As this book is about military history, the written accounts of the Romans about their successes and achievements in Roman Britain are going to of necessity dominate the narrative: this is unavoidable. Archaeology is good at recovering the remains of processes that took a considerable time, if only because the occupation of a Roman fort over 400 years is likely to have generated more discarded archaeological evidence (i.e. rubbish), than the overnight stay of an army on the march to a battle site further north or a naval encounter in the middle of the English Channel. Archaeology is just not good at detecting small scale and short-lived historical events, although admittedly cataclysmic events such as the sacking and burning of Colchester and London, or an entire battlefield as at the site at Kalkriese or Kalefeld in Germany, leave their marks in the archaeological record, even if the relevant historical account may not have survived.
We are thus most of the time at the mercy of the historical accounts as left to us by the Romans in word or more rarely images (the Iron Age has so far failed to produce equally identifiable ‘speaking’ records for us to recognize). ‘At the mercy’ might not be too strong a term in this context. With the exception of Julius Caesar’s Commentaries, we have no records of Romano-British events from the perspective of an eyewitness. Most of our literary sources were written by Mediterranean residents, living for the most part many years and often generations after the events described. Even the accounts of Tacitus of the governorship of his father-in-law Agricola and Dio’s account of Severus’ campaigns in Britain were composed more than a decade after the events described. None of these would be considered by historians of more recent periods as primary accounts. Tacitus may have been using eyewitness accounts for some of the events described in the Histories and perhaps the memories of his father-in-law, and there are several occasions in Cassius Dio, where he claims to have been an eyewitness to events in the first half of the third century. However, with increasing distance from the event (be that chronological or geographical) most of the accounts had to be secondary in nature, using earlier accounts, official documents or possibly even oral traditions, written by authors selecting material of variable quality to compose a historical account of their own choosing.
This leads us to the second problem: none of our surviving records offer a history of Roman Britain. All relevant passages (including Agricola’s exploits in Britain) were culled from larger documents, usually providing a much wider context, most commonly an account of the history of the Roman Empire at a certain period or a biography of a series of Emperors/Imperial officers. In some cases we are dealing with speeches in praise of individuals, including the Emperors themselves. In other cases the material comes from a text, prose or poetry to which the historical events are all but coincidental. Some sources focus thematically: e.g. a history of early Christianity. This contextualization is important. Developments in Britain are rarely of vital importance to the wider Empire; usually the British sections are just short chapters within a larger whole.
Anybody who has tried to gain an understanding of the military situation of the Anglo-Scottish/Irish wars under Edward I, or the Stuart kings from any social history of Britain under the Plantagenets/Stuarts/Tudors, will be able to relate how much or rather how little of the information of interest to a military historian gets included in these accounts. A similar case applies to ancient sources – Tacitus does not hide his intention to offer a view different to other historians, who he accuses of falsifying the record for various reasons (Hist. I and Annals I). But in addition to his stated aims in his prooemia, Tacitus’ Annals are constructed around the concept of a lost libertas, offering first and foremost a social study of Rome and the early Emperors and how Roman society was corrupted by their rule; in the process of which he destroyed the reputations of Tiberius, Caligula and Nero. His account lives through lots of vivid details and characterizations. He makes an excellent point and makes his point in extremely well written language – and he has rightly been named as one of the world’s best social historians. After all, his account of life under a reign of terror must be one of the best psychological sketches attempted on the topic.
However, and this is worth keeping in mind, whatever has been expressed above in a positive way, can also be expressed negatively: Tacitus is not good at sustained annalistic narrative of factual events. This may seem boring (which is possibly why he didn’t use it), but for a military historian, it is essential for a detailed analysis of the progress of a campaign. So the very features that make Tacitus an excellent social historian, work against him in the military arena.
A similar point can be made for nearly all writings with a historical content. The author’s thematic choices influence how and why events are included and how they are treated, or whether they were excluded as irrelevant. None of this should come as a surprise to a modern historian. Selection of material is a very important tool of the trade and thus it is hardly surprising to find that it was already practised in Antiquity. After all, few of us would expect an Argentinian history of the Malvinas to culminate in praise of the British forces in the Falklands, or a sympathetic biography of Winston Churchill to be complimentary of the election outcome of 1945. It is, however, interesting that we credit ancient historians with more universal interests than we would expect of their modern counterparts.
Another point is well worth making, or rather repeating: historical writing in Antiquity is meant to entertain, as well as (morally) educate. This may sound a lot more unusual than people give it credit for, and at the same time a lot less usual than people experience nowadays. There is a tradition in European historical writing of the well-researched academic tome, several inches thick, with lengthy annotations at the bottom of the page or the end of the chapter. It is usually prepared by learned historians based at well-established universities, and is sold at exorbitant prices by the specialist bookshop or bought directly from the academic publisher. It is not a requirement of the genre, but many of these volumes are written in difficult, if not impenetrable prose, that cures insomnia; and with some of the older examples you find that one hundred years after publication the pages after 250 are still not cut.
Many of these works are rightly considered pinnacles of scholarship; their presentation of the period carefully balanced and attempting objectivity of account and presentation throughout. But despite the immense amount of time and skill that has gone into writing these books, they tend to have a very limited readership, and their price and language give the appearance of elitism, whether intended or not.
In the other school of writing history, the books are frequently found to be less than 200 pages long, mostly soft-backed and (lavishly) illustrated. There is little referencing to the origins of the material discussed and instead of a comprehensive bibliography they contain a few pages of suggested ‘further reading’. But these books tend to be well written (possibly because often they are written by professional writers such as journalists rather than university based academics), entertaining, reasonably priced and frequently very popular with at least that part of the population that is interested in history, including the writers of the large tomes discussed above.
The two cases illustrate the two extremes within the genre of historical writing. There are many books (including, it is hoped, this volume) that sit somewhere in-between and will satisfy both tastes, of the academic and of the non-university-based lover of history.
It would be easy to deride the second school as a recent development of a ‘dumbed down society’ dallying in ‘education lite’. But most teachers would agree that it is much easier to educate on any subject, if you are simultaneously able to entertain and capture your audience. This realization is centuries old, and thus unsurprisingly, entertaining historical writing is the older of the two schools. While the methodology and technique of the worthy tomes was mainly developed in the late 18th and 19th centuries, the readable and entertaining account has been with us since Herodotos. Few of the ancient writers are particularly good at, or interested in, giving direct references to their sources, but most of the surviving texts know ‘how to spin a good yarn’ (problems with the translations not withstanding) and tell their story well. It is thus no coincidence that in English (as well as French and German) the words for historical writing and storytelling are closely related.
All historians write history on the basis of their own experiences and the times they live in. This may not be readily apparent when the books are first published, as the readership is likely to share the writers’ preconceptions, but if a historical account ‘feels funny’ or seems ‘dated’ when read thirty years later, then it is likely that the experiences of the readership have changed from those of the writer and his/her original readers. Ten years ago, it was fashionable to talk about the falsification of the historical record or the use of propaganda on display in historical (and visual) sources. Both terms refer to a wilful selection of material and the order (or way) in which it is presented in a source. It is an old adage; that just because events occasionally happen in a particular order, it does not follow that they were caused in that order or because of each other (this belief is sometimes referred to as ‘post hoc ergo propter hoc’). A black cat crossing the road is not usually the cause of bad luck to the witness, despite all proverbial statements to the contrary. Consequently if you are trying to give the reasons why certain events happened, it may be better as a historian (or storyteller) to change to a causal rather than a chronological order of events or occasions.
The selection of which material to include is a very personal decision by the writer; this may not be readily apparent at the time of reading. Cassius Dio for example has a tendency to be critical of women in positions of power, especially those who act without proper advice from male relatives. It is possible to blame this on his experience of living through two minority reigns, when the Roman Empire was run by empresses as regents for their small sons (Iulia Maesa and Iulia Soaemias for Elagabalus (218/222) and Iulia Mamaea for Severus Alexander (222/235)). Alternatively it could be due to his cultural background as a Greek from Asia Minor, where this sort of behaviour was not widely approved of (if you think this unlikely, reflect on what Winston Churchill thought of Nancy Astor as an M P, and how Wellington would have reacted to the idea of women in Parliament). However, before we think of these fairly transparent sources of ‘inspiration’, also consider the option that he may be reacting to his possible childhood experiences (domineering mother, bossy sister – the list of possibilities is endless). In the absence of any autobiographical musings or surviving diaries/personal letters by Cassius Dio, all or none of these scenarios may be correct – but, to pick up on an earlier point, all these scenarios would be valid.
Identifying bias in the written record can be comparatively easy, for example when Eusebius appears unable to identify any positive characteristics in Constantine’s opponents. Identifying the problem is, however, different from explaining the reasons for this bias, especially if it is less obvious than Eusebius’ hero worship. It is then all the more important to discuss the differences in the varying accounts instead of just brushing over them, or summarily declaring one account superior to the other. By exploring the differences we should be able to find out more about the events and the writers involved.
Every now and then the current readers of ancient sources suspect, or in some cases are able to prove, that source material has been intentionally misconstrued or misrepresented to make a political point. In these cases we need to ask age old questions about motive and opportunity, keeping in mind that it is possible that the material may have been subconsciously influenced by the general climate of the time: no falsification or misleading may have been intended; after all the difference between propaganda and blind belief is bound up with whether the author believes what he says.
On the other hand the motives for misrepresentation might be the need for decisive argument in a political speech at a critical point in Rome’s history (Tacitus’ Agricola and Cicero’s Philippics would probably qualify). Opportunity to use this device to good effect might reflect a chance of writing an account or a laudatory speech (such as the panegyrics of the late Empire) for a particularly important person, who may then reward a flattering description of events.
Whatever their motivation, the ancient writers who told the (hi)stories discussed in the current volume, wanted to impart something that they thought important: they saw their stories as part of a wider context. This may have influenced how the stories were told and how much detail (and more importantly which details) was included. This is important if a classical author thought that military success might be dependent on the moral character of the ruler at the time (as Herodian and Eusebius do), or on the character of the general in charge (as Tacitus suggests in the Agricola). Ancient writers saw parallels in how an Emperor dealt with the assassination of his mother and the partially illegal, but unnecessarily cruel integration of a client kingdom into the Roman province (as Tacitus did with regard to Nero’s murder of Agrippina and his treatment of Boudicca).
To underline these parallels, authors like Tacitus or Suetonius used literary ‘tricks’, citing passages from plays that readers should be familiar with, to suggest to readers how they would like to see a particular event judged. Recasting the entire scene, so that it mirrors or imitates the events of the possible parallel is another. These analogies are often what made these ancient accounts more readable and easier to understand, but to a modern historian they can make them more unreliable as our small sample of surviving ancient literature might make us miss important clues. There are also expectations to fulfil, a case in military history is the imitation of Alexander the Great, thought by many Romans to be the greatest general ever – thus any good general should betray traces of Alexander – Pompey did this by adopting a hair style reminiscent of Alexander and calling himself Magnus (‘Great’). Another way is to credit a general with similar experiences to Alexander (the stories told about the birth of Scipio Africanus and Alexander were strikingly similar). But it can also happen, by crediting a general with using the same stratagems as Alexander. All of this can make ancient battle descriptions extremely hard to judge – how much is real, how much is flattery?
In addition, there are standard ways in which a battle can be fought; infantry in the middle, cavalry on the side: the aim is to encircle the other army or make it run away, by breaking through its lines. This might even reflect reality. If this is how battles are supposed to be fought according to the handbooks or ‘established wisdom’, the majority of generals would probably have done so. It is also possible, if no better information was available, that the writers copied this standard engagement pattern, or just referred to it obliquely, by picking out short scenes, rather than risk boring the audience by giving a blow-by-blow account.
The ultimate literary tool is the battle speech. It is well known that these speeches are unlikely to ever have been the great set pieces of oratory we see in historical writing, if only as without a PA system, very few people would have heard the general and the remainder would have become restless. On the other hand, speeches by the enemy are unlikely to have been available as transcripts after the event. However, battle speeches offer the writer the opportunity to summarize the main differences between the two parties, both in their demands, as well as in their approach to war (discipline vs enthusiasm; righteous need for retribution vs untamed bloodlust). It also gives them the opportunity to portray the moral superiority expected of a Roman general in order to win a battle against Rome’s enemies. The enemies’ speeches on the other hand give the Roman writers a way to voice the arguments against Rome, in a form that would have been at times borderline treasonous if uttered from the mouth of a Roman (‘They create a desert and call it peace’ is probably the most striking example (Tacitus: Calgacus’ speech in the Agricola)).
The one thing to remember about these speeches is that they are the equivalent to our analysis chapter in a campaign description, but they are not authentic: they present the author’s view, not the view of the players.
The problem is that unlike the situation for the history of Rome generally (for most of the history of Roman Britain our sources are so thin) we have problems independently verifying the events mentioned, let alone to actually make more than educated guesses whether a divergent view reflects deliberate falsification of history, a genuine mistake or the personal choice of the author, based on experiences unknown to us.
On the other hand for Roman Britain, unlike the senatorial debates in the city of Rome, we have a further control, the archaeological record. Archaeology, as we have seen, may have its limitations, but occasionally it offers additional insights, such as the dendro-date for Carlisle or the construction dates for the Saxon Shore forts that can change our views of a particular passage or more often broaden our understanding and contextualize events that we would otherwise find ourselves at a loss to interpret.