Fisher G. Between Empires. Arabs, Romans, and Sasanians in Late Antiquity

Подождите немного. Документ загружается.

a nomadic lifestyle, with its attendant inevitable connotations of

barbarism which traced their roots to as far back as the Epic of

Gilgamesh. The name ‘Arab’ could also be used as an ethnic label to

refer to the inhabitants of ‘Arab’ polities or principalities such as the

Nabataean kingdom, or simply to describe those who originated from

the place commonly and loosely known then, as now, as ‘Arabia’.In

the third century

AD, the terms ‘Saracen’ and ‘t



ayyāyē’ superseded

‘Arab’ to become the common broad terms used to refer to the Arab

allies of the Romans and Sasanians. Today, the name Arab describes,

for many people, Arabic-speaking Muslims inhabiting the nations of

the Middle East. All of these terms—tribe, Arab, Saracen, t



ayyāyē—

are of variable meaning and I will discuss them in more detail

throughout. At the outset, however, it is useful to be clear about the

frequent appearance of the word ‘Arab’ in this study. I use it as the

most common and convenient label to refer to those peoples from

the general geographical region of the Arabian peninsula, including

the frontier regions bordering on northern Arabia, whom the soldiers,

writers, monks, priests, and others of the Roman Empire encountered

and described as such, without any desire to express a particular

opinion on ethnicity, nationality, or identity.

A number of Arab tribal groupings or individual leaders came

under Roman or Sasanian pressure in Late Antiquity, including the

Jafnids, the Nas



rids, and larger, less precisely deﬁned groups such as

Maʿadd, Mud

ar, Tanūkh or Salīh

. The sources for these groups vary

in their quantity, quality, and reliability. The most prominent of the

Arabs in contact with the Romans in the sixth century were the

Jafnids, for whom we possess the most abundant source material; to

the east, the Nas



rids had long been allies of Sasanian Iran, and, in the

sixth century, the H

ujrids, Ma‘add

, and others came to the attention

of Roman writers as a result of Constantinople’s diplomatic adven-

tures in the Arabian peninsula. The H

ujrids were, during the same

period, also clients of the H

imyarite kingdom, and Maʿadd also seems

to have been contested by the H

imyarites, the Romans, and the

Nas



rids at different times. Numerous named and unnamed Arab

ﬁgures appear in Byzantine sources, whose afﬁliation to a particular

group or tribe, if any, is unknown. While I consider a wide range of

evidence here for the various Arabs who appear in late antique source

However, Ma‘add may refer to a ‘type’ of culture, rather than a speciﬁc group of

people; for this unresolved problem, see Ch. 3.

2 Introduction

material, the main interest of this study is in the Jafnids and the

Nas



rids, both of whom found themselves pressured to a signiﬁcant

extent by the wide-ranging effects of the ongoing competition

between the Romans and Sasanian Iran.

JAFNID OR GHASSĀN: ARAB ELITES

The names Ghassān, Lakhm, and Kinda have become closely asso-

ciated with the individuals who make up the Arab dynasties known as

the Jafnids, Nas



rids, and H

ujrids in modern scholarship.

This ter-

minology—Ghassān for Jafnid, for example —can be misleading, and

it is perhaps more desirable to prefer the terms used to refer to the

dynasties or ‘elites’ themselves, that is, the Jafnids, Nas



rids, and

ujrids, rather than descriptions such as ‘the Ghassānid leader al-

ārith’. Labels such as ‘Jafnid’ are not without their own difﬁculties,

since, like Nas



rid and H

ujrid, the name is derived from a putative

ancestor (‘Jafna’, ‘Nas



r’, ‘H

ujr’) about whom little or anything may

actually be known, and contemporary Graeco-Roman authors do not

talk of ‘the Jafnids’, but prefer to use the names of individuals, such as

al-H

ārith (Arethas) or al-Mundhir (Alamoundaros) for reasons

which are unclear. No matter what label we choose to apply, however,

it is important to make the distinction between elites and the groups

which they led.

First, it is clear that in relation to the empires of Late

Antiquity, at least, power rested with the individuals, and not the

larger groups of people. The Romans, Sasanians, and the leaders of

imyar preferred to create relationships with individuals, not wider

groups of people, and Justinian’s agreement with al-H

ārith the Jafnid,

for example, was a highly personal one. Al-H

ārith obtained a major

proportion of his power from this individual relationship with the

The works of Irfan Shahid have helped to cement this association.

‘Elite’ is best applied here to describe a minority in a position of power, prestige,

prominence, or leadership such as that held by the Jafnids. See, for further discussion

and deﬁnitions, J. Haldon and L. I. Conrad (eds.), The Byzantine and Early Islamic

Near East, vi, Elites Old and New in the Early Islamic Near East (Princeton, 2004), esp.

J. Haldon, ‘Introduction: elites old and new in the Byzantine and early Islamic Near

East’,1–12; M. R. Salzman and C. Rapp (eds.), Elites in Late Antiquity, Arethusa, 33/3

(2000), 315–468; C. Wickham, Framing the Early Middle Ages: Europe and the

Mediterranean 400–800 (Oxford, 2005), 153–258.

Introduction 3

head of the Roman Empire—a position which he had presumably

derived initially from his earlier ability to win conﬂicts, and protect

water sources, pasturage, or other objects vital to the success of those

around him, as well as fend off contenders for his own position. The

ujrid leaders in northern Arabia and the Nas



rids within the Sasa-

nian Empire enjoyed similar links with the H

imyarites and Sasanians.

On the basis of our evidence, it was these individual elite leaders who

created personal agreements with states and empires, and who acted

as the ‘brokers’ or conduits between the tribe and the state.

A second reason to discuss elites separately from groups concerns

questions of origin and leadership. To talk of ‘the Ghassānid leader al-

ārith’ supposes that al-H

ārith led Ghassān in the sixth century, but

in actual fact hardly anything is known about Ghassān’s whereabouts

or composition.

Ghassān is described in an inscription from south

Arabia, and the name, or one like it, appears in a letter written by

Simeon of Beth Arshām and perhaps also again in relation to a

monastery in the sixth-century Letter of the Archimandrites.

Other

than these examples, there is no contemporary information for

Ghassān and it is not clear if the Jafnid leaders represented by al-

ārith and his descendants, who are generally associated with Ghas-

sān, maintained their leadership over it in the sixth century. An

analogy can be drawn with Shammar, a federated group of tribes

active in Mesopotamia between the seventeenth and nineteenth cen-

turies. In the seventeenth century the chief of Shammar took them

out of Najd, in central Arabia, and in the nineteenth his descendants

ruled a confederation of Shammar and other allied tribes in Meso-

potamia. Some of Shammar, in the meantime, had stayed in Najd,

and the Shammar leaders in Mesopotamia effectively ruled a

Cf. comments made by Beck about the Qashqai, a tribal group in modern Iran:

‘Discussion of “the Qashqai” as a political force ought more properly to read, “the

Qashqai khans”, for they were the power brokers who dealt directly with other powers

in the region. Much “Qashqai” history is therefore a history of the Qashqai khans.’ See

L. Beck, ‘Iran and the Qashqai tribal confederacy’, in R. Tapper (ed.), The Conﬂict of

Tribe and State in Iran and Afghanistan (London, 1983), 284–313, at 305.

This issue is discussed in more detail in Ch. 3. Most recently, Christian Robin has

argued for the separation of Jafnid and Ghassānin‘Les Arabes de H

imyar, des

«Romains» et des Perses (III

–VI

siècles de l’ère chrétienne)’, Semitica et Classica,1

(2008), 167–208, and previously advanced a similar argument for Kinda/H

ujrids in ‘Le

royaume H

ujride, dit «royaume de Kinda», entre H

imyar et Byzance’, CRAI (1996),

665–714.

Ch. 2, below.

4 Introduction

confederation where Shammar was, in fact, the minority.

Whether

ornotthissuggestsanaccuratepictureofGhassān—whereby the

Jafnids and parts of Ghassān travelled north and enrolled the sup-

port of new tribes, whilst a signiﬁcant number stayed behind in

south Arabia—it does demonstrate why scholars should be careful

about attaching the names of the Jafnids to GhassānortheNas



rids

to Lakhm, because this can lead to the erroneous impression that

they were the leaders of politically cohesive ‘kingdoms’ or ‘states’.As

this study shows, it is clear that the power rested with the individual,

not the group. Furthermore, while the natur e of the Jafnids’ power

and their political identity developed throughout the sixth century,

and would go on to exhibit some features we might normally

associate with states, it is clear that the people under their leadership

did not constitute a state, and any development trending in this

direction was capped by the realities of Roman power.

In this

study, I use the terms Jafnid, Nas



rid, and H

ujrid to describe the

groups of elite individuals in leadership positions described by

Graeco-Roman authors, and thus to make the distinction between

them and whichever people lay under their control.

How can we examine these elites? As with the Germanic peoples in

the west, analysing the Jafnids or Nas



rids is a challenge, since in-

formation about them tends to be scarce, difﬁcult to assess, or

unreliable before such people come to the attention of contemporary

authors within the Roman Empire. Even then, ethnographic or lit-

erary convention could inﬂuence which of their activities were de-

scribed by Roman authors. As a result, there is very little information

on the origins of the ruling dynasties and the exact means by which

they came to power. Complications arise from these difﬁculties since

it is only through the activities of the elites that we can begin to learn

anything about the otherwise anonymous groups of people under

their control—whether these were Ghassān, Lakhm, or any number

of others—who are rarely, if ever, discussed by our sources.

Reliable information on the Nas



rids is particularly sparse. A certain

ʿAmr, ‘king of Lakhm’, appears on the late third-century Paikuli

inscription from Kurdistan as an ally or vassal of the Sasanian

R. Hoyland, ‘Late Roman Provincia Arabia, Monophysite Monks and Arab tribes:

a problem of centre and periphery’, Semitica et Classica, 2 (2009), 117–39, at 118.

Robin, ‘Les Arabes de H

imyar’, 193, applies the term ‘principauté’, which evokes

the power of the ‘prince’ or individual, in contrast to that of the state.

Introduction 5

Emperor Narseh (293–302), providing some clues to the longevity of

the relationship between the Nas



rids and Iran.

The nineteenth-

century German scholar Gustav Rothstein attempted to reconstruct

the history of the Nas



rid dynasty based on Muslim Arab sources, but

his version remains speculative because of the lack of a consistent

body of contemporary supporting source material.

The most active

leader of the Nas



rids was al-Mundhir (?505–554), who makes occa-

sional appearances in Roman sources as a result of his enmity with

the Jafnids in the mid-sixth century. Menander describes the efforts

of al-Mundhir’s successor, ʿAmr, in diplomatic initiatives with the

Romans around the time of the treaty of 561/2; the last Nas



rid ruler at

al-H

īrah, al-Nu‘mān (583–c.602) appears in contemporary ecclesias-

tical sources, due to his late ‘conversion’ to Christianity, but beyond

this, there is little other information. For the H

ujrids the main source

of information comprises a small number of south Arabian inscrip-

tions which have allowed scholars to reconstruct some aspects of their

relationship with their patrons, the kingdom of H

imyar. Through

additional brief mentions in late Roman sources, we are given a

glimpse into their diplomatic liaisons with Constantinople and their

involvement with Maʿadd.

The Jafnids are relatively well-documented in comparison to the

Nas



rids or the H

ujrids. Nevertheless, the early history of the interac-

tion between the Jafnids and the Roman Empire is poorly understood.

Ghassān, from whom the Jafnids are believed to have originated, were

active in south Arabia in the third century; an inscription from this

region, dated to approximately

AD 260, refers to the ‘kings of the

peoples of Ghassān’, alongside those of three other groups.

If the

Jafnids indeed originated in the southern part of the Arabian penin-

sula, the most straightforward explanation for the ways in which they

came into contact with Rome is some form of migration. This might

not necessarily have been because of a wholesale movement or as a

H. Humbach and P. O. Skaervo, The Sassanian Inscription of Paikuli iii/1

(Wiesbaden, 1983), 71, paragraph 91.

G. Rothstein, Die Dynastie der Lah

miden in al-H

ira. Ein Versuch zur arabisch-

persischen Geschichte zur Zeit der Sasaniden (Berlin, 1899), 60–9 (generally), 70–125

(for speciﬁc Nas



rid rulers).

R. G. Hoyland, ‘Arab kings, Arab tribes, Arabic texts and the beginnings of

(Muslim) Arab historical memory in late Roman inscriptions’, in H. Cotton,

R. Hoyland, J. Price, and D. Wasserstein (eds.), From Hellenism to Islam: Cultural

and Linguistic Change in the Roman Near East (Cambridge, 2009), 374–400, at 376.

6 Introduction

result of a major event, but could also have occurred as part of the

natural movement of peoples who probably depended on seasonal

resources, such as oases and pasturage, for some of their sustenance

and income. Hoyland suggests that alongside the traditions of pre-

Islamic tribal movements there are numerous indications (for example,

from changes in the names appearing in ancient north Arabian grafﬁti)

that these migration stories, the products of much later writers, may

reﬂect a partial truth.

We can only speculate on the activities of the

Jafnids before 500, but around this date the consolidation of power

around the leading members of the family began to take shape, with

support from Rome. Theophanes reports that an individual named

Jabala was worsted in an encounter with Roman forces, after which

there was peace and, a generation later, we read in the histories of

Procopius about the elevation of his son, al-H

ārith, under the Emperor

Justinian.

Over time, a signiﬁcant degree of political power crystal-

lised around the Jafnids. After al-H

ārith, who ruled or led between

approximately 529 and 569, his son al-Mundhir took over, until his

arrest and exile in 582. He was succeeded by al-Nuʿmān, one of his

sons, whose exact fate is unknown. Whatever the precise reasons for

the movements of Arabs north and into the Fertile Crescent, the case of

the Jafnids shows that strong individuals could and did come to the

attention of Romeand the Sasanians, through their raids in the southern

frontier areas or through their wars with each other which occasionally

prompted the intervention of imperial forces.

RATIONALE AND BACKGROUND

What can be learned from studying these late antique Arab elites? In

the broadest and most obvious sense, they stand as the immediate

antecedents of those who achieved far greater historical prominence

Id., ‘Epigraphy and the emergence of Arab identity’, in P. Sijpesteijn, L. Sundelin,

S. Tovar, and A. Zomeño (eds.), From Andalusia to Khurasan: Documents from the

Medieval Islamic World (Leiden, 2007), 219–42, at 225–6.

Theoph. Chron. 141; al-H

ārith: Proc. BP 1.17.46. Note that Shahid has suggested

that the rogue adventurer Amorkesos (Imruʾ l-Qays), who appears in the Roman

province of Arabia in about 470, may have been a ‘member’ of Ghassān, although this

is very speculative. I. Shahid, Byzantium and the Arabs in the Fifth Century

(Washington, 1989), 61.

Introduction 7

after the advent of Islam. Yet in spite of this the Jafnids, Nas



rids, and

others here are rarely given adequate attention in studies of either the

Muslim Arabs or Roman Late Antiquity.

They sit uneasily in a

conceptual no-man’s land, perceived to ﬁt imperfectly into the

broader topics of late Roman imperial and frontier history or the

history of the Muslims after the seventh-century conquest. As such,

they tend to ﬁnd themselves incorporated into larger works, or simply

ignored altogether. Difﬁcult to trace archaeologically, and with only

occasional appearances in literary sources, it is tempting to dismiss

them as irrelevant or at best tangential to the main historical narrative

of the late antique Near East.

Yet to dismiss the ﬁfth- and sixth-century Arabs in this way is to

give up a promising opportunity to address a set of compelling

historical problems. Even though the source material is generally

poor in quantity and difﬁcult to assess, the Jafnids rank among the

better-documented groups of Arabs, and examining their interaction

with Rome and the ways in which they dealt with aspects of that

relationship provides a glimpse into evolving Arab identities, however

imprecise, before the mid-seventh century. We can use a study of the

Jafnids, complemented by information on the Nas



rids and the H

uj-

rids, to provide a further angle on some pressing questions, most

notably whether or not the Arabs, in any form, had any political

cohesion or political identity before the seventh century. Another

question which can be posed is whether or not they spoke Arabic

and, if so, whether or not this made any difference to their ethnic

identities before the Arabic language became deeply connected, over

time, to Islam and a new type of Arab ruling elite. Questions about

religious, linguistic, ethnic, and cultural identities are a core research

interest in studies of Late Antiquity, form a key background element

to this study, and are developed in detail in Chapter 4.

We can also

Underscored in the introduction to R. G. Hoyland, Arabia and the Arabs: From

the Bronze Age to the Coming of Islam (London, 2001), 1. A good example of this

problem is the recent work of H. Börm, Prokop und die Perser. Untersuchungen zu den

römisch-sasaniden Kontakten in der ausgehenden Spätantike (Stuttgart, 2007), who

notes, 209, that the Arabs are beyond the scope of his study, referring readers to

Shahid instead.

The literature is extensive, but see (in addition to material citied in the notes

below), essays in S. Mitchell and G. Greatrex (eds.), Ethnicity and Culture in Late

Antiquity (London, 2000); R. Miles (ed.), Constructing Identities in Late Antiquity

(London, 1999); essays in G. Clarke and D. Harrison (eds.), Identities in the Eastern

Mediterranean in Antiquity: Proceedings of a Conference held at the Humanities

8 Introduction

use the Jafnids and the Nas



rids to investigate additional facets of

another enduring historical problem, by situating them in the context

of studies which straddle the conceptual division represented by the

Muslim conquests of the 630s. Crone and Cook,

Garth Fowden,

Millar,

Bowersock

and numerous other authors

have all ad-

dressed the persistence of tradition, refashioning of old into new,

and innovation which characterised a large part of the social, political,

cultural, and religious life of the period between the ﬁfth and seventh

centuries. It is both promising and worthwhile to situate the Jafnids

and Nas



rids within this ongoing debate which, itself, is part of the

wider recognition of the permeability of the traditional divisions

between historical periods in both east and west in the ancient

world.

Within works concerned with Arabs in general or with broader

historical themes or ideas, the Jafnids and the Nas



rids play only an

occasional role. Few studies are devoted to them, and it is somewhat

indicative of the difﬁculty and scarcity of the sources for the Nas



rids

Research Centre in Canberra 10–12 November 1997, Mediterranean Archaeology,11

(1999); F. Millar, A Greek Roman Empire: Power and Belief under Theodosius II (408–

450) (Berkeley, 2006), which deals extensively with the question of regional identities

in the ﬁfth-century Near East, at 84–129; also essays in E. Digeser and R. M. Frakes

(eds.), Religious Identity in Late Antiquity (Toronto, 2006). Most recently, H. Cotton

et al. (eds.), From Hellenism to Islam.

P. Crone and M. Cook, Hagarism: The Making of the Islamic World (Cambridge,

1977).

G. Fowden, Qusayr ‘Amra: Art and the Umayyad Elite in Late Antique Syria

(Berkeley, 2004).

F. Millar, ‘The Theodosian Empire (408–450) and the Arabs: Saracens or

Ishmaelites?’, in E. Gruen (ed.), Cultural Borrowings and Ethnic Appropriations in

Antiquity (Stuttgart, 2005), 297–314.

G. W. Bowersock, Hellenism in Late Antiquity (Cambridge, 1990), esp. 75–6;

most recently, id, Mosaics as History: The Near East from Late Antiquity to Islam

(Cambridge, Mass., 2006).

Brieﬂy, see, for example: on Islam, F. M. Donner, ‘The Background to Islam’,in

M. Maas (ed.), The Cambridge Companion to the Age of Justinian (Cambridge, 2005),

510–34; on material culture, A. Walmsley, Early Islamic Syria: An Archaeological

Assessment (London, 2007); on Christianity, R. Schick, The Christian Communities of

Palestine from Byzantine to Islamic Rule (Princeton, 1995).

Inﬂuentially, P. Brown, The World of Late Antiquity: From Marcus Aurelius to

Muhammad (London, 1971); id., The Rise of Western Christendom: Triumph and

Diversity,

AD 200–1000 (2nd edn, Oxford, 2003); collected essays in G. W. Bowersock,

P. Brown, and O. Grabar (eds.), Interpreting Late Antiquity: Essays on the Postclassical

World (Cambridge, Mass., 2001); the numerous works on the formation of the early

medieval world in the multi-volume Transformation of the Roman World series

produced by the European Science Foundation.

Introduction 9

in particular that Rothstein’s short work is still the only study on the

group over a century later. In the last thirty years, even with the rapid

growth of interest in the late antique world, only a very small number

of works have appeared that are signiﬁcantly concerned with the

topic. In 1982, Sartre’s Trois études made a comprehensive attempt

to assess the many aspects of the relationship between the Roman

Empire and the various Arab tribes and individuals whom it encoun-

tered on its southern frontier.

Ghassān featured prominently in this

discussion, and aside from exploring the political and administrative

aspects of Roman engagement with the Jafnids, Sartre also attempted

to identify the archaeological sites which Muslim authors such as

Yāqūt associated with the group, including their reputed base at

Jabiya, but with limited success.

Sartre also offered an examination

of the ‘Ghassānid’ involvement with Christianity. Previously, John

Trimingham’s Christianity among the Arabs in Pre-Islamic Times had

made the most comprehensive effort, since the works of François Nau

and other French Orientalists,

to analyse the spread of Christianity

among the Arabs in the ancient world.

After a lengthy period with

Trimingham’s work as the only monographic treatment of the topic,

Hainthaler’s Christliche Araber vor dem Islam was published in

2007.

For a variety of reasons, the most prominent work on the Jafnids is

the monumental study of Irfan Shahid.

In the last two decades,

Shahid set out to update and re-work Theodor Nöldeke’s study of the

Jafnids, originally published in 1887.

One of the aims of Shahid’s

enterprise was to establish the place of the Jafnids and Ghassān within

his speciﬁc concept of late antique history, through an exploration of

M. Sartre, Trois études sur l’Arabie romaine et byzantine (Brussels, 1982), 120–99.

Ibid. 177–89.

F. Nau, Les Arabes chrétiens de Mésopotamie et de Syrie du VII

au VIII

siècle

(Paris, 1933); also, P. H. Charles, Le christianisme des Arabes nomades sur le limes et

dans le désert Syro-Mésopotamien aux alentours de l’Hégire (Paris, 1936).

J. S. Trimingham, Christianity among the Arabs in Pre-Islamic Times

(New York, 1979).

T. Hainthaler, Christliche Araber vor dem Islam (Leuven, 2007).

Irfan Shahid’s multi-volume study began with Rome and the Arabs: A Prolego-

menon to the Study of Byzantium and the Arabs (Washington, 1984). This work was

followed by three others: Byzantium and the Arabs in the Fourth Century (Washing-

ton, 1995), Byzantium and the Arabs in the Fifth Century, and Byzantium and the

Arabs in the Sixth Century, 2 vols (Washington, 1995, 2002, and 2010).

T. Nöldeke, Die Ghassānischen Fürsten aus dem Hause Gafnas (Berlin, 1887).

10 Introduction

their role as allies of Rome, as participants in the religious disputes of

the sixth century, and through what he saw as a widespread building

programme. Shahid has also explored, sometimes problematically,

aspects of continuity between the Jafnids and the early Muslims in

Syria and Palestine.

Shahid’s ambitious project has attracted a great

deal of attention, and Rome and the Arabs and its successors have

occupied centre stage for some time, playing a key role, too, in

cementing the popular connection between Ghassān and the Jafnid

leaders. Yet Shahid’s efforts pose signiﬁcant problems, not least in his

choice and use of sources and his ideological desire to present a

particular image of Ghassān as a staunchly miaphysite Christian

Arab group, loyal to the Roman Empire, and important to Constan-

tinople’s ability to defend its frontiers from the Sasanians and their

Arab allies. As such, he over-interprets the source material and

assigns too much importance to (for example) the role and position

of phylarchs.

Despite these challenges, Shahid’s study is the ﬁrst

major work on the Jafnids since that of Nöldeke. He should be

recognised for raising the proﬁle of Rome’s Arab allies in scholarship,

as well as for collecting and making available for others an extensive

and disparate collection of literary source material.

In a review article concerning Shahid for the Journal of Roman

Archaeology, Whittow, echoing concerns aired by Michael Whitby,

confronted the probability that Arab allies of the two empires never

exerted a signiﬁcant inﬂuence on their affairs—that the fact that the

‘Arabs appear marginal [in the sources] is because they were’ .

The

positive contribution made by Ghassān as the group under Jafnid

leadership, in particular, as well as more generally by allied Arabs, was

In addition to Byzantium and the Arabs, see, for example: I. Shahid, ‘Ghassānid

and Umayyad structures: a case of Byzance après Byzance’, in P. Canivet and

J.-P. Rey-Coquais (eds.), La Syrie de Byzance à l’Islam, VII

–VIII

siècles. Actes du

Colloque international Lyon–Maison de l’Orient Méditerranéen Paris–Institut du

Monde Arabe. 11–15 septembre 1990 (Damascus, 1992), 299–308. See too a recent

review of Shahid ’s work, A. Walmsley, ‘Review of Byzantium and the Arabs in the

Sixth Century, 2/1’, Ancient Near Eastem Stadies, 44 (2007), 155–6, at 156, which

assesses his contribution to the continuity debate.

e.g. Shahid, Fourth Century,518,Fifth Century,500–1, and Sixth Century, ii. 1, 26.

M. Whittow, ‘Rome and the Jafnids: writing the history of a 6th-c. tribal

dynasty’, in J. H. Humphrey (ed.), The Roman and Byzantine Near East: Some Recent

Archaeological Research (Portsmouth, RI, 1995–2002), ii. 207–24, at 219, commenting

on M. Whitby, ‘Greek historical writing after Procopius: variety and vitality’,in

G. R. D. King and Averil Cameron (eds.), The Byzantine and Early Islamic Near

East,i:Problems in the Literary Source Material (Princeton, 1992), 25–80, at 80.

Introduction 11