7 Egypt

We depart from the Ionians’ opinion, and here is what we personally have to say about the matter: Egypt is all that country which is inhabited by Egyptians, just as Cilicia is inhabited by Cilicians and Assyria by Assyrians, and we know of no border which is properly so called between Asia and Libya except the borders of the Egyptians. But if we follow the belief of the Greeks, we shall consider the whole of Egypt, beginning from the Cataracts and the city of Elephantine, to be divided into two parts, and to have both names; one part of it belonging to Libya, and the other to Asia. For the Nile, beginning from the Cataracts, runs through the middle of Egypt separating it into two parts on its way to the sea.¹

To counter the belief “of the Ionians” that the Nile marked the separation between the continents of Asia and Africa, Herodotus felt compelled to argue that like any other place, Egypt was a single country on a single continent inhabited by a single people. Today, a quick look at a satellite image leaves no doubt that Herodotus was right. The “Ionians’ opinion,” however, perceiving the Nile as a sea that could separate two continents, highlights the importance of the river in the country’s existence. To a great extent, Egypt was the Nile, or rather the Nile Valley and its Delta. This did not escape Herodotus, of course, and he insisted strongly on the ways in which Egypt’s life and civilization were based on the Nile and its annual flood.² However, it was also what marked, in his eyes, the limits of the country’s resemblance to other countries: he famously described at great length how and why the land of the Nile was in every aspect entirely different from the rest of the world. The longevity of his judgment can be measured by the energy still expended by students of late Roman Egypt to convince nonspecialists that it was a province of the empire like any other.³ In his Egypt in Late Antiquity, the only comprehensive treatment of the subject to this day, Roger Bagnall devotes the entire concluding chapter to demonstrating that what his book describes is “a Mediterranean Society.”⁴

It is undeniable that by the fourth century Egypt had become totally integrated in the Roman empire, whatever allowance one might make for specific traits arising from local history and culture. The discussion concerning the degree to which Egypt can be seen as representative of the Roman empire as a whole, however, is still alive and well. Understandably so, because even though the case for official, administrative, and socioeconomic integration is strong, it is not entirely by chance that it has been so slow to convince most scholars. If its perception as “different” has persisted with such strength, it is because it rests on two important realities.

The first is objective and has to do with the country’s geography. Egypt is made up of a narrow strip of cultivable land slivering through the desert. This valley has been cut through the sandstone by the Nile, whose annual flood brought with it and deposited a layer of silt that boosted the soil’s fertility. This aesthetically unique landscape also involved both a high degree of isolation and an equally high ease of penetration and control, as the desert margins formed a very effective border and the navigable river created an easy link with the Mediterranean for a country whose length was more than fifty times its average width. As a result of this, Egypt had proverbial wealth based primarily on its extraordinary fertility and could support a very dense population for the standards of the time.

The second reality determining this view of Egypt’s uniqueness within the empire is historiographical and lies with the sources through which we know and write its history. The papyri preserved by its dry climate offer evidence of a kind unknown for the rest of the ancient world. They document different areas of life from literary sources (or shed a different light on them), they are not ideal or normative but the result of practice, and they capture strata of society and forms of activity that might otherwise have remained unknown.⁵ Using these sources, historians of Roman Egypt have constructed a narrative that effectively sets the country apart from other areas of the empire, if only because of the research questions asked and the subjects treated, inevitably dictated by the sources at hand.

Yet the gift of the papyri has also had the detrimental effect of sending to sleep the need for other types of evidence, which have proved extremely fruitful in the study of other provinces. The most obvious victim of this has been archaeology. The vast amount of archaeological work carried out in the country since the late nineteenth century has focused to a large extent on the pharaonic and Ptolemaic remains. For later periods, it has concentrated on monumental remains and, more specifically for Late Antiquity, overwhelmingly on monastic art and architecture. Generally, archaeology in Egypt has not been driven by the same questions as in other areas of the empire, namely, society, economy, and institutions, areas for which scholars have predictably turned to the papyri. Only recently has there been a rising awareness not only of the significance of archaeological data but also of the importance of recording the archaeological context and the artifactual nature of the papyri themselves, important information that had hitherto been overlooked because the text they are inscribed with was considered to be the only bearer of meaning.⁶

Literary evidence from late antique Egypt is mostly of ecclesiastical or monastic origin, and much of it comes from Alexandria and the surrounding region. At the beginning of the period, it is essentially in Greek, but from the sixth century onward there is a rise in the literary production in Coptic in cities of the valley, also mainly within the ecclesiastical realm.⁷ The bias toward religious questions that is created by this situation is reinforced by the external literature dealing with Egypt, which tends to focus on monasticism or the Christological controversies in which the country was such an important actor.

To many ancient historians, Egypt justifiably looks like a treasure-house of sources. Yet much remains to be done. Few attempts have been made to combine the various types of evidence in order to gain a fuller picture of late antique Egypt, except for the fourth and early fifth centuries, for which Roger Bagnall has offered a full and thorough account based on all the available sources. The period after 450, although not exactly neglected, has not been covered in the same systematic way. In part, this is because research on late antique Egypt was long fragmented among different disciplines that did not communicate very much: Coptic studies, papyrology, patristics, legal and administrative history, monastic archaeology, and history of monasticism, among others. Integration between the various groups of scholars working on this society has been slow to come, but it is now well under way, and some very interesting work is being carried out. Another area where improvement is not only possible but necessary is the quality, variety, and quantity of the archaeological evidence collected, for instance by shifting part of the weight of late antique archaeology from monastic to other types of settlements and structures. Also, as has been demonstrated by several scholars in recent years, it is possible to enhance our understanding of papyri by taking their materiality and archaeological context into account. Finally, making this extraordinary evidence more widely accessible to nonspecialists is an important task that has only recently begun to be addressed by specialists of late antique Egypt.⁸

Late Antiquity is a period of important political transformations that affected Egypt in various ways. The adoption of Christianity by Constantine and the initiation of an imperial policy of control over religious matters brought about a significant development, namely a much stronger connection between the political and the religious history of the empire, both institutionally and in terms of events; tensions between center and periphery increasingly crystallized around issues of religion and were carried by religious institutions and officials. Because of its economic importance and the cultural dominance of Alexandria in the eastern Mediterranean, Egypt pioneered those developments, followed closely by Syria and partly Palestine.

The move of the imperial capital to Constantinople involved a new destination for the annual grain delivery intended for the annona, which Egypt had previously provided for Rome. The convoy of ships normally left Egypt for the capital from the port of Alexandria, which gave the local authorities there tremendous leverage on important issues played out at the time of its departure. During the Arian crisis, for example, there were allegations that the archbishop of Alexandria, Athanasius, was planning to prevent the grain fleet (σιτοπομπία) from leaving the port.⁹ While Dioscorus of Alexandria was in exile and his see occupied by Proterios, the supporters of the former are again said to have threatened to prevent the departure of the grain fleet and were taken seriously to the point that the emperor Marcian had the fleet depart from Pelusium instead of Alexandria to ensure it would reach the capital.¹⁰ Thus Egypt—or rather Alexandria—could use its role as granary of the capital to push forward its position during times of deteriorating relations with the imperial center.

The first such crisis was in the fourth century, when the see of Alexandria with its bishop Athanasius was at the center of the Arian controversy. In the fifth century, it was once again at the forefront of contestation of the imperial line with Dioscorus of Alexandria, in a rift that went much deeper than the first and was ultimately to persist until the end of Roman rule in the country. Even while successive emperors from Constantine to Justinian reinforced the institutional position of the church in the empire, the Egyptian hierarchy was using this newfound power to contest imperial authority. The result was a series of tense episodes between Alexandria and Constantinople, and imperial attempts to bring the country to conformity, first under Justinian and then under Heraclius, had no other result than to reinforce the opposition’s determination.

From the sixth century onward, an important part of the Egyptian church, which included the majority of the monastic institutions, was openly in conflict with the imperial line in terms of doctrine, to the point that it created and maintained a parallel hierarchy that mirrored that of the imperial church. There were not only two patriarchs in Alexandria but also two bishops for Valley sees, one Chalcedonian, the other not.¹¹ Even though this found explicit justification in their disagreement concerning theological questions, it was not a purely religious phenomenon: it was an act of direct defiance toward imperial power, and the energy deployed by successive emperors to bring the Church of Alexandria back into their orbit leaves little doubt that it was understood in that way. Apart from the tension on that front, the sixth century was one of overall political stability, marked by Justinian’s series of administrative reforms. On the contrary, the seventh century was from the beginning one of instability, first with the Alexandrian revolts against Maurice, then with Egypt’s involvement in the civil unrest under Phocas, and finally with its reluctant alignment with Heraclius and the subsequent vexations it underwent during the Monothelite crisis.

The period of the Persian occupation from 619 to 629 is poorly documented, even though a number of papyri have survived, including a decent number of Pahlavi papyri produced by the Sasanian occupants (around 950 known to this day). The paleographical difficulties they present, however, have prevented their systematic exploitation as sources for the period, while the documents produced by the local population do not bring out any distinctive traits. The Pahlavi papyri deal primarily with military supply, although there are also some private letters indicating that the Persian presence was not exclusively male and military, but that some soldiers at least were in Egypt with their families. On the whole, the evidence we have is remarkable for the absence of any sign of cultural interaction between the Persians and the local population.¹² Neither the texts nor any archaeological data give the impression of a major disruption.

Another blank in our knowledge of the country is the ten-odd years between the departure of the Persians and the arrival of the Arabs. We have fewer than 100 dated texts for a period longer than ten years—even though a large number of undated papyri could potentially date from this period. Several later texts refer to this period, but they are strongly biased by hindsight and partiality. Many are dominated by their hostility to the figure of Kyros, who is claimed to have been appointed both as patriarch and as provincial governor by Heraclius, and by accounts of the persecutions he unleashed against the non-Chalcedonians. Conversely, those sources give a very positive view of Benjamin, who was at the time the anti-Chalcedonian patriarch and who had to spend a number of years in exile because of the persecutions. After the defeat of the Romans, Benjamin is presented in anti-Chalcedonian sources as the main interlocutor of the new Muslim rulers.

In the years 641–642, an Arab force led by ‘Amr ibn al-’Āṣ conquered the country with relative ease. Alexandria was briefly recaptured by Roman forces in 645 but definitively lost in 646. The provincial governor and the army settled in the newly built military camp of Fusṭāṭ, next to the ancient city of Babylon at the apex of the Delta, a very strategic position from where both the valley and the Delta could be easily controlled. Like the sixth century, the first period of Arab rule was marked by the combination of administrative reform with a relative political stability that hardly reflected the situation outside Egypt, where the Caliphate was torn by civil wars. The reappointment of ‘Amr as governor by Mu‘āwiya in 659 has been understood as an attempt to keep the province under control, and it seems to have been successful.¹³ Only from the beginning of the eighth century did the situation gradually begin to change, but that lies beyond the scope of this book.

As already mentioned, the Nile was central in the life of the country, if only because it created a number of preconditions that determined to a large extent the forms of exchange in the country. First of those was, of course, the fertility brought about by the annually renewed silt deposit, which boosted Egypt’s agricultural productive capacity. Combined with the ease of transport by river toward the Mediterranean and also within the country itself, it gave Egypt a huge advantage and required a relatively low level of investment in transport and irrigation infrastructures. Even so, as early as the Ptolemaic period and through to Umayyad times, artificial irrigation was extensively used to gain additional land on the valley’s margins.

Agriculture was not Egypt’s only resource. The deserts on both sides of the valley, despite the impression of void they give on a satellite image, played an important part in the country’s economic life. The eastern desert, also known as the Arabian desert, was rich in mineral resources, and throughout the Roman period, its gold mines were exploited, as were various mines for gems such as amethyst, beryl, or jasper. It is especially well known today for the imperial quarries at Mons Porphyrites and Mons Claudianus, which provided, respectively, the red porphyry that became the imperial color in Late Antiquity and the gray granite used in imperial building projects. Other quarries were less heavily exploited and provided a variety of different stones and colored marbles, also often used in imperial contexts.¹⁴ Those two important quarries were protected by the army, and the settlements of soldiers and workers created there are plentifully documented by the ostraka and inscriptions the inhabitants left behind. Their production seems to have been exported for the most part: for their own monuments, the Egyptians used mainly the sandstone and limestone of the valley’s cliffs, as well as the red granite quarried at Aswān, even though some important architectural elements were sometimes made of imported stone.¹⁵

The western desert, known as the Libyan desert, is less rich in minerals, even though it produced alum, and just south of Alexandria, the rich natron deposits of the Wādī al-Natrūn were the empire’s main source for this mineral, essential for glassmaking. Several oases from the Fayyūm in the north to Khargha in the south had important settlements with some agricultural production, especially in Dakhla, and they are now being more systematically investigated.¹⁶ As yet, however, very little is known on the period between the fourth century, when they seem to have rapidly declined, and the reoccupation during the Middle Ages.

The annual flood of the Nile had another important effect on social life. It lasted for three months every year, beginning after the harvest period in early June and ending in early September with the new sowing season. This also marked the beginning of the Egyptian year. The result was that Egypt’s seasons did not work entirely in tune with those of other provinces, as there was a long period of agricultural inactivity when much of the social life and religious festivals took place. Much traveling and trading was also done in that period, as transport was made easier by the water. Finally, the flood determined in large part the form of the habitat, in that most settlements were perched on small mounds that allowed the houses to be above the water.

Because of its fertility and the barrenness of the surrounding areas, the Nile Valley was very densely populated throughout antiquity. Estimates of Egypt’s population in Roman times range from 2.5 to more than 4 million inhabitants, and it is generally accepted that they were supported by local production and were moreover able to produce sufficient surplus to supply the annona.¹⁷ Even though the Antonine plague in the second half of the second century did hit Egypt severely in some areas, there seems to have been almost complete recovery, and Egypt remained one of the most densely populated areas in the ancient world.¹⁸ This density varied in the valley, and it has been suggested that it was higher in the Delta and in the narrower parts of the valley that were easier to exploit.¹⁹

This population, still ethnically diverse at the arrival of the Romans, had become reasonably homogeneous by the fourth century, and even more so by the time of the Arab conquest. The bilingual, Greek- and Coptic-speaking inhabitants of the valley do not seem to have differentiated between themselves as they had under the Ptolemies and in the early Roman period.²⁰ On the other hand, they mention with ever greater insistence members of nomadic groups such as Saracens, Nobades, Blemmyes, and Berbers. Saracens are mentioned from the fourth century in various documents and inscriptions, both in the valley and in the Great Oasis, and seem to have been integrated with the local population even though identified ethnically. One Ἰωάννης Σαλαγηνός (for Σαρακηνός) even acts as scribe for a dromedarius of a garrison stationed at Kysis, in the Kharga Oasis, in the fourth century:²¹

Παταῦτ δρομεδ(άριος) Λατοπολίτων Ἰσοκράτει ἐπιμελητῇ ιδ ἰνδ(ικτίωνος) οἴνου. ἔσχομεν παρὰ σοῦ οἴνου ξέστας εἴκοσι ὑπὲρ τῆς ἱερᾶς ἀννώνης μό(νον)

Ἰωάνης Σαρακηνὸς ἀξιωθεὶς ἔγραψα ὑπὲρ αὐτοῦ.

Pataut, dromedarius of (the garrison of) the Latopolites to Isocrates epimelētēs of wine for the 14th indiction. We have received from you twenty xestai of wine for the sacred annona, 20 only.

I, John the Saracen, was found worthy to write for him.

In the sixth century, the Saracens, like the Blemmyes, seem to have become more threatening in the valley itself, not only on the margins and the desert regions. In the draft of a long petition to an unidentified official by a group of citizens of Antaiopolis, they are both accused of raiding the city of Antaiopolis and characterised as ἀλιτήριοι (criminals).²² The theme of heightened insecurity in the sixth century runs through much of the literature and is often invoked to explain some of the changes visible between the beginning and the end of the century. Why it was easier for nomads to raid towns is not entirely clear. Study of the Eastern Desert has shown that from the fourth century the presence of the army there was drastically reduced, and there are hardly any traces of it after the fifth century. The army’s main job was to keep away the raiders, and the effects of its withdrawal seem almost mechanical.

In part, the weakening of security and protection may have been connected with the depopulation brought about by the plague of 541–542. Yet it is difficult to assess the effect of the plague on the population of Egypt, as we have no estimates of its size either before or after the spread of the disease.²³ According to John of Ephesus, it spread to Egypt from Ethiopia and Himyar (South Arabia). “When the majority of the people had perished, to the point that Egypt came to be deprived of its inhabitants, ruined and deserted, it fell upon Alexandria and killed a throng of people.”²⁴ Procopius says the disease spread throughout the Mediterranean from Pelusium,²⁵ which concords perfectly with John’s account because that is precisely where it would have arrived if traveling up the Red Sea from Ethiopia. Considering that the papyri hardly mention it, it is tempting to imagine that, having traveled mainly by sea, the plague did not have as devastating an effect on the valley as in other areas, despite John’s assertion that Egypt was left empty.²⁶ Indeed, even though an eyewitness, John was traveling by land from Alexandria to Constantinople at the moment of the outbreak and thus did not see what was happening in the Valley. It is entirely possible that by “Egypt” he meant essentially the Delta.

It is true, as Peter Sarris has argued, that a number of objective signs indicating a strong fall in population numbers are valid for Egypt, as they are for other areas. One of these, however, is imperial monetary policy, and that would not have varied from province to province. The other is the rise in the security of tenants in land leases, which indicates a readiness of landowners to make concessions resulting from a need for labor.²⁷ Even if we accept this admittedly tenuous sign, it could conceivably have been prompted by a sharp fall of population in the north of the country and a subsequent fear that lands might be abandoned in the Valley in order to obtain better conditions in the Delta.

The only occurrence of loimos in the papyri, precisely a short time after 542, is in a petition of the inhabitants of Aphroditō to the empress Theodora, where they are complaining about the way the officials of Antaiopolis are treating them on tax matters. However, there are two problems with this reference: first, it is partly restored (only the genitive ending ]μου actually remains), and second, it occurs in a very rhetorical passage where the empress is expected to “bring a healing hand” to the disease (ὑγιοῦσθαι διὰ χειρὸς τῆς εὐσεβεστάτης ἡμῶν δεσποίνης)—which thus lends it a moral rather than a medical meaning.²⁸ One hardly imagines such a metaphor being used if the plague was raging in the country.

This absence of the plague in the papyri is intriguing and cannot be dismissed by saying that the documentation is too haphazard.²⁹ In contrast to the fifth century, the sixth century yielded large archives, and even though they are not exhaustive, it is implausible that only the nonpreserved documents should have mentioned an event such as the plague if it had hit the valley as fiercely as it had hit other areas. In his study of a fiscal register from Aphroditō, Constantin Zuckerman maintains that the impact in Constantinople was high, despite the continued provision of the annona, which should not be taken as a sign that the population of the capital had not been affected by the disease.³⁰ He does not use this evidence, however, to discuss the impact of the plague on the population levels of Egypt. It seems clear that for production to continue at the same scale in 542 and 543, there must have been sufficient manpower to carry it out. In 546–547, Aphroditō was still supplying the same amount of wheat as before the outbreak.³¹ Later, in the late seventh and early eighth century, requisitions of men for construction work in Fusṭāṭ, but also in Jerusalem and Damascus, were made in Middle and Upper Egypt, which could indicate that the Nile Valley had more manpower to supply than did Syria and Palestine.³²

That said, there are indeed signs of fragility in the second half of the sixth century. The document that mentions Saracen and Blemmye raids on Aphroditō, a draft of a petition to an unspecified authority, also shows a rise in the arbitrariness of tax collection and complains about the number of irregular taxes.³³ This is not an isolated case and could be the sign of a certain diminution of the tax base.

In terms of settlement, however, the only clearly recognizable trend where more systematic work has been carried out is a certain redistribution of the population and its concentration in larger centers, especially in rural areas. While cities generally diminished in size, some rural sites were abandoned and others became larger, so that the earlier hierarchy of size was slowly undermined, and the frontier between formally recognized cities and larger rural towns became more and more blurred. The strong economic specialization of villages seems to have eroded as time went by, and by the sixth century, a wider variety of economic activities can be observed there than earlier. At the same time, some rural, agricultural functions gradually moved into urban areas, so that the various types of settlement were more integrated within the wider agricultural economy.³⁴ Only hints of this are archaeologically detectable in Egypt, but the phenomenon of the partial ruralization of cities and the economic expansion of villages is much more clearly observable in neighboring North Africa and Syria, where archaeological evidence has been better exploited.³⁵

A parallel phenomenon touched the monastic world from the late sixth century. The multiplicity of small monasteries seems to have gradually given way to larger, richer, and more powerful structures, some of which functioned as independent settlements, owning and exploiting their own land and, at least in the Arab period, having their own dependent tenants.³⁶ The two documentary archives usually studied for the sixth century, those of Oxyrhynchos and Aphroditō (aka the “Apion” and “Dioskoros” archives), reflect a situation that had not yet reached that point. In the Oxyrhynchite, despite the large Apion estate that dominated agricultural production and relations in the nome,³⁷ the number of small shrines and monasteries was high. The same can be said of Aphroditō, where we also have the possibility of comparing it with the eighth-century situation. By then, the number of monastic centers had diminished, and their status was enhanced. Such monastic centers kept archives, and several are preserved today, especially from Middle and Upper Egypt. The famous Bawīṭ, for instance, has yielded papyri whose edition has already produced three volumes and a number of articles.³⁸

What has generally been seen as the prime sign of “the decline of cities,” namely, the slow demonumentalization of city centers in favor of urban models based on very different principles, is not something that strikes only modern scholars. In the previously mentioned petition of 566 from Antaiopolis, the perception of this phenomenon as a form of decline is very explicitly expressed—by a group of what were presumably the city’s affluent inhabitants. After describing an “old” Blemmye raid, the text continues:³⁹

After that, while the imperial and public baths had been very large to that day, today all that is left in the city is a private bath functioning for the poor, who can bathe there for a couple of nummi.

The demonumentalization, however, was only relative. Large colonnaded streets and classical-looking public buildings and temples were no longer the favorites of the day, and indeed smaller public buildings became more common. Yet building activity did not stop; on the contrary, it went on unremittingly. The multiplication of churches and chapels from the end of the fifth century onward is a very striking phenomenon. Even though some of those were private foundations, they all had a public function: at least once a year, the bishop officiated there.⁴⁰ Towns like Aphroditō could have as many as twenty-six churches and chapels, and this is without counting monastic establishments. In 535, Oxyrhynchos had at least twenty-five functioning churches, and most probably more than thirty of them.⁴¹ Neither were those churches all tiny, as a simple glimpse at the plates of Peter Grossmann’s compendium of Christian architecture in Egypt can easily show.⁴² In Oxyrhynchos itself, the church of St. Philoxenos was expanded in the late sixth or early seventh century, an operation that involved the delivery of 120 new column bases and capitals, which makes for a rather large church for a provincial city.⁴³

The foundation of Fusṭāṭ marked a new move toward urban monumentality. The conqueror of Egypt, ‘Amr ibn al-’Āṣ, is said to have built a mosque at the site where his tent stood during the Arab siege of the Roman fort of Babylon. By the mid-eighth century, this mosque had been altered and extended twice, under governors Maslama ibn Mukhallad al-Anṣārī and Qurra ibn Sharīk.⁴⁴ It is still standing today and known as the “mosque of ‘Amr.” At the beginning of the eighth century, the governor Qurra ibn Sharīk was also building a palace in the city for which men and materials were being requisitioned in the valley, as testified by demand letters preserved in the archive of the pagarch of Aphroditō Basilios.⁴⁵

Regarding split palm branches and palm-leaves. In the name of God. Qurra ibn Sharīk, the governor, to Basilios, the pagarch of kome Aphroditō. We have issued an assessment for provisions on account of the roofing of the buildings of the palace that is being built at the order of the Commander of the Faithful (i.e. the caliph) along the inner harbour at Phossaton … under Ata, son of Abderaaman and Yahya, son of Andala, in the present eighth indiction, and we have made an assessment through your pagarchy of provisions in accordance with the attached requisition notice, and after drawing up the orders (entagia) for these things we sent them to you. Thus once you receive the present letter, work towards fulfillment of the orders and send the needed provisions to Babylon on account of the said palace, as has been described. This was written on the twenty-third of the month of Thoth in the eighth indiction….

Despite their gradual physical transformation, urban centers remained nodal points in the life of the country, in great part because they retained their administrative functions as intermediaries between the center and the rural areas. The administration—civil, military, and ecclesiastical—is an area that has been abundantly studied and described for Egypt because of the important complement of information brought by the papyri, and I shall not repeat at length what can easily be found elsewhere;⁴⁶ rather, I shall highlight some aspects I find important or insufficiently studied to date.

The imperial administration was centered in Alexandria and relayed by several officials to the three provinces created by Diocletian, and below them to the lower levels of local administration. In a similar fashion, the caliphal administration was based in Fusṭāṭ, with a provincial governor whose authority rested ultimately on the network of district governors still called pagarchs. As often in large empires, the local levels of administration were extremely important in defining the form of the relation between state and individuals or local communities. The local elites were initially city magistrates, and increasingly, as the traditional structures of the ancient city became obsolete, members of those groups of local notables can be found in various roles ranging from pagarch in the capital of a pagarchy to protokometes in a village, but also bishop or some rank in the clergy. Some were lawyers or had some judicial function such as defensor civitatis. It is not easy to follow the evolution of this pattern through time, as the necessary groundwork of collecting and analyzing the sources has not yet been carried out. For the same reason, it is also difficult to tell whether the changes in settlement or the Arab conquest changed the composition of those groups, or whether the same individuals and families adapted to new circumstances and remained in power. There are some examples of notables passing the Arab conquest without casualties, and even a case of one being promoted to a higher rank,⁴⁷ but only a systematic long-term study based on a multitude of dispersed sources can analyze and describe Egyptian society as a whole and the transformations it underwent during those crucial centuries.

In many respects, the ecclesiastical administration in Egypt was built on the civil one, based on a network of cities that were the bases of episcopal sees. One important difference, however, was the absence of a regional level centralizing groups of bishoprics under the intermediate hierarchical echelon of the metropolitan. This may have initially been felt as unnecessary because of the peculiar geographical structure of the diocese, but with time, it resulted in a very monarchical ecclesiastical structure, with the patriarch having absolute and direct control over bishops and being surrounded by courtiers whose real influence extended well beyond their nominal ecclesiastical rank.⁴⁸ Perhaps unsurprisingly in such a structure, simony was very common.⁴⁹ Another feature of Egypt’s ecclesiastical administration from the late fifth century onward is the parallel Chalcedonian and non-Chalcedonian hierarchies resulting from the church’s division after the Council of Chalcedon. It is clear from the papyrological evidence that this meant not only that there were two competing patriarchs in Alexandria but also that episcopal sees as remote from the capital as Hermonthis in Upper Egypt could have two competing bishops.

Like everywhere, ecclesiastical careers became more and more prestigious with time and most certainly attracted members of the local social elites. However, this is not easy to substantiate from the surviving evidence without a more systematic study. As clerics often dropped their patronyms in favor of the title they bore, it is difficult to determine their family background in the individual sources in which they appear. However, many seem to have had some basic education and fortune or income. Bishop Hierakiōn of Oxyrhynchos, for instance, had founded one of the churches of the city, presumably from his own funds.⁵⁰ After the Arab conquest, the church became the main institution that could confer social capital on Christians, who still formed the large majority of the population—although the administration retained its allure, and still in the tenth century, many Christians were working for it.

Throughout Late Antiquity, the higher echelons of the ecclesiastical hierarchy were increasingly chosen from the monastic milieu, and by the eighth century, bishops had almost always previously been monks.⁵¹ On the one hand, this facilitated the control of bishops over the monastic institutions that tended to function with some autonomy; on the other hand, it allowed the monastic world to brand the secular church with a strongly monastic character. This was probably reinforced by the fact that non-Chalcedonian bishops often resided in monasteries to avoid confrontation with their Chalcedonian counterparts in the official episcopal city.⁵²

The various administrative structures ensured political, social, and fiscal control; they also allowed the organization and maintenance of infrastructures that would make such control easier and at the same time facilitate economic activity. Evidence for this is less abundant and has been less systematically studied than for the period before Diocletian, but it is clear that efforts were made in a number of areas.

Throughout Late Antiquity, there was much investment in maintaining and even expanding the irrigation infrastructure, which had allowed successive rulers of Egypt since the pharaohs to improve agricultural productivity. As the population of Constantinople rose very quickly after its foundation, the pressure for annona wheat must have been strong. The multiplication of saqiya-type waterwheels in the fifth and sixth centuries testifies to further attempts to extend the cultivable area.⁵³ This drive to encroach onto the desert is strikingly visible today on satellite images of the margins of the Valley—although the expansion is not necessarily located in the same areas as in antiquity. Another way to improve productivity, which also demanded good irrigation, was the rotation of crops, traditionally attributed to the Arabs⁵⁴ but already practiced from Roman times.⁵⁵ The monitoring of the flood also continued, and it is a tribute to its importance that the famous abbot Shenoute attempted to appropriate its symbolic value by presenting himself as being in control of the flood.⁵⁶

One area of state investment was the protection and maintenance of the communication system. Since the Nile ensured north-south communication, land routes were mainly needed for desert transport to and between the oases and especially to the mines and quarries of the Eastern Desert and to the Red Sea. Access to the Red Sea could also be gained through the canal from the eastern branch of the Nile Delta to the port of Klysma, known as Trajan’s canal because of that emperor’s re-excavation of it after it had fallen into disuse. The canal is usually considered to have been left unused by “the Byzantines” and put back into use by the Arabs in the later seventh century. New archaeological and survey work in the area by John Cooper, together with a more systematic analysis of the sources, suggest otherwise. There is also documentary evidence for maintenance work on the canal until well into the fifth century. It is not clear if at that time it was open along its entire length, but the parallel increase in activity at Klysma indicates that the canal must have been active at the time. When it was, according to a tradition recounted by the ninth-century historian Ibn ‘Abd al-Ḥakam, reopened by the Arabs, its closure was described as recent and temporary. ‘Amr, we are told, suggested the reopening to the caliph ‘Umar as a solution to a drought in the Ḥijāz, in the following terms:

You know that before Islam, ships used to come to us carrying traders of the people of Egypt. When we conquered Egypt, that canal was cut, having been blocked off, and the traders had abandoned it.⁵⁷

In his Historia Francorum, Gregory of Tours mentions the canal as being in operation—although admittedly his description is somewhat vague.⁵⁸ Reopened to facilitate the grain supply to the Ḥijāz, it was, again according to Ibn ‘Abd al-Ḥakam, blocked up by al-Manṣūr in the eighth century to cut supplies to the Ḥijāz and the anti-‘Abbāsid rebels there.⁵⁹

Coherent though it sounds, Ibn ‘Abd al-Ḥakam’s explanation is a little intriguing. If the grain for the Ḥijāz was because of a drought, this would suggest that normally the Ḥijāz did not need that grain. Also, if that grain came from Egypt, it did not depend entirely on the availability of the canal—and conversely, closing the canal did not entirely cut off the Ḥijāz from Egypt. Admittedly, the land routes through the desert were less straightforward but not impracticable. Grain was produced in the Delta but also largely in the Valley, which would have involved shipping it north to the canal, through the canal to Klysma, and then from Klysma back south to Medina, a rather laborious voyage. The most important reason for the continued maintenance of the canal in that period probably lies elsewhere. We know from the papyri that from the 660s onward, Klysma had been turned into a shipyard for the construction of the Arab raiding fleet and probably, with time, for the regular fleet.⁶⁰ This protected the building site from Byzantine raids, as the Red Sea ports were inaccessible to them, but made the canal necessary if the Arab fleet was to reach the Mediterranean.

The mountainous eastern desert was cut through by valleys that had allowed the construction of a series of routes between sites in the valley and ports on the Red Sea. One of the most important and best documented was the route from Kainopolis/Qena on the bend of the Nile to the port of Myos Hormos/Quṣayr al-Qadīm.⁶¹ Beyond the access they provided to the mines and quarries exploited in the eastern desert, those routes were important for the eastern trade, especially with India but also South Arabia and Ethiopia, from where such goods as ivory, ebony, incense, pepper, and spices were imported and then traveled up the Nile to the ports on the Mediterranean, Alexandria and Pelusium.

The land route from Myos Hormos to Qena, well documented for the early Roman period, seems to have been abandoned after the third century, and the praesidia housing the army units that guarded it fell into disrepair. This followed the abandonment of the port of Myos Hormos at that time, although at the more southern Berenikē, there are signs of occupation until the sixth century. Activity also continued at the more northern port of Abū Shar, where a fort was built in the fourth century for a military unit of about 200, as well as various military installations along the road from the port to Qena on the bend on the Nile via the quarry of Mons Porphyrites.⁶² These were active through the fifth century and slowly abandoned. The garrison and outposts protected against desert raiders not only the goods and the people that traveled on that road (a Latin military inscription found on the site mentions mercatores)⁶³ but also probably the nearby monasteries of St. Anthony and St. Paul. In the late fifth and sixth centuries, a Christian settlement including a church, possibly a monastery, developed at the fort.

Even though it was not a big settlement, the port of Abū Shar seems to have participated in the trade with India. A Greek inscription of the fifth or sixth century found in the southwest corner of the fort mentions one of its practitioners:

† ἐγὼ Ἀνδρέας … ἰνδικοπλεύσ[της] ἦλθον ὧδε … Παῦν[ι.] ἰνδ(ικτίωνος) θ †

I, Andreas, traveler to India, came here … Pauni., 9th indiction.

Although there is much skepticism about whether traders really went to India or simply to Ethiopia and Somalia,⁶⁴ archaeological evidence from India increasingly vindicates those who gave themselves the epithet.⁶⁵ The little-cited account of the Theban scholastikos who visited India at the end of the fourth century, preserved in the guise of a letter by Palladios in his De gentibus Indiae et Bragmanibus, shows not only that Axum and Adulis were clearly distinguished from “real” India at the time but also that exchange with India was regular and that a Roman could reasonably expect to be able to communicate in Sri Lanka, presumably because of the permanent presence of interpreters.⁶⁶ Confusion has probably been introduced by the fact that the actual crossing of the Indian Ocean was undertaken from Axumite or Yemenite ports, where travelers seem to have transboarded after sailing down the Red Sea. The scholastikos, for instance, boarded a ship with an Indian crew at an Axumite port.⁶⁷ Thus Ethiopia and Yemen were part of “the India voyage” and were even sometimes qualified as “India,” but that does not mean that they were mistaken for India in Late Antiquity.⁶⁸

The foundation of ‘Aydhab south of Berenikē in the later seventh century allowed the new rulers, who now possessed both coasts of the Red Sea, to establish their control over it. It soon became an important port for the transfer of slaves from the Sudan to work in the mines of the Ḥijāz, and from the ninth century onward, it was established as an important commercial port also used for the exploitation of the Sudanese gold mines, the ḥajj, and the India trade. The route to the Nile initially went across the Wādī al-’Allāqī to Aswān, and in the eleventh century, as the Fāṭimids tried to gain control of trade with India through ‘Aydhab, it was moved northward to arrive at Qūṣ on the bend of the Nile, boosting the city into prominence.⁶⁹

It is very difficult to give a global assessment of the state of the Egyptian economy, whether in agriculture, manufacturing, or trade, from the late sixth century to the end of the period covered by this book, as there has been no systematic exploitation of the full range of documents available. This is in great part because they come in three different languages and are handwritten and difficult to read and understand. Recent work is beginning to reverse this situation. The most important impetus for the study of the later part of late antique Egypt (sixth to eighth centuries) was undoubtedly Jean Gascou’s enormously influential and controversial essay on the Apion estate, which sparked a lively debate among papyrologists and specialists of late antique social history, which in turn has led much of the subsequent research on sixth-century Egypt to focus on the issue of landholding patterns, especially the development and function of large estates.⁷⁰ A new impulse was given to the study of postconquest Egypt by the creation in 2002 of the International Society for Arabic Papyrology and its regular conferences that bring together those working on documentary material of the Arab period in any language.⁷¹ Unsurprisingly, much of the work to date has focused to a great extent on fiscal questions, as this is the subject of the overwhelming majority of preserved documents. The subject also allows one to study the administration and thus the formation or organization of the state and the army. This has generally resulted in an insistence on the increasing fiscal burden the Arab conquest represented for most Egyptians.

Surprisingly, however, there has been no analysis of the economic effect on the Egyptian agricultural system of the end of the annona. An embolē continued to be levied to cater to the soldiers (μωαγαρίται) in Fusṭāṭ, but this could hardly have equaled the quantities needed for the population of Constantinople. The assertion that a form of annona continued to cater for the needs of the Holy Cities is not supported by any contemporary evidence,⁷² and the fact that Ibn ‘Abd al-Ḥakam mentions an emergency levy under ‘Amr ibn al-’Āṣ and the caliph ‘Umār because of a drought touching Medina (discussed earlier) clearly indicates that it was not initially conceived as a regular extractive practice. Besides, contrary to Rome, the Caliphate also controlled Mesopotamia, which meant that Egypt was no longer the one and only fertile province of the empire. Ibn ‘Abd al-Ḥakam’s account does mention regular imports from Egypt to the Ḥijāz, not only grain but also textiles, lentils, and other goods. In any case, there was certainly a substantially lower demand for Egyptian grain after the Arab conquest, the consequences of which still need to be investigated—whether, for example, it meant the abandonment of cultivable areas and transfer of manpower to different sectors, or whether the same amount of land and effort produced a surplus that allowed the population to deal with the new tax demands of the Arab rulers, and the new Arab rulers to express such demands.

More generally, there is a strong need for systematic exploration of the period between the sixth century and the tenth, when Fāṭimid Egypt emerged as a considerably rich province. The conditions for that wealth were slowly created in its early post-Roman years, even if they could be exploited to their full potential only under the political conditions created by the Fāṭimid Caliphate. This is, of course, a story that takes us much beyond Late Antiquity.

Egypt has been at the center of debates on literacy in the ancient world because of the detailed evidence offered by the papyri. Ewa Wipszycka has argued very convincingly for a relatively high degree of literacy in the country compared with what is generally estimated for the ancient world.⁷³ Indeed, a simple glimpse at the sort of texts found on papyri makes it clear that this was a society that functioned to a great extent on written documents. Orders of delivery and receipts were written out for transactions that even today might have remained oral. For example, the steward of a given church would write out an order of delivery to a wine merchant to deliver a certain amount of wine to such and such social group. This would be followed by a receipt written out by the recipients and archived at the church. Presumably, the merchant also made a receipt to the steward who paid him for the wine delivered. Such orders and receipts were made even for small sets of nails to repair a stable door or small quantities of tin and lead for the repair of kitchenware.⁷⁴ Two monasteries on the Apion estate in the Oxyrhynchite could share a waterwheel located on the premises of one of them, but both would contribute to its maintenance, and records of their contributions were kept.⁷⁵

One could argue that so important a role afforded to the written word in the conclusion of everyday transactions denotes a fundamental lack of trust within society and a high degree of latent conflict. There is also, as Wipszycka points out, the long bureaucratic tradition of temple scribes who noted and archived everything. Whatever the case, some basic literacy was clearly necessary to carry out a great number of activities. That this literacy was often indeed rather basic is also clear from the variety of grammatical and stylistic registers of privately written documents such as letters. The way in which it was acquired has been described in some detail by Raffaella Cribiore, at least for the early period, and new papyri and ostraka containing school exercises are regularly published.

What this documentation makes clear is that there was large-scale bilingual education, which means that, contrary to a common model for bilingual societies, Egyptians did not learn to write only in the dominant language while using the vernacular orally; rather, they were biliterate, and even if most of them must have had a stronger and a weaker language, whether this was Greek or Coptic was not immediately obvious but depended on a number of factors. Collection and analysis of the equivalent documentation in Arabic could shed much light on the gradual linguistic transition from Coptic to Arabic after the Arab conquest: yet another task that remains to be undertaken.

Education was not confined to teaching basic literacy for everyday needs, however. Jean Gascou describes Hermoupolis in the early seventh century as “un centre d’études littéraires et juridiques fort développé.”⁷⁶ The focus on Alexandria as a center of learning has mostly obscured the fact that several cities in the Nile Valley were important centers of learning in their own right. Not only had they produced famous figures like Nonnos of Panopolis but also they sustained a cultural life of some quality, considering their provincial status and their distance from major centers. Individuals like Dioskoros and the anonymous scholastikos from Thebes who traveled to India and described the habits of the Brahmans were not exceptions: they constituted the typical educated elite of the country. This group was still thriving in the second half of the seventh century under Mu’āwiya and controlled the local administration, like Papas, the pagarch of Apollōnos Anō, whose archive from the 660s has been preserved, and his various correspondents, like the notary Helladios or Plato, the pagarch of neighboring Latopolis.⁷⁷

From early on, books in the form of codices replaced rolls, even on papyrus. Although the passage from roll to codex has often been associated with the symbolic role of the book in Christianity, the relation is far from obvious. The capacity of the codex to hold a larger quantity of text and its practicality for reference offer a more plausible explanation;⁷⁸ they also made it a popular form in the administration, in particular for tax registers.⁷⁹ Literary codices were produced with a basic division of labor, which indicates that a relatively high quantity of them were made. Text, illustration when there was any, and binding were done by different individuals, who seem to have worked individually, taking something akin to postal orders. Orders were sent to them, and the final product was sent back: papyri have preserved a number of such documents. The fascinating dossier of Frange, a bookbinder who in the eighth century had set up shop in an ancient Theban tomb, sheds new light on the entire process. Among the documents are orders sent to him for specific jobs, as well as his own orders and receipts for the materials necessary to his trade.⁸⁰

We preserve today the trace of a variety of book collections about which we usually have very little information. Some of those, like the personal library of Dioskoros of Aphroditō, are connected to otherwise known dossiers and can be understood in their context. Others, however, survive only as fragments of individual works⁸¹ or in the form of fragmentary book lists on papyri,⁸² and much less can be said about them—they could be anything from an order to a copyist to part of a library catalogue, from a reading list to a list of borrowed books. Despite their shortcomings, those lists give us an idea of the most popular works in late antique Egypt. Generally, lists are either “classical” or “Christian,” although this is not entirely consistent. Classical book collections, moreover, do not necessarily denote someone with little Christian interest: the most well-known case is Dioskoros, whose father had founded a monastery and who quotes biblical texts regularly in his works.⁸³ A catalogue in Coptic of the library of Apa Elias in the Theban region contains biblical, patristic, and hagiographical texts and a “book of medicine.”⁸⁴ Other technical books and treatises are known, and their role in producing the knowledge necessary in various economic activities and in trade was without doubt larger than reflected in current scholarship.

That books circulated and were read is attested by a number of letters from the Theban region that offer a vivid image of book-related practice in the seventh and eighth centuries. For example, a monk of St. Phoibammon wrote to an anchorite in the seventh or eighth century:⁸⁵

The book that I sent you … can it be sent back to me, so that I can read it during the vigil of Saint Apa Phoibammon.

In the sixth century, before the monastery moved to its final location on the ruins of the temple of Hatshepsut, a certain David wrote to a monk who had placed an order with him:⁸⁶

Recto: Before [all things] I greet my [beloved] brother Kouloudje. [When] I left [thee] (thou saidst), “Write the Deuteronomy.” Now, I did not write it, but [I] have written the Leviticus and the Numbers in their order. If I am able, I shall write the Deuteronomy.

Verso: Give it to the master, Kouloudj, from David, the most humble sinner. Farewell in the Lord. The Holy Trinity.

As has often been pointed out, the copying seems to have been mainly done by monks, but nothing really indicates that provincial or rural monasteries had proper scriptoria like the well-known medieval ones. More plausibly, some establishments hosted one or two monks who worked individually, as the letter shows.

Late Antiquity is often treated as a single, chronologically undifferentiated unit, and this has also been the case for Egypt. In some cases, scholars cover the entire period from the late third to the seventh century;⁸⁷ in others, they concentrate on a given period within Late Antiquity, generally the fourth or the sixth century.⁸⁸ Implicitly, this reflects the fact that the period covered by the expression is one of very important transformations, for which the fifth century seems to have been pivotal. This is clearly recognized, to the point that Andrea Giardina suggested calling Egypt “late antique” until the mid-fifth century and “Byzantine” after that⁸⁹ (a suggestion that has found favor against the old periodization calling “Byzantine” the period from Diocletian to the Arab conquest). There has been as yet no full survey work on Egypt during the so-called “long” Late Antiquity, which includes the Umayyad period, although the sections devoted to Egypt in Chris Wickham’s Framing the Early Middle Ages can largely stand as an independent introduction to the country’s history over those centuries.⁹⁰ Very much like for the rest of the Mediterranean, the divide is clearly visible between the early fourth century, when the country is still recognizably Roman, and the late seventh to early eighth century, when it is unambiguously on the medieval side of the border. Christianity, a key element in definitions of Late Antiquity whether explicitly or implicitly, is in the fourth century a rising force, fighting—very efficiently—for position and power; in the seventh century, it is established and totalizing, especially in Egypt and Syria, where it is beginning to feel the threat of Islam.

The differences between the beginning and the end of the period are not necessarily linked to Christianity, however. Many of those changes happened over the late fifth and early sixth centuries within an empire that was already Christian. The evolution described before of the settlement pattern toward greater concentration and the concomitant developments in administrative structures probably have little to do with religious questions. The rise of large estates and rich landowning aristocracies in the sixth century, a theme that has been very prominent in recent research largely because of the conservation of the Apion archive, also happened independently from the influence of Christianity. Even though the economic importance given to this phenomenon as a specificity of the sixth century seems greatly exaggerated, considering that in the early empire there were also great landowners with large estates producing industrial amounts of agricultural goods and shipping them across the Mediterranean, there were a number of qualitative changes that gave sixth-century oikoi a different social function. The fact that they levied some taxes, for example, has been at the center of heated debates as to whether they were acting as agents of the state⁹¹ or usurping the state’s attributions and thus undermining its power and authority.⁹² To some extent, the two suggestions are not entirely incompatible. As research on late medieval and early modern regional aristocracies has repeatedly shown, such groups were both using their status as representatives of the state locally to acquire authority and wealth, and then used those to undermine and control the sovereign’s power. It was not in their interest, however, for the system of the state to collapse, because their own power was based on it. After all, consular families like the Apions can be said to be one of the parts that constitute “the state” in the first place. Even though we are not, as was once argued,⁹³ witnessing the beginnings of feudalism in sixth-century Egypt, this is a marked departure from the Roman municipal system toward one based on a centralized monarchy both relying on and resisted by powerful aristocracies, a form that appears as distinctly medieval and early modern.

Like everywhere in the empire, the rise of the influence of the church was exponential throughout Late Antiquity. By the sixth century, it was a central institution in the political, economic, social, and cultural life of the country.⁹⁴ From the late fifth century, it began marking the cities and the surrounding landscape with a multitude of shrines of various sizes, in parallel with the rise of the cult of saints that gave an impetus for the construction of those new cultic buildings.⁹⁵ By the mid-sixth century, the physical aspect of both urban and rural space had been transformed and so had the temporal experience of that space, since the shrines, deliberately different in form from temples, were also visited by communities at given times of the year for the eponymous feasts.⁹⁶ Within the church, the monastic element was steadily on the rise, so that what had initially been a movement of contestation and rejection of institutional hierarchy and authority was by the sixth century becoming the moving force behind the institutional church itself, and the higher echelons of church administration were more and more occupied by monks. This was supported by the massive production, in the form of holy biographies and sermons, of a discourse that presented ascetic virtue as the best quality a churchman could exhibit. Monasteries were also landowning institutions with some financial means. By the seventh century, they had also largely appropriated the cult of the saints and used its persuasive power to obtain donations of cattle, consumables, land, and urban property.⁹⁷ In the early Arab period, they seem to have served as centers for the levy of taxes and thus acquired a status that allowed them to exert control over the corresponding communities.

The political power of the church was, of course, curbed by the Arab conquest, since the new governmental structures did not take account of ecclesiastical sensitivities. Within the Christian communities themselves, however, its power eventually came to a zenith, as it was the only institution officially representing the Christians within the Caliphate. This was probably much slower than usually thought, however, and did not come about during the period covered by this book.⁹⁸ In the seventh century, local officials and tax collectors were lay Christians, and they were the most important individuals in everyday life for most of the population.

I have not given here a full account of late antique Egypt—an impossible task within the allotted space. Instead, I chose to highlight a number of subjects that have received less attention in recent general overviews. By looking from a slightly different angle at a country that is generally associated, for this period, with monasticism, administration, taxes, the Apions, Dioskoros, Alexandria, and Coptic art, and has often been seen as a closed entity, I hope to have shown that it was also a dynamic participant in the life of the empires of which it was part and that its fate was dialectically rather than passively linked to theirs.