Evidence for Household Industries on Thera and Kea

These are not 'workshops' - not, that is, the center of a single specialized craft - but households where one or more (and usually more) industrial activities took place.

The problem of how to identify industrial activity in a domestic context is addressed: what sorts of material evidence have we? Some artifacts are readily identifiable, but many others which probably had an industrial function are not specialized enough in form to suggest this, except by contextual association. By investigating whole assemblages, we will have more success in distinguishing between domestic and industrial utensils. We can then progress towards an analysis of patterns of household industries at settlements where large areas have been excavated and recorded.

Despite obvious differences, Akrotiri and Ayia Irini prove to have features in common. One important shared element is the separation of activities, at least in the large houses: residential rooms are kept separate from storage and industrial rooms, usually by putting them on a different floor.

On Kea certainly, and perhaps also on Thera, work-areas could house a variety of activities - up to three or four crafts or industries within a single room. Individual households did not specialize. But the evidence suggests that the settlement at Ayia Irini as a whole may have specialized in producing certain manufactured items, perhaps for export.

INTRODUCTION

Both in Crete and on the Mainland, the palaces supported resident full-time craftsmen, some of whose workshops have been recognized. Craft specialization was not necessarily confined to the palatial centers. Even where one cannot locate the workshop itself, one may deduce its existence from the products. For example, the distinctive local pottery of Akrotiri (Marthari 1987) was certainly made by specialist potters on Thera, probably in the vicinity of the town.

Attempts to develop a methodology for the identification of workshops include recent studies (Branigan 1983; Eveley 1988; Tournavitou 1988) which highlight the problems, and offer much stricter definitions than have hitherto prevailed. A room is now accepted as a workshop if it contains waste material and rejects, together with examples of some of the following: raw material, unfinished objects, specialized equipment, and perhaps a store of objects all of the same type representing finished products. Such workshops are usually the center of a single specialized craft or industry.

Within this strict definition of the term, there is at Akrotiri no single building or room which might be called a workshop. Authoritative claims for their identification (Doumas 1983, 53; S. Marinatos 1970, 43; 1974, 29) are based on evidence for the existence of work areas in each house. At Ayia Irini on Kea, just about every building in the Late Cycladic I-II settlement provides evidence suggestive of one or more (and nearly always more) industrial and/or craft activities.

Clear distinctions need to be drawn between craft specialization and domestic production. One of the important differences between non-palatial permanent workshops and domestic workshops, it has been suggested (Tournavitou 1988, 447), is that the latter were used only seasonally or sporadically, in response to particular household needs, and did not provide the whole livelihood of the people concerned. It is questionable, however, whether they should be termed 'workshops' at all, if they catered only to the needs of the individual households in which they were located. The vaguer term 'work area' (Eveley 1988, 399) meets the case better. Both work areas and permanent workshops (the latter involving craft specialization) can coexist in a non-palatial domestic setting. Some work areas may indeed have been permanent workshops, but have not produced sufficient evidence to meet the strict archaeological criteria for identification as such. The term 'Household Industries' serves as a convenient catch-all, leaving open the question of specialization and permanence.

Unlike the study of the specialized workshop, the analysis of patterns of household industries requires large areas of settlement sites to have been excavated and published. There are few such sites. Gournia and Palaikastro in Crete, and Phylakopi on Milos are the obvious candidates, all excavated in the early years of this century. The data available to us are quite inadequate. Objects which may or may not be 'industrial' can only be understood by examining the whole context in which they were used.

Ayia Irini on Kea (Fig. 1) provides the sorts of information that we need. We can investigate contexts, reconstitute whole assemblages, and ultimately produce distribution plans and charts which should have some statistical validity. But it also presents difficulties of a sort that are familiar from other sites with multi-storeyed buildings: collapse of the houses spilled the contents of upper floors into the rooms below, sometimes over a wide trajectory. In such cases, apparent assemblages should be viewed with a critical eye. Even so, the presence of certain types of objects in the same house at the same time, though perhaps on different floors, can still be very suggestive.

How does one identify industrial activity in a domestic context? More particularly, how does one distinguish objects having an industrial application from those whose use is purely domestic? Georgiou (1983, 75) points out that industrial vessels tend to be so complex that their use must have been limited to a single specific purpose. For Ayia Irini, she has produced a valuable corpus (1986) of ceramic objects used in home industries, crafts, or specialized domestic processes, as distinct from domestic ware for cooking, eating, and storage. She is clearly right that specialized industrial objects are essential for certain specific industrial processes, and are thus easily identifiable. The difficulty lies in differentiating between those many other less specialized objects which might have had a variety of either domestic or industrial uses.

It is quite possible that in the houses at Ayia Irini, particularly in those rather few rooms which had fixed hearths, domestic and small-scale industrial activities could have gone side by side. Although they may not have shared precisely the same equipment, they may have had some of the same sorts of equipment in common. The heating of liquids, for example, may be necessary for both domestic and industrial purposes, and a vessel such as the tripod jar could have been used in either. At Mycenaean Pylos, the vessels and utensils needed for the perfume industry were precisely the same as those found in an ordinary kitchen (Shelmerdine 1985, 47).

The way to proceed is to investigate whole contexts, which would include any architectural peculiarities or fixed features, built hearths or other evidence of deliberate localized use of fire, any raw materials, and the complete assemblage of man-made artifacts. The evidence from Akrotiri will be invaluable, since its unique state of preservation may provide reliable information which elsewhere is much more ambiguous.

TYPES OF MATERIAL EVIDENCE

Not surprisingly, the objects by which we can demonstrate household industrial or craft activity are the same sorts as those used to identify workshops. They all fall into one or other of the following broad categories: waste products, raw material, unfinished products, specialized vessels, other specialized equipment, and tools. Those types of vessels or tools which could be multi-functional should be assumed to be domestic, unless there are good grounds for thinking them industrial.

     Waste products (Fig. 2):

Akrotiri (Warren 1979) and Ayia Irini have produced drill cores from the manufacture of stone vessels, plus good examples of the finished products. Cores and vessels at both sites include local and imported stones. Sometimes, perhaps frequently, drill cores were recycled as pestles, and so may be found in contexts far removed from the work area where they originated.

On Kea, there are also a good many waste pieces of deer antler, a few pieces of slag, and many lumps of litharge.

     Raw material:

Most notable at Ayia Irini are fragments of copper ingots. Many samples of minerals have also been found, particularly in House A (Cummer and Schofield 1984, Appendix 2). Some relate to metallurgy, but more were probably used for producing pigments. A bowl containing red pigment is reported in the West House at Akrotiri (N. Marinatos 1983, 5).

     Unfinished products:

Both sites have produced unfinished stone vessels: a large red marble jar from Room 6 of the House of the Ladies at Akrotiri (Warren 1978, 1979), and a few examples from Ayia Irini, including a purple rhyton (Fig. 3) found in Room 16 of House A (Cummer and Schofield 1984, no. 396). Neither vessel was found with workshop debris, and neither room in any way conforms to the current definition of a lapidary's workshop. At most, it may be thought likely that the intention was to finish them somewhere within the buildings where they were found. If so, at least part of the process of manufacturing stone vessels may at non-palatial island sites have been a household industry (albeit a very skilled one), rather than a craft confined to specialist workshops.

     Specialized vessels:

Ayia Irini has more fireboxes (Fig. 4) than any other site. Georgiou (1986, 4-22) published 90 of them, and knew of fewer than ten at Akrotiri. She argues for a function connected with the manufacture of aromatic substances by a purifying technique, and suggests (1983, 86) that tripod pots and cauldrons might have had related functions. Warren (1987) further suggests that her Type c spreading bowls and the strainer jars (Georgiou 1986, nos. 147-151, 173-176) may likewise have been used in perfumery. Fireboxes often cluster in small numbers, and their wide distribution on Kea indicates that they were used in almost every household.

Almost as numerous are the crucibles (Fig. 5), most of which were apparently used on Kea to melt down bronze. They too have a wide distribution, suggesting that though this sort of metal-working may have been a skilled occupation, it was not a particularly specialized one. Crucibles do not feature among the published material from Akrotiri.

Many other vessels may have had an industrial function, but are either not specialized enough to suggest this except by contextual association, or are so peculiar that we are unable to determine what they were used for.

     Other specialized equipment:

Some clearly relate to metal-working, such as stone moulds. A very fine mould (Fig. 6) from House A at Ayia Irini has five faces, each for producing a different tool (Cummer and Schofield 1984, no. 978). Other types of objects may be related to metal-working, such as stone ladles, lead spoons, and terracotta firestands. One of the few known matching pairs of the latter came from Room Delta 2 at Akrotiri (S. Marinatos 1971, Pl. 29a, 101a), and a single fragment from the site is also illustrated (S. Marinatos 1969, Pl. 38). Several examples from Ayia Irini are published by Georgiou (1986, 23-28), who points out that on Kea they are never found in pairs, and argues that they were used in an industrial process.

Other specialized equipment includes terracotta loom weights and spindle whorls, and potter's wheel disks. Pottery production cannot be considered a household industry: the thousands of locally-made pots at both sites must be the output of more or less full-time potters equipped with kilns, located presumably at the extremities of or outside the towns. The presence of a few potter's wheel disks (Fig. 7) in houses at Ayia Irini, mostly in Middle Cycladic contexts (Georgiou 1986, 36-39) is therefore interesting. It would be worth investigating whether potters and metal-workers could have been members of the same families, or even the same people, as has been suggested for Crete (MacGillivray 1987, 277-278).

     Tools:

It is difficult to know exactly what most of them were used for. Microscopic examination of wear patterns may give clues, but investigation of contexts might sometimes reveal more.

Large numbers of bone tools and at least two bronze borers are recorded from Akrotiri (S. Marinatos 1974, 34, Pl. 84b, c; 1976, Pl. 52e). Among the bone tools at Ayia Irini studied by Krzyszkowska (Cummer and Schofield 1984, 43-45, Pl. 32), many are awls or scrapers, which may be related to some industrial activity, such as leather-working (Fig. 8). There are also a few bone knives, but for both these and the bronze blades it is not possible to say whether their use was domestic, industrial, or both.

Stone tools present the same problems. Relatively large numbers of blades and flakes indicate that obsidian was used in the Late Bronze Age, though precisely what for is another question (cf. Torrence in Davis 1986, 93). Pounder-grinders (Fig. 9), mortars, and pestles are all likely to be multi-functional. Very often they are probably associated with food preparation, like those found in the mill installations at Akrotiri (Doumas, 1983, 54; S. Marinatos 1970, 13-15, Pl. 10-12; 1972, 22-24, Pl. 40, 42). But contexts suggest that some had an industrial use, as in basement rooms 4/4a in the West House (S. Marinatos 1974, 29-31, Pl. 64-66). As already noted, there is evidence for the grinding of pigments on both Thera and Kea, and probably several of the mortars, pestles and pounders were so used. An unpublished marble pestle at Ayia Irini is encrusted with red colouring matter, and two pounder-grinders evidently used for grinding pigments were found upstairs in the West House together with the bowl containing red colouring matter (S. Marinatos 1974, 27, Pl. 59).

     Other vessels:

Many pots of less specialized shapes may have had an industrial function, suggested by their find-spots. One such is a miniature jug, locally produced at Ayia Irini, and presumably used for pouring small quantities of liquids. They do not conform to a standard size, or to an absolutely standard shape (Fig. 10). Contexts suggest that at least sometimes, possibly always, they could have had an industrial use.

Nor should we overlook the ubiquitous conical cup. Found in their thousands, they must surely be multi-functional; but it is worth our while to check carefully those contexts where they cluster in significant quantities. (By significant, I mean groups of more than 250; anything fewer than that is commonplace, at least on Kea).

I deliberately avoid the question of a possible ritual use for these cups (N. Marinatos 1983, 15-16). The thesis (N. Marinatos 1984) that in both Crete and Thera there was a connection between ritual and economy, including industrial activities, may be applicable to Kea also (Cummer and Schofield 1984, 39). For the present, however, let us concentrate exclusively on the industrial element. In any case, though conical cups have sometimes been found in Crete in what seem to be purely ritual contexts, it is presumptuous to assume that they necessarily fulfilled the same purposes both in Crete and outside it. It is one thing to speak of fireboxes, a highly specialized shape which surely did always have basically the same function, and quite another thing to talk of conical cups - an unspecialized shape which could and demonstrably did fulfill a variety of functions, some of them industrial at Ayia Irini.

A few cups served as paint-pots and contain colouring matter. Many probably served as lamps, as indicated by scorch marks at their rims. A significant number were used in conjunction with fire, but not in the same ways: patterns of charring vary. As with other ambiguous artifacts, we should investigate the contexts to determine whether conical cups may frequently have served an industrial purpose.

CONTEXTS

Three areas at Akrotiri were identified by Marinatos as 'workshops' and are described in some detail in the preliminary reports: Room Gamma 1 (1970, 43-44), Basement Room 4/4a in the West House (1974, 29-31, Pl. 63b-67a), and on the south side of Xeste 3 (1976, 22).

Gamma 1, which contained a hearth, a clay chest, and benches, produced several grooved stone hammers, and many other stone tools, as did the adjoining Room Gamma 2. It was considered to be a work area possibly of coppersmiths, but no metal or metallurgical equipment was found, and the identification is unlikely. The so-called 'workshop' area of Xeste 3, likewise, was characterized by large numbers of stone hammers and other tools. Use of the working areas both here and in Gamma 1 and 2 seems to have been confined to the brief period between the earthquake and the eruption. The work in question probably was simply part of the programme of demolition and repair of buildings, for which there is much evidence (Doumas, 1983, 134).

The situation in the West House was quite different. Basement Room 4/4a is interpreted both as the kitchen and as a metallurgical workshop. There is no mention of a hearth. Contents of the room included a stone mortar, pounder-grinders and other stone tools; a fragment possibly of keroussite ore; a jar; two 'cooking pots', one of them containing a heavy whitish substance; and a bowl caked with the same substance. The process is tentatively interpreted as cupellation of silver. Scientific analysis should settle the question; meanwhile some sort of metallurgical work area is a possibility here. The discovery of a stone mould (S. Marinatos 1974, 34, Pl. 83c) in the West House (in an unspecified context) is further evidence of an interest in metal-working.

For Ayia Irini, a useful starting point is provided by Georgiou's tables showing the distribution of fireboxes and firestands (1986, 6, 11, 12, 25). House A (Cummer and Schofield 1984) immediately stands out. The eastern half of House A contained the residential quarters and reception areas on the ground floor and upper storey, with storage and working areas in the cellars beneath. The west side of the complex was more obviously a working area, at both cellar and ground floor level.

My Tables I-III illustrate the distribution within five architectural units in House A of materials, vessels, equipment and tools which were or may have been associated with industrial activities. These are particularly striking assemblages, but similar objects were found in other rooms (and in earlier or later deposits in these same rooms). One of the assemblages, in Rooms A.17 and A.18, dates to Late Cycladic I, and thus approximately contemporary with the houses at Akrotiri; the others all date to Late Cycladic II.

Room A.7 (which may also have contained a domestic shrine) had several storage pithoi fixed in place. This room and its ante-room A.11 are at ground level, and probably did not have another storey above. The finds thus represent actual room contents. The other rooms are all cellars, and the finds represent boh cellar contents and material collapsed from the floor(s) above. Generally there is more collapse material, but one cannot always be sure which is which. Room 16 had a large jar set into its floor with ashes nearby. 

In summary, House A has produced evidence of several industries, including metal-working, pigment-grinding, antler-working, lapidary activity, manufacture of aromatics, and textile. Metal-working is the most prominent, perhaps partly because it is the easiest to demonstrate, since the equipment is specialized. There is no real reason to doubt that these industries were carried on within House A itself, though in some cases one may doubt that they were practiced within the architectural unit where the equipment was found. The area above basement rooms A.19 - A.32 and A.25 - A.27, for example, included large pottery store-rooms, and it is likely that the crucibles and other specialized equipment were also stored there, for use in other parts of the building.

There is no good evidence that storage and industry shared the same space. Store-rooms were for storage, not work. But work-rooms may have housed a variety of activities. Every room that contained clear evidence of industrial and/or craft activity produced equipment associated with more than one - usually three or four. As for the distinction between domestic and industrial activities, final answers must await detailed statistical analyses. It is likely that both sorts of work could co-exist within the same architectural unit, but not often within the same room, unless sharing of a hearth was necessary. Some ceramic assemblages which are commonly labeled 'domestic' may prove to be industrial.

The working areas within House A concentrate on the west side, away from the public and residential wing. At Akrotiri, residential areas are separated from work and storage rooms by putting the former on the upper floors and the latter generally on the ground floor (Doumas 1983, 53, 54; Shaw 1978, 433). House A differs from Theran houses in major ways, not least in having deep cellars with one or occasionally two storeys above. The cellars share the storage and some of the industrial functions of the Theran ground-floor and semi-basement rooms; but those industrial and craft activities which require good light were almost certainly centered on ground-floor rooms, which are the equivalent of the Theran upper storey. At Akrotiri, weaving is the only craft attested on the upper floors (Doumas 1983, 117).

Household industries were not confined to House A. House L includes a courtyard with a hearth full of bits of bronze, while the rooms immediately to the east, L.12 and L.13, produced metallurgical equipment, and also fireboxes. To the north, Rooms L.3 and L.7 (which may or may not be part of the same house) likewise contained crucibles and fireboxes, plus material probably associated with the grinding of pigments, and about 1,150 conical cups.

One other architectural unit at Ayia Irini has a striking assemblage. Rooms W.23, W.24, W.25 and W.33 all contain fireboxes. W.23 has in addition a concentration of lead objects and of litharge, ten miniature jugs, and 627 conical cups.

Georgiou (1986, 53) concludes that there was no industrial quarter in the town, and probably no full-time specialist craftsmen (except perhaps potters). Rather, the wide dispersal of industrial activities suggests part-timers in family workshops. Something os the sort has been suggested for Crete (van Effenterre 1983). We may add that, while it is true that material relating to at least some industrial activity has been found in every Late Cycladic I-II building at Ayia Irini, only House A has convincing evidence for almost the complete range of crafts and industries which can be demonstrated on the site; and each industrial unit there may have supported three or four different industries.

Among the industrial equipment represented at Ayia Irini, fireboxes and metallurgical gear are particularly common, occurring in most houses. Whatever expert knowledge was required for recasting bronze or manufacturing aromatics was shared by just about every household. Ayia Irini as a settlement appears to have specialized in certain industries, and this may be related to external trade. Further research will show whether we are justified in drawing a distinction between crafts and industries (manufacture of stone bowls, weaving, etc.) which produced only for home consumption, and those deliberately producing surpluses for export.

In a sense, Ayia Irini can be viewed as one big 'workshop'. If the town as a whole manufactured many of its own essential needs, and in addition specialized in producing certain manufactured items for export, maybe other towns did likewise. On Kea, the nature of those industries required specialized equipment that is easily identifiable in the archaeological record. Elsewhere, if the equipment was less specialized in shape, it may be harder to identify. The craftsmen of Akrotiri clearly did not share the special interest in metallurgy or the production of aromatics. Whether they specialized in industrial activities of their own can only be determined by detailed analysis of the assemblages and their contexts.

Addendum

Aspects of the question raised in my last paragraph, whether the people of Akrotiri specialized in industrial activities, were addressed by several participants at the Congress, who supplied much valuable new evidence. Their answers vary, depending on the craft under discussion. Production of the chipped stone tools, for example, seems not to have involved specialists, and the range of tools themselves does not imply that they were designed for specialized industries (Moundrea-Agrafioti 1990). But a case is made for the existence of specialist weavers, and of a permanent textile workshop in the West House (Tzachili 1990), and another in Complex A (Michailidou 1990). Of particular interest are the association of 450 loom weights with a set of lead weights in Room 3 of the West House and a similar combination in Sector A, and the suggested relationship between lead weights and household industries (Michailidou 1990).

Comparisons with Kea are instructive. Davis (1984, 161-163) records fewer than 200 loom weights in the LC I town, about half of which came from the areas of Houses L and M: enough, he believes, to suggest craft specialization. But neither these houses, nor House A (Table II), nor any other building at Ayia Irini in LC I - II has produced the huge numbers of loom weights found in the West House.

As at Akrotiri, lead weights at Ayia Irini (Petruso, forthcoming) occur in most of the building complexes. But they are usually isolated finds (never more than three found together), and the very heavy weights found at Akrotiri are lacking on Kea. These circumstances reinforce Michailidou's suggestions that weighting was a commonplace activity related to home industries, and that the heaviest lead disks were used for the textile industry at Akrotiri. She points to the evidence of the Linear B tablets that metals and aromatic substances (among other commodities) were evaluated by weight. Clearly, lead weights must also be taken into account in future studies of the industrial assemblages at Ayia Irini.

-----------------------------------------------

----------------------------------------------