As to the troops already on the spot, Reille was ordered to defend the line of the Bayas river, until the armies of the South and Centre should have had time to get past his rear and reach Vittoria. Gazan was ordered to collect the whole of the Army of the South at Arminion, behind Miranda, drawing in at once the considerable detachment which he had left beyond the Ebro. In this position he was to wait till D’Erlon, with the Army of the Centre, who had to move up from Haro, ten miles to the south, should have arrived and have got on to the great chaussée. He was then to follow him, acting as rearguard of the whole force. Between Arminion and Vittoria the road passes for two miles through the very narrow defile of Puebla, the bottle-neck through which the Zadorra river cuts its way from the upland plain of Vittoria to the lower level of the Ebro valley.
D’Erlon, starting at dawn from Haro, reached Arminion at 10 o’clock in the morning of the 19th, and pushed up the defile: the head of his column was just emerging from its northern end when a heavy cannonade began to be heard in the west. It continued all the time that the Army of the South was pressing up the defile, and grew nearer. This, of course, marked the approach of Wellington, driving the covering force under Reille before him toward the Zadorra. The British commander-in-chief had slept the night at Berberena near Osma, and had there drafted a set of orders which considerably modified his original scheme: probably the change was due to topographical information, newly garnered up from the countryside. Instead of sending Graham’s column via Orduña, to cut in on the flank of the chaussée behind Vittoria, he had resolved to send it by a shorter route, a mere country road which goes by Luna, Santa Eulalia, and Jocano to Murguia—the village which it had been ordered on the previous day to reach via Orduña and the high road. Presumably it had been discovered that this would save time,—the advantage of using a first-rate track being more than counterbalanced by the fact that the Orduña road was not only ten miles longer but crossed and recrossed by steep slopes the main sierra, which forms the watershed between Biscay and Alava. Or possibly it was only the discovery that the Luna-Jocano route could be taken by artillery that settled the matter: if it had been reported useless for wheeled traffic the old orders might have stood. At any rate, the turning movement, which was to take Graham into the rear of Vittoria, was made south of the main mountain chain, and not north of it[532].
Meanwhile, though Graham diverged north-eastward, the rest of the Army moved straight forward from the valley of the Omecillo to that of the Bayas in four columns, all parallel to each other, and all moving by country roads. The 3rd Division was ordered from Berberena to Carcamo—the 7th followed behind it. The 4th Division with D’Urban’s cavalry in front, and the Light Division with V. Alten’s hussars in front, were directed on Subijana and Pobes, keeping in close and constant communication with each other. Behind them came the cavalry reserve—R. Hill’s, Grant’s, and Ponsonby’s brigades—also the heavy artillery. Hill’s column, which had now come up into touch with the leading divisions, kept to the high road from Osma and Espejo towards the Puente Lara and the Ebro. But only cavalry reconnaissances went as far as the river—the mass of the corps turned off eastward when it had passed Espejo, and moved by Salinas de Añana, so as to come out into the valley of the Bayas south of the route of the Light Division. It thus became the right wing of the army which was deploying for the frontal attack.
It has been mentioned above that a local Spanish force from the Cantabrian mountains had joined Wellington at Medina de Pomar; this was the so-called ‘division’, some 3,500 bayonets, of the great guerrillero of the coast-land, Longa, now no more an irregular but a titular colonel, while his partida had been reorganized as four battalions of light infantry. Longa was a tough and persistent fighter—a case of the ‘survival of the fittest’ among many insurgent chiefs who had perished. His men were veteran mountaineers, indefatigable marchers, and skilled skirmishers, if rudimentary in their drill and equipment. Wellington during the ensuing campaign gave more work to them than to any other Spanish troops that were at his disposal—save the Estremaduran division of Morillo, old comrades of Hill’s corps, to which they had always been attached since 1811. The use which Wellington now made of Longa’s men was to employ them as a light covering shield for Graham’s turning force. They had been sent across the hills from Quincoces[533] on the 17th to occupy Orduña—from thence they descended on the 19th on to Murguia, thus placing themselves at the head of the turning column. The object of the arrangement was that, if the enemy should detect the column, he would imagine it to be a Spanish demonstration, and not suspect that a heavy British force lay hid behind the familiar guerrilleros. While Longa was thus brought in sideways, to form the head of Graham’s column, the other Spanish force which Wellington was employing was also deflected to join the main army. Giron’s Galicians, as has been mentioned above, had been sent by a long sweep through the sierras to demonstrate against Bilbao. They had reached Valmaseda on the 18th, and their approach had alarmed all the French garrisons of Biscay. Now, having put themselves in evidence in the north, they were suddenly recalled, ordered to march by Amurrio on Orduña and Murguia, and so to fall into the rear of Graham’s column. The distances were considerable, the roads steep, and Giron only came up with the Anglo-Portuguese army on the afternoon of the battle of Vittoria, in which (unlike Longa) he was too late to take any part. Somewhere on his march from Aguilar to Valmaseda he picked up a reinforcement, the very small Asturian division of Porlier—three battalions or 2,400 men—which Mendizabal had sent to join him from the blockade of Santoña. This raised the Galician Army to a total of some 14,000 bayonets.
The 19th June was a very critical day, as no one knew better than Wellington. The problem was whether, starting with the heads of his column facing the line of the Bayas, where Reille had rallied his three divisions and was standing at bay, he could drive in the detaining force, and cross the Zadorra in pursuit of it, fast enough to surprise some part of the French army still in its march up from the south. And, as an equally important problem, there was the question whether Graham, marching on Murguia, could reach the upper Zadorra and cut the great road north of Vittoria, before the French were in position to cover it. If these operations could be carried out, there would be a scrambling fight scattered over much ground, rather than a regular pitched battle. If they could not, there would be a formal general action on the 20th or 21st against an enemy established in position—unless indeed King Joseph should choose to continue his retreat without fighting, which Wellington thought quite possible[534].
Wellington is censured by some critics, including Napier[535], for not making a swifter advance on the 19th. It is said that a little more haste would have enabled him to get to Vittoria as soon as the enemy, and to force him to fight in dislocated disorder on ground which he had not chosen. This seems unjustifiable. The distance between the camps of the Allied Army, in front of Osma and Espejo, and Vittoria is some twenty miles—difficult ground, with the Bayas river flowing through the midst of it and a formidable position held by 15,000 French behind that stream. As Reille, conscious how much depended on his gaining time for the King to retreat from Miranda, was determined to detain the Allied Army as long as he could, it would have been useless to try to drive him away by a light attack. The rear of each of Wellington’s columns was trailing many miles behind the leading brigade. It was necessary to bring up against Reille a force sufficient to make serious resistance impossible, and this Wellington did, pushing forward not only the 4th and Light Divisions, but Hill’s column in support, on the southern flank. It took time to get them deployed, and the attack was opened by a cannonade. By the time that a general advance was ordered Reille had begun to retire—he did so very neatly, and crossed the Zadorra by the four nearest bridges without any appreciable loss. By this time the afternoon had arrived, the rest of the French army had passed the defile of Puebla, and Wellington judged the hour too late for the commencement of a pitched battle[536]. Moreover, he did not intend to fight without the co-operation of Graham’s column, and the latter was not where he would have wished it to be. The day had been very rainy, the cross roads were bad, and by some error of staff-work the head of the column had not received the countermarching orders directing it to march direct on Murguia, but had started in accordance with the earlier plan on to the Orduña road; the 1st and 5th Divisions had gone some way upon it before they were recalled and counter-marched into the right path[537]. Owing to blocks and bad weather they only reached Jocano, some six or eight miles east of Osma, by six in the evening. Murguia was still nine miles ahead, the rain was still falling in torrents, and Graham ordered the divisions to halt and encamp for the night. The fate of the campaign was not to be fought out that day, nor on the next, but on the third morning—that of June 21st, 1813.
SECTION XXXVI: CHAPTER VII
THE BATTLE OF VITTORIA. JUNE 21, 1813
(A) The First Stage
The plain of Vittoria, into which the French army debouched on the afternoon of June 19th, is a plain only by comparison with the high hills which surround it on all sides, being an oval expanse of rolling ground drained by the swift and narrow Zadorra river, which runs on its north-eastern side. Only in its northern section, near the city, does it show really flat ground. It is about twelve miles long from north-east to south-west, and varies from six to eight miles in breadth. The Zadorra is one of those mountain streams which twist in numberless loops and bends of alternate shallows and deep pools, in order to get round rocks or spurs which stand in the way of their direct course[538]. At one point seven miles down-stream from Vittoria it indulges in a complete ‘hairpin-bend’ in which are the bridges of Tres Puentes and Villodas, as it circles round a precipitous knoll. At several other spots it executes minor loops in its tortuous course. The little city of Vittoria stands on an isolated rising ground at its northern end, very visible from all directions, and dominating the whole upland with two prominent church spires at its highest point. The great road from France enters the plain of Vittoria and the valley of the Zadorra three miles north-east of the city, descending from the defile of Salinas, a long and difficult pass in which Mina and other guerrilleros had executed some of their most daring raids on French convoys. After passing Vittoria the road keeps to the middle of the upland in a westerly direction, and issues from it by the defile of La Puebla, where the Zadorra cuts its way through the Sierra de Andia in order to join the Ebro. There is not much more than room for road and river in the gorge, which is dominated by the heights of La Puebla, a spur of the Andia, on the east, and by a corresponding but lower range, the end of the heights of Morillas on the west.