When Wellington’s forces appeared before the 30-foot-high bastions of Badajoz in May 1811, it was the first time in almost two decades, stretching back to 1793 when Britain first entered the war against revolutionary France, that a British army had begun a major land-based siege operation against a fortress town in Continental Europe. Wellington’s army, however, would more than make up for this in the ensuing years, by becoming involved in a series of large-scale sieges over the final three years of the Peninsular War. This would prove to be a brutal and often horrific return to the stage of European siege warfare, for both British soldiers and Spanish civilians alike. For the majority within Wellington’s army, too, Spain represented their first direct experience of siege.

Yet as new as siege warfare was to most of Wellington’s officers and men, they were nevertheless engaging in one of the oldest and most enduring forms of warfare, with long-established forms, customs and rituals. The fundamentals of both fortification and siege operations in the Napoleonic era were still very much anchored in the high age of European siege warfare in the first half of the eighteenth century. And the customary laws of war governing sieges were handed down from one generation of soldiers to the next, although not without evolving interpretations, practices and sentiments.

Spanning the history of siege warfare across the long eighteenth century, this chapter sets out important groundwork and context for the chapters to follow. The first half outlines the nature of ‘the siege’ as the classic form of old-regime positional warfare – its operational forms, temporal and spatial dimensions, and rituals and customary laws of war – and charts the relative historic decline of breach assaults and siege-related massacres in the eighteenth century. The second half shifts to the Revolutionary-Napoleonic era. It identifies the continuing importance of siege operations beyond their earlier high point in European military affairs; introduces our key British case studies in the colonial world; and ends with an overview of British and French sieges in the Peninsular War – the epicentre of Napoleonic siege warfare and sack.

Siege and Its Forms

From c.1680 to the mid-eighteenth century, sieges dominated European warfare to a degree rarely if ever matched in European history.Footnote ¹ This was the golden age of sieges, especially during the long wars of Louis XIV.Footnote ² The War of the Spanish Succession alone (1701–14) saw at least 115 sieges staged in the Low Countries, northern France, the Rhineland, Italy and the Iberian Peninsula.Footnote ³ From the British perspective, this war also represented an unprecedented military commitment to Continental Europe, with the British taking a ‘greater proportional role’ in the conflict than in any other during the long eighteenth century, including the Napoleonic Wars.Footnote ⁴ This commitment included a central role in the siege warfare conducted in the Low Countries and Spain, exemplified in the military career of the Duke of Marlborough, who not only commanded four major battles but oversaw about thirty sieges in the northern theatre.Footnote ⁵ Three decades after the Sun King’s death, sieges still remained fundamental to European military campaigns, playing an integral role in the War of the Austrian Succession (1740–48). Alongside Maurice de Saxe’s celebrated victory over the British at the Battle of Fontenoy (1745), French armies besieged and captured numerous fortresses in the Low Countries, Saxe writing of ‘that rage for sieges which prevails at present’.Footnote ⁶

The broad operational form of sieges in this period reflected changes in artillery and fortifications over the previous centuries. With advances in gunpowder and artillery ultimately rendering obsolete the high and relatively thin walls of medieval castles, a new type of fortification evolved in the fifteenth and sixteenth centuries – the trace italienne – which spread northwards. Ultimately, an entire new geometry of ‘military urbanism’ emerged. This reached its apogee in the late seventeenth and early eighteenth centuries under the two master military engineers and rivals, the Dutchman, Menno van Coehoorn, and the Frenchman, Sébastien Le Prestre de Vauban, the latter designing the famous double line of fortresses that protected France’s northern borders.Footnote ⁷

The basic model of an eighteenth-century fortified town was as follows. The surrounding main curtain walls comprised thick earth ramparts, with parapets on top. Bastions extended outwards from the walls, their faces and flanks providing wide angled fields of artillery fire. A wide ditch encircled the walls and bastions: the inner face (scarp) was lined with bricks and stones, the outer face known as the counterscarp. Beyond the ditch was a ledge or walkway (covered way), where infantry could form the first line of defence, protected by a palisade and the embankment of the rim of the glacis, the sloping open ground that formed the outer perimeter. Fortress towns could also include a formidable bastioned citadel, and various outworks might be added to complete the defences.Footnote ⁸

Whilst ‘open’ towns or those with relatively weak defences could be attacked and captured almost immediately by surprise, stealth, storm or escalade (the use of ladders over walls), a strong fortress normally had to be besieged. Derived from siège (seat or chair), to lay siege in the military context meant to ‘sit down before a place’ and to ‘chuse a position from which you may commence the necessary operations to attack and get possession of it’.Footnote ⁹ The classic form of European positional warfare against a fortress constituted a regular siege, or what was referred to in the early eighteenth century as the ‘siege in form’. No single person did more to establish and regularise this than Vauban, Louis XIV’s leading military engineer, who directed at least forty-eight sieges and wrote what became the eighteenth century’s most influential siege manual (De l’attaque et de la défense des places, first published in 1737).Footnote ¹⁰ Long after his death in 1707, notwithstanding ongoing debate, modifications, and differing military engineering traditions, Vauban continued to be recognised as the single greatest authority on fortifications and siege craft, his methodology studied in military colleges for artillery and engineering officers throughout Europe, including the Royal Military Academy at Woolwich, established in 1741. There, John Muller presided as ‘Professor of Artillery and Fortification’, producing his own siege manuals from the 1740s, intended for both British officers and general readers, in which he openly acknowledged his great debt to Continental masters, especially the ‘celebrated Vauban’.Footnote ¹¹

The classic methodology of a regular siege began with the attacking forces surrounding the fortified town (investment) and establishing inward- and outward-facing defences (lines of circumvallation and contravallation, respectively).Footnote ¹² The key moment was ‘breaking ground’, digging the first of a series of trenches running parallel to the defensive walls. With the establishment of the first parallel, siege batteries were set up, normally concentrating their fire on the fortress guns. After the first parallel, the attacking forces would then advance by means of sapping, digging an approach trench that often followed a zigzag pattern to prevent enemy enfilade fire. Once the sapping had advanced far enough, a second parallel trench was dug, and the breaching batteries advanced accordingly. A third parallel might then be dug. The ultimate goal was to get the breaching batteries as close as possible, ideally to the very rim of the covered way, to concentrate short-range fire on the walls to create practicable breaches – that is, an opening wide enough to allow assault parties to enter – thus precipitating either capitulation or a general storm.Footnote ¹³

This, then, was the ideal formal siege of eighteenth-century Europe, dependent upon heavy-calibre cannon and technical expertise. But regular sieges varied in scale and duration, from minor to major military events, and could last from a few days to many months. Not all regular sieges in practice adhered as strictly to the same methods, nor advanced as far along the operational march as others. And commanders improvised relative to their resources and the circumstances. Regular sieges were more common in Western Europe – especially in the fortress-dense regions of northern France, the Low Countries and northern Italy – than in Central and Eastern Europe. In the Silesian Wars of the 1740s and the Central European theatres of the Seven Years War (1756–63), the state of roads, fortresses and engineering corps did not favour highly technical sieges as in the West.Footnote ¹⁴ Whilst Frederick the Great was well versed in classic fortification and siege theory, and built new fortresses and strengthened existing ones, his practical siege record, with few exceptions, was characterised by limitations and shortcomings: blockades and bombardments, bluster and abandonment, and a general distrust of engineers.Footnote ¹⁵ The slow march of a formal siege did not suit Frederick’s disposition nor vision of war as one of rapid manoeuvre and field actions.

It follows that the term ‘siege’, as used by contemporaries, had ‘various grades’ of meaning, from formal siege operations to blockades and containments.Footnote ¹⁶ Indeed, eighteenth-century Europeans often distinguished between regular sieges and blockades, and between sieges and bombardments.Footnote ¹⁷ This remained part and parcel of the conception and language of urban positional warfare in the Napoleonic era. Charles James, in his Military Dictionary for British military officers, which passed through four editions between 1802 and 1816, included separate entries for ‘siege’, ‘blockade’ and ‘bombardment’.Footnote ¹⁸ At one level, this was no mere semantic splitting of hairs in the age of dictionaries, as each represented a distinct tactical form of positional warfare in its own right. A blockade meant surrounding a town and cutting off its supplies, relying on slow strangulation over time to bring about submission through starvation or disease. On the other hand, a ‘bombardment’ sought to bring about the surrender of a fortress town as quickly as possible: artillery, mortars and sometimes incendiaries were fired into the town itself, often deliberately terrorising and targeting civilians. Bombardment might be used if the attacking forces lacked not only the necessary time but also the specialised skills and equipment necessary to lay down a formal siege. Bombardment was also the method favoured by joint army-navy operations to reduce coastal towns and fortresses.

Nevertheless, there is also a danger here in making too fine and consistent a distinction relative to the operational realities of war. Whilst blockades and bombardments existed as distinct and singular approaches to bringing about the fall of a town in their own right, without recourse to a formal siege, commanders could switch between forms, from a regular siege to a blockade or vice versa. More importantly, though, the lines between these tactical forms often became blurred and fluid in practice. Commanders might simultaneously deploy siege batteries against town walls and bombardment over the walls to hasten surrender; and starvation and disease were invariably part and parcel of long sieges, irrespective of whether they were secondary to, or merely a consequence of, formal siege operations.

The Laws of War: Surrender and Sack

Sieges represented not only the most structured form of eighteenth-century warfare but also the most highly ritualised and rule-bound. Sieges were conducted according to rules and norms that had evolved over the centuries. Both the besiegers and the besieged operated within a framework shaped by traditions of natural, divine and military law, and by a pan-European military honours system grounded in chivalric traditions and aristocratic martial culture. In an age before binding international treaty law, these unwritten rules, commonly referred to as ‘laws of war’ by contemporaries, were mutually recognisable and culturally self-regulated, underpinned by the principle of reciprocity.Footnote ¹⁹

Time-honoured protocols marked out the changing tempos and rhythms of ‘siege time’. Before a siege commenced, it was customary for the attacking commander to summon the military governor of the fortress town as to whether they wished to negotiate capitulation terms. A governor would typically confer with a ‘council of defence’, which included senior officers and sometimes municipal representatives. The initial offer to treat was nearly always declined, with the siege then progressing. After the initial summons, the attacking commander would typically extend further offers to treat as the siege unfolded, but the longer a town held out, the weaker its negotiating position became, with the likelihood of generous capitulation terms diminishing, setting in train a process of escalation. The critical juncture of a siege, and most dangerous phase for the garrison and civilians, was when a practicable breach was made in the outer walls. This conventionally signified a defence taken to the ‘last extremity’. The defenders then faced a stark choice: capitulate, or fight on, with the latter decision running the very high risk, if the defences were overwhelmed, of the garrison being put to the sword and the town sacked, including the murder and rape of civilians.

For the governor and garrison, military honour and duty were of paramount importance in how they conducted their defence and at what stage they surrendered.Footnote ²⁰ A governor who easily lost a fortress, or who capitulated relatively early in a siege, ran the risk of accusations of dereliction of duty or premature surrender, bringing with it dishonour and even the prospect of execution on the orders of an aggrieved sovereign. This happened on rare occasions in the eighteenth century: Frederick the Great, for instance, had the commander of Glatz tried and executed after the fortress was taken by surprise assault by the Austrians in 1760.Footnote ²¹ An honourable defence was generally considered to be one of brave and determined resistance that lasted until the defensive structures became untenable, or when the ammunition or food supplies ran out, or when a practicable breach had been made – and there was no hope of relief arriving at the eleventh hour. A governor who wished to treat would signify this by ‘beating the chamade’ and raising a white flag. Representatives of the opposing commanders were then empowered to negotiate capitulation terms. The final capitulation document, signed by both parties, set out in a series of articles the terms accorded to the garrison and the inhabitants, and what would happen at the handover of the town.Footnote ²² Civilians were normally guaranteed that no harm would befall their persons or property when the enemy took possession. For garrisons, the best outcome was to depart the town with full military honours, bearing arms amidst much fanfare and ceremony, with all private possessions in tow, and free to fight another day upon reaching a designated town or fortress, although officers sometimes served a period of parole before returning to arms. The worst surrender outcome was to be made prisoners of war, which involved the garrison evacuating the town with partial or no honours bestowed, and ceremonially surrendering their arms.Footnote ²³

But a horrific fate could befall garrisons and towns if they refused to surrender at the point of a practicable breach and the besiegers were forced to storm the breaches. The act of storming was the pivotal threshold moment, a grey zone where the fate of both the garrison and the inhabitants hung in the balance. If the town fell to a general storm, the garrison and the inhabitants forfeited their protective rights under the laws of war. Under these circumstances, customary law permitted two practices: putting the garrison to the sword and sacking the city at the expense of civilian property and lives. Whilst garrison troops might manage to take refuge in a citadel, or attempt to ‘surrender at discretion’, effectively throwing themselves on the mercy of the conqueror, civilians had neither of these military options available to them, their bodies and property fully exposed to the stormers.

These practices were traditionally justified on a number of grounds. First, in refusing to surrender in the face of a practicable breach, the defenders, including any civilians who had assisted in the defence, were deemed responsible for the needless loss of life that the besieging forces incurred in storming, thereby legitimising retaliation or vengeance on the part of the attackers. Second, it was seen as a form of collective and exemplary punishment, a political calculation that was designed to terrorise future garrisons and towns into surrendering sooner rather than later, thus sparing the besiegers the precious commodities of both men and time. Third, plundering the town acted as an incentive, reward and indemnity for the losses and hardship that soldiers endured in being forced to take the place by storm.Footnote ²⁴

These customary laws of war were as old as siege warfare itself, acknowledged with recourse to both biblical and classical examples by numerous early modern European jurists, from Francisco de Vitoria (1485–1546) and Alberico Gentili (1552–1608) to Hugo Grotius (1583–1645).Footnote ²⁵ As Grotius noted on the granting of quarter: ‘In besieged Towns, the Romans observed this Custom before the battering Ram struck the walls.’ This principle, ‘is still observed, viz. in weak Towns, before the playing of the Batteries; and in fortified Cities, before the giving of a Storm’.Footnote ²⁶ If the storm was successful, sack invariably followed in its wake. As Gentili put it: ‘Cities are sacked when taken; they are not sacked when surrendered.’Footnote ²⁷ Whilst jurists stressed restraint, morality and proportionality in siege warfare, they nevertheless upheld the principle of strategic necessity. In such circumstances, conceded Vitoria, sacking a besieged town was ‘not unlawful’, even if it gave licence for soldiers to ‘commit every kind of inhuman savagery and cruelty, murdering and torturing the innocent, deflowering young girls, raping women, and pillaging churches’.Footnote ²⁸

Yet, the very fact that sieges were so rule bound can sometimes lead historians to assume that attitudes to these customary laws of war were fixed and unchanging over time. Robin Clifton, for example, writing of Oliver Cromwell’s sacks of Drogheda and Wexford in 1649 asserts that: ‘As late as the nineteenth century, and to a commander as eminent as the Duke of Wellington, these rules were accepted as normal, reasonable and lawful.’Footnote ²⁹ Moreover, sack is often taken as a historic constant, Philip Dwyer writing: ‘The sacking of towns, during which soldiers committed murder and rape … was part and parcel of eighteenth-century warfare.’Footnote ³⁰ Certainly, one needs to distinguish here between towns that were sacked at the end of a siege and poorly protected or largely undefended towns or villages that were almost immediately attacked and subjected to plunder and atrocities by armies or troops in other military contexts – an important distinction we shall return to later. But the sack of towns in the context of sieges was very rare in the eighteenth century – certainly in Western Europe. Indeed, tracing the genealogy of sieges over the early modern era reveals changing norms and patterns of constraint and excess.

Restraint and Violence in Old-Regime Sieges

Relative to eighteenth-century siege warfare, the massacre of garrisons and the sack of towns was much more common to sieges of the sixteenth and seventeenth centuries. Sack was a feature of the Italian Wars in the first half of the sixteenth century.Footnote ³¹ The Eighty Years War between the Spanish and the Dutch saw numerous sieges that ended in garrisons being put to the sword, or towns sacked, or both – especially at the hands of the Duke of Alba.Footnote ³² When Haarlem surrendered in 1573, Alba, despite previous assurances to the contrary, ordered the summary execution of the garrison: 2,000 soldiers had their throats cut, and the remainder were drowned in the river.Footnote ³³ Within the English context, the most notorious case was Oliver Cromwell’s sack of the Irish town of Drogheda in September 1649 during the Civil Wars. An estimated 3,500 soldiers and civilians were massacred, the majority of victims being garrison troops who were shown no quarter during the storm itself or summarily executed in cold blood afterwards.Footnote ³⁴ Between 1,500 and 2,000 soldiers and civilians were then massacred by Cromwell’s troops the following month at the siege of Wexford.Footnote ³⁵ But it was the Thirty Years War in Germany, which witnessed the seventeenth century’s greatest siege atrocity: the Imperial army’s sack of the Protestant city of Magdeburg in May 1631, which claimed an estimated 20,000 lives and left the city a smouldering ruin.Footnote ³⁶

Yet after the religious wars of the seventeenth century, there was a marked decline in siege-related massacres. Inter-European warfare between 1648 and the Revolutionary-Napoleonic Wars has traditionally been characterised by historians as one of relative restraint and moderation. In his classic study of war in the ‘Age of Reason’, Christopher Duffy conceived of the period as one of ‘remarkable restraint in the conduct of warfare’, with the wars generally ‘fought in a manner that reflected the urbanity and reasonableness of the age’.Footnote ³⁷ Deconfessionalisation, the Enlightenment, the post-Westphalian state system, the rise of professional, uniformed and regularly paid state armies, the regularisation of requisitioning, and more comprehensive and enforceable military justice systems, have all been seen as contributing to more ‘limited’ forms of warfare.Footnote ³⁸

Within the context of war and cultural change, the ideal of ‘humanity’ emerged as a core principle amongst a growing number of eighteenth-century military publicists and educationalists. ‘Making good war’, as Christy Pichichero has argued, became a guiding ideal within the Military Enlightenment, with the nature of war increasingly theorised, historicised and reflected upon.Footnote ³⁹ Informed by prevailing conceptions of historical progress, civility and moral sentiment, enlightened military circles advocated a humane and compassionate approach to the welfare of soldiers and to the conduct of war more generally. The Enlightenment’s most famous jurist on the laws of war, Emer de Vattel, in his seminal work, The Law of Nations (1758), set out rational and ethical principles for wartime conduct between states. Like most enlightened writers on war, Vattel endorsed the principle of military necessity; and his conception of laws of war strictly applied to nations and enemies deemed ‘civilised’ – a fundamental distinction we shall return to later. But as to the general conduct of contemporary European war, Vattel hailed a prevailing spirit of ‘moderation’, ‘generosity’ and ‘politeness’ relative to centuries past.Footnote ⁴⁰

This included the treatment of civilians. The term ‘civilian’ in its modern sense did not exist in the eighteenth century.Footnote ⁴¹ ‘Civilian’ in the English language meant a person who practised or studied civil law, as set out in the many editions of Dr Johnson’s Dictionary from 1755.Footnote ⁴² The word began to take on its modern meaning only in the first decades of the nineteenth century, and then it is not one we find in British soldiers’ accounts of the Peninsular War. When identifying those not directly involved in military affairs, contemporary writers referred to ‘inhabitants’ or specific categories of age, gender or occupation: women, children, the elderly, the infirm, peasants, burghers, men of letters, the religious, etc.Footnote ⁴³ The essential principle of ‘distinction’ was that these groups did not normally bear arms, or resist the enemy, and consequently should suffer neither injury nor death at the hands of soldiers, Vattel declaring this a ‘maxim of justice and humanity’.Footnote ⁴⁴ Reflecting on the relationship between war and civilians, Vattel concluded: ‘At present war is carried on by regular troops; the people, the peasants, the inhabitants of towns and villages do not concern themselves in it, and generally have nothing to fear from the enemy’s arms.’Footnote ⁴⁵

Of course, scholarship over the past twenty years or more has challenged how ‘limited’ and ‘civilised’ European warfare was in practice after 1648, especially with regards to civilians.Footnote ⁴⁶ Marlborough’s devastation of Bavaria in 1704; the British suppression of the Jacobite uprising in 1746; the French pillaging of Hanover under the Duke of Richelieu in 1757; and Russian pillage and plunder in East Prussia during the Seven Years War; not to mention the violence of European armies within the colonial sphere – all are stark reminders of the limits of humanity in war and of the traumas visited upon civilian populations in the eighteenth century. Perhaps, 400,000 civilians died during the War of the Austrian Succession; an even greater number suffered during the Seven Years War.Footnote ⁴⁷

But what of violence within the most common and regulated form of eighteenth-century warfare? Historians have traditionally considered old-regime sieges to be the very model of restraint. For Geoffrey Best, siege warfare ‘had about it something of the formalized manoeuvring of a courtly dance’, with the elaborate and formalised capitulation rituals exemplifying the ‘gentlemanly and self-limiting sides of eighteenth-century warfare’.Footnote ⁴⁸ This idea of the siege as epitomising courtly etiquette was bound up with the age of the Sun King, where sieges seemingly adhered to a set choreography and script, a cultural as well as military event, with Louis XIV and his courtiers sometimes spectators.Footnote ⁴⁹ Louis was particularly drawn to the stage-managed nature of sieges, as if they were the war spectacle equivalent of Versailles itself.Footnote ⁵⁰ In attendance at the 1691 siege of Mons, he ordered the Régiment du Roi to play a concert for the women of the town.Footnote ⁵¹ Moreover, courtly ritual merged with the idea that scientific laws governed the outcomes of sieges. Part of Vauban’s legend was that he famously set out a timetable for the number of days for a typical siege – 41.Footnote ⁵² He saw this as only a general guide for provisioning, but it was accepted by some contemporaries as dogma, and has often been taken literally by historians as a schedule that could predict a siege’s conclusion to the very day.Footnote ⁵³ The ‘Vaubanian siege’ came to represent the aspirational scientific siege, where the result was deemed predictable and inevitable. Frederick the Great, no great exponent of siege warfare himself, nevertheless wrote: ‘The art of conducting sieges has become a trade like that of carpenter or clock-maker. Certain infallible rules have been established, and we follow an unvarying routine, applying the identical procedure in the same cases.’Footnote ⁵⁴ Furthermore, there was an important humanitarian dimension to Vauban’s conception of siege warfare, wanting to minimise not only time and costs but human lives too, famously pleading to ‘burn more powder and spill less blood’.Footnote ⁵⁵ Thus, aristocratic formalism, geometry and mathematics, and civility and humanity – all combined to produce an idealised model of ‘enlightened siege’.

Yet as Jamel Ostwald has shown, old-regime sieges did not always match high-minded Vaubanian scientific and humane ideals, with impatient siege commanders pushing for faster and more vigorous methods, often against the advice of engineers, which invariably produced a higher toll in human lives and urban destruction.Footnote ⁵⁶ Eighteenth-century sieges could be brutal and bloody affairs for the attacking troops. The 1708 siege of Lille, for example, was the costliest Allied siege of the War of the Spanish Succession, claiming some 12,000 dead or wounded amongst the besieging force.Footnote ⁵⁷

Civilians continued to suffer in eighteenth-century sieges.Footnote ⁵⁸ Indeed, within conventional military operations, sieges presented the greatest challenge to Vattel’s distinction between separate military and civilian spheres, although sieges could simultaneously reinforce and erode those divisions. Civilians were often more vulnerable than the garrison, and their needs were rendered subordinate to the principles of military necessity and the priorities of the town’s military authorities. The management of a besieged town’s limited food supplies favoured garrisons over inhabitants, reflected in the medieval practice of expelling civilians deemed to be ‘useless mouths’.Footnote ⁵⁹ This tradition was still practised at times in the eighteenth century, as did forcing refugees back into a fortress, as de Saxe did at the siege of Tournai in 1745, knowing full well that this put added pressure on the food supplies.Footnote ⁶⁰ Civilians could be pressured and targeted by the attacking forces to turn on garrisons to capitulate. Before besieging Ghent in 1708, for example, Marlborough threatened a deputation of civil authorities that ‘they must either assist us against the garrison or expect we should use all manner of extremity to reduce them to their duty’.Footnote ⁶¹ Civilians continued to suffer from the slow and silent killers of disease and famine in long sieges. And civilians and their homes were especially vulnerable to bombardment, a practice that Vauban opposed but one that siege commanders could draw upon to press for a resolution.Footnote ⁶² Frederick the Great’s 1760 destruction of Dresden, in particular, drew widespread outrage.Footnote ⁶³ The morality of bombardment became part of enlightened debate about limiting the impact of war on civilians; one that historians have pointed out was a forerunner to debates over twentieth-century aerial bombardment of cities.Footnote ⁶⁴

Nor were civilians always passive victims, for they could be actively involved in the defence of towns to varying degrees. Despite the rise of military professionalism in the eighteenth century, siege defences were not always left to garrisons alone. The mathematician, Guillaume Le Blond, in his widely consulted manual on siege warfare, encouraged defending governors to mobilise the population in what we might term a ‘levée en masse in miniature’: craft workers were to help with their skills; others were expected to put out fires, transport materials and build retrenchments; the religious were to attend to the sick and wounded; and ‘nuns and the poor women may be employed, in washing the soldiers’ linen, in making lint, etc.’.Footnote ⁶⁵ In sum, the inhabitants were expected to play an important support role, short of taking up arms. Yet in some cases, this traditional distinction between soldier and civilian was crossed, when town militias were deployed, or when civilians took it upon themselves to fight with weapons. Indeed, the level of popular armed mobilisation at the sieges of Turin (1706) and Barcelona (1714) would not have been out of place during the Revolutionary-Napoleonic Wars.Footnote ⁶⁶ This was hardly siege as ‘limited’ warfare.

Nonetheless, there was one fundamentally important change when it came to siege violence: besieged garrisons and civilians in the eighteenth century were less likely to experience the trauma of sack than previous generations. John Lynn has pointed to a shift in siege surrender rituals that took hold in the early eighteenth century, with capitulation becoming more managed and predictable, increasingly ending on honourable and humane terms.Footnote ⁶⁷ Technological and cultural changes coalesced: advances in the destructive firepower of siege trains reduced the likelihood of successfully defending a breach; and there was an understanding that a respectable or honourable defence was reached when the walls were breached, if not earlier, when the covered way was taken by the attacking forces.Footnote ⁶⁸ Duffy has remarked that ‘last-ditch stands went out of fashion in the one-and-a-half centuries which intervened between the end of the Wars of Religion and the coming of the French Revolution – the temper of the times was too cool and too rational to inspire people to sacrifice themselves before the march of a siege’.Footnote ⁶⁹ In 1710, Louis XIV reportedly chastised an overly zealous officer defending Béthune: ‘playing the hero at the wrong time is the way to becoming a laughing stock’.Footnote ⁷⁰

Yet, there are geographical distinctions and nuances that need to be drawn here. This shift was most marked in Western Europe, and in some theatres, more so than others. The emergence of this trend was especially evident in northern France and the Low Countries, the epicentre of old-regime siege warfare. Of the many sieges conducted in the Low Countries during the War of the Spanish Succession, the overwhelming majority ended in capitulation.Footnote ⁷¹ Tellingly, the Duke of Marlborough conducted twenty-six sieges in the Low Countries during the war without a single town being taken by storm or subjected to sack.Footnote ⁷² All ended with negotiated capitulation. John Millner, a sergeant in the Irish Foot, recorded in his journal of Marlborough’s campaigns in Flanders: ‘of all the cities and towns taken throughout the whole of the war on our side thereof, none stood a storm, but at the last extream surrender’d on reasonable conditions’.Footnote ⁷³

Turning to the Iberian theatre of the same war, we find a more varied picture.Footnote ⁷⁴ This is especially pertinent given it was the last series of major sieges waged in Spain before the Peninsular War, our central focus a century later. The majority of sieges in Iberia during the War of the Spanish Succession ended in negotiated capitulation either before or at the moment of a practicable breach, with surrender sometimes coming very early in proceedings.Footnote ⁷⁵ Nevertheless, there were some famous obstinate defences of breaches, notably at the sieges of Denia (1707), Xativa (1707) and Barcelona (1714), the first two cases involving British garrison troops.Footnote ⁷⁶ Yet, few besieged towns ultimately fell to storm, and such cases were not marked by the slaughter of garrisons, or wholesale sack and large-scale atrocities, although there was certainly plunder and some violence to civilians.Footnote ⁷⁷ The noted exception was the sack and destruction of Xativa by Bourbon troops, within the broader context of the suppression of the Valencian rebellion against Philip V. This three-week siege ended in storm and street fighting, the English garrison and town inhabitants having mounted an obstinate defence. Whilst the garrison withdrew into the citadel (capitulating some days later), the inhabitants were left at the mercy of Bourbon troops. The exemplary punishment did not end there: the Duke of Berwick had the town razed to the ground – ‘in order to impress terror, and by an example of severity to prevent the like obstinacy’.Footnote ⁷⁸

Duffy notes a ‘tradition of popular resistance’ within Spanish siege warfare, one dramatically illustrated at Xativa and on a far greater scale at the 1714 siege of Barcelona.Footnote ⁷⁹ At Barcelona, the last major siege of the war, sack was a very near run thing. The siege proved a protracted and brutal affair, the Duke of Berwick’s forces pitted against the entire city, with men, women and the religious involved in the defence. On 11 September, after weeks of mounting frustration at the defenders’ obstinacy and repeated refusal to surrender, Berwick ordered a final general storm of the breaches; the attacking forces managed to establish themselves inside the city walls, but fierce fighting continued. At this extremely late and desperate juncture, the city’s authorities finally sought to treat. Given the eleventh hour, and that he had been forced to storm, Berwick claimed that it was now in his ‘power to put everything to the sword’, but he nevertheless agreed to terms, promising the Catalans that ‘their lives would be safe, and even that there should be no plunder’.Footnote ⁸⁰ And so it came to pass – Barcelona was spared the horror of sack.Footnote ⁸¹

Inspired by the 1714 siege of Barcelona, the famous French landscape engraver, Jacques Rigaud (1681–1754), produced a series entitled, ‘Répresentations des actions les plus considérables du siège d’une place’.Footnote ⁸² Originally published in Paris in 1732, English versions with extended descriptions were reproduced in England over the eighteenth century.Footnote ⁸³ The series comprised six plates, each representing a key stage in the unfolding of a siege, beginning with the ‘The Opening of Trenches’ and ending with ‘The Place given up to Plunder’. But Barcelona in 1714 was not given up to plunder, and Rigaud acknowledged the rarity of sack in his description of the engraving: ‘Tis very seldom that a place will expose and abandon itself to the fury of the soldiers. This representation is added to give some idea of the cruelty of such a bloody scene.’Footnote ⁸⁴ The final plate – depicting the assault of women, civilians pleading for mercy, and house fighting between soldiers and civilians – largely belonged to sieges of a by-gone era.

This was partly why the later French sack of Bergen-op-Zoom in 1747 during the War of the Austrian Succession created such outrage – it was the exception to the norm.Footnote ⁸⁵ During the war, most fortresses in the Low Countries surrendered to French forces with relative ease; that was, until the sixty-four-day siege of Bergen-op-Zoom late in the war. British battalions were amongst the defenders. The town was eventually taken by a surprise attack against a breach the Dutch deemed impracticable. The town was sacked in the aftermath. According to one account, the French commander, Ulrich Woldemar de Löwendal, ‘took all the necessary precautions to lessen the evil and to quell it as soon as possible’.Footnote ⁸⁶ Yet, an estimated 2,000 civilians were killed.Footnote ⁸⁷ The sack shocked contemporary opinion, including that of military men who were present. ‘My heart bleeds when I think of the inhumanities and cruelties committed after the assault upon the poor inhabitants of Bergen-op-Zoom’, wrote one British officer.Footnote ⁸⁸ Marshal Saxe reportedly quipped to Louis XV that the king had to either hang Löwendal or make him a marshal – the latter prevailed.Footnote ⁸⁹

It was in the East, in Russia’s wars, where sack remained part of eighteenth-century siege warfare, especially along the Turkish frontiers in the Balkans. At the siege of Narva (1704) during the Great Northern War, Peter the Great’s troops massacred most of the Swedish garrison and killed civilians, after the commander refused to accept Peter’s surrender offer after a breach had been made.Footnote ⁹⁰ The most savage siege massacres occurred against the Ottoman Turks, with restraints lowered amidst historically entrenched religious and cultural enmities. There were large-scale massacres in the Danubian frontier regions across the Russian-Turkish Wars of the eighteenth century, where both garrisons and the inhabitants were put to the sword, sometimes even after formal capitulation. In the Austro-Russian-Turkish War of 1737–39, at the siege of Ochakov in 1737, the Russians massacred much of the garrison despite a white flag and the fortress capitulating.Footnote ⁹¹ And as we shall see, the worst European siege massacres of the century occurred during the Russo-Turkish Wars of 1787–92.

But looking back over the greater eighteenth century, the sack of towns had largely disappeared from siege warfare amongst European belligerents. This was a remarkable historical phenomenon given the sheer number and importance of sieges during the first half of the century. As a sign of this historic shift, Vattel did not see cause to address the specific question of sack violence to civilians, unlike his sixteenth- and seventeenth-century counterparts, Gentili, for example, having devoted an entire section in his De jure belli libri tres to ‘The Destruction and Sacking of Cities’.Footnote ⁹² Nor is it surprising that later eighteenth-century military writers noted that the age-old customary law of sack was passing out of customary usage. In 1769, Stephen Payne Adye, a British artillery officer and deputy judge advocate, wrote in his influential A Treatise on Courts Martial: ‘Towns taken are never now given up to plunder, as was formerly the practice among the Ancients.’Footnote ⁹³ And casting his eye over the nature of contemporary warfare, Jacques-Antoine-Hippolyte, comte de Guibert, the most influential of all late eighteenth-century military writers, observed in his famous Essai général de tactique (1770) that wars were now ‘less barbarous and cruel’, noting among other changes that ‘towns are not sacked’.Footnote ⁹⁴ A new normative framework had evolved over the course of the eighteenth century, but one that would come under stress during the Revolutionary-Napoleonic Wars.

The Death of the Siege?

In Laurence Sterne’s best-selling novel, The Life and Opinions of Tristram Shandy (1759–67), written during the Seven Years War but set during the War of the Spanish Succession, Uncle Toby’s obsession with fortifications and re-enacting the Duke of Marlborough’s sieges in the back garden of his country estate, with the aid of his faithful servant and fellow veteran, Corporal Trim, provides a running source of comical satire. The siege in Tristram Shandy has been read as a ‘curious and fascinating relic of an increasingly obsolete conception of war’.Footnote ⁹⁵

Critics had been predicting the death of the siege from the mid-eighteenth century. Ironically, the very success of French sieges in the War of the Austrian Succession, most notably the ease in which most fortresses in the Low Countries fell, led to a re-evaluation of the military effectiveness and financial cost of permanent fortresses.Footnote ⁹⁶ Saxe, in his famous Mes Rêveries, attacked the ‘present erroneous policy of fortifying towns’, but conceded that ‘there is hardly a single person who will concur with me in opinion; so prevailing and so absolute is custom’.Footnote ⁹⁷ Those dissenting voices, however, would grow in the decades to come, especially in light of the Seven Years War. French failures there produced considerable post-war soul searching within French military circles between traditionalists and reformists. The ensuing debates amongst artillerists and engineers included the role of permanent fortifications and siege craft in future wars, which fed into the Military Enlightenment more generally.Footnote ⁹⁸ Guibert dismissed fortresses and siege warfare as static, routinised and overly reliant on technology.Footnote ⁹⁹ Henry Lloyd, too, assigned sieges only a limited role in military campaigns, concluding: ‘Sieges are attended with so great expense, and so much loss of time, and men, that they ought never to be taken without the utmost necessity.’Footnote ¹⁰⁰ From 1776, the marquis de Montalembert, as a way to shift the balance back in favour of fortresses over besieging armies, tried (unsuccessfully) to persuade the French engineering fraternity of the merits of a radical redesign of fortresses and of dramatically enhancing their defensive firepower.Footnote ¹⁰¹ Then, in 1781, the Austrian Emperor, Joseph II, did away with permanent fortresses altogether, dismantling most of the famed Barrier Fortresses in the Austrian Netherlands.Footnote ¹⁰² All this seemed to herald the dying days of fortresses and sieges in European warfare.

Yet after the Seven Years War, many of these assumptions would remain untested for decades to come, given that there was no general war in Continental Europe until the Revolutionary Wars of the 1790s. Moreover, the global dimensions of the Seven Years War reveal the important strategic and symbolic roles that sieges played within the sphere of Anglo-French and Anglo-Spanish imperial rivalry. The successful British siege of Louisbourg in 1758 was a pivotal moment in the French loss of Canada; and in India, the British besieged and took Pondicherry from the French in 1760–61 (and took it again in 1778). In 1762, the British then struck at the heart of both the Spanish Caribbean and the Spanish East Indies through two siege operations, capturing Havana and then only a few months later Manila. And of course, the last major military action of the American Revolutionary War was Cornwallis’s surrender at Yorktown in 1781, following a siege where French engineers and gunners played a crucial role.Footnote ¹⁰³ Moreover, as the British surrendered at Yorktown, a dramatically long siege was still unfolding much closer to home – the Great Siege of Gibraltar (1779–83).Footnote ¹⁰⁴ Sieges, in fact, had a long after life beyond the mid-eighteenth century, with the sieges of the Revolutionary-Napoleonic Wars a crucial part of that history.

The Return of Siege: The Revolutionary-Napoleonic Wars

Of course, mass armies, rapid movement and field battles dominated the campaigns of the French Revolutionary and Napoleonic Wars, to a degree unparalleled in early modern European history. The figures are staggering, Tim Blanning highlighting that there were 713 battles between 1792 and 1815, against 2,659 over the previous three centuries.Footnote ¹⁰⁵ It was not just the frequency of the battles that stood out, but their sheer scale, too, the Battle of Leipzig in October 1813 involving more than half a million soldiers.Footnote ¹⁰⁶ Battle had well and truly eclipsed siege as the primary form of conventional warfare. As a sign of this shift, Napoleon, in his own words, ‘fought sixty battles’, but had direct involvement in only a handful of sieges; whereas Louis XIV had been present at some twenty sieges but never witnessed a major battle.Footnote ¹⁰⁷ Reflecting on the epoch’s wars, Carl von Clausewitz was no champion of siege warfare, deriding its mechanistic and geometrical form and seeing the seizure of fortresses as a relic of the past, at best a ‘necessary evil’ with very limited strategic value.Footnote ¹⁰⁸ Unlike siege warfare, it was in the tactical realm of the field and battle, where ‘everything is more mobile, and psychological forces, individual differences, and chance play a more influential role’.Footnote ¹⁰⁹ In a Clausewitzian conception of war, the siege was not a true test of the higher moral and intellectual dimensions of war; a stage not worthy of romantic genius ‘which rises above all rules’.Footnote ¹¹⁰

Yet for all this, fortresses and sieges still mattered throughout the wars: strategically, politically and symbolically.Footnote ¹¹¹ This was evident from the very outbreak of the Revolutionary Wars. Allied planning for the invasion of France in 1792 took into account France’s north-east fortresses.Footnote ¹¹² Whilst both Longwy and Verdun quickly fell, the fall of Verdun on 2 September had profound consequences on the course of radical politics and popular militancy within France, news of its capture being the trigger for the notorious September Massacres in Paris.Footnote ¹¹³ A year later, Republican forces lay siege to Toulon (7 September–19 December 1793), regaining possession of this important Mediterranean naval port and arsenal from Royalist and Anglo-Spanish forces, with a young Bonaparte playing a starring role.Footnote ¹¹⁴ And the Austrian surrender of Charleroi on the Sambre, on the evening of 25 June 1794, played a role in France’s famous victory over the Austrians at the Battle of Fleurus the following day, freeing up General Hatry’s besieging forces to serve in the line of battle.Footnote ¹¹⁵ Moreover, there were early and painful reminders of the impact of sieges on soldiers and civilians. At the 1793 siege of Mainz, the French garrison and town inhabitants endured a month-long bombardment before surrendering.Footnote ¹¹⁶ Goethe was present with the Prussian forces and registered his shock at the suffering of civilians and the level of destruction: ‘Ashes and ruins were all that was left of what it had cost centuries to build up.’Footnote ¹¹⁷ It was a harbinger of what was to come over the next two decades.

Over those ensuing years, given the length and geographical sweep of the Revolutionary and Napoleonic Wars, land-based siege operations were conducted across much of Europe at one time or another: from the Low Countries, France and the Rhineland, to northern and southern Italy, and from the Iberian Peninsula to northern Germany, Prussia and Poland, and even extending across to Syria in 1799, with Bonaparte’s two-month-long siege of Acre and the short and bloody siege of Jaffa.Footnote ¹¹⁸ These sieges varied in form and duration, and across time and place, from short bombardments to long starvation blockades such as Mantua in 1796–97, and from the horrendous urban street fighting of Zaragoza to classic Vaubanian style sieges such as the 1807 siege of Danzig. Spain stood out for its number of regular sieges, concentrated during the Peninsular War, with bombardment and blockade strategies much more common elsewhere, notwithstanding some notable exceptions. And the most important single factor in shaping the chronology, geography and identities of the besieger and besieged was the military fortunes of the French Republic and then the rise and fall of the Napoleonic Empire itself. Indeed, having begun as the besieged in 1792–93, the French would end as the besieged in 1812–15. Yet in the long interim, the French were generally the besieger, with sieges part and parcel of French conquests.

With respect to the Napoleonic Wars from 1803 to 1815, leaving aside for now the all-important Spanish theatre, we can identify general chronological and geographical shifts in siege activity across the years. Earlier, during the Revolutionary Wars, siege warfare had returned to one of its traditional heartlands – northern Italy – with two famous sieges in particular: the eight-month-long French siege and blockade of the Austrian-held fortress of Mantua (June 1796–February 1797); and the Austrian siege of French-held Genoa (19 April–4 June 1800) during Napoleon’s second Italian campaign.Footnote ¹¹⁹ Come the Napoleonic Wars, with the French invasion of the Kingdom of Naples in 1806, there were a number of regular, but smaller scale, sieges in the south. First, Marshal André Masséna took four months to besiege and capture the coastal fortress of Gaeta (4 March–18 July 1806).Footnote ¹²⁰ From late 1806, the French then besieged Amantea in southern Calabria, which capitulated on 7 February 1807;Footnote ¹²¹ and the following February, the French captured the fortress of Scilla on the Strait of Messina, the small British garrison forced to evacuate after eight days of cannon fire had ruined the defences.Footnote ¹²²

In Central and Eastern Europe, siege operations came in two main phases during the Napoleonic Wars, the first during the French conquests of 1806–07, and the second in the aftermath of the retreat from Russia and during the collapse of the French Empire in 1813–14. Between the Battle of Jena in October 1806 and the Treaties of Tilsit in early July 1807, the French conducted a series of blockades and sieges of Prussian fortresses in Pomerania, Silesia and Poland. In the wake of Jena, French marshals were dispersed to chase the remnants of the Prussian army as it raced eastwards, or to contain and capture Prussian troops that had taken refuge in fortresses and cities to the rear.Footnote ¹²³ French siege operations then shifted south to Silesia, capturing the Prussian fortresses in the Oder Valley, including Glogau, Breslau, Brieg and Schweidnitz, which were generally blockaded and then bombarded once the siege trains arrived.Footnote ¹²⁴ Napoleon’s advance into Poland at the end of 1806 brought with it further sieges and blockades, including against the Prussian towns of Kolberg and Graudenz.Footnote ¹²⁵ But the centre piece of siege operations was the Baltic port of Danzig. Napoleon entrusted this to Marshal François Lefebvre, who had 45,000 troops at his disposal, the cream of the French engineering corps and a siege train of 100 guns. Beginning in March 1807, the siege was conducted according to classic eighteenth-century lines, with the French digging three sets of parallel trenches. It took ten weeks before the Prussians finally capitulated, the French entering the city on 27 May.Footnote ¹²⁶

Siege operations returned to Eastern and Central Europe in the final years of the war. This time the tables were turned. Fortresses played a strategic role in Napoleon’s desperate attempt to retain a foothold in the east. In the wake of the Russian retreat, Napoleon left 50,000 troops garrisoned in fortresses in Poland and East Prussia.Footnote ¹²⁷ They were soon besieged and blockaded by Russian forces in the early months of 1813.Footnote ¹²⁸ An epic struggle took place at the siege of Danzig, where an Imperial garrison held out Russian forces for almost a year, from January to November 1813.Footnote ¹²⁹ Following his defeat at Leipzig in October 1813, Napoleon then left garrisons behind in fortresses and cities along the Elbe. Hopelessly isolated, Napoleon’s troops in Germany were doomed. Dresden capitulated on 11 November 1813, followed soon by the Elbe fortresses of Torgau (10 January 1814) and Wittenberg (13 January).Footnote ¹³⁰ At Hamburg and Magdeburg, the besieged French remained defiant to the end, capitulating only in May after Napoleon’s fall and abdication.Footnote ¹³¹

With the allied invasion of France in 1814, siege operations returned to the Rhine, the Low Countries, and France’s northern and eastern borders, for the first time since the early years of the Revolutionary Wars. But whilst the French fortress belts figured in Allied invasion plans, the core objectives were to penetrate quickly into France, defeat Napoleon’s field armies, and to take Paris. Footnote ¹³² There was little desire to conduct regular sieges, with fortresses generally blockaded, contained or loosely observed, with hopes that some might fall to surprise attack or capitulate after a brief bombardment.Footnote ¹³³ The half-hearted operations of the Prussians against the Moselle and Saar fortresses in January 1814 are a case in point, Gebhard Leberecht von Blücher cautioning his commanders to spare ‘ammunition for battle’ and not to ‘expend any powder’.Footnote ¹³⁴ In the Low Countries, there were several failed attempts by the Prussians and British to besiege Antwerp in January–February 1814, and the British tried to take Bergen-op-Zoom by a surprise attack on the night of 8 March 1814, but succeeding in only getting themselves captured.Footnote ¹³⁵

The last period of siege activity came in the summer of 1815. Once again battles and the French capital loomed first and foremost with the allies, but fortresses could not be ignored, as the aftermath of Waterloo demonstrated. As Wellington and Blücher pushed on for Paris, subsidiary forces were deployed to deal with fortresses. The British took Cambrai via escalade on the night of 24 June, and Peronne surrendered on 26 June after a defensive outwork was stormed.Footnote ¹³⁶ The Prussian second corps was tasked with capturing frontier fortresses left in the rear: Maubeuge, Landrecies, Philippeville, Mariembourg and Rocroi. Each of these fell in quick succession to bombardment throughout July and August. For the task, the Prussians had been assigned British help: Richard Henegan was in charge of the siege train, and Alexander Dickson commanded six companies of British artillery.Footnote ¹³⁷ It was rather fitting that Dickson was involved in the last phase of siege operations of the Napoleonic Wars. Not only had he played a crucial role in all the major British sieges of the Peninsular War, placed in charge of the royal artillery under Wellington in 1812–13,Footnote ¹³⁸ but also as a younger captain, he had commanded the British artillery detachment at the 1807 siege of Montevideo.Footnote ¹³⁹ Dickson’s career captured the importance of siege operations to the British army in this era, and their global dimensions and interconnections.

British Siege Operations before the Peninsular War

When Britain entered the war against Revolutionary France in 1793, and promptly despatched the Duke of York’s expeditionary army to the Low Countries, this was the first time in three decades, since the days of the Seven Years War, that a British army had campaigned in Continental Europe. Despite this long absence, the British quickly found themselves a siege: in fact, the Duke of York oversaw two sieges in 1793, although with limited and mixed results. Success did come in the first instance, with the capitulation of Valenciennes (13 June–28 July), although this joint Anglo-Austrian operation was only nominally under the command of the Duke of York, with the Austrian general, Ferraris, overseeing the technical direction of much of the operation. But farce later ensued in the second instance when British, Austrian, Hanoverian and Hessian troops under the command of the Duke of York besieged Dunkirk in August–September 1793: the promised siege guns failed to arrive, and York was forced to lift the siege to much embarrassment.Footnote ¹⁴⁰

From this rather humiliating return to the stage of Continental siege warfare, it would be almost another twenty years, until the Peninsular War, before another British commander conducted a major land-based siege in Continental Europe. Given this, there is a long and persistent historical tradition of dismissing Britain’s siege record in the interim as having almost no relevance to what would transpire in Spain. A hundred years ago, Charles Oman prefaced a discussion of British sieges in the Peninsula by declaring that this record ‘need hardly be mentioned’.Footnote ¹⁴¹ This, Oman assured his readers, consisted of nothing more than blockades, bombardments and the ‘battering of old-fashioned native forts in India’. All this made for a rather poor apprenticeship for what the British would confront in Spain, where they ran up against European fortress towns obstinately defended by Napoleon’s garrisons. Siege warfare in Spain in 1811, one historian has written, came as a ‘complete novelty to both officers and men of the British army’.Footnote ¹⁴²

Yet, this requires some qualifications. Certainly, following the withdrawal of the Duke of York’s forces from the Continent in early 1795, British siege operations in Europe were directed against coastal fortresses and towns. These were generally joint army-navy operations and often amounted to bombardments rather than formal sieges, ranging from the sieges of Bastia and Calvi in Corsica in 1794 to the blockade of Malta in 1798–1800, and from the bombardment and capture of neutral Copenhagen in 1807, to the later bombardment of Flushing in August 1809 during the ill-fated Walcheren expedition.Footnote ¹⁴³

But a different picture emerges if we look at British siege operations beyond Europe. As Bruce Collins has shown, the notion that Britain had not been involved in any recent sieges of note prior to the Peninsular War needs reassessing, especially in light of India, most notably during the Fourth Anglo-Mysore War (1799) and the Second Anglo-Maratha War (1803–05).Footnote ¹⁴⁴ These were crucial years in the dramatic and rapid expansion of the British East India Company into a quasi-state and major territorial and military power beyond Bengal. And sieges played a role in these conquests. Whilst the British did indeed take many hilltop fortresses with relative ease, some were formally besieged, and two operations are of particular interest in the context of this book: Seringapatam in 1799 and Gawilghur in 1803. These involved soldiers who later served in Spain, most famously Arthur Wellesley, the future Duke of Wellington, and ended in breach assaults and sack.Footnote ¹⁴⁵

Of all the British sieges in India in this period, that of Seringapatam in April–May 1799 stands out in terms of its scale, resources and historical importance. This siege brought to a dramatic conclusion Britain’s intermittent and three-decade long conflict with the southern Kingdom of Mysore, where the British fought a series of wars against the father and son Muslim rulers, Haider Ali and Tipu Sultan. The third war, against Tipu in 1790–92, had ended before the walls of the Mysorean capital, Seringapatam, with Tipu agreeing to peace terms with Cornwallis, ceding half of his territory in the process.Footnote ¹⁴⁶ The final war, in 1799, was very much unfinished business for the Company. Triggered by fears that Tipu might finally realise his long-cherished dream of a formal alliance with the French, the British campaign against Mysore commenced in February 1799.Footnote ¹⁴⁷ Under the command of Major-General George Harris, this was a major operation, involving some 43,000 soldiers across multiple army groups, comprising Company and Royal regiments, and a large contingent from the Nizam of Hyderabad (which Colonel Arthur Wellesley commanded), with most of the soldiers being Company sepoys and Indian allies.Footnote ¹⁴⁸ The objective was besieging and seizing Seringapatam, which Harris’s main army reached on 5 April. For the task ahead, Harris was taking no chances, having brought a siege train of forty heavy guns and seven howitzers.Footnote ¹⁴⁹ Seringapatam was situated on a small island in the Kaveri River, with a fortress-citadel situated on the western end of the island. The siege proper began on 17 April and culminated in a practicable breach on 3 May, which was stormed the following day. Tipu was famously killed in the fighting, and the city was sacked.Footnote ¹⁵⁰

With the conquest of Mysore, the Company increasingly turned its attention to internal divisions within the powerful Maratha Confederacy that controlled much of central, western and northern India, giving rise to the Second Anglo-Maratha War. The British deployed two armies in the opening campaigns of 1803: General Gerard Lake commanded in the north, in Hindustan, against the Maratha prince, Daulat Rao Scindia; whilst Arthur Wellesley, now a Major-General, commanded in the south, in the Deccan, against the forces of both Scindia and the Raja of Berar.Footnote ¹⁵¹ Whilst the key British military objectives were to defeat the enemy in the field, fortresses came into calculation as they were important sites of princely power and wealth.Footnote ¹⁵² In the case of Wellesley, the decisive moment of the 1803 campaign was, of course, his famous victory at the Battle of Assaye on 23 September, but the campaign had nevertheless begun with the three-day siege of the fortress of Ahmednagar in August, and the final act was the siege of Gawilghur in December 1803, the formidable mountain fortress of the Raja of Berar. This was a four-day siege, hardly on the scale of Seringapatam, but it was a very challenging one, with the British having to drag their siege guns through mountainous country. Batteries were established before the walls of Gawilghur on 12 December and firing commenced the following day. The outer walls were breached on the evening of 14 December, and stormed the next morning, with massacre and sack following.Footnote ¹⁵³ This short but violent siege was the last that Arthur Wellesley oversaw before the Peninsular War.

Before turning to Spain, however, there is another siege that is rarely mentioned, if at all, within the context of the British siege record prior to the Peninsular War – the 1807 siege of Montevideo. This was not a major military event like Seringapatam, but it was a regular siege, nevertheless, which involved British forces fighting the Spanish as their centuries old traditional enemy and Imperial rival; it involved regiments and personnel who later fought in the sieges in Spain; and most importantly, it was a fortified town that the British took by storm.

The siege took place within the extraordinary context of the British invasion of the River Plate in 1806–07, which began in unexpected circumstances, when Sir Hope Popham, without authorisation, sailed across the Atlantic from the Cape to the River Plate, with a small force of 1,500 men under the command of William Beresford. In June 1806, Beresford captured Buenos Aires, a largely open city, only for it to be retaken by Spanish forces in August, with Beresford and his men taken prisoner. It all ended nearly a year later, in controversy and humiliation, when Lieutenant-General John Whitelocke surrendered whilst trying to recapture Buenos Aires in July 1807 – the British were forced to evacuate from the River Plate, and Whitelocke was court-martialled upon returning to England.Footnote ¹⁵⁴

In the interim, however, the British did enjoy one success story – the siege and capture of Montevideo in February 1807. The operation was commanded by Brigadier-General Samuel Auchmuty, who had set sail from England the previous October. With reinforcements from the Cape, his total force came to about 5,000 men. Montevideo sat on a small rocky Peninsula, surrounded by the sea on three sides, with its eastern landward side protected by 15-foot-high curtain walls, bastions and a citadel. The siege began on 22 January, supported by Rear-Admiral Charles Sterling’s fleet. Entrenchments were dug and batteries established, the citadel and defences fired upon, and parts of the city bombarded. On 3 February, a practicable breach was stormed and the city fell to the British.Footnote ¹⁵⁵

Upon returning to Britain, a number of the regiments involved at Montevideo were among the expeditionary force that sailed from Cork in July 1808 under the command of Arthur Wellesley destined for the Peninsular War. They now found themselves in the rather unfamiliar position of going to liberate rather than fight the Spanish. As Sergeant William Lawrence of the 40th Regiment recalled: ‘The nation we had recently been fighting in Montevideo, Buenos Aires and Colonia, was now calling for our assistance to drive the French out of their country.’Footnote ¹⁵⁶ Four years later, Lawrence was among the stormers at Badajoz.

Sieges in the Peninsular War

From 1808, for the first time since the War of the Spanish Succession, siege warfare returned to the Iberian Peninsula on a major scale. In a highly complex, drawn out and shifting theatre of war, sieges operated alongside conventional field battles, a war of manoeuvre, and asymmetrical warfare between French regulars and Spanish guerrillas. The importance of fortress towns for the French as an army of occupation in insurrectionary regions, combined with geography, sparse food resources, poor roads and limited land corridors for the rapid movement of armies, meant that around thirty regular sieges took place between 1808 and 1813.Footnote ¹⁵⁷

Philip Haythornthwaite lists twenty-nine sieges for the Peninsular War, comprising twenty-eight regular sieges of varying scales and duration, and the 1813 blockade of Pamplona.Footnote ¹⁵⁸ The overwhelming majority occurred in Spain, with only two sieges in Portugal (Almeida in 1810 and Campo Mayor in 1811). Some cities and towns were besieged multiple times over the course of the war, with siege a recurring experience for those inhabitants and defenders. Notably, Zaragoza was besieged twice by the French (15 June–14 August 1808; 20 December 1808–20 February 1809); Gerona, after a failed coup de main on 20–21 June 1808, was besieged twice (24 July–16 August 1808; 6 June–10 December 1809); Ciudad Rodrigo was besieged once by the French (26 April–9 July 1810) and once by the British (7–20 January 1812); and San Sebastian was besieged twice by the British (7–27 July and 24–31 August 1813). Badajoz was unique, besieged on four separate occasions: first by the French (26 January–10 March 1811) and then three times by the British (6–12 May 1811; 27 May–10 June 1811; 16 March–6 April 1812).

The chronological and geographical patterns of the sieges reflected the broad campaign and strategic shifts of the war. The French waged the overwhelming majority of offensive sieges early in the war, against Spanish-held fortress towns, whilst the British took the offensive lead in the final years of the war, against French-held fortress towns. Laying siege to cities and towns became an integral part of the French strategy for conquering and pacifying Spain. Indeed, following their invasion of Spain in early 1808, the French waged the first fifteen sieges of the war to early 1811. The great sieges of Zaragoza and Gerona dominated in 1808 and 1809, amidst the wave of regional insurrection following the Madrid uprising of May 1808. The most intensive years for offensive French siege warfare were 1810–12. Amongst French marshals and generals, it was Louis-Gabriel Suchet who oversaw the greatest number of siege operations. As part of his efforts to subjugate Catalonia and Valencia, he waged a succession of sieges in eastern Spain between April 1810 and early 1812: Lerida, Mequinenza, Tarragona, Saguntum, Tortosa, Valencia and Peñíscola.Footnote ¹⁵⁹

Turning to the British side of things, the 30,000-strong expeditionary force that landed in Portugal in August 1808 was not only the largest British troop deployment to the Continent during the Revolutionary-Napoleonic Wars, but the largest British military commitment to Continental Europe since Marlborough’s days.Footnote ¹⁶⁰ That commitment only grew over the ensuing years, with Wellington commanding 62,000 British troops in the Peninsula in October 1813,Footnote ¹⁶¹ many of whom by that stage had fought not only a series of famous battles, but a succession of bloody sieges. Whilst the first troops arrived in 1808, it was not until three years later that the British began their first siege operations. This reflected the fact that after Sir John Moore’s campaign in late 1808 and early 1809, and Wellington’s offensive into the Tagus Valley that ended with the Battle of Talavera (July 1809), the British were on the back foot, adopting a defensive strategy in Portugal in the middle years of the war (1810–11), culminating in the Lines of Torres Vedras. But after having seen off Masséna’s invasion of Portugal in 1811, Wellington shifted into offensive operations into Spain again, which gave rise to siege warfare.

Haythornthwaite lists nine British sieges during the war: Badajoz (on three occasions), Ciudad Rodrigo, the Salamanca Forts, Burgos, Sebastian (twice) and finally the blockade of Pamplona. This does not include some minor operations, such as the brief siege of Olivenza (9–15 April 1811), where a French garrison of only 400 surrendered after a small siege train had battered a breach in the walls.Footnote ¹⁶² Nor does it include the controversial siege of Tarragona in June 1813, which formed part of the British campaign on the east coast of Spain, removed from Wellington’s main forces and theatres of operation. Tarragona was besieged by British, Spanish and Sicilian forces under the command of General Sir George Murray, but the siege was lifted, the guns abandoned and Murray court martialled for his conduct.Footnote ¹⁶³

Wellington’s sieges in Spain were part of a broader strategy to facilitate a war of movement and battle, maintain lines of communication or to contain and capture French strongholds left behind the lines of advancing British field operations.Footnote ¹⁶⁴ The major British sieges of the war were concentrated in two key areas: the Portuguese–Spanish borderlands, and northern Spain. Above all, before Wellington could launch offensives into the heartland of Spain, it was essential to capture both Ciudad Rodrigo and Badajoz, the strategically important borderland fortresses that oversaw the central and southern land corridors between Spain and Portugal. These two fortresses represented the ‘keys to Spain’.Footnote ¹⁶⁵ This was precisely why they were so contested during the war. With these two fortresses captured in early 1812, Wellington then pushed into central and northern Spain. Wellington’s great victory at the Battle of Salamanca (22 July 1812), ultimately took the British all the way to Madrid, from where Wellington then pressed northward to besiege (unsuccessfully) the fortress of Burgos in September and October 1812. A French counteroffensive, however, forced the British to retreat from Madrid and Burgos in late 1812 and winter in the borderlands of Portugal. The following year, a new British offensive took them all the way into northern Spain and ultimately into France itself, bringing the Peninsular War to an end. Following the British victory at the Battle of Vitoria (21 June 1813), the remains of the defeated French army withdrew into the Pyrenees, leaving two French garrisons in the north: the two sieges of San Sebastian and the Anglo-Spanish blockade of Pamplona (25 June–31 October 1813) then ensued. Once over the Pyrenees, Wellington’s forces besieged Bayonne in southern France from late February 1813, news of Napoleon’s abdication ending the siege.

The Eye of the Storm: Sack in Spain

The Peninsular War was not only the heart of regular sieges in the Napoleonic Wars, but it was also where sack made a notorious return to European siege warfare, after having largely vanished over the previous century. Indeed, the sacks in Spain were part of a broader cluster of infamous siege-related massacres and sacks that occurred in the late eighteenth and early nineteenth centuries, involving European armies across different regional and global settings and inter-cultural contexts. We find this first appearing in a region traditionally associated with siege violence: on the European frontiers of the Ottoman Empire, during the Russo-Turkish War of 1787–92, with Potemkin’s 1788 sack of Ochakov that may have claimed 10,000 lives, and Alexander Suvorov’s even bloodier 1790 sack of Ishmail with possibly 25,000 victims.Footnote ¹⁶⁶ Although not in the context of a siege, Russian forces under Suvorov later sacked the Warsaw suburb of Praga in April 1794, during the Kosciuszko Rising, with somewhere between 12,000 and 20,000 lives lost.Footnote ¹⁶⁷ In 1799, during his Egyptian/Syrian campaign, Napoleon besieged, stormed and sacked the town of Jaffa (3–7 March), with the cold-blooded mass shooting of Turkish prisoners marking its aftermath.Footnote ¹⁶⁸ And in India, that very same year, the British stormed and sacked Seringapatam. But it was not only sieges within the European colonial sphere or between European and Ottoman forces where sack occurred. As we shall later see, siege operations conducted within Continental Europe during the Napoleonic Wars nearly always ended with capitulation – as they had in the eighteenth century. But there was one noted exception – the war in Spain – which was both Napoleonic Europe’s most siege-intensive theatre of war, and the site of its most prolonged wartime brutalities and excesses.

Before outlining these sacks, however, it is important to place them within the broader spectrum of massacres and atrocities in the Peninsular War, arising across different military circumstances and operational contexts, the vast majority of which were not a result of siege warfare. As Philip Dwyer has observed, there were different types of massacres throughout the Revolutionary-Napoleonic Wars, each shaped by particular circumstances, time and place.Footnote ¹⁶⁹ This is no less true of ‘sack’ itself, a term that historians often use in a general sense to describe the fate of towns where troops committed plunder and perpetrated atrocities, but one that belies significant contextual differentiations therein. The Peninsular War, in particular, experienced the full gamut of massacres and atrocities that occurred during the Napoleonic Wars, a broader reflection of the nature of the war south of the Pyrenees.

The first sacks of towns during the Peninsular War, which were among the worst of the war, had nothing to do with sieges. As Charles Esdaile has noted, French armies sacked a number of Spanish towns that happened to be next to field battles and engagements, where Spanish forces were defeated or routed. This was the case with the sacks of Córdoba (30 June 1808), Medina de Rio Seco (14 July 1808), Burgos (10 November 1808) and Uclés (13 January 1809).Footnote ¹⁷⁰ Outside Córdoba, the French routed a Spanish force made up largely of armed peasants and town citizens; in all other cases, the French defeated and dispersed regular Spanish armies.Footnote ¹⁷¹ The broader and evolving context here was regional insurrection throughout the Spanish provinces in the aftermath of the Madrid uprising in May 1808, the French withdrawal to the Ebro following their defeat at the Battle of Bailén on 19 July 1808, and then Napoleon’s sweeping counteroffensive in late 1808. In the aftermath of the battles in question, there was no organised armed defence of the towns, with the French not even forced to escalade or storm.Footnote ¹⁷² Of the sack of Burgos, where the French poured into the town in pursuit of fleeing Spanish soldiers, Comte Miot de Melito wrote: ‘The churches were sacked, the streets were choked with the dead and the dying; in fact we witnessed all the horrors of an assault, although the town had made no defence!’Footnote ¹⁷³ Early the following year, across the border in Portugal, the French sacked Oporto on 28 March 1809, after Marshal Nicolas Jean-de-Dieu Soult’s forces attacked and overran the long lines of defensive earthworks that had been thrown up over the surrounding hills just to the north of the city. Richard Henegan, an English commissar, was present: ‘What pen can trace the atrocities that were perpetrated in every quarter of the city at this awful moment.’Footnote ¹⁷⁴

But the most common type of sack during the Peninsular War did not arise out of conventional warfare at all, but rather from the French military repression of rural insurrection and the petite guerre, as was also notably the case in Calabria from 1806, and the Tyrol in 1809.Footnote ¹⁷⁵ David Bell has argued that whilst the savage pacification processes of French republican forces in the Vendée – exemplified by Turreau’s infamous ‘hell columns’ – were never repeated on that scale and degree of totalisation in the Napoleonic Wars, the Vendée nevertheless created a ‘matrix of French experience’ for dealing with revolt and partisan warfare thereafter.Footnote ¹⁷⁶ Napoleonic military repression was framed as yet another war on ‘brigands’, where the laws of war essentially did not apply, the ‘enemy’ identified as the population at large, not merely those under arms.Footnote ¹⁷⁷ Napoleon himself, in 1806, explained to his brother, Joseph, the politics regarding the systematic destruction of villages in Calabria: ‘That is how we must treat villages that rebel. This is the law of war, but it is also a duty laid down by politics.’Footnote ¹⁷⁸ In the war against the guerrillas in Spain, villages were routinely sacked and razed, prisoners killed and the inhabitants massacred, often justified as collective reprisals against atrocities inflicted on Imperial troops.Footnote ¹⁷⁹ In 1810, in a pattern repeated elsewhere in Spain, when Marshal Soult pushed into Andalusia to deal with popular resistance, he ordered those under arms to be shot, and the towns of Algodonales and Grazalema were made examples of.Footnote ¹⁸⁰

As was traditionally the case in European warfare, the inhabitants of villages and towns also suffered plunder and violence at the hands of retreating and demoralised armies, especially when supply chains broke down and deprivation set in. This was exacerbated by the fact that Napoleon’s armies, following the custom of their French Revolutionary predecessors, were largely dependent on living off the land, which was especially challenging in the Iberian Peninsula given its geography and limited resources.Footnote ¹⁸¹ The worst case of this type of marauding and violence occurred when Masséna’s Army of Portugal found itself in the ‘scorched earth’ of central Portugal in 1810 and early 1811 before the Lines of Torres Vedras and then retreated to the Spanish border in March 1811, leaving behind a trail of atrocities and ruined villages and towns.Footnote ¹⁸² On a much lesser scale, retreating British armies in the war were also guilty of plunder and destruction, the most infamous case being Sir John Moore’s retreat to Corunna through Galicia in the winter of 1808–09. The British officer, Robert Blakeney, described the town of Bembibre as having ‘exhibited all the appearance of a place lately stormed and pillaged’.Footnote ¹⁸³

Finally, we have ‘classic’ sacks in Spain arising from siege warfare. As we shall see in Chapter 2, even in the Peninsular War, most sieges did not end in sack but rather capitulation. Nevertheless, both French and British forces sacked a number of towns over the course of the war’s long succession of sieges. These sacks were not random: in every instance, the defending garrisons refused to capitulate in the face of practicable breaches, necessitating a general assault by the attacking forces. Massacre and sack followed to varying degrees in the wake of the French storming of Lerida (1810), Tarragona (1811) and Castro-Urdiales (1813).Footnote ¹⁸⁴ The first two occurred under Suchet during his campaigns to conquer and subjugate eastern Spain in 1810 and 1811. At Lerida, the inhabitants were exposed to plunder and atrocities after two breaches were stormed on the night of 13 May, Suchet writing of ‘scenes of disorder, inevitable in a city taken by assault’; the inhabitants were later herded into the citadel and bombarded to force the garrison’s hand into capitulating.Footnote ¹⁸⁵ The worst French massacre and sack occurred in June 1811 at the siege of Tarragona on the Catalonian coast. This fiercely contested two-month-long siege ended when Suchet’s forces stormed the breaches of the city’s upper defences, putting the garrison to the sword and sacking the town.Footnote ¹⁸⁶ Two years later, on the Bay of Biscay, French and Italian troops under the command of General Maximilien Foy besieged and sacked the port town of Castro-Urdiales, after storming a 60-foot-wide breach in the curtain wall; civilians were killed and raped in the sack, but most of the Spanish garrison, which had retreated to the citadel, was evacuated onto British ships.Footnote ¹⁸⁷ But, of course, the storming and sack of besieged towns in the Peninsular War was not the work of Napoleon’s forces alone. Of the four major sieges that the British conducted in Spain in 1812 and 1813 – Ciudad Rodrigo, Badajoz, Burgos and San Sebastian – three ultimately ended in sack, with the siege of Burgos abandoned.

And so it transpired that exactly a century after the War of the Spanish Succession, British soldiers once again found themselves in Spain, embroiled in a long and complex war, with sieges playing a significant role. Except this was a rather different kind of war. Notwithstanding obstinate defences and breach assaults in Spain a century earlier, the conventions of eighteenth-century siege warfare now faced new challenges – for a new generation of soldiers. A hundred years, too, separated the exploits of Britain’s two greatest generals of the long eighteenth century: Marlborough and Wellington. Long after the passing of the high age of European siege warfare, one in which Marlborough played a key role, Wellington still commanded his own fair share of sieges, in India and Spain. But whereas Marlborough, who admittedly did not serve in Spain, was never forced to take a fortress town by breach assault, nor deal with sack in the aftermath, Wellington, as he lamented in 1813, had unfortunately overseen all too many siege storms throughout his career.Footnote ¹⁸⁸

In keeping with Marlborough’s record, in Laurence Sterne’s mid-eighteenth-century novel, Tristram Shandy, Uncle Toby never has to stage the general storming of a fortress town in his back garden as he re-enacts Marlborough’s sieges. Uncle Toby, for all his obsession with sieges and love of war, is also at pains to stress that he deeply feels the human suffering of war, that alongside bravery, glory and honour, ‘tis another thing to reflect on the miseries of war; – to view the desolation of whole countries’.Footnote ¹⁸⁹ Toby’s offsider, Corporal Trim, also professes his own humanity as a soldier: ‘I never refused quarter in my life to any man who cried out for it; – but to a woman or child, continued Trim, before I would level my musket at them, I would lose my life a thousand times.’Footnote ¹⁹⁰ Yet, the character of Trim, espousing eighteenth-century ideals of humanity in war, never faced the trials of storm and sack. For Wellington’s all too real officers and soldiers it was another matter entirely. Storming breaches would prove to be the ultimate test of not only their martial courage and ability to manage fear and rage but also of their humanity in the face of enemy soldiers and town inhabitants. But before the storm came no surrender.