A history of the Peninsular War, Vol. 1, 1807-1809

by Charles Oman

Castaños had placed the troops under his immediate command at Tarazona and Cascante, which were destined to form the left and centre of his position: the remainder of it, from Cascante to Tudela, was allotted to the Aragonese and to the Murcian division of the Army of Andalusia, which had been across the Ebro in O’Neille’s company, and was now returning with him. Till they came up Castaños had only under his hand two complete divisions of his ‘Army of the Centre,’ and some small fragments of two others. The complete divisions were those of Grimarest (No. 2) and La Peña (No. 4), each of which had been increased in numbers but not in efficiency by having allotted to it some of the battalions of the ‘Army of Castile,’ which had been dissolved for its bad conduct at Logroño on October 26. There had at last begun to arrive at the front a considerable part of the other two Andalusian divisions, which had first been detained beyond the Sierra Morena by the Junta of Seville, and then kept some time in Madrid to complete their equipment. Two battalions of these belated troops had at last appeared on October 30, and ten more had since come up[486]. But the bulk of the 1st and 3rd Divisions was still absent, and no more than 5,500 men from them had been added to Castaños’ army. The mixed brigade formed from these late arrivals seems to have been under General Villariezo, of the 1st Division. The whole force amounted to about 28,000 men, of whom 3,000 were horsemen, for the army of Andalusia was stronger in the cavalry arm than any other of the Spanish hosts. But of these the Murcian and Valencian division of Roca (formerly that of Llamas) was with O’Neille, and had not yet reached the field; while five battalions, from the dissolved Castilian army, were far away on the left in the mountains of Soria, whither Castaños had detached them under General Cartaojal, with orders to observe the French corps which was coming up on his rear.

The other half of the Spanish army consisted of the missing division of the Army of the Centre—that of Roca—and the two divisions belonging to Palafox—those of O’Neille and Saint March—the former composed mainly of Aragonese[487], the latter almost entirely of Valencian troops. None of the Aragonese reserves from the great camp at Saragossa had yet come upon the scene. But the two divisions in the field were very strong—they must have had at least 17,000 men in their ranks. On November 1 they were more than 18,000 strong, and, two months after—when they had passed through the disaster of Tudela, and had endured ten days of the murderous siege of Saragossa—they still showed 14,000 bayonets. We cannot calculate them at less than 17,000 men for the battle of November 23. On the other hand, there were hardly 600 cavalry in the whole corps.

It would appear then that Castaños must have had some 45,000 men in line, between Tarazona and Tudela, when Lannes came up against him[488]. The French marshal, on the other hand, had about 34,000. On the difference in quality between the two armies we have no need to dilate: even the two divisions of the conscripts of 1807, which served in Moncey’s corps, were old soldiers compared to the armies of Aragon and Castile, or a great part of that of Andalusia. Moreover, as in all the earlier battles of the Peninsular War, the Spaniards were hopelessly outmatched in the cavalry arm. There was no force that could stop the 4,500 or 5,000 horsemen of Colbert, Digeon, and Wathier[489].

The position Tudela—Tarazona, which Castaños intended to hold, is of enormous length—about ten and a half miles in all. Clearly 45,000 men in the close order that prevailed in the early nineteenth century were inadequate to hold it all in proper strength. Yet if the points on which the French were about to attack could be ascertained in good time, the distances were not so great but that the army could concentrate on any portion of the line within three hours. But to make this practicable, it was necessary firstly that Castaños should keep in close touch with the enemy by means of his cavalry—he had quite enough for the purpose—and secondly that he should have all his men massed at suitable points, from which they could march out to the designated fighting-ground at short notice. The Spanish troops were, now as always, so slow at manœuvring that the experiment would be a dangerous one, but this was the only way in which the chosen position could possibly be held. The ground was not unfavourable; it consisted of a line of gentle hills along the south bank of the river Queiles, which commanded a good view over the rolling plain across which the French had to advance. On the extreme right was the town of Tudela, covered by a bold hill—the Cerro de Santa Barbara—which overhangs the Ebro. Thence two long ridges, the hills of Santa Quiteria and Cabezo Malla, extend for some two and a half miles in a well-marked line: this section formed the right of the position. From the left of the Cabezo Malla as far as the little town of Cascante—four miles—the ground is less favourable; indeed, it is fairly flat, and the line is indicated mainly by the Queiles and its irrigation-cuts, behind which the Spanish centre was to form[490]. From Cascante westward as far as Tarazona—a distance of four miles or a little over—the position is better marked, a spur of the Sierra de Moncayo coming down in a gentle slope all along the southern bank of the little Queiles. The centre, between the Cabezo Malla and Cascante, was obviously the weak point in the position, as the only obstacle to the enemy’s advance was the river, which was fordable by all arms at every point along this dangerous four miles.

The disaster which Castaños was to suffer may be ascribed to two mistakes, one of which was entirely within his own control, while the other was due to the stupidity of O’Neille. With 3,000 cavalry in hand, the Commander-in-chief ought to have known of every movement of the French for many hours before they drew near to the position. It would then have been in his power to concentrate on those parts of the line where the attack was about to be delivered. But instead of sending out his horse ten miles to the front, Castaños kept them with the infantry[491], and the first notice of the approach of Lannes was only given when, at nine in the morning, a regiment of Wathier’s cavalry rode right up to the town of Tudela, driving in the outposts and causing great confusion. To the second cause of disaster we have already had occasion to allude: on the night of the twenty-second O’Neille had (contrary to his orders) encamped north of the Ebro. His 17,000 men began to defile over the bridge next morning in a leisurely fashion, and were still only making their way to their designated positions when Lannes attacked. In fact the Spanish line of battle was never formed as had been planned: the various brigades of the Army of Aragon were hurried one after another on to the heights south-west of Tudela, but entirely without system or order: the lower ground to the left of the Cabezo Malla was never occupied at all, and remained as a gap in the centre of the line all through the battle.

Lannes, who was aware that the Spaniards were intending to fight at Tudela, had marched at dawn from his camps in front of Alfaro in two columns. One, composed of Moncey’s corps, with Wathier’s cavalry at its head, came by the high-road near the Ebro. The other, composed of the two independent cavalry brigades of Colbert and Digeon, and of Lagrange’s division, was more to the west, and headed for Cascante. The Marshal had no intention of attacking the left of the Spanish line in the direction of Tarazona, which he left entirely to itself. He met not a single Spanish vedette till Wathier’s cavalry ran into the pickets immediately outside Tudela.

Castaños was in the town, engaged in hurrying the march of the Aragonese troops across the great bridge of the Ebro, when the fusillade broke out. The unexpected sound of musketry threw the troops into great excitement, for they were jammed in the narrow mediaeval lanes of Tudela when the sounds of battle came rolling down from the Cerro de Santa Barbara. The Commander-in-chief himself was caught between two regiments and could not push his way out to the field for some time. But the men were quite ready to fight, and hurried to the front as fast as they were able. Roca’s Valencian division (the 5th of the Andalusian army) had been the first to cross the Ebro: it was pushed up to the Cerro de Santa Barbara, and reached its summit just in time to beat off the leading brigade, one from Morlot’s division, which was ascending the hill from the other side. Saint March’s battalions, who had crossed the bridge next after Roca, were fortunate enough to be able to deploy and occupy the hill of Santa Quiteria before they were attacked. But O’Neille’s Aragonese and Murcians were less lucky: they only succeeded in seizing the Cabezo Malla ridge after driving off the skirmishers of Maurice Mathieu’s French division, which had come up next in succession to Morlot, and was just preparing to mount the slope. But the position was just saved, and the Army of Aragon was by ten o’clock formed up along the hills, with its right overhanging the Ebro and its left—quite in the air—established on the Cabezo Malla. The front was somewhat over two miles in length, and quite defensible; but the troops were in great disorder after their hurried march, and the generals were appalled to find that the Army of the Centre had not moved up to join them, and that there was a gap of three miles between the Cabezo Malla and the nearest of the Andalusian divisions. Castaños perceived this fact and rode off, too late, to bring up La Peña from Cascante to fill the void. Palafox was not on the field: he had gone off at daybreak (still in high dudgeon that his scheme for an attack by Pampeluna had been overruled) and was far on the road to Saragossa.

It is clear that Lannes’ first attack was unpremeditated and ill-arranged: he had been tempted to strike when his vanguard only had come up, because he saw the Spanish position half empty and the Aragonese divisions struggling up in disorder to occupy it. Hence came his first check: but the preliminary skirmish had revealed to him the existence of the fatal gap between the two Spanish armies, and he was now ready to utilize it. While Castaños was riding for Cascante, the divisions of Musnier, Grandjean, and Lagrange were coming upon the field, and Lannes was preparing for a second and more serious attack.

Meanwhile the fortune of the day was being settled on the left. When the army of Lannes appeared in the plain, La Peña at Cascante should have marched at once towards the Aragonese, and Grimarest and Villariezo from Tarazona should have moved on Cascante to replace La Peña’s division at that place. Neither of them stirred, though the situation was obvious, and though they presently received orders from Castaños to close in to their right. La Peña was the most guilty, for the whole battle-field was under his eye: he would not move because he had before him Digeon’s and Colbert’s cavalry, and was afraid to march across their front in the open plain, protected only by the shallow Queiles. He had 8,000 or 9,000 Andalusian and Castilian infantry, and 1,500 horse, but allowed himself to be neutralized by two brigades of dragoons. All that he did in response to the summons to move eastward was to send two battalions to occupy the hamlet of Urzante, a mile to his right. There was still a space of three miles between him and Saint March. This scandalous and cowardly inaction is in keeping with the man’s later career: it was he who in 1811 betrayed Graham at Barossa, and fled back into safety instead of stopping to assist his allies. On this occasion he lay for four hours motionless, while he watched the French forming up for a second attack on the Army of Aragon. Cowed by the 3,000 dragoons in his front, he made no attempt to march on the Cabezo Malla to O’Neille’s assistance. Grimarest’s conduct was almost equally bad: he was further from the scene of fighting, and could not, like La Peña, see the field: but it is sufficient to say that he received Castaños’ order to march on Cascante at noon, and that he did not reach that place—four miles distant—till dusk.

The Commander-in-chief himself was most unlucky: he started for Cascante about noon, intending to force his divisional generals to draw near the battle-field. But as he was crossing the gap between O’Neille and La Peña he was sighted by some French cavalry, who were cautiously pushing forward through the unoccupied ground. He and his staff were chased far to the rear by this reconnoitring party, and only shook them off by riding hard and scattering among the olive groves. Unable to reach Cascante, he was returning towards Tudela, when he received a hasty note from General Roca to the effect that the right wing of the army had been broken, and the heights of Santa Barbara lost.