Archaeological and biological evidences for the Justinianic plague:
the “garage Lux” in Alexandria?

Introduction

In March 2020, the state of health emergency is declared in France, trying to limit the spread of Covid-19, an epidemic “that has become an immediate, pressing reality”¹. This speech, promising the resistance of the French territory against the disease, will have marked the spirits, the commentators insisting on the urgency of a phenomenon considered unprecedented.

If this dramatization may be justified, the situation is not unique. The late few years have shown the epidemic risks that humans are facing – AIDS, Ebola, Zika and more recently Covid-19 or monkeypox, new or re-emerging pathologies that awaken ancestral fears: those of disease, of a contagion that cannot be contained and finally of mass mortality on a global scale. The epidemic history of human societies confirms the threat of such conjuncture, with recurrent trends from one to another, both in the disorganization of society and in the deep and lasting fear that it provokes. It was even worse at the time of the first epidemics described in literature, when the specific cause was unknown and the medicine quite powerless. The successive waves of such pathologies, of varying duration and intensity, have left many scars on people’s minds due to the demographic, social and economic upheaval they have engendered.

Since the very end of the 19th century, we have known that the bacillus Yersinia pestis is responsible for plague epidemics² whose propagation and dissemination mechanisms, still incompletely understood today, have allowed a massive extension and consequently unprecedented catastrophes. Described by ancient authors from the sixth century of our era and responsible for three pandemics of several centuries each, it is a “bacterial anthropozoonosis”³, “whose pathogenicity for humans is the highest in the bacterial world”⁴. It is therefore a pathology that has attracted the attention of researchers. Burial areas, graves and the individuals they contain are, through archaeological and anthropological indicators, the ideal witnesses of mortality crises. Since the 1980’s, research on mortality crises and epidemics has been promoted by the development of preventive archaeology and the improvement of techniques for excavating funerary areas, particularly at the end of the 1990’s, when the discovery of several crisis cemeteries demonstrated the informative value of an anthropological study of the individuals buried in these contexts⁵. These funeral complexes have allowed the development of syntheses on past plague epidemics, from an archaeo-anthropological point of view, and have given rise to studies of the specific composition by sex or age of the series from these peculiar environments⁶, and analyses of the funerary treatments of plague victims⁷. The evolution of research methods and the development of interdisciplinary studies have allowed significant progress, concerning particularly on the spread of Yersinia pestis epidemics and its transmission modes⁸, but also on its genetic signature and its mutations.

Recent works are being carried out by the ANR “PSCHEET” program⁹, a program aiming to understand the paths and routes of transmission of epidemics, and to discuss their social, cultural and biological impacts. It is an integral part of the present interdisciplinary movement, mixing the fields of anthropology, archaeology and genetics. These include debates about the non-selective recruitment of the plague, challenged by recent anthropological studies showing a potential influence of age, gender or health status¹⁰. The project is based on knowledge acquired for the Medieval period that allow to work on poorly known ancient plagues, as the Justinianic plague. The study of the “Garage Lux” in Alexandria, a Christian cemetery of about 300 individuals in simple and mass graves, is part of this approach, and could place Alexandria on the main route of transmission of the first plague pandemic, or give a clue for an Egyptian origin of the first plague pandemic.

State of knowledge about the Justinianic plague

Ancient authors are particularly prolific on plague epidemics that took place in the medieval and modern periods. They provide more or less precise knowledge about plague episodes, although it is necessary to moderate their comments because of each author’s overstatements. They nevertheless provide a partial understanding of the demographic impact of plague epidemics, but in different ways, since some assert that the disease was very selective, and others that it affected all age groups and both sexes without exception. In this sense, their exploitation alone does not provide an answer to the selectivity and impacts of the plague on the populations. The parish registers can, in rare cases, provide some details, but none remain from the time of the Justinianic plague. Therefore, archaeological sources, the study of mass graves and their recruitment, are imperative.

Archaeology of the deaths

A mortality crisis can be identified by specific archaeo-anthropological criteria, related to the number of deaths in a short period of time¹¹. Individual burials cannot provide any indication on this subject, but plural burials¹², with two individuals or more, can turn out to be typical of a mortality crisis if it can be shown that the different subjects have been buried simultaneously (or quasi-simultaneously). This multiple burial type is defined by the simultaneous deposits of several individuals within the same tomb (or pit), as opposed to collective multi-stage burials, in which inhumations succeed one another and are spread out over time (often over a long period of time). However, a mortality crisis (particularly an epidemic) doesn’t necessarily lead to multiple simultaneous burials for, at the beginning or at the end of the epidemic episode, the number of dead is low and manageable, and the usual funerary treatment with individual burials may be maintained¹³.

The decomposition of a body begins at its first place of burial, theoretically the place of “primary burial”. It is a relatively rapid phenomenon; therefore, the simultaneous deposits of several bodies implies that occur before the decay of organic matter and the beginning of the dislocations begun – a question of a few days, very few weeks at the most. The simultaneous burials are thus ‘archaeologically simultaneous”, which means that few days can be left between two deposits. De facto, it is considered that the simultaneity of the deposits implies a simultaneity, or quasi-simultaneity, of the deaths. The timing of this simultaneity/quasi-simultaneity is dependent on the conditions of conservation, on the temperature (cold slowing down the decay process), on possible anaerobic environments (peat bogs or desiccation are preventing putrefaction). Funerary practices are also to be taken into account, certain treatments slowing down, or even blocking the decomposition of soft tissues (airtight lead coffin, practice of mummification, injection of antibacterial or antifungal products). On the other hand, several bodies found in the same secondary burial (dislocated or partially dislocated skeletons) do not indicate that they were buried at the same time, nor the synchronicity of the deaths.

Several studies deal with past epidemics, but only a handful of them focus on the effects of mass mortality on funeral practices. The increase of the number of deaths leads sometimes leads to a more or less gradual abandonment of standard practices for more functional processes¹⁴. The simultaneity of the deposits can be determined through archaeo-anthropological evidence: mostly the conservation of anatomical connections, especially the most fragile, unstable joints (hands, patellae, temporo-mandibular, etc.). If the individuals are buried at the same time, the joints will indeed be maintained in a strict manner and the bones, after the decomposition, will only move vertically, when the underlying soft surfaces disappear. The absence of sedimentation between individuals in the same burial is also an indication of simultaneous deposits¹⁵.

The archaeological study of the disposal of the dead within mass graves is only a part of the work on death crises and mass burials. It allows the observation of the “management” of the crisis and of the possible changes in funerary practices sometimes carried out of sense of urgency, but their cause can only be determined by an anthropological study of the bone collection from such funerary assemblages, whose data of which must be compared with populations whose deaths did not occur in times of crisis.

Paleodemographic and pathological studies

Archaeo-anthropological studies of osteological collections from crisis cemeteries consist firstly of estimating the biological sex and the age at death of all the individuals¹⁶. This work allows researchers, in a second step, to establish a mortality profile and to estimate the rate of masculinity of a population¹⁷. The age and sex distributions thus obtained are compared with data from a so-called “natural” or attritional mortality, determined from pre-Jennerian or archaic populations, dating from before the industrial revolution¹⁸, or a model, for instance Ledermann’s standard life-tables (1969), as recommended by Sellier (1996). Non-adults Age at death is usually dealt with through five-year age classes, except for the first ones (0 and 1-4 years of age), but any other age-groups are possible. For each age-group, probability of death is calculated and then compared to Ledermann’s mortality figures, allowing the establishment of a mortality profile in search of “demographic anomalies”¹⁹, which are differences from the model than can be interpreted in terms of mortality crisis, burial selection, etc. The estimation of the age and sex of individuals is essential in every palaeobiological study. Recent anthropological studies of crises cemeteries have shown that some mortality crisis can cause a selection in terms of age and sometimes of sex, depending on the initial composition of the population and on the nature of the crisis (epidemic, famine, etc.). For historical or archaeological plague samples, there is an under-representation of immatures less than 1 year of age, and an over-representation of all the other non-adult individuals (in terms of probability of death) whereas in a natural mortality, the minimum probability of death is found in the 10-14 age-group. The current study of the osteological population of the “Garage Lux” burial ground could therefore highlight demographic anomalies within the collection and help us to understand the nature of this specific mortality²⁰.

The palaeobiological study of skeletons from such contexts shows an absence of stigmata related to interpersonal violence, excluding a war or a massacre crisis. As far as paleopathological conditions are concerned, the action of infectious germs is too quick, during an epidemic, to let bone lesions appear, except during low-lethal epidemics such as leprosy or tuberculosis. Non-specific stress markers (cribra orbitalia, cranial hyperostosis, periosteal bony formation on the tibia, dental enamel linear hypoplasia) are peculiar to any epidemic but they can indicate deficiencies and a vulnerable state to lead a higher prevalence of diseases during lifetime. Those markers, associated with the absence of traumatic lesions, can give evidence to a mortality crisis due to an epidemic and show the degree of adaptation of a society to mass mortality. The health status of the “Garage Lux” will be presented according to the number of individuals by category of the nosological framework, and not according to the number of lesions. In addition, lesions too close to a taphonomic fracture or alteration, or incomplete ones, are not considered as valid.

Paleomicrobiological studies

The study of ancient diseases requires biological and molecular analyses, which have been developed recently but are the only ones leading to precise identification of a specific pathogen. Indeed, if the taphonomic, anthropological and demographic data give clues for a mortality crisis, the definite diagnosis can only be confirmed by the molecular analysis of bone or dental tissues from such contexts. Detection of pathogenic microorganisms can be managed by complete genomic analysis, amplification of ancient DNA by polymerase chain reaction (PCR) on bone or dental samples, or by microscopic observation of the sample for characteristic pathogenic antigens or ectoparasites.

Yersinia pestis DNA is most easily found in the dental pulp of ancient specimens. The first studies aimed at the identification of three main varieties – antiqua, medievalis and orientalis – each one has been proposed to be responsible for one of the three historical plague pandemics²¹, with reference to the supposed geographical origins of each epidemic and the distribution of the variants. This hypothesis was disproved in 2004 by genotyping the bacterium in dental pulp samples, showing that only the orientalis variety was found in ancient samples²². However, there is still some debate about this hypothesis, with some claiming that the three pandemics were caused by ancestral strains of Yersinia pestis that do not correspond to any of the current biotypes²³. These three biotypes are distinguished by their glycerol fermentation and nitrate reduction: antiqua is capable of both, while medievalis does not ferment glycerol. However, all three strains show similar virulence and symptoms. Their dangerousness comes from the fact that they multiply very quickly in an extracellular situation, protecting the bacteria from antibodies, preventing the inflammatory reaction and phagocytosis²⁴.

In 2013, a study by Harbeck et al. suggested that the three historical plague pandemics, caused by Yersinia pestis, originated in Asia, specifically China, due to the position of their DNA on the phylogenetic tree of the bacillus²⁵. However, in 2013, Cui et al. proposed that one of the branches (the “Angola” branch) originated in Africa, spread to Europe, and caused the first pandemic; this remains a hypothesis today. Recent work has shown that the Justinianic plague was distinct from already known plagues²⁶ and that different strains of Yersinia pestis, whose virulence could be exacerbated by the presence of certain genes²⁷, were circulating in Europe in the sixth and seventh centuries²⁸.

Historical studies

The symptoms of the plague are described in ancient and medical literature, especially after the Justinianic plague. Although they can vary, some symptoms are common to all populations. In the 5th c. BC, Thucydides describes a plague-like epidemic which has never be proven to be the plague itself (before the first recognized pandemic, the Justinianic one). He mentions the symptoms of the main forms of the pathology (bubonic, pulmonary and septicemic, mentioning the fear of the epidemic, the ineffectiveness of the remedies and the doctors, the differences with other pathologies, the damages in the cities, depression, observation of the sick persons (their thirst, their breath, the blood in their saliva)²⁹.

Historical sources mentioning the Justinianic plague indicate an African origin of the first plague pandemic, Egyptian³⁰ or Ethiopian. Attested in 541 AD in the Egyptian port of Pelusium, located in the eastern delta of the Nile, it spread then to the rest of the Mediterranean basin through maritime trade, rapidly affecting other Egyptian ports, then Palestine, Syria and Constantinople³¹. No data on its extension beyond Persia has been received. From Constantinople, the plague spreads from 541 AD to Illyria and Spain, then to Italy and Marseille the following year³².

For more than two centuries, the plague was present in the Mediterranean region, returning in successive waves until 767 AD; the chronicles of the time count fifteen waves, ten of which affected present-day Europe³³. Thus, the second wave, from 558 to 561 AD, reached Constantinople once again, penetrated westward through the city of Ravenna, reached Genoa and was limited to Italy. The third wave, in 570 AD, entered in Genoa and Marseille and hit the eastern part of Gaul and Italy more severely, reaching Constantinople in 573-574 AD. The fourth and fifth waves (580-582 AD and 588-591 AD) began in Narbonne and only affected Western Europe, quite exclusively the Spanish, Italian and Gallic coasts. The sixth plague outbreak, from 599 to 600 AD, affected Ravenna, Rome and Marseille. The next nine waves, from 608 AC to the 760’s AD, were more limited in the regions they affected and in their impact, mainly concerning the cities of Marseille, Arles, Rome and Pavia around 654 AD, the region of Narbonne 14 years later, then Carthage, Naples, Calabria and Sicily in 746 AD, before reaching the southern part of Italy in 767 AD³⁴.

The expansion of the Justinianic plague would then have been geographically limited to the Mediterranean basin and mainly the coastal parts, its penetration inland being limited to the major river routes (Rhone, Loire, Elba)³⁵, probably related to by population densities and movements. However, this picture is totally dependent on the fragmentary and often emphatic content of the chronicles of the period. On the other hand, those chronicles do not give any clue for the cause of the disappearance of the disease in the 8th century, which remains unexplained to this day, other than the possibility that the pathology did not find “the necessary conditions for its persistence”. If such writings contain many details about the repercussions on the societies affected by the plague, it must be taken with caution because of the bias and exaggerations they may contain, influenced by the author’s point of view³⁶.

If the debate around the impacts of the Justinianic plague on societies is still ongoing, the major point of contention concerns the origins and transmission routes of this epidemic. Indeed, the population models do not fit with historical sources, the biological and genomic data pointing rather to an Asian origin of the plague³⁷ where the medieval authors suggest an Egyptian or Ethiopian origin. The study of the “Garage Lux” in Alexandria could provide an element of response to this debate. It is indeed justified by the discovery of multiple burials during the rescue excavations conducted by the Centre d’Études Aleaxndrines (CEAlex) between 2000 and 2002, and is coping with the funerary organization of the necropolis, the possible number of crisis, the implantation and possible simultaneity of the tombs, and detailed archaeo-anthropological analysis, the nature of its mortality profile (through life tables and figures of probabilities of death) completed by aDNA analysis.

The “Garage Lux” in Alexandria: a preliminary assessment

After a period of political unrest during the second half of the 3rd century, Alexandria resumed its commercial expansion during the 4th and 5th centuries thanks to its geographical location³⁸, although weakened by several earthquakes, including the famous “universal cataclysm” narrated by Ammian Marcellin³⁹. In addition, the city is corrupted by internal divisions, especially religious ones, which lead to the looting and destruction of many public places or buildings, such as the Serapeum or the Caesareum. In this context of difficulties, a new episode of unrest occured at the end of the 5th century: the beginning of an epidemic of plague, causing an unprecedented mortality crisis in a very short time. It posed a challenge for the management of this episode by the Alexandrian society and the traces it may have left in the urban extension of the city⁴⁰.

The study of the “Garage Lux” in Alexandria could provide some answers or hypotheses, firstly thanks to the study of the simultaneity of the burials and a discussion on funerary practices in this context (number of bodies in a pit, orientations of bodies, etc.). The stratigraphy of the cemetery and its chronology must also be established in order to see if the site could fit in the chronology of the Justinianic plague. A biological study (mortality profile, pathological lesions) is conducted in the a third phase, allowing researchers to eliminate the hypothesis of a war or a massacre, researching a specific demographic signature revealing of a particular epidemic and discussing the health status of the population. This work will have to be completed by DNA analysis to confirm or deny the presence of Yersinia pestis, and thus the possibility to place Alexandria on the main route of transmission of the first plague pandemic.

“Garage Lux”: presentation and localization

The site of “Garage Lux” is located in the center of the ancient Ptolemaic city between the present-day streets of Shakhour Pacha, el Qala, Safia Zaghloul and Sultan Hussein, in the Brouchion district. The site is surrounded by other areas of archaeological excavations carried out by the CEAlex and has been the place of rescue excavations between 2000 and 2002⁴¹, led by the CEAlex in collaboration with the Egyptian Antiquities Directorate. Nearly 650 m² had been destroyed by the property developer before excavations could take place, leading to a substantial archaeological material found outside the stratigraphy. The site of “Lux”, resting (as the entire region) on a limestone massif, has been divided into five archaeological sectors where four occupation phases have been distinguished.

The first phase is chronologically situated at the end of Antiquity. Indeed, the site has re-occupied the area of the ancient sanctuary of the Ceasareum, a monumental complex ordered by Cleopatra in honor of Mark Anthony and completed under Augustus. It is the only Alexandrian sanctuary whose location is absolutely known, thanks to the two obelisks that marked its entrance, named “Cleopatra’s needles” and found near the northern part of the site⁴². The excavations carried out by the CEAlex between 1992 and 2002 on four sites – the “Garage Lux”, the “Billiardo Palace”, the “Cinéma Park” and the “Cinéma Majestic” – revealed a surface area of more than two hectares. The time of destruction of the Caesareum is uncertain, probably at the end of the third century or the beginning of the fourth because of the numerous civil troubles already mentioned.

The second phase of “Lux” is illustrated by a hydraulic network from the earliest times of Alexandria, illustrated by a drainage system for waste water with hyponoma⁴³. During Antiquity, wells were built in public spaces, as well as cisterns where urban water was accessible. This reorganization of the public spaces according to a new urbanistic conception is illustrated at “Lux”, where hydraulic structures (cisterns, wells, collectors) have been discovered, associated with pipes. During the study of the architectural elements, it has been suggested that the cisterns may have been used to protect the drinking water from contamination, water being recognized as a major factor in the spread of disease during Antiquity⁴⁴.

The third occupation phase corresponds to Lux cemetery and is the focus of this paper. It is spread over three of the five sectors defined by the excavators. The stratigraphy of the funerary complex of “Garage Lux” is quite complicated, present on more than 1.50 m. high with an important number of burials. This gives to the site a major place in the Alexandrian history, since it is the first cemetery located within the city wall⁴⁵. Moreover, two wooden crosses were found associated with burials, definitively identifying the “Garage Lux” as a Christian cemetery established on at least two levels (two phases). The first state of the cemetery was implanted under the ground level of the hydraulic state, which was overburdened and thus destroyed probably to recover the materials and spaces. However, there is no trace of an associated church, leading to question its existence.

The last state of the “Garage Lux” is a modern one with a market place, dating from the 19th century, especially buildings specialized in the preparation of fish.

A Christian cemetery

A Christian cemetery attached to a church?

The “Garage Lux” cemetery could have been attached to a church. Indeed, textual sources mention the Kaisareion, a church that could be located on the site of an ancient imperial sanctuary. It would have been founded in the 4th c. AD by the arian bishop Gregory (339-345) for the emperor Constance II: έν χρόνοις Κωνσταντίου έδοξεν αύτό οίκοδομηθηναι, Γρηγορίου του Μελιτιανου και ‘Αρειανου άρξαμένου⁴⁶ and όν ώκοδόμησετόπον ό βασιλεύς; Τώ Κυρίω ώ καί τόν οίκον πεποίηκας⁴⁷. This is the megalè Ekklésia of Alexandria attested from the 4th to the 10th c. AD. It would have been the seat of religious ceremonies until its destruction by fire in 913, but would also have served as a refuge for the poor and foreigners and as a hospital. Several buildings would have been connected, such as a factura, a zygostate and a tachograph cabinet⁴⁸.

No archaeological evidence have been found for this church, but its presence and its possible link to the “Garage Lux” burial ground show that plague cemeteries were not necessarily set apart from the cities. Moreover, it would be the first documented cemetery within the city walls.

Study and dating of the archaeological material

The dating of the archaeological material seems to fit in the chronology of the Justinianic plague. Indeed, the ceramic fragments from the different levels are dated between the 5th c. AD for the oldest, and the 7th c. AD for the levels just above the most recent tombs.

The coins associated with some of the burials give the same chronological range. Some glass ware was also found associated with the skeletons. Two vases, blown in a mould and found in burials from the 3rd phase can be dated from the late 6th-7th AD (according to the shape of the tulip necks and the bottom⁴⁹).

Two Christian crosses were found in the second sector of the cemetery, in connection with a multiple burial including an immature of less than 1 year of age and three adults aged between 20 and 29 years, whose sex is undetermined. Given their condition, the study of crosses is relatively complicated. They seem to be of the same type (copper alloy, dimensions, shape, proportions of the arms); they are two so-called “Latin” crosses because the lower arm is longer than the others; the upper right arm is also slightly longer than the horizontal ones and is perhaps surmounted by a suspension ring⁵⁰. The two crosses would come from different origins: the first would be Roman, the second Byzantine (fig. 1). Also linked to a burial from the first phase of the cemetery, there is a fragment of wall coating with engraved crosses (fig. 2). They do not seem to correspond to gravestones. The shape of the crosses is very simple.

Archaeological evidence

During the excavations and the discovery of the multiple (simultaneous) burials from the “Garage Lux” cemetery, several hypotheses have been proposed to explain their presence: a phase of reorganization of the space for the installation of new burials, or the setting up of collective tombs in response to a particular event, which is our main hypothesis aiming to the identification of the site as a crisis cemetery.

Simultaneous burials

To study the anthropological evidences for an epidemic, the topography and relative position of each bone has been observed (consistent or not with their original anatomical position). This was complemented by taphonomic observations and a reconstruction of the chronology of disarticulations, which reveals the bodies’ dynamics inside the burials (the dynamic of the decay).

The taphonomic observation of the multiple burials of “Garage Lux” support the hypothesis of the simultaneity (or quasi-simultaneity) of the corpse deposits: maintaining of anatomical connections with no (or very few) displacements, preservation of the skeletal position despite the superpositions, and the absence of sediments between individuals. There is also a special burial organization with an optimization of the space of the graves: when adults and children are together, immature subjects are placed in the empty spaces between the adults bodies (fig. 3). Each individuals was probably wrapped in a flexible container, like a shroud, since black fibers have been observed around and under the bones in most burials.

Biological studies

Burial “recruitment”

Started in June 2022, the osteological study takes into account around 304 individuals, most of them poorly preserved. The adult population of the “Garage Lux” represents half of them, most of them in an average state of conservation that allowed for primary sexual diagnosis on the coxal bones^51[51] for only 21 of them (9 women, 12 men, 69 undetermined). For the time being, we cannot discuss any age or sex selection as was proposed in other plague cemeteries; it will be done once the study is completed and the secondary sexual diagnosis made. As for age, we have here again a majority of indeterminate adult individuals (57); the others are a majority of young adults (15 individuals between 20 and 29 years of age, and 4 between 20 and 39), and eventually not so many older adults (fig. 4). However, this information has to be put in perspective and, given the number of undetermined adults, we should probably discard the representativeness of those figures for the adult population of “Garage Lux”.

The non-adult population from the “Garage Lux” does not appear to follow a “normal” mortality, compared for instance to Ledermann’s model (fig. 5). Indeed, the 0-1 year old  age-group presents really low values, whereas in a natural scheme this class is susceptible to die prematurely. Several hypotheses are possible: bone conservation (sometimes different according to the age), physico-chemical factors, the excavations’ urgency or disturbances occurred during excavations. An exclusion of the very young immatures is unlikely in Alexandria, as no such organization has been observed in the earliest or contemporary Alexandrian cemeteries. It has been suggested that, during the medieval epidemics, the youngest children would have been left to public organizations, but it is not traceable as there are no records related to such practices at that time. The quotients of 5 to 14 years old are over-represented according to the expected results and suggest an increase in mortality sufficient to be noticed in the related age-group. Although the curve is not entirely similar to a plague epidemic pattern, those results can be compared to other plague cemeteries. For instance, at the Clos des Cordeliers (5th-6th centuries, France), the probabilities of death for 1-4 year old and 20-29 year old individuals are rather consistent with a natural mortality. Same calculation for the site of Lunel-Viel (6th-7th centuries, France), where the non-adults’ distribution shows a small number of young children compared to the adolescents’ figures. In Lunel-Viel, PCR analysis had provided a Mediterranean genome of Yersinia pestis, demonstrating that the plague was responsible for that mortality crisis⁵².

Health status

In “Garage Lux”, some stress markers have been observed, with a majority of dental enamel linear hypoplasia, some Cribra orbitalia and a very few lesions of cranial hyperostosis. There is no distinction according to sex, but some differences appear according to the age at death: the histogram (fig. 6) show two different groups. For the non-adult individuals, in accordance with previous anthropological studies and medical literature, children under 5 years old have no cranial or orbital stress-markers.

The overall health status and pathology of the “Garage Lux” population is quite standard (tab. 1). Like in other anthropological studies, degenerative joint pathologies are the most numerous with 44 individuals with arthrosis and 2 osteoarthritis, quite a high frequency. Among these individuals, young adults between 20 and 29 years of age have the most higher arthrosis frequency, mainly in the vertebrae, ribs and knees. This is quite consistent with the economy of the city itself, with two labor-intensive harbors which can explain such lesions, even among the youngest. In fact, Schmörl nodules or hernias are often associated with osteoarthritis.

Injuries are illustrated by fractures, especially of the ribs and bones of the upper limbs. Specific infections indicate possible cases of tuberculosis (fig. 7) and meningitis. Among the non-specific infections, there are a few cases of osteomyelitis and abscesses, most of which are characterized by the presence of porosity or diffuse periosteal bone production (elsewhere than on the tibia).

Two cases of cancer have been identified at the moment by the presence of lytic lesions.

Several injuries, such as traumatic ligament injuries or abscesses and/or cysts, are reported but cannot be precisely defined, requiring radiographic equipment to which we do not have access.

The study of oral pathologies is also quite classical. There is a high prevalence of caries followed by a large number of ante-mortem tooth losses. There is a concordance between these pathologies, since the presence of an abscess or an ante-mortem tooth loss in an individual is systematically associated with that of caries. Ante-mortem tooth losses do not concern solely the older individuals; it is more common for the 20-29 year old subjects than for those over 40. As for dental wear, we have some cases of worn teeth by bruxism; the degree of tooth-wear seems very often related to the age of the “Garage Lux” individuals, which is not very surprising (the older individuals have the highest degrees of wear, which are very rarely found in the younger ones).

The “Garage Lux” individuals’ general health status can thus excludes the hypothesis of interpersonal violence or armed conflict. It shows a very low presence of specific infections that could explain the death of individuals, confirming our hypothesis of an epidemic-like plague that caused this mortality crisis.

The stratigraphic study

The stratigraphic levels

The study of the stratigraphy allowed us to observe that the creation of multiple burials was not the Alexandrians’ first response to the mortality crisis. The very first burials (at the level of the previous hydraulic structures) are cist burials. Two are located in sector 2, in the lower levels of the cemetery. The bones have not been collected and preserved, so we cannot say how many individuals they contained nor even calculate a minimal number (NMI). The subsequent extension of the cemetery counts a majority of individual burials and a few double cist burials (two individuals deposited simultaneously). Three graves stand out and are clearly different from the others from this level, all located in sector 2 – the largest and deepest area, and comprising between three and ten individuals⁵³. Unfortunately the skeletons from the burial of ten have not been collected and preserved and the excavation information concerning them is incomplete, so nothing can be said about their health status or a possible simultaneity of the ten corpse deposits. The second level of the cemetery also contains many individual burials, interspersed with some scarce double burials and a multiple burial of seven individuals⁵⁴. The third level contains a majority of multiple burials, before returning in a final phase (level 4) with numerous individual burials.

The establishment of stratigraphic levels makes it possible to identify four phases of use of the cemetery of the “Garage Lux”, succeeding to a first (deeper) phase of installation of first burials. The level 1 counts a large majority of individual burials. 
The intermediate levels (2 and 3) see the implantation of many multiple burials before returning to level 4 with many individual burials. It seems that the mortality crisis has been characterized by a gradual change in funeral practices, parallel to the increasing number of the dead at the peak of the crisis. It is possible to speak of a rupture compared to the classical funerary organization characterized by individual burials and the beginnings of a mortality crisis with the presence of some double and multiple burials.

The funerary organization of the cemetery

The types of tombs vary between pits, burial vault and only one amphora burial for a child (enchytrismos). The tombs were built on the same model and with similar material: the pits are dug in the embankments, according to the size of the individual to be buried. The grave is covered by marble or limestone slabs large enough to rest on the edges of the grave. The individuals are buried lying on their back (dorsal decubitus), head to the West and feet to the East, with the exception of one prone individual, possibly disturbed. The number of bodies change from one tomb to another, from one body to ten. A minimum number (MNI) of 266 individuals have been established at the time of the excavations, which we can raise to a minimum of 304 individuals after our inventory of the bone material. A preliminary osteological study was carried out on 164 of them between 2001 and 2002 by G. Alix for the second and third sectors, and by J. Siguoirt for the fourth one. Those first osteological studies counted 22 mass graves and 186 individual burials scattered all over the site.

The study of the “Garage Lux” was justified by the discovery of multiple burials (with more than one individual) on the three areas concerned by the cemetery, associated with a large number of individual burials. There are at least 145 single burials for 67 multiple burials, two “reductions” and eight burials that could not be defined due to the lack of documentations and the poor bone preservation. During the excavations and discoveries of such multiple burials, several hypotheses were proposed to explain their presence: either they were put in place to meet a particular event or need, either they correspond to a phase of redevelopment of the burial space. For us, the first hypothesis is the one to be confirmed.

According to the map realized by Patricia Rifa Abou el Nil in 2022⁵⁵, burials do not appear to be arranged with a particular pattern, as multiple burials are not separated from the individual ones (fig. 8). It shows that during an epidemic, the crisis does not necessarily lead to the exclusion of the dead, but also can allow the use of an already existing graveyard, a fact also observed for some burial grounds from the beginning of the second pandemic. This plan also clearly shows differences of burial orientation according to the different phases of the cemetery. It is particularly clear in sectors 3 and 4 – where topographical data are the most completed: burials are oriented on a South-West/ North-East axis in the oldest phases, and on a North-South axis for some graves of the most recent phase. Such an shift in burial orientation according to chronological period has precedents in Egypt, not related to mortality crisis, for instance in sector 3 of the Gabbari necropolis⁵⁶, studied by Romain Séguier⁵⁷.

Conclusion

The site of the “Garage Lux”, first cemetery found within the city walls and potentially attached to a church, has delivered around 304 individuals in both individual and multiple simultaneous burials, suggesting mortality crisis. The archaeological material points to a chronological span of time from the 5th to the 7th c. AD, corresponding to the Justinianic plague which is the best hypothesis for that mortality crisis episode affecting the “Garage Lux” cemetery. The burials are spread over four levels that can correspond to several stages or waves of the epidemic. During the first phase, classical funerary practices are maintained (individual burial, etc.) but when the number of deaths increases, multiple burials are implanted. Within those multiple burials, the simultaneity or quasi-simultaneity of the deposits is proven by the preserved anatomical connections, the absence of sediment between the individuals, the preservation of their original position and a rationalization of the space since the children are placed in empty spaces between the adults. The biological study of the skeletons rejects the hypothesis of an armed conflict and suggests the action of an epidemic, whose bacterial agent could be identified only by future aDNA analysis.

Acknowledgements

This article takes place within a PhD contract (CNRS) carried out jointly at the University Paris 1 Panthéon-Sorbonne (ED112), the Alexandrian Studies Center (CNRS/Ifao, UAR 3134 CEAlex) and the Musée de l’Homme (UMR 7206 EA, ABBA team). The research was funded by the French National Research Agency (PSCHEET project; ANR-19- CE27-0012).

I would like to thank all the people who helped me during the study of the osteological collections and the writing of this article.

I want to thank first my PhD directors, Dominique Castex, Marie-Dominique Nenna and Pascal Sellier for sharing their knowledge, their kindness, their advice and their corrections.

I express my gratitude to Thomas Faucher, director of the CEAlex, and to the entire CEAlex team, without whom the access to the osteological collections and to the excavations data would not have been possible, as well as for their welcome and support. A special thanks to Olivier Lempereur, Aude Simony and Patricia Rifa Abou el-Nil for their expertise and their work on the excavation documents and the archaeological material of the “garage Lux”, and to Etienne Forestier for his photographic work and his explanations about the use of the photographic equipment. Thanks to all the members of the Shallalat team for their availability, their kindness and their daily work.

I also express my gratitude to Dominique Benazeth for her expertise on Christian markers.

References

Primary sources

• Ammien Marcellin: Roman History, XXVI.10, 15-19.
• Athanase: Apology to Constance, 17-18, ed.: Szymusiak, 107 et 109.
• Épiphane: Panarion, II, 69, 2, ed. : Holl., 153, in : Gascou 2020, 49.
• Évagre: Histoire ecclésiastique, IV.30, ed. : Du Cerf, coll, Sources Chrétiennes, 424.
• Michel le Syrien: Chronicles, I, VII.1, ed. : E. Leroux, Paris, 347.
• Procopius de Césarée: Histoire de la guerre contre les Perses, II.22.
• Pseudo-Denys: Chroniques de Zuqnîn, XXXII, Liverpool University Press, 198.
• Strabon: Géographie, XVII, 14, trad. Laudenbach, CUF, Paris, 2015.
• Thucydide: Histoire de la guerre du Péloponnèse, XLVIII-XLIX, trad. Romilly, Bodin, Well, Les Belles Lettres, Paris, 1990, 270-275.

Secondary sources

• Al Qahtani, S. J., Hector, M. P. and Liversidge, H. M. (2010): “Brief communication: the London atlas of human tooth development and eruption”, American Journal of PhysicalAnthropology, 142 (3), 180-191.
• Audouin Rouzeau, F. (2003): Les chemins de la peste. Le rat, la puce et l’homme, Rennes.
• Biraben, J.-N. (1975): Les Hommes et la peste en France et dans les pays européens et méditerranéens. I – La peste dans l’histoire, Paris.
• Bizot, B., Castex, D., Reynaud, P. and Signoli, M. (dir.) (2005): Saison d’une peste (avril-septembre 1590) : le cimetière des Fédons à Lambesc (Bouches-du-Rhône), Paris.
• Bruzek, J. and Murail, P. (2006): “Methodology and reliability of sex determination from the skeleton”, in: Schmitt, A., Cunha, E. and Pinheiro, J., ed.: Forensic Anthropology andMedicine, Totowa, 225-242.
• Bruzek, J., Santos, F., Dutailly, B., Murail, P. and Cunha, E. (2017): “Validation and reliability of the sex estimation of the human os coxae using freely available DSP2 software for bioarchaeology and forensic anthropology”, American Journal of Physical Anthropology, 164 (2), 440-449.
• Cabuzuelo, U. and Castex, D. (1994): “Le cimetière Saint-Pierre à Dreux (Eure-et-Loire)”, Dolmens, sarcophages et pierres tombales. Les pratiques funéraires en Eure-et-Loire dela Préhistoire à nos jours, Chartres, 68-69.
• Castex, D. (1994): Mortalité, morbidité et gestion de l’espace funéraire au cours du haut Moyen Âge, PhD Thesis, Bordeaux 1 University.
• Castex, D. (2007): “Les anomalies démographiques : clefs d’interprétation des cimetières d’épidémies en archéologie”, in: Castex, D. and Cartron, I., ed.: Épidémies et crises demortalité du passé,  Pessac, 109-138.
• Castex, D. and Cartron, I. (2007): Épidémies et crises de mortalité du passé, Pessac.
• Castex, D. (2008): “Identification and interpretation of historical cemeteries linked to epidemics”, in: Raoult, D. and Drancourt, M., ed.: Paleomicrobiology Past Human Infections, Berlin, 23-48.
• Castex, D. and Kacki, S. (2013): “Funérailles en temps d’épidémies”, Les nouvelles de l’archéologie, 132, 23-29.
• Castex, D. and Kacki, S. (2022): “Bring out your dead”, in: Knüseul, C. J. and Schotsmans E. M. J. ed.: The Routledge Handbook of Archaeothanatology, Londres, 331-352.
• Decobert, C. and Empereur, J.-Y (1998): Alexandrie Médiévale I, Études Alexandrines, 3, Alexandrie.
• Demeure, C. E. and Carniel, E. (2009): “Yersinia pestis”, Revue francophone des Laboratoires, 415, 33-39.
• Demeure, C. E., Dussurget, O., Mas Fiol, G. and Le Guern, A.-S. (2019): “Yersinia pestis and plague: an updated view and evolution, virulence determinants, immune subversion, vaccination and diagnostics”, Genes and Immunity, 20 (5), 357-370.
• Devignat, R. (1951): “Variétés de l’espèce Pasteurella pestis”, Wull World Health Organ, 4, 27, 247-263.
• Dewitte, S. N. (2006): The paleodemography of the Black Death 1347-1351, PhD Thesis, Pennsylvania State University.
• Dewitte, S. N. (2014): “the anthropology of plague: insights from biological analyses of epidemic cemeteries”, The medieval Globe, 1, 97-123.
• Dewitte, S. N. and Wood, J. W. (2008): “Selectivity of Black Death mortality with respect to preexisting health” Proceedings of the National Academy of Sciences, 105, 1436-1441.
• Drancourt, M., Roux, V., Dang, L. V., Tran-Hung, L., Castex, D., Chenal-Francisque, V., Ogata, H., Fournier, E., Crubezy and E., Raoult, D. (2004): “Genotyping, Orientalis-like Yersinia pestis and plague pandemics”, Emerg Infect Dis, 191, 607-611.
• Duday, H. [2005] (2007): “Les preuves archéologiques d’une crise brutale de mortalité : simultanéité du dépôt de cadavres, simultanéité des décès ?”, in: Castex D., Cartron, I., ed.: Épidémies et crises de mortalité du passé, Pessac, 15-21.
• Feldman, M., Harbeck, M., Keller, M., Spyrou, M. A., Rott A., Trautmann B., Scholz, H. C., Päffgen, B., Peters, J., McCormick, M., Bos, K., Herbig, A. and Krause, J. (2016): “A high-coverage Yersinia pestis genome from a sixth-century Justinianic plague victim”, Molecular Biology and Evolution, 33,  11,  2911-2923.
• Gascou, J. (2020): Églises et chapelles d’Alexandrie byzantine, Recherches de topographie cultuelle Studia Papyrologica et Aegyptiaca Parisina (StudPAP), Paris.
• Georges, P., Blanchard, P. [2005] (2007): “Les sépultures multiples du ‘35, rue de Sarrebourg’ à Bourges (18) : discussion du contexte et interprétations envisagées”, in: Castex, D. and Cartron, I., ed.: Épidémies et crises de mortalité du passé, Pessac, 147-168.
• Haas, C. (2006): Alexandria in Late Antiquity. Topography and Social Conflict, Baltimore and London.
• Hairi, I., ed. (2012): Du Nil à Alexandrie. Histoire d’eaux, Alexandrie, exposition au Musée Tessé, 26 nov. 2011-27 mai 2012.
• Harbeck, M., Seifert, L., Hänsch, S., Wagner, D. M., Birdsell D., Parise, K. L., Wiechmann, I., Grupe, G., Thomas, A., Keim, P., Zöller, L., Bramanti, B., Riehm, J. M. and Scholz, H. C. (2013): “Yersinia pestis DNA from skeletal remains from the 6th century AD reveals insights into Justinianic plague”, PLoS Pathogens, 9 (5), [En ligne] https://www.nature.com/articles/s41467-023-38393-w [consulté le 27/02/2024] .
• Kacki, S. and Castex, D. (2012): “Réflexions sur la variété des modalités funéraires en temps d’épidémie. L’exemple de la Peste noire en contextes urbain et rural”, ArchéologieMédiévale, 42, 1-21.
• Kacki, S. (2017): “Influence de l’état sanitaire de populations du passé sur la mortalité en temps de peste : contribution à la paléoépidémiologie”, BMSAP, 29 (3-4), 202-212.
• Keller, M., Spyrou, M. A., Scheib, C. L., Neumann, G. U., Kröpelin, A., Haas-Gebhard, B., Päffgen, B., Haberstroh, J., Ribera I Lacomba, A., Raynaud, C., Cessford, C., Durand, R., Stadler, P., Nägele, K., Bates, J. S., Trautmann, B., Inskip, S. A., Peters, J., Robb, J. E., Kivisild, T., Castex, D., McCormick, M., Bos, K. I., Harbeck, M., Herbiget, A. and Krause, J. (2018): “Ancient Yersinia pestis genomes from across Western Europe reveal early diversification during the First Pandemic (541-750)”, BioRwiv, 481226.
• Le Goff, J. and Biraben, J.-N. (1969): “La peste dans le Haut Moyen Âge”, Annales. Économies, Sociétés, Civilisations. 24^e année, 1484-1510.
• Leclerc, J. and Tarrete, J. (1988): “Sépulture”, in: Leroi-Gourhan, A., ed.: Dictionnaire de la Préhistoire, Paris, 963-964.
• Leroi-Gourhan, A., dir. (1988): Dictionnaire de la Préhistoire, Paris.
• Morelli, G., Song, Y., Mazzoni, C. J., Eppinger, M., Roumagnac, P., Wagner, D. M., Feldkamp, M., Kusecek, B., Vogler, A. J., Li, Y., Cui, Y., Thomson, N. R., Jombart, T., Leblois, R., Lichtner, P., Rahalison, L., Petersen, J. M., Balloux, F., Keim, P., Wirth, T., Ravel, J., Yang, R., Carniel, E. and Achtman, M. (2010): “Yersinia pestis genome sequencing identifies patterns of global phylogenetic diversity”, Nat Genet, 42, 1140-1143.
• Nenna, M.-D. (2016) “Alexandrie, des siècles d’histoire urbaine”, Dossiers Archéologie, special number 374, 2-12.
• Schmitt, A. (2002): “Estimation de l’âge au décès des sujets adultes à partir du squelette : des raisons d’espérer”, Bulletins et Mémoires de la Société d’Anthropologie de Paris, 14, 51-73.
• Sellier P., Castex D. and Houet, F. (1995): “Principes de distribution des classes d’âge non adultes et mise en évidence des anomalies démographiques dans les populations archéologiques”, Biodémographie et évolution humaine, Actes du colloque de Florence, avril 1995.
• Sellier, P. (1996): “La mise en évidence d’anomalies démographiques et leur interprétation : population, recrutement et pratiques funéraires du tumulus de Courtesault”, in: Piningre, J.-F., ed.: Nécropoles et société au premier âge du Fer, Paris, 188-202.
• Signoli, M. (1998): Étude anthropologique de crises démographiques en contexte épidémique : aspects paléo et biodémographiques de la peste en Provence. Phd thesis, Aix-Marseilles University.
• Straley, S. C. and Cibull, M. L. (1989): “Differential clearance and host-pathogen interactions of YopE- and YopK – YopL – Yersinia pestis in BALB/c mince”, Infection andimmunity, 57, 1200-1210.
• Tzortzis, S. and Rigeade, C. (2008): “La posture des cadavres de pestiférés : reflet d’une persistance et/ ou d’une transgression des normes funéraires ?”, Corps, 5, 23-32.
• Wagner D. M., Klunk J., Harbeck M., Devault A., Waglechner N., Sahl J. W., Enk J., Birdsell D. N., Kuch M., Limubao C., Poinar D., Pearson T., Fourment M., Golding B., Riehm J. M., Earn D. J. D., Dewitte S. N., Rouillard J.-M., Grupe G., Wiechmann I., Bliska J. B., Keim P. S., Scholz H. C., Holmes E. C. and Poinar H. (2014): “Yersinia pestisand the Plague of Justinian 541-543 AD: a genomic analysis”, The Lancet Infectious Diseases, 14, 319-326.
• Yersin, A. (1899): “Rapport sur la peste bubonique à Nhatrang (Annam)”. Annales de l’Institut Pasteur, XIII, 1899, 226-246; 229-230.

Notes

1. “qui est devenue une réalité immédiate, pressante”, Vie Publique, 16 mars 2020 : https://www.vie-publique.fr/discours/273933-emmanuel-macron-16-mars-2020-coronavirus-confinement-municipales.
2. Yersin 1894.
3. Kacki 2017.
4. Demeure & Carniel 2009, 25.
5. Cabuzuelo & Castex 1994; Castex 1994; Signoli 1998; Bizot et al. 2005.
6. Castex 2007.
7. Tzortis & Rigeade 2008; Kacki & Castex 2022.
8. Demeure et al. 2019, 357-358.
9. Peste et sociétés humaines : émergence, évolution et transformations bio-culturelles. 
   ANR-19-CE27-0012-PSCHEET, direction D. Castex.
10. Dewitte & Wood 2008; Dewitte 2010, 2014.
11. Duday 2007.
12. Leclerc & Tarrête 1988.
13. Georges & Blanchard 2007.
14. Castex & Kacki 2022.
15. Castex 2008; Castex & Kacki 2013.
16. Schmitt 2002; Bruzek & Murail 2006; Al Qahtani et al. 2010.
17. Castex 2007, 110.
18. Masset 1973; Sellier 1996.
19. Sellier, ibid.; Sellier et al. 1995; Castex 2008.
20. Castex 2007, 111.
21. Devignat 1951.
22. Drancourt 2004.
23. Harbeck et al. 2013.
24. Straley & Cibull 1989; Kacki 2016.
25. Morelli et al. 2010; Kacki, ibid.
26. Wagner et al. 2014.
27. Feldman et al. 2016.
28. Keller et al. 2018.
29. Thucydide, Histoire de la guerre du Péloponnèse, XLVIII-XLIX. Trad. Romilly, Bodin, Well, Paris, ed. Les Belles Lettres, Paris, 1990, 270-275.
30. Évagre le Scholastique, Histoire ecclésiastique, IV.30; Jean d’Éphèse, Chronicles, extract in Michel the Syrian, Chronicles, tome I, VII.1 and Pseudo-Denys, Chronicles of Zuqnîn, XXXII; Procope de Césarée, War Historyagainst Persians, II.22.
31. Le Goff & Biraben 1969; Biraben 1975.
32. Audoin-Rouzeau, ibid., p. 23.
33. Le Goff & Biraben 1969.
34. Biraben 1975.
35. Kacki 2016.
36. Biraben 1975.
37. Harbeck et al. 2013.
38. Haas 2006.
39. Ammian Marcellin, Roman history, book XXVI, chap. X, 15-19.
40. Decobert & Empereur 1998; Nenna 2016.
41. The garage was destroyed for the construction of a commercial building.
42. Gascou 2020.
43. Network of underground water sewers dug into the rock.
44. Hairy 2012.
45. Gascou 2020.
46. “The Arian bishop Gregory the Melitian began this building to the glory of Constance”. Épiphane, Panarion, II, 69, 2, ed. Holl.,153, in: Gascou, ibid., p. 49.
47. “Gregory the Melitian and Arianus began to build a cult place to the glory of Constance”. Athanase, Apology toConstance, 17-18, ed. Szymusiak, 107 and 109, in: Gascou, ibid., p. 49.
48. A factura is an imperial fabric of weapons, a zygostate is defined as a coin weigher and a tachograph cabinet is writing workshop.
49. Writing communication from Marie-Dominique Nenna (April 26th, 2023), to whom we address our many thanks.
50. Writing communication from Dominique Benazeth (January 25th, 2023), to whom we address our many thanks.
51. DSP: Bruzek et al. 2017.
52. Castex 2007; 2008.
53. Ft. 2132 (4 individuals), Ft. 2164 (3 individuals) et Ft. 2165 (10 individuals).
54. Ft. 2121.
55. We offer many cordial thanks to Patricia Rifa for her outstanding work and her cheerful reception.
56. The district of Gabbari is located around 800 m away from the ancient wall, in the Necropolis described by Strabo in 25 BC (Geography, XVII). It was discovered in 1996 during the construction of a road linking the western gate of Alexandria to the Cairo highway and revealed the presence of collective underground graves, studied by the CEAlex and the Egyptian Antiquities Service between 1997 and 2000. It is a necropolis used in the long term (3th c. BC to 7th c. AD) where there are, among others, numerous ancient hypogea, loculi, cremations, and, as in the “Garage Lux”, tombs in a pit or burials. The sector 3 of the necropolis contained 198 individuals in 174 graves.
57. See the volume Necropolis 4, to be published.