SECTION XVI: CHAPTER V

CONCENTRATION OF THE FRENCH ARMIES: THE KING TAKES THE OFFENSIVE: COMBATS OF TORRIJOS AND CASA DE SALINAS

It is now necessary to turn to the French camp, in order to realize the course of events which had led to the concentration of such a formidable force in the environs of Toledo. Down to the twenty-second of July Joseph and his adviser Jourdan had remained in complete ignorance of the advance of Wellesley upon Plasencia, and seem to have been perfectly free from any apprehension that Madrid was in danger. Since their return from their fruitless pursuit of the army of La Mancha, they had been spending most of their energy in a controversy with Soult. The Duke of Dalmatia, not content with the command of the three army corps which Napoleon had put at his disposal, had been penning elaborate dispatches to the King to demand that the greater part of the remaining French troops in Spain should be used to co-operate in his projected campaign against the English in Portugal. He wrote on July 13 to urge on Joseph the necessity (1) of drawing large detachments from the armies of Aragon and Catalonia, in order to form a corps of observation in the kingdom of Leon to support his own rear; (2) of placing another strong detachment at Plasencia to cover his flank; (3) of transferring every regiment that could be spared from Madrid and New Castile to Salvatierra on the Tormes, just south of Salamanca, in order to form a reserve close in his rear, which he might call up, if necessary, to strengthen the 60,000 men whom he already had in hand. He also demanded that Joseph should send him at once 200,000 francs to spend on the fortification of Zamora, Toro, and other places on the Douro, as also 500,000 francs more for the present expenses of the 2nd,[p. 495] 5th, and 6th Corps. If this were granted him, together with 2,000,000 rations of flour, and a battering-train of at least forty-eight heavy guns for the sieges of Ciudad Rodrigo and Almeida, he thought that he should be in a position to deliver a serious attack on Northern Portugal, and ultimately to drive the British army into the sea[622].

On the day upon which the Duke of Dalmatia made these comprehensive demands upon King Joseph, the British army had been for ten days in Spain, and was preparing to advance from Plasencia on Madrid. It was therefore an exquisitely inappropriate moment at which to demand that the greater part of the King’s central reserve should be sent off from the capital to the neighbourhood of Salamanca. There were other parts of Soult’s lists of requisitions which were equally impracticable. It is clear that Suchet could not have spared a man from Aragon, and that St. Cyr, with the siege of Gerona on his hands, would have found it absolutely impossible to make large detachments from Catalonia. Even if he and Suchet had been able to send off troops to Leon, they would have taken months to reach the Galician frontier. The demand for 700,000 francs in hard cash was also most unpalatable: King Joseph was at this moment in the direst straits for money: his brother could send him nothing while the Austrian war was in progress, and as he was not in proper military possession of any large district of Spain, he was at this moment in a condition of hopeless bankruptcy. He confessed to Soult that he was living from hand to mouth, by the pitiful expedient of melting down and coining the silver plate in the royal palace at Madrid.

Jourdan therefore replied, in the King’s behalf, to Soult that he must do his best with the 60,000 men already at his disposition, that no troops from Catalonia, Aragon, or Madrid could be spared, and that money could not be found. All that could be given was the battering-train that had been demanded, 600,000 rations of biscuit, and an authorization to raise forced contributions in Old Castile. For the protection of his flanks and his communications the Marshal must utilize Kellermann’s dragoons and the other unattached troops in the valley of the[p. 496] Douro, a force which if raised to 12,000 men by detachments from the 5th or 6th Corps could keep La Romana and the Galicians in check[623].

It is curious to note how entirely ignorant both Soult and the King were as to the real dangers of the moment. Soult had drawn up, and Joseph acceded to[624], a plan for the siege of Ciudad Rodrigo, and an invasion of Northern Portugal—operations which would take long weeks of preparation—at the time when Madrid was in imminent danger from the combined armies of Wellesley, Cuesta, and Venegas. The Marshal’s plan was perfectly correct from the point of view of the higher strategy—the main objective of the French was certainly the British army, and it would have been highly advisable to invade Northern Portugal with 60,000 men in the front line, and 40,000 in support, if the circumstances of the moment had permitted it. But these circumstances were hidden alike from Soult and the King, owing to the impossibility of obtaining accurate information of the movements of the allies. The fundamental difficulty of all French operations in the Peninsula was that the commanders could never discover the whereabouts of the enemy till he actually came in contact with their outposts. Hence it chanced that Soult was planning, and Joseph approving, a campaign on the borders of Northern Portugal, at the precise moment when the British were on the march for Talavera.

It was actually not until July 22 that the King’s eyes were at last unsealed. Victor having come into collision with the cavalry of Wellesley’s advanced guard, sent news to Madrid that the British army had joined Cuesta, and had reached the Alberche. On the same day, by a fortunate chance, there also arrived in the capital another emissary of Soult, with a message much less impracticable than that which had last been sent. This was General Foy, whom the Duke of Dalmatia had dispatched on July 19, after receiving very definite rumours that the British were moving in the valley of the Tagus, and[p. 497] not approaching Old Castile[625]. The Marshal sent word that in this case he must of course concert a common plan of operations with the King, and abandon any immediate action against Portugal. He suggested that his best plan would be to concentrate his three corps at Salamanca, and to march against the flank and rear of the English by way of Bejar and the Puerto de Ba?os. If the King could cover Madrid for a time with the 1st and 4th Corps, he would undertake to present himself in force upon Wellesley’s line of communications, a move which must infallibly stop the advance of the allies towards the capital. If they hesitated a moment after his arrival at Plasencia, they would be caught between two fires, and might be not merely checked but surrounded and destroyed. Soult added, however, that he could not move till the 2nd Corps had received the long-promised provision of artillery which was on its way from Madrid, and till he had rallied Ney’s troops, who were still at Astorga, close to the foot of the Galician mountains.

Napoleon, at a later date, criticized this plan severely, declaring that Soult ought to have marched on Madrid to join the King, and not on Plasencia. He grounded his objections to the scheme on the strategical principle that combined operations on external lines should be avoided. ‘The march of Marshal Soult,’ he wrote, ‘was both dangerous and useless—dangerous, because the other army might be beaten (as happened at Talavera) before he could succour it, so that the safety of all my armies in Spain was compromised: useless, because the English had nothing to fear; they could get behind the Tagus in three hours; and whether they crossed at Talavera or at Almaraz, or anywhere else, they could secure a safe line of retreat on Badajoz.’ Against this criticism the defence made by both Soult and King Joseph was that it would have required a much longer time to bring the three corps from the Douro to Madrid than to Plasencia; that it would have taken them at least ten days to reach Madrid, and that during those days the King and his army might have been beaten and driven out of the capital by the united forces of Wellesley, Cuesta, and Venegas. It was, of course, impossible to foresee on[p. 498] July 22 that Wellesley would refuse to pursue Victor beyond Talavera, or that Venegas would let Sebastiani slip away from him. Accordingly King Joseph and Jourdan fell in with Soult’s suggestion, because they thought that he would come sooner into the field if he marched on Plasencia, and would remove the pressure of the British army from them at a comparatively early date. As a matter of fact, he took a much longer time to reach Plasencia than they had expected: they had hoped that he might be there on July 27, while his vanguard only reached the place on August 1, and his main body on the second and third[626]. But it seems clear that the expectation that he would intervene on the earlier date was far too sanguine. Soult dared not move till his three corps were well closed up, and since Ney had to come all the way from Astorga, it would have been impossible in any case to mass the army at Plasencia much earlier than was actually done. Napoleon’s remark that Soult could not hope to catch or surround the British army seems more convincing than his criticism of the march on Plasencia. If the passes of the Sierra de Gata had been properly held, and prompt news had been transmitted to Talavera that the French were on the move from the valley of the Douro, Wellesley would have had ample time to cover himself, by crossing the Tagus and transferring his army to the line of operations, Truxillo-Badajoz. The British general always defended himself by this plea: and complained that those who spoke of him as being ‘cut off from Portugal,’ by the arrival of Soult at Plasencia, forgot that he had as good a base at Elvas and Badajoz as at Abrantes.

But we must not look too far forward into the later stages of the campaign. It is enough to say that Jourdan and Joseph sent back Foy to rejoin Soult, on the same day that he had reached Madrid, bearing the orders that the Marshal was to collect his three corps with the greatest possible haste, and to march by Salamanca on Plasencia, where they trusted that he might present himself on the twenty-seventh or twenty-eighth of the current month. Meanwhile it was necessary to hold[p. 499] back Cuesta and Wellesley till the Duke of Dalmatia’s operations in their rear began to produce their effect. The only possible way of doing this was to concentrate in all haste every available man in New Castile, and to cover Madrid as long as possible. This massing of the French forces turned out to be perfectly feasible, since Venegas had neglected to press in upon Sebastiani, so that it was possible to withdraw the whole 4th Corps from in front of him, and to send it to reinforce Victor, without any immediate danger. Accordingly, the 1st Corps was directed to fall back from its perilous advanced position on the Alberche, and to draw near to Toledo: Sebastiani was told to abandon Madridejos and La Mancha, and to hasten by forced marches toward the same point: while the King himself resolved to leave Madrid with the slenderest of garrisons, and to carry the rest of the central reserve to the general rendezvous. Accordingly, he left only one brigade of Dessolles’ division, with a few of his untrustworthy Spanish levies, to hold the capital: the total did not amount to much over 4,000 men, and General Belliard, the governor of the city, was warned that he must be prepared to retreat into the Retiro forts, with his troops and the whole body of the Afrancesados and their families, if anything untoward should occur. For it was possible that an insurrection might break out, or that Venegas might succeed in slipping into Madrid by the roads from the east, or again, that Wilson (whose column had been heard of at Escalona and was believed to be much larger than was actually the case), might attempt a coup de main from the west. Leaving Belliard in this dangerous and responsible position, the King marched out upon the twenty-third with the remaining brigade of Dessolles’s division, the infantry and cavalry of his French Guard, two squadrons of chasseurs and fourteen guns, a force of some 5,800 men[627]. He had reached Navalcarnero, with the intention of joining Victor on the Alberche, when he received the news that[p. 500] the Marshal had retired towards Toledo, and was lying at Bargas behind the Guadarrama river. Here Joseph joined him on the morning of July 25.

On their concentration a force of 46,000 men was collected, Victor having brought up 23,000, the King 5,800, and Sebastiani 17,500. The latter had placed four of the six Polish battalions of Valence’s division in Toledo, and was therefore short by 3,000 bayonets of the total force of his corps. With such a mass of good troops at their disposition, Joseph, Jourdan, and Victor were all agreed that it was right to fall upon the Spaniards without delay. They were astonished to find that the British army was not in their front, but only Cuesta’s troops. They had expected to see the whole allied host before them, and were overjoyed to discover that the Estremadurans alone had pushed forward to Torrijos and Santa Ollala. Instead, therefore, of being obliged to fight a defensive battle behind the river Guadarrama, it was in their power to take the offensive.

This was done without delay: on the morning of July 26 the French army advanced on Torrijos, with the 1st Corps at the head of the column. But Cuesta, when once he had discovered the strength of the force in his front, had resolved to retreat. Victor found opposed to him only the division of Zayas and two cavalry regiments, which had been told off to cover the withdrawal of the Estremaduran army. The Marshal sent out against this rearguard the chasseurs of Merlin and the dragoons of Latour-Maubourg, who drove in the Spanish horse, almost exterminating the unfortunate regiment of Villaviciosa, which, in retiring, chanced to blunder against the high stone walls of some enclosures from which exit was difficult[628]. Zayas then went to the rear, and retired towards the cavalry division of Albuquerque, which Cuesta hastily sent to his assistance. The French cavalry took some time to re-form for a second attack, and their infantry was still far off. The Spanish rear[p. 501]guard therefore, covered by Albuquerque’s horse, had time enough to fall back on the main body, which was already in full retreat. Their cavalry then followed, and being not very strenuously pursued by Merlin and Latour-Maubourg, got off in safety. The whole army, marching at the best of its speed, and in considerable disorder, finally reached the Alberche without being caught up by the enemy. Cuesta found the British divisions of Sherbrooke and Mackenzie guarding the river: Wellesley had sent them forward when he heard of the approach of the French, and had placed the former on the hills above the further side of the bridge, to cover the passage, and the latter in reserve. He rode out himself to meet the Spanish general, and begged him to carry his army beyond the Alberche, as it would be extremely dangerous to be caught with such an obstacle behind him, and no means of retreat save a long bridge and three fords. But Cuesta tempted providence by declaring that he should encamp on the further bank, as his troops were too exhausted to risk the long defile across the bridge after dark. His sullen anger against Wellesley for refusing to follow him on the twenty-fourth was still smouldering in his breast, and the English were convinced that he remained on the wrong side of the river out of pure perversity, merely because his colleague pressed him to put himself in safety. He consented, however, to retreat next morning to the position which Wellesley had selected in front of Talavera.

The French made no appearance that night, though they might well have done so, and the Spanish army, bivouacing confusedly in the narrow slip of flat ground between the heights and the Alberche, enjoyed undisturbed rest during the hours of darkness. It is impossible not to marvel at the slackness with which Victor conducted the pursuit: he had twelve regiments of splendid cavalry to the front[629], and could undoubtedly have pressed the Estremadurans hard if he had chosen to do so. Cuesta’s retreating columns were in such a state of confusion and disorder that a vigorous assault on their rear might have caused a general débandade. But after driving in Zayas in the[p. 502] early morning, Victor moved very slowly, and did not even attempt to roll up Albuquerque’s cavalry rearguard, though he could have assailed it with very superior numbers. When taxed with sloth by Marshal Jourdan, he merely defended himself by saying that the horses were tired, and that the infantry was still too far to the rear to make it right for him to begin a combat which might develop into a general engagement. But it is hard to see that he would have risked anything by pressing in upon Albuquerque, for if Cuesta had halted his whole army in order to support his rearguard, there was nothing to prevent the French cavalry from drawing off, and refusing to close till the main body of the 1st Corps should come up.

Thanks to Victor’s slackness the Spaniards secured an unmolested retreat across the Alberche on the following morning. It is said that Cuesta, in sheer perversity and reluctance to listen to any advice proffered him by Wellesley, delayed for some hours before he would retreat, and that when at last he yielded to the pressing solicitations of his colleague he remarked to his staff ‘that he had made the Englishman go down on his knees’ before consenting.

All through the morning hours of the twenty-seventh the Army of Estremadura was pouring across the bridge and the fords, not in the best order. They had almost all passed, when about noon the French cavalry began to appear in their front. When the enemy at last began to press forward in strength, Wellesley directed Sherbrooke’s and Mackenzie’s divisions to prepare to evacuate their positions on the eastern bank, which they did as soon as the last of the Spaniards had got into safety. The first division passed at the bridge, the third at the fords near the village of Cazalegas: then Sherbrooke marched by the high-road towards Talavera, while Mackenzie, who had been told off as the rearguard, remained with Anson’s light horse near the ruined Casa de Salinas, a mile to the west of the Alberche.

It may seem strange that Wellesley made no attempt to dispute the passage of the river, but the ground was hopelessly indefensible. The left bank (Victor’s old position of July 22) completely commands the right, the one being high, the other both low and entirely destitute of artillery positions. More[p. 503]over, a great part of the terrain was thickly strewn with woods and olive plantations, which made it impossible to obtain any general view of the country-side. They would have given splendid cover for an army advancing to storm the heights on the French bank, but were anything but an advantage to an army on the defensive. For, unable to hold the actual river bank because of the commanding hills on the further side, such an army would have been forced to form its line some way from the water, and the tangled cover down by the brink of the stream would have given the enemy every facility for pushing troops across, and for pressing them into the midst of the defender’s position without exposing them to his fire. Wellington had examined the line of the Alberche upon the twenty-fourth and twenty-fifth, and had pronounced it absolutely untenable; ‘no position could be worse,’ he wrote to O’Donoju[630], but he had discovered one of a very different kind a little to the rear, and had already settled the way in which it was to be occupied. It presented so many advantages that even Cuesta had consented to accept it as a good fighting-ground, and the Estremaduran army was at this very moment occupied in arraying itself along that part of the line which had been allotted to it. Sherbrooke’s division was retiring across the plain to fall into the section which Wellesley had chosen for it, and Hill’s and Campbell’s troops were moving to their designated ground. Only Mackenzie and the light cavalry had yet to be established in their post.

In the act of withdrawing, this division became involved in an unfortunate combat, which bid fair for a moment to develop into a disaster. Its two brigades had been halted close to the ruined house called the Casa de Salinas, in ground covered partly with underwood and partly with olive groves. The cavalry had been withdrawn to the rear, as it was impossible to use it for vedettes in such a locality. The infantry was supposed to have a chain of pickets thrown out in its front, but it would appear that they must have been badly placed: as one eye-witness confesses, ‘we were by no means such good soldiers in those days as succeeding campaigns made us, and sufficient precautions had not been taken to ascertain what was passing in the wood[631],’ and[p. 504] between it and the ford below Cazalegas. French cavalry alone had hitherto been seen, and from cavalry Mackenzie’s troops were certainly safe in the tangled ground where they were now lying.

But already Victor’s infantry had reached the front, and its leading division, that of Lapisse, had forded the Alberche far to the north, and had entered the woods without being observed by the outlying pickets of Mackenzie’s left brigade[632]. It had even escaped the notice of Wellesley himself, who had just mounted the roof of the ruined Casa de Salinas, the only point in the neighbourhood from which anything like a general view of the country-side could be secured. While he was intent on watching the heights above the Alberche in his front, and the cavalry vedettes descending from them, the enemy’s infantry was stealing in upon his left.

Lapisse had promptly discovered the line of British outposts, and had succeeded in drawing out his division in battle order before it was observed. He had deployed one regiment, the 16th Léger, as a front line, while the rest of his twelve battalions were coming on in support.

While, therefore, Wellesley was still unconscious that the enemy was close upon him, a brisk fire of musketry broke out upon his left front. It was the French advance driving in the pickets of Donkin’s brigade. The division had barely time to stand to its arms—some men are said to have been killed before they had risen from the ground—and the Commander-in-chief had hardly descended from the roof and mounted his charger, when the enemy was upon them. The assault fell upon the whole front of Donkin’s brigade, and on the left regiment (the 2/31st) of that of Mackenzie himself. So furious and unexpected was it, that the 87th, 88th, and 31st were all broken, and driven some way to the rear, losing about eighty prisoners. It was[p. 505] fortunate that the French advance did not strike the whole line, but only its left and centre. The 1/45th, which was just outside the limit of Lapisse’s attack, stood firm, and on it Wellesley re-formed the 31st, while, a little further to the north, the half-battalion of the 5/60th also held its ground and served as a rallying-point for the 87th and 88th. The steadiness of the 1/45th and 5/60th saved the situation; covered by them the division retired from the woods and formed up in the plain, where Anson’s light horsemen came to their aid and guarded their flanks. The French still pressed furiously forward, sending out two batteries of horse artillery to gall the retreating columns, but they had done their worst, and during the hours of the late afternoon Mackenzie’s infantry fell back slowly and in order to the points of the position which had been assigned to them. Donkin’s brigade took post in the second line behind the German Legion, while Mackenzie’s own three regiments passed through the Guards and formed up in their rear. Their total loss in the combat of Casa de Salinas had been 440 men—the French casualties must have been comparatively insignificant—probably not 100 in all[633].

From the moment when the fray had begun in the woods till dusk, the noise of battle never stopped, for on arriving in front of the allied position, the French artillery drew up and commenced a hot, but not very effective, fire against those of the troops who held the most advanced stations. As the cannonade continued, the different regiments were seen hurrying to their battle-posts, for, although the arrangements had all been made, some brigades, not expecting a fight till the morrow, had still to take up their allotted ground.

‘The men, as they formed and faced the enemy, looked pale, but the officers riding along their line, only two deep, on which all our hopes depended, observed that they appeared not less tranquil than determined. In the meanwhile the departing sun showed by his rays the immense masses moving towards us, and the last glimmering of the light proved their direction to be across our front, toward the left. The darkness, only broken[p. 506] in upon by the bursting shells and the flashes of the French guns, closed quickly upon us, and it was the opinion of many that the enemy would rest till the morning[634].’

Such, however, was not to be the case: there was to be hard fighting in front of Talavera before the hour of midnight had arrived.