Recent Researches in Mechanics

The Birth of a Late Iron Age Fortification. A Preliminary Archaeological and Osteological Approach VLAD VINTILĂ ZIRRA1 & VALENTIN DUMITRAŞCU2 Institute of Archaeology 11 Henri Coandă Str., , [email protected]; [email protected]

Abstract: The contribution proposed here presents a large context (or unit) connected to the erection of the burnt brick wall which fortified the western part of the proto-historical site at Bâzdâna “La Cetate”, Dolj County, Ro- mania. The context in question, which dates to the approximate 300 BC period, allows us to interpret it in all probability as the expression of a very elaborate ceremonial for the founding and consecration of the fortification structure that the local community had undertaken to build. The clear stratigraphy, the archaeological and zooar- chaeological material and the remains of a human infant skeleton, each and every one, contribute to configuring the respective ritual in most probable outlines. Key-Words: fortification with burnt adobe bricks, propitiatory founding/consecrating ritual, zooarchaeological and anthropological study

1 Importance of the Problem western terrace of the river; it stretched over an Founding and consecration rituals are part of the area of circa 1.5 ha, was triangular in shape and its general human behaviour to such an extent that they northern and southern sides were delimited by very become implicit for the majority of individuals and steep adjoining ridges. The settlement was fortified their significance may, therefore, even go almost un- on its weastern side by considerable joint efforts at noticed, in spite of people’s direct participation in a certain moment in time (very probably towards the them. But they remain topical though their sacred end of the 4th c. BC, according to the testimony of the dimension is most noticeably overridden by the lay archaeological materials (Fig. 1). sense in societies of the Western type. Without aiming to discuss in depth about the sacred- 3 The Brick Fortification and its profane binome [1], we shall present, in the follow- Structure ing paragraphs, a series of factual observations made The structure in question was erected at the level of a as a result of the archaeological research done on a slightly earlier habitation area, of which there are no proto-historical site of south-western Romania; these extant traces of proper habitation features apart from will be accompanied by an attempt to characterize some pretty substantial remains of mainly hand-made the main directions followed in performing the cer- pottery. The fortification was shaped as an oblong emony in question. crescent (of circa 125 m) and, to the north and south, it was delimited by the already mentioned steep ridges. It consisted of two paraments of burnt adobe bricks. 2 Short Description of the Site The bricks, which were treated with vegetal remains Systematic archaeological research began at Bâzdâ- to remove the clay grease, were placed transversely na, com. Calopăr, Co. Dolj, in the location known as to the fortification axis, on either side of the two para- “La Cetate” in 1981 and continued, with insignificant ments and they had similar size modules: L. 35-40 breaks until 2005. In spite of its obviously great sig- cm, l. 20 cm, and 10-12 cm in depth, weighing an av- nificance, the publications of this archaeological site erage of 6-8 kg. In their overwhelming majority, they have been sparse and partial so far [2]. were burnt on all sides as well as inside their body; We have to do with a Getian fortified settlement with they were deposited in the structure of the fortifica- two obvious main habitation phases (in addition to the tion after having been burnt in another location [3]. sparse pre-historical occupation-phases of the Early The bricks were laid in place as piles, on top of each Bronze Age, the Coţofeni culture): the first one spans other, without using any kind of binding and without from the latter half of the 4th century to the first half splicing them. A great amount of thoroughly burnt of the 3rd c. BC; the second one covers the first c. BC clayey earth (emplecton) was deposited between the until the mid first c. AD; there is a hiatus between them paraments, which made the resulting specific mass of when the site was uninhabited. The initial, unfortified the structure unusually light. settlement was founded on an erosion plateau on the The fortification was circa 2,8 – 3 m high and had a

ISBN: 978-1-61804-020-6 359 Recent Researches in Mechanics

Figure 1: Bâzdâna “La Cetate”, com. Calopăr, Co. Dolj, 1993-2004 excavations defence trench circa 3 m deep beyond the exterior there were sections where the maximum number of parament and a berm of approximately 0,5 m. It is overlapping still standing bricks included as many as hard to estimate the height of the burnt brick wall, 11 or 12 (Fig. 2). One obvious question to raise is but the observations made during the on site research how stable such a large scale edifice might be, in the indicated that, especially at the northern extremity, absence of mortar, of any signs of a systematic foun-

ISBN: 978-1-61804-020-6 360 Recent Researches in Mechanics dation layout or brick splicing [4]. The answer is to be found most prob- ably in the design and construction of a wooden scaffolding to support the struc- ture both horizontally under the para- ments and raw stone fillings of the walls and vertically, between the paraments and at the back of the interior parament [5].

4 The Unit at the Foot of the Brick Wall The general layout drawing of the exca- vations (Fig. 1) shows that we only had access to the northern and southern ends Figure 2: View of the interior bunt brick parament, (trench X/2002) of the fortification (owing to the present- day situation of the land ownership); here were excavated several trenches. The rest of the central body of the for- tification was only investigated by non- destructive methods (magnetometry and GPR), which is why we have no certain- ties about the existence of the context to be discussed in what follows. Neverthe- less, the direct observations obtained by on-site research are valid for an area of over 100 square meters. Under the en- tire fortification (for about 3-4 metres in length) and at the back of the interior parament could be noticed a continuous stretch of ash and charcoal, the result of strong combustion, which formed a con- sistent layer, with an average thickness of circa 5-8 cm.

This is the place to recall that the inner Figure 3: Intentionally abandoned bowls and spindle whorl, (trench parament which faced the settlement X/2002) was supported by a kind of stabilization earthwork or backfill structure, which also permitted access to the top of the wall. In the layer in question excavations brought to light significant amounts of ceramic material. This included both vessels intentionally broken on site and several items of pottery that the on site inspection revealed to have been deliber- ately deposited at the foot of the defence structure and became slumped under the pressure of the construction erected on top of them. Open-shaped ceramic vessels (such as the bowl or kantharos type, Fig. 3) were seen to predominate at the northern end of the unit, especially, whereas the more narrow-shaped ves- sels (of the jar or cup type) were seen to Figure 4: Intentionally abandoned bag-shape pot on fireplace, (trench be more numerous towards the southern IX/2002)

ISBN: 978-1-61804-020-6 361 Recent Researches in Mechanics end of the unit (Fig. 4). In addition to the substantial left one lacks its extremities and the right one has amount of pottery, there were tens of vessels of var- only the acromial extremity. Parts of both scapulae rying dimensions, some of them being restorable in were identified, containing the glenoid cavity. Only their entirety, others only in part; they were found to- the left proximal humerus was found. gether with some spindle whorls and three hearths in From the pelvis we have recovered the left ischium the same layer [6] of ash and charcoal. In addition to and the femur is represented by a proximal epiphysis the goods and installations mentioned, this layer also that could not be lateralised. allowed us to document numerous animal remains, some of them with retraceable anatomical connec- The age at death was estimated based on the eruption tions, but never whole skeletons [7]. In addition, of the first permanent molar. According to Ubelaker slightly higher than the layer in question, in the mass [9], the degree of teeth eruption and development in- of the embankment at the back of the inner parament dicates an age at death of 6 ± 2 years. According to was found part of a child’s skeleton, which, as indi- Sjøvold [10], the individual is placed in the boundary cated by the on-site observations, was not given the of infans I / infans II age classes. regular funerary treatment. It is likely to have been Due to the young age, we could not estimate the sex deposited there at the same time as the construction of the child. There are no indicators of pathology. work of the fortification began.

6 Animal Bones Study 5 Description of the Human Skeleton The bones come from a feature located behind and The skeleton belongs to a young individual and it is below the interior brick parament. The surface of not complete, for various causes, such as the fragil- the context is about 32 square meters, at 1,10-1,40 ity of the bones and their partial destruction by fire. m depth, in the trench XIII/2004. The assemblage There are evident burning traces at the level of the is dominated by three species: pig, cattle and sheep/ maxillae, the mandible, and two vertebral bodies. goat. The other animals (horse, dog, red deer, roe The levels of burning vary from half carbonised to deer and mussel) are present with very few remnants. fully carbonised (completely black) [8]. The remains The remains show typical fresh bone breakage (36%) were carefully recovered, washed in fine-mesh sieves and cut marks (7,21%). Much of them are gnawed by (1 - 0,5 mm) and restored as far possible. carnivores (22,68%), most probably dogs, and part of The skull is fragmented, but important elements have them are burned in different degrees (6,18%). There been recovered. Of the temporal bones, only the right are also parts of animals in anatomical connection, petrous pyramid was preserved. During sieving, we and some of the elements, found at distance from have identified also the hammer (malleus). The oc- each other, could be refitted. Ages were calculated cipital is represented by a fragment of the basilar part on the basis of tooth eruption and wear, as well as the and also a fragment of the right lateral part. The max- stage of epiphyses fusion [11]. illae are represented by the alveolar process, broken The pig is represented by four individuals (based on at the level of the right medial incisive. The ethmoid the mandibles), one less than a year, two about two is represented by fragments of the cribriform plate years old and the fourth about three years old. The and the lateral masses. The sphenoid is almost com- pig remnants consist mostly in head elements (two plete, lacking the pterygoid processes and parts of the cranial and eight mandible fragments) and other parts greater and lesser wings. of the axial skeleton (a piece of atlas, five ribs). The From the mandible, we have recovered the right ram- forelimb is practically absent, except for two scap- us and part of the right body, with the first permanent ulae. In exchange, hindlimb elements from at least molar in the alveolus, and part of the left body, with three individuals are present (two hip bones, two the deciduous first and second premolars in their al- femora, three tibiae, two fibulae and two calcanei). veoli. The rest of the teeth are also present, but loose. Cattle are present with three individuals (based on as- From the vertebral column we have identified a mini- tragali), two of them more than than 4 years old and mum number of four cervical vertebrae, including the third probably younger. The only head elements the atlas, and eight vertebrae from the thoracic and/or are a loose upper molar, a nasal bone and a mandible lumbar regions (vertebral bodies and arches). fragment. The axial skeleton is represented by mini- mum three cervical vertebrae, a thoracic vertebra The ribs are also fragmented. We have recognised six and two ribs. From the forelimb we have identified left ribs (including the first and the second), five right two radii. There are hindlimb elements from three ribs and twelve fragments with uncertain symmetry. individuals (two femora, two tibiae, three astragali, The clavicles are present, in a fragmented stage. The a naviculo-cuboid, a metatarsal and two phalanges).

ISBN: 978-1-61804-020-6 362 Recent Researches in Mechanics

It was difficult to separate sheep and goat bones, so we will consider them as a single TAXON NISP % NISP MNI % MNI category. A minimum number of four indi- Pig (Sus domesticus) 31 32 4 25 viduals were estimated on the basis of tibi- ae. One of them is younger than 15 months, Cattle (Bos taurus) 30 31 3 18,8 two are older then 20 months and the fourth is older then four years. From the head we Sheep/Goat (Ovis aries/Capra hircus) 28 29 4 25 have identified three elements, a small horn- core fragment, part of the left upper jaw and Horse (Equus caballus) 3 3 1 6,3 a loose inferior molar. The axial skeleton Dog (Canis familiaris) 2 2 1 6,3 consists in three vertebrae and two ribs. The fore limb is represented solely by an ulna. Red deer (Cervus elaphus) 1 1 1 6,3 The hind limb, as in the case of pig and cat- tle, is better represented. We have identified Roe deer (Capreolus capreolus) 1 1 1 6,3 two hip bones, two femora, four tibiae and one metatarsal with its phalanges. Mussel (Unio sp.) 1 1 1 6,3 The horse is present with one metatarsal, TOTAL 97 100 16 100 broken in three pieces and gnawed by Tabel 1: Animal species representation (NISP – number of carnivores. identified specimens, MNI – minimum number of individuals) Two bones from a dog aged less than a year and a half were identified, an unfused tibia Sus domesticus Bos taurus Ovis/Capra and a complete metacarpal. Skeletal elements NISP MNE NISP MNE NISP MNE One red deer metatarsal was recovered, burned and showing a high degree of weath- Horncore/Antler 1 1 ering. Cranium 2 2 1 1 1 1 The roe deer is present only with a piece of Mandible 8 4 1 1 antler that exhibits no human activity. Loose teeth 1 1 1 1 1 1 The only worked artefact is a mussel shell with abrasion on one of the extremities. Vertebra 1 1 5 4 3 3 Ribs 5 3 2 2 3 2 7 Conclusions. An Attempt to Scapula 2 2 Recover Data about the Fortifi- cation Consecration Ritual Radius 2 2 Undoubtedly, the archaeological situation 1 1 1 1 that we have just briefly outlined of the area Ulna surrounding the cremated brick wall erected Metacarpus in the “La Cetate” fortified settlement can- not be the result of ordinary habitation pat- Pelvis 2 2 6 2 terns (if we except the deposits which pre- Femur 2 2 4 2 2 2 dated the erection of the wall and underlay it) and this is shown by numerous occasions Tibia 3 3 2 2 4 4 and examples. There is, at the same time, a clear difference from the „normal” levels of Fibula 2 2 the archaeological deposits inside the settle- Tarsal bones 2 2 6 6 1 1 ment in question, which resulted from the place being inhabited for about 80-100 years Metatarsus 1 1 1 1 in the period of this site’s early settlement. Phalanges I-III 2 2 4 4 The unit found at the foot of the fortifica- tion is a constitutive part of it and actually Long bone shaft 1 1 2 1 1 1 marks the beginning of these construction works. The archaeological material, which TOTAL 31 25 30 26 29 24 was found and collected in perfectly clear Table 2: Skeletal element representation for the main species stratigraphical conditions, were deposited or (NISP – Number of Identified Specimens, MNE – Minimum left there in a short, precisely delimited pe- Number of Elements)

ISBN: 978-1-61804-020-6 363 Recent Researches in Mechanics

riod of time (a few hours, perhaps even a few days); Thraco-Dacica 5, 1984, 92-110; idem, Preocupări the fortification structure acted as a kind of seal in în domeniul metalurgiei bronzului în aşezarea geto- respect to the context described. Chronologically dacă de la Bâzdâna, jud. Dolj, 7-8, 1988- and typologically, this is an excellent opportunity for 1989, 15-28; Vlad V. Zirra, D. Pop, Săpăturile de la making associations in order to establish the set of Bâzdâna-Calopăr, „La Cetate”, jud. Dolj – Campa- ceramic forms at the end of the 4th century BC. nia 1993, Arhivele Olteniei, S.N. 10, 1995, 13-27; Vlad V. Zirra, Dan Pop, Sorin Oanţă, On the other hand, the unit or context provides us Şantierul Bâz- with arguments, which are hard to refute, in favour dâna, comuna Calopăr, punctul „La Cetate”, judeţul , AO, S.N. 11, 1996, 5-20. of a ritual or ceremony destined to inaugurate the Dolj. Campania 1994 construction of a monumental defence structure [3] For a contrary opinion, which supports the idea which, in all probability, had a prestige value for the that the bricks which make up the paraments were community that put in place such a project. burned after they were put in place in the masonry structure, see Mircea Babeş, The way the pottery (which came in large amounts Despre fortificaţiile , SCIVA and had various forms, being hand-made or wheel- „Cetăţii Jidovilor“ de la Coţofenii din Dos 48, 1997, 3, 199-236. He mentions a constructively turned) was found or abandoned is very enlightening, similar fortification, dated to the same age and also in addition to the fact that vessels with large, open situated in the Dolj Co., at Coţofenii din Dos (Vlad forms, especially bowls meant for solid or liquid edi- Zirra et. alii, „ bles, predominate at the wall’s northern end and, at La station gétique fortifiée de «Cetatea , Dacia the southern end of the fortification predominate nar- Jidovilor» (Coţofenii din Dos, dép. de Dolj)“ N.S. XXXVII, Bucureşti 1993, 79-157. row-shape vessels, pitcher or cups, meant for keeping or consuming beverages. [4] Attempts at levelling the ground on which the paraments were placed by bringing earth from other We need only imagine further that the „feast”, which locations and compacting the ground levels were the local community organized, with large amounts identified only in random places. of goods, edibles, physical and emotional energy was part of the fortification itself, just as the oblation- [5] Because the wooden mass has been consumed du- goods, meant for gaining the goodwill of the deities ring the long years, it is hard to identify the wooden for a safely enduring building, which could legitimate scaffolding, which probably consisted of thin tree- a prominent position among other settlements merged trunks and (perhaps) of large branches. with the mass of fortification materials. Thus, the [6] Two of them were easily observable, while the human sacrifice (a child in this case) is part and parcel third was largely damaged. of the founding and consecration ritual. From this perspective, the consistent cremation level at the foot [7] The palaeofauna material unfortunately comes of the fortification can be interpreted as a purification almost exclusively from a single trench (S XIII/2004); before beginning the construction of the defensive unfortunately most of it has been lost at present. structure. It is possible, though, for the cremation in [8] M. C. Stiner, S. L. Kuhn, 1995, Differential Bur- question to have been aimed at making water-proof ning, Recrystallization, and Fragmentation of Archa- the soil on which the wall was to be erected. eological Bone, Journal of Archaeological Science, 22, 223–237. Acknowledgement: This paper is suported by the Sec- [9] D. H. Ubelaker, 1989, Human Skeletal Remains: torial Operational Programme Human Resources De- Excavation, Analysis, Interpretation, (2nd Ed.), velopment (SOP HRD), financed from the European Washington, DC: Taraxacum. Social Fund and by the Romanian Government under the contract number SOP HRD/89/1.5/S/59758. [10] T. Sjøvold, 1990, Estimation of stature from long bones utilizing the line of organic correlation, in : References Human Evolution, vol. 5, p. 431-447. [1] A. Van Gennep, The Passage Rites, University of Chicago Press, 1960; E. Cizek, Istoria Romei, 2002. [11] E. Schmid, 1972, Atlas of Animal Bones, for Prehistorians, Archaeologists and Quaternary Geo- [2] C. M. Tătulea, Aşezarea geto-dacică fortificată de logists, Elsevier Publishing Company. la Bâzdâna, judeţul Dolj. Consideraţii preliminare,

ISBN: 978-1-61804-020-6 364