The exceptional density of long-term Neolithic settlements in Thessaly was recognized from the beginning of the century. In the decades that followed, successive surveys gradually increased the number of sites identified, bringing the total to more than 300. In 1984, Halstead exploited an already impressive corpus and offered the first thorough analysis of settlement patterns in Thessaly (Halstead 1984). Although the details of his analyses have remained unpublished, the main diachronic and synchronic conclusions can be found in several papers (e.g., Halstead 1977, 1981a, 1989a, 1989b, 1994, 1995). Not much can presently be added to his analyses of western Thessaly, the Karditsa plain. In the meantime, however, Gallis had resumed surveys in eastern Thessaly with his collaborators, leading to the publication of a systematic ‘Atlas of prehistoric settlements in eastern Thessaly’ (the ATAE). This included several newly discovered sites and refined chronological attributions, as well as various statistics on the chronological distribution of sites, the duration of their occupation, their size, etc. (Gallis 1992).

Relying on Gallis' Atlas and recent geomorphological fieldwork (van Andel et al. 1995), van Andel and Runnels published another study that concentrated on the palaeo flood-plains of the Larissa basin. Though more restricted in scope than the previous analyses, its conclusions differed and were the basis of important theoretical developments on the causes and dynamics of the Neolithic expansion (van Andel and Runnels 1995).

According to some recent essays on the Neolithic, the latter should be viewed primarily as a mental, symbolic mutation: a new conception of the supernatural world, a different way of interacting with ‘divinities’ (Cauvin 1997; Hodder 1990). Whether or not this ‘symbolic revolution’ is considered as a driving force, it cannot be denied that the technical, economic and social transformations were sustained by transformations in the symbolic and ritual domains. The latter are often expressed in spectacular ways: the early PPNB monumental buildings such as the ‘Skull building’ at Çayönü (Özdoğan and Özdoğan 1990), the ‘temples’ with greater than life-size anthropomorphic pillars at Nevali Çori (Hauptmann 1993), the 8-metre high tower of Jericho (Kenyon 1957). These early collective monuments are echoed, in later contexts, by the equally monumental graves and megalithic buildings of the Atlantic façade. In parallel, the highly structured representations on the house walls at Çatal Hüyük demonstrate that this ‘symbolic revolution’, whatever its exact interpretation (e.g., Cauvin 1997; Forest 1993; Hodder 1990), had also penetrated within the more private, domestic sphere.

Yet, no equivalent of these spectacular symbolic expressions has thus far been found in the Neolithic of Greece. If a spiritual mutation is to be recognized, it can possibly be sought in the profusion of small, ‘non-utilitarian’ objects – miniatures, models, ornaments, precious stone artefacts, incised clay tablets, and so forth – that characterize the Neolithic of Greece and that of its Near Eastern and Balkanic counterparts.

An elusive Mesolithic: absence of evidence or evidence of absence?

Here, I shall use the term Mesolithic in its chronological sense, to designate early Holocene hunter-gatherer assemblages. The period under consideration spans between c. 9500 and 8000 BP uncalibrated, or c. 8700 to 7000 BC in calendar years. Detailed data-oriented presentations have been offered elsewhere (Perlès 1990a, 1995; Runnels 1995), so I shall focus on issues directly relevant to the problem of the origins of the Neolithic.

The most salient characteristic of the Mesolithic in Greece is how poorly it is known, and how few sites are recorded. Diverging opinions about the significance of this scarcity have led to opposing views on the origins of the Neolithic in Greece. I shall argue that Mesolithic Greece was indeed sparsely populated, and that this low demography rules out the hypothesis of a purely indigenous shift to agriculture.

The few sites known to date concentrate in two main regions: north-east Attica and the Argolid, in eastern Greece, Corfu, the coastal plains of the Acheron and the Preveza region in north-western Greece (fig. 2.1). So far, only four sites have been excavated and published: Sidari in Corfu, Franchthi in the southern Argolid, Zaïmis in Attica and Ulbrich also in the Argolid. The important site of Theopetra, in Thessaly, with a sequence spanning the Palaeolithic, Mesolithic and Neolithic, is currently under excavation by N. Kyparissi-Apostolika.

Paradoxically, even if the Near Eastern origin of several domesticated species is now well established, the precise origin of the farmers themselves remains as elusive as ever. No satisfactory link has been established with any specific region of the Near or Middle East, and the most obvious candidate, western Turkey, has not yet provided evidence of Neolithic settlements as early as those of Greece.

The ‘random’ parallels between Greece and the Near East

Depending on whether one envisions a rapid or a slow movement of expansion, the reference sites for comparisons differ. According to the model of Ammerman and Cavalli-Sforza, western Anatolia and Greece belong to the same isochron. This would imply a roughly simultaneous expansion of farming groups in the two areas (Ammerman and Cavalli-Sforza 1984). Given the dates of the earliest Neolithic sites in Greece, the best comparenda should thus belong to the Final Pre-Pottery Neolithic B (PPNB) and the Early Pottery Neolithic of the Near East, that is, to phase 5 (6900–6400 cal BC) in the general chronological scheme of Aurenche et al. (1987; see also Cauvin 1985).

Yet, if the spread of farming groups were a slow movement, it would be more appropriate to turn to earlier sites, of the late phase 4 (late PPNB, c. 7600–6900 BC). The Late PPNB was indeed qualified by Cauvin (1989: 19; 1994: 107ff.) as the period of the ‘great exodus’, when Initial Neolithic farming groups started to migrate out of the ‘nuclear zone’ and to colonize the Taurus, central Anatolia, the deserts and the temperate Mediterranean zone.