Unconformist Archaeology: Papers in honour of Paolo Biagi 9781407311463, 9781407341163

Table of contents :
Cover
Title Page
Copyright
Contents
Preface
Part 1: Italy
Chapter 1: RAPID RIVERS, SLOW FORESTS:
EARLY NEOLITHIC CHARCOAL FROM NORTHERN ITALY
Part 2: Europe
Chapter 2: MESOLITHIC OBSIDIAN NETWORKS IN THE AEGEAN
Chapter 3: CLIMATE CHANGES AND SOCIAL RESPONSES IN THE EARLY PREHISTORY OF
GREECE: A COMPLEX AND DYNAMIC INTERACTION
Chapter 4: THE 1965–1968 EXCAVATIONS AT SCHELA CLADOVEI (ROMANIA) REVISITED
Chapter 5: THE NEOLITHIC TRANSITION IN EUROPE:
SOME COMMENTS ON GAPS, CONTACTS, ARRHYTHMIC MODEL, GENETICS
Chapter 6: REMAINS OF A SPECIAL ‘PERSONALITY’ FROM THE COPPER AGE
OF THE EASTERN CARPATHIAN BASIN
Chapter 7: PASS THE SKELETON KEY… ANIMALS IN AN EARLY COPPER AGE INHUMATION
BURIAL FROM PUSZTATASKONY–LEDENCE I, HUNGARY
Chapter 8: SMALL BUT SPECIAL: THE ISLAND OF PALAGRUŽA IN THE 3rd MILLENNIUM BC
ADRIATIC
Chapter 9: THE BIRTH OF THE HERO
THE EMERGENCE OF A SOCIAL TYPE IN THE 4th MILLENNIUM BC
Chapter 10: THE SIBYLLINE BOOK OF EUROPEAN ETHNOARCHAEOLOGY
Part 3: Asia
Chapter 11: THREE END-SCRAPERS FROM MOHENJO-DARO (PAKISTAN)
Chapter 12: MULTIPLE TECHNOLOGICAL CHOICES FOR THE CERAMIC BANGLES OF THE
EARLY AND MATURE INDUS CIVILIZATION (PAKISTAN)

Citation preview

BAR S2528 2013

Unconformist Archaeology Papers in honour of Paolo Biagi

STARNINI (Ed)

Edited by

Elisabetta Starnini

UNCONFORMIST ARCHAEOLOGY

B A R Starnini 2528 cover.indd 1

BAR International Series 2528 2013 18/06/2013 13:24:39

Unconformist Archaeology Papers in honour of Paolo Biagi

Edited by

Elisabetta Starnini

BAR International Series 2528 2013

ISBN 9781407311463 paperback ISBN 9781407341163 e-format DOI https://doi.org/10.30861/9781407311463 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

CONTENTS

Preface

iii

Italy 1. Rapid Rivers, Slow Forests: Early Neolithic charcoal from northern Italy Renato Nisbet

3

Europe 2. Mesolithic Obsidian Networks in the Aegean Małgorzata Kaczanowska, Janusz K. Kozłowski

17

3. Climate Changes and Social Responses in the Early Prehistory of Greece: a complex and dynamic interaction Nikos Efstratiou

27

4. The 1965–1968 Excavations at Schela Cladovei (Romania) Revisited Adina Boroneanț, Clive Bonsall

35

5. The Neolithic Transition in Europe: some comments on gaps, contacts, arrhythmic model, genetics Jean Guilaine

55

6. Remains of a Special ‘Personality’ from the Copper Age of the Eastern Carpathian Basin Pál Raczky

65

7. Pass the Skeleton Key… Animals in an Early Copper Age inhumation burial from Pusztataskony–Ledence I, Hungary László Bartosiewicz, Zsófia Eszter Kovács, Balázs Farkas

77

8. Small but Special: the island of Palagruža in the 3rd millennium BC Adriatic Stašo Forenbaher

89

9. The Birth of the Hero: the emergence of a social type in the 4th millennium BC Svend Hansen

101

10. The Sibylline Book of European Ethnoarchaeology John Nandriş

113

Asia 11. Three End-scrapers from Mohenjo-Daro (Pakistan) Massimo Vidale, Anna Siviero, Giancarlo Sidoti, Giuseppe Guida, Gianfranco Priori

119

12. Multiple Technological Choices for the Ceramic Bangles of the Early and Mature Indus Civilization (Pakistan) Michela Spataro

131

i

UNCONFORMIST ARCHAEOLOGY

Top: August 1990, exploration of the Kara Dağ, Hakkâri Province, Turkish Kurdistan. Photos by E. Starnini. Bottom left: taking some rest on the excursion to Laghetti di Bruffione (Brescia Province, Italy). Photo by E. Starnini, July 2010. Bottom right: Paolo Biagi examining a soil profile near the Alpine Mesolithic site of Laghetti del Crestoso (Brescia Province, Italy). Photo by E. Starnini, July 2011.

ii

PREFACE

From time to time God causes men to be born – and thou art one of them – who have a lust to go abroad at the risk of their lives and discover news [...]. These souls are very few; and of these few, not more than ten are of the best. [Rudyard Kipling, Kim, Chapter 9] I knew about Paolo Biagi from his scientific writings in the early 1980s, when I was a young BA student at the University of Genova, well before meeting him personally. At that time, in my opinion, that of a beginner, he was already one of the very few prehistoric archaeologists in Italy able to write convincing papers in that field. I was soon to be very lucky indeed, because the year after, in 1981 at the age of 33, Paolo was appointed Researcher in Prehistory at my Faculty, and I could benefit first hand from his teachings and tutoring for my BA and MA dissertations. At that time, he was one of the few young scholars in our Universities not only to have a PhD degree, but also to have received it from a foreign University. In fact, as his close friends know well, he studied for his doctorate at University College, London, and he was always very proud to tell us that HRH The Princess Anne in person presented him with his diploma. I was very sorry for myself, yet happy for him when, in 1988, he left Genova to take up the position of Associate Professor in Palaeoethnology at Ca’ Foscari University of Venice, where he continued his career becoming eventually, in 2002, full Professor in Prehistory and Protohistory, and where he still works. However, our cooperation did not end; we have continued working together on several projects in spite of the distance separating Genova and Venice. Nemo propheta acceptus est in patria sua, quoting the Evangelists of the New Testament: nothing is truer of Paolo Biagi who, as John Nandris (Paolo’s PhD supervisor at UCL) reminds us in his paper in this volume, in 2009 was elected as an Honorary Fellow of the Society of Antiquaries of London, founded in 1717 as the second oldest learned society in Britain after the Royal Society, ‘in recognition of his original work in the field, studying the environmental contexts and lithic technologies of a wide range of human culture, from Europe to the Indian sub-continent’. Time passes in Italy but things are slow and difficult to change: in the Italian academic top positions, there are still people without a PhD, in spite of the fact that our University system desperately needs internationalization and the capability to gather funds and students on the international ‘market’. Actually, Paolo Biagi is one of the few professors in our field who encourages students to open their minds and go abroad to widen their knowledge and experience. For many years now, his research activity has been conducted almost exclusively abroad and covers many topics that are the focus of current international scientific debate.

iii

Paolo is the author of more than 250 papers and books, many of which are written in English and published abroad. He is, without doubt, one of the few outstanding Italian personalities working in prehistory and enjoys an international reputation for his research. Outstanding not only for the quality and quantity of his tireless scientific activity in his chosen discipline, but also – as echoed in the title I have chosen for this book – for his honest and direct approach to research. He consciously and deliberately refuses to conform to fashionable theories and short-lived trends that from time to time invade archaeology, firmly rejecting any mystification of data used for the interpretation of the past. This attitude has undoubtedly earned him many enemies and made it more difficult to obtain funding for his research projects. But as the saying goes, ‘many enemies, much honour’. Indeed, I learned a great deal from his nonconformism, which has served to enhance my appreciation of his scientific personality and has acted as a stimulus orienting my personal research ethic and approach to scientific activity. If he has a ‘weakness’, it is his readiness to trust those who appear to share his great enthusiasm for his subject, especially someone who approaches him proposing new collaborative scientific enterprises. This has caused him many disappointments and delusions, yet reveals his trusting nature and innate curiosity, a reflection of his openness to new stimuli accompanied by a fearlessness and intellectual curiosity about new ideas. The photographs I have chosen for the book cover and opposite depict Paolo’s favourite environment, that of the high elevations of the mountains, and reflect his soul and his particular personality. This honorary volume brings together papers by his closest, most sincere and longstanding friends, former students and colleagues who have shared with him research experiences on different geographical regions and topics. I extend my warmest thanks to all of those who accepted, with great enthusiasm, the invitation to write a tribute for his 65th birthday. The papers cover almost all the fields of Paolo Biagi’s scientific activity, extending from the prehistory of northern Italy, the Adriatic and Balkans to the Indus Valley, including some hot topics such as the Mesolithic-Neolithic transition and the Neolithization of the Mediterranean Basin. Several contributions are more philosophical and theoretical, and offer an original view of the human past. I owe a debt of gratitude to my good friend, Barbara Voytek, who corrected the English of those papers written by non-native speakers. At the beginning, she and I had

intended to write a joint contribution for the book, but in the end we realized that this would not be possible without causing further delay to the enterprise; regretfully, therefore, we gave up on the idea. Nevertheless, I do hope that Paolo will appreciate our efforts put into this book.

Special mention is owed to Clive Bonsall who, besides co-authoring a paper for the book, also took charge of the final editorial checks, language revision and typesetting, thereby greatly improving the quality of the volume. Elisabetta Starnini Genova, 30 January 2013

iv

ITALY

Chapter 1 RAPID RIVERS, SLOW FORESTS: EARLY NEOLITHIC CHARCOAL FROM NORTHERN ITALY

Renato NISBET* *Dipartimento di Studi Umanistici, Università Ca’ Foscari, Venezia, Italy Abstract The paper first reviews some theories about the mobility/sedentariness of Early Neolithic groups in relation to the spread of forests across Europe, then takes into account the data referring to human settlement in northern Italy. Using the available record of on-site charcoal analyses, some suggestions are put forward. First, charcoal diagrams, documenting the systematic use of riparian vegetation, show almost invariably that the first farmers used open areas, close to rivers. Second, there is no reason to assume that these open spots formed as a consequence of deliberate man-made clearings over a long timespan; they were pre-existent to settlement. Third, this would imply that early farmers tended to avoid movement through woodland. Key words: Early Neolithic, charcoal, mobility

Introduction The spread of the earliest farming communities throughout continental Europe has been the object of countless papers for more than a century – that is, since it was first suggested that the loessic belt covering a large part of the European plains was quite unfavourable to the expansion of forests, thus allowing the diffusion of wide open areas, meadows and grassland (Penck 1887). The Steppenheide-Theorie (Gradmann 1906), therefore, was the first logical attempt at understanding the way agriculture moved across the continent from distant Asiatic nuclei. More and more distribution maps of Early Neolithic activity showed its progression along the valleys and/or the loessic belt which, according to various theories, were more suitable for agricultural purposes. These ‘light’ soils were considered most desirable to early farming for either hoe or plough agriculture. Furthermore, loessic soils on interfluves, where drainage conditions would have been unsuitable to the dense, closed forest of the Atlantic chronozone, would have provided excellent settlement conditions. Woodland palaeoecological studies developed with the increase of pollen research, leading first to the conclusion that dense forests covered the majority of the continent and eventually to the concept of Neolithic ‘landnam’ clearances (Iversen 1941). The whole problem of soil, vegetation cover and Early Neolithic settlement would be subsequently reviewed in a well known polemical paper by G. Clark (1945). Using the more recent pollen and soil data available at that time, he demonstrated that the ways to Central Europe for the early farmers were to be found within the forests, and not on open areas. Pollen data proved that loess was able to sustain closed forest vegetation. Within such a landscape, agriculture first spread, stone axes being the evidence of a strong human impact on a

forested environment and the conclusion was drawn that the light, loessic soils of the forests had been chosen because of their high fertility. Since then, many new data and new models have been developed. Following several authors, farming appeared in Europe with such an elaborate know-how that intensive agriculture was already possible and practised by the first farmers, and local but substantial change in the forest cover occurred subsequently, which can now be identified owing to new models of interpretation of macro- and microremains. Some evidence of early agriculture in the 7th millennium cal BC (i.e. during Mesolithic time) has also been suggested (Tinner et al. 2007). In general, this kind of evidence is ephemeral, based upon single pollen grains in peat or lake sediment and doubtful dates, and should be treated with caution (Behre 2007). Settlement and forests A controversial aspect concerns the ease of movement of people, products and ideas throughout the dense forests covering Europe during the Atlantic period. The problem of the real extension and density of the canopy has been long argued. This is not only because its compactness determines the amount of light on the ground, therefore favouring or hindering agricultural or pastoral activities (light being a crucial factor in the growth of the herb layer), but also because a high density of trees would cause obvious difficulties in the movements of mobile or semi-mobile societies. When small groups of farmers, with cattle, move from residential to peripheral camps, or to an entirely unknown area, a further problem arises. In an environment characterized, at least during many months of the year at 3

R. Nisbet

Figure 1.1 Location of sites mentioned in the text. Dots, sites with charcoal analyses (see diagrams): 1 Alba; 2 Pizzo di Bodio; 3 Cecima; 4 Travo; 5 Ostiano; 6 Isorella; 7 Vhò; 8 Fiorano; 9 Roncade; 10 Fagnigola; 11 Valer; 12 Sammardenchia. Squares, sites with huts, ‘villages’: A Lugo di Romagna; B Lugo di Grezzana; C Castel Grande. European latitudes, by a cover of trees reaching or exceeding 40m, hiding the sky, and by a view from the ground that would barely attain 40 or 50m (Tilley 2007), any planned movement inevitably would be slowed down or even hampered. Apart from this obvious but apparently neglected aspect, a major difficulty arises relative to form and size of the Early Neolithic settlement. In Central Europe several authors have argued that clusters of longhouses were commonly frequented for long periods of time:

Many archaeologists have long debated the degree of sedentism or mobility in Early Neolithic times. In his survey on the Mesolithic–Neolithic transition, Barker (2006) argues that in the Mediterranean countries much of the previous style of life survived in the early ‘farming’ communities, maintaining a high degree of mobility and exploiting seasonal resources over wide areas, with ‘a mix of foraging and farming’. In these early forager–farmer societies, from Dalmatia to Portugal, the impact on the landscape ‘was essentially the same as those of their forebears’. In summary, Barker points to the difficulty of establishing a sound definition of 7th–6th millennium communities:

The length of Early–Middle Neolithic site occupations varies but is often of the order of several centuries; some large settlements were occupied for more than four hundred years (Bogaard 2004, 13, quoting Lüning 1997 and 2000).

... all of whom can be categorized as Early Neolithic […] differed enormously in their degrees of sedentism, subsistence behaviour, diet, community organization, and ties to the land (Barker 2006, 350–356).

However, the evidence could be interpreted in a rather different way. According to the same author:

This appears not to be an isolated opinion, but rather a shared view. For Dalmatia, and within the Impressed Ware culture, Forenbaher and Miracle (2005, 524) maintain that, ‘Rather than establishing permanent settlements [Early Neolithic people] may have made short-term, seasonal camps in caves and the open-air’. For the western side of the Alps and Provence, Beeching (2003, 177) established that:

What appears to be a dense concentration of longhouses may represent the replacement of a single structure over time or include very few contemporary longhouses separated by considerable distances … many LBK settlements appear to consist of one to a few longhouses at any one time (Bogaard 2004, 13). 4

Rapid Rivers, Slow Forests ... the settlement model appears no different from that of the Mesolithic. As in the latter period, it would have been more a case of travelling the land, a question of natural mobility, exploration and exploitation of the environment, rather than the existence of a natural frontier (author's transl.).1

also beech) has been thought to have been the dominant primeval Holocene woodland. A debate has arisen on the presence of frequent open areas, formed by the natural regeneration of the forest or – as has been suggested – by the deliberate activity of Mesolithic hunters, and possibly maintained by the subsequent action of browsing herbivores. However, it is commonly believed that the original forest was substantially undisturbed at least until the 5th millennium cal BC. According to most French palynologists, during the spread of the first farming communities the whole of Western Europe was covered by une mer forestière in which it is difficult to perceive large open areas, and some of these were to be found only on wet spots, along rocky outcrops, after natural fires or in the mountains (de Beaulieu and Goeury 2004). Early signs of human impact on the local vegetation (end 7th–beginning 6th millennium cal BC) are sometimes observed. But in the case of several pollen sites of northeast France, anthropogenic pollen indicators never form a continuous curve. Rather, they point to a short or very short period of land use quite consistent with a mobile economy:

For the Cardial of the Franco-Iberian region, Guilaine and Manen (2007, 37) stated that, ‘Recent research puts the accent on the variety of economic systems adopted but also on an organization based on a mobile system of resource exploitation’. For the whole Early Neolithic of southern France and the Western Mediterranean, Sénépart and Beeching (2009, 304) found that: The Early Neolithic settlement of Western Europe seems more varied. Because of its diversity, it probably reflects a gradient in the colonization of new areas, which does not proceed from one same place of origin and especially not the same place of arrival (author's transl.).2 For the Balkans, the rapid spread, demonstrated by radiocarbon dates (Biagi et al. 2005) also suggests a high mobility of the Early Neolithic farmers. A different situation might, however, be found along the western Adriatic coasts, where the ditched enclosures of Tavoliere (southern Italy), rectangular structures and shallow pits (Calabria, Apulia) have been claimed to belong to a sedentary way of life that lasted for centuries. Some authors, however, have urged more caution. Whittle (1996, 295, 299) stressed that:

For over a millennium, short phases of human impact are described, among which the woodland shows no apparent change detectable by pollen analysis (Richard and Ruffaldi 2004, 123; author's transl.).3 In other areas further east, for example the continental Balkans, similar arguments have been used suggesting there is no evidence of human influence on the vegetation before 5000 cal BC, therefore dating to the Middle Neolithic.

Few if any of these [Tavoliere enclosures] seem to indicate prolonged or large occupations […]. Other occupations may rather suggest impermanence and movement [...]. The enclosures may mark points of coming and going, places of fixed interest in a fluid landscape.

This is not to say that the transition did not occur with the arrival of the first farmers but that it was not of sufficient intensity to be detected upon a landscape scale […]. The conclusion has therefore to be drawn that human impact was local in scale and did not affect regional patterns of vegetation. These results demonstrate that population densities at the Neolithic transition were low and that farming techniques had little or no impact on the existing vegetation (Willis and Bennett 1994, 237–238).

Similarly, Barker (2006, 349) understands these same features, ‘in terms of semi-mobile forager communities developing a commitment to agriculture’. Summing up, it is therefore evident that this aspect of Early Neolithic mobility is controversial owing to: 1. The poor preservation of many sites.

It seems there is a general agreement among palynologists in relation to the first episodes of well-defined clearings, as they appear only within the already well consolidated areas of Middle or Late Neolithic settlement. In the north Italian Apennines: Evidence for significant disturbance of forests, burning, flint scatters and soil erosion all indicate much more intensive activity in the mountain zone commencing in the Late Neolithic and becoming more intensive in the Chalcolithic to Bronze Age (Lowe et al. 1994).

2. The wide differences in behaviour, social organization, objective diversities of the exploited landscape (climate, soil conditions, geomorphology, flora and fauna). 3. The model of ‘neolithization’ accepted by the different authors, in terms of either a local acculturation or a colonization from outside. Whatever the case, there are no particular grounds for accepting permanent, long-lasting settlement as a normal lifestyle in the Early Neolithic. Spread of forests

The earliest pastures in the eastern Alps could not have been formed before 3400 cal BC (therefore of Late Neolithic age), though distinct clearances are clearly shown only from the Bronze Age onwards (Oeggl 1994).

Similar difficulties arise when one takes into account the opinions of many palynologists concerning the character of the forests during the 7th and 6th millennia cal BC, in the Middle Holocene. A mixed deciduous woodland, mostly formed by oak, lime, ash, elm (in some eastern regions of Central Europe

There are, however, some different views claiming a more decisive and disruptive role of early farmers on the vegetation, at least at a local level. Central European LBK sites (5500 cal BC) would offer examples of high potential for human impact on the vegetation, as imposed by the Neolithic way of life (arable and pastoral farming). This has 5

R. Nisbet left ‘a distinct imprint [in pollen records] on the European landscape’ (Kalis et al. 2003, 34), because Early Neolithic groups were full-time farmers cultivating their crops on permanent fields close to permanent settlements. The new economy would require open areas and, therefore, extensive clearance in a previously wooded landscape. As the AP/NAP (tree/herbs) ratio shows no particular changes before the Late Bronze Age, the authors dismiss this frequently used index as evidence of human impact on vegetation. However, we accept their arguments only if it could be proved that, 1) early farmers lived for long periods within the forests, and 2) they were fully sedentary. As we have already seen, the evidence for this is much disputed. More data will be presented suggesting a different explanation, at least for the northern Italy plains, which would better fit with the archaeological data relating to the penetration of 5th millennium cal BC populations.

Moreover, in many Early Neolithic sites of the Po Plain the economic basis is largely of a ‘Mesolithic’ type: The data concerning the Early Neolithic Fiorano communities should be brought up to date on the basis of new excavations, even if the presence of cattle breeding and agriculture is certain, their overall importance, as regards hunting, fishing and harvesting activities, is apparently scarce (Bagolini 1981, 194; author's transl.).5 As already noted for other parts of Europe, therefore, the evidence from northern Italy is contradictory, pointing in the majority of cases to mobile or semi-mobile groups, in others to the presence of huts and structured, long-lasting spaces of use. If this is the case, we should not be surprised to get conflicting evidence also from the environmental data (particularly those relating to vegetation) within Early Neolithic sites, as these should be referred in the different sites to very different histories – that is, in terms of land use depending on how long communities locally manipulated and transformed their setting.

Early Neolithic settlement in northern Italy In the last twenty years, the Early Neolithic settlement of northern Italy has been reviewed by several authors (e.g. Bagolini and Biagi 1987, 1990; Bagolini 1990; Biagi et al. 1993; Binder 2000; Improta and Pessina 2000; Pessina 2000). The focus has been on the definition of different cultures, chronologies, ceramic and lithic aspects. Therefore, the specific problem of mobility/sedentism has been touched on only as a secondary feature of the still controversial Mesolithic/Early Neolithic boundary. In regard to the settlements, the interpretation of many of the most common Early Neolithic features (particularly pits) is far from being agreed. In particular, there is today a strong reaction against considering the large shallow pits as pit dwellings or fondi di capanne, as has been usual in the past. They may well represent archaeological structures of a mobile or semi-mobile group of farmers repeatedly visiting the same site, even for diverse cultural reasons (Pearce 2008). A recent study carried out on the dwellings and related features of the first farmers in northern Italy (Cavulli 2008) reconsidered this topic. Out of c. 150 Early Neolithic sites (20% of which are rockshelter or cave sites), Cavulli found evidence of huts (three in total: Lugo di Romagna, Lugo di Grezzana and Castel Grande, the latter in Canton Ticino, Figure 1.1), trenches and palisades, postholes, channels, storage pits and terraces. In his opinion, the final reports on some sites like Lugo di Romagna, Lugo di Grezzana, Savignano, Sammardenchia, Campegine and Bazzarola would furnish a new picture of the Early Neolithic settlement, with villages, palisades, floors and a large number of minor features probably linked to agricultural practices. In his conclusions (Cavulli 2008, 346), however, he casts some doubt on the whole matter:

Early Neolithic vegetation in northern Italy There is little palynological evidence of the Mesolithic/Neolithic transition in northern Italy, particularly in the plains. The most extensive research was carried out some 30 years ago in the Alps (Schneider 1978, 1985; Schneider and Tobolski 1985); in general, the pollen diagrams relate to off-site deposits (lakes, peat bogs, etc.). In her comprehensive review, R. Schneider (1985) published a table of the sites in northern Italy that were mentioned in the palynological literature up to that time. Out of 137 sites, 61 were on flat land or valley bottoms below 500m (i.e. where most of the Early Neolithic sites are located), but only in 14 cases is the Atlantic chronozone represented (Schneider 1985, table 1, 92–97). According to Schneider and Tobolski (1985), evidence of small clearances and agriculture can be seen in some diagrams around Lago di Ganna at a date between 4000 and 3200 cal BC. More extensive open areas are found around Lago di Biandronno, but only after 4140 cal BC, i.e. during the Middle and Late Neolithic. The Atlantic forest, before the Neolithic presence was dense and dark: As a whole, this period [the late Atlantic chronozone] is characterized by a dense forest cover … Finally [toward 6300 cal BC] the forest seems to have been totally closed (Schneider 1978, 81; author's transl.).6 In this site, 2km away from the well known Early Neolithic site of Isolino di Varese, no evidence of contemporaneous clearances has been detected so far. We should mention, however, a different point of view by Accorsi et al. (1996). According to these authors, the Emilia Romagna Plain (the eastern segment of the Po Plain) was generally characterized by a relatively lighter AP (tree) cover than expected since the Pre-Boreal period. In particular, from the Boreal onwards, man-made clearings are present and no particular variations in the forest composition are detected through the entire Holocene. They suggest a very early (Boreal) human influence on the forest, and maintain that the landscape was a mix of woody and open

Nevertheless one wonders on the one hand if the first Neolithic communities were not sedentary as so far presumed, or on the other hand if the dwellings were poorly structured and made of light building materials, such as tents, which leave faint traces on the ground … Some data could indirectly confirm a scarce sedentariness, while others categorically refute it (author's transl.).4 6

Rapid Rivers, Slow Forests areas, the latter either natural, related to river environments, or anthropic. In addition, the same authors admit that Boreal and Atlantic forests were fairly similar in composition, with an increase of dense, dark cover (particularly due to Tilia and Ulmus) in the latter period. These conditions changed from the Sub-Boreal phase, when the landscape was possibly a patchwork of closed forest remnants, park-like areas and more open sites that clearly match the charcoal and seed database for the same (Sub-Boreal) period. We should then conclude that human disturbance at the Mesolithic/Neolithic transition, if any, was ephemeral and short-lived. Accordingly, Caramiello et al. (1996) reach similar conclusions in their review of the evidence for human impact in pollen diagrams of the western North Italian Plain. Using the AP/NAP ratio for the Atlantic chronozone (between 65 and 90%) they conclude that farming had an influence on the woodland vegetation particularly around 3800 cal BC, i.e. during the Middle–Late Neolithic, and not before. On-site pollen analyses have been published in the last 20 years. We take into account the Early Neolithic site of Alba (Southern Piedmont), Cooperativa dei Lavoratori, where both pollen and charred wood were analyzed (pollen: Caramiello and Zeme 1995; charcoal: Motella De Carlo 1995). Two samples (US 28 and US 31) might be referred to the Atlantic period. According to the stratigraphic description US 28 belongs to, ‘a sequence of natural sedimentation preceding the first occupation, 7 while US 31 is ‘a flood deposit … which on archaeological evidence dates to the Middle–Late Neolithic (third phase)’ (Venturino Gambari et al. 1995, 73; author's transl.).8 The authors argue that:

(charcoal, seeds, molluscs, bones, etc.) in order to gain a better understanding of Early Neolithic settlement. We shall take into account 12 Early Neolithic charcoal diagrams from the whole of the North Italian Plain (and surroundings hills or valleys). These have been presented, when data are available, as ubiquity (number of occurrences of a particular taxon in the sample total, Figure 1.2) or, otherwise, as a percentage of the total number of fragments (Figure 1.3). There are some significant features common to all diagrams: 1. Oak (Quercus sp.) and ash (Fraxinus sp.) are dominant, followed by plum/Prunus, crab apple trees (Maloideae, Prunoideae) and hazel (Corylus sp.). 2. Riparian trees like aspen (Populus sp.) and alder (Alnus sp.) are always present. 3. Dense woodland, shade-tolerant trees like lime (Tilia sp.), maple (Acer sp.) and elm (Ulmus sp.) are generally much less well represented, both as site-presence (ubiquity) and number of fragments. There is a general agreement (reviewed by Castelletti and Rottoli 1997) that these charcoal associations relate to growth in open areas, close to rivers, lakes or flooded valley bottoms. In some cases it has been argued that this environment may represent the regeneration phase following a long period of clearances and agricultural pressure on the local woodland by arable and pastoral farmers (in the majority of these sites cereal grains are also present). This belief rests, evidently, on the assumption – in our opinion not always supported by sufficient archaeological evidence – that the sites where human communities first settled, a) needed to be cleared because the forest was dense and unsuitable for any form of farming, and b) show a charcoal spectrum that refers to the last phase when, after decades of woodland management, lightdemanding trees were eventually growing at the forest margins. However, many authors (e.g. Vera 2000; Rackham 2003; Kreuz 2008) have convincingly demonstrated, on the basis of the growth physiology of the shoots, that a mixed oak woodland is permeated by light. We should therefore consider the Querco–Fraxinetum represented in our charcoal diagrams as a somewhat open association. Moreover, Fraxinus normally grows on river terraces and the frequency of its wood could be related to the greater presence of the trees around the sites and/or to a deliberate selection of its leaves for use as fodder.

... the low values of human presence seem to show a slight impact through the whole sequence from the Neolithic to the Bronze Age, even if at the time between the Neolithic and the Copper age higher values have been noticed (Caramiello and Zeme 1995, 244; author's transl.).9 In particular, US 28 (which seems to precede the first settlement) shows higher percentages of pine, juniper and hazel and in lesser quantity ash, alder, lime and aspen. This therefore points convincingly to the existence of fairly open, light-demanding vegetation before the human land occupation. Charcoal and forests In this same site of Alba, charcoal from fireplaces was analyzed by Motella De Carlo (1995). The published sample was collected in US 34, possibly of Early Neolithic age. The results broadly confirm the pollen data have a very high percentage of ash, followed by oak, aspen, plum, crab apple/pear. Quite rightly, the author concluded that this charcoal assemblage could be related to riparian and open areas, though she decided in favour of a strong anthropogenic impact on the local vegetation. Charcoal diagrams provide good evidence for the reconstruction of the local woodland composition, integrating with pollen data. Moreover, in northern Italy the scarcity of on-site pollen diagrams requires other environmental data obtained by good sampling strategies

Man may have contributed toward its distribution by felling trees, as ash behaves as a pioneer tree. Therefore, in a later phase of settlement ash values could be higher than at the beginning and lower during periods of abandonment (Castelletti and Maspero 1992, 304; author's transl.).10 While this association, along with members of the Rosaceae family and hazel, points to open areas, the constant presence of riverside trees, namely alder and aspen, clearly indicates that most of the Early Neolithic sites were chosen for their location close to a stream. This particular aspect has been observed by many authors from throughout Europe. 7

R. Nisbet

Figure 1.2. Charcoal diagrams, Early Neolithic. Presence percent (ubiquity). In the Roncade diagram Quercus** – Quercus sez. robur; Quercus* – Quercus sp.; Σ – number of samples. In Central Europe ‘the location of Early–Middle Neolithic sites along river valleys […] may reflect a preference for relatively open vegetation’ (Bogaard 2004, 14). In the Western Mediterranean countries the Cardial Neolithic, along with its well known coastal sites, shows an ‘early penetration into more continental domains, in particular along the main fluvial routes’ (Guilaine and Manen 2007, 37). In the Balkans, Körös sites:

... are usually flat and lined up along river courses […] Körös houses were single-room, rectangular structures with gabled roofs and wattle-and-daub or reed walls. Despite such apparently substantial structures, there seem to have been frequent shifts of settlements and perhaps only a semi-sedentary settlement pattern (Gronenborn 1999, 145). Nowadays in the Po Plain, the riparian vegetation is mostly formed by aspen and willow colonizing alluvial sandy soils, with a sparse undergrowth of elder (Sambucus) and dogwood (Cornus). In the surrounding areas, a xeric 8

Rapid Rivers, Slow Forests

Figure 1.3. Charcoal diagrams, Early Neolithic. Fragments percent. Conclusions

grassland is frequently found on drier and gravelly soils. According to Mondino (2007), in pre-industrial times a transition between the Quercus–Carpinus association and the riparian underbrush was formed by alder, aspen and pedunculate oak, growing on periodically flooded areas. These open belts along the rivers, undoubtedly present since the Early Holocene, were well known to the first migrants and their animals as they could provide water, food, fodder and good, well drained soil for short-lived camps. They provided excellent natural, pre-settlement locations, easily found both from the streams themselves and from the woodland hedges where the undergrowth did not constitute a problem for the movement.

Notwithstanding several millennia of intense use of woodland, Medieval Europe was still covered by a dense forest (Higounet 1966). According to this author, in some regions the woodland was reduced somewhat in extent, but about 70% of Europe remained densely wooded: the ideal environment for hermits, robbers, monks or anybody having some reason to disappear for months or years, certainly not for travellers. And even if we accept Wickham’s criticism of his justified refusal of the late 1st millennium AD European woodland as a horrendum desertum dominated by silvae densissimae, we should conclude with the same author 9

R. Nisbet (Wickham 1990) that the true, last deforestations occurred only in post-Carolingian times. At that time, part of north Italy had already suffered major deforestation during the Bronze Age – that is, the ‘Terramare’ period – when the first proto-urban cultures largely changed the extent and shape of the previous forests. A site density of 9–10 per km 2 has been calculated (Cremaschi 1997) over a period of 4–5 centuries, with some larger sites reaching 10 hectares with 200–300 inhabitants. Requiring a huge use of timber, Terramare point evidently to the existence of large forests surviving in the region even after some millennia of Middle–Late Neolithic farming, when an increase in site numbers, demographic expansion and fixed villages caused the first wide and well documented expansion of open areas in the Holocene due to human activity. Considering the early north Italian Neolithic, it should be stressed that sites with conclusive evidence of a long period of occupation, with a sound 14 C-based chronology, are extremely rare. For the greater part of them, the archaeological features are usually connected to pottery and flint scatters, post- and stake-holes, small pits, ditches, shallow depressions and other similar small features. It is thus not easy to see in this evidence, a ‘village’ or a lasting settlement. In order for a forest site, colonized by a resident group of Neolithic immigrants, to show changes in the charcoal sample composition (thus displaying with sufficient evidence a new spectrum, formed by trees favoured by the protracted human presence) several decades have to go by from its early occupation. In such a case, we should find a shift, in the charcoal composition from the use of close, dense forest trees, to a more open landscape. However, anthracology shows a quite different picture, representing an environment formed by heliophilous (lightdemanding) trees and, together with these plants, others from riverbanks and gravelly or sandy soils. This choice of fluvial terraces for occupation is by no means surprising. It has been described in Central Europe by several authors and recently Bogaard stated that: ... it seems plausible that river valleys provided an important form of seasonal pasturage for livestock in an otherwise wooded environment. The proximity of Early–Middle Neolithic settlements to river valleys, therefore, may relate primarily to animal rather than to crop husbandry (Bogaard 2004, 158). We should add to these priorities the need to avoid dense forests for the movement of a colonist group. In the North Italian Plain, river terraces were constantly walked and chosen for settlement. So it happened for almost all Early Neolithic sites of the Friuli flatland (‘the first farmers settled on natural embankments of ancient riverbeds’, Fontana 2000, 222; author's transl.),11 where water was provided by the presence of resurgence streams. The same applies to many central Po Plain sites, like those of Isorella (Starnini et al. 2000; this site was previously considered by the present author to be a forest site), Vhò (Castelletti and Maspero 1992), Savignano (Bernabò Brea et al. 1990), Ostiano (Biagi 1995), Travo (Bernabò Brea et al. 1984), etc. Interestingly, strong physiographic similarities can be found in the Late Mesolithic sites, which are generally

located close to river terraces, watercourses or stream confluences (Franco 2011). This would raise the question of a possible competition for the same environment between Mesolithic tribes and the Neolithic newcomers. In conclusion, if a site were occupied over a period of time spanning centuries, then we can agree that important changes in the local vegetation would have occurred. As such, the charcoal evidence would reflect these changes in terms of an increase of light-demanding trees and bushes, hazel, plum, etc. In such a case, however, we should also find clear proof of a stable settlement in terms of artefacts, huts, and other features. If the site were only a temporary one, frequented for years or a few decades at most, this short period of time would not be sufficient to substantially alter the local vegetation, allowing the growth and expansion of fringe species and their subsequent use in the last fireplaces kindled on site. As the archaeological evidence seems, in most cases, so scarce, we have to admit that our charcoal diagrams should be related to a cover that the moving group found – or, better, chose – on azonal vegetation (Kreuz 1990), such as would be present on river terraces and barren soils, and by no means the result of a deliberate long-term manipulation of the local landscape. If this frequent presence of naturally formed, large, open areas frequented by the Neolithic ‘demic’ diffusion in north Italy since the middle of the 5th millennium cal BC were an immediate consequence, or a relic, of the 8.2 ka BP ‘climate event’ in the Po Plain, as seems to be the case in other parts of Europe (e.g. in the Balkans: Budja 2007), or the result of much more limited, local situations, is a difficult question that still needs to be answered. That is certainly not to say that, during the Early Neolithic, more or less wide, open areas of anthropic origin created for crop or pasture were totally absent or that permanent settlement never existed. In some instances, geomorphology and soil studies around a small number of sites have provided evidence of important, possibly manmade disturbances of which the case of the already mentioned Lugo di Grezzana site is perhaps the most impressive (Cavulli et al. 2002). The main purpose of the present paper is to show, using on-site data from charcoal, that the similarities in the diagrams can be interpreted according to a different model, envisaging also the presence of short-period occupation sites, which had a low impact on the local vegetation and a preference for pre-existing open areas along streams and rivers. In this model there is at least one good reason for the periodical, seasonally-determined abandonment of the camp, i.e. recurrent flooding, as suggested by Whittle (1997). These arguments do not seem to be in contradiction with many current archaeological data and the well known choice of fluvial routes by early farming communities. If the streams themselves were actively used as a preferred solution to movement, it has not yet been fully recognized but has in some way already been suggested even in the case of Late Mesolithic hunters (Franco 2011, 184). Unfortunately, the evidence of north Italian Neolithic canoes does not start before the 3rd millennium cal BC and mostly relates to lake travelling (Barbaglio 2006; Rufino 2009). Therefore, the idea of the existence and use of early fluvial routes should await stronger evidence. 10

Rapid Rivers, Slow Forests Notes 1. ‘... le statut des implantations [du Néolithique ancien] ne semble pas différer de celles du Mésolithique. Comme à cette époque, il s’agirait plutôt d’un territoire parcouru et d’une mobilité naturelle d’exploration et d’exploitation du milieu que d’un front pionnier’. 2. ‘L’habitat du Néolithique ancien de l’Ouest européen apparait plus varié […] Par sa diversité, il est le reflet probable de gradients dans la conquête de nouveaux espaces qui ne procède pas d’un même lieu d’origine et surtout pas d’un même lieu d’arrivée’. 3. ‘Pendant plus d’un millénaire, sont décrites de courtes phases d’anthropisation entre lesquelles s’intercalent des épisodes forestiers où apparemment aucune trace d’influence humaine n’est perceptible par l’analyse pollinique’. 4. ‘Sorge tuttavia il dubbio da un lato che le prime comunità neolitiche non fossero sedentarie quanto finora supposto e dall’altro lato che le abitazioni fossero poco strutturate e costituite da materiali leggeri, come ripari-tenda, che lasciano scarse tracce sul terreno […] Alcuni dati […] potrebbero confermare indirettamente una scarsa sedentarietà, mentre altri la confutano in modo categorico’. 5. ‘I dati sull’economia delle popolazioni di Fiorano necessiterebbero di un aggiornamento sulla base di nuovi scavi anche se è comunque certa la presenza di allevamento e di agricoltura, seppure con un peso complessivo apparentemente ancora scarso rispetto alle attività di caccia, pesca e raccolta’. 6. ‘Dieser Abschnitt ist im gesamten durch einen starken Waldschluβ gekennzeichnet […] Endlich scheint sich der Wald vollständig geschlossen zu haben’. 7. ‘una sequenza di depositi naturali preesistenti alla prima occupazione’. 8. ‘un evento alluvionale […] che i materiali archeologici assegnano al Neolitico medio-recente (III fase)’. 9. ‘... i valori bassi dell’indice di frequentazione antropica sembrano indicare una modesta influenza dell’uomo lungo tutta la sequenza temporale dal Neolitico all’età del Bronzo, anche se nel periodo fra Neolitico e Eneolitico si osservano valori un po’ più elevati’. 10.‘L’uomo può aver contribuito alla sua diffusione col taglio degli alberi, perché il frassino ha caratteristiche di albero pioniere. Quindi, nelle fasi non iniziali dell’insediamento i valori di frassino potrebbero essere superiori che all’inizio e diminuire poi nelle fasi di abbandono’. 11. ‘i protoagricoltori si stabilirono sugli argini naturali dei paleoalvei’.

Acknowledgments I should like to thank the Editor for inviting me to be a contributor to this volume, and Barbara A. Voytek who read the paper and kindly improved the English version of the text. Rob Scaife made some valuable suggestions. Paolo himself has provided materials through the years and I still remember with pleasure the beginning of the story, a hot summer some 40 years ago, in a small railway station in the middle of the Friuli Plain, where we first met when digging some Early Neolithic pits at Fagnigola. References Accorsi, C.A., Bandini Mazzanti, M., Mercuri, A.M., Rivalenti, C. and Trevisan Grandi, G. 1996. Holocene

forest pollen vegetation of the Po Plain – northern Italy (Emilia Romagna data). Allionia 34, 233–276. Bagolini, B. 1981. Il neolitico di Spilamberto – S. Cesario e le prime comunità agricole padane. In B. Bagolini (ed.), Il Neolitico e l’età del Rame. Ricerca a Spilamberto – S. Cesario 1977–1980, 189–216. Vignola, Cassa di Risparmio. Bagolini, B. 1990. Contacts entre les courants danubiens et Méditerranéens en Italie du Nord. In D. Cahen and M. Otte (eds), Rubané et Cardial. Actes du Colloque de Liège, Etudes et Recherches Archéologiques de l’Université de Liège 39, 73–82. Bagolini, B. and Biagi, P. 1987. Il Neolitico dell’Emilia Romagna. Atti della XXVI riunione scientifica dell’Istituto Italiano di Preistoria e Protostoria ‘Il Neolitico in Italia’, Firenze 1985, vol. 1, 218–227. Firenze. Bagolini, B. and Biagi, P. 1990. The radiocarbon chronology of the Neolithic and Copper Age of northern Italy. Oxford Journal of Archaeology 9(1), 1–24. Barbaglio, F. 2006. Le imbarcazioni monossili: aspetti di ricerca, restauro e museologia. Unpublished MA thesis, Università Cattolica Sacro Cuore, Milano. Barker, G. 2006. The Agricultural Revolution in Prehistory. Why did Foragers become Farmers? Oxford, Oxford University Press. Beeching, A. 2003. Mobilité et société néolithiques dans les Alpes occidentales et la France méridionale. Preistoria Alpina 39, 175–187. Behre, K-H. 2007. Evidence for Mesolithic agriculture in and around Central Europe? Vegetation History and Archaeobotany 16, 203–219. Bernabò Brea, M., Cattani, M., Conversi, R., Cremaschi, M., Nisbet, R. and Ricci, C. 1984. L’insediamento neolitico di Cassa di Risparmio a Travo (PC). Preistoria Alpina 20, 59–80. Bernabò Brea, M., Steffè, G. and Giusberti, G. 1990. Il Neolitico antico a Savignano. In Nel segno dell’elefante: Geologia, Paleontologia e Archeologia nel territorio di Savignano sul Panaro, 71–134. Savignano sul Panaro. Biagi, P. (ed.) 1995. L’insediamento neolitico di OstianoDugali Alti (Cremona nel suo contesto ambientale ed economico. Monografie di Natura Bresciana 22. Brescia. Biagi, P., Shennan, S. and Spataro, M. 2005. Rapid rivers and slow seas? New data for the radiocarbon chronology of the Balkan peninsula. In L. Nikolova, J. Fritz and J. Higgins (eds), Prehistoric Archaeology & Anthropological Theory and Education, 41–52. Salt Lake City, International Institute of Anthopology. Biagi, P., Starnini, E. and Voytek, B. 1993. The Late Mesolithic and Early Neolithic settlement of northern Italy: recent considerations. Porocilo o raziskovanju paleolita, neolita in eneolita v Sloveniji 21, 45–67. Binder, D. 2000. Mesolithic and Neolithic interaction in southern France and northern Italy: new data and current hypotheses. In T.D. Price (ed.), Europe’s First Farmers, 117–143. Cambridge, Cambridge University Press. Bogaard, A. 2004. Neolithic Farming in Central Europe. London, Routledge. Budja, M. 2007. The 8200 cal BP ‘climate event’ and the process of neolithisation in south-eastern Europe. 11

R. Nisbet Documenta Praehistorica 34, 191–201. Caramiello, R., Siniscalco, C. and Arobba, D. 1996. Human impact on the western Po Valley vegetation in the Holocene. Allionia 34, 149–163. Caramiello, R. and Zeme, A. 1995. Analisi archeopalinologica in sequenze stratigrafiche comprese tra il Neolitico e l’età del Bronzo. In M. Venturino Gambari (ed.), Navigatori e contadini. Alba e la valle del Tanaro nella preistoria. Quaderni della Soprintendenza Archeologica del Piemonte, Monografie 4, 239–244. Alba, Famija Albèisa. Carugati, M.G. 1993. I resti vegetali macroscopici. In B. Bagolini, M.G. Carugati, A. Ferrari and A. Pessina, Fagnigola Bosco Mantova (Azzano Decimo – Pordenone). Notizie preliminari sull’intervento 1991. Atti della Società di Preistoria e Protostoria del FriuliVenezia Giulia 7(1992), 47–64. Trieste. Carugati, M.G. 1994. Nota sui resti vegetali carbonizzati del sito neolitico di Valer (Azzano Decimo – Pordenone). Atti della Società di Preistoria e Protostoria del FriuliVenezia Giulia 8 (1993), 115–120. Trieste. Castelletti, L. and Carugati, M.G. 1994. I resti vegetali del sito neolitico di Sammardenchia di Pozzuolo del Friuli (Udine). Atti della XXIX Riunione scientifica dell’Istituto Italiano di Preistoria ‘Preistoria e Protostoria del Friuli-Venezia Giulia e dell’Istria’, Trieste, Istria, 28–30 settembre 1990, Firenze, 167–184. Castelletti, L. and Madella, M. 1994. Appendice 3. In D.G. Banchieri and C. Balista, Note sugli scavi di Pizzo di Bodio (Varese) 1985–88. Preistoria Alpina 27(1991), 236–238. Castelletti, L. and Maspero, A. 1992. Analisi di resti vegetali di Campo Ceresole del Vhò di Piadena e di altri siti neolitici padani. Natura Bresciana 27, 289–305. Castelletti, L. and Rottoli, M. 1997. New data on Neolithic agriculture and environment in northern Italy. Preistoria Alpina 33, 57–61. Cavulli, F. 2008. Abitare il Neolitico. Le più antiche strutture antropiche del Neolitico in Italia Settentrionale. Preistoria Alpina 43 / Supplemento I. Trento. Cavulli, F., Angelucci, D.E. and Pedrotti, A. 2002. La successione stratigrafica di Lugo di Grezzana (Verona). Preistoria Alpina 38, 89–107. Clark, G. 1945. Farmers and forests in Neolithic Europe. Antiquity 74, 57–71. Cremaschi, M. 1997. Terramare e paesaggio padano. In M. Bernabò Brea, A. Cardarelli and M. Cremaschi (eds), Le Terramare. La più antica civiltà padana, 107–136. Milano, Electa. de Beaulieu, J-L. and Goeury, C. 2004. Les premiers signes de l’anthropisation dans les Alpes françaises d’après l’analyse pollinique. In H. Richard (ed.), Néolithisation précoce. Premières traces d’anthropisation du couvert végétal à partir des données polliniques, 163–171. Besançon, Presses Universitaires Franche-Comté. Fontana, A. 2000. Siti e ambienti neolitici nella pianura friulana. In A. Pessina and G. Muscio (eds), La neolitizzazione tra oriente e occidente, Atti del Convegno di Studi, Udine, 23–24 aprile 1999, 213–230. Udine, Museo Friulano Storia Naturale. Forenbaher, S. and Miracle, P.T. 2005. The spread of

farming in the eastern Adriatic. Antiquity 79, 514–528. Franco, C. 2011. La fine del Mesolitico in Italia. Identità culturale e distribuzione territoriale degli ultimi cacciatori-raccoglitori. Trieste, Società per la Preistoria e Protostoria della regione Friuli-Venezia Giulia, Quaderno 13. Gradmann, R. 1906. Beziehung zwischen Pflanzengeographie und Siedlungsgeschichte. Geographische Zeitschrift 94, 305–325. Gronenborn, D. 1999. A variation on a basic theme: the transition to farming in southern Central Europe. Journal of World Prehistory 13(2), 123–210. Guilaine, J. and Manen, C. 2007. From Mesolithic to Early Neolithic in the Western Mediterranean. In A. Whittle and V. Cummings (eds), Going Over. The Mesolithic–Neolithic Transition in North-West Europe, 21–51. Oxford, Oxford University Press/British Academy. Higounet, C. 1966. Les forêts de l’Europe occidentale du Ve au XIe siècle. In Agricoltura e mondo rurale in Occidente nell’Alto Medioevo, 343–398. Spoleto, Centro Italiano di studi sull’Alto Medioevo, 13. Improta, S. and Pessina, A. 2000. La neolitizzazione dell’Italia settentrionale. Il nuovo quadro cronologico. In A. Pessina and G. Muscio (eds), Settemila anni fa il primo pane, 107–115. Catalogo della mostra dicembre 1988–maggio 1999, Comune di Udine e Museo Friulano di Storia Naturale. Iversen, J. 1941. The influence of prehistoric man on vegetation. Danmarks Geologiske. Undersøgelse Series 2, no. 6, 1–25. Kalis, A.J., Merkt, J. and Wunderlich, J. 2003. Environmental changes during the Holocene climatic optimum in Central Europe – human impact and natural causes. Quaternary Science Reviews 22, 33–79. Kreuz, A. 1990. Die ersten Bauern Mitteleuropas. Eine archäobotanische Untersuchung zu Umwelt und Landwirtschaft der ältesten Bandkeramik. Analecta Praehistorica Leidensia 23. Leiden, Leiden University Press. Kreuz, A. 2008. Closed forest or open woodland as natural vegetation in the surroundings of Linearbandkeramik settlements? Vegetation History and Archaeobotany 17, 51–64. Lowe, J.J., Branch, N. and Watson, C. 1994. The chronology of human disturbance of the vegetation of the northern Apennines during the Holocene. In P. Biagi and J. Nandris (eds), Highland Zone Exploitation in Southern Europe, 169–187. Monografie di Natura Bresciana 20. Brescia, Museo Civico di Scienze Naturali. Lüning, J. 1997. Wohin mit der Bandkeramik? Programmatische Bemerkungen zu einen allgemeinen Problem am Beispiel Hessens. In C. Becker, M-L. Dunkelmann, C. Metzer-Nebelsick, H. Peter-Röcher and B. Terzan (eds), Chronos. Beiträge zur prähistorischen Archäologie zwischen Nord- und Südosteuropa. Festschrift Bernhard Hänsel, 23–57. Internationale Archäologie, Studia Honoraria 1. Espelkamp, Marie Leidorf. Lüning, J. 2000. Steinzeitliche Bauern in Deutschland – die Landwirtschaft im Neolithikum. Bonn, Habelt. 12

Rapid Rivers, Slow Forests Mondino, G.P. (ed.) 2007. Flora e vegetazione del Piemonte. Savigliano, Edizioni L’Artistica. Motella De Carlo, S. 1995. Paleoecologia ad Alba nella preistoria. Indagine sui macroresti vegetali. In M. Venturino Gambari (ed.), Navigatori e contadini. Alba e la valle del Tanaro nella preistoria, 245–255. Quaderni della Soprintendenza Archeologica del Piemonte, Monografie 4. Alba, Famija Albèisa. Nisbet, R. 1982. Cecima (PV). Capanna del Neolitico Inferiore padano. Analisi archeobotaniche. Notiziario Soprintendenza Archeologica della Lombardia 1982, 15. Nisbet, R. 1984. Analisi paleobotaniche. In M. Bernabò Brea, M. Cattani., R. Conversi, M. Cremaschi, R. Nisbet and C. Ricci 1984. L’insediamento neolitico della Cassa di Risparmio di Travo (PC). Preistoria Alpina 20, 78–79. Nisbet, R. 1995. I resti macrobotanici. In P. Biagi (ed.), L’insediamento neolitico di Ostiano-Dugali Alti (Cremona) nel suo contesto ambientale ed economico, 104–106. Monografie di Natura Bresciana 22. Brescia, Museo Civico di Scienze Naturali. Nisbet, R. 2000. Aspetti forestali e agricoltura ad Isorella. In E. Starnini, F. Ghisotti, A. Girod and R. Nisbet, Nuovi dati sul Neolitico antico della Pianura Padana centrale dal sito di Isorella (Brescia). In A. Pessina and G. Muscio (eds), La Neolitizzazione tra oriente e occidente, Atti del Convegno di Studi, Udine, 23–24 aprile 1999, 231–255. Udine, Museo Friulano di Storia Naturale. Oeggl, K. 1994. The palynological record of human impact on highland zone ecosystems. In P. Biagi and J. Nandris (eds), Highland Zone Exploitation in Southern Europe, 107–122. Monografie di Natura Bresciana 20. Brescia, Museo Civico di Scienze Naturali. Pearce, M. 2008. Structured deposition in Early Neolithic northern Italy. Journal of Mediterranean Archaeology 21(1), 19–33. Penck, A. 1887. Das Deutsche Reich. In A. Kirchhoff (ed.), Länderkunde des Erdteils Europa I (cited in G. Clark 1945). Pessina, A. 2000. Aspetti culturali e problematiche del primo Neolitico dell’Italia settentrionale. In A. Pessina and G. Muscio (eds), Settemila anni fa il primo pane, 95–105. Catalogo della mostra dicembre 1988–maggio 1999. Udine, Comune di Udine e Museo Friulano di Storia Naturale. Pignatelli, O. and Rottoli, M. 1996. Analisi archeobotaniche. In E. Bianchin Citton, (ed.), Indagine interdisciplinare nell’insediamento neolitico di Roncade (Treviso) – Località Biancade, 113–118. Quaderni di Archeologia del Veneto 12. Venice, Canova. Rackham, O. 2003. Ancient Woodland: its History, Vegetation and Uses in England. Dalbeattie, Castelpoint Press. Richard, H. and Ruffaldi, P. 2004. Premières traces polliniques d’influence de l’homme sur le couvert végétal de l’Est de la France. In H. Richard (ed.), Néolithisation précoce. Premières traces d’anthropisation du couvert végétal à partir des données polliniques, 117–125. Besançon, Presses Universitaires de Franche-Comté. Rufino, R. 2009. Per un’archeologia dell’altomedioevo fluviale. Le imbarcazioni monossili della pianura padana. Unpublished MA thesis, Università Ca’ Foscari of Venice.

Schneider, R. 1978. Pollenanalytische Untersuchungen zur Kenntnis der spätund postglazialen Vegetationsgeschichte am Südrand der Alpen zwischen Turin und Varese (Italien). Botanische Jahrbücher, Systematik 100(1), 26–109. Schneider, R. 1985. Palynologic research in the southern and southeastern Alps between Torino and Trieste. Dissertationes Botanicae 87, 83–103. Schneider, R. and Tobolski, K. 1985. Lago di Ganna – Late Glacial and Holocene environments of a lake in the southern Alps. Dissertationes Botanicae 87, 229–271. Sénépart, I. and Beeching, A. 2009. De la maison au village dans le Néolithique du Sud de la France et de l’Ouest méditerranéen – Essai de synthèse. In A. Beeching and I. Sénépart (eds), De la maison au village. L’habitat néolithique dans le Sud de la France et le Nord-Ouest méditerranéen. Actes de la Table Ronde des 23 et 24 mai 2003, 303–310. Paris, Société Préhistorique Française. Starnini, E., Ghisotti, F., Girod, A. and Nisbet, R. 2000. Nuovi dati sul Neolitico antico della Pianura Padana centrale dal sito di Isorella (Brescia). In A. Pessina and G. Muscio (eds), La neolitizzazione tra oriente e occidente, Atti del Convegno di Studi, Udine, 23–24 aprile 1999, 231–255. Udine, Museo Friulano di Storia Naturale. Tilley, C. 2007. The Neolithic sensory revolution: monumentality and the experience of landscape. In A. Whittle and V. Cummings (eds), Going over. The Mesolithic–Neolithic Transition in North-West Europe, 329–345. Oxford, Oxford University Press/British Academy. Tinner, W., Nielsen, E.H. and Lotter, A.F. 2007. Mesolithic agriculture in Switzerland? A critical review of the evidence. Quaternary Science Reviews 26, 1416–1431. Venturino Gambari, M., Gambari, F.M. and Davite, C. 1995. L’indagine archeologica. In M. Venturino Gambari (ed.), Navigatori e contadini. Alba e la valle del Tanaro nella preistoria, 57–104. Quaderni della Soprintendenza Archeologica del Piemonte, Monografie 4. Alba, Famija Albèisa. Vera, F.W.M. 2000. Grazing Ecology and Forest History. Wallingford (UK), CABI Publishing. Whittle, A. 1996. Europe in the Neolithic: The Creation of New Worlds. Cambridge, Cambridge University Press. Whittle, A. 1997. Moving on and moving around: Neolithic settlement mobility. In P. Topping (ed.), Neolithic Landscapes, 15–22. Oxford, Oxbow. Wickham, C. 1990. European forests in the early Middle Ages: landscape and land clearance. In L’ambiente vegetale nell’alto Medieovo: Settimane di studio del Centro italiano di studi sull’alto Medioevo 37, 479–545. Willis, K.J. and Bennet, K.D. 1994. The Neolithic transition – fact or fiction? Palaeoecological evidence from the Balkans. The Holocene 4(3), 326–330. *

*

*

*

*

Author’s address: Dipartimento di Studi Umanistici, Palazzo Malcantòn Marcorà, Dorsoduro 3484/D, Università Ca’ Foscari, I-30123, Venezia, Italy E-mail: [email protected] 13

EUROPE

Chapter 2 MESOLITHIC OBSIDIAN NETWORKS IN THE AEGEAN

Małgorzata KACZANOWSKA* and Janusz K. KOZŁOWSKI* * Institute of Archaeology, Jagiellonian University, Kraków, Poland Abstract The paper deals with the Aegean obsidian extraction and diffusion from two main sources on Melos and on Ghiali (Yali) in the Mesolithic. The range of obsidian diffusion, particularly Melian, in the Aegean Basin overlaps with the specific cultural province embracing the islands from the Sporades in the north, Cyclades in the west and Dodecanese in the south. The ‘Aegean Mesolithic’ was a hermetic system that was based on the exploitation of marine and insular resources. The diffusion of elements of the ‘Neolithic package’ even in the early phase of the ‘Aegean Mesolithic’ and – in its final phase (Aceramic Neolithic on Crete) – confirms the relations with the Eastern Mediterranean. Key words: Aegean, Mesolithic, Melian and Ghialian obsidian, Neolithization

The Aegean Mesolithic as a taxonomic unit The present paper discusses the use of obsidian by Early Holocene foragers of the Aegean Basin. On the interregional scale these communities exploited first of all obsidian from the island of Melos. The sources of obsidian on Melos have been investigated and described by C. Renfrew et al. (1965), R. Torrence (1986), C. Renfrew and A. Aspinall (1990), A. Arias et al. (2006) and other researchers. Another, more regional source of obsidian was the island of Ghiali (Yali) whose deposits were investigated and described by H.G. Buchholz and E. Althaus (1982) and by A. Sampson and I. Liritzis (1998). The third source of obsidian in the Aegean Basin – the island of Antiparos – was of no importance before the Late Neolithic (Cann and Renfrew 1964). The settling of the islands of the Aegean began as early as the Late Pleistocene; on some islands of the northern Sporades (Alonissos, Psathoura, Gramiza, Kyra-Panagia, Pappous) and on Euboea traces of occupation were dated to the Middle Palaeolithic (Sampson 2006), whereas on the Cyclades probable traces of the Upper Palaeolithic come from the period of marine regression of the Last Glacial Maximum (the Gravettian on the island of Kythnos, site KT-21 – Sampson et al. 2010, and probably on Naxos at Stelida – Séferiadès 1983). Recent discoveries on the island of Limnos (Efstratiou et al. 2013) indicate the presence of Epigravettian occupation in the northern part of the Aegean Sea Basin in the Late Pleistocene. Trans-Aegean communication routes during the postLGM period are confirmed by the cultural Epigravettian unity that embraced territories from the Adriatic Basin, across the Balkans, as far as southwestern Anatolia (Kozłowski and Kaczanowska 2004). It was only the ridge of the Taurus Mountains that constituted a boundary of this unity in the east. East of the Taurus Mountains extended the province of the Near East Late Palaeolithic units (within the

Kebaran/Natufian sequence: S.K. Kozłowski 1994; Aurenche and Kozłowski 2005). The Epigravettian (Figure 2.1) was the basis for the formation of Early Holocene cultural units both in eastern Greece and the Aegean islands. Marine contacts between continental Greece and the Cyclades during the Epigravettian are documented by the occurrence of obsidian artefacts as early as the Late Pleistocene lithic phase VI in the Franchthi Cave (Perlès 1987) dated to 10.8–10.2 ka BP (Perlès 1995). However, on the island of Melos, traces of occupation that would correspond to the Late Palaeolithic have not been discovered. If obsidian blades and tools from the KT-21 site on Kythnos were associated with the Gravettian or the Epigravettian, this would be an argument in support of even earlier exploitation of Melian obsidian (Figure 2.2). At the very beginning of the Holocene, in the eastern part of the Mediterranean Sea Basin we are dealing with the differentiation of the Epigravettian province into several distinct zones (Figure 2.3): 1. In the Adriatic Basin (Italy, western Balkans) the Sauveterrian develops (Broglio 1980; Biagi 1991, 2003; Martini and Tozzi 1996). 2. In the central and eastern Balkans together with continental Greece the local Epigravettian tradition continues to function with technological changes, i.e. a greater importance attached to flake technology and splintered technique, and stylistic changes manifested in the smaller role of microliths (which in this region played a vital role at the very end of the Palaeolithic: Koumouzelis et al. 2003), and an increased role of denticulated-notched tools. These phenomena show some similarity to the process of formation of the Lepenski Vir culture (Kaczanowska and Kozłowski 2003). 17

M. Kaczanowska, J.K. Kozłowski

Figure 2.1. Map of the Late/Final Palaeolithic taxonomic units in the Eastern Mediterranean. 3. On the southern and western coast of Anatolia another Epigravettian group developed, namely the Antalyan (Yalçinkaya et al. 2002); similar assemblages are known from Lemnos Island and, probably, from the region of the Dardanelles Strait. 4. Around the western and eastern coasts of the Black Sea as far as the Marmara Sea Basin there are industries with the ‘steppe’ Epigravettian tradition (Gatsov and Özdögan 1994; Gatsov 2001). 5. The fifth zone is the Aegean islands where the Epigravettian tradition (Kaczanowska and Kozłowski 2006) underwent specific modifications resulting from, on the one hand, insular isolation and, on the other hand, the effect of sea-borne communication over large distances that is documented, among others, by the circulation of Melian and Ghialian obsidian. This zone is described as the ‘Aegean Mesolithic’ (Kozłowski and Kaczanowska 2009). So far the ‘Aegean Mesolithic’ has been identified in the Cyclades (Kythnos), in the northern Sporades (Youra) and north of the Dodecanese (Ikaria). Recent discoveries by A.A. Ammerman on Cyprus (Aspros and Nissi Beach) show that a similar tradition, with almost identical technology and stylistics of chipped stone artefacts preceded the Aceramic Neolithic of Cyprus (Ammerman et al. 2007, 2008). This tradition was, in all likelihood, later than layer 2 at the site of Aetokremnos (Ammerman and Noller 2005).

A characteristic feature of the ‘Aegean Mesolithic’ is the gradual replacement of blade technique by flake and splintered techniques, decreasing importance of microliths that were replaced by backed inserts on flakes, increased importance of denticulated-notched tools and retouched flakes whose proportion could be as much as half of the retouched tools in a given tool inventory. In terms of chronology the tradition of the ‘Aegean Mesolithic’ spans the period from the middle of the 9th millennium cal BC (Maroulas on Kythnos: Facorellis et al. 2010; Figure 2.4) until the 8th/7th millennium cal BC. The latter time interval is documented by finds from the lowest layer at Knossos on Crete (Kaczanowska and Kozłowski 2011). 1 It is noteworthy that elements of the ‘Neolithic package’ (together with the semi-sedentary or, even, sedentary way of life) appear not only in the final phase registered on Crete, where their broadest range was recorded, but some of these elements appear already in the early phase known from Maroulas. Thus, in the evolution of the ‘Aegean Mesolithic’ three stages can be distinguished: 1. The early stage dated at Maroulas on Kythnos in the first half the 9th millennium cal BC. Because of the large standard deviation of the obsidian hydratation dates from Kerame on Ikaria, we can only hypothetically ascribe this site to the same chronological stage as Maroulas (Sampson et al. 2008). 2. The late stage (9th–8th millennium cal BC) represented primarily by the upper layers in the Cyclope Cave on the 18

Mesolithic Obsidian Networks island of Gioura in the northern Sporades (Sampson 2008), and eventually, unpublished until now surface sites, explored by A. Sampson on Naxos and Chalki. 3. The stage transitional to the Neolithic represented by the lowermost, aceramic layers at Knossos on Crete, dated to the turn of 8th/7th millennium cal BC.

Figure 2.2. Circulation of Melian obsidian during the Late/Final Palaeolithic.

Obsidian deposits and methods of their exploitation Thanks to the investigations carried out by C. Renfrew et al. (1965) and subsequent surveys and analyses (Shelford et al. 1982; Torrence 1986) obsidian outcrops on the island of Melos (Figure 2.5) have been relatively well explored. Two main outcrop areas have been distinguished connected with extensive workshops: the zone situated on the northern coast of the Milou Bay on the slopes of the Nichia plateau (in the vicinity of a locality of Adamantas), and on the southern side of the Milou Bay in the mountainous part of the island, around Dhemenegaki (Figures 2.6–2.7). The former deposit area is easily accessible from the seaward side, while the latter is situated on a plateau with plunging cliffs on all sides. Analyses of finds from the Lower Mesolithic layers (lithic phase VII) and the Upper Mesolithic layers (lithic phase VIII) in the Franchthi Cave using neutron activation analysis have shown that all the specimens examined come from the deposits near Adamantas; the only exception is a specimen from the Upper Mesolithic layer (A242) which may come from the vicinity of Dhemenegaki (Renfrew and Aspinall 1990). It should be added that Neolithic specimens from Franchthi come, in all likelihood, from the latter deposit. Surface surveys indicate that obsidian processing – the traces of which are huge workshops on the slope of the Nichia massif and in the area around Dhemenegaki – aimed at manufacture of macroblade cores and bifacial tools by pressure technique; therefore the workshops are unquestionably of Neolithic or Early Bronze Age date. In the Nichia massif extensive areas are covered with waste or failed cores; moreover, the superimposition of such areas

can be clearly seen in some profiles on the slopes of Nichia. In all these workshops, from the lowest to the uppermost, the technique of processing evidences the Neolithic or the Bronze Age. Around Dhemenegaki strong erosional processes meant that remains of workshops were strewn widely on the slopes and do not form distinct concentrations. So far the exploration of remains of workshops on Melos has provided no evidence of pre-Neolithic extraction. The only traces of the presence of a Mesolithic population could be the surface materials collected at the site of Agios Sostos by Evangelina Maniami on the southern side of the Milou Bay (Figure 2.8). This site provided, among others, two short end-scrapers (one of them on a splintered piece) and a thick, bilaterally retouched denticulated tool. These specimens are diagnostic of the Aegean Mesolithic. At Early Mesolithic sites, e.g. on Kythnos, obsidian was brought as small, partially decorticated concretions, or as pebble fragments usually not larger than 3cm in diameter. The extraction of obsidians from primary deposits, i.e. blocks or concretions within tuffitic matrix or its detritus can be ruled out. In these forms obsidian occurs frequently in the Nichia area as well as around Dhemenegaki. Its concretions reach considerable size. For these reasons it seems more plausible that the Mesolithic population exploited secondary deposits on the beaches near Adamantas, especially epiclastic deposits that today occur at a fairly shallow depth below sea level, and which in the Early Holocene were accessible on the broad continental shelf (Figure 2.9). A total of 245 pebbles and chunks were recovered from an area of 1m2 on the seabed near the shore. These fall into the following size classes: