Introduction1

1The excavation of the building complex, located to the north-west of the citadel at Mycenae, was first undertaken by Papadimitriou and Petsas during several months in 1950 and 1951, uncovering a series of rooms filled with ceramics (Papadimitriou & Petsas 1950; 1951). The current excavation program was initiated in 2000 to complete the work of the early excavators for publication and to expand the excavation for a better understanding of the complex, its pottery, and the settlement of Mycenae during the LH IIIA2 period. The work of each season since has revealed a structure of several levels set on and built into the sloped hillside, all or part of that originally covered with an upper story (Fig. 20.1). The earliest understood use of the rooms excavated in the 1950s was pottery storage, then little by little were revealed, during the new project, signs of ceramic production, elements of industrial complexity, and purely domestic contexts. Well established at this point through systematic recovery of archaeological evidence during ten held seasons and the analysis of finds, both old and new, are the uses of Petsas House, as well as the date and cause of its destruction late in the LH IIIA2 period, due to an earthquake and localised fires (Iakovidis 2000; 2001; 2002; 2003; 2004; 2005; 2006; 2007; 2008; 2012; Petrakos 2013; Shelton 2010; 2015).

2The pottery of Petsas House is important for the study of Mycenaean ceramics, especially in LH IIIA2, stylistically and otherwise, and as an assemblage that potentially demonstrates the level and nature of consumer demand – the quantity and variety of vases, and other terra cotta objects for that matter, necessary to be in production, in storage, and available for consumption in any or all use contexts (Shelton 2009: 221-228). Because of the variety of wares and vessel types available together with vastly different levels of production values in practice, I believe that the ceramics produced by the potters at Petsas House reflect a population-wide consumer interest, both horizontally across the settlement and vertically among different socio-economic levels. The corpus also exemplifies terrifically the nature of the changing relationship among strata of the society and the effects of production and consumer strategies in ceramics. This to me suggests a significant socio-economic status for the individuals responsible for the design, production, and distribution of such diverse market potential.

3With this opportunity, I would like to emphasise the character and status of the inhabitants of Petsas House through an examination of their residence and business, physically and conceptually, and through the features and material remains associated with the functional character of the complex. In this way, I would like to attempt a reconstruction of the demographics of a settlement workshop by populating the spaces and activities of individual Mycenaeans at work and at home and in so doing, begin to explore the social dimensions of pottery production.

1. The House

4Our current understanding of the architecture and its design indicates an organisation of the building into functional units, as might be expected in a Late Bronze Age context, combining habitation use, craft production, and large-scale storage. The domestic areas, both preserved and reconstructed, are of surprisingly different characters. The closeness in space of diverse functional elements, like a grand staircase and grain stores, is also unusual. The building design also exhibits some consideration for the delivery and movement of goods with wide well-made stairs and a ramp for wheeled traffic or heavy loads; a strong indication that part of this structure, at least, was intended to be actively accessible, both from outside and from within, and was built with craft accommodations in mind. There was ample space for living and work activities, accommodating a significant number of people. It is also quite possible that the structure was at least used, if not owned as well, by a single-family unit. The complexity of the structure indicates functional and demographic complexity as well.

5Petsas House was a terraced two-story structure with habitation over the ground-floor storerooms based on material recovered high in the destruction fill, clearly fallen from above, most importantly wall painting fragments, many from an extensive fresco dump (Shelton 2015b), and the excavation of several potential points of access to a now-missing upper story. However, this structure also has different levels of habitation, conceptually and literally. A purely domestic context was encountered in the building’s interior at the ground floor level in Room Sigma, and food stores were found adjacent to the ramped entrance in Room Ypsilon, where at least II pithoi were built into the floor of the room. A significant amount of foodstuffs could be stored here indicating wealth and bounty for the residents or simply many mouths to feed. The possible second story not withstanding, the bulk of the evidence for human activity in Petsas House suggests primarily industrial use even as firm evidence for habitation confirms that this structure was also used as a home by generations of family members.

6Only the south-east comer of Room Sigma was preserved, built into and up against the bedrock of the upper terrace2. The room was damaged first in a violent earthquake destruction followed by a localised fire, and was further destroyed by terracing activity and dumping during the Hellenistic period in this part of the slope. Sigma’s domestic function is reconstructed from a distinct lack of artefacts recovered in it, with only five small, undecorated cups (two conical cups and three shallow cups with vertical loop-handles) for consumption and a tripod cooking pot in fragments on the floor near the east wall. The tripod had burned at an extremely high temperature so that the silicate elements in the clay vitrified and the vessel exploded across the room. Additionally, the walls were plastered and painted in solid red and blue, the only ground level area with painted surfaces.

7The floor of the room was bedrock, simply flattened, with a thin layer of soil packed into it in spots. Remarkable is an infant burial in a small and shallow pit3 in the bedrock under floor level. Only two tiny fragments of bone were recovered. The date of the ceramics suggests that the burial occurred soon after the construction of the building in LH IIIA1 and perhaps as much as three quarters of a century before the final destruction of the building, several generations later. Within the pit were uncovered three whole vases: a conical cup, a miniature conical cup with linear decoration, and an exceptional miniature ‘sippy’ goblet with ivy decoration and fashioned with a built-in lid and strainer mouthpiece at a right angle to the ring-handle (Fig. 20.2). I believe this exceptional vase confirms that the original inhabitants of Petsas House were involved in ceramic production. The goblet was constructed with tremendous care and thought for the needs of an infant or small child, and I suggest that it was given as a token of very real feeling to accompany the lost member of this family, to provide eternal nourishment, and perhaps to represent a craft and art that this child would never produce. Together with the vases was a necklace of beads: gold spherical and ivy-shaped with granulated detail, lapis lazuli, and blue glass. The imported materials and advanced technology exhibited in the beads, signal a rather unusual and extravagant gift for an infant and a marker of potential wealth and status for the family. This deposit to me is the most ‘human’ of all the features encountered so far in our excavation. Something about the death of a child shines a light on the lives that were impacted and the ordinary events of birth and death in an extraordinary context.

8The reconstructed habitation areas indicate several options for the make-up of the residents. On the one hand, there are characteristics of building construction and decoration associated with the affluence of an elite population and yet, also domestic space of straight forward and basic make-up with a minimum of material objects. The variety of evidence could indicate a complex social context as the household and activities within it, in this case further complicated in association with workshop spaces, are inseparable from the broader cultural context (Nevett 2010: 4-5). There is strong evidence for generations of craftsmen, and likely craftswomen, from the outset living in Petsas House, but there are also clear signs of heterogeneous levels of socio-economic status.

1.1. Storage

9One of the primary uses of the building, clearly recognised since its initial discovery, was for the storage of mass quantities of pottery, the best example being Storeroom Alpha that contained upwards of 500 vases of 20 different shapes, over 90 % of them decorated fineware, all with a strong piriform or ‘pedestalled’ profile. The vases are of very high production quality and could reflect the work of a single potter, a specialist in this characteristic shape of LH IIIA2. Even though the majority are highly decorated and represent a substantial investment in production time, the output could represent as little as a month’s labour for an experienced potter working fulltime (Knappett 2001; Whitelaw 2001). The vessels are primarily closed, fineware storage vessels (stirrup and piriform jars of at least three graduated sizes) but include a number of open vases, and a few undecorated examples of the analogous shapes. There are a few pieces that seem to be shaped with slightly less precision and are consistently produced in inferior quality clay. The decorative patterns and syntax are rather simple and include a narrow range of motives, but are also completely typical of the period. The variation observed in the execution of the decorative motives indicates several painters working on the vases of a single master-potter and perhaps one or two apprentices. The specialisation in the piriform vessel profile contributes to the sense of extreme organisation in storage practices, exhibited in pots arranged on shelves by shape and size around three sides of the room (Fig. 20.3). The 500+ vases were packed in, which would seem to suggest limited available space. The large structure was equipped, however, with numerous storage rooms, albeit, also filled in a similar manner. A more likely cause was the intentional organisation of the material by shape and size, potentially as a form of inventory control, for ease of retrieval, or simply personal preference.

10The organisation of the room content, of course, could be a matter of function as related to shape. This seems to be the case in other rooms of the building, such as Storeroom Epsilon, where predominantly undecorated vessels were stored, especially open shapes for drinking and serving, including kylikes and bowls of a vast size/scale range. Some of these examples are similar enough in fine execution to see further specialist production, while the huge number of rather carelessly produced carinated kylikes could have been produced by any number of individuals in a range of skill levels – since precise shaping was not apparently the necessary goal. The proportion overall of undecorated to decorated vases, and of kylikes to all the other shapes, open and closed, is comparable to the proportions observed culture-wide in many and various use-contexts. It is highly unlikely that all of these vases were intended for use in this building itself (unlike those at Pylos for example, see Whitelaw 2001), although there may have been ample facilities for group feasting on the upper floor that have not survived. The large number of fineware storage vessels would certainly be redundant all in this same closed and limited context. The majority of the vases recovered from Petsas House were new or unused and should reflect, at the very least, the ‘cottage’ industry of this group of individuals, whether a traditional kin-based family group or a group of craftsmen, such as a guild.

1.2. Production

11An examination of some technical details of the ceramic craft may provide information about modes of production and individual elements of style. With such a corpus, such a vast inventory of vases, there is bound to be a higher percentage of errors or non-standardised details, those not usually observed in the more ‘polished’ or ‘finished’ output that ended up in use contexts. These details include the sometimes awkward combination of wheel-thrown and handmade elements, unintentional paint drips and runs, fingerprints and tool marks (Fig. 20.4), and the introduction of new shapes and decoration, sometimes with better or worse results. We are also able to recognise inventory practices in action where a simple but effective marking system was used to count the numerous undecorated vases. The manufactured output of Petsas House suggests a good deal of mass production, especially of a limited range of shapes that must be significant to the needs of the culture, or at least an interested segment of it, while there continues alongside the production of a few individual and highly original pieces, often requiring a great expense of labour. New, experimental shapes are also produced complimenting older, more traditional shapes, rather than simply replacing them. Most remarkable is the variety of ceramics produced and stored here that exhibits a breadth of specialisation in a full range of techniques: fineware of course, decorated and undecorated, but also cookwares, and figures, all made on the wheel, together with coarse vessels built up at least initially from coils, and hundreds of figurines and many miniature vases all fashioned by hand.

12The exact location of the production centre is still speculative but I believe that the main potting area may have been in Room Pi in association with a large stone slab that forms a platform or bench installation built into the north-east corner of the room, and near a convenient and necessary water source – a well in the southern part of the room. A series of post-holes found in recent excavation lend further support to this reconstruction as they may be related to the use of a potting-wheel in this space. Another important area for the production of pottery was a large yard or court to the west of the storerooms and, at the edge of that space, a newly discovered kiln.

13Room Π, the Well: The greatest quantity of ceramic material found so far comes from the well context (Fig. 20.5). This material is clearly related to the pottery vessels in storage as demonstrated through ceramic joins between the well and deposits in other rooms in the building. Excavated to a depth of 12.35 m, the deposit is characterised by a fill of moist red-brown soil together with fragments of charcoal, burned mudbrick, and frescoes – building debris – mixed with animal bone, clay, a vast quantity of pottery, and other small finds of ceramic, stone, ivory, glass, and fragments of Linear B tablets. The deposit is the result of a single, extended clean-up event after the initial destruction of the building4. Many thousands of whole and restorable vases have been recovered from the well deposit; the majority is undecorated fine ware, especially kylikes of several varieties (Shelton 2014), together with some exceptional decorated pieces and a good deal of cookware. It is great luck that we have such a range of materials in this deposit, which represent the building blocks of activity and life in the structure: frescoes on timber-frame walls, pottery of all types, classes, and functions, figurines and other small finds both sympolic and decorative, debris from habitation as well as possibly, some of its inhabitants. An on-going study of the faunal material from the well has revealed in the bottom levels of the well dog remains of various sizes and at least two species, several clearly fully articulated and likely victims of the earthquake destruction. In this instance, we can reanimate an unexpected population in a workshop context, but one for which we have tangible evidence unlike that for their human companions.

2. Craftspeople/Artists

14By late in the 14^th c. BC, these potters at work in Petsas House were producing vessels and other ceramic objects for most segments of the population, and for a wide variety of consumption patterns and functions, including thousands of kylikes, perhaps as many as 8,000 at the time of the building’s destruction. Over 95 % of the kylikes produced were undecorated and roughly two-thirds of those were ones with a carinated profile (FS 267). The variety and qualities of available wares suggest a consistent consumer need among a widespread proportion of the population, and at different levels of the social hierarchy. Some of these vessel shapes, like the carinated kylix (Fig. 20.6), were clearly mass-produced. However, this bulk quantity would not necessarily have required either more or less of an investment in time and labour than the often larger, usually decorated, and always better-made pieces (Fig. 20.7). The fact that the production of labour-intensive individual pieces appears to have occurred simultaneous to the mass-production of other vessels would suggest a greater number of craftsmen at work, at least for potting, since a much smaller percentage of the vases was decorated, whether by polishing or painting. The necessary variety of production qualities could also accommodate potters at various skill levels, while the variety of wares would encourage specialisation and a division of production tasks that could be more easily organised into a limited workshop space. More essential for the determination of this quantity and variety of production, especially in the limits of a household context, is consumer demand rather than individual artistic choice.

15Potters are often thought of in quite simple terms, almost always anonymous, even though they produce such an important part of our daily needs. Perhaps it is due to the banality of the everyday consumables they produce, only ever elevated to a ‘minor art’ at best. The ubiquitous and common nature of the product, projects back onto its maker. Pottery production is also considered a craft and rarely an art although the design and production of most Mycenaean pieces was truly an art – highly unlikely though, to have ever been considered or treated as art at the time – the possible exception being pictorial pottery, which nonetheless was so limited in comparable production numbers. This influences us to imagine, or reconstruct, craftsmen of simple, straightforward means and status; working in a cottage industry of local importance. This is a very false impression indeed when we look at the evidence for the trade of pottery to destinations and cultures all over the Mediterranean and the evidence of foreign contact and knowledge of cross-cultural preferences and consumer demand exhibited by the potters themselves.

3. Markets

16The material excavated at Petsas House indicates a relationship with different segments of the local population – the surrounding settlement, the elite, and the palatial administration, as well as wider spheres of interest, both regionally, and internationally. Most of the ceramics recovered in the excavation are for local consumption surely; there is ample material for use in the Mycenaean home as well as in the tomb. Whether this is free market production and consumption by a range of consumers, or a more controlled, centralised production is debatable; support in part for the latter comes from the Linear B evidence, albeit very fragmentary and spotty. The tablets suggest some level of palatial interest, potentially control; unless, of course, we consider the possibility of elite and/or merchant literacy, and perhaps we should.

17The production of figurines (Fig. 20.8) and related objects should also be of primarily local interest and consumer need. Although it is completely logical that ceramic objects of all types would be produced by potters, it is indeed not always the case, especially where industrialisation and specialisation are dominant. The technological process of sculpting the clay to form the small-scale figurines is different from the throwing of wheel-made vessels. However, in Mycenaean vases, elements, such as handles and spouts, are formed by hand in very much the same manner as figurines are worked. Wheel-thrown figures are in fact almost a hybrid of ceramic-forming techniques and often include appendages or elements that could be confused with spouts among vase fragments (French 1985)5. I would argue that in figurine manufacture, we have represented other members of the potting family, especially women, who in ethnographic studies are more likely to produce handmade and coil-made pottery rather than use the fast-moving wheel. It is also interesting that in these same studies, figurative and animal forms are a common response when new marketing opportunities become available (Vincentelli 2006; 2008: 4).

18Based in part on the sophisticated architecture and its painted decoration, one might assume that the inhabitants of Petsas House must have been of elite status, certainly to access and participate in the exchange of commodities transported over long distances and across cultural boundaries. The ceramic evidence suggests a sophistication of pottery design and production ‘to order’ that would require a knowledge of overseas market preferences. This also indicates the possibility of a real distribution centre within the Mycenaean settlement context. Vases that were designed for export to Cyprus, and elsewhere, and the presence at Petsas House of foreign exotica also suggest a rather high status for these individuals; so does the clear connection to the palace.

19So, it is very likely that we have in Storeroom Alpha, at least, the products of a master-potter, a specialist in the piriform profile, and his or her assistants. There is no direct indication of gender in the production of these particular vases or in other contexts and the very few partial fingerprints left on vases are not conclusive indicators. The number of miniature vases, many of them wheel-made, could indicate individuals with smaller hands, either women or even children. In a traditional production environment, especially in the household-based workshop, women produce most of the pottery and children grow up with constant exposure to the craft, often being incorporated into production at a very early age. Craft skills are both inherited and observed; they shape the body, creating movements and gestures that are learned often early on among traditional female potters (Vincentelli 2008: 8). Only when ceramic production becomes industrialised do men assume the bulk of the production and usually a great deal of specialisation and labour organisation is evident (Costin 1991). As I have observed elsewhere (Shelton 2010), in Petsas House, the nature of the pottery itself seems to signal a transitional period from production for the home, the settlement, and the bulk of the population, to one of a mandated or more restricted industry with more specialised and focused consumer needs, and perhaps a more limited clientele. The demographics of the workshop at Petsas House may also have been in transition from multi-generational family members to a stratified group of professional craftsmen. Construction phases observed later in the structure’s lifecycle, which alter circulation patterns and access, could perhaps reflect functional changes in the make-up of the workforce and their relationship to household and workshop spaces. However, there remained a rather limited area for production even if we imagine some of our workshop employees living off-site.

20The potters of Petsas House, at the time of its destruction at least, were likely to be primarily men based on traditional societies where potting is a fulltime enterprise, foot-wheeled technology predominates, and the updraught kiln is consistently in use (Carlton 2008: 5). These ethnographic studies indicate that female potters are more likely to use hand-building and open-fire techniques. Nonetheless, even when women are rare in pottery production, they are involved in clay processing, the removal of pots from the wheel, drying and firing (as well as buying and selling) (Vincentelli 2008: 4, 8-9; Carlton 2008: 4). Where ceramic production continues to be based around the household, rather than removed to detached workshops, women tend to be involved in the production process, with pottery formed by men often decorated by women (Carlton 2008: 5-8)6. The variety of ceramic wares and products, found at Petsas House, is more impressive still than are the number of objects available at the moment of the destruction, when the structure came crashing down. Even the mass production of thousands of vessels and tens of thousand of kylikes could be accomplished by a relatively small number of potters, especially ones working fulltime. There is indeed evidence for specialisation and mass production of pottery in this context, however, there is also evidence for the continued production of a variety of ceramic wares and products – many at the hands, I believe, of a broader and more traditional cross-section of producers. These are the individuals that we want to know more about in order to gain insight into their lives and the conditions and environment in which they existed. It is not an easy task but one that can be pieced together from bits of disparate evidence and modeled through ethnographic studies. Petsas House is a huge architectural and industrial complex that was also a family residence. It seems straight forward enough, but this is not just any family of the settlement at Mycenae. The pottery, the tablets, and the foreign contacts indicate a significant relationship with other community members and with other communities, with the Palace and with the wider market economy of the Late Bronze Age Mediterranean.