Siege warfare is a rather neglected part of Napoleonic studies, since it is generally assumed that the great Emperor had 'revolutionised' the art of war to the extent that sieges were no longer important. In 1800 he had scornfully by-passed Fort Bard, on the Italian frontier, in order to strike deep into the heart of the country and fight a decisive battle at Marengo. If the campaign had been fought by some of his eighteenth century predecessors, it would doubtless have settled down into a long siege on the frontier, and stayed there.

This stereotype is misleading, however, since Napoleon was actually very interested in fortresses and siegecraft. As a gunner and the victor of Toulon, after all, he was fully conversant with the scientific aspects. In his 'classic' campaigns of invasion such as Austerlitz or Jena, moreover, he made it a rule never to move anywhere without having a strong fortification at least once in every four days' march along his line of communication (or Route de l'Armee.

These fortifications had many roles: as centres to control conquered areas or to collect supplies and convoys and defend them against enemy raids. They would also be rallying points for reinforcements moving up to the front, who would themselves normally provide the garrisons without taking soldiers directly out of the front line. And if the worst should happen, these forts would provide rallying points and depots for Napoleon's retreating army if it were defeated.

In the case of Wellington, however, the general impression is that he was still an eighteenth century type of commander -- taking everything very slowly and carefully, and placing great importance on siege warfare - but actually this too is very far from the truth. Wellington was in reality a very mobile and aggressive commander, and often a gambler (as Michael Glover, among others, has ably argued in his numerous Peninsular histories [1] ), Wellington was excellent at seizing fleeting opportunities to make an attack, which was an art he had learned in his Indian battles at Assaye and Argaum, and then carried forward to Oporto (1809), Salamanca (1812) and Vittoria (1813).

It was only in the period between Oporto and Salamanca -- or more accurately from Talavera, which was a failed offensive that followed soon after Oporto - that he was forced to adopt a more defensive posture for the security of Portugal, which has been called Wellington's 'Cautious System'. It is this period of almost three years in 1809-11 which has branded him as a defensive general in many people's eyes.

In fact, however, even during this 'cautious' phase he made several attacks on the French fortresses covering the exits from Portugal -- usually Badajoz -- but it was largely because he failed to win these sieges before 1812 that he was unable to burst out into the Spanish plains for more adventurous manoeuvres.

The trouble was that Wellington was really not much good at siege warfare. It was perhaps his greatest weakness as a commander. Just as he had learned the power of a sudden offensive in the field, in India, he had also found in that theatre -- at the sieges of Dummul, Arrakerry, Armednuggur and Gawilghur -- that any Indian fortress could be carried very quickly by a sudden storming.

Unfortunately the same was not true against the scientific French, and these 'colonial style' assaults often bounced off carefully-prepared defences manned by determined troops with knowledgeable commanders.

The Role of Siege in Napoleonic Warfare
The MYTH	The REALITY
Napoleon's deep invasions removed the need for fortresses.	Napoleon always cemented his L of C with fortresses.
The 'mobile' French didn't need to run sieges in either attack or defence.	They often did run them, and were very good at them.
The British had a "plodding 18th century-style" army, ie were good at sieges.	Wellington, at least, was poor in siege warfare (= more like the French image!).
The French used the column but the British used the line (in tactics).	Wellington used big columns repeatedly in his sieges, but the French engineered in lines.

In the Peninsula there were always grave difficulties of logistics and these became acute whenever it was necessary to collect the vast mass of materiel, artillery and ammunition required for a siege. The navy might quibble over the release of its guns for land service; the roads would be exceptionally tested by the exceptional demands of the siege train, and considerable powers of foresight would be required to predict the probable need for such items as shovels, gabions, scaling ladders or building timber.

There was also a sorry shortage of trained engineers, with the Corps of Sappers and Miners being created only towards the end of the war, after the Royal Military Artificers had been shown to be inadequate in both numbers and skills during the important earlier sieges. Engineer officers had a distressing habit of being hit by enemy fire at the very start of digging operations or, in the case of the great Badajoz storming, at the start of the assault they were intended to direct throughout its duration. [2]

In these circumstances Wellington often found that his strategic needs were too pressing to wait for the realistic speed of these siege operations, so he would be tempted to insist on dangerous haste. This in turn implied a willingness to expend the lives of his troops in the interests of a quick success, and there were some horrible scenes of carnage in the ditches at the base of ill-prepared breaches.

The third Badajoz siege in April 1812, in particular, is notorious less for the very heavy toll of 1,300 casualties lost during its preliminary engineering phase than for the 3,700 casualties suffered in the final storming. This total of 5,000 men -- the equivalent to an entire Division -- is the type of loss to be expected in a major battle; and after Talavera it was in fact the second longest butcher's bill that Wellington had ever suffered up to that point. [3]

Against this, it may perhaps be argued that Wellington made constant subconscious calculations about just how many lives he could afford to gamble, in order to win an important fortress in good time. No one can deny that he was a numerate general, just as he was certainly an experienced one.

Thus although he wept openly at the horrific scale of casualties in the eventual capture of Badajoz, he surely also realised in that moment that the plains of Leon lay open to him as a direct result of the sacrifice. The capture of Badajoz opened the road to Madrid for the first time since the strategic defeat of the Talavera campaign almost three years earlier, and although the victory of Salamanca would cost a further 5.200 allied casualties, it surely more than made up for the relative balance of losses as between the allies and the French.

If it is true that Wellington really did base his gambles in siegecraft upon some basis of subconsciously-perceived 'vital statistics', then it would surely be interesting to find out just what those statistics might have been.

In an attempt to find out, I looked up the classic Journal of Sieges in Spain by Sir John T Jones.' Within its pages there are details of all the British sieges between 1811 and 1814, although not quite all of them were Wellington's. Jones' selection includes Bergen op Zoom in the Netherlands, one in southern Italy and a couple on Walcheren island, as well as the French attack on Tarifa, near Gibraltar, which was the only Peninsular siege where the British (as opposed to their Spanish allies) were the defenders. However, I don't think it is illegitimate to include these 'non Wellington' cases, since their general conditions seem to have been very consistent with what Wellington himself was doing in his attacks on fortresses.

From these statistics we can see just how attractive it must have been to lay siege to a fortress, insofar as almost 40% of the forts attacked were pretty small and weak, and of these almost 80% surrendered before they were stormed at all.(See Table 1) Of the remaining forts that were 'big', a third was attacked by an army which seems to have been of adequate strength and fully prepared.' Apart from their superior artillery and engineer assets, these could normally mount three or four assault columns in the final storming, which was enough to swamp the defences in 80% of cases.

Hence the British can scarcely be faulted in the cases of attacks against small forts, or against big forts attacked by adequate armies - i.e. slightly more than a half of the total - and in fact the only 'big' fortress which ultimately survived attack by what I have called an 'adequate' army was Bergen op Zoom (ie Not one of Wellington's sieges); whereas the only 'small' fortress which survived attack was Tarifa (when it was the French attacking a British garrison). So far so good the casualties in these assaults may have been horrific, and far more than Vauban would have been happy to accept - but the final strategic success represented by the capture of the fort surely goes a long way to absolve Wellington of serious blame.

The real problem comes when we turn to the 9 cases when big fortresses were attacked by small armies. Wellington is shown in his worst light whenever this happened, and in the first attempt against Badajoz in May 1811 he was so slow to develop a convincing attack that Soult's relieving army was able to arrive before a storming could even be mounted. At Pampluna in July 1813 Wellington decided on a blockade rather than any sort of siege, although he betrayed his frustration by issuing the most blood-curdling instructions that his Spanish allies should hang the entire garrison if they should ever manage to catch 'em.

Soon after this Wellington's frustration came to boiling point when Graham's ill-prepared first effort to storm San Sebastian ended ignominiously, just at the moment when Marshal Soult (once again!) was dealing a shrewd surprise blow across the inland mountains.

Overall, only in 22% of the 'Small Army vs Big Fort' attacks was the fort finally captured, which is an almost diametrically opposite to the result from the 80% success of 'Adequate Army vs Big Fort' attacks. The need for numerous attack fronts is certainly confirmed (See Table 2) when we analyse the stormings in terms of the number of assault columns used in each case.

Thus we only have 2 examples of stormings against small forts but, it is notable that one (at Almaraz) was successful when it used 3 columns - thereby splitting the defender's attention and fire while the other (at Tarifa) failed in a single column. The same is true with the 'big forts attacked by adequate armies', where 80% of the assaults were in 3 or 4 columns - although admittedly the one attack in a single column (at San Sebastian) was also a success. Equally the generally unsuccessful cases of 'big forts attacked by small armies' were all in only one or two columns, and showed a poor record of success.

Let us consider the success of each individual column .(See Table 3: Success of each Column): The very difficult nature of actual stormings is shown by the fact that over half of them ended in total failure, and less than 30% of the assaults were 'completely' successful against big forts. Even in multiple assaults which together led to the capture of the fort, the majority of the columns might fail and even, as 'Feints', may have been excpected to fail.

Finally, let's look at the casualties suffered within each of the storming columns. Contrary to our expectations, we find that the least successful assaults tended to suffer relatively light casualties, whereas the more effective attacks were often by far the bloodiest. Perhaps this shows that a failed attack tended to be over quickly although in several cases, such as the second siege of Badajoz, the storming party stayed in the ditch for a full hour, trying to find a viable escalade, before it retired. It was certainly the most determined troops who pressed on into danger and paid the price, and particularly in the case of 'partial' successes, which tended to be bloodier than 'complete' ones.

NOTES