Version classiqueVersion mobile

Le Péloponnèse

 | 
Josette Renard

Revenons à nos moutons. Surface Survey and the Peloponnese in the Late and Final Neolithic1

William G. Cavanagh

Texte intégral

  • 1 Acknowledgements : I wish to thank Josette Renard for the invitation to deliver this paper, Chris (...)
  • 2 Halstead 1981, 326-7 ; cf. also 1994

1I wish to start with an observation of Paul Halstead's published in 1981 «... the later Neolithic explosion of settlement in... southern Greece is perhaps less the resuit of population pressure on arable land forcing the colonisation of marginal areas than of developments such as olive cultiva tion or the milking of livestock making these areas more attractive for settlement »2. I believe that this insight helps make sense of the results of recent intensive survey in the Peloponnese. We find a high proportion of sites, which seem (agriculturally speaking) to be located in marginal locations, large numbers of cave sites, and many ‘sites’ (notably those discovered through archaeological survey), which are small, evidently shortlived, and probably serving some specialised purpose (at the very least not permanently occupied village sites). Other explanations may be possible, but many archaeologists have interpreted this evidence as indicating a greater emphasis on animal husbandry. Some have argued for a graduai demographie expansion pushing the population to exploit more marginal lands. This démographie argument is not entirely persuasive : firstly because the evidence is weak for a sufficient increase in population in S Greece at this time, and secondly because much good arable land was not settled. As an alternative this early colonization of secondary land can be understood as the resuit of a cultural preference by one sector of the population of LN/FN Greece, which placed particular value on stockraising. The preference for such a way of life would have become possible only through the development of a symbiotic relationship with communities, whose economy placed greater emphasis on arable farming. The dialectic of relationships between such communities would have significant consequences for developments during this long and important phase in Greece's history.

  • 3 Zachos 1996a
  • 4 E.g. Kephala Coleman 1977, esp. 106, pl. 67 ; Alepotrypa, Papathanassopoulos 1996, 223 pl. 30
  • 5 Blackman 1997. 73
  • 6 Cherry 1981, 1990
  • 7 Sampson 1984, 1987, 1988a, 1992, 1996

2The Late and especially the Final Neolithic periods have come into better focus only in relatively recent years. Not least because the beautifully painted pottery styles of the Neolithic seem to disappear as the period advances, and the pottery assemblages become dominated by coarse wares, there is a sense of decline. To be set against this, however, the period sees important developments, which were to be critical for the evolution of the Bronze Age. Zachos, for example, has recently pointed to the significant innovations in metallurgy during these phases3, and one might point to other technologies, such as the production of polished marble vases4. Moreover, particularly in S Greece and the Peloponnese, there seems to have been an expansion of settlement (fig. 1). Nevertheless the significant differences between S Greece and N Greece persist through the LN and FN periods : we find in the south neither the density of settlement, nor the very large scale settlements, such as that at Makriyalos5, that were characteristic of Greece from Thessaly northwards. Rather the pattern seems closer to that which Cherry, in particular, has recognised in the smaller Aegean islands during LN6, and the important work by Sampson on Euboea and in the Dodecanese7 points in a similar direction. We would suggest that the pattern of change they have characterised is not restricted to the islands, but applies more generally to S Greece and, in particular, to the Peloponnese. Recent survey evidence has brought this into clearer focus.

3To anticipate the argument three basic patterns of settlement development for the Peloponnese during LN-FN will be advanced :

  1. Expansion in the areas settled and increase in the variety of types of site (‘the expansion model’)
  2. No expansion (‘the continuity model’)
  3. Decline relative to EN and MN occupation (‘the settlement décliné model’)

4The results of the various surveys, where possible, will be examined in the light of other environmental evidence. Having assembled this evidence an attempt will be made to review the explanations which have been advanced and to place in context our understanding of the social and economic forces behind this critical phase in the prehistory of Greece.

fig 1 Map of Late and Final Neolithic sites in the Peloponnese (see appendix for liste of sites) By David Taylor.

  • 8 See recently Weisshaar 1979a and b, Warren and Hankey 1989, Coleman 1992, Perlès and Demoule 1993, (...)
  • 9 Treuil et al. 1989, 127
  • 10 Treuil ét al. 1989, 112 Table 1 ; Demoule and Perlès 1993, 366 fig. 2

5This is not the place to discuss the chronological terminology and dating of this period8. Terms such as Final Neolithic or Late Neolithic II are not universally approved, and the general linking, for the whole Aegean area, of rather distinct styles such as various forms of pattern-burnished pottery and crusted wares, has been criticised9. Certain regional groupings, notably the Attic/Kephala/Aegina group, have been clearly defined, but for much of S Greece there is still scope for refinement of the chronology. Thus on the basis of the stratigraphy of specific sites Sampson has proposed four subphases within LN, whilst Demoule and Perlès have recently presented a tripartite division for the whole of Greece. Nevertheless, when dealing with poorly preserved surface material from survey it is frequently difficult to be more précisé than a very general date. It is clear that the LN occupies an immensely long span of time : in calibrated years it might start around 5000 BC and end just before 3000 BC, a duration of, say, 1500-2000 years10. It is, thus, comparable, in length, with the whole of the Bronze Age. If for this reason alone, I believe there are dangers conflating the LN with the EBA : the distribution and density of settlement in the one period is very différent from that in the other, and the evidence for social complexity in EH II is very different from what we can observe of LN/FN Greece, and more especially S Greece.

The expansion model

Laconia Survey (fig. 2)

  • 11 Renard 1989

6The first point to make is that there is no evidence for settlement in the survey area before the Final Neolithic period. As far as the earlier part of the Neolithic is concerned our picture of central Laconia has not changed in the last 50 years : Kouphovouno continues to be the only site known with certainty to have been occupied11. This site covers some 4 ha in area, though we cannot yet say how large it was at any given phase of its life down to the EBA.

fig 2 Map of LN/FN sites from the Laconia Survey area and major soil types.By David Taylor.

7In the Final Neolithic and possibly into EHI we can point to finds from some 12 locations (TABLE 1)

Table 1 : FN/EHI find spots from Laconia Survey

Table 1 : FN/EHI find spots from Laconia Survey

8Of these the site on the hill of Plakia, E48, was richest in finds ; it produced a variety of pottery including types for the consumption of food, cooking and storage pithoi. A great variety of chipped stone artefacts was recovered with representatives of the whole chaîne opératoire-. 11 cores, primary, secondary and tertiary flakes, large numbers of blades, rejuve— nation pieces, debris. Specific types included tanged points, scrapers, backed pieces and a piercer. Polished stone tools included a flat axe, polishers and a pounder. The site was judged to have covered about 0.6 ha in extent. Of the remaining sites E77 and E81 appear to be satellites, perhaps special activity sites, probably to be linked with E48, though in truth the finds recovered were few and their date uncertain. B116, N363, L401, R429 and T480 and 481 have yielded stone assemblages consistent with a LN/FN date, but either no pottery or undatable sherds ; Bill and probably U487/489 have stone tools within the range LN-EHI. These other sites have either suffered so severely from erosion that their pottery has not survived, or they served some special function (meaning that they were not permanent habitation sites). The ‘non-site’ 10496 yielded a lot of sherds, from a few pots, concen— trated in a small area.

9The particular observation to which I wish to draw your attention is the distribution of these sites within the landscape (fig. 2). The two major clusters, E48 + E77 and E81, and Bill + B116, stand out because they are situated on the poorest soils in the whole region, on limestone out— crops, which today support very thin terrarossa soils allowing a sparse maquis/garrigue vegetation. There is a long history of Holocene soil erosion in Southern Greece and in the area, but it is unlikely that these limestone outcrops ever supported deeper soils. Likewise none of the remaining sites occupies primary agricultural land. Their distribution is markedly different from that of the Early Bronze Age, when settlements occupy the fertile sectors of the area which are still exploited as arable today, notably the ‘neogene’ maris, which dominate the southern part of the survey area, and the schist soils in the north.

10Taken at face value these observations suggest two inferences :

  1. The Neolithic colonisation of the hinterland of Laconia did not precede the LN/FN period.
  2. Although indicating the first identified human activity in the area, the sites occupied conspicuously passed over those types of soil and landform, which were favoured throughout the rest of history, and preferred locations which, in later times, have been used for defence or for sheep-folds.

11These conclusions appeared, to me at least, paradoxical : why should the first settlers in this large tract of land go for the least attractive part ?

Soutbern Argolid (fig. 3)

  • 12 Runnels et al. 1995, 7, 89-91 ; Jameson et al. 1994, 435-6, 468, 471, 477, 482-3, 488, 510, 521, 5 (...)
  • 13 Op. cit. 522
  • 14 Jameson et al. 1994, 346
  • 15 Op. cit. 347.

12The finds of the Southern Argolid Regional Survey indicate a number of sites of LN (D3, E5 ( ?), E14, G9) and FN date (A65 [Halieis], C15, C25, C29, F14, G7, G9 ; another 33 EH sites might have a small FN compo— nent)12. It has been suggested that the Didima Cave (D3), the Mouzaki Cave (E14), Kotena Cave (G9) and Koufo (G11, FN component small13) may have been used as shepherding sites perhaps « milking stands and overnight shelters located in the grazing lands a day's walk from Franchthi Cave »14. They suggest « A new settlement pattern characterizes the Final Neolithic... new sites (with)... celts, stone ornaments, millstones, and an obsidian blade industry in association with coarse handmade pottery suggest... farmsteads or small hamlets... »15, in addition to Franchthi itself this description fits C15 and C29 particularly well. Tanged bifacial points, similar to those from LS, were found at C29, D3 and E3.

  • 16 Jameson et al.1994, 232 fig.4.8,341-8
  • 17 Op.cit 346-7

13This number of LN-FN sites represents a distinct increase from the pre— ceding periods : EN is reported only from the site of Franchthi Cave16, whilst a few sherds of MN pottery were found in Mouzaki Cave (E14) and Didima Cave (D3) indicating, it is suggested, not « solely population growth, but... a shift in the mode of exploitation in the region », where the value of caves for livestock is underlined. FN is considered the culmination of a process of diversification started in MN, and partly under the pressure of the loss of prime arable land to marine transgression17.

Fig. 3 a. Maps of Final neolithic sites from the area of the Southern Argolid Survey. After Jameson et al. 1994, 233 fig. 4.10.

Fig. 3 b. Maps of Final neolithic sites from the area of the Southern Argolid Survey. After Jameson et al. 1994, 233 fig. 4.11.

  • 18 Bottema 1990
  • 19 Bottema 1990, 123-4
  • 20 Moody et al. 1996, 291, referring lo NW Crete

14Pollen evidence of relevance to the period under investigation has been recovered from Koilada Bay18. Zone i of the core covers an immensely long period (c. 6°700-3°200 BP, roughly MN-LBA), and it is suggested that « the very low arboreal pollen values in Khilada Zone i point to vegetations that were by nature of a very open character caused by almost steppic conditions ». The effects of human activity were clearly visible in the diagram firstly through the unexpectedly high counts of cerealia-type pollen, probably introduced through threshing very close to the coring site19, and secondly, perhaps, by the large numbers of charcoal particles, indicating fires. The last two observations apply to activity in the immediate vicinity, in the fields immediately surrounding Franchthi Cave, whereas the near steppe-like conditions must reflect the general aridity of this part of Greece. The non— cultivated landscape of the LN would not have been markedly different from that of before, even though evidence from further south indicates that conditions in the Late to Final Neolithic may have been wetter and colder20.

  • 21 Jahns 1993
  • 22 Jahns 1993, 200.
  • 23 Jahns 1993, 200.
  • 24 Moody et al. 1996 : 286.

15The pollen core from Lake Lerna21 gives support to these conclusions, and indeed a finer definition. The earliest zone I (c. 5°700-5°200 BC) contai— ned taxa resistant to high temperature and aridity, but sensitive to frost ; an open deciduous woodland is indicated, confirming the Koilada picture. Whilst there might be some human impact, the open character of the vegetation and the large numbers of Mediterranean species can be explained by the climatic conditions of the period. Zone II (c. 5°200-3°600 BC) is marked by denser oak woods and a decrease in non-arboreal pollen ; this change was not recorded in either the Koilada or the Kleonai cores, but the NW Cretan observations may support a climatic explanation. Certainly LN seems to be a period of limited activity at the adjacent site of Lerna (of course the Kephala Cave, Aria and Tiryns, not far away, have LN) and in zone II « no human impact can be proven from the pollen record »22. The first clear sign of human impact comes c. 3°600 BC, though from the archaeologist's point of view subzone IIIa (c. 3600-800 BC) spans a frustratingly long and varied period of human settlement. It is suggested, however, that woodland clea— rance increased from c. 2°500 BC and olive shows a continuous curve which rises slowly to 27%23. Again the NW Cretan pollen record indicates a jump in olive pollen production in the LN-FN period and then a graduai increase24, compatible with human exploitation, first of wild and then of domesticated trees.

  • 25 van Andel et al. 1990, 382.

16In the southern Argolid it would appear that agricultural settlement during the FN did not produce a soil erosion horizon that could be traced : « Evidence for soil erosion, however, is lacking (sc. in FN and EBA) until the end of the 3rd millennium when Pikrodhafni debris flows covered the valleys of those drainages that were occupied by settlers »25.

Berbati-Limnes (fig. 4)

17From the Berbati-Limnes Survey Johnson has recorded 19 Neolithic find spots, as well as scattered finds from 24 tracts. Of the 19 only two, FS400 and FS23 could be dated to MN. The former, the largest Neolithic site in the area, produced finds of EN, MN, LN and FN date. As a long-lived centre it might be compared with Kouphovouno, just outside the survey area in central Laconia or with Asea (see below).

  • 26 Johnson 1996, 37.
  • 27 Johnson 1996, 63.

18The expansion in the FN saw « the beginnings of site hierarchy... since three sites (FS 400, 39 and 12) each occupying a central or dominant position in the Berbati valley, the Limnes area and the Miyio valley respectively, are large and rich in finds compared to the surrounding small sites »26. Thus of the group associated with the site of Vigliza (FS 39), Johnson comments that arable agriculture was no doubt possible, but the choice of locations implies an increasing role for pastoralism27. Likewise in the third cluster FS 12 seems to be a small agricultural site, FS 31 produced only lithics and FS 28 may have been important for herding.

  • 28 Wells et al. 1990, 222 ; Demitrack et al 1990, 384 ; van Andel et al. 1990, 384 ; Zangger 1992, 17
  • 29 Zangger 1992, 19.
  • 30 Rackham 1990, 341.
  • 31 Rackham and Moody 1996, 18-24.

19It has been suggested that the FN and EH expansion of settlement observed in the Berbati-Limnes area contributed to a significant soil-erosion event recorded in the Argive plain28. The 5 m thick LN/EH alluvial unit, which was identified by Zangger on the coastal fringe close to Argos and Tiryns, consisted of coarse, poorly sorted sediments like those of the Pleis— tocene fans. It has been dated as post MN, because it overlies a MN site 14C dated to 6°240±125 BP (i. e. before or immediately after 5000 BC), and pre 2500 BC, as the site and the alluvium were eroded at the time of maximum transgression of the sea, itself dated by a radiocarbon estimate from another core. This large time span (2500 years) for an event described as catastro— phic, probably one event, and which must have happened within 30 years29, makes archaeological interpretation difficult : it might equally well belong to MN, LN, FN, EHI or even EHII. The extent to which landclearance might have caused soil erosion is controversial. Rackham has commented « It is often assumed that regions are deforested by cutting down trees ; that trees, and only trees, protect the landscape from erosion ; that there was no significant grazing or browsing before domestic animais ; and that goats destroy all vegetation. All these are theories, not truisms, and need to be demonstrated, not assumed »30. Other processes can come into play, and Rackham has suggested that rare but catastrophic deluges might also be an important agent in eroding landscapes in Greece31. On present evidence i should be cautious of interpreting the Berbati-Limnes LN/FN expansion as indicating a very extensive deforestation ; certainly some clearance of land for arable is indicated, though pastoralism is also indicated.

Asea Valley (fig. 5)

  • 32 Forsén et al. 1996, 85.
  • 33 Forsén et al. 1996, 88.

20« In the 1994 season, four Neolithic sites were discovered, the largest being of Middle Neolithic (MN) date, while the three others, of fairly small and uniform size, produced material which is datable to the Late Neolithic (LN) period and/or the Final Neolithic (FN) »32. Thus in addition to Asea itself another fairly large MN site (currently designated S16) has been found. On the other hand the LN and/or FN sites are much smaller than the big MN site : S2, with obsidian but no diagnostic pottery, is a ‘special-purpose site’, whilst the other two, S23 and S41, the former is near good arable land and copious springs, the latter interpreted as a small Neolithic farm ; more generally there develops in the LN-FN period « more, but smaller, sites using more marginal land »33.

  • 34 Forsén 1996.

21In her recent very careful review of the evidence from Asea34, Forsén has shown that the LN and FN material forms a smaller proportion of the finds than that of the earlier, especially the MN period. Unfortunately we cannot be sure if this is due to Holmberg's selecting an unrepresentative proportion of the sherds (note that LN/FN pottery is commonly very coarse, and may have been discarded for this reason), or really indicates that the site shrank in size. In brief there may be slightly more FN/LN sites, but they seem smaller in size.

  • 35 Forsén et al. 1996, 83.
  • 36 Op. cit. 85.

22Initial appraisal suggests that the major alluvial sediments in the valley plain were most probably deposited in the early Holocene35, and therefore would not be a consequence of human settlement. Certain localised deposits observed in gully sections revealed that parts, at least, of some sites could be buried by accumulations of sediment36.

Fig. 4.Map of neolithic findsports, Berbati Limnes, Survey After Johnson 1996, 39 fig. 2.

  • 37 I thank Jim Roy for this point.

23To the north of Asea we should mention the important site of Sakovouni, and the nearby Kamenitsa Cave. It seems natural to associate these sites, like the similar sites elsewhere in the Peloponnese, with an increased emphasis on animal husbandry. Appropriate, no doubt, to Arcadia, though the severe winters of upland Arcadia do present a problem to the Arcadian shepherd37.

Fig 5 Map of neolithic sites from Asea Valley Survey After Forsèn et al, 1996 86 fig 11.

Pylos Regional Archaeological Project (fig. 1)

  • 38 Davis et al. 1997, 417, 438-9.
  • 39 Treuil 1983, 44-5 ; McDonald et al. 1972, 130-1.

24Within the survey area Late or Final Neolithic material has been recovered by excavation from the Cave of Nestor, Voidokoilia and Chora Katavo— thra. An ambiguous pottery fabric which may be of Neolithic or may be of Middle Helladic date was found at Ano Englianos and four other locations38. So far no confirmed reports of EN or MN have been published from the area. Indeed, if we widen our scope to the whole of SW Peloponnese, the dating of pottery from Malthi to EN and MN is far from certain39, and of the sites published in the Minnesota Messenia Expedition only the isolated cave site at Velika, Kokora Troupa, was thought to have sherds clearly earlier than Late Neolithic. The PRAP data, therefore, confirm the impression already gained from MME that there was a modest expansion of settlement in the Later Neolithic period with open sites and also a number of cave sites occupied ; McDonald and Hope Simpson comment that sites can be located in remote and isolated gorges above what are now perennial rivers (in McDonald and Rapp 1972, 131).

  • 40 Kraft et al. 1980 ; Zangger et al. 1997, 559.
  • 41 Wright 1972.
  • 42 Kraft et al. 1980.
  • 43 See also Zangger et al 1997, 588.
  • 44 Zangger et al. 1997, 588 zone OS I.
  • 45 Op. cit. 589.

25Geomorphological cores have shown that at the beginning of the Holo— cene the Bay of Navarino extended 5 km north of its present limit ; massive deposition during the early Holocene produced an alluvial floodplain with deposits up to 24 m thick40. This seems too early for human intervention. Of the various pollen cores taken in Osmanaga lagoon and the Bay of Navarino, that established by Wright starts rather too late (c. 2°000 BC) to be of help for FN/LN41. Zone B2 in the Navarino core42, perhaps 6°500-5°500 BC (7°500-6°700 BP), shows a marked increase in pine forest, possibly to be linked with the spread of Pinus halepensis in coastal regions43. Again it would be surprising if there were any marked human impact on these develo— pments. Regrettably none of the cores produced good evidence for the period 5°000 to 2°000 BC44 as the sediments were coarse and unsuitable for pollen, but a first notable impact by man is recognisable at 390 cm in the PRAP core from Osmanaga, dating c. 2°000 BC. It is suggested that this impact occurred later than elsewhere in Greece45, and by implication that human influence in LN, FN and even EH was slight.

Continuity Model

Metbana (fig. 6)

  • 46 MS9. 40 and MS61. 35 may be the tangs of tanged points, Mee et al. 1997, 176 fig. 11. 40 21 and 19 (...)

26The acropolis site of Palaiokastro (MS10) has produced slight evidence for settlement in the MN period. No finds could be dated with certainty to LN and FN, and the distinctive FN/LN obsidian and flint points were not found46. There was, then, very little activity at any time in the Neolithic ; contrast the great expansion in the occupation of the peninsula during the Early Bronze Age (fig. 6).

Settlement Decline Model

Nemea Valley (fig. 7)

  • 47 Wright et al. 1990, 609.
  • 48 Cherry et al. 1988, 174.

27In general terms « EN and MN pottery of standard northeast Pelopon— nesian types is well attested, while the LN and FN periods are scarcely repre— sented at all »47. This generalisation is documented in Cherry et al. 1988. EN is known at Phlius, Tsoungiza, Nemea and at Sites 3, 201 and 10 ; MN at Phlius, Tsoungiza, and sites 700 and 702 ; sparse remains of LN/FN at Phlius and sites 506 and 702. Finds of LN barbed-and-tanged projectile points « may reflect hunting trips made by Neolithic peoples who resided outside the survey area »48. The richest Neolithic site (702 overlooking the Tretos pass) produced almost 4°000 sherds, the great majority of which were datable to MN. A few small, poorly preserved sherds could have been related to ‘Corinthian Grey Ware’, and so LN, whilst some sherds in a dark red fabric and a possible ‘cheese bowl’ fragment could be FN. Chipped stone tools were surprisingly few, but a large chert triangular projectile point, probably of FN date, is comparable with LS R429/1 (Carter and Ydo 1996, 156-7). In brief far from an expansion in LN/FN we see a decline.

  • 49 Atherden et al 1993.
  • 50 Atherden et al. 1993, 355.

28The pollen column from Sto Kephalari, Kleonai49 was just outside the east limit of the Nemea Valley Archaeological Project, as close as one can hope in a country as arid as Greece. Sadly zone 3 (SKK 3), which effectively includes LN-FN, produced very little pollen, and we cannot draw any conclusions about the vegetation in the region. The earlier zone 2 (SKK 2) indicates the sort of open woodland environment also evident in the Lake Lerna and the Koilada sequences. It is suggested that « agricultural activity in the Neolithic was limited and scattered »50.

fig. 6 Map of Early Helladic sites from the Methana Survey After Mee et al. 1997, 44 fig. 4.1.

Reliability of the Evidence

  • 51 E.g. Zangger et al. 1997, 568.
  • 52 Note, for example, the MN site discovered by Zangger in the Argive floodplain : Zangger 1992, 17.
  • 53 Cf. van Andel and Runnels 1995, 487.

29Before we can begin any attempt to interpret these trends the question of the reliability of the surface survey evidence needs to be addressed. In the first place we can be sure that sites have been lost through erosion, plough damage and other destructive processes. It must be noted in the first place that these processes have not acted evenly over time. Plough damage and possibly soil erosion were probably not so significant throughout the Neolithic ; and it can be argued that the effects of mechanized agriculture, over just the last 50 years, have been very much worse than destructive processes in the more distant past. We have observed these effects ourselves in the course of the Laconia Survey : site P269, for example, from which we collec— ted over 200 sherds, obsidian and ground stone tools in the 1°980s, was found when revisited in 1992 and 1993 to have virtually no artefacts left at ail. Similar stories can be told of sites elsewhere51. The burial of sites through alluviation and rise in sea level can be significant considerations for surveys in basins, floodplains and coastal areas52, though their effects can perhaps be overstated for the Greek landmass as a whole53. On the other hand the burial of sites through the action of soil animais does not seem to be a significant factor in Greece.

  • 54 E.g. Cherry et al. 1988, Whitelaw 1991, James et al. 1994, Zangger et at. 1997, esp. 569-76.
  • 55 Mee et al. 1997, 25.
  • 56 Zangger et al. 1997, 568.
  • 57 James et al. 1994, 414.

30The question of the effects of soil loss on site preservation and surface artefacts has been studied in a number of different contexts54. Conditions vary considerably from one régime to another. On Methana « there is no evidence of major Holocene erosion »55, whilst in the Pylos Region «... enhanced erosion and redeposition are likely to have removed a substan— tial portion of the evidence of early habitation where mari forms the bedrock »56. The Methana study emphasized « the importance of defining the mode of erosion as well as assessing its quantity... appreciable erosion of soil by surface wash may not resuit directly in significant redistribution of artifacts »57. Certainly we are not in a position to quantify the original numbers of sites in Neolithic Greece (and it may be doubted we ever will be), but detailed study does suggest that the archaeological record is not unreliable.Note, for example, the MN site discovered by Zangger in the Argive floodplain : Zangger 1992, 17.

fig. 7 Distribution of neolithic finds from the Nemea region After Cherry et al 1988 173 fig. 12.

31Moreover, where possible we have referred to the pollen evidence as corroboration that, in the Neolithic, man's relatively slight impact on the vegetation does not suggest high levels of population : independent evidence that survey does not seriously misrepresent the historical reality. It is, however, sufficient for the argument of settlement expansion firstly for us to be confident that the relative numbers of EN, MN and LN/FN have not been unduly biased by destruction through time, and secondly that the types of site and the environments in which they were found do not misrepresent the past. The fact that small single period sites of EN and MN date have been found in a number of the surveys, even in highly eroded environments such as the Berbati-Limnes and Nemea regions, lends support to this wor— king hypothesis.

32To a degree the different methodologies developed by the various surveys may lead to contrasting results. Thus it would be unwise to draw refined conclusions on the basis of the estimates of absolute site size from one pro— ject to another, as different methods of collection and treatment lead to différent estimates. Relative scales (small, medium, large) may be more relia— ble. Likewise, we try to make inferences about the function of sites on the basis of the artefacts recovered, but there are particular problems in inter— preting negative evidence. Is the absence of pottery from a find spot due to its original function (perhaps a chipping site or a pastoralists' camp) or has the pottery been completely destroyed by weathering ? Does the absence of querns from a site mean that they were never used there, that the site is not an ‘agricultural’ one, or is it simply an accident of survival or retrieval ? Our inferences increase in confidence where confirmed by other types of evidence (for example in the second case the site setting. and its suitability for agricultural exploitation), where combined with information from prior understanding (for example in the first case other information about pat— terns of obsidian processing at the time) and by a good sample of cases.

Overview

  • 58 E.g. Bottema 1990, 124 ; Moody et ai 1994, 287-8.
  • 59 Bottema 1994, 56.

33Granted these provisos, it is possible to offer a number of generalisations. Whilst there is evidence for some increase in the number of sites during the LN and especially the FN periods in the Peloponnese, there are also regional variations. It is difficult to put forward reasons that are not merely ad hoc to explain why within the NE Peloponnese, the S Argolid and Berbati Limnes surveys should show an expansion, Methana a stable pattern and the Nemea Valley a decrease in the number of sites. Even at its most marked (Berbati-Limnes) the expansion does not seem dramatic, particularly if we bear in mind that a number of the find spots do not imply permanent habitation. The pollen evidence goes some way to confirm this picture, with some, but relatively slight, evidence for human impact on the environmentduring the Neolithic as a whole. For the FN period in particular the Koilada column seems to imply an essential continuity from the earlier phases and the Lake Lerna column a more densely wooded environment. There is nothing from the Peloponnese to match the increase in olive pollen recorded in NW Crete. A number of analysts have commented on horizons of charcoal and have indicated in certain pollen spectra, an ecology of plants, notably those of the heather and cistus garrigue, which have a competitive advantage from fire58 ; it is not clear if the increase in pine should be mentioned in this context. Of course fires can start naturally, but fire-setting to improve fodder cannot be ruled out, and would fit in with the suggestions of an increasing emphasis, in the FN period, on sheep and goats. More generally, and referring to evidence from all Greece, Bottema has suggested that around 6°500 BP (say 5°400 BC), there was a spread of hop hornbeam (Ostrya carpinofolia) and eastern hornbeam (Carpinus orientalis). The one species might benefit from land clearance and the other from grazing, but these indications are insufficient to argue for an anthropogenic cause59.

  • 60 Whitelaw 1991, esp. 207.
  • 61 Evans 1971.

34Sites vary in size and function from villages and hamlets, to single farmsteads, caves, rock-shelters, obsidian scatters, perhaps chipping floors, and possibly even hunting sites. Estimating the size of a settlement from surface remains is especially difficult for this period. Whitelaw's careful review of the FN sites of Kephala and Paoura on the island of Keos suggests that, although it was originally thought that the settlement of Kephala cove— red the whole headland (2.2 ha), 0.7 ha would be a more realistic estimate60. Paoura may have been 1.5-2 ha. Both were small villages with populations, probably, of fewer than 100 individuals. In the light of these the 3.25 ha estimate for FS400, the largest FN site in the Berbati-Limnes survey, seems large, and other sites, such as Kouphovouno in Laconia, might have been of the same size ; Late Neolithic Knossos has been estimated as 5 ha61. Even the largest settlements are unlikely to have supported more than 1°000 or so inhabitants. On any reconstruction, and making every allowance for the loss of evidence through erosion and under-reporting, the population pool in the Peloponnese must have been very small.

Interpretations

Crops

  • 62 Cherry 1988, 23 ; Jameson et al. 1994, 343 ; Halstead 1994, 211 ; Zachos 1996b.
  • 63 Diament 1974, 363-6.
  • 64 Halstead 1981, 326.
  • 65 Sampson 1988a.

35Recently the LN/FN period has been seen as one of population growth62, though before the advent of survey data the evidence, if anything, pointed to a decline63. A more nuanced interpretation must take account not only of the ‘continuity’ and ‘decline’ models, but also of cases like the Asea Valley where, although the number of sites is greater in LN/FN, than in MN, the size of the sites diminishes. Moreover the duration of the MN period is considerably shorter (perhaps 500-1000 years) than LN/FN (1500-2000 years), and consequently simple site counts must be corrected to allow for the time over which they occur. In addition MN sites often appear rich in artefacts, and seem to have been rather long-lived, in contrast with the sparse representation and relatively short-lived duration of many LN/FN sites. Finally we must bear in mind the nature of the LN/FN finds, and in particular ‘special-purpose sites’ and the use of cave sites, which in some cases, such as Franchthi and Alepotrypa, may have been used all the year round64, but in others may have been seasonally occupied65. In brief, expansion of settlement, in the sense of the exploitation of areas that were pre— viously unoccupied, is clearly indicated by the evidence, but this need not imply a significant increase in population.

  • 66 Halstead 1981, van Andel and Runnels 1995.
  • 67 In summary Sherratt 1994, 170-1 ; Sherratt 1981 ; Bogucki, 1988, esp. 85-8, has argued that second (...)
  • 68 Pullen, 1992.
  • 69 Van Andel and Runnels 1987, 70-1 ; Jameson et al. 1994, 343-8.
  • 70 Van Andel and Runnels 1995.

36If, then, there is little evidence for pressure of population growth forcing people to occupy secondary land, an alternative explanation is needed. One such has Iooked to a shift from highly intensive methods of hoe cultivation, carried out close to the settlement and involving weeding and manuring, to extensive cultivation methods66. The secondary products revolution, as witnessed in C. Europe, including the use of the light plough and the first wheeled vehicles, is placed by Andrew Sherratt after c. 3500 BC, late to explain a shift in LN, though not impossibly so for FN Greece67. As yet direct evidence for plough agriculture in Greece does not pre-date the EBA68. Emphasis has also been placed on the importance of spring-fed agriculture, as an insurance against years of poor rainfall, in deciding the.ocation of earlier Neolithic sites in S Greece69. The authors have extended their argument, to explain the density of Neolithic settlement in the Thessalian and other European floodplains, by the exploitation of annual Spring inundations70. It is to be hoped that seed samples recovered from these sites might be tested for the types of weed associated with irrigation, as suggested by Glynis Jones' important research into irrigated agriculture.

Pastoralism

  • 71 With reference to the Classical period in Greece, Forbes 1995, 329.

37These authors would, then, see changes in the systems of plant agriculture as an important determinant in the changes in the later Neolithic. This may be part of the explanation. All the same, the evidence of site location in LN/FN indicates an interest in the exploitation of resources in addition to arable. To describe these locations as marginal may be to privilege the traditional farmer's values. For a community, which ascribed greater worth to animal stock, access to browse would be advantageous. As Forbes has recently had reason to stress71, « Greek pastoralism's raison d'être is primarily predicated on the existence of, first, a large proportion of the landscape that is otherwise mostly unproductive of humanly consumable food... (and) second, a significant amount of by-products of agriculture which humans are likewise incapable of converting into energy ».

  • 72 Halstead 1981, 1986 ; Cherry 1988.
  • 73 Halstead 1994, 196.
  • 74 Hubbard 1995.
  • 75 1986, 102-3.
  • 76 Papathanassoupoulos 1996, 39-40 ; Sampson 1988a.
  • 77 E.g. Chang and Koster 1986, 112-4.

38Halstead and Cherry72 have emphasised the difficultes presented to specialised transhumance in early farming Greece : the more densely wooded environment, making the herding of sheep difficult, and absence of larger populated centres making the marketing of specialised products nugatory. Moreover it is argued that « the high mountain grasslands, particularly the lush pastures of the north Greek mountains, have been greatly extended by human interference »73. They suggest that, towards the end of the Neolithic, these brakes were off. More extensive farming opened up the landscape, and significant centres of population were developing. Both arguments are stronger for Northern Greece than the South. We have already seen that the pollen record, at least for some parts of the Peloponnese, indicate a rather open environment throughout the Neolithic, and maquis, garrigue and steppe were significant components of the landscape from the early Holocene. Human impact is hard to discern, indeed the Lerna sequence indicates, if anything, a more densely wooded environment in the later Neolithic. An extent of grazing suitable for small cattle is, incidentally, also witnessed by the flourishing population of fallow deer, chamois and roe deer testified (if not fully represented) in the faunal assemblages of Greek Neolithic sites74. Likewise, it has been suggested above, the case for a demographie expansion and the development of concentrated centres of population in S Greece is not strong. Certainly an increase in pastoralism, as Chang and Koster have commented75, need not imply deforestation, and agriculture and pastoralism can complement rather than compete with each other. The value of cave sites and rock shelters to pastoralists have repeatedly been emphasised, by both archaeologists76 and ethnographers77, for habitation, as shelters — especially in winter and during the lambing period —, and for the storage of both animal fodder and grain, nuts and other produce. A noteworthy feature of a number of sites is the pairing of cave and open sites : Voidho— koilia and the Cave of Nestor, Sakovouni and its cave near Kamenitsa, Franchthi, Aria and there are others elsewhere in Greece. Likewise, whilst the ‘non-sites’ of LN/FN date are open to various interpretations, there are certainly pastoral uses such as folds, pens and wind screens, which could happily join the list of candidates.

Transhumance and agropastoralism

  • 78 Halstead 1996, 32.
  • 79 See for example the various examples in McCorkle 1992.
  • 80 Van den Brink et al. 1995, esp. 382-3, 385.

39Opinions differ about the nature and extent of transhumance at this time. Certainly the notion of sheep-ranching on the scale known in postmedieval Spain and Italy is hopelessly anachronistic, and has very properly been ruled out of consideration (Lewthwaite 1981, 1984). An economy of ‘pure pastoralism’ (which probably does not exist) and transhumance on a massive scale are not at issue in the context of LN/FN Greece. As Halstead has clearly established «... archaeozoological evidence of species composition, season of slaughter, body size, mortality and anatomical representation offers virtually no support for any of the three basic practices of recent pastoralists large-scale herding, favourable nutrition and specialization for exchange »78. Any model of the prehistoric economy looking to recent historical parallels or to those from the ethnographie present will inevitably need to factor out the population densities and large-scale markets, which condition the parallels. Nevertheless terms such as ‘agropastoralism’ although imprecise may be useful in indicating a range of possibilites79. In sub-Saharan Africa, for example, as one moves from areas dominated by low and highly variable rainfall, to high and more stable rainfall patterns, the local économies move from those dependent on fully mobile livestock, to systems associated with some form of crop production : « the more ‘sedenta— rised’ pastoralists of the southern Sahel... practice restricted seasonal movements within zones of 30 to 50 kilometres »80. The intention here is to restate the commonplace, that various systems of land-tenure, and various social devices to negotiate access to common resources, can exist between different communities competing for land in the same area.

fig. 8. Plot of the percentage of sheep and goats (out of total domesticates) against percentage of sheep (out of sheep + goat). Key : filled circles — cave sites ; open circles — ‘marginal’ open settlements of Kastri and Kephala ; squares — other open settlements (after Halstead 1996, 31 fig.2).

  • 81 Halstead 1996, esp. 31 fig. 2.
  • 82 Perevolotsky 1992.
  • 83 Primov, 1992, 56.
  • 84 Jullien 1981.
  • 85 Spitaels 1982.
  • 86 Lambert 1972.

40In his analysis of the archaeozoological data from prehistoric sites in Greece (fig. 8), Halstead81 has demonstrated that the bone assemblages from cave sites and ‘marginal’ open sites (largely LN/FN) are overwhelmingly dominated by sheep and goat. If a lesser proportion of sheep than at other open settlements, nevertheless sheep still form about 50% of the total. The choice of animais, which are raised in agropastoral communities, is gover— ned by a complex of factors82. In addition to economic, and environmental factors, and the needs of the animais themselves, a taste for mutton could also explain the high proportion of sheep83. Large numbers of sheep and goats do not, of themselves, imply seasonal movement. Indeed there is evidence to indicate that some of these communities were not very mobile. Jullien, in his report on the faunal assemblage from the Kitsos Cave refers to a « pourcentage notable de porcs traduisant un nomadisme faible des éleveurs »84, whilst the occurrence in cave sites of impediments such as large storage jars also argues against high mobility over long ranges. In suggesting that the occupants of certain LN sites were agropastoralists I wish, therefore, to avoid the implication of extended transhumance. All the same the location of cave sites such as Kitsos or Velika Kokora Troupi indicate a very strong emphasis on animais. And there is always a local context. To take a case from Attica rather than the Peloponnese, the cave at Kitsos is close to the metal-working and mining site at Thorikos85, and there may even have been N occupation on the bare and unpromising islet of Makronisos86, just across the water from Thorikos. Two or three sites within relatively easy reach of each other, whose resources, site catchments and archaeology indicate quite different if probably interrelated economies.

Economie pluralism ?

41If, as I have argued, the changes observed during this long time span are not easily understood in terms of ecological stress or population pressure, then a cultural explanation would seem to be demanded. A cultural value was placed on the holding and raising of animal stock, which had not been the case before. Secondary products, their storability and their exchangeability may well be important. We have had reason to stress the prominence in LN/FN assemblages of storage vessels and pithoi. These can be used to store cereals of course, but we should not forget the importance of storing cheeses and meat in oil or brine, and other methods of preserving perishable foods such as salting, pickling, smoking and drying.

  • 87 McCorkle 1992b.
  • 88 Halstead 1992.
  • 89 Halstead 1994, 207.
  • 90 I am thinking here of the Dirou caves whose ritual significance seems to extend beyond a merely lo (...)

42Spécialisation implies a sacrifice of self-sufficiency, which runs counter to our view that the peasant household aims to meet all its own needs. There may have been purely economic advantages for agropastoralists : animal husbandry and crop cultivation, although complementary in some respects, notoriously can compete with each other for labour and for land87. Small household herds are inefficient, they can demand attention when the fields need cultivating, and threaten damage to crops unless carefully contained. Larger flocks herded away from the arable are more effective : a large flock can still be controlled by just one shepherd, and given the low level of population and the small size of most LN/FN sites, labour may have been the major constraint on production. Nevertheless an important part of the story must lie in the relationship between the communities operating at different levels : the Neolithic sites of the MN tradition for example and the new colonising sites. The assumption in the past has been that an essential identity of community links the two, with the former representing the top of a hierarchy of sites. There are other possibilites, which might include semi-autonomous groups e.njoying a symbiotic relationship. Halstead has developed a model for the exchange of livestock between households for the Neolithic of Thessaly88, with its very dense concentration of villages. The effectiveness of this, he argues, was under— mined during the LN precisely by the colonisation of marginal lands89. Whilst this interpretation works well for Thessaly, it seems less easily applicable to S Greece. Exchange networks involving metals, obsidian, polished stone axes, stone vessels, millstones and ceramics are also marked features of LN-FN Greece. For these to work on a supra-regional scale, across the Balkans and the Aegean, they would also need to be effective on the local scale revealed, not least through regional survey. Could, then, one key to the LN/FN transformations in S Greece involve exchange between different communities ? These might be segmentary, each would have had a different emphasis in production, each may have had its own regional centres of cuit90, and each had access to the networks of exchange wirnessed by exotirs. No doubt these exchange mechanisms could help buffer the communities against poor harvests, though the mobility of stock would offer an alternative safety net, namely to up and move to better pastures in a land that was still thinly populated.

  • 91 Lee 1979, esp. 56-7, 76-86, 401-31 ; this is hardly the place to discuss the revisior.ist views of (...)

43Such exchanges need not depend on the existence of large centres of population in the region. Indeed it is not uncommen even for hunter— gatherer communities to develop trading and client relationships with nearby farmers or ranchers. A classic example is the ! Kung San who traded with the Bantu, operated cattle-posts under the Tswana mafisa system, supplied labour to such posts, and worked as trackers to hunting groups ; only later came work in the mines and the militarization of these groups91.

44Lee has also pointed out how such relationships can be incorporated into the continuai adaptations of subsistence strategy to years of low rainfall and to its marked local variability.

  • 92 Reddy 1997.

45All researchers would agree, at least, that many questions remain to be resolved, and the excavation of a settlement site with good contextualised evidence including faunal and other environmental data is an important next priority. Reddy's classic case study of the relationship between agriculture and pastoralism in the Late Harrapan of Gujarat has indicated just how valuable palaeoethnobotanical reconstruction can be in investigating this type of question92.

46William G. CAVANAGH

47Department of Archaeology

48University of Nottingham

49Nottingham, NG7 2RD Royaume-Uni

50e-mail : mailto:Bill.Cavanagh@nottingham.ac.uk

Bibliographie

Bibliography

Atherden M., Hall Jean, and Wright James c. 1993. « a pollen diagram from the northeast Peloponnese, Greece : implications for vegetation history and archaeo— logy », The Holocene 3 : 351-356.

Alram-Stern Eva, 1996. Die ägäische Frühzeit. 2 Serie : Forschungsbericht 1975-1993. 1. Band Das Neolithikum in Griechenland mit Ausnahme von Kreta und Zypern. Vienna.

Blackman D., 1997. « Archaeology in Greece 1996-97 », Arch. Repts. 43 : 1-125.

Blegen C.W, 1937. Prosymna : The Helladic Settlement Preceding the Argive Heraeum. Cambridge, Mass.

Bogucki P., 1988. Forest Farmers and Stockherders : Early Agriculture and its Conse— quences in North-Central Europe. Cambridge.

Bottema S., 1990. « Holocene environment of the Southern Argolid : a pollen core from Kiladha Bay » in T. J. Wilkinson and S. T. Duhon, Franchthi Paralia : the Sediments, Stratigraphy, and Offshore Investigations (Excavations at Franchthi Cave, fasc. 6) : 117-138.

Bottema S., 1994. « The prehistoric environment of Greece : a review of the palyno— logical record » in Kardulias 1994 : 45-68.

Bottema S., Entjes-Nieborg G., and van Zeist W (eds), 1990. Man's Role in the Shaping of the Mediterranean Landscape. Rotterdam.

Carter T. and Ydo M., 1996. « The chipped and ground stone » in Cavanagh et al 1996 : 141-182.

Cavanagh WG., Crouwel J., Catling R., and Shipley G., 1996. Continuity and Change in a Greek Rural Landscape : The Laconia Survey II. London.

Cherry J., 1981. « Pattern and process in the earliest colonisation of the Mediterranean islands », PPS 47 (1981) 41-68.

Cherry J., 1988. « Pastoralism and the role of animais in the pre— and protohistoric economies of the Aegean » in C. R. Whittaker (ed.) Pastoral Economies in Classical Antiquity : 6-34. Cambridge Philiological Society Suppl. Vol. 14. Cambridge.

Cherryj„ 1990 « The first colonization of the Mediterranean islands : a review of recent research », JMA 3 : 145-221.

Coleman J. E., 1977. Keos i. Kephala. A Late Neolithic Settlement and Cemetery. Princeton.

Coleman J. E., 1992. « Greece, the Aegean » in R. W Ehrich (ed), Chronologiesin Old World Archaeology (third edition) 1 : 247-288, II : 203-221. Chicago.

Diament S., 1974. The Later Village Farming Stage in Southern Greece. PhD Dissertation, University of Cincinnati.

Evans J. D., 1971. « Neolithic Knossos — the growth of a settlement », PPS 37 : 267-276.

Forbes H., 1995. « The identification of pastoralist sites within the context of estatebased agriculture in ancient Greece : beyond the ‘transhumance versus agropastoralism’ debate », BSA 90 : 325-338.

Forsén J., 1996. « Prehistoric Asea revisited », Opuscula Atheniensia 21 : 41-72.

Forsén J., Forsén B., and Lavento M., 1996. « The Asea Valley Survey : a preliminary report of the 1994 season », Opuscula Atbeniensia 21 : 74-97.

Halstead P., 1981. « Counting sheep in Neolithic and Bronze Age Greece » in i. Hodder, G. Isaac and N. Hammond (eds.), Pattern of the Past : Studies in Honour of David Clarke : 307-339.

Halstead P., 1992. « From reciprocity to redistribution : modelling the exchange of livestock in Neolithic Greece », Anthropozoologica 16 : 19-30.

Halstead P., 1994. « The North-South divide : regional paths to complexity in Prehistoric Greece » in C. Mathers and S. Stoddart (eds.), Development and decline in the Mediterranean 195-219. Sheffield.

Halstead P., 1996. « Pastoralism or household herding ? Problems of scale and specialization in early Greek animal husbandry », World Archaeology 28 : 20-42.

Holmberg E. J., 1944. The Swedish Excavations at Asea in Arcadia. Lund.

Hubbard R. N. L. B., 1995. « Fallow deer in prehistoric Greece, and the analogy between faunal spectra and pollen analyses », Antiquity 69 : 527-538.

Immerwahr S. A., 1971. The Athenian Agora XIII : the Neolithic and Bronze Ages. Princeton.

Jacobsen T. W, 1973. « Excavations at Franchthi Cave, 1969-1971. Part 2. », Hesperia 42 : 253-283.

Jahns S., 1990. « Preliminary notes on human influence and the history of vegetation in S. Dalmatia and S. Greece » in Bottema et al. 1990 : 333-340.

Jahns S., 1993. « On the Holocene vegetation history of the Argive plain (Peloponnese, Southern Greece) », Vegetation History and Archaeobotany 2 : 187-203.

James P. A., Mee C. B., and Taylor G. J., 1994. « Soil erosion and the archaeological landscape of Methana, Greece « Journal of Field Archaeology 21 : 395-416.

Jameson M. H., Runnels C. N., and van Andel T. H., 1994. A Greek Countryside : the Southern Argolidfrom Prehistory to the Present Day. Stanford.

Johnson M., 1996. « The Berbati-Limnes Archaeological Survey. The Neolithic period » in Wells and Runnels 1996 : 37-73.

Jullien R., 1981. « La faune des vertébrés à l'exclusion de l'homme, des oiseaux et des poissons » in Lambert 1981 : 569-590.

Kardulias P. N. (ed.), 1994. Beyond the Site : Regional Studies in the Aegean Area. Lanham.

Katsarou S. and Sampson A., 1939. « H ανασκαφκή έρευνα στο Σπήλαιο των Λιμών στα Καστρια Καλαβρυτών », AAA 22 : 161-170.

Korres G., 1990. « Excavations in the region of Pylos » in J.-P. Descoeudres (ed.), Eumousia : Ceramic and Iconographie Studies in Honour of Alexander Cambitoglou (MedArch Supplément) : 1-11.

Kraft J. C., Rapp G. R., and Aschenbrenner S. E., 1980. « Late Holocene palaeo— geomorphic reconstructions in the area of the Bay of Navarino : Sandy Pylos », Journal of Archaeological Science 7 : 187-210.

Lambert N., 1972. « Vestiges préhistoriques dans l'île de Makronissos », BCH 96 : 873-81.

Lambert N., 1981. La Grotte préhistorique de Kitsos (Attique). I : Missions 1968-1978. Recherche sur les grands civilisations, Synthèse 7. Ecole française d'Athènes.

Lee R. B., 1979. The ! Kung San : Men, Women and Work in a Foraging Society. Cambridge.

Lewthwaite J., 1981. « Plain tails from the hills : transhumance in Mediterranean archaeology » in A. Sheridan and g. Bailey (eds.), Economic Archaeology [BAR International Series 96] : 57-66.

Lewthwaite J., 1984. « Pastore, Padrone : the social dimensions of pastoralism in prenuragic Sardinia » in W H. Waldren, r. Chapman, J. Lewthwaite and r.-c. Kennard (eds.), The Deya Conference on Prehistory : Early Settelement in the Western Mediterranean Islands and their Periferal Areas I [BAR International Series 202] : 251-268.

Lolos Y. G., 1994. Пύλος ημαθόεις η πρωτέυουσα τοι Νέστορος και η γύρω περιοχή Athens.

McCorkle C., 1992a. Plants, Animais and People : Agropastoral Systems Research. Boulder, Westview Press.

McCorkle C., 1992b. « The agropastoral dialectic and the organization of labor in a Quecha community » in McCorkle 1992a : 77-97.

McCormick F., 1992. « Early faunal evidence for dairying », OJA 11 : 201-209.

McDonald w a. and Rapp g. r. (eds.), 1972. The Minnesota Messenia Expedition. Minneapolis.

McDonald W A. and Wilkie N. C., 1992. Excavations at Nichoria in Southwest Greece II : The Bronze Age Occupation (Minneapolis).

Mee C. B. and Forbes H. (eds), 1997. A Rough and Rocky Place : the landscape and settlement history of the Methana Peninsula, Greece. Liverpool.

Moody J., Rackham O., and Rapp G., 1996. « Environmental archaeology of prehistoric NW Crete » Journal of Field Archaeology 23 : 273-297.

Papathanassopoulos G. A., 1971a. « Dirou Cave : the excavations of 1970-71 », AAA 4 : 12-26.

Papathanassopoulos G. A., 1971b. « Dirou Cave », AAA 4 : 149-153.

Papathanassopoulos G. A., 1971c. « Dirou Cave », AAA 4 : 289-303.

Papathanassopoulos G. A., (ed.) 1996. Neolithic Culture in Greece. Athens.

Perevolotsky A., 1992. « Integration versus conflict : crops and goats in Piura, Peru » in McCorkle 1992a : 23-50.

Primov G., 1992. « The role of goats in agropastoral production systems of the Brazilian Sertào » in McCorkle 1992a : 51-75.

Pullen D„ 1992. « Ox and plow in the EBA Aegean >, AJA 96 : 45-54.

Rackham O., 1990. « The greening of Myrtos » in Bottema et al. 1990 : 341-348.

Rackham O. and Moody J., 1996. The Making of the Cretan Landscape. Manchester.

Reddy S. N., 1997. « If the threshing floor could talk : integration of agriculture and pastoralism during the Late Harrapan in Gujarat, India », Journal of Anthropo— logical Archaeology 16 : 162-187.

Renard J., 1989. Le site Néolithique et Helladique Ancien de Kouphovouno (Laconie) : Fouilles de 0-W von Vacano (Aegaeum 4) (Liège).

Renfrew c, 1972. The Emergence of Civilization : the Cyclades and the Aegean in the Third Millennium BC. London.

Säflund G., 1965. Excavations at Berbati 1936-37. Stockholm.

Sampson A., 1982. « Σπήλαιο τοῦ Νέστσρος », PAE : 175-187.

Sampson A., 1984. « The Neolithic of the Dodecanese and Aegean Neolithic culture », BSA 79 : 239-249.

Sampson A., 1987. H Νεολιθική περίοδος στα Δωδεκάνιησα. Athens.

Sampson A., 1988a. « Periodic and seasonal usage of two Neolithic caves in Rhodes » in S. Dietz, and i. Papachristodoulou (eds.), Archaeology in the Dodecanese : 11-16. Copenhagen.

Sampson A., 1988b. H Νεολιθική Κατοίκιση στο Ґυαλί της Νισρου. Athens.

Sampson A., 1992. « Late Neolithic remains at Tharrounia, Euboea : a model for the seasonal use of settlements and caves », BSA 87 : 61-101.

Sampson A., 1996. « La grotte du Cyclope : un abri de pêcheurs préhistoriques ? », Archéologia 328 : 54-59.

Sherratt A., 1981. « Plough and pastoralism : aspects of the secondary products revolution » in N. Hammond, I. Hodder and G. Isaac (eds.), Pattern of the Past : 261-305. Cambridge.

Sherratt A., 1994. « The transformation of early agrarian Europe : the Later Neolithic and Copper Ages, 4°500-2°500 BC » in B. Cunliffe (ed.), The Oxford Illustrated Prehistory of Europe : 167-201.

Shott R. J., 1992. « On recent trends in the anthropology of foragers : Kalahari revisionism and its archaeological implications », Man, the Journal of the Royal AnthropologicalInstitute N.S. 27 : 843-871.

Spitaels P., 1982a. « Final Neolithic pottery from Thorikos » in P. Spitaels (ed.), Studies in South Attica I (Miscellanea Graeca 5) : 8-44. Gent.

Touchais G., 1980. « La céramique néolithique de l'Aspis », Études Argiennes : BCH Supplément 6 : 1-40.

Touchais G., 1981. « L'antre Corycien I. Le matériel néolthique », BCH Supplément 7 : 95-257.

Treuil R, Darcque P., Poursat J.-C., Touchais G, 1989. Les civilisations Égéennes. Paris.

Treuil R., 1983. Le Néolithique et le Bronze Ancien égéens : les problèmes stratigraphiques et chronologiques, les techniques, les hommes. Bibliothèque des Écoles Françaises d'Athènes et de Rome 248. Paris.

Valmin N. S., 1938. The Swedish Messenia Expedition. Lund.

van Andel T. and Runnels C., 1995. « The earliest Farmers in Europe », Antiquity 69 : 481-500.

Van Andel T., Zangger E., and Demitrack a., 1990. « Land use and soil erosion in prehistoric and historical Greece »,Journal of Field Archaeology 17 : 379-396.

Warren R and Hankeyv, 1989.Aegean Bronze Age Chronology. Bristol.

Weisshaar H.-J., 1979a. « Ausgrabungen auf der Pevkakia-Magula und der Beginn der frühen Bronzezeit in Griechenland », Archäologisches Korrespondenzblatt 9 : 385-392.

Weisshaar H.-J., 1979b. « Nordgriechischer Import im kupferzeitlichen Thessalien », JRGZM 26 : 114-130.

Wells B. and Runnels C., 1996. The Berbati-Limnes Archaeological Survey 1988-1990. Acta Instituti Atheniensis Regni Suecae vol. 44. Stockholm.

Wright H. E„ 1972. « Vegetation history » in McDonald et al. 1972 : 188-199.

Wright J. C., Cherry J. F., Davis J. L„ Mantzourani E., Sutton S. B., and Sutton R. F., 1990. « The Nemea Valley Archaeological Project : a Preliminary Report », Hesperia 59 : 579-659.

Zangger E., 1992. « Assessing the natural resources for agriculture » in B. Wells (ed.), Agriculture in Ancient Greece. Skrifter utgivna av Svenska Institutet i Athen, quarto series, 42 : 13-19.

Zangger E., 1993. The Geoarchaeology of the Argolid (Deutsches Archäologisches Institut Athen, Argolis II). Berlin.

Zachos C., 1987. Ayios Dhimitrios, a Prehistoric Settlement in the Southwestern Pelopon— nesos : the Neolithic and Early Helladic Periods. PhD dissertation, University of Boston.

Zachos C., 1996a. « Metallurgy » in Papathanassopoulos 1996 : 140-143.

Zachos C., 1996b. « The Western Peloponnese » in Papathanassopoulos 1996 : 78-79.

Notes

1 Acknowledgements : I wish to thank Josette Renard for the invitation to deliver this paper, Chris Mee, Catherine Perlés, Jim Roy, Gilles Touchais for helpful comments, and Hamish Forbes for reading and improving the manuscript and saving it from a number of errors. Figures 1 and 2 are by David Taylor, to whom I am also very grateful

2 Halstead 1981, 326-7 ; cf. also 1994

3 Zachos 1996a

4 E.g. Kephala Coleman 1977, esp. 106, pl. 67 ; Alepotrypa, Papathanassopoulos 1996, 223 pl. 30

5 Blackman 1997. 73

6 Cherry 1981, 1990

7 Sampson 1984, 1987, 1988a, 1992, 1996

8 See recently Weisshaar 1979a and b, Warren and Hankey 1989, Coleman 1992, Perlès and Demoule 1993, Sampson 1989, Zachos 1987

9 Treuil et al. 1989, 127

10 Treuil ét al. 1989, 112 Table 1 ; Demoule and Perlès 1993, 366 fig. 2

11 Renard 1989

12 Runnels et al. 1995, 7, 89-91 ; Jameson et al. 1994, 435-6, 468, 471, 477, 482-3, 488, 510, 521, 522, 534

13 Op. cit. 522

14 Jameson et al. 1994, 346

15 Op. cit. 347.

16 Jameson et al.1994, 232 fig.4.8,341-8

17 Op.cit 346-7

18 Bottema 1990

19 Bottema 1990, 123-4

20 Moody et al. 1996, 291, referring lo NW Crete

21 Jahns 1993

22 Jahns 1993, 200.

23 Jahns 1993, 200.

24 Moody et al. 1996 : 286.

25 van Andel et al. 1990, 382.

26 Johnson 1996, 37.

27 Johnson 1996, 63.

28 Wells et al. 1990, 222 ; Demitrack et al 1990, 384 ; van Andel et al. 1990, 384 ; Zangger 1992, 17.

29 Zangger 1992, 19.

30 Rackham 1990, 341.

31 Rackham and Moody 1996, 18-24.

32 Forsén et al. 1996, 85.

33 Forsén et al. 1996, 88.

34 Forsén 1996.

35 Forsén et al. 1996, 83.

36 Op. cit. 85.

37 I thank Jim Roy for this point.

38 Davis et al. 1997, 417, 438-9.

39 Treuil 1983, 44-5 ; McDonald et al. 1972, 130-1.

40 Kraft et al. 1980 ; Zangger et al. 1997, 559.

41 Wright 1972.

42 Kraft et al. 1980.

43 See also Zangger et al 1997, 588.

44 Zangger et al. 1997, 588 zone OS I.

45 Op. cit. 589.

46 MS9. 40 and MS61. 35 may be the tangs of tanged points, Mee et al. 1997, 176 fig. 11. 40 21 and 191 fig. 11. 55 9 ; see also MS56. V35, op. cit.. 188 fig. 11. 52 8, MS63. 07 and MS64. V45, op. cit.. 191 fig. 11. 55 20-1.

47 Wright et al. 1990, 609.

48 Cherry et al. 1988, 174.

49 Atherden et al 1993.

50 Atherden et al. 1993, 355.

51 E.g. Zangger et al. 1997, 568.

52 Note, for example, the MN site discovered by Zangger in the Argive floodplain : Zangger 1992, 17.

53 Cf. van Andel and Runnels 1995, 487.

54 E.g. Cherry et al. 1988, Whitelaw 1991, James et al. 1994, Zangger et at. 1997, esp. 569-76.

55 Mee et al. 1997, 25.

56 Zangger et al. 1997, 568.

57 James et al. 1994, 414.

58 E.g. Bottema 1990, 124 ; Moody et ai 1994, 287-8.

59 Bottema 1994, 56.

60 Whitelaw 1991, esp. 207.

61 Evans 1971.

62 Cherry 1988, 23 ; Jameson et al. 1994, 343 ; Halstead 1994, 211 ; Zachos 1996b.

63 Diament 1974, 363-6.

64 Halstead 1981, 326.

65 Sampson 1988a.

66 Halstead 1981, van Andel and Runnels 1995.

67 In summary Sherratt 1994, 170-1 ; Sherratt 1981 ; Bogucki, 1988, esp. 85-8, has argued that secondary products of cows were exploited during the primary Neolithic in N Europe (that is to say a period approximately contemporary with the LN in Greece), though see also McCormick 1992. Cows formed a much smaller component in the Neolithic economy of S Greece, and their husbandry is not well understood. Thanks to Hamish Forbes who drew my attention to this point.

68 Pullen, 1992.

69 Van Andel and Runnels 1987, 70-1 ; Jameson et al. 1994, 343-8.

70 Van Andel and Runnels 1995.

71 With reference to the Classical period in Greece, Forbes 1995, 329.

72 Halstead 1981, 1986 ; Cherry 1988.

73 Halstead 1994, 196.

74 Hubbard 1995.

75 1986, 102-3.

76 Papathanassoupoulos 1996, 39-40 ; Sampson 1988a.

77 E.g. Chang and Koster 1986, 112-4.

78 Halstead 1996, 32.

79 See for example the various examples in McCorkle 1992.

80 Van den Brink et al. 1995, esp. 382-3, 385.

81 Halstead 1996, esp. 31 fig. 2.

82 Perevolotsky 1992.

83 Primov, 1992, 56.

84 Jullien 1981.

85 Spitaels 1982.

86 Lambert 1972.

87 McCorkle 1992b.

88 Halstead 1992.

89 Halstead 1994, 207.

90 I am thinking here of the Dirou caves whose ritual significance seems to extend beyond a merely local reach.

91 Lee 1979, esp. 56-7, 76-86, 401-31 ; this is hardly the place to discuss the revisior.ist views of K ;lahari foragers, see the review Shott 1992, though they can only strengthen an argument which looks to the symbiotic relationships between différent cultures.

92 Reddy 1997.

Table des illustrations

Légende fig 1 Map of Late and Final Neolithic sites in the Peloponnese (see appendix for liste of sites) By David Taylor.
URL http://books.openedition.org/pur/docannexe/image/20606/img-1.png
Fichier image/png, 260k
Légende fig 2 Map of LN/FN sites from the Laconia Survey area and major soil types.By David Taylor.
URL http://books.openedition.org/pur/docannexe/image/20606/img-2.png
Fichier image/png, 399k
Titre Table 1 : FN/EHI find spots from Laconia Survey
URL http://books.openedition.org/pur/docannexe/image/20606/img-3.png
Fichier image/png, 145k
Légende Fig. 3 a. Maps of Final neolithic sites from the area of the Southern Argolid Survey. After Jameson et al. 1994, 233 fig. 4.10.
URL http://books.openedition.org/pur/docannexe/image/20606/img-4.png
Fichier image/png, 386k
Légende Fig. 3 b. Maps of Final neolithic sites from the area of the Southern Argolid Survey. After Jameson et al. 1994, 233 fig. 4.11.
URL http://books.openedition.org/pur/docannexe/image/20606/img-5.png
Fichier image/png, 404k
Légende Fig. 4.Map of neolithic findsports, Berbati Limnes, Survey After Johnson 1996, 39 fig. 2.
URL http://books.openedition.org/pur/docannexe/image/20606/img-6.png
Fichier image/png, 343k
Légende Fig 5 Map of neolithic sites from Asea Valley Survey After Forsèn et al, 1996 86 fig 11.
URL http://books.openedition.org/pur/docannexe/image/20606/img-7.png
Fichier image/png, 239k
Légende fig. 6 Map of Early Helladic sites from the Methana Survey After Mee et al. 1997, 44 fig. 4.1.
URL http://books.openedition.org/pur/docannexe/image/20606/img-8.png
Fichier image/png, 286k
Légende fig. 7 Distribution of neolithic finds from the Nemea region After Cherry et al 1988 173 fig. 12.
URL http://books.openedition.org/pur/docannexe/image/20606/img-9.png
Fichier image/png, 603k
Légende fig. 8. Plot of the percentage of sheep and goats (out of total domesticates) against percentage of sheep (out of sheep + goat). Key : filled circles — cave sites ; open circles — ‘marginal’ open settlements of Kastri and Kephala ; squares — other open settlements (after Halstead 1996, 31 fig.2).
URL http://books.openedition.org/pur/docannexe/image/20606/img-10.png
Fichier image/png, 49k

© Presses universitaires de Rennes, 1999

Conditions d’utilisation : http://www.openedition.org/6540

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search