• 1 For quantitative data on the relative frequencies of flask fabrics and decorative motifs in each re (...)
• 2 Chovanec 2016; Chovanec 2018.

1Flasks are small, highly decorated, closed vessels, approximately 25 cm high and manufactured in various fabrics across Cyprus for approximately 500 years, from the beginning of the Early Cypriot (EC) to the Middle Cypriot (MC) period. Invariably round-mouthed, these vessels could be securely sealed, easily transported, and probably contained one or more precious substances. They comprise a small percentage of overall pottery assemblages but are relatively common in tombs, where they and their contents may have played an essential role in funerary ceremonies. Their incidence in mortuary contexts may have been higher as flasks are typically soft-fired, making preservation and identification during excavation difficult, especially given the compromised integrity of many tombs. The incidence of flasks as a percentage of domestic assemblages remains contentious due to the relatively small number of excavated settlements and normal processes of assemblage attrition during settlement use and abandonment.1 Although no residue analysis has been conducted, organic deposits have been found in other small Bronze Age (BA) containers from Cyprus, indicating that the flasks may have held similar substances. These substances could include a variety of perfumed oils, medicinal compounds, psychotropic substances, and similar mixtures.2

• 3 E.g. Conkey, Hastorf 1990; Carr, Neitzel 1995; Sackett 1985; Sackett 1990; Weissner 1983; Weissner (...)

2While many flasks lack context, the form, fabrics and decorative style of provenanced vessels reveal a degree of similarity and variation through time and across space, which suggests that they are products of regional workshops. It is possible to suggest further that certain motif combinations conveyed specific information to consumers, including the vessels’ origin and contents. Numerous studies on ‘style’ in archaeology have been valuable in developing models of relationships between pottery and communities of different types and different scales of time, space and social context.3 Function must also be considered; however, style and function form the basis for variability in archaeological material. In this study, the concept of style is applied to decoration, form and manufacturing processes. These can be used as social markers, modes of communication or indicators of individual creativity or group conformity. Ultimately, however, the identification of potters remains contentious and the subjective nature of an investigation into Bronze Age pottery decoration needs to be acknowledged at the outset.

• 4 See Carigiet 2022.

3This paper presents the results of a detailed study of 540 flasks manufactured in the monochrome fabrics of the EC and MC periods (painted flasks are not included).4 It is based on all published provenanced and unprovenanced vessels, with a focus on vessel form and structure and the selection and combination of decorative motifs to identify regional styles.

• 5 Stewart, Åström 1992, p. 64.
• 6 See most recently Bolger, Webb 2013.

4The earliest BA flasks date to the PhiliaEC period (2400/2350-2250 BC). Manufactured in Red Polished Philia ware (RPP), they belong to Stewart’s Type VAa5 and comprise a relatively small percentage of the overall RPP assemblage. They form a homogeneous group with globular bodies, rounded bases, relatively short, usually concave necks and flared or continuous rims. Large perforations appear singly or together on one side or, less frequently, in opposition. Differential firing occurs on some flasks, but these are otherwise consistent with wholly red-fired RPP vessels, which generally have highly polished surfaces and well-levigated, soft-fired fabrics with small amounts of gritty temper.6

• 7 Webb, Frankel 2013.

5Decorative motifs are generally infrequent and minimal on RPP.7 Flasks, however, are always decorated and thus stand apart from the rest of the assemblage. Motifs include parallel lines, unframed parallel line groups, multiple zigzags and chevrons. No two flasks are identical despite this relatively restricted array of motifs, although it is possible to identify several groups of closely related vessels. The first group (fig. 1a), from Philia, Nicosia Ayia Paraskevi and Marki, have a herringbone (or chevron) collar at the neck/body join, delineating the neck from the body and a running zigzag (or chevron) encircling the middle of the neck and the body. The two flasks from Philia have a single visible perforation, possibly one of an opposed pair. The Ayia Paraskevi and Marki flasks have string holes on one side of the upper neck. The second group (fig. 1b), from Philia, Ayia Paraskevi and Deneia, have herringbone collars and angled parallel line groups on the neck. However, the body decoration (angled and horizontal parallel line groups, multiple zigzags, fringed motifs and more running herringbone) is more diverse. The flasks from Philia and Deneia have paired string holes, while those from Ayia Paraskevi have single opposed perforations. The final group (fig. 1c) includes two flasks from Ayia Paraskevi decorated with angled parallel line groups on the neck, parallel lines with a fringe at the neck-body junction and a zigzag on the body. One has double parallel lines framed with angled dashes on the base.

• 8 Webb, Frankel 1999; Webb, Frankel 2008; Frankel, Webb 2012; Crewe 2014; Crewe 2015.
• 9 Swiny 1989, p. 18; Manning 1993, p. 48; Webb 2014a, p. 215.
• 10 See most recently Dikomitou-Eliadou 2019; Dikomitou-Eliadou 2021; Dikomitou-Eliadou, Zomeni 2017.
• 11 Dikomitou-Eliadou, Zomeni 2017.
• 12 Webb 2014a, p. 215.

6While slight differences in decoration (particularly body motifs) perhaps served to identify individual flasks or may reflect normal variability in the production of handmade vessels, the overall homogeneity of this cohort suggests significant conformity of design structure within a broadly uniform RPP assemblage. Homogeneity has been identified as a defining feature of the PhiliaEC, with current views suggesting a widely dispersed, cohesive system with a common material culture and shared cultural practices, including food consumption and burial rites.8 The level of uniformity in flask production and the RPP assemblage further suggests that RPP was manufactured in a limited number of locales by potters adhering to widely accepted models and subsequently distributed to communities across the island.9 This appears to be confirmed by recent petrographic and chemical analyses, which have identified an island-wide, compositionally uniform, calcareous RPP fabric that accounts for 75% of the analysed sample.10 The Ovgos Valley and Vasilia on the north coast are potential candidates as both are known centres of PhiliaEC settlement with access to good-quality calcareous clays.11 While smaller numbers of igneous RPP fabrics at Marki and other PhiliaEC sites suggest the existence of other local or regional workshops in which potters attempted to produce morphologically similar pots, flasks appear only in the calcareous fabric. They are likely to have been produced in one or more northern workshops.12 Their wide distribution suggests the existence of extensive networks linking settlements across the island and a broad degree of cohesion in the function of these small vessels and their contents, specifically in funerary contexts.

• 13 See Webb, Frankel 2013.
• 14 Georgiou, Webb, Frankel 2011, pp. 358-362; Webb, Frankel 2013; Webb 2014a.
• 15 Webb 2014b, pp. 219, 222.

7In contrast to the ceramic uniformity of the PhiliaEC, pottery assemblages from the succeeding ECI-II period (ca 2250-2150 BC) show a marked degree of regional variation.13 Red Polished (RP) ware, derived from RPP, was the dominant fabric in the north, while Red Polished Mottled (RPm) ware accounts for the majority of ceramic production at sites along the south coast from Anoyira in the west to Pyla in the east, and Drab Polished (DP) ware emerged in the west with several minor fabrics, including Black Polished (BP), Brown Polished (BrP) and Red Polished South Coast (RPSC) wares, also in restricted use. The ceramic style zones developed across the island in ECI-II have been linked to a breakdown of the earlier PhiliaEC system and the emergence of well-defined regional entities with more localised primary connections and boundaries of different kinds, promoting new practices of local and regional identity.14 Most utilitarian pottery appears to have been produced in local, village-based workshops, with more specialised, high-value vessels, including flasks, manufactured at regional centres, possibly alongside their contents.15

• 16 Stewart, Åström 1992, pp. 65-67.
• 17 Georgiou, Webb, Frankel 2011, pp. 347-349, figs. 9.12-9.13.

8ECI-II flasks are probably of Stewart’s type VAc.16 Most are similarly sized and consistently taller than those of the PhiliaEC, with an estimated capacity almost double that of PhiliaEC flasks. As a vessel type, they comprise 4.4% of the excavated funerary assemblage at Karmi but occur more frequently at Vounous (approximately 7.8%) and Psematismenos (9.2%).17 The most common shape is the so-called “leather form”, with a globular body and rounded base. Some vessels from the south coast, however, have flat bases. Tapered necks and everted rims occur more frequently than other neck/rim types. String holes are standard and often opposed but again show some regional diversity. Lugs are present on some flasks and, on rare occasions, co-occur with string holes. This arrangement, however, was short-lived and confined to vessels from the north coast. Most flasks were manufactured in RP and its variants. The remaining vessels are BrP or early versions of BP. Although the motifs found on PhiliaEC flasks remained dominant (parallel lines, zigzags and parallel line groups), there is a greater variety in their placement, and additional elements, including framed motifs, circles and diamonds, complement the standard designs.

• 18 Webb et al. 2009.
• 19 See Webb 2018; Webb 2020.
• 20 Webb et al. 2009, p. 248.

11The smaller sample of flasks from the cemetery at Karmi Lapatsa (fig. 2b) is also relatively homogeneous in shape and decorated with motifs like those found at Vounous and, more generally, on the north coast in this period. The first two vessels (without context, Dep 13.3c) have close decorative parallels with Vounous flasks, suggesting they were manufactured there. The remaining flasks (Tomb 14A.16 & 17) exhibit typical north coast traits, and more specifically, in the case of Tomb 14A.16, they parallel Vounous18 while Tomb 14A.17 shows similarities with flasks from Lapithos.19 The short-lived ear-lug flasks (fig. 2c) with both string holes and lugs on the upper body also appear at Vounous and Karmi, although probably manufactured at Vounous. Together with commonalities in mortuary practices and other aspects of material culture, these similarities in flask form and decoration suggest strong connections between these communities, with the small hill villages of Karmi probably established as offshoots of the larger settlement at Vounous.20

• 21 Eccleston, Frankel, Webb 2011, pp. 263-265, 269; Webb 2014a, p. 217.
• 22 See Georgiou, Webb, Frankel 2011, pp. 347-351, fig. 9.10.
• 23 Georgiou, Webb, Frankel 2011, pp. 346-347.

12While north coast flasks show a solid regional signature in ECI-II, the central south coast flasks are more diverse at both the site and regional levels. The highly varied Psematismenos flask cohort (fig. 2d) does not conform to Stewart’s flask typology, which was devised almost entirely based on pottery from the north and centre of the island and has little in common with other aspects of north-coast flask production. The Psematismenos flasks were manufactured in Red Polished I-II (incised) (RPI-II[inc]), BrP and RPSC fabrics. Their elemental composition is distinct from that of the locally-produced RPmI-II ware, suggesting the use of non-local clay sources; they also show considerable internal diversity in the proportions of different trace elements and appear to reflect the use of a variety of finer clay sources collected from sedimentary deposits along the south coastal strip.21 They were likely brought to the site from several production centres. Given their frequent appearance in burial deposits,22 it is also likely that they and their contents were primarily, if not exclusively, acquired for use in funerary ceremonies. Flasks are the only incised vessels found in the tombs at Psematismenos and the only vessels on which use-wear is consistently absent, suggesting that they were new at the time of deposition and intended for a single-use, specialised purpose associated with the burial event.23

• 24 Webb 2014a, p. 219.
• 25 Webb, Knapp 2021.

15ECI-II flasks were produced in several regional workshops on both the north and south coasts (and presumably elsewhere) and appear to have mainly moved within their regions of manufacture and only occasionally beyond. RPI-II(inc), RPSC and BrP, all fabrics reserved for specific vessel types, circulated primarily along the south coast, with some movement beyond this core area to the central lowlands, where both RPSC and BrP have been found at Marki and Alambra.24 The production and distribution of flasks, possibly associated with their contents, are best seen in emerging specialised production locales, increased regional mobility and multiple small-scale interactions between larger and smaller settlements.25

• 26 Steel 2004; Webb et al. 2009.
• 27 Webb 2014a.
• 28 Herscher 1973; Crewe 2014.

16ECIII-MCI (ca 2150-1850 BC) was a transitional period with a significant rise in population and settlement density, growing resource competition and a renegotiation of social relationships across the island.26 This appears to have led to more extensive interaction between the central lowlands and the north coast.27 The west and the Karpas Peninsula, however, remained relatively isolated.28

• 29 Stewart, Åström 1992, pp. 68-73.

18The earlier leather-form vessels were replaced in ECIII-MCI by gourd-shaped flasks with piriform or globular bodies and rounded bases of Stewart’s type VBa.29 Rim types now included bell-shaped and continuous versions, and necks tend to be straight rather than tapered. String holes continue to be found most often on the rim rather than the upper neck, but there is an increase in the number of vessels with single lugs on the rim. Overall, flask height increased, and body diameter reduced. The rim aperture is also smaller than on PhiliaEC and ECI-II flasks, and there is an overall reduction in vessel capacity (fig. 3). Flasks now varied more in size and included both diminutive examples and those far exceeding the norm. Parallel lines are the dominant decorative motif, used in isolation or combined with secondary motifs, such as circles, framed motifs and zigzags. However, regional preferences in motif choice are visible, with parallel lines used more frequently on the north coast and zigzags on the south coast.

• 30 Washbourne 2000.
• 31 See Dikaios 1940; Webb et al. 2009.
• 32 Herscher 1991.

20Seven RPIII flasks are decorated with a distinctive net pattern (fig. 4a). This pattern, which Washbourne30 suggests may be a schematic representation of a woven net used to transport or store these round-based vessels, occurs more generally on flasks and other vessel types from the north and centre of the island in this period.31 The flasks with this specific motif form, however, have been found only at Vounous, Palealona, Lapithos, Dikomo Mavro Nero and Ayia Paraskevi, suggesting relatively limited movement from one or more workshops on the north coast, most likely located at Vounous. The use of the motif in its more generic form on other vessel types on the south coast (Episkopi, Kalavasos, Sotira) suggests, as Herscher32 has argued, some iconographic connections between the two areas.

• 33 Brewster 2004.
• 34 Webb 2018, pp. 213-214; Webb 2020, p. 476.
• 35 Frankel, Webb 2007, pp. 154-156.
• 36 Webb 2019.
• 37 Webb 2020, p. 476.

21Production of BP, including flasks (fig. 4b), began on the north coast in ECI-II. By MCI, however, Deneia had become the primary production centre,33 and most BP vessels found in tombs at Lapithos are likely to have travelled there from Deneia.34 BP flasks produced at Deneia differ from those manufactured on the north coast. They are usually smaller, occasionally flat-based and decorated with deeply incised, typically idiosyncratic ‘Deneia’ (RP and BP) motifs.35 These vessels probably moved along a well-defined route northward from Deneia through the Agirda pass, perhaps established as part of Lapithos’ copper procurement network.36 As Webb37 has noted, the vessels which moved between Deneia and Lapithos were principally fine table wares and cooking pots rather than storage or transport vessels, probably reflecting the movement of people rather than commodities and the presence of people from Deneia at Lapithos and vice versa.

• 38 Frankel, Webb 2012; Webb 2019; Webb 2014a.
• 39 Frankel, Webb 2006, p. 147.

25The homogeneity visible in PhiliaEC material culture extends to all facets of PhiliaEC flask manufacture: fabric, surface treatment, shape and the limited range and placement of decorative motifs. These flasks moved within a widely dispersed, inclusive system with vessels produced at larger settlements and distributed to smaller villages through established interregional networks, possibly in reciprocal exchange for copper.38 This situation changed radically in ECI-II when regionally specific ceramic signatures became visible across the island. The flask assemblage shows significant differences in form and decorative style from north to south and between fabrics, with recovery contexts suggesting multiple small-scale interactions between settlements within regions but relatively limited movement between areas. Locally produced RPSC and BrP flasks were dispersed along the south coast, and RPI-II and BP flasks were on the central north coast and, to a lesser extent, the central lowlands. At Marki, the only settlement with excavated deposits of ECI-II in the central lowlands, the ceramic assemblage belongs to the south-central style zone, but the presence of a small quantity of north-coast RPI-II indicates some north/central movement of pottery and other goods.39 Notably, while ECI-II levels at Marki produced an RPmI-II assemblage typical of the south/central style zone, the RPSC, BrP and RPI-II(inc) flasks found on the south coast rarely occur, if at all at Marki or elsewhere in the centre of the island and appear to have been distributed exclusively in the south.

• 40 Webb 2017, p. 172.
• 41 Webb 2014a, p. 222.
• 42 Herscher 1991, pp. 47-48.

26The recovery contexts of ECIII-MCI flasks suggest different patterns of interaction and a greater degree of inter-regional movement. The isolation that characterised the ECI-II period gave way in ECIII-MCI to an increasing homogeneity in material culture and a more extensive interaction sphere in the central lowlands and the north coast. Flask production continued in local fabrics across the island, probably in dedicated workshops associated with their contents’ production. RPIII vessels, including flasks, were transported more widely between regions than other wares, with finely decorated vessels moving south and west to sites like Erimi, where RPIII comprises 11.2% of the published tomb assemblage.40 However, secondary or ‘minor’ fabrics are less often found beyond their manufacturing domain. Their distribution patterns may have operated within reciprocal competing networks separate from those of the more dominant fabrics,41 perhaps reflecting an organised trade in geographically specific commodities. A greater flask size and form variety in this period suggests more diversity in their contents and possibly in their function. For example, the small BP flasks may have contained a liquid that required dispensing one drop at a time. The small mouth aperture would have allowed for this and the use of a stopper. The much larger flasks, particularly those from the Karpas peninsula, have broader mouth openings and opposed perforations below the rim, suggesting they were sealed differently. In this case, the contents were clearly produced, stored, transported and perhaps used in larger quantities than those in smaller vessels. Whatever these contents were, they may have been more readily available in this region. This, in turn, suggests a complex system of production and distribution likely involving the growing ‘commercialisation’ of these special-purpose vessels and, quite possibly, an increase in the variety of their contents. As Herscher42 has argued, with reference to the R&BP flasks, the distinctive shape and decorative style of these vessels make them instantly recognisable and, as such, may have served as a ‘trademark’ or ‘guarantee’ of the nature or quality of their contents or/and their place of origin.