Preamble

The idea to produce a collective book centred on the transformation—in some cases, the fall—of Roman frontiers, over the entire duration of the ancient phase of the Empire and on all the regions that comprised it came up during a discussion between the two editors of this special issue of HiMA, which was held on the sidelines of the 25th Congress of Roman Frontier Studies/Limes Congress (Nijmegen, Netherlands, 21-27 August 2022). Their motivation for such a project was dictated not only by the lack of such a work to date, but also by their common interest in the mutation of border systems and life on the margins of the Empire in Late Antiquity.1 In addition, both also share an interest in the very notion of a border in the Roman era and, in particular, the meaning of the different Latin and Greek terms used by archaeologists and historians to describe it.2

They themselves were convinced of the interest of their project, but they did not know, initially, what the reaction of their colleagues would be, knowing that specialists on the subject usually focus their research on the origin and functioning of border systems, and not on their metamorphosis or their disappearance. To their great surprise, the reaction of their colleagues was quite enthusiastic (all those we asked responded favourably to the call) and the very rich final result confirms what the editors initially thought: this special issue of HiMA on The Limits of Conquest and Transformations of Roman Imperial Frontiers constitutes a true historiographical moment, obviously not because of the contributions of both editors, whose sole purpose is to introduce the dossier, but through the papers of the other authors. All of them are the leading specialists in their field. Obviously, other regions and mechanisms than those discussed could have been addressed, but this would have been far too much material for a single issue of a scientific journal. Nevertheless, it is one of the objectives of this special issue to push for the production of an even more complete synthesis on the topic in future.

And why did we choose a journal—moreover, a French one—to host this work? For two reasons: to improve the distribution of the book, given that libraries around the world subscribe to scientific journals such as HiMA (while not all books are purchased), and to make the main topic better known in the French-speaking world (even if the contributions are in English), perhaps even in the Romance-speaking world in general (as the journal is French), which works less on Roman imperial borders than other linguistic spaces. We don’t know if these objectives will be fully achieved—we really hope so!—but that’s what motivated our choice!

Before ‘giving the floor’ to the contributors of this special issue, the editors thought it would be interesting to quickly address some fundamental elements, in order to clarify for readers what we mean by the frontier in the context of the Roman world. This introduction will therefore discuss the vocabulary used by the Romans, focusing particularly on a term more often associated with the reality described here than others, and which is often very poorly defined or used in contemporary historical publications: limes. As for my co-editor’s chapter, it also focuses on another commonplace often misinterpreted in the general historiography on the Romans, namely the role of walls in border defence systems.

The words for the profane border3

The ‘border’ is far from being a secondary notion in Roman history, the very myth of the founding of Rome being intrinsically linked to the tracing of the sacred boundaries that constituted the pomoerium. Given the central and fundamental nature of this notion, there is a whole historiographical tradition on real or symbolic borders, frontiers, limits, margins, and extremities in the Roman world, and a significant part of this tradition revolves around the study of the ‘centre-periphery’ relationship.4 At the very root of this interest—for which there is no equivalent in the Greek world—in what delimits the power of the Urbs and its leaders lies the difficulty of translating into Greek and Latin all of the realities that contemporary languages conceptualise today through the aforementioned suite of words. If we focus only on secular borders, given that this is the topic of this special issue, we mainly identify—considering both material and ideal borders—five groups of Greek nouns (ὅρος and all associated terms, τέρμων/τέρμα as well as πεῖραρ/πείρας/πέρας and their respective derivatives, ἐσχατιά/ἐσχατιή, λιμιτόν) and four Latin words (terminus, finis, ripa, limes) to designate the different nuances.

The most common Greek term is certainly ὅρος which refers, in its primary meaning, to a boundary stone; by extension and like different terms built on the same root, it came to mean ‘border’ in all senses, both literal and figurative.5 Tέρμων/τέρμα and all the associated words are very close to it, in that they also refer to the idea of a physical limit, but in a less precise way than a boundary.6 For their part, πεῖραρ/πείρας/πέρας refers to “what is unsurpassable, unless it is no longer in the element considered”,7 while ἐσχατιά/ἐσχατιή corresponds to “the place at the extreme edge, the end of the world or of a country, where the beyond is different”,8 that is to say at the frontier area of a territory. As for λιμιτόν, it is only a late transliteration of Latin limes, which is defined below.9

The quadripartite diagram that we can sketch out for the Latin side, largely inspired by Roman ‘gromatic’ literature (Roman agrimensores/surveyors), is not too far removed from the Greek concepts. Thus, terminus designates the terminal marking the limit of a property or a territory. Associated with imperii, this word would become only a simple metaphor equivalent to the expression finis imperii, the external borders of the Roman Empire having never known defined boundaries.10 The word finis would refer precisely to an “elusive and floating limit on the periphery of the orbis Romanus”.11 As for ripa and limes, which sometimes work in pairs in ancient texts,12 they are generally associated with natural borders such as the bank of a river, on the one hand, and with artificial fortified borders, on the other hand, although the situation is somewhat more complex with regard to the second of the two nouns.13

What was the limes for the Romans?

Of all the terms that have just been mentioned, limes (and its Greek equivalent λιμιτόν) is the one on which the most ink has been spilled, while remaining the one whose definition remains the most uncertain and subject to discussion. The situation can be explained, at least in part, by the absence of a full study taking into consideration all the available sources, throughout Antiquity, in a context where almost all attempts at a definition thus far have mainly sought to characterise the limes during the High Empire, even though two-thirds of the references to it date from 4th-6th centuries.14 We are therefore witnessing an anachronistic processing of the topic, as if the limes had been a single administrative and military reality during the six centuries of the ancient phase of the Empire. Furthermore, the definition that has been most commonly accepted since the 18th century, that of limes as a linear defensive system, composed of walls and fortresses, aimed at marking the physical limits of the Empire and preventing external enemies from penetrating it at will, is today refuted by many scholars.

If Edward Luttwak15 was the scholar who theorised this so-called ‘grand strategy’ in its most fully developed form, his conclusions should be understood as the culmination of a long historiographical tradition that sought to interpret the Roman imperial model through the lens of the functioning of contemporary nation-states, a process in which the contribution of the Reichs-Limeskommission (1890-1937) cannot be overlooked.16 The first major challenge to the limes as a linear militarised border resulting from a policy of the central power was articulated in the works of Charles Richard Whittaker17 and Benjamin Isaac,18 who, without answering all the questions, proposed alternative definitions to the grand strategy theory. Of course, the Romans may indeed have sometimes conceptualised their external frontiers in the form of linear fortifications,19 but this is not the precise meaning of the word limes.

To summarise the details of the model proposed by the second of these historians, given that it has the advantage of proposing a clear chronology of limes in its military definition (because it is a polysemous word): in the 1st century 2 it designated a road made by and for the army, like those built during the Germanic campaigns under Augustus; between the end of the 1st and the 3rd centuries, the same word came to correspond to a provincial land border, without, however, referring to any military structure whatsoever; from the 4th century, it finally became the name of a border district, placed under the command of a dux as well as a military bureaucracy, but without ever referring to the defensive system which constituted it.20 This model is not perfect, given that Benjamin Isaac did not use contradictory sources and relied on translations of ancient authors which served his hypothesis, without always returning to the Latin or Greek text,21 but we can still rely on its approximate timeline and conclude that there would therefore be several definitions of limes over time. However, it is not impossible that the original meanings remained while new ones were added, thus making the military limes of Late Antiquity a multifaceted reality.

The limes during the transition to the Middle Ages era

The truly polysemous character of the term limes during the last centuries of Antiquity explains the difficulty in finding definitions that fit perfectly, since, as already mentioned, the specialists who have worked on the issue until now have used the sources from this period in an anachronistic way, in order to fill the gaps in previous sources. Paradoxically, the limes of Late Antiquity, although the best documented, has been the least studied. Indeed, the Later Romans provide us with precise indications of their own perception of the concept, including the evolution of its use, as in the case of the definition given in Justinian’s Institutes, which was published in AD 533, in order to explain the ius postliminium (viz. the legal guarantee that a citizen could return to his original status and rights, after a modification of these rights, generally captivity, under foreign legislation):22

It was said the postliminium is derived from limen (threshold) and post (after), which explains why we say that the person who has been captured by the enemy and has come back into our territories has returned by postliminium. For just as the threshold forms the boundary of houses (domibus finis), so the ancients represented the boundary of the Empire (imperii finis) as a threshold. And this is also the origin of the term limes, signifying a kind of finis and terminus. Thus postliminium means that the captive returns by the same threshold at which he was lost. A captive who is recovered after a victory over the enemy is deemed to have returned by postliminium.

For the description of the ius postliminium itself, the author(s) of the Institutes relied on the ancient jurist,23 and for the etymological part on the earlier ‘gromatic’ literature.24 According to this book of the Corpus juris civilis, ‘limes’ would thus be etymologically close to ‘limen’ (the Latin word corresponding to the threshold of a door, therefore the passage from one space to another, which can be opened or closed), while referring at the same time to the notion of finis (theoretical control zone and/or contact zone) and terminus (real boundary). We can therefore deduce that the Late Antique limes, in its administrative and defensive dimension, referred to both a cross-border district under military jurisdiction, serving as a buffer zone between the provincial world in a strict sense and the territories totally outside the legal influence of the imperium, as well as the monitored crossing points used to control the roads in this district, but without ever designating the defensive infrastructure which accompanies them or the road network itself.

We cannot be absolutely certain, because this would once again amount to transposing a later reality onto the past, but there is a strong presumption that this Late Antique limes, which should be understood as a transit area from one political reality to another, is the direct ancestor of a structure well known in the Middle Ages: the march. Indeed, the terms limes and marc(h)(i)a (which is a Germanism) are synonyms in early mediaeval sources, while the duces are still then the military authorities of these districts, which meets the positions of Charles Richard Whittaker.25 Moreover, who other than the duces could take care of the management of the Late Antique defensive devices and systems beyond the external provincial borders (like in the eastern Pannonian plain or in Transdanubian Dacia)? It is therefore not so incoherent to think of border districts, limites, which would encompass both sides of this border.

Regarding the multiplication of limites in Northern Africa as toponyms,26 we should probably simply see them as so many small military districts, independent of each other, but operating in a network, along the roads, according to water points, in a hostile desert environment. On this point, the Later Roman occupation certainly did not evolve much compared to that of the end of the Republic and the Early Empire in the region, other than by disseminating the notion of limes to qualify all the outposts concerned.27 In addition to taking into consideration the chronology, we cannot ignore provincial, regional, or even local characteristics. There was not a uniform Roman imperial limes, but a lot of very different Roman imperial limites!

The plan of this special issue and acknowledgements

The editors are convinced that the programme of this special issue of the HiMA journal will contribute to the case, by helping to define the Roman imperial frontiers in time and space, by approaching them from the point of view of their transformation, or even of their disappearance, by following this plan (omitting the editors’ contribution):

• the failure of the conquest of Germania Magna and the creation of the frontier in Lower Germany: Erik Graafstal;
• the establishment of the Litus Saxonicum: Sofie Vanhoutte and Tony Wilmott;
• the consolidation of the Rhine frontier in Late Antiquity: Marion Brüggler;
• the end of the frontier in Upper Germany and in Raetia: Markus Scholz and Christof Flügel;
• the evacuation of Dacia Trajana/Felix and the creation of the frontier in the Lower Danube: Florian Matei-Popescu and Ioan C. Opriș;
• the transformation of the border area in Mesopotamia, with regard to the economic situation: Rocco Palermo;
• the creation of the Claustra Alpium Iuliarum: Josip Višnjić;
• the Anastasian Wall and other Early Byzantine and Bulgarian defence systems in the Balkans: James Crow;
• the transformation of the eastern frontier, with regards to the Islamic conquest: John Haldon.

In addition to thanking the authors, I would like to express, together with Rob Collins, our sincere gratitude to the editorial committee of the journal, in particular to Jean-Christophe Couvenhes and Giusto Traina, as well as the Presses universitaires de Franche-Comté for having enthusiastically accepted the project and for their support throughout the process. They also thank the University of Lille, the HARTIS-UMR 9028 research centre (Univ. Lille, CNRS, French Ministry of Culture, Inrap), and Newcastle University for their financial support, without which the issue would never have seen the light of day, at least not in this format.