At the beginning of the twentieth century the Camden Professor of Ancient History, Francis Haverfield, was compiling evidence for a history of Roman Britain. As a classicist and ancient historian he was well versed in the Greek and Latin ancient sources; on the other hand the increasing amount of archaeological data from excavations in Scotland, Hadrian’s Wall, Wales, but also the numerous finds of Roman material from London or the villa sites in the south of England lacked chronological order beyond the information available through the finds of coins and the few inscriptions naming Emperors recovered at the site.
He decided that the literary sources and their accounts could provide, at the time, the most reliable chronological framework for the study of Roman Britain. Thus a collection of relevant historical texts was created, a handlist. This is the basis even today of our understanding of the history of Roman Britain, and it is small wonder that its current (expanded) reincarnations – such as Stanley Ireland’s much reprinted Roman Britain: A Sourcebook (2008) can be found on the shelves of most Romano-British archaeologists.
Many, but by no means all of these sources were written by Romans (or at least holders of Roman citizenship), who came from the Mediterranean world, having grown up in various provinces, before pursuing careers in the centre of power, Rome itself, either as politicians, or as writers patronized by politicians and Emperors. Relatively few of them offer irrefutable evidence for first-hand knowledge of the British Isles, and none have a claim to Britain as their home. Consequently the reason their writings mentioned Britain could be many and various, but frequently were not derived from a need to satisfy ethnographical or topographical curiosity. More commonly Britain was the stage on which events important to the Central Government took place. If the interest of the Central Government shifted, the focus shifted and Britain could drop off the historical radar of the Mediterranean historians for long periods.
It is possible that the above picture is exaggerated. After all the writers that we have available today are not the sum total of the literature accessible at the time, but probably only a small percentage of the accounts and writings that may have been around. A new source, whether epigraphic, literary or even numismatical (not to speak of the archaeological evidence, which has slightly different problems) has the power to completely change our perception of events and as such suggests caution in trusting the statements of any one source too implicitly.
Since the first handlist of historical sources was compiled, Romano-British archaeology has tried to expand its knowledge base in all directions and the material that is today available to scholars is infinitely richer than anything that Haverfield could have hoped to analyze. We have, for most of the island, sequences of pottery that create a historical backbone independent (or nearly so) of the historical sources. These sequences allow us to write an account of the Roman period for areas such as Yorkshire, that received little mention by Roman writers; in fact in the early 1990s Martin Millett argued for the irrelevance of much of the written record for the history of Roman Britain (1990).
In the end, however, a coin is just a coin and a pot sherd is just a pot sherd. To advance from a catalogue of archaeological material found at a site and the dimensions of a building excavated, to an account of what we think it was like to live in Roman Britain at point x in time, is a matter of interpretation, and thus dependent on the context in which the archaeologist/historian operates. Archaeological interpretation reflects our own experiences and expectations, often without us realizing it. This is why students today spend a substantial part of their time at university learning archaeological theory – a frequently little loved, but very important aspect of their skills, which (hopefully) informs them about the origins of some of the interpretations they learn and the various levels of reliability of interpretation, as well as some basic concepts of logical deduction. One of the most valuable lessons they can take away from these courses is the realization that with regard to the past, the use of the term ‘truth’ as a fixed value may be ultimately unachievable. The amount of material recovered, which is usually considered to be between one per cent and 0.1 per cent of what was originally there, makes it unlikely to ever get to the bottom of what really happened at any given time in Roman Britain – in this respect archaeology and modern forensics have very different expectations. The best any researcher can hope for in reconstructing the past is a ‘valid’ interpretation. The difference is easy and important. Truth is what computer people call binary – it either is or it isn’t. A valid statement by contrast is any statement that takes account of all pieces of evidence as they are known at the time.
To use a case in point: in a typical crime novel a murder victim is found shot in a locked house; none of the other six guests in the house can have come or left since the victim was last seen alive. Without any further information all you can hope to achieve is to produce a list of valid suspects: the number of people in the house with the victim (including the victim itself, as at this point no information has been presented to rule out suicide).
Only with further information can you hope to reduce the number of valid suspects. In our case, if the victim was shot in the back, we can rule out suicide. In a satisfactory crime novel, the outcome will result in the final denouement, that one person who cannot be ruled out and for whom clear evidence exists that links them to the crime.
Roman archaeology (like the archaeology of many other periods), can rarely hope to get beyond the creation of the list of likely suspects, and the most we can do is to present a particularly attractive scenario – but it remains just that, one scenario of many.
What not all archaeologists of today realize, but what was already clear to Haverfield and his colleagues, is that the written sources are equally prone to interpretation. None of the texts that were excerpted by the early scholars of Roman Britain exist in isolation. They are the products of the time in which they were written, formed by the intention, knowledge and experiences of their writers, and their content is limited by what the writer and its audience deemed relevant and/or interesting.
Even the way in which something is phrased may have been decided by consideration of style. It may have been deemed a lot more relevant that the paragraph ‘flows well’, rather than provide utterly accurate but perhaps boring reading. Latin literature was as much, and probably more than modern prose, governed by considerations of acceptable forms of oratory. We may have rules about avoiding repetition, and not starting sentences with ‘and’, to name a very few; but to a Roman writer, clearly structured prose, carefully polished and displaying full command of Latin or Greek, often showing that they could imitate styles of other writers or turn a good phrase was a form of defining their identity as a truly cultured human being (hence the use of the term ‘humanitas’ for this type of display of education).
A lot of the study of ancient history and classics in the last 100 years has been devoted to better understanding the few bits of ancient literature that we have. Only rarely do we still try and establish the best text in terms of what the Latin or Greek actually says. Much more important to classicists today is what it all means. Any parent of a teenager, or any teenager with a parent, will be able to vouch for the fact that these two statements are not synonyms and that a certain amount of care is needed even today to find out what is being said and what it means.
So why should this outline take up the first two pages on a military history of Roman Britain? Part of it is due to the history of the subject: as more information on Roman Britain became available, it has become more and more common to specialize. Today you can have specialists on Roman pottery, on Roman glass, on Roman villas, as well as on Roman military equipment. Few of us are of necessity true generalists and it can be argued that the level of information is now such, that it is pretty impossible for one person to keep up with everything that is written on all fields of Roman Britain. Perhaps this is one reason why there are now so few detailed general accounts of Roman Britain of similar depth to those written by Sheppard Frere (last updated 1987) or Peter Salway (1981) in the late sixties or early eighties. This is not to say that there aren’t numerous accounts of Roman Britain on the market; they are usually addressed to beginners or as a general introduction and some (names have been purposely omitted at this point) clearly believe this to be an excuse to reproduce in an uncritical fashion material culled from books several decades old, as little of importance can have changed.
One of the side effects of this expansion of the subject is an inability (for reasons of time as often as not) to engage with material from a neighbouring sphere of expertise. If the deadline is looming, it is only too tempting to drop the trawl through the literature beyond the must read section (and while a lot of care has been taken with this book, the author is sure that there are important articles she will have overlooked), especially in adjoining fields. What this book hopes to achieve, is to point to some of the studies in neighbouring fields (especially classics) that influence our understanding of Roman Britain and especially its military history. This book will be looking at familiar sources, and then comparing them with data from archaeological reports, but also from studies into classical literature and ancient history as well as some other areas. It will thus try to return to first principles in order to determine how much of what we thought we knew about Roman Britain can really be relied on. The resulting accounts may debunk or at least give us pause for thought over some cherished details, some of which may have been based on little else but the Chinese whispers in a century old discipline: this may bring an almost emotional shock to some. Nevertheless, if successful, it should leave our subject with stronger foundations on which, in future, a more reliable structure can be built.
The point of this exercise is not to discredit one source or the other, but to raise occasionally a note of caution in view of our changed understanding of how and why literature was written and read in Rome, and to occasionally provide alternative scenarios that may on present evidence be equally valid. It is hoped that in future it will be possible to gain more information and limit or refine the scenarios, but for the moment the idea is to underline that there are alternatives or at least problems with some of the existing versions of our established history of Roman Britain, which should offer opportunities for further study.
Even in a book this length and with the limited existing amount of evidence for Roman Britain, it is not possible to cover every aspect of the military of Roman Britain. The material covered was chosen, either for its intrinsic importance, or because it illustrates a particular problem of the historical reconstruction. Hopefully, I should be able to demonstrate the way the argument can develop, when all data is combined and induce somebody to take the argument into areas I have not covered. I have, over the years, tried to make myself familiar with the arguments of my friends and colleagues who specialize in ancient literature, but I am the first to admit that while I spent some time studying Latin and Greek to university level and am still teaching Latin, I am not a philologist in the true sense of the word, or of a calibre approaching some of the experts in the field, who have helped me with my enquiries over the years. I hope I have done justice to their considerable efforts to make me understand the finer points of Roman and Greek literature and apologize for any misrepresentation; it derives not from ill-will, but from a passion for making a wider audience understand the importance of their work.