Skip to content
BY 4.0 license Open Access Published by De Gruyter Open Access April 15, 2021

A Most Notable Dwelling: The Domus Romana and the Urban Topography of Roman Melite

  • Benedict Lowe EMAIL logo
From the journal Open Archaeology


This paper offers an overview of Roman Melite: the paper examines the epigraphic evidence for the topography of the urban centres of Gaulos and Melite in particular, the activities of wealthy benefactors and the civic government of the municipia through benefactions to the Temples to Apollo and Proserpina and dedications to the Imperial Cult. There is only limited evidence for the buildings themselves apart from the Domus Romana that was discovered on the outskirts of Rabat in 1881. The urban area appears to have been in decline by the fourth century AD despite the presence of a Late Roman see and Byzantine officials. The paper concludes with the abandonment of the island as a consequence of the Aghlabid sack of 870.

1 Introduction

The islands of Malta and Gozo first enter Roman history in 256 BC when a Roman army returning from Africa raided the island. Thirty-seven years later during the war with Hannibal, the Roman General Tiberius Sempronius Longus annexed the islands in preparation for his abortive assault on Carthage. After this brief interlude Malta disappears from the broader panoply of Roman history appearing only occasionally in reference to the trial of Verres, the shipwreck of St Paul, a breed of small dog that was unique to the island and the quality of its textiles.

The fullest description of Roman Malta was written by the Sicilian Greek historian Diodorus Siculus. Describing Malta in c. 40 BC Diodorus described the wealth of the island and its inhabitants (V.12.1–3): he refers to eponymous towns on each of the islands and singled out for particular praise the number and excellence of the harbours, as well as the skill of its artisans – in particular the textile workers and the opulence of the houses.

2 The Roman Municipium

According to Orosius (Hist. 4.8.5) the consul Atilius – either A. Atilius Caiatinus (cos. 258 BC) or C. Atilius Regulus (cos. 257 BC) – ‘rove about and destroyed the noble Sicilian islands of Lipara and Melite’ – ‘Atilius consul Liparam Melitamque insulaes Siciliae nobiles peruagatus euertit’. Cnaeus Naevius (Bell. Pun. IV.31–32) describes the devastation wrought by the Romans who laid waste to the island by fire and slaughter – ‘the Roman crossed over to the island of Malta, he devastates, sacks and burns the entire island, he brought to an end the business of the enemy’ – ‘transit Melitam Romanus insulam integram; urit populatur vastat, rem hostium concinnat’. Despite bringing the business of the enemy to an end – ‘rem hostium concinnat’ – Atilius’ attack seems to have been no more than a raid and it was not until the outbreak of the Second Punic War that the Romans resolved on conquering the islands. In 218 BC in preparation for his assault on Carthage, the Roman Consul, Tiberius Sempronius Longus occupied the island. According to Livy (21.51.1–2) the consul departed from Lilybaeum for Malta where upon his arrival the Carthaginian commander Hamilcar surrendered the town and the island together with nearly 2,000 soldiers – arriving Hamilcar, the son of Gisgo, the commander of the garrison, with a little less than two thousand soldiers, surrendered the town and the island – ‘advenienti Hamilcar, Gisgonis filius, praefectus praesidii, cum paulo minus duobus milibus militum oppidumque cum insula traditur’. After only a few days Longus returned to Lilybaeum selling into slavery the prisoners he had taken and releasing those of noble birth. Hamilcar's surrender ensured an easy transition that underpinned the demographic continuity from Punic to Roman rule.[1] Thereafter the islands were placed under the jurisdiction of Sicily with the status of an allied city or ‘civitas foederata’. Silius Italicus (Pun. 14.251) lists Melite amongst the allies who supported the Romans during the siege of Syracuse in 213–212 BC.

According to Diodorus (V.12.1–4) Melite, Gaulos and Cercina each possessed an eponymous urban centre as well as harbours that provided ships with protection against adverse weather. According to Pseudo-Skylax writing in the fourth century BC there were two settlements on Melite: the eponymous city of Melite and a port (Periplus 111). The two are echoed in the description of Ptolemy in the second century AD: as well as Melite, a second city is named Chersonesos, and there are two temples, the Temple of Hera and the Temple of Heracles (Geography 4.3.13). The location of each of these has been the subject of considerable debate, although Melite is normally located in the vicinity of Mdina-Rabat.[2] Despite the presence of 372 tombs generally situated along the western and south-western parts of Rabat (Said-Zammit, 2001, p. 128; Sagona, 2015, p. 220), little trace remains of the urban fabric or population.[3] Pottery recovered from excavations from the Museum Esplanade in front of Domus Romana ranges from the fourth century BC to the fourth century AD (Anastasi, 2019, pp. 19–21).

Something is known of the civic organisation of Melite about the time of the Roman conquest thanks to two inscriptions honouring Demetrius, the son of Diodotus of Syracuse (IG XIV.952, 953). The inscriptions were discovered in Rome before 1549 and were erected sometime in the first century and a half of Roman rule in honour of Demetrius by the Council and Assemblies of Akragas and Melite (Bonanno, 2017). The Council (synkletos) and People (demos) of Melite resolved to declare Demetrius and his progeny proxenos and benefactor – πρóξɛνον καὶ ɛὐɛργέτη – of the people of Melite, because of his virtue and benevolence towards the Maltese. Three magistrates are named in the decree: the priest or hierothutes Hiketas, and two archontes Hereas and Kotes. No details were given of Demetrius’ actions of behalf of Melite, although proxenia was an honorific by which a polity expressed gratitude towards a foreigner for services rendered in his place of residence. It is possible that the two archontes are equivalent to the Punic suffetes. A Punic inscription from Gozo of uncertain date (sometime between the fourth and second centuries BC) refers to the existence of Punic magistrates or rabs drawn from members of the senate, ɔdrm (CIS I.132). The inscription also records the actions of the collective legislative body of the people of Gaulos – m Gwl – in this case renovating four temples: two to unidentified deities, one to Astarte and one to an otherwise unknown deity Ṣdmbl (Heltzer, 1993, pp. 202–203).

Like any other Roman town Mdina must have had its share of public buildings. In 1881 Caruana mentioned the presence of the ruins of a theatre and temple on Triq il-Villegaignon (Caruana, 1881, pp. 10–11, 1882, p. 89). This is the same temple that is referred to in a second century AD inscription that was found near the Benedictine monastery of St Peter on Triq il-Villegaignon in 1747. The inscription records the construction of a temple of Apollo and the payment for the podium, floor, 4 columns of the portico and the flanking pilasters (CIL X.7495). The name of the individual responsible for the benefaction is unidentified although was evidently of some distinction being described as ‘primus omnium Melitensium’ – the first of all the Melitenses. Another inscription found near the same monastery in 1868 records the construction of a marble temple by Claudius Iustus, the patron of the municipality (CIL X.8318). The name of the deity to whom the temple is dedicated is missing, however, the proximity of the two inscriptions may suggest that the two refer to the same building (Bonanno, 2005, p. 206; Ashby, 1915, pp. 31–32).

In 1613 an inscription was discovered on Mtarfa Hill recording a Temple of Proserpina: the inscription was found along with other remains of the temple on Mtarfa Hill where a statue of St Nicholas now stands (Abela, 1647, p. 209; Caruana, 1882, pp. 88–89, 135–136). The inscription, dated to the Augustan period, records the restoration of the Temple of Proserpina by Chrestion, a freedman of Augustus and procurator of the Maltese Islands (CIL X.7494; Busuttil, 2016, pp. 60–61; Bonanno, 2005, p. 204). Proserpina was a goddess particularly popular in Sicily: the whole island was consecrated to Ceres and Libera (Proserpina) (Cic. Ver. 2.4.106). The temple was evidently in a state of disrepair as Chrestion ‘restored the columns, gables and walls that were in a state of imminent collapse due to age’ – ‘columnas cum fastigiis et parietibus vetustate in ruinam imminentis restituit.’ According to Dio Cassius (53.15.3) the emperor sent procurators – either freedmen or equites – to all provinces where they were responsible for collecting public revenue and disbursing funds according to the instructions given to them. Under Augustus and Tiberius Procuratores Augusti were part of the ‘familia Caesaris’ and were responsible for administering imperial property (Sherwin-White, 1939, pp. 14–15): according to Tacitus (Ann. 4.15.2) when the procurator of Asia, Lucilius Capito, was accused of by the cities of Asia of exceeding his authority, the Emperor stated that the procurator only had jurisdiction over imperial slaves and revenues – ‘servitia et pecunias familiares’.

Nothing survives of these buildings, although in 1647 Abela reported that the streets and especially the square in front of the Cathedral were littered with fragments of marble: then along all the streets of [Mdina] you can see marble columns, others whole, others in pieces, cornices, pedestals, and capitals, and other vestiges of ancient buildings – ‘Indi per tutte le strade di essa [Mdina] si vedono colonne di marmo, altre intere, altre in pezzi, cornicioni, piedestalli, e capitelli, e altri vestigie di fabriche antiche [...]’ (Abela, 1647, p. 32; Cardona, 2008, pp. 41–42). As recently as the late nineteenth century fragments of marble and other architectural remains were still to be found in the vicinity of the Benedictine monastery and the Casa Azzopardi nearby and are depicted in the drawings of Gatto Murina street by Michele Bellanti (Cardona, 2008, p. 45). However these were all found before laws were introduced to protect the archaeological heritage: in the 1680s Gregorio Carafa – the Grand Master of the Hospitallers – the Order of St John – ordered the recycling of the remains of the Temple of Proserpina to decorate the Auberge d’Italie and the remains of the temple of Apollo were stolen by private collectors between 1710 and 1747 (Mercieca, 2014, p. 36; Cardona, 2008, p. 42; Bugeja, 2004, pp. 65–66). When Caruana visited the site in 1881 he found that no remains were left (Caruana 1882, pp. 88–89). In 2002 part of a possible podium of the temple was uncovered on Triq il-Villegainon (Testa, 2002) and remains uncovered in a limited excavation in Pjazza Mesquita may come from an earlier temple of Melqart or Baal-Hammon (Bonanno, 2016, p. 248).

Of the two temples mentioned by Ptolemy, the temple of Hera/Juno is referred to by Cicero when he recounts that the notorious governor of Sicily, Caius Verres plundered the temple of Juno that was situated on a promontory (Verr. 2.4.103–104). The temple was located at Tas-Silġ where excavations since 1963 by the Missione Archeologica Italiana and the University of Malta have uncovered dedications confirming the association of the temple with Juno-Astarte (Amadasi Guzzo, 1993). There may also have been a shrine to the Imperial Cult on the Island: a Greek inscription from the reign of Tiberius records a dedication to the deified Augustus by the eques Lucius Castricius Prudens (IG XIV.601; Busuttil, 1972, p. 165). Several inscriptions from Gozo refer to priests and a priestess of the Imperial cult. Lutatia, a sacerdos augustae, and her husband M. Livius Optatus, a flamen of Gaulos, made a dedication to the Livia – the mother of the Emperor Tiberius – in the guise of the goddess Ceres (CIL X.7501) (Figure 1).[4] The location of the second temple – dedicated to Hercules/Melqart – is unclear, although may have been at Ras ir-Raħeb. The site was visited by Themistocles Zammit in 1922 who observed a tessellated pavement and scarce structural remains that were worthy of future investigation (MAR, 1922–1923, p. 5). Unfortunately the site does not match the co-ordinates given by Ptolemy and the dating and function of the site remains unclear, however, excavations between November 1961– May 1962 uncovered limestone foundation blocks together with votive figurines and an ivory plaque dating to the last quarter of the sixth century BC decorated with the depiction of a boar, an animal that is familiar from Hercules’ slaying of the Erymanthian Boar (Vella, 2002, pp. 88–89, 90–91; Buhagiar, 1988, pp. 70–72; Bonanno, 2016, pp. 254–259). The existence of a cult of Hercules/Melqart on the island is borne out by two bilingual marble cippi of unknown provenance that record dedications to Hercules/Melqart in Greek and Phoenician (CIS I.122, 122bis).

Both Melite and Gaulos had the status of municipia and several inscriptions refer to decurions. It is not clear when either town was granted this status, although Gaulos was enrolled in the tribus Quirina. An inscription of unknown provenance but used for some time as a statue base in the Citadel in Victoria-Rabat and dating to the second half of the second century AD refers to Varenianus Lutatius Natalis Aemilianus as patron of the municipium (CIL X.7506) (Figure 2). There is limited evidence for the existence of decurions in Melite. In gratitude for financing the construction of the Temple of Apollo referred to earlier the Decurions voted to erect a statue in honour of the benefactor (CIL X.7495). Similarly the inscription honouring Claudius Iustus is erected by the Decurions (CIL X.8318). Iustus is identified as patron of the municipium and as quattuovir. L. Castricius Prudens held the positions of ἂρξας (archon) and πρῶτος of Melite (IG XIV.601). Πρῶτος usually signified an outstanding member of the community and it is not clear if it was a magistracy per se or merely an honorific (Busittil, 1972, pp. 162–163) – perhaps equivalent to Publius, the chief of the island, who entertained St Paul after his shipwreck on the island (Acts 28.7). Prudens may have been Patron of the community representing the interests of the Maltese in Rome (Busittil, 1972, pp. 163–164). Cicero recounts the story of a wealthy citizen of Melite, Diodorus, who was resident in Lilybaeum at the time of Verres’ governorship when his collection of silver plate became the object of Verres’ avarice (Cic. Ver. 2.4.39–41). Not only was Diodorus wealthy and enjoyed close ties with both Malta and Sicily, but was also able to appeal to friends in Rome for assistance. It is not clear why Diodorus had relocated to Lilybaeum – perhaps he had married a local girl, or he may have moved to take advantage of the greater social and commercial opportunities presented by that port city as opposed to Melite (Busuttil, 1968, pp. 33–34). Cicero refers to an additional Maltese citizen engaged with wider Roman politics: in the autumn of 46 BC Cicero wrote to Rex – either Q. Marcius Rex or P. Rupilius Rex – on behalf of his old friend A. Licinius Aristotelis of Melite, a long-standing host of mine, and in addition connected by great intimate acquaintance – ‘antiquissimus est hospes meus et praeterea coniunctus magno usu familiaritatis’ (Ad Fam. 13.52). In the letter Cicero states that he had secured Aristotelis’ liberty from Caesar and that he had remained loyal to the cause for longer than either Cicero or Rex – perhaps referring to his support for Pompey and Cicero's securing of a pardon from Caesar (Busuttil, 1967; Cassia, 2008, p. 157). Aristotelis must have been a man of some wealth and attainment to have been included within the circle of Cicero's friends. Evidence is more plentiful for Gozo with several inscriptions referring the presence of Decurions: in AD 211 the ordo decurionum honoured Geta the short-lived brother of Caracalla (CIL X.7503). In the second half of the second century AD in accordance with a decree of the Decurions the people of Gaulos – the plebs Gaulitana – set up an inscription honouring C. Vallius Postumus (CIL X.7507; Azzopardi, 2006) (Figure 3). Postumus was both patron of the municipium and flamen perpetuus of the deified Hadrian, as well as having held all the offices in the civic administration of Gaulos including having been Quinquennalis. Postumus is named as patron of the municipium in an inscription honouring him and his son, M. Vallius Rufus (CIL X.7508). Two fragmentary dedications to the emperors Constantius and Maximian in AD 305–306 were erected by Pollio and Rufus who are identified as tresvirs and duovirs (CIL X.7504-7505). It is not clear if the plebs Gaulitana were equivalent to the Punic m. No trace remains of any public buildings associated with the administration of the municipia.

Figure 1 Dedication to the empress Livia (CIL X.7501) (Photo: author, courtesy of the Gozo Museum of Archaeology).
Figure 1

Dedication to the empress Livia (CIL X.7501) (Photo: author, courtesy of the Gozo Museum of Archaeology).

Figure 2 Statue base referring to Varenianus Lutatius Natalis Aemilianus as patron of the municipium of Gaulos (CIL X.7506) (Photo: author, courtesy of the Gozo Museum of Archaeology).
Figure 2

Statue base referring to Varenianus Lutatius Natalis Aemilianus as patron of the municipium of Gaulos (CIL X.7506) (Photo: author, courtesy of the Gozo Museum of Archaeology).

Figure 3 Inscription honouring C. Vallius Postumus (CIL X.7507) (Photo: author, courtesy of the Gozo Museum of Archaeology).
Figure 3

Inscription honouring C. Vallius Postumus (CIL X.7507) (Photo: author, courtesy of the Gozo Museum of Archaeology).

Until the discovery of the Domus Romana in 1881 there was little evidence to support Diodorus’ description of the elaborate houses on the the island. The Domus Romana is located at the south-western periphery of Mdina overlooking Mtarfa to the west. It was here on February 3, 1881 that fragments of a mosaic were discovered during the planting of trees along the esplanade fronting the fortifications of medieval Mdina (Bugeja, 2004, p. 53). News of the discovery was brought to the Superintendent of Public Works and subsequent excavations under the direction of Antonio Caruana, the Librarian of the Public Library, revealed a luxurious suburban mansion with elaborate mosaics arranged around a peristyle (Caruana, 1882, p. 91).

Construction of the Domus was dated from the last quarter of the second century or first quarter of the first century BC (Bonanno, 2018, p. 255; Sagona, 2015, p. 268). The house is built along Hellenistic lines centering upon a colonnaded courtyard with sixteen painted Doric columns, around which the principal reception rooms were arranged. Peristyle houses originated in Greece where they are known from the fifth century BC before spreading across the wider Greek world during the Hellenistic period. Over 90 such houses have been excavated on Delos mostly postdating 167 BC when the Romans placed the island under the jurisdiction of Athens. The incomplete layout of the Domus means that it is impossible to ascertain the relationship of the house to adjoining properties. Antonio Caselli (2008, pp. 36–37) has suggested that the axis of the property lay to the north-west of the peristyle with rooms A and B serving as atrium and tablinum accessed through the fauces to the west.

The Domus is structured around the peristyle that was decorated with a mosaic depicting two doves drinking from a bowl surrounded by a geometric tromp l’oeil border (Figure 4). Mosaics depicting doves are found elsewhere for example from an oecus or dining room in the House of the Mosaic Doves (VIII.2.34) in Pompeii and the palace of the emperor Hadrian at Tivoli. They may have copied a masterpiece from Pergamum by Sosus that is described by the elder Pliny (Nat. 36.60).[5] Fragments of fresco were found by Zammit in 1922 (MAR, 1921–1922, p. 4). The fragments are Pompeian First Style – a design that imitated the appearance of exotic marble and was popular from the third century BC until c. 80 BC although it continued to be used in Pompeii until the destruction of the town in AD 79. It created an impression of refined dignity and may have been intended to evoke the appearance of Greek public buildings (Lorenz, 2015, pp. 256–257) highlighting the lack of distinction between the public and private functions of Roman houses.

Figure 4 Peristyle mosaic, Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 4

Peristyle mosaic, Domus Romana (Photo: author, courtesy of Heritage Malta).

The Domus is also notable for the discovery of several fragments of statues including several members of the Imperial family dating c AD 50 – the emperor Claudius (Figure 5), the emperor's daughter Claudia Antonia (Figure 6), and a young boy wearing a toga – perhaps the young, future emperor, Nero shortly after his adoption by Claudius in AD 49 (Figure 7). Statues of the good emperors or members of the imperial family are quite common in private houses, for example, the statue of Augustus from a villa on via Labicana, and that of Livia from the Villa of the Mysteries in Pompeii with the most extreme example coming from the Late Imperial villa at Chiragan on the banks of the river Garonne. During the nineteenth century more than 200 sculptures were excavated from the Imperial villa dating from the first century BC to the fourth/fifth centuries AD including a collection of statues of emperors from Augustus to Philip the Arab. The villa may have been owned by a member of the local elite with ties to Rome – perhaps during the revolt of Postumus c AD 260 (Attanasio, Bruno, & Prochaska, 2016, pp. 173).[6] The reasons for such a collection are unknown – perhaps reflecting a debt of gratitude, or purported relationship to the emperor in question – thus the Villa of the Mysteries is thought to have belonged to the family of Livia. Whilst no such connection can be shown in the case of the the Domus, it is tempting to see the building serving a public function, perhaps as the residence of a magistrate (Bonanno, 1997, p. 64, 2018, pp. 255–256, 258; Cassia, 2008, p. 141; Bruno, 2009, pp. 32–34). According to Vitruvius (6.5.2) a house should reflect the status of its owner with the houses of magistrates modeled upon the public buildings within which they fulfil their duties – ‘because in their houses public counsel and private judgements and decisions are often made’ – ‘quod in domibus eorum saepius et publica consilia et privata iudicia arbitriaque conficiuntur.’ Several houses in Pompeii and Herculaneum imitate elements of public architecture with the use of columns in particular blurring the distinction between private and public space (Wallace-Hadrill, 1988, pp. 59–69). There is little to connect the houses in Pompeii and Herculaneum with nobiles – the only known senator from either town is M. Nonius Balbus, the Proconsul of Crete and Cyrenaica, and the only extant houses that can be connected with senators are the Villa of the Papyri at Herculaneum and Villa A at Oplontis. Instead they are probably the houses of members of the municipal elite – decuriones – and rich freedmen. According to Cicero's On Duties (1.139) a man's dignity could be enhanced by an appropriate house – ‘ornanda enim est dignitas domo.’ The house of the wealthy freedman Trimalchio includes architectural elements emulating the houses of his aristocratic neighbours – private baths, porticoes and an atrium decorated with autobiographical paintings and scenes from the Iliad and Odyssey (Satyricon 29, 72–73). Despite the aspirations of his house the owner of the Domus Romana may not have been as wealthy as first appears: rather than using marble for the columns of the peristyle the appearance of faux marble is created using limestone covered with a thin layer of painted stucco (Bonanno, 2018, p. 256).

The importance of the remains led to the construction of a small museum building to protect part of the ancient house. The expansion of the museum building in the early 1920s led to further excavations by Sir Themistocles Zammit between 1920–1925 concentrating on the area to the north-west of the Roman Domus where Zammit uncovered remains of humbler houses and a section of Roman road dating to the third-fourth centuries AD (Figure 8). The buildings were coarsely built using reused material (MAR, 1924–1925, p. 3). Some decorative elements were found including fragments of inscriptions and statuary although their original setting is unknown (MAR, 1922–1923, p. 5–6). The area may have been used for industrial activity: finds of needles and loom weights used in textile manufacturing (MAR, 1924–1925, p. 4); clay moulds used for crucibles suggests that metalworking took place (Sagona, 2015, p. 271; MAR, 1922–1923, p. 6).

Figure 5 Statue of the Emperor Claudius from the Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 5

Statue of the Emperor Claudius from the Domus Romana (Photo: author, courtesy of Heritage Malta).

Figure 6 Statue of Claudia Antonia from the Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 6

Statue of Claudia Antonia from the Domus Romana (Photo: author, courtesy of Heritage Malta).

Figure 7 Statue of a young boy wearing a toga – the future emperor Nero – from the Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 7

Statue of a young boy wearing a toga – the future emperor Nero – from the Domus Romana (Photo: author, courtesy of Heritage Malta).

Figure 8 Housing to the north-west of the Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 8

Housing to the north-west of the Domus Romana (Photo: author, courtesy of Heritage Malta).

The Domus appears to have extended to the east. Unfortunately no record was made of any finds recovered in 1889 when a road was cut through the site to the Museum Railway Station between Rabat and Mtarfa (Zammit, 1922, p. 131). Although Zammit excavated three trenches in the fields to the east of the train station road, few details were published. Zammit reported a large platform made of reused materials, a cistern and a room supported by four pillars (Zammit, 1922, p. 132). In June 2019 the Melite Civitas Romana Project conducted remote sensing profiles using Ground-Penetrating Radar in the fields to the north and northwest of the Domus (Areas A and B) and along the façade of the museum building (Area E) in order to ascertain the extent and location of Zammit's excavations and to place these excavations within the broader urban fabric (Brown et al., 2019).

The square and public parking in the Melita Esplanade have yielded extensive archaeological remains. The most significant discoveries were made in two short seasons conducted by staff from UCLA in 1983–1984. The excavations have only recently been published by Maxine Anastasi and found a short section of street with the frontage of two buildings constructed of rectangular ashlar blocks facing onto the street (Anastasi, 2019, pp. 12–13). The two structures appear to have been constructed in the first century AD. The buildings were abandoned in the middle of the fourth century: the latest pottery dates to the early-middle of the century, and the last coin dates to the reign of Constantius II (AD 337–361) (Anastasi, 2019, p.19).

Fragmentary remains of other houses have been found at several locations in Rabat. Caruana reported finds of houses and mosaic floors from 1867 in Triq Bir il-Ljun (Caruana, 1882, pp. 91–92; Ashby, 1915, pp. 39–47). In 1965–66 and 1969 remains of a substantial Roman house were discovered in Saqqajja Square (Saqqajja I and II) with floors supported by upturned amphorae. Finds of African Red Slip ware dates the house to the first-second centuries AD (Sagona, 2015, pp. 272–273). In 2019 a possible Roman domus was discovered at Triq il-Mitħna in Rabat with several large ashlar blocks and rock-cut rooms arranged around a central axis (Annual Report, 2019, p. 29). Several wealthy houses may have stood around the fringes of the Roman town. Scattered Roman finds from the vicinity of Ix-Xagħra in Ġnien is-Sultan suggest the existence of an elaborate building with a shrine, wall foundations, a large water tank, pottery, coloured marble, tiles, mosaic fragments and three cornice slabs (Ashby, 1915, pp. 48; Sagona, 2015, pp. 273–274). Abela refers to the remains of a Roman building – variously identified as either bathhouse or mausoleum – a stone's throw from the city walls at Għajn Ħammiem (Abela, 1647, pp. 33–36; Teuma, 2005, pp. 88–89; MAR, 1922–1923, p. 7; Zammit, 1922, pp. 133–134). Remains of a gate or tower of the city wall were found on Triq Nikol Sawra (Saqqajja II) with a porticoed building containing shops and dwellings (Sagona, 2015, p. 273). Large ashlar blocks and a rectangular rock-cut pit of possible ritual use dating to the second-third centuries AD have been found in the Sta Margherita Cemetery (Annual Report, 2006, p. 11).

The date of the abandonment of the Domus is unclear. The state of the excavations provides little evidence for anything but the most general dating of the house. Several of the pieces of Samian pottery recovered by Zammit bore makers’ marks dating to the first century AD (Zammit, 1923, p. 223). At some point the house was remodeled with the mosaics in rooms A and B to the north-west and north of the peristyle being raised and an early pavement being visible beneath the floor of room A (Caselli, 2008, p. 35). The building may have been abandoned as early as the second half of the first century AD (Bonanno, 2005, p. 214). Hasty repairs using cheaper ceramic tiles were carried out in the mosaic pavement of the oecus or tablinum at the end of the first century AD or beginning of the second century AD (Figure 9). Occupation seems to have continued at least in the vicinity however. Caruana (1881, p. 8) reported finding coins dating to the third and fourth centuries AD and several fourth century coins were found during Zammit's excavations of the houses to the north of the Domus (MAR, 1924–1925, p. 4). According to Zammit the most common pottery was imitated sigillata dating to the fourth century AD (MAR, 1921–1922, p. 4). Rubble and fallen blocks suggest that the buildings uncovered in the Melita Esplanade had collapsed by the second half of the fourth century (Anastasi, 2019, p. 19).

Figure 9 Repaired mosaic from the Domus Romana (Photo: author, courtesy of Heritage Malta).
Figure 9

Repaired mosaic from the Domus Romana (Photo: author, courtesy of Heritage Malta).

That Melite continued to be the seat of a Bishop is clear from two letters of Pope Gregory the Great: in AD 592 the Pope wrote to Lucillus, the Bishop of Melite to resolve a problem concerning Church property in Africa that was administered by the Bishop (Ep. 2.36). In AD 598 a further letter was sent to Johannes the Bishop of Syracuse requesting intervention by the Defensor Siciliae – the official responsible for civil disputes and administering ecclesiastical patrimony – to deal with Lucillus’ appropriation of church property – furnishings and money (Ep. 9.25). A Byzantine notitia episcopatum dating between AD 730–780 lists Melite as a suffragan see within the province of Sicily (Brown, 1975, p. 80). Remains of a Byzantine monastery were found in a field in front of the Abbatija tad-Dejr catacomb: a fragment of a polichrome mosaic in black, red, yellow and white tesserae were found as well as fragments of limestone columns (MAR, 1933–1937, p. 9). The islands – now referred to collectively as Gaudomelete – are mentioned by the Patriarch Nikephoros as the seat of a Byzantine military official – a δούξ. In AD 636–637 following an abortive coup against the Emperor Heraclius the emperor's nephew Theodorus was exiled to the islands with instructions that one of his feet should be amputated on his arrival (Niceph. Hist. Comp. 24.14–17; Brown, 1975, p. 76; Buhagiar, 1997, p. 119). A no-longer extant seal refers to the presence of a Byzantine Duke or drungarios – a local military commander usually in command of a fleet (Schlumberger, 1900, p. 492). Like other islands in the central Mediterranean Malta became increasingly exposed to attack as Byzantine power declined.

According to the fourteenth century geographer Muhammad bin Abd al-Munim al Himyari the islands were devastated by the Aghlabid invasion in AD 870 and remained uninhabited until they were resettled by the Arabs after AD 1048–1049 (Cooperson, 2015; Bruno, 2009, p. 11). According to al Himyari the Muslims captured the fortress of Malta, taking the Byzantine governor Amros (Ambrosios?) prisoner. They demolished the fortress and looted and desecrated whatever they could not take away (Brincat, 1995, p. 11). The geographer Ibn Hawqal – visiting Sicily in the tenth century – says that Malta was empty except for savage donkeys, sheep and honey, and was visited only occasionally by Arabs seeking to exploit the Island's resources (Luttrell, 1987, pp. 158–159). Recent archaeological work is throwing into question such a prolonged period of abandonment. In 2008 excavations at the base of the sixteenth century town walls at Mdina revealed occupation dating to the Early Bronze Age, Early and Late Middle Ages. The early Medieval structure consisted of the glacis of a city wall situated further up the slope under the later walls and cut by a fifteenth century artillery platform (Bruno et al., 2018, p. 112).[7]

That the site of the Domus was definitely abandoned by the eleventh century is clear from Zammit's discovery of Saracenic burials along the front of the museum building (Zammit, 1923, p.219; MAR, 1921–1922, p. 3; MAR, 1924–1925, p. 4). The cemetery contained at least 245 burials aligned east-west. The burials give little indication of date – other than the tombs themselves the only find was a silver ring with the inscription ‘Rabbī Allāh wāḥid’ – ‘My Lord is one God’ (Zammit, 1923, p. 219, 1925, p. 2). The burials are primarily pit or cist graves surmounted by prismatic tombstones of a type known as mqabriyyah that originated in the Maghreb. They are mostly carved from local Globerigerina Limestone and bear a Nash or Kufic inscription usually from the Koran although some tombs gave the details of the deceased (Thane, 2016, pp. 221–222; Rossi, 1929, pp. 437–444).

3 Conclusions

Epigraphic evidence for the involvement of Patroni and Primi Melitensium reflects the participation of the local elite in the funding of public buildings as means of social advancement and status symbolism. Wealthy individuals like Chrestion and Claudius Iustus invested in the maintenance of public buildings. Apart from Chrestion and Prudens the individuals named in inscriptions were neither magistrates nor imperial officials – perhaps the political isolation of the islands and the lack of a significant urban government meant that the burden for public munificence fell elsewhere. It is striking that the Temple of Proserpina should be neglected to such an extent and Chrestion's intervention is a reflection of the wider program of investment and religious revival that underpins the Augustan period.

The continued presence of Late Roman Bishops and Byzantine officials implies the continued existence of an urban centre at Melite, however, the abandonment of the Domus Romana and the neighbouring structures to the north and south raises the likelihood that the urban centre was already in decline by the fourth century AD. Late Roman Bishops could be residual and the presence of a Byzantine military official suggests the island was a peripheral location exposed to Muslim attack. The site of the Domus was vacant by the ninth century when the area was reused as a necropolis.

Article note: This article is a part of the Special Issue on At the Crossroads of the Mediterranean: Malta and the Central Mediterranean During the Roman Period, edited by Davide Tanasi, David Cardona, & Robert Brown.



Museums Annual Reports. Reports on the Working of Government Departments. 1904–2002. Available online:

Annual Report:

Superintendence of Cultural Heritage. 2003 onwards. Annual Report. Available online:


This research stems from the work of the Melite Civitas Romana Project – an international collaboration between Heritage Malta, the University of South Florida, the University of North Alabama and Intercontinental Archaeology and I would like to acknowledge the assistance of all our institutional partners including the Superintendence of Cultural Heritage for their support of our research. I would particularly like to thank successive Deans of the College of Arts and Sciences at UNA, Dr Carmen Burkhalter and Dr Sarah Lynn Baird, and the Provost, Dr Ross Alexander, for their financial support. I would like to thank my colleagues David Cardona, Robert Brown, Andrew Wilkinson and Davide Tanasi for their comments, as well as Professor Anthony Bonanno and Walter Prochaska for their help in providing references.

  1. Conflict of Interest: Author states no conflict of interest.


Abela, G. (1647). Della descrittione di Malta isola nel mare Siciliano con le sue antichita ed altre notitie. Malta: Paolo Bonacota.Search in Google Scholar

Amadasi Guzzo, M. G. (1993). Divinità Fenicie a Tas-Silġ, Malta – I Dati Epigrafici. Journal of Mediterranean Studies, 3(2), 205–214.Search in Google Scholar

Anastasi, M. (2019). Pottery from Roman Malta. Oxford: Archaeopress. in Google Scholar

Ashby, T. (1915). Roman Malta. The Journal of Roman Studies, 5, 23–80. in Google Scholar

Attanasio, D., Bruno, M., & Prochaska, W. (2016). The Marbles of the Roman Villa at Chiragan at Martres-Tolosane (Gallia Narbonensis). Archaologischer Anzeiger, 1, 169–200.Search in Google Scholar

Azzopardi, G. (2002). Quintius Lutatius Longinus and his wife Lunia in a funerary inscription from Gozo. Malta Archaeological Review, 6, 50–52.Search in Google Scholar

Azzopardi, G. (2005). The Vallii in Gozo. The Malta Historical Society, 14(2), 230–238.Search in Google Scholar

Balty, J.-C. (2015). Sculpture Romaine et Domaine Impérial de Chiragan (Haute-Garonne). Revue Archéologique, 1, 131–136.Search in Google Scholar

Bonanno, A. (1997). The Imperial portraits from Malta: their contextual significance. In J. Bouzek & I. Ondřejová (Eds.), Roman Portraits: Artistic and Literary (pp. 62–64). Mainz: Verlag Philipp von Zabern.Search in Google Scholar

Bonanno, A. (2004). Malta during Phoenician, Roman and Byzantine times: outside influence and original traits. In K. Gambin (Ed.), Malta - Roots of a Nation. The Development of Malta from and Island People to an Island Nation (pp. 45–54). Valletta: Heritage Malta.Search in Google Scholar

Bonanno, A. (2005). Malta: Phoenician, Punic and Roman. Malta: Midsea Books.Search in Google Scholar

Bonanno, A. (2016). The Cult of Hercules in Roman Malta: a discussion of the evidence. Melita Classica, 3, 243–263.Search in Google Scholar

Bonanno, A. (2017). A friend in high places: Demetrios, son of Diodotos, the Syracusan. Melita Classica, 4, 11–58.Search in Google Scholar

Bonanno, A. (2018). Roman Villas in the Maltese Archipelago. In A. Marzano & G. Métraux (Eds.), The Roman Villa in the Mediterranean Basin (pp. 255–265). Cambridge: Cambridge University Press. in Google Scholar

Brincat, J. M. (1995). Malta 870–1054: Al-Himyari's Account and its Linguistic Implications. Valletta: Said International.Search in Google Scholar

Brown, R., D. Cardona, D., De Giorgi, L., Leucci, G., Lowe, B., Persico, R., Tanasi, D., & Wilkinson, A. (2019). Melite Civitas Romana project: Preliminary results from GPR survey. 2019 IMEKO TC4 International Conference on Metrology for Archaeology and Cultural Heritage, MetroArchaeo 2019. (pp. 351–354). Retrieved from: (accessed 28 June, 2020).Search in Google Scholar

Brown, T. S. (1975). Byzantine Malta: A Discussion of the Sources. In A. T. Luttrell (Ed.), Medieval Malta. Studies on Malta before the Knights (pp. 71–87). London: British School at Rome.Search in Google Scholar

Bruno, B. (2009). Roman and Byzantine Malta: Trade and Economy. Malta: Midsea Books.Search in Google Scholar

Bruno, B., & Cutajar, N. (2018). Malta between the Ninth and Tenth Century – two early medieval contexts. Archeologia Medievale, 45, 111–122.Search in Google Scholar

Bugeja, A. (2004). Ancient mosaic pavement discovered at Notabile: a historical appreciation. Melita Historica, 15(1), 53–66.Search in Google Scholar

Buhagiar, M. (1988). Two Archaeological Sites – Ras ir-RaĦeb, Malta, and Ras il-Wardija, Gozo. Melita Historica, 10(1), 69–87.Search in Google Scholar

Buhagiar, M. (1997). Gozo in late Roman, Byzantine and Muslim times. Melita Historica, 12(2), 113–129.Search in Google Scholar

Busuttil, J. (1967). Aulus Licinius Aristotelis. Melita Historica, 4(4), 257–259.Search in Google Scholar

Busuttil, J. (1968). Diodorus Melitensis. Melita Historica, 5(1), 32–35.Search in Google Scholar

Busuttil, J. (1972). Lucius Castricius Prudens. Journal of the Faculty of Arts, 5(2), 162–167.Search in Google Scholar

Busuttil, J. (2015). The Chrestion Inscription. Treasures of Malta, 62, 60–63.Search in Google Scholar

Cardona, D. (2008). The known unknown: identification, provenancing, and relocation of pieces of decorative architecture from Roman public buildings and other private structures in Malta. Malta Archaeological Review, 9, 40–50.Search in Google Scholar

Caruana, A. A. (1881). Recent Discoveries at Notabile. Malta: Government Printing Office.Search in Google Scholar

Caruana, A. A. (1882). Report on the Phoenician and Roman Antiquities in the Group of Islands of Malta. Malta: Government Printing Office.Search in Google Scholar

Caselli, A. (2008). Revisiting the Roman domus in Rabat, Malta, through a consideration of its mosaic flooring. Malta Archaeological Review, 9, 32–39.Search in Google Scholar

Cassia, M. (2008). L’arcipelago maltese sotto il dominio romano. In A. Bonanno & P. Militello (Eds.), Malta in the Hybleans, the Hybleans in Malta (pp. 133–194). Palermo: Officina di Studi Medievali.Search in Google Scholar

Cooperson, M. (2015). Al-Ḥimyarī's Account of Medieval Malta. A Reconsideration. In H. Amstutz, A. Dorn, M. Müller, M. Ronsdorf, & S. Uljas (Eds.), Fuzzy Boundaries. Festschrift für Antonio Loprieno (Vol. 1, pp. 347–351). Hamburg: Widmaier Verlag.Search in Google Scholar

Galea Debono, F. (2006). Traces of Bronze Age Settlement unearthed in Rabat. Times Malta. Retrieved from: (accessed 14 June, 2020).Search in Google Scholar

Heltzer, M. (1993). The Inscription CIS, 1, 132 from Gozo and the Political Structure of the Island in the Punic Period. Journal of Mediterranean Studies, 3(2), 198–204.Search in Google Scholar

Lorenz, K. (2015). Wall Painting. In B. Borg (Ed.), A Companion to Roman Art (pp. 252–267). London: Wiley-Blackwell. in Google Scholar

Mercieca, S. (2014). The Proserpina Temple and the History of its Chrestion Inscription. Treasures of Malta, 61, 33–39.Search in Google Scholar

Rossi, E. (1929). Le Lapidi Sepolcrali Arabo-Musulmane di Malta. Rivista degli Studi Orientali, 12, 428–444.Search in Google Scholar

Sagona, C. (2015). The Archaeology of Malta. From the Neolithic through the Roman Period. Cambridge: Cambridge University Press. in Google Scholar

Said-Zammit, G. (2001). The Phoenician and Punic Necropoleis of Rabat, Malta. Melita Historica, 13(2), 117–146.Search in Google Scholar

Schlumberger, G. (1900). Sceaux Byzantins Inédits (Quatrième Série). Revue des Etudes Grecques, 13(55), 467–492. in Google Scholar

Sherwin-White, A. N. (1939). Procurator Augusti. Papers of the British School at Rome, 15, 11–26. in Google Scholar

Tansey, P. (2008). L. Sempronius Atratinus Aug. Imp. Zeitschrift für Papyrologie und Epigraphik, 165, 304–306.Search in Google Scholar

Testa, M. (2002). New find at Mdina most important so far in old capital. Retrieved from: (accessed 17 June, 2020).Search in Google Scholar

Teuma, E. (2005). Qanat, Saqqajja & Roman Aqueduct System at Rabat, Malta. Proceedings of History Week, 2003, 75–93.Search in Google Scholar

Thake, C. (2016). Envisioning the Orient: The New Muslim Cemetery in Malta. Muqarnas, 33(1), 221–251. in Google Scholar

Vella, B. (2015). Reflections on the Literary Sources on Byzantine Malta. Melita Classica, 2, 115–119.Search in Google Scholar

Vella, C., & M. Spiteri. (2009). Archaeological Remains at Gheriexem Street, Rabat (Malta). FastiOnline. Retrieved from: (accessed 21 October, 2020).Search in Google Scholar

Vella, N. (2002). The lie of the land: Ptolemy's temple of Hercules in Malta. Ancient Near Eastern Studies, 39, 83–112. in Google Scholar

Vella, N. (2005). Phoenician and Punic Malta. Journal of Roman Archaeology, 18, 436–450. in Google Scholar

Vella, N. (2013). Vases, bones and two Phoenician inscriptions. In O. Loretz, S. Ribichini, G. E. Watson, & J. A. Zamora (Eds.), Ritual, Religion and Reason: Studies in the Ancient World in Honour of Paolo Xella (pp. 589–605). Münster: Ugarit-Verlag.Search in Google Scholar

Wallace-Hadrill, A. (1988). The Social Structure of the Roman House. Papers of the British School at Rome, 56, 43–97. in Google Scholar

Zammit, T. (1922). Excavations in Malta. The Antiquaries Journal, 2(2), 131–134. in Google Scholar

Zammit, T. (1923). Excavations at Rabat, Malta. The Antiquaries Journal, 3(3), 219–225. in Google Scholar

Zammit, T. (1925). Saracenic Remains in Malta. Melita, 5, 1–3.10.1017/S0003581500006727Search in Google Scholar

Received: 2020-07-27
Accepted: 2020-11-15
Published Online: 2021-04-15

© 2021 Benedict Lowe published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 23.9.2023 from
Scroll to top button