For the past few decades, studies on slavery in late antiquity have been primarily concerned with the question of whether slavery was, in fact, present during the period, or whether late antique slavery slowly declined and transformed into so-called medieval serfdom. For many years the latter proposition enjoyed favour among historians. Now, the picture is quite different. The majority of recent studies on late antique slavery confirm that slavery was alive and well during the period. Foundational studies such as Chris Wickham’s monumental analysis of the medieval period, followed more recently by Alice Rio’s focused study of early medieval slavery, Youval Rotman’s reconstruction of Byzantine slavery, and finally Kyle Harper’s extensive survey of late ancient slavery come to similar conclusions: the model or paradigm of ‘transition’, with its roots in nineteenth-century Marxist economic theory (especially from Marx and Engels), has outlived its usefulness for understanding labour and modes of production in the late antique world.

Located somewhere between ancient fiction and Christian hagiography, the story of Euphemia and the Goth recounts the tale of a young girl from Edessa, Euphemia, whose widowed mother, Sophia, is deceived and manipulated into letting an unnamed Gothic soldier marry her daughter. Later, as the story develops, events take a turn for the worse, and the pregnant Euphemia is taken away to the Goth’s homeland, only to find that he is already married. She is then given as a slave to the Goth’s wife, and suffers terrible abuse before being miraculously rescued.

The story is set in ca. 395 ce Edessa, in the context of the invasion of Mesopotamia by the Huns, although possibly composed decades later in the fifth century. As in some other cases from Syriac literature, Euphemia and the Goth reads in many ways like a tale from the genre of the Greek novel.

This study demonstrates how the macrocosm of slavery, along with its violent measures of discipline and punishment, was assimilated into and reproduced by Christian ascetic culture of late antique Syria, in both urban and rural spaces. After investigating the nature of slaveholding practices in the cities and outlying villages of the region, the study asks how the type of slavery that was characteristic of Syria may have affected the ascetic practices of fourth- and fifth-century monks who lived on the fringes of cities and villages. Finally, in order to understand the ascetic afterlife of slavery in Christian Syria and Mesopotamia, the study also explores the nature and ascetic function of slavery in some later sources of the sixth and even seventh centuries. This exploration into the later centuries gauges the potency of discourses and practices of slavery. It also serves to make us aware of just how powerful, pervasive, and persistent doulological discourse and discursive shifts were in late antiquity.

The concept of asceticism is notoriously difficult to define, and we still cannot say that there is one ‘accepted’ definition. What we have are varying representations and propositions from ancient authors of the ascetic project, each with different emphases and goals in mind.1 Rather than straightjacketing an ancient author or literary work within a preconceived notion of asceticism, it should rather be asked how the author understands and utilises ascetic discourse and practice, and how this particular utilisation intersects with other related religious, cultural and political concepts and phenomena. One phenomenon with which asceticism has been shown to interact is that of violence. But the relationship between violence and ascetic discourse and practice has been shown to be quite complex.2 The notion that asceticism and acts of violence are not necessarily mutually exclusive has gained much ground in scholarly circles, and rightfully so.

This volume is the outcome of a seminar on issues relating to changing patterns of population mobility and settlement in southern Africa. The seminar was held in January 1996 at the Grahamstown campus of Rhodes University in South Africa, and was arranged jointly by the Institute for Social and Economic Research at Rhodes University and the International African Institute. We have organised the order of the chapters in the book, as well as our discussion of them, in terms of the broad thematic sequence of population mobility, settlement and policy implications.

THE NEED FOR A NEW APPROACH

Tt is now widely conceded that human movement is definitive of social life more often than it is exceptional in the contemporary world’ (Appadurai 1995: 215). One might add that this is especially so in what is broadly known as the Third World, where political instability and its correlate of weak states often result in governments resorting to violence to achieve obedience from their citizens (Migdal 1988: ch. 1). In addition to widespread circular labour migration, urbanisation and movement within urban areas, in Africa as a whole, over the last half century or so, more than fifty million people have been uprooted, as shown in Table 1.1.

In southern Africa, such widespread human movement has been occasioned by economic forces (including widespread migrant labour to the mines and the urban areas), environmental pressures (such as the harsh droughts of the 1980s), political policies (notably apartheid in South Africa, socialist-based villagisation policies in Mozambique, and security villages in Angola and the then Rhodesia during the wars of decolonisation) and flight from the oppression and civil war in Angola, Mozambique, Namibia, South Africa and Zimbabwe. Some of the more prominent instances of population movement in southern Africa have been:

ANGOLA. Between 1961 and 1972 approximately 700,000 people were moved into concentrated settlements for purposes of military security during theAngolan war of liberation against the Portuguese (Niddrie 1974: 56-7, 76). During the period 1961 to 1992 the average number of Angolan refugees in Zaire was 315,932, peaking at 620,000 in 1978, while during the period 1966 to 1992 the average number of Angolan refugees in Zambia was 48,604, peaking at 102,500 in 1991 (Hansen this volume).