SECTION XVI: CHAPTER VIII

THE RETREAT FROM TALAVERA

When the dawn of July 29 had arrived, the plain and the rolling hills in front of the allied position were seen to be absolutely deserted. No trace of the French army was visible save the heaps of dead upon the further side of the Porti?a: the wounded had been carried off, with the exception of those who had fallen within the British lines, and so become prisoners of war. It was soon discovered that the enemy had left a screen of cavalry along the western bank of the Alberche: but whether his main body lay close behind the stream, or had retired towards Madrid, could not be ascertained without making a reconnaissance in force. Such an operation was beyond Wellesley’s power on the morning after the battle. He was neither able nor willing to send out a large detachment to beat up the enemy’s camps, with the object of ascertaining his situation and intentions. The British army was utterly exhausted: on the preceding day the men had fought upon half-rations: when the contest was over they had found that only a third of a ration had been issued: this scanty pittance was sent up to the regiments in the evening, as they still lay in battle-order on the ground that they had held during the day. Water was almost equally deficient: it was difficult to procure: nothing but the wells of the few houses in the rear of the position being available. Only on the morning of the twenty-ninth, when the departure of the enemy had become certain, were the troops allowed to return to their old bivouacs in the rear, and there to seek repose. Even then it was only a minority of the men who could be spared from duty. The gathering in of the vast numbers wounded—French as well as English—and their removal into Talavera demanded such enormous fatigue-parties that the larger number of the survivors had to be told off to[p. 560] this work and were denied the rest that they had so well earned.

It is certain that the British army could have done nothing upon the twenty-ninth even if their commander had desired to push forward against the enemy. The men were not only tired out by two days of battle, but half-starved in addition. But Wellesley was far from feeling any wish to pursue the French. His infantry had suffered so dreadfully that he could not dream of exposing them to the ordeal of another engagement till they had been granted a respite for the refreshment of body and spirit. Of his divisions only that of A. Campbell—the smallest of the four—was practically intact. The others had suffered paralysing losses—in Hill’s ranks one man out of every four had been stricken down, in Mackenzie’s one man in every three, while Sherbrooke’s frightful casualty-list showed that nearly two men out of five were missing from the ranks. Never, save at Albuera, was such slaughter on the side of the victors seen again during the whole course of the Peninsular War. ‘The extreme fatigue of the troops,’ wrote Wellesley, ‘the want of provisions, and the number of wounded to be taken care of, have prevented me from moving from my position[685].’

On the morning of the twenty-ninth the depleted strength of the army was partly compensated by the arrival of the first of those reinforcements from Lisbon which Wellesley had been anxiously expecting. At about six o’clock Robert Craufurd came upon the scene with the three regiments of his Light Brigade—all old battalions who had shared in Moore’s Corunna campaign. He was accompanied by a battery of horse artillery (A troop), the first unit of that arm which came under Wellesley’s command. But the Light Brigade were almost as weary as their comrades who had fought in the battle: they had only reached Talavera by a forced march of unexampled severity. Hearing at Navalmoral that the two armies were in presence, Robert Craufurd had hurried forward with almost incredible swiftness. Dropping his baggage and a few weakly men at Oropesa he had marched forty-three miles in twenty-two hours, though the day was hot and every soldier carried some fifty pounds’ weight upon his back. All day long the cannon was heard growling in the distance,[p. 561] and at short intervals the brigade kept meeting parties of Spanish fugitives, interspersed with British sutlers and commissaries, who gave the most dismal accounts of the progress of the fight. In spite of his desperate efforts to get up in time Craufurd reached the field thirteen hours too late, and heard to his intense chagrin that the battle had been won without his aid[686]. Weary though his men were, they were at once hurried to the front, to relieve A. Campbell’s division on the line of advanced posts. There they found plenty of employment in burying the dead, and in gathering up the French wounded, whom it was necessary to protect from the fury of the Spanish peasantry.

The arrival of Craufurd’s brigade did something towards filling up the terrible gap in the ranks of the British infantry, but was far from enabling Wellesley to assume the offensive. Indeed the advent of fresh troops only accentuated the difficulty of feeding the army. Corn was still almost unobtainable; the supplies from the Vera de Plasencia showed no signs of appearing, and even oxen for the meat-ration, which had hitherto been obtainable in fair quantities, were beginning to run short. Nothing was to be had from Talavera itself, where Victor had exhausted all the available food many weeks before, nor could any assistance be got from the Spanish army, who were themselves commencing to feel the pinch of starvation.

All Wellesley’s hopes at this juncture were founded on the idea that the diversion of Venegas upon the Upper Tagus would force the French host in his front to break up, in order to save Madrid from an attack in the rear. The army of La Mancha had failed to keep Sebastiani in check, and to prevent him from appearing on the field of Talavera. But since the enemy had concentrated every available man for the battle, it was certain that Venegas had now no hostile force in his front, and that the way to the capital was open to him. If he had[p. 562] pushed on either by Aranjuez or by Toledo, he must now be close to the capital, and King Joseph would be obliged to detach a large force against him. That detachment once made, the army behind the Alberche would be so much weakened that it would be unable to face the British and Cuesta. If it offered fight, it must be beaten: if it retired, the allies would follow it up and drive it away in a direction which would prevent it from rejoining the troops that had been sent against Venegas. On the twenty-ninth Wellesley was under the impression that the army of La Mancha had already brought pressure to bear upon the French, for a false report had reached him that on the previous day it had captured Toledo. His dispatches written after the arrival of this rumour indicate an intention of moving forward on the thirtieth or thirty-first. The King, he says, must now detach troops against Venegas. This being so, it will be necessary to induce Cuesta to advance, supporting him with the British army ‘as soon as it shall be a little rested and refreshed after two days of the hardest fighting that I have ever been a party to. We shall certainly move towards Madrid, if not interrupted by some accident on our flank[687].’

The last words of this sentence are of great importance, since they show that already upon the day after Talavera Wellesley was beginning to be uneasy about his left flank. Some time before the battle he had received news from the north, to the effect that both Ney and Kellermann had returned to the valley of the Douro, after evacuating Galicia and the Asturias[688]. He had therefore to take into consideration the chance that the enemy might move southward, and fall upon his line of communication with Portugal, not only with the corps of Soult, but with a large additional force. Unfortunately the information that had reached him from the plains of Leon had been to the[p. 563] effect that Ney’s and Kellermann’s troops were much reduced in numbers and efficiency, so that even when they had joined Soult the total of the French field army upon the Douro would not much exceed 20,000 men[689]. This misconception affected all his plans: for if the hostile force about Salamanca, Zamora, and Benavente was no greater than was reported, it followed that any expedition sent against his own communications could not be more than 12,000 or 15,000 strong, since Soult would be forced to leave a containing force in front of Beresford and Del Parque, who now lay in the direction of Almeida and Ciudad Rodrigo. Any French advance against Bejar and Plasencia, therefore, would, as Wellesley supposed, be a mere raid, executed by a comparatively small force. He doubted whether Soult dared undertake such an operation: ‘the enemy,’ he wrote, ‘would not like to venture through the passes into Estremadura, having me on one side of him, and you [Beresford] and Romana upon the other[690].’ He was therefore not much disturbed in mind about the movements of the French in the valley of the Douro. If he had but known that not 20,000 men but 50,000 men were now concentrating at Salamanca, his feelings would have been far different. But it was not till some days later that it began to dawn upon him that Soult was far stronger than he had supposed, and that there might be serious danger to be feared from this quarter. Meanwhile he hoped to prevent any advance of the French in the direction of Plasencia, by causing a strong demonstration to be made in the valley of the Douro. He wrote to Beresford that he must contrive to arrange for joint action with La Romana and the Army of Galicia. If they appeared in strength in the direction of Ciudad Rodrigo, the Duke of Dalmatia might be deterred from making any movement to the south. If, however, the Spaniards proved helpless or impracticable, the Portuguese army would have to confine itself to the defence of its own frontier.

On the morning of July 30 Wellesley received the first[p. 564] definite information which led him to conclude that the French forces from the north were actually contemplating the raid upon his communications which on the preceding day he had regarded as doubtful. The Marquis Del Reino, whom, as it will be remembered, Cuesta had sent to the Puerto de Ba?os with two weak battalions, reported that troops from the Douro valley were threatening his front. At the same time messages were received from the Alcaldes of Fuente Roble and Los Santos, places on the road between Salamanca and Bejar, to the effect that they had received orders from Soult to prepare 12,000 and 24,000 rations respectively, for troops due to arrive on July 28. The numbers given counted for little in Wellesley’s estimation, since it is the commonest thing in the world for generals to requisition food for a far larger force than they actually bring with them. But at least it seemed clear that some considerable detachment from Salamanca was on its way towards the Puerto de Ba?os. In consequence of this fact Wellesley wrote to the Spanish government, and also informed Cuesta, that in the event of a serious attempt of the enemy to cut his communications, he should ‘move so as to take care of himself,’ and do his best to preserve Portugal[691]—in other words, that he should abandon the projected march on Madrid which had been his main purpose on the preceding day. He was still, however, under the impression that Soult had no very large force with him, as is sufficiently shown by the fact that on the thirty-first he suggested to Cuesta that it would be well to detach one of his divisions—say 5,000 men—to strengthen the insignificant force which was already in position at the Puerto de Ba?os. ‘I still think,’ he wrote, ‘that the movements of General Beresford with the Portuguese army on the frontier, and that of the Duque del Parque from Ciudad Rodrigo, combined with the natural difficulties of the country, and the defence by the Marquis Del Reino, may delay the enemy’s advance till the arrival of your division[692].’ It is clear that when he[p. 565] wrote in these terms Wellesley was still labouring under the delusion that Soult’s advance was a mere raid executed by one or two divisions, and not a serious operation carried out by a large army.

While Wellesley was spending the three days which followed the battle of the twenty-eighth in resting his men and pondering over his next move, the enemies whom he had defeated at Talavera were in a state of even greater uncertainty and indecision. By daylight on July 29, as we have already seen, the whole French army had retired behind the Alberche, leaving only a screen of cavalry upon its western bank. The King was under the impression that Wellesley and Cuesta would probably follow him up ere the day had passed, and drew up his whole force along that same line of heights which Victor had occupied upon the twenty-second and twenty-third of the month. But when nothing appeared in his front during the morning hours save a few vedettes, he realized that he might count upon a short respite, and took new measures. After sending off to his brother the Emperor a most flagrantly mendacious account of the battle of Talavera[693], he proceeded to divide up his army. As Wellington had foreseen, he detached a large force to hold back Venegas and the army of La Mancha, who were at last coming into the field upon his flank. He was bound to do so, under pain of imperilling the safety of Madrid.

It is time to cast a glance at the operations of the incompetent general whose sloth and disobedience had wrecked the plan that Wellesley and Cuesta had drawn out at their con[p. 566]ference near Almaraz. On July 16 Venegas had begun to move forward from El Moral, Valdepe?as, and Santa Cruz de Mudela, in accordance with the directions that had been sent him. He occupied Manzanares and Daimiel, and then came into collision with Sebastiani’s cavalry at Villaharta and Herencia, for the 4th Corps had not yet begun to withdraw towards Madrid. Owing to the profound ignorance in which the enemy still lay as to the advance of Wellesley and Cuesta, Sebastiani had not, on the nineteenth, received any order to fall back or to join Victor and the King. Thus, when pressed by the advanced troops of Venegas, he did not retire, but held his ground, and showed every intention of accepting battle. Learning from the peasantry that he had the whole of the 4th Corps in front of him, and might have to deal with nearly 20,000 men, the Spanish general halted, and refused to advance further. In so doing he was fulfilling the spirit of the instructions that had been sent him, for Cuesta and Wellesley had wished him to detain Sebastiani and keep in touch with him—not to attack him or to fight a pitched battle. They had taken it for granted that the Frenchman would receive early news of their own advance, and would already be in retreat before Venegas came up with him. But it was not till July 22, as we have already seen, that Victor and King Joseph obtained certain intelligence of the march of the allies upon Talavera. Until the orders for a retreat arrived from Madrid, the 4th Corps was kept in its old position at Madridejos, and courted rather than avoided an engagement with the army of La Mancha[694].

Venegas, after summoning his divisional generals to a council of war, refused to attack Sebastiani, and wisely, for his 23,000 men would certainly have been beaten by the 20,000 Frenchmen who still lay in front of him. From the nineteenth to the twenty-second the two armies faced each other across the upper Guadiana, each waiting for the other to move. Late on the twenty-third, however, Sebastiani received his orders to evacuate La Mancha, and to hasten to Toledo in order to join Victor[p. 567] and the King, in a combined assault upon Wellesley and Cuesta.

It was on the next day that Venegas committed the ruinous error which was to wreck the fate of the whole campaign. On the morning of the twenty-fourth the 4th Corps had disappeared from his front: instead of following closely in the rear of Sebastiani with all speed, and molesting his retreat, as his orders prescribed, he made no attempt to prevent the 4th Corps from moving off, nor did he execute that rapid flanking march on Aranjuez or Fuentedue?as which his instructions prescribed. He moved forward at a snail’s pace, having first sent off to Cuesta an argumentative letter, in which he begged for leave to direct his advance on Toledo instead of on the points which had been named in his orders. On the twenty-sixth he received an answer, in which his Commander-in-chief authorized him to make his own choice between the route by Aranjuez and that by Toledo.

Venegas had already committed the fatal error of letting Sebastiani slip away unmolested: he now hesitated between the idea of carrying out his own plan, and that of obeying Cuesta’s original orders, and after much hesitation sent his first division under General Lacy towards Toledo, while he himself, with the other four, marched by Tembleque upon Aranjuez. So slow and cautious was their advance that Lacy only arrived in front of Toledo on July 28—the day that the battle of Talavera was fought, while Venegas himself occupied Aranjuez twenty-four hours later, on the morning of the twenty-ninth. He had taken six days to cross the sixty miles of open rolling plain which lie between the Guadiana and the Tagus, though he had been absolutely unopposed by the enemy whom he had allowed to slip away from his front. Sebastiani had marched at the rate of twenty miles a day when he retired from Madridejos to Toledo, Venegas and Lacy followed at the rate of ten and twelve miles a day respectively. Yet the special duty imposed on the army of La Mancha had been to keep in touch with the 4th Corps. Further comment is hardly necessary.

On the morning of the day when Wellesley was assailed by the forces of Victor and King Joseph, General Lacy appeared[p. 568] in front of Toledo. The town was held by 3,000 men of Valence’s Polish division: it is practically impregnable against any attack from the south, presenting to that side a front of sheer cliff, overhanging the river, and accessible only by two fortified bridges. To make any impression on the place Lacy would have had to cross the Tagus at some other point, and then might have beset the comparatively weak northern front with considerable chances of success. But he contented himself with demonstrating against the bridges, and discharging some fruitless cannon-shot across the river. General Valence, the Governor of Toledo, reported to Jourdan that he was attacked, and his message, reaching the battle-field of Talavera after Victor’s second repulse, had a certain amount of influence on the action of King Joseph. The place was never for a moment in danger, as Lacy made no attempt to pass the Tagus in order to press his attack home.

On the following morning (July 29) Venegas reached the other great passage of the Tagus, at Aranjuez, with two of his divisions, and occupied the place after driving out a few French vedettes. He pressed his cavalry forward to the line of the Tajuna, and ere nightfall some of them had penetrated almost as far as Valdemoro, the village half way between Aranjuez and Madrid. No signs of any serious hostile force could be discovered, and secret friends in the capital sent notice that they were being held down by a very weak garrison, consisting of no more than a single French brigade and a handful of the King’s Spanish levies. There was everything to tempt Venegas to execute that rapid march upon the capital which had been prescribed in his original orders, but instead of doing so this wretched officer halted for eight whole days at Aranjuez [July 29 to August 5].

On the day after Talavera Jourdan and Joseph had not yet discovered the whereabouts of the main body of the army of La Mancha: but Lacy had made such a noisy demonstration in front of Toledo that they were inclined to believe that his chief must be close behind him. Accordingly the garrison of Toledo was reinforced by the missing brigade of Valence’s Polish division, and raised to the strength of 4,700 men. The King, with the rest of Sebastiani’s corps and his own Guards and[p. 569] reserves, marched to Santa Ollala, and on the next day [July 30] placed himself at Bargas, a few miles in rear of Toledo. In this position he would have been wholly unable to protect Madrid, if Venegas had pressed forward on that same morning from Aranjuez, for that place is actually nearer to the capital than the village at which Joseph had fixed his head quarters. The sloth displayed by the Spanish general was the only thing which preserved Madrid from capture. On August 1, apprised of the fact that the main body of the army of La Mancha was at Aranjuez and not before Toledo, Joseph transferred his army to Illescas, a point from which he would be able to attack Venegas in flank, if the latter should move forward. Only Milhaud’s division of dragoons was thrown forward to Valdemoro, on the direct road from Aranjuez to Madrid: it drove out of the village a regiment of Spanish horse, which reported to Venegas that there was now a heavy force in his front. For the next four days the King’s troops and the army of Venegas retained their respective positions, each waiting for the other to move. The Spaniard had realized that his chance of capturing Madrid had gone by, and remained in a state of indecision at Aranjuez. Joseph was waiting for definite news of the movements of Wellesley and Cuesta, before risking an attack on the army of La Mancha. He saw that it had abandoned the offensive, and did not wish to move off from his central position at Illescas till he was sure that Victor was not in need of any help. Yet he was so disturbed as to the general state of affairs that he sent orders to General Belliard at Madrid to evacuate all non-combatants and civilians on to Valladolid, and to prepare to shut himself up in the Retiro.

The doings of Victor, during the five days after he had separated from the King, require a more lengthy consideration. Left behind upon the Alberche with the 1st Corps, which the casualties of the battle had reduced to no more than 18,000 men, he felt himself in a perilous position: if the allies should advance, he could do no more than endeavour to retard their march on Madrid. Whether he could count on any further aid from the King and Sebastiani would depend on the wholly problematical movements of Venegas. Somewhat to his surprise Wellesley and Cuesta remained quiescent not only on the[p. 570] twenty-ninth but on the thirtieth of July. But an alarm now came from another quarter: it will be remembered that the enterprising Sir Robert Wilson with 4,000 men, partly Spaniards, partly Portuguese of the Lusitanian Legion, had moved parallel with Wellesley’s northern flank during the advance to Talavera. On the day of the battle he had ‘marched to the cannon’ as a good officer should, and had actually approached Cazalegas, at the back of the French army, in the course of the afternoon. Learning of the results of the fight, he had turned back to his old path upon the twenty-ninth, and had entered Escalona on the upper Alberche. At this place he was behind Victor’s flank, and lay only thirty-eight miles from Madrid. There was no French force between him and the capital, and if only his division had been a little stronger he would have been justified in making a raid upon the city, relying for aid upon the insurrection that would indubitably have broken out the moment that he presented himself before its gates.

It was reported to Victor on the thirtieth not only that Wilson was at Escalona, but also that he was at the head of a strong Portuguese division, estimated at 8,000 or 10,000 men. The Marshal determined that he could not venture to leave such a force upon his rear while the armies of Wellesley and Cuesta were in his front, and fell back ten miles to Maqueda on the high road to Madrid. On the following day, still uneasy as to his position, he retired still further, to Santa Cruz, and wrote to King Joseph that he might be forced to continue his retreat as far as Mostoles, almost in the suburbs of Madrid [Aug. 2]. He was so badly informed as to the movements of the allies, that he not only warned the King that Wilson was threatening Madrid, but assured him that the British army from Talavera had broken up from its cantonments and was advancing along the Alberche towards the capital[695]. Joseph, better instructed as to the actual situation of affairs, replied by assuring him that Wellesley and Cuesta were far more likely to be retreating on Almaraz than marching on Madrid, as they must have heard ere now of Soult’s advance on Plasencia. He ordered the[p. 571] Marshal to fall back no further, and to send a division to feel for Wilson at Escalona. On detaching Villatte to execute this reconnaissance [Aug. 5] Victor was surprised to find that Sir Robert’s little force had already evacuated its advanced position, and had retreated into the mountains. For the last four days indeed Victor had been fighting with shadows—for the British and Estremaduran armies had never passed the Alberche, while Wilson had absconded from Escalona on receiving from Wellesley the news that Soult had been heard of at the Puerto de Ba?os. In consequence of the needless march of the 1st Corps to Maqueda and Santa Cruz, the allied generals were able to withdraw unmolested, and even unobserved, from Talavera, and were far upon their way down the Tagus before their absence was suspected. The erratic movements of Victor may be excused in part by the uniform difficulty in obtaining accurate information which the French always experienced in Spain. But even when this allowance is made, it must be confessed that his operations do not tend to give us any very high idea of his strategical ability. He was clearly one of those generals, of the class denounced by Napoleon, qui se font des tableaux, who argue on insufficient data, and take a long time to be convinced of the error of their original hypothesis.

Neither Victor nor King Joseph, therefore, exercised any influence over the doings of Wellesley and Cuesta at Talavera between the 29th of July and the 3rd of August. The allies worked out their plans undisturbed by any interference on the part of the old enemies whom they had beaten on the battle day. Down to August 1 the British general had been unconvinced by the rumours of Soult’s approach, at the head of a large army, which were persistently arriving from the secret agents in the direction of Salamanca[696]. It was only on the evening of that day that he received news so precise, and so threatening, that he found himself forced to abandon for the moment any intention of pushing on towards Madrid, in consequence of the impending attack on the line of his communications[p. 572] with Portugal. It was announced to him that the vanguard of the French army from the north had actually entered Bejar on the twenty-ninth and was driving in the trifling force under the Marquis Del Reino, which Cuesta had sent to the Puerto de Ba?os.

Whatever might be the force at Soult’s disposal—and Wellesley was still under the delusion that it amounted at most to a single corps of 12,000 or 15,000 men—it was impossible to allow the French to establish themselves between the British army and Portugal. If they were at Bejar on the twenty-ninth they might easily reach Plasencia on the thirty-first. On receiving the news Cuesta, who had hitherto shown the greatest reluctance to divide his army, detached his 5th division under Bassecourt, with orders to set out at the greatest possible speed, and join the Marquis Del Reino. This move was tardy and useless, for it is four long marches from Talavera to Plasencia, so that Bassecourt must arrive too late to hold the defiles. If he found the French already established on the river Alagon, his 5,000 men would be utterly inadequate to ‘contain’ double or triple that number of Soult’s troops. As a matter of fact the enemy had entered Plasencia on the afternoon of August 1, before the Spanish division had even commenced its movement to the west[697].

On the morning of August 2 Wellesley and Cuesta held a long and stormy conference. The Captain-General proposed that Wellesley should detach half his force to assist Bassecourt, and stay with the remainder at Talavera, in order to support the Army of Estremadura against any renewed attack by Victor and King Joseph[698]. The English commander refused to divide[p. 573] his force—he had only 18,000 effectives even after Craufurd had joined him, and such a small body would not bear division. But he offered either to march against Soult with his entire host, or to remain at Talavera if his colleague preferred to set out for Plasencia with his main body. Cuesta chose the former alternative, and on the morning of the third Wellesley moved out with every available man, intending to attack the enemy at the earliest opportunity. He was still under the impression that he would have to deal with no more than a single French corps, and was confident of the result. His only fear was that Victor might descend upon Talavera in his absence, and that Cuesta might evacuate the place on being attacked. If this should happen, the English hospitals, in which there lay nearly 5,000 wounded, might fall into the hands of the enemy. On halting at Oropesa he sent back a note to O’Donoju, the chief of the staff of the Estremaduran army, begging him to send off westward all the British wounded who were in a condition to travel. He asked that country carts might be requisitioned for their assistance, if no transport could be spared by the Spanish troops[699].

Wellesley was setting out with 18,000 men to attack not the mere 15,000 men that he believed to be in his front, but three whole corps d’armée, with a strength of 50,000 sabres and bayonets. In his long career there were many dangerous crises, but this was perhaps the most perilous of all. If he had remained for a little longer in ignorance of the real situation, he might have found himself involved in a contest in which defeat was certain and destruction highly probable.

The real situation in his front was as follows. On receiving the dispatch from Madrid which permitted him to execute his projected march upon Plasencia, Soult had begun to concentrate his army [July 24]. Mortier and the 5th Corps were already in march for Salamanca in pursuance of earlier orders: they arrived in its neighbourhood the same day on which Foy brought the King’s orders to his chief. The 2nd Corps was already massed upon the Tormes, and ready to move the moment that it should receive the supply of artillery which had been so long upon its way from Madrid. Ney and the[p. 574] 6th Corps from Benavente and Astorga had far to come: they only reached Salamanca on July 31; if we remember that the distance from Astorga to the concentration point was no less than ninety miles we cease to wonder at their tardy arrival.

Soult had strict orders from the Emperor to march with his troops well closed up, and not to risk the danger of being caught with his corps strung out at distances which would permit of their being met and defeated in detail[700]. He was therefore entirely justified in refusing to move until the 6th Corps should be in supporting distance of the rest of his army, and the 2nd Corps should have received the cannon which were needed to replace the pieces that they had lost in Portugal. For this reason we must regard as unfounded all the vehement reproaches heaped upon him by Joseph and Jourdan during the acrimonious correspondence that followed upon the end of the campaign. It would have been wrong to start the 5th Corps upon its way to Plasencia till the 2nd Corps was ready to follow, and the much needed guns only came into Salamanca on the twenty-ninth, though their approach had been reported on the preceding day.

We cannot therefore blame Soult for sloth or slackness when we find that he started Mortier upon his way on July 27, and followed him with his own corps upon July 30, the day after the guns arrived, and the day before Ney and his troops were due to reach Salamanca from the north.

The order of march was as follows: the vanguard was composed of the whole corps of Mortier, nearly 17,000 strong[701], reinforced by three brigades of dragoons under Lahoussaye and Lorges with a strength of 2,000 sabres. The 2nd Corps followed; though it started three days later than the 5th it was gradually gaining ground on the vanguard all through the march, as it had no fighting to do or reconnaissances to execute.[p. 575] Hence it was only twenty-four hours behind Mortier in arriving at Plasencia. Its strength was 18,000 men, even after it had detached the brigades of dragoons to strengthen the vanguard, and placed five battalions at the disposal of General Kellermann[702]. During its stay at Zamora and Toro it had picked up a mass of convalescents and details, who had not taken part in its Galician campaign. The rear was formed by Ney’s troops, which started from Salamanca only one day behind the 2nd Corps. The infantry was not complete, as a brigade of 3,000 men was left behind on the Douro, to assist Kellermann in holding down the kingdom of Leon. Hence, even including a brigade of Lorges’ dragoons, the 6th Corps had only some 12,500 men on the march. The whole army, therefore, as it will be seen, was about 50,000 strong.

Just before he marched from Salamanca Soult had heard that Beresford’s Portuguese were commencing to show themselves in force in the direction of Almeida, while Del Parque’s small division at Ciudad Rodrigo was beginning to be reinforced by troops descending from the mountains of Galicia. Trusting that the danger from this quarter might not prove imminent, the Marshal left in observation of the allies only the remains of the force that Kellermann had brought back from the Asturias—the 5th division of dragoons and a few battalions of infantry, strengthened by the five battalions from the 2nd Corps and the one brigade detached from Ney. The whole did not amount to more than 9,000 or 10,000 men, scattered along the whole front from Astorga to Salamanca. It was clear that much was risked in this direction, for Beresford and Del Parque could concentrate over 20,000 troops for an attack on any point that they might select. But Soult was prepared to accept the chances of war in the Douro valley, rightly thinking that if he could crush Wellesley’s army on the Tagus any losses in the north could easily be repaired. It would matter little if the[p. 576] Spaniards and Portuguese occupied Salamanca, or even Valladolid, after the British had been destroyed.

Mortier, starting on July 27, on the road by Fuente Roble and Los Santos, made two marches without coming in touch with any enemy. It was only on the third day that he met at La Calzada the vedettes of the trifling force under the Marquis Del Reino which Cuesta had sent to hold the Puerto de Ba?os. After chasing them through Bejar, the Marshal came upon their supports drawn up in the pass [July 30]. Del Reino thought himself obliged to fight, though he had but four battalions with a total of 2,500 or 3,000 bayonets[703]. He was of course dislodged with ease by the overwhelming numbers which Mortier turned against him—the first division of the 5th Corps alone sufficed to drive him through the pass. Thereupon he retired down the Alagon, and after sending news of his defeat to Cuesta fell back to Almaraz, where he took up the bridge of boats and removed it to the southern bank of the Tagus.

Having cleared the passes upon the thirtieth, the 5th Corps advanced to Candelaria and Ba?os de Bejar upon the thirty-first, and entered Plasencia on the first of August. Here Mortier captured 334 of Wellesley’s sick, who had been left behind as being incapable of removal. On the preceding day the town had been full of British detachments: the place was the half-way house between Portugal and Talavera, and many commissaries, isolated officers going to or from the front, and details marching to join their corps, had been collected there. Captain Pattison, the senior officer present, withdrew to Zarza, with every man that could march, when he heard of Mortier’s[p. 577] approach, taking with him a convoy which had recently arrived from Abrantes. But he was obliged to leave behind him a considerable amount of corn, just collected from the Vera, which had been destined for Wellesley’s army. The whole civil population of Plasencia fled to the hills, in obedience to an order of the local Junta, and the British soldiers in the hospital were the only living beings whom the French vanguard found in the city. The men of the 5th Corps plundered the deserted houses, as was but natural, but behaved with much humanity to the captured invalids[704].

After seizing Plasencia Mortier halted for a day, in obedience to Soult’s orders, that he might allow the 2nd Corps to close up before he pressed in any further towards Wellesley. The Duke of Dalmatia was determined to run no risks, when dealing with an adversary so enterprising as his old enemy of Oporto. On August 2 he himself and the leading divisions of his corps reached Plasencia: the rest were close behind. On the same afternoon, therefore, the advance could be resumed, and Mortier set out on the high road towards Almaraz and Talavera, having eight regiments of horse—3,000 men—in his front. He slept that night at Malpartida, seven miles in advance of Plasencia, and moved on next morning to the line of the Tietar and the village of Toril. One of his reconnoitring parties approached the bridge of Almaraz and found it broken: another reached Navalmoral. He was now drawing very close to Wellesley, who had encamped that day at Oropesa, and was only thirty miles away: indeed the British and the French cavalry came in contact that evening in front of Navalmoral.

On August 3, by a curious coincidence, each Commander-in-chief was at last informed of his adversary’s strength and intentions by a captured dispatch. A Spanish messenger was arrested by Soult’s cavalry, while bearing a letter from Wellesley to General Erskine dated August 1. In this document there was an account of the battle of Talavera, which had[p. 578] hitherto been unknown to Soult. But the most important clause of it was a request to Erskine to find out whether the rumours reporting the advance of 12,000 French towards the Puerto de Ba?os were correct. The Duke of Dalmatia thus discovered that his adversary, only two days before, was grossly underrating the numbers of the army that was marching against his rear. He was led on to hope that Wellesley would presently advance against him with inferior numbers, and court destruction by attacking the united 2nd and 5th Corps[705].

This indeed might have come to pass had not the allies on the same day become possessed of a French dispatch which revealed to them the real situation of affairs. Some guerrillas in the neighbourhood of Avila intercepted a friar, who was an agent of King Joseph, and was bearing a letter from him to Soult. They brought the paper to Cuesta on August 3: it contained not only an account of the King’s plans and projects, but orders for the Marshal, which mentioned Ney and the 6th Corps, and showed that the force marching on Plasencia was at least double the strength that Wellesley had expected[706]. This letter Cuesta sent on to his colleague with laudable promptness; it reached the British commander in time to save him from taking the irreparable step of marching from Oropesa to Navalmoral, where the vanguard of Mortier’s cavalry had just been met by the vedettes of Cotton’s light horse. Wellesley had actually written to Bassecourt to bid him halt at Centinello till he himself should arrive, and then to join him in an attack on the French[707], when he was handed the intercepted letter which showed that Soult had at least 30,000 men in hand.

[p. 579]

This unpalatable news changed the whole prospect of affairs: it would be mad to assail such an enemy with a force consisting of no more than 18,000 British troops and Bassecourt’s 5,000 Spaniards. Wellesley had therefore to reconsider the whole situation, and to dictate a new plan of campaign at very short notice, since his cavalry were actually in touch with the enemy at the distance of a single day’s march from Oropesa. On the morrow he must either fight or fly. The situation was made more complicated by the fact that Cuesta, when forwarding the French dispatch, had sent information to the effect that he considered his own situation at Talavera so much compromised that he was about to retreat at once, with the design of crossing the Tagus at Almaraz, and of taking up once more his old line of communications, which ran by Truxillo to Badajoz. It may be asked why the Captain-General did not adopt the simpler course of crossing the Tagus at Talavera, and moving under cover of the river, instead of executing the long flank march by Oropesa to Almaraz on the exposed bank, where the French were known to be in movement. The answer, however, is simple and conclusive: the paths which lead southward from Talavera are impracticable for artillery and wheeled vehicles. Infantry alone could have retreated by the route which climbs up to the Puerto de San Vincente, the main pass of this section of the Sierra de Guadalupe: nor was the track along the edge of the river from Talavera to Arzobispo any better fitted for the transport of a large army. It is this want of any adequate communication with the south which makes Talavera such a dangerous position: no retreat from it is possible save that by the road to Oropesa, unless the retiring army is prepared to sacrifice all its impedimenta.

Cuesta has been criticized in the most savage style by many English writers, from Lord Londonderry and Napier downwards, for his hasty departure from Talavera. It is fair to state in his defence the fact that if he had tarried any longer in his present position he might have been cut off not merely from Almaraz—that passage was already impracticable—but also from the bridge of Arzobispo, the only other crossing of the Tagus by which artillery and heavy wagons can pass southward. If he had started on the fourth instead of the third he[p. 580] might have found Mortier and Soult interposed between him and this last line of retreat. He would then have been forced to abandon all his matériel, and to hurry back to Talavera, in order to take the break-neck track to the Puerto de San Vincente. But there was every reason to believe that Victor might arrive in front of Talavera on the evening of the fourth or the morning of the fifth, so that this last road to safety might have been already blocked. Thus the Spanish army, if it had started on the fourth for Oropesa, might have found itself caught between the two French corps, and vowed to inevitable destruction. As a matter of fact Victor moved slowly and cautiously, and only reached Talavera on the sixth—but this could not possibly have been foreseen. We cannot therefore blame Cuesta’s precipitate departure upon the night of August 3.

His main body marched under cover of the darkness to Oropesa, where they arrived, much wearied and in some disorder, on the following morning. He left Zayas’s division and Albuquerque’s horse as a rearguard, to hold Talavera till midday on the fourth, with orders to make a semblance of resistance and to detain Victor for a few hours if he should appear. But no hostile force showed itself: by his unwise retreat to Santa Cruz the Marshal had drawn back so far from the enemy that he could not take advantage of their retrograde movement when it became known to him. Villatte’s division and Beaumont’s cavalry only reached Talavera on the morning of the sixth.

The departure of the Estremaduran army had one deplorable result. It exposed the English hospitals at Talavera, with their 4,000 wounded, to capture by the enemy. Wellesley, before he had marched off, had given orders that all the men capable of being moved should be sent off towards Plasencia and Portugal as soon as possible. But he had no transport that could cope with the task of transferring such a mass of invalids towards his base. He wrote from Oropesa begging Cuesta to requisition carts from the country-side for this purpose[708]. But it was notorious that carts were not to be had—all Wellesley’s letters for the last three weeks were full of complaints to the effect that he could not procure them by money[p. 581] or by force. When the Spaniards were themselves departing, bag and baggage, it was an inopportune moment at which to ask them to provide transport: yet since the British wounded had been left to their care they were bound in honour to do all that could be done to save them. It is said that Cuesta made over[709] no more than seven ox-carts and a few mules to Colonel Mackinnon, the officer charged with the task of evacuating the hospitals. These and about forty vehicles of various kinds belonging to the British themselves were all that could be procured for the use of the wounded. They could only accommodate a tithe of the serious cases: the men with hurts of less consequence were forced to set out upon their feet. ‘The road to Oropesa,’ writes one of their fellow sufferers, ‘was covered with our poor limping bloodless soldiers. On crutches or sticks, with blankets thrown over them, they hobbled woefully along. For the moment panic terror lent them a force inconsistent with their debility and their fresh wounds. Some died by the road, others, unable to get further than Oropesa, afterwards fell into the hands of the enemy[710].’ The rest trailed onward to the bridge of Arzobispo, where Wellesley provided transport for many of them by unloading baggage-wagons, and ultimately reached Truxillo, at which place the new hospitals were established. Of the whole 4,000 about 1,500 had been left at Talavera as hopeless or dangerous cases, and these became the captives of the French: 2,000 drifted in, at various times, to Truxillo: the remaining 500 expired by the wayside or were taken by the French in the villages where they had dropped down[711].

Long before Cuesta and his host had arrived at Oropesa,[p. 582] Wellesley had made up his mind that the only course open to him was to abandon the march towards Navalmoral and Almaraz, and to turn aside to the bridge of Arzobispo. As the French were known to be at Navalmoral, it would have been impossible to force a passage to Almaraz without a battle. If the enemy were to be estimated at two corps, or 30,000 men, according to the indications of the intercepted letter, they would probably be able to detain the Anglo-Spanish army till Victor should arrive from the rear. For, without accepting a pitched battle, they would be strong enough to harass and check the allies, and to prevent them from reaching Almaraz till the 1st Corps should come upon the scene. ‘I was not certain,’ wrote Wellesley to Beresford two days later, ‘that Ney was not with Soult: and I was certain that, if not with him, he was at no great distance. We should therefore have had a battle to fight in order to gain the road to Almaraz—Plasencia was then out of the question—and if Victor had followed Cuesta, as he ought to have done, another battle, probably, before the bridge could be re-established[712]. Then it was to be considered that, Cuesta having left Talavera, the bridge of Arzobispo would have been open to the enemy’s enterprise: if they had destroyed it, while we had failed in forcing Soult at Navalmoral, we were gone.’

It is impossible not to bow before Wellesley’s reasoning. The French critics object that only Mortier was at Navalmoral on August 4, Soult being twenty miles behind him at Bazagona on the Tietar, so that it would have been possible for the British army to have driven back the 19,000 men of the Duke of Treviso, and to have forced its way to Almaraz[713]. But even if Wellesley had fought a successful action with Mortier on August 4, Soult would certainly have joined his colleague on the fifth, before the bridge could have been repaired, or at any rate before the whole Anglo-Spanish army and all its impedimenta could have crossed the Tagus. If attacked during their passage by the 37,000 men of the 2nd and 5th Corps they[p. 583] would have fared badly. Wellesley was perfectly correct in his decision; indeed the only point in which he was deceived was that he believed the enemy in his front to be Soult’s and Ney’s Corps, whereas they were in reality those of Soult and Mortier. Ney only reached Plasencia on August 4, and did not join the main body of the army till two days later.

When Wellesley and Cuesta met at Oropesa, early on the morning of August 4, they found themselves as usual engaged in a heated controversy. The British general had directed his divisions to hold themselves ready to march on the bridge of Arzobispo without further delay. Cuesta on the other hand had been attacked by a recrudescence of his old disease, the mania for fighting pitched battles[714]. He proposed that the allied armies should remain on the north bank of the Tagus, adopt a good defensive position, and defy Soult to attack them. Wellesley would not listen for a moment to this project, and finally declared that in spite of all arguments to the contrary, he should cross the Tagus that day at the head of his army. The two generals parted in wrath, and at six o’clock the British commenced their march to Arzobispo, only nine miles distant; the whole force crossed its bridge before evening, and established itself in bivouac on the south side of the river.

Cuesta remained at Oropesa for the whole day of August 4, and was there joined both by Bassecourt, who had fallen back from Centinello, and by Zayas and Albuquerque, who had evacuated Talavera at noon and made a forced march to join their chief. He appeared disposed to fight even though his ally had abandoned him. In the afternoon Mortier’s cavalry pressed in against him. He turned fiercely upon them, deployed a whole division of infantry and 1,200 horse in their front, and drove them back towards their supports. This vigorous action had a result that could not have been foreseen: Mortier jumped to the conclusion that he was himself about to be attacked by the whole Spanish army—perhaps by Wellesley also[715]. He[p. 584] halted the 5th Corps in advance of Navalmoral, and wrote to implore Soult to come up to his aid without delay. The Duke of Dalmatia hurried up with all speed, and on August 5 brought the 2nd Corps to Casatejada, only six miles in the rear of his colleague. Ney, following with a like promptness, advanced that day to Malpartida, a march behind the position of Soult.

On the sixth, therefore, the whole army from the Douro was practically concentrated, and Soult and Mortier advanced against Cuesta with Ney close in their rear. They found that they were too late: after remaining in battle order in front of the bridge of Arzobispo during the whole of the fifth, courting the attack which Mortier had been too cautious to deliver, the Captain-General had crossed the Tagus that night, and had occupied its further bank. He had left in front of the bridge only a small rearguard, which retired after a skirmish with the advanced cavalry of the 5th Corps. For once Cuesta had found luck upon his side; if Mortier had ventured to assail him on the fifth, and had forced him to an engagement, in a position from which retreat was difficult, and with the Tagus at his back, his situation would have been most perilous. For even if he had kept the 5th Corps at bay, he could not easily have withdrawn in face of it, and Soult would have been upon him on the next morning. In escaping across the narrow bridge of Arzobispo his losses must have been terrible: indeed the greater part of his army might have been destroyed.

Finding, on the evening of August 6, that both the British and the Estremaduran armies were now covered by the Tagus, whose line they appeared determined to defend, Soult was forced to think out a new plan of campaign. His original design of taking the allies in the rear and cutting off their retreat had miscarried: he must now either halt and recognize that his march had failed in its main purpose, or else deliver a frontal attack upon the line of the Tagus. The bridge of Almaraz was broken, and troops (the detachment of the Marquis Del Reino) were visible behind it. The bridge of Arzobispo was not destroyed, but the Spaniards were obviously ready to defend it. It was barricaded, the mediaeval towers in its midst were manned by a detachment of infantry, and a battery for twelve guns had been placed in an earthwork erected on a knoll thirty yards in[p. 585] its rear, so as to sweep all the approaches. Considerable forces both of cavalry and of infantry were visible on the hillsides and in the villages of the southern bank. Cuesta, in fact, while proposing to fall back with his main body to Meza de Ibor and Deleytosa, in order to recover his communication with his base at Badajoz, had left behind a strong rearguard, consisting of Bassecourt’s infantry division and Albuquerque’s six regiments of cavalry, a force of 5,000 bayonets and nearly 3,000 sabres. They were ordered to defend the bridge and the neighbouring ford of Azutan till further orders should reach them. The ground was very strong; indeed the ford was the one perilous point, and as that passage was narrow and hard to find, Cuesta trusted that it might be maintained even against very superior numbers. So formidable did the defence appear that Soult halted during the whole day of August 7, while he took stock of the Spanish positions, and sought up-stream and down-stream for means of passage other than the bridge. He was not at first aware of the existence of the ford: it was only revealed to him by the imprudence of the Spanish cavalry, who rode their horses far into the stream when watering them, thus showing that there were long shallows projecting from the southern bank. By a careful search at night the French intelligence-officers discovered that the river was only deep for a few yards under their own bank[716]: for the rest of its breadth there were only two or three feet of water. Having found the point, not far from the bridge, where the more dangerous part of the channel was fordable, they advised the Marshal that the passage of the river would present no insurmountable difficulties. Soult resolved to deliver an assault both on the bridge and on the ford upon the morning of August 8. Nor was it only at Arzobispo that he determined to force the line of the Tagus. He directed Ney, who was bringing up his rear at the head of the 6th Corps, to turn aside to the broken bridge of[p. 586] Almaraz, and to endeavour to cross the river by aid of a ford which was said to exist in that neighbourhood. Sketch-maps were sent to the Marshal in order to enable him to locate the exact point of passage—it would seem that they must have been very faulty.

Meanwhile Wellesley had passed the Tagus four days and Cuesta three days before the Marshal’s attack was ready, and both had been granted time to proceed far upon their way. It was fortunate that they were not hurried, for the road from Arzobispo to Meza de Ibor and thence to Deleytosa and Jaraicejo, though passable for guns and wheeled vehicles, was steep and in a deplorable condition of disrepair. It took Wellesley two days to march from the bridge to Meza de Ibor, a distance of only seventeen miles, because of the endless trouble caused by his artillery. There were places where he had practically to remake the roadway, and others where whole companies of infantry had to be turned on to haul the cannon up slopes where the half-starved horses could make no headway. These exertions were all the more exhausting because the men were falling into a state of great bodily weakness from insufficient supplies. Even at Talavera they had on many days received no more than half rations: but after passing Oropesa regular distributions of food ceased altogether for some time: there were still a few slaughter-oxen with the army, but bread or biscuit was unobtainable, and the troops had to maintain themselves on what they could scrape up from the thinly peopled and rugged country-side. A diet of overripe garbanzos, parched to the hardness of bullets, was all that many could obtain. Better foragers eked them out with honeycomb stolen from the peasants’ hives, and pork got by shooting the half-wild pigs which roam in troops among the woods on the mountain side. Many, in the ravenous eagerness of hunger, ate the meat warm and raw, and contracted choleraic complaints from their unwholesome feeding[717].

Divining that Soult would probably make a dash at Almaraz as well as at Arzobispo, Wellesley sent on ahead of his main body the brigade of Robert Craufurd, to which he attached Donkin’s[p. 587] much depleted regiments, in order to make up a small division. As they were unhampered by guns or baggage this detachment reached Almaraz on the sixth, after a fifteen hours’ forced march on the preceding day. They took over charge of the broken bridge and the ford from the Spanish troops of the Marquis Del Reino, and proceeded to entrench themselves in the excellent positions overlooking the point where the river was passable. Thus Ney, when he reached Almaraz on the following day, found the enemy already established opposite him, and ready to dispute the crossing. About 4,000 British troops and 1,500 Spanish troops were holding the river bank: immediately at their backs was the narrow and eminently defensible defile of Mirabete, which completely commands the road to Truxillo: it was an even stronger position than that which covered the ford and the ruined bridge.

On August 7 therefore Wellesley considered himself in a comparatively satisfactory situation. The passage at Almaraz was held by a vanguard consisting of the best troops in the army. Two divisions, the cavalry, and all the guns had traversed the worst part of the road, and had reached Deleytosa, only nine miles behind Craufurd’s position. If the French should attack on the following day, the main body could reinforce the light brigade in a few hours. One division, in the rear, was holding the position of Meza de Ibor, which Wellesley did not wish to evacuate until the Spanish army was ready to occupy it. He had discovered that there were points between Arzobispo and Almaraz where the passage of the Tagus was not wholly impracticable for small bodies of infantry[718], and dreaded that the enemy might throw a detachment across the stream to make a dash for the Meza. If this position had been lost the communication between the two armies would have been broken.

Cuesta, meanwhile, was engaged in the steep and stony mountain road over which Wellesley had toiled on the 5th and the 6th of August. His vanguard was now close to Meza de Ibor: the rest of the army was strung out between that point and Val de la Casa: the Captain-General himself had his head quarters on the night of the seventh at Peraleda de Garbin, ten miles[p. 588] west of Arzobispo. Bassecourt and Albuquerque were still covering the rear, with Mortier’s corps now plainly visible in their front. On their steadiness depended the safety of the whole army, for Cuesta had more baggage and more guns[719] than Wellesley, and therefore the road over the hills was even more trying to him than to his colleague. There was a congestion of wheeled transport at certain spots on the road which created hopeless confusion, and barred the march of the cavalry and even of the infantry divisions. It was only removed by setting whole battalions to work to drag the wagons out of the way. Cuesta’s ultimate destination was the Meza de Ibor, a position of unparalleled strength, which could be held even after the enemy had crossed the Tagus. That they would ultimately win their way over the river was certain, for already news had arrived that Victor, after reaching Talavera on Aug. 6, had pushed infantry over its bridge on the road to Herencia and Aldea Nueva. Troops coming from this direction would outflank the Arzobispo position, and compel Albuquerque to abandon it. Even without cavalry or guns this detachment of the 1st Corps would be strong enough to dislodge the guard of the bridge, by falling upon its rear, while Mortier was attacking it in front. As the cavalry of Victor and Soult had met, half way between Oropesa and Talavera, upon the afternoon of the seventh, the two marshals were now in full communication, and able to concert any plans that they might please for joint operations.

The Duke of Dalmatia, however, preferred to win all the credit for himself, and attacked without allowing his colleague’s troops time to approach the Spanish position. It was fortunate for Albuquerque that the rivalry of the two hostile commanders saved him from the joint assault, which would have been far more ruinous to him than the actual combat of Aug. 8 was destined to prove.

Having full knowledge of the existence and the locality of the ford of Azutan, Soult had resolved to launch his main attack upon this point, while directing only a subsidiary attack upon the fortified bridge. This last was only to be pushed[p. 589] home in case the troops sent against the ford should succeed in making good their footing upon the further bank. A careful observation of the Spanish lines showed that both Albuquerque and Bassecourt were holding back the main body of their divisions at some distance from the water’s edge, in the groves around the three villages of Pedrosa, Burgillo, and Azutan. There was only a single regiment of cavalry watching the river bank, and two or three battalions of infantry manning the towers of the bridge of Arzobispo and the redoubt in its rear. The Spaniards showed every sign of a blind confidence in the strength of their position behind the broad but shallow Tagus.

Knowing their habits, Soult selected for the moment of his attack the hour of the siesta. It was between one and two o’clock in the afternoon when he bade his columns, which had been drawn up under cover, and at some distance from the water’s edge, to advance to force the passage. For the assault upon the ford he had collected the whole of his cavalry, no less than twelve regiments. Lahoussaye’s dragoons formed the van, then came Lorges’ brigade, then the division of light horse belonging to the 2nd Corps, in the rear the corps-cavalry of Mortier. This mass of 4,000 horsemen was to be followed by the first brigade of Girard’s infantry division of the 5th Corps, while its second brigade was to assault the bridge, when Lahoussaye and Lorges should have won the passage of the ford and have established themselves on the flank of the Spanish defences. Gazan’s division, the second of the 5th Corps, was to support Girard, while the masses of the infantry of the 2nd Corps remained in reserve. All the light artillery of the army was to gallop down to the water’s edge at various selected points, when the attacking columns were first put in movement, and to distract the attention of the enemy’s guns so far as lay in their power.

At about 1.30 P.M. Caulaincourt’s brigade of Lahoussaye’s dragoons, a force of about 600 sabres, sallied out from its cover behind the village of Arzobispo, and moved down to the ford at a sharp trot. It plunged into the water, had passed the deeper part of the channel almost before the Spaniards had guessed its intention, and soon reached the shallows on the opposite bank. The only hostile force ready to meet it[p. 590] was a single regiment (the 1st Estremaduran Hussars) which was watching the ford, and a battalion of infantry which Bassecourt sent down in haste from the redoubt behind the bridge. A fierce charge of Caulaincourt’s dragoons dispersed and routed the Spanish horse; after they had been driven off the victors turned upon the battalion, which tried to form square on their approach, but was late in finishing its man?uvre. It was assailed before the rear side had been formed, broken up, and cut to pieces.

Soult had thus gained a precious half-hour, during which the remainder of his cavalry, squadron after squadron, came pouring over the ford, and began to form up on the southern bank. When several regiments had passed he also let loose the infantry brigade which was to attack the bridge. So narrow was the approach that only a single battalion (the 1st of the 40th of the line) could deliver the assault. But the tirailleur companies of several other battalions, and two batteries of horse artillery, opened a lateral fire from various points of the northern bank, to distract the Spaniards from the frontal attack. The fraction of Bassecourt’s division which was in position at the bridge and the redoubt had already been completely cowed by seeing Lahoussaye’s cavalry forming up in their flank and rear. If they waited to resist the infantry attack, it was clear that they would be cut off from their sole line of retreat by the dragoons. They abandoned their positions after firing a couple of scattering volleys, and fled eastward along the river bank towards the village of Azutan. The heavy guns in the redoubt were left behind, and fell into the hands of Caulaincourt. Girard’s infantry was therefore able to cross the river almost without loss, two regiments at the bridge, two at the ford which the cavalry had already utilized. A few men were drowned in the second column, having strayed into deep water by swerving to the right or left of the proper route.

Meanwhile Albuquerque’s horse and Bassecourt’s second brigade, roused from their ill-timed siesta, were pouring out of the villages which had sheltered them from the noontide heat. The infantry—four battalions apparently—drew up beside a wood, on the slope a mile above the bridge, and waited to be[p. 591] attacked. The cavalry, however, came on in one great mass, and charged down upon Lahoussaye’s division, which was covering the deployment of the rest of the French horse. Albuquerque’s only thought was to engage the enemy before he had succeeded in passing the whole of his squadrons over the ford. Vainly hoping to atone for his previous slackness by haste that came too late, he had hurried his five regiments forward as soon as the men could saddle and bridle their horses. Fractions of the different corps were mixed together, and no proper first or second line had been formed. The whole mass—some 2,500 sabres—in great disorder, galloped down upon the two brigades of Lahoussaye, and engaged them for a short time. But Lorges’ dragoons and part of Soult’s light horse were now at hand to aid the leading division; the Spaniards were beset in flank as well as in front, and broke after the first shock. Albuquerque, who showed plenty of useless personal courage, tried in vain to rally them on the 2nd Estremaduran Hussars, the only regiment which remained intact. It was borne away by the backrush of the rest, and scattering over the hillsides the whole body fled westward and northward, some towards Peraleda de Garbin, others towards Pedrosa. Bassecourt’s infantry went off to the rear as soon as they saw their comrades routed, and took to the hills. By keeping to rocky ground they suffered comparatively little loss.

The French urged the pursuit of Albuquerque’s fugitive horsemen for many miles, chasing them as far as the defile of La Estrella in the Sierra de Guadalupe in one direction, and beyond Val de la Casa in the other. On the latter road the chase only ceased when the dragoons came upon the divisions of Henestrosa and Zayas, from Cuesta’s main army, drawn up across their path. The losses of the Spaniards were very considerable—600 men and 400 horses were captured, and over 800 killed and wounded. One flag was taken, that of the regiment cut to pieces by Lahoussaye’s dragoons at the commencement of the fighting. The pieces in the redoubt, and the divisional battery of Albuquerque, 16 guns in all, were lost. By an additional mischance the French also recovered fourteen of their own seventeen guns that had been taken at Talavera. Cuesta had not been able to utilize these pieces for want of[p. 592] gunners: they were trailing along in the rear of his army, very indifferently horsed, when the French dragoons swept along the road to Peraleda. On the approach of the pursuers they were abandoned by the wayside. This capture enabled Soult to assert that he had taken in all 30 cannon, and emboldened Sebastiani, a few weeks later, to declare that he had never lost his guns at Talavera[720]. Having recovered them he could exhibit them—all save two or three—in evidence of his mendacious statement.

Soult declared in his official report that his cavalry had lost only 28 killed and 83 wounded, his artillery 4 wounded, his infantry hardly a man, save some few drowned at the ford.

The rout of the Spanish rearguard and the capture of the bridge of Arzobispo gave Soult a foothold on the southern bank of the Tagus, but little more. The road by which he could now advance against the allies was detestable—we have already seen how its cliffs and ravines had tried the British and the Estremaduran armies. To reach Cuesta’s new position on the Meza de Ibor the Duke of Dalmatia would have had to make a two days’ march through these defiles, dragging his guns with him. His cavalry he would have been forced to leave behind him, as there would have been no means of employing it in the mountains. Meanwhile Wellesley had established himself in the ground which he had selected behind the broken bridge of Almaraz, and Cuesta had got the whole of his infantry and half his artillery over the Ibor stream and arrayed them on the Meza, where the rocky slopes are impregnable against a frontal attack, if the defending army shows ordinary determination[721].[p. 593] All through the ninth and the morning of the tenth the Spaniards were dragging the rest of their guns and their baggage up the steep zigzag path between the river and the summit of the plateau, and it was not till the end of the latter day that everything was in position. It is probable therefore that if Soult had pressed his pursuit with all possible speed, he might have captured some of the Spanish impedimenta on the morning of the tenth. But there were defiles between Peraleda and the Ibor river where Cuesta’s rearguard might possibly have detained him till the guns and baggage were in safety[722].

The Duke of Dalmatia, however, paused at the bridge of Arzobispo before committing himself to a second advance against the allies. He was averse to making an isolated attack upon the admirable position now occupied by the Estremaduran army, and wished to combine it with a simultaneous assault upon the British. It will be remembered that he had detached Ney’s corps from the rear of his line of march, and ordered it to attempt the passage of the Tagus at Almaraz, by the ford which he knew to exist close to the ruined bridge. He also wrote to Victor to desire him to push forward the two infantry divisions which had crossed the river at Talavera, and to direct them on Mohedas and Alia, so as to turn Cuesta’s flank by a long circuitous march among the rugged summits of the Sierra de Guadalupe.

Neither of these subsidiary movements was carried out. One division of Ney’s corps, and Fournier’s brigade of dragoons reached Almaraz on Aug. 8: the other division and the light cavalry had followed the 2nd Corps so closely that it had passed Navalmoral on its way eastward, and had to make a long counter-march. It was not till the ninth or tenth therefore that[p. 594] the Duke of Elchingen would have been in a position to attempt the passage of the Tagus. Craufurd’s detachment had been established at Mirabete, behind the broken bridge, since Aug. 6, and two days later the main body of the British army had reached Deleytosa, where it was within a few hours’ march of the vanguard, and perfectly ready to support it. If Ney had endeavoured to pass the Tagus on the ninth or tenth with his 12,500 men, it is clear that the head of his column must have been destroyed, for the ford was narrow and difficult, and indeed barely passable for infantry even in the middle of August[723]. But the Marshal did not even attempt the passage, for the simple reason that his intelligence officers failed to discover the ford, and reported to him that none existed. He sent word to Soult that the scheme was impracticable, and drawing back from the water’s edge concentrated his whole corps at Navalmoral [Aug. 9].

Victor, at the other end of the French line, showed no desire to adventure his infantry among the defiles of the Sierra de Guadalupe, without guns or cavalry, and refused to move up into the mountains in order to turn Cuesta’s right flank. Thus the whole plan concerted by the Duke of Dalmatia for a general attack on the allies came to an ignominious conclusion.

It would appear, indeed, that his chance of inflicting a serious blow on the enemy had passed away long ere he brought the 2nd and 5th Corps down to the bridge of Arzobispo. It was on the fifth, when Mortier refused to close with Cuesta and allowed him to withdraw across the Tagus, that Soult had lost his best opportunity. On that day the Spaniards were still on the[p. 595] wrong side of the river, and the British vanguard had not yet reached the broken bridge of Almaraz. If Mortier had engaged the army of Cuesta, and Ney had found and attacked the ford at Almaraz before Craufurd’s arrival, the position of the allies would have been forlorn indeed. But on the fifth Soult had not yet discovered the real position of affairs; and the head of Ney’s corps was only just debouching from Plasencia, two long marches from Almaraz. In short ‘the fog of war,’ as a modern writer has happily called it, was still lying thick about the combatants, and Soult’s best chance was gone before he was even aware of it.

On August 9, matters looked far less promising, even though the bridge of Arzobispo had been won. Since Ney sent word that he could not cross at Almaraz, while Victor declined to commit himself to any schemes for an advance into the eastern mountains, Soult saw that he must construct another scheme of operations. His own preference was for a march into Portugal by way of Coria and Castello Branco. Such an attack upon Wellesley’s base, made by the 50,000 men of the 2nd, 5th, and 6th Corps, would compel the British to abandon Almaraz, to give up their connexion with Cuesta, and to march in haste by Truxillo, Caceres, and Portalegre on Abrantes, in order to cover Lisbon. It was even possible that, if the invading army made great haste, it might reach Abrantes before the British: in that case Wellesley would be forced to keep to the southern bank of the Tagus and cross it at Santarem, comparatively close to the capital. Thus all Central Portugal might be won without a battle, and Lisbon itself might fall ere the campaign ended, since the 20,000 men of the British general, even when aided by the local levies, could not (as Soult supposed) hold back three French corps d’armée[724]. There was another alternative[p. 596] possible—to march not on Lisbon but on Ciudad Rodrigo and Almeida, and to invade Portugal by the northern road. But this plan would take a longer time to execute, and promised less decisive results.

But even before the combat of Arzobispo had taken place, Joseph and Jourdan had determined that they would not permit Soult to carry out any schemes of advance against Portugal. They could show very good grounds for their decision. If the Duke of Dalmatia marched off to attack Lisbon, he would leave the 1st and 4th Corps and the King’s reserve,—less than 50,000 men in all, after the losses of Talavera,—opposed to Cuesta, Wellesley, and Venegas, who between them would have at least 75,000[725]. If the British army should refuse to be drawn away towards Portugal, and should recross the Tagus at Almaraz with Cuesta in its wake, the situation would be deplorable. Victor would be exposed, just as he had been on July 22 and 23, to a joint attack from the two armies. And on this occasion Sebastiani and the King would not be able to bring him help, for they were now closely engaged with Venegas near Aranjuez. If they moved away from the front of the army of La Mancha, Madrid would be lost in two days. If they did not so move, Wellesley and Cuesta might crush Victor, or drive him away on some eccentric line of retreat which would uncover the capital. Jourdan therefore, writing in the name of Joseph, had informed Soult in a dispatch dated Aug. 8, that it was impossible to permit him to march on Portugal, as his departure would uncover Madrid and probably bring about a fatal disaster. He also urged that the exhaustion of the troops rendered a halt necessary, and that it would be impossible to feed them, if they advanced into the stony wilderness on the borders of Portugal before they had collected magazines. For the present the King would be contented to keep the allies in check, without seeking to attack or disperse them, until the weather began to grow cooler and the troops had rested from their fatigues.
Map of the campaign of Talavera

Enlarge  THE CAMPAIGN OF TALAVERA
JULY-AUGUST 1809

As if intending to put it out of Soult’s power to undertake his projected expedition into Portugal, Jourdan and Joseph [p. 597]now proceeded to deprive him of the control of one of his three army corps. They authorized Ney to recross the mountains and to return to Salamanca, in order to protect the plains of Leon from the incursions of the Spaniards of Galicia. Deprived of such a large section of his army, Soult would be unable to march against Abrantes, as he so much desired to do. There were good military reasons, too, for sending off Ney in this direction: Kellermann kept reporting that La Romana was on the move, and that unless promptly succoured he should find himself obliged to abandon Benavente and Zamora and to fall back on Valladolid. The Spaniards from Ciudad Rodrigo had already taken the offensive, and Del Parque’s advanced guard had even seized Salamanca.

Ney accepted with alacrity the chance of withdrawing himself from the immediate control of his old enemy Soult; he received his permission to return to Leon on Aug. 9: on the tenth his whole corps was on the move, and on the eleventh he had retired to Plasencia. On the following day he plunged into the passes and made for Salamanca with all possible speed[726].

While the 6th Corps was dispatched to the north, the King directed Soult to take up, with the rest of his troops, a defensive position opposite the allied armies on the central Tagus. The 2nd Corps was to occupy Plasencia, the 5th to watch the passages at Almaraz and Arzobispo, while keeping a detachment at Talavera. Thus all Soult’s plans for an active campaign were shattered, and he was told off to act as a ‘containing force.’ Meanwhile Joseph drew Victor and the 1st Corps away from Talavera, towards Toledo and La Mancha, with the intention of bringing them into play against Venegas. For just as[p. 598] Soult had always ‘an eye on Portugal,’ so Joseph had always ‘an eye on Madrid.’ He could not feel secure so long as a Spanish army lay near Toledo or Aranjuez, only two marches from the gates of his capital, and was determined to dislodge it from this threatening position before taking any other operation in hand. He had accepted as true rumours to the effect that part of Cuesta’s troops had retired in the direction of Oca?a[727] to join the army of La Mancha, and even that 6,000 British[728] had been detached in this same direction. Thus he had persuaded himself that Venegas had 40,000 men, and was desirous of drawing in Victor to his head quarters before delivering his attack, thinking that Sebastiani and the central reserve would be too weak for the task.