines of communication with France. A more realistic estimate, therefore, of the French force that might move to succour the garrison at Tarragona was nearer 15,000 to 18,000, and that was precisely Wellington’s point.
Despite the troop withdrawals by Bentinck in late April, Murray’s force still numbered in excess of 21,000 men. He was, not unreasonably, keen to embark every man. However, he could only take as many men as the naval transports would allow. He resolved to take, as directed in Wellington’s memorandum, the Anglo-Sicilian force complete and as many of Whittingham’s and Roche’s men as could be carried. In the end his entire force numbered just over 17,000 (see Appendix I) and Roche would have been devastated to have been left off the final manifest. Murray had received a number of artillery reinforcements in the months leading up to the operation but the exact breakdown of the siege ordnance and field guns he took to Tarragona is unclear.6 Elio’s 2nd Army moved to fill the defensive positions vacated by the Anglo-Sicilians and Whittingham’s division, while those of Del Parque moved from the border region of Andalusia and deployed into the areas around Yecla and Chinchilla when, in turn, these were vacated by Elio’s men. These two Spanish armies combined numbered about 43,000 and were fully armed, resupplied (to a degree) and poised to advance at the first sign that the French line to their front was being weakened.
Murray’s force began embarkation on 28 May and sailed three days later. Suchet was well aware of the allied preparations and his brigadesized reaction force, under the command of General Pannetier, monitored and shadowed the fleet as it sailed with favourable winds up the coast past Pannetier’s initial positions at Valencia. Suspicions that the objective was Tarragona were confirmed on 2 June when the fleet sheltered under the lee of Cape Salou, 15 kilometres south of the city. General Copons met with Murray and, despite reservations, agreed to concentrate 7,000 men (about 12 battalions and some cavalry) 10 kilometres to the north and east of Tarragona. This constituted all the available troops of the 1st Army, as the balance of 8,000 men were required to hold the Spanish garrisons in the interior; a problem not exclusive to the French. Two of these battalions under Colonel Lauder were earmarked to join a small force (2/67th, De Roll-Dillon’s Battalion and two guns under Captain Arabin), tasked by Murray to move under Colonel Prevost directly by land to the Col-de-Balaguer, where they were to take control of the defile and capture the small fort of San Filipe located there. This was a sensible precaution as the defile controlled the coast road from the lower Ebro and this was the only suitable road for artillery guns, heavy baggage and ammunition wagons in the area. The only alternative was an interior route which was circuitous, narrow and treacherous for the heavier wheeled wagons and carts. Capture of the fort was ‘desirable’ in Murray’s orders, domination of the road was the overriding mission and this could be achieved at a number of areas to the north of the fort itself. Hallowell sent a strong naval force to assist them in their task.7 Prevost arrived late on 3 June and conducted a reconnaissance while Arabin’s two light guns deployed in support and fired a few shrapnel shells in order to keep the defenders’ heads down. The next day some heavier ordnance arrived and the battering began but with little effect. It was not until five heavy 24-pounder guns and two heavy mortars arrived, having been landed from the Stromboli, that the siege began in earnest. The ordnance opened on 7 June and one of the mortar rounds hit the secondary magazine, and the resulting explosion was enough to convince the demoralised garrison to surrender.8 It was a good start for Murray’s force. They now controlled the main access road from the south.
Back at Tarragona, however, the main effort was not progressing quite so well. At Salou Bay the force began disembarking early on 3 June and by mid-afternoon the entire force and 22 guns (a mix of field and siege) and a large proportion of the stores and ammunition had been brought ashore. The investment of Tarragona was complete that night. Murray conducted a fairly thorough reconnaissance of the place with his chief engineer, Major John Thackeray, and his Commander Royal Artillery (CRA), Major John Williamson. They concluded, not surprisingly given the terrain, that an attack from the west was the best option. The garrison, under the command of General Antoine Bertoletti, consisted of about 1,600 men including one French (20th Line) and one Italian (7th Line) battalion. It was an inadequate-sized force and the Milanese commander was unlikely to cause any surprises. The sight of this large besieging force complete with a number of siege guns sapped the garrison’s morale and fostered a feeling of hopelessness. Nevertheless, Bertoletti did his best to raise the spirits of his men, unaware at this stage that his best ally was in fact the allied commander himself.
Two batteries were marked out and work began immediately on the positions adjacent to the road running parallel to the Francoli River. Tom Scott, the brigade major to the CRA, noted on 5 June that:
This day we made some advance and in the evening threw four 8-in Howitzers into a battery on the sea beach, about 900 yards from the body of the place and another battery of two 24s near the main road, directed towards Fort Reale, an outwork 400 yards from the line wall, and which will be taken before our breaching batteries are established; indeed in my opinion the work ought to have been stormed at once, as from our unexpected arrival the French were not prepared to defend the place, as they will be by the delay on our part.9
It was a view shared by many in Murray’s army. In the lower part of the city, where the repairs from the French siege had been rather hastily completed, Bertoletti had positioned a mere handful of men. In the Fort Royal and the bastion of San Carlos, which had also been completely closed, a single company about 100 strong constituted the total defence force. They only had a single heavy gun in each strongpoint, which was all they could man. Murray’s pessimism, which he had demonstrated prior to the embarkation of his force, resurfaced almost immediately. It clearly affected his judgment and the corresponding plan of action. On 7 June, four days after the troops and stores were landed, he would write to Wellington:
I am afraid we have undertaken more than we are able to perform, but to execute your Lordship’s orders, I will persevere as long as prudence will permit. I have as yet no certain information about Suchet’s movements nor of De Caen [Decaen] to the eastward but there are reports of both, and if they prove true in five of six days I may be attacked by a force ultimately superior, without the hopes or chance of retreat.10
Despite Murray’s concerns, the siege continued and on 6 June the two batteries had opened against the two structures in the lower town. The next night Murray ordered the construction of a third battery near the Francoli Fort and that was completed, manned and ready the next day. The fire from the three batteries was steady and effective. Nevertheless, the infantry were keen to get on with the assault, as they were well aware that defenders given time would retrench the defences behind the breached area and construct all sorts of hidden surprises to counter the attack. Many infantry officers began to voice their disapproval at the methods and lack of urgency of the technical services. Scott noted in his diary:
Skelton’s fire from the Howitzers as well as Cox’s from the advance appeared to do much damage to the work, but this very piddling work, and every one crying: “Why haven’t you more guns up?” However what officers of the line generally know about sieges is regulated by the quantity of noise and fire they hear and see, without reflecting upon the nature of the ground, the situation of the place or the difficulties to be overcome, and conceive it as easy work for the Engineers and Artillery to proceed in their duty, as the common routine of Infantry parades.11
The guns in the third battery had enfiladed the Fort Royal, levelled the parapet and rendered the single gun inoperative. At dusk on 8 June, Thackeray reported a practicable breach but then went on to caveat that such an assault would be virtually worthless as ‘he could turn the immediate possession to no account, and that retaining it would cost the lives of many men’.12 He went on to propose a new plan that, he concluded, would take about two weeks to produce another breach in the upper city walls. This entire episode raises many questions. Why did Thackeray not realize this impenetrability during his initial engineer reconnaissance? Why did the artillery advisor not have a say or express an opinion on the matter?13 How did Thackeray come to this conclusion? After all, the French siege in 1811 had established the third parallel in the lower city area after it was captured. The fortifications, ruins and houses in the lower city area had provided the French with good cover and the walls between the San Pablo and San Juan bastions had been breached during the siege and not properly repaired.14 So on what premise Thackeray came to his extraordinary conclusion is baffling; it was not, but certainly should have been, raised during Murray’s subsequent court martial. Confronted with this report, Murray really only had two options. Firstly, to raise the siege as he clearly felt that the combined French forces would descend on his location in far less than two weeks or, secondly, to continue the siege against the lower town, as it was originally planned, in the hope that it was concluded before the French arrived. It is, therefore, difficult to comprehend why he elected to follow Thackeray’s recommendation, which was not an option, and nothing in his letters or his subsequent testimony enlightens the mystery. Murray’s explanation is thoroughly confusing: ‘Every day was to be regretted’, he wrote, ‘but as the state of the fort was now such that it could be taken when convenient, I consented to defer the attack, and directed that the fire upon it should continue only sufficiently to prevent its re-establishment.’15
Scott noted in his diary that the engineers needed to up their game and that ‘terrible invectives thrown out against the sister Corps; cannot help concurring in opinion that they are rather slack and in our present situation, a little more activity would be highly commendable’. He went on to add that ‘there appears a general apathy, arising too from the apparent indecision of Sir John’.16 There was a growing feeling of discontent across the force. Hallowell, Clinton and Mackenzie were beside themselves with anger at the way the siege was being conducted and by the way Murray kept them all in the dark with regard to his intentions. They all considered the lower defences in the town assailable by escalade but Murray was having none of it. Orders were given to land more siege ordnance, ammunition and stores and this was undertaken over the next few days. The more that was disembarked, the more that would have to be re-embarked should the siege need to be raised. To all intents and purposes Murray had just dug his own grave.
On 9 June Murray wrote again to Wellington, informing him of the success of Prevost’s brigade in capturing the Fort San Filipe at Col-de-Balaguer and emphasising the importance of this action in closing the road from Tortosa to Tarragona.17 He made no mention of his revised plan nor did he make any reference to his earlier concerns and he did not provide an update as to the whereabouts and strengths of the French forces. On 10 June, while the two new batteries were being constructed and equipped, Murray decided to ride out to meet Copons at Torre dem Barra. A number of points emerged from this meeting. Firstly, Murray confirmed that Copons had only 8,500 men in the field; secondly, Copons clearly stated that he was unable, on his own, to hold off the French force marching towards Tarragona from Barcelona; thirdly, that Murray agreed in principle to support Copons along the line of the River Gaya with no more than 9,000 men.18 The next day Murray rode out again to meet Copons at Vendrills to discuss the finer detail of that support and to conduct a more detailed reconnaissance. Before leaving his headquarters he had summoned Clinton and given him very clear orders:
Early in the morning of the 11th, I received orders to go to headquarters, and my division to be in readiness to march; on arriving at head-quarters, I found Sir John going out, he said to me that he was going to the river Gaya to look for a position to meet the enemy, who was advancing from the side of Barcelona. He said to me that he left me in temporary command of the siege, and added, if Major Thackeray reports the breach in the Fuerte Reale [Fort Royal] to be practicable, let it be stormed.19
Murray, on arrival at the banks of the River Gaya, was less than impressed with the defensive qualities of the position but nevertheless agreed to send six battalions the following morning.20 Meanwhile, back at Tarragona at about midday, Thackeray reported to Clinton that the breach was practicable and the temporary commander gave the orders for the assault at 10 pm that evening. Expectations were high and the infantry and engineers completed their final preparations. Murray returned to his headquarters in the early evening and Clinton briefed him on the situation and sought confirmation that the attack was to continue. He was told that the attack was to go ahead. Clinton returned to brief the commanders and ensured that everything was organized before returning to Murray’s headquarters to report that all was in readiness. On arrival he was astounded to discover that Murray, based on an intelligence report about the movement of Suchet’s forces from the south, had reversed the order. A young captain on the staff was immediately sent to Fort Olivo to prevent the signal rocket being fired to instigate the assault. He arrived just in time. Morale plummeted across the force. Scott’s account summed up the mood:
At night the 10th Regiment with three companies of the Levy, headed by General M’Kenzie were paraded for the purpose of storming the outwork, and a strong diversion was to have been made by the division under Gen Clinton, as well as from Whittingham’s Corps. 10 was the hour fixed, and every one in anxious expectation of seeing it, when at 1/2 past 9 an order arrived to postpone it till 11. This was far from being relished, but how dreadfully disappointed and vexed was every one in half an hour after to find it was not to take place. A murmur passed along the troops, and every one expressed themselves in strong language against the horrid indecision of Sir John, and no one more so than Gen M’Kenzie.21
Murray was clearly shaken by this intelligence report but it is difficult to determine exactly which report caused the commander to decide the game was up. Intelligence reports of French troop numbers and their locations had been at the forefront of Murray’s mind long before the expedition set sail. They are the driver for Murray’s subsequent decisionmaking and deserve detailed investigation. In terms of Suchet’s force moving north it was Consul Tupper who on 5 June passed early intelligence from his extensive group of informants across Valencia that the French were weakening their forces south and west of the city and preparing to move north the next day. He went to conclude, in no official capacity, that ‘I think it very likely that Suchet will retreat with his army upon Aragon and leave only a small garrison at Tortosa’.22 Two days later another source (Sr. Larente) passed more information, for which he was paid cash by Tupper, which clarified that Suchet was leaving Valencia with 9,000 men towards Tortosa and, one supposes therefore, not withdrawing from his forward positions on the River Júcar.23 This approximate figure was corroborated the following day with a report from Captain Charles Adam RN (HMS Invincible) who had received information from the mayor of Tortosa that 2,000 men had arrived at Tortosa and that another 6,000 were due.24 The next day, Captain Alexander Sharpe RN, in HMS Hyacinth off the coast at Sagunto, passed intelligence of considerable troop movements ‘estimated at 10,000 which have left this province’.25 Finally, on 11 June at 3 pm, information from a local peasant confirmed that Suchet had arrived at Tortosa with 8,000 men and that they marched again today ‘and may be at Reus tomorrow’.26 Other reports on Suchet’s and Pannetier’s movements followed but as these would have arrived after Murray made his decision to end the siege they are not considered here.
With regard to the French forces descending from the north, both Copons and Cabanes had provided periodic updates and intelligence estimates throughout May. However, the first on-the-ground estimate comes from Whittingham’s cavalry reconnaissance on 8 June in which he claims that ‘the enemy in force from 4 to 5,000 men have put themselves in motion’.27 Two days later a letter by Colonel Joseph Manso, a Spanish officer, confirmed that ‘this evening 10,000 men with 200 cavalry and 14 pieces of artillery would begin their march from [Barcelona]’.28 The next day Murray received two reports: the first from Whittingham that the force was ‘from 7 to 8,000 infantry and 3 to 400 horse with 6 pieces of cannon’, the second from Manso corroborating the numbers at 8,000 infantry and 200 horse.29 A far less reliable source arrived at Murray’s headquarters at midday on 11 June and indicated that the force was ‘supposed to be 10,000 men’.30 This formed the basis of Donkin’s subsequent report to Murray which he received on return from the reconnaissance on the Gaya and, it is assumed, as Murray never stated definitively, the one which forced the allied commander to change his mind and raise the siege.
From these two sets of reports it is fairly conclusive that Decaen had despatched 8,500 or so men (under the command of Maurice Mathieu) and Suchet had a similar number. However, it appears that Murray was convinced that the figure of 8,000 for Suchet did not include the 3,000 men of Pannetier, which is not entirely unreasonable, and that the Barcelona column was nearer 10,500 if not 13,000, which is less realistic. Furthermore, Donkin’s report of 11 June went on to speculate that Suchet’s force could be at Reus tomorrow and also ‘as De Caen, if he marched this day to Villa Franca, can also reach Reus tomorrow’.31 Thus according to this assessment prepared by Donkin at 3 pm and forwarded to Murray in the field, both Suchet’s and Decaen’s forces would, most likely, have united a few miles from Tarragona at some stage on 12 June. It was enough to convince Murray that the game was up. The order was given to raise the siege and embark the force and it is from this point on, until he is relieved of command by Bentinck a few days later, that Murray appears to lose his head.
In fact Murray had already determined on 9 June that, in all probability, he would have to raise the siege and not unreasonably asked Quartermaster-General Donkin to make plans accordingly.32 However, Murray did not, for reasons which are not clear, order Donkin to put this contingency into operation and it also appears that very little seems to have been ordered or undertaken in the early hours of 12 June. Williamson recalled that at half an hour before midnight a staff officer came to summon him to Murray’s headquarters. On arrival the artillery commander was informed of the decision and told to ‘get the guns immediately off’.33 He informed Murray that it would take time to rally working parties, disengage the guns and move them back towards the beach. It would be daylight before they were half way to the embarkation point and they would pass a portion of the road which was under fire from the fort and this would inevitably result in considerable loss of men. Williamson went on to state that he could bring the guns off safely the following night and Murray agreed. Early on 12 June Donkin received Murray’s orders:
Very soon after daylight on the 12th, I went to take Sir John Murray’s orders; he then told me, that he was sorry he had not been able to get the battering guns down during the night, and that he should be unable to bring them away. He then acquainted me with the order in which he meant to embark his troops, which was, that he should first withdraw General Whittingham’s corps, that General Clinton should cover the march of it with his division, and that General Clinton’s division should then retire, withdrawing his picquets, destroying the gun carriages and spiking the guns. This Sir John Murray communicated to me as his intention; but not as an order for execution.34
As Donkin returned he bumped into Williamson and asked him if he had the means to conduct the destruction of the siege guns. Williamson was beside himself and, after ordering officers to prepare to carry out the order, he went straight to Murray’s headquarters to plead his case. It is unclear if he was given an audience but what is clear is that Murray was now convinced that he only had until dusk to embark the force. At 10 am orders were sent to both Clinton and Mackenzie to embark their troops and for all the guns to be spiked or abandoned in the batteries in their areas of responsibility. Both officers recall receiving a verbal order and putting it into effect but neither can recall the time the guns and certain carriages were destroyed or burned.35 Curiously, Murray was adamant that he had not given the go-ahead for the guns to be spiked. At around midday Hallowell was at Murray’s headquarters, along with Donkin, who stated it was much earlier, at about 8 am, when a Spanish officer arrived with a letter from Copons containing up-to-date intelligence as to the strength and whereabouts of Maurice Mathieu’s force. It confirmed that at seven that morning the force, of about 5,000 men, was still at Vilafranca [Vilafranca Del Penedès], approximately 25 kilometres and a good day’s march from Tarragona. Donkin dismissed the report as wholly unreliable to which Hallowell suggested that he should send a cavalry reconnaissance out to confirm the information. Donkin refused but word had already leaked from the headquarters that the likelihood of the French force arriving on 12 June was very remote. In the early afternoon General Mackenzie and Colonel Adam tried to convince Murray to march out and attack Maurice Mathieu. Murray rejected the proposal out of hand, quoting paragraph 13 of Wellington’s memorandum as his rationale, at which point Mackenzie agreed that ‘there was no answering that’.36
Frederick Bentinck, who had been deployed to the north of Tarragona to monitor the approach roads from Barcelona, arrived back at Tarragona with his cavalry at about midday:
I remained behind till near seven, to the best of my recollection, between six and seven o’clock, in order to gain any information I could. I could not learn from the peasants that the enemy was nearer Villa Franca [Vilafranca]; I reached head-quarters between eleven and twelve o’clock, I found there part of my brigade that had remained there, had been embarked, and that the remainder of the 20th were upon the point of embarkation. I also heard that the guns were ordered to be spiked. No question was asked of me respecting any information that I had of the movements of the enemy; soon afterwards I heard Sir John Murray say that he had not given the order to spike the guns, and at about the same time, General Donkin gave me an order to stop the re-embarkation of the remainder of the 20th, who had not got on board, and to march with them and the rest of the cavalry, and the two six-pounders which had come with me to Coll de Ballaguer, which order, I carried into execution between two and three o’clock that same afternoon.37
This is an interesting statement by Bentinck. Firstly, it clearly indicates that both Murray and Donkin had made up their collective minds on the location, strengths and intentions of the French and made no attempt to enquire from the senior cavalry commander as to any additional intelligence he might possess on the issue. Secondly, it indicates confusion as to whether or not the order to spike the guns had been given and disseminated. As it turned out, the decision to send Bentinck some of the cavalry and half of the 2/27th to the pass at Balaguer was in fact Hallowell’s idea from a letter he penned to Murray at 9 am in which he assured him that ‘their retreat would be secure, and their embarkation certain’.38
Clinton’s division was the last to embark; when they were on board and with no sign of either Suchet or Maurice Mathieu, the boats were sent back to the shore by Hallowell and the mules, engineer stores, ammunition, and some rations were loaded along with some of the field guns and other siege ordnance which had been left on the beach. There are conflicting reports as to how long this final loading continued; however, it certainly appears to have gone on until early the next morning. When the fleet finally sailed at about 9 am, the French garrison ventured out, somewhat bewildered, to discover eight 24-pounder guns, five 10-inch mortars and five 5-inch mortars had been abandoned in the five batteries, along with amounts of shot and shell.39 On the beach they found the detritus of a fleeing army including some badly needed rations. Scott, who had not had the time to complete his diary for two days, now penned his entry for 13 June as he watched the French garrison pick over the abandoned spoils:
After a glorious night’s rest rose with much better spirits, but not without imprecating the brute, who has destroyed our hopes of displaying to the world we are not that inactive army, that we have hitherto unfortunately been. Alas! Our sun has set and nothing but a gloomy cloud of despair hangs over us. Whither we go, I care not, but the more solitary the place, the more adapted to our luckless situation; for we cannot but be reviled wherever we go – 22,000 flying before an Army of 8000!! Was Britain ever so disgraced before? Never were her soldiers so debased. Plainly could we perceive the enemy dismantling our batteries and carts innumerable [and] employed in removing our numerous stores left behind. How shameful; plenty of time had we to prevent all this, but the stigma, Thank God, rests upon the shoulders of him, whom I hope will be for ever disgraced, if not hung or shot, which he undoubtedly merits.40