Whose Dark Age?

With the possible, but dubious, exception of Homer (see pp. 24-5), ancient authors preserved no memory of the four or so centuries that followed the collapse of the Mycenaean palaces. Save for the locations of the impressive citadels whose remains continued to be visible, they display little recollection of the Mycenaean period either, but at least here we have the testimony of the Linear B tablets to supplement the material record. For the succeeding centuries we have only archaeological evidence, though recent decades of archaeological exploration have seen impressive advances in our knowledge of this period. This has persuaded some that the term “Dark Age,” conventionally employed to designate the period ca. 1200-750, is now a misnomer. The Dark Age, it is argued, is defined less by the material conditions of the immediate postMycenaean period as it is by our ignorance of those conditions on account of the absence of contemporary written documents. Since, however, archaeology has stepped in to compensate for the lack of literary evidence, the Dark Age has now been illuminated. That we know more about this period now than we did several decades ago is indisputable. Nevertheless, while it would be futile to deny that some continuities are traceable across the centuries of darkness, such information as has come to light serves only to confirm a general picture of isolation, introversion, and instability for mainland Greece and the islands of the Aegean (Cyprus and, to a lesser extent, Crete weathered the crisis with more resilience).

The data in Figure 3.4 are taken from a survey, conducted thirty years ago, of sites in southern-central Greece (including the Cyclades but excluding the Ionian islands, Epirus, Macedonia, Thrace, and Thessaly). What emerges is a steep decline in the number of archaeologically identifiable sites, beginning in LHIIIC and continuing to a low point in the succeeding Submycenaean phase (ca. 1070-1000). There is then a recovery in the Protogeometric period (ca. 1000-900) and a further increase in the Geometric period (ca. 900-700), though the total number of sites is still well below the level attested for LHIIIB, prior to the palatial destructions.

It would be unwise to assume a direct correlation between site numbers and population levels. As we have seen, the settlement area around the citadel of Tiryns was far more extensive in LHIIIC than it had been in LHIIIB and this suggests that the decline in the overall number of sites in the twelfth century was due in part to processes of settlement nucleation whereby small rural sites were abandoned in favor of larger, more defensible locations. But were settle-

Ment nucleation the only explanation, we would expect the few sites that are attested in the eleventh century to be even larger, and this is not generally what we find. Furthermore, while ongoing excavations (for example, in East Locris and the Cyclades) have revealed greater continuity in settlement than was apparent even thirty years ago, it still remains the case that the sites that continued to be occupied throughout the Dark Age were only sparsely settled. Numbers of graves are a notoriously unreliable indicator for population levels but the fact that the number of eleventh-century graves at Argos is less than half the number attested for the tenth century is at least consistent with the settlement evidence. An estimate of between 600 and 1,200 for the Dark Age population of Argos is probably in the right range if we regard the lower figure as an absolute maximum for the eleventh century.

Argos was among the larger Dark Age settlements. More typical, perhaps, is Nikhoria in Messenia. A site of some importance in the Late Bronze Age - it may even have been the administrative center of the “Further Province” mentioned in the Pylos tablets - Nikhoria suffered a destruction ca. 1200 though was not entirely abandoned. In its first Dark Age phase (ca. 1075-975), simple one-room dwellings were erected in haphazard fashion directly above the ruined foundations of the Mycenaean settlement. The excavators estimated that Nikhoria was at this time home to some thirteen or fourteen families, or eighty-five to ninety people in all. In the next phase (ca. 975-800), clusters of apsidal houses sprang up, providing home for up to forty families, or 200 people, though the community appears to have dwindled again in the final phase (ca. 800-750) with an estimated population of 100 people, or twenty families. There is no particular reason to suppose that there were more populous settlements in Messenia at this time: survey results suggest that the Dark Age population of the region was little more than 10 percent of what it had been during LHIIIB. What, then, had happened to the Bronze Age population of Messenia?

It is possible that some died in the disturbances ca. 1200, but this cannot be the whole story. People evidently left and since there are few indications in eleventh-century Greece for the emergence of new settlements, it is not unlikely that some of them were among the settlers who established new communities in Asia Minor in the eleventh and tenth centuries (see pp. 99-100). Another possibility is that part of the population adopted a more transient way of life within Greece, occupying on a temporary or seasonal basis sites in marginal areas. The peripheral location of such sites and the non-permanent nature of their occupation would certainly make them correspondingly more difficult to identify in the archaeological record. Telling in this respect, perhaps, is the scarcity of identifiable cult locations in this period, notwithstanding the clear evidence that the religious ideas and practices we find attested in the later Archaic and Classical periods preserve important continuities with those of Bronze Age Greece. Evidence for continuous cultic activity across the Dark Age is indicated for only a handful of sites: Kalapodhi in Phocis, Kato Symi, and perhaps the Diktaian Cave on Crete. More sanctuaries appear in the course of the eleventh and tenth centuries (e. g. Amyclae in Laconia, Isthmia, Kombothekra and Olympia in Elis,

Mende-Poseidi in the Chalcidice, Mount Hymettus and Munychia in Attica, the Polis Cave and Aetos on Ithaca, and perhaps Tegea in Arcadia, the Argive Heraion, and the Samian Heraion), but it is not until the eighth century that there is a veritable explosion in the numbers of permanent cult places. At several sanctuaries where we believe there to have been Bronze Age cultic activity - Agia Irini on Ceos, the Maleatas sanctuary at Epidaurus in the Argolid, and perhaps Delphi - there is a complete lack of evidence for Dark Age worship.

The hypothesis of a more mobile population is closely connected to the view that subsistence strategies in the Dark Age were dominated by pastoralism. The evidence for pastoralism is, admittedly, far from overwhelming and is predominantly based on faunal analysis from Nikhoria which indicates a sharp rise in bovine consumption between the latter part of the thirteenth century and the Dark Age. Pastoralism, it is argued, would have been an obvious subsistence option when agricultural regimes were threatened by the instability that accompanied the palatial destructions but it would also have required greater mobility and the seasonal or periodic occupation of sites as grazing areas became exhausted. Furthermore, a meat-rich diet was less nutritionally beneficial than one based on grain, pulses, and vegetables and this dietary change would have had deleterious effects on reproduction rates, thus contributing further to demographic decline.

Whatever the merits of the pastoralist hypothesis, it is clear that not everybody adopted a more transient lifestyle. The pottery styles of LHIIIC continue many Mycenaean traditions but they also display for the first time regional differentiations that are maintained into subsequent centuries and that are normally interpreted as indications for isolation and introversion. The vast majority of the material at Nikhoria - as at many other sites - is locally produced, and there are very few imports from other Greek regions. Similarly, while the diffusion of iron technology from Cyprus and the appearance of Near Eastern imports in tenth - and ninth-century graves from Attica, Euboea, Crete, and the Dodecanese testify to a certain maintenance of overseas contacts, there can be little doubt that such contacts were far less intensive and frequent than they had been in the Mycenaean period or than they would prove to be from the eighth century onwards.

It was not all doom and gloom. The site most frequently invoked by those who deny that there was a true Dark Age in Greece is Lefkandi, excavated since 1964 by Greek and British archaeologists. To the east of the modern village, the Xeropolis headland was first settled on a permanent basis towards the end of the Early Bronze Age (ca. 2000). Rebuilt ca. 1200, perhaps by refugees fleeing the disturbances on the mainland, recent excavations have revealed that the settlement continued to be occupied into the Early Iron Age, while Submycenaean cist graves appear in the Skoubris plot, situated on the hills to the northwest of Xeropolis.

Shortly after 1000, an impressive apsidal building, forty-five meters long and ten meters wide, was constructed on a leveled rock platform in the Toumba plot, southeast of the Skoubris cemetery. With plastered mud brick walls standing on stone foundations, its gabled thatched roof was supported by a row of internal central columns and by an exterior wooden colonnade, making it not only the earliest known peripteral (colonnaded) building in Greece but also the only structure in the whole period that can justifiably warrant being described as monumental. Divided into five rooms, two shafts, 2.75 meters deep, were discovered in the middle of the largest, central room. In the more southerly shaft were the skeletons of four horses, two with iron bits still in their mouths. In the other were two burials. One was the cremation burial of a male, aged thirty to forty-five: his ashes had been gathered in a linen cloth and placed within a twelfth-century bronze amphora of probably Cypriot manufacture; closed by a bronze bowl, the urn was accompanied by a whetstone and an iron razor, spearhead, and sword, the latter originally sheathed in a wooden scabbard. The other burial was the supine inhumation of a female, aged twenty-five to thirty, who wore bronze and iron pins, a gold and faience necklace, an elec-trum ring and a gold ring, gold hair-spirals, and a gold brassiere. By her head was an ivory-handled iron knife, which suggested to the excavators that she might have been sacrificed during the funeral of the warrior whose remains were found beside her.

The interpretation of the building is fraught with difficulties - not least because it appears to have been partially dismantled shortly after its construction and deliberately buried beneath a huge earth tumulus. Many believe that it was the house of a “big man” - that is, a leader of a small, fairly egalitarian community who achieved his status on account of his military prowess and competitive generosity (see further pp. 129-34). The status of the big man is always fragile, rarely outliving him, and the suggestion is that the obliteration of the Toumba building, immediately after its occupant’s death, visibly signaled the end of his authority. The excavators, however, have noted that there are scorch marks on the bedrock below the clay floor of the building, suggesting that the funeral took place before the construction of the building - a conclusion that may be strengthened by the observation that there is a slightly wider gap between the two internal supports either side of the shafts than between any of the other supporting posts. They suggest that the building was designed as a type of mausoleum but that it was buried out of fear after some unforeseen circumstance - perhaps subsidence in the zone.

Immediately after the construction of the tumulus, a new cemetery was laid out at its eastern end. The second half of the tenth century is, in fact, the great heyday of Lefkandi: the influence of Attic - alongside Thessalian and Cypriot - styles on the local pottery is particularly noticeable and the practice of inurned cremation may also derive from Attica, but the most distinctive feature of this period is the voluminous presence of precious metals, especially gold, and of imported luxury items from Egypt, Cyprus, and the Near East. Those imports continue into the ninth century, even if local ceramic styles become more resistant to the influence of other Greek regions, but around 825 the excavated cemeteries are abandoned and Xeropolis appears to go into decline. The settlement becomes open once more to influences from Attica, Thessaly, and now

Corinth in the second half of the eighth century, before being for the most part abandoned ca. 700 (though see p. 7).

How typical is Lefkandi? The human resources that the warrior evidently commanded, the wealth that the community - or at least part of it - possessed, and the long-distance contacts that are attested with Egypt and the Near East are unmatched on this scale anywhere else in Greece during this period. It is always possible that there are further sites like Lefkandi waiting to be discovered, although it would be hard to claim that the archaeology of Early Iron Age Greece has been neglected in recent decades and the few parallels that have been suggested for the Toumba building - for instance, Megaron B at Thermon in Aetolia - hardly measure up. There is some debate about the size and the nature of the community that lived at Lefkandi. Assuming that the excavated graves are a representative sample, the population should have been smaller than Argos and perhaps closer in size to the community at Nikhoria. One estimate would even put the population below fifty. On the other hand, it has been calculated that the tumulus that covered the Toumba building would have required between 500 and 2,000 days of human labor. Unless the inhabitants of Lefkandi could rely on outside labor, it is difficult to imagine that they could have accomplished the task with so small a population. While it is possible that there are many more cemeteries awaiting discovery in the area around Lefkandi, it is also not unthinkable that a substantial part of the local population has not been recognized in the archaeological record because it was not afforded formal burial - a possibility that has been proposed for nearby Attica in this period (see pp. 80-1).

Those who argue for a small community tend to envisage it as broadly egalitarian, save for the pre-eminent authority of the warrior buried beneath the Toumba building. Yet there are some indications to the contrary. Figure 3.5 shows the percentage of graves that contained metal objects in the three principal cemeteries at Lefkandi, while Figure 3.6 tabulates the average number of metal artifacts in those graves together with the standard deviation (a measurement which serves as an index of variance). The figure of 60 percent for the earliest graves in the Palia Perivolia cemetery is not terribly significant because the sample includes only five graves. What is more interesting is that far fewer of the Late Protogeometric (ca. 950-900) and Subprotogeometric I-II (ca. 900-825) graves in this plot contained metal objects than is the case with either the Skoubris or the Toumba cemeteries. Furthermore, the average number of

Figure 3.5 Percentages of graves with metal items at Lefkandi by cemetery and period

Figure 3.6 Average number of metal items in graves at Lefkandi by cemetery and period together with the standard deviation around the mean

Artifacts in graves in the Palia Perivolia cemetery is well below that in the other two burial grounds and there is far less variance between the grave assemblages there. Mere numbers cannot capture the wealth of some of the burials in the Toumba cemetery, especially in its later phases: Tomb 36, for example, contained a gold diadem and ten gold attachments, alongside two gold rings, three bronze fibulae, a bronze bracelet, and pieces of faience, amber, and crystal. Nevertheless, the differential in wealth between - and, in the case of the Skou-bris and Toumba plots, within - cemeteries might lead one to suspect that the community at Lefkandi qualifies at least as a ranked society (see p. 130). Set within a broader context, however, the apparent exceptionality of Lefkandi only indicates that in the Dark Age an aggregately lower amount of wealth was distributed less evenly between communities than was the case in either the Mycenaean period or the period from the eighth century onwards. Lefkandi hardly serves to refute the concept of a Dark Age.

Unless, that is, the Dark Age is simply a historiographical mirage, generated by faulty chronological reckoning. This is precisely the suggestion made in 1991 by a consortium of historians and archaeologists. Struck by the seemingly seamless resumption, after three or four centuries, of certain craft skills such as ivoryworking, figured pictorial representations on pottery, monumental stone architecture, and even literacy itself, they wondered whether the palatial destructions may not have taken place considerably later, thereby compressing the period of time normally allotted to the Dark Age. In their view, the source of the problem lies with Egyptian chronology, upon which all the dating-systems of the Mediterranean and Near East are, to a greater or lesser degree, dependent. The relative chronology of Egypt is derived from late epitomes of a text, originally written ca. 280 by an Egyptian high priest named Manetho, which records the names of all the kings of the thirty dynasties together with the number of years that they reigned. This chronology is then pegged to absolute dates by what is known as the Sothic theory. The Egyptian year began with the flooding of the Nile. Ideally this should coincide with the rising of the Sirius star on the eastern horizon immediately before dawn - something that did, in fact, occur in 1321 and 139 CE - but since the solar and lunar years are not the same length, the two events would have diverged from one another at a rate of one day every four years unless the Egyptians practiced intercalation (e. g. the insertion of an extra day every fourth year). Thus, when Egyptian documents name the calendar date on which Sirius rose, we are able to count the number of days that separated this event from the New Year, multiply the result by four and arrive at the number of years that had elapsed since the start of the Sothic cycle. If the same document also records in which regnal year of which pharaoh an event took place, then we are able to synchronize Manetho’s list with absolute dates.

The objections to this system are that we cannot be sure the Egyptians did not practice intercalation, that the rising of Sirius would have been visible on different days depending on where one was in Egypt, and that Manetho’s king list may not be reliable in the first place. By some ingenious number-crunching - and a few wild guesses - it is proposed that the dates of the eighteenth, nineteenth, and twentieth dynasties should be lowered by about 250 years. The consequence for the Aegean is that the Mycenaean period would then have ended ca. 900, slashing in half at one stroke the Dark Age. The thesis offers a salutary reminder about the contingent nature of our chronology (see p. 39), but Egyptologists have disputed the claim that New Kingdom/Late Period chronology is entirely dependent upon Sothic dating and Greek archaeologists note not only that it is difficult to squeeze the eight chronological sub-phases of the Geometric style into a mere one hundred years but that radiocarbon dates broadly seem to support the traditional chronology. In fact, recent dendrochronological samples of charcoal matched against radiocarbon dates from Assiros in Macedonia may suggest that the traditional ceramic chronology for Early Iron Age Greece should even be raised by about 100 years.

Perhaps the ultimate failing of this chronological challenge, however, is the belief that a Dark Age is in some sense problematic and needs to be explained away. Dark Ages are attested for various historical periods around the world and only those who are unremittingly committed to a unidirectional evolutionist view of progress would seek to attribute every one of them to present-day ignorance. There were, as we shall see, many important underlying continuities that spanned the Dark Age, but it remains the case that Greece in the eighth, seventh, and sixth centuries was a very different place from what it had been in the Late Bronze Age. The explanation for this undoubtedly lies in the unstable conditions of the Dark Age, when centralized authority vanished, when obligations and loyalties lay nearer to home, and when self-reliance became essential rather than desirable.