Boudicca is not just a historical figure – in modern Britain she is an icon, a term sometimes defined as cultural short-hand for a focus of stories dealing in this case with various aspects of British self-definition, whether in the form of statues at Westminster or Cardiff, films and books of various qualities, or a quotation in rock videos or as labels for pubs and ale. She is to some a feminist role model, for others a freedom fighter, or a fighter against governmental injustice.
Most of the stories told about her today have a positive slant, which is surprising, because what they celebrate is after all the perpetrator of an all out massacre of two, perhaps three, towns full of civilians, most of whom had little to do with the injustices she and her tribe the Iceni felt they had suffered. Usually the wholesale slaughter of innocent bystanders causes revulsion in witnesses (even if at third and fourth hand); here it has caused celebrations as a hero. This is not the place for a moralistic essay on the rights or wrongs of this approach, but this chapter is going to look at the evidence for Boudicca’s Rebellion, where this heroic celebration comes from, and compare it with the archaeological evidence for the Iceni in Norfolk.
As we have seen the form of administrative integration of a tribe into the Roman Empire could vary, and in addition to direct control, Aulus Plautius and his successors set up at least three, but possibly more client kingdoms. The term ‘client kingdom’ has sometimes been equated with the modern concept of a satellite state, but the relationship with Rome was more complex. Client status is first and foremost a treaty agreement between the Roman state and a particular foreign ruler, who is called by the Romans ‘rex et amicus SPQR’, a king and friend of the Senate and People of Rome. Technically it is a treaty of mutual support: Rome supports the king’s internal claim to power, while the king aids Roman interests in his region. Both offer mutual defence against common enemies. It is worth stressing though, that in practice these treaties were not equal; Rome expected a substantial level of obedience, but on its part could (and frequently did) renege on the deal whenever it suited its political interest, foreign or domestic. This obedience to Rome could be further insured by the provision of hostages, such as the ‘offer’ of raising some of the client king’s family members in Rome. Client kingdoms are furthermore not strictly independent in a military sense, as the Romans could provide military protection for the king and Roman interest in the form of military bases – one of the best known of these military bases is the legionary fortress of the legio XI Claudia in Charax in the Bosporanian kingdom on the Black Sea, a longstanding client kingdom of Rome in the Crimea.
As the client treaty was with a particular king, rather than the tribe, the situation was fluid and had to be renegotiated with every change of ruler. This could lead to competition by various sons of a ruling king in gaining the support of the current administration in Rome. We assume that the princes that approached Augustus, Caligula and Claudius at various points before 43 were actually trying to gain Rome’s support for becoming the next ‘rex et amicus’ within their own tribe.
In the aftermath of the Roman conquest of Britain, we hear of three client kings, the most famous being Togidubnus/Cogidubnus of Chichester/Fishbourne. His client status is mentioned in Tacitus’ Agricola (chapter 14), which describes him as follows:
Certain civitates were given to King Cogidubnus, who remained most loyal (to Rome) to within our own memory, following the long established custom of the Roman people of using even kings to make others slaves.
Archaeologically his existence is confirmed by the reading and reconstruction of one inscription from Chichester (RIB 91) which calls him REG MAGN BRIT (rex magnus Britannorum – great king of the Britons). His association with Fishbourne palace, while often repeated, is far from secure and more recently he has been associated by some with Silchester; in truth we know very little about him or his actions during his apparently long reign.
Two further client kings are mentioned in the Roman historic sources; one is Queen Cartimandua of the Brigantes, the only bona fide woman ruler in her own right that we know of in Roman Britain. The other is Prasutagus of the Iceni. As we have seen in the last chapter, apart from these three client kingdoms, the rest of the South of England, usually assumed to be the area south of the Fosse Way (which in contrast to older research is no longer seen as a Roman frontier by modern scholars) was administered directly by Rome. In addition to the military, which had come from the Continent with the conquest or possibly as later reinforcements, increasingly large numbers of immigrants arrived from other areas of the Roman Empire. They were not necessarily Roman citizens themselves, and they appear to have come from a wide range of cultures within the Empire; but their interests were deeply bound up with the success of Britain as a Roman province. Substantial numbers of incomers settled in the emerging cities. London was at this point an informal trading centre with excellent infrastructure links including the harbour on the Thames, as well as a safe river crossing. The second largest Roman town was the colonia at Camoludunum/Colchester, which was set up on the site of the first Roman legionary fortress, which in turn occupied the site of a pre-existing oppidum. This was also the site of the temple to Claudius, and thus most likely the meeting place of the provincial assembly. Provincial assemblies have sometimes been described as early forms of provincial parliaments, but this far exaggerates their power. Their main role was as an assembly of the leading men of the province, who were charged with the proper organization and celebration of the Imperial cult within the province, the cost of which was expected to be born by the members of the assembly. In addition, they could send letters and embassies, usually of a congratulatory character, to the Emperor. We know that elsewhere in the Roman world they voted honorary statues for outgoing governors (and one assumes procurators), as well as occasionally sending letters of complaint about departing officers.
Elsewhere in Britain some Iron Age oppida were slowly changing into Roman civitas, capitals with a more Roman character. This process is particularly well understood for Silchester, but can also be seen in operation at Verulamium and Chichester to name but a few. In fact research in Silchester over the last 20 years suggests that it might have rivalled Colchester and London in size. The military units may have changed garrison frequently at this time, but at least the larger forts and fortresses managed to attract civilian settlements around them. On the other hand in many parts of the province rural Iron Age settlement patterns appear to have continued with very little change.
Archaeologically the Iceni are associated with a culturally distinct late pre-Roman and Roman Iron Age group that can be recognized north of the Waveney and Ouse in East Anglia, and which extends westwards into the Fenlands at least as far as Stonea.
While its base metal work, which is dominated by horse harness, is closely associated with what used to be called ‘Celtic tradition’ and showed some links to the South and the Continent, other characteristics set them apart from their neighbours. Unlike the neighbouring Aylesford-Swarling culture to the south (which has frequently been associated with the Trinovantes and Catuvellauni), this group did not use wheel-made pottery and appears to have had markedly different burial customs, as there are few of the cremation burials that have been found further south.
Like their neighbours to the south and west, however, they struck coins, but they never apparently copied the Roman coinage and its patterns in the manner of the Eastern and Southern Dynasties. In the western half of the group’s distribution a series of sanctuaries/areas of ritual deposition have been excavated. Some, such as the one at Fison Way in Thetford, are enclosures with central round houses and shafts for the repeated deposition of carefully selected material. Others like Snettisham appear to be larger areas, where especially gold and silver were deposited in the open.
Within the area, at least four subdivisions can be identified archaeologically with at least four, perhaps five oppida i.e. proto-towns. The most enigmatic is Sedgeford, which appears to cover an area of circa eight square miles including areas of ritual significance and deposition as well as settlement zones. Further south are more traditional enclosed oppida such as Caistor St Edmund, under the later Venta Icenorum, Stonea/Chatteris in the Fenlands, Saham Toney/Ashill and Thetford. The last three are all found in the western part of the area, the so-called Breckland, traditionally the most fertile land, while the eastern districts are even more low-lying and waterlogged over large areas. The fertility of the Breckland might also explain the substantial amount of disposable wealth that can be recognized archaeologically from the Middle Iron Age (c. 200BC) onwards. In addition to the oppida and sanctuaries, there are a number of smaller sites, both defended (such as the forts at Holkham, Warham, South Creake, Narborough and Thetford) and open smaller sites, the majority of which appear to have been abandoned shortly after the Boudiccan rebellion.
It seems that by the time of the Claudian conquest the Iceni had merged into a single tribe with multiple centres (e.g. coins were minted at Thetford, Saham Toney and Needham), of which the wealthiest group and thus presumably also the most powerful lived in the western Breckland.
By the time of the conquest coins were struck with the inscription ANTED and slightly later further ones were marked ECEN, both were originally read as abbreviated royal names. John Davies of Norwich Museum, however, considers the ECEN inscription as an indication of the tribal name rather than an individual, making the Iceni one of the few tribes to advertize themselves in that way.
While it is possible that the Cenimagni mentioned by Caesar during his second campaign were predecessors of the later Iceni, the first time we hear about the Iceni themselves is around 48. Aulus Plautius’ successor Ostorius Scapula suggested disarming some of the British tribes (including the Iceni), who (or at least some of whom) took offence at this, and started a revolt. Ostorius Scapula sent the army in, the rebels withdrew into a fort with only a single narrow access point and eventually the Roman army stormed the fort, and thus ended the rebellion. There is no archaeological evidence for this rebellion to date, and suggestions that this fort may have been Stonea or Holkham are based on general topographical consideration (large size, difficult access), rather than being based on archaeological confirmation of a battle site. Archaeologically we can identify a number of Claudian military installations in Norfolk, which may represent the initial garrison after the conquest or the response to the uprising. Many of these sites are positioned close to Icenian settlements. Unfortunately, as many of these sites are known from aerial archaeology only, it is not always clear which of these forts are Claudian and which represent the later response to the Boudiccan uprising. One presumably Claudian site, Woodcock Hall, appears to be sited next to the Saham Toney oppidum.
According to J. Davies’ chronology (Davies & Robinson 2009, 54f.) at the same time as the first rebellion, silver coins naming Prasutagus began to be struck, which might suggest that the Romans decided that the Iceni may be better administered through a client king, rather than directly. Alternatively Prasutagus may have been chosen at the time of the conquest as the client king. It could then be argued that Prasutagus would have had to face consequences in not keeping his compatriots under better control, but the source (Tacitus’ Annals) is too curt to allow us any insights into the context and background of the uprising, thus either scenario remains a possibility. At no point in the historical or numismatic evidence is Boudicca mentioned in any form, let alone as a co-ruler. She differs thus markedly from her neighbour to the west, Cartimandua of the Brigantes, where it appears that Cartimandua is in charge, while Venutius, at least in Tacitus’ eyes, is very much the consort. Unfortunately, the Brigantes never struck coins thus depriving us of a possible insight into their view on who the Brigantian ruler really was. We also do not have any evidence for the political preferences of Boudicca; Prasutagus is described as a trusted friend of Rome, but it has been pointed out that just as in the case of Venutius and Cartimandua, the two rulers might have represented opposite ends of the political spectrum.
We hear nothing about the Iceni between this uprising and the events of the Boudiccan rebellion either in Tacitus’ accounts or any other literary source. Archaeologically it is possible to point to a number of Roman imports arriving at the Iron Age sites, which demonstrate that there was more than diplomatic contact between the client kingdom and the directly administered part of the province. These included items of considerable value such as the Crownthorpe Hoard, a large collection of copper-alloy vessels, representing a near complete set of vessels needed for a Roman style drinking party (Davies & Robinson 2009, 33).
There are four main accounts of the events in the middle of Nero’s reign and especially the Boudiccan uprising – one by Tacitus in the Agricola, a much more detailed one in the Annals both written about 40–50 years after the events, and one by Suetonius in his ‘Lives of the Caesars’, which dates probably to the second quarter of the second century AD. The last one was written by Cassius Dio in the first half of the third century AD.
For reasons which will become apparent, we will start with Suetonius. We know that Suetonius was the ab epistulis of Hadrian, the man who controlled Hadrian’s Latin correspondence and who was removed from office by Hadrian because Suetonius had behaved to the Empress Sabina ‘with greater familiarity than the etiquette of the court required’ (SHA Hadrian 11.3). Before this he had been ‘a studiis’ and ‘a bibliothecis’, two posts which gave Suetonius access to the official archives and libraries in the city, and thus theoretically unrivalled access to the original records – if he had the time or the inclination in his position to actually use them.
Amongst other works, Suetonius has left us a series of biographies on the first twelve Caesars, i.e. the Emperors from Caesar to Trajan. While the biographies can be examined from many different perspectives, and certainly some elements are doubtful in their veracity (for example exactly how would Suetonius know what Tiberius got up to in his bedroom?), there is an underlying subject matter, which is ‘what makes a good or bad Emperor?’ and one way the twelve biographies can be read is as a series of examples for and against a particular definition of the role of Emperor. To elaborate the issue, the biographies focus on the life of the Emperor himself and his direct actions to reveal his character, less on events that he only indirectly influenced, thus a lot of ‘provincial history’ is excluded from the biographies.
Suetonius focused his biography very much on Nero’s achievements or embarrassments in Rome; his policies with regard to the provinces covers a very short chapter (18): apparently Nero lacked any wish to extend the Empire and even considered withdrawing from Britain.
Suetonius later returns to Britain (chapter 39), when he lists among the disasters that struck during Nero’s reign ‘a disaster in Britain, where two important towns were sacked.’ By belittling the event, (note that Boudicca doesn’t even get mentioned by name) in the middle of detailed accounts of the organisation of games and public performances and the burning of Rome, Suetonius creates the image of an Emperor who had his priorities completely wrong.
The longest account of the uprising is given by Cassius Dio, who wrote a history of the Roman Empire in the first half of the third century. Cassius Dio had been brought up in Asia Minor, modern western Turkey before becoming a member of the Senate following a long career in the service of the Roman Empire, which had taken him mostly to Eastern provinces, but amongst which his northernmost appointment was as governor of Pannonia Superior (modern Hungary) c. 226/8, when his excessive discipline and attitude caused problems with the army of the province. As a result the animosity of the army appears to have led to him not entering Rome for his second consulate in 229; instead he returned to his home province. The Histories appear to have been written over a period of about twenty years from 207 onwards. While Cassius Dio’s original accounts appear to have been quite detailed only books 36–60 (dealing with 68/10BC), 55–60 (dealing with 9BC/46AD) and part of books 79–80 (217/c. 220) survive; the rest of his work mostly endures as an abbreviation (epitome) by Johannes Xiphilinus (and an even shorter version by Johannes Zonaras), both historians of the Byzantine Empire of the eleventh and twelfth century respectively. Xiphilinus is famously not really summarizing the text, but selecting material here and there for reasons that for the most part elude us, omitting substantial sections of the original without trace or reference, which makes it hard to determine whether the choice of material was Xiphilinus’ or Cassius Dio’s.
The surviving account, whether by Cassius Dio or Xiphilinus, is dominated by two battle speeches, which can be summarized by Suetonius Paullinus’ ‘Romans win because it is their birth right’ and the speech of Boudicca, which suggests that Britain ought to win because the Britons were the more manly warriors and they wanted their freedom back, especially their freedom from taxation. The latter point is important, because in Dio’s account the Boudiccan uprising, which is again credited with destroying only two unnamed Roman cities, is depicted as mainly a tax revolt, precipitated by the money-grabbing attitude of the governor, and especially Seneca (who at the time of the event was still the chief advisor to Nero). The description of a tall Boudicca, red-haired, in long loose, multi-coloured clothes with a hoarse voice, holding a spear would have been to Cassius Dio very much the epitome of everything uncivilized, in many ways as shocking as the first punks were to the British establishment of the 1970s.
There is little that is likeable about Cassius Dio’s Boudicca; she was wild, uncivilized and needed putting down, a service that Suetonius Paullinus promptly rendered. Any further British attempts at continuing warfare ended with Boudicca, when she died of illness – a very unheroic death indeed.
Overall Cassius Dio’s account apportions little blame to the Romans, but when it does, it focuses on Seneca’s money lending trickeries, which create an echo to his earlier description of Nero’s extravagance and his resulting constant need for money in the aftermath of the murder of his mother Agrippina. For Cassius Dio the uprising is just one of those things the barbarians in the provinces would do, and a proper Emperor would have dealt with it properly and kept his court in better control to avoid this situation arising in the first place. If anything, Boudicca’s speech seems to imply that the situation could have been avoided if Nero was a bit less effeminate and less given to entertainment. It is not clear if Cassius Dio is here necessarily commenting only on Nero, or if he was also using the opportunity to pass judgement on the rule of Elagabalus in the early third century, which he witnessed and which would have been fresh on people’s minds at the time of writing and whose combination of strong/controlling mother and decadent son provided a good parallel to the events of the first century AD. Either way it is worth noting that neither Cassius Dio nor Suetonius suggested that Boudicca deserved admiration.
This would suggest that the account most likely to have generated the hero worship described at the beginning of the chapter must have been written by Tacitus. Chronologically, the short passage in the Agricola (chapter 14–16) is closest to the events. But it only states that Suetonius Paullinus after two successful years as governor had laid himself open to attack from the rear, when he decided to attack Mona (Anglesey).
The fault for the uprising is seen as general exploitation and mismanagement by the Roman administration, and Boudicca is named as the leader, but in this account only the destruction of the colonia – thus most likely Colchester, is mentioned. Earlier in chapter 5 of the Agricola coloniae (plural!) are mentioned, but this probably reflects the requirements of a strident punch line ‘trucidati veterani, incensae coloniae, intersaepti exercitus’ (veterans cut down, coloniae burned and armies surrounded), which is otherwise all in the plural. An interesting fact is Tacitus’ comment that after a single battle order was restored:
But many retained their weapons. They were influenced by consciousness of their guilt as rebels and by their personal fear of the legate – in case this otherwise excellent man would take high-handed measures against those who surrendered and would punish every offence with undue severity, as if it was a personal affront (Agricola, chapter 16, trans. A.R. Birley).
Here too, Boudicca was just a rebel, who objected to Roman rule and sought freedom, but there is no suggestion that Tacitus saw this as a just cause. Noteworthy for Tacitus was Agricola’s involvement; he served as a tribune to Suetonius Paullinus (Agricola, chapter 5) during the rebellion. It was also a crucial part of his concise but strident history of Britain up to the arrival of Agricola as a governor.
The search for the origin of Boudicca as a British hero has thus led us finally to Tacitus’ Annals. He (Annals XIV, 29ff) begins his account with a description of Suetonius’ governorship, how in competition with Corbulo (a general who was at the time very successfully operating in Armenia, now Eastern Turkey/Georgia), he had decided to conquer Mona/Anglesey. Tacitus’ account suggests that Paullinus was in fact in the mopping-up stages of this operation, when Boudicca’s uprising began.
Tacitus saw the origin of the problem in the death of King Prasutagus and his will, in which he named Nero joint heir with his daughters. The resulting military/police operation by the Romans is outlined in a few short shocking phrases: Nero makes himself sole heir to the kingdom, the Icenian kingdom is despoiled by centurions, the king’s house by Roman slaves as if he had been conquered. Boudicca, the widow, is flogged, the daughters raped, the leading Icenians are robbed of their possessions and the relatives of the king are treated like slaves.
In response to Boudicca’s call for revenge, the Trinovantes and unnamed others join the uprising to regain their freedom. Money problems are mentioned as additional motivation when the priests of the temple of Claudius (which would be the provincial assembly discussed above) are described as being plundered of their funds in the name of religion. The people blamed for these problems are the military and the veterans, who were exacerbating the danger to themselves by leaving the colony without defences.
The procurator, Decianus Catus, who as the financial officer must have ordered the ill-judged intervention at the Icenian royal seat, is only mentioned by Tacitus as doing nothing except supplying a mere 200 insufficiently armed men in defence to diffuse the situation.
Consequently Camulodunum was destroyed by the attack; the temple of Claudius which became the refuge for the Roman population held out for two days longer, but eventually was also burnt to the ground. The next attack was on legio IX Hispana and its commander Petillius Cerialis, who had tried to come to the rescue of the colonia. As a result the infantry was nearly completely destroyed: only the cavalry and the commander managed to escape.
In the meantime Suetonius Paullinus, returning without the army from Anglesey, had reached London, but finding the town undefendable, abandoned the site to its fate. London was destroyed and Verulamium plundered, resulting in c. 70,000 slain Romans, while apparently the military sites were left unmolested or at least unmentioned by Tacitus. In the meantime Suetonius had pulled part of the legio XIV Gemina and legio XX Valeria Victrix as well as the nearby auxiliaries together and offered Boudicca battle at a site described as follows: a position approached through a kind of corridor, and protected in the rear by a forest, so that the British could not attack from any side but the open plain in the front.
Paullinus deployed his troops in a standard battle line: legions in the centre, flanked by the auxiliaries, while the cavalry is stationed at the ends of both wings. The British are described as disorganized and had even brought their wives as witnesses, who they positioned at the far end of the battlefield. We are given two speeches by the opposing commanders: Boudicca’s is given first and focuses briefly on the fact that they are trying to gain liberty from Roman rapaciousness and that the gods were on their side. And thus it was their destiny to conquer now or fall. The speech culminates in the punchline: ‘Id mulieri destinatum, viverent viri et servirent.’ – This is the decision of this woman, men might choose to live, but live as slaves. Suetonius by contrast is pragmatic when he exhorted the troops to do their duty and prove their superior fighting skills, despite being in the minority.
The battle was fought and the lack of discipline led to the break of the British battle line and the slaughter of all British warriors, women and horses within reach of the Romans. In a typical Tacitean style we are informed that at the end 80,000 British were dead, while only 400 Romans were killed. Boudicca commits suicide by taking poison. Poenius Postumus, the camp prefect of the legio II Augusta, who had refused a direct order and not brought his legion to the battle, committed suicide by an unknown method.
Taken out of context, the way Boudicca is represented does not seem very positive. However, when put into the context of the wider narrative in Book XIV of the Annals, her courage and the portrayal of wronged wife and mother gain force. Book XIV starts with the description of Nero’s murder of his own mother Agrippina, which went unpunished and indeed was celebrated by a Senate in fear of the Emperor. The narrative then turns to Nero’s behaviour in Rome, and the lavish amusements he indulged in, while Corbulo fought the Emperor’s wars in Armenia. The next year’s narrative opened with the Boudiccan uprising, before returning to Rome and the descriptions of Nero’s reign of terror, first against the Senate and later against the population as a whole, ending with the divorce and murder of his wife Octavia. All three women eventually lose their lives because of Nero’s actions, but the way they deal with their fate differs: Agrippina did not control Nero when she could; his wife refused to offer any resistance; even the Roman senators who suffered similar humiliations and who could have done something about it (for example by using the army under their command or attacking the Emperor during a meeting, or by organizing a conspiracy) took the insults quietly and acquiesced; until finally, eight years later the German bodyguards and Vitellius were finally willing to make a stand. In Book XIV, Boudicca is the only one (male or female) who responded to the insults and terror with ‘manly courage’ by standing up for herself and her cause, a talent for which Tacitus professes admiration throughout the Annals, especially when it was doomed to failure.
As the punchline said: ‘Id mulieri destinatum, viverent viri et servirent.’ – This is the decision of this woman, men might choose to live, but live as slaves. Tacitus clearly did intend this line to be a comment on the situation in Rome; but caged in an enemy’s battle speech, it gains in power, because an outsider has to show the Romans how to behave. Thus it was not a comment on or probably by Boudicca, but Tacitus’ verdict on the behaviour of the elite of Rome under a reign of terror.
The positive view of Boudicca that pervades so many of the modern accounts absorbed this undercurrent consciously or unconsciously. It also fed on a series of historic coincidences. The Boudiccan account in the Annals first becomes available again in Britain in the sixteenth century, first indirectly through the histories of Hector Boece and Ralph Hollinshead and later through the English translation by Henry Savile in 1591. This coincided with the reigns of queens both in England and Scotland and offered numerous opportunities to the contemporaries for some flattering comparisons.
A second era of huge popularity of the Boudicca story was the Victorian period, when Boudicca the wronged wife and mother, was identified with Britannia (and to a lesser extent Victoria). In a period when all over Europe Classical literature was used to find and redefine national heroes, such as Vercingetorix in France, Arminius in Germany and the Batavians in Holland, Boudicca filled a void presenting a heroine who could be interpreted as fighting for British liberty and justice, and who was much less problematic than the Caratacus story, which was intrinsically linked to (the morally questionable) Queen Cartimandua and her surrendering him to the Romans.
So which one of these accounts offers the real Boudicca? David Braund (1996, 145) summarized the problem very strikingly:
To seek the historical reality behind or beneath these images of powerful women in Britain is largely to miss the point. The characterizations of Tacitus and Dio tell us little about Boudicca […] but they speak volumes about these authors’ attitudes to women in power.
Given how much the account is defining Boudicca through her actions, rather than her words, one has to wonder how far the account has been shaped/distorted to reflect these very different views of her.
Having reviewed the different accounts, the second part of this chapter is to fill out the scarce accounts with some more details. Reconstructing a believable military history, which offers times and places in a clear order and makes this campaign reconstructable is beset with difficulties. None of the accounts presented above give us a time scale. We know that the main fighting took place in 61, because of the dates given by Tacitus. But as he starts his account with setting the scene and describing events that fall well before the fighting, it is unclear if the death of Prasutagus fell in the year before and if the disastrous visit by Roman troops is separated by a short or long timespan from the sacking of Colchester. Problems with the timeline continue as there is no indication of the time elapsed between the different events. Are the sacking of London and Colchester weeks apart or just 10 days? How long after the fall of London was the battle fought? In some of these issues, the archaeology can actually add further data, but this may not reduce the level of confusion surrounding the events described.
Boudicca is at her most visible in Colchester and London: the destruction layers in many parts of the towns are thick packets of burnt matter. In addition the excavations in Colchester have provided evidence of methodical looting and breaking up of statues before the town was systematically fired (e.g. Lion Walk and especially St Mary’s (Crummy 1997, 80–82; Aldhouse-Greene 2006, 186)). This suggests a substantial amount of time spent in the town. Furthermore, at two locations in East Anglia fragments of the Claudian equestrian statue from Colchester were found, the head in the River Alde and one of the hoofs at Ashill near Saham Toney (Davies 2009, 65), which suggests that at least part of the army had time to return home with its loot. A further fact not mentioned by either Tacitus or Dio is that Colchester had at Gosbecks a nearby auxiliary fort, which could have held c. 500 soldiers; whether these soldiers were actually present at the time, is again impossible to tell.
A further question is the site of the battle of the legio IX Hispana. Cerialis was supposedly coming to the rescue of Colchester. There would have been very little chance had he brought the legion from Lincoln; the marching times involved would have made that near impossible. Tacitus suggests in his text that the sack of Colchester came first, but this may be caused by his way of presenting the story in a clearer, more striking way, a literary trick that he has been proven to employ elsewhere. It is hardly surprising then that Aldhouse-Greene suggested that the attack of the legio IX Hispana actually happened DURING the British operations against Colchester (2006, 186).
Cerialis may of course have come from Longthorpe, where a vexillation fortress often associated with the legio IX Hispania is known. On the other hand as it was during the campaigning season, he may already have been on route to or from campaigning elsewhere. Tacitus suggests in his account of the aftermath of the uprising that Nero sent 2,000 legionaries from the Continent and so made up the losses in the forces. If this is accurate, it does not sound as if the whole of legio IX Hispana was involved in the encounter against Boudicca (in which case losing all but the cavalry would have been losses in the region of 4000 soldiers). It raises the next question: how big was Boudicca’s army in Colchester?
After Colchester, Boudiccan destruction is also reported from Chelmsford on the road to London (Waite 2007, 101), but these destruction levels are apparently not universal throughout Essex. The Roman small town of Elms Farm, Heybridge, which developed from Iron Age predecessors and contained an indigenous sanctuary, did not apparently sustain any noticeable damage during the uprising (Atkinson & Preston 1998, 92–98, where the earliest phase Late Iron Age to Roman transition is dated 50BC to 70AD with no noticeable interruptions).
When we get to London, the evidence is clearer: the destruction horizon is known to cover substantial parts of central London, both sides of the Walbrook, as well as an area in Southwark, as shown by the Jubilee Line excavations in the Southwark area (Bird 2004, 51). But if the rebellion crossed the river, why did they not go south or west to Silchester?
Tacitus mentions three towns by name in the Annals: Colchester, London and Verulamium. Suetonius and Dio clearly state two were destroyed. It would be facile to suggest that Cassius Dio has made a mistake; he appears to have got the number from a similar source as Suetonius – and Suetonius is potentially the historian with the best access to the official archives. It may be better to assume that there was some doubt over the third town, i.e. Verulamium. Tacitus only mentions this as being sacked.
Interestingly in her account of the archaeology of Verulamium, Ros Niblett (2001, 67) draws attention to the lack of evidence for the widespread destruction.
Verulamium has not so far produced evidence for contemporary burning comparable to that found at Colchester. At Verulamium the only buildings that were definitely destroyed at about this time were the workshops in insula XIV; the bathhouse in insula XIX may also have been damaged but it was not totally destroyed. Burnt deposits at low levels in excavations in other parts of the Roman town, most of them found in the 1930s, have since been shown to contain pottery made in kilns in Highgate Wood, north-west of London, and dating from not earlier than c. 80AD. All this confirms the impression that at the time of the revolt Verulamium was still very much an emerging town where there was, as yet, comparatively little to destroy in terms of ‘Roman’ buildings. Lack of excavations makes it difficult to gauge the effects on ‘native’ farmsteads in the immediate area, although on the sites that have seen excavation – Gorhambury, Park Street and Prae Wood – it is difficult to identify destruction layers that can be attributed to Boudicca with any degree of confidence.
On the other hand over the last few years excavations at Silchester continue to encounter evidence for destruction that appears to date to the mid first century and which is extensive enough to relate to Boudicca, rather than any accidental fire (Fulford & Timby 2002, 568–569). This, combined with the confirmed presence of Boudiccan destruction deposits in Putney, Brentford and Staines, all on the route to Silchester, suggests that in addition to the destruction mentioned in the literary sources, there may well have been a further path of destruction aiming west from London rather than north (Fuentes 1983, 216–7).
Furthermore, unless confronted with the excessive remains of burning seen in some parts of London and Colchester, it is difficult to differentiate between an accidental fire (caused for example by oil catching fire in a pan or a lightning strike, or a fire deliberately set to level a site prior to new building) and deliberate enemy destruction. Human remains are a rare find in most destruction deposits, as most people would return and reoccupy the plot and bury the dead, if only to avoid the spread of diseases. Finding a burnt layer is thus not necessarily an indicator of enemy action.
But the question reaches further: which other tribes but the Trinovantes and the Iceni took part in the revolt? Apart from the Claudian statue pieces, there is little evidence of the revolt in the Icenian heartland; in fact, it is much easier to trace its aftermath than the actual revolt. Pottery industries operating in eastern Cambridgeshire before and after the uprising show little sign of interruption (Gibson & Lucas 2002). There appears to be little evidence of destruction further west amongst the Corieltauvi or the Brigantian territories and for the most part the areas south of the Thames appear to be untouched as well.
This leaves the question of the final battle site. Most commentators agree that it should lie somewhere along Watling Street, but in the absence of a clear line of march it is worthwhile calling to mind that this reconstruction is based on the fact that we know Paullinus left in search of more troops, most of which were still coming back from Wales and that the British army was last seen at Verulamium just on Watling Street – thus suggesting that they may have progressed further along it. We again have no indication of how much time extended between these two events or even if the British army really all operated in a single unit – this is at best a good guess – and thus any of the predictions of a battle site, be it at Mancetter as Webster (1978) suggested or the more recently suggested High Cross (Waite 2007, 173–181) depend on this first suggestion being correct. It should perhaps also be mentioned that Nick Fuentes, S. Kaye and David Bird argue that if Suetonius Paullinus hoped to join up with legio II Augusta from Exeter, then it would be more reasonable to expect the battle to have taken place close to Staines (where burning is associated with Boudicca) or Silchester, rather than in the Midlands (Fuentes 1983, 314–316; Bird 2004, 26; Kaye 2010, 30–33). The written evidence allows for both scenarios, and in the absence of any firmly identified traces of the uprising or the battle at either location, an open mind may well be a valuable asset for future researchers.
The only time we get a bit of topographical detail is the actual description of the battle site. The site is described as artis faucibus (in the style of the entrance corridor to a Roman house). This is usually translated as ‘defile’, but might just as easily suggest that a plain narrowed before widening out again (perhaps by woodland or slopes) and with a wood behind as cover (and incidentally a serious obstacle in case of flight). A recent computer analysis of this information against a map of southern England suggested 263 possible battle sites covering most of the Midlands and the Thames valley (Kaye 2010, 32). This in itself suggests that the information provided in the literary sources is too generic to identify the site itself.
Paullinus’ battle lines are arranged as:
cavalry – auxiliary – legions – auxiliary – cavalry
However, the accounts stress that Paullinus kept the army for as long as possible on the narrow protected ground, making sure that the British forces could only attack from the front. Eventually the infantry attacked in wedge formation, while the cavalry rode the heaviest opposition down. It is hard to assess if this is really how things happened, but if it did, then Suetonius Paullinus decided to play this battle exactly by the textbook, taking absolutely no chances, but going for a battle line that had been tried and tested throughout Antiquity. The battle as described in the Annals suggests that it was lost by the British through a lack of discipline rather than any sophisticated tactics employed by either side. The fact that Cassius Dio’s description stresses the long duration of the battle indicates that it might not have been as easy a victory as suggested by Tacitus.
The only figures that are given are that about 70,000 or 80,000 (according to Dio) Romans died in the uprising and that Boudicca lost 80,000 warriors in the final battle. Unfortunately, these numbers are not reliable, as very few from ancient battle accounts ever are. There are certain rules to what the expected tally should look like: usually, the winner has to have lost fewer people than the loser. So 80,000 British dead at the final battle means probably little more than ‘heavy losses but in total the same as our side lost during the revolt and a few more’. This helps little in our reconstruction of the size of Boudicca’s army at the time, nor indeed for Suetonius Paullinus’ army which consisted of one and a bit legions (the legion from Exeter had decided not to march and the legio IX Hispana had already lost in an earlier battle, leaving two legions, which may have been understrength due to the earlier campaigning in Wales) as well as auxiliaries from the nearby forts – we may be talking about 10,000–12,000 fighting men, but there is just not enough data.
It would be good to be able to discuss this battle in more detail, but the description of the battle is so short and cursory that the points that would really interest a military historian remain once again obscure due to the lack of sources. The accounts differ again as to what happened afterwards. According to Dio, sufficiently large numbers of the British army escaped the battlefield and were preparing to fight again, when Boudicca fell ill and died. Apparently she was awarded a costly burial, before the Britons dispersed to their homes. According to Tacitus Boudicca killed herself by poison (Annals XIV, 37).