I wish to thank D. Nadali and A. Polcaro, co-directors of the Italian Archaeological Expedition to Tell Zurghul/Nigin (MAIN), for the opportunity to study the Ubaid pottery materials at the site. I am deeply grateful to R. L. Zettler, K. Blanchard (respectively Associate Curator-in-Charge and Fowler/Van Santvoord Keeper of Collections of the Near East Section at Penn Museum of Philadelphia) and G. Brereton (previous curator of ancient Mesopotamian section at the British Museum of London) for allowing me to study the Ubaid materials from Tell al-Muqayyar/Ur stored in their museums. I wish to thank J. S. Baldi and K. Abu Jayyab for inviting me to participate in this volume. I am much indebted to the three anonymous reviewers who contributed with their comments to significantly improve my manuscript.
- 1 The term ‘Ubaid’ is used here only when referring to the ceramic repertoire and exclusively to indi (...)
1This paper provides an updated overview of the Ubaid pottery materials1 recovered at the site of Tell Zurghul. The analysis here proposed is based on a typochronological approach, differing from the style-based approach adopted so far in the analysis of Ubaid pottery of southern Mesopotamia (Oates 1960, 1987, 2010).
2This updated analysis of the Ubaid pottery in southern Mesopotamia is part of a larger work reassessing the entire Ubaid sequence (Volpi 2021). The specific focus of this article is the final phase of the period, the one immediately preceding the End of the World, considered as the structural changes that occur during the Late Chalcolithic period in southern Mesopotamia.
- 2 I prefer the definition of ‘Early’ and ‘Middle Chalcolithic’ periods (instead of Ubaid period) for (...)
- 3 Recent radiocarbon dates are available for northern Mesopotamia and in particular for the autonomou (...)
3The Middle Chalcolithic period in southern Mesopotamia2 is dated between the end of the 6th millenium (ca. 5300-5200 BCE) and the second half of the 5th millennium BCE (ca. 4400-4300 BCE). Considering that no recent radiocarbon dates are available at the moment either for the excavations in Area B of Tell Zurghul or for the coeval sites currently under excavation in the region, for southern Mesopotamia we have to rely on dates from the 1990s from Tell el-‘Oueili, placing the Ubaid 4 phase between ca. 4800-4400 BCE (Forest 1996: 387).3
4About three decades after the last excavations in the prehistoric settlements of southern Mesopotamia, the new excavations in the Area B of Tell Zurghul have again raised the question of what the Ubaid in southern Mesopotamia is (fig. 1a).
Fig. 1 – A. Location of the sites cited in the text; B. Satellite image of Tell Zurghul/Nigin with the location of Area B; C. Limits and fillings of the so-called ‘Koldewey’s pit’.
MAIN, Missione Archeologica Italiana a Nigin.
5Meanwhile, excavations have concentrated on other regions of Greater Mesopotamia, bringing to light the prehistoric events of these regions. This has created a biased but beneficial situation, especially in the analysis of the local diversity of the regions of Greater Mesopotamia involved in the Ubaid phenomenon, considered as a phenomenon of shared formal traits (e.g. the so-called Black-on-buff pottery and the tripartite architecture) over a vast geographical area encompassing the entire Greater Mesopotamia (Henrickson and Thuesen 1989; Carter and Philip 2010). In this scenario, the ‘core-area’ of the phenomenon (i.e. southern Mesopotamia) was not investigated, while new data were accumulated on the ‘peripheral’ regions. This resulted in the development of models and theories (i.e. the ‘interaction spheres’ model proposed by G. J. Stein [2010] and the ‘transformations without a center’ model by P. Karsgaard [2010]) questioning the ‘core-periphery’ approach previously proposed, models and theories that should now be considered in the reanalysis of the Ubaid of southern Mesopotamia.
6In the light of these new approaches, I claim that even the Ubaid of southern Mesopotamia should be analysed as a local phenomenon with its own peculiarities before being considered in a wider geographical perspective.
7The Italian Archaeological Expedition to Nigin/Tell Zurghul (MAIN) has been working on the site since 2015 (fig. 1b). The expedition is co-directed by D. Nadali (Sapienza University of Rome) and A. Polcaro (University of Perugia). Tell Zurghul is located in the modern province of Dhi Qar in southern Iraq, ca. 7 km south-east of Tell al-Hiba/Lagash. The site has an extension of ca. 70 ha, and it is composed of two main mounds surrounded by an extensive lower town. The main period of occupation is the 3rd millennium BCE: the site belongs to the ancient State of Lagash, and it is identified with the Sumerian city of Nigin. According to cuneiform sources, extensive building activities were carried out by the rulers of the first and second dynasties of Lagash (Nadali and Polcaro 2020; Nadali and Verderame 2021). In addition to those dated to the 5th millennium BCE (discussed in this article; see also: Volpi 2020a; Volpi and Fragnoli 2021), traces of occupations dated to the transition between the 4th and the 3rd millennium BCE (Pizzimenti 2020b) and to the very beginning of the 2nd millennium BCE have also been found (Casadei and Oselini 2020; Pizzimenti 2020a).
8The lowest mound of the site, called Mound B, has been investigated so far in three campaigns in 2015, 2017 and 2019. The excavations in Area B have been carried out using the open-area method.
9The area was previously investigated in 1887 for two weeks by R. Koldewey (Koldewey 1887). Remains of Koldewey’s excavations have been recovered on the south-eastern slope (length 35 m; width 3.5 m). Moreover, a pit on the top of the mound (length 6.5 m; width 5.5 m) could probably be attributed to his activities. A small sounding (depth 1.6 m) has been opened in 2019 within the filling of the pit, but its bottom has not been reached yet (fig. 1c).
10The excavations were therefore concentrated on the north-western portion of the mound. Three superimposed architectural phases were discovered immediately under the surface (fig. 2a). Based on the material culture, all the three architectural phases are assigned to the Ubaid 4 phase as defined by Oates (1960).
Fig. 2 – A. Reconstructive section of the architectural phases (phases 1-3) identified in the excavation of Area B at Tell Zurghul/Nigin (computer graphics by L. Volpi and L. Radicioni); B. Composite plan of the same architectural phases with the identification of the cross-section line (A-A1).
MAIN, Missione Archeologica Italiana a Nigin.
11The architectural remains of phase 1 are poorly preserved and only two fragments of overlapped floors (L.250, L.253) belonging to a rectangular room were found (W.251-255; fig. 2b; for assignment of loci to various phases and sub-phases, see table 1).
Table 1 – Summary table with the indication of the phasing, sub-phasing, loci, and periodisation of the archaeological remains recovered in Area B at Tell Zurghul.
Phase |
Sub-phase |
Loci |
Phase 1 |
|
L.250, L.253, W.251, W.252, W.254, W.255 |
Phase 2 |
a |
L.27, L.263, L.271, W.26/270, W.28, W.272 |
b |
L.273, L.364, L.375, I.373, W.256, W.264, W.330, W.356, W.362 |
Phase 3 |
a |
L.258, W.257, W.275, W.361, W.370 |
b |
L.278, L.366, W.276, W.277, W.309, W.315, W.318, W.321, W.363 |
12Phase 2 is divided into two sub-phases. The walls of sub-phase 2a (W.26/270, W.28) trace the orientation of the previous walls of sub-phase 2b (W.264, W.330, W.356, W.362). Belonging to the latter, a floor level (L.364) with in situ materials was found. Unfortunately, the architecture of phase 2 is compromised by the pit on top of the mound (fig. 2b).
13In 2015, below a wall fragment belonging to sub-phase 2b, a deposit of five complete censers was found. While the rest of the floor was cleaned and only small pottery sherds were recovered, these censers were found complete, and it has been suggested by excavators that they were deliberately left in place underneath sub-phase 2b, with the aim of obliterating them and rendering them unusable. A similar deposit of two further censers was found in 2017 under some wall fragments belonging to sub-phase 3a. Considering that the censers found close comparisons with similar vessels recovered exclusively within Temple VI at Tell Abu Shahrein/Eridu in southern Mesopotamia (Safar et al. 1981: fig. 74), the excavators suggested a parallel function as a temple for, at least, the buildings of phases 2 and 3 at Tell Zurghul (Nadali 2020: 4; fig. 2b).
14Phase 3 has been exposed so far only in the northern sector of the excavation area, and it is further subdivided into two sub-phases. The recovered walls, especially those of sub-phase 3b in the Area B North sounding (W.309, W.315, W.318, W.321), present a different dimension and a slightly different plan orientation compared to the ones of phase 2. The recovery of 79 clay net sinkers, associated with a large number of fish bones, would suggest a use of these rooms as an annex to the main temple or as a storage area dedicated to specific operations related to fishing activities. However, the layout and function of phase 3 still remains to be clarified (Caselli and Titolo 2020: 184; Nadali 2020: 6-7).
15All sherds analysed in the present study come from reliable archaeological contexts: directly from floor levels (e.g. L.364 of phase 2b), or from stratigraphic units (fillings) immediately above floor levels and comprised between one sub-phase and the other. The time span is therefore limited according to these stratigraphic principles. Materials from later pits have been excluded, with the exception of those selected from the Koldewey’s pit on top of the mound.
16The ceramic repertoire consists of ca. 4,500 BCE pottery sherds, of which 1,038 are diagnostic sherds (entire or complete vessels, rim and base sherds, handles, lugs, spouts and pouring lips, and painted or decorated sherds; fig. 3a). Entire or complete vessels are attested in very limited numbers, while most of the repertoire (more than 75% of the total) is composed of rim sherds.
Fig. 3 – A. Pie charts with indications of sherd counts from the excavations in Area B at Tell Zurghul/Nigin; B. Ratio between painted and unpainted published repertoires in the cited sites; C. ‘Envelope method’ for comparing the so-called bell-shaped bowls and the so-called bell-shaped beakers.
17The repertoire was first divided into open and closed shapes. Open shapes account for 55% of the total number of diagnostic sherds; closed shapes account for only 9%. The remaining 36% is composed of decorated sherds and all those sherds which cannot be assigned with a certain degree of confidence.
18As for the painted/unpainted repertoire, the former one amounts to ca. 22.5% of the total number of diagnostic sherds, while the unpainted repertoire is more than 77%. Although slightly higher, the number is comparable with the data from coeval levels at Tell el ‘Uwaili and Tell Abu Shahrein/Eridu, where the percentage of the painted repertoire is around 12-13% of the total (Safar et al. 1981: 160; Huot 1987: 296).
19In published data, the ratio between painted and unpainted fragments is frequently in favour of painted fragments, whereas the relevance of unpainted fragments should not be underestimated (fig. 3b). As for Tell Zurghul, it is possible to notice that the unpainted assemblage is morphologically related to the painted one (the same shapes and types are generally recognised as painted and unpainted).
20The criteria of pottery classification adopted in the present paper are based on an ‘intuitive’ typology, namely on the examination of the vessel shape in its entirety and then considering the variability of individual attributes (Read 1989: 164-166; Cocchi Genick 2012: 16-17; Rice 2015: 223). This object-based typology has been selected for two reasons: because the production, although well crafted, is basically hand-made and household-produced; because ancient potters, before standardised productions, presumably perceived their vessels as a whole and not as a list of attributes to be assembled in the manufacture of the vessels (Cocchi Genick 2012: 16-17).
21The broader category for the ceramic classification is the shape, intended as a general and ‘ideal’ definition of the vessel function (e.g. bowls, cups/beakers, holemouth jars, necked jars); the type is the minimum taxonomic unit, and it is considered as a unique aggregate of diagnostic attributes (Hill and Evans 1972; Cocchi Genick 2012: 14-15; D’Andrea 2014: vol. II, 166-170; Bortolini 2016: 652).
- 4 The counting is still in progress for the entire pottery repertoire from Tell Zurghul. At present, (...)
22When possible, the degree of similarity within each type is tested with the ‘envelope method’ (Orton 1987; fig. 3c). Sherd profiles are superimposed at the same scale to highlight dimensional or morphological variations within each type. For ceramic materials from Tell Zurghul, when pottery materials are found in situ they are counted with the method of the ‘Minimum Number of Vessels’, a qualitative and quantitative evaluation for relating the number of sherds to the minimum number of vessels present in each context (Voss and Allen 2010; Rice 2015: 261-264).4
23The most distinctive types are described below from the earliest to the more recent ones, i.e. from those attested in phase 3 to those recovered only within the filling of the pit on top of the mound (figs. 4-6).
- 5 For the Tell al-Muqayyar/Ur vessels, see also figs. 8-9.
- 6 Comparisons are limited exclusively to sites in southern Mesopotamia in order to perform a ceramic (...)
24The pottery of phase 3 is characterised by five types, attested so far only in this phase. Two types of bowls have been identified. The first one is the so-called assiette à marli bowl, a globular bowl with a flattened lip, a widely thickened-
outside rim, and a rounded base (fig. 4a-d). This is a very distinctive type of Oates’ Ubaid 4 phase, and it is mainly attested at Tell el ‘Uwaili (Lebeau 1983: 81, pl. III, V, 1). Assiette à marli bowls are also evidenced at Tell Abu Shahrein/Eridu (Temple Sounding, TS levels VII-VI, Cemetery area, and Hut Sounding, HS levels VII-III; Safar et al. 1981: figs. 75.4-9, 76.1, 77.4, 77.6-7, 129.9, 129.13, 130.6-7, 131.10-12, 132.21), at Tell al-Muqayyar/Ur (Pit F Graves, PFG/Q, Z, LL, OO, and TT; Woolley 1955: 90-100),5 at Warka/Uruk (Sounding K XVII Schnitt I, levels 6-5; Boehmer 1972: Taf. 46, 60, Taf. 47, 87-88), and at Tell Uqair (Sounding IV; Lloyd and Safar. 1943: pl. XXIb.15-16).6
Fig. 4 – Pottery types distinctive of specific architectural phases identified in the excavations of Area B at Tell Zurghul/Nigin.
MAIN, Missione Archeologica Italiana a Nigin.
25The second type is a globular bowl with rounded lip and with a ridge on the inner wall close to the rim (fig. 4e). This kind of bowls is present also at Tell el ‘Uwaili, in levels assigned to both Oates’ Ubaid 3 and Ubaid 4 levels (Lebeau 1983: pl. XI, 6; Lebeau 1991: pl. VIII, 3, 13), and at Tell Abu Shahrein/Eridu (TS lev. IX-VIII, and HS lev. VII-III; Safar et al. 1981: figs. 129.15, 130.11-12, 132.19).
26Two closed shapes are distinctive of this phase. The first one is a holemouth jar with rounded lip. These holemouth jars present some knobs (fig. 4f) or a ridge on the external surface (fig. 4g). Similar vessels have been found at Tell el ‘Uwaili in Ubaid 4 levels (Lebeau 1983: pl. XVIII, 2), at Tell Abu Shahrein/Eridu (HS lev. VII-V; Safar et al. 1981: figs. 129.29, 130.24), and at Warka/Uruk (Grabung III and VI and in Eanna Sounding, lev. XVIIIa-XVIIb-XVIIa; respectively, Boehmer 1972: Taf. 56, 510, Taf. 58, 659 and Hermansen 1990: fig. 6.44), and at Tell Uqair (Sounding IV; Lloyd and Safar 1943: pl. XXIb, 4).
27The second is a small jar (probably, a bottle) with rounded or thinned lip, and with a short, vertical or slightly everted neck (fig. 4h). This type is known only by fragments, and it can be paralleled with sherds from Warka/Uruk (Eanna Sounding, lev. XVIIIa-XVIIIb; Hermansen 1990: fig. 11.99-100; Sounding K XVII Schnitt I, lev. 5-1, and Grabung IV Boehmer 1972: Taf. 48, 107, Taf. 53, 283, Taf. 57, 569).
28Two bowl types occur in both phases 3 and 2 only. They are the slightly carinated bowls with externally-bevelled lip (fig. 4i-j), attested also at Tell al-Muqayyar/Ur (PFG/JJ; Woolley 1955: 97) and at Warka/Uruk (Eanna Sounding, lev. XIXba-XV; Hermansen 1990: fig. 1.1; Sürenhagen 1986: 55, 395), and the so-called bell-shaped bowls with a thinned lip, an everted rim, and a bell-shaped profile of the body (fig. 4k-l). Although there is no clear distinction in the diameter range between this kind of bowl and the so-called bell-shaped beakers (a similar type classified as a cup/beaker because of its shorter diameter and its shallow body), bell-shaped bowls have a diameter greater than 15 cm (fig. 3c). The type is attested also at Tell el ‘Uwaili (Ubaid 3 and Ubaid 4 levels, as well as in Sounding Y 27, lev. 7; see Lebeau 1983: pl. X.2; Calvet 1983: pl. XII.6; Lebeau 1991: pl. IV), at Tell Abu Shahrein/Eridu (TS XIV-VI, HS VII-VI and in the Cemetery area; Safar et al. 1981: figs. 74.11, 78.15), and at Tell al-Muqayyar/Ur (PFG/R and SS; Woolley 1955: 91, 100). The general impression is that the bell-shaped bowls are a type slightly older than the bell-shaped beakers, but unfortunately the data are not enough for the moment to confirm this suggestion.
29Very few pottery types are distinctive of phases 2 and 1 only. This is partly due to the poor state of preservation of the architectural remains, particularly of phase 1.
30The most interesting one is a type of holemouth jar with an internally-bevelled lip, a knob close to the outer rim and a body with a globular profile (fig. 4m-4n). Similar vessels are attested at Tell el ‘Uwaili (eastern mound, lev. D-G; Calvet 1991: pl. XVIII.181), at Tell Abu Shahrein/Eridu (HS, lev. V-II; Safar et al. 1981: fig. 130.23) and at Warka/Uruk (Eanna Sounding, lev. XIXa-XVIIIb; Hermansen 1990: fig. 8.75).
31The so-called hollow clay cones, characterised by a rounded or thinned lip, an everted rim and a point-shaped base have also been found at Tell Zurghul for the moment only in phase 2 (fig. 4o-4p). Hollow clay cones have been found also at Tell el ‘Uwaili (Ubaid 4 levels; Lebeau 1983: pl. XXII.1-3), at Tell Abu Shahrein/Eridu (HS, lev. V-I; Safar et al. 1981: figs. 130.1, 131.1, 132.1-2, 133.1) and at Tell Uqair (unstratified context; Lloyd and Safar 1943: pl. XXII.15-16).
32Lastly, within the filling of the pit on top of the mound, a type of bottom base has been recovered, probably pertaining to globular cups (fig. 4q). This kind of bottom base is a distinctive ceramic trait of the Ubaid 5/LC1 phase, as attested at Tell el ‘Uwaili (eastern mound, liv. C; Calvet 1991: pl. X.100-103) and at Warka/Uruk (Eanna Sounding, lev. XVIIIa-XVIIb; Hermansen 1990: fig. 11.110-112, 11.114; Sounding K XVII Schnitt I, liv. 6-2 and K XVII Schnitt II, liv. 7-1; Grabung III; Grabung IV; Grabung VI; Boehmer 1972: Taf. 54, 420, Taf. 58, 673-674). This type is so far attested at Tell Zurghul only within the filling of the trench, while it is never attested in any of the architectural phases.
33The rest of the repertoire is attested in the entire architectural sequence discovered at Tell Zurghul (phases 3-1). Bowls with both rounded and thinned lip are two of the most common types of bowls attested. The rim of these kind of bowls can be almost vertical (fig. 5a) or everted (fig. 5b-5c). The distribution of these types is very common also in the other Ubaid sites of the region, and their diffusion in the entire Ubaid sequence makes it unnecessary to report comparisons in detail.
Fig. 5 – Pottery types identified in all the three phases (phases 1-3) in the excavations of Area B at Tell Zurghul/Nigin.
MAIN, Missione Archeologica Italiana a Nigin.
34Other types of bowls present an internally-bevelled lip and a slightly carinated body (fig. 5d), as well as a flattened lip with plain (fig. 5e), thickened-inside (fig. 5f) or thickened-outside rim (fig. 5g). Bowls with flattened lip and thickened-inside rims are attested also at Tell el ‘Uwaili (Ubaid 4 levels; Lebeau 1983: pl. XIV.7-8), at Tell Abu Shahrein/Eridu (HS lev. VII-I; Safar et al. 1981: figs. 129.6, 131.8, 132.17, 133.10), and at Warka/Uruk (Eanna Sounding, lev. XIXb-a-XVIIIa; Sounding K XVII Schnitt I, lev. 6-1; Schnitt II, lev. 5-2; Grabung III and IV; see respectively Hermansen 1990: fig. 1.7-10; Boehmer 1972: Taf. 44, 4-7, Taf. 47, 82-84, Taf. 50, 148-149, Taf. 56, 468, 532, Taf. 57, 602).
35Bowls with thickened-inside rims are attested at Tell el ‘Uwaili (Ubaid 4 levels; Lebeau, 1983: pl. IV.1, IV.9), at Tell Abu Shahrein/Eridu (TS lev. VI-XIII, HS VII-I; Safar et al. 1981: figs. 76.2, 84.24, 89.4, 129.3, 130.9), and at Warka/Uruk (Eanna Sounding, lev. XIXba-XVIIIa; Sounding K XVII Schnitt I, lev. 6-1; Schnitt II, lev. 5-2; Grabung III and IV; see respectively Hermansen 1990: fig. 2.16; Boehmer 1972: Taf. 44, 10-11).
36Two common types are represented by globular cups with a thinned lip, and a vertical rim. These vessels can have both a flat (fig. 5h) or a ring base (fig. 5i). Considering that most of the repertoire is composed by rim sherds (fig. 5j), it is not always possible to understand which of the two types most of the sherds belongs to. Parallels for both these types can be found at Tell el ‘Uwaili (eastern mound, lev. C-G; Calvet 1991: pl. IX.93-97, X.99, X.102, X.106; and Ubaid 4 levels; Lebeau 1983: pl. VII.1-9, VIII.1, VIII.3, XXI.5-16, XXI.4), at Tell Abu Shahrein/Eridu (TS, lev. VII-VI, Cemetery area, and in HS, lev. II; Safar et al. 1981: figs. 74.3, 78.1-2, 78.4-8), at Tell al-Muqayyar/Ur (PFG/N, K, P, U, Ubis, X, CC, HH, JJ, MM, TT, as well as in the filling of the Pit, depth 3.50-3.00 m; Woolley, 1955: 88-100), at Warka/Uruk (Eanna Sounding lev.XVIIIa-XIVb; Hermansen 1990: fig. 11.110-111, 11.115; Sürenhagen 1986: 55.371; 58.399; sounding K XVII Schnitt I, lev. 4; Grabung IV; Boehmer, 1972: Taf. 46, 42-54, Taf. 49, 115-123, Taf. 51, 183-184, Taf. 56, 492-496, 525-527, Taf. 57, 580-589, 625-626, Taf. 58, 647-658), Tello/Girsu (unstratified materials; Parrot 1948: fig. 9), and at Tell Uqair (Sounding IV; Lloyd and Safar 1943: pl. XIXb.1-22).
37Another type of cup presents a pronounced carination in the middle of the body (fig. 5k). This type always has a ring base, and it is attested also at Tell el ‘Uwaili (eastern mound, liv. C-G, Ubaid 4 levels, and in Sounding Y 27, liv. 8; respectively Calvet 1991: pl. IX.92; Lebeau 1983: pl. VIII.2; Calvet 1983: pl. XII.1), at Tell Abu Shahrein/Eridu (TS, lev. VII-VI and in the Cemetery area; Safar et al. 1981: fig. 78.3, 78.7), at Tell al-Muqayyar/Ur (PFG/P, S, T, U, W, X, Z, AA, CC, HH, JJ MM, and TT, F, as well as in the filling of the Pit, depth 3.50-3.00 m; Woolley 1955: 89-100), at Warka/Uruk (Eanna Sounding, lev. XIXa-XVIIIb; Hermansen 1990: fig. 6.49, 6.51; Sounding K XVII Schnitt I, liv. 6-5; Boehmer 1972: Taf. 46, 50, 53); at Tell al-Ubaid (Trench, Hut Settlement; Hall and Woolley, 1927: pl. XLIX, T.O. 254), at Tello/Girsu (unstratified materials; Parrot 1948: fig. 9b), and at Tell Uqair (Sounding IV; Lloyd and Safar 1943: pl. XIXb.9, XIXb.12, XIXb.19).
38Bell-shaped beakers are also attested through the entire sequence of Tell Zurghul (fig. 5l-51m). Their presence in the other sites of the region is testified at Tell el ‘Uwaili (Ubaid 4 levels; Lebeau 1983: pl. X.3-8), at Tell Abu Shahrein/Eridu (TS, lev. VII-VI and in the Cemetery area; Safar et al. 1981: figs. 74.1, 78.11, 80.1), and at Tell al-Muqayyar/Ur (PFG/R; Woolley 1955: 91).
39As for closed shapes, one type of jar is characterised by a rounded lip, a vertical rim, and a short, vertical, not-distinct neck (fig. 5n). Strong comparisons with this type are not easily found, but it can be possibly paralleled with some sherds from Tell el ‘Uwaili (Ubaid 4 levels; Lebeau 1983: pl. XVII.5), and from Warka/Uruk (Eanna Sounding, lev. XVIIIa-XVII-XVI; Hermansen 1990: fig. 10.88).
40One of the most common types of jars is a necked jar with a rounded or thinned lip, a slightly everted neck, and a globular or slightly flattened body (fig. 5o-5r). In this case as well, considering that most of the sherds pertain to the upper part of the vessel, it is not always possible to distinguish between jars with globular and slightly flattened bodies. This is a common type, and it is present also at Tell el ‘Uwaili (eastern mound, liv. C-G; Calvet 1991: pl. XIII.131-134, XVI.170-171; Ubaid 4 and Ubaid 3 levels; Lebeau 1983: pl. XVI.2-5, XVII.8-10; Lebeau 1991: pl. IX.9-18), at Tell Abu Shahrein/Eridu (TS, lev. VII-VI, HS, lev. VII-I, and Cemetery area; Safar et al. 1981: figs. 74.8, 79.1, 79.8-9, 80.7, 80.9), at Tell al-Muqayyar/Ur (PFG/GG; Woolley 1955: 96-97), at Warka/Uruk (Eanna Sounding, lev. XIXa-XVIIIb-XVI; Hermansen 1990: fig. 9.77-85; Sounding K XVII Schnitt I, liv. 6-2 and Schnitt II, liv. 7-2; Grabung III; Grabung IV; Grabung VI; see Boehmer 1972: Taf. 50, 163-169, Taf. 51, 304-309, Taf. 57, 569-577), at Tell al-Ubaid (Cemetery area, Gr. 9; Hall and Woolley 1927: pl. XLIX, T.O. 253), at Tell Uqair (Sounding IV; Lloyd and Safar 1943: pl. XXI.5-6) and at Tulul al-Baqarat (TB7; Bruno 2020: tav. 33, 12-19).
41Sherds of the so-called censers have been found in all the three architectural phases. This kind of vessel presents a rounded or flattened lip, a vertical rim, and a distinct, vertical, or slightly everted neck. These vessels have also a carinated-ridged shoulder and a cylindrical body with triangular fenestrations (fig. 5s-t). Similar censers are attested only at Tell Abu Shahrein/Eridu (TS, lev. VII-VI; Safar et al. 1981: fig. 74.13, 74.15).
Fig. 6 – Main pottery types and their relative occurrence in the architectural phases of Area B at Tell Zurghul/Nigin.
42The detailed description of the main ceramic types recovered at Tell Zurghul has enabled us to contextualise the repertoire on the basis of the other sites of the region (table 2).
Table 2 – Comparative chronology chart with the indication of the stratigraphic sequences cited in the text. The arrow below lev. B-A in the eastern mound of Tell el ‘Uwaili indicates that the levels are probably more recent than the others included in phase c.
43With the aim of better understanding the Ubaid ceramic development of the Middle Chalcolithic period in southern Mesopotamia, I reanalysed the repertoire from all the coeval sites with in situ materials. The purpose is to realise an updated typochronological analysis linking the typology with the stratigraphy of the sites, considering the entire repertoire known to date. Therefore, whenever possible, the provenance of all the vessels and sherds published for the site under evaluation has been reconstructed. Each site has its own issues regarding the stratigraphical and the documentary situation, which makes the study even more challenging. In the following section, I highlight some relevant aspects to consider from site to site.
44Tell Abu Shahrein/Eridu is the guide-site for the Ubaid 4 phase. In Oates’ reconstruction (1960), the levels VI-VII of the Temple Sounding (TS), the Hut Sounding (HS) levels, and the graves excavated in the Cemetery area all belong to the Ubaid 4 phase.
45As for the ceramic repertoire, two points are critical: the morphological analysis is based on typological families (or groups) and not on types (see Oates 1960: pl. IV; Safar et al. 1981: 148, fig. 72). Typological families are groups of types clustered together by common morphological features (Peroni 1998: 11-12; Rice 2015: 229-237). However, such distinctions are not appropriate for a strict ceramic periodisation. The second point is related to TS, lev. VII and the lack of the pottery repertoire directly recovered from the temple. Due to its similarity with the TS, lev. VI, the quantitative analysis for lev. VII was considered superfluous (Safar et al. 1981: 154). However, this has created a gap in the sequence, and it is therefore not possible to reconsider the differences between TS, lev. VI and VII, as well as the similarities and/or differences with the end of the phase Ubaid 3 (evidenced by TS, lev. VIII).
46Tell el ‘Uwaili is the second guide-site for the Ubaid pottery materials. The excavations led to the identification of two additional phases (Ubaid 0 and Ubaid 5) besides the existing ones. Structures belonging to the Ubaid 4 phase were investigated in the 1976, 1978, 1981 and 1983 campaigns by extensive excavations (squares X, Y, Z, AA, BB 27-31; Forest 1983: 71-80; Forest 1987: 18-20; see also Huot et al. 1978) and in the deep sounding Y 27 (levels 1-8; Calvet 1983: 25-26). The architectural evidence identified was firstly divided into five levels, from the most recent (lev. 1) to the oldest (lev. 5); they were then grouped into three sub-phases within the Ubaid 4 horizon. Sub-phase 1 (previous lev. 2 and 3) is characterised by a caisson terracing wall; sub-phase 2 (previous lev. 4) is characterised by a multi-cellular building called the granary; sub-phase 3 (previous lev. 5) is composed of a tripartite building and some cellular structures placed in front of the tripartite building (called annexes), for which a functional interpretation as grain storage cells has been proposed (Forest 1983: 73-74, figs. 2-6). Lev. 1, to which no sub-phase was assigned, was identified on the surface and it is characterised by the presence of four pottery kilns (Forest 1983: 71; Huot et al. 1978: 206-207).
- 7 Starting from 1983 campaign, the numbering system was changed, and the sherds were indicated as e.g (...)
47The main issue with the Ubaid 4 excavation at Tell el ‘Uwaili is related to the publishing choices, particularly regarding the numbering system adopted for the 1976-1981 excavation campaigns. Ubaid 4 pottery has been published in detail, but the numbering system for the earlier campaigns (e.g. 76.693 or E 1025) does not allow us to identify the provenance of each sherd according to the stratigraphy of the area (Lebeau 1983).7
- 8 Levels B-A in the eastern mound of Tell el ‘Uwaili (not discussed here) are the most recent levels (...)
48The Ubaid 5 phase was discovered during the 1985 campaign in the eastern mound. Levels C-G, characterised by few architectural remains, presented a peculiar pottery repertoire that “se présentait de façon tout à fait différente” from the Ubaid 4 repertoire (Calvet 1991: 162). This phase is also called Terminal Ubaid and it is considered as contemporary with the LC 1 phase in northern Mesopotamia (see Rothman 2001: table 1.2).8
49Excavations at Tell el ‘Uwaili have been resumed since 2019 by the French expedition (Vallet 2020).
- 9 The zero point is the water level identified in the deepest layer within Pit F (Woolley 1955: tav. (...)
50At Tell al-Muqayyar/Ur, the earliest occupation of the site has been identified in the lowest levels of Pit F (also named Area PFT), a deep trench of ca. 15 × 25 m (depth of ca. 18 m) investigated in the 1929-1930 campaign. In the layers comprised between 1.10 m and 6.10 m asl,9 49 graves with Ubaid pottery materials have been identified. According to their stratigraphic position and their ceramic assemblage, C. L. Woolley identified three phases, called ‘Ur al-Ubaid I-III’, with the phase ‘Ur al-Ubaid I’ being the earliest (Woolley 1955: 19; 87-102).
51For reasons of space, I am not going to describe in detail the subdivision proposed by Woolley (1955: 19, 146-147). Nevertheless, I would like to highlight a few aspects:
- the stratigraphy proposed in the published section (Woolley 1955: pl. 73) can be reinterpreted, as described later;
- the ceramic analysis (Woolley 1955: 146-147) is based on the typology established by Woolley.
52However, his classification is now outdated; moreover, the typologies proposed by Woolley for his excavations at Tell al-Muqayyar/Ur and at Tell al-Ubaid are based on standardised representations of vessels and they do not correspond to any actual pottery specimen (see also Volpi 2020b). It is therefore difficult to reanalyse the pottery repertoire from the Ubaid graves according to the published data.
- 10 See also Dittmann 2006 and Sürenhagen 1999.
53In particular, P. P. Vertesalji has reinterpreted the Ubaid graves from Tell al-Muqayyar/Ur in terms of their chronological attribution (Vertesalji 1984: 30-88).10 The author proposed a distinction into four sub-phases (1-4; the sub-phases 2 and 3 further subdivided into a and b), with sub-phase 4, which is the most recent, almost corresponding to the ‘Ur al-Ubaid III’ phase indicated by Woolley (Vertesalji 1984: 36, table 1). His work is based on a revaluation of the pottery types identified by Woolley and an updated chronological seriation of the painted decorations. Although the proposal by Vertesalji is worth mentioning, such a strict distinction (especially regarding sub-phases 2a, 2b, 3a and 3b) is hardly verifiable, and, in my opinion, the chronological value given to the painted decorations in his reconstruction is overestimated.
54In the debate about the relative chronology of the Pit F graves I propose two approaches, one based on a restudy of Woolley’s data, and one based on a first-hand analysis of the pottery vessels from the Pit F graves stored at the British Museum of London and at the Penn Museum of Philadelphia.
55Old data by Woolley’s excavations are reanalysed with the use of an updated seriation table considering both the graves’ stratigraphic position and the pottery types identified (fig. 7). Six distinct ceramic groups (1-4) are identified, with two groups (A-B) composed of ‘long-duration’ types. Group 1 consists of types aU.22, aU.45, aU.10b, aU.50, aU.27 and aU.6, identified in the graves PFG/SS, UU, OO, MM, NN, QQ, HH, LL. These are the oldest graves in the sequence, found at an altitude between 1.10 and 1.90 m asl. At the stratigraphic level, the graves were all excavated within the Flood stratum; only one of them (PFG/SS) seems to be cut into the underlying clay layer (the graves PFG/PP and RR also seem to be excavated within the clay layer, but their associated ceramic repertoire falls into group A of the long-lasting types). The ceramic repertoire reveals that type aU.6 corresponds to the assiette à marli bowls, a characteristic type of the Ubaid 4 phase, as attested at Tell el ‘Uwaili, Tell Abu Shahrein/Eridu and Tell Zurghul. Although there is only one specimen in PFG/LL, its presence allows this group to be assigned to an early Ubaid 4 phase. In comparison with the sub-phases identified by Vertesalji, the types belonging to this group were all placed in the sub-phase 2a-b.
Fig. 7 – Cross-occurrence seriation table with the indication of the graves (PFG) identified in the excavations of the Pit F at Tell al-Muqayyar/Ur and the pottery types (aU.) proposed by Woolley (1955: 146-147).
56Group 2 consists of types aU.42, aU.41, aU.8, aU.5, aU.16, aU.48, aU.52, aU.28, aU.17, aU.19, aU.29, aU.26, and aU.49, found in the graves PFG/KK, TT, R, CC, EE, AA, AAbis, BB, T, P, Z, U, V. These graves are placed between 2.20 and 3.20 m asl, and they were all found within the ‘Flood stratum’. Although the ceramic repertoire of this group is clearly different from that of groups 1, 3 and 4, it is possible that, as Woolley already stated, the graves found within the ‘Flood stratum’ approximately 2.20 m upwards are mostly contemporary (corresponding partially with the ‘Ur al-Ubaid II’ sub-phase proposed by Woolley 1955: 68-69, pl. 73). These graves were probably excavated within the silty layer from a surface formed over the ‘Flood stratum’ once the flooding had ended.
57Groups 3 and 4 are clearly distinguished based on the pottery repertoire. Group 3 is characterised by types aU.20, aU.44, aU.9a, aU.24, and aU.9b (found in the graves PFG/FF, W, X, GG, L, J, K, N, M), while Group 4 is characterised by types aU.13 and aU.12 (graves PFG/A, B, C, E, D, F).
58Stratigraphically, the graves of Group 3 lie between 3.60 and 5.00 m asl, and they are in an intermediate level between the layer with organic material and the ‘Flood stratum’. They could have been excavated from the surface above, namely the interface between the ‘kiln stratum’/’wasters layer’ and the layer with organic material.
59The graves of Group 4 lie between 5.80 and 6.30 m asl, and they are all located above the interface with the ‘kiln stratum’/’wasters layer’. They were therefore excavated from an overlying surface. In terms of ceramics, the characteristic of Group 3 is the bowl on a high foot (type aU.9a and b), whereas the presence of type aU.13 (beaker with button base) is distinctive of Group 4 and is found in the graves PFG/A, B and C.
60These four groups are clearly distinguished based on the stratigraphy and the ceramic repertoire. Regarding the long-lasting groups of types (A and B), Group A presents many types attested in the graves of Groups 1 to 3 (with types aU.43, aU.47 and aU.32 attested in the graves of Groups 1 and 2), while Group B has types attested only in the graves of Groups 2 and 3. Interestingly, no types of Group A or Group B were found in the graves of Group 4.
61At the same time, and particularly considering the problems related to Woolley’s typological classification described above, I decided to analyse first-hand the pottery vessels from the PFG graves of Pit F stored at the British Museum of London and at the Penn Museum of Philadelphia. These vessels are published here for the first time, and they are relevant as part of a more in-depth study of the Ubaid ceramic repertoire of southern Mesopotamia (figs. 8-9). The repertoire is characterised by a high percentage of cups with a thinned lip, a vertical rim, a globular or flattened body, and a ring base. Noteworthy is the presence of a bell-shaped bowl in the assemblage of PFG/R, and of the assiette à marli bowls within PFG/Q, Z, LL, OO, and TT. All these graves are within the Groups 1 and 2 described above, suggesting the anteriority of these types with respect to the ones attested in the later part of the Ubaid 4 phase (Group 3; Group 4 can be probably assigned to the Ubaid 5/LC 1 phase as proposed by Calvet 1991, see below).
Fig. 8 – Pottery repertoire from some of the graves (PFG) in the Pit F at Tell al-Muqayyar/Ur, analysed by the author at the British Museum of London and at the Penn Museum of Philadelphia.
Fig. 9 – Pottery repertoire from some of the graves (PFG) in the Pit F at Tell al-Muqayyar/Ur, analysed by the author at the British Museum of London and at the Penn Museum of Philadelphia.
62Other sites (such as Warka/Uruk, Tello/Girsu, Tell Uqair, and Tulul al-Baqarat) are included in the analysis. However, for various reasons they are not as relevant as the above-mentioned ones.
63At Warka/Uruk, Ubaid prehistoric levels have been found in the Deep Sounding of the Eanna Area, excavated beneath the central hall of the Limestone temple reaching a depth of ca. 19 m (Eichmann 1989: 39-44). Thirteen levels (from VII to XIX) have been evidenced, identifying a series of architectural structures (VIII2-X; XIIa-XIIb; XIVa-XVI) alternated by levelling layers (VIII1; XI; XIII; XVIIa; XVIIIa; Hermansen 1990: 2; Sürenhagen 1986: 19, fig. 1b). Other contemporary levels have been found in the Anu Ziggurat Area (K XVII; Boehmer 1972). Two trenches (Schnitt I and Schnitt II) and a series of deep soundings (numbered from III to VI) were investigated, unearthing prehistoric materials close to those found at levels XVIII-XII of the Eanna area. As for the pottery materials, the repertoire is mostly contemporary with the Ubaid 5/LC1 materials recovered at Tell el ‘Uwaili. The repertoire from the Eanna Precinct Area was excavated in the 1930s, and it has suffered from an arbitrary selection. However, it was stored at the Vorderasiatisches Museum of Berlin and recently reanalysed by Sürenhagen (1986) and Hermansen (1990). Warka/Uruk is the only site in which the ratio between the painted and unpainted published repertoire favours the unpainted (see fig. 3b).
64Before the excavations at Tell Zurghul, Tello/Girsu was the only site in the Lagash region with Ubaid pottery materials in situ (de Genouillac 1934; Parrot 1948; see also Huh 2008: 23; as for Tell al-Hiba/Lagash, only two Ubaid sherds from surface layers are known from the entire site; Hansen 1970: 244; S. Renette pers. comm.). The main materials were recovered from the excavation in Tell de l’Est, particularly in the Bridge area. Unfortunately, published Ubaid pottery materials from Tello/Girsu are very scarce (Genouillac 1934: pl. 26-34; Parrot 1948: figs. 8-9; see also Pariselle 1983). Since 2015, an expedition from the British Museum of London has resumed work at the Tello/Girsu site. Starting in 2017, the works also encompassed a restoration project on the Bridge, during which several soundings were carried out bringing to light new Ubaid ceramic materials (Rey 2019; see also British Museum (2018). Iraq Scheme: Tello [Online]. Available at: https://www.britishmuseum.org/our-work/international/iraq-scheme/tello).
65The site of Tell Uqair was excavated only for two campaigns in 1940-1941 by an Iraqi expedition (Lloyd and Safar 1943). The excavations have exposed in Sounding IV a domestic structure separated from a huge public building by a street (Lloyd and Safar 1943: 135-138). A deep sounding in the area between the house and the building has revealed seven archaeological levels, some of them earlier than the buildings. However, the ceramic repertoire is quite homogeneous, and it is therefore not possible to recognise further subdivisions according to the stratigraphy (Lloyd and Safar 1943: table XVII). The entire repertoire has been assigned to the Ubaid 4 phase.
66Lastly, Tūlūl al-Baqarat is a site under excavation by an expedition of the University of Turin. The toponym is related to a series of different mounds (TB1-TB8), presenting various occupational periods. TB7 and TB8 have revealed until now the most ancient evidence, from survey and from soundings in TB7 (S1-S4, together with the excavations of Building A in S3; Lippolis 2020a, 2020b; Quirico 2020). The ceramic evidence from phases II-I of Building A and from soundings S1-S4 points toward a dating around the transition between the Middle and the Late Chalcolithic period (Ubaid 5/LC1) and the beginning of the Late Chalcolithic period (LC2; Bruno 2020: 367-371; see also di Michele 2016; Lippolis et al. 2019: 142-149).
67In the previous sections, I described in detail the pottery repertoire from the new excavations at Tell Zurghul, together with an outline of the main contemporary sites in the southern Mesopotamian region.
68Here, considering the abovementioned data, I propose a ceramic periodisation based on the use of a statistical-combinatory analysis represented by a cross-occurrence (type-context) seriation table (figs. 10-12). For the Ubaid pottery, it has been employed for single-site analyses (e.g. Safar et al. 1981: 148, fig. 72; Vertesalji 1984: 36, table 1), but this is the first time (to my best knowledge) it is applied to the entire Ubaid repertoire from all the sites with in situ materials in southern Mesopotamia.
Fig. 10 – Cross-occurrence seriation table of the entire Ubaid repertoire from all the sites with in situ materials in southern Mesopotamia. Distinctive types of phase a and cluster 1.
The numbers in the vertical (y) axis refers to the archaeological sites: 1. Tell Zurghul; 2. Tell el ‘Uwaili; 3. Tell Abu Shahrein/Eridu; 4. Tell al-Muqayyar/Ur; 5. Warka/Uruk; 6. Tell Uqair; 7. Tulul al-Baqarat. Vessels are not to scale.
Fig. 11 – Cross-occurrence seriation table of the entire Ubaid repertoire from all the sites with in situ materials in southern Mesopotamia. Distinctive types of phases b and cluster 2.
The types of phase c are beyond the scope of this paper and are therefore not included here. The numbers in the vertical (y) axis refer to the archaeological sites: 1. Tell Zurghul; 2. Tell el ‘Uwaili; 3. Tell Abu Shahrein/Eridu; 4. Tell al-Muqayyar/Ur; 5. Warka/Uruk; 6. Tell Uqair; 7. Tulul al-Baqarat. Vessels are not to scale.
Fig. 12 – Cross-occurrence seriation table of the entire Ubaid repertoire from all the sites with in situ materials in southern Mesopotamia. Types identified in the entire sequence (from phase a to phase c).
The numbers in the vertical (y) axis refer to the archaeological sites: 1. Tell Zurghul; 2. Tell el ‘Uwaili; 3. Tell Abu Shahrein/Eridu; 4. Tell al-Muqayyar/Ur; 5. Warka/Uruk; 6. Tell Uqair; 7. Tulul al-Baqarat. Vessels are not to scale.
69The method is mainly based on the typological analysis of the repertoire, with the idea of identifying series of recurring types found in association within different archaeological contexts (Dunnell 1970; Duff 1996; Peroni 1998). As described by Vacca, “these associations generate the relative chronological series, that is, series of archaeological contexts placed in a chronological sequence on the basis of their degree of similarity” (Vacca 2020: 184).
70The cross-occurrence table presents on the vertical (y) axis the archaeological contexts in order from the earliest (at the top) to the most recent one (at the bottom). This is the first critical aspect to consider: until enough absolute dates are furnished, relationships between the contexts of different sites are necessarily relative, and their position (above or below) in the cross-occurrence table of each context is also relative. What must be ensured is that the stratigraphic position of the archaeological contexts of each site is internally respected.
71On the horizontal (x) axis, the identified ceramic types are shown. They are also ordered from left (earliest) to right (most recent). The horizontal axes are subject to shifting, with the principle of placing the types recovered in the same contexts as close as possible.
72Thus, groups (or clusters) of types in chronologically-close contexts are visualised. These clusters identify homogeneous archaeological phases, distinguished by characteristic and exclusive ceramic types (Vacca 2020: 186).
73The limitations of this method require some discussion here. The value of the archaeological phases identified is not in question; the problem is that they are bounded by the quality and the quantity of the data, and how accurate the typological analysis is (the second aspect is influenced by the first).
74For the Ubaid pottery, and particularly for the later part of the Middle Chalcolithic period here discussed, the issues related to data quality and quantity have already been described in detail. I want to emphasise here a few aspects again.
75The first relates to the quality of data, particularly regarding the sites of Tell el ‘Uwaili and Tell Abu Shahrein/Eridu. Ubaid 4 materials from Tell el ‘Uwaili have been extensively published, but there is no possibility at present to investigate a potential sub-division among the architectural phases identified. As for Tell Abu Shahrein/Eridu, the lack of repertoire from TS, lev. VII makes it impossible to better understand the continuity/discontinuity between levels VI and VIII.
76The second aspect is related to the quantity of data: in some cases, the types identified in the typological analysis are composed of a single vessel or sherd. This compromises the quality of the investigation, but it is currently the only information available, and it is due to the selection and publication choices, as seen e.g. from Tell al-Muqayyar/Ur. A further aspect is the publication of mainly painted vessels and fragments, which undermines the knowledge of the unpainted repertoire.
77For the Middle Chalcolithic period, the outcomes illustrated in the tables (figs. 10-12) show the existence of two phases (a and b, marked by solid-line rectangles). The two clusters (1 and 2) marked by dashed-line rectangles represents the ‘long-duration’ types in common between only two phases, while cluster 3 (fig. 12) shows some of the types attested for the entire sequence under analysis.
78It is relevant to underline that the figures present only the most distinctive types for each cluster and phase.
79In the analysis of the cross-occurrence tables, some interesting aspects emerge:
- Phase a does not correspond to the Ubaid 3 and Ubaid 4 phases previously described by Oates (1960); it is instead a combination of elements from both phases. This means that, morphologically, the Ubaid 3 phase is hardly recognisable.
- Although not shown in fig. 10, phase a is preceded by a cluster of a few ‘long-duration’ types that connect this ceramic phase with the earlier one. Among these types there are the famous tortoise or lentil jars (one was found in Tell Abu Shahrein/Eridu, TS lev. VIII; Safar et al. 1981: fig. 82.9). Therefore, types recovered in the archaeological levels assigned to the Ubaid 3 phase belong to both phase a and this previous cluster of ‘long-duration’ types. The levels assigned to the Ubaid 3 phase are therefore in connection with both the earlier and the later assemblages (Ubaid 2 and Ubaid 4 termed by Oates), but so far they do not seem to represent an ‘archaeological phase’ on their own.11
- As opposed to the situation of phase a, phase b is coherent with the Ubaid 5/LC1 phase as described by Calvet (1991). The types attested in the levels assigned to phase b are characteristic and exclusive of this phase. This means that, from a morphological point of view, there exists a pottery phase that is chronologically at the transition between the end of the Middle and the beginning of the Late Chalcolithic period (chronologically defined as Ubaid 5/LC1).
- Cluster 1 is characterised by ‘long-duration’ types showing a strong continuity between phases a and b. The same can be said for cluster 2, meaning that also the passage between the Middle and Late Chalcolithic period in southern Mesopotamia is characterised by strong continuity at least in terms of ceramic shapes and types. The presence of a high number of ‘long-duration’ types has already been evidenced by Oates (1960: 47; see also fig. 12).
- Lastly, the so-called phase c, outside of the scope of this paper and therefore not described here in detail, is relevant due to the presence of a number of types in continuity with those of cluster 2 (fig. 11), together with the introduction of other typical Late Chalcolithic types.
80The analysis here proposed is based on the morphological aspects of the repertoire. This element represents the basic distinction from the previous periodisation (Ubaid phases 0-5), which was mainly based on the decorative aspects of the repertoire.
81Focusing on the later stage of the Ubaid pottery sequence, it was decided to pay greater attention to the morphological rather than the decorative aspects of the repertoire for two reasons: from the one hand, a morphological correlation between the painted and the unpainted repertoire was generally observed, at least at Tell Zurghul/Nigin. Same types are attested in both painted and unpainted versions. Once a type is identified, regardless of being painted or not, it can therefore be included in the typochronological seriation. For this reason, it seems reasonable to enlarge the amount of analysable ceramic materials including in the analysis the undecorated repertoire by focusing on its morphological aspects.
82On the other hand, late Ubaid ceramics of southern Mesopotamia are characterised by a simplification of decorative motifs, as already pointed out by Oates (1960). Painted decorations are frequently limited to simple geometric patterns (e.g. single or multiple horizontal bands), and this partly affects a classification of the decorative motifs for chronological determination, because in most of the cases they are too general to allow for further investigation at the level of chronological sub-phases.
83On the basis of the aforementioned observations, this article has attempted to propose a reliable chronological classification based on the morphological aspects of the Ubaid ceramic repertoire.
84In the article, three points concerning the analysis of the Ubaid pottery have been highlighted, particularly regarding the chronological periodisation of the Middle Chalcolithic period.
85The first aspect is related to the updated publication of the repertoire from the ongoing excavations at Tell Zurghul/Nigin. The analysed materials here discussed are meant as an update of the preliminary results published for the 2015 and 2017 campaigns (Volpi 2020a). The repertoire from the three architectural phases identified can all be assigned to the Ubaid 4 phase described by Oates, although some types are attested in phase 3 only, and they may be relevant for proposing an updated ceramic periodisation for the period.
86The second and the third aspects are immediately consequential, because they are related to the reanalysis of the contexts and the pottery materials from the coeval sites in southern Mesopotamia, as well as to the realisation of a cross-occurrence seriation table for identifying the correspondence between ceramic types and contexts, and, therefore, delineating an updated distinction into archaeological phases.
87It has been demonstrated that all the type-sites used so far for a ceramic chronological determination are affected by issues regarding the selection of the published materials, the provenance of the vessels and/or the stratigraphic determination of the contexts. This is not meant as a criticism of the first archaeologists interested in the late prehistory of Mesopotamia; to the contrary, in reconsidering their analyses it is still possible to work on their data to achieve new results. A similar remark was already made by Oates in 1960, reanalysing the materials from Tell al-‘Ubaid and Tell al-Muqayyar/Ur after the study of the Tell Abu Shahrein/Eridu repertoire: “It is, however, no criticism of Woolley to say that, like all pioneers, he raised more questions than he could solve” (Oates 1960: 32).
88The proposed distinction into ceramic phases takes into account only the last part of the period, but it is in line with the future research perspectives on the Ubaid pottery, following what was expressed by J. S. Baldi in the first preliminary report of the new excavations at Tell el ‘Uwaili: “[…] one of the main objectives of the ceramic studies is (now and in the years to come) the definition of a typochronology surely anchored to stratigraphy, so as to establish which types distinctively characterize the various periods of late prehistory in southern Mesopotamia” (Baldi 2020: 114).
89The analysis of the new materials from Tell Zurghul, together with the updated ceramic periodisation here proposed, can be considered as an attempt to rediscuss the chronology of this formative period of the Mesopotamian prehistory.