BATTLE OF VITTORIA. ROUT OF THE FRENCH
When Picton and the 3rd Division, followed by the one available brigade of the 7th Division, came pouring across the Zadorra on the side of Mendoza, while Kempt debouched from the knoll of Yruna, and Vandeleur crossed the bridge of Villodas, the position of Leval’s division became desperate. It was about to be attacked in flank by four brigades and in front by two more, and being one of the weaker divisions of the Army of the South, only 4,500 bayonets, was outnumbered threefold. Its original reserve (half Daricau’s division) had gone off to the Puebla heights hours before: the general army-reserve (Villatte’s division) had been sent away in the same direction by Jourdan’s last orders. The nearest disposable and intact French troops were Darmagnac’s division of the Army of the Centre—two miles to the rear, in position by Zuazo: the other divisions of that army had (as we have seen) gone off on a wholly unnecessary excursion to watch the Trevino road. The left wing and centre of the Army of the South was absorbed in the task of keeping back Hill, and had just begun the counter-attack upon him which Jourdan had ordered an hour before.
The sudden change in the situation, caused by the very rapid advance of Picton and the brigades that helped him, was all too evident to King Joseph and his chief of the staff, as they stood on the hill of Ariñez. The whole force of the 3rd division struck diagonally across the short space between the river and Leval’s position—Brisbane’s brigade and Power’s Portuguese making for the French flank, while Colville, higher up the stream, made for the rear of the hill, in the direction of the village of Margarita. Kempt followed Brisbane in second line, Grant’s brigade of the 7th Division, when it crossed at Mendoza, came on behind Colville. So did Vandeleur, from Villodas, after he had pulled down the obstructions and got his men over the narrow bridge. Nor was this all—the 4th Division, so long halted on the scrubby hillside opposite the left-hand of the two Nanclares bridges, suddenly started to descend the slope at the double-quick, Stubbs’s Portuguese brigade leading.
Jourdan had to make a ‘lightning change’ in all his dispositions. Leval, obviously doomed if he did not retire quickly, was told to evacuate his hill and fall back past Ariñez, into which he threw a regiment to cover his retreat, on to the heights behind it. The two brigades of Daricau and Conroux, which had stood on the other side of the high road from Leval, were to make a parallel movement back to the same line of heights. The other brigades of Daricau and Conroux, with Maransin—now deeply engaged some with O’Callaghan and others with the 50th and 92nd—were to abandon the attack which they had just begun, and which had somewhat pushed back the British advance. They too must go back to the slopes behind Ariñez. It would take longer to recall Villatte, who was now far up on the crest of the mountain to the left, engaged with Morillo’s Spaniards and the 71st. But this attack also must be broken off. Lastly, to fill the gap between Leval’s new position and the Zadorra, the Army of the Centre must come forward and hold Margarita, or if that was impossible, the hill and village of La Hermandad behind it. But only Darmagnac’s division was immediately available for this task, Cassagne’s having to be brought back from the eccentric counter-march toward the Trevino road, to which it had been committed an hour before. In this way a new line of battle would be formed, reaching from the Zadorra near Margarita across the high road at Gomecha, to the heights above Zumelzu on the left. It was at best a hazardous business to order a fighting-line more than two miles long, and bitterly engaged with the enemy at several points, to withdraw to an unsurveyed position a mile in its rear, where there was practically no reserve waiting to receive it. For on the slopes above Ariñez there was at that moment nothing but Pierre Soult’s light horse, and Treillard’s dragoons, with two batteries of artillery[571]. The new front had to be constructed from troops falling back in haste and closely pursued by the enemy, combining with other troops coming in from various directions, viz. the two divisions of the Army of the Centre. And when the line should be re-formed, what was to prevent its left from being turned once more by the Allied troops on the Puebla heights, or its right by columns crossing the middle Zadorra behind it[572]. For there would still be a gap of two miles between Margarita village and the nearest troops of the Army of Portugal, who were now engaged with Graham at Abechuco and Arriaga.
To speak plainly, the second French line was never properly formed, especially on its left; but a better front was made, and a stronger stand, than might perhaps have been expected, though the confusion caused by hasty and imperfect alignment was destined in the end to be fatal.
On the extreme left Villatte had been caught by the order of recall at the moment when he was delivering his attack on Morillo and the British 71st. He had reached the crest as ordered, had formed up across it, and then had marched on a narrow front against the Allies. Both British and Spanish were in some disorder when he came in upon them—they had now been fighting for four hours, and in successive engagements had driven first Maransin’s and then St. Pol’s brigades for two miles over very steep and rocky ground. At the moment when Villatte came upon the scene, the Allied advance had just reached a broad dip in the crest, which it would have to cross if its progress were to be continued. The Spaniards were on the right and the 71st and light companies on the left, or northern, part of the ridge. It would have been a suitable moment to halt, and re-form the line before continuing to press forward over dangerous ground. But the officer who had succeeded Cadogan in command[573] was set on ‘keeping the French upon the run,’ and recklessly ordered the tired troops to plunge down the steep side of the declivity and carry the opposite slope. He was apparently ignorant that fresh French troops were just coming to the front—several eye-witnesses say that a column in light-coloured overcoats with white shako-covers, which had been noticed on the right, was taken for a Spanish detachment[574]. At any rate, the 71st crossed the dip—four companies in its centre, the remainder at its upper end—and was suddenly met not only by a charging column in front, but by an attack in flank and almost in rear. The first volley brought down 200 men—the shattered battalion recoiled, and remounted its own slope in utter disorder, leaving some forty prisoners in the hands of the enemy. These were the only British soldiers who fell into the hands of the French that day[575]. Fortunately the 50th, coming up from the rear, was just in time to form up along the edge of the dip and cover the retreat, and was joined soon after by the 92nd, who had been facing another separate French unit lower down the slopes and to the left flank. Seeing their opponents move off for no visible reason (they had received no doubt the general order to retire), the Highland regiment had pushed up on to the crest and joined the 50th and the Spaniards.
Villatte, still ignorant that the whole French army was falling back, tried to improve his success over the 71st into a general repulse of the Allied force upon the crest, and ordered his leading regiment to cross the dip and attack the troops upon the opposite sky-line. They suffered the same fate as the 71st and from the same cause: the climb was steep among stones and furze, they were received with two devastating volleys when they neared the top of the declivity, and then charged by the 50th and 92nd—the column broke, rolled down hill, and went to the rear. A second but less vigorous attack was made by another French regiment, and repulsed with the same ease—the wrecks of the 71st joining in the defence this time. Villatte then brought up a third regiment, but this was only a feint—the attack never developed, and while it was hanging fire the whole division swung round to the rear and went off—Jourdan’s general order to retreat had at last been received, and Villatte was falling back to the new line[576]. Cameron of the 92nd, now in command on the heights, followed him up, as did the Spaniards. But there was to be no more serious fighting upon the Puebla mountain: the French gave way whenever they were pressed[577].
Long before Villatte’s fight on the high ground had come to an end, the engagement at the other end of the French line had taken an unfavourable turn. The battle in this direction fell into three separate sections. Close to the Zadorra, Colville, with the left-hand brigade of the 3rd Division, was pushing up towards Margarita, while Darmagnac, from the heights of Zuazo, was making for it from the other side. It had taken some time to file Colville’s battalions across the ford, and deploy them for the advance, and the French brigade of Darmagnac’s division got into the village first, and made a strong defence there, while the German brigade occupied La Hermandad in its rear. Colville was held in check, suffered heavily, and could not get forward. But after half an hour’s deadly fighting the enemy gave way, not only because of the frontal pressure, but because the troops on his left (Leval’s division of the Army of the South) had been defeated by Picton and were retiring, thereby exposing the flank of Darmagnac’s line. D’Erlon drew back Chassé’s much thinned brigade half a mile, to the better defensive ground formed by the village of La Hermandad and the height above it, where his German brigade was already in position: this was an integral part of the new line on which Jourdan had determined to fight, while Margarita was on low ground, and too far to the front. Colville’s brigade, like its adversaries much maltreated[578], was replaced by Grant’s brigade of the 7th Division in front line[579], while Vandeleur’s of the Light Division followed in support. They had now in front of them not only Darmagnac’s but Cassagne’s division, which had come back from its fruitless excursion to the Trevino road, and had joined the other section of the Army of the Centre[580], taking up ground in second line.
Meanwhile the really decisive blow of the whole battle was being delivered by Picton, a thousand yards farther to the south, in and above Ariñez. The striking force here consisted of Power’s Portuguese brigade on the left, and of Brisbane’s British brigade on the right, opposite the village. Kempt’s half of the Light Division had followed Picton faithfully in his diagonal movement across the slopes, and was close behind Brisbane. Farther to the right the new front of attack of Wellington’s army was only beginning to form itself—the 4th Division had deployed after crossing the upper bridge of Nanclares, and was now coming on in an échelon of brigades—Stubbs’s Portuguese in the front échelon, then W. Anson, last Skerrett. They extended from the high road southwards, and were getting into touch with Hill’s column, which after the French evacuated the height behind Subijana had also deployed for the advance—the 2nd division having now thrown forward Byng’s brigade on its left, with O’Callaghan’s next it, and Ashworth’s Portuguese in second line. Silveira’s division remained in reserve. The cavalry of the centre column had crossed after the infantry—R. Hill, Ponsonby, Victor Alten, and Grant by the upper bridge of Nanclares, D’Urban’s Portuguese by the lower. They deployed on each side of the high road east of the river, behind the 4th Division, ground suitable for horsemen being nowhere else visible. On the heights of Puebla there still remained Cadogan’s brigade (now under Cameron of the 92nd) and Morillo’s Spaniards. This detached force, which was hard in pursuit of Villatte’s retreating column, was decidedly ahead of the rest of the army, and well placed for striking at the new French flank, but it was tired and had fought hard already for many hours.
When the 4th Division had passed the upper bridge of Nanclares, and before the cavalry began to cross the Zadorra, Colonel Dickson had by Wellington’s orders commenced to bring forward the reserve artillery. Very few British batteries had yet come into action, the broken nature of the ground preventing them from keeping close to their divisions. Hence it came to pass that there was by this time an accumulation of guns in the centre: during the rest of the battle it was employed in mass, many divisional batteries joining the artillery reserve, and a formidable line of guns being presently developed along the heights which had been the original position of Gazan’s centre and right. Some of them were to the north of the high road, on Leval’s hill: some to the south, where Daricau’s front brigade had stood when the action commenced. As soon as they had come up, they began to pound the French infantry on the opposite hill. Here General Tirlet had a still more powerful artillery force in action—all the guns of the front line had got back in safety save one belonging to the horse artillery battery which had been placed opposite the bridge of Nanclares[581], and three batteries from the reserve were already in position. The cannonade on both sides was fierce—but it was the infantry which had to settle the matter, and the disadvantage to the French was that their troops, much hustled and disarranged while retreating into the new position, never properly settled down into it—especially Conroux’s and Daricau’s divisions, which had been divided into separate brigades by the way in which Gazan had dealt with them at the commencement of the action, never got into regular divisional order again, and fought piecemeal by regiments.
The decisive point was on the ground at and above Ariñez, which was held by Leval’s division, with one regiment of Daricau’s (103rd Line) on their left. The village, low down on the slope, was held by Mocquery’s brigade—Morgan’s was in support with the guns, higher up and more to the right. Picton attacked with Power’s Portuguese on his left, Brisbane’s brigade on his right, and Kempt’s brigade of the Light Division in support, except that some companies of the 1/95th had been thrown out in front of Brisbane’s line, and led the whole attack. The Riflemen rushed at the village, penetrated into it, and were evicted after a fierce tussle, by a French battalion charging in mass down the street. But immediately behind came the 88th and 74th. The former, attacking to their right of the village[582], completely smashed the French regiment which came down to meet them in a close-fire combat, and drove them in disorder up the hill, while the latter carried Ariñez itself and swept onward through it. The 45th, farther to the right, attacked and drove off the regiment of Daricau’s division which was flanking Leval. Power’s Portuguese would seem to have got engaged with Morgan’s brigade, on the left of the village; it gave way before them, when the 74th had stormed Ariñez and the Connaught Rangers had broken the neighbouring column. Leval’s routed troops appear to have swept to the rear rather in a southerly direction, and to their left of the high road, so as to leave the beginnings of a gap between them and Cassagne’s division of the Army of the Centre, which was coming up to occupy the ridge north of Gomecha, in the new position.
The complete breach in the French centre made by Picton’s capture of Ariñez, and the driving of Leval out of his position above it, had the immediate effect of compelling Darmagnac’s division to conform to the retreat, by falling back from Margarita on to La Hermandad and the hills behind it on the one wing, while the confused line of Daricau’s, Conroux’s, and Maransin’s troops, on the other hand, had to retire to the level of Gomecha, though the 4th and 2nd Divisions were not yet far enough forward to be able to press them. Nearly all the French guns appear to have been carried back to the new position, which may roughly be described as extending from Hermandad on the Zadorra by Zuazo and Gomecha to the hills in front of Esquivel. It was quite as good as the Margarita-Ariñez-Zumelzu line which had just been forced by Picton’s central attack.
It took some little time for Wellington to organize his next advance; the troops which had forced the Ariñez position had to be re-formed, and it was necessary to allow the 4th and 2nd Divisions to come up level with them, and to bring forward Dickson’s mass of artillery to a more advanced line, to batter the enemy before the next infantry assault was let loose. The only point where close fighting seems to have continued during this interval was on the extreme left, where Lord Dalhousie, after the French had left Margarita, was pressing forward Grant’s brigade of his own division, supported by Vandeleur’s brigade of the Light Division, against D’Erlon’s new position, where Neuenstein’s brigade of five German battalions lay in and about La Hermandad, with Chassé’s dilapidated regiments in reserve behind. There was a very bitter struggle at this point, rendered costly to the advancing British by the superiority of the French artillery—D’Erlon had now at least two batteries in action—Dalhousie only his own six divisional guns, those of Cairnes. Grant’s brigade, after advancing some 300 yards under a very heavy fire, came to a stand, and took cover in a deep broad ditch only 200 yards from the French front. According to an eye-witness Dalhousie hesitated for a moment as to whether a further advance was possible[583], and had the matter settled for him by the sudden charge of Vandeleur’s brigade, which came up at full speed, carried the 7th Division battalions in the ditch along with it in its impetus, and stormed La Hermandad in ten minutes. The German defenders—Baden, Nassau, and Frankfurt battalions—reeled back in disorder, and retreated to the crest of the heights behind, where Cassagne’s division, hitherto not engaged, picked them up. D’Erlon succeeded in forming some sort of a new line from Crispijana near the Zadorra to Zuazo, where his left should have joined the right of the Army of the South. It is curious to note that while Grant’s brigade lost heavily in this combat (330 casualties), Vandeleur’s, which carried on the attack to success, suffered hardly at all (38 casualties). Their German opponents were very badly punished, having lost 620 men, including 250 prisoners, in defending La Hermandad against the two British brigades[584].
While Wellington, after the first breaking of the French line, was preparing under cover of the cannonade of Dickson’s guns for the assault on their new position in front of Zuazo and Gomecha, General Graham was developing his attack on the Army of Portugal and the French line of retreat, but not with the energy that might have been expected from the victor of Barrosa.
He had, as it will be remembered, sent Oswald and the 5th Division against the bridge of Gamarra Mayor, and Longa’s Spaniards against that of Durana, while he himself remained with the 1st Division, Pack’s and Bradford’s Portuguese, and the bulk of his cavalry on the high road, facing the bridge of Arriaga and its outlying bulwark the village of Abechuco. We should have expected that the main attack would be delivered at this point, but nothing of the sort took place. When the noise of Oswald’s heavy fighting at Gamarra Mayor had begun to grow loud, Graham directed the two Light Battalions of the German Legion (under Colonel Halkett) to clear the French out of Abechuco. This they did with trifling loss—1 officer and 51 men, capturing several guns in the village[585]. But Graham made no subsequent attempt to improve his success by forcing the bridge behind—as is sufficiently witnessed by the fact that of his remaining battalions the Guards’ Brigade had no casualties that day, and the three Line battalions of the K.G.L. one killed and one wounded. Before drawing up in front of Arriaga he had sent Bradford’s Portuguese, for a short time, to demonstrate to his right, toward the bridge of Yurre; but he called them back after a space, and placed them to the right of Abechuco, continuing the general line of the First Division. Bradford’s battalions lost precisely 4 men killed and 9 wounded. It is clear, therefore, that Graham never attacked the Arriaga position at all. Why he massed 4,000 British and 4,000 Portuguese infantry on this front—not to speak of two brigades of cavalry—and then never used them, it is hard to make out.
We know, it is true, that not only Graham but Wellington himself over-estimated the strength of Reille’s force. They did not know that Maucune had gone away in the dark, in charge of the great convoy, and thought that Foy’s and Taupin’s divisions were the only troops of the Army of Portugal which had not rejoined. Arguing that he had four infantry divisions in front of him (though they were really only two), Graham no doubt did well to be cautious—but he was much more than that. It was at least his duty to detain and engage as many of the enemy’s troops as was possible—and he certainly did not do so.
There was opposite him at Arriaga one single infantry division—Sarrut’s, and he did so little to employ it that Reille dared—after observing the British movements for some time—to take away Sarrut’s second brigade (that of Fririon) for use as a central reserve, which he posted at Betonio a mile back from the river, leaving Menne’s brigade alone—not much over 2,000 bayonets—opposed to the whole 1st Division, Bradford, and Pack. It is true that Menne had heavy cavalry supports—Digeon’s dragoons and Curto’s brigade of Mermet’s light horse—on one flank, and Boyer’s dragoons not far away on the other. But cavalry in 1813 were not troops which could defend a bridge or the line of a river. There was also a good deal of French artillery present—a more important fact under the existing circumstances. For Reille had still twenty guns ranged along the river[586], beside those which were detached on the flank with Casapalacios’ Spaniards. But Graham had almost as many—the three batteries of Lawson, Ramsay, and Dubourdieu—and of these the two last, ranged opposite Arriaga bridge, and pounding the village behind it, quite held their own against the opposing guns.
It can only be supposed that Graham, in refraining from any serious attack along the high road, was obeying in too literal a fashion Wellington’s orders not to commit himself to close fighting in the low ground, and to regulate his movements by those of the columns on his right (Picton’s and Dalhousie’s divisions). When these had worked their way up the Zadorra to his neighbourhood he did advance. But it was then so late that the enemy in front of him was able to get away, without any very disastrous losses.
While Graham kept quiet on the high road, Oswald was engaged in a very different style at Gamarra Mayor, where after his first capture of the village, he made at least three desperate attempts to force the bridge, held most obstinately by Lamartinière’s division. The passage was taken and retaken, but no lodgement on the southern bank could be made. After Robinson’s brigade had exhausted itself, Oswald put in the 3/1st from Hay’s brigade and some of his Portuguese[587]. But no success was obtained, though both sides suffered heavily. The casualty rolls of the 5th Division show a loss of 38 officers and 515 men—those of their French opponents 38 officers and 558 men. Practically all on both sides fell in the murderous fighting up and down Gamarra bridge. The forces were so equally balanced—each about 6,000 bayonets and one divisional battery—that on such a narrow front decision was impossible when both fought their best. The only way of attacking the bridge was by pushing straight down the narrow street of the village from the British side, and across an open field from the French. Both parties had guns trained upon its ends, which blew to pieces any column-head that debouched. There were no fords anywhere near, and the banks for some distance up-stream and down-stream were lined by the skirmishers of both sides, taking what cover they could find, and doing their best to keep down each other’s fire. ‘It plainly appeared this day that the enemy had formed a sort of determination not to be beat: we never saw them stand so vigorous before,’ writes a diarist from the ranks, in Robinson’s brigade[588].
There was an absolute deadlock at Gamarra Mayor till nearly five o’clock in the afternoon. At Durana things went otherwise: Longa, though hampered by his lack of guns, ended by pushing the Franco-Spanish brigade across the bridge, and then for some way down the south side of the Zadorra. The retreating party then made a stand behind a ravine and brook some half-mile farther on, where they were flanked by a brigade of Mermet’s light cavalry, as well as by their own five squadrons, and supported by the French battalion of the 3rd Line which had been in their quarter of the field from the first. Longa was unable to push them farther—probably for fear of lending his flank to cavalry charges, and gained no further ground till the general retreat of the French army began. But he had effectively cut King Joseph’s communication with France by seizing Durana—and this was the governing factor of the whole fight, since the enemy had now only the Pampeluna road by which he could retreat. If Joseph had owned some infantry reserves, he could (no doubt) have driven Longa away; but he had not a man to spare in any part of the field, and things were going so badly with the Army of the South that he had no attention to spare for the Army of Portugal.
It must have been about four o’clock before Wellington, having rearranged his line and brought up his artillery, determined to renew the general attack on the French right and centre. Joseph had brought up to the new position (extending from Crispijana on the left by Zuazo to the heights in front of Esquivel) the whole of the infantry of the Armies of the South and Centre, which now formed one rather irregular line. The only infantry reserve was the six weak battalions of the Royal Guard—perhaps 2,500 bayonets[589], placed on the high road in front of Vittoria—there was also a mass of cavalry in reserve, but this was of as little use for the defence of a hill-position as was Wellington’s for the assault on it. There were now in line Tilly’s division of dragoons which had been brought back from its useless excursion on the Logroño road, and Pierre Soult’s light horse, both of Gazan’s army, with Treillard’s dragoons, Avy’s Chasseurs, and the two cavalry regiments of the Royal Guard, all from the Army of the Centre—in all some 4,500 sabres.
The artillery, however, was very strong, and—deployed in a long line on both sides of the high road—was already sweeping all the slopes in front. There were present 46 guns of the Army of the South (all that it owned save one piece lost at Ariñez and three absent with Digeon’s dragoons), twelve guns of the Army of the Centre, and 18 from the reserve of the Army of Portugal—76 in all[590]. Dickson would appear to have brought up against them very nearly the same number: 54 British, 18 Portuguese, and 3 Spanish guns, when the last of the reserve batteries had got across the Zadorra and come forward into line—a total of 75 pieces[591]. The cannonade was the fiercest ever known in the Peninsula—each side was mainly trying to pound the enemy’s infantry—a task more easy for the French than the Allied gunners, since the assailant had to come up the open hillside, and the assailed was partly screened by woods (especially in front of Gomecha) and dips in the rough ground which he was holding.
The French line was now formed by Cassagne’s division on the extreme right, with one regiment (the 16th Léger) in Crispijana, and the others extending to meet Darmagnac’s much depleted battalions which were in and about Zuazo. Leval ought to have been in touch with Darmagnac, but obviously was not, the ground on each side of the high road being held by guns only, with cavalry in support some way behind. For after losing Ariñez Leval’s infantry had inclined much to their left. But on the other flank Villatte had, as it seems, inclined somewhat to the right, for having lost the heights of La Puebla, he could not prevent Cameron and Morillo from pressing along their crest and getting behind his new position: they were edging past his flank all through this period of the action.
The long front of the British advance started with Colville’s brigade—now once more in front line—opposite Crispijana, and was continued southward by those of Grant, Power, Brisbane, Stubbs, Byng, and O’Callaghan, while Vandeleur, Kempt, W. Anson, Skerrett, and Ashworth were formed in support, with Silveira’s division and the cavalry in third line. The missing brigades of the 7th Division were not yet on the field, possibly not even across the Zadorra, for neither of them lost a man that day. The advance of the line was a splendid spectacle, recorded with notes of admiration by many who witnessed it from the hill of Ariñez or the heights of La Puebla.
The French artillery fire was heavy, and in some sections of the line very murderous—Power’s and Stubbs’s Portuguese brigades were special sufferers. But the infantry defence was not resolute: on many parts of the front it was obviously very weak. The enemy was already a beaten army, he had been turned out of two positions, the news had got round that the road to France had been cut, and that Reille’s small force was in grave danger of losing the line of the Zadorra—in which case the whole army would find itself attacked in the rear. It is clear from the French narratives that the infantry did not support the guns in front line as they should. The reports of the Army of the Centre speak of being turned on their right by a column which kept near the river and took Crispijana—obviously Colville’s brigade. As the 16th Léger in that village only suffered a loss that day of one officer and 26 men, its resistance cannot have been very serious. But D’Erlon’s divisions were also outflanked on their left—by clouds of skirmishers drifting in by the wood and broken ground about Gomecha, who turned on the line of artillery, and began shooting down the gunners from flank and rear[592]. Obviously there was a gap along the high road, by which these light troops must have penetrated, and as obviously it was caused by Leval having sheered off to the south. For the artillery report of Gazan’s army also speaks of being turned on its right wing—‘enveloped by skirmishers who had got into Gomecha and were in rear of the position, which was being also attacked frontally.’ Nor is this all, ‘the mass on the mountain (Cameron and Morillo) descended on the left flank and rear of the Army of the South before it had time to form again: the artillery found itself without support.’ If so, where were the four and a half divisions of infantry which should have been protecting it?
The only possible deduction from all our narratives is that Gazan’s army made no real stand on the Gomecha-Esquivel position, and retired the moment that the attack drew near. And the person mainly responsible for the retreat was the Army-Commander himself, whose very unconvincing account of this phase of the action is that ‘the right flank of the line was continually being outflanked: I received no further orders about the taking up of the position of which the King had spoken; the enemy was getting near the gates of Vittoria (!), and so I had to continue my movement toward that town, after having taken up a position by Zuazo, with the intention of covering with my right-hand division and my guns the retreat of the rest of the army, which without this help would have been hopelessly compromised. At this time I had only lost four guns, abandoned on the extreme left of the line: the artillery was intact, and the army had suffered no greater loss than it had inflicted on the enemy.’
Reading this artful narrative, we note that (1) Gazan evacuated on his own responsibility a position that the King had definitely ordered him to take up—because he had ‘no further orders’; (2) the continued ‘turning of his right’ could only have resulted from the defeat of his own troops on and about the high road—which was in his sector, not in D’Erlon’s, Leval was across it when the last phase of the battle began; (3) D’Erlon held Zuazo and complained that there was nothing on his left, which was completely turned on the side where Leval ought to have been but was not; (4) Gazan retreated without any serious loss having compelled him to do so. This is obviously the fact. Villatte’s whole division had less than 300 casualties, Leval’s under 800, Daricau’s under 850: the only troops hard hit were Maransin’s brigade, Rey’s brigade of Conroux’s division, and the 103rd Line in Daricau’s. Moreover, the 4th and 2nd British Divisions, the troops immediately opposed to Gazan’s main line, had insignificant losses in this part of the action, Byng’s brigade under 150, Ashworth’s just 23, Anson’s 90, Skerrett’s 22; the only appreciable loss had been that of Stubbs’s Portuguese—about 240. For the heavy casualty lists of O’Callaghan’s, Cadogan’s, and Brisbane’s battalions had all been suffered in the earlier phases of the fight[593].
The fact is that Gazan went off without orders, and left the King and D’Erlon in the lurch. Jourdan is telling the exact truth when he says that ‘General Gazan, instead of conducting his divisions to the position indicated, swerved strongly to the right, marching in retreat, so as to link up with Villatte; he continued to draw away, following the foot-hills of the mountain [of Puebla], leaving the high road and Vittoria far to his left, and a vast gap between himself and Count D’Erlon’[594].
No doubt the breach made in the French centre when Picton stormed the heights behind Ariñez had been irreparable from the first; and no doubt also the flanking movement of Morillo and Cameron on the mountain must have dislodged the Army of the South, if it had waited to come into frontal action with Cole and Hill. But Gazan showed complete disregard of all interests save his own, and went off in comparatively good condition, without orders, leaving the King, D’Erlon, and Reille to get out of the mess as best they might.
Some of the British narratives tend to show that many parts of Gazan’s line had never properly settled down into the Gomecha-Esquivel position, having reached it in such bad order that they would have required more time than was granted them to re-form. An officer present with the skirmishing line of the 4th Division writes in his absolutely contemporary diary: ‘From the time when our guns began to open, and to throw shells almost into the rear of the enemy’s height, we saw him begin to fall back in haste from his position. We [4th Division] marched on at a great pace in column. From that moment the affair became a mere hunt. Our rapid advance almost cut off four or five French battalions—they made some resistance at first, but soon dissolved and ran pell-mell, like a swarm of bees, up the steep hill, from which they began to fire down on us. We disregarded their fire, and kept on advancing—in order to carry out our main object: broken troops are easy game. When we found the enemy in this second position there was heavy artillery fire. Since his left wing was somewhat refused, we advanced in échelon from the right. When we got within musketry range, we found he had gone off out of sight. After that we drove him out of one position after another till at last we were near the gates of Vittoria[595].’ Several other 4th Division and 2nd Division narratives agree in stating that after the storming of Ariñez the French never made anything like a solid stand on the Gomecha-Esquivel heights. But (as has been said above) the best proof of it is the British casualty list, which shows that the reserve line was hardly under fire, and that the front line had very moderate losses, except at one or two points.
D’Erlon made a more creditable resistance on the French right, but he was obviously doomed if he should linger long after Gazan had gone off. After losing Crispijana and Zuazo he made a last stand on the slopes a mile in front of Vittoria, between Ali and Armentia, to which the whole of his own artillery and that of the reserve, and perhaps also some guns of the Army of the South, retired in time to take up a final position[596]. For some short space they maintained a furious fire on the 3rd Division troops which were following them up, so as to allow their infantry to re-form. But it was but for a few minutes—the column near the Zadorra (Grant, Colville, and Vandeleur) got round the flank of the village of Ali, and the line of guns was obviously in danger if it remained any longer in action. Just at this moment D’Erlon received the King’s orders for a general retreat by the route of Salvatierra. The high road to France had ceased to be available since Longa got across it in the earlier afternoon. Any attempt to force it open would obviously have taken much time, and might well have failed, since the French were everywhere closely pressed by the pursuing British. Jourdan judged the idea of reopening the passage to be hopelessly impracticable, and ordered the retreat on Pampeluna as the only possible policy, though (as his report owns) he was aware of the badness of the Salvatierra road, and doubted his power to carry off his guns, transport, and convoy of refugees by such a second-rate track. But it was the only one open, and there was no choice.
The orders issued were that the Train and Park should get off at once, that the Army of the South should retreat by the country paths south of Vittoria, the Army of the Centre by those north of it. The Army of Portugal was to hang on to its position till D’Erlon’s troops had passed its rear, and then follow them as best it could. All the cross-tracks indicated to the three armies ended by converging on the Salvatierra road east of Vittoria, so that a hopeless confusion was assured for the moment when three separate streams of retreating troops should meet, and struggle for the use of one narrow and inadequate thoroughfare. But as a matter of fact the chaos began long before that time was reached, for the road was blocked or ever the three armies got near it. The order for the retreat of the Park and convoys had been issued far too late—Gazan says that he had advised Jourdan to give it two hours before, when the first positions had been abandoned. But the Marshal had apparently high confidence in his power to hold the Hermandad-Gomecha-Esquivel line; at any rate, he had given no such command. The noise of battle rolling ever closer to Vittoria had warned the mixed multitude of civil and military hangers-on of the army who were waiting by their carriages, carts, pack-mules, and fourgons, in the open fields east of Vittoria, that the French army was being driven in. Many of those who were not under military discipline had begun to push ahead on the Salvatierra road, the moment that the alarming news flew around that the great chaussée leading to France had been blocked. The Park, however—which all day long had been sending up reserve ammunition to the front, and even one or two improvised sections of guns—had naturally remained waiting for orders. So had the immense accumulation of divisional and regimental baggage, the convoy of treasure which had arrived from Bayonne on the 19th, and the heavy carriages of the King’s personal caravan, stuffed with the plate and pictures of the palace at Madrid. And of the miscellaneous French and Spanish hangers-on of the Court—ministers, courtiers, clerks, commissaries, contractors, and the ladies who in legitimate or illegitimate capacities followed them—few had dared to go off unescorted on an early start. There was a vast accumulation of distracted womenfolk—nous étions un bordel ambulant, said one French eye-witness—some crammed together in travelling coaches with children and servants, others riding the spare horses or mules of the men to whom they belonged. When the orders for general retreat were shouted around, among the fields where the multitude had been waiting, three thousand vehicles of one sort and another tried simultaneously to get into one or another of the five field-paths which go east from Vittoria, all of which ultimately debouch into the single narrow Salvatierra road. A dozen blocks and upsets had occurred before the first ten minutes were over, and chaos supervened. Many carriages and waggons never got off on a road at all. Into the rear of this confusion there came charging, a few minutes later, batteries of artillery going at high speed, and strings of caissons, which had been horsed up and had started away early from the Park. Of course they could not get through—but the thrust which they delivered, before they came to an enforced stop, jammed the crowd of vehicles in front of them into a still more hopeless block. Dozens of carriages broke down—whereupon light-fingered fugitives began to help themselves to all that was spilt. Of course camp followers began the game, but Jourdan says that a large proportion of the French prisoners that day were soldiers who s’amusaient à piller in and about Vittoria; one of the official reports from the Army of Portugal remarks that the retreating cavalry joined in the ‘general pillage of the valuables of the army,’ and other narrators mention that the treasure-fourgons had already been broken into long before the English came upon the scene[597]. Be this as it may, there was no long delay before the terrors of the stampede culminated with the arrival of several squadrons of Grant’s hussars, who had penetrated by the gap between D’Erlon’s and Gazan’s lines of retreat, and had made a short cut through the suburbs of the town. Of the chaos that followed the firing of the first pistol shots which heralded the charge of the British light cavalry, we must not speak till we have dealt with the last fight of the Army of Portugal, on the extreme French right.
Enlarge Battle of VITTORIA
Till Grant’s and Colville’s brigades broke into Crispijana and Ali, the line of Reille’s gallant defence along the Zadorra had remained intact. Only artillery fire was going on opposite the bridge of Arriaga, where the 1st Division and the two Portuguese brigades had halted by Graham’s orders at some distance from the river, and had never advanced. The two cavalry brigades were behind them. At Gamarra Mayor the 5th Division had failed to force a passage, though it had inflicted on Lamartinière’s troops as heavy a loss as it suffered. Longa was held up half a mile beyond Durana, though he had successfully cut the Bayonne chaussée. There was considerable cannonade and skirmishing fire going on, upon a front of three miles, but no further progress reported. The whole scene, however, was changed from the moment when Grant’s and Colville’s brigades, turning the flank of D’Erlon’s line close to the Zadorra, came sweeping over the hills by Ali into the very rear of Reille’s line. The Army of Portugal had been directed to hold the bridges and keep back Graham, until the rest of the French host had got off. But the danger now came not from the front but from the rear. In half an hour more the advancing British columns would be level with the bridge of Arriaga and surrounding the infantry that held it. Reille determined on instant retreat, the only course open to him: so far as was possible it was covered by the very ample provision of cavalry which he had in hand. Digeon, who has left a good account of the crisis, made several desperate charges, to hold back the advancing British while Menne’s brigade was escaping from Arriaga. One was against infantry, which formed square and beat off the dragoons with no difficulty[598]—the other against hussars, apparently two squadrons of the 15th, which had turned northward from the suburbs of Vittoria and tried to ride in and cut off the retreat of Menne’s battalions[599]. They were beaten back, and the infantry scrambled off, leaving behind them Sarrut, their divisional general, mortally wounded as the retreat began. Moreover, all the guns in and about Arriaga had to be abandoned in the fields, where they could make no rapid progress. The gunners unhitched the horses, and escaped as best they could.
Reille had drawn up in front of Betonio the small infantry reserve (Fririon’s brigade) which he had wisely provided for himself, flanked by Boyer’s dragoons and Curto’s brigade of light cavalry. The object of this stand was not only to give Menne’s and Digeon’s troops a nucleus on which they could rally, but to gain time for Lamartinière to draw off from in front of Gamarra Mayor bridge, where he was still hotly engaged with Oswald’s division. This infantry got away in better order than most French troops on that day, and even brought off its divisional battery, though that of the cavalry which had been co-operating with it had to be left behind not far from the river-bank, having got into a marshy bottom where it stuck fast. The Franco-Spaniards of Casapalacios and their attendant cavalry escaped over the hills east of Durana, pursued by Longa, who took many prisoners from them.
When Lamartinière’s division had come in, Reille made a rapid retreat to the woods of Zurbano, a mile and a half behind Betonio, which promised good cover. He was now being pursued by the whole of Graham’s corps, which had crossed the Zadorra when the bridges were abandoned. Pack’s brigade, followed by the 1st Division and Bradford, advanced on Arriaga; they were somewhat late owing to slow filing over the bridge. At Gamarra Oswald sent in the pursuit of Lamartinière two squadrons of light dragoons[600] which had been attached to his column; these were followed by the rest of their brigade, which Graham sent up from the main road to join them, and also by Bock’s German dragoons. The object of using the smaller and more remote bridge was that cavalry crossing by it had a better chance of getting into Reille’s rear than they would have secured by passing at Arriaga. The much exhausted 5th Division followed the cavalry.
Having reached the edge of the woods, Reille ordered the bulk of his troops to push on hard, by the two parallel roads which traverse them, keeping Fririon’s brigade in hand as a fighting rearguard. The rather disordered columns were emerging on the east side of the woods, and streaming into and past the village of Zurbano, when the leading squadrons of the 12th and 16th Light Dragoons came in upon them. These squadrons had avoided entangling themselves in the trees till the French rearguard had passed on, but prepared to charge the moment they got into open ground, though the main body of the brigade had not come up. They found opposed to them a regiment of Boyer’s dragoons, supported by another of hussars, which they charged but did not break[601]. But on the coming up of the rear squadrons, the attack was renewed with success. The French cavalry gave way, but only to clear the front of the 36th Line of Fririon’s brigade, which was in square outside Zurbano. The British light dragoons swept down on the square, but were completely repulsed by its steady fire. This gained time for the rest of Reille’s troops to make off, and the pursuit slackened. But the bulk of the French went off in such haste that they abandoned four guns of Lamartinière’s artillery, and took away with them only the remaining two—the sole pieces that escaped that day of the immense train of Joseph’s three armies. Some hundreds of stragglers were taken, but no single unit of the retiring force was cut off or captured whole. Reille wisely kept his army, so long as was possible, on the side-paths by Arbulo and Oreytia, before debouching into the main Salvatierra road, which was seething over with the wrecks of the other armies and the convoys. Hence he succeeded in escaping the utter confusion into which the rest fell, and, finally, when he turned into the main track, was able to constitute himself a rearguard to the whole. Graham’s pursuit of him seems to have been slow and cautious—no troops indeed ever came near the retreating columns except the two light dragoon regiments, the Caçador battalions of Pack and Bradford, and some of Longa’s skirmishers, who (as Reille mentions) followed him along the hills on his left, shooting down into the retreating masses, but not attempting to break in. The 5th Division appears to have followed not much farther than the open ground beyond the woods of Zurbano, where it halted and encamped after eight o’clock in some bean-fields. Nor does it seem that the 1st Division got more than a league or so beyond the Zadorra[602]. The Caçadores and cavalry, however, did not halt till they reached El Burgo, four or five miles farther on.
The scene was very different on the other side of Vittoria, where D’Erlon’s army was pushing its way, in utter disorder, through the fields and by-paths over which the Parks and convoy were trying in vain to get off, and Gazan’s (farther to the south) was making a dash over ground of the most tiresome sort. For in the rugged tract east of Esquivel and Armentia such paths as there were mostly ran in the wrong direction—north and south instead of east and west—and it was necessary to disregard them and to strike across country. Six successive ravines lay in the way—marshy bottoms in which ran trifling brooks descending toward the Zadorra—and several woods on the ridges between the ravines. Hill’s and Cole’s skirmishers were pressing in the rear, and above, on the heights of Puebla, Morillo’s and Cameron’s troops could be seen hurrying along with the intention of getting ahead of the retreating masses. The confusion growing worse every moment—for companies and battalions each struck out for the easiest line of retreat without regard for their neighbours—Gazan gave orders to abandon all the artillery, which was getting embogged, battery after battery, in the ravines, and gave leave for every unit to shift for itself—general sauve-qui-peut. The horses were unhitched from the guns and caissons, many of the infantry threw off their packs, and the army went off broadcast, some in the direction of Metauco and Arbulo, others by village paths more to the south. The general stream finally flowed into the Salvatierra road, where it was covered by the Army of Portugal, which was making a much more orderly retreat[603]. English eye-witnesses of this part of the battle complain bitterly that no horsemen ever came up to assist the wearied 2nd Division infantry in the pursuit, and maintain that thousands of prisoners might have been taken by a few squadrons[604]. But all the cavalry seems to have been directed on Vittoria by the high road, and save Grant’s hussar brigade none of it came into action. This is sufficiently proved by its casualty lists—in R. Hill’s, Ponsonby’s, Long’s, Victor Alten’s, and Fane’s brigades that day the total losses were one man killed and eleven wounded! Only Anson’s Light Dragoons in Graham’s corps and Grant’s regiments in the centre got into the fighting at all. The rugged ground, it is true, was unfavourable for cavalry action in regular order, but it was almost as unfavourable for disordered infantry escaping over ravines and ditches. Something was wrong here in the general direction of the mounted arm—perhaps it suffered from the want of a responsible cavalry leader—the brigadiers dared not act for themselves, and Bock (the senior cavalry officer) was not only short-sighted in the extreme, but absent from the main battle all day in Graham’s corps. One cannot but suspect that Wellington’s thunderings in previous years, against reckless cavalry action, were always present in the minds of colonels and brigadiers who had a chance before them. And possibly, in the end, there was more gained by the avoidance of mistakes of rashness than lost by the missing of opportunities, if we take the war as a whole. But at Vittoria it would most certainly appear that the great mass of British cavalry might as well have been on the other side of the Ebro for all the good that it accomplished.