Adding a battalion of foot-gendarmes from the evacuated Biscay garrisons, and another of local National guards, Villatte had over 17,000 infantry—a force as big as Reille’s or Clausel’s lieutenancies. There was some weak stuff among them[842], but the greater part were experienced troops in nowise inferior to the units of the marching divisions. It is hard to see why Soult did not make up a tenth division out of the best of them: this would still have left Villatte 12,000 men, and would have been very useful in the fighting army. But apparently, as will be seen during the narrative of the campaign of July-August, he intended to use the Reserve in an active fashion, and was foiled by the caution or timidity of Villatte, who discharged the orders given him in a very half-hearted way. This mass of troops was really quite capable of being used as a fighting unit—it had its four batteries of field artillery, the foreign cavalry were allotted to it, and a proportion of sappers and engineers. Even after leaving a detachment—say the Afrancesados and one or two of the weakest French battalions—to aid the conscripts to garrison Bayonne, it could have taken the field with 15,000 men of all arms[843].
To assume the offensive with a recently beaten army is dangerous. Many of the regiments which Soult had to use had only a month before recrossed the Pyrenees in a state of complete disorder: they had been entirely out of hand, and had been guilty of outrages among the French peasantry which recalled their worst doings in Spain. The greater number of the senior officers had applied for service in the Army of Germany the moment that they had crossed the frontier: it was refused to nearly all of them. The junior officers would have made similar applications, if they had thought that it would be of any use. Every one of every rank was cursing King Joseph, luck, the weather, the supply services, the War Ministry at Paris, and his own immediate hierarchical superior: a sentiment d’ineptie générale prevailed[844]. A great deal of this demoralization was mere nervous exhaustion, resulting from long marches and semi-starvation extending over many weeks. After a fortnight of comparative rest and more or less regular rations it commenced to subside. The whole army consisted of veteran troops proud of their regimental honour and their past victories: they soon persuaded themselves that they had never been given a fair chance in the recent campaign—Soult’s clever and malicious proclamation exactly hit off their state of mind. It was easy to lay all the blame on bad generalship, and to plead that a good half of the troops had not fought at Vittoria: in full force and under a competent leader they would have made mincemeat of the ‘motley levies’ of Wellington. When the whole host was reassembled, after Clausel’s arrival, every one could see that the numerical loss in the late disaster had been very moderate: there were more French troops assembled on the Bidassoa than any member of any of the armies had ever seen concentrated before. Shame and anger replaced dejection in their minds: even the marauders and deserters came back to the Eagles by thousands. Soult wrote to the Minister of War that the disquieting state of indiscipline which he had discovered on his arrival was subsiding, that marauding had ceased with the distribution of regular rations, and that the morale of the army was satisfactory.
The weakest part of the reorganization was, of course, the transport. The army had lost all its wheeled vehicles and many of its animals in the disaster of Vittoria: they had not been replaced in any adequate fashion during the four short weeks that followed. By extraordinary efforts the Commissariat had found food for every one, so long as the main body of the army was encamped between Bayonne and the Bidassoa. But it would be impossible to start it properly equipped for a long mountain campaign far from its base. Soult took the desperate risk of starting on July 23 with only four days’ rations in hand with the marching columns: all that would be wanted later was to come up by successive convoys sent out from the neighbourhood of Bayonne. And it was little better with munitions for infantry and artillery alike. The campaign became a time-problem: unless a decisive success were achieved within the first four or five days, there was grave danger of the army being brought to a stop. It would be helpless if it should have used up all its cartridges in several days of continuous and severe, but indecisive, fighting: or if bad weather should prevent the regular appearance of convoys from the rear. This Soult knew, but thought that his arrangements were calculated to secure that quick and decisive victory which would justify all taking of risks.
The plan of campaign which the Marshal chose was one of the three alternative schemes which the unlucky Jourdan had formulated during his last days at the front[845]. It will be remembered that they were (1) to endeavour to raise the siege of St. Sebastian by massing every available man on the Bidassoa, and striking at Giron and Graham—who was wrongly supposed to have only one British division with him. After driving back Graham as far as Tolosa on the high road, it would be possible to turn southward and relieve Pampeluna: Clausel’s corps, meanwhile, should demonstrate in Navarre, to distract Wellington from sending his reserves from the south to the main point of danger. Both Jourdan on July 5 and Soult on July 23rd believed, the former correctly, the latter wrongly, that there were three British divisions engaged in the blockade of Pampeluna.
(2) To leave a corps of observation on the Bidassoa to contain Graham, while the rest of the army struck at Pampeluna by the route of Roncesvalles. This would have the advantage of raising the siege of that place at once, though the relief of St. Sebastian would be deferred. But, Jourdan added, it had three disadvantages—the road was bad and might conceivably prove impracticable for artillery: the bulk of the British divisions would probably be found concentrated for a fight to cover the blockade: and Navarre was a country in which no army could scrape together food enough to live upon.
(3) To leave a corps of observation on the Bidassoa, and convey the rest of the army into Aragon, by the pass of Jaca, to join Suchet and operate with overwhelming forces on the Ebro against Wellington’s flank and rear. This, the operation which Jourdan regarded as the best of the three, had since become impossible, because Paris had abandoned Saragossa, and Suchet had taken his army to Catalonia: on July 17 he was at Tarragona.
Soult’s plan was in essence Jourdan’s second alternative; but he complicated it by dividing the army which was to strike at Pampeluna into two columns. Two-thirds of the whole (Reille and Clausel) was to march by the Roncesvalles passes; D’Erlon with the remainder was to force the pass of Maya and to converge on Pampeluna by the Bastan and the Col de Velate. Jourdan’s memorandum contained a special caution against the dissemination of columns, ‘je pense que toute opération de plusieurs corps isolés ne réussira pas.’ How frequently in military history two columns starting from distant bases have failed to meet on the appointed day is known to every student. It is seldom that a complete success is secured—as at Königgrätz. Much more frequently the enemy against whom the concentration was planned has beaten one corps before the other—delayed by one of the uncountable chances of war—has come upon the scene. And this was to be the case with Soult before Pampeluna.
The advantages of Soult’s scheme were very comprehensible. The greatest was that, owing to the existence of a first-class road from Bayonne to St. Jean-Pied-du-Port, he could concentrate the main body of his army upon his extreme left wing long before Wellington could make the corresponding counter-move. In the early days of the operation he could calculate on having an immense superiority of force on the Roncesvalles front. It was more doubtful what would happen on the third or fourth day, if Wellington divined his enemy’s plan at the first moment, and ordered a general concentration before Pampeluna, as Jourdan had taken for granted that he would. But Soult considered that he had found means to make such a concentration impossible, by sending D’Erlon and his 20,000 men to pierce the allied left-centre at Maya, and to sweep down the Bastan. If D’Erlon broke in on the first day, and got possession of the positions that dominate the road-system of the Bastan—Elizondo and the Col de Velate—the northern divisions of Wellington’s army would only be able to join the southern divisions in front of Pampeluna by immense détours to the west—by Santesteban and the Pass of Donna Maria, or even by Tolosa and Yrurzun. In all human probability they would arrive too late.
The British commander, as we shall see, did not divine Soult’s whole purpose at the first moment, and therefore his general concentration was ordered a day later than it might have been. And the officers who were in charge of his extreme right at Roncesvalles gave way quicker than he had intended. The campaign became a most interesting time problem, which was settled in favour of Wellington in the end, by the fact that his first-arriving northern reserves got to the positions in front of Pampeluna a day before Soult’s secondary column under D’Erlon came into touch with the French main body. The subordinate officers on both sides made some extraordinary mistakes, which compromised the plans of their superiors: the worst of them were in the line of slow and irregular transmission of news, which in a mountain country, where all side communications were difficult, and no general view of the situation could be taken by the commander’s own eye, sometimes led to ruinous miscalculations at head-quarters. It cannot be said that one side suffered more than the other from this negligence of subordinates.
To proceed to the details. Soult’s left wing at St. Jean-Pied-du-Port was ready to start, Conroux and Clausel’s old divisions having been cantoned in the neighbourhood of that fortress ever since they arrived in France; they were only one long march from their objective, the gap of Roncesvalles. At the other end of the line, D’Erlon’s three divisions were also within one long march of the Pass of Maya—they already lay with their vanguard at Urdax and their last rear brigade at Espelette. The difficulty lay with the third ‘lieutenancy’; Reille’s three divisions of the old Army of Portugal on July 20th were on the lower Bidassoa, holding the front opposite Graham and Giron; they had to be got to St. Jean-Pied-du-Port, to join Clausel and co-operate in the main attack[846]. To draw them out of their existing positions without arousing too much attention at Wellington’s head-quarters was comparatively easy. On the night of the 19th-20th four brigades of Villatte’s reserve replaced Reille’s troops at the outposts. On the following day the three fighting divisions concentrated at St. Jean de Luz and started to march eastward, with orders to cross the Nive at Cambo, and to get into the great chaussée by Urcaray and Hellette, which leads to St. Jean-Pied-du-Port. Drawing up a far too optimistic time-chart, Soult had hoped that they would all be in the neighbourhood of St. Jean-Pied-du-Port by the morning of the 22nd. As the distance was not less than 50 miles, this would have been very hard work in any case, for a long column marching with a large train of artillery, as Reille was bringing up beside his own pieces three batteries for Clausel, who had arrived gunless from Aragon. But the main cause of delay was quite different: though the Bayonne-St. Jean-Pied-du-Port chaussée was excellent, the local cross-road from St. Jean de Luz by St. Pée to the bridge of Cambo was not, and torrential rains on the 20th made it into a quagmire[847]. The troops got on very slowly, and at night the bridge at Cambo was carried away by a spate, leaving the bulk of Lamartinière’s division on the wrong side of the water. It was repaired; but the general result was that the head of Reille’s column only neared its destination late at night on the 22nd, while the rear division and the guns did not get up till the 24th. The Marshal had been waiting for them since the morning of the 21st, and had hoped to start operations on the 23rd. He had been joined by his two cavalry divisions, brought up from their cantonments between the Nive and the Adour on that day, so that Reille’s delays put the movement of all the other corps a day late.