Mycenaean feasting on Tsoungiza at Ancient Nemea

Mycenaean feasting on Tsoungiza at Ancient Nemea

Mary K. Dabney

ABSTRACT

This paper presents a ceremonial feasting deposit from Late Helladic IIIA2 Tsoungiza. The dominance of head and foot bones from at least six cattle suggests on-site butchery, with the possibility that the meat was distributed for consumption elsewhere. The pottery fulfills most of the criteria proposed here for recognizing feasting activities in ceramic assemblages. A ceramic female figure, similar to those from sanctuaries at Phylakopi and Mycenae, ties the feasting to religious rituals. It is suggested that regional feasts contributed to maintaining political and economic alliances within the area around Mycenae.

**********

Ancient Nemea is located in the northeast Peloponnese at the head of the Nemea Valley, outside the Argolid but within two hours walking distance of Mycenae. (1) Excavation of the Bronze Age settlement on the hill of Tsoungiza at Ancient Nemea took place from 1984 to 1986 as part of the Nemea Valley Archaeological Project (NVAP). One of the project goals is to study how the settlement was incorporated into larger social systems during different periods of occupation. After a period of abandonment in the Middle Bronze Age, the site was occupied continuously from the late Middle Helladic (MH) through the Late Helladic (LH) period. The number of structures found in the excavated area of the settlement increased from six during the Early Mycenaean (late MH through LH II) era to 10 in the Late Mycenaean (LH III) era (Fig. 1). Evidence from NVAP’s surface survey and from excavations conducted by the University of California at Berkeley in deep trenches underneath the Classical Sanctuary of Zeus in the river valley below Tsoungiza suggests an even greater increase in Late Mycenaean settlement size. (2) Large refuse dumps of Late Mycenaean remains were found throughout the excavated area. One early LH IIIB1 refuse dump contained an estimated 1,400-2,100 vessels representing nearly the full range of ceramic vessel forms known at Mycenae. (3)

[FIGURE 1 OMITTED]

What brought about this change to a larger, denser settlement in the LH III period? The answer may lie in the remains of the earliest of these LH III refuse dumps, in excavation unit (EU) 9, the earliest excavated layers of which contained pottery dating to LH IIIA2 (early). (4)

FAUNAL EVIDENCE

The faunal remains from this deposit are distinctive. Cattle make up half of the identified material, with pig, sheep, and goat accounting for most of the remainder, but there are also a few specimens of dog, ass, and red deer. Half of the material exhibited traces of gnawing, indicating that the bone was accessible to scavengers (probably domestic dogs or pigs) either before or after incorporation in the dump. A quarter of the assemblage (including bones of cattle, pig, goat, dog, and ass) bore traces of burning, however, and butchery marks were observed on ca. 4% of the bones (including those of cattle, pig, and dog), leaving no doubt that much or all of the material was butchered and discarded by humans.

In terms of anatomical representation, the remains of pig and sheep/ goat include most parts of the carcass; those of dog, ass, and deer are too few for analysis, but remains of cattle are heavily biased toward the head and feet (Table 1). This bias is apparent whether bones are quantified in terms of minimum numbers of anatomical units (MinAU) or minimum numbers of individuals (MNI). Anatomical representation may be shaped by a number of factors, including archaeological retrieval and post-depositional attrition, as well as pre-depositional human behavior. Thus, the absence of such small body parts as the phalanges of pig and sheep/goat might plausibly be attributed to retrieval loss, but the “missing” body parts of cattle are not small and the routine use of sieving at Nemea seems to have ensured fairly complete recovery of identifiable fragments of this large taxon. Similarly, although the assemblage has been extensively gnawed, the missing and scarce body parts include some of the most robust (e.g., distal humerus, distal tibia) as well as the most vulnerable (e.g., proximal humerus). (5) Thus there can be little doubt that the biased anatomical representation of cattle is the result of selective human behavior.

Deposits dominated by head and foot bones of cattle are relatively common in Roman and medieval towns in northwest Europe. (6) Such assemblages are usually interpreted as primary butchery waste discarded by specialist butchers. Because animals are butchered in large numbers, different stages of carcass-processing tend to be separated in space and time and, as a result, the waste from different stages tends to be discarded in different contexts. In this case, however, butchery marks and types of fragmentation observed on cattle limb bones suggest that the discarded material does not represent primary butchery waste, but rather carcass parts that had been processed for the extraction of bone marrow. The faunal material thus arguably represents waste from food preparation or consumption, rather than primary butchery.

Segregated discard of particular body parts may also occur for symbolic rather than practical reasons, as is widely illustrated by bone deposits from early historical sanctuary sites in various parts of the Mediterranean. (7) In such sanctuary contexts, the highly symbolic nature of bone deposition is often highlighted by the selection of a particular taxon, age group, body part, or side of body, or by selective treatment (e.g., burning, rapid burial, lack of fragmentation). In the present case, the frequency of gnawing indicates that the assemblage as a whole was not accorded immediate burial, and the incidence of gnawing, burning, and old breaks is broadly similar for cattle, pig, and sheep/goat. On the other hand, because experimental data show that most attrition of faunal remains by gnawing takes place within a few hours of discard, burial may have taken place shortly after exposure. (8)

There is no evidence, therefore, other than the selective deposition of heads and feet, that bones of cattle were treated in a distinctive fashion. The age of the cattle represented is varied, including at least one infant, one juvenile, and two adults. Moreover, despite the preponderance of heads and feet, other body parts of cattle are generally underrepresented rather than absent altogether. Although individual body parts display a bias toward left- or right-sided specimens, no consistent pattern of selection was observed. In spite of the association with a possible ceremonial activity (see below), therefore, no faunal evidence suggests that the selective deposition of cattle heads and feet was an act of great symbolic significance. It might be argued that this deposit represents those parts of the carcass not selected for ceremonial treatment, but again the range of body parts missing or underrepresented is too large to offer active support for such an interpretation.

The scale of carcass-processing represented here is difficult to estimate. Surviving and recovered remains of cattle include parts of at least six individuals, but this figure is probably an underestimate of the actual number discarded. The excavators estimated that less than half of the dump was excavated. Widespread gnawing is likely to have resulted in some losses, and, even within the head and foot categories, the body parts listed in Table 1 account for only a minority of those expected for six cattle. The duration and number of episodes of deposition are also uncertain, although the condition of the associated ceramics suggests, at most, just a few episodes. Even if only one or two cattle were deposited in each episode, however, a significant quantity of meat would have been involved, which, taken in conjunction with the ceramic evidence, suggests consumption on a large scale. In modern Greece, prior to the introduction of electric refrigerators, cattle were rarely consumed in villages and were normally sold for slaughter in the towns. (9) The scale of meat consumption represented by this deposit is thus certainly consistent with a major feast or series of feasts.

An unresolved question, at this stage, is the fate of the parts of cattle not found in this deposit. As has already been noted, the range of parts missing does not favor destruction in a burned sacrifice, (10) nor does ongoing study of the rest of the LH III faunal assemblage from Tsoungiza indicate deposition elsewhere on the site. A third and intriguing possibility is that the feasts taking place at Tsoungiza involved drinking, sacrifice, and slaughter, followed by some consumption on-site (at least of bone marrow), but that most of the meat from the slaughtered animals was distributed to participants from other settlements for eventual consumption in their own communities. This tentative interpretation would imply that feasting at Nemea had political significance on a regional scale, a possibility further discussed below.

CERAMIC EVIDENCE

The identification of ceramic remains resulting from feasting is a difficult task in the absence of a set of vessels or accompanying decorative motifs that are peculiar to that activity. This is true not only for Mycenaean Greece, but for other past cultures as well. (11) Although Stocker and Davis have identified a kind of miniature kylix as an indicator of ritual feasting at Pylos,12 for the most part Mycenaean feasts seem to have employed pottery that was also employed on an everyday basis. Since Mycenaean settlement deposits are in general dominated by precisely the sorts of serving vessels (such as kylikes) that would presumably be used in feasts, it is useful to attempt to develop criteria to identify ceramic remains from feasting. A combination of characteristics, not ceramic evidence alone, will probably be needed to identify such a deposit. In the case of the material from EU 9, for example, it is important to keep in mind the clear evidence for the butchering of large quantities of meat (see above).

An important preliminary question to ask is whether it is reasonable to expect to find large deposits of ceramics from feasts, since the vessels would retain their utility after the meal was completed, and might continue in use afterward. Deposition of intact vessels and even deliberate destruction of complete vessels are known practices associated with funerary meals, (13) but are not well established in other contexts. Although vessels might be deliberately destroyed during the course of a feast or afterward, as argued below, we need not insist on ritual breakage to suspect that a large feast will leave behind clear ceramic evidence. The presence of large numbers of people and the consumption of wine itself virtually guarantees a number of broken vessels. Moreover, if the participants traveled some distance to take part in the feast and were provided pottery by their host, many would probably discard the vessels before returning home.

As a first step in the development of a model, the sorts of vessels likely to be present in a deposit of ceramics resulting from feasting activities should be considered. It appears sufficiently settled that wine and meat were important components in Mycenaean feasting. The association of the kylix with feasting is agreed upon by virtually all writers; we ought, therefore, to expect that kylikes will compose a large percentage of such a deposit. Further, we can expect that jugs or other pouring vessels for the wine would be present. If ceramic vessels were used for cooking, cooking pots should be present as well, especially if meat was prepared in the form of a stew. (14) If pottery was used for serving meat dishes, an obvious possible form is the Furumark shape (FS) 295, a shallow angular bowl (SAB). Although Tournavitou has suggested that the SAB is well designed for serving hot foods, (15) the very large number of such vessels found in room 21, part of the suite of pantries at the palace at Pylos, has been little remarked on. With nearly 1,100 examples from this room alone, it is the second most common shape after the kylix in the pantries and a logical candidate as a serving vessel for meat dishes. (16) Evidence for the function of this vessel as a plate for meat is provided by a recent organic-residue analysis of a SAB from a LH IIIB context at Mycenae that indicated traces of fat. (17)

The ceramics from feasting activities should thus exhibit most of the following criteria:

1. Deviation from the norm in the amount and kind of decoration. If provision of food and drink was the primary purpose served, and the vessels themselves were not valued souvenirs of the feast, elaborate decoration would be unnecessary. A higher-than-usual percentage of unpainted pottery might be found, or painted pots might exhibit a lower-than-usual percentage of patterned examples. The same features could also be observed if the sponsor of the feast wanted to deemphasize social differences and create an atmosphere of equality among the guests.

2. A higher-than-usual percentage of open shapes used to serve food and drink. The kylix should be especially abundant. Based on later practice at Pylos, we might also expect to find substantial numbers of SABs among the open shapes; shallow cups may be present as well, since over 1,000 of them were present in room 21. Dippers and kraters for serving wine are also likely finds.

3. The presence of closed shapes in the form of jugs. Cooking pots of various sorts should be prevalent, especially if meat was prepared in stews.

4. The presence of vessels peculiarly associated with ritual feasting. In the case of Mycenaean Greece, a possible candidate is the miniature kylix with high-swung handles, as argued by Stocker and Davis elsewhere in this volume.

5. A restricted range of vessels. If a deposit is formed primarily from feasting activity, it is likely to exhibit fewer types of vessels than a deposit representing everyday activity, especially in terms of utilitarian vessels employed for processing food or other products.

6. The possible presence of oversized pots for dignitaries or for shared consumption among participants.18 Both kylikes and unusually large SABs were encountered in the pantries at Pylos. The very few large kylikes present in comparison to kylikes of average size might indicate that sharing occurred only within a select group. (19)

We should now reflect briefly on the process of deposition, beginning with a theoretically ideal situation and then working toward the messier reality likely to confront an excavator. Ideally, a fresh pit would be dug to hold all of the debris from a feast and closed immediately thereafter with a layer of sterile soil. The participants would carefully gather all of the discarded bones and uneaten food and deposit them into the pit; all pots, vessels of other materials, and utensils used, whether broken or unbroken, would be cast in as well. Ceramics from such a deposit would be recognizable archaeologically by very high “mendability” into whole pots, a prevalence of shapes associated with feasting, a lack of shapes associated with other activities, complete chronological homogeneity of shape and decoration, and minimal wear from weathering processes.

We have no reason, however, to believe that refuse dumps such as those found at Tsoungiza were used exclusively for refuse from large-scale feasts. A realistic model designed to characterize feasting activity needs to take into account the likelihood that two streams of waste might well be flowing into a dump, one derived from feasting and the other from daily use. One can anticipate that the stream of waste from daily activities will in general be more fragmentary, exhibit greater differences in wear, and contain a broader range of the types of vessels in use at the site, not only those associated with eating and drinking. Even if we can develop criteria to distinguish between ceramics from feasting and daily meals, the latter stream of waste may “dilute” the distinctiveness of the waste resulting from feasts. If the feasts are small and infrequent, one can anticipate that their remains in a dump will probably be archaeologically indistinguishable from those of daily meals. In addition, the possibility of a third stream of fill used to cover decaying bones and meat is discussed below.

Distinguishing depositional features from large feasts can be suggested. The first is a general lack of soil matrix between sherds, since substantial quantities of pottery will be deposited at once. In a “dual-use” refuse pit, then, we can anticipate finding substantial areas of densely packed sherds and bones reflective of individual feasts, separated by layers associated with everyday activity, with fewer sherds and more soil matrix. A second feature that can be expected is less general weathering of surfaces, since fewer sherds will be exposed to the effects of weathering if they are dumped in a heap than if they are deposited in other ways.

Obvious disparities and potential difficulties with the model should be considered, along with some reasonable modifications. First, the LH IIIA2 (early) deposit in EU 9 is not composed primarily of whole vessels, although it includes some vessels that are entirely or nearly complete, and it contains substantially more restorable pottery than, for example, a LH IIIB1 deposit excavated in EU 2. Second, the deposit is not chronologically homogeneous: nearly 5% of the sherds belong to earlier periods, a finding that does not appear to be the result of the excavators’ cutting into earlier strata. By contrast, fewer than 1% of the sherds in the LH IIIB1 refuse pit in EU 2 belonged to earlier periods.

A rough index of mendable pottery in deposits with broken vessels is the percentage of sherds remaining after mending is complete. After sherds from earlier periods were excluded, the EU 9 deposit, after mending, contained roughly 79% of the initial number of sherds excavated; the EU 2 deposit had 86%. Not only were more joins found in the EU 9 deposit than in the EU 2 deposit, but numerous additional joins could have been made between body sherds in the former deposit had unlimited amounts of time been available. Moreover, only a portion of the EU 9 deposit was recovered, in contrast to the EU 2 deposit, which was completely excavated (the area to the east of EU 9 was private property not purchased for the project). This point has a bearing on any argument about the numbers of relatively whole pots and the amount of faunal material present. The deposit ranged in depth from 0.57 to 1.04 m in a 4 x 4 [m.sup.2] area; the total volume of the deposit was about 13.5 [m.sup.3]. The percentage of the total deposit excavated is not calculable, but the increasing numbers of sherds found as one moves eastward in successive 1-m-wide strips suggest that a very substantial portion of the deposit lies to the east of the excavated area. The basal levels of EU 9 (SU 1584, 1558, and 1559), which form an area of roughly comparable thickness across the trench, show this clearly: a 4-m-long strip along the E729 grid line running from N459 to N463 contained 699 sherds weighing 8.235 kg; the strip along the E730 grid line, 1,112 sherds weighing 13.616 kg; the strip along E731, 1,215 sherds weighing 14.670 kg; and the strip along E732, 2,297 sherds weighing 22.640 kg. Were the area to the east to be excavated, many more pieces could presumably be joined to those already excavated, since it was not unusual to find pieces from the same pot in the four adjacent units.

That the deposit accumulated in a relatively short period of time and was not built up gradually as the result of daily disposal of rubbish is indicated by several lines of evidence. The excavators noted a number of very dense and thick “lenses” of sherds during the course of excavations; one such lens appears in a section drawing. Little matrix was observed between sherds in these lenses, consistent with dumping large quantities of sherds at once. The deposit in general and the more completely preserved vessels in particular exhibit considerably less surface weathering than the pottery in the EU 2 deposit dating to LH IIIB1, which is consistent with bulk disposal and fairly rapid burial. The EU 9 deposit was excavated in 59 units. Only 15% (9 out of 59) of the units contained heavily worn potsherds and 36% showed light wear; 63% of the units contained large sherds. Of the 37% containing only small to medium-sized sherds, many were from the basal level of the deposit where the sherds were in good condition, but broken into small pieces–almost as though an area had been leveled off and then walked on, breaking the sherds lying on its surface into small fragments. Moreover, the period to which the pottery belongs is a short one, accorded no more than 20-30 years in most absolute chronologies. (20) The deposit was stratified beneath a thin layer of LH IIIA2 (late), so it could not have accumulated over a very long period of time.

Roughly 5% of the sherds belong to earlier periods–an amount somewhat unexpected in a primary disposal context but one consistent with deposition associated with feasting. For symbolic as well as practical reasons, debris from important ceremonies is often destroyed, displayed, or buried rather than merely being left discarded on the ground. A striking Mycenaean example is the series of bone groups from burned sacrifices placed around the Palace of Nestor at Pylos, (21) and such depositional practices were commonplace in later Greek sanctuaries. In this case, refuse from feasting may have been disposed of in the EU 9 pit and then covered with a layer of soil. The great majority of the earlier ceramic material from the pit dates to EH II and III; areas with considerable remains from these periods are immediately at hand to the west and north as a source of fill dirt. A sequence of feasts could thus leave evidence in the form of thick lenses of sherds and bones separated by areas of fill with higher concentrations of earlier sherds, a scenario compatible with the situation in the EU 9 deposit.

In order to assess how well the EU 9 deposit fits the above model for identifying feasting, it is necessary to compare it to other deposits (Table 2). Unfortunately, no settlement deposits from LH IIIA2 (early) have been characterized statistically in a manner comparable to this deposit; few closed settlement deposits of this period have in fact been identified. In order to have some basis of comparison, however, Table 2 presents statistics for deposits from Asine, Tsoungiza, Mycenae, and Korakou. Usage of these statistics is fraught with difficulties, since at least some sherds (particularly unpainted and solidly painted sherds) from most of the Mycenae deposits were discarded before being analyzed. In addition, because Mycenae was a preeminent site, it is more likely to have remains of feasting activities than a smaller site, and thus using deposits from Mycenae to establish a norm for settlements of all sizes is questionable; deposits from specialized storage areas such as the Petsas House at Mycenae and the so-called Potter’s Shop at Zygouries have, for this reason, not been employed. Pottery from Nichoria and Asine is referred to in the following discussion, but it has not generally been quantified in a manner similar to the deposits from Tsoungiza and Mycenae. (22) Although plentiful material is available from Tiryns for LH IIIA2 (late) and LH IIIB1, it too has not been quantified in a readily comparable manner. (23)

As would be expected in any settlement deposit, open shapes are more common in the EU 9 deposit than closed ones. Of the painted pottery, there is a higher percentage of open shapes in the EU 9 deposit than in EU 2 (LH IIIB1): 84.5% vs. 75.5%. In the larger and probably more representative deposits from Mycenae (the Atreus Bothros, Citadel House room 3, South House room 22, and Causeway deposits), the percentage of open painted vessels is typically around 60%. At Korakou, which as a small, nonpalatial site probably offers the best comparison to Tsoungiza, about 80% of the painted sherds are from open shapes.

The unusual nature of the EU 9 deposit is indicated by the low frequency of patterned sherds and sherds with linear decoration, and the high incidence of solidly painted sherds (criterion 1, above). Fewer than 10% of the sherds from the EU 9 deposit bear a pattern, and only 21% have linear decoration. In no other published Mycenaean deposit of LH IIIA-B, with the possible exception of the LH IIB-IIIA1 strata from Asine, does the percentage of painted, patterned sherds fall below 20%. (24) For those deposits where most or all of the painted pottery was saved (Tsoungiza EU 2, Korakou, and Mycenae rooms 3 and 22, and the Causeway deposit), the percentage of patterned pottery typically amounts to 20-25%, and sherds with linear decoration typically comprise half or more of the total painted sherds. Because of the large number of kylikes, roughly 67% of the painted sherds in the EU 9 deposit are solidly painted. It can be argued that, in the absence of settlement groups of comparable date, these percentages relate more to chronology than function, but until additional LH IIIA2 (early) settlements are found, one must work with available evidence. (25)

As the labor-intensive practice of burnishing the surfaces of drinking cups declined during LH IIIA, the application of hard red paint to the surface might have served a similar aesthetic purpose by giving them a shinier, more metallic appearance, and would have lessened the porosity of body walls. One might speculate that the needs of public feasts such as those postulated here demanded greater quantities and more efficient production of drinking vessels, especially if the vessels were broken soon after being used.

The kylix is by far the most common painted shape in EU 9 (criterion 2). Although it is sometimes difficult to distinguish between rims of solidly painted kylikes and those of stemmed bowls, painted kylikes certainly comprise no fewer than 50%, and possibly closer to 55%, of all painted sherds. By contrast, only 7.5% of the LH IIIB1 assemblage in EU 2 is made up of painted kylikes. The bulk of the painted kylikes in EU 9 belong to FS 264, the rounded kylix typical of LH IIIA2, and are solidly painted (Fig. 2); a substantial number of patterned kylikes are present as well, although there are at least 10 solidly painted kylikes for every patterned one. The kylix is also the form most prevalent among the unpainted sherds: 72% of diagnostic fineware sherds represent various kylix forms, mostly FS 267, the angular kylix, and the rounded kylix in both its deep (FS 264) and shallow (FS 266) variants (Fig. 3). When all unpainted features (e.g., rims, handles, bases) are taken into account, kylikes amount to over 55% of the diagnostic sherds. The percentage of unpainted vessels is less striking when compared to deposits with available statistics for unpainted ware: in the EU 2 deposit, 69% of the fine features belonged to kylix forms and 49.5% of diagnostic sherds; at Mycenae, in the Citadel House, room 3, Wardle notes that unpainted kylikes account for about half the total number of vessels represented in the unpainted pottery; in the South House, room 22, unpainted kylikes account for 62% of all unpainted features. (26)

[FIGURES 2-3 OMITTED]

Another popular open serving vessel present in this deposit is the stemmed bowl, FS 304 (Fig. 4); other than several examples decorated with stipple pattern, nearly all of these are solidly painted. They comprise a minimum of 4.6% of the painted sherds, but because of the difficulty in distinguishing between some rims and body sherds of these vessels and those of solidly painted kylikes, the true proportion is probably closer to 10%. The only other open shape among the painted sherds that accounts for more than 2% of the total is a peculiar deep cup (similar to FS 214) with a solidly painted interior.

[FIGURE 4 OMITTED]

Of the unpainted pottery, the SAB (FS 295, Fig. 5) accounts for 2.7% of the fine features and 1.1% of all unpainted features; the same percentages apply to the unpainted shallow cup. Dippers make up 2.3% and 0.9% of the unpainted fine features and total unpainted features, respectively. Apart from some conical cups (FS 204), which may have served as lamps, only a small number of open vessels are present. These percentages are comparable to those of the EU 2 deposit. The raw percentages conceal the fact, however, that the remains of the SABs could often be mended into whole profiles or at least very substantial portions of the vessels.

[FIGURE 5 OMITTED]

Among closed vessels (criterion 3), large and medium-sized jugs and hydrias are the most common shapes represented in painted and unpainted assemblages in EU 9. Sherds from painted jugs and hydrias make up 4% of all painted sherds (cf. the EU 2 deposit with 9%). At Mycenae, in the Citadel House room 3 deposit, painted jugs comprise 4.9% of the painted sherds; in the South House room 22 deposit, 3.7%. At Korakou, jugs make up roughly 2% of the total painted sherds in the East Alley deposit. Among the unpainted pottery in EU 9, sherds from the jug/hydria make up 15% of all diagnostic sherds, in comparison to 10.5% of all features in the EU 2 deposit. In the South House room 22 deposit at Mycenae, the jug/hydria makes up only 7.5% of the unpainted features; although exact figures are not provided, the percentage in the Citadel House room 3 deposit appears to be even lower. (27) The unpainted jug/hydria may thus be more common in the Tsoungiza EU 9 deposit than is usual.

Cooking vessels of various kinds are present in abundance, including tripod cooking pots, at least three kinds of cooking jar, and two possible cooking lids. In both the EU 2 and EU 9 deposits, cooking vessels make up roughly half of the medium-coarse pottery and a tenth of all unpainted sherds. The latter figure is somewhat more than the 6.6% seen in the South House room 22 deposit at Mycenae.

In terms of the presence of special ritual vessels (criterion 4), this deposit contains a single example of the miniature kylix with high handles (FS 272; Fig. 6) that Stocker and Davis (this volume) connect to ritual feasting at Pylos. The rim diameter is only ca. 8 cm, about half the normal rim diameter of a kylix. This vessel is the only identified example of a miniature kylix at Tsoungiza in LH III levels, and its appearance may represent a link to sacrificial practices and feasting.

[FIGURE 6 OMITTED]

The EU 9 deposit includes relatively few medium-coarse and coarse utilitarian vessels (criterion 5). For example, nearly a fifth of the coarse diagnostic sherds from the EU 2 deposit come from vats, tubs, and large basins; because the sherds from these sorts of vessels in EU 9 were in general much smaller and harder to classify than those from the EU 2 deposit, exact percentages are impossible to determine, but comparable vessels appear to account for only 10% of the EU 9 deposit.

Finally, oversized versions of some pots are present (criterion 6). A small number of kylix sherds have rim diameters of more than 20 cm. The FS 295 SABs present in this deposit are on the whole larger than usual. In the EU 2 deposit dating to LH IIIB1, almost all of the SABs have rim diameters that fall between 13 and 16 cm; most of the examples in the EU 9 deposit range from 15 to 18 cm, but a number of examples are considerably larger, with diameters up to 23 cm. The larger specimens are not simply a reflection of their period since the average rim diameter, even in LH IIIA2, ranges from around 15 to 18 cm.

In conclusion, although the partial excavation of the EU 9 deposit and the difficulties in finding comparable data must be considered, the pottery in the EU 9 deposit nevertheless fulfills most of the criteria proposed for recognizing feasting activities in ceramic assemblages.

FEMALE FIGURE AND OTHER EVIDENCE

The presence of a fragmentary ceramic female figure (Figs. 7; 8:a) makes the Tsoungiza deposit stand out as unique. The figure belongs to Elizabeth French’s type A, best known from the cult centers at Mycenae and Phylakopi. (28) Only the lower two-thirds was recovered, with attachment scars for the arms on its sides; its restored height is 45 cm. The date of the pottery with which it was deposited confirms a date of LH IIIA for the first appearance of these types of figures, surmised by French from the decoration on figures found in later contexts. (29) A comparable figure, smaller but approximately contemporary, was discovered at Pylos. (30)

[FIGURE 7 OMITTED]

At Tsoungiza, no associated architectural context has been identified for the figure. The only excavated structure at the site with LH IIIA2 destruction debris is later (LH IIIA2 [late]) than the EU 9 deposit. This finding is not surprising, considering that the cult centers where similar figures occur were not built until LH IIIA2 and later, when the palatial centers were already established. (31) Although it is possible that the figure was originally housed in a structure that was either not preserved or outside the area excavated, (32) there is no evidence for a built cult center at Tsoungiza.

The EU 9 deposit also contained other types of figurines (Fig. 8), including two unusual “breadmaker” figurines, three female figurines, and one bovine figurine. Also found were a ceramic bead, fragments of seven ground stone tools (millstones, handstones, and a whetstone), and one chert and three obsidian retouched chipped stone tools. Plant remains from the deposit are typical of most deposits at Tsoungiza and are not significant because only a few species were found due to poor preservation. (33)

[FIGURE 8 OMITTED]

CONCLUSIONS

How are we to understand the social significance of these archaeological remains? The character of the pottery and faunal remains suggests that most of the refuse was deposited shortly after use in a single event or a series of similar specialized events. The predominance of plain vessels used for serving food and drink combined with faunal remains from carcass-processing and the presence of religious display items indicates that this event was a feast. (34) The quantity of cattle consumed suggests that there were many participants in the feast, while the consumption of large animals such as adult cattle may indicate that much of the meat (and perhaps drink) for the feast was provided by a single sponsor or, at most, a very restricted number of sponsors. (35) Thus, there are grounds for interpreting the feast(s) at Ancient Nemea in terms of conspicuous generosity by the high-ranking sponsor(s), rather than collectively organized commensality. The large quantity of meat consumed, the small size of the settlement, and the “missing” parts of the cattle also highlight the possibility of a regional feast, involving participants from a number of different settlements. At regional feasts, the elite guests typically take away large portions of food to redistribute in their own villages. (36) The presence of the female figure and other distinctive figurines suggests that this feast was associated with a religious activity. The disposal of the figure along with the remains of feasting may point to a different ceremonial role for this figure than for later figures found in cult centers elsewhere.

The ceremonial feasting deposit described above provides information about the relationship of Ancient Nemea to external social and economic systems. This deposit marks a turning point in the settlement’s history, coinciding with the beginning of a period of extensive growth. The person or people who provided the resources for this ceremonial feast created an obligation of reciprocity in those who partook of the feast. The provider(s) not only provided the food and drink but also the containers in which the food and drink were served. This required also acquiring and transporting the vessels from the point of production. Kim Shelton has suggested that the Petsas House at Mycenae, which contained a figure and pottery similar to the material in the Tsoungiza deposit, was a warehouse from which such ceramics were distributed. (37) The feast may have been conducted in such a way that the containers as well as their contents were removed from circulation in the community by their disposal. The disposal of the feasting vessels prevented the participants from reciprocating in kind. The provider(s) would thereby have created a situation entailing reciprocity in other spheres such as consolidation of obligations as trading partners or political allegiance. (38)

By inviting people to partake of the feast, the provider was able to bring affiliated people in the region into a closer social relationship. In studying the use of feasts to mobilize collective labor, Michael Dietler and Ingrid Herbich found that feasts act as a means of converting the raw materials of the feast, agrarian wealth, into social and economic prestige. (39) Providing a feast is a powerful means of expanding political power, one not lost on modern-day politicians and lobbyists. To this one must add the provider’s control over a potentially even more significant aspect of social relations, the symbolic belief system as represented by the female figure. The provider’s act of introducing and then disposing of the female figure heightens the meaning of the event represented by this deposit. In the context of relations between Ancient Nemea and the closest larger center, Mycenae, the feasting deposit at Tsoungiza might be evidence of the forging of closer social and economic relations that contributed to the growth of the relative importance of both sites within the region during the Mycenaean era.

The settlement at Tsoungiza was only one of a number of settlements that lay on the northern periphery of Mycenae. Settlements to the east of Tsoungiza at Kleonai and Zygouries, and to the west at Phlious, Ayia Irini, and the settlement associated with the Aidonia cemetery might also have been drawn into Mycenae’s social and economic sphere of influence. As Pia de Fidio argues,40 palace and villages were engaged in a dynamic relationship, not merely one in which the center commanded and the periphery obeyed; within villages themselves, hierarchies emerged. Competition among the elites at these sites could have driven both the practice of feasting and the establishment of public ritual. The rotation of regional feasts involving elites from a number of different centers would have contributed to maintaining regional political and economic alliances.

TABLE 1. ANATOMICAL REPRESENTATION OF

FAUNAL REMAINS IN THE EU 9 DEPOSIT

Cattle Pig Sheep/Goat

Anatomical Part MinAU MNI MinAU MNI MinAU MNI

Head

Horn/Antler 2 1 — — 2 1

Mandible 3 3 3 3 5 4

Upper Forelimb

Scapula 1 1 1 1 0 0

Humerus proximal 1 1 3 3 0 0

Humerus distal 0 0 2 2 3 2

Radius proximal 1 1 0 0 1 1

Ulna proximal 0 0 2 2 3 2

Radius distal 0 0 1 1 1 1

Upper Hindlimb

Pelvis 1 1 1 1 1 1

Femur proximal 0 0 0 0 1 1

Femur distal 1 1 1 1 1 1

Tibia proximal 3 2 1 1 3 3

Tibia distal 2 1 1 1 4 2

Astragalus 1 1 0 0 1 1

Calcaneum 1 1 0 0 0 0

Foot

Metacarpal proximal 7 5 0 0 0 0

Metacarpal distal 7 4 0 0 0 0

Metatarsal proximal 7 6 0 0 2 2

Metatarsal distal 7 4 0 0 2 2

Phalanx 1 5 2 0 0 0 0

Phalanx 2 1 1 0 0 0 0

Phalanx 3 3 2 0 0 0 0

Total 54 6 16 4 30 5

MinAU = minimum number of anatomical units; MNI = minimum number of

individuals.

TABLE 2. CERAMIC REMAINS FROM TSOUNGIZA AND RELATED SITES

Patterned Linear

Site/Sample (%) (%)

Asine, room D, str. 2 Total 10.4 Total 0.5

N (painted) 884 Open 7.7 Open 0.5

N (unpainted) 3,271 Closed 2.7 Closed 0.0

Tsoungiza, EU 9 Total 9.4 Total 21.2

N (painted) 2,160 Open 6.2 Open 12.4

N (unpainted) 8,442 Closed 3.2 Closed 8.8

Tsoungiza, EU 2 Total 24.0 Total 48.3

N (painted) 2,626 Open 18.8 Open 35.0

N (unpainted) 15,123 Closed 5.2 Closed 13.3

Mycenae, Atreus Total 78.1 Total 21.9

Bothros Open 48.0 Open 11.0

N (painted) 2,094 * Closed 30.1 Closed 10.9

Mycenae, terrace on Total 68.4 Total 31.6

the Atreus ridge Open 65.0 Open 31.6

N (painted) 674 Closed 3.4 Closed 0

Mycenae, terrace below Total 68.0 Total 32.0

the House of Shields Open 54.4 Open 32.0

N (painted) 206 Closed 13.6 Closed 0

Mycenae, dromos of Total 62.3 Total 31.1

tomb 505 Open 57.4 Open 24.6

N (painted) 61 Closed 4.9 Closed 6.5

Mycenae, Prehistoric Total 82.2 Total 16.2

Cemetery, central Open 53.7 Open 7.3

N (painted) 1,871 Closed 28.5 Closed 8.9

Mycenae, room 3 Total 20.5 Total 69.7

N (painted) 878 Open 17.3 Open 23.1

N (unpainted) ca. 7,000 Closed 3.2 Closed 46.6

Mycenae, room 22 Total 25.6 Total 53.5

N (painted) 1,240 Open 20.3 Open 23.5

N (unpainted) 6,830 Closed 5.3 Closed 30.0

Mycenae, Causeway Total 21.1 Total 74.1

N (painted) 825 Open 17.5 Open 33.0

N (unpainted) 3,604 Closed 3.6 Closed 41.1

Korakou, Total 21.8 Total 46.9

East Alley I-IV Open 20.2 Open 34.7

N (painted) 729 ** Closed 1.6 Closed 12.2

Solidly Unidentifiable

Painted

Site/Sample (%) (%)

Asine, room D, str. 2 Total >80

N (painted) 884 N/A

N (unpainted) 3,271

Tsoungiza, EU 9 Total 66.8 Total 2.6

N (painted) 2,160 Open 64.2 Open 1.7

N (unpainted) 8,442 Closed 2.6 Closed 0.9

Tsoungiza, EU 2 Total 18.5 Total 9.2

N (painted) 2,626 Open 14.5 Open 7.2

N (unpainted) 15,123 Closed 4.0 Closed 2.0

Mycenae, Atreus

Bothros N/A N/A

N (painted) 2,094 *

Mycenae, terrace on

the Atreus ridge N/A N/A

N (painted) 674

Mycenae, terrace below

the House of Shields N/A N/A

N (painted) 206

Mycenae, dromos of Total 6.6

tomb 505 Open 6.6 N/A

N (painted) 61 Closed 0

Mycenae, Prehistoric Total 1.6

Cemetery, central Open 1.5 N/A

N (painted) 1,871 Closed 0.1

Mycenae, room 3 Total 9.8

N (painted) 878 Open 8.5 N/A

N (unpainted) ca. 7,000 Closed 1.3

Mycenae, room 22 Total 20.9

N (painted) 1,240 Open 13.5 N/A

N (unpainted) 6,830 Closed 7.4

Mycenae, Causeway Total 4.8

N (painted) 825 Open 4.4 N/A

N (unpainted) 3,604 Closed 0.4

Korakou, Total 31.3

East Alley I-IV Open N/A N/A

N (painted) 729 ** Closed N/A

Painted Open

vs. Closed Painted vs.

Site/Sample Shapes % Unpainted (%)

Asine, room D, str. 2

N (painted) 884 N/A Painted 21.3

N (unpainted) 3,271 Unpainted 78.7

Tsoungiza, EU 9

N (painted) 2,160 Open 84.5 Painted 20.4

N (unpainted) 8,442 Closed 15.5 Unpainted 79.6

Tsoungiza, EU 2

N (painted) 2,626 Open 75.5 Painted 14.8

N (unpainted) 15,123 Closed 24.5 Unpainted 85.2

Mycenae, Atreus

Bothros Open 59 N/A

N (painted) 2,094 * Closed 41

Mycenae, terrace on

the Atreus ridge Open 96.6 N/A

N (painted) 674 Closed 3.4

Mycenae, terrace below

the House of Shields Open 86.4 N/A

N (painted) 206 Closed 13.6

Mycenae, dromos of

tomb 505 Open 88.6 N/A

N (painted) 61 Closed 11.4

Mycenae, Prehistoric

Cemetery, central Open 62.5 N/A

N (painted) 1,871 Closed 37.5

Mycenae, room 3

N (painted) 878 Open 48.9 Painted 11.1

N (unpainted) ca. 7,000 Closed 51.1 Unpainted 88.9

Mycenae, room 22

N (painted) 1,240 Open 57.3 Painted 15.4

N (unpainted) 6,830 Closed 42.7 Unpainted 84.6

Mycenae, Causeway

N (painted) 825 Open 54.9 Painted 18.6

N (unpainted) 3,604 Closed 45.1 Unpainted 81.4

Korakou,

East Alley I-IV Open 79.8 N/A

N (painted) 729 ** Closed 20.2

Percentages are of the total number of painted sherds; N/A = not

available or not able to be calculated.

For Tsoungiza, figures reflect numbers after mending and extraction of

earlier sherds. In conformity with the Tsoungiza calculations,

characteristic handles from Mycenae are counted with the patterned.

The “unidentifiable” category includes sherds that had traces of

paint, but could not be definitely assigned to the three main

categories of patterned, linear, or solidly painted. Unpatterned rims

and bases assigned to particular shapes at Mycenae (in published

tables) are assumed to have been linear unless explicitly identified

as solidly painted.

Percentages were calculated using the following sources: Asine, room D,

stratum 2 (LH IIB-IIIA1): Frizell 1980, pp. 34-41 (all sherds kept);

Tsoungiza, EU 2 (LH IIIB1): Thomas 1992, pp. 520-525, figs. 2:1-2:7

(all sherds kept); Mycenae, Atreus Bothros (LH IIIA1): French 1964,

pp. 260-261 (unpainted and most solidly painted discarded); Mycenae,

terrace on the Atreus ridge (LH IIIA2 [late]): French 1965, p. 200

(only the “best and most typical pieces kept”); Mycenae, terrace below

the House of Shields (LH IIIA2 [late]): French 1965, p. 201 (much

discarded); Mycenae, dromos of tomb 505 (LH IIIA2 [late]): French

1965, p. 202 (at least a third of the originally catalogued sherds lost

by the time of analysis; most solidly painted and unpainted sherds

appear to have been discarded); Mycenae, Prehistoric Cemetery, central

(LH IIIB1): French 1966, p. 235 (all unpainted and most linear sherds

discarded); Mycenae, Citadel House, room 3 (LH IIIB1): Wardle 1969,

p. 279 (painted sherds kept; unpainted sherds given a preliminary

analysis, but many discarded before final study); Mycenae, South House,

room 22 (LH IIIB1): Mountjoy 1976, p. 110 (all sherds kept); Mycenae,

Causeway (LH IIIB2): Wardle 1973, p. 320 (painted sherds kept;

unpainted sherds given a preliminary analysis, but many discarded

before final study); Korakou, East Alley I-IV (LH IIIB1): Rutter 1974,

pp. 102-103, fig. 27 (coarse body sherds were discarded; other classes

may have been partially discarded).

* Apparently the great majority of the solidly painted sherds were

jettisoned before French’s study, and so the few remaining have not

been taken into account. Those would bring the total number of painted

sherds to 2,133.

** In the Korakou East Alley deposit, the number of patterned and

linear sherds assigned to LH IIIB is 501; solidly painted sherds

present in the preserved collection number 228. Because the solidly

painted sherds were not broken down into open or closed shapes, the

percentage of open vs. closed shapes is based only on the patterned

and linear sherds.

(1.) Wright et al. 1990, p. 581, fig. 1. The Nemea Valley Archaeological Project was sponsored by Bryn Mawr College under the auspices of the American School of Classical Studies at Athens and directed by James C. Wright. Funding was provided by the National Endowment for the Humanities, the National Geographic Society, the Institute for Aegean Prehistory, and private donors. The final publication of the Late Helladic III settlement by Mary K. Dabney, with studies of ceramics by Patrick Thomas, faunal remains by Paul Halstead, botanical remains by Julie M. Hansen, ground stone tools by Kathleen Krattenmaker, and chipped stone tools by Anna Karabatsolis, is in preparation.

We wish to thank Brian Hayden, Jeremy B. Rutter, and James C. Wright for their comments and suggestions on this article.

(2.) Cherry, Davis, and Mantzourani 2000, http://classics.uc.edu/NVAP/ MycNVAP.html.

(3.) Thomas 1992, pp. 25-267; Dabney 1997, pp. 469-470.

(4.) The deposit in EU 9 contains stratigraphic units (SU) 1536, 1540, 1554-1559, 1581-1584, 1588, and 1589 (for the location of EU 9, see Fig. 1). Although some overlap with LH IIIA1 is possible, the material most convincingly dates to LH IIIA2 (early).

(5.) Brain 1981, pp. 21-23.

(6.) E.g., Maltby 1985, p. 55.

(7.) See references in Isaakidou et al. 2002.

(8.) Munson and Garniewicz 2003.

(9.) P. Halstead, unpublished interviews in Macedonia, Epirus, the Peloponnese, the Cyclades, and Crete.

(10.) Cf. Isaakidou et al. 2002.

(11.) For example, see the discussion concerning Mayan feasting in LeCount 2001, esp. pp. 946-948; see also Blitz 1993 for a study of Mississippian feasting.

(12.) Stocker and Davis, this volume.

(13.) See Hamilakis 1998, pp. 119-126, for a review of relevant evidence.

(14.) Tzedakis and Martlew (1999, pp. 84, 103, 108, 110) have analyzed the organic residue in vessels from Minoan Crete, showing that stews of vegetables, meat, and olive oil were common; similar evidence specifically relating to LH III Thebes, Mycenae, and Midea is also presented (pp. 115-135).

(15.) Tournavitou 1995, p. 90, fig. 25:4.

(16.) Palace of Nestor I, pp. 129-132.

(17.) Tzedakis and Martlew 1999, p. 133; the nature of the fat could not, however, be determined. Another possible candidate for a serving vessel is the stemmed bowl (FS 304). This is a shape that first appears in a one-handled form in LH IIIA1 and becomes common in a two-handled form in LH IIIA2.

(18.) Blitz (1993) notes in a study of Mississippian feasting that although vessel types from village/nonfeasting contexts were not significantly different from those in presumed feasting contexts associated with the structures on the mounds, pots in use in the latter contexts tended to be larger. Shared consumption would help to reinforce a sense of community.

(19.) For the SABs, see Palace of Nestor I, p. 356; the largest has a rim diameter of 22.5 cm, height of 8 cm, and measured capacity of 1.4 l. Blegen and Rawson (Palace of Nestor I, p. 371) also report only six “very large” (type 29h) and six “giant” (type 29i) kylikes from rooms 19 and 20, in contrast to nearly 3,000 other kylikes from the site.

(20.) Wiener 1998.

(21.) Isaakidou et al. 2002; Stocker and Davis, this volume.

(22.) A statistical breakdown is provided for two admittedly small Nichoria units dating to LH IIIB2: see Shelmerdine 1992, pp. 510-511. Frizell (1980) does quantify LH IIB-IIIA1 strata at Asine, but the frequencies of solidly painted sherds are not explicitly given.

(23.) See Schonfeld 1988.

(24.) By making some assumptions about Frizell’s (1980, pp. 34-50) counting procedures, it appears that 11% of the painted sherds from stratum 2 under room D were patterned and fewer than 1% had linear decoration.

(25.) Although exact numbers are not provided, Martin (1992, p. 488) states that in a pure LH IIIA1 unit at Nichoria, 2% of the sherds were decorated (presumably this figure includes patterned and linear examples), 11% solidly painted, 67% plain (“fine” in terms of terminology employed in characterizing the Tsoungiza deposits), and 20% coarse (“medium-coarse” and “coarse” at Tsoungiza). It is not clear whether these figures are calculated before or after mending. With the data from Tsoungiza aggregated in this fashion, “decorated” sherds would amount to 7.2% of the total deposit, solidly painted to 16.4%, “plain” to 51.4%, and “coarse” to 25%. The prevalence of solidly painted vessels at Nichoria (mostly goblets) may be an indication that this form of decoration is much more common than has been supposed and may continue into LH IIIA2; Shelmerdine (1992, p. 496) notes that solidly painted kylikes and stemmed bowls, taken together, outnumber patterned kylikes in levels she dates to LH IIIA2 (early) by a 3:1 ratio. For Asine, Frizell (1980, pp. 120-121) states only that solidly painted decoration is the most frequent type observed in LH IIB-IIIA1 levels.

(26.) For the Citadel House, see Wardle 1969, p. 280. For the South House, see Mountjoy 1976, p. 111, table 3.

(27.) Wardle 1969, pp. 280-282; Mountjoy 1976, p. 111.

(28.) Moore and Taylour 1999, pp. 46-47; French 1985, pp. 214-216, 221.

(29.) French 1971, pp. 103-105, 109.

(30.) Palace of Nestor III, p. 159, fig. 232:a-c.

(31.) Wright 1996, p. 61.

(32.) Wright 1996, pp. 69-70.

(33.) J. M. Hansen (pers. comm.).

(34.) Cf. Hayden 2001, pp. 40-41.

(35.) These suggestions are based on the assumption that domestic animals were not in communal ownership.

(36.) B. Hayden (pers. comm.).

(37.) Shelton 2002.

(38.) Hayden 2001, pp. 58-59.

(39.) Dietler and Herbich 2001, p. 252.

(40.) See de Fidio 2001, pp. 23-24.

REFERENCES

Blitz, J. 1993. “Big Pots for Big Shots: Feasting and Storage in a Mississippian Community,” AmerAnt 58, pp. 80-96.

Brain, C. K. 1981. The Hunters or the Hunted? An Introduction to African Cave Taphonomy, Chicago.

Cherry, J. F., J. L. Davis, and E. Mantzourani. 2000. “The Nemea Valley Archaeological Project Archaeological Survey: Internet Edition” (http://classics.uc.edu/NVAP/).

Dabney, M. 1997. “Craft Product Consumption as an Economic Indicator of Site Status in Regional Studies,” in TEXNH: Craftsmen, Craftswomen, and Craftsmanship in the Aegean Bronze Age. Proceedings of the 6th International Aegean Conference, Philadelphia (Aegaeum 16), ed. R. Laffineur and P. P. Betancourt, Liege, pp. 467-471.

de Fidio, P. 2001. “Centralization and Its Limits in the Mycenaean Palatial System,” in Economy and Politics in the Mycenaean Palace States (Cambridge Philological Society, Suppl. 27), ed. S. Voutsaki and J. Killen, Cambridge, pp. 15-24.

Dietler, M., and I. Herbich. 2001. “Feasts and Labor Mobilization: Dissecting a Fundamental Economic Practice,” in Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, ed. M. Dietler and B. Hayden, Washington, D.C., pp. 240-264.

French, E. 1964. “Late Helladic IIIA 1 Pottery from Mycenae,” BSA 59, pp. 241-261.

–. 1965. “Late Helladic IIIA 2 Pottery from Mycenae,” BSA 60, pp. 159-202.

–. 1966. “A Group of Late Helladic IIIB 1 Pottery from Mycenae,” BSA 61, pp. 216-238.

–. 1971. “The Development of Mycenaean Terracotta Figurines,” BSA 66, pp. 101-187.

–. 1985. “The Figures and Figurines,” in C. Renfrew, The Archaeology of Cult: The Sanctuary at Phylakopi (BSA Suppl. 18), London, pp. 209-280.

Frizell, B. 1980. An Early Mycenaean Settlement at Asine: The Late Helladic IIB-III A:1 Pottery, Goteborg.

Hamilakis, Y. 1998. “Eating the Dead: Mortuary Feasting and the Politics of Memory in the Aegean Bronze Age Societies,” in Cemetery and Society in the Aegean Bronze Age (Sheffield Studies in Aegean Archaeology 1), ed. K. Branigan, Sheffield, pp. 115-132.

Hayden, B. 2001. “Fabulous Feasts: A Prolegomenon to the Importance of Feasting,” in Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, ed. M. Dietler and B. Hayden, Washington, D.C., pp. 23-64.

Isaakidou, V., P. Halstead, J. Davis, and S. Stocker. 2002. “Burnt Animal Sacrifice in Late Bronze Age Greece: New Evidence from the Mycenaean ‘Palace of Nestor,’ Pylos,” Antiquity 76, pp. 86-92.

LeCount, L. 2001. “Like Water for Chocolate: Feasting and Political Ritual among the Late Classic Maya at Xunantunich, Belize,” American Anthropologist 103, pp. 935-953.

Maltby, M. 1985. “Patterns in Faunal Assemblage Variability,” in Beyond Domestication in Prehistoric Europe: Investigations in Subsistence Archaeology and Social Complexity, ed. G. Barker and C. Gamble, New York, pp. 33-74.

Martin, S. L. 1992. “Mycenaean Pottery from the Settlement, Part II: The Late Helladic IIIA1 Pottery,” in Excavations at Nichoria in Southwest Greece II: The Bronze Age Occupation, ed. W. A. McDonald and N. C. Wilkie, Minneapolis, pp. 488-494.

Moore, A. D., and W. D. Taylour. 1999. Well Built Mycenae 10: The Temple Complex, Oxford.

Mountjoy, P. 1976. “Late Helladic IIIB 1 Pottery Dating the Construction of the South House at Mycenae,” BSA 71, pp. 77-111.

Munson, P., and R. Garniewicz. 2003. “Age-Mediated Survivorship of Ungulate Mandibles and Teeth in Canid-Ravaged Faunal Assemblages,” JAS 30, pp. 405-416.

Palace of Nestor I = C. W. Blegen and M. Rawson, The Palace of Nestor at Pylos in Western Messenia I: The Buildings and Their Contents, Princeton 1966.

Palace of Nestor III = C. W. Blegen, M. Rawson, W. Taylour, and W. P. Donovan, The Palace of Nestor at Pylos in Western Messenia III: Acropolis and Lower Town: Tholoi, Grave Circle, and Chamber Tombs, Discoveries outside the Citadel, Princeton 1973.

Rutter, J. B. 1974. “The Late Helladic IIIB and IIIC Periods at Korakou and Gonia in the Corinthia” (diss. Univ. of Pennsylvania).

Schonfeld, G. 1988. “Bericht zur bemalten mykenischen Keramik: Ausgrabungen in Tiryns 1982/83,” AA 1988, pp. 153-211.

Shelmerdine, C. W. 1992. “Mycenaean Pottery from the Settlement, Part III: Late Helladic IIIA2-IIIB2 Pottery,” in Excavations at Nichoria in Southwest Greece II: The Bronze Age Occupation, ed. W. A. McDonald and N. C. Wilkie, Minneapolis, pp. 495-517.

Shelton, K. 2002. “The Reopening of Petsas House: The Study of a Ceramic Warehouse and the New Excavations at Mycenae,” AJA 106, p. 259 (abstract).

Thomas, P. M. 1992. “LH IIIB:1 Pottery from Tsoungiza and Zygouries” (diss. Univ. of North Carolina, Chapel Hill).

Tournavitou, I. 1995. The “Ivory Houses” at Mycenae (BSA Suppl. 24), London.

Tzedakis, Y., and H. Martlew, eds. 1999. Minoans and Mycenaeans: Flavours of Their Time, Athens.

Wardle, K. A. 1969. “A Group of Late Helladic IIIB 1 Pottery from within the Citadel at Mycenae,” BSA 64, pp. 261-298.

–. 1973. “A Group of Late Helladic IIIB 2 Pottery from within the Citadel at Mycenae: The Causeway Deposit,” BSA 68, pp. 297-348.

Wiener, M. H. 1998. “The Absolute Chronology of Late Helladic IIIA2,” in Sardinian and Aegean Chronology: Towards the Resolution of Relative and Absolute Dating in the Mediterranean. Proceedings of the International Colloquium “Sardinian Stratigraphy and Mediterranean Chronology,” Tufts University (Studies in Sardinian Archaeology 5), ed. M. S. Balmuth and R. H. Tykot, Oxford, pp. 309-319.

Wright, J. C. 1996. “The Spatial Configuration of Belief: The Archaeology of Mycenaean Religion,” in Placing the Gods: Sanctuaries and Sacred Space in Ancient Greece, ed. S. E. Alcock and R. Osborne, Oxford, pp. 37-78.

Wright, J. C., J. F. Cherry, J. L. Davis, E. Mantzourani, S. B. Sutton, and R. F. Sutton Jr. 1990. “The Nemea Valley Archaeological Project: A Preliminary Report,” Hesperia 59, pp. 579-659.

Mary K. Dabney

BRYN MAWR COLLEGE

DEPARTMENT OF CLASSICAL AND

NEAR EASTERN ARCHAEOLOGY

BRYN MAWR, PENNSYLVANIA 19010-2899

mdabney@brynmawr.edu

Paul Halstead

UNIVERSITY OF SHEFFIELD

DEPARTMENT OF ARCHAEOLOGY

SHEFFIELD S1 4ET

UNITED KINGDOM

P.Halstead@sheffield.ac.uk

Patrick Thomas

UNIVERSITY OF EVANSVILLE

DEPARTMENT OF ARCHAEOLOGY AND ART HISTORY

EVANSVILLE, INDIANA 47722

pt4@evansville.edu

COPYRIGHT 2004 The American School of Classical Studies at Athens (ASCSA)

COPYRIGHT 2005 Gale Group