Walls of Rome
Walls of Rome
Walls of Rome
FORTRESS 71
NIC FIELDS
First published in Great Britain in 2008 by Osprey Publishing, Midland House, West Way, Botley, Oxford OX2 OPH, United Kingdom 443 Park Avenue South, New York, NY 10016, USA Email: info@ospreypublishing.com
2008 Osprey Publishing Ltd.
ARTIST'S NOTE
Readers may care to note that the original paintings from which the colour plates in this book were prepared are available for private sale. All reproduction copyright whatsoever is retained by the Publishers. All enquiries should be addressed to: Peter Dennis, Fieldhead, The Park, Mansfield, Notts, NG18 2AT, UK The Publishers regret that they can enter into no correspondence upon this matter.
All rights reserved. Apart from any fair dealing for the purpose of private study, research, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, electrical, chemical, mechanical, optical, photocopying, recording or otherwise, without the prior written permission of the copyright owner. Enquiries should be addressed to the Publishers. A CIP catalogue record for this book is available from the British Library. ISBN 978 184603 198 4 Editorial by llios Publishing, Oxford, UK (www.iliospublishing.com) Page layout by Ken Vail Graphic Design, Cambridge, UK (kvgd.com) Cartography by The Map Studio, Romsey, UK Typeset in Sabon and Myriad Pro Index by Alan Thatcher Originated by United Graphic Pte Ltd, Singapore Printed and bound in China through Bookbuilders 08 09 10 11 12 10 9 8 7 6 5 4 3 2 1
FOR A CATALOGUE OF ALL BOOKS PUBLISHED BY OSPREY MILITARY AND AVIATION PLEASE CONTAG: NORTH AMERICA Osprey Direct, c/o Random House Distribution Center, 400 Hahn Road, Westminster, MD 21157 Email: info@ospreydirect.com ALL OTHER REGIONS Osprey Direct UK, PO Box 140, Wellingborough, Northants, NN8 2FA, UK Email: info@ospreydirect.co.uk
www.ospreypublishing.com
CONTENTS
INTRODUCTION CHRONOLOGY ROME'S EARLY DEFENCES
The seven hills. First defences. The Servian wall
4
5
11
AURELIAN'S WALL
Aurelian's Rome. Tracing the circuit. Design of the wall The method of construction The anatomy of the wall The function of the wall After Aurelian
23
40
45
AURELIAN'S LEGACY
The trace italienne Garibaldi's Rome
53
61
62
63
64
INTRODUCTION
The walls of Rome evolved over many centuries. The first early ditches and banks were thrown up by Rome's founding fathers. In the 4th century BC the Roman king Servius Tullius created what became known as the Servian wall, built of tufa stone and featuring a number of gates. Servius's creation would serve Rome well during the Second Punic War (218-201 BC), its formidable strength warding off siege by Hannibal's forces. As the power of Rome grew, so did its capital, which expanded beyond the limits of the Servian wall. A long period of peace followed the founding of the empire, but in the third century AD new threats appeared. Barbarian raiders lay waiting on the borders of the empire, and economic crisis brought it almost to the point of collapse. The emperor Aurelian (AD 214-75), by stupendous military exertions, physically reunited the Roman empire under his iron rule. However, it was an empire battered and traumatized, and for the first time since Hannibal had ridden up to Porta Collina, the city of Rome itself had become vulnerable. This situation led to Aurelian's greatest monumental achievement - Aurelian's wall, built between AD 271 and 275. Still bearing his name to this day, it was erected to protect Rome following its narrow escape from a Germanic incursion that had penetrated deep into the Italian peninsula. In AD 307, barely 30 years after the completion of the wall, the usurper Maxentius, faced with the prospect of defending Rome against two Roman armies - one led by Severus, the duly appointed western Caesar, and the other by Galerius, the eastern Augustus - reorganized the Aurelianic defences. This he did by doubling their height, blocking several lesser entrances and strengthening a number of the remaining gateways. According to Lactantius, he 'began the digging of a ditch but did not complete it' (De mortibus persecutorum 27). A hundred or so years later, in the first decade of the 5th century AD, the defences were again reorganized by Stilicho, the regent of Honorius (r. AD 395-423). They proved an effective defence against two sieges by the Goths under Alaric, but failed to withstand the third attempt (AD 410). Nevertheless, Aurelian's wall continued to playa significant part in the history of Rome thereafter. Repaired twice in the mid 5th and early 6th centuries, the wall played a crucial role in the sieges and counter-sieges of the Gothic wars of Iustinianus (Justinian), during which it was twice repaired and strengthened by Belisarius (AD 537 and AD 546).
4
Although embellished, strengthened and restored many times over, it was Aurelian's original structure that remained the basis of Rome's defences down to the mid 19th century, when Garibaldi's pro-Italian unification forces (who had overthrown Pope Pius IX and declared Rome a republic) managed for some time to withstand the French army coming to the Pope's aid. Today the remains of his wall are still discernible along much of the original circuit. Aurelian's wall is arguably the best preserved of all city walls in the Roman empire, and even the present-day traveller cannot help but be impressed by the majesty of the imposing ruins.
Aurelian's wall along Viale Metronia between the Metrobia and Latina gates - a general view looking south. (Author's collection)
CHRONOLOGY
4th century Be The llkm-Iong Servian wall is built around the city of Rome
AD
235
244 248 249
Murder of the emperor Severus Alexander - beginning of the period known as the 3rd Century Crisis First campaign of Shapur, King of Persia, against Rome The Roman emperor Philip the Arab celebrates the millennium of Rome Goths cross the Danube Shapur's second campaign against Rome - Antioch is sacked Goths invade the Balkans Third campaign of Shapur - destruction of Dura-Europus; Franks cross the lower Rhine Fourth campaign of Shapur against Rome; establishment of Gallic empire by breakaway provinces of Gaul, Britannia and Hispania The Heruli sack Athens The Alamanni invade Italy; Goths invade the Balkans, but Claudius defeats them at Naissus (Nis)
5
AD AD
AD AD AD AD
AD
AD AD
AD
270 271 272 274 275 324 395 410 455 493 536
The death of Claudius - Aurelian is proclaimed emperor; the Iuthungi invade Italy The Vandals invade Pannonia; the Palmyrene empress Zenobia invades Syria and Asia Minor; construction begins on Aurelian's wall in Rome Aurelian recovers Egypt, and campaigns against the Palmyrene empire (the former provinces of Syria, Palestine and Egypt) Aurelian quashes the Gallic empire Murder of Aurelian Constantinus becomes sole emperor - foundation of Constantinople (Istanbul) Death of Emperor Theodosius - the empire is split into the east (Arcadius) and west (Honorius) The Gothic king Alaric takes Rome - the city is pillaged for three days The Vandals under Gaiseric capture Rome and occupy it for 14 days Foundation of the Ostrogothic Kingdom of Italy by Theodoric The emperor Justinian's general Belisarius retakes Rome from the Ostrogoths, the first of many struggles for control of the city
AD
AD
AD
AD
AD
AD
AD
AD AD AD
The Servian wall north-east of the Viminal in Piazza dei Cinquecento, a general view looking north-west. Observe the non-alignment of vertical join between two stretches of the wall. (Author's collection)
t
5~0 ~ardS
500 meters
which lay south of an island in the river, was overlooked by a group of hills that harboured an adequate number of fresh-water springs. The hills themselves, which rise from the Latium Plain, were well wooded, fairly precipitous and defensible. The site, therefore, afforded some protection against floods, predators and the like. Cicero may have once boasted 'that Romulus had from the outset the divine inspiration to make his city the seat of a mighty empire' (De re publica 2.10), but in the early days of its career nothing seemed to single out for future greatness a puny riverine settlement that long lay dormant. In these obscure times Rome was allied with other Latin settlements in Latium, and the seasonal battles that preoccupied the Latins were little more than internal squabbles over cattle rustling, water rights, and arable land.
7
A close-up shot of the Servian wall in Piazza dei Cinquento, showing its dry stone construction using ashlar blocks of yellow and grey tufa. (Author's collection)
A bronze head (Thessalonica, Archaeological Museum, 4303) of Severus Alexander (r. AD 222-35). His assassination would lead to a half-century of anarchy in and around the empire. (Author's collection)
featured a number of sanctuaries, such as that dedicated to the mysterious Quirinus. The latter was identified by the Romans with both Mars and his miraculous son, the deified Romulus. Known as the 'plebeian' hill, the Aventine sat outside the pomoerium - a ritual furrow made by a yoked bull and cow, so marking the area of a sacredly constituted city - until the early Principate. This hill was the site of the cult of Ceres, Libera and Liber Pater. The temple (496 Be), which was adorned with terracotta decorations executed by Greek artists, also functioned as the headquarters of the plebeian aediles and contained their archives as well as copies of senatus consulta, or decrees of the Senate.
First defences
Legend has it that Remus was killed when he mockingly leapt over the fortifications that Romulus was constructing on the Palatine. While Remus desired to build on the Aventine, Romulus much preferred the Palatine, and traces of a palisade defence dating to around that period have been found on this hill. As for the first defences of the city as a whole, it was said that they were erected during the phase of Etruscan domination.
The marble sarcophagus known as the 'Grande Ludovisi' (Rome, Palazzo Altemps, 8574), depicting Herennius Etruscus, son of Decius (r. AD 249-51), riding against the Goths at Abrittus. (Esther Carre)
A marble bust (Paris, Musee du Louvre, MR511) of Gallienus (r. AD 253-68). The fact he survived the ignominious capture of his father and the widespread unrest that inevitably followed suggests that he was a singular man. (Esther Carre).
Traditionally the last three kings of Rome were Etruscan, and it was the second of these, Servius Tullius (r. 579-534 BC), who was believed by later Romans to have constructed a massive stone wall around their city. The historian Livy (1.36.1, 44.3), writing under Augustus, reports that the project had been planned by Tarquinius Priscus but was eventually carried out by his son-in-law and successor. Livy, like other writers of the early Principate, believed that the wall of Servius Tullius could be identified with the stone enceinte that could still be seen in his day encircling the Capitol, Palatine, Caelian, Quirinal, Viminal, Aventine and part of the Esquiline. By the end of the 1st century BC this wall had long been out of use, appeared to be of great antiquity, and could thus be identified with the only early defences of the city mentioned in the historical tradition. But Livy and his fellow historians were mistaken. The earliest bank-and-ditch defences of Rome, which may date to circa 540 BC, did not form a complete circuit around the city but only protected areas vulnerable to attack or raiding. The massive earth agger or rampart associated with the early ditch seems not to have been erected before about 480 BC, and probably no later than about 450 BC (Todd 1978: 14). It was during this period that Rome, along with other Latin cities, fought a series of petty and inconclusive wars with the neighbouring highlanders, the Aequi, the Volsci, and the Hernici, who threatened to overrun Latium. It was not until the end of the fifth century BC that the most formidable of these warlike mountain tribes, the Volsci, had been pushed out of the small, but rich coastal plain.
10
No projecting towers were provided either at the time of the original fortifications or later. The gateways seem to have been simple openings, the single entranceways being covered by towers placed against the internal wall face, a gate-type that persisted until the 1st century BC in the Italian peninsula (Todd 1978: 19). There were later modifications to the Servian wall. According to Appian (Bellum civilia 1.66) the consuls of 87 BC, faced by the renegade army of Caius Marius, tried to strengthen the city defences by digging new ditches, restoring the wall and creating emplacements for artillery. The subsequent history of the wall, however, appears to have been one of progressive decay and dilapidation. Repairs to the Servian wall were not an option to Aurelian. It had largely been subsumed and obscured by subsequent building, and even by the reign of Augustus its exact line was uncertain.
BELOW LEFT A member of the re-enactment group Quinta equipped and dressed as a 3rd-century cavalryman. His bronze scale armour and wooden oval shield are based on evidence from Dura-Europus. (Author's collection) BELOW RIGHT The monumental arch carrying the Aqua Antoniniana, in a general view looking south in Via di San Sebastiano towards the rear of Porta Appia. Later Maxentian rebuilding would see this serving as an inner gate. (Author's collection)
11
The assassination (in March AD 235) and replacement of Severus Alexander by a tough career soldier from Thrace, Maximinus Thrax (r. AD 235-38), was a stark reminder that the empire needed emperors who knew the army. An equestrian outside the ruling clique, Maximinus had exploited the opportunities of the Severan army to gain numerous senior appointments. However, the senatorial aristocracy could not agree to this particular appointment, and, after an eyeball-to-eyeball confrontation, they managed to face the army down. The subsequent run of emperors - the three Gordiani, Decius, Trebonianus Gallus, Valerianus and Gallienus - was one of 'gentlemen officers'. Yet their military misfortunes would finally destroy the prestige of the Augustan system, leaving military rule as the only alternative. Maximinus, the Thracian soldier of obscure birth and exclusively military experience, had set the trend whereby the army called the shots, putting forward their own commanders as new emperors. As the 3rd century AD progressed, the number of senior army positions held by men of senatorial rank gradually declined, and this move away from the traditional mixed military and civilian career would gather momentum under Gallienus. Far more opportunities lay open to equestrians, especially those who campaigned under the emperor himself. The equestrian officers who now dominated the army were in many respects career soldiers, owing their advancement purely to their military record and the favour of the ruling emperor. Perverse as it may seem, it was usually these men who plotted to murder an emperor and nominated a usurper from their group. Several of the most successful emperors of the second half of the century came from a virtual junta of professional officers from the Danubian provinces (hence the loose term 'Illyrian'), men of obscure origins but undoubted military ability who worked their way up through the crisis years to the highest commands and then doggedly fought invaders and each other.
a rock-cut relief from Naqs-e Rustam. Two Roman emperors are shown submitting to the king on horseback, one of whom is Philip, who sues for peace on bended knee. A further Persian offensive led to the occupation of Armenia, the devastation of Syria, and the capture of Antioch (AD 252), the great commercial capital of Hellenic Syria. Roman Antioch had never before fallen to an enemy. It was retaken with some difficulty, but henceforth Antioch was to be a piece in the strategic Rome-Persia game of chess, the theatre of which had now shifted alarmingly from the Euphrates, the boundary between the olive and the date, to only a short distance inland from the Mediterranean, the cradle of GraecoRoman culture. The third campaign of Shapur culminated in the capture and destruction of the border fortress of Dura-Europus (AD 256). It was never reoccupied. In the aforementioned rock-cut relief from Naqs-e Rustam the other emperor is Valerianus (r. AD 253-60), he and his army having surrendered to Shapur after being humiliated at Edessa (AD 260). Although Valerianus is portrayed standing up, his hands are held by the king, a reference to the fact he was taken prisoner, an ignominy that had never previously befallen a Roman emperor. To reinforce the insult, Shapur is said to have used the captive emperor as a human mounting block. One lurid story even claims that after Valerianus died in miserable servitude, he was flayed and his skin, dyed crimson, was stuffed with straw and put on public display. With the Roman army of the east in utter disarray, it was left to Septimius Odaenathus, ruler of Palmyra, or Tadmor to give this Roman protectorate its Semitic name, to play the major role in forcing Shapur to withdraw from Roman territory. With his father disappeared into Persian captivity, Gallienus, who had been installed as co-emperor seven years earlier, now assumed full power. His area of effective control, however, was confined to Italy, Dalmatia, Greece, the western Danube and Africa, and though he managed to hold on to power for a further eight years, he was never able to reassert his authority over the whole empire.
II
The Castra Praetoria, the camp for the Praetorian Guard, was erected in AD 23 and was a reflection of the rise to prominence of L. Aelius Seianus, commonly known as Sejanus, the highly ambitious Praetorian prefect under Tiberius. As Tacitus dryly notes, the 'command of the guard had hitherto been of slight importance. Sejanus enhanced it by concentrating the guard cohorts, scattered about Rome, in one camp' (Annates 4.2.1). The new camp stood at the north-eastern edge of the city on the Viminal hill and enclosed an area of just 16.72 hectares, about two-thirds the size of a contemporary legionary fortress accommodating two legions. The original Tiberian curtains, of brick-faced concrete, stood some 35m high and supported a rampart-walk protected by battlements. These were then heightened, probably by Caracalla (r. AD 211-17), and subsequently repaired and given loftier towers by Gordianus III following the chaos of AD 238. The next major change was when Aurelian decided to fortify Rome, consequently
15
A Claud ian monumental arch carrying the Aqua Claudia-Anio Novus, subsequently incorporated into Aurelian's wall to form the portals of Porta Praenestina-Labicana. (Author's collection)
the opportunities of what seemed limitless reserves of booty in the Roman provinces, the earlier Germanic tribes had fused into tribal confederations. By the early 3rd century three major groupings had emerged. The Suevic tribes of central Germania, with whom Caesar had first made contact (he had been unnerved by their ferocity), had formed into the Alamanni; those of the lower Rhine group into the Franks; and the sea-peoples at the mouth of the Elbe and Wesser into the Saxons. Though still of loose internal unity, the scale of military expedition these confederations could now mount was of an entirely new order, beyond what the existing Roman frontier defences had been designed to deal with. Whether built in stone, timber, or earth and turf, whether consisting of a military way or a line of a river, these fixed frontier lines (limites) separated those outside from those within, those becoming romanized from their still barbarous neighbours. Trade and contact persisted, but it had been geographically channelled through supervised customs and crossing points. These physical barriers, therefore, had not been intended as impregnable fortifications or fighting platforms. On the contrary, they had been designed for surveillance and active, forward defence against anticipated raids or low-level incursions. As any fighting was intended to take place in the immediate zone beyond Roman territory, concentrated attacks could easily penetrate these defences. The other area where the situation was changing to Rome's detriment was the lower Danube region. Starting in the late 2nd century the Gothic peoples in their tribal groupings had begun to shift slowly south-east from the Baltic littoral toward the Black Sea steppe. In AD 249, while Roman armies were occupied in civil war elsewhere, the Goths seized their opportunity and penetrated parts of Thrace and Asia Minor. Decius (r. AD 249-51), along with his son and heir, were both ambushed and cut down by this new foe. He was the first emperor to die in battle against enemies outside the empire.
16
The Papal fortress of Castel Sant'Angelo, which incorporates the cylindrical brick drum of the Mausoleum of Hadrian sited and built
(AD
135-39) so as to impress
'Super-heavy' cavalry
Palmyra had for long patrolled and policed the eastward caravan routes on which its prosperity depended. This was a pertinent preparation for military power. In other respects, also, the Semitic, semi-Hellenized Palmyra was well qualified to fill the role of Roman sword-bearer in the east. The Sassanid army relied extensively on noble cavalry, the cJibanarii ('oven-men', ct. Greek klibanos, baking oven). These were horsemen, as their name suggests, fully encased in metal scale armour and mounted on horses protected by housings of leather or thick felt. The Palmyrene army also deployed heavy-armoured cavalry, the cataphractarii. By comparison, however, the Palmyrene cataphractus was a fully armoured man aboard a horse that was also usually armoured, but not necessarily so. Both Sassanid and Palmyrene horsemen, however, were armed with a heavy spear some 3.65m in length and held two-handed without a shield. The contus (Greek kont6s) was a weapon for shock action, being driven home with the full thrust of the body behind it. The greater weight of men, horse and equipment meant their charge was considered to be more powerful than that of conventional cavalry.
northern Italy. The Iuthungi defeated Aurelian at Placentia (Piacenza) and advanced, apparently irresistibly, down the Via Aemilia into central Italy, threatening Rome itself. Panic gripped the city, for no significant force stood between it and the invaders. Rome had long since outgrown and built over its ancient city walls. Besides, the vast empire and the strength of its legions had long been an ample buffer to protect the city, along with the rest of the Italian peninsula, from external threats. As the terrified, un-walled capital hurriedly made what preparations it could, the emperor regrouped his battered army and was able to turn back the invasion at Fanum Fortunae (Fano), then destroyed it completely in the open plains near Ticinum (Pavia). In recognition of this triumph he assumed the title Germanicus Maximus. Yet Aurelian was so alarmed that he ordered the immediate construction of a defensive circuit around Rome (spring AD 271), the famous wall that is still associated with his name to this day.1 During his short reign Aurelian had to deal with a number of challenges to his imperial authority, the greatest of which came in spring AD 272 when several eastern provinces of the empire were annexed by Septimia Zenobia, queen of Palmyra. By expanding into the power vacuum of the east, her late husband Odaenathus, though loyally defending the empire against Persia, had in fact adroitly created for himself a position of independence in the caravan city of Palmyra. A later author, looking back in disdain on the recent past, may have moaned that 'the ruler of Palmyra thought himself our equal' (Panegyrici Latini 8.10), but in AD 261 Gallienus had belatedly appointed Odaenathus vice-regent of the east, declaring him corrector totius orientis; he could do little else. The Palmyrene prince thus held the supreme command of all the armed forces in the east, with full authority over the provincial governors of the entire region from Asia Minor to Egypt. As a result of this command Odaenathus assumed the title dux Romanorum.
1
See especially, Zosimus 1.49.2, Aurelius Victor Liber de Caesaribus 35.7, Epitome 36.6, Eutropius 9.15.1, SHA Aurelian 39.2, Jerome Chronicle 223, and Malalas 12.30.
18
In AD 267 Odaenathus' talented widow Zenobia (Bath-Zabbai in Aramaic) inherited his position of unprecedented power in the Roman east and waited for an opportunity to break completely with Rome. And so while Claudius and Aurelius were preoccupied with the Goths in the mountains of northern Thrace, she easily secured Arabia and Iudaea (spring AD 270). Then as the Iuthungi overran northern Italy and threatened the capital, she overran much of Egypt (autumn AD 270). Next up was Syria and most of Asia Minor, including Galatia (spring AD 271). Aurelian, though at first conciliatory, later felt obliged to reassert Roman authority. After assembling a substantial expeditionary force in Asia Minor, he quickly vanquished the formidable cavalry-army of Zenobia in two battles. Zabdas (Zabda), Zenobia's general, was unable to hold Antioch (spring AD 272) and made a second stand at Emesa (Horns). Here the Palmyrene cataphractarii drove Aurelian's cavalry from the field, but the emperor won the battle during their absence and the remnants of the Palmyrene forces soon found themselves beleaguered in Palmyra. Palmyra fell to Aurelian, despite Zenobia's efforts to involve Persia (summer AD 272). Zenobia was captured as she attempted to cross the Euphrates, but Aurelian spared her along with the city. The following year, after successfully defeating the Carpi along the Danube, the emperor was incensed when he heard that Palmyra had revolted and had slaughtered the Roman garrison installed there. He then executed a well-conducted foray that surprised the defenders, captured Palmyra, and mercilessly razed it. From that time the City of Palms sank into quiet oblivion to become an unimportant provincial town on the outskirts of the Roman empire. Zenobia, however, lived to walk in Aurelian's triumph (autumn AD 274) and ended her life as a fashionable Roman hostess with a pension and a villa. With the affairs of the east firmly under his grip, Aurelian now turned his attention west (summer AD 274), specifically to the sub-empire of Gaul, Britannia and Iberia, which had been pursuing its independent course with some success for well over a decade. The emperor Postumus (r. AD 260-68) had been lynched by his own soldiers and the present ruler, Tetricus, believed that he could rule over his own Gallic empire just as Postumus had done before. Together with his young son of the same name he had managed to hold out for nearly three years, but now it was all up for him, though Aurelian did spare the lives of father and son - they both featured alongside Zenobia as the star attraction in Aurelian's magnificent extravaganza (Eutropius 9.13.2) - even going so far as re-confirming their senatorial status and granting the elder Tetricus a civil administrative post in Italy. The political aberration of parallel rulers holding sway in different parts of the empire, which had persisted for nearly a decade and a half since the
A close-up shot of Via Ardeatina, running south-east out of Porta Ardeatina, showing the agger or embankment, basalt metalling, and one of the gutters. (Author's collection)
19
capture of Valerianus, was at an end. By the springtime of AD 275 unity seemed to have been restored to the empire. Aurelian had rightly taken the title restitutor orbis, 'restorer of the world', and everything seemed to suggest that the burly soldier-emperor was in complete control of things. He was even able to work on bolstering the economy through the most comprehensive overhaul of the imperial monetary system since the reign of Augustus. But for an emperor of the 3rd century, danger was always lurking just around the corner. Aurelian, who was marching eastward through Thrace to wage war on the Persians, was assassinated in the vicinity of Byzantium by his senior officers: so disappeared from history an emperor who had done everything to halt the decline of the empire. Aurelian reigned for just five years and two months, but under his energetic rule the empire had been granted time to recover. Like Gallienus before him, Aurelian realized that the empire could only be protected if the static concepts of frontier defence were abandoned. With the deployment of field armies - Aurelian placed his confidence in the cavalry corps developed by Gallienus - there was now a conscious shift towards strategic mobility.
Aurelian's army
Internal instability had led to losses and defeats on all imperial frontiers and further encouraged local rebellions and military coups. Each emperor was required to campaign with little respite, since he could rarely afford to entrust the command of an army to a potential rival. When the emperor was required to campaign in one theatre of operations there was a great danger that other parts of the empire, feeling their own difficulties were being neglected, would create a rival. It was Gallienus who developed the weapon with which his Illyrian successors fought off Persians and Germanic tribes alike. This was what was known at the time as the 'elite army', namely a mobile force not committed to frontier defence. Made up of detachments (vexillationes) drawn from frontier units in Britannia and on the Rhine and Danube, this force operated independently and was perhaps the forerunner of the 4th-century comitatenses, or field armies. The elite army In the 50 years from the assassination of Alexander Severus to the temporary establishment of peace under Diocletianus, there was an 'elite army' permanently in the field. It was not always exactly the same army that consisted of exactly the same units. Successive emperors commanded armies composed of vexillationes from various different legiones, cohortes and alae of the provincial garrisons, the choice of troops depending of course upon the location of the almost perpetual wars and availability of manpower. Although nothing new, these vexillationes, as opposed to whole legions, had now become the standard combat formation. The legion-based army of the Principate was designed primarily for delivering powerful offensive strikes at specific fixed targets. In the military context of the 3rd century, however, cavalry were fast becoming increasingly important in the defence of the empire and the struggle against rebels and usurpers. In both cases, mobility was essential. To move an army from the Rhine to Rome took eight weeks - and to the Euphrates six months. Roman armies could no longer choose the time and place for their battles and mount a campaign with the advantage of time and planning on their side. The days of overt imperialism were over, a time when
20
tribal aggression in any particular sector could be anticipated and neutralized outside Roman territory. Now the encounters were all too often sprung upon the emperor, by barbarians or by fellow Romans - sometimes simultaneously.
Gallienus's cavalry corps Sometime around AD 255, when he was defending the Rhine frontier and there was a desperate need for rapid movement, Gallienus created a cavalry corps (de Blois 1976: 26). He almost certainly employed the corps as part of the army hastily gathered together for the campaign against the Alamanni some three years later. This 'elite army' was quite small, Gallienus having assembled vexillationes from the static garrisons on the Rhine, Britannia, Pannonia and Noricum, and brought legio II Parthica and the Praetorian Guard from Rome. From about AD 260 the cavalry corps was stationed at Mediolanum (Milan) under its single commander Aureolus, whose task, according to Zosimus (1.40.1), was to prevent the anticipated invasion by Postumus across the Alps from Gaul. The threat from the breakaway Gallic empire was probably not Gallienus's sole concern, however. The much more pressing reason for occupying Mediolanum in considerable strength, with emphasis on mobility, was the threat posed by the Alamanni immediately to the north in Raetia. Though little is known about Gallienus's cavalry corps, it is likely that he seconded his troopers, undoubtedly horsemen of proven ability and skill, from existing units. It is known, for instance, that he extracted men from the mounted troops stationed in Dalmatia, the equites Dalmatae. Besides the regular alae and cohortes equitatae of the provincial garrisons, there were tribal contingents available also, such as the Mauri and Osrhoeni recruited by Alexander Severus in the east and brought to Rhine by Maximinus (SHA Severus Alexander 61.8, two Maximini 11.1, 7, Herodian 7.2.1-2). If they had enlisted for 25 years, they would have had a few years left to serve when Gallienus was seeking experienced horsemen (Southern-Dixon 2000: 12). These horsemen were certainly brigaded together, but it is not known how they fought together on the battlefield. They seem to have employed different, specialized fighting techniques: the Mauri, nimble horsemen of legendary ferocity, were armed with javelins and the Osrhoeni, as befitting eastern horse-archers, with the powerful composite bow. Individual units may have been employed for different purposes, but the cavalry had only one commander, and this unity of command implies unity of operation. It also facilitated potential usurpations, since the commander of the cavalry corps had an excellent power base at his immediate disposal. As the most influential, hence most dangerous, subject in the embattled empire, the brilliant but capricious Aureolus could not resist the temptation to rebel against Gallienus, but he did not succeed to the throne; he merely cleared the path to it for Claudius, before being murdered himself. The questions of whether the cavalry corps survived and whether Gallienus is really the innovator behind the 4th-century comitatenses are unanswerable, given the lack of contemporaneous evidence. A Byzantine chronicler, George Cedrenus, states quite firmly that the emperor was the founder of the first cavalry army, emphasizing 'the Roman army having previously been largely infantry' (Compendium Historiarum 454). Not all scholars would agree with his judgement, Tomlin (1989: 223), for instance, pointing out the independent cavalry forces that won victories for Trajan and Septimius Severus. The Byzantine writers had the benefit of hindsight and were accustomed to the use of cavalry armies from the time of Constantinus (r. AD 306-37) onwards; therefore it seems natural that any army composed purely of horsemen, which
21
Aurelian's wall along Viale di Porta Ardeatina between the Ardeatina and Appia gates, in a general view looking north that shows the regularly placed, projecting towers. (Author's collection)
was moreover not part of any provincial garrison but answerable via its commander to the emperor alone, would seem to be a direct forerunner of the later cavalry armies. De Blois (1976: 28) points out that the cavalry were no longer stationed at Mediolanum after about AD 285. By this time the Gallic empire had been quashed and Gaul was back again in the fold of the empire. De Blois takes the view that Gallienus's cavalry corps was not unlike the vexillationes employed in other wars, brought together temporarily for a specific purpose and disbanded when that purpose had been fulfilled. This view is shared by Ferrill (1986: 32), who thinks that Gallienus had no permanent policy in mind. Southern and Dixon (2000: 13), on the other hand, raise a minor point that possibly contradicts this thesis and, in part, goes some way to rehabilitate the opinions of older scholars. Numismatic evidence demonstrates that the title given to Gallienus's cavalry corps was simply equites, rather than ala or the less permanent vexillatio. This use of the non-specific title possibly signifies that the corps was not intended to function after the fashion of the provincial alae, but at the same time it was not intended to function as another vexillatio. An inscription (ILS 569), dating to the year after Gallienus' murder, preserves this distinction, whatever it may mean, by listing vexillationes adque equites side by side. Yet, given the current state of evidence, it is not possible to refute or endorse the theory that Gallienus's cavalry corps was intended to form the first permanent cavalry army, the precursor of the comitatenses. Gallienus originally developed the cavalry corps not from any comprehensive plan but in answer to his need for mobility on the Rhine, and then adapted the use of this mobile force to the multiple desperate situations facing him in the ensuing years. Legend claimed there had been 30 usurpers during his comparatively long reign, so the permanent survival of the cavalry corps could have been almost accidental at first, then regularized by custom afterwards. Southern and Dixon (2000: 14) suggest that its disappearance from Mediolanum is a possible indication that after AD 285 the corps was permanently in the field with the ruling emperor, employed in a similar fashion to the later comitatenses. It may have been used by Claudius against the Alamanni, who invaded Italy through Raetia just after Gallienus's death. After initial defeats, Claudius appointed Aurelian 'commander-in-chief of the cavalry' (SHA Aurelian 18.1). There is no
The brick-faced concrete construction on the interior of a tower in Viale Giotte. Once properly set, it created a homogeneous mass that was less vulnerable to collapse than dry stone construction. (Author's collection)
proof that this command embraced Gallienus's cavalry corps, but it is at least likely that the remnants of it formed the rump of Aurelian's cavalry corps. There were certainly Dalmatian and Mauritanian cavalry units in his corps, just as there were in that of Gallienus. When Claudius despatched Aurelius to tackle the Gothic incursion of AD 269, his sizeable command certainly included the Dalmatian cavalry, which he used to great effect (Zosimus 1.45, SHA Claudius 11.3-9). Likewise Aurelian, as emperor, used not only the Dalmatians that had distinguished themselves under his leadership in the Gothic wars, but also Mauritanian horsemen to defeat the formidable Palmyrene cataphractarii (Zosimus 1.50.3-51.1).
A close-up shot of the west gate-tower of Porta Appia, showing the rampart-walk and battlements; note the height of the merlons. (Author's collection)
AURELIAN'S WALL
Aurelian's Rome
The 3rd century was above all a world dominated by armies. The emperors, created by these armies, were almost exclusively men of comparatively humble origin promoted on merit rather than social standing. In this martial climate, the senatorial aristocracy in Rome lost its pride of place. It no longer retained the sole access to political power, still less to control it. But now emperors no longer resided or were made at Rome. It was more practical for emperors raised in the field surrounded by their own troops, as most were, to appoint men from among their own kind, men like Aurelian himself. Following the general rule of the day, his accession had been an army coup, set in a military camp, and marked by ceremonial acclamations hailing the new emperor as imperator.
From the late 2nd century onwards the centre of power in the empire had become increasingly peripatetic, following the emperor as he spent even more time in the frontier zones. 'Rome', as the conceptual capital of the empire, thus became divorced from the physical city of seven hills, or, as Herodian properly puts it, 'Rome is where the emperor is' (1.6.5, cf. 2.10.9). The emperor's presence on campaign often necessitated the elevation of his provincial headquarters into de facto regional 'capitals', that is, imperial centres in the frontier zones, often associated with regional branches of the imperial mint. This process would ultimately culminate in the foundation of Constantinople as a 'New Rome' on the Bosporus. All roads did not lead to the old Rome, yet the attention that Aurelian lavished on major building projects in Rome, which not only included the city walls but also a new camp to house the urban cohorts, is not so much a comment on the strategic or political importance of the capital as on his conviction that Rome still mattered symbolically. Before he departed for the Danubian front to deal with a renewed barbarian threat, the emperor personally oversaw the necessary arrangements for the building project to get swiftly under way (SHA Aurelian 22.1, Malalas 12.30).
economy of resources cannot be cited as the reason for the choice of line. Wherever deviations from the old customs boundary can be postulated, therefore, there existed sound strategic reasons for the line chosen (Watson 2004: 146). Strategy also demanded that the river bank itself should be strengthened to connect the fortifications on each side of the Tiber. Two stretches of Aurelian's wall were therefore built on the topmost of three embankment tiers along the east bank: one in the south, of some 800m; the other approximately three times as long, linking the trans-Tibertine walls with Porta Flaminia in the north. The circuit thus incorporated all the urban bridges within the fortifications, with the possible exception of the Pons Aelius and perhaps the Pons Neronianus, if the latter had not already been demolished by Aurelian's time. Though the evidence is wanting, it is highly likely that the fortifications reached across the Pons Aelius, incorporating the great cylindrical drum of the Mausoleum of Hadrian (now Castel Sant'Angelo) on the west bank, thereby making a bridgehead of this imposing structure. This is certainly the case in Procopius's day, when this structure was a bridgehead fort surrounded by strong walls, and as such attracted the attention of the besieging Ostrogoths. When it was first turned over to its new function Procopius does not say, but as the dramatic events he describes first-hand make plain (Wars 5.22.12-25), it would have made little strategic sense for Aurelian to have left the mausoleum and the bridge outside the fortification system.
Roads
The most famous legacy of the Romans, roads provided direct, well-maintained routes along which the army could move with ease. An important aspect of Rome's absorption of conquered territory was to construct roads linking new colonies to Rome. In Italy itself, the roads tended to follow its conquests both in time and space. With the annexation ofVeii (396 BC), we observe Rome's citizen-army move against the rest of Latium (Latin War, 341-328 BC) and then up into the central Apennine fastness of the Samnites (Samnite wars, 327-304 BC and 298-290 BC). Next came the turn of Etruria and Umbria to the north (viae Aurelia, Cassia, Flaminia and Aemilia), with Campania to the south soon to follow, then Lucania and the Greek poleis of southern Italy (Via Appia). Built on a monumental scale, these roads combined practical utility with visually impressive statements of power. Among the most important roads were the Via Salaria, the 'salt road', which led north from large salt pans situated at the mouth of the Tiber, and the Via Appia, called by the poet Statius (Silvae 4.3) the regina viarum, which ran for a total of 132 Roman miles to Capua.
An interior view of Aurelian's wall in Viale Carlo Felice between the Amphitheatrum Castrense and Porta Asinaria, showing the Maxentian galleried wall sitting upon the Aurelianic gallery. (Author's collection)
25
RIGHT The Amphitheatrum Castrense, built under Septimius Severus (r. AD 197-211), continued to function but was adapted to fit into Aurelian's wall. This exterior view looking north-west in Viale Castrense shows the blocked southern arcades. (Author's collection)
OPPOSITE PAGE The tomb of M. Virgilius Eurysaces was subsequently incorporated into the central tower of Porta Praenestina-Labicana. Pope Gregory XVI (1838) ordered the gateway to be demolished so that both the funerary monument and aqueduct might be clearly visible. (Author's collection)
BELOW The Castra Pretoria looking south at the junction of Viale del Policlinico with Piazza Girolamo Fabrizio. The camp still houses the barracks of Rome's resident military garrison. (Author's collection)
artillery. Gateways, of course, were potential weak points. They, too, became more heavily defended, often with flanking towers or towers on either side of a single, narrow entranceway. Usually a broad ditch, or ditches, surrounded the whole work. One very important factor for this change in defensive architecture was the shift in the nature and location of warfare. Whereas warfare had previously been conducted on or beyond the frontiers of the empire, in the 3rd century, as we have already discussed, the theatre of war shifted to being largely within the provinces. For instance, the cities in Gaul, when rebuilding and castramentation took place after the barbarian invasions of the middle decades of the century, saw their urban space typically contract. This reduction in size is illustrated by Augustodunum (Autun), whose Augustan walls enclosing 200 hectares were now supplemented by an inner circuit covering just ten. The larger circuit continued to stand, but it had probably little defensive value. While in the Principate cities had not required circuits, they now started to acquire powerful, as opposed to merely prestigious, urban fortifications and to change their appearance into the walled city typical of late antiquity.
RIGHT The Pyramid (and tomb) of Caius Cestius (built c. 18 Be), looking north-west in Piazzale Ostiense. This is the most striking example of a monument being adapted without ceremony to fit into the city walls. (Author's collection)
BELOW LEFT Towers along the Viale di Porta Tiburtina, looking north-west in Piazza Tiburtino towards Porta Tiburtina. Observe the original, Aurelianic doorway, and subsequent insertion of marble gun-ports, in the first tower. (Author's collection) BELOW RIGHT The north gate-tower of Porta Tiburtina; an interior view looking north-west in Piazza di San Lorenzo, showing the doorway giving access to the rampart-walk. (Author's collection)
28
An exterior view of Porta Appia, looking north in Via Appia Antica. Originally, like all Type I gateways, Porta Appia was equipped with a double-span archway flanked by round-fronted towers. (Author's collection)
various sizes and then sub-divided, usually into smaller triangles, and placed in position with the vertex at the inner concrete core of a wall and the base facing outwards. For Aurelian's wall the bricks and tiles, as Vitruvius (2.3.2) recommends, were reused and thus weathered. Most were Hadrianic, as indicated by the stamps, though some were as late as Severan, and probably came from buildings demolished to make room for the passage of the wall. There was a potential weakness in this type of wall construction method, specifically at the junction of the shallow facing and the concrete core. Although the builders of Aurelian's wall used material for their facing with a long tail that could be well held by the core, as an extra security, one or several horizontal bonding courses were also used at regular vertical intervals. The material used in these courses was large tiles. These reached further back into the core than the facing bricks or tiles themselves, and helped to key in the facing more securely. Bonding courses, a minor but nonetheless significant change in defensive architecture, also served as a means of levelling a wall during its construction. The wall itself was constructed in short segments, measuring
An exterior view of Porta Ostiensis East, looking north-north-west in Piazzale Ostiense. Like Porta Appia, this was a Type I gateway and thus served a main axial road into Rome. (Author's collection)
4.5 to 6m in length, 1.3 to 1.8m in height and extending right through the thickness (Richmond 1930: 60). The absence of putlog holes implies the builders worked from the wall top as construction proceeded, or perhaps from free-standing scaffolding. All things being equal, the resulting structure was tough and durable, capable of withstanding the ravages of weather and the shock of earthquake. Yet the quality of workmanship varied considerably. It is interesting to note that Vitruvius (2.8.7), a military engineer under Caesar and later Octavian (the future Augustus), complains that in his day builders, eager for speedy results, attended only to the facing and botched the core. Such a common human weakness was still apparent in Aurelian's day. In places, great care was taken to pack in the concrete tightly behind the facing. For much of the circuit, however, the haste of the construction and the inexperience of the builders are evident. In places, for instance, the bricks and tiles were of insufficient depth to permit proper bonding to the core, allowing the facing to sheer off over time.
Human resources As in times of the Principate, the most obvious labour force for this type of project would have been the army. Aurelian, however, could not spare the manpower. According to the Byzantine writer John Malalas (12.30), active in the late 6th century, Aurelian therefore drafted the city guilds (collegia) to carry out the actual building work, perhaps under the supervision of a small cadre of military personnel. The use of the collegia as conscript labour was an innovation imposed on Aurelian by the circumstances of the time, but in the next century it would become increasingly common. In return for this undertaking, the collegia were granted the right to bear the name Aureliani in their official titles.
The simplicity of the overall design and the high level of standardization imposed on almost every aspect of construction were necessary to workmen who lacked the expertise and discipline of military engineering. Not surprisingly this simplicity and uniformity also helped to save time and expense, even if, like Rome, the wall was not built in a day. Certainly, the project as a whole occupied the rest of Aurelian's reign, and indeed remained unfinished at his death in the autumn of AD 275. Malalas (12.30) states that Aurelian's wall was finished in a very short time, and implies that this happened within the emperor's reign, but Zosimus says (1.49.2) it was finally completed under Probus, who, after all, was a man very much in Aurelian's own mould. Probably the bulk of the project was completed under Aurelian but the whole not actually finished until the reign of Probus, a period of six years from conception to completion.
Exterior view of Porta Latina, looking west in Via Latina. This Type II gateway was altered under Honorius, whereby the single-span archway was reduced in width. The right-hand tower is original, but that on the left is Belisarian in date. (Author's collection)
Curtains The foundations were laid in a 4m-wide trench, and of varying depth, revetted by wooden shuttering, which was in many sections left in situ as the concrete hardened. To accommodate undulating terrain, the footings were sometimes stepped, faced with tiles or blocks of tufa and left exposed above ground. In this way the top of the foundation was maintained at a fairly uniform level.
OPPOSITE PAGE Interior view of Porta Tiburtina, looking south-east in Piazza di San Lorenzo. This Type II gateway began life as a monumental arch, seen here, erected under Augustus (r. 27 Be-AD 14) to carry the Aqua lulia-Tepula-Marcia over Via Tiburtina. (Author's collection)
31
Incorporating the monumental arches carrying the Aqua Claudia-Anio Novus over viae Praenestina and Labicana, Porta Praenestina-Labicana was effectively a double-Type II gateway. This interior view shows the architrave inscription celebrating its imperial sponsor, Claudius I. (Author's collection)
Above this solid base the curtain-wall of brick- or tile-faced concrete stood 6.1m high and 3.65m thick, which in turn supported a rampart-walk made of fine concrete with a string-course of tiles on the outer face of the wall. The rampart-walk was protected by wide-set and somewhat irregular battlements, which raised the total height of the superstructure from the outside to just short of 8m (Watson 2004: 147). On certain sections of the circuit the structure of the curtains are of a quite different type. Either side of Porta Asinaria in the south, and east of Porta Pinciana in the north, the curtain-wall is solid only to a height of about 3m, upon which base was constructed a low, barrel-vaulted gallery supporting the rampart-walk and the battlements at the standard height. The gallery was equipped with loopholes for archers. It is not clear why this is so, and the difference may represent nothing more than the work of different labourers, perhaps even military personnel. Still, as we shall see, this was a striking anticipation, albeit on a much smaller scale, of the later Maxentian curtains. For economic and strategic reasons many pre-existing buildings and older structures were incorporated within the fabric of Aurelian's wall. Of these, the most outstanding are: the retaining walls of the Horti Aciliorum and Horti Sallustiani in the north; the curtains and towers of the Castra Praetoria, which had to be raised (the camp itself still retained its military function); the side of a tenement block near the north-east angle of the Castra Praetoria, embedded in the wall fabric with its windows filled in; a short stretch of the Aqua Claudia-Anio Novus on either side of Porta Praenestina-Labicana; the early 3rd-century Amphitheatrum Castrense, which still functioned but had its southern arcades bricked up; and several tombs, most notably that of M. Virgilius Eurysaces and the Pyramid of Caius Cestius. As already noted, the Mausoleum of Hadrian may also be counted in this list. In total, approximately one-tenth of the entire circuit was accounted for in this way (Todd 1978: 28). Of those parts that were truly Aurelianic, the most distinctive deviation from the blue print was to be seen in the riverine curtains. Confident that the river afforded sufficient security, the circuit had been erected here almost
32
entirely devoid of towers, at any rate for long stretches. Yet according to the experienced military judgement of Procopius (in all probability this was actually the opinion of his beloved commander Belisarius), the riverside sector 'was especially vulnerable' (Wars 6.9.16).
Towers For the greater part of its length, the circuit was studded at regular intervals of about every 30m by a system of 381 towers. With very few exceptions the Aurelianic towers were simple, in keeping with the economic policy adopted by the planners. They were uniformly rectangular in shape, measuring 7.6m across, projecting 3.35m in front of the curtains and flushed with the back, and rising some 4.5m above the rampart-walk. In most cases the towers consisted of a solid mass of brick- or tile-faced concrete to the level of the rampart-walk, from where there was access to a triple barrel-vaulted chamber with a central stairway leading up to a crenellated open terrace (Watson 2004: 147). The chambers were usually equipped with two round-headed windows facing forward for the use of ballistae, with another such window on either side to allow these machines to swivel 90 degrees. It is important to note that the standard ballista of this period was a twin-armed torsion engine, which shot bolts (iaculi) and not stones (Marsden 1969: 188-89). These projectiles normally had an iron head, pyramidal in cross-section, which was fixed to a wooden body, mainly ash, equipped with three maple wood vanes or flights. On the galleried sections of the wall, the towers were differently planned, though their external aspect remained largely unaltered. These towers also projected 3.5m in front of the curtain-wall, but were only solid up to the height of the gallery floor, that is, to about 3m. From the gallery, which passed through the rear of the tower, a staircase ran round the other three sides and gave access to an upper-level artillery chamber and the rampart-walk. The chamber was of the same dimensions as the majority of other towers, but its internal arrangement was cramped by the staircase (Todd 1978: 32-33). In neither type of tower was there any means of access from the ground to the upper chamber or the rampart-walk. Access to the upper works could only be gained by stairs associated with gateways. With access to the wall solely confined to gate-towers, the defenders could pass along it only by the rampart-walk. The principal aim in this was to control the unwanted interference of civilians, whose presence might impede the defenders during an emergency. Gateways Aurelian's wall originally had 18 gateways, of which nine have survived, though it was actually pierced by as many as 29 entrances if the numerous posterns (small side gates) are taken into account. As the builders were clearly working to a carefully predetermined plan, these openings may be divided into four distinct types. First there were four great gateways (Type I: the Flaminia, Appia, Ostiensis East and Portuensis gates), each originally equipped with a double-span archway set in a two-storied curtain-wall faced with travertine and flanked by roundfronted towers. These gateways served the four main axial roads leading into the city: the Via Flaminia in from the north, the Via Appia from the south and the two main roads either side of the Tiber that led to the two ports of Rome, the Via Ostiensis on the east bank and the Via Portuensis on the west bank.
33
LEFT
An exterior view of Porta Ardeatina, looking north-west in Via Ardeatina. Like all Type III gateways, Porta Ardeatina was simply a glorified postern. (Author's collection)
RIGHT An exterior view of the blocked postern opposite Via dei Marsi, looking south-west in Via di Porta Labicana. Observe the double relieving arches above the travertine door-head. (Author's collection)
The second category (Type II: the Salaria, Nomentana, Tiburtina and Latina gates) consisted of a single-span archway, again set in a two-storied curtain-wall, but without the travertine facing, and flanked by round-fronted towers. These gateways served roads of secondary importance, the viae Salaria, Nomentana, Tiburtina and Latina. Of interest is Porta Tiburtina on the east of the city. This started life as a monumental arch of the Augustan period erected to carry the Aqua Iulia-Tepula-Marcia over Via Tiburtina, which was then incorporated into Aurelian's wall. Porta Praenestina-Labicana, which incorporated and transformed the monumental arches that Claudius I had erected to carry the Aqua Claudia-Anio Novus over the viae Praenestina and Labicana, was a special case. The two roads made for separate arches in the aqueduct, the funerary monument of M. Virgilius Eurysaces, pistor et redemptor ('baker and public contractor'), lying between them. It was, therefore, effectively two gateways of Type II juxtaposed
EI
U1
sharing a central, round-fronted tower that incorporated the massive tomb, minus its front. The latter was constructed like a bread store with a frieze illustrating the work in the occupant's bakeries. The gateway at the head of the Pons Aelius was probably also of Type II, though almost nothing is known about it. Porta Aurelia-San Pancrazio, spanning the Via Aurelia Vetus on the crest of the Janiculum hill, may also belong to this category, though its demolition in 1643 to make way for the new fortifications of Pope Urbanus VIII has left insufficient evidence. The third category (Type III: the Pinciana, Chiusa, Asinaria, Metrobia, Ardeatina and Septimiana gates) consisted of a single-span archway in the curtain-wall between two ordinary rectangular wall-towers at the usual interval. These gateways gave access to roads only used by local traffic. Devoid of flanking towers these gateways were originally scarcely more than posterns, though, as we shall see, several of them received more serious treatment in subsequent phases. Porta Pinciana on the north side of the city was an unusual example of a Type III. Offset in order to accommodate the oblique angle of the road passing through it, the Via Pinciana, the entranceway was apparently guarded by a single, rather narrow, round-fronted tower on the east side. The fourth and final category (Type IV) encompasses the large number of anonymous posterns and doorways in Aurelian's wall, some of which probably served private needs. In addition to the small portal sometimes referred to as Porta Ostiensis West, four original posterns and two wickets are known in the wall. They were all similarly constructed, their narrow openings being surmounted by flat lintels and in two cases by flat arches, of travertine blocks. Above these door-heads were double relieving arches. Most, if not all, were blocked up at a very early date, probably under Maxentius (Richmond 1930: 219-21, 229-35, 247). Posterns, no fewer than five in number, also pierced the stretch of ri'ver wall from the Pons Agrippae up towards Porta Flaminia. These served the key ferry crossings and landing quays. Due to their commercial importance, these posterns remained open much longer than their landward counterparts.
invaders long enough for a relief force to be sent to the city's defence. These barbarians had no great expertise in siege warfare, an understandable failing among peoples who had no fortified cities of their own. What is more, they were slow to learn the techniques necessary for the building of siege machines, and even slower to develop the cohesion and deliberation needed to conduct a protracted siege (Thompson 1965: 131-40). The walls of Rome, therefore, were not designed to withstand concerted attack from an army equipped with sophisticated engines of war. The large number of entrances clearly demonstrates the truth of this. The point is underlined by the fact that so many of the posterns were closed and the remaining gateways strengthened when the military circumstances altered to increase the likelihood of siege warfare. Certain tactical flaws in the design and construction of Aurelian's wall, which once again point to the lack of experience in this kind of construction on the part of the civilian workers involved, indicate that the function of the wall was as much a psychological deterrent as a physical barrier. These flaws are most obvious at those places where pre-existing structures have been incorporated. A glaring example is the total lack of communication along certain sections of the Aqua Claudia-Anio Novus around Porta Praenestina-Labicana and at the tenement house just to the north of this. The junction of the circuit with the north-west angle of the Castra Praetoria and the re-entrant east of Porta Ostiensis East also created unnecessary weak points (Richmond 1930: 67, 242-45, 248). However, despite these apparent functional faults, the design of Aurelian's wall was clearly made with artillery defence in mind. The provision of windows in the gate- and wall-towers for the use of ballistae was a relatively innovative idea. This design further reinforces the anticipated nature of the attack the defences were intended to withstand. The artillery system had a limited range in the area directly in front of the wall itself, thus providing effective deterrence rather than meaningful defence. Nor was it possible to defend much more than a single sector of the circuit at anyone time in this fashion.
The junction of Aurelian's wall with the north-west angle of (astra Praetoria, looking west-south-west in Viale del Policlinico towards Porta Nomentana. This was one of the weak points in the enceinte. (Author's collection)
37
Ii
38
To defend the city from all sides, as would be required in a formal siege, by arming every artillery emplacement would have required a complement of well over 700 ballistae in working order, together with the experienced military personnel to man them. It is highly unlikely that appreciable contingents of ballistarii were ever permanently stationed in Rome (Todd 1978: 34). Indeed, when faced with the serious prospect of a siege Honorius reverted instead to a primary reliance on archers, using artillery only as a reserve. Aurelian's wall thus represents 'a formidable barrier not a fighting platform' (Richmond 1930: 67).
After Aurelian
The threat of barbarian invasion that prompted Aurelian to build his wall around Rome did not materialize. A measure of stability was created by perhaps the greatest of the Illyrian soldier-emperors, Diocletianus (r. AD 284-305), who gradually developed a system of power sharing known as the Tetrarchy. In its evolved form, there were two senior emperors, each known as Augustus, who ruled the eastern and western provinces respectively, assisted by a junior colleague or Caesar. In Diocletianus's view, and who are we to argue, the empire had grown too large and too complex to be governed by one man. This idea was nothing new. Valerianus had divided the provinces on a geographical basis between himself and his son Gallienus, to whom he allotted the west. Yet Diocletianus's tetrarchic system went further. For not only was it designed to provide enough commanders to deal with several crises simultaneously, but by nominating the Caesares as successors to their senior colleagues it served to prevent civil war by providing for the ambitions of all men with armies. In time the two Caesares would became the new Augusti and appoint two new Caesares of their own. The Senate played no role in either the selection of emperors or the governance of the empire.
A marble bust (Istanbul, Arkeoloji Muzesi) of Diocletianus (r. AD 284-305) from Nicomedia (lzmit). His military coup would mark the start of a new phase in Roman history. (Author's collection)
39
By AD 293 the system was up and running. Diocletianus and his co-emperor Maximianus, another Illyrian soldier, were the highest-ranking executives. Diocletianus reigned in the east, with Galerius as his Caesar; Maximianus controlled the west, with Constantius Chlorus as Caesar. The arrangements were sealed by dynastic marriages and the adoption of Diocletianus's family name Valerius, and widely advertised on coins and in official panegyric. Naturally, all four claimed the right to be worshipped as gods. Then, without precedent, but probably with that natural human longing for the peace of retirement, Diocletianus abdicated his position (1 May AD 305). He felt that he had been in power long enough and there were ample safeguards to keep his new form of government in place. The restless old Maximianus, however, surrendered the purple only with extreme reluctance. The walls of Rome now played an increasingly important military role, albeit against internal as opposed to external foes. In the successive refurbishments and strengthening that Aurelian's wall received, its function was altered to meet the new military climate, turning Rome into a fortress.
RIGHT A marble bust (Paris, Musee du Louvre, MA3522 bis) of Maxentius (r. AD 306-12) from Langres. Self-styled conservator urbis suae, Maxentius carried out an assertive building programme, involving the curtains, towers and gateways, in a short space of time. (Esther Carre)
RIGHT Pons Mulvius (Milvian Bridge), looking upstream from the west bank of the Tiber. This was the site of the decisive clash (28 October AD 312) between the western rivals, Constantinus and Maxentius. (Author's collection)
40
LEFT The Maxentian galleried wall just south of Port Latina. Brick-faced concrete construction was ideally suited for the building of galleries and, moreover, did not require highly skilled labour. (Author's collection)
RIGHT The interior of the Maxentian galleried wall, immediately south of Porta Metrobia. To the left can be made out a concave parapet housing an arrow-slit. (Author's collection)
Constantinus's army, although outnumbered, was battle-hardened and confident. Maxentius's army was thrown back in confusion and, as it retreated across the Tiber, the pontoon bridge collapsed. Maxentius and the Praetorian Guard were drowned in the swollen river, a scene flamboyantly depicted on the Arch of Constantine, the triumphal arch erected near the Colosseum in AD 315 to commemorate Constantinus's victory over his rival 'by the inspiration of the divinity' (instinctu divinitatis ). The anxious Senate subserviently welcomed Constantinus as liberator of Rome and passively proclaimed him the sole emperor of the west. The Milvian Bridge would become one of the most famous battles in Roman history, mainly because of Constantinus's pronouncement afterwards that he owed his victory to the God of the Christians. There are three versions of this remarkable story. It was some weeks prior to the battle, so he told his panegyrist Eusebius (Vita Constantini 1.28-29), bishop of Caesarea Palestinae, that he saw a sign
An exterior view of Porta Asinaria, looking north-north-west in Via Sannio, showing Maxentian rebuilding. Richmond suggests this gateway was distinguished in this monumental fashion because it lay on the road leading to the new Circus Maxentius. (Author's collection)
41
in the sky at midday, a cross of light superimposed on the sun. He took this as a sign of victory from the god whose symbol was the cross, stating under oath that he saw the words 'By this sign you will be victorious' (Hac signa victor eris) written in stars around the cross. The night before the battle Christ appeared to him and instructed him to put this heavenly symbol on the standards of his army. He apparently did so with the desired results. Yet in a much earlier version of the same story (Historia Ecclesia 9.9), Eusebius does not mention any such vision, and is content to liken the victory, and in particular the engulfment of Maxentius's pagan troops in the Tiber, to the fate of Pharaoh's chariots at the crossing of the Red Sea. According to the more credible version, that offered by Lactantius (De mortibus persecutorum 44.5-6), the night before the battle Constantinus dreamt that he was ordered to put the sign of Christ on the soldiers' shields. On awaking the next morning he put his faith to the test when he ordered his men to paint the chi-rho monogram (XP), the Greek initials of Christ superimposed, on their shields. Whatever it was that happened to him before the battle, there is no doubt that Constantinus showed conspicuous favour to the Christians, then a vocal if small sect amongst many others, and continued to wear the symbol for Christ against every hostile power he faced. What is more, 'the New Rome which is the city of Constantinus', namely Constantinople, was formally dedicated by the emperor to the Holy Trinity and to the Mother of God (11 May AD 330).
LEFT
The Maxentian galleried wall along Viale Carlo Felice. Piercing the gallery curtain at irregular intervals are arrowslits set in concave parapets. (Author's collection)
RIGHT A tower in Viale Castrense, an exterior view looking west. Besides blocking up the round-headed windows, the Maxentian builders left this tower pretty much in its original, Aurelianic state. (Author's collection)
42
The walls of Rome, AD 312, showing the main roads into the city, aqueducts and key sites
- - - Aurelian Wall - - - Servian Wall - - Road - - - - Aqueduct Postern
1 mile
I I
1km
In the front face of the gallery, arrow-slits occurred at irregular intervals. The total height of the curtains from footings to merlons was now more than 15m, and in some places nearly 20m (Todd 1978: 49). The construction process of the new work was broadly similar to that of the original structure, though the bricks and tiles used in the facing were not so scrupulously chosen and the bonding courses were omitted. This minor decline from the original higher standards is probably to be attributed to haste rather than to mere negligence or shoddy workmanship. Nevertheless, the overall standard of the work is still consistently high for so massive an undertaking. As in the wall of the first period, the Maxentian work was not entirely uniform in construction throughout its entire length. One variant form existed in the south-west corner of the circuit, where the wall ran in front of the Aqua Claudia-Anio Novus. The Maxentian wall reached the level of the aqueduct and to support the new structure the Aurelianic curtains were extended to their rear and buttresses placed against the back of the heightened wall. The towers in this sector had been sited between the piers of the
43
aqueduct so that access to the Maxentian wall could be had from these without the users climbing over or walking on the aqueduct channels. This unusually elaborate arrangement was thus a painstaking device to ensure that the flow of water in these two conduits was in no way interfered with. Now that the wall had been raised considerably, the city possessed defences that only the most determined and prolonged siege could hope to penetrate. The galleried curtains meant that men and machines could be rushed to whatever sector required them. Furthermore, the enormous height of the upper rampart-walk gave not only improved vision to the defenders, but it also enabled a smaller number of them to engage and hold down an attacking group, thereby allowing greater flexibility of movement among those manning the wall. It is difficult to see how any ancient army could invest and take by siege operations an enceinte of this scale on a 19km circuit, and, as we shall see, the 'later history of Rome was to prove the truth of this and the worth of the wall of Aurelian and Maxentius' (Todd 1978: 50).
The heightening of the curtains obviously brought changes to the towers, since they now rose to the level of the tower tops. But in the case of many of the towers, the resultant modifications were inconsiderable. In a number of surviving examples it is clear that the Maxentian workers left the Aurelianic towers much as they were, so that their crenellated open terraces projected forward of the wall at the same level as the new crenellated parapet. Others, however, were substantially enlarged to take account of the new associated curtain-wall at the rear. In many, a large chamber covered by a hipped roof was built over the older open terrace, the parapet and merlons having been removed and the four windows below them blocked. These new upper chambers normally contained three round-headed windows in front, one to each side and two at the back, with a door giving access to the rampart-walk. In a small number of cases a more radical rebuilding was carried out. The old upper works were taken down to the level of the former
LEFT A tower in Corso d'italia, an exterior view looking south-east. Here the Maxentian builders have raised the tower one storey, and added new round-headed windows and a hipped roof. (Author's collection)
RIGHT An exterior view of the east gate-tower of Porta Asinaria, looking north in Via Sannio, showing a rectangular Aurelianic tower and a Maxentian round-fronted addition. (Author's collection)
44
open terrace. Over this was then erected a new storey, covered like the other heightened towers with a hipped roof. There are no discernible tactical or topographical reasons as to why the Maxentian workers gave such different treatments to the towers. As with the construction of the galleried wall, it may simply be a sign of haste. As Todd (1978: 52) points out, this ambitious scheme, begun probably in AD 307 when Rome was under threat, was still incomplete in AD 312 when Constantinus arrived from the north.
Exterior view of Porta Pinciana, looking south in Piazza Brasile. The round-fronted tower on the right was built onto this Type III gateway under Maxentius. (Author's collection)
45
sJaMol
TOWERS
2. In the first Maxentian example reconstructed here, the original parapet and merlons were removed and the four windows below them blocked. The builders then finished the modifications by simply raising the tower and adding a new crenellated open terrace. 3. In this more elaborate, second Maxentian example, a large chamber covered by a hipped roof has been built over the older open terrace, the parapet and merlons having been removed and the four windows below them blocked. The new upper chamber contains three round-headed windows in front, one to each side and two at the back, with a doorway allowing access to the new rampart-walk.
1. Aurelianic towers. Projecting towers, which now dominated enceintes, provided enfilading fire and advance positions for light artillery. The ballista was a two-armed torsion engine that fired bolts. With a range of some 400m, if used carefully, it could keep an enemy from coming in close to the defences. The towers themselves were uniformly rectangular in shape, measuring 7.6m across, projecting 3.35m in front of the curtain-wall and flushed with the back, and rising some 4.5m above the rampart-walk. This particular Aurelianic example has a triple, barrel-vaulted chamber with a central stairway leading up to its crenellated open terrace. The chamber is equipped with two round-headed windows facing forward for the use of its two ballistae, with another such window on either side to allow the machines to swivel 90 degrees.
but events decided otherwise, thus forcing the builders to concentrate their efforts on the curtains and towers until Constantinus's victory brought Maxentius's programme (and his life) to a halt.
47
LEFT The north face of the marble plinth on the Obelisk of Karnak, Istanbul. It shows Theodosius I (r. AD 379-95) with his two (ineffectual) sons, Arcadius (left) and Honorius (right). The latter two would inherit, respectively, the eastern and western halves of the empire. (Author's collection)
RIGHT West face of the marble plinth on the Obelisk of Karnak, Istanbul. Theodosius, with his two sons and Valentinianus II, the 'western emperor', are shown receiving tribute from Gothic envoys. (Author's collection)
In the face of continued intransigence Alaric marched on Rome for a third time and broke in at Porta Salaria under the cover of darkness (24 August AD 410). The Goths plundered the city for three days but did comparatively little collateral damage. Alaric once more withdrew, this time heading south. He planned to cross to Africa, but died before this. Honorius was momentarily aghast. The oddly domestic lifestyle he led in the remote but easily defensible Ravenna is illustrated, according to a bizarre anecdote of Procopius (Wars 3.2.25-26), by his reactions on being told Rome had perished. The emperor exclaimed in perplexity that it had just taken food from his hands. Matters were quickly explained to him and he sighed with relief: his enormous pet chicken, which was named 'Rome', was still in the very best of health.
48
towers of rectangular plan and built of re-used stone. The combined effect of the Honorian towers and the stone curtains was not only to strengthen but also to enhance the external appearance of these major gateways. Other gateways were also included in this work of blocking. Porta Tiburtina, originally a monumental arch carrying the Aqua Iulia-Tepula-Marcia over Via Tiburtina, had its entranceway reduced to a similar width of 4m, while Porta Asinaria was blocked altogether by brickwork, never to be reopened. Porta Chiusa, which had been narrowed by Maxentius to only 3.6m, was now also entirely blocked and never again re-used. All the surviving single-portal gateways, and several of those now vanished, were provided with curtains in travertine.
LEFT
Under Honorius the doubleportalled Porta Ostiensis East was reduced to a single portal. This was effected by the insertion of a one-storey curtain of travertine provided, as made manifest by the slot shown here/ with a portcullis. (Author's collection)
RIGHT
An exterior view of Porta Tiburtina/ looking south-east in Via Tiburtina Antica/ showing the Honorian alterations. Its portal was reduced in width by the insertion of a wellconstructed travertine gatehouse. (Author's collection)
49
Ll)
II
The walls of Rome represent at once both the most emblematic and the most enduring of Aurelianic monuments. Indeed nothing else so eloquently demonstrates that, by Aurelian's day, the empire was now on the back foot. The wall itself was a massive obstacle, comprising 19km of brick-faced concrete nearly 3.7m thick and 8m high (rising to more than 15m after the reorganization of Maxentius), with 381 enfilading towers and 18 gateways. This scene, set during the reign of Honorius (AD 395-423), illustrates a stretch of the wall from the Tiber to Porta Ostiensis East. The gateway itself has been modified as part of the
Some of the gateways were afforded somewhat different treatment. At Porta Ostiensis East the new-style facing was applied but to the old round-fronted towers. An extra storey was also added here. Two gateways, the Salaria and Nomentana, were not given new facings in stone at all, thought the first was heightened by an extra storey, as was Porta Pinciana. The ceremonial character of the Honorian treatment of the gateways was further emphasized, in some cases at least, by the handsome inscriptions on their new stone-faced curtains. As already mentioned, three instances are known, of which one (elL 6.1190), on Porta Tiburtina, remains in situ, and another (elL 6.1189), from Porta Praenestina-Labicana, survives in part and has been reconstructed at the site of the gateway (Todd 1978: 61,63-64).
An ivory diptych (Paris, Musee du Louvre OA9063), c. AD 532, known as the 'Barberini Ivory'. The horseman probably represents a triumphant lustinianus I (r. AD 527-65), whose grand ambitions included the recovery of Italy. (Esther Carre)
51
A late 5th-century mosaic (Ravenna, Palace of Theodoric, Portico A frs. 4-6), depicting an Ostrogothic horseman hunting a wild boar. In battle light spears were the primary Ostrogoth weapon, with iron swords, of the long-bladed double-edged Sassanid type, serving as a secondary weapon. (Author's collection)
Having conquered Vandal Africa in a lightning campaign, Belisarius, with just 7,500 regulars and his indispensable bucellarii, took Ostrogothic Sicily (AD 535) then Naples by siege and Rome by negotiation (AD 536). Successfully defending Rome for a year against Vitigis (r. AD 536-40), Belisarius next fought his way up the peninsula and occupied the former imperial capital of Ravenna (AD 540). The rapid conquest, however, was only superficial. The weak and divided Ostrogothic leadership had contributed much to Belisarius's spectacular success, and with the emergence of an energetic and able leader in Totila (r. AD 54152) the military balance swung back in favour of the Ostrogoths. From the main Ostrogothic settlements north of the Po, Totila quickly re-conquered Italy and Sicily, except for a few coastal strongholds. Moreover, Persian attacks on the empire meant the imperial army in Italy was starved of men and materiel. Worse still, the army lost Belisarius, who was recalled to command on the eastern front. Prior to the arrival of Belisarius, the alterations to Aurelian's wall were matters of repair and refurbishing rather than major reconstruction. After the earthquake of AD 442 large cracks appeared in the southern sector of the wall between the Appia and Metrobia gates. Porta Appia itself was affected by the earthquake and required large-scale repairs. Certain parts of the city walls were now given buttresses, notably to the east of Porta Latina and of Porta Appia. Against the threat of the returning Ostrogoths Belisarius reconstructed those parts of the city walls that had suffered damage or decay, equipping each merIon of the battlements with a spur wall so as to cover the exposed left side of defenders from missile fire, a defensive device not hitherto used on Aurelian's wall. Drafting local workers, he also dug a ditch or ditches around the city, of which no trace now survives. In front of some of the gateways large man-traps were set. Known as wolves (lupi), these were a form of spiked drawbridge that were designed to be dropped on assaulting troops. As the aqueducts of the city could also provide a means of entry for the enemy, Belisarius had them sealed off by filling their channels with masonry for a considerable distance. All this is told by Procopius (Wars 5.14.15, 19.18, 21.19-22). He was present during the Ostrogothic siege of Rome, being a civil servant who served in a logistical capacity on the staff of Belisarius. Procopius, a civilian who obviously had an eye for military affairs, also tells us (Wars 5.21.14, 18) that bolt-shooting ballistae were installed in the towers and that stone-throwing onagri were mounted on the curtains. A bolt-shooter fired large body-piercing bolts (iaculi), whereas a stone-thrower simply relied on the weight of its projectile to crush the target. Both machines were an important factor in the successful defence of Rome, playing havoc with the Ostrogothic machinery, which was in any case held at some distance from the wall by the new ditch-system, and serving as effective anti-personnel weapons. Procopius says (Wars 5.23.9-11) that a lone Ostrogothic archer was shot by a bolt from an engine mounted on a tower, the missile passing through his cuirass and body, pinning him firmly to the tree he was standing next to. After Vitigis withdrew, further repairs took place before Rome once again came under threat. An Ostrogothic force under Totila was by treachery allowed to break into the city at Porta Asinaria (17 December AD 545). The king,
52
recognIzIng the immense strength of the wall, set about demolishing large stretches of it, but did not proceed far with the work before Belisarius, who was now back in Italy, recovered Rome in the following spring. The damage caused by Totila's men was restored, and some of these repairs can still be seen in the facing of the curtains, where re-used blocks of travertine and marble have been thrust into the gashed fabric. At the towers, a botching job was carried out, the stone blocks being simply pushed up against the earlier facing to form a rudimentary buttress. Nevertheless, when Totila again invested Rome (summer AD 548), the battered old city held out surprisingly well.
AURELIAN'S LEGACY
The cities of Ostrogothic Italy, in general, had suffered devastation from siege and counter-siege in Iustinianus's opportunistic war of re-conquest (AD 536-54), and the Rome that Procopius had once visited had likewise suffered grievously from the interminable hostilities. Paradoxically, the emperor's attempt to bring back Roman rule to Rome did more damage than the barbarian visitations the city had endured so far. The heavy reliance on the strategy of blockade meant that some sieges were protracted and Rome, as we have seen, twice underwent lengthy investments of a year's duration in the space of a decade. Behind the city walls the population had already shrunk from around 800,000 in Maxentius's time to perhaps 80,000 under the Ostrogothic kings, most of them concentrated in the west of the city near the bend of the Tiber, from the foot of the Palatine and the Capitol down to the river and on the west bank in the Janiculum quarter. Many of the old senatorial families had already died out. The Gothic war then ruined many of the families that survived, who abandoned their urban villas and rural estates and took refuge in the eastern empire. Around its dwindling population, meanwhile, the physical city decayed. The ancient city prefect (praefectus urbi) still held office under the Ostrogothic kings and after the Gothic war, and under him were officials in charge of building maintenance and restoration as well as dedicated funds for purchasing the necessary bricks and mortar. Procopius found the Romans to be 'lovers of their city' (philopolides) beyond all others and noted that during the previous Ostrogothic regime 'they had mostly preserved the city's buildings and their adornments even though under barbarian rule' (Wars 4.22.5-6). The new construction mostly comprised churches. The Papacy presided over a building programme that left most of the Augustan regions shimmering with spacious new basilicas. The popes themselves bore most of the cost, as they had the necessary means. As the senatorial aristocracy abandoned Rome and central Italy they bequeathed their cherished estates to the see of Rome, which became the greatest landowner in the peninsula and in Sicily as well. With these buildings each pontiff impressed his stamp on Rome, as its secular as well as its spiritual ruler. As such they also frequently managed secular urban construction. By the time of Gregory the Great (AD 590-604), stability and a level of prosperity had returned, and Rome had taken on its medieval dress.
An angle-tower in Via Casilina, in an exterior view looking west. Belisarian restoration work is easily recognized when re-used blocks of travertine and marble have been thrust into the gashed fabric of the city walls. (Author's collection)
53
54
An exterior view of Porta Flaminia, looking south-southeast at the junction of Piazzale Flaminio and Via Flaminio. Its present state, apart from the side arches (1877), represents the gateway erected for Pope Pius IV Medici (1559-65). It is now known as Porta del Popolo. (Author's collection)
The Bastione del Sangallo (1538-42), an exterior view looking west in Viale di Porta Ardeatina. Quadrilateral, angled bastions were designed to resist artillery bombardment rather than human assault; the age of 'vertical defence' had passed. (Author's collection)
A splayed embrasure (above) and a vaulted casemate (below) at the Bastione del Sangallo. Built squat and solid, two of its sides pointed outwards and carried heavy artillery, while the other two stood at right angles to the main wall and bristled with anti-personnel weapons. (Author's collection)
the crest, while the thickness entailed by height rendered attack by siege engines less effective. Counter-weighted machines (tension artillery) threw projectiles that struck only glancing blows at such walls; spring-powered machines (torsion artillery), though working with a flat trajectory, were intrinsically under-powered. Even stone-firing bombards had made little impression upon the art of siege warfare. The only certain means of bringing down a wall was to attack it at its base by mining, a laborious task that ditches and moats readily defeated, and that was also open to the riposte of counter-mining. The new cannon could be brought rapidly into action close to a wall, and then handled to fire accurately in a predictable arc of impact; their advent effectively transferred the effect of mining to combustible artillery. Compact iron cannonballs, directed at the base of a wall in a horizontal pattern of attack that did not vary in height, rapidly cut a channel in the stonework. The cumulative effect was to use the physics of the wall against itself: the higher the wall, the more quickly it would become unstable and the wider the breach it left when it toppled. French powder makers and gun casters had reshaped the slowfiring and very immobile bombard into an efficient prototype of the modern gun. Lighter, more manoeuvrable cannons firing an energetic new form of powder created a destructive weapons system. And so with relative ease Charles's state-of-the-art cannon had knocked down walls that had stood stoutly for many centuries, thereby making good his claim to the Kingdom of Naples. Italy, appalled at the easiness of the trans-Alpine king's triumphal march to Naples, would soon become the new school of not only experienced master masons, but also experts in mathematics and engineering. It was Giuliano da Sangallo, with his brother Antonio, who founded the first and most important of the Italian fortification 'families', an extremely competitive group of Mafia-like bands that were contained by ties of blood, companionship and patronage. These not only included the Sanmicheli, Savorgnano, Peruzzi, Genga and Antonelli, but also such unlikely practitioners as Leonardo da Vinci, who, in spite of his conviction that war was bestialissima pazzia, became inspector of fortresses to Cesare Borgia (1502), and Buonarroti Michelangelo, who, as Commissary General of Fortifications, equipped his native Florence with new defences (1527-29).
56
The Bastion in Viale Aurelio Saffi, erected under Pope Urbanus VIII (1643), equipped with gun-ports and loops. Guns of all calibres could be either fired over the parapet, or concealed and fired through embrasures cut into the parapet. (Author's collection)
Michelangelo, who is now remembered chiefly for his titanic struggles with blocks of marble and the ceiling of the Sistine Chapel, was also renowned in his own day as a military engineer. In 1545, in a course of a heated argument with Antonio da Sangallo minor, the renowned military architect employed by the Farnese family, he gave vent to the astonishing statement that 'I do not know very much about painting and sculpture, but I have gained a great experience of fortifications, and I have already proved that I know more about them than you and the whole tribe of the Sangallos' (Clausse 1901: 2.351). It was Antonio da Sangallo minor who had been hired by the Pope to add 18 powerful bastions to Aurelian's wall in addition to five for the defence of the Vatican (1538-42). Those that were erected were done so along the southern sector of the wall, an area of Rome most vulnerable to attack. The Bastione del Sangallo, near Porta Ardeatina, probably represents the acme of 16th-century Italian military architecture. The cost, however, was astronomical. The scheme to surround Rome with a belt of bastions was abandoned when the construction of this one bastion alone was found to have cost 44,000 ducats (Parker 1996: 12). As high walls were extremely vulnerable to the law of ballistics, new walls to resist the cannon therefore needed to stand low. However, a fortress so built was open to escalade, the rushing forward of a storming-party with ladders to sweep over the crest and into the fortress interior by surprise attack. The new system of fortification had to incorporate features that resisted bombardment and, at the same time, held the enemy's foot soldiers at bay. The solution to this problem of surrendering height while acquiring depth was the solid angular bastion. Strong enough not to be battered shapeless by a concentration of enemy fire, this wallhigh structure stood well forward of the main wall, where it dominated the ditch or moat, and served as a firing platform for gunpowder weapons. The most suitable design proved to have four faces. Two of these formed a wedge that pointed out toward the surrounding countryside so as to present a glancing surface to enemy fire, and where big ordnance could be mounted to fire out across the glacis. The other two faces, those that joined the wedge to the main wall at right angles, from the ramparts of which defenders could use small-calibre firearms, both hand-held and mounted, served to sweep the ditch and stretches of curtain between bastions. The bastions should be built of stone, though brick was an acceptable substitute, backed and filled with
57
58
A bronze equestrian statue of Garibaldi (1807-82), Piazza Gramsci, Siena. It was 'his fortune never to take full part in the common prose life of civilised men', wrote George Macaulay Trevelyan. (Author's collection)
rammed earth to better absorb the shock of shot, the whole constituting a structure of immense solidity so as to provide both a rock-solid cannon platform and a sloping outer face on which impacting shot would make the least possible impression. The German artist Albrecht Durer, having studied in Italy, took the blueprint for this style of gunpowder fortifications north. He published the first treatise on the new defensive system, which spread across Europe under the name trace italienne. Over a period of 50 years the quintessential bastion neutralized many of the advantages of improved cannon and returned siege warfare to a new equilibrium. The trace italienne would develop into the fearful geometry of fortification associated with the most famous of French military engineers, Vauban.
Garibaldi's Rome
In February 1849, some 19 centuries after its demise, the Roman Republic was revived. In France the ambitious new president of the Second Republic, Louis-Napoleon Bonaparte (before long Napoleon III, Emperor of the French), dispatched an army to restore the Pope and 'liberate' Rome from the handful of dangerous radicals who, as he saw it, had forced themselves
A manikin (Marsala, Museo Civico) dressed as a Garibaldino at the time of the defence of Rome. The idiosyncratic shirt had evolved, six years earlier, out of a requisitioned stock of bright red overalls destined for slaughterhouse workers. (Author's collection)
upon the unwilling citizens. On 27 April Garibaldi led his followers into Rome through streets packed with people shouting his name. He entered the city riding a white horse and wearing a black slouch hat and a swirling white poncho, which was flung back to show his celebrated red shirt. Behind him clattered his 'brigand-band' of red-shirted followers, the Garibaldini. The Roman commander was General Avezzana, and of the nearly 20,000 men under his command, the Garibaldini constituted only a small fraction. But Giuseppe Garibaldi (1807-82) is remembered as the defender of the brief Roman Republic. On 29 April, with Avezzana's approval, Garibaldi occupied the Villa Corsini, a private house set in gardens just outside and perched on a hill above the western sector of the city walls. The following day the French marched lackadaisically up to Rome's Janiculum quarter, assured by the reports of an easy entrance into the city. Garibaldi sent his men, seasoned Garibaldini and new Roman recruits alike, racing downhill to repulse them. Initially the French held their ground, but when Garibaldi personally led a second charge, they turned and fled. A French representative negotiated a cease-fire that allowed the French army to remain in situ as a shield against an Austrian (Hapsburg) army poised to the north, or so the Romans were told. Meanwhile, the army of the Bourbon king of Naples was menacing Rome from the south. Garibaldi went to meet it under the command of Colonel Pietro Roselli. There was a desperate, inconclusive engagement at Velletri, where Garibaldi came close to being killed: he and his horse were thrown down and badly trampled by some of his own retreating horsemen. Having been dragged clear of a tangled heap of fallen horses and men, he returned to Rome with his battered band. On 1 June the French general Charles Oudinot, his army now heavy reinforced, gave notice that he was ending the armistice. The Romans, understanding themselves to have three days to prepare, were taken completely by surprise when, on 2 June, the French, determined not to be beaten, occupied the undefended Villa Corsini. Garibaldi was given the task of recapturing it. The battle for Villa Corsini, which took place on 3 June, was a terrible one. For 17 hours, from dawn to dusk on a sweltering hot day, Garibaldi sent wave after wave of men up the rising ground between the city walls and the villa, through its narrow garden gate and up the steeply sloping drive towards the front of the four-storey villa, where from every window, balcony and terrace the French were firing on them. Twice the villa was taken 'at the point of the bayonet', but each time the French, who could approach it under the cover of trees to the rear, swiftly retook it. With the French immovably entrenched in the hilltop villa the fall of Rome was inevitable.
60
For another month the Republic held out. Garibaldi commanded the defence of the most desperately beleaguered section of the city walls. On the night of 29 June the French, having completed their siege lines, launched the final offensive. For two hours Garibaldi valiantly led the defenders as they struggled to hold back the assault. At last, as the western sector collapsed under the French bombardment and the invaders came pouring through the breach, Garibaldi rode over the Tiber to the Capitol where the Assembly was in session. Rome was lost and the great political experiment had failed. But when he walked into the chamber covered in blood, sweat and dust, the Assembly rose as one man and cheered him. The republican government surrendered, but Garibaldi, fated to become Europe's greatest republicans, did not.
The Servian wall, at Stazione Centrale Roma Termini. Despite its incongruous mise en scene at a McDonald's restaurant, the short stretch of wall here can be studied at close-quarters. (Author's collection)
61
French. Further up the river and just beyond the Vatican is the papal fortress of Castel Sant'Angelo, part of which is the mausoleum that Hadrian himself designed, its massive drum being later incorporated into the city's defences. Another location worth visiting is Stazione Roma Centrale Termini. Between the railway station and the Museo Nazionale Romano, in Piazza dei Cinquento, stands the best-preserved section of the so-called Servian wall. Beneath the station itself, surrounded by the chairs and tables of McDonald's, two very short sections of this wall can be closely examined too.
Ballista/ballistae Ballistarii
Bonding courses
Bucellarii
Caesar
Foederati
Header
Bal/ista bolt.
'One-thousand paces' - a Roman mile (1,618 yards/1.48km). 'Wild ass' - a single-armed torsion engine throwing stones. A low narrow defensive wall, usually with crenels (open part) and merlons (closed part), along the upper outer edge of the curtains. Volcanic sand giving strength when mixed in cement. Stone block placed horizontally with its length parallel to the length of a wall (cf. header). Grey-white stone suitable and popular for building both in the Roman period and today. A porous rock formed of calcium carbonate (chalk) deposited from springs.
Abbreviations
elL lLS PBSR SHA TAPA Wars
62
T. Mommsen et aI., Corpus Inscriptionum Latinarum (Berlin, 1862- ) H. Dessau, Inscriptiones Latinae Selectae (Berlin, 1892-1916) Proceedings of the British School at Rome Scriptores Historiae Augustae (London, 1932) Transactions of the American Philological Association Procopius, History of the Wars (London, 1919)
BIBLIOGRAPHY
Barker, J.W. Justinian and the Later Roman Empire (London: University of Wisconsin Press, 1966) de Blois, L. The Policy of the Emperor Gallienus (Leiden: E.J. Brill, 1976) Downey, G. 'Aurelian's victory over Zenobia at Immae, AD 272', TAPA 81: pp. 57-68 (1950) Elton, H. Warfare in Roman Europe, AD 350-425 (Oxford: Clarendon Press, 1996, 1997) Heather, P.J. The Goths (Oxford: Clarendon Press, 1996) Hughes-Hallett, L. Heroes: Saviours, Traitors and Supermen (London: Fourth Estate, chapter 7, 2004) Jones, A.H.M. The Later Roman Empire: a Social, Economic and Administrative Survey, 2 vols (Oxford: Clarendon Press, 1964) Marsden, E.W. Greek and Roman Artillery: Historical Development (Oxford: Clarendon Press, 1969) Marsden, E.W. Greek and Roman Artillery: Technical Treatises (Oxford: Clarendon Press, 1971) Osier, J. 'The emergence of the third century equestrian military commanders' Latomus 36: pp. 674-87 (1977) Parker, G. The Military Revolution: Military Innovation and the Rise of the West, 1500-1800, 2nd edition (Cambridge: Cambridge University Press, 1996) Rankov, N.B. The Praetorian Guard, Elite series no. 50 (Oxford: Osprey Publishing, 1994) Richardson, L. A New Topographical Dictionary of Ancient Rome (Baltimore: Johns Hopkins University Press, 1992) Richmond, LA. 'The relation of the Praetorian Camp to Aurelian's Wall of Rome', PBSR 10: pp. 12-22 (1927) Richmond, LA. The City Walls of Imperial Rome: An Account of its Architectural Development from Aurelian to Narses (Oxford: Clarendon Press, 1930) Rostovtzeff, M.L Social and Economic History of the Roman Empire, 2nd edition (Oxford: Clarendon Press, 1957) Saunders, R.T. A Biography of the Emperor Aurelian, AD 270-275, Ph.D. dissertation (University of Cincinnati, 1991) Southern, P. The Roman Empire from Severus to Constantine (London: Routledge, 2001) Southern, P. and K.R. Dixon The Late Roman Army (London: Routledge, 1996, 2000) Stoneman, R. Palmyra and its Empire: Zenobia's Revolt against Rome (Ann Arbor: University of Michigan Press, 1992) Thompson, E.A. The Early Germans (Oxford: Clarendon Press, 1965) Todd, M. The Walls of Rome (London: Elek Books, 1978) Todd, M. The Northern Barbarians (Oxford: Clarendon Press, 1987) Tomlin, R.S.O. 'The late Roman Empire', in General Sir John Hackett (ed.) Warfare in the Ancient World, pp. 222-49 (London: Guild Publishing, 1989) Trevelyan, G.M. Garibaldi's Defence of the Roman Republic (London: Longmans, Green, & Co., 1907) Ward-Perkins, J.B. From Classical Antiquity to the Middle Ages: Urban Public Building in Northern and Central Italy, AD 300-850 (Oxford: Clarendon Press, 1984) Watson, A. Aurelian and the Third Century (London: Routledge, 1999,2004) Williams, S. Diocletian and the Roman Recovery (London: Routledge, 1985, 2000) Wolfram, H. History of the Goths (Berkeley: University of California Press, 1988)
63
Design, technology and history of key fortresses, strategic positions and defensive systems
Photographs
Unrivalled detail
_ Colour maps
OSPREY
PUBLISHING
5 1 895
781846 031984