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

The Perseverance of Archaeology: New Data from a Rescue Investigation at Triq Fejġel in Rabat and its Contribution to the Punic and Roman Maltese Funerary Context

  • Kevin Borda EMAIL logo , Bernardette Mercieca-Spiteri , Paolo Spadaro and Carlo Veca
From the journal Open Archaeology


The Maltese funerary context during the Punic and Roman times is documented from discoveries and archaeological reports primarily from the twentieth century. Notwithstanding, documentation standards in the first half of the last century were such as to provide limited archaeological data to properly understand the context, phasing and ritual. The combination of robust policy-driven archaeological monitoring procedures together with a scientific excavation of reported discoveries is essential to provide fresh archaeological data which must necessarily be published within adequate time frames. This will by no small means contribute to the formulation of a proper national research agenda by identifying lacunae as well as giving rise to new research questions. This study draws attention to the survival of archaeology seen as limited stratigraphic contexts that have persevered through the centuries and the continuous exploitation of the site. It is a case study of the application of a stratigraphic scientific approach to a recent archaeological discovery during archaeological monitoring, providing ample data with regard to funerary reuse and associated practices and rituals together with an in-depth osteological observation of skeletal remains therein discovered.

1 Introduction

The discovery of the tomb in 2019, which will be discussed in this study, is a direct result of the predictive modelling and precautionary approach being adopted by the superintendence of cultural heritage (SCH), which is based on the limiting of ground disturbance and the imposing of archaeology monitoring during any development works within the area of archaeological importance (AAI) for Rabat and Mdina in view of the high probability of new archaeological discoveries. The SCH uses this approach during the fulfilling of its statutory obligations in the evaluation of planning applications affecting known archaeologically sensitive areas; in this instance, the urban areas of the village of Rabat and the neighbouring city of Mdina. Through historic documentary research and continuous archaeological discoveries, it has been clearly established that the present urban extents of Rabat and Mdina overlap the Roman period settlement and its necropolis, with the original demarcation between the two still clearly visible in the modern-day topography (Triq Santa Rita – “Triq” is the Maltese word for street – St Rita Street, a straight wide road, is built on the fill of the Roman period ditch which demarcated the world of the living from that of the dead, with all burials being located to its South). In effect, this demarcation appears to have already been established during the preceding Phoenician/Punic periods, while not enough data exists to understand land uses in the preceding Prehistoric and Bronze age periods.

The continued archaeological discoveries within this urban area led to the planning authority (PA) establishing an AAI for Rabat and Mdina, in 1998. This AAI is a scheduled area (scheduling is the statutory protection of heritage sites and buildings, provision for which falls under the remit of the Planning Act 2016 – previously 2010 and 1992) intended to provide all the necessary regulatory safeguards for the protection of the archaeological resource of the area from development and any other negative impacts. The establishing of the AAI permitted the introduction of a regulated approach towards archaeology protection when assessing the development planning applications for the area, with clear monitoring procedures introduced to identify, record and safeguard archaeological discoveries during the construction projects. The consequent high incidence of archaeological discoveries that were the direct result of these adopted procedures led to the need for an extension of the existing AAI. In 2013, SCH and PA cooperated by sharing data relative to archaeological discoveries for the period 1998–2012 within the established AAI and its immediate environs; these new discoveries were individually scheduled, and the AAI was further extended to include areas of archaeological interest that were previously under-represented. Following the 2013 publication of the AAI, continued and improved monitoring procedures have resulted in further archaeological discoveries within the AAI, which is the subject of this study (Figure 1).

Figure 1 
               Site at Triq Fejġel in the Rabat context.
Figure 1

Site at Triq Fejġel in the Rabat context.

2 The Site: General View

The site is located within an urban area that was newly developed between 1970 and 1980, prior to which the area was predominantly rural. There is no reporting of any archaeological discoveries within the site or its immediate environs during this period and if anything, the discovery appears to be located at a distance from the major necropolis identified in the Rabat district (at distances that vary from 250 to 230 m). The site lies along the gentle slope of the North/West side of the Taċ-Ċagħki Hill (wherein the most extensive necropolis is located ranging in date from the Phoenician/Punic to the Paleo-Christian), halfway between St Paul’s Missionary School (a Late Roman catacomb complex) and L-Abatija tad-Dejr (an extensive Paleo-Christian Catacomb later reused as an early Christian cultic centre). Due South of the site, at a distance of 45 m, is located a cluster of archaeological features consisting of a quarry area, a mutilated catacomb, cart-ruts and two rock-cut chamber tombs, all of which were scheduled when the AAI was extended in 2013. Further features recorded in the last decade include clusters of rock-cut trenches, showing a primary agrarian use of the landscape to the western side of the hill (Vjal il-Ħaddiem – “Vjal” is the Maltese word for avenue – Worker’s Avenue). The permitted redevelopment of the site under archaeological monitoring, which consisted of a garage, led to the discovery of the remains of this tomb. This is a veritable testimony of the perseverance of archaeology, where stratigraphic contexts have survived despite the number of interventions happening on-site throughout the centuries – quarrying, rifling, unregulated construction – until recently. More specifically, the tomb provided a limited stratigraphic context measuring 0.225 m3 (this volume refers to the surviving deposits discovered at the western part of the tomb), which has however yielded sufficient data to trace its use over a span of 400 years, showing cultural changes and different burial practices.

Within the site itself, apart from the tomb, several rock-cut features had been uncovered, which include ten rectangular-shaped rock-cut trenches, which probably had an agricultural function, and evident quarrying marks. The most significant features recorded on-site are the truncated tomb (the subject of this report), a rock-cut feature which partially redefines the South side of the tomb (its direct relationship with the tomb has not yet been fully established) and the quarry area to the North/West of the plot. The latter feature has only been partially exposed as it continues underneath a third-party property (Figure 2).

Figure 2 
               General layout of the plot (courtesty of QPML), plan and section/elevation of the tomb in Triq Fejġel.
Figure 2

General layout of the plot (courtesty of QPML), plan and section/elevation of the tomb in Triq Fejġel.

The tomb could not be fully assessed as a portion of it partially lies underneath a third-party property (Figure 14). Nevertheless, it is clear that the tomb is characterized by a roughly rectangular shaft (which traces are more clearly defined on its southern side) and an elongated chamber also rectangular in shape with rounded corners. The tomb was found partially truncated with the ceiling of the chamber nearly destroyed and severely damaged. Clear evidence of disturbance within the chamber, leading to its partial destruction, is also given by an evident cut within the original deposits subsequently filled with rubble, debris and inconsistent material resulting from the looting of the tomb (Figure 2 and Figure 13). Although the tomb cannot clearly be associated with one of the classified layouts, it seems to be typologically close to plans 6 and 7 as they were defined by Sagona (Sagona, 2002, pp. 237–251, p. 323, figure 3, p. 653, figure 333). Both types appear to have been in use during phases III and IV, but one should note that the chronological distinction is far from conclusive as those layouts are also encountered in tombs dating to earlier or later phases (type 7 appears to be mainly in use for tombs dating from the sixth/fifth century BC, which is the same chronological range wherein the tomb described in this report falls into).

The rock-cut feature, already referred to above, is part of a cluster of trenches, mainly featuring rectangular and elongated shapes and following a North/West–South/East orientation. Such trenches are traditionally associated with agricultural exploitation of the landscape and the site in question might have been marked by this change when the funerary function was no longer considered as the primary purpose of the area. The specific trench, located along the southern side of the truncated tomb, was possibly cut at a later stage; although a direct relationship and stratigraphic connection with the tomb cannot be fully established (the southern side of the tomb is found truncated and partially disturbed by machinery). Even though such redefinition of the landscape cannot be chronologically defined, the scarce fragments collected from the investigated feature point to a Late Roman date (potsherds in SU07 are comparable to forms dating to the second/third century AD). The different types of trenches may also suggest a continuous use of the area over a considerable period of time.

The last significant feature uncovered on-site is identified as a quarry area located along the western boundary of the plot and partially underneath a third-party property. The evident quarry marks along the eastern side (three shallow and roughly square-shaped cuts), including the steep and the vertical cuts noted in a rectangular-shaped area, point to a quarrying function. The area has further significance in marking the use and chronology for the area as it appears to have been covered (at least partially at its northern side) around the second century BC by an ashy deposit full of chunks of charcoal, within which three unguentaria forms were found (Sp. Finds 16–17–10). This might be contemporary to one of the burial phases identified within the chamber of the tomb and possibly connected with some type of ritual activity including a cremation burial, although no evidence of cremated bones has been uncovered so far.

The site, despite its small extent and the state of conservation, shows how significantly the landscape transformation can be marked by human interaction with the environment. It displays, as a small sample, the exploitation of the natural bedrock from a productive/industrial use (quarry), the adaptation and carving of the rock to hold the deceased (tomb), with the funerary function persisting over a number of distinct and un-associated phases, and furthermore reusing the same natural source to create an agricultural landscape (trenches).

3 Materials and Methods

The focus of the report is the analytical approach of the archaeological investigation carried out during rescue procedures, which led to the scientific collection of cultural and osteological materials from the site in question.

Cultural materials are discussed following their identification as special finds, considering the funerary context wherein objects have a ritual significance in accordance with the burial practice. Preliminary to the cultural and osteological material analyses, the stratigraphic contexts identified during the investigation were processed in order to establish a relative sequence of events within the site. After the full comprehension of the stratigraphy of the site, an in-depth analysis of the skeletal assemblage and special finds was necessary. The aim of the osteological study for this site was to identify the individual burials within the stratigraphic sequence of the tomb. The study of the skeletal assemblage included the understanding of demographic data and pathological evidence. In addition, the identification of the individuals and their association with the finds were crucial to the interpretation of the sequence of events.

In parallel, the macro-analysis of the finds involved the description and their comparison with known forms from other Maltese contexts, in order to attempt an absolute chronological sequence of the objects found within the site.

The results of the post-excavation analysis led to the identification of three main phases which will be discussed. Despite the fragmentation of the analysed data and the partial loss of the primary deposits, the cultural and osteological materials collected clearly showed a sequence of events reflecting the burial of several individuals and deposition of artefacts within the tomb. Consequently, three main phases were created to highlight the main actions as recorded during the investigation and later included within chronological limits (Figure 3). The dating range, although wide and not definite, is the result of the known chronological sequence wherein the major type and forms are distributed.

Figure 3 
               Plans of phases and overview.
Figure 3

Plans of phases and overview.

4 The Osteological Description

The skeletal remains from this site are not the typical assemblages found from tombs of this period or other periods for that matter. Despite having a small volume and depth of stratigraphic material, the number of individuals identified is rather high. The uniqueness of this assemblage is evident in the fragmentary nature of the skeletons, since no complete skeleton was found in articulation. Instead, most of the burials identified included parts of the skeleton in situ, with the larger parts of the skeleton lost or moved in antiquity. The very nature of this action, points to an often-chaotic use of the tomb and possibly a frequent one.

The following is a description of the methods used to assess the skeletal remains discovered inside the chamber tomb and the results of the study carried out. This description presents the main identifiable burials and gives an overview of the demographic representation.

4.1 Method

Since the skeletal assemblage was highly fragmentary in nature, the study required that the methods used created order in the data collection and interpretation. The parameters available to create this order included (1) the aspect of spatial assessment of the bones, i.e. the connectivity of certain bones which may represent an individual, (2) the taphonomic understanding of the bones as well as the associated grave goods, in order to make sense of the fragmented skeletons and deposits, (3) the demographic results; i.e. the identification of bones pertaining to adult males or females, immature individuals of a given age range, and (4) the pathological or morphological differences within the skeletal assemblage.

The age evaluation for adult individuals consisted mainly of the assessment of cranial suture closure (Buikstra & Ubelaker, 1994). Other bones (such as the pelvis) used for ageing adults were on the whole missing or too fragmented to assess. When the bones were present and in good condition, sexually dimorphic traits at the cranium (Buikstra & Ubelaker, 1994) were used to suggest the biological sex of the individual. A lack of suitable pelvic bones, as noted above, prevented the assessment of sex.

For the age evaluation of immature individuals, the methods used consisted mainly of (1) taking measurements of complete bones, which were then compared with the established results from other studies and (2) assessing the morphological development of the bones and comparing them with the established developmental stages in the maturing skeleton. The various methods mentioned in the subsequent text were taken from Schaefer, Black, and Scheuer (2009). Ageing through dental development in situ of the mandible or maxilla was for the majority of the immature burials not possible due to the high fragmentation of these bones.

The bones were also observed for pathological conditions following in particular the descriptions provided by Ortner (2003) and Roberts and Manchester (2005).

4.2 The Skeletal Data

The burials are listed in chronological order from the earliest burials to the latest burials in the description below and in Table 1. Each burial is identifiable with a skeletal unit or group of bones, which is thought to represent one individual. Due to the intensity of use, and thus disturbance to the burials, these are not represented by intact articulated skeletons. Instead they are represented by closely connected bones. These bones, as will be shown below, may be represented in different contexts and thus one burial may have more than one context (SU) number. As mentioned earlier, the possibility of joining groups of closely connected bones to the common burial units (or individuals) was only possible after intensive analysis of the human remains at the post-excavation stage, together with the assessment of the drawing records and photo images. The criteria assigned to understand the burial units include the assessment of age, especially for the immature individuals, as well as morphology, size and pathology of the individual bones.

Table 1

Burial catalogue

Burial phase Skeletal ref. Context ref. Age Sex
Early phase I Skeleton 2 49, 54, 55 Adult Male?
Skeleton 1 49, 54 Middle adult Male?
Skeleton 5 49 Older adult Female?
Skull A 49 Adult Male?
Early phase I or intermediate phase II Other burial 49 11–15 months Unknown
Intermediate phase II Skeleton 6 30, 37, 47, 48, 49, 51 Middle adult Male?
Skeleton 7 30, 37, 39, 44 Adult Unknown
Infant burial 1 44, 48, 49 2–3 years Unknown
Infant burial 2 30, 34, 36, 39 2–3 years Unknown
Infant burial 3 30, 31, 32, 34, 38 38–40 weeks Foetus/Newborn Unknown
Infant burial 4 37, 39, 44, 45, 46, 47, 50 38–40 weeks Foetus/Newborn Unknown
Other burial 30, 44, 49 3–5 years Unknown
Other burial 29, 44, 47 13–16 years Unknown
Other burial 44 13–16 years Unknown
Late phase III Articulated feet 29, 30, 31, 32, 33 Adult Unknown
Skull 1 29, 30, 33 Young adult Unknown

The skeletal data revealed that the minimum number of individuals for this tomb was calculated at 16, with 8 individuals being immature individuals and another 8 individuals being adults. The summary of the demographic data is also provided in Table 2.

Table 2

Demographic data

Age group Unsexed Female? Male? Total for each age group Total for immature individuals and adult individuals
Foetus < newborn c. 38–40 weeks 2 2 8
Infant c. 11–15 months 1 1
Infant c. 2–3 years 2 2
Young child c. 3–5 years 1 1
Adolescent c. 13–16 years 2 2
Adult – unknown age group 2 2 4 8
Young adult 1 1
Middle adult 2 2
Old adult 1 1
Total for group according to sex identification 11 1 4 16

The earliest evidence of burials (early phase I) found at this tomb consisted of human remains pertaining to four adult individuals: (1) Skeleton 2 (contexts 49, 54 and 55), (2) Skeleton 1 (contexts 49 and 54), (3) Skeleton 5 (context 49) and (4) Skull A (context 49). The human remains are clustered into four groups, and each group containing one cranium.

The human remains in three of these clusters, including the cranium for each (mentioned above) relate to the upper part of the human body, comprising bones pertaining to the thorax, spine and upper limbs, in some instance also the pelvis. During their identification on-site, some of these bones were articulated or in anatomical position, others were found disarticulated.

These three individuals have been interpreted to have been buried in a supine position (on the back) and with an orientation of West to East. The lower part of the bodies would have been situated in the area of the tomb which was heavily disturbed by the recent development works.

The individual skeletal units on plan are found one next to another across the entire stretch of the width of the chamber from South to North (Figure 3). Further evidence, described below, shows that the four individuals were probably buried at different times.

4.2.1 Skeleton 2 (Contexts 49, 54 and 55)

The human remains associated with Skeleton 2 (contexts 49/54/55) furthest North were less articulated, and the bones, including the fragmented cranium, seem to have been shoved or shifted from the central area of the chamber to the North where the archaeologists and osteologist identified them. The lack of articulation is being interpreted as evidence that the human remains had already been skeletonized or had already reached advanced stages of decomposition because probably the bones dislodged from articulation during their movement to the North side of the chamber (Figure 4). This action of disturbance has also been noticed with a trefoil jug (Sp. Find 14) as described below.

Figure 4 
                     Uncovering of early phase I with skeletons 1 and 2.
Figure 4

Uncovering of early phase I with skeletons 1 and 2.

The articulating cervical and upper thoracic vertebrae noted just above the bedrock in this area (context 55) are also being associated with Skeleton 2. Probably this particular articulated spine survived the movements, unlike the rest of the upper skeleton, due to its anatomical location being at the deepest position of a skeleton in supine position.

  • Estimation of age. The cranial and post-cranial bones associated with this skeleton are fully developed and therefore indicate that the individual was an adult. The cranium was too fragmented to assess for the degree of suture fusion and therefore no further ageing could be indicated for this individual.

  • Determination of sex. Despite the cranium being found in a highly fragmented state, it was possible to determine the sex of this individual by assessing the different sexually dimorphic cranial features present, including the supra-orbital margins and supra-orbital ridge following Buikstra and Ubelaker (1994). These indicated the individual is probably male.

  • Identification of pathology. Although not extensively studied, one tooth, the right maxillary central incisor, was observed to include pathological evidence. This consisted of a central groove incised in the superior margins of the tooth. This groove is possibly indicative of the use of the mouth as a tool or aid to an activity being carried out by the individual during life such as processing of fabric, resulting in this incisal groove on the incisor (Schulz, 1977, pp. 87–91; Ubelaker, 1999, p. 100; Roberts & Manchester, 2005, pp. 80–81).

4.2.2 Skeleton 1 (Contexts 49 and 54)

The human remains found to the West of Skeleton 2 (contexts 49/54) have been identified as Skeleton 1 (contexts 49/54). This latter burial is interpreted as a burial occurring sometime after the burial of the individual identified as Skeleton 2. The evidence for this lies in the bone positions belonging to this skeleton which were found above those of Skeleton 2 and Sp. Find 14 as described above (Figure 4).

Skeleton 1 presents some articulation, but like skeleton 2 most of the remains are disarticulated. Despite this, the bones are roughly in the position within the skeletal unit, probably indicating that the individual’s skeletal remains remained in the original place of burial within the tomb. The disarticulation of the bones is on the other hand interpreted as relating to the burials that occur in the second phase of burial above Skeleton 1 (described below).

  • Estimation of age. The cranial and post-cranial bones associated with this skeleton are fully developed and therefore indicate that the individual was an adult. The cranial fragments were also assessed for the degree of suture fusion (Buikstra & Ubelaker, 1994), and these indicated the individual was a middle adult.

  • Determination of sex. Present among the cranial fragments was the right mastoid process, which indicated the individual is probably a male (following Buikstra & Ubelaker, 1994).

  • Identification of pathology. From initial study, the permanent teeth, which have been associated with this individual, were observed to include certain pathological conditions including (1) moderate occlusal wear to the mandibular molars, maxillary incisors and maxillary canines and (2) a carious lesion at a mandibular first molar of unknown side.

4.2.3 Skeleton 5 (Context 49)

The human remains found at the southern extent of the chamber have been identified as Skeleton 5 (context 49). This skeleton displays evidence of having been disturbed by the second phase of burials occurring above it – the disturbance includes bones such as the cranium and mandible having shifted/rolled towards the east. The skeletal unit of this upper body is more contained within its area of burial, and, therefore, it is being interpreted that the burial was carried out in this part of the chamber (Figures 5 and 6).

  • Estimation of age. The cranial and post-cranial bones of Skeleton 5 are fully developed and therefore indicate that the individual was an adult. The cranium is largely intact and complete and was assessed for degree of suture fusion (Buikstra & Ubelaker, 1994). The assessment indicated the individual is a middle-to-older adult. The evidence of this individual being an older adult also stems from the pathological observation at the maxilla and mandible and other evidence in the spine, which are described below.

  • Determination of sex. The cranium of Skeleton 5 strongly suggests this individual is a female. The interpretation resulted from the assessment of the supra-orbital margins, supra-orbital ridge and mastoid process following Buikstra and Ubelaker (1994).

  • Identification of pathology. The vertebrae presented evidence of minor osteophytes at the articular facets, synonymous with spinal osteoarthritis (Ortner, 2003, p. 549). Also, one of the thoracic vertebrae was noticed to have a flattened vertebral body – this seems to be synonymous with the compression of the spine at an extreme level. The maxilla and mandible presented no dentition within the alveoli (sockets). Since most of the alveoli of both maxilla and mandible were observed to be reported, it can be firmly interpreted that most of the dentition were lost during the individual’s life, an indication the individual was probably rather advanced in age (Ortner, 2003, pp. 604–605). The only dentition identified within the context of Skeleton 5 consisted of a single root stump found exfoliated. The root stump seems to indicate that the tooth crown was worn down heavily possibly through use when it was still in occlusion of the mandible or maxilla and may possibly be indicative of another pathological condition such as a carious lesion (Hillson, 2005, p. 293).

Figure 5 
                     Skeleton 5 in early phase I.
Figure 5

Skeleton 5 in early phase I.

Figure 6 
                     Detail of Skeleton 5 in early phase I.
Figure 6

Detail of Skeleton 5 in early phase I.

4.2.4 Skull A (Context 49)

The fourth individual identified within the early phase I only consists of a cranium. Other disarticulated bones close by include a clavicle and a rib, which may be associated with this fourth individual. The cranium’s position between Skeletons 1 and 5 (described above) can only allow speculation that a fourth burial was originally laid in the area between Skeletons 1 and 5. Probably the disturbance to the skeleton associated with Skull A (the fourth individual) is related to the second phase of burial (described below).

  • Estimation of age. The cranium indicates that the individual was fully developed and therefore an adult.

  • Determination of sex. The morphological characteristics, in particular the right mastoid process, of skull A seems to indicate that this individual was probably a male (Buikstra & Ubelaker, 1994).

  • Identification of pathology. An isolated tooth found in close proximity to Skull A was identified as being an upper lateral incisor and observed to have diagonal wear. This type of wear could have been an intentional modification to the tooth, or else unintentionally caused by a repeated action which involved the tooth as a tool in this action, in both circumstances when the tooth was still present in the maxillary socket (Burnett & Irish, 2017).

4.2.5 Other Evidence of Burials

In addition to the evidence of four individuals presented above, four immature bones were identified at context 49. These were found close to “skull A” and “Skeleton 1”. These four bones include a cranial bone, vertebra and two hand bones. There is a possibility that these bones belong to the same individual, but this cannot be ascertained entirely because the bones were found disarticulated. The cranial bone consisted of the pars basilaris (part of the occipital bone) which was measured and associated with an infant probably aged around 11–15 months following the data by Scheuer and MacLaughlin-Black (Schaefer et al., 2009, p. 13). There is uncertainty whether this individual may have been buried in the early phase I or whether they are bones which have infiltrated the context from the intermediate phase II (described below).

The second phase of burial (intermediate phase II) is characterized by further inhumation burials. One of the differences observed with the previous phase is that the orientation of the skeletons is in a North-to-South direction. Another difference with the previous burial phase is the clear presence of immature individuals buried with the adults (Figure 3).

From the evidence observed during the excavation, it seems that the second phase of burial disturbs the skeletons in the earlier phase. The movement of the bones from the first phase indicates the individuals were skeletonized when the next inhumations were laid in the second phase.

There are at least two adult burials identified in this phase: (1) skeleton 6 (contexts 30, 37, 47, 48, 49 and 51), (2) Skeleton 7 (contexts 30, 37, 39 and 44) and a number of infant burials identified: (1) infant burial 1 (contexts 48, 49 and 44), (2) infant burial 2 (contexts 36, 39, 34 and 30), (3) infant burial 3 (contexts 30, 31, 32, 34 and 38) and (4) infant burial 4 (contexts 37, 39, 44, 45, 46, 47 and 50).

4.2.6 Skeleton 6 (Contexts 30, 37, 47, 48, 49 and 51)

The earliest skeleton found in this second phase is Skeleton 6. This skeleton is probably disturbing the skeletons found in the earlier phase; therefore, Skeletons 1 and 2 described above as well as Skeleton 5.

Skeleton 6 is not complete and is only represented by different skeletal portions found across the width of the chamber. At the centre of the chamber, at context 49, a radius and ulna from the right arm were identified parallel to one another and are synonymous with a right arm. Right hand bones have also been identified in contexts 49 and 48 nearby. To the southern extent of the chamber and at the same level, a cranium and mandible, at the beginning of the excavation as context 37, have been identified as possibly pertaining to Skeleton 6. Together with the cranium, some other bones from the upper body were identified, including a humerus, ribs and scapula. Similarly, at the northern extent of the chamber at contexts 51/47, two articulated feet (left and right) were found. The Skull and feet are isolated and do not seem to belong to earlier or later burials and therefore has been interpreted as probably being part of Skeleton 6.

The individual is laid with the cranium positioned at the North and the rest of the skeleton oriented towards the South, thus occupying probably the entire width of the burial chamber. Of particular interest are the stones found adjacent and just under the cranium and mandible in context 37. At the other end of the chamber underneath the articulated feet, other stones were identified (context 52). One can perhaps speculate that these stones were placed in preparation for the burial of the individual.

The limited articulated evidence identified for this individual is probably a result of disturbance from burials occurring after this burial. These later burials probably took place after the body was skeletonized, since the bones probably dislodged easily from their anatomical location.

  • Estimation of age. The post-cranial bones associated with this skeleton are fully developed and therefore indicate that the individual was an adult. The degree of cranial suture fusion indicates the individual was probably a middle adult (Buikstra & Ubelaker, 1994).

  • Determination of sex. The cranium indicates the individual was probably a male – this is indicated through the observation of the supra-orbital margins and the mastoid process (Buikstra & Ubelaker, 1994).

  • Identification of pathology. Both the upper and lower teeth presented medium to high levels of wear. The pedal phalanges (toe bones) found in context 47 (if we are to associate them with SU 51 and also this individual) indicate the individual suffered from severe osteoarthritis, since the phalanges were found to have evidence of eburnation and osteophytes at the articulate surfaces (Ortner, 2003, pp. 547–549).

4.2.7 Skeleton 7 (Contexts 30, 37, 39 and 44)

The second adult individual buried in this second phase is buried in supine position over Skeleton 6 along the width of the chamber. Unlike the latter adult skeleton, Skeleton 7 is found laid in a South-to-North orientation with the feet position at the northern extent of the chamber (Figure 7).

Figure 7 
                     Intermediate phase II: Infant burial 2 on large pot SU36 and partial uncovering of Skeleton 7 with special finds.
Figure 7

Intermediate phase II: Infant burial 2 on large pot SU36 and partial uncovering of Skeleton 7 with special finds.

The bones present include a left and probable right femur parallel to each other and a tibia (context 39), parallel tibia and fibula from the left side (context 44) which are continuous with the previous bones described, and foot bones from the left and right side (contexts 30, 37, 39 and 44). A left pelvis (context 44) which is positioned along the western wall of the chamber may also possibly belong to this individual.

The foot bones (contexts 30, 39 and 44) and distal tibia (context 44) seem particularly smaller than the average size for a “typical” adult, even though these are fully developed bones. Therefore, the association of these bones with the same individual was more confidently ascertained. In addition, contexts 44 and 37 both presented a pathological condition called “symphalangism” (described below) which could be an evidence that these pertained to the same individual. Nevertheless, it cannot be excluded either because two individuals may have had this pathological condition.

A number of hand bones have been identified in contexts 39, 44 and 46 – these seem consistent with the concentration in the left and right hands of two individuals, possibly belonging to the two adults described above in the second burial phase (Skeletons 6 and 7).

  • Estimation of age. The skeletal remains indicate the individual was a fully developed adult. No further age group could be assigned to the individual because the cranium and pelvis were largely missing or fragmented.

  • Determination of sex. The determination of the sex was not possible for this individual because no reliable sexually dimorphic bones were found.

  • Identification of pathology. Two instances of ankylosis of the “interphalangeal joints” (fusion between phalanges) were identified. The pathological bones in both instances included the distal and intermediate foot phalanges also noted in the above description. This pathological evidence is often referred to as “symphalangism,” a condition that can occur in the toes or fingers (Nakashima, Hojo, Suzuki, & Ijishi, 1995, pp. 275–278). A possible diagnosis of the fusion of joint bone surfaces can be osteoarthritis (Roberts & Manchester, 2005, p. 135)

4.2.8 Other Burials (Adults)

To the southern extent of the chamber, a cluster of bones were identified (context 50) within a cut (context 53). This cluster of bones consisted of mainly adult bones including cranial bones, a mandible, upper vertebrae, fragments of a humerus and ulna, hand and foot bones, and a patella. An immature bone was also identified, and this is described below in infant burial 4. The nature of this cluster of bones remains highly unclear. What we do know is that the remains are disarticulated. There is a chance that these adult remains belong to Skeleton 7 (described above), because these bones are not found in the latter’s skeletal assemblage. A few of the bones, in particular the hand bones, may also be related to Skeleton 6 (described above).

Other adult bones included a small number of ribs and sternum within context 36, on the pottery sherds, alongside the immature bones (femurs) identified for infant burial 2 described below. Although not articulating, the ribs and sternum were found in close proximity indicating they may have been from the same adult individual, possibly Skeleton 6 or 7 or another individual within the intermediate phase II or late phase III.

A fragment from the left pelvis of an adult individual and a manual phalanx was also identified close by in context 43. These bones are probably disarticulated remains from Skeleton 7 described above.

Context 46 also included some carpals and a rib belonging to an adult individual. These remains are probably associated with Skeletons 6 and 7 but may have disarticulated from the earlier contexts.

4.2.9 Infant Burial 1 (Contexts 44, 48, 49)

Along the western wall of the chamber in close proximity to the cranium of context 37/Skeleton 6, Sp. Find 12 and Sp. Find 13, a concentration of immature human remains were identified. Although these were not identified as articulating, the close connection of the bones and the presence of small bones consistent with one individual of the same age, strongly indicate that an immature individual was buried in this location. The remains found in context 48 which are associated with this skeleton include cranial bones, vertebrae, a humerus, a radius, hand bones, pelvic bones, a fibula and foot bones.

Some of the bones from this infant burial were noted to be stained green, probably residue from Sp. Find 18 (metal disc) which was found under this skeleton.

A number of disarticulated adult remains were found above this immature burial, namely, a femur and foot bones, and the left pelvis possibly identifiable with Skeleton 7 (mentioned above) from context 44.

  • Estimation of age. Through various measurements carried out for the long bones as well as the developmental stages of the individual bones, this burial has been aged to 2–3 years (Schaefer et al., 2009).

  • Determination of sex. The sex identification could not be identified since this individual is a very young individual (non-adult).

  • Identification of pathology. No pathology was observable on the human remains of this individual.

4.2.10 Infant Burial 2 (Contexts 30, 34, 36, 39)

Laid above context 39 (where part of Skeleton 7 was situated) were the remains of an immature individual referred to here as context 34. The bones identified for this individual included a group of ribs found lying next to one another and a group of vertebrae as well as a fragment of a mandible.

At the same level of context 34, another set of bones identified as context 36 was noted. These other bones, a left and right femur, seem to have been placed parallel to one another and laid on a fragmented pottery vessel. These seem to be of the same age as the immature bones identified in context 34, described above. Evidence of this infant is also seen within context 30 – this consisted of a few bones including some vertebrae and a manual phalanx.

It is interesting that there are no duplicate bones with infant burial 1 described above. This study cannot confirm whether the skeletal remains of infant burial 1 and infant burial 2 may in fact be the same individual.

The general hypothesis is that the remains described below are a secondary treatment to skeletal remains – especially for the laying of the femurs on the “reconstructed” pottery sherds (context 36).

  • Estimation of age. Using the approximate measurements of the femora and comparing them to the data of Maresh found in Schaefer et al. (2009), this burial has been aged to 2–3 years.

  • Determination of sex. The sex identification could not be identified, since this individual is a very young individual (non-adult).

  • Identification of pathology. No pathology was observable on the femora.

4.2.11 Infant Burial 3 (Contexts 30, 31, 32, 34 and 38)

Probably contemporary to infant burial 2 is another infant burial. The remains of individual are mainly found within a coarse ware container (Sp. Finds 4 and 6; Figures 810). The bones within this container were found within a soil deposit with assigned context 31. The bones included cranial bones, a right mandible, vertebrae, ribs, a clavicle, a scapula, left arm bones (humerus, ulna and radius), hand bones, pelvic bones, a femur and foot bones.

Figure 8 
                     Uncovering of skull 1 (far left); the infant burial 3 (West of skull 1).
Figure 8

Uncovering of skull 1 (far left); the infant burial 3 (West of skull 1).

Figure 9 
                     Detail of the infant burial 3 within the cooking pots (Sp. Finds 4, 6 and 7).
Figure 9

Detail of the infant burial 3 within the cooking pots (Sp. Finds 4, 6 and 7).

Figure 10 
                     Partially opened cooking pots (Sp. Finds 4 and 6), uncovering contents of infant burial 3 and Special Find 7.
Figure 10

Partially opened cooking pots (Sp. Finds 4 and 6), uncovering contents of infant burial 3 and Special Find 7.

There are enough bones from the skeleton to suggest that the individual was buried in the container. But there is evidence that secondary treatment was carried out to this burial, and as a consequence, some of the bones were lost or spilled out of the container. Evidence of this was found when a few bones were identified in the fill of a small pot (Sp. Find 7) also buried in the coarse ware container (Sp. Finds 4 and 6). These bones consisted of a cranial bone fragment, a rib, a vertebra and a metatarsal (context 31). Other contexts, outside the container but in close proximity, presented remains that probably belong to this same individual. These include contexts 30, 32, 34 and 38.

  • Estimation of age. Through various measurements carried out for the long bones as well as the developmental stages of the individual bones, the infant in this burial has been identified as a foetus/newborn aged 38–40 weeks (Schaefer et al., 2009).

  • Determination of sex. The sex could not be identified because this individual is very young (non-adult).

  • Identification of pathology. No pathology was observable on the human remains.

4.2.12 Infant Burial 4 (Contexts 37, 39, 44, 45, 46, 47, 50)

Other similar skeletal remains to infant burial 3 have been identified. Although not much can be concluded on this burial, the presence of the bones in the deeper contexts (37, 39, 44, 45, 46, 47 and 50) suggests that these bones are not part of the infant identified above (infant burial 3) but represents another burial probably interred at the start of the intermediate phase II. The clear disarticulated nature of these bones suggests the burial was disturbed during the interments that followed (Skeletons 6 and 7). The bones include cranial bones, vertebrae, ribs, a clavicle, a humerus, a left and right ulna, a phalanx, pelvic bones, a femur and fibula.

  • Estimation of age. Through various measurements carried out for the long bones as well as the developmental stages of the individual bones, the bones have been attributed to a foetus/newborn aged 38–40 weeks (Schaefer et al., 2009).

  • Determination of sex. The sex could not be identified since this individual is very young (non-adult).

  • Identification of pathology. No pathology was observable on the human remains.

4.2.13 Other Burials (Immature)

During the excavation and analysis of the bones, further immature bones were identified, which fall into older age groups from those of the infant burials described above.

  1. The first is the remains of a 3- to 5-year-old child of unknown sex and without any particular pathological evidence. The bones included a left femur and a fibula identified in close proximity to the right ulna of Skeleton 6 (described above) as well as a mandible (context 49). Other bones also aged 3- to 5-year-old included a vertebra and manual phalanx (both in context 30) and a calcaneus (context 44). The ageing was assigned through the measurements of the long bones (Schaefer et al., 2009) and through the dental development observed at the mandible (White, 2000).

  2. Human remains of two older immature individuals, probably aged 13–16 years, were also identified. These included two vertebrae and pelvic bone fragment (context 47), two left scapulae, hand phalanges and pelvic bone fragments (context 44), an ulna and a metacarpal (context 29).

The age of these individuals was estimated by comparison of size and development (Schaefer et al., 2009).

The last phase (Late Phase III) mostly consists of highly disturbed burials, probably because these were exposed to modern disturbance during development. Despite this exposure, there is still evidence that at least two adult individuals were buried in this phase (Figure 3). These individuals are represented by (1) articulated feet (contexts 29, 30, 31, 32 and 33) and (2) Skull 1 (contexts 29, 30 and 33). Both sets of bones are found over one another at the South of the chamber, indicating the probable position of the individuals at burial, as described below.

4.2.14 Articulated Feet (Contexts 29, 30, 31, 32 and 33)

To the South of the chamber, just beside infant burial 3, but at a slightly higher surface, left and right articulating foot bones of an adult were discovered together with distal fragments of the lower limbs (contexts 32 and 33). Although one cannot further identify the bones for age and sex, these articulating bones are the evidence that another individual had been present probably in a North-to-South orientation. Other foot bones were found in contexts 29, 30 and 31 – these are probably the bones missing from the articulate feet identified above.

4.2.15 Skull 1 (Contexts 29, 30 and 33)

Above the articulated feet mentioned earlier, a Skull and mandible (context 30) had been identified. Within context 30, other bones have been identified including ribs, vertebrae, fragments of a scapula and a humerus. These bones although not found articulated are interpreted as the latest individual buried in the tomb. The orientation of this individual would have been a south-to-north orientation. The cranial sutures at the cranium indicate this individual was a young adult (Buikstra & Ubelaker, 1994). Other bones found in contexts 33 and 29 which include cranial fragments and other bones associated with the upper body can probably be associated with this skeleton.

4.2.16 Other Human Remains

Other disarticulated adult human remains were also identified in this phase. These include bones from context 34 and consist of vertebrae, a carpal, a patella and a fragment from a radius. The position of these bones at the centre of the chamber may mean that these bones were part of the individuals described earlier in this phase, especially since these specific bones do not appear with any of the assemblage in this phase.

Another set of bones included a cluster identified as context 35 from the western side of the chamber. In this context, cranial fragments of a young adult were identified. It is interesting to note that these cranial fragments may have formed part of Skull 1 (described above), but physical joining of these remains was not conclusive, leaving interpretation open. Also present in this context were other adult bones including fragments of a mandible and tibia, metacarpals, hand phalanges and a vertebra.

The earlier deposits, contexts 27 and 28, also included some fragmented adult human remains. Not much can be said about these bones other than that they are disarticulated and probably belong to an individual buried in the late phase III. The bones vary from cranial bones to post-cranial bones.

5 Catalogue of Finds

The finds are listed according to the numbering given during the investigation and to the defined cluster of objects by stratigraphic unit. The following information is provided for each object: the unique code (special find number), the relating context (SU number) where relevant, a brief description of its characteristics and the reference to the known forms and types to define chronology. These finds are crucial to the interpretation of the site.

Special Finds 1 and 2 (Figure 11; supposed overall height between 10 and 15 cm; red colour fabric 2.5YR 4/6, very pale brown slip 10YR 8/2) are part of the same small trefoil jug, featuring a flat disc-like base and a rounded section handle. The broken vessel is found in one of the most disturbed deposits (context 29), possibly relating to the burials (or remnants of them) laying at the southern side of the chamber. Similarities are found with form V: 1b, dating to phase V (100 BC/50 AD – Sagona, 2002, figure 139, 53). Within the same deposit, a bronze pin with part of the eyelet (length 5.8 cm, diam. 0.2 cm) was recovered.

Figure 11 
               Selected special finds.
Figure 11

Selected special finds.

Special Find 3 (Figures 11 and 12; diam. at rim 7.2 cm, diam. at base 4.4 cm; reddish yellow fabric 5YR 7/6, darker and crispy core, pinkish white slip 7.5YR 8/2), a small bowl with an almost flat and everted rim, a flat string-cut base and a carinated body. Similarities with form IV–V: 3d, dating from late phase IV to early phase V – 100 BC (Sagona, 2002, figure 48, 4, figure 160, 1, figure 322, 3, figure 347, 13). The bowl was recovered from the first recognizable deposit covering the whole western side of the chamber (SU30, identified as a disturbed brownish layer, preserving a Skull at its South side and other bones at the Central/North side).

Figure 12 
               Selected drawings of finds.
Figure 12

Selected drawings of finds.

Special Finds 4 and 6 (Figures 812) forming a container identified as a secondary deposition and preserving infant bones, Special Find 7 and a few fragments of a rim. Special Find 4 (which fragments have high quantity of chalky efflorescence, reddish-brown colour fabric with a darker core and clear burnt traces due to its use, min. diam. 15 cm) is the partial rounded base of a cooking pot, which might be associated with other fragments, including those of a rim, found within the container and underneath it (context 47). The combined fragments of base and rim appear to have close similarities with form VI: 1, dating to phase VI – 50 century AD, which features an almost flattened and everted rim as the fragments found within the burial container (Sagona, 2002, p. 222, figure 349, 53). Therefore, both the fragments of the base and those of the rim might represent a single vessel, which is known in the archaeological literature as caccabium, which is mainly used for infant burials (past references to the Rabat area are found in: MAR 1950–1951, Necropolis in Triq Ferris, tomb No. 2, p. 378; Temi Zammit’s Notebook III, pp. 99–100, record of a tomb accidentally discovered in Rabat in 1911–1912, wherein a caccabium is reported – the tomb features an identified rectangular space opposite the tomb, containing three cinerary urns. Zammit, 1909–1912; further reference in Sagona, 2002, p. 962). The cooking pot appears to be very common among the local types, with its use ranging from the fifth/fourth century BC to the first century AD, and thus it is not chronologically useful. Further parallels are found at Tas-Silġ contexts (where it is mainly present between the fourth and the second century BC, see Quercia, 2002, p. 407, figure 1, Olla A1 type) and with other Maltese funerary assemblages (see Bulebel, where it is found in a context dating to the first century AD; Anastasi, 2019: Type C1, pp. 77–79, figure 41). The base of the burial container identified as Sp. Find 6 (fragments have high quantity of chalky efflorescence, a light red fabric; diam. rim 32 cm, diam. at base, height 5 cm), is a cooking shallow pan having a wide, flat base with a short everted and slightly curved walls ending in a simple rim. Strong comparisons are found with form III–IV: 1, typical of phase III–IV, 300–100 BC (Sagona, 2002, figure 72, 11, figure 112, 13, figure 255, 1, figure 345, 30). It also appears to be quite common in the assemblage from Tas-Silġ (comparisons are found with African plates dating from third century BC and first century AD; Quercia, 2002, p. 415, Figure 4) and from Bulebel (dating between the end of the first and the early mid-third century AD; Anastasi, 2019, C17. 1–3, pp. 85–86, figure 45). Inside this container was also found Sp. Find 7, consisting of a wide bowl with everted rim, thin carinated walls on the lower part of the body and a flattened base (fabric not examined, height 7.4 cm, diam. 7.5 cm, width 9.8 cm). There are similarities with Special Find 3, as both are compared with form IV–V 3d, dating from the late phase IV to early phase V – 100 BC (Sagona, 2002, figure 48, 4, figure 113, 12, figure 160, 1, figure 253, 6).

Figure 13 
               General view of the truncated chamber with the unexcavated deposits.
Figure 13

General view of the truncated chamber with the unexcavated deposits.

Figure 14 
               General view of the central part of the chamber, the shaft and the third-party wall partially blocking the shaft.
Figure 14

General view of the central part of the chamber, the shaft and the third-party wall partially blocking the shaft.

In Special Find 8 (very dark colour and probably burnt, originally might have been of a blue colour; diam. 1.1 cm, height 0.9 cm), a bead found within context 34.

Pot in SU35 (inadequately fired, crisp and gritty ware, fabric is pale brown 2.5Y 8/2), a juglet or small flask (no special number assigned) having a disc-flat base, a cup-like mouth neck, simple rim and ovoid section handle starting from the neck. It appears to be a part of those wares produced from phase III onwards; the flask may have strong similarities with form IV: 1a–e, dated to phase IV – 300/100 BC; Sagona, 2002, pp. 143–145, p. 666, figure 346, 10–13).

Potsherds SU36 (Figure 7) used as base for laying two bones of an infant. Such fragments are consistent with sherds recovered from one of the first very disturbed layers (context 27) filling the truncated tomb (the potsherds feature a wide curvature which might be consistent with a large vessel, possibly of an amphora type, in association with child’s burials).

Special Find 10 is the fragment of an unguentarium (brown fabric 7.5YR 4/4, dark brown slip 7.5YR 3/2 and darker black slip traces; base is preserved, residual height 5.2 cm, diam. base 1.6 cm; very fine ware, possibly imported object). Similarities are found with form IV: 1a, suggesting a date to the second-century BC – late phase IV to early phase V; Sagona, 2002, figure 200, 6, figure 346, 17. Within the same context is also found Special Find 11, consisting of two fragmented bronze pins or brooches (1: with a hole, length 8.3 cm, diam. 0.2 cm; 2: length 7.3 cm, diam. 0.2 cm).

Special Find 12 (Figures 11 and 15; red fabric 10 R 5/8, light red slip 2.5YR 7/6; diam. 9.3 cm, height 10.5) is a spouted flask or guttus having a flaring neck, narrow at the juncture with the shoulder, two vivid red bands around the body under the handle and the spout, and another one in the lower body. It appears to be a variant of form V: 1a–b, dating to phase V – 100 BC/50 AD (Sagona, 2002, figure 274, 7, figure 141, 7, figure 139, 51).

Figure 15 
               Details of Special Finds 12, 13 and 15 in intermediate phase II, and partial uncovering of Special Find 14 in early phase I.
Figure 15

Details of Special Finds 12, 13 and 15 in intermediate phase II, and partial uncovering of Special Find 14 in early phase I.

Special Find 13 (Figures 11 and 15; height 10 cm, diam. body 5.7; pale yellow fabric 5Y 8/2) is a trefoil-mouthed juglet with swelling neck and round-sectioned handle with dimple at the bottom; although these types of juglets feature slight changes in the long-term, similarities are found between forms IV and VI (Sagona, 2002, pp. 140–142. It may have stronger comparisons with form VI: 1b; figure 305, 5–7–8). It is difficult to date with certainty, as it might range between phases IV and V (300/100 BC – 100 BC/50 AD). Its position being close to the spouted flask may suggest Phase V.

Special Find 14 (Figures 4, 11, 12, 15, 16; found in context 46, considered the same as context 49. Reconstructed, residual height 25/26 cm, fabric yellowish red 5YR 5/6, and a slight darker slip, although not sufficiently preserved; reddish-brown and white medium size inclusions) is a trefoil jug with lifted up and sloping down rims, which appears to be pinched to form the spout, elongated and narrow body, thin and equidistant painted bands on the neck and shoulder. Strong similarities are found with form III–IV: 1a–b, which can be dated to late phase III to early phase IV, 410/300 BC (Sagona, 2002, figure 302, 1, figure 255, 3, figure 239, 9, figure 127, 6, figure 130, 8, figure 45, 1). It is important to note that the trefoil jug had been disturbed and partially destroyed to make space for later burials. Other fragments consistent with the trefoil jug and including part of the handle and the base were found to the North side of the chamber, within the deposit context 54 (given to a second phase of articulated skeletons within context 49 and similar to it) and in the very last level of context 49, underneath the articulated spine (articulating cervical and upper thoracic vertebrae being associated with skeleton 2, see osteological description for reference).

Figure 16 
               General view of early phase I (Sp. Finds 14 and 22) and intermediate phase II (Sp. Finds 18, 19 and 21).
Figure 16

General view of early phase I (Sp. Finds 14 and 22) and intermediate phase II (Sp. Finds 18, 19 and 21).

Special Find 15 (Figures 11 and 15; diam. body 8 cm, height 7.8 cm; red fabric 2.5YR 5/6) is a simple open bowl with a curved profile and a small flat, disc base; it is defined as a Romano-Punic cup made of well-baked fabric with next to no inclusions, similar to form VI: 1a, dating to late phase V or VI – first-century BC/first-century AD onwards (Sagona, 2002, figure 221, 9).

Special Finds 16, 17 and 20 are three unguentaria recovered from a layer of ashy dark brown soil (context 25) identified within an area of quarrying activities (Feature 14). The most preserved appears to have a dark reddish-brown fabric; height 12.5 cm, diam. rim 2.2 cm. The unguentaria appear to have strong similarities with form IV: 1-b, which are mostly dated to the second-century BC (Sagona, 2002, figure 301, 5–6–11–12–13–16, figure 206, 2–3–7). Even though there is limited evidence in the quarry area (feature 14) which was heavily disturbed by the foundations of the adjacent property, one can suggest a later use of this area in conjunction with one of the most intense phases (intermediate phase II) recorded within the tomb. The quarry area might have served as a ritual space.

Special Find 18 (Figures 4, 11 and 16; possibly a small mirror, diam. 5.4 cm, thickness 0.3 cm), a metal disc placed on top of Special Find 21, a small plate characterized by a thick and flat rim, rounded off at its extremity, with the rim having almost a horizontal stance or slightly tilted (diam. 9 cm, red fabric 10 R 5/8, very pale brown slip 10YR 8/2). It appears to be very similar to plates in form IV–V: 1b, dated mostly in late phase III to phase IV (410/300 – 300/100 BC), although similarities are shared with even later forms (Sagona, 2002, p. 213. It is important in its association with pottery forms of phase IV and very likely as part of an internment happening after the late phase III).

Special Find 19 (Figures 11, 15 and 16; width 9.8 cm, height 2.2 cm; red fabric 2.5YR 5/6) is an oil lamp with broad arched back, which spouts occupying only a small part of the rim and squashed sharply into the bowl. Strong similarities with form IV: 1b, within the phase IV – 300/100 BC (Sagona, 2002, figure 297, 6–8; but also, with form IV: 1a, wherein the use of soft orange ware has been recorded). It was found turned upside down and very likely fell from its original position on top of the body.

Special Find 22 (Figures 4, 11 and 16; diam. 16 cm, pale brown slip 2.5Y 8/2, fabric yellowish brown, with large red and brown inclusions, medium to high frequency) is a plate having wide rim, broad and flat floor, well-finished and wheel made with thick walls. Strong similarities with form III–IV: 1, which is found in association with pottery from late phase III to early phase IV (410/300 BC; Sagona, 2002, figure 148, 5, figure 85, 2, figure 83, 17, figure 47, 14; clays are heavy, crispy and brown to light red with fine or medium gritty inclusions, the rims range from 8 to 19 cm), but also in Tas-Silġ contexts where it appears already in the fifth century (Quercia, 2011, p. 434 and p. 436, figure 1:1; the type appears to be more common in Tas-Silġ contexts dating to the fourth and third century BC).

Special Find 23 (Figures 4 and 11; diam. max 7.8 cm, residual height 7.6 cm, red fabric 10R 5/8, very pale brown slip 10YR 8/2) is a flask, similar to form IV: 1c; which shows a development of the rim into a wide and shallow cupped mouth (usually with curved profile and up to 13 cm high) and dating to phase IV – 300/100 BC (Sagona, 2002: usually made of soft refined reddish-yellow clays, soft orange, with pale-slipped surface).

Special Find 24 (context 50) is a bronze pin, with part of the eyelet still present (length 8.8 cm, width 0.2 cm).

Beads (context 58): two dark burnt beads (diam. 1 cm, height 0.8 cm); two bone beads (diam. 0.5 cm, height 0.6 cm); two glass azure and blue colours (diam. 0.8 cm, height 0.7 cm).

Amphora (Figure 11; No SU context) having an egg-shaped body with one missing handle and rim, while the lower part of the body is painted with four thin lines as decoration (pale reddish-yellow fabric with darker grey core, thin pale slip; diam. body 35 cm, residual height 52 cm). Close similarities are found with form III: 1 and form III–IV, the latter is still common during the early phase IV (300 BC; Sagona, 2002, figure 28, 2). No chronological definition between two phases, as they tend to have egg-shaped to ovoid bodies, often wider below the middle, the overall shape starts to be more elongated and cylindrical with bases usually round. Pertaining to a more defined phase IV is the fragment of rim very likely referring to the amphora found in context 28. The rim is square cut, having strong similarities with form III–IV: 1 (Sagona, 2002, figure 95, 2–5).

Potsherds SU7 (Figure 12) are two fragments of rim of a cooking pot having a square-shaped rim everted and probably a roundish/ovoid body of casserole forms. It can be associated with one fragment of lid, having a simple rim. The rim of the cooking pot has a biscuit fabric, very dark greyish, coarse ware, external surface red, possibly with light cream slip. The lid appears very similar, although the fabric was not examined. The lid is closely comparable with North African types (see Hayes Form 196), having a thickened rim edge, which appears to be a common North Tunisian series dating from the late second to mid-third century AD (Examples of hooked lids found in major Maltese context and comparable to Hayes types are fully described in Anastasi, 2019, pp. 96–99, figures 51 and 52. See also reference to Bonifay, 2004, p. 221) while the rim might be part of a casserole or bowl form, locally produced, having close similarities with Hayes 184 (probably second to third century AD; Hayes, 1972, pp. 203–209, figures 35 and 36).

6 Discussion: The Archaeological Context

The archaeological investigation, although being carried out when the chamber tomb was almost fully exposed, resulted in the recovery of a remarkable quantity of material culture, including skeletal remains, pottery and metal artefacts, all within shallow layers of deposits, which were heavily disturbed throughout the centuries.

The stratigraphic excavation identified an extensive deposit (defined by three layers), which was probably filling the entire truncated chamber and the shaft. The fill consisted of heavily fragmented cultural material mixed with stones and plastic, defining several events of disturbance, the most recent probably being concurrent to the construction of the garage. A clear action of cutting into the original layers of the chamber was identified later during the investigation and referred to the actual disturbance of the tomb and possibly to its destruction (Figure 2). Beside the original deposits recorded within the chamber, the rifling activity had also left two rounded ashy traces within the shaft (Figure 17; referring to perishable objects or cinerary urns having a 40-cm diameter and comparable to a discovery recorded by Zammit in his Notebook III, wherein a tomb was accidentally discovered in 1911–1912 in Rabat, consisting of a rectangular space opposite the chamber containing three cinerary urns). Notwithstanding the chaotic activity of rifling within the chamber, less than 0.50 m of the original deposits preserving human remains were identified at the western side of the chamber. The preliminary analysis of the finds revealed evidence of continuous interment inside the tomb from the fifth-/fourth-century BC until the first-century AD. The remnants of human remains and objects, being spared by the destructing/looting activity, give enough data to attempt a relative sequence of interments during three main phases. Preliminary to any description of the context and the material culture, it is crucial to define the limits of the analysis, which are given by the high and consistent disturbance recorded in the deposits, which resulted in shallow and restricted archaeological layers.

Early phase I (410/300 BC – 300/100 BC) (Figures 36 and 16). This first phase is identified by the few articulated skeletal remains found in the very last deposits within the chamber, specifically three crania and associated upper skeletal remains as well as remnants of a fourth burial. In association with these bones, three groups of finds (Sp. Finds 14 and 22, beads in context 58) are considered significant in providing the earliest date for the interments. Crucial for the interpretation of these first burials is the stratigraphic action being recorded within the chamber. This consists of a deliberate disruption/breaking of the trefoil jug (Sp. Find 14) centrally placed, with joining fragments found scattered across the northern side of the chamber mostly next to the cranium associated with Skeleton 2; and in the lowest level of the deposit underneath the articulated spine associated with Skeleton 2 described below. The disruption action is also reflected in the skeletal remains, particularly with Skeleton 2. It is noticed in the archaeological record that parts of the skeleton were moved (possibly even intentionally) to the northern side of the chamber, where the archaeologists found them. An action of disturbance is also clearly noted in the plate next to the trefoil jug (Sp. Find 22). This plate is broken in situ and partially scattered towards the southern side of the chamber. The two vessels have a strong relationship also in terms of chronology, as both appear very similar to forms dating to the fifth-/fourth-century BC. These first burials are disturbed again at a later stage by a new interment being placed between the broken trefoil jug (Sp. Find 14) and the Skull to the North associated with Skeleton 2. This new individual, Skeleton 1, was accompanied by a small flask (Sp. Find 23) placed on the right side of the chest and dating to the fourth-/second-century BC. As suggested by the position and the chronology of the flask, this burial might be marking a new action of interring inside the chamber, still following a West-to-East orientation of the individual and possibly belonging to the same cultural group. However, such action indicates the first disruption of the original interments, revealing the subsequent chaotic reuse of the tomb itself as shown in the later phases described below.

Intermediate Phase II (300/100 BC – 100 BC/50 AD) (Figures 3, 7, 9, 10 and 15). As evidenced by the last interment recorded in the early phase, the chamber is now constantly marked by chaotic actions of burying new individuals. The distinction with the previous phase is supported by a change in the orientation of the laying out of the individuals. These are now following a North-to-South or South-to-North orientation. The sequence of layers as identified during the archaeological investigation became very fragmented, resulting in full skeletons hardly being preserved, and instead resulting in clusters of articulated bones being found and, on a few occasions, clearly in association with grave goods. The first cluster is characterized by a few fragmented bones identified as Skeleton 6, consisting mainly of a cranium, a right arm and feet, associated with a small plate and a metal disc (Sp. Finds 18 and 21), possibly a mirror, which might be an evocative reference to personal beauty care. The small plate is widely used in the Maltese funerary contexts and its chronology does not give a reliable reference point for this phase as it was in use from the fifth- to the second-century BC. However, it clearly marks a transitional moment within the tomb, where the practise of disrupting the previous burials is replaced, and instead the individuals are inserted on older burials or in spaces available in the tomb. This is immediately followed by a second individual, skeleton 7 characterized by the lower limbs, which is stratigraphically and spatially very close to another cluster of finds (Sp. Finds 12, 13 and 15), mostly dateable within the second-century BC and first-centuryAD.

Considering the earliest date for the first individual and the cluster of unguentaria recovered from the quarry area (Sp. Finds 16, 17 and 20. See previous paragraph), the second-century BC might be marking a phase of intense burial activity within the site. This suggestion is further emphasized by the numerous clusters of articulated skeletal remains successively interred within very shallow deposits. Among all the recorded clusters, two are particularly significant as a marker of disrupting actions. These consist of an infant burial placed within a coarse ware container (Sp. Finds 4, 6 and 7) identified as infant burial 3, and the bones with fragments of a large pot (SU36) identified as infant burial 2. They both indicate secondary burials happening after the initial interment was disrupted and moved from its original position. It is even more significant since they relate to infants/children who were previously buried in large vessels such as a cooking pot or an amphora-type vessel, as the cluster of joining potshards (Ref. to catalogue as large pot) seems to suggest. The first cooking pot is evidently showing an action of recovering a previous broken vessel and pairing it with another cooking shallow pan to be assembled as a funerary container consisting of a base and lid, which includes the human remains of a foetus or newborn (infant burial 3) together with a small bowl (Sp. Find 7) (a significant and recent comparison is found with a cremation burial of an infant within a cooking pot, uncovered during the archaeological investigation at St Paul’s Catacombs; see Cardona, 2019, Figure 17, SP285, p. 25). The creation of this burial container shows an evident conscious action of respecting a former burial and the small bowl which was introduced at this stage, might be dating evidence for this action which can be considered as a secondary deposition/repositioning of the former burial (in the cooking pot; Sp. Find 4) after the second-century BC and before the first-century AD. The second group of joining fragments, although being less diagnostic, provides interesting evidence on the existence inside the tomb of another large vessel, possibly relating to an amphora-type commonly used for burials (Sagona, 2002, pp. 87–93). The evidence suggests the presence of a long phase mainly characterized by disruptive actions, repositioning of articulated bones and secondary interment of infants, happening most probably around the second-century BC and being concluded before the first-century AD (the fragmentation of the skeletal remains, the pottery and the state of conservation of the deposits allow only a macro interpretation of a long intermediate phase characterized by an intense use of the tomb).

Figure 17 
               Location of the ashy traces within the shaft.
Figure 17

Location of the ashy traces within the shaft.

Late phase III (100 BC/50 AD) (Figures 3 and 8). After having identified an extensive intermediate phase of continuous chaotic burial actions, it is highly probable that Skull 1 found within the first identifiable burial layer is to be considered as the last interment inside the chamber. Although no evidence survives, it cannot be excluded that further burials were made here – the episodes of disturbance during agricultural use and development may have disrupted this later evidence. Its funerary goods consist of a broken open bowl (Sp. Find 3), which is typologically close to the bowl inside the cooking pot container (Sp. Find 7), and a small fragmented jug (Sp. Finds 1 and 2). The skeletal remains were clearly covering the cooking pot container, sealing the previous phases and the secondary infants’ burials. Although the stratigraphic connections were highly disrupted by the later destruction/looting activity, it is clear that this last individual was buried before or not later than the first-century AD, and possibly closing the long-term use of the tomb. Unfortunately, this last phase is partially recorded and heavily disturbed, de facto limiting our full comprehension and interpretation of the actions happening at the very end of the tomb’s life and the events leading to its sealing, abandonment and later destruction.

7 Conclusion

The primary aim of this report is to provide a comprehensive presentation of the archaeological data recovered together with a preliminary analysis which can contribute to the knowledge of the Rabat area. The discovery draws attention to the complexity of the continuous reuse within a funerary context and draws out two important considerations.

First, the sequence of events clearly shows two points of demarcation within the tomb; a previous burial context (1) in the early phase I, which is suddenly disrupted by a change, possibly cultural, that introduces a series of events happening immediately before and after the second-century BC; followed by a phase marked by infant burials (2). This might be evidence of a broader cultural transformation within the funerary area of Rabat, the understanding of which may require further and deeper analysis, particularly, when comparing the early burial context with the intermediate and later ones. The first suggests a grave layout commonly and widely encountered in the Maltese Punic funerary context, featuring two or more individual with jewellery (beads), a trefoil jug for pouring liquids and a plate for ritual offerings; while the contexts that follow mark a shift in the approach to the tomb, which is again occupied and reused by individuals buried almost exclusively with small objects, and together with children.

The second consideration this site brings up is the relationship between the participants in the setting of funerary practices, thus the relationship between the corpse (and any associated grave goods) with the living in the form of rituals carried out. The evidence identified through this tomb points to a relationship that does not end with the primary burial of an individual who has recently died but depicts a setting of interaction between the living and the dead. This is seen in the movement of bones and objects, possibly even reburial of an infant in a pottery vessel. Even though the reopening of the tomb and re-organizing of the remains of deceased individuals cause breakages to the skeletal units, the individual bones and the grave goods, there is still an act of care and remembrance towards the dead individuals, in that they are recognized and reintegrated among more recently deceased individuals of the society.

The collected data from the tomb at Triq Fejġel should be of primary importance for future scientific analysis, such as radiocarbon dating (C14), aDNA, lipid analysis, isotope analysis, among others that will add crucial information to the understanding of the wider context of the area during the Roman period.

The stratigraphic excavation of this chamber has shown not only the wealth of data that can be extracted from limited stratigraphic contexts but also the need to have accurate, regulated and comprehensive monitoring procedures to ensure the timely identification of archaeological remains and their subsequent investigation, recording and publication. The importance of this discovery goes beyond academic considerations; it is equally important as a case study for proper heritage management. Established monitoring procedures have come increasingly under attack on the grounds of being excessive and irrelevant. It is often argued that in already developed sites, no archaeology survives and thus such procedures are not required. This site shows the fallacy of this argument and validates the existing monitoring procedures. The site is a testament to the perseverance of the archaeological resource and the need to ensure that the planning process fully adopts appropriate mitigation measures. Without constant surveillance, the limited volume of the stratigraphic context would have been bulldozed away in a matter of minutes, leading to the loss of important archaeological data. This report is the fruition of a well-established procedural process across the different units within SCH. The identification of the site potential and application of appropriate monitoring conditions at planning consultation stage, the issuing of monitoring terms of reference, which reflect the particular site specifications and the investigation, recording, post-excavation and treatment of finds. The publication of data is the fulfilment of the mission statement of the SCH – “The SCH’s mission” is to fulfil the duties of the state in ensuring the protection and accessibility of Malta’s cultural heritage.

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.


The Superintendence of Cultural Heritage acknowledges the contribution by the archaeology monitors approved by the same Superintendence and present on-site, QPML, particularly Ms Tamsin Marie Kingswell and Ms Emily Ogden, for the identification of the site which has permitted this investigation. The Superintendence would also like to thank the undergraduate students from the Department of Classics and Archaeology at the University of Malta who have assisted in the excavation and post-excavation during their summer placement – Ms Renita Raina El Masri, Mr Sandro Grech, Mr Daniel Polidano, Ms Emaily Slater, Ms Rochelle Lorraine Xerri, and Ms Julia Zerafa.

  1. Funding information: The archaeological excavation and post-excavation, together with the writing of the paper was not funded by any outside sources, and was resourced through internal funds of the Superintendence of Cultural Heritage.

  2. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

  3. Conflict of interest: The authors state no conflicts of interest.

  4. Data availability statement: All data generated or analysed during this study are included in this published article.


Anastasi, M. (2019). Pottery from Roman Malta, Malta archaeological review supplement 1. Oxford: Archaeopress.10.2307/j.ctv131bsz6Search in Google Scholar

Bonifay, M. (2004). Etudes sur la céramique romaine tardive d’Afrique. Oxford: Archaeopress.10.30861/9781841716510Search in Google Scholar

Buikstra, J. E., & Ubelaker, D. H. (1994). Standards for data collection from human skeletal remains. Fayetteville: Arkansas Archaeological Survey.Search in Google Scholar

Burnett, S. E., & Irish, J. D. (2017). Introduction to a world view of bioculturally modified teeth. In S. E. Burnett & J. D. Irish (Eds.), A world view of bioculturally modified teeth (pp. 1–16). Gainesville: University Press of Florida.10.5744/florida/9780813054834.001.0001Search in Google Scholar

Cardona, D. (2019). Excavations at St. Paul’s Catacombs: New evidence at an old site. In Mitteilungen zur Christlichen Aarchäologie, Band 25/2019. Wien: Österreichische Akademie der Wissenschaften.10.1553/micha25s9Search in Google Scholar

Hayes J. W. (1972). Late Roman pottery. London: The British School at Rome.Search in Google Scholar

Hillson, S. (2005). Teeth (2nd edn.). Cambridge: Cambridge University Press.10.1017/CBO9780511614477Search in Google Scholar

Munsell washable soil colour charts 2009 edition.Search in Google Scholar

Nakashima, T., Hojo, T., Suzuki, K., Ijishi, M. (1995). Symphalangism (two phalanges) in the digits of the Japanese foot, Annals of Anatomy – Anatomischer Anzeiger, 177(3), 275–278.10.1016/S0940-9602(11)80199-8Search in Google Scholar

Ortner, D. J. (2003). Identification of pathological conditions in human skeletal remains (2nd edn.). London: Elsevier.Search in Google Scholar

Quercia, A. (2002). La ceramica da fuoco del santuario di Tas Silg (Malta): tipi attestati e proposte interpretative. In M. G. Amadasi, M. Liverani, & P. Matthiae (Eds.), Da Pyrgi a Mozia. Studi sull’archeologia del Mediterraneo in memoria di Antonia Ciasca, 403–424. Roma: Consiglio Nazionale delle Ricerche.Search in Google Scholar

Quercia, A. (2011). Typological and morphological remarks upon some vessels in the repertoire of pottery in Punic Malta. In C. Sagona (Ed.), Ceramic of the phoenician-punic world; collected essays (pp. 433–450). Bari: Edipuglia.Search in Google Scholar

Roberts, C. & Manchester, K. (2005). The archaeology of disease (3rd edn.). United Kingdom: Sutton Publishing.Search in Google Scholar

Sagona, C. (2002). The archaeology of Punic Malta, ancient near eastern studies, supplement 9. Leuven: Peeters.Search in Google Scholar

Schaefer, M., Black S., & Scheuer, L. (2009). Juvenile osteology: A laboratory and field manual. Amsterdam: Academic Press.Search in Google Scholar

Schulz, P. D. (1977). Task activity and anterior tooth grooving in prehistoric California Indians. American Journal of Physical Anthropology, 46(1), 87–91.10.1002/ajpa.1330460112Search in Google Scholar

Ubelaker, D. H. (1999). Human skeletal remains: Excavation, analysis and interpretation (3rd edn.). Washington: Taraxacum.Search in Google Scholar

Van Beek, G. C. (1983). Dental morphology: An illustrated guide (2nd edn.). Oxford: Butterworth-Heinemann.Search in Google Scholar

White, T. D. (2000). Human osteology (2nd edn.). California: Academic Press.Search in Google Scholar

White, T. D., & Folkens, P. A. (2005). The human bone manual. Amsterdam: Elsevier Academic Press.Search in Google Scholar

Zammit, T. (1909–1912). Archaeological field notes, notebook III. National Museum of Archaeology, Valletta.Search in Google Scholar

Received: 2020-07-30
Revised: 2021-01-08
Accepted: 2021-01-25
Published Online: 2021-04-15

© 2021 Kevin Borda et al., published by De Gruyter

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

Downloaded on 27.9.2023 from
Scroll to top button