SUCHET ON THE EBRO. THE FALL OF TORTOSA.
DECEMBER 1810-MARCH 1811
In the last pages of the third volume of this work we brought the history of the campaigns of Suchet and Macdonald in Aragon and Catalonia as far as December 12th, 1810, the day on which the Marshal came down to the lower Ebro at the head of the field-divisions of the 7th Corps, in order to cover the long-delayed siege of Tortosa, which his colleague, the commander of the Army of Aragon, was about to take in hand.
For five months, ever since August 1810, as it will be remembered, Suchet had been waiting to commence the attack on Tortosa, which he was not strong enough to conduct with his own resources, since the besieger would have not only to execute his own operations, but to fend off at the same time all attempts by the Spanish armies of Catalonia and Valencia to relieve a fortress which was equally important to each of them. For Tortosa commanded the one land route by which Catalonia and Valencia were still able to communicate with each other; no other bridge on the Ebro was in Spanish hands. It was highly desirable to keep the line of defence in eastern Spain unbroken, and Tortosa lay at its narrowest and most dangerous point. The Emperor Napoleon attached an immense importance to its capture, and considered that after its fall, and that of Tarragona, the conquest of Valencia and the termination of the war on this side of the Peninsula would be at hand. Already in September 1810 he was looking forward to all these events as matters of the near future[290]. But though Tortosa was actually to be captured in January 1811, Tarragona in June, and Valencia in December, the long-foreseen consummation was never to come to pass. For two years more, down to the very end of[p. 228] the war, the indomitable Catalans were destined to maintain their independence, despite of all the successes of the French arms.
Tortosa was not quite so indispensable in reality as it looked on the map. For though it would have been a point of absolutely vital importance to the Spaniards, if they had been compelled to conduct all their operations by land, it must not be forgotten that the British command of the Mediterranean gave an alternative route for communication between north and south, which the French could never touch. Long after Tortosa had fallen, troops were repeatedly and easily transferred from Catalonia to Valencia, and vice versa, despite of the fact that Suchet had completely mastered all the land routes. Indeed it may be said that the sea passage from one to the other was always preferable to the route through Tortosa, for the country about both banks of the lower Ebro was rugged, barren, and thinly peopled, and Tortosa itself—a decayed city of 10,000 inhabitants—was the only place in this region which presented resources of any kind for an army on the march. It is, moreover, twenty miles from the sea, and not a port; strange to say, there is no decent harbour at the mouth of the Ebro. Though the river is intermittently navigable for a good many miles in its lower course, its estuary has never been a point through which trade discharges itself to the sea. The commerce of southern Catalonia goes to Tarragona, sixty miles to the north of the Ebro mouth, that of Northern Valencia to Peniscola, thirty miles to its south. Tortosa, indeed, was more important to the French than to the Spaniards, since Napoleon’s armies were tied down to land routes, and, if ever they were to make a lodgement in Valencia, would find the possession of that city and its bridge absolutely necessary, if they were to keep in touch with the troops left in Catalonia.
At last, in December 1810, Macdonald had accomplished the preliminary duties which rendered it possible for him to move on Tortosa. He had revictualled Barcelona, so that it would be safe for some months from famine, and he had repaired the gap in the French lines in northern Catalonia, which had been caused by Henry O’Donnell’s daring expedition to La Bispal in the preceding September. Though the Spanish army had not been[p. 229] crushed, nor indeed seriously injured, it had been thrust aside for the moment: Macdonald had brought down three divisions, over 15,000 men, to Mora on the Ebro, some twenty-five miles to the north of Tortosa, and with them he was prepared to act as a covering force to the projected siege. He was strong enough to render any attempt at relief by the Catalan army impossible, for a force sufficient to beat him could not be collected by the enemy, unless they abandoned all their outlying posts; and this was practically impossible, for local feeling in the Principality was too strong to allow of the withdrawal of the smaller Spanish detachments from the various valleys and small towns which they covered. There was still the Valencian army to be considered; but Suchet considered that he could himself deal with that unfortunate and oft-defeated force; he had already proved its weakness at the series of engagements in August[291] in which he had so effectually scattered the levies of the dictator José Caro. One division would probably suffice to keep the Valencians at bay, while with two others the actual siege of Tortosa could be taken in hand. It was the Catalan army alone which had made Suchet anxious, and he was now to be relieved of all care on that side by Macdonald.
The preparations for the siege, it will be remembered, had long been in progress; ever since August magazines and material had been accumulating at Xerta. They had been brought down the Ebro from Mequinenza, whenever the water was high enough to allow of navigation—not without much difficulty, and occasional petty disasters when the Catalan miqueletes made a pounce upon an exposed convoy[292]. Fifty-two heavy guns were now lying ready at Xerta, with 30,000 rounds of ammunition for them, and 90,000 lb. more of powder. To set in motion the besieging army and this very large battering-train, it only required that the covering forces on each side of the Ebro should take up their positions. When Macdonald had brought up his troops to Mora, and undertaken to move them to Perello, the junction point of the two roads from Tortosa to Tarragona, nothing more remained to be settled. If the Catalan army should try to attack him it would certainly be beaten, being far too weak[p. 230] to face 15,000 French troops concentrated in one body. For the restraining of the Valencians, on the other side of the Ebro, General Musnier with 7,000 men was placed at Uldecona, twelve miles beyond Tortosa on the great road to the south.
These arrangements being made, Suchet crossed the Ebro at Xerta by his pontoon bridge on December 15th with twelve battalions, and made a sweeping circular march to shut in Tortosa on the northern side, while five battalions more, under General Abbé, moved by the other bank to block the bridge-head, by which the city communicates with Valencia and the south. The operation was completed without resistance, save at the Col de Alba, where a post of 600 Catalans, placed to cover the road between Tortosa and the pass of Balaguer, was discovered. This small force was driven into Tortosa after a skirmish; two small convoys on their way from Tarragona by sea had to turn back, because the way into the city was closed.
The siege of Tortosa was remarkable for its swift progress and complete success—it can only be compared with Wellington’s capture of Ciudad Rodrigo, just a year later, in this respect. The Army of Aragon arrived in front of the city on December 16th—the surrender took place on January 2nd. At first sight the problem set before Suchet appeared by no means likely to receive such a rapid solution, for Tortosa as a fortress had many strong points. The city lies on and around the end of a spur of the Sierra de Alba, which runs down to the bank of the Ebro. This spur at its termination is many-headed: three ravines divide it into four separate hills, of which the highest is that crowned by the castle. The other three die down into undulations, and leave between them and the Ebro a comparatively flat space, on which the lower part of the town is built. Its higher quarters lie on the slopes where the various hills begin to rise from the level. The old enceinte of Tortosa had consisted of a mediaeval wall running across the hills and down the ravines between them, so as to enclose nothing but the inhabited parts of the city. But after a famous siege during the War of the Spanish Succession (1708), a number of outer works had been constructed to crown the culminating summit of each hill, and keep future besiegers at a distance, while part of the inner ancient fortifications had been transformed into a series of bastions, so far as the ground allowed.
[p. 231]The outer works were: (1) Starting from the river, on the extreme north-east front, a large hornwork called Las Tenazas[293], crowning the hill which lies most to the left, protecting the suburb of Remolinos, and commanding the flat ground as far as the river, from which it is only 400 yards distant. (2) A second hornwork covered the outer side of the castle hill: this was called El Bonete; there was a deep ravine on each side of it. (3) A sort of outer enceinte with three bastions (La Victoria, El Cristo, and Cruces) runs along the summit of the fourth hill, which is broader than the rest and expands into a plateau; the old town wall formed a second line in rear of this advanced front. (4) The last hill, that most towards the south-east, was crowned by a strong closed fort, named after the Duke of Orleans, the general who had captured the city for Philip V in the year 1708. Between the Orleans hill and the river there was no high ground; this was the only open front of the city which was approachable on a level without any natural hindrances. Here the enceinte had no outer works; it consisted of two large bastions, San Pedro and San Juan, with a demi-lune named El Temple projecting between them. This last was a low work of no great strength: the curtain behind it, which joined San Pedro and San Juan, was a mere shell with no terrace or room for guns. The other (south-western) side of the city was sufficiently protected by the broad and swift river, 300 yards in width; behind it was only the ancient wall, but, as this was wholly inaccessible, its weakness did not matter. In the middle of the river front, which was 1,200 yards long, was a gate leading to the great bridge of boats. This was protected on the further bank by a little tête-du-pont, built in the form of a ravelin and well armed with artillery.
The garrison consisted, when the siege began, of 7,179 men, including 600 artillery and a weak battalion of Urban Guards. The regiments were drawn partly from the Catalonian and partly from the Valencian army, there being seven battalions from the former and four from the latter[294]. The governor was[p. 232] Major-General Lilli, Conde de Alacha, an old officer who had won some credit, two years back, by bringing off his small corps intact after the battle of Tudela, when he had been cut off by the French in the mountains[295]. Unfortunately it is impossible to argue that because an officer has made a skilful retreat he is a good fighting-man. Lilli, indeed, proved the reverse on this occasion—all that can be said in his defence is that he was old and in bad health. His conduct during the siege was vacillating and inexplicable; he more than once declared that he would give over charge of the defence to the second in command, Brigadier-General Yriarte. But he would then appear again, and countermand all Yriarte’s orders. In the end he capitulated on his own account, against the wishes and without the knowledge of the brigadier.
The problem in poliorcetics set before Suchet and his engineers at Tortosa had some resemblance to that which was to confront Wellington at Badajoz a few months later. Given a town to besiege which is partly built on heights crowned by forts and partly in the flat, is it better to attack one of the forts, which if taken commands the whole town, or to start against the lower front? The latter will be easier to assail, but after its fall the strongholds on the heights may hold out as independent fortresses. Whereas if a dominating work on a high-lying hill is taken in hand, it may make a very hard and long resistance, but, if once it is captured, the whole city is overlooked by it and must surrender. The answer to the problem seems to be that all depends on the governor and the garrison; if he and they are weak, it will suffice to attack the easier front, for when the lower town has fallen the morale will be so shaken that they will surrender, without attempting to hold the upper works. But if they are active and obstinate, it may be worth while to aim[p. 233] at the most commanding position, which, if captured, absolutely compels the capitulation of the whole place. Yet there remains the danger that the crucial fort may be so strong that no effort can take it from a determined defender. This happened to Wellington at the two earlier sieges of Badajoz; the fall of San Cristobal would have brought about most inevitably the surrender of the city. But it proved too hard a nut to crack, in the time and with the means at the English general’s disposal. Suchet at Tortosa struck at the easiest front, taking the risk that he might have to conduct a second siege of the upper works (as he had to do at Tarragona, his next leaguer). When he had breached the weakest point, but before a storm had been tried, the imbecile governor capitulated.
The weak front of Tortosa was the bastion of San Pedro and the demi-lune of El Temple on its flank. They were not properly supported or flanked by any works on higher ground, for Fort Orleans (on the nearest hill) did not command all the flats in front of San Pedro, and moreover Suchet intended to give this fort so much business on its own account that its gunners would have little attention to spare for the attack on the lower ground to their right. There were two additional advantages: the soil in front of San Pedro was soft river mould, very easy to dig; and moreover this bastion could be enfiladed by batteries on the other side of the Ebro. Such batteries had nothing to fear, if properly constructed, from the guns of the little tête-du-pont; while on the river front of the city there were no cannon at all—they could not be mounted on the mediaeval walls.
On the 16th, 17th, and 18th Suchet was employed in bringing up his siege-guns, choosing the emplacements of his camps, and constructing flying-bridges across the river, both above and below the fortress. On the 19th active operations started, with the development of a false attack on Fort Orleans, whose attention it was necessary to distract. The construction of a first parallel against this work was begun with some ostentation, and had to be carried out under a furious fire from its artillery. On the night of the 20th-21st the real business commenced: 2,300 men crept across the flat ground by the river opposite San Pedro, and threw up an entrenchment within 160 yards of its ditch. They were undiscovered and unopposed, for the night[p. 234] was dark and windy, and the Spaniards had no outposts beyond the walls, and kept bad watch. At dawn 500 paces of trench had been constructed, and a safe access to the parallel had been contrived, by connecting one of its ends with a ravine which cuts across the flat a little to the rear, and whose bottom could not be searched by the guns of the place. The trench was sloped away on the right, so as not to be enfiladed by downward fire from Fort Orleans, which would have commanded it supposing it had been drawn exactly parallel to the front of San Pedro. This was a tremendously advantageous start; seldom has a besieger been able to begin his works at such a short distance from his objective.
Next morning the new trench became visible to the Spaniards, who turned all the artillery of the neighbouring front upon it, but to little effect, for the soil was soft and the French had dug deep. A sortie was made from the demi-lune of El Temple, but was driven off with loss. The only successful effort of the Spaniards on this day was that the guns of Fort Orleans succeeded in destroying part of the trenches of the false attack in front of them, and drove out the workers. This was of little consequence, as Suchet was not really aiming at the fort. On the 22nd and 23rd December the main attack was urged with a celerity that seemed appalling to the defenders; despite of a heavy fire of musketry as well as of artillery, approaches were pushed forward from the first parallel, to within 80 yards of the bastion of San Pedro and 110 yards of the Temple demi-lune. The works opposite Fort Orleans were repaired and extended. On the 24th the two approaches in the plain were connected by a long trench, which formed the second parallel, only 60 yards from the walls. On the 25th new zigzags, thrown out from this line, reached the glacis of San Pedro. The garrison made two sorties in the night to hinder this advance at all hazards, but failed, being driven off by the musketry from the second parallel without much difficulty: the force employed, 300 men, was too small. Meanwhile the artillery and sappers were constructing ten batteries, four—and these the largest—in the main frontal attack, but four others on the heights over against Fort Orleans, and two beyond the river, whose special purpose was to enfilade the front of San Pedro,[p. 235] and also to play upon the bridge of boats which joined the city to the tête-du-pont. The trenches on the heights before Fort Orleans were also strengthened, and a second parallel constructed in front of the first, at some loss of life; but it was necessary to keep the enemy on this front employed, or he would have interfered too much with the real attack in the plain.
All this had been accomplished before the French had fired a single gun; there is hardly another instance to be quoted of a siege in which the assailants got up to the glacis, and crowned the covered way, without any assistance from their artillery. It seems that the gunnery of the defence must have been exceptionally bad, and the sorties hitherto had been small and feeble—quite insufficient to interfere with, much less to destroy, the approaches. Noting batteries sketched out on several points and approaching completion, Yriarte saw that he must at all costs try to delay the opening of the adversary’s fire, and on the night of the 27th-28th made two sallies in considerable force—600 men came out of the Rastro gate, beyond Fort Orleans, to attack the upper trenches, as many more issued from San Pedro against the main attack. The first-named sortie was an entire failure—most of the men lay down and began to skirmish with the trench-guard before they had got near the works; a very few reached them, and were killed on the parapet. But the attack in the plain was a serious one, and pressed home. The lodgement in the covered way was captured and destroyed, and the Spaniards penetrated to the second parallel, and captured part of it for a time. They were finally driven out by a reserve of four battalions commanded by General Abbé, but not before they had done considerable damage. The besiegers had to spend the following day and night (28-29 December) in repairing the trenches and parapets, and getting a fresh lodgement in the covered way. On the morning of the latter day the ten siege-batteries opened simultaneously with 45 guns, and very soon gained a marked superiority over the fire of the defence. The cannon of Fort Orleans and the bastion of San Juan were silenced, as were also those of the Temple demi-lune and San Pedro. The bridge of boats was nearly destroyed. Next night, the Spanish fire being crushed, the third parallel was[p. 236] constructed on the very brink of the ditch of San Pedro, and within 25 yards of the wall of the bastion. The mortar batteries were employed in distributing a rain of projectiles in the streets behind the attacked front, to prevent the besieged from constructing defences and barricades on which they might fall back when the wall was breached. Though not altogether successful—for the Spaniards succeeded in building some traverses and in blocking and loopholing many houses—the bombardment caused many casualties, and cowed the population, who evacuated this quarter and sought refuge in the interior of the town. The interior of Fort Orleans was also shelled with some effect: its garrison retired to their bomb-proofs, and kept very quiet, making little attempt to repair the injuries to its outer wall, or to replace the injured cannon.
On the night of the 30th the French succeeded in getting down from the third parallel into the ditch of San Pedro, with the object of mining the scarp and blowing down sufficient debris to fill the ditch. Their first party was driven out again by the fire of two guns which had been brought up to enfilade the ditch from the extreme flank of the bastion. But on the following morning all the siege-batteries were turned on to these guns, and destroyed them. Meanwhile the Spaniards had abandoned the tête-du-pont, taking the men away by boat, and throwing the guns into the water, except three injured and spiked pieces. The ditch was occupied during the day (December 31st) and the miners got to work, so little incommoded by the fire of the defenders, who were hardly visible on the wall, that they lost only two men killed while establishing themselves in their dangerous position. Their most serious hindrance came from the good quality of the masonry which they were attacking—it was mediaeval work and as hard as iron. The decisive stroke at this point, however, was to be given by the artillery, and on the night of the 31st a battery for four 24-pounders was commenced in the third parallel—only 25 yards away from............