Roman Warlords and the Early Medieval World

Please download to get full document.

View again

of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Information Report

Arts & Culture


Views: 0 | Pages: 14

Extension: PDF | Download: 0

This is the non peer reviewed version of the paper published as Carr, D. (2019) 'Roman Warlords and the Early Medieval World', in Christie, H. and Kasten, M. (eds.) Current Approaches to People Places and Things in the Early Medieval Period;
  1 Roman Warlords and the Early Medieval World  Douglas Carr  Newcastle University Abstract: Societies of the early medieval west are often characterised by being pervasively militarised, traditionally seen in opposition to the Roman Empire’s separation of civilian and military spheres. This has led to the view that Roman civilian provincial elites were incapable of actively partaking in the creation of the early medieval world. In this paper, I present an interpretative framework built around the concept of warlordism and an archaeological case study as evidence for the potential of these elites to actively resist or support the transformation that turned the Western Roman Empire into early medieval Europe.  Keywords: Warlordism; Elites; Roman-Medieval Transition; Britain  Introduction: This paper brings together both the work that I presented at EMASS 2018 and further research that came out of that work. The aim of this paper is to situate the civilian Roman elite in their proper place as every bit the equal in terms of violence as the early medieval ‘superthugs’ described by Richard Reece (2000: 3). This paper consists of three parts: firstly, an examination of the current place of warlordism within Roman studies and what is meant by the terms warlord and warlordism; secondly, I present the case for an interpretative framework of warlordism for Roman civilian elites; and thirdly, I look at an archaeological case study for this interpretative framework of warlords. Despite its Roman focus, this is highly relevant to understanding the beginnings of early medieval Europe, as it removes the impression of the late Roman elite as mere passive bystanders to this transition and places them in their proper place as active agents in this transition. Warlordism in Roman Studies: Previous application of warlordism to late Roman studies has been through the usage of what can be termed the generalissimo model. This model draws upon twentieth century Chinese warlordism, which remains the most well-defined form of warlordism (Hills 1997: 36-37). Sheridan (1966) has defined these Chinese warlords as having exercised independent regional governmental control, maintained through the authority of military command. This is termed the generalissimo model, as it features military commanders using their control of violence to achieve political objectives (Pye 1971: 8). Chinese warlordism is interpreted as the result of a terminally weak state unable to prevent independent action by its servants (MacKinlay 2000: 53), and explanatorily functions as a bridge between a previously unified monolithic society and a diverse and disunited one (Pye 1971: 8). Both MacGeorge (2002) and Wijnendaele (2016 and 2017) have discussed the actions of specific figures in the Late Roman West through the generalissimo model. Both authors (MacGeorge 2002: 33 and Wijnendaele 2017: 430-33) have cited the relinquishment of de facto  but not de jure  command by the Emperor to powerful generals, who could no longer be simply dismissed, as key to the development of this model. Wijnendaele (2017: 437-40) cites the behaviour of Aegidius, Marcellinus and Ricimer as examples of this process, their army’s loyalty to their commander provided the independent powerbase these generals required. The important distinction between these warlords and the usurpers of the third century is that these generals did not seek to acquire the imperial throne. Beyond that limited to Chinese warlordism discussed above, there has been little critical engagement with what a warlord is. The terms warlord and warlordism have however found usage amongst many authors {1}, but there has been a lack of critical engagement with what the terms mean. This point should not be construed as any disagreement on my part with the meaning those authors have intended, merely that these complex terms require  2 proper explanation. The work of Collins’ (2012) provides an example whereby the transformation of frontier communities along Hadrian’s Wall during the fifth century is characterised as warlordism and understood archaeologically but does not engage with the literature surrounding warlordism. At best as in the case of MacGeorge (2002) and Wijnendaele (2016 and 2017) warlordism provides a valuable tool in studying the political history of the late Roman Empire, at worst it provides merely a pejorative term with which to describe complex and largely elusive figures of the late and immediately post-Roman period. Although rightly caution must be exercised when applying any modern conception of warlordism to the ancient world (Wijnendaele 2016: 187-78), a critical understanding of the terminology allows an appropriate description of figures within the late Roman and early medieval worlds. This does then leave us with the question, what exactly is a warlord? Warlordism beyond the generalissimo model: The terms warlord and warlordism are somewhat vague (Freeman 2015: 791 and Ahram and King 2012: 170). In an attempt to reach an understanding of these terms, it is necessary to look outside both archaeology and history. Various definitions that have been offered include: •   Armed groups using violence and economic exploitation to rule independent fiefdoms (Vinci 2007) •   Autocratic authorities with power rooted in a local monopoly on violence (McCormick and Fritz 2009: 83) •   Personality-based, strongman-centred, patron-client networks that challenge the development of formal centralised state authority (Wang 2008: 174) The variety in definition reflects the variety in usage of the terms warlord and warlordism. The first English usage is by Emerson (1856: 176), to describe the transition of the basis of aristocratic power from military to political or economic strength. The term was then subsequently used, as discussed above, to describe the independent military commanders active in twentieth century China. Between the 1960s and 1990s warlord grew to encompass a wide array of sub-state actors, ranging from insurgents to criminals (Nourzhanov 2005: 109-10). The most comprehensive attempt at a definition has been that of Marten (2007: 48), who identified four shared characteristics: 1.   Armed men to seize control over small areas in the absence of central authority 2.   Motivation rooted in self-interest rather than ideology 3.   Authority based on personal charisma and ties of patronage 4.   Personal rule resulting in the localisation of the economy Warlord is a deeply pejorative term, evoking brutality and criminality (MacKinlay 2000: 48) and considered a deeply archaic form of societal organisation (Blair and Kalmanovitz 2016: 430). Warlords then are taken to represent the archetypal illegitimate actor, but this is entirely reliant on a paradigm where the state is assumed as the dominant and only legitimate societal form (Freeman 2015: 791 and Goetze 2016: 138). The warlord is reduced to a mere hindrance to the centralised authority’s presumed legitimate, legal and benevolent monopoly on violence (MacKinlay 2000: 53 and Ahram and King 2012: 171). It is unhelpful to understand warlords in such a limited and negative way, as like any form of social organisation they are neither inherently positive or negative (Freeman 2015: 802). Although warlords do operate autonomously this does not entail either an antagonistic or a non-existent relationship with central authority (Freeman 2015: 796), as warlords are differentiated from revolutionaries as they seek to acquire status within a system rather than alter that system (Ahram and King 2012: 172). Warlords have value to the state as they can mediate between a central authority and local actors, this sub-contraction of power by central authority representing an admission that local magnates can more effectively exert control (Ahram and King 2012: 173-76). As such warlords operate in regions where political power has devolved from the centre to the periphery, gaining local control and keeping centralised authority at bay (Blair and Kalmanovitz 2016: 428). Warlords do also fulfil state-like functions by providing order (Blair and Kalmanovitz 2016: 429), often in a way that matches local requirements in a way the central state cannot (Ahram and King 2012: 173). This should not be understood as mere altruism, as a warlord must enjoy a degree of legitimacy in the eyes of the community they claim to lead (Nourzhanov 2005: 110). Warlordism can be understood less as an  3 exploitation of chaos and more as a local societal effort to restore stability in the midst of anarchy (Jackson 2003: 147). Local Roman Elites as Warlords: The story of Ecdicius (Sidonius The Letters  3.3) provides a good starting point; Ecdicius a Romano-Gallic landowner raised a private army to combat Gothic warriors besieging  Augustonemetum  and defeated them. Whilst the actions of Ecdicius could be dismissed as a singular event in an unusual situation, this would be to ignore an important element in the experience of civilian Roman elites. The importance of the story of Ecdicius is that a landowner raised a private army and defended a local territory outside of the formal structures of the Roman world. Whittaker and Garnsey (1997: 311) saw warlordism developing in the Late Roman West as a result of the weakness of imperial governance and the growth of rural patronage. Whilst I would agree with the underlying causes of late Roman warlordism Whittaker and Garnsey (1997) outlined, these are not features exclusive the late Roman period. The Roman Empire was a ‘truly minimal state’ (Bang 2007: 13), local magnates supplementing patronage with violence in order to bolster their power was not a result of declining imperial control but rather a long-term feature of a state largely uninterested in protecting the majority of its subjects. The smaller scale militarised society of the early medieval period (James 1997: 19) I would argue is an organic development from the Late Roman West rather than an alien system imposed by incomers. By the seventh century the elites of Western Europe were no longer supplying wealth to a central authority so that authority could employ armed men but were instead directly supplying those armed men (Heather 2001: 441). Local Roman elites were integral to the administration of the provinces they resided in (Greatrex 2015: 36), a fundamental feature of Roman administration (Bang 2007: 27). These elites drew their legitimacy from their local power (Vanderspoel 2009: 432) and their wealth from control of land and its produce (Bang 2007: 25 and Brown 2012: 3). It is unsurprising that Van Dam (1985: 14) described Romano-Gallic society as composed of local tyrannies ruled by those with local authority, control of land, surrounded by armed retainers and upon whom the majority of the population were dependent. Van Dam’s characterisation of Romano-Gallic society is very close to the definitions of warlordism discussed above. Dependence upon elites was largely expressed through patronage, a practice ubiquitous in the Roman world. At best, patronage provided a safeguard against economic or social upheaval (Garnsey and Woolf 1989: 154). Patronage refers to a powerful individual providing care and protection over a subordinate in return for their acceptance of the patron’s jurisdiction and providing physical or material support when required (Whittaker and Garnsey 1997: 293). Local elites exploited state and local systems of patronage in order to bolster their position (Hopwood 1989: 184), local grievances could only be aired outside these small communities with the support of the right patron (Kelly 1997: 156). Patronage of course was not only a vertical practice, the demonstration of status and the creation or renewal of alliances horizontally amongst elites through shared social activities is also important (Ellis 1991: 199 and Mathisen 1993: 14). Maintenance of law and order has usually been assumed to be the responsibility of the Roman army (Whitby 2001: 477), this obscures the far greater role played by local authorities (Fuhrmann 2016: 300). Policing was the responsibility of a local magistrate, such as the  praefectus arcednis latronciniis  of  Noviodunum  ( CIL  13.5010), and therefore the local elite (Brown 2012: 4-5). Whilst urban centres may have had some designated enforcers (Shaw 1993: 319), ad-hoc posses likely played the greatest role (Fuhrmann 2012: 52). The army would only intervene in truly exceptional circumstances (Carrié 2005: 287) and for day-to-day suppression of banditry it was the armed retainers of local elites that the Roman state relied on (Bachrach 1972: 490). Defence against banditry was enshrined in Roman law as a legitimate reason for self-defence ( Codex Theodosianus  7.8.14 and 9.14.2). The Roman legal system emphasised self-help, including the responsibility of the plaintiff to produce a defendant, by force if necessary (Harries 2007: 106). Add this emphasis on self-help to a lack of law enforcement (Shaw 1993: 308) and the only force available to fill this gap were private retainers (Fuhrmann 2016: 297). Sidonius ( The Letters  3.12.1-3) himself summarily seized and tortured those he finds desecrating his grandfather’s grave. Banditry, at a low-level, was endemic throughout Roman history (Whittaker and Garnsey 1997: 308). The term bandit included not just highwaymen, but also private feuding, urban riots and the activities of usurpers (Shaw 1993: 307-09), any group using illegitimate violence risked acquiring the label of latrones  (Fagan 2011: 477-78). Despite the widespread nature of banditry, the breadth of the Empire meant that central authority could do little and was a rare site, with local arrangements predominating (Shaw 1993: 308). Travellers simply defended themselves, Galen (  De Anatomicis Administrationibus  1.2.221-22) dismisses a roadside corpse as simply a  4 brigand killed in self-defence, even the heavily taxed merchants of Palmyra were responsible for their own protection (Bang 2007: 13-14). The relationship between local elites and bandits was occasionally less clear cut than that envisaged in law, elites could protect bandits (Hopwood 1989: 182) and recruit them as instruments of private violence (Shaw 1993: 324). The practice of local elites protecting bandits apparently was widespread enough to necessitate specific laws ( Codex Theodosianus  9.29.1-2 and Codex Justinianus  9.39.2). There are several other reasons why Roman elites might have chosen to surround themselves with armed retainers. In the late Roman period there existed great and scattered concentrations of land (Esmonde Cleary 2013: 436). Perhaps one of the best-known examples are the estates of Melania the Younger whose lands scattered across the Empire yielded an annual income of 120,000 solidi ( The Life of Melania the Younger   10-15). Such a situation does beg that question of how the income raised from those lands was transported and put at the disposal of their owner. In a world, as already discussed, rife with banditry the movement of any wealth must have required trusted and armed servants. Additionally rents, whether paid in cash or kind are ultimately enforced by the threat of unpleasant consequences (Bang 2007: 39-40), consequences a landowner would have to enforce. Slavery definitely existed in the late Roman period, but its exact character is difficult to discern (Whittaker and Garnsey 1997: 295). Regardless slavery as a system does require the employment of individuals willing to inflict horrendous violence in order to keep people in servitude. A case from fourth century  Hispania  aptly demonstrates this, two slaves were beaten to death on the orders of their mistress (Brown 2012: 61), and such an action signifies the existence of employees willing to inflict such violence at the behest of their employer. The role of the forces of civilian elites in usurpations may also be an overlooked phenomenon. Perhaps one of the most well-known examples of this would be the resistance mounted against Constantine III in  Hispania  by cousins of the Emperor Honorius, who raised an army from amongst their tenants and slaves (Zosimus  Nova Historia  6.4.3). In the work of Gildas it is possible that there is another example, it is stated that Magnus Maximus took all Britain’s ‘soldiery and armed bands’ (  De Exidio Britanniae  2.14), perhaps referencing the usage of elites’ armed retainers to bolster the forces available to a usurper. Now I believe the case has been established that local civilian Roman elites required armed retainers, they question is did they possess them? Two things would have been required: 1.   Arms – encompassing a great variety of artefacts from the sword to the wood axe 2.   Retainers – that is followers capable of and willing to inflict violence To start with arms, there is a long-standing view that inhabitants of the Roman Empire outside the military or civil service were prohibited from possessing arms. This view is entirely false and was categorically refuted by Brunt (1975). The restrictions often quoted from the  Digest   (48.6.1-12) do not prohibit the possession of arms, merely they restrict the quantities and usages of those arms, but the use of arms for self-defence is both expected and condoned within the  Digest  . A prohibition on arms was likely a local peculiarity (Hopwood 1989: 179); whilst it would have been usual to wander a city visibly armed to travel through rural areas unarmed was deeply unwise (Lintott 1968: 23 and Fuhrman 2016: 297). Hunting was highly popular amongst late Roman elites (Esmonde Cleary 2013: 242), evident from the profusion of domestic mosaics featuring hunting scenes (Ellis 1991: 124). Hunting provides an entirely reasonable explanation for the possession of arms and the existence of a group of armed specialists surrounding elite figures (Drinkwater 2001: 143). Of course, the tools of the huntsman could easily be turned against intransigent tenants or local bandits if required. Retinues of private armed groups in the Roman world are well attested and necessitated by conditions of generalised violence (Whittaker 1993a: 137). The size of these retinues is impossible to determine, but law codes of the successor kingdoms may give some indication and reference bands of up to five men (  Rothairs  Edict   19 and The Laws of King Ratchis  2.10.6), similar in number to the six Sarmatians Theophanes employed as escorts (Matthews 2006: 50). The Germanic model of retinues is the most familiar with the magnate exercising control over their kin group and calling upon their family and clients to provide violent services (Gil Egea 2003: 493), but parallel developments within the Roman world demonstrate it as a Roman concept too (Whittaker 1993b: 290). The term bucellarii  although srcinally referring to an elite cavalry unit, comes to be applied to describe troops recruited and upkept privately by military commanders (Liebeschuetz 1986: 468). It appears that landowners began to recruit their own bucellarii  despite official threats of punishment (Schmitt 1994: 167). Gascou (1976) goes too far in seeing in bucellarii  the beginnings of feudalism, but their employment by landowners represents an increasingly blurred division between the forces of the state and the forces of powerful local magnates.  5 The demilitarisation of late Roman local elites has been vastly overstated (Halsall 2007: 495). The Roman world was characterised by widespread violence, both privately and state orchestrated (Garnsey and Woolf 1989), with landowners large and small using violence to bolster their economic and social position. Emperors were not omnipotent (Kelly 1997: 157), their desire to regulate the activities of local elites and their ability, let alone desire, to enforce a total monopoly on violence is debateable, as long as taxes arrived, and order was maintained elites were granted an almost entirely free hand (Shaw 1993: 317). The definitions of warlordism provided by Marten (2007), Vinci (2007), Wang (2008) and McCormick and Fritz (2009) discussed above fit with the elements of Roman local elite life I have outlined here. Examples from the literary sources such as Ecdicius could be written off as isolated examples but in reality, reflect only those whose deeds were recorded (Liebeschuetz 2007: 489). Considering local late Roman elites as warlords is essential to dispelling the image of them as mere passive bystanders to the end of the Roman world. Case Study: Militarisation and Roman Britain The section above outlined an interpretative perspective for the late Roman elite but is based largely on textual evidence levied from throughout the Roman world. To study this archaeologically however, a process must be identified that represents a visible manifestation of aspects of Roman civilian elites that allow their characterisation as warlords. This process is militarisation, defined by Esmonde Cleary (2013: 60-90) as the adoption of military symbols by the wider population and the expression of elite power through a military vocabulary at the expense of traditional means. Roman Britain has been chosen as a case study for the application of the process as it lacks the textual evidence for figures such as Ecdicius. Burial practice formed a large part of the evidence mobilised for Northern Gaul, the well-known burials containing weaponry and dress fittings previously interpreted as Germanic warriors, Esmonde Cleary (2013) convincingly argues these were merely a minority variation on Gallo-Roman provincial burial practice. In Roman Britain, there are few burials containing supposedly official dress accessories and only two burials possibly contain weapons {2}. The weapon burials are both from Dyke Hills, a disturbed grave with an axe (Booth 2014) is the most convincing but as Theuws (2009) has highlighted axes are in no way distinctly military. The second burial is associated with some possible iron objects disposed of by the discoverer in a river (Kirk and Leeds 1953: 67), some have chosen to view these as weapons. Burials with official dress accessories are more common than those with weapons but still far rarer than on the continent (Collins 2017: 31). The lack of burial evidence requires a more thorough examination of the distribution and context of finds of official dress accessories in Roman Britain to find evidence of militarisation. Two artefact types are examined here, Hawkes and Dunning belt fittings and crossbow brooches. The late Roman zoomorphic belt fittings found in Britain were first seriously worked on by Hawkes and Dunning (1961), since then a far greater number have become known and a total of 430 were assembled for this research. These 430 examples consist of 237 Portable Antiquities Scheme finds and 193 from excavations that can as a minimum be assigned to a particular site. The distribution of these artefacts (Fig. 1) shows a distribution largely away from the major garrisons of  Britannia , Hadrian’s Wall and the Saxon Shore forts. The sites that the excavated finds come from equally do not reflect a military interpretation (Fig. 2), with most coming from either urban or rural contexts. The military finds are disproportionally represented by finds from Richborough, 62.5 per cent of the military finds are from this one site. The rural finds (Fig. 3) are largely drawn from either nucleated settlements or isolated farmsteads or villas; hardly where the garrison of  Britannia  would be expected to be found. A further 27 finds come from Roman burials, representing 11 individuals. A further 35 examples come from Early Medieval contexts, 29 of these from burials and six from non-burial contexts {3}. Neither the distribution nor contextual information for the vast majority of these belt fittings demonstrates support for the suggestion that these are entirely military dress accessories. The assumption that they are has led to the invention of barbarian field armies (Hawkes 1974) and semi-official vicarius  (Leahy 2007) or tribal (Laycock 2008) militias. Any such interpretation means accepting the idea that the garrison of  Britannia  were dispersed throughout the villas and farms of the diocese . Some types are without a doubt official, such as Type IVA, but these represent a truly tiny proportion of the dataset {4}. As the evidence does not reflect a military interpretation, I will suggest that the interpretative framework I have presented for late Roman civilian elites above does provide an explanation. These belt fittings are clearly emulating late Roman official dress, without directly copying it, and cannot be directly linked to any official group. Instead, what we can observe is the appearance of military style belts that are appropriating the symbolic language of military authority in late
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!