This site is a preliminary synthesis of the results of the Nemea Valley Archaeological Project as they pertain to the Greek Late Bronze Age. Most of the illustrations are large, intentionally so that they can be downloaded and enlarged for closer inspection by users of the site. Not recommended for those without hard Ethernet connections.

Mycenaean Nemea: The Results of the Nemea Valley Archaeological Project
Archaeological Survey

Intensive archaeological survey was conducted in the valley of Ancient Nemea and adjacent areas during three summer campaigns, 1984-86. More extensive survey in 1989 examined areas left unstudied in 1984-86, resulting in virtually total coverage of the valley of Ancient Nemea itself, southern parts of the Tretos Pass, and the valley of Xerokambos/Tourkovrisi to the south of the town of New Nemea. Despite its proximity to Mycenae, the area was archaeologically not well known before investigations sponsored by the Nemea Valley Archaeological Project (NVAP). A Gazetteer of Aegean Civilisation in the Bronze Age notes only two Late Helladic sites, those of Tsoungiza, and Phlious, and the only other Mycenaean finds published between the date of its appearance in 1979 and the beginning of researches sponsored by NVAP were those from excavations in the Sanctuary of Zeus (see below). With the exception of Tsoungiza, finds were either exceedingly limited, as at Phlious to a single kylix stem and a steatite whorl. In contrast NVAP identified Mycenaean finds at twenty-five different sites and several dozen off-site locations within the study area of the project.[1]

In the three seasons of intensive survey artifacts were initially collected from tracts, units of homogeneous vegetation and conditions of visibility defined in the field and mapped by survey teams; densities of artifacts in each tract were estimated on the basis of counts of pottery and tile recorded by team members as they walked over each tract. Maps showing artifact densities and the distribution of artifacts of particular dates were subsequently prepared and on this basis approximately 100 high density concentrations of artifacts were defined as sites. Further collections of artifacts were made at these locations, using a variety of collection systems including 1) Transects and Grabs; 2) Grids; and 3) Field Grabs and Field Middles.

Two basic types of data are consequently available for the study of Mycenaean settlement and land use patterns in the Nemea area. The first consists of distributions of artifacts collected from tracts in the course of initial field walking. In some cases, Mycenaean artifacts found in such circumstances are from locations later defined as sites. In many other instances, however, artifacts were found only in very small quantities, suggesting that explanations for their presence in the landscape should be sought in processes other than those associated with permanent long-term occupation of the locations where they were found. This is also true of instances in which very small quantities of artifacts were found at sites where the preponderance of artifacts collected belonged to other periods. Making sense of the overall patterns in the distribution of Mycenaean artifacts thus demands careful analysis of the data, analysis which takes into account not only find spots of the artifacts, but the quantities of artifacts found, the spatial extent of their distribution, and the mode of surface collection employed to collect them.

Quantities of Finds at Sites and Their Distribution

Pottery and other artifacts that can be dated to the Late Bronze Age (i.e., to the Mycenaean period) were found at approximately twenty-five sites widely distributed throughout the study area, not including Tsoungiza and the Sanctuary of Zeus. At most of these sites, only a few sherds of Mycenaean pottery were found (fewer than five), densities not significantly higher than at those locations where Mycenaean pottery was found not associated with sites (see below). More substantial numbers of Mycenaean artifacts were found at only eight of the twenty-five sites.

These eight sites are fairly evenly distributed throughout the study area: three (205 and 209/213) in the Tretos Pass; two (503 and 923) in the Xerokampos/Kato Aspria area; two at the northern end of the Nemea Valley (922 and 925), one near the middle of the valley (003); and one (400) at the south end of the valley, west of Tsoungiza, in the pass between the valleys of Old and New Nemea.[2] To these can be added the sites of Tsoungiza, where Mycenaean finds have been recognized in surface collections (see below), as well as in excavations, and in the Sanctuary of Zeus. All are located at relatively low elevations, most near modern transportation routes.

Site Elevation Setting
003 374 knoll; western slopes of Nemea valley
205 300 knoll near road from Tretos Pass to Heraklion
213 260 bluff overlooking Tretos Pass
400 440 knoll in pass; between Heraklion and Nemea valleys
503 428 knoll in pass between Tourkovrisi and Palaiohano
922 345 knoll east of road to Vrahati
923 280 interfluve at intersection of roads to Phyktia and Zaharias
925 400 ridge near road linking Nemea valley to Cleonai

Table 1. Setting and elevation of sites where significant quantities of Mycenaean finds have been recognized.

The Size of Sites and Site Hierarchies

Sites where the most Mycenaean artifacts have been found in both the northern and southern parts of the study region are also the largest in terms of the area over which the Mycenaean finds are distributed.[3] Virtually all sites defined at Nemea are multi-component, in the sense that artifacts of more than a single period are represented at them. The overall size of the site is thus an unreliable basis on which to estimate the size of the site in any specific period represented among the finds from the site. For this reason, estimates of sizes used here refer to the area over which finds specifically of the Late Bronze Age were recorded.

Size estimates of this sort will to some extent reflect the method used to collect and record finds at the site: generally speaking, methods of collection, such as the Grabs and Transect system, that only grossly record information about the findspots of specific artifacts may produce higher estimates for the size of any given component of a site than a more fine-tuned system of collection (e.g., a 10-m. grid). Since the Grabs and Transects method is incapable of fixing the location of collected finds more precisely than to a quarter of the total area of the site, in a worst case scenario the estimated size of a spatially restricted component may be greated inflated if it is located near the center of the site. The larger the site, the greater the potential for size inflation.

The Field Middle and Grab system can also fail to estimate accurately the extent of small components within a large site, since the locations of artifacts cannot be fixed more closely than to field itself, typically an area of approximately one hectare. More precise are estimates derived from a gridded collection of a site, but even these must be used with caution. In instances where surface finds are distributed discontinuously across the site, the total size of the site can be calculated in two ways: 1) as the sum total of the areas of all the grid squares in which Mycenaean finds have been identified; or 2) discontinuities can be ignored and the size of the site calculated as the area of the polygon defined by the edges of the overall distribution of Mycenaean finds. Because these methods can result in very different calculations of the size of the Mycenaean component, one a minimum, the other a maximum; both estimates are here reported.

Site Minimum Size Maximum Size Collection Method
003 0.84 Transects and Grabs
205 0.60 Transects and Grabs
213 0.50 Modified Transects and Grabs
400 0.32 1.96 20-m. Grid
503 0.30 0.30 10-m. Grid
922 0.28 1.50 20-m. Grid
923 1.7 1.7 20-m. Grid
925 4.24 Field Middle and Grab

Table 2. Estimates of the size of the Mycenaean component at sites where where significant quantities of Mycenaean finds have been recognized.

The size of all Mycenaean components seems to be smaller than two ha., while some are much smaller. The exceptionally high estimate of the size of Site 925 seems to be principally the result of the Field Middle and Grab collection procedure employed at the site and probably is a gross overestimate of the size of the Mycenaean component. Late Helladic artifacts were found to be concentrated in two distinct locations, one on the top of the ridge (see description of the site on-line) in an area ca. 200 m. x 170 m., the second ca. 150 m. to the northwest, in an area ca. 120 m. x 70 m. Despite the extent of the finds, they were few in number and there are reasons to believe that subsurface archaeological deposits are considerably smaller in extent than surface distributions imply. Modern agricultural is intensive in both locations where Mycenaean artifacts were found; all artifacts were very worn and appear to have been transported some distance by erosion. There are thus good reasons to doubt that the area of Site 925 in the Mycenaean period was much, if any, larger than the others listed in Table 1.2.

None of the eight sites seems to be nearly as large as Tsoungiza, which in size dominates the valley of Ancient Nemea, and indeed the entire study area: 7.5 ha. is a conservative estimate for the extent of Mycenaean surface artifacts there (ca. 250 m. x 300 m.). The vicinity of Tsoungiza was intensively surveyed according to the same procedures employed elsewhere in the study region; after tract collection, extensive areas north, west, and east of excavated parts of the site were investigated in an attempt to define more precisely its size.[4] For this purpose a 20-m. grid was used, tied to the coordinate system of the excavation grid (the gridded areas were assigned site numbers 906, 907, 933, and 934 so that information from them could be integrated with the databases used for surface collections at other sites. The estimated size of Tsoungiza is based on the analysis of finds both from tracts and these gridded collections.

Other evidence supports the hypothesis that the dominant focus of settlement in the study area was located at the southern end of the valley of Ancient Nemea. Immediately east of Tsoungiza, substantial numbers of Mycenaean finds have been found in the Sanctuary of Zeus in three principal locations: in an area ca. 140 m. N-S and 160 m. E-W around the Temple of Zeus; in an area ca. 20 m. x 40 m. in size to the SW; and within an isolated 20-m. grid square ca. 120 m. W of the temple. But in no place has a pure Mycenaean stratum yet been excavated and in most locations finds appear to be heavily worn and transported by erosion (N.L. Klein, "Prehistoric Material from the Sanctuary of Zeus at Nemea" [unpublished ms. 1987]). In addition to Mycenaean finds from sites, there is also an abundance of Mycenaean offsite finds at the head of the valley, where they are more common than in other parts of the survey areaÑi.e., south of the village of Ancient Nemea, but particularly to the east and northeast of the Sanctuary of Zeus.

All available evidence suggests, therefore, that there existed in Mycenaean times a clear size hierarchy of sites in the area examined by the NVAP survey, with the site of Tsoungiza and associated activity areas at the head, or the southern end, of the valley of Old Nemea dominant. It seems also clear that any further investigation into the role that Tsoungiza played within the Mycenaean settlement system of the Argolid and Corinthia must take into account not only the remains at Tsoungiza, but also those in the Sanctuary of Zeus as well as lower density distributions in surrounding fields. In this regard, the nature of the activities conducted in the Sanctuary of Zeus and their relationship to Tsoungiza is critical to any assessment of the situation, but is difficult to determine. Some of the excavated Mycenaean finds in the Sanctuary of Zeus cannot have eroded from Tsoungiza itself. And their quantity is so substantial that it is difficult to imagine that all were carried here through human agency, through some process, such as manuring, unconnected with settlement.

On the other hand, finds from the Sanctuary of Zeus need to be considered within the larger context of Mycenaean finds at the southern end of the valley. Their presence within the confines of the later sanctuary seems hardly unusual, although it has previously been suggested a Mycenaean cult place existed preceded the historical sanctuary. The finds from the excavations are not themselves suggestive of religious activities, and what evidence there may be for a Mycenaean cult place in the Valley of Old Nemea comes from Tsoungiza. At present, a more conservative interpretation of the evidence might interpret the finds from the Sanctuary of Zeus and elsewhere in the vicinity as the remains of short-term activity areas associated with the occupation of Tsoungiza that, in part, but not exclusively, in conjunction with agricultural practices such as manuring (see below), led to the creation of a "halo" of Mycenaean artifacts in the vicinity of the site.

It is impossible to be determine with certainty how far from Tsoungiza lay sites similar to it in size and complexity. Areas between the Xerokambos Valley and the Argive Plain are still little explored. Zygouries, to the northeast, in the valley of Ancient Kleonai, was as large or larger. To the west, NVAP explored only small parts of the valley of New Nemea around the Panayia Rachiotissa ridge, the historical acropolis of Phlius. Very small quantities of finds (fewer than three artifacts in any given tract) were discovered in several places on the ridge, and in its foothills to the north, west, and south. The overburden of historical strata is such that would be difficult to say more about the character of Mycenaean occupation here without extensive excavation. Substantial quantities of Mycenaean remains have also been reported from the site of Ayia Irini, on the western outskirts of the modern town of New Nemea, outside the NVAP study region.

Periods and Continuity of Use

The break between the Middle and Late Bronze Ages as reflected in finds from both survey and excavation is dramatic. Middle Helladic finds are extremely rare in the Nemea area before the Middle Helladic/Late Helladic transitional phases [Following EH III, the site of Tsoungiza also appears to have been abandoned until the time of the Shaft Graves and the picture that can be gleaned from artifacts in the Sanctuary of Zeus appears to be similar]. Almost all sites where later MH finds were recognized (6 of 7 sites) show evidence of continued use in the Late Bronze Age, while LH artifacts have been identified at nearly twenty additional sites. The Early Mycenaean period is well-represented both on-site and off-site throughout the study region.[5] There is an equally strong break in the artifactual record at the end of the Bronze Age. Both Protogeometric and Geometric finds have been observed at only one of the 25 sites where LH artifacts have been identified; Geometric finds have been noted only at an additional three. At most sites there is no trace of occupation or re-use before the Archaic period, when eleven of the sites with LH finds again show evidence of activity.

Although much of the pottery collected from sites is datable no more closely than to the Late Helladic period, the individual trajectories of the major Mycenaean sites defined by survey suggest that most were used for long periods of time; both Early and Late Mycenaean pottery has been found at the majority.

Site Phases Represented
205 LH II, LH IIIA2/B1
209/213 LH I, LH II, LH IIIA2, LH IIIA2/B1
400 LH IIIA2
922 LH

Table 1.3. Trajectories of principal Mycenaean sites.

After the resettlement of the Nemea area at the beginning of the Mycenaean period, both surface and excavated finds suggest that occupation was continuous until the 12th c. It is of interest for the evalution of evidence from other surveys that it has been possible for us to recognize every major stage of the Mycenaean period in surface collections, belying concerns that any particular phase of the LBA might be archaeologically difficult to identify because of the low chronological diagnosticity of its ceramics. The study of artifacts from sites refines somewhat the picture deduced from artifacts collected from tracts (Table 1.4). In both types of collections far fewer artifacts can be dated to the early Mycenaean period than to LH IIIA2-B (Table 1.5). As in excavated deposits from Tsoungiza, the LH IIIC period is only faintly represented in surface collections.

Periods in Tract Collections Periods in Site Collections

Table 1.4. Periods represented among finds from tract collections and site collections.

Tracts 5 5 16 72
(5) (5) (16) (73)
Sites 12 16 18 126
(7) (9) (10) (73)

Table 1.5. Number of sherds from tract and site collections by period. Possible sherds in parentheses.

Nature of Finds

A variety of types of pottery, both closed and open, is represented in surface collections from sites. Open shapes, particularly stemmed cups, dominate. Table 1.6 presents statistics for pottery from sites, a representative sample of which was catalogued.

Open Shape Number %
goblets, kylikes, stemmed cups and bowls 94 41
other cups and bowls 18 8
deep bowls 6 3
kraters 4 2
dippers 4 2
basin 1 0
other open 6 3
Subtotal 160 69
Closed Shape Number %
bridge-spouted jar 1 0
alabastron 1 0
jugs/jars 13 6
cookpots 27 12
other closed 6 3
Subtotal 48 21
Other Shape Number %
brazier 2 1
jug/cup 1 0
bowl/pitcher 1 0
pitcher 1 0
other bases 8 4
other rims 6 3
other body 4 2
handle not assignable to shape 23 10
Total 231

Table 1.6. Shapes represented in site collections.

In most instances Mycenaean pottery found in the course of initial fieldwalking comes from tracts at or near locations where sites were later defined. In several instances, Mycenaean sherds were found in tracts in small numbers and not concentrated in any particularly location. One such cluster of tracts lies immediately northeast of the Sanctuary of Zeus. The Mycenaean finds from these tracts (Table 1.7), like collections from sites, consists largely of small open vessels.

Open Shape Number %
goblets, kylikes, stemmed cups and bowls 6 46
other cups and bowls 2 13
Other Number %
handles 4 26
body sherd 1 7
Total 13

Table 1.7. Shapes represented in off-site tracts northeast of the Sanctuary of Zeus.

Assessing the Reliability of Artifact Distributions

Soil units in the study area were mapped in 1986 and 1989 by Anne Demitrack.[6] One object of her study was to provide a basis for judging the extent to which alluvial soils may today be masking prehistoric land surfaces. Soil units were initially dated with reference to artifacts found in exposed sections or collected from sites situated on top of the soil unit (see Demitrack in Wright et al. 1990, p. 588 etc.). Three Holocene soil units were identified: H1, the oldest, deposited some time between the Early Neolithic period and the Early Bronze Age; both H2 and H3 are post-prehistoric. Subsequent mapping of the distributions of artifacts collected from sites and tracts confirms this picture. In the valley of Ancient Nemea, no site where Mycenaean finds was collected located on either the H2 or H3 soil unit.[7] Nor is any tract where Mycenaean finds were collected situated in its entirety on the H2 or H3 unit. Parts of the site of Tsoungiza that have been excavated are all located on marl and colluvium; in contrast, the Sanctuary of Zeus is entirely covered by the H2 soil unit, the earlier portion of which buried the Classical remains of the Sanctuary. Inspection of soil maps of the Nemea Valley suggest that there are other parts of the valley bottom where prehistoric remains could also be buried, particularly at its southern end, beneath the H2 or H3 soil unitsÑa fact that needs to be considered in estimating the size of the prehistoric settlement at Tsoungiza and its relationship to the excavated remains in the Sanctuary of Zeus.

The picture in the Xerokambos Valley is similar.[8] Substantial parts of the valley floor are blanketed with post-prehistoric alluvium and Mycenaean artifacts were not found either in tracts or at sites entirely located on the H2 or H3 soil units.

Distributional Patterns of Mycenaean Artifacts Relative to Current and Recent Agriculture

In attempt to measure the extent of land brought under cultivation in recent times, under agricultural regimes that have emphasized the cultivation of vines and olives, maps, based on aerial photographs, were prepared, under the supervision of Susan B. Sutton, showing the extent of cultivation in the Nemea area at three points in time, ca. 1943, ca. 1970, and ca. 1980. In each case a similar amount of land was cultivated. For our purposes it is interesting that Mycenaean sites as well as low-density distributions of artifacts have been found in virtually every part of the Nemea area that has been in cultivation since the 1940s. Few currently cultivated parts of the Nemea area failed to produce any Mycenaean remains, although they are rare in the Xerokambos Valley, in the areas of Stanotopaki and Tourkoilias (on the western edges of the Tretos Pass), and on the slopes of Mt. Liofata, at the east side of the valley of Ancient Nemea.[9] The evidence suggests that the scale of land use in the Late Bronze Age was not dissimilar to that in the 20th century and that then, as now, virtually the same types of land were under cultivation. Since many of the areas that have been under cultivation in the 20th century are most suitable, or only suitable, for olive cultivation, it seems difficult to avoid the conclusion that this crop was a principal component of the Mycenaean agricultural system.