• 1 Albertazzi, Muti forthcoming; Muti et al. forthcoming.

2From this perspective, it is crucial to focus on textile actors (producers and consumers) thereby promoting multi-scalar approaches to the analysis of the early textile environment,1 in order to set it into a wider analytical framework and bring these actors from the background to the centre of our attention.

• 2 Bill, Johansson 2010; Hyrkäs 2016.

3Over the last decade, start-up entrepreneurship has been referred to as a form of business venturing, that intends to grow large by adopting a combination of businesshypothesis-driven experimentation, iterative product releases, and validated learning. Sociological debate has described start-up entrepreneurship as a cultural phenomenon, involving the popular role of founders and the creation of an identitarian narrative. These specific organizations tend to represent themselves as small communities with worldwide ambitions and need myths and ritual elements to sustain themselves and grow.2 From a broader perspective, both the concepts of start-up and community of practice have common traits, and may both help us in outlining the emergence of new forms of community identity and new forms of labour organization in Middle Bronze Age Cyprus.

• 3 Bombardieri 2017; Bombardieri, Graziadio 2019; Hadjisavvas 2017; Swiny 1986; Swiny et al. 2003.

5The features and processes of the biophysical environment provide a background for and interact with human socio-economic and cultural systems, affecting also the lifeways of textile actors. Textile actors and their lifeways are embedded in a biophysical landscape that shapes and is shaped by socio-economic and cultural systems. Located in the southern region of Cyprus, the Kouris catchment extends over 300 km², from the southern side of the Troodos Massif to the Mediterranean Sea. Most of the area is drained by the Kouris, Limnatis and Kryos Rivers and their tributaries, originating in the Troodos Mountains and in the sedimentary hills. Located in the middle Kouris valley, Erimi-Laonin tou Porakou is one of the four major Bronze Age settlements with evidence of textile activities, along with Sotira-Kaminoudhia (Early Cypriot period = EC), Episkopi-Phaneromeni (late Middle Cypriot to Late Cypriot I period) and the urban centre of Alassa (Late Cypriot period = LC).3

• 4 Bombardieri 2017, p. 349.

6Since 2009, Erimi-Laonin tou Porakou has been systematically investigated by the Italian Erimi Archaeological Project, as a joint research initiative of the University of Turin (Department of Humanities) and the Department of Antiquities of Cyprus. Based on currently available chronological evidence, Erimi-Laonin tou Porakou was occupied during two main periods (Periods 1 and 2). To date, the earlier Period 2 is the best documented, characterized by a short sequence of two consecutive phases of occupation (Phases A and B) dating to the Middle Bronze Age. The following Period 1 occupation follows a lengthy hiatus and is related to a sporadic use of the area during the Hellenistic and Roman periods.4

• 5 Bombardieri 2017, p. 20-72.

8Analysis of the built environment revealed significant differences and analogies between the two areas of the settlement. The workshop complex is coherently organized into discrete units and open-air spaces, as the result of an original homogeneous design. The domestic quarter, likely formed by agglutination of roofed units and open areas, is characterized by long rectangular rooms arranged around rectangular or quadrangular courtyards, connected by entrances and passages. Although the planning and organization of the inhabited space differed, the same building materials and techniques were utilized for the construction of linear architectural spaces in both the settlement areas. The natural limestone bedrock was carefully worked to construct roofed units in both the workshop complex and residential area, where the rectangular building units are carved into the limestone bedrock, thus creating slightly sunken floors. The same building technique was used in the quadrangular open-air spaces of the workshop complex, where installations like emplacements and basins with varying depths and openings were realized.5

• 6 Albertini, Monaco 2017; Bombardieri 2017, p. 358-360, p. 73-128.
• 7 Christofi et al. 2015.

9An extramural funerary area extends towards the East and South of the settlement. The evidence points to an extended necropolis area, with two distinct and contemporary tomb clusters. The southern cluster has been extensively exposed in Area E, where a series of rock-cut tombs was excavated. Human remains were recovered in Tombs 228, 229, 230, 248, 328, 427-430, 464-467 all containing multiple burials. Concerning mortuary practices, the skeletal remains were disarticulated and commingled, suggesting a degree of secondary treatment involving intentional bone manipulation and relocation within the grave.6 A further funerary cluster is located to the east of the settlement (Vounaros cluster). Recent salvage excavations by the Department of Antiquities of Cyprus in the area of Ypsonas-Vounaros (about 400 m East of the hilltop) exposed three graves (Tombs 35-37) belonging to the same chronological horizon as those excavated in Area E, all of which yielded fragmentary human skeletal remains.7

• 8 Boronina et al. 2003.

10Textile processing has been identified as one of the main activities carried out in the workshop complex at Erimi. There is clear evidence for textile dyeing documented from both the occupational phases, as attested by installations for heating and water processing, cooking pots, and containers for liquids. As to the water supply, the Kouris river is the most obvious and prominent hydrological feature of the valley. It is a braided type river, and its continuous flow was interrupted only recently in 1988, by the construction of a modern dam; groundwater is an additional source of freshwater. Several springs characterise the upper part of the catchment; in total, 64 springs have been mapped within the Kouris watershed.8 To the south and southwest of Erimi-Laonin tou Porakou, the presence of two possible minor accumulation basins has been hypothesised, and until 1980 a small stream was recorded on the maps.

• 9 Bombardieri et al. 2015; Muti et al. forthcoming.

11Open spaces in the workshop complex display the high occurrence of liquid processing installations, mainly basins of different types, and drainage channels. These installations allowed people working in this area to process and use an enormous quantity of water and other liquid substances: a total, full capacity of 8 000 litres for the largest basins.9

• 10 Carrera et al. 2009.

12The site location seems particularly relevant to the activities performed. We know that the productive complex is located on the top of the hill, in an area which is the most exposed to the wind. Specifically, the largest open area (courtyard WA IV) is characterized by nine basins and a series of supports for informal structures, such as temporary shelters for workstations or – most probably – for drying processed fibers rather than textiles. This hypothesis appears to be confirmed by the location of these spaces in the northernmost wing of the complex, with direct exposure to wind channelled within the river valley. This phenomenon of channelling has been observed elsewhere and refers to the tendency for winds within a valley to blow parallel to the valley axis for a variety of wind directions above ridge height (Fig. 2).10

• 11 Scirè Calabrisotto et al. 2017; Muti 2017a, p. 23-25; see also Hoysepyan, Willcox 2008.
• 12 Kim et al. 2005.

13Finally, the archaeobotanical evidence strongly suggests that varieties of local wild plants were collected and processed in the productive complex. The most significant and interesting findings are represented by the large amount of mineralised seeds remarkably concentrated in the sediments of some of the basins in unit SA I of the workshop complex, during earliest Phase B. Morphological traits allow the seeds to be identified to the family of Boraginaceae.11 It was demonstrated that among the Boraginaceae, plants belonging to Echium, Lithospermum and Alkanna are well known for producing pigments that can be extracted principally from their roots.12 At Erimi the peculiar occurrence and accumulation of botanical remains from these species in significant contexts and specific features in the workshop seems to suggest that selected plants were intentionally collected, accumulated and processed in specific areas of this productive complex. This evidence, together with the presence of work installations (basins, channels) and associated material (spindle whorls, grindstones, cooking and pouring vessels), strengthens the hypothesis that these plants could have been collected for their dyeing properties.

• 13 Barber 1991, p. 240.

14In sum, interpretation of the evidence strongly suggests that the settlement was deliberately established in this area because of its particular natural potentialities. The availability of wild plants with desirable dyeing properties, the accessibility to various water sources, the position of the hill and its exposure to channelled wind appear to be basic elements in the establishment of a textile production activity, as argued for analogous cases.13

• 14 Muti 2017b, p. 226.

15Homogeneous datasets from settlement and funerary contexts across Cyprus suggest that the textile tool assemblages mainly consist of terracotta spindle whorls and loom weights. Thus, spinning and weaving are basic in the textile chaîne opératoire at MBA Erimi. If compared with coeval contexts, medium-weighted and fast whorls are prevalent at Erimi and in the Kouris valley. Spinning seems thus oriented to the production of thin and medium-thin yarn, likely privileging the use of short-stapled fibres. Indeed, the most recurrent types of spindle whorls recovered from the site are the medium and medium/heavy spindle whorls designed to spin short fibres. But, the recovery of four heavy/very heavy spindle whorls demonstrates that other type of fibres, not strictly shortstapled, were also processed.14 Concerning the weaving assemblage, only one type of loom weight is known for MC period, i.e. the conical weights. As these artefacts were made of dried or low fired clay, they do not often survive in the archaeological record. At Erimi, a concentration of loom weights was found in room SA XII. As to the relationship between weights and thicknesses, these loom weights were part of the same set, suitable for producing thick fabrics (Fig. 3-4).

• 15 McCartney 2017.

16Together with the wide diffusion of aforementioned primary textile tools the evidence for additional processing tools forms a coherent picture. Particularly, the prominence of the pièces esquillées and utilized pieces in the chipped stone assemblage suggests that they may have been associated with textile production. Although use-wear analysis is needed to confirm this, such tools would have served well for breaking flax straw into pieces following retting or the chopping of plants before preparation of dyes or could have been used as scutching knives to remove straw pieces from the inner fibre.15

• 16 Muti et al. forthcoming.

17With regard to fibre, both wool and flax are attested in the area. While the lack of relevant data from Erimi and the absence of kill-off pattern obtained for the ovicaprids in the Kouris valley data-sets point to exploitation exclusively aimed at wool production, indirect evidence suggests that it could have been significant.16

• 17 Lucaks, Pastor 1988; Alt, Pichler 1998; Molnar 2011.

18Preparation of plant fibre for a variety of purposes including cordage, basket and/or textile manufacture is a widely attested task in ancient as well as in modern populations.17

• 18 Bonfiglioli et al. 2004, p. 449.
• 19 Alt, Pichler 1998, p. 390.

19As any other type of physical activity, it involves energy expenditure with variable demands for muscles and joints, and, in some circumstances, the integrity of the bone and dental tissues may be additionally affected. However, wherever other activity-related stress markers such as entheseal changes (indicative of the extent of muscle use) and osteoarthritis (degradation of joint surfaces resulting in bone modification) have been extensively debated, extra-masticatory dental modifications have not received comparable scholarly attention. This gap in knowledge may be ascribed to several reasons such as the accidental recurrence of these lesions. Indeed, not all the activities performed by past populations involved the use of the teeth as tools. Furthermore, some of these dental defects (e.g. chipping) may be attributed to a variety of events including traumatic episodes.18 Generally speaking, in absence of information on their environmental contexts the interpretation of these evidences results to be more challenging.19

• 20 Lucaks, Pastor 1988.

20Aside from general dental attrition consisting in a progressive diminishing of dental enamel due to the frictional contact between the mandibular and maxillary teeth during mastication, two categories of extra-masticatory dental defects associated to plant fibre processing were identified in the archaeological skeletal assemblage.20 They consist of:

• 23 Bombardieri 2017, p. 113-128.

22Interestingly, an unusual wear pattern detected on the lingual surface has been evidenced on a right maxillary central incisor recovered in the chamber Tomb 428, excavated in 2014,23 and will be discussed below in a greater detail.

• 24 Bombardieri 2017, p. 113-128.

24To the West of the dromos, at a mid-point between the chamber entrance and the pit tomb T 427, a small structure was found (T428, US 5), directly carved into the limestone bedrock, used for the ritual display of ceramic vessels, which were found still in situ.24

• 25 Fissore 2017, p. 245-248; Muti 2017b, p. 231-232; Douglas, Muti 2019, p. 22-23.

25An assemblage of 64 objects comes from this deposit, mostly sitting on the chamber floor. The majority of the objects deposited are ceramic vessels, including Red Polished, Red Polished Coarse ware and Drab Polished small and medium size bowls, juglets and jars, with few examples of imported Red Polished III juglets. In addition, four clay spindle whorls with or without incised decoration, along with personal ornamental objects and one complete copper spearhead with rat-tail tang have been recovered within the deposit on the chamber floor (Fig. 6).25

• 26 Albertini, Monaco 2017, p. 319.
• 27 Osterholtz 2015, p. 82.
• 28 Buikstra, Ubelaker 1994.
• 29 Buikstra, Ubelaker 1994.
• 30 Brothwell 1981; Miles 1963; Lovejoy 1985.

28Tomb 428 contained the fragmentary and commingled human remains of at least four individuals, including three adults, two in their 30s (skulls 1 and 2) and one in its 40s (skull 3), and one adolescent (skull 5).26 The number of the deceased was calculated through a combined feature-based approach, namely by calculating the number of the most recurrent anatomical elements.27 Where available, sexually dimorphic cranial traits were used to assess the sex of the deceased on the basis of the Buikstra and Ubelaker’s Standards.28 Only one pair of os coxa was recovered and, thus, utilized to estimate the sex of one of the occupants, which was certainly a male.29 Since the absence of other, more reliable skeletal indicators, age-at-death was estimated on the basis of the stage of eruption and development of the teeth other than on the basis of the dental wear.30 All the dental specimens recovered in the sub-mentioned tomb were macroscopically screened for specific dental defects attributable to the use of teeth as tools (Tab. 1).

• 31 Watson, Haas 2017; Irish 1997.
• 32 Turner, Machado 1983; Tenga et al. 2016.
• 33 Osterholtz 2015, p. 115, 151.

31The discovery of a single case of LSAMAT within the human skeletal remains exhumed at Middle Bronze Age Erimi may be deemed peculiar and worthy of further attention. LSAMAT have been rarely recorded in archaeologically derived human collections and, where they are, the incidence ranges from 45 %31 to 85 %.32 This figure alone is enough reason for claiming that the frequency observed for the Erimi osteological materials is atypical. Indeed, if this unusual wear pattern was correlated to the mere consumption of any kind of less processed foods, then why does only a single individual exhibit this peculiar dental wear pattern in Erimi, along with few cases from the other coeval and closely located Cypriot archaeological contexts? This extra-masticatory dental lesion was detected in only other two cases in Bronze Age Cypriot contexts: in a maxillary dentition of an adult male deposed in Tomb 25A (MC period) from Episkopi-Phaneromeni and in two isolated teeth discovered in the Tomb 65 (ECIII-MCII) at Kalavasos Village Tombs (Fig. 10-11).33

• 34 Osterholtz 2015, p. 254-255.
• 35 Osterholtz 2015, p. 255.

32In both the cases, Osterholtz recommends caution in the interpretation of these features since the poor preservation status of the Cypriot skeletal materials.34 Nevertheless, she points out that these atypical wear patterns seem to occur in the MC period, during a phase of increasing social and economic changes which led to labour specialization and so, possibly, to the involvement of teeth in a variety of activities.35

• 36 Turner, Machado 1983; Irish 1997; Irish, Tuner 1987; Tenga et al. 2016; Watson, Haas 2017; Pechenki (...)
• 37 Berbesque et al. 2012.
• 38 Pechenkina et al. 2002, p. 29.

33Thus, for the Erimi extra-masticatory dental lesion, we favour the suggestion, supported by other studies, of a long-term, repetitive practice assisted by the use of the teeth.36 There is scholarly consensus that the specific tasks that produce this lesion are the peeling and the shredding of plant fibre; but the pulling and holding of any abrasive materials is possible.37 Pechenkina et al. suggest that these tasks were preliminary to the “processing of a cord or a sinew during weaving or basket-making”.38 More likely, we may presume that the authors refer to fibre processing rather than to weaving (Fig. 9).

• 39 Turner, Machado 1983; Irish 1997; Irish, Tuner 1987.
• 40 Turner, Machado 1983.
• 41 Irish 1997.
• 42 Turner, Machado 1983, p. 128; Irish 1997, p. 145.
• 43 Irish 1997, p. 145.

34There are numerous hypotheses regarding which species of plant is responsible for these types of lesions. Studies have documented a strong correlation between the occurrence of LSAMAT and higher rates of caries.39 In these cases, tuberous plants rich in carbohydrates were deemed as responsible of this evidence: manioc in the case of the Archaic Corond (Brazil)40 and sugar cane in the case of a Senegalese series of the early 20^th century.41 The two aforementioned species are believed to have been widely cultivated among these populations; thus, the high rate of caries observed would be the consequence of their consumption.42 To corroborate this theory, Irish found that the diameter of the sugar cane which ranges between 2.5 to 5.0 cm is compatible with this type of abrasion.43

• 44 Watson, Haas 2017; Pechenkina et al. 2002; Liu et al. 2010; Tenga et al. 2016.

35In the same way, Watson does not rule out the hypothesis that the processing of “one or more progenitor species of endemic tubers” explains the frequency of LSAMAT in the examined Archaic Titicaca Basin osteological series (Peru) but the association between LSAMAT and carious lesions is not established.44

• 45 Albertini, Monaco 2017.
• 46 Scirè Calabrisotto 2017, p. 303.
• 47 Turner, Machado 1983, p. 129.

36Examination of 128 teeth from the Erimi osteological collection have revealed only eight carious lesions yielding a rate of 6.2 %.45 This finding is not anomalous since it deals with a community which based its subsistence on non-cariogenic foods, such as terrestrial animal proteins and plants; but it is sufficient to rule out the hypothesis that a similar, cariogenic plant was consumed and/or processed by the ancient inhabitants of the site.46 The same low rate of caries was observed in other hunter-gatherer groups mentioned (Titicaca Basin, Peru; Alfadena, Southern Italy; Xinjiany, Northwest China). Indeed, the only exception to this scenario is represented by the Corond sample which yielded an unexpected high frequency of caries although they were a “meat-eating population”.47

• 48 Watson, Haas 2017.
• 49 Muti 2017b, p. 227.

37In sum, physical anthropologists agree that a comprehensive understanding of these features can be only assured through the combined use of archaeological and ethnographical records.48 Textile tools assemblages from Middle Bronze Age workshop complex and cemetery at Erimi suggest that the processing of short fibres (e.g. wool) was certainly practiced and preferred to the long fibre ones, as demonstrated by the high percentage attestation of medium and medium/heavy spindle whorls. It is important to stress that it is impossible to achieve a complete identification of the fibres processed at Erimi only on the base of the spindle whorl répertoire. From this perspective, the evidence of archaeobotanical analysis from the workshop complex helps us in drawing the picture, as argued above. In fact, the significant presence of dyeing plants remains in specific processing basins in the workshop complex would corroborate the hypothesis of the processing of wool, which absorbs pigments better than other fibres. On the other hand, the low incidence of heavy whorls, ideal for spinning so-called bast-fibres (e.g. linen) and not suitable for short-fibres, indicates that this type of processing was also carried out, possibly at a smaller scale.49 Fibres like wool are softer and inappropriate to be processed with teeth because of their thinness (10-130 µm). Bast-fibres, instead, had to be decorticated (the process of scraping the fibrous materials from the plant) before being cleaned and then processed. It is plausible that this preliminary step, namely the peeling and the reduction into small pieces of the plant fibres, was accomplished by using the teeth; and the low incidence of LSAMAT on the dental remains from Erimi would be consistent with this picture.

• 50 Buccellati 2005 and 2013; Bombardieri 2019.
• 51 The model of habitus, as derived from a controversial philosophical concept, and further adapted to (...)
• 52 Bombardieri, Muti 2018b.

38Current evidence leads to argue for a long formative process of supra-household social structures throughout the Middle Bronze Age in Cyprus, without significant external triggers, that culminates in the more complex institution of the MC III-LC I period. We may assume that during this period some significant elements became strongly apparent in the archaeological record, such as control and management of the industrial production, central productive complexes, development of inter-regional trade networks, and production of prestige/symbolic objects. These are fundamental prerequisites in the subsequent process of state formation in Cyprus. The major driving force of this phenomenon appears to be the gradual development of a network of specialized industrial districts and centres and, specifically, the emergence of increasing specialized labour. In this view, the emergence of social functions in supra-household contexts provides a powerful cohesive tool, when the group begins to extend beyond the level of acquaintanceship.50 Thus, the proliferation of specializations and different functions act as a social glue that cultivates a sense of communal belonging (or ‘solidarity’). During the transition of MC society from household-based to supra-household driven communities, social functions reflect a status specifically related to the technological habitus.51 When a textile ‘start-up’ is established, the status of textile specialized producers gradually emerges. These textile actors become a new social entity involving much more than the technical knowledge or skill associated with undertaking a specialized task: they constitute a textile community of practice.52 Members share common symbols and are involved in a set of relationships that develop and solidify over time. These symbols and relationships give a sense of a common and shared identity, beyond gender and individual connotations.

• 53 Gleba 2007.
• 54 Cottica 2007.
• 55 Douglas, Muti 2019.
• 56 Webb 2015, p. 381-383.
• 57 Crewe 1998, p. 38; Douglas, Muti 2019, p. 26.

39As to gender implications, the symbolism of textile tools is the result of a long process of development all over the ancient Mediterranean,53 involving concepts traditionally associated with the female domain, along with consistent collective ideas constituted mythically (e.g. the metaphor of destiny personified by Fates).54 This major cultural stream and scholarly tradition has strongly affected studies of Cypriote Prehistoric evidence, thus creating a frequent a priori gendering of textile work.55 Webb argues that Middle Bronze Age women, men, and children most likely contributed to workshop-level production; and therefore, there may have been less rigid task differentiation at this time, although she maintains that spindle whorls as textile tools share a close relationship with females.56 Stating the difficulties of associating individual, osteologically sexed bodies with objects in multiple burial funerary contexts, Douglas and Muti observed that only 19 spindle whorls could be attributed to women over a total of seven sexed tombs in the whole MC.57

• 58 Parker Pearson 2005, p. 43-44; Andreou 2015, p. 345.
• 59 Peltenburg 1981, p. 23; Bombardieri 2014, p. 50-51; 2017, p. 239-245; Bombardieri, Muti 2018b, p. 3 (...)

41As to the social role of the individual, it has been argued that communities with specialized functions, such as Erimi, adopted personal adornments or bear special objects to communicate and maintain community identity.58 At Erimi the set of symbolic objects that might possess symbolic capital rather than being common-use objects, or in addition to their functional value, appears significantly related to the textile environment. This is the case of decorated double spindle whorl found in Tomb 231 and comb-shaped picrolite pendants found in units SA II and SA XII of the workshop complex and in Tomb 240, which may be connected with better known terracotta comb models. These objects have been interpreted as reproductions of wool carding combs.59

• 60 Frankel, Webb 2006, p. 175; Douglas, Muti 2019, p. 27.

42In addition, during the MC period spindle whorls are recurrently placed in proximity to the deceased, or even over their bodies as documented in the case for Tomb 231 at Erimi, in a similar way to ornaments or metal objects commonly considered as markers of the social identity (Fig. 12). While groups of whorls have been also identified as spinners’ tool kits in specific burials, recurrent decorative patterns on grouped whorls have been related to kin-groups, thus suggesting that diffusion of similar design patterns among neighboring areas might imply an economic and social relationship.60