Walsh The Archaeology of Mediterranean Landscapes

The Archaeology of Mediterranean
Landscapes
This volume presents a comprehensive review of palaeoenvironmental evi

dence and its incorporation with landscape archaeology from across the
Mediterranean. A fundamental aim of this book is to bridge the intellectual
and methodological gaps between those with a background in archaeol
ogy and ancient history, and those who work in the palaeoenvironmental
sciences. The aim of this volume is twofold: first, to provide archaeologists
and landscape historians with a comprehensive overview of recent palaeo
environmental research across the Mediterranean, and second, to consider
ways in which this type of research can be integrated with what might be
considered ‘mainstream’ or ‘cultural’ archaeology. This volume takes a
thematic approach, assessing the ways in which environmental evidence is
employed in different landscape types, from coastal zones via rivers and
wetlands to islands and mountainous areas. This volume also presents ana
lyses of how people have interacted with soils and vegetation, and revisits
the key questions of human culpability in the creation of so-called degraded
landscapes in the Mediterranean. It covers chronological periods from the
Early Neolithic to the end of the Roman period.
Kevin Walsh is senior lecturer in the Department of Archaeology at the

University of York. He has completed research in the Southern French Alps,
the Roman mill at Barbegal near the Camargue, the Sainte Victoire Mountain
near Aix-en-Provence, and at Stymphalos in the northern Peloponnese. He
is co-editor of two books: Interpretation of Sites and Material Culture from
Mid-High Altitude Mountain Environments and Mediterranean Landscape
Archaeology 2: Environmental Reconstruction in Mediterranean Landscape
Archaeology.
The
Archaeology of
Mediterranean
Landscapes
Human–Environment
Interaction from the Neolithic
to the Roman Period
Kevin Walsh
University of York
32 Avenue of the Americas, New York, NY 10013–2473, USA
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9780521853019
© Kevin Walsh 2014
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2014
Printed in the United States of America
A catalog record for this publication is available from the British Library.
Library of Congress Cataloging in Publication data
Walsh, Kevin, 1963–
The archaeology of Mediterranean landscapes : human–environment interaction from the
Neolithic to the Roman period / Kevin Walsh, University of York.
pages cm
Includes bibliographical references and index.
ISBN 978-0-521-85301-9 (hardback)
1. Human ecology – Mediterranean Region. 2. Landscape archaeology – Mediterranean
Region. 3. Human geography – Mediterranean Region. 4. Excavations
(Archaeology) – Mediterranean Region. 5. Mediterranean Region –
Civilization. 6. Civilization, Ancient. I. Title.
GF541.W35 2014
937–dc23 2013007621
ISBN 978-0-521-85301-9 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy

of URLs for external or third-party Internet Web sites referred to in this publication,
and does not guarantee that any content on such Web sites is, or will remain,
accurate or appropriate.
For Flo, who opened my eyes to so much … and kept me going by
regularly asking, ‘C’est quand que tu termines ce livre…?’
Contents
Acknowledgements page xi
1 Introduction 1
Mediterraneanism 2
Frameworks for the assessment of human–environment
engagements 4
Environmental knowledge and cultural ecologies 7
2 From Geology to Biology: Defining the Mediterranean 10
Fundamental geological and biological characteristics 10
Basic climatic and biogeographical characteristics 11
Tectonics and the creation and destruction of niches 14
Hellenistic and Roman catastrophes 24
Tectonic legacies 27
3 Sea-Level Change and Coastal Settlement – Human
Engagements with Littoral Environments 30
Introduction 30
Characterising the Mediterranean 30
Maritime processes 31
Coastal processes 33
Examples of Mediterranean coastal change 37
Higher-energy events 38
Pre- and proto-historic coastal exploitation 39
Coastal exploitation: The development of ports and harbours
(Bronze Age onwards) 46
Early ports and harbours 49
Classical coasts and harbours 51
Controlling permeability 64
4 Rivers and Wetlands 68
Studying Mediterranean rivers and wetlands: Research
questions and approaches 69
vii
viii Contents
Characteristics of Mediterranean rivers 70

Springs and karst 78
Wetlands 79
Alluvial geoarchaeology: People and climate 81
Alluvial landscapes and farming in Anatolia and Greece 85
Aspects of alluvial archaeology in Italy 88
Late proto-historic and classical alluvial and hydrological
landscapes 92
Urban alluvial geoarchaeology: Glanum, Rome, and
Gordion 94
Glanum 94
Rome 96
Gordion 100
Environmental knowledge in dynamic alluvial and
wetland zones 101
Human engagements with Mediterranean wetlands 104
Hydromythology 109
The Pontine Marshes: Roman ‘relationships’ with a
wetland 111
Wetlands and disease 113
Discussion: Responses to hydrological variability 115
5 Environmental Change: Degradation and Resilience 119
Approaches and research questions 119
The phytological context 121
The Fall from Eden 122
Landscape change around the Mediterranean 122
Anthropogenic and climatic impact: Views from around
the Mediterranean 133
Southern Mediterranean 135
The Near East 137
Anatolia and Greece 140
People and environment in Italian landscapes 149
People and environment in southern French landscapes 153
People and environment in Spanish landscapes 159
6 Working and Managing Mediterranean Environments 171
Lifeways in Mediterranean environments 171
Clearance, terracing, and the creation of the sustainable
Mediterranean landscape 172
The role of fire 181
Agricultural and productive vegetation 183
Woodland and landscape management (Dehesa and
other systems) 189
Environmental change and social geoarchaeology 195
The Sainte Victoire: Changing patterns of interaction
with environment 196
Contents ix
The Roman watermill at Barbegal 201

Discussion: The human scale of interaction with past
environmental processes 204
7 Islands: Biogeography, Settlement, and Interaction 210
Introduction 210
Mediterranean islands and island biogeography 213
Colonisation of Mediterranean islands 215
Settlement, economy, and insularity 218
Aspects of insular environmental and cultural change 219
Malta 219
Smaller islands 224
The Balearics 231
Discussion: Are islands different? 235
8 Mountain Economies and Environmental Change 242
Introduction: Vertical spaces, cyclical time 242
Defining mountain landscapes 244
Human–landscape engagements across Mediterranean
mountains 247
Greece and Anatolia 250
Italy, France, and Spain (the Apennines, Alps, and
Pyrenees) 255
The Alps 257
Southern French Alps 259
The Pyrenees 265
Patchy porosity: Mediterranean mountains and variable
integration 268
Conclusion 277
9 Conclusions – The Mediterranean Mosaic: Persistent and
Incongruent Environmental Knowledge 280
Retrodicting human engagement with the landscape 282
Revisiting Mediterranean environmental problems 284
Mediterranean landscapes: Past, present, and future… 289
Bibliography 293
Index 355
Acknowledgements
Special thanks are due to several people for reading draft chapters. In par
ticular, Tony (A. G.) Brown for reading an entire draft, and the following
for their invaluable comments on various chapters: Eleni Asouti, Geoff
Bailey, Andrew Bevan, John Bintliff, Will Fletcher, Helen Goodchild,
Allan Hall, Bruce Hitchner, Bernard Knapp, Caroline Malone, Christophe
Morhange, and Terry O’Connor.
I am indebted to friends and colleagues in the Department of
Archaeology at the University of York for their support over the years,
and also to the University of York for awarding me an anniversary lecture
ship grant that permitted extended research leave for the initial research
for this book.
I would also like to thank friends and colleagues in the Centre Camille
Jullian (Aix-en-Provence) for their continuing support; in particular the
Bibliothèque d’Antiquité d’Aix has been one of the key resources for
the research for this volume. I am also indebted to Philippe Leveau who
directed my postdoctoral research at the Centre Camille Jullian and has
continued to support me ever since.
Thank you to the following for providing and/or helping with fig
ures and photos: Greg Aldrete, Jean-Francois Berger, Andrew Bevan,
Philippe Boissinot, Tony (A. G.) Brown, Karl Butzer, Shirley Cefai, Gaëtan
Congès, Jacques-Louis de Beaulieu, Carlo Giraudi, Itamar Greenberg and
Ehud Galili, Fred Guiter, Bruce Hitchner, Nick Marriner, Anne Mather,
Berengere Perez, Lisa Rayar-Bregou, Santiagio Riera-Mora, Dorit Sivan,
Cynthianne Spiteri, Iain Stewart, Stathis C Stiros, Jon Swogger (and the
Çatalhöyük Research Project), Sebastian Vogel, Jamie Woodward, and
Eberhard Zangger. I would particularly like to thank Kieron Nieven for
xi
xii Acknowledgements
his work on many of the figures in this volume, and Gordon Wallace for
help with copy-editing.
Special thanks go to Jean-Marie Gassend for allowing me to use (and
convert) his watercolours into black-and-white images. I would also
like to thank my copy editor, Luane Hutchinson, for her work on the
manuscript.
1
Introduction
One motivation for writing this book is to bridge the conceptual and
methodological gaps for those with a background in archaeology and
ancient history, and those who work in the palaeoenvironmental sciences;
different groups of researchers who all share a passion for Mediterranean
landscapes. Therefore, the aim is twofold: to provide archaeologists and
historians with a comprehensive overview of recent palaeoenvironmental
research across the Mediterranean, and second, to consider ways in which
this research can be integrated with what might be considered ‘main
stream’ or ‘cultural’ archaeology. This synthesis is structured in such a
way that readers can ‘jump’ to the geographical or thematic sections
that are of particular interest to them. In addition, the landscape the
matic approach (with each chapter addressing a landscape type or con
nected themes) is designed to provide readers working in or researching
a given landscape type access to modern environmental studies in those
areas. Therefore, most of the chapters in this book follow a similar form.
The first section in each chapter provides an overview of how each land
scape/environment type has been studied, followed by a resumé (which
is largely descriptive) of the principal findings of this research. Finally,
latter sections of most chapters provide integrated assessments of some
archaeological and palaeoenvironmental projects from across the region.
The aim is not to define a sequential development of the Mediterranean
environment and its peoples; this book is more concerned with the ways
in which different peoples have interacted with different landscapes at
different times. The examples comprise case studies from the beginning
of agriculture to the end of the classical periods. This time span has been
chosen, as much archaeological (especially landscape survey) and palaeo
environmental research focusses on this chronological range. That is not
1
2 Introductio
to say that there is a dearth of medieval evidence. In fact, another rea

son for ending the temporal perspective of this volume at the end of the
Roman period is also due to the author’s own chronological research
interests.
Mediterraneanism
The Mediterranean is the only region in the world that gives its name to a
climate type. Although this volume is concerned with the Mediterranean
geographical region, Mediterranean environments exist in California,
Chile, the Cape (South Africa), as well as South and Western Australia
(Allen 2001: ch. 1). Consequently, the cultural significance of studying
Mediterranean environments is of global relevance.
There have been many helpful discussions of Mediterraneanism in
recent years, most notably in the book edited by William Harris (2005b).
The key point is that, after much debate, most people who carry out
Mediterranean research believe that a pan-Mediterraneanist framework
is reasonable and useful, in part due to the shared environmental charac
teristics, but also because of the obvious connected histories and cultural
developments across the region.
At one level, the sheer variety of landscapes across the Mediterranean
(a region where Europe, Africa, and Asia meet) implies that there cannot
be a singular Mediterranean. However, there is a set of similar envir
onmental characteristics, in particular, similar geological structures and
climatic cycles. There are, of course, important fluctuations in average
temperatures, precipitation, and vegetation. However, such variations
are not just spread across the region as a whole but can occur within
subregions due to considerable local variations in topography. These fea
tures are considered in Chapter 2.
If one could provide a straightforward definition of a typical
Mediterranean environment, we would emphasise the dramatic differ
ences in landscape forms that exist within relatively small spaces. Between
the Alps in the north and the Atlas mountains in the south, there are
plains, wetlands, arid zones, forests, barren lands, and, perhaps most
importantly, an incredible variation in coastal landforms, and within the
sea, there are of course the islands. If we were to draw a transect across
any part of the Mediterranean region, most if not all of these landscape
types would be available – this sequence or group of landscapes is what
Mediterraneanis 3
defines the Mediterranean. Most of these environments are dealt with

via thematic chapters. Coastal environments are presented in Chapter 3,
with an assessment of background changes in sea level, and an analysis
of their variation across the Mediterranean. The ways in which different
societies have engaged with coasts and the sea is dealt with in the sec
ond part of that chapter. Alluvial and fluvial systems are the subjects of
Chapter 4. Here, descriptions of alluvial processes are presented along
with studies of how Mediterranean people have engaged with rivers
and wetlands. Although a related issue, the problem of aridity and areas
where water supply is restricted or unpredictable is part of the subject
matter of Chapters 5 and 6, where erosion, soils, and wider issues in
Mediterranean geoarchaeology are considered. This analysis incorporates
the assessment of vegetation histories and human engagements with veg
etation from the Neolithic through to the Roman period. An overriding
theme (discussed by others, e.g. Grove & Rackham 2001) is the notion
of landscape degradation or the ‘Fall from Eden’. This Genesis myth is
founded on the notion that people in the past adopted an instrumental
attitude to the landscape (i.e. exploited it without always caring for it),
and this negligence was punished (in a codified form) via the story of the
expulsion from Eden. Chapters 7 and 8 consider the range of processes
discussed in the preceding chapters, but develop specific assessments
of the ‘bounded’ and quintessentially Mediterranean islands and then
mountains.
As the chapters unfold, the reader will probably appreciate that
any notion of a single Mediterranean, with homogenous responses to
similar environmental processes and common economic strategies, is
largely misplaced (J. G. Manning & Morris 2007). If we accept that
each environmental niche and its constitutive processes have a role as
a non-human agent (Latour 1997), contributing to the development
and continual reconstitution of cultures, then, on that basis alone,
we cannot argue for a homogenous Mediterranean culture and inte
grated systems of environmental manipulation. However, it is possible
to idealise a particular type of ‘typical’ Mediterranean physical geog
raphy. This idealised Mediterranean is sometimes conceptualised as a
framework over which variations in cultural development are evident,
but where the environmental framework apparently influences these
variations. As French historian Henry Laurens (2010: 59–60) suggests,
with chronological variations from area to area, Mediterranean peoples
4 Introductio
have experienced the same processes that have profoundly transformed

the Mediterranean landscape. Agricultural production expresses these
shared traits with similar crops (e.g. wheat, olive, and vine) and land
scape features (such as terracing). The ways in which people engage
with the environment are influenced by the possibilities that occur nat
urally within a given space. However, the form of human engagement
with that space is contingent upon a wide range of cultural processes.
Each environmental niche is characterised by its potential and its limita
tions. The manner in which different peoples impose their layered cul
tural values on that environment, and develop their awareness of nature,
clearly influences the ways in which landscapes evolve. Whilst any kind
of environmental determinism is quite understandably frowned upon,
we cannot underplay the impact of structural geology (topography),
climate, vegetation, and hydrology on settlement, economy, ideology,
myths, and culture across the Mediterranean. The complex tectonic
processes and extreme topographical variations within relatively small
spaces have always had a profound effect on where people can live and
work. Anyone who has travelled along the coastline of Italy, Greece,
and the larger Mediterranean islands cannot have failed to notice the
ever-present mountain ranges in the middle distance, and the sheer cliffs
dominating much of the coast. Large portions of these coastlines are
uninhabitable or, at best, unsuited for agriculture or even pastoralism.
Settlements are nested in the areas adjacent to faults or relatively flat
zones that have evolved as rivers and streams, which in turn have depos
ited sediments and yielded a more ‘useful’ environment. If we accept
that the processes that characterise these heterogeneous environments
contribute to the construction of lifeways and culture in their broadest
sense, then past human experiences and activities in these landscapes
cannot be assessed via material cultural alone; we should also consider
environmental processes in the development of a ‘symmetrical anthro
pology’ (Latour 1997).
Frameworks for the Assessment of Human–Environment

Engagements
When we think about how people have interacted with an environ

ment over time, we often consider the choices that they made regard
ing settlement location, landscape management (both in terms of the
Frameworks for Assessment of Engagements 5
geomorphic system and vegetation), choice of crops, extraction of rocks

and minerals, and the impact that these activities had on the landscape.
Archaeology and environmental archaeology are often concerned with
researching and explaining change in cultural and environmental sys
tems. Quite logically, we are interested in identifying and explaining
changes in society and landscape and, most importantly for this vol
ume, changes in the environment. Environmental archaeologists look
for impact on the vegetation system or phases of erosion. When we
do identify a phase of environmental stability, such a phase is identi
fied because it represents a change or a rupture. However, periods of
stability and maintenance of specific activities are just as important. We
need to consider why people chose to settle and establish settlements in
particular topographic situations within a given landscape. Many settle
ments have been continually occupied for centuries and even millennia.
Briefly abandoned extant settlements often attract reuse; in some land
scapes, we may want to consider different forms of inertia, as well as
processes of change and adaptation. One relevant notion that has seen
much discussion in recent years is ‘resilience’ (Butzer 2005; Redman
2005). Here, resilience theory does not imply that environments auton
omously maintain equilibrium, but that societies who engage with these
environments develop strategies for ensuring the persistence and prod
uctivity of a given niche.
We are often told that humans adapt to changes in the environment.
What do we mean by this, and how can we be sure what adaptations were
made? The environmental processes that we measure (eroded sediments,
proxy data for vegetation change, etc.) are not necessarily representative
of events witnessed by and responded to by past peoples. Whilst evidence
for aridification caused by climate change, in the form of a reduction in
precipitation, might induce settlement shift over a relatively long period,
how do we demonstrate responses to shorter-term events, such as soil
erosion? Substantial sedimentary units can be deposited by a few severe
storms – events that would have been a common occurrence even within
ostensible phases of environmental stability.
Modern, Western notions and perceptions of the environment are
regularly informed by instrumental economic philosophies. Modern sci
ence, with its roots in the Enlightenment, employs a discourse; a way of
interpreting and discussing the world that is so different to the numer
ous forms of environmental understanding that would have existed in
6 Introductio
the past. In order to engage with this issue, the final section of most
of the core chapters will examine how past societies may have engaged
with the environmental processes that we believe are significant and
relevant.
The other modern view of nature/landscape is one dominated by
a Romantic aesthetic (see Johnson 2007: ch. 2). Again, this is in part
a consequence of a disembedded relationship with the natural world:
landscapes are places that we visit and engage with at an ideological level
where perspectivism is all-important. This relationship with landscape
characterises certain postprocessual approaches in landscape archaeology,
in particular, phenomenological strands (Tilley 1994); approaches that
appear to be underpinned by a Romantic notion of the countryside as
destination and distraction, rather than a place of work and engagement
with the sometimes harsh realities of the natural environment (Bintliff
2009; Flemming 2006). These approaches are often more detached
from the reality of past lifeways than the environmental science that
they often attempt to critique. Such approaches are not as common in
Mediterranean archaeology (for an exception, see Hamilton et al. 2006),
where emphasis is placed on assessing human impact on the environment
or the economic potential of a landscape, and how this might have varied
with climate change and/or human impact.
As a number of recent works have demonstrated, a significant under
lying theme in Mediterranean landscape archaeology is the notion of
the ‘Fall from Eden’, or the culpability of humanity in the destruction
of a once supposedly pristine landscape (Grove & Rackham 2001).
Recent narratives also attempt to demonstrate how the characterisation
of Mediterranean environments as marginal and degraded has been mis
placed. Horden and Purcell (2000: ch. V) believe that whilst certain
Mediterranean niches are not always productive in isolation (in the sense
that they easily generate surpluses), once we see the different niches as
nodes within an integrated network of production, the whole is so much
greater than the sum of the parts. Whilst these more recent frameworks
are useful, we also need to consider how different groups in different
societies in the past engaged with these landscapes. For example, some
societies saw their relationship with nature as a conflict or battle, such
as that which might have been held in Mesopotamian society (Hughes
1994a: 34).
Environmental Knowledge and Cultural Ecologies 7
Environmental Knowledge and Cultural E

cologies
Rather than provide a comprehensive overview of the development of

cultural and historical ecology (dealt with in a number of publications,
e.g. Balée 2006; Crumley 1994b; Meyer & Crumley 2012; Sutton &
Anderson 2004), this section identifies some key tenets that underpin
the approach adopted in this book.
The origins of most human ecological strands of thought lay with cul
tural ecology, which is directly associated with functional anthropology
(Steward 1955). Cultural possibilists who worked within a cultural eco
logical framework suggested that certain peoples, in particular, hunter-
gatherers, were constrained by their environments. Steward in particular
developed these ideas and moved towards assessments of cultural evo
lution, emphasising adaptation and stability with the investigation of
change in hunter-gatherer groups in North America (Bettinger 1991:
44–5). Some of these ideas were then adopted by archaeologists, and the
fact its use is often associated with an under-theorised form of processual
archaeology should not detract from the value of approaches that adopt
a cultural ecological framework. One notion, which was applied by some
archaeologists, was the culture-area concept, whereby technologies and
human lifeways were apparently correlated with the nature of the envir
onmental context within which societies developed (Clark 1968).
An early example of an unsophisticated cultural ecological interpreta
tion of a historical process was the contention that the fall of Rome was
an ecological catastrophe partly caused by a misuse of resources resulting
from poor knowledge or information (Sutton & Anderson 2004: 3).
As argued at certain points in this volume, what is more likely is that, at
certain times and places, the environmental knowledge, articulated via
macro-political and economic forces, was at odds with the environmental
experiences and concomitant knowledge of the peoples who lived and
worked in these different landscapes.
Cultural ecology assesses environmental knowledge, that is, how
people understand and engage with their landscape and environment.
The notion of adaptive strategies, where groups of people develop
technologies that facilitate life and, in particular, food production in a
given environment, is important. D. O. Henry’s (1994) work in south
ern Jordan is one example of such an approach. This type of approach
8 Introductio
does not assume that technologies and human lifeways will be repeated
in landscapes characterised by identical or similar sets of environmental
characteristics. As noted above, Mediterranean cultures do share certain
forms of landscape-management strategies, but these strategies are con
tingent upon historical, cultural, and economic processes that vary across
time and space. Responses to changes in the environment do tend to
be controlled by the ability of social institutions to adapt. As Bettinger,
Richerson, and Boyd (2009) suggest in their assessment of constraints
on the development of agriculture, it was the gradual evolution of cer
tain social institutions that limited the speed of the uptake of farming
in some regions. A key question is how was environmental knowledge
applied in the past, and by whom? People are not separate from ecologi
cal systems; they are participants in environmental processes, and as such,
human participation in environmental change is quite natural (Walters
& Vayda 2009: 536). At a wider level, a cultural ecological approach
can also inform the study of landscapes where there is a dearth of mate
rial evidence, or in landscapes that are considered difficult to manage
and in some ways ‘unattractive’, such as arid zones or mountains. Here,
the premise is that each society’s engagement with the environment is
dynamic. Consequently, if we can elucidate the manner in which past
peoples manipulated and responded to their environments, then this is
an effective scheme for the investigation of past cultures and the transi
tions or changes in culture across a given landscape. Finally, resilience
theory offers a way of conceptualising the relationships between different
spatial and temporal scales of cultural processes (Redman 2005). Here,
resilience theory presents a scheme for investigations of the relationships
between small-scale, localised groups of people (e.g. individual farms)
and how they relate to extensive hierarchical structures (e.g. the Roman
Empire or its regional authority). Of most interest is the notion that
successful environmental exploitation strategies only work if people can
adapt. However, if local engagements with environments are controlled
by entrenched political forces during periods of environmental change,
and local people are unable to respond effectively to these changes, then
such a situation might contribute to local and regional societal insta
bility. When local, potentially small-and-fast, adaptive strategies are sti
fled by slow-responding, large-scale hierarchies, such as certain empires,
then environmental problems might ensue. Conversely, certain hierar
chical organisations might impose or apply new forms of environmental
Environmental Knowledge and Cultural Ecologies 9
management that are successful and enthusiastically adopted by local

people. It is the appreciation of different forms of human ecology that
allows us to move away from the original conceptions of cultural ecology,
perhaps best characterised by the definition of ‘culture areas’ (Steward
2005).
In summary, the discussions developed in some of the following
chapters are informed by the frameworks considered in this chapter.
The ultimate aim is to identify trends and trajectories in Mediterranean
landscapes from the Neolithic to the Roman period, sometimes offering
resumés of published environmental research and presenting syntheses of
this type of research with related archaeological information. Part of the
approach includes the evaluation of the range of human–environment
interactions across the Mediterranean, where environmental evidence
can be deployed in assessments of human–environment engagements,
and, where possible, to consider scenarios where variations in forms of
environmental knowledge could have been responsible for stresses, rup
tures, and resilience in the wide range of cultures that have lived and
worked in these dynamic landscapes.
2
From Geology to Biology: Defining the Mediterranean
This chapter comprises a brief overview of the geological and biogeograph-

ical contexts of the Mediterranean. There is also a brief description of cat-
astrophic processes, and some analysis of associated human responses to
these phenomena. As the principal aim of this volume is the discussion of
more mundane environmental processes, the assessment of catastrophic
events is kept to a minimum, partly because there are a number of special-
ised volumes that deal admirably with these processes, and a single chapter
cannot do justice to this increasingly popular area of research (Ambraseys
2009; Balmuth, Chester, & Johnston 2003; Nur & Burgess 2008).
Fundamental Geological and Biological Characteristics
The Mediterranean is defined in part by its geology. The Iberian, Eurasian,

Arabian, and African plates; their associated faults; and mountain chains
situated within relatively short distances from coastlines explain the enor-
mous variation and complexity of Mediterranean landscapes. This geo-
logical crossroads is also important from a biological perspective, as flora
and fauna (including humans) have moved from Asia and Africa, and
then onwards between the Near East and Europe.
There are many books that provide detailed descriptions of the geo-
logical foundations and processes that characterise the Mediterranean
(e.g. Dixon & Robertson 1996; Jolivet et al. 2008; Stanley & Wezel
1985). In simple terms, the Mediterranean comprises boundary zones
between the Eurasian, African, and Arabian plates (Allen 2001: 48).
Consequently, much of the Mediterranean comprises undulating topog-
raphy and mountainous areas. The eastern half of the Mediterranean
possesses a series of active fault lines that also include dormant and active
10
Fundamental Geological and Biological Characteristics 11
volcanoes (Mather 2009) (Fig. 2.1). This half of the Mediterranean is

much more tectonically active than the western half, with Italy and Sicily
constituting the boundary. To a certain extent, the distribution of volca-
nic activity follows this arc, as well as the Calabrian Arc, which includes
Vesuvius. This geological configuration leads many authors to character-
ise the Mediterranean region as inherently unstable, with this instability
contributing to degradation processes and the evolution of reputed mar-
ginal landscapes or niches. As with the other environmental processes
assessed in this book, the relative impact that geological processes have
had on human societies in the past has been the subject of much debate
in recent years (Grove & Rackham 2001; Horden & Purcell 2000).
Major tectonic events tend to be high-magnitude, low-frequency
events, but these processes then determine the conditions for long-term,
low-magnitude processes, such as soil erosion. First, there are the char-
acteristics of the solid geological layer, which for the sake of simplifica-
tion can be divided into alkaline and acidic categories. Despite the fact
that limestone is omnipresent, most Mediterranean subregions incorpor
ate a wide range of rocks, including sandstones, granites, lavas, flysch,
and shales. Soft rocks, including conglomerates, marls, and clay, are also
common. The intrusive igneous and metamorphic rocks comprise min-
erals or rock types that have been economically valuable in the past, such
as marble, jadeite, or obsidian. The geological substrates influence the
properties and evolution of soils and their potential for erosion. An ero-
sion pattern is then influenced by climate (phases of relative humidity or
aridity), vegetation development, and pedogenesis.
Basic Climatic and Biogeographical Characteristics

The Mediterranean possesses climatic characteristics that are quite rare
in that there are relatively few regions on the planet that experience such
dramatic seasonal shifts in dominant pressure systems and concomitant
weather patterns (Harding, Palutikof, & Holt 2009: 69). There are, of
course, variations across the Mediterranean. The western area is influ-
enced by Atlantic systems, and seasonal temperature variations are less
severe than in the east, this area being influenced by central European
and Asian systems. The south-eastern area is the most arid. Within the
Mediterranean, relatively small and localised depressions can develop.
This is more common in the central to north-eastern zones, the most
famous being the Gulf of Genoa depressions. The important aspects of
2.1. The principal geological structures of the Mediterranean. (By permission of Oxford University Press, Mather, A. (2009), Tectonic setting and landscape development,
in J. Woodward (ed.), The Physical Geography of the Mediterranean, fig. 1.1(b), ch. 1, p. 6.)
Fundamental Geological and Biological Characteristics 13
NE
NW
SW
SE

2.2. The biogeographical zones of the Mediterranean (after Blondel et al. 2010).
climate change comprise adjustments in average seasonal temperature

and precipitation levels. The Mediterranean climate can be considered as
transitional between cold temperate and dry tropical, with the northern
part of the region characterised by colder, windier conditions than the
south. The other key defining characteristics of the Mediterranean biocli-
matic regime are the distribution and timing of rainfall, with the eastern
area often experiencing five to six months each year without rainfall, as
opposed to two months in the western area (Blondel et al. 2010: 12).
As discussed in the previous chapter, defining the boundaries of the
Mediterranean is a difficult task, but the distribution of plants such as
olive and holm oak with sclerophyllous leaf structures (thick leaves that
preserve moisture) is one widely agreed bioindicator. However, as with
all organisms, its geographical distribution has fluctuated. Consequently,
the combination of climatic and biological characteristics and the cre-
ation of bioclimatic criteria is one of the most useful frameworks for
the definition of the region. The intersection of these different features
then constitutes the biogeographical characteristics. The four princi-
ple biogeographical zones are the four quarters of the Mediterranean,
that is, north-west, north-east, south-east, and south-west (Blondel &
Aronson 1999; Blondel et al. 2010) (Fig. 2.2). Average temperatures
across these biogeographical zones do of course vary. Whilst average
summer temperatures across the Mediterranean are usually between
14 From Geology to Biology: Defining the Mediterranean
25°C and 30°C, averages are higher and often go above these tempera-
tures in the eastern and southern zones. This trend in higher average
temperatures in the eastern and southern zones is also true for average
winter temperatures. Topographic situation also effects average tem-
perature, with these averages being lower at higher altitude. For exam-
ple, Madrid, at 667 m, has mean January temperatures between 2.7°C
and 9.7°C, whilst Valencia, at 11 m, has corresponding temperatures of
7°C and 16.1°C (AEMET; Allen 2001). These variations in climate and
topographic situation clearly affect soil, plant, and animal distributions.
As Blondel et al. (2010: 103–13) have demonstrated, the most effec-
tive way of appreciating biogeographic variability within a zone is to
examine a transect from the coasts towards the interior high altitudes.
In Blondel’s north-west transect (the south of France), we move from
coastal wetlands (the Camargue; discussed in this volume, Chapter 4),
through the hilly zones with the evergreen shrublands (elements in the
discussion of Chapters 5 and 6), through to the interior (often moun-
tainous) areas with mixed evergreen vegetation (part of the discussions
in Chapters, 4, 5, and 8). In Blondel’s south-eastern Quadrant (the
Lebanon), the coastal zone (Chapter 3) comprises dune-tufted grasses,
then, as we move into the hinterland, evergreen woodlands are present.
Then, in the more mountainous zones, the vegetation comprises decid-
uous oak and pine woodlands, with cedar and juniper above these wood-
lands. Moving down, towards the east, a more arid-type Mediterranean
landscape is evident, comprising shrublands dominated by wormwoods.
Lower down, the Beqa’a Valley is a fertile cultivated zone. Even further
to the east, we come to desert.
The characterisations presented above refer to current climate data.
There have of course been a number of significant climatic fluctuations
in the Mediterranean during the Holocene, and these are considered at
different points in the book. Most importantly, the recent palaeoclimatic
research now allows us to understand the development of these regional
climate regimes during the Holocene (Brayshaw, Rambeau, & Smith
2011; Finné et al. 2011; Magny et al. 2011).
Tectonics and the Creation and Destruction of Niches
Although tectonic processes and associated volcanic activity can damage

parts of landscapes, render coastal sites unusable, and lead to loss of life,
Tectonics and the Creation and Destruction of Niches 15
we should not forget that tectonic activity creates the undulating, varied
topography that structures the development of numerous environmental
niches that are home to an extensive range of flora and fauna. This mosaic
characterises many parts of the Mediterranean, and has thus led to the
need for people to develop complex forms of environmental knowledge
that operate at the local level. This characteristic of Mediterranean land-
scapes is dealt with in a number of chapters. Here, the immediate conse-
quences of seismic activity are considered (Fig. 2.3).
Catastrophes, in the form of earthquakes and volcanic eruptions,
make for good television, popular books, as well as scholarly journals
(e.g. Keys 1999; Ryan & Pitman 2000; Silva, Sintubin, & Reicherter
2011; Wilson 2001). Catastrophism became unfashionable for some
time in archaeology, but in recent years, its study has taken on a higher
profile. The citation of catastrophic events as explanations for rup-
tures or significant changes in ancient societies, along with attempts
to identify the realities behind certain myths such as the Great Flood
or destruction of Atlantis, are central themes in such research. More
recently, Horden and Purcell (2000: 300) have appropriately observed
that ‘disaster history continues to fascinate, and needs no misleading
promotion by causal association with the language of popular science’.
Commenting specifically on the eruption of Vesuvius, they assert that
the productive landscape in this area was resilient (ibid.: 306). However,
resilience is time dependent in that evidence for people returning to
a landscape a generation after the catastrophe may appear swift, but
the generation that experienced the event may not have been able to
demonstrate resilience. One effect of such an eruption might have been
a change in topography as volcanic sediment covered the landscape.
However, whilst the volcanic sediments around Vesuvius are up to 15
m deep, the eruption blanketed the AD 79 topography and left a new
surface that essentially reflected the original topography (see Fig. 2.4 for
a map of sites referred to in this chapter). Despite the fact that the post-
AD 79 topography was, in a way, moulded onto the ancient landscape,
the volcanic material did have a significant effect on the surrounding
environment. These deep deposits would have had an impact on the
fluvial and alluvial elements across the landscape (Fig. 2.5) (Vogel &
Märker 2010; Vogel, Märker, & Seiler 2011).
One thread running through the following chapters is that of tem
poral scale and the extent to which we can develop discourses that
2.3. Major tectonic structures and associated seismicity of the Mediterranean. (By permission of Oxford University Press, Stewart, I., and Morhange, C. (2009), Coastal
geomorphology and sea-level change, in J. Woodward (ed.), The Physical Geography of the Mediterranean, fig. 13.1a&b, p. 387.)
Sea of Azov
Bay
of
Biscay
Black Sea
Gulf of
Lions
Adriatic
Sea
1
Balearic
Sea
Gulf of Aegean Sea
Tyrrhenian Sea Taranto
11
16 13
Ionian Sea 5
Strait of Gilbraltar
10
3 15 1412
2 4
Sea of Crete
78 96
Mediterranean Sea
Gulf of Sidra
2.4. Map of sites referred to in text. 1: Vesuvius, 2: Ayios Dhimitrios, 3: Voidokilia/Area of Messenia/Pylos, 4: Santorini, 5: Lake Gölhisar, 6: Mochlos, 7: Delphinos,
8: Phaistos, 9: Gouves, 10: Midea/Mycenae/Tiryns, 11: Pyrgos, 12: Cnidus, 13: Pamukkale, 14: Kos, 15: Sparta, 16: Delphi.
2.5. Vogel et al. post-AD 79 volcanic deposits of Somma-Vesuvius to reconstruct the pre-AD 79 topography of
the Sarno River plain (Italy). Top: present-day environment. Bottom: pre-AD 79 environment. (By permission
of Geologica Carpathica, Vogel, S., Märker, M., and Seiler, F. (2011), Revised modelling of the post-AD 79
volcanic deposits of Somma-Vesuvius to reconstruct the pre-AD 79 topography of the Sarno River plain (Italy).
Geologica Carpathica 62(1), 5–16, fig. 5.)
evaluate possible human engagements with various environmental pro-

cesses. Whilst we should accept that catastrophes cannot usually be
blamed or cited as direct causes of societal and cultural changes, we
cannot reject the notion that certain catastrophic events would have had
profound repercussions for individuals, families, villages, and even towns
at certain points in the past. Despite the fact that average earthquake
magnitudes across the Mediterranean are not large, the Mediterranean
has had a relatively high population density for much of its past, and
therefore the potential hazard/risk has been greater (Mather 2009: 11).
In addition, tectonic processes give rise to certain landscape qualities
that have deeply positive cultural connotations: the ‘structural insta-
bility’ of the Mediterranean also gives rise to features such as thermal
springs and sulphur beds that possess powerful and often-positive cul-
tural associations.
The historical and archaeological study of seismology and volcanism
is complex (Guidoboni & Ebel 2009; Jones & Stiros 2000), and there
are many problems with the chronologies of these events and thereby the
identification of their impact on specific human communities. Moreover,
the geographical spread of research is patchy, with some areas, such as
Algeria, not having benefited from much detailed study (Benouar 2004).
There is little doubt that some volcanic eruptions have had a profound
impact on communities: the immediate destructive power of high-mag-
nitude eruptions is doubtless. At the Bronze Age site of Ayios Dhimitrios
in south-west Greece, it appears that the site’s occupants fled without
having time to gather valuable items (Zachos 1996: 169). Some 60 km
away, a similar set of events have also been identified at Voidokilia during
the Early Bronze Age (3200–2050 BC). These sites may represent some
of the earliest evidence that we have for destruction by an earthquake,
including a direct impact on the population (ibid.).
What is more problematic is an assessment of the longer-term con-
sequences of these events. Even where we have ancient written sources,
in particular for the AD 79 eruption of Vesuvius, we cannot be sure of
the longer-term effects. The eruptions of Vesuvius and Santorini are the
two obvious candidates for discussion within Mediterranean archaeol-
ogy. Some would argue that the effects of some significant eruptions
were felt on a continental or even a global scale, with ash clouds blocking
solar radiation and reducing temperatures (Grattan & Gilbertson 2000;
LaMoreaux 1995).
The precise definition of the chronological framework is essential for

any wider analysis of the relationship of an event with wider sociopolit-
ical and cultural processes. There is probably no other single event in
Mediterranean archaeology that has engendered such an extensive and
detailed debate regarding its date than the eruption of Santorini/Thera
(Thera being one of the islands that constitute the Santorini group)
(S. W. Manning 1999; Zielinski & Germani 1998). The issue of chro-
nology is important, as there is the posited high date of 1628 BC for the
eruption (Kuniholm 1996; S. W. Manning 1999), and then a low date
of around 1520 BC (Warren 1996). The most convincing chronology is
that based on the analysis of a series of secure, radiocarbon dates which
places the eruption at 1650–1620 BC (possibly c. 1628 BC) (Bruins
et al. 2008; S. W. Manning et al. 2002). The eruption itself was a multi
phase event comprised of a Plinian explosive phase (an eruption type
named after Pliny the Younger’s description of the ash and gas column
from Vesuvius which reaches the stratosphere), followed by the base-
surge phase when seawater connected with the magma. The final phase
saw the crater wall giving way and clouds of hot ash and gas sweeping
outwards laterally across the island (Friedrich 2004: 75–6). Recent esti-
mates suggest that some 60 km3 of material would have been ejected
by the Theran eruption (this is half as big again as earlier estimates)
(Sigurdsson et al. 2006).
When we discuss the environmental impact of any phenomenon,
whether it is seasonal erosion or the consequences of a ‘catastrophic’
event such as the eruption on Santorini, we must consider what we mean
by ‘environmental impact’. Although it has been difficult to demonstrate
that Cretan sites were destroyed by tsunamis or even ash fallouts emanat-
ing from the Santorini eruption, it was argued that nuées ardentes (highly
heated gas clouds) could have reached eastern Crete (I. G. Nixon 1985),
whilst the tephra fallout may well have covered an area from the Black
Sea down to the south-eastern Mediterranean (Gulchard et al. 1996).
Santorini ash has been found across the wider region, from the Nile Valley
to the Black Sea (ibid.). There is evidence for the creation of homoge-
neous (redeposited) sediment – ‘homogenite’ – which has been found in
more than 50 deep-sea cores. This layer (up to 10 m thick on the abyssal
plains and more than 20 m to the west of the Mediterranean Ridge) must
represent a sudden, high-magnitude event (Cita, Camerlenghi, & Rimoldi
1996: 158), which some authors argue to be the Santorini eruption (Cita
& Aloisi 2000). A similar phenomenon has been identified in the cen-
tral Ionian Sea (Hieke & Werner 2000). The fact that on certain sites,
such as Rhodes, there is no break in Minoan activity (Keller, Rehren, &
Stadlbauer 1990) suggests that the effects of the Theran eruption were
not disastrous in the short term (Minoura et al. 2000: 59). Evidence
from Lake Gölhisar in south-west Turkey demonstrates that the erup-
tion had little impact on the vegetation in this part of the Mediterranean
(Eastwood et al. 2002). Excavations at the Mochlos site on Crete pro-
duced a tephrastratigraphy which demonstrated that the eruption took
place towards the end of the Late Minoan IA period, whilst the collapse
of Minoan society started somewhat later during the Late Minoan IB
period. There is also some doubt as to whether the acidity peaks seen
in some of the ice cores directly relate to the Santorini eruption, whilst
the reduced growth rings seen in some tree-ring records may have been
caused by other environmental processes; in fact, enhanced tree growth
may have occurred during this period (Kuniholm 1996).
There is no doubt that the eruption had a profound impact on
Santorini itself; the pre-eruption Bronze Age landscape would have been
quite different, with greater topographical variation (Heiken, McCoy, &
Sheridan 1990; McCoy & Heiken 2000a). In addition, the vegetation
on the island would have been far more varied. Charcoal evidence points
to the presence of vine, oak, pine, and tamarisk; moreover, tree roots
have been found in buried soils. Today, there is remarkably little tree or
shrub vegetation on the island (Asouti 2003a: 473). On Crete, there
is little evidence to suggest that the Santorini eruption was responsible
for vegetation degradation; moreover, there is no sedimentary evidence
within the Delphinos (a north-western Cretan coastal marsh) core for a
tsunami. In the Delphinos pollen diagram, there is some suggestion that
crop cultivation ‘stagnated’ after the eruption and that olive production
levels possibly did decline (Bottema & Sarpaki 2003: 747). Any modifica-
tions in vegetation and settlement patterns were more likely the result of
socio-economic changes which may have in some small part been influ-
enced by the eruption; an event that possibly ‘nudged’ an already fragile
social system just over the edge. One elegant assessment of the intersec-
tion of culture and environment suggests that changes in pottery style
after catastrophes articulate a new attitude towards the natural world,
in this instance, the sea. Rather than representing the impact of a possi-
ble Thera-generated tsunami, one possibility is that the motifs represent
maritime life that was lost or damaged by pumice fall, as this would have
had a serious impact on the productivity of the sea around Crete. In par-
ticular, deep-water cephalopods and gastropods are presented – creatures
that require clear, clean water, suggesting that Minoans avoided consum-
ing marine products as a consequence of post-eruption marine pollution
(Bicknell 2000: 101).
Despite the lack of evidence for the immediate impact of the Theran
eruption on Cretan society, it is possible that a combination of earth-
quakes and tsunamis (possibly following volcanic activity) did have con-
sequences later on for Crete, at the end of the Neopalatial period (LM
IB, fifteenth century BC), and then again during the Late Minoan IIIB
period (thirteenth century BC) (Vallianou 1996: 153). Anti-seismic con-
struction methods appear to have been employed at Phaistos (central-
southern Crete) from the Old Palace period (phase IB). The building
excavated by Vallianou at Pitdiai, just to the south of Phaistos, collapsed
and was abandoned during the Late Minoan IB period. Elsewhere in
northern and eastern Crete, there were a number of destruction phases
recognised on archaeological sites dated to the Late Minoan IB period.
Excavations at a Minoan settlement now some 90 m from the present
coast and the harbour at Gouves revealed a layer with large quantities of
pumice and evidence for flooding. This destruction layer has been dated
to the Late Minoan IIIB period (1350/40–1190 BC or 1340/1330–
1190 BC) (ibid.). Driessen (1998) considers that the Theran eruption
may have damaged food-production systems, thus undermining ruling
elites on Crete and leading to the development of decentralised networks
of power, which in turn left the way open for an increase in Mycenaean
influence. Settlement patterns changed and moved so much during the
pre- and proto-historic periods that it seems unlikely that the only (or
even dominant) factors were environmental/subsistence criteria (Wallace
2010: 53). These high-magnitude events may well have exposed inher-
ent weaknesses in Minoan society; weaknesses that could not be resolved,
perhaps partly because of the nature of insular societies (see Chapter 7)
and associated forms of cultural, political, and environmental knowledge.
Recent studies suggest that Minoan palatial society was unstable due to
its complex hierarchical structure and its dependence on agricultural spe-
cialisation and extensive cropping (Driessen 2002; see Hamilakis 1996).
A widespread economic and cultural downturn across the Aegean,
and the start of the ‘Bronze Age Dark Ages’, have been blamed on an
earthquake ‘storm’ – a series of earthquakes in relatively short succes-

sion that occurred in the Eastern Mediterranean between c. 1225 and
1175 BC. This research is based on an analysis of modern seismological
processes and a reanalysis of evidence for earthquake impact on a series
of Late Bronze Age sites in the Aegean (Nur & Cline 2000). There
appears to have been a phase of substantial destruction at Midea in the
Peloponnese at the end of the Late Helladic IIIB2 period (c. 1190 BC).
This is contemporary with destruction events at Mycenae and Tiryns.
The authors suggest that the inhabitants of this site may have fled in
panic to Tiryns, whilst it is likely that the site was reoccupied during the
Late Helladic IIIC period. The west gate of the Acropolis appears to have
been destroyed during an earthquake. A young girl was found crushed
by fallen stones; she appears to be a victim of an earthquake (Ånstöm &
Demakopoulou 1996: 37–9). On the hill of Pyrgos (the Homeric town
of Kynos), there is evidence of storerooms that had been damaged by
earthquakes. The pottery indicates an early Late Helladic IIIC date for
one earthquake followed by a more ‘decisive’ event later on during this
period. The discovery of marine fossils even leads the authors to suggest
that these may have been swept in by a tsunami: the building is only 16
m above the present sea level, and some 100 m away from the shoreline.
After this destruction phase, the buildings were reused, although the
quality of construction was poorer. These new floors also contain baby
burials in pits (Dakoronia 1996: 42). A horizon dated to Late Helladic
IIIB c. 1250 BC comprised skeletons covered with fallen stones. It is also
possible that the preceding construction (heavy terraces to support steep
slopes) was a precaution against earthquakes. Repairs after earthquakes
do seem possible; for example, mud poured into moulded spaces allows
repairs within walls that were still partly standing (E. B. French 1996:
51). There are a number of other sites that comprise similar sequences
of events, including the palace at Thebes which was totally destroyed
during the thirteenth century BC (LH IIIB) (Sampson 1996; Spondilis
1996).
In the north-east Peloponnese, nearly 100 Bronze Age sites have
been found. Many of these sites were destroyed not long after 1200 BC,
although some appear to have been evacuated without being destroyed.
In Messenia, there were about 150 to 200 Late Helladic IIIB settle-
ments. Most of these were quite small and were abandoned after the
burning of the Palace of Nestor at Pylos. Some dismiss the importance
of catastrophic natural events, and whilst citing geological catastrophe

as the direct cause is problematic, there are also problems in replac-
ing this with another unique cause. However, the possible reasons are
varied, and might even include the impact of a new form of warfare
employed by invading tribes (Drews 1993). The other fact that might
have contributed to the downturn in settlement and economic activ-
ity during this period is climate change: the possibility that this region
witnessed a phase of relative aridity (L. D. Brandon 2012). Whilst cata-
strophic events would have affected Bronze Age peoples in the region,
they would not have directly led to the destruction of these cultures.
These environmental processes may well have tested certain forms of
sociocultural resilience. Some food resources may have been affected
by catastrophic events combined with a new, relatively arid climate.
Ideological and ritual mechanisms may well have emerged that helped
mitigate these problems. These ideological mechanisms may well have
reinforced the social memory of such disasters and led people to ques-
tion the ability of social elites, or even the Gods, to moderate the threats
posed by the natural world.
Hellenistic and Roman Catastrophes

Tectonic activity can directly affect buildings (Fig. 2.6) as well land-
scape structure and concomitant processes, such as soil erosion and
watercourse development. An extreme example of the importance of
such events comes from Western Anatolia where 12 important cities,
along with many minor settlements, were damaged or destroyed by an
earthquake in AD 17 (Ambraseys 2009: 105–8). On sites where there is
evidence for repeated impact of earthquakes, we have to consider how
people responded. It is these sites, where people chose to stay, that prob-
ably represent a form of resignation; people accepting that seismic activ-
ity is a part of life. Cnidus in south-west Turkey was situated on a fault.
Here, two powerful earthquakes took place, the first during the Late
Hellenistic Period (late second–third century BC), the second during the
fifth century AD (possibly AD 459) (Altunel et al. 2003). This type of
scenario is repeated across many sites in the region, including Hierapolis
(Pamukkale, Turkey) (Hancock & Altunel 1997). One characteristic of
this latter site is the presence of offset irrigation channels, thus revealing
how relatively minor events would have damaged the all-important
hydraulic elements here.

2.6. Fallen columns at the Temple of Zeus, Olympia, Greece. Probably toppled by a sixth-century AD earth-
quake (photo: author).
We should not forget the other environmental consequences of seis-

mic activity. Whilst many geoarchaeologists have claimed climatic and/or
anthropogenic causes for erosion, there is no doubt that seismic events,
with their resultant reconfiguration of topography, would have caused
erosion (Papanastassiou, Gaki-Papanastassiou, & Maroukian 2005). In
some instances, economic responses to earthquakes might be identi-
fied. A series of earthquakes is known to have affected Kos (south-east
Aegean) between c. 200 BC and c. AD 200, with a particularly severe
and catastrophic event taking place in 198 BC. It seems that this event
was followed by a substantial increase in the monetary supply (based on
detailed numismatic research), which suggests that this society required
new financial resources to support the reconstruction of Kos. The fun-
damental question asked by Höghammar (2010) is whether these repairs
were successfully carried out over a short period, or whether they were
protracted, perhaps testing the resilience of this society.
In most instances, individual earthquakes, even if they are very pow-

erful, have had little effect on political and cultural trajectories of past
civilisations, unless such an event acts as a catalyst and accentuates an
existing political or cultural weakness within a society. However, contin-
ued tectonic activity must have had some influence on the ways in which
people engaged with the landscape, and in some instances, the cumula-
tive effect of these events would have had some effect on economic and
cultural systems.
More often than not, it is felt that earthquakes induce immediate and
sometimes calamitous impacts on society, via building collapse or some
other dramatic consequence. Earthquakes can also engender medium-
term environmental impacts. Subsidence and uplift influence geomor-
phology and vegetation, and earthquakes will have damaged arid zone
irrigation systems, with a change in water supply resulting in vegetation
changes. The analyses of pollen from Dead Sea stratigraphic units that
include seismites (sedimentary units disturbed by seismic shaking) allow
an assessment of post-earthquake vegetation change. After the 31 BC
earthquake, there was a reduction in agricultural production, signalled
by reductions in olive, Pistacia, Vitis, and Juglans pollen. An increase in
the extent of the desert could also have been another result of this event
(Leroy et al. 2010).
A key characteristic of effective environmental knowledge is predic-
tion. For example, Sparta experienced a number of earthquakes, but these
significant events were separated by many centuries, or even millennia,
which was clearly significant in terms of the collective memory and envir
onmental knowledge associated with such processes. Major events are
dated to 2500 BC and 464 BC, with relatively minor earthquakes tak-
ing place at 3900 BC, 2000 BC, 550 AD, and 1000 AD. Non-periodic
events, such as those around Sparta (Finkel et al. 2002), are only predict-
able if they are recorded and a pattern is recognised, with knowledge of
the patterns passed on through collective memory. The 464 BC earth-
quake in the southern Peloponnese is considered to have been centred
on a fault close to Sparta (Armijo, Lyon-Caen, & Papanastassiou 1991),
and Thucydides described how the helots took advantage of this event to
mount a revolt (Urbainczyk 2008: 24). The unpredictable and powerful
nature of the earthquake may well have exposed a weakness in Spartan
society, thus creating the ideal moment for revolt. Governing elites may
well attempt prediction and mitigation of environmental problems, and
Tectonic Legacies 27
if they should fail in these acts, such failures might reveal or accentuate
other weaknesses within a society, such as the Spartan Helot conflicts.
Tectonic Legacies
Despite the questions over the relative impact of seismic and volcanic
events on Mediterranean civilisations, at the very least we must accept
the importance of tectonic processes in the formation of Mediterranean
landscapes – not merely in terms of defining a topographic base layer, but
also in terms of Holocene tectonic events affecting sea level, fresh-water
supplies, sedimentation processes, and, as a product of these processes,
vegetation development. Tectonic landscapes are effectively full of sur-
prises; within short distances, spring-fed basins with relatively lush veg-
etation abut arid, eroded land surfaces (Fig. 2.7). Active fault zones are
characterised by varied topography, intermountain basins with springs,
and good quality soils that develop on alluvium (Trifonov & Karakhanian
2004: 293–4). Small patches of wheat will grow adjacent to slopes cov-
ered in rubble with a few persistent pines. As Trifonov and Karakhanian
(2004: 290) suggest, active faults will also comprise elements and natu-
ral radioactivity that can have negative consequences for plants, animals,
and people, whilst other faults might even furnish unusual conditions,
such as the emission of gasses with ‘narcotic’ effects; a configuration that
might explain the location and evolution of Delphi as an Oracle (De
Boer & Hale 2000). Technically, active zones can also render water sup-
plies unreliable, with some evidence suggesting earthquakes have been
responsible for changes in groundwater supply, leading to the abandon-
ment of certain major sites on Crete (Gorokhovich 2005).
Whilst the longue durée has traditionally been employed in the charac-
terisation of environmental process, we should not forget that all human–
environment interactions are also constituted by medium and short-term
actions. In any assessment of human response to a given process, the
analysis must explicitly consider the spatial and temporal scales in which
humanity is studied. The great danger with many historical discourses is
that the language employed by researchers renders the scales of analysis
opaque, even invisible. Horden and Purcell, amongst other recent writ-
ers, produce a textual melange that avoids an explicit declaration regard-
ing their scales of interest. There is nothing inherently wrong in writing
a history or archaeology that considers societal processes in the longue
(a)
(b)
2.7. (a) Typical Mediterranean limestone topography with geological folds contributing to the creation of mul-
tiple niches (Sainte Victoire, nr. Aix-en-Provence) (photo: author). (b) Watercolour of the geological structure
of the Sainte Victoire (permission granted by J.-M. Gassend).
Tectonic Legacies 29
durée, in the same way that we should not see fault in those who attempt
to write an histoire événementielle. However, authors should outline and
provide a rationale for their chosen temporal and spatial scales of analysis.
Such a declaration, so many decades after the publication of Braudel’s
histories, may seem unnecessary. However, the articulation of our scales
of analysis is primordial if we are to understand the range of possible
responses to all environmental processes.
3
Sea-Level Change and Coastal Settlement – Human

Engagements with Littoral Environments
Introduction
It is links across the sea that create the very essence of Mediterraneanism
(Bresson, 2005). The study of the dynamism of the Mediterranean Sea,
environmentally and culturally, is a prerequisite for any historical and
archaeological endeavour in the region. The Mediterranean Sea is in itself
as important in the minds of people as the lands that abut it (Guilaine
1994). At one point, classical populations seem to have identified 27
component seas within the Mediterranean (V. Burr cited in Harris 2005a:
15). The permeability of the coastline, that is, the ability to move easily
between terra firma and the sea, is influenced by environmental change
and a society’s ability to manipulate access to and from the sea. ‘In the
relatively tideless Mediterranean, the shore is narrow – a line, a bound-
ary, a margin, a place where opposites meet’ (Buxton 1994: 102).
This chapter starts with an overview of the principal environmental
processes that contribute to the development of coastlines, and then
assesses the development of human use and modification of these areas.
Characterising the Mediterranean
The virtual absence of tides across most of the Mediterranean results

in the absence of the diurnal cycles of coastal life that are prevalent in
most other regions around the planet. This day-to-day environmental
stability in the Mediterranean has its advantages in that the location of
settlements, moorings, ports, and harbours is relatively straightforward.
However, medium and long-term processes can render life difficult, or
even impossible, especially in areas where the increasing progradation
30
Maritime Processes 31
(land advance caused by sediment deposition on river deltas) from about

6,000 years ago resulted in significant coastline changes.
Relative sea-level change, whether caused by eustacy or crustal (tec-
tonic and isostatic) movements, are the key processes, but seasonal
storms are another problem, and whilst people on the coasts were safer
than those at sea, these events would have coloured everyone’s percep-
tion of the sea and the littoral band. In some ways, the Greek percep-
tion of the sea was similar to that held of the mountains; these were
environments where specific types of activity such as fishing, trading, and
travelling took place (ibid.: 97). The coastline, however, is the interface
between two remarkably different worlds; these are environments on the
edge – not marginal per se, but zones where a change in one particular
environmental parameter or process can test the resilience of any soci-
ety’s engagement with that landscape.
Maritime Processes
The Mediterranean is a small ocean that has a direct connection with

the Atlantic Ocean. The characteristics of the outflow and inflow from
the Mediterranean towards the Atlantic have changed throughout the
Holocene.
The present outflow layers (referred to as Mediterranean Outflow
Water (MOW)) are situated at depths of 800 and 1300 m, and were estab-
lished between 7,500 and 5,500 years ago (Schonfeld & Zahn 2000). On
the north-eastern edge of the Mediterranean lies the Black Sea, which
is connected to the Mediterranean via the Sea of Marmara (Fig. 3.1).
Prior to the Holocene, the Sea of Marmara was a freshwater lake, but it
was inundated by the Mediterranean about 12,000 years ago (Cagatay
et al. 2000), and in turn, the Black Sea was flooded by Mediterranean/
Marmaran waters. Some have argued that this took place relatively quickly
about 7,000 years ago (Ryan et al. 1997). However, the most recent
reassessment of the evidence clearly suggests that there was not a cata-
strophic event (Aksu et al. 2002). More specifically, the absence of under-
water archaeological evidence dated to the Late Pleistocene through to
the Early Holocene certainly leads us to examine the argument for such a
catastrophe at this time (Yanko-Hombach, Mudie, & Gilbert 2011).
Changes in climate, inflows of freshwater, and sediments – all have
an impact on the ecological productivity of the regional seas across the
Sea of Azov
Bay
of
Biscay
24
Black Sea
23
2219 25 33
Gulf of 4
Lions
Sea of Marmara
Adriatic
26 36 Sea
32 Ap
Balearic uli
Sea a 21
Gulf of Aegean Sea
Tyrrhenian Sea Taranto 9 8
7
Calabria 1
15 17 Ionian Sea
34
35 13 18 31
Strait of Gilbraltar 10
27 12
11 2
Sea of Crete
16
20 30
28
14 29
Mediterranean Sea 5
6
Gulf of Sidra
3.1. The Seas of the Mediterranean and map with sites referred to in the chapter. Sea of Marmara, Straits of Gibraltar, Gibraltar, Rhodes, Apulia, Calabrian Island
arc, Adriatic and Tyrrhenian Sea. 1: Palairos-Pogonia, 2: Santorini, 3: Gouves, 4: Cassis, 5: Atlit-Yam, 6: Tel Nami/Tel Dor/Caesarea, 7: Petromagoula/Pevkakia-
Magoula and Lolkos, 8: Dimini bay/Volos, 9: Sesklo, 10: Franchthi Cave, 11: Messenian peninsula, 12: Pylos/Palace of Nestor, 13: Tiryns/Mycenae/Argive plain, 14:
Kommos/Matala Bay, 15: Gulf of Cádiz, 16: Kition/Larnaca, 17: Peloro Peninsula/Tindari, 18: Delos, 19: Marseille, 20: Phalasarna, 21: Troy, 22: Fos, 23: Ravenna
(North-east Italy), 24: Aquilea, 25: Fréjus, 26: Ostia/Portus, 27: Carthage, 28: Mahdia, 29: Tyre, 30: Beirut, 31: Menderes Delta/Priene/Miletos, 32: Cumae, 33:
Pisa, 34: Ephesus, 35: Baelo Claudia, 36: Castelporziano, Lazio.
Coastal Processes 33
Mediterranean. More specifically, plankton and foraminifera produc-

tion have always been affected by wider changes in climate (Piva et al.
2008). As with much environmental research, the work carried out on
the Mediterranean Sea informs current studies of nutrient problems
across the region (e.g. Benoit et al. 2005; Jeftic, Keckes, & Pernetta
1996). The Mediterranean has never been as ecologically productive as
other seas and oceans around the world. Very stable strata impede the
regeneration of nutrients from lower depths. Consequently, fish catches
in the Mediterranean are relatively small compared to the major oceans
around the world. For a book that is in part concerned with human
impact on the environment, one topic that is missing (due to a dearth
of relevant research) is the chronology and nature of human impact on
marine ecosystems during the prehistoric and classical periods (Erlandson
& Fitzpatrick 2006; Jackson et al. 2001). Changes in sea temperature or
even salinity would have affected societies and economies, although spe-
cific responses to such processes are difficult to identify.
Within the Mediterranean, the formation of a sapropel (organic-rich
dark sediments found across the seabed) took place between 9,000 and
6,000 years ago, this period corresponding with the so-called monsoon
maximum. After this, from c. 6500 BC, this deposition was interrupted
and resulted in improved deep-water ventilation that coincided with
cooling over the Aegean and Adriatic. Another similar cooling event
occurred at c. 4000 BC (Rohling et al. 2009: 57). There is some dis-
cussion of the precise mechanisms that lead to sapropel formation, with
some arguing that climatically controlled stratification of the sea leads to
improved preservation of the sapropel organic matter (De Lange et al.
2008). Periods of lower sea level and reduced sediment input seem to
correspond to arid periods, which have been identified across the Western
Mediterranean from the late fifth millennia BC onwards. There is a two-
layered flow of waters in the Strait of Gibraltar: a surface inflow of low-
saline water and the deeper outflow of cooler saline (Mediterranean)
water (Goy, Zazo, & Dabrio 2003).
Coastal Processes
Many processes and configurations of the different seas are directly

affected by tectonic activity (Yaltrak et al. 2000), with geodynamic pro-
cesses defining many coastlines (see Fig. 2.3) (Stewart & Morhange
34 Sea-Level Change and Coastal Settlement

3.2. Watercolour of Crete, illustrating the uplift at the western end of the island (by permission of J.-M.
Gassend).
2009). Coastline development is the result of the complex interplay of

eustatic, glacio-hydro-isostatic and tectonic processes. This combination
results in regionally heterogeneous changes. For example, uplift rates at
Taormina and Scilla (the toe of Italy) over the last six millennia were about
100% greater than the long-term average, whilst the last 2,000 years at
the Briatico site (on the northern edge of the toe of Italy), eustatic sea
level rise has been compensated for by tectonic uplift (Lambeck et al.
2004).
Although the actual coastline is influenced by changes in sea level and
sedimentary inputs derived from terrestrial zones, it is of course a com-
plex geological formation, largely dominated by limestone, and highly
influenced by crustal processes that structure relative changes in sea
level. The most useful general description of Mediterranean coastlines
is that they are ‘…a nested set of marginal-sea coasts – narrow shelves
fronting steep hinterlands along the shores of restricted seas enclosed by
major land masses and island chains’ (Stewart & Morhange 2009: 386).
Coastal Processes 35
Whereas eustatic sea-level rise has followed a relatively steady trend,

local, relative sea-level changes are specific to each part of the coast.
Even across one relatively small island, there will be a variety of sea-level
histories.
The input of sediments along river systems is also of profound impor-
tance in the development of coastal plains, and even the eruption of vol-
canoes, such as Vesuvius, can result in shoreline progradation (Pescatore
et al. 2001). Whilst eustatic sea-level rise has continued (due to the
continued input of glacial and ice-sheet melt waters) throughout the
Holocene, tectonic activity can lead to either exaggerated relative sea-
level rise, when land moves downwards, or relative sea-level fall, when
land moves upwards. In some parts of the Mediterranean, such as across
Crete, there is a complex interplay of these processes, with uplift in the
west and down-warping in the east occurring across the same island
(Fig. 3.2).
Coastlines tend to be characterised by either solid geology or soft
geology (clastic coasts where sedimentary processes dominate). These
characteristics determine how changes in relative sea level can be studied
on a given coastline. However, in both environment types, microscopic
and macroscopic floral and faunal species are identified in the assess-
ment of sea-level changes. Soft coasts, with their sedimentary inputs,
yield informative environmental records; the sediments themselves, plus
the macro- and micro- fossils found therein, along with any chemical
signatures, provide a range of useful information regarding natural and
cultural processes. However, on coastlines characterised by solid geol-
ogy, our knowledge of the biological zonation of floral and faunal spe-
cies across the different littoral zones (from sublittoral to supralittoral) is
important (Laborel & Laborel-Deguen 1994). It is necessary to note that
the border between midlittoral and sublittoral (or infralittoral) is referred
to as the ‘biological sea level’ (Laborel 1986). The investigation of wave
notches is common on rocky coastlines (Fig. 3.3). The measurement of
the position/heights of archaeological remains is also noteworthy. For
example, some Roman archaeological remains along the coastlines of
Croatia and southern Turkey are now about 1.5 m below the present
sea level. This demonstrates how regional subsidence has occurred since
the Roman period. As we move south to the Ionian Sea, tectonic influ-
ences become even more important (Fouache & Dalongeville 2003:
471–6). Subsidence on the north-eastern coast of Crete has been such

3.3. Typical Mediterranean wave notches (characteristic of a tideless sea) on an uplifted coastline (Côte d’Azur,
France) that indicate past relative sea-level stands (photo: author).
that wave notches have been erased, and vermitide development has
been prevented. It is likely that the overall eustatic rise on the Northern
Mediterranean is about 50 cm since the start of the Holocene (Pirazzoli
1976).
Many coastlines have changed their forms as sediments have been
transported from inland areas down towards the sea. The processes of
aggradation and progradation have been studied across the Mediterranean
(Marriner & Morhange 2007). The study of sedimentary facies through
boreholes and/or excavated trenches in coastal locales allows us to
develop an impression of how land has ‘moved’ seawards, and relative sea
levels have fallen. More specifically, the study of biological sea-level indi-
cators, both macrofauna (marine molluscs) and micro fauna (foramin-
ifera and ostracods) retrieved from sedimentary units permits the analysis
of changes in marine environments over time (e.g. from lagoonal, fresh
to saline conditions).
Examples of Mediterranean Coastal Change 37
Examples of Mediterranean Coastal Change
One common feature of many reports on sea-level change is the use of

figures of overall relative sea-level change and, therefore, mean rates of
sea-level change. On Gibraltar, one study has demonstrated that tectonic
uplift has led to a series of stepped, uplifted marine terraces (Rodriguez-
Vidal et al. 2004), and that over the last 200,000 years, the mean uplift
rate has been 0.05±0.01 mm. Such statistics are problematic for archae-
ologists, as these temporal scales have little relevance for understanding
human interaction with coastal environments. Mean rates may hide the
fact that there have been periods when sea-level change was extremely
rapid. Such swift changes are common in the Mediterranean, where tec-
tonic events can have an immediate and dramatic effect on relative sea
level. For example, a series of uplifted notches along the north-east east
coast of the island of Rhodes were shown to be up to 6,000 years old.
A 1-m subsidence occurred as part of an earthquake in 227 BC, which
some argue was responsible for the collapse of the giant statue Colossus
(Kontogianni, Tsoulos, & Stiros 2002: 301). Slipways in the harbour
were also destroyed at this time. Older slipways have been identified
which correspond with a sea level between 2.05 and 3.1 m higher than it
is today. New slipways were then constructed, possibly built with a view
to counteracting seismic land subsidence (Fig. 3.4) (ibid.: 305). This
process – and human response to it – clearly differs to that on coastlines
where changes are far more gradual. It is, however, important to avoid
suggestions that sudden, high-energy events had catastrophic or substan-
tial affects on wider societal and economic processes (Morhange et al.
2013).
As with many environmental processes, the variation in spatial and
temporal resolution and the magnitude of the impact of littoral processes
vary considerably. Sudden tectonic movements along the coast and con-
comitant changes in relative sea level would have had an immediate
impact on coastal activities, whilst responses to gradual changes in sea
level, which lead to the emergence of new coastal topographies, would
have been perceived in a different manner. Changes in the coastline thus
present archaeologists with an interesting problem in the study of human
perception of temporal scales: the longue durée, as represented by eustatic
sea-level rise, and the short-term (événementielle) process constituted by
tectonic processes.
drowned
slipway
227BC
stepped 3
upper notch
seismic
formation
uplifts
notch elevation above

present sea level (m)
fall Sea level rise
2
main seismic uplift

1
after 227BC
7 6 5 4 3 2 1 0
time (1000 yr BP)
3.4. Conceptual model for the relative sea-level changes responsible for the notch uplift
along the north-east Rhodes coast. (By permission of Elsevier BV., Kontogianni, Villy A.,
Tsoulos, Nikos, Stiros, Stathis C. (2002), Coastal uplift, earthquakes and active faulting
of Rhodes Island (Aegean Arc): Modelling based on geodetic inversion, Marine Geology
186(3–4), fig. 5, p. 306.)
Higher-Energy Events
One type of high-magnitude event that has taken place across the
Mediterranean throughout the Holocene is tsunamis. For example, just
within the Sea of Marmara, there is evidence for 30 probable tsunamis hav-
ing taken place over the last 2,000 years (Yalcner et al. 2002). Tsunamis
can be caused by volcanic eruptions as well as submarine slumps (caused
by seismic activity). Despite the fact that there is little doubt that tsuna-
mis did take place in the pre- and proto-historic past, the identification of
tsunami deposits is notoriously difficult, as they are very similar to sedi-
ments deposited by storm-surge flooding (Stewart & Morhange 2009:
400). Accumulations of large boulders can be deposited by high-powered
waves, such as those studied on the Apulian coast (Mastronuzzi & Sansò
2000, 2004). Despite the inherent problems in the accurate identifica-
tion of tsunami deposits, there are places where tsunami sediments have
been identified. Several have been identified from sedimentary units in
the Huelva Estuary (Gulf of Cadiz, south-western Spain), a zone that
is not strictly Mediterranean, as it is situated to the west of the Straits
of Gibraltar. The tsunami deposits comprise coarse material (especially
marine molluscs) and heavy metals that were clearly derived from outer
coastal or continental shelf environments. A mid-to-late fourth century
Pre- and Proto-Historic Coastal Exploitation 39
AD event is the earliest in the sequence of tsunamis identified here

(Morales et al. 2008). Not all parts of the Mediterranean are exposed
to the same level of tsunami risk. As Soloviev’s (2000: 15) analyses of
Mediterranean tsunamis demonstrates, recorded tsunamis are more
common in the northern half of the region, with a concentration in the
Aegean, around the Calabrian Island arc, and other higher-risk areas in
the Eastern Mediterranean, the Adriatic, and Tyrrhenian Sea.
Evidence for the Santorini tsunami (a direct consequence of the c.
1620 BC eruption) has been identified on the seabed between Santorini
and Crete (McCoy & Heiken 2000a, 2000b). The tsunami may have
been some 40 m high (Monaghan, Bicknell, & Humble 1994). One sim-
ulation suggests that the impact of the wave would have been restricted to
the South Aegean, although a low wave would have arrived in other areas
(Pareschi, Favalli, & Boschi 2006). A number of numerical models of the
pyroclastic flow from the eruption and ensuing tsunami have been pro-
duced (Monaghan et al. 1994). Another simulation, based on the pres-
ence of tsunami deposits around the Aegean and Eastern Mediterranean,
suggests that Crete and the coast of western Turkey may have experi-
enced tsunami waves with heights of 5 m and above. A layer of sediment
was also found at Gouves (30–90 m inland from a Minoan harbour), 15
km to the east of Knossos (Minoura et al. 2000: 60).
Pre- and Proto-Historic Coastal Exploitation
Mediterranean coasts could be considered as the most significant envi-

ronment for the study and understanding of Mediterranean Early
Neolithic societies. Some consider that the majority of Neolithic col-
onisation took place via the sea; the large number of Cardial Neolithic
sites along Mediterranean coastlines might support such a hypothesis
(Guilaine 2003: 161, 181) (Fig. 3.5), whilst the movement of farming
(associated with Impressed Ware) along the eastern Adriatic up to north-
ern Dalmatia is also evident (Forenbaher & Miracle 2005).
Lower sea levels would have reduced the distances between continen-
tal landmasses, the archipelagos, and their islands. We should remem-
ber that many sites that could represent the movement of people would
have been lost or masked by sedimentation. For example, Thrace, which
some consider to have been a ‘missing link’ in the story of the move-
ment of agriculture around the Mediterranean, was subject to a marine
18
17
63 57 39
66-67 60 56 53-55 38
37 15
62 14
64-65 35 3316
68-70 61 52 28-31 13
73 36 34 32 26
58 24 6
71 51 23
72 59 27
80 22 5
42 11
79 41 21 12 7-9
78 74 40 19-20
75 47
46
43 10
76 45 1
77 44
48
49 50
2
4
3
1: Mersin; 2: Parekklisha-Shillourokambos; 3: Akrotiri-Aetokremnos; 4: Kissonerga-Mylouthkia; 5: Ilipinar; 6: Yarimburgaz; 7: Nessonis; 8: Achilleion; 9: Theopetra;

10: Franchthi; 11: Sidari; 12: Konispol Cave; 13: Crvena Stijena; 14: Odmut; 15: Obre; 16: Vela Špila; 17: Pupićina; 18: Edera; 19-20: Neolithic villages at Matera
including Trasano; 21: Torre Sabea; 22: Torre Canne; 23: Ostuni; 24: Grotta del Guardiano; 25: Balsignano; 26: Pulo di Molfetta; 27: Rendina; 28: Lagnano da Piede;
29: Masseria La Quercia; 30: Coppa Nevigata; 31: Masseria Candelaro; 32: Guadone; 33: Prato Don Michele; 34: Torre Sinello; 35: Fontanelle; 36: Tricalle;
37: Maddalena di Muccia; 38: Ripabiancadi Monterado; 39: Fornace Cappuccini; 40: Capo Alfiere; 41: Favella; 42: Grotta San Michele di Saracena; 43: Umbro; 44:
Stentinello; 45: Grotta del Kronio; 46: Grotta dell’Uzzo; 47: Castellaro Vecchio; 48: Għajn Abdul; 49: Skorba; 50: Għar Dalam; 51: Le Caprine di Montecelio;
52: La Marmotta; 53: Arma dello Stefanin; 54: Grotta Pollera; 55: Arene Candide; 56: Riparo Mochi; 57: Alba; 58: Basi; 59: Grotta Verde; 60: Pendimoun;
61: Caucade; 62: Chateauneuf; 63: Grotta l’Aigle; 64: Peiro Signado; 65: Pont-de-Roque-Haute; 66: Camprafaud; 67: Grotte Gazel; 68: Jean Cros; 69: Dourgne;
70: Leucate; 71: Les Guixeres deVilobi; 72: L’Esquerda Roques del Pany; 73: Chaves; 74: Cova de l’Or; 75: Cova de les Cendres; 76:Cariguela;

77: Los Castillejos de Montefrió; 78: Almonda; 79: Calderião; 80: Eira Pedrinha.
3.5. Many of the key Cardial sites from across the Mediterranean. (Figure produced by Cynthianne Spiteri.
By permission of Debono Spiteri, C. (2012), The transition to agriculture in the Western Mediterranean:
Evidence from pots, Unpublished PhD thesis, University of York; p. 28, fig. 2.3.)
t ransgression about 2,900 years ago, and consequently, any Neolithic

sites would have been destroyed (Ammerman et al. 2008).
For coastal communities, the impact of sea-temperature change could
have had serious consequences. Although it is difficult to model palaeo-
marine resource changes, there is no doubt that temperature changes
and concomitant affects on water circulation (in all dimensions) would
have led to changes in marine animal presence, abundance, and seasonal
movement. Research around the Mediterranean clearly demonstrates
that sea-surface temperature changes did take place, one example being
the winter cooling in the southern Adriatic at c. 4000 BC (Sangiorgi
et al. 2003). Such a change could have been an impetus for increased
emphasis on terrestrial resource production.
Whilst the transition to agriculture was taking place across much of the
Mediterranean, sea levels had not achieved the point (and relative stabil-
ity) that characterises the modern Mediterranean coastline. Substantial
eustatic sea-level rise continued until about 6,000 years ago. During the
Early Neolithic, islands continued to be drowned, and coastlines were
gradually submerged. At about 6000 BC, the distance between Corsica
and Sardinia was only about 6 or 7 km, whereas today it is about 12 km.
Settling in an area susceptible to marine inundation is a potential risk
if the benefits of locating in such a zone are outweighed by the threats

posed by the hazard.
For all periods, the use of archaeological sites as sea-level markers has
enormous potential, in that such phenomenon not only serve as an (indi-
rect) record of sea-level change, but they also represent human inter-
action with littoral environments. However, we should be aware that
the chronological resolution offered by such markers is often imprecise
(Auriemma & Solinas 2009). One project that has investigated some of
the most enigmatic underwater sites is that at Atlit-Yam. The growth
of fishing villages was undoubtedly one of the most significant develop-
ments along the Neolithic coastline. The Pre-Pottery Neolithic site of
Atlit-Yam is now situated about 300–400 m off the Carmel coast under
8–12 m of sea (Galili, Zviely, & Weinstein-Evron 2005; Galili et al.
1993). During the sixth to fifth millennia BC, a particular economic sys-
tem developed on the coastal plain where hunting, fishing, and gathering
were practiced alongside agriculture and grazing (Gopher 1993: 62). A
fishing-farming settlement should not necessarily be located adjacent to
the sea, as sea spray would reduce the productivity of the land for agri-
culture. Therefore, in order to maximise the productive capacity of the
area around the village, it should be located a few hundred metres inland.
However, if fishing is productive, and the village is not reliant on agricul-
tural produce, then a coastal location might make more sense. Another
possibility would be a split community. At Atlit-Yam, evidence for fishing
comes from artefactual remains (bone points, hooks, and possible net
gauges) and some 6,000 fish remains (Galili et al. 2002: 180). Across the
Eastern Mediterranean, inland agro-pastoral sites predate the develop-
ment of the first fishing villages. A few centuries after the development of
sites such as Atlit-Yam, similar villages developed on Cyprus and in parts
of the Aegean. Galili et al. (2002: 190) believe that the spread of fishing
villages (and the combination of agricultural with marine resource gath-
ering) may be a consequence of certain groups exhausting the agricultural
capacity of their lands, or ‘…it may have been the natural outcome of the
meeting and merging of agriculture and husbandry with a well-developed
local Mesolithic-Early Neolithic maritime subsistence system’.
The earliest known well in the world was found at Atlit-Yam. When
established, the sea level adjacent to the village was some 17 m below its
present height; within four centuries, it had risen by about 9 m (Nir 1997:
148). The rate of rise then decreased, with stabilisation at its present level

3.6. The now-submerged Neolithic well at Atlit Yam. (By permission of photographer, Mr. Itamar Greenberg;
excavator, Ehud Galili. The Institution Israel Antiquities Authority.)
taking place during the latter half of the sixth millennium BC. The rapid
sea-level rise that occurred during the early phases of activity at Atlit-Yam
must have been apparent within a human generation. As sea levels rose,
storms may have deposited sand into the well, with the well then used
as a refuse pit, with seeds from spices, pistachio nuts, grape seeds, and
also specimens of the granary weevil (Sitophilus granarius) found therein
(Kislev, Hartmann, & Galili 2004: 1302).
Coastal wells must not be dug too deep in order to avoid the incur-
sion of saline waters. This operation is complex, as at least 30–40 cm
of water was required in the well if water was to be drawn using jars.
Consequently, it is assumed that the depth of the wells was some 0.3–0.4
m below the water table. Moreover, the base of the well should have
been between 0.1 and 0.2 m above the average sea level when the well
was in use (Fig. 3.6). The excavation of the well at the Tel Nami site on
the Carmel coast (early second millennium BC) demonstrated that the
lowest course of this well is at 0.7 m below the present sea level. This well
is now some 100 m inland from the present coastline (Sivan et al. 2001:
107–8) (Fig.3.7). Despite these apparently successful attempts to deal
8 8
A
6 Surf zone Swash zone Building 6
4 4
2 2
S.L.
0 0
–2 Living floor –2
–4 –4
–6 –6
4 G.W.L. Ground water level Water B 4

Built–up well well
3 3
Rock–cut well
2 2
1 1
S.L. G.W.L.
0 0
–1 –1
Water level within the well
–2 –2
Bottom of the well
–3 –3
7 7
6 a b b1 C 6
Sand
5 5
4 Bedrock 4
3 3
2 2
1 Shipwreck 1
S.L.
0 0
a1
–1 –1
–2 –2
a2
–3 –3
–4 Anchor at time of wreckage –4
–5 Anchor after settlement –5
–6 –6
3.7. Three examples of archaeological sea-level indicators. (a) Living floors provide upper bound. In this
research, palaeo sea level is assumed to have stood at least 2 m below the floor levels so as to be beyond the swash
zone. (b) Ancient wells provide upper and lower bounds at sea level. Coastal wells have to be dug to a minimal
depth in order to avoid salinisation, but they still have to be effective at the lowest water levels. The inferred bot-
tom depth of ancient coastal wells along the Israeli coast was 0.3±0.4 m below the water table (in order to draw
clear water when using jars). This implies that the base of the well was about 0.1±0.2 m above mean sea level at
the usage period. For lower bound, we adopt a level of 1 m below sea level. (c) The dispersion line of shipwrecks
and heavy objects from the wreckage approximates the palaeocoastline. Analogous to modern observations, we
assume that approximately 1.5±2 m of sand covered the wrecked objects, and their present dispersion at water
depths not shallower than 23 m imply that palaeo sea level at the time of the wreckage was close to the pre-
sent level. (By permission of Elsevier, Sivan, D., Wdowinski, S., Lambeck, K., Galili, E., and Raban A. (2001),
Holocene sea-level changes along the Mediterranean coast of Israel, based on archaeological observations and
numerical model. Palaeogeography, Palaeoclimatology, Palaeoecology 167(1–2), fig. 2, p. 105.)
rias
Se Xe
sk
oli
tis
max. transgression Volos
lolkos
3000 B.C.
Dimini
fault
Petromagoula 300 B.C. present
coast
Demetrias Pevkakia
wall N
1 km

3.8. Coastline advancement at Volos – an area where settlements appear to have
‘followed’ the advancing coastline (by permission of E. Zangger).
with changing sea levels, these settlements did fall out of use and were
even destroyed. The reasons were perhaps political, military, or perhaps
the result of technical problems with the wells, such as wall collapse and
salinisation (Nir 1997).
At Petromagoula, located on what would have been the coastline
between Pefkakia and Dimini (Andreou, Fotiadis, & Kotsakis 1996: 549),
activity increased during the Late and Final Neolithic. New activity in the
hinterland of Volos Bay may have caused alluviation that had an impact
on the Dimini bay (Halstead 1984, cited in Andreou et al. 1996). Later
on, sites migrated, following the prograding coastline – thus maintaining
direct links with the sea. Zangger considers that this explains the pres-
ence of single-phase sites in this zone. For example, the Neolithic site of
Sesklo (6000–4400 BC) is now 8 km from the coast. During the Early
Bronze Age (3000 BC), the shoreline had advanced by almost 2 km,
and the site of Petromagoula would have been on the coast at that time
(Zangger 1991: 3) (Fig. 3.8). The sites at Pevkakia-Magoula and Lolkos
were located on higher ground, either side of the embayment, and their
relationship with the coast has not changed over the last three-and-half
millennia. Consequently, there was variation in the choice of settlement
locations in this coastal area; the combination of fixed settlements, such
as those at Pevkakia-Magoula and Lolkos, compared to the single-phase
settlements that followed the prograding shoreline. Here, we see how
people responded to gradual coastal change, a process that may have had
a generational or multigenerational resolution (ibid.).
Perhaps the most famous prehistoric coastal site in the Mediterranean
is Franchthi Cave. This area has largely been influenced by eustatic
rather than crustal processes, any possible effects of tectonic activity are
either minor or not apparent given the time range and time scale of the
Franchthi sequence and local topography. From about 13,000 BP, the
rate of sea-level rise was as high as 5 cm/year. Therefore, during a life-
time of 40 years, the sea level would have risen by about 2 m – a signif-
icant change in the coastal landscape. This rate of sea-level rise slowed
during the Early Holocene, although 20 m of eustatic rise has occurred
during the last seven millennia (van Andel 1987: 33). However, these
estimations are based on regional or global models; there is no complete
local curve for the southern Argolid.
The sea around Franchthi was probably more productive between
9,000 and 7,000 years ago, when marine resources were heavily exploited.
Sapropel deposition at this time supports such a conclusion (ibid.: 53) in
that sapropels possess higher concentrations of organic carbon and
appear to enhance productivity due to nitrogen fixation (Katsouras et al.
2010).
The beach, or ‘Paralia’ site, at Franchthi was first occupied during
the Early Neolithic. Between 6500 and 5000 BC, sea levels in front of
Franchthi Cave were 11 m lower than they are today (van Andel 1987:
34). Therefore, a more extensive beach was exposed during the Early
Neolithic, with the sea between 1 and 2 km further out, thus increasing
the possibility of aeolian deposition across the site (Wilkinson & Duhon
1990: 77). Most of the beach site appears to have been abandoned by
the end of the Middle Neolithic, although occupation at the cave site
continued into the Late Neolithic.
The changes in sea level at Franchthi and other Neolithic sites were var-
ied in their rate and nature. However, eustatic sea-level rise was a feature
of Neolithic coastal life, and would have affected settlement infrastruc-
tures, as well as the range of resources available to early farmers – farmers
who undoubtedly continued to exploit marine resources. We know that

shellfish use at Franchthi continued up until the end of the Neolithic at
c. 3000 BC (Shackleton & van Andel 1986: 130). Here and elsewhere,
these changes in sea level may well have helped push Neolithic farmers
towards a greater reliance on terrestrial foods. However, there is little
doubt that Neolithic exploitation of coastlines did encompass the contin-
uation of earlier food gathering strategies, including the intensification
of fishing. The concomitant development of coastal villages, with specific
forms of environmental knowledge, such as well digging, might have
been threatened by the continuing eustatic sea-level rise (which was still
relatively rapid for the earlier Neolithic) as well as variations in marine
productivity.
Coastal Exploitation: The Development of Ports and Harbours

(Bronze Age Onwards)
Whilst there is plenty of evidence for mixed coastal economies, evidence

for early boat mooring or beaching is negligible. Although movement
of people and materials undoubtedly took place via the sea during the
Neolithic, the assessment of the harbouring or beaching of boats is
inferred rather than demonstrated. As we move into the Bronze Age, the
first conclusive evidence for technological intervention along coastlines
appears.
The late third and second millennia witnessed a fundamental change
in some societies’ relationships with the coast. The emergence of deep-
hulled sailing ships during the late third millennium BC would have
changed the way in which the sea was perceived (Broodbank 2000: 341).
The development of complex trading networks and the need for perma-
nent ports and harbours led to an intensification of the management of
coastal environmental processes.
The coastal geomorphology of many parts of the Northern
Mediterranean provides a number of natural harbours (protected inlets,
estuaries, and lagoons). On the other hand, the Southern Mediterranean
has far fewer natural ports and harbours, and therefore these facilities
tend to be constructed. Fortunately, such construction is relatively
straightforward compared to Atlantic coasts due to the absence of tides.
Moreover, in many places (including Tunisia and Libya), relative sea lev-
els were quite stable during the later Holocene (Anzidei et al. 2011).
Coastal Exploitation 47
However, one fundamental problem was sedimentation and the infilling

of harbours. If a port or harbour was built near a river delta, it was often
located to the west of the delta, as long-shore action in the Mediterranean
often moves in a broadly west–east (counterclockwise) direction.
There were obviously certain types of coast that were deemed unsuit-
able for ports and harbours. For example, the Messenian peninsula
(south-west Peloponnese), comprises for the most part a rugged and
treacherous coastline, and this probably explains the relative lack of
Mycenaean sites here (Higgins 1966: 23). However, the easily erodible
coastline in this area may have been lost as sea levels rose, thus destroy-
ing any harbours (Higgins 1966). One notable harbour that does exist
in this area is Pylos (Messenia, Peloponnese) (Davis 2007). Located 5
km to the south-west of the Palace of Nestor, the harbour was situated
500 m inland and comprised a rectangular basin, along with a reser-
voir that was created through the damming of a river. Clean water from
this reservoir passed through an artificial channel, and this helped pre-
vent the reservoir from silting up (Zangger 2001; Zangger et al. 1997)
(Fig. 3.9). Tiryns, a Late Bronze Age citadel on the Argive plain in the
Peloponnese, possesses extant foundations that date to c. 1250 BC (LH
IIIB) (Zangger 1994). As with many low-lying plains, the Argive Plain
has undergone dramatic changes during the Holocene. At about 2500
BC, the sea-level transgression achieved its maximum, and the coast was
some 1.5 km further inland than it is today, only 300 m from the early
site at Tiryns. Large quantities of alluvial material just to the south of
Tiryns led to coastal progradation, and by the Late Bronze Age, 1 km
separated the palace from the coast (Fig. 3.9). During the Early Bronze
Age, a lower town at the south-western foot of the palace knoll was
traversed by a stream. An artificial levee was built to protect part of the
town from flooding. Eventually, this area was abandoned and covered by
thick alluvial deposits (ibid.: 196–8). During the Late Holocene (IIIB2–
IIIC periods), there were significant changes in the alluvial system that
affected the lower town on the plain below the palace at Tiryns. A stream
that once flowed to the south moved to the north of the palace knoll,
and about 4 m of sediment was deposited to the north and east of the
citadel. Houses were then built in the ancient stream bed, and a dam
was built, possibly in response to these dramatic flooding and erosional
events. The dam and a canal served to divert the stream to the south,
away from Tiryns (ibid.). At Tiryns, this early example of environmental
Argive
Plain
dam
si
nes
Ma
Tiryns canal
coastline
around 1200 BC
tholos
tombs
present
coastline
Nauplion
0 1
km N
floodwater and
sediment
lake
harbour
flow of clean water
canal
sediment deposit
harbour
original course
harbour
of the river
200 entrance
water
overflow
40
0 2 km N

3.9. Bronze Age coastal management systems at Tiryns (top) and Pylos (Palace of Nestor) (bottom) (by
permission of E. Zangger).
Early Ports and Harbours 49
management, which endeavoured to reduce the ever-present problem of

sedimentation and coastal alluvial processes, is in some ways a precursor
to the complex marine engineering that developed during the classical
and Roman periods.
Early Ports and Harbours
The development of ports on Mediterranean islands during the Bronze

Age was essential for integration into wider Mediterranean networks, and
in some cases, this led to islands such as Crete establishing themselves
as regional economic powers (Gkiasta 2010). Crete comprises a wide
range of coastal environment types and related environmental processes,
from progradation to tectonic uplift and downwarping. These environ-
mental complexities lead to the development of diverse responses. One
coastal process which is relatively rare across the Mediterranean is sand-
dune development. At the Minoan harbour town of Kommos (southern
Crete), sand dunes developed during the second and first millennium
BC. Eventually, the Greek sanctuary here was covered with sand after it
went out of use at c. AD 150 (Gifford 1995: 49). Sand-dune develop-
ment is initiated by the exposure of a large expanse of sand to wind. In
this instance, tectonic activity was the key to a change in environmental
processes. A rapid tectonic uplift event may well have exposed sand to
aeolian action across a previously submerged beach.
Others suggest that this part of Crete has experienced an uninter-
rupted sea-level rise. The presence of now submerged Roman fish tanks
and tombs implies that the sea level was 1.2 m lower 2,000 years ago
than it is today, and that between the fifth and the end of the first millen-
nia BC, relative sea levels changed on two occasions, moving from 3.8 m
to 2.5 m below the present sea level (Watrous et al. 1993: 204–5). At its
greatest extent, the shoreline would have been a further 90 m out to the
west. The fact that these measurements come from Matala Bay, a rocky
embayment just to the south of the Masara Plain, suggests that quite
distinct processes were taking place. Kommos was potentially subject to
a variety of different destructive processes. A large portion of one of the
major buildings at Kommos (building-T) seems to have been destroyed
by the sea. However, it is possible that a more immediate threat (dur-
ing LM I) was an earthquake, which may have damaged much of this
impressive building (J. W. Shaw 2006: 33–5). Moreover, sea levels could
have risen significantly during the occupation of Kommos. Floor levels

in building-T (J. W. Shaw & Shaw 1993) were raised by more than 1
m, most likely a response to the rising sea level, or rather the highest
reach of the waves (J. W. Shaw 2006: 56). Much of the cultural interest
associated with the assessment of these natural processes relies on a tight
chronology, that is, proof that these environmental events were actually
witnessed and responded to by the people living on the site. The research
at Kommos and its environs, perhaps more than most other sites, allows
us to assess the intersection of cultural and environmental processes.
Whilst the environmental processes demanded technological and cultural
responses, the environment did not seem to have an adverse impact on
the settlement and its people. A series of actions mitigated changes in sea
level and the movement of sand, in particular, the architectural modifica-
tions mentioned above.
Unsurprisingly, where possible, ports and harbours exploited natural lit-
toral topography. A major harbour on the south-eastern coast of Cyprus –
Kition – lay within a large protected bay. Established by the Late Bronze
Age, activity continued up until the end of the Phoenician period. There
are at least two harbours in this area; the first, in the northern part of this
area, is associated with the Late Bronze Age town, whilst the second, to
the east, is associated with the classical town. The choice of location is
explained by the fact that this area is protected from strong winds, and
is the closest point to Egypt and the Levantine coast (Collombier 1988:
35–7). From 4000 to 2100 BC, the deposition of fine sediments ulti-
mately led to the development of a coastal spit that seems to have been in
place at about 600 BC. The main archaeological site is now some 500–600
m from the shore, and an earthquake towards the end of the eleventh
century BC seems to have been responsible for the temporary demise of
the Late Bronze Age town. The town then developed again by the ninth
century BC, expanding across a much wider area (ibid.: 40–1). Adjacent to
Larnaca on Cyprus, the inner harbour of Bamboula (now 400 m inland)
was connected to Lichines (an ancient marine embayment) during the
Phoenician period (eighth to fourth centuries BC). The lagoon area was
gradually isolated from the sea, although the bay at Larnaca was still con-
nected. Ultimately, the lagoon was isolated from the sea, and uplift in this
area prevented access to the harbour (Morhange et al. 2000: 223).
The uplifted Peloro Peninsula on the north-east coast of Sicily is
another area that has witnessed significant changes during the Holocene.
Classical Coasts and Harbours 51
Today, there is no natural harbour, although in the past, and in particular

during the fifth century BC, the Pantano Piccolo was connected to the
Tyrrhenian Sea via a 120-m wide, 1-m deep channel. This harbour was
probably adequate for 320 ships (Bottari & Carveni 2009). A similar sit-
uation was also observed to the east at Tindari, where palaeotopographic
reconstruction has identified the probable location of a sheltered inlet
that would have been used during the fourth century BC (Bottari et al.
2009).
One example of a response to sea-level rise which is verifiable archae-
ologically comes from the Tel Dor site (Israel), dated to c. late thirteenth
century BC. This site included slipways and flushing channels that fed
seawater into an industrial area. Solution notches on these features were
used to infer past sea levels. Three phases of construction might indicate
how people responded to the rising sea level through the construction of
higher courses of stonework at +1.04, +1.57, and +2.53 m (Sivan et al.
2001: 108). Here, we see how cultural and environmental temporalities
did intersect, with a society responding to a specific change in their land-
scape. As stated earlier, it does seem that, in some cases, we can identify
landscapes or sites where societies were not only aware of a risk, but were
willing and able to negotiate the risk, as the benefits of remaining on
a site, or indeed cultural attachment to a site, led them to manage the
environmental hazard. These Bronze Age examples represent a range
of opportunistic exploitations of naturally occurring harbour locations,
as well as the early development of technological interventions in the
coastal landscape – engineering works that became all important during
the classical and Roman periods.
Classical Coasts and Harbours
The creation of artificial ports and harbours developed apace during the
classical periods, these installations forming the nodes in the quintes-
sential ‘connected’ Mediterranean (Horden & Purcell 2000: 391–5); a
space where everything Mediterranean (economy, culture, myths, flora,
and fauna) transited via ports and harbours. Whilst many Bronze Age
ports and harbours merely modified existing natural emplacements, the
classical and Roman periods saw the emergence of complex engineering
and management of harbour facilities. As with all periods, Mediterranean
sites were subject to a variety of environmental processes, and one

3.10. Relative sea-level rise at the harbour at Kenchreai (Peloponnese, Greece) (photo: author).
notable characteristic that we should not forget is that sea level at a num-
ber of Graeco-Roman sites was lower than it is today (Fig. 3.10); for
example, the sea was some 2 m lower when Delos was active (Duchêne,
Dalongeville, & Bernier 2001: 174). However, despite this rise, relative
sea levels in a number of regions, including Sardinia and parts of the
Adriatic coast (Florido et al. 2011) would have been comparatively sta-
ble. In some areas, the last 2,000 years has seen subsidence ranging from
about 1.5 m to between 0.63 and 0.89 mm per year (Antonioli et al.
2007). As stated earlier, this rate of subsidence might not have been
stable and linear. Moreover, subsidence may have been less than this,
with some suggesting a figure for the Roman sea level of about −50 cm
(Evelpidou et al. 2012).
We can categorise a range of environment types within which different
ports and harbours were situated in the past (Fig. 3.11). The first cate-
gory comprises protected estuarine inlets that are usually at the mouth of
drainage basins (e.g. Marseille and Phalasarna), or more open bays such
Unstable coasts Stable coasts

Submerged Uplifted Buried Buried Buried Buried Eroded
harbours harbours urban landlocked fluvial lagoonal harbours
harbours harbours harbours harbours
- Antioch
-Alexandria -Aigeira - Acre - Enkomi - Aquileia - Coppa Nevigata - Ampurias
- Baia - Lechaion - Beirut - Ephesus - Gaza - Cuma - Caesarea (outer
- Eastern Canopus - Phalasarna - Byzantium/Istanbul - Kalopsidha - Minturnae - Frejus harbour)
- Egnazia - Seleucia Pierea - Cartagena - Leptis Magna - Narbonne - Lattara
- Helike - Kition Bamboula - Miletos - Naucratis
- Herakleion - Marseilles - Malta - Ostia (Sardinia)
- Kenchreai - Naples - Priene - Pelusium
- Megisti - Olbia -Troy - Rome
- Miseno - Piraeus - Salamina - Sevilla
- Pozzuoli - Sidon - Schedia
-Toulon - Thebes
-Tyre - Valencia
- Zaragoza
3.11. Non-exhaustive list of harbours classed into seven groups. Categories of coastline types vis-à-vis impor-
tant Mediterranean harbours. (By permission of Elsevier. Marriner, N., and Morhange, C. (2007), Geoscience
of ancient Mediterranean harbours. Earth-Science Reviews 80(3–4), fig. 8.)
as at Troy. Some harbours or ports are built on the edges of deltas, such
as Fos (south of France), Ravenna (north-east Italy), and Aquileia; others
at the mouths of rivers and wetlands, with Fréjus (south of France), Ostia
(close to Rome), and Carthage (Tunisia) falling into this category. These
different environments have a direct effect on the preservation potential
of environmental and archaeological materials (Marriner & Morhange
2007) (Fig. 3.12). Those harbours built on rocky coasts often have
Greek or Phoenician origins. From the fourth century BC onwards, the
Greeks and Romans used less favourable environments for their harbours
as the technology for managing these environments developed (ibid.).
This technology included dredging, empoldering, and actually mov-
ing docks (Morhange et al. 1999: 147–8). For example, at Phalasarna,
on the western part of Crete, where fortifications were built during the
second half of the fourth century BC, the harbour started to silt up at
c. 67 BC, thus resulting in the possible blocking of the harbour entrance.
However, this low-magnitude process could have been managed; it was
the uplift event attributed to AD 365 that ensured the demise of this
harbour (Pirazzoli et al. 1992). The presence of solution notches on this
part of the coast does suggest that the sea level was stable for relatively
long periods. The uplift event probably occurred over an extremely short
period (either minutes or days) (Hadjidaki 1988: 466). Two canals that
once connected the harbour to the sea are now on dry land. On the side
of one of the channels, solution notches indicate that the sea level was
Proximity to the coastline Position relative to sea level
Buried urban harbours Submerged harbours

(calanques, pocket beaches (e.g. Alexandria, Portus Julius, Baia,
e.g. Byzantium, Marseilles, Kenchreai, Amathus)
Naples)
Uplifted harbours
Buried land-locked harbours (e.g. Phalasarna,
Medium preservation potential

(rias, deltas, estuaries Corinthus Lechaion)
Good preservation potential
e.g. Troy, Ephesus, Priene)

Sedimentary
environments
Buried lagoonal harbours
(e.g. Frejus, Cuma)
Buried river harbours

(e.g. Aquileia, Minturnae, Roma)
Taphonomy
preservation
potential
Poor
Eroded harbours
(e.g. Caesarea Maritima (outer basin),
Ampurias' Roman port)
3.12. Ancient harbour classification based on four variables: (1) proximity to the coast-
line, (2) position relative to sea level, (3) sedimentary environments, and (4) taphonomy.
(By permission of Elsevier. Marriner, N., and Morhange, C. (2007), Geoscience of ancient
Mediterranean harbours. Earth-Science Reviews 80(3–4), fig. 7.)
6.6 m higher when the channel was in use. It is possible that the Romans
deliberately blocked the channel in 67 BC, as some consider Phalasarna
to have been a port used by pirates (Hadjidaki 1988: 476). Some inter-
pret the Hellenistic port here as a ‘cothon’, dug in zones where natural,
sheltered harbours did not exist (Blackman 1982a, 1982b; Frost 1995).
These were often dug behind the coast in soft sedimentary environments,
such as lagoons, as was the case at Carthage. A channel would then be
excavated linking the cothon to the open sea. It is likely that the primary
aim was to create a port that was located within a town’s protective walls
(Carayon 2005: 11). Perhaps it is more reasonable to suggest that such
towns would have developed on or adjacent to natural sheltered ports if

such places had existed. As noted earlier, there are more natural harbours
in the Northern Mediterranean, whilst the southern coastlines (where
many of the cothon lie) tend not to provide such favourable environ-
ments (ibid.).
The term ‘cothon’ is particularly associated with Carthage, where a
channel linked a rectangular basin with the open sea. The area around
Carthage was initially covered with a dune blanket at the start of the
Holocene, from the Middle to the Late Holocene eustatic sea-level rise
eroded the headland. This was followed by the reworking of these sedi-
ments and the development of a spit, and the emergence of a lagoon
which was to become the setting for the Late Punic circular harbour
(Gifford, Rapp, & Vitali 1992: 585). Carthage also comprises a circular
basin. A similar configuration existed at the port of Mahdia to the south
of Carthage. The analysis of sedimentological (abiotic and biotic) data
demonstrated that the sediments beneath the Late Punic cothon were
from a lagoonal environment that could have provided a useable natural
site for a harbour, thus demonstrating the probability that the Cothon
was not the first harbour on this site (Vitali et al. 1992).
One of the tasks of modern coastal geoarchaeology in the
Mediterranean is to verify or question ancient descriptions of specific
ports, harbours, and their settings. For example, much work has been
done on the harbour of the Phoenician city of Tyre. Tyre, which was
once an island, is connected to the mainland by a tombolo (a sand isth-
mus), which was certainly formed by the nineth century BC, attested
to by the presence of the Phoenician necropolis of Al Bass (Carmona &
Ruiz 2008). However, Marriner, Goiran, and Morhange (2008) suggest
a later date of 332 BC, although the conditions for tombolo develop-
ment existed before this date. Ancient texts implied the existence of four
harbours at Tyre (Marriner, Morhange, & Carayon 2008). The central
area of Tyre’s seaport is now buried beneath the modern and medieval
city centres (ibid.: 1282). The northern harbour had Bronze Age origins,
and was protected by an aeolianite ridge system. The harbour continued
to develop during the Iron Age, and then the Hellenistic and Roman
artificially closed harbours were developed, as suggested by the fine tex-
ture of sedimentary units dated to this period (ibid.: 1292). As we have
already seen, the silting up of harbours is one of the most serious envir
onmental problems across the Mediterranean, and extensive dredging
of the harbour at Tyre took place during the Greco-Roman periods

(Marriner & Morhange 2006). Sedimentary evidence from Beirut also
demonstrates the development of a managed Roman harbour (Marriner
Morhange, & Saghieh-Beydoun 2008: 2510). Using controlled currents
was one method employed for de-silting. Such a system existed for the
inner south harbour at Tyre (Blackman 1982b: 199). In fact, harbours
are themselves sedimentary traps to the extent that accumulation rates
within harbours can be as much as 10 to 20 times higher than on natu-
rally prograding coasts (Marriner & Morhange 2007: 175).
On a regional scale, similar processes were also seen on the Menderes
Delta in western Turkey (Aegean coast) where the ancient ports of
Priene and Miletos suffered from silting. Whilst Priene’s harbour was
moved, Miletos appears to have lost its importance (Brückner 2001:
124). Geophysical work has shown how the harbour possessed a shallow
mouth with a deeper inner area.
In some areas, which may have been perceived as potentially danger-
ous for navigation, harbours were not necessarily built, and we should
not forget the importance of small ‘sheltering’ islands that could be used
as a refuge by sailors during storms (Greaves 2000). The exploitation
and manipulation of littoral zones that possess natural characteristics
ideal for the development of ports and harbours is quite understandable.
However, it is essential that geoarchaeologists verify the development of
such installations, and the nature of any interaction between people and
the littoral environment. For example, it was once argued that a Roman
harbour at Cumae lay within a lagoon (Paget 1968). However, recent
geoarchaeological research demonstrates that during the Roman period,
this part of the coast would have been far from ideal for the installation
of a harbour. From the second century BC through to the second cen-
tury AD, sandbars would have rendered navigation along this part of the
coast quite dangerous. Therefore, this area was characterised by a large
and potentially useful beach, but with difficult access. The substantial
quantities of pottery and structural remains indicate that the area was
exploited, but probably not as a port (Stefaniuk et al. 2005: 58).
The development of complex forms of environmental knowledge
is nowhere more evident than in the field of harbour and port con-
struction and maintenance. Possibly the most significant development
in ancient harbour construction was the Roman invention of hydrau-
lic cement (pozzolana). This could set under water, thus facilitating the
construction of harbours in areas where people had been unable to build

before (Blackman 1982b: 185). Unsurprisingly, ancient engineers had
a clear grasp of the processes that could undermine port and harbour
structures. The use of rubble cushions to protect sea walls that were
built on sand from under trenching caused by wave currents is known
at Caesarea (Raban 1988: 187). Moreover, Roman engineers used poz-
zolana concrete to build moles in order to create an offshore harbour
(Oleson et al. 2004; Raban 1992).
We have to ask if the palaeoenvironmental record always implies
‘improvement’ or rather ‘maintenance’ in the face of hostile environ-
mental processes. The rate of silting in ports at the mouths of rivers or in
sheltered natural coves (such as Marseille) obviously varies. As colonies
developed during the proto-historic periods, centres such as Marseille
became home to a wide variety of peoples – some, permanent residents,
others, itinerant traders. These different groups would have possessed
different attitudes to and understandings of the natural environment, but
would have had an expectation that major ports and harbours would be
accessible and operational. The management and maintenance of a port
environment in some ways represents a step towards the professionalisa-
tion of environmental management and the direct engagement with the
natural processes that contribute to riskiness in such a milieu. However,
there are certain processes affecting ports and coastal settlement, such
as tectonic movement and extreme sedimentation, that no society can
manage, and the only form of possible ‘improvement’ is mobility and
rebuilding.
The courses and sedimentary regimes of the rivers Arno and Auser
around Pisa changed a great deal during the later Holocene (Bruni
& Cosci 2003). Catastrophic floods on the Pisan coast seem to have
resulted in the movement of harbour activity to the north during the
Roman period (Lippi et al. 2007). During the second and first centuries
BC, the coastline was 7 km landward of its current position; prograda-
tion is thought to have started towards the end of the first century BC
(Pasquinucci & Rossetti 1988: 137). During the late Republican and
Early Imperial periods, it is likely that Pisa’s port was attached to the
city itself, and landing areas for boats have been hypothesised in the
city and along the River Arno at San Piero a Grado. As ships increased
in size, and the Arno and its delta prograded, the relocation of the port
just to the north of Livorno took place (ibid.: 139). A similar difficulty

3.13. Watercolour reconstruction of the development of the port at Marseille. Note how the ancient Greek
and Roman port is now inland, with the silting up of the port area having caused coastal advance, thus forcing
the development of the modern port out towards the south (by permission of J.-M. Gassend).
was experienced at the harbours of Ephesus and Smyrna, which were

also built near river mouths, where the silting up of the harbours forced
the seaward migration of these cities (Brückner 1997). Ephesus actually
went out of use during the Late Roman period (Blackman 1982b: 186).
Other sites suffered similar problems, including Aquileia, which may
have been replaced by Grado, situated on a coastal bar at the mouth of a
lagoon; Ravenna and Narbo were also subject to such processes. Silting
also affected Ostia, with a new port built about 3 km to the north of the
Tiber’s mouth. However, southerly winds and currents moved sediments
towards the harbour, and resulted in its eventual closure, despite the
construction of a mole designed to protect the area (ibid.: 187).
One of the most studied harbours in the Mediterranean is that at
Marseille. Founded by the Phoceans during the seventh century BC, it is
still in use today (albeit in a different form and environment) (Fig. 3.13),
whilst the harbour at Fos, to the west, was established some five centuries
later, and went out of use during the Middle Ages (Morhange et al. 2003;
Morhange et al. 1999: 145).
The ancient harbour at Marseille is now on dry land. Silting up of
the area has ‘forced’ the modern harbour to move seaward, whilst the
harbour at Fos is now under water. In order to understand these early
urban ports, we need to consider the impact that people had on the
landscape around the port. One of the most critical processes is erosion,
which leads to the silting of harbours. The earliest evidence in the port
area at Marseille includes Late Neolithic shell middens, but there is little
evidence for Neolithic erosion into what became the harbour zone. The
first ‘environmental event’ took place during Early to Mid-Bronze Age
(2020–1681 BC) when there appears to have been some silting up of the
area. Also, at about 1390–1043 BC, a shell midden was created. The next
major process took place from 600 BC onwards, when erosion from the
surrounding slopes seems to have become a problem. These sediments
included ‘urban refuse’, and are clearly indicative of an anthropogeni-
cally induced process, with pollution tolerant species of marine mollusc
appearing in these units. During the Roman period, soil erosion into the
harbour area continued. It seems that the Greek (Phocean) sediments
were derived from unpaved urban surfaces, whilst the Roman units were
the result of run-off across paved or cemented areas in the city. Although
there was some reduction in sedimentation at about AD 50, the Romans
were forced to dredge the harbour, cutting into the existing Greek sedi-
mentary units (Morhange et al. 1999: 151). This environmental history
of a specific port presents a series of environmental processes, or even
problems, that ancient peoples had to manage. Whilst these processes
were extremely difficult to control, they were probably just accepted as a
‘natural’ consequence of activity in this type of environment, and envir
onmental managers were resigned to the need for mitigation.
A markedly different situation existed just to the west of Marseille at
Fos-sur-Mer. Underwater archaeological research has revealed a complex
series of remains comprising a necropolis with stele and objects dated
to the end of the second or the beginning of the third century AD.
These sites were situated on two different sandbars that ran parallel to
the coast with a lagoon between the sandbars; both bars are now sub-
merged. The closest bar to the coast supported the boat construction
yard, whilst a necropolis was situated in a literally liminal position on
the second sandbar. Cores from this area demonstrate that the ancient
sea level (represented by a layer of pebbles dated to between 410 BC

and 180 AD) was about 2 m lower than it is today (Gassends & Maillet
2004).
On the Spanish littoral, relatively few Roman port constructions have
been found on the coast of Roman Baetica (Andalucia, Spain). Either
remains have disappeared because of coastal change, or alternative meth-
ods and locations for loading and unloading ships were employed. For
example, geoarchaeological investigations at the possible harbour bay
associated with the Phoenician site of Toscanos demonstrated that, dur-
ing the eighth century BC, the bay was partially filled, and the centre of
the bay was about 6–7 m deep. At c. 400 AD, a sand barrier existed on
the shore of the bay, with the coastal delta prograding since the Roman
period (Hoffman & Schulz 1988: 60–2). It is possible that the Romans
did not choose to employ their harbour-building technology in this area,
as sand beaches and navigable rivers may well have provided the sim-
plest solution. In certain situations where ships could not be beached,
slaves may well have waded and brought goods ashore, as is suggested
for the beach at Belo (Hohlfelder 1976: 467–8). Harbour installations
were only built where they were needed for economic or environmen-
tal reasons. Manageable, natural landing places would probably have
been quite common, but they are perhaps not always easy to identify.
We should not assume that Rome imposed technology for technology’s
sake; technological solutions were only employed in locations where the
environment did not proffer a natural solution within given economi-
cally instrumental parameters.
Whilst Rome’s original port was at Ostia, a new port to the north
was inaugurated a little before AD 64. Named Portus, it would serve as
the port of Imperial Rome, and would remain in use well into the Late
Antique period. From a geoarchaeological perspective, perhaps the most
intriguing characteristic of the sediments on the delta is the fact that
Roman structures are often covered with a clayey unit which implies
that sediments accumulated at quite a slow rate (Keay et al. 2006: 18).
Moreover, some geomorphological evidence points to the establishment
of the Republican settlement on a stable shoreline. Nevertheless, any
activity in this area had to negotiate incredibly complex and persistent
alluvial sedimentation (Bellotti et al. 1994; Mikhailova et al. 1998).
Coastal advance did not begin until the Imperial period (Keay et al.
2006: 26–9). This sequence demonstrates that the potential hazard for
Roman sites was not necessarily extreme, and this would have influenced
the manner in which Roman engineers responded to this landscape. In
fact, the engineering works themselves might have exacerbated any envir
onmental risk; once the harbour was established, its moles would have
had a dramatic effect on the deposition of sand from long-shore drift.
In fact, it is possible that the difference in coastal advance (400 m to the
north, 1400 m to the south) was a result of the moles disrupting this
process (ibid.: 30).
The northern quay of the Claudian harbour was situated in a land-
scape comprised of sand dunes, with salt and freshwater lagoons between
these dunes. Artificial wells were dug along the first line of dunes and
lagoons, and this evidence is taken to indicate activity in this zone from
the second century BC onwards.
There clearly comes a point in time when people, and engineers in
particular, are aware of the potential environmental problems, such as
sedimentation within a harbour (Figs. 3.14 & 3.15) (Morelli, Paroli,
& Verduchi 2006: 247). Either it was accepted that this would happen,
or the intensity and even the nature of the process was not understood.
Trajan modified the harbour during the second century AD, and it is
clear from the sedimentary evidence that this harbour was immediately
vulnerable to silting (Giraudi, Tata, & Paroli 2009), with the connection
between the sea and port basins only remaining open until AD 230–450
(Goiran et al. 2009).
The Roman investment in harbours was entirely logical, and we
should not forget that the imperative for the management of the inter-
face between land and sea resulted in the most impressive engineering
works in the ancient Mediterranean. It was not just the scale of the sites,
but innovations such as hydraulic concrete seen on a number of sites,
including Baiae and Portus Iulius in Italy (C. J. Brandon, Robert, &
John 2008).
Up until this point, emphasis has been placed on the study of har-
bour/port installations, with no consideration of other uses of the coast.
Specific niche resources, such as saltings, were present across much of the
Mediterranean, and variations in sea level would have affected the ability
to control the drying process. Salt production would have been impor-
tant from quite early on. The ancient Greek word for salt was the same as
for sea (Powell 1996: 12). In addition, people in coastal towns (or vicus)
would have embraced a variety of coastal economic activities.

3.14. The harbour at Portus. Geological map of the study area, including the Claudius and
Trajan harbour. (A) Fiumicino branch dammed by beach ridges; (B) mouth of the secondary
branch of the Fiumicino; (C) northern entrance of the Claudius harbour dammed by beach
ridges. (By permission of John Wiley and Sons, Giraudi, C. et al. (2009), Late Holocene
evolution of Tiber river delta and geoarchaeology of Claudius and Trajan harbour, Rome,
Geoarchaeology 24(3), fig. 2.
At Castelporziano, Lazio, Italy, an important vicus site has revealed

a series of substantial buildings, including a fish farm (Claridge 2007).
The coast as an attractive leisure or residential area was not a recent
innovation. The Laurentine Shore was Rome’s coastal zone, and com-
prised luxurious villas and the vicus mentioned above. This area was
situated in front of an existing dune ridge. Coastal progradation took
place during the Roman occupation phase on the Laurentine Shore, and

3.15. Watercolour reconstruction of the Claudian and Trajan ports. Note the different hypothesised recon-
structions from Schmiedt and Castagnoli (by permission of J.-M. Gassend).
subsequent settlements were established on this new land. However, it

seems unlikely that the dunes were reactivated before the end of the
Roman Period (Bicket et al. 2010). Situated between the ports of Ostia
and Portus, these buildings and the associated dune environment cor-
roborate the argument that this shoreline was relatively stable during
the Roman period. The sea level was relatively constant, and new sand
source areas were probably not exposed. Consequently, it is likely that
dune development was insignificant (Rendell et al. 2007). However, we
should not forget that dunes can move suddenly if the holding vegeta-
tion is reduced and storms whip up the newly exposed sand.
This Roman coastline was structured via the intersection of several
complex environmental, economic, political, and cultural processes, all
which combined in the development of this thalassocracy (Purcell 1998:
7). Despite the fact that this coastal area was defined (by some) as a wil-
derness, it emerged as a domain of otium (a concept of leisure that was
an alternative to the world of political and military power) (ibid.: 1).

However, this domain was also defined via its relationship to the sea and
the ever-changing coastline. Seaside retreats, as described by Pliny in his
letter to Gallus (Epistles 2.17) (Goalen & Fortenberry 2002), could only
exist once economic profits had been extracted via the successful man-
agement or mitigation of littoral processes.
Controlling Permeability
All environmental processes operate across a range of temporal and spa-

tial scales. Relative sea-level rise might have been perceptible at a gen-
erational level; the results of tectonic uplift or downwarping were often
perceptible immediately. These natural processes would have had dif-
ferent consequences for people, and their influence on the structuring
of everyday lifeways diverse. As well as emphasising the study of ports
and harbours as gateways facilitating connections and economic oppor-
tunities, we should put equal emphasis on the assessment of the com-
plex technological engagements, and related environmental knowledge,
associated with these places (Horden & Purcell 2000: 391–5). These
engagements reveal the underlying environmental constraints that act
upon certain forms of coastal activity. The development of ports and
harbours would have emerged as a new part of humanity’s relationship
with the coast and the sea, with people endeavouring to control spe-
cific coastal locales – locales that would have been subject to the natural
changes highlighted above.
At Atlit-Yam, we saw how the rising sea level not only inundated the
village (probably long after its abandonment) but initially spoilt access
to fresh water from the wells. Moving to a new site may have been
unproblematic, although the rediscovery of a freshwater source might
have been the most challenging task facing these people. The mooring
or beaching of boats during the Neolithic would have probably taken
place on suitable beaches, or against natural geological shelves (wharfs)
or in embayments (as natural harbours). Moving into the Bronze Age,
the development of early ports and harbours required the development
of mitigation schemes to protect these places from environmentally ini-
tiated redundancy. As noted at the outset, the Mediterranean Sea influ-
enced the nature of relationships between the various subregions along
Controlling Permeability 65
its coastlines. Changes in relative sea level, whether eustatic, crustal, or

changes in coastal configuration because of sedimentary deposition and
consequent progradation, would have had an impact on coastal socie-
ties as well as those in the hinterlands. Moreover, the activities of those
in the hinterlands would have influenced the nature of environmental
change along the coast, with eroded sediments eventually making their
way to the littoral, resulting in the steady seawards march of deltas in
some areas, or contributing to the silting up of ports and harbours. The
Mediterranean, more than most regions around the world, is defined by
this complex interplay of environmental, economic, and cultural pro-
cesses that lie between the mountains and the sea. Whilst tectonic activ-
ity could not be mitigated, some silting could. The infilling of harbours
was managed via dredging or controlled water flow. However, in some
cases, no amount of dredging could save harbours from the unrelenting
advance of the coastline. Sedimentation was so serious in certain areas
that some harbours have never been found by archaeologists, including
that at Luna (north-west Italy) (Bini et al. 2012). Lechaion (the western
harbour of ancient Corinth) was unfortunately exposed to tectonic uplift
and silting, with evidence for mounds of dredged silts still apparent today
(Morhange et al. 2012).
Small harbours were often constructed in hazardous areas, such as
capes, where unpredictable winds and currents forced boats and ships to
take shelter. In some instances, sanctuaries developed close by (Blackman
1982b: 188), representing a particular intersection of cultural and envir
onmental processes, where the ever-present risks associated with the mar-
itime environment were articulated via religious as well as technological
engagements with the landscape.
One criterion for the location of ports and harbours is their situa-
tion vis-à-vis sea currents and predominant winds; not all places along
a coast will be subject to the same currents and winds, thus proffering
access to the natural maritime ‘highways’. Direct access to certain cur-
rents and winds would have been critical prior to the development of
the ability to tack and sail against the wind. The wind was even referred
to as the Devil until recently in some parts of the Aegean (Powell 1996:
10). Specific proxies that allow us to identify changes in wind regimes
are rare. We need to consider to what extent the environment, in the
form of climate (wind levels, wind directions, etc.) and sea level (with
its direct effects on coastal configuration), would have been different.
46
44
42
40
38
36
34
32
30
-5 0 5 10 15 20 25 30 35
Main path Mesoscale wind-induced
Secondary path / recirculation
Mesoscale (instability)
Interannual path
3.16. Surface circulation of waters in the Mediterranean (after credits CLS/Pujol/2006: http://www.aviso.
oceanobs.com/ – permission granted).
Morton (2001: 6) considers that sea conditions in the Mediterranean

have not changed significantly since antiquity (Fig. 3.16). However, there
is no doubt that shifting weather systems (a corollary of climate changes)
would have resulted in variations in wind and current patterns, at the very
least altering the seasonality and timing of certain winds. We can infer
changes in wind direction based on variations in the movement and posi-
tion of seasonal pressure systems, although much modern research covers
the more recent past (Barriendos 1997; Barriendos & Llasat 2003). In
the extreme Western Mediterranean, anticyclonic conditions facilitate the
flow of superficial Atlantic water into the Mediterranean and littoral drift
increases, which enhance progradation, whilst cyclonic conditions create
the opposite conditions (Zazo et al. 1994). One argument that involves
a claim for a change in the direction of trade winds around the Near East
c. 7,000 years ago relates to the evidence for increased humidity during
this period, which could have been caused by a reversal in the relation-
ship between the anticyclones in north-east Europe and the Azores (Issar
et al. 1992). The Mediterranean comprises a transitional climate. During
the summer, it is usually influenced by the Azores anticyclone, and in
the winter, southward-moving depressions from Europe create instability
and regular north-westerly winds. Research into Little Ice Age climatic
changes implies that winter wind speeds in the Mediterranean would
have increased during such periods of climatic deterioration (Raible et al.
Controlling Permeability 67
2007). In fact, some routes might have been easier to navigate during the
winter months (Morton 2001: 233), although the summer would have
been the time of intense maritime activity with the etesians (ancient strong
and unpredictable winds) blowing from May to October. The nature of
sea currents was so influential that they contributed to the definition of
boundaries on dry land. Some regions, although close to one another
geographically, were ‘separated’ because different sea lanes had to be used.
For example, the North African regions of Maghrib and the Maghrib al-
Aqsa were treated as two distinct zones (B. D. Shaw 2006: 9).
The complexity of maritime and littoral processes resulted in the
development of complex forms of environmental knowledge, ranging
from navigation through to harbour construction and maintenance. The
agency of currents, winds, tectonic activity, and erosion have influenced
the very structure of Mediterranean regions, economies, and relation-
ships with the sea and coast. The combination of these environmental
and socio-economic processes constitute a core group of traits that char-
acterise the very essence of Mediterraneanism.
The growth and development of technologies and environmental
knowledge that would facilitate movement of people, materials, goods,
and ideas was of course a process that underpinned the emergence of a
connected Mediterranean. These connections (Horden & Purcell 2000:
ch. 5) demanded a change from passive to active engagements with the
environment. This process of technological intervention was probably
more intense across these coastal environments than in the other land-
scape types dealt with in the following chapters.
4
Rivers and Wetlands
Floods are a fairly common occurrence and cause serious damage across
large areas of land. Severe loss of life in Mediterranean floods does occur
today, and even when no lives are lost, the economic cost and distress for
thousands of ordinary people is beyond doubt. The 2002 flood on the
Gardon in the south of France claimed the lives of 21 people and caused
extensive damage. Although this flood was higher than those recorded
in written records, it was not the highest vis-à-vis floods recorded in
the sedimentary records. Deposits from caves 17–19 m above the nor-
mal river (base) level, which is 3 m above the 2002 flood height, show
how extreme events have taken place in the past (Sheffer et al. 2008).
The impact of short-cycle events (that often last between 24 and 48
hours) is something that geoarchaeologists can rarely identify with confi-
dence. However, these events influence humanity’s relationship with the
environment.
This chapter presents a description of the main types of hydrological
systems present in the Mediterranean. Issues of climatic variation and
precipitation regimes are considered, as well as the instability and unre-
liability of water supplies in some parts of the region. A more detailed
analysis of erosion histories and human relationships with the soil system
appears in Chapters 5 and 6. The examples presented here range from
small-scale, individual site-based studies, through to large-scale examples
that assess the archaeology of human settlement on or adjacent to impor-
tant Mediterranean rivers and wetlands.
68
Research Questions and Approaches 69
Studying Mediterranean Rivers and Wetlands: Research

Questions and Approaches
Rather than rehearse methodologies and approaches that are covered

in a number of books, in particular A.G. Brown’s (1997) comprehen-
sive assessment of alluvial geoarchaeology, this section will consider the
nature of the dominant research questions that comprise modern allu-
vial geoarchaeology (see also Howard & Macklin 1999). However, two
essential definitions are required: ‘fluvial’ and ‘alluvial’. The first cor-
responds to processes directly associated with a body of water (usually
a river) and its channel; the second, the sediments transported by that
body of water and then deposited, producing a terrace or flood plain for
example.
A relatively early exposition of research questions that are still cur-
rent in Mediterranean alluvial geoarchaeology was presented by Judson
in 1963. He stated that we needed to know whether changes in the
landscape were caused by people or ‘by some non-human cause’, and
that such changes may have had an impact on human exploitation of
a given landscape (ibid.: 287). His dating of the river terraces along
the Gornalunga Valley in central Sicily suggested a post-325 BC date
for the 8–10 m terraces based on the identification of a burial within
the terrace deposits. Judson demonstrated that prior to the Greek
burial, the valley was 10 m deeper than it is today. Then, subsequent
to that burial, the stream would have flowed some 10 m higher than its
modern level. Questions of alluvial ‘stability’ and ‘instability’, and the
probable influence of changes in topographic form along river systems,
are the dominant research questions half a century on from Judson’s
work. Our methods are more refined, but the fundamental research
questions regarding people’s relationships with the landscape have
not altered a great deal. An example of a valuable study that adopts
this approach is the work carried out around Basilicata, Southern Italy
(Piccarreta et al. 2011). Here, we see that periods were characterised
by extensive flood (slackwater) deposits: 5200–4800 BC, 2800–2550
BC, 2300–2100 BC, 1400–1100 BC, 350 BC–150 AD, and 300–680
AD. These sediments imply the movement of water beyond the banks
of the river, and that these events occurred during colder, wetter cli-
matic phases, whilst incision of the water body further into its bed
70 Rivers and Wetlands
took place during warmer periods. The phases of aggradation appear

to have been more intense during the last two millennia – possibly a
consequence of increased human activity. This geomorphological study
legitimately presents human activity as a process (Fig. 4.1). However,
archaeologists need to move beyond narratives where environmental
reconstruction divorces people from their environment, where peo-
ple constitute another forcing mechanism or emerge as an amorphous
physical process, implicitly or explicitly characterised as energy inputs
and outputs.
The majority of alluvial geoarchaeological/geomorphological work
tends to present the alluvial system as a sedimentary record which
serves as a proxy indicator for climatic or anthropogenic impact on a
landscape (see Bintliff 2002 for a review of such approaches). Some
studies also consider human responses to changes in river characteris-
tics, position, and/or form. One example, which moves beyond these
research questions, is the work carried out at Fiume di Sotto di Troina
River Valley, north-central Sicily (see Fig. 4.2 for a map with sites
referred to in this chapter). One aim of this project has been the assess-
ment of the possible impact of past agricultural systems on the Sicilian
landscape. In particular, the erosion history of the valley was investi-
gated. The lack of Neolithic material in this study area is possibly a
result of destruction or burial of Neolithic sites by colluvial and alluvial
sediments, as these early farming sites were located on the flood plain
(Ayala & French 2003; Leighton 1999). The models of human impact
and erosion suggest that the opening up of the landscape for intensive
pastoralism resulted in a greater potential for soil movement and that
this ‘…would have left a landscape more resistant to regeneration after
abandonment’ (Ayala & French 2005: 164–5). The issues of regen-
eration or more specifically resilience are notions that are addressed
throughout this volume.
Characteristics of Mediterranean Rivers

One notable characteristic of Mediterranean rivers is their steep gradients
across tectonic landforms, in particular, mountain ranges that embrace
the sources of many rivers (Mather 2009: 17). Watercourses are often
constrained by topography, and the distances between the source and the
coast can be quite short, thus producing the potential for fast-flowing
rivers with concomitant sedimentary loads comprised of coarse materials
Monticchio Glacial Mediterranean Human

record (41º N 16º E) Lake Levels Variations
Variations flooding phases pressure
9000
(Allen et al., 2002)
(Giraudi, 2005b)
(Holzhauser et al., 2005)
(Zielhofer et al., 2008)
(Macklin et al., 2010)
(Thorndycraft and Beni
Mesolithic
(Giraudi, 1998)
8000
(Primavera et al., in press)
201
10)
7000
Benito, 2006a)
Neolithic
6000
Calibrated years BP
5000
(advance phases)
Eneolithic Bronze
4000
3000
Greeks &
Romans
2000
(Magny et al., 2007)

(Giraudi, 2004)
Medieval Modern
1000
Age
0
(42º N 13º E)
Calderone
Tunisia
Greece
Spain
Crete
higher
higher
higher
higher
advance
lower
lower
lower
lower
retreat
0
0.5
1
60
80
100
Relative Probability Arboreal Alimini Fucino Mezzano Accesa Aletsch
High
Very
Low
high
(as in figure 2) pollen (%) 40º N 18º E 42º N 13º E 43º N 12º E 43º N 11º E
46º N 8º E
4.1. Cumulative probability density functions of 14C dates associated with major flooding in Basilicata plotted alongside
human and several palaeoenvironmental information: pollen records of vegetation change (Allen et al., 2002); hydro-
logical records from central-southern Italy lakes (Giraudi, 1998, 2004; Magny et al., 2007; Primavera et al., 2011; Alps
and Apennine glacier variations (Giraudi, 2005; Holzhauser et al., 2005); and Mediterranean Holocene fluvial activity
(Thorndycraft & Benito 2006; Zielhofer & Faust 2008; Macklin et al. 2010). (By permission of Elsevier, Piccarreta, M.,
Caldara, M., Capolongo, D., and Boenzi, F. (2011) Holocene geomorphic activity related to climatic change and human
impact in Basilicata, southern Italy, Geomorphology 128(3–4), fig. 5, p. 143.)

Sea of Azov
Bay
of
Biscay
24
25 23
26 Black Sea
1 9
7 10
5 2827
Gulf of
Lions
30 29
40 22 Adriatic
34
12 13 Sea
35 32
Balearic
3
Sea
33 Gulf of
1620 Aegean Sea
11 18
Calabria 8 19
17 37 15
Ionian Sea 38
4 36
2 39
31
Sea of Crete 6
14
Mediterranean Sea
21
Gulf of Sidra
4.2. Rivers, wetlands, and sites referred to in the chapter. 1: Gardon, 2: Gornalunga Valley, 3: Basilicata, 4: Fiume di Sotto di Troina River Valley, 5: Cecina River,
6: Lefkosia, 7: Lez delta, upon which the city of Lattara, 8: The Acheloös Delta, 9: Fontaine du Vaucluse, 10: Camargue/Vallée des Baux/Glanum/Le Carrelet, 11:
Coto Doñana, 12: Ebro Delta (Spain), 13: Pontine Marshes, 14: Sebkhet Kelbia, 15: Amuq Valley, 16: Piera, 17: River Aguas and Guadalentín depression, 18: Cacchiavia
Valley, 19: Konya plain/Çatalhöyük, 20: Penios River, Thessaly – Platia Magoula Zarkou, 21: Erani Terrace, 22: Biferno Valley, 23: The Po-Venetian plain, 24: Lagozza di
Besnate, 25: Campo Ceresole, 26: Villaggio Grande, 27: Ancona, 28: Suasa, Ostra, 29: Narce, 30: Fiora, Marta, and Treia rivers, 31: mid-Medjerda floodplain/Simitthus,
32: Gordion, 33: Velia, 34: Stobi, 35: Klidhi bridge, 36: River Xerias/Inachos River valley, 37: Copais, 38: Stymphalos, 39: Marshes at Lerna, 40: Lake Fucino.

4.3. Large boulders and stones are easily transported down many Mediterranean river courses with relatively
steep gradients over short distances. The River Golo, Corsica (photo: author).
(Fig. 4.3). Moreover, past tectonic events could also have led to signifi-
cant changes in hydrological regimes (Caputo, Bravard, & Helly 1994).
Tectonic processes are fundamentally important in structuring drainage
basins (Mather 2009: 23).
The lower Cecina River (Tuscany, central Italy) is an example of a tec-
tonically controlled river system, where channel instability and migration
may have discouraged permanent, extensive human settlement in parts
of the valley prior to the late medieval period (Benvenuti et al. 2008).
There is some evidence for Roman activity in the middle Cecina Valley
(Camin & McCall 2002), and the absence of sites in parts of this valley
may well be a function of geomorphological processes that have either
destroyed or masked archaeological remains (Terrenato & Ammerman
1996). There is little doubt that settlement during the Etruscan and
Roman periods in the wider Volaterrae region was relatively stable, and
there is little suggestion that the environment constrained agricultural

activity (Terrenato 1998). In addition, much activity would have been
centred on the coast (Pasquinucci & Menchelli 1999). It is possible
that some of the settlement patterns in this area can be explained by
either avoidance of certain parts of the valley or differential survival and
destruction of sites. In either case, an assessment of the alluvial regime
is a necessity.
Sudden tectonic events may have literally turned water supplies on
and off in some places. At Lefkosia, Nicosia, Cyprus, the flood plain
is characterised by an extensive series of flood deposits, with palaeosols
representing hiatuses between flooding events/phases. Moreover, the
change from an aggrading river system to an entrenched or down-cut
system is especially interesting, as such a regime change could have been
quite sudden – the product of tectonic activity. This situation demon-
strates the need for precise dating through the sequence if we are to
calculate dates of flooding and sediment accumulation rates. However,
dating on such sequences is never straightforward, and demands sensi-
ble inferences regarding the development of such a flood plain and the
nature of human interaction with that area (Newell, Stone, & Harrison
2004: 75).
Generally, river flows arriving on the Northern Mediterranean coast-
line tend to be perennial, although many smaller, inland rivers will often
run dry during the summer months. Moreover, Northern Mediterranean
rivers tend to have extensive catchments and drainage networks, whilst
Southern Mediterranean rivers have less extensive networks, and are
sometimes supplied by sources beyond the Mediterranean area, such
as the Nile (Allen 2001: 65). As the Mediterranean climate is charac-
terised by strong seasonal variation – dry summers with wet/stormy
autumns – this has obvious ramifications for river systems. Many riv-
ers are quite ‘flashy’, in that they will flood during autumn storms. A
number of Mediterranean rivers have their sources in mountainous areas,
and consequently their flow rates increase during the spring snowmelt
period. One problem that archaeologists must address is the extent to
which we can identify human responses to these ‘normal’ annual cycles.
Much geoarchaeological work is quite understandably concerned with
the identification and analysis of phases that are extra-normal, that is,
decennial or secular variations from supposedly normal flow rates and
flooding regimes.
30m
a b c)
Thalwag
Vegatated bar
Shoal or bar
Minor channels (e)
30m 30m
e f
4.4. River types (A) straight, (B) sinuous, (C) meandering, (D) braided, (E) anastomos-
ing, and (F) anabranching. (By permission of Cambridge University Press, Brown, A.G.
(1997), Alluvial Geoarchaeology, fig. 3.3.)
Mediterranean rivers can be placed into two basic categories: single

channel (sometimes, meandering) or braided (Fig. 4.4). There are of
course variations within these simplified categories. However, one reason
for this simplification is to introduce the notion that rivers, and par-
ticularly Mediterranean rivers, can undergo quite profound transforma-
tions, evolving from single-channelled rivers to braided (often straight
rivers with several small channels separated by bars that may be flooded
from time to time) and anastomosing (channelled rivers with bars that
are normally stable) (Fig. 4.5). Such changes would not only have had
(a)
(b)
4.5. Typical braided rivers – (Top) The Durance flows from the Southern French Alps into Provence (aer-
ial photo, by permission of Centre Camille Julian); (Bottom) The River Andarax view from Los Millares (SE
Spain) (photo: author).
an important impact on the physical nature of the alluvial landscape,

but they would also have had serious repercussions for people living on
or near the river. For example, the Lez delta in southern France, upon
which the city of Lattara (near Montpellier) was established during the
sixth century BC, started to develop from the Neolithic period onwards
(Blanchemanche et al. 2004: 159). As the Lez plain expanded, this pro-
vided Neolithic populations with new land. Therefore, one of the most
notable aspects of fluvial and alluvial environments is their capacity to
evolve as landscape forms. River channels can migrate and change their
breadth and depth, whilst terraces can form and radically change the
relationship of a river to its surrounding landscape. In the same area, we
also see how a river channel can also influence the structure of field sys-
tems. Here, the digging of early Iron Age ditches and then the location
of Roman rural settlement probably represent responses to changes in
the alluvial environment (ibid.: 171).
River channels not only contribute to the configuration of activities
in the landscape, but they also act as communication routes and bound-
aries, and are often imbued with deep religious or mythological mean-
ings (Frisone, 2012). One of the mythical twelve labours associated with
the hero Heracles includes the diversion of a river to clean the Augeian
Stables. The silting up of the river Acheloös caused a border dispute
between Acarnania and Aetolia (western Greece). The construction of
embankments and channels remedying this dispute was considered one
of Heracles’ acts (Salowey 1994: 77–8). The changes that can occur in a
river system can thus have an effect on any one of these aspects, and there-
fore the role of rivers and wetlands is primordial within any landscape.
The migration of river channels has serious taphonomic consequences
(Fig. 4.6), given that the position of a river course will obviously have
influenced the location of settlement sites and had repercussions for the
position of any adjacent flood plain. We should note that geoarchaeol-
ogy is not just concerned with ‘reconstructing’ past environments, it
also has a key role to play in understanding landscape taphonomy, that
is, assessing areas of site survival and destruction (Clevis et al. 2006).
The destruction of archaeological sites by meandering rivers stands as
an important example of direct impact upon society that should then
allow us to consider human awareness (or lack of awareness) of envi-
ronmental dynamics and hazard – a subject dealt with later in this
chapter.

4.6. Block diagram of a typical flood plain with its associated fluvial and alluvial features. (By permission of
Cambridge University Press, Brown, A.G. (1997), Alluvial Geoarchaeology, fig. 1.1.)
Springs and Karst

Although this chapter explicitly emphasises the study of rivers and
wetlands, there are other noteworthy hydrological elements, in partic-
ular, springs/thermal springs, and smaller streams, which are of fun-
damental importance in Mediterranean landscapes, feeding rivers and
wetlands.
Even though karstification is predominantly found on limestone areas,
it can also occur on other geologies (Fig. 4.7). It comprises the chemical
dissolution of rock by water, thus creating fissures, sinkholes, caves, and
underground streams. Phreatic (subterranean multidirectional) streams
create passages, and where water still flows in these passages, it can
re-emerge into the open landscape. Where these springs emerge from
depth under great pressure, they are referred to as Vauclusian springs
(resurgences), after the southern French Fontaine du Vaucluse (Lewin
& Woodward 2009: 296). Tufa (cold water) and travertine (often hot
water) are carbonate precipitates, the analysis of which can contribute to
the production of environmental histories. For example, some research
has revealed high tufa growth rates during the sixth and fifth millennia
BC in the south of France (Vaudour 1994), with certain areas seeing
high levels of tufa growth during the climatic optimum, but declining
after c. 2000 BC (Pedley 2009: 240). Recent work has presented a more

4.7. A typical Mediterranean karst system region showing material inputs, stores, and outputs and associated
processes in the vadose phreatic zones (modified from Ford and Williams and Gillieson 1996). The phreatic zone
lies below the water table (dashed line). Active tectonics in the Mediterranean can produce hydrothermal inputs
into karst systems (By permission of Oxford University Press, Lewin, J., and Woodward, J. (2009), Karst geo-
morphology and environmental change, in J. C. Woodward (ed.), The Physical Geography of the Mediterranean,
fig. 10.8, p. 297.)
complex picture of travertine development, with production continuing

in some landscapes until the Middle Ages, although the Neolithic saw
substantial changes in the travertine production regime (Ollivier et al.
2006).
Most importantly, karst landscapes comprise hydrological systems that
are hidden from the eye, with water moving underground and emerg-
ing in unexpected places – thus, the set of myths associated with these
landscapes and their water sources. These cultural aspects are considered
in more detail in the latter parts of the chapter.
Wetlands
Some of the most important wetlands in the Mediterranean include the
Camargue in the south of France, and abutting areas such as the Vallée
des Baux, the Coto Doñana and Ebro Delta (Spain), the Pontine Marshes
(Italy), and Sebkhet Kelbia (Tunisia).
Humid zones Small wetland zones on lake edges

within
depressions at
head of
catchment Artificial wet zones
Pond zone Wet zones on

floodplains
Marsh and
peat zones
Managed agricultural wetland
Managed brackish wetland
Littoral marsh
and lagoons
4.8. A transect of different wetland environment types.
As with any environment type, the characteristics of wetlands vary.

We can divide wetlands into categories founded on different develop-
mental stages: marshes comprise shallow water (fresh through to saline)
and contain grasses, rushes, reeds, and sedges; fenlands comprise devel-
oping peat areas and associated vegetation (Mitsch & Gosselink 1993).
These characteristics depend on the levels and cycles of water supply,
how close the wetland is to the coast, and the sedimentary inputs from
the fluvial systems feeding the wetland (Fig. 4.8). Some wetlands are
flooded for the entire year, whilst others may only flood seasonally.
Wetland systems are dynamic, not only over the long durée but also
on the annual scale. As with any environmental system, we need to ask
to what extent changes in the system are the results of climate change
and/or human manipulation. More intriguing from our perspective is
the debate regarding the utility of wetlands versus the potential dan-
gers or risks, especially that of disease (see Horden & Purcell 2000, ch.
VI.5). Some see wetlands as niches that are environmentally marginal
and dangerous due to the presence of malaria. A number of everyday
Alluvial Geoarchaeology: People and Climate 81
terms characterise wetlands in a pejorative manner: one can be swamped

with work, or one can be bogged down (Mitsch & Gosselink 1993:
12). However, many wetland zones are incredibly rich in resources
(especially in terms of fish, fowl, and flora), and malaria has not always
been present. Moreover, some wetlands, especially their edges, have
been used for agriculture (Morelli et al. 2008), and some societies have
endeavoured to drain wetlands in order to gain permanent agricultural
lands.
Alluvial Geoarchaeology: People and Climate
Much research into alluvial histories has considered the long durée
(Quaternary and Pleistocene) with a view to assessing riverine
responses to changes in climate (I. C. Fuller et al. 1998; Macklin et al.
2002; Macklin & Woodward 2009). As with many environmental pro-
cesses, the frequency of flooding events is related to climatic cycles.
Many models from across the Mediterranean articulate correlations
between changes in climate and river regimes. This section presents
some brief examples of issues addressed by recent alluvial research
around the Mediterranean, before moving on to more detailed regional
summaries.
In the Southern Mediterranean (Tunisia and Morocco), rivers appear
to be more dynamic during drier periods than during cooler, wetter
periods, which is the case in the Northern Mediterranean. A phase of rel-
ative aridity at about 4500–4000 BC interrupted an otherwise continu-
ous Neolithic presence in the Moroccan dry lands in the south-west and
north-east areas of the country (Zielhofer & Linstädter 2006). By the
end of the Neolithic (c. 2800 BC), there was an increase in fluvial activ-
ity in Tunisia. However, the relative dearth of archaeological material
for the post-Neolithic periods makes correlations between environmen-
tal changes and human activities difficult (Zielhofer, Faust, & Linstädter
2008).
Changes in flood-plain dynamics and, in particular, the course of the
river will have had important consequences for past peoples and their
engagements with and movement around that landscape (Fig. 4.9).
In the Amuq Valley, situated between the Tigris-Euphrates and the
Mediterranean, late Chalcolithic sites were situated on a low-energy

4.9. Idealised and simplified flood plain sedimentary systems: (a) a low-sinuosity sandy-
braided flood plain, (b) an intermediate sinuosity anastomosing flood plain, and (c) a sin-
uous avulsion-dominated flood plain These complex variations can occur across space but
also evolve in the same space over time, thus influencing the ways in which people engage
with that landscape. (By permission of Cambridge University Press, Brown, A.G. (1997),
Alluvial Geoarchaeology, fig. 1.2, p. 22.)
flood plain. This environment appears to have remained stable until

the late first millennium BC. A change to more powerful sedimen-
tation could imply an increase in flooding (and the strength of these
flows), or a migration of the river channel towards the area where the
geoarchaeological work was undertaken. Either way, this part of the
landscape clearly changed come the classical period. Increased settle-
ment activity in the surrounding hills could have been responsible for
the contraction of the lake and change in the alluvial system during
the late first millennium BC or early first millennium AD (Yener et al.
2000).
On reading most alluvial geoarchaeological reports from the
Mediterranean, we often see a narrative which (quite understandably)
places greater emphasis on human impact on alluvial systems and sedi-
mentation from the Neolithic onwards. Typically, this is when landscape
degradation is considered to have increased substantially. In northern
Piera (Macedonia, Greece), the study of two streams (the Gerakaris and
Agios Dimitrios) revealed long periods of relative stability and soil for-
mation, with infrequent but intense aggradation (the build-up of sedi-
ment), incision (the down-cutting into the bed of the river), and lateral
migration (movement across the flood plain). An alluvial unit in the
Gerakaris Valley was deposited during the final Neolithic, followed by
‘at least 1350 years of landscape stability’ (Krahtopoulou 2000: 22–3).
The next alluvial unit is dated to the period 2464–799 BC, followed by
‘stability’ for at least 1,100 years. Between the third and sixth centu-
ries AD (Late Roman period), a fill was deposited in the valley, and the
stream migrated to the north-east. A similar process was also identified
in the Agios Dimitrios stream, although channel migration does not
appear to have occurred here. Krahtopoulou (2000: 25) argues that
Early and Late Holocene alluvial activity was most likely controlled by
climatic factors, whilst Neolithic through to proto-historic changes in
these alluvial systems may have been influenced by both climate and
cultural factors. What we do see here are relatively long periods of sta-
bility and probable landscape resilience, with processes such as chan-
nel migration possibly forcing some changes in the location of human
activity.
In the Vera basin in south-east Spain, braided rivers, which existed
during the Pleistocene, developed into meandering river systems with
associated terraces during the Holocene. The development of the river
terrace sequence on the River Aguas (which feeds into the Vera basin)
is dated using both radiometric methods and archaeological material.
Evidence for Chalcolithic and Roman activity was found in one terrace,
with erosion phases following in both cases. The fact that terrace depo-
sition in this area is mirrored elsewhere in Spain led Schulte to suggest
that climate or sea-level control were probably the dominant factors,
rather than anthropogenic influences. Evidence for Neolithic land use –
and thereby impact – is considered insignificant. However, during the
Chalcolithic (4000–2300 BC), it appears that erosion may have been
caused by human activity (Schulte 2002: 96–7), whilst post-Roman
deposition may have been caused by climatic factors (Chapman et al.
1998 cited in Schulte 2002).
As stated earlier, the characteristics of rivers are directly affected by
tectonic movements. For example, during the last few thousand years,
8–10 m of subsidence in the Calabria region (Italy) has prevented
modern streams from incising (the streams have not down-cut, find-
ing the quickest route to the sea) (Abbott & Valastro 1995: 200).
Agricultural practices, and the consequent input of fine-grained sed-
iment into the fluvial system, may also explain this lack of incision.
Here, the complex interplay of natural and anthropic influences has
combined to change the nature of environments within this area over
time. Moreover, the fact that this pattern was not repeated in all of the
valleys in the region means that each community will have adapted and
developed environmental knowledge specific to their valley and river
system. In the Cacchiavia Valley, there have been periods of slow, fine-
grained sedimentation but with some short periods of incision (ibid.:
201). At c. 3500 BC, a phase of rapid aggradation comprised of silty-
clays occurred, followed by incision from about 3000 BC. Depending
on the rapidity of the changes, people may well have had to adapt
to this changing environment; for example, an incised river might be
more difficult to traverse, thus affecting movement around the valley.
Another phase of aggradation took place up to 2700 BC, and then
slow sedimentation continued for about another millennium. By 800
BC, an incision phase had taken place. River sedimentation has differ-
ent consequences along the length of a valley, and understanding the
links between what happens high up in a catchment vis-à-vis conse-
quences for the lower end of the river is a vital form of environmental
knowledge, as aggradation in valleys, such as in Calabria, will result
in progradation of the coastline. When combined with poor drainage,

this results in the formation of marshes and swamps. Such a devel-
opment may have had serious consequences for the local population
as agricultural land decreased and the potential for disease increased
(ibid.: 202–3).
These few brief examples exhibit the dominant narratives in alluvial
geoarchaeology/geomorphology. More often than not, we are obliged
to identify the causes of erosion or ‘instability’ within alluvial systems.
Prior to the development of agriculture, climate is often identified as
the cause, with people emerging as the key drivers of erosion after this.
The ‘negative’ impact of farming on vegetation and the pedosphere
undoubtedly did result in erosion in many instances, and this subject
will be dealt with in Chapters 5 and 6. We should, however, note here
that we sometimes lose sight of the fact that environmental stability and
resilience would have characterised long periods in many Mediterranean
landscapes, with areas witnessing phases of stability after the Neolithic.
Consequently, alluvial geoarchaeology can help elucidate the conditions
within in which certain activities and societies developed. For example,
the landscape conditions associated with the development of farming.
Alluvial Landscapes and Farming in Anatolia

and Greece
The importance of alluvial landscapes for the development of early agri-
culture cannot be understated. ‘Intensive but localised cultivation of
cereal crops on alluvial wetlands is thought to have provided the ecolog-
ical basis for the primary Neolithic settlement that spread across south-
west Asia and southeast Europe’ (Roberts & Rosen 2009: 393). In many
areas, the introduction of farming occurred on flood plains due to the
intrinsic fertility of these zones. However, activity on many flood plains
carries with it an inherent risk – that of flooding.
Çatalhöyük was located next to a branch of the Çarşamba River with
zones of marshy flood basins and marl hummocks (ibid.: 396). The site
developed as an artificial hill in order to avoid flooding on a poorly drained
flood plain (Roberts, Boyer, & Parish 1996) (Fig. 4.10). The presence
of ‘silica-skeletons’ or phytoliths (produced within the cells of crops)
in large quantities is considered an indicator of irrigated farming or of
farming in wet conditions. Despite the fact that the Konya plain (where
Çatalhöyük is situated) would have been wetter during the Neolithic,

4.10. Reconstruction of the Çatalhöyük landscape (reconstruction by Jon Swogger; by permission of Çatalhöyük
Research Project).
an initial study showed that there were surprisingly few silica skeletons,
suggesting that wheat was grown on better-drained soils away from the
site itself (Rosen 2005: 211). However, a more recent study of phyto-
liths indicates that wet farming of wheat was indeed undertaken (Shillito
2011). We should still consider the possibility that the area around the
site could have been flooded for parts of the year, and whilst this niche
offered a series of wetland resources, arable agriculture and pastoralism
might not have been pursued in all zones around Çatalhöyük.
Neolithic mobility, where people responded to changes in specific
environmental processes, was the norm. In fact, from a human ecologi-
cal perspective, mobility is a basic strategy in the maintenance of environ-
mental resilience, even amongst agriculturalists. The flexible application
of environmental knowledge across different niches was part of an
environmental culture inherited from a time when mobility was a way
of life. The landscape around Çatalhöyük did change with time, with
alluvial deposition stopping between the Neolithic and the Chalcolithic.

Archaeological survey demonstrates that there were increases in popula-
tion in the Near East from the Early Bronze Age, and the Konya Basin
was no exception, with increases in the number and sizes of sites on the
Konya plain (Boyer, Roberts, & Baird 2006: 683–9). It appears that
people in this area successfully managed a complex hydrological land-
scape with its changes in river regime and variation in soil salinity.
In Thessaly (northern Greece), Neolithic settlement first occurred on
the flood plain of the Penios River during an aggradation phase, possibly
caused by soil erosion, although it seems likely that this erosion was not
caused by early farmers (Bintliff 2002). People were probably attracted
to the annually deposited silts (van Andel, Gallis, & Toufexis 1992: 131).
Subsequent to this early phase, occupation mounds developed as ‘human
accumulation’ overtook the natural rate of sedimentation; that is, even
though the deposition of alluvial sediments continued, the settlement
sites on the plain maintained levels above the accumulating plain.
During the Early and Middle Neolithic, the development of the Platia
Magoula Zarkou site (situated on a mound some 5 m high and 200
m across) was contemporary with the deposition of several metres of
alluvial sands that represent flooding events (ibid.: 138). Early farmers
exploited the moist silts deposited by floods. Despite the risks associ-
ated with working in an active flood plain, which included poor yields
on wet areas (van Andel, Gallis, & Toufexis 1995: 140), wetland edges
and flood plains would have been more suitable for pre-plough culti-
vation (Bintliff et al. 2006). Consequently, permanent, year-round set-
tlement may not have been possible (Whittle 2002: 17). A void in the
site distribution pattern might represent the agricultural land exploited
by the sites closest to the lake (Perlès 2001: 130). A similar process has
been identified in the Near East, where alluviation increased during the
Middle Holocene. However, the sediments deposited by these streams,
such as those around the Erani Terrace (the Shephela foothills on the
central coastal plain of Israel), were fine, silty units that were character-
istic of a flood-water farming system (Rosen 2007: 86–8). In both of
these areas, we can see how locating on the edge of a zone susceptible
to flooding was important. During the Early Bronze Age, flood-water
farming here became impossible as the wadi started to incise. Individual
subsistence farmers could probably recover after one or two years of
low rainfall. However, more complex societies with a high number of
‘non-producers’ may have been less resilient (Rosen 1995: 32–3). As the
streams cut down over time and water supply to the land was reduced,
the ‘buffer provided by floodwater farming’ diminished (ibid.: 39).
In some instances, Thessalian sites were located some 8–10 km from
rivers and streams (Perlès 2001: 135). Certain environmental charac-
teristics in this region may thus have had a ‘repulsive role’ during the
Neolithic (Perlès 1999, 2001); some areas may have been too wet.
Statistical analyses assessing the nature of the settlement pattern imply
that each village would have had a roughly 450 ha territory. As settlement
density on the poor-quality soils (Ayia Sofia soil) was lower, this sug-
gests that these people were aware of variations of soil quality. However,
Perlès (2001: 143) does not feel that soil quality influenced settlement
choice; it is possible that seasonally flooded zones, along with marshes
and arid zones were avoided. The highest density of sites occurred in
areas that were never flooded and in environments that Perlès consid-
ers ‘homogenous’. However, homogeneity is a problematic notion, as
people undoubtedly created ecological patches, thus rendering these
landscapes heterogeneous. The fact that sites in this area appear to have
been close to one another implies a common notion of what environ-
ment type was the most productive, and collective or shared resources
would have been a cultural characteristic within these landscapes. In
central Greece, the settlement distribution was different, with dispersed
farms prevalent. Although a common environment type appears to have
been chosen, with Early to Middle Neolithic sites also situated adjacent
to wetland fields (Bintliff et al. 2006: 671–2), it is possible that in drier
climates, areas of high water tables with their associated soils formed
discrete zones – quite different to the plains of Thessaly and areas fur-
ther north (Bintliff 2012: 60). These environmental characteristics had a
direct influence on settlement density and distribution.
Aspects of Alluvial Archaeology in Italy

A great deal of alluvial geoarchaeological research has been carried out
in Italy, and some of this has been combined with landscape survey. The
Biferno Valley (central Italy) landscape survey comprised a geoarchaeo-
logical element. Here, seven sedimentary units were identified, the first
being Neolithic/Bronze Age, then classical, medieval, early post-medi-
eval, nineteenth century, early twentieth century, and then, finally, the late
twentieth century. During the Early Holocene, the Biferno developed
from a single meandering channel system to an aggraded one (Hunt

1995: 73–4). Most of the settlement sites with later Neolithic pottery
were situated in the lower valley, whilst sites with later Neolithic blades
tended to be close to the river. Most of the Bronze Age sites were found
in the middle valley (ibid.: 138). Therefore, it seems likely that certain
activities took place in specific locations along the course of the river.
Unsurprisingly, activities directly intersect with the environment. Whilst
the identification of specific activities might not be possible, we can assess
a range of potential relationships with natural environmental processes.
For example, those sites with pottery would have been situated on a
part of the river that was slower moving, and thereby represent activ-
ities in zones where river behaviour was more predictable. We should
not forget that many riverside activities would have been temporary, and
one response to a change in the alluvial system would have been quite
simply to move somewhere else. During the Iron Age, there appears
to have been a trend towards settlement in the lower part of the val-
ley (ibid.: 162), then during the classical period, an intensive aggrading
regime developed, centred on the Samnite and Early Roman periods. An
increase in activity during the Roman period (between the third century
BC and the first to second centuries AD) seems to be contemporary with
a phase of aggradation (Barker & Hunt 2003: 187). This landscape pro-
cess, lacking detailed chronological resolution, may have taken place over
a period of time that had no relevance for the communities living and
working in this area. At best, such a process may represent periodic soil
erosion. Despite these chronological issues, it should be apparent that
changing fluvial and alluvial regimes would have engendered changes in
activities that took place along the river.
The biggest alluvial basin in Italy is the Po-Venetian Plain. It comprises
71% of the plain zones in that country (Pellegrini 1979 quoted in
Marchetti 2002: 361). The development of the Po plain, with its net-
work of meanders, is best appreciated from aerial photographs or satellite
imagery (Tozzi 1993) (Fig. 4.11). Even a cursory examination of these
documents should convince anyone that this landscape has witnessed a
complicated alluvial history that has influenced the distribution of settle-
ments, the structure of field systems, and the day-to-day lives of ordinary
people for millennia.
The Po is navigable up to 257 km inland from the sea, plus it is
linked with a series of major lakes (Como, Lago Maggiore, and Garda)

4.11. The Po plain with its complex fluvial network – LANDSATT image (courtesy of the U.S. Geological
Survey).
(Mikhailova 2002: 370). Perhaps the most noteworthy characteristic of

the Po delta is its progradation. Between the Etruscan period (sixth cen-
tury BC) and 1600 AD, the delta advance rate was 450 m/100 years, but
after AD 1600 (during the Little Ice Age), it was 7 km/100 years (ibid.:
375). As already noted, such rates of environmental change may not have
been constant, although some transformations would have been percep-
tible on a generational scale, and we would expect to see the influence on
the chronological spread of sites in some areas, with a leading edge trend
of more recent sites following the prograding delta outwards.
A number of surveys in Italy have demonstrated how Early to Middle
Neolithic settlements were situated above important rivers (Malone
2003: 258). The Vhò culture (5340–3990 BC) settled on terraces and
hill slopes on the northern Po plain. The Lagozza culture (4000–3300
BC) was dominant across much of the eastern Po plain with the devel-
opment of wooden platforms on lake and riverside areas, such as that at
Lagozza di Besnate (Malone 2003: 263–4), articulating a specific form

of environmental knowledge and response to the character of this land-
scape. During the Early Neolithic, people on the Po plain maintained
so-called Mesolithic food-gathering strategies (hunting and fishing), but
they also developed agriculture. Freshwater fish, molluscs, deer, boar,
chamois, beaver, wolf, and turtle were present on sites such as Campo
Ceresole (ibid.: 271). Areas such as the Po plain might not have wit-
nessed a swift or complete transition to farming partly due to the richness
of the naturally occurring food resources in these areas. Farming in more
marginal and arid zones to the south would have made sense as part of a
new risk-buffering mechanism that comprised greater specialisation than
in the north. We can see how this region, despite what seem to be quite
impressive landscape changes across the Po plain in terms of meanders
and evidence for progradation on the coast, was a rich and stable source
of food resources.
By the Middle to Late Bronze Age (1650–1200 BC), settlements of
the Terramare culture (the villages are referred to as ‘Terramaras’ in the
cental Po valley) developed a landscape dominated by cereal fields, pas-
tures, and meadows that appear to have been suddenly abandoned at c.
1150 BC. The possibility of flooding may have encouraged the building
of houses on wooden piles between 1600 and 1400 BC. Later, the houses
were built directly on the ground, without piles (Mercuri et al. 2006: 56).
Many of these sites are associated with a palaeochannel that was active
whilst the site was occupied. A series of 45 wells that showed signs of hav-
ing been regularly re-excavated after collapse was found on the Villaggio
Grande site. These wells followed the local water table, which in turn was
linked directly to the Po River. Towards the end of the Bronze Age, an arid
phase developed, and the flow rate of the Po was reduced. This phase is
recorded over much of Europe, with lake levels falling and alpine glaciers
retreating (Cremaschi, Pizzi, & Valsecchi 2006: 95). The development
of a drier environment is inferred from the presence of Cichorioideae (a
plant that usually inhabits dry areas), which fits with a picture of increased
human pressure on the landscape (Mercuri et al. 2012). The loss of a
reliable water supply may have led to the abandonment of this area, or
at the very least led to changes in the way in which the landscape was
used. Here, we see how certain sets of practices (including elements of the
settlement and economic system) are not persistent. We do not know if
environmental changes were directly responsible for these changes, as the
identification of correlation (i.e. climate change coinciding with changes

in settlement and economy) is not explanation.
Late Proto-Historic and Classical Alluvial and

Hydrological Landscapes
As complex (urbanised) societies developed, settlements obviously
took on greater permanency, and therefore responses to changes in
river regimes demanded technological intervention in one form or
another. However, evidence is often elusive, such as in the Nemea Valley
(Peloponnese), where it seems that after the phases of abandonment or
settlement reduction, some form of drainage system might have been
required at certain times in order for agriculture to flourish in this area.
This was probably the case during the Early Christian and Byzantine
periods (Wright et al. 1990: 644–5).
One example that provides archaeological evidence of responses to Iron
Age flooding comes from Campania (Italy). The alluvial ‘crisis’ during the
sixth to fifth centuries BC took the form of rising river and stream levels
on several sites. At the Piano di Sorrento site (Naples), the first phase com-
prises Chalcolithic tombs adjacent to a watercourse. A sixth- to fourth-
century BC site was then subject to severe erosion from the stream. The
size of some of the architectural blocks moved downstream attests to the
force of the major flooding events at this time. A second (fourth cen-
tury BC) structure was built above the stream, and appears to have been
regularly covered with flood deposits. An artisan zone on this site was
abandoned during the second century BC due to flooding. Then, as a
consequence of the AD 79 eruption of Vesuvius, surrounding hillsides
became unstable and a layer of up to 3 m was deposited on the site (Albore
Livadie 2003: 360). Here, there appear to be three ‘events’ that proba-
bly had immediate consequences for people. However, these events took
place over a relatively long period, and our scale of analysis dictates how we
assess resilience in such an area. Whilst there is no doubt that people were
affected by specific events, as possibly indicated by the abandonment of the
artisan zone, over the longer durée, activity appears resilient.
Changes in the form and characteristics of a river channel will have
greater repercussions for large, stable, fixed settlements. The change from
a meandering to a braided river is a common feature of Mediterranean
fluvial systems. Such a change will not necessarily have caused problems
for nearby settlements (although navigability might become difficult).
The causes of such a change can be many and varied. In the Marche
region, such a transformation may not have been caused by climate
but rather by a change in sediment supply. Variations in sediment sup-
ply can result not only from quite spatially specific variance in geology,
but also human interference in the geomorphic system. Deforestation
by Neolithic communities (Malone et al. 1992) is one possible cause
(Coltorti 1993: 321–2). The Roman towns of Suasa (along the middle
Cesano) and Ostra (along the middle Misa) were partly constructed over
Holocene flood-plain deposits, with some buildings at Ostra built on
terraces produced by the meandering system, thus implying that this sys-
tem could have existed up until the Roman period and perhaps even later
(ibid.: 317). Moreover, the terrace upon which the Romans built was
probably stable and not subject to flooding.
At Narce (Lazio), there are two alluvial terraces, the largest of which
is about 4 m above the present river, whilst the lower terrace stands at
about 1.5 m (Cherkauer 1976: 108). Just prior to the Roman period,
two alluvial units were deposited, and then during the Roman period
itself, the stream migrated towards the eastern side of the valley, followed
by renewed migration to the west. Therefore, it is clear that channel
migration occurred during the Faliscan settlement at Narce (ibid.: 115–
7). Brown and Ellis (1996) argue that whilst climate variation had some
impact on the Fiora, Marta, and Treia rivers, it seems more likely that
changes in the density and nature of human activity had a greater impact
on these rivers’ regimes and caused higher rates of overbank deposition,
especially as settlement density increased during the Early Roman period.
Though the dating of alluvial sediments has always posed a fundamental
problem, the use of palaeomagnetic and optically stimulated lumines-
cence (OSL) perhaps offers greater accuracy.
During the long period over which the Roman Empire existed, there
were obviously substantial changes in some rivers, and people would have
had to adapt to those changes. We know that alluvial terraces formed in
a number of Mediterranean landscapes during the Roman period, espe-
cially in south-east Italy and parts of Sicily (Neboit-Guilhot & Lespez
2006: 340). There is no doubt that flood mitigation strategies existed
(a more detailed discussion of urban flooding is presented below),
and one specific response comprised the planting of alder trees along
riverbanks as a flood-protection measure (Meiggs 1982: 376). Purcell
(2005) considers the dynamic relationship between the periods of urban
stasis, characterised by architecture and social practices that constitute

social memory, and the dynamic changing environment, in particular,
that beyond the city walls over which societies have little control. There
are numerous instances of towns and cities that have been damaged,
destroyed, or gradually eroded by natural processes.
North African alluvial systems are fundamentally different to their
southern European counterparts. Rivers will often run dry or, at best, have
a minimal flow. Therefore, they rely on seasonal rainfall patterns that vary
from year to year. Moreover, some rainfall events might be so extreme that
severe flash flooding will not only cause damage, but also result in too
much water arriving at once, with water actually being lost from the system
and not being stored or exploited by people. As with the examples above,
the question of phases of stability and instability also underpins studies of
river systems in the Southern Mediterranean. For example, the study of the
mid-Medjerda flood plain (northern Tunisia) demonstrated geomorphic
stability via a period of soil formation that occurred between 6,000 and
4,700 years ago. This was followed by what the authors refer to a ‘Mid-
Holocene climatic collapse’ (Zielhofer et al. 2004: 859) when alluvial activ-
ity increased due to an aridification of the climate (a situation where surface
run-off increased due to sparse vegetation). During the Roman period,
the area around the flood plain appears to have been stable, even though
agricultural activity was intense. During the Roman occupation, settlement
was concentrated on the flood plains and the hills of the Medjerda River
catchment. Zielhofer and Faust (2003: 210) tell us that ‘Archaeological
evidence suggests that Roman farming techniques in North Africa were
adjusted to the natural conditions’. Whilst it is unsurprising that farming
settlements should be located along a fertile river catchment in an oth-
erwise arid region, we should be wary of assertions that imply a ‘natural’
adjustment, when responses to and engagements with any environment
are heterogeneous. There is little doubt that Roman North Africa was an
extremely productive part of the Empire and that the management of the
flood plain was in many ways a macro-economic success.
Urban Alluvial Geoarchaeology: Glanum, Rome, and Gordion
Glanum
Whilst the vast majority of this book deals with rural landscapes, more
often than not, floods and human responses to fluvial systems are
Urban Alluvial Geoarchaeology: Glanum, Rome, and Gordion 95

4.12. Glanum in Provence located within a small limestone valley that is subject to periodic flooding (photo:
Lisa Rayar-Bregou).
registered in urban environments. Understandably, the probability of

loss of life and damage to property is higher within urban environments
than rural ones.
In the south of France, the topographical situation of Glanum prof-
fers an interesting example. The town is located at the opening of two
valleys within a complex karstic zone with plentiful water (Fig. 4.12),
and has therefore always been exposed to intensive alluvial processes.
These valleys were particularly susceptible to flooding from the end of
the Neolithic to the end of the Iron Age (Jorda & Provansal 1989). This
Iron Age town comprised features designed to mitigate flooding and
conserve water for the dry periods. Basins were constructed in the Vallon
Saint-Clerg during the second half of the second century BC (Agusta-
Boularot & Gazenbeek 2003: 105). The construction of drainage sys-
tems was a common occurrence on many late proto-historic sites in the
region. The collection and storage of water on the site was so important
that the water cisterns, known as the ‘nymphée’, were monumental in
design (Agusta-Boularot et al. 2004: 31). Another hydrological structure
was the dam (built during the Roman phase of occupation) that stored
water during the summer. Geomorphologists initially suggested that it
did not seem logical to settle in this area, as the presence of a ‘flashy’
stream regime posed an obvious threat, whilst archaeologists consider
that such a situation, with available water in an otherwise relatively dry
landscape, was an obvious choice for a settlement (Provansal 2006). As
much as any other site, Glanum emphasises the risks that people will take
in order to ensure an adequate water supply, being a resource so often
unpredictable across the Mediterranean.
Rome
Undoubtedly, one of the most culturally significant rivers in the
Mediterranean is the Tiber. The river was fundamental for the devel-
opment of Rome as an artery that facilitated links between the differ-
ent parts of the city, and then beyond as the key communication route.
Settlements manifestly developed along the river, beyond the walls of the
city (Patterson et al. 2000) and would have been subject to flooding.
However, whilst we should not underplay the impact of floods on rural
communities, urban groups do not have the same flexibility of response
to major floods in that they are constrained by urban space and architec-
ture. In certain situations, rural communities can move and settle new
ground, whilst city dwellers must rely on fixed technologies, planning,
and rules of ownership that may or may not successfully mitigate flood-
ing. Moreover, urban dwellers need to consider and manage processes
that affect the catchment of their river, in some cases processes well
beyond their control, especially if the town is within the catchment of a
mountainous area, as is often the case in the Mediterranean. For exam-
ple, changes in the alluvial regimes of the Central Apennines would have
been important for alluvial processes downstream around Rome. Such
changes have been correlated with Tiber flooding events around Rome.
At Campo Imperatore, in the Central Apennines, the first terrace overlies
a soil that is dated to 2830–2410 BC, and the second is dated to 190
BC–AD 10. The second terrace perhaps corresponds with the high num-
ber of historical accounts for extensive flooding from the second century
BC through the second century AD. This was then followed by another
phase of increased flooding from the fifth to ninth centuries AD. There
is little documentary evidence for floods during the fourth and third cen-
turies BC, nor for the third and fifth centuries AD.
It appears that the most significant floods were those that occurred
during the second and first centuries BC (Giraudi 2005: 771). The loca-
tion of Roman residential areas above the flood plain suggests that floods
must have threatened property on a number of occasions. Moreover,
certain public buildings, such as theatres and athletic centres, were sit-
uated on the flood plain (Heiken, Funiciello, & De Rita 2005: 63). To
some extent, these were expendable, in the sense that they were not
permanently occupied by large numbers of people or they were easier to
protect. Over time, starting in the third century BC, the Romans con-
structed a major drainage system, the Cloaca Maxima, which took flood
waters away from the city. Underground drainage systems (Cuniculi)
were in fact initially developed by the Etruscans (Judson & Kahane
1963).
The central zone of the Tiber delta prograded between the first cen-
tury BC and the first century AD, and phases of flooding here correspond
with the majority of the erosion phases identified in the Tiber catchment
in the Central Apennines. Consequently, environmental changes in the
mountainous area were having an effect on the flood plain around Rome.
As ever, the question is what caused these changes in the alluvial regime?
The first candidate for any such change is an evolving glacier system that
fed the catchment source. We know that the medieval glacial advances
did correspond with the alluvial phases identified at Campo Imperatore
and flooding of the Tiber (Giraudi 2005: 772).
Le Gall (1953) produced the first comprehensive list of floods that
affected Rome. He detailed 25 floods for a 500-year period (the last
300 years of the Republic and the first two centuries of the Empire), a
period which broadly corresponds with Giraudi’s (2005) second century
BC to second century AD alluvial phase, and activity that was probably
caused by changes in the precipitation regime. Contemporary written
accounts spanning 800 years describe 33 years in which floods occurred
(Aldrete 2007: 14). The peak in the number of recorded floods falls
between 200 BC and AD 200 (ibid.: 74). The Campus Martius (which
would have included temple of Apollo and the altar of Mars) is one area
that was frequently flooded. Extreme floods (20 m above sea level) would
have covered many of the important parts of the city, including major
political, public, and entertainment structures (Figs. 4.13 and 4.14).
Most floods take place from November through to February. Frequent
low-level floods (e.g. 10 m above sea level) would have a greater impact

4.13. Rome with 10 m flood height indicated (by permission of G. S. Aldrete).
on the construction of environmental knowledge and concomitant mit-

igation strategies, not just for city dwellers, but also for those whose
farmland around Rome was damaged.
The fact that the peak of settlement activity in the Tiber Valley
(Patterson, di Giuseppe, & Witcher 2004) does coincide with peaks in
recorded and inferred flood activity raises a number of issues. First, does
the increased number of floods reflect increased reporting of such events
due to high population levels and concomitant economic concerns with
the impact of floods? Alternatively, if the increase in flooding was real
(Giraudi 2005; Giraudi et al. 2009), then resilient forms of environmental
knowledge and mitigation must have developed in the Tiber catchment

4.14. Rome with 20 m flood height and principle public buildings (by permission of G. S. Aldrete).
that permitted the maintenance of successful agricultural regimes here.

Moreover, flood silts could have enhanced the agricultural potential of
certain areas, rendering them more attractive.
Rome undoubtedly made the link between rainfall in the mountains
and consequent flooding in Rome. Moreover, we know that a number
of lakes were altered/managed as part of mitigation strategies. Ateius
Capito and L. Arruntius were charged with the development of projects
designed to modify the course of the Tiber and its tributaries (Leveau
2008b). Therefore, to what extent were any works concerned with the
entire fluvial system? Remote sensing has shown that a Roman drainage
system may have been established in the mid-Tiber Valley (De Meo et al.
2003). If the aim of river-channel and drainage-system management was

to protect Rome, and thus concentrated on upstream modifications, what
happened downstream? We could also consider whether settlements to
the east of Rome, and towards the coast, were left to suffer the conse-
quences of upstream mitigation strategies if the protection of Rome was
the priority. One obvious outcome for the coastal zones, in particular for
ports, was the coastal advancement of sediments, resulting in the silting
up of these areas – a topic dealt with in the previous chapter.
Obviously, other important towns were subject to flooding. There is
evidence of serious floods around Pisa during the period from the sec-
ond century BC through to the fifth century AD, in particular, of levee
crevassing, where high-energy events and their sediments break levees
deposited on the outside of a bank. Such an event could be classed as a
high-magnitude, low-frequency event, with instantaneous repercussions
for people in the immediate vicinity. In this instance, the moorings in the
Etruscan and then Roman harbour would have been at risk. However,
despite these problems, the harbour was reused each time (Benvenuti
et al. 2006). Whilst the Pisan example above gives us an insight into a
Roman response to a threat posed by flooding within a fluvial setting,
the response to alluvial risk across settlements is best answered through
the study of archaeological sites that have obviously been affected by
river systems.
Gordion
The Anatolian Iron Age city of Gordion, situated in the Sakarya Valley
to the south-west of Ankara, was covered with alluvial sediments dur-
ing its occupation. After its abandonment, the site was partly destroyed
and buried by these alluvial processes. Situated on a flat-topped mound,
there were precursors and successors to the Phrygian city, although it is
the Iron Age settlement that constitutes the most significant archaeo-
logical evidence (Marsh 1999: 164). Five phases of alluvial activity that
affected the site have been identified. The Iron Age city was built on a silt
layer, interpreted as a palaeosol, which at that time constituted the flood
plain. This may have been a rich agricultural soil, with the city expanding
onto an area that had been used for agriculture. After the early Phrygian
period, the river aggraded with 5 m of sediment deposited in some areas.
A dated stump from the base of this unit produced a 14C estimate of
the seventh to sixth centuries BC (ibid.: 167). Sedimentation continued
Environmental Knowledge in Dynamic Alluvial and Wetland Z
ones 101
until the present day; 5 m over c. 2,500 years is not substantial, and in
situations such as this, we have to ask if the sedimentation actually com-
prised rapid, catastrophic floods. If so, were there periods when this unit
was truncated, and parts of it redeposited? Direct evidence for a response
to sediment deposition takes the form of a stone ‘walkway’ built during
the Roman period on top of a 1-m thick deposit covering a Phrygian
structure, linking the site to the river. One of the most intriguing geoar-
chaeological elements on this site is the evidence for ‘an Iron Age earth-
moving project’. Here, 10 ha of artificially dumped sediment was moved
from the river’s edge or the river itself. Marsh interprets this as evidence
of a channel-widening project. This earth could also have helped raise
the next phase of construction above the flood plain (ibid.: 168). The
lowest parts of the site were abandoned when this phase of aggradation
was taking place. There is also evidence for floods that actually destroyed
structures along the river’s edge.
The examples above give some notion of the direct engagements that
people have had with Mediterranean alluvial and fluvial processes. The
subsequent sections deal with specific examples of how past environmen-
tal knowledge has developed vis-à-vis wetlands and associated alluvial
processes.
Environmental Knowledge in Dynamic Alluvial

and Wetland Zones
The Mediterranean comprises zones of plentiful or even excess water, as

well as the other extreme: parched, semi-arid landscapes. Mediterranean
societies have often found ways to manage and alleviate problems within
these zones. In those areas where water is present in excessive quantities,
such as wetlands, or where the supply of water is unreliable, settlement
activity has often waxed and waned. At times, certain societies have liter-
ally moved water from one place to another – Roman aqueducts being
the most prominent form of water-supply technology (Hodge 1995).
Water, in terms of its supply and movement, is perhaps the most muta-
ble characteristic of any landscape. Rainfall patterns vary, springs can be
unreliable, and rivers can flood, run still, or dry up totally. Moreover,
with time, the very course of a river, or the extent and nature of a wet-
land can change. Here, we will consider some of the themes broached in
the previous sections via assessments of environmental knowledge that

4.15. Mycenaean Bridge spanning a dried out streambed (photo: author).
have produced hybrids – technological and cultural mechanisms that

promote successful lifeways in these landscapes.
Occasionally in the Mediterranean, we come across ancient bridges
that are seemingly incongruous, as they do not span running water
(Fig. 4.15). There are three principal explanations: (1) the bridge is tra-
versing a gulley obstacle that has rarely seen running water; (2) water
flow is seasonal, largely occurring during the autumn through to spring;
(3) the watercourse has dried up. Consequently, we need to consider the
nature of the landscape at the time of construction. For example, today,
a remnant arch of Klidhi Bridge sits on the Thessaloniki Plain, Greece,
with water not having run under this bridge for some time (Ghilardi et al.
2010). The relationship of the ruins of the Roman bridge at Simitthus
(mid-Medjerda flood plain in northern Tunisia) with the ancient courses
of the river and the sediments deposited by these courses allows us to
consider how and when the river changed course (Zielhofer & Faust
2003: 212). The bridge appears to have been in a useable state between
AD 112 and the third century AD. As the Medjerda River meandered
ones 103
during this period, it damaged the bridge, which had to be repaired on

several occasions. Associated roads had to be re-routed as a consequence
of changes in the river. Therefore, whilst the river clearly changed course
during the period, the surrounding flood plain appears to have been sta-
ble up until the post-Roman period (ibid.).
The River Xerias (‘xeros’ denoting to ‘dry’ in Greek) on the Argos
plain is an excellent example of the relationship between the physical
(fluvial regime) characteristics of a watercourse and its etymology. The
Xerias is a torrential river, which is often dry but has produced brutal
and devastating floods in the past. Floods were of such amplitude that
a series of substantial flood defence walls were established during the
fifth to fourth centuries BC. The presence of a second-to-third-century
AD pottery kiln near the piles of one Roman bridge indicates that this
bridge must not have functioned while the kiln was in use. The pres-
ence of two first-century BC Hellenistic tombs just to the north of
another ‘fossilised’ Roman bridge indicates the presence of an ancient
arm of the Xerias (Fouache 1999: 176–7). Bridge redundancy is a func-
tion of not only a change in channel position, but also changes in cli-
mate that comprise a reduction in rainfall as well as changes to spring
outflows. Redundancy might only be seasonal in many Mediterranean
landscapes, with some watercourses traversable on foot during the
summer months, and some bridges serving during periods of water
flow, normally between the autumn and spring. The seasonal variation
of hydrologic processes is a fundamental agency in the development of
human–environment interactions and the cyclical nature of cultural and
economic activity.
There is substantial spatial and temporal variability in flooding (and
the consequences of floods), especially in areas such as south-east Spain
(Hooke 2006: 313). As with all environmental processes, temporal cycles
of varying length are crucial to the human experience of these. One fun-
damental problem that all palaeoenvironmental scientists and archaeolo-
gists must confront is that of teasing out weather events from climate
(average temperatures and rainfall). The intra-annual temporal spread
of rainfall will influence the probability of flooding and erosion events.
If, within two separate years, the same annual rainfall is measured, but
in one of those years, that rainfall is concentrated within a small number
of high-magnitude events, then these events will result in much higher
levels of flooding and erosion (Wittenberg et al. 2007).
We need to consider the notion that riverside sites were a corollary

of so-called Holocene climatic stability (Burroughs 2005: 245). Whilst
proxy records may indicate the possibility or even probability of relative
stability, we cannot be sure to what extent the weather was stable and pre-
dictable. The evidence for flooding on many sites needs to be considered
within the context of how hazards (and risks) were assessed in the past.
The greater cycles of alluvial time are comprised of shorter, annual cycles
which include seasonal, low-level floods associated with autumn storms,
then summer drying of seasonal watercourses and the spring melt waters
and rains refilling the rivers. However, we must ask to what extent were
people aware of high-magnitude event cycles and possessed a notion that
was in any way similar to the modern assessment of cyclical events as fall-
ing within a defined cycle (such as so-called 50- or 500-year events).
In central Spain, there were periods with ‘clusters’ of floods on the
Tagus River. The fact that floods tend to cluster is of profound impor-
tance for any attempt to understand cycles of human response. It is
apparent that there were periods when the probability of a flood occur-
ring was relatively high. Therefore, we might expect to be able to identify
phases of human responsiveness to the clustered flood regimes (Benito
et al. 2003). It is unclear as to whether flood frequency increases during
periods of a colder, wetter climate or during periods of climatic warm-
ing. There is no doubt that floods occur under both of these regimes.
One climatically driven scenario is founded on the correlation between
Atlantic-influenced and Mediterranean-influenced river systems across
Spain. Here, the Middle to Late Holocene periods, centred on 2820–
2440 BC and 865–350 BC, are characterised by increased flooding.
Despite the difference in the flood-producing weather conditions char-
acteristic of the Atlantic and Mediterranean hydro-climatic regions, the
radiocarbon date clusters do coincide over the last 3,000 years (Benito
et al. 2008: 75). The 2820–2440 BC phase is characterised as a ‘warm-
ing-dry period’ when flooding appears to have been common. As far as
past people were concerned, it may well have been sudden, high-magni-
tude events that were more of a threat. This type of event is not neces-
sarily identified within slackwater deposits (ibid.).
Human Engagements with Mediterranean Wetlands

Archaeological research in Mediterranean wetlands reveals the devel-
opment of certain forms of environmental knowledge and historical
ones 105
ecologies. The evidence from some wetlands allows us to assess responses

to environmental processes, thus showing that we cannot necessarily
identify a homogenous wetland cultural ecology for any period in the
past.
We have to assess the hierarchy or ordering of these ecologies, and
how ideological and political processes directly influence this layering.
This form of characterisation is referred to as ‘Panarchy’ by resilience
theorists, where Panarchy describes cross-scale interactions, wherein ‘the
resilience of a system at a particular focal scale will depend on the influ-
ences from states and dynamics at different spatial and temporal scales.
For example, external oppressive politics, invasions, market shifts, or
global climate…’ (Walker et al. 2004: 5). In addition, social elites may
have one view of how an environment operates and should be managed,
whilst those who actually live and work in that environment have differ-
ent, spatially specific forms of environmental knowledge. These notions
inform the assessments of wetland landscapes presented below.
Some of the most ancient evidence for human management of hydro-
logic processes comes from the Peloponnese and central Greece. These
areas comprise the full range of characteristics discussed so far in this
chapter: rivers, springs (on karst topography), wetlands, and semi-arid
areas. Possibly the oldest (functioning) piece of hydraulic engineering
in Europe is that associated with the Bronze Age city of Tiryns on the
Argive Plain. Here, a large dam was constructed in order to divert a
watercourse, and protect the city from flooding. This system also com-
prised an artificial canal which was about 1.5 km long (Zangger 2001:
127). A very different problem characterised the area around Copais in
central Greece. This riverless zone once had lakes and a wetland that cov-
ered an area of 150 km2. Here, the Myceneans constructed a drainage
system designed to reclaim land and perhaps reduce the risk of disease.
During the fourteenth and thirteenth centuries BC, they constructed
a canal on the northern edge of the plain. This canal was some 25 km
long, 40 m wide, and 2–3 m deep. It transported water from several dif-
ferent watercourses away from this zone (ibid.: 130). Looking at specific
areas around the lake, it seems that places, such as the Kephissos Valley,
were drained naturally or via human intervention (Farinetti 2009: 105),
leaving behind some areas of good-quality agricultural land. According
to Strabo, the waters of Lake Copais threatened the town of Copai. The
natural sinkholes around this, and other lakes, were often blocked by
sediment, thus resulting in the raising of water levels. The area around
the lake was occupied since at least the Neolithic, followed by a substan-
tial increase in activity during the Late Helladic period. Recent work
suggests successful and extensive drainage of the area centred on the
Mycenaean fortress of Gla. Two key elements constituted by the north
and south canals diverted the course of the river along the extreme north
and southern edges of the lake, the aim being to force these waters to run
into the katavothrai (sinkholes) (Iakovidis 2001: 155). The structures
associated with the management of the lake (Cyclopean-style dykes)
tend to be dated to the fourteenth and thirteenth centuries BC (Knauss
1985). The aim of the management scheme was not only to limit the
extent of flood waters, but also to create new agricultural land. A cat-
astrophic event might have taken place at the end of the Late Helladic
period. Some have argued that the lake may well have suddenly emptied
via the katavothres during the Hellenistic period, in particular at around
338 BC. This could have left behind a disease-ridden marsh (Châtelain
2007: 212). However, extensive research employing GIS analyses of site
distributions and reconstructions of the past lake extent do not really
support this (Farinetti 2009).
One Greek lake and wetland that has seen important transformations
over time is Stymphalos. Modern Stymphalos is situated in the Arcadian
Mountains. Best known as the site of Heracles’ sixth labour of killing the
Stymphalian birds, the remains of the mountain city of Stymphalos are
situated on the edge of the lake. However, the location of the first early
city (c. 700–375 BC) (Gourley & Williams 2005) is unknown. The lake
at Stymphalos is a dynamic and hybrid water system – enigmatic as both a
natural and cultural feature. The waxing and waning of the lake has been
controlled by a complex series of natural processes as well as technologi-
cal interventions, including management of the sinkhole on the south-
ern edge of the lake. The lake is largely fed by a series of springs along
its northern edge, the ancient town having been built around some of
these springs (Fig. 4.16). Other springs are situated along a line higher
up the south-facing slopes of the mountains that define the northern
edge of the Stymphalos area. The precise nature of the links between
the Stymphalos, Scotini, and Alea poljes has interested archaeologists
and geologists alike. The subterranean routes followed by these chan-
nels emerge at the Kephalovryso and Douka Vryssi springs and in the
Inachos River valley (Crouch 2004: 115; Morfis et al. 1985) (Fig. 4.17).
ones 107

4.16. The ancient city (top) and spring (bottom) at Lake Stymphalos (Peloponnese, Greece).
Whilst wetlands might harbour disease, the benefits provided by a guaranteed water-supply
are obvious. Moreover, the ritual/cultural importance of water and springs contributes to
settlement persistence in many areas such as this (photos: author).
The flow of water is considered intermittent. However, these processes

are poorly understood. Today’s wetland is an artefact of centuries of
human manipulation and management. Moreover, its modern character-
istics might lead us to believe that water has been incessantly bountiful,
(K) Doline
K43
(S)Spring
S7 Stymfalia Underground
Channels
Mountains
K5
Basins
Scotini
K6
BASINS
North
Alea
K8
S146
Argos
H I G H L A N D S Argive
S110
Plain
S112
Gulf of Argos
S111
S113

4.17. The karst-spring subterranean network of the north-west Peloponnese –
Underground connections between Tripoli Plain and Argolis coast springs. (By permission
of Oxford University Press, Crouch, D. P. (2004), Geology and Settlement: Greco-Roman
Patterns, fig. 4.2.)
accessible, and easy to manage, and that agriculture in this landscape has
always been a relatively straightforward endeavour.
Whereas pre-Roman management of the lake and springs probably
did take place, the first incontrovertible evidence for intervention in the
system dates to the second century AD, when the lake level was con-
trolled through the construction of an aqueduct that provided water to
the major city of Corinth 45 km to the north-east. There is no doubt that
large fluctuations in the water level occurred before this date, as coring
undertaken in the area presents a history of changes in lake level going
back through the Holocene (Brown & Walsh, forthcoming; Unkel et al.
2011).
ones 109
The hydrological importance of Stymphalos would have been trans-

formed during the Roman period when the Hadrianic aqueduct to sup-
ply water for Corinth was built. The manipulation of this plentiful water
supply (more specifically, the spring at Driza, just to the north of Lake
Stymphalos; Lolos 1997: 276) by Roman technology altered the very
nature and meaning of water at Stymphalos. Local people’s engage-
ment with the lake and surrounding springs and their associations with
sanctuaries and deities would have changed. The capture of this source
must have regulated inputs into the lake and certain components of the
hydrological system around Stymphalos. Such a structure not only cre-
ated a physical link between the source and consumers of the water (in
this instance Corinth), but it may well have also changed the nature of
cultural and ideological links between the source area and the consuming
city. This change in a community’s or society’s relationship with water
would have been true in any landscape where such a feat of hydraulic
engineering had been undertaken. In Greece alone, there were about 25
aqueducts, plus a dozen across the Greek islands (ibid.: 303–12).
Hydromythology
Perhaps more than any other cultural feature, the sanctuary articulates
a hybrid or intersection of economic, religious, and environmental pro-
cesses within a landscape. As Jost (1996: 217) argues for Arcadia, ‘…
certain places seem destined to be considered sacred, and certain types
of landscape attract cults of one divinity rather than another’. Springs
are of particular importance, and there is no denying the influence that
the distribution (across both time and space) of water has on all aspects
of life. In addition, the deities associated with flooding are another vital
component in many classical landscapes. The plains of eastern Arcadia
comprise a calcareous soil covered with clay and alluvium that does not
drain well, and it is the limestone fissures (the katavothra) that provide a
natural drainage system. Thus, these natural features are managed, and
were often associated with Artemis (a goddess associated with dampness)
as well as Poseidon. Artemis was celebrated as the goddess of the marsh-
land at Stymphalos (ibid.: 220).
There is no doubt that hydrological processes and their associated
landscapes were understood via forms of environmental knowledge that
were different to our own. However, one characteristic that is as true
in the past as it is today is the fact that environmental knowledge varies
depending on one’s role within society. For Herodotus, a marshland was

a veritable ecosystem, characterised by its flora and fauna. A marshland
could be seen as a contact zone between two opposite types of water
(saline and fresh), or a wetland could be the product of a river having
retreated. Herodotus also noted the ‘calmness of waters, where there is
an absence of activity; a sleeping body of water’ (Châtelain 2007: 30–1).
Human engagement with any environment is constructed via different
forms of environmental knowledge, and these vary not only across the
different layers of society but also across time. Moreover, environmental
knowledge of rivers and wetlands would have been partly constituted by
mythological characterisations of these features and associated characters
from the myths. There are some 30 important ‘mythological’ rivers in
Greece. ‘The mountains formed an important part of the mythological
landscape but they remained inanimate, whereas the rivers were charac-
ters in the myths, often their protagonists, as well as essential features of
the scenery’ (Brewster 1997: 2–3).
Knowledge of environmental potential is partly influenced by the cul-
tural and cosmological ideas associated with a given landscape type. In
the ancient world, these were mediated via mythical-religious processes
and manifestations (deities, shrines, temples, and stories) (Clendenon
2009; Retallack 2008). There are a number of mythological references
to Mediterranean wetlands that imply that these areas were considered
repulsive or problematic. The depictions of Heracles slaying the multi-
headed serpent of Hydra on Corinthian pottery dating to the period
c. 630–570 BC may well be a reference to attempts to drain Marshes
at Lerna in the Argolid (Grmek & Gourevitch 1998). Also, the slaying
of the birds at the lake at Stymphalos (Heracles’ sixth labour) can be
interpreted as a metaphor for the dangers associated with wetlands. The
possibility that Heracles was (or represents) an infamous hydrological/
agricultural engineer of sorts is another intriguing perspective (Châtelain
2007). There is little doubt that drainage was a problem at Stymphalos,
rendering use of the land difficult. One myth refers to the blocking of
the lake’s sinkhole – considered to have been the result of Artemis’s
actions. She had been angered by the Stymphalians who had stopped
worshipping her, and also because their use of the land had damaged the
preferred environment of the animals dear to Artemis (Burford 1993:
165). This intersection of mythological narratives – archaeological and
environmental proxy data – allows us to address how environmental
The Pontine Marshes: Roman ‘Relationships’ with a W
etland 111
knowledge may have varied across time and space. This in turn informs
our (pre)historical ecological assessment, as myths, albeit unverifiable or
even unlikely as actual historic events, do articulate understanding, and
therefore possible engagements between peoples and wetlands that have
resulted in their modern-day characteristics.
The Pontine Marshes: Roman ‘Relationships’ with a Wetland
One of the most famous Mediterranean wetlands is the Pontine Marshes,

an area that probably did witness a waxing and waning of human activity
for much of the Holocene. Roman interventions in this landscape have
been much discussed, and a brief overview is presented here.
The Mezzaluna core (from the north-eastern part of the Pontine
plain) gives us an indication of how this landscape developed prior to
the Roman period. The palaeoecological evidence points towards the
development of macchia-type vegetation from c. 4,500 years ago, with
alder carr and willow implying the presence of a relatively humid area
at the start of the first millennium BC. Other palynological evidence
implies the development of cereal growth from the Late Bronze Age
(Attema, Burgers, & van Leusen 2010: 35–6). The core from the Lago
di Fogliano suggests some agricultural activities around this lake from the
Early Bronze Age, and the gradual drying of the lake starting at the end
of the Bronze Age or the start of the Iron Age. The Roman period saw
the emergence of a drier agricultural zone, with some evidence for olive
cultivation on the Mont Lepini slopes (ibid.: 154). During the fifth and
fourth centuries BC, Roman colonies were established along the hilltops
of the Lepine scarp on the north-east edge of the Pontine. Roman colo-
nisation changed the Pontine region from a sparsely settled and unman-
aged environment into a landscape that saw dense settlement in some
areas (Attema & Delvigne 2000). The lower graben zone is only about 1
m above sea level, and constitutes the core zone of the Pontine Marshes.
Despite this inherent environmental disadvantage, the area was settled at
the end of the fourth century BC. Although the chronology for centuria-
tion is not entirely clear, it would probably have post-dated this fourth-
century settlement (de Haas 2011). Attema and Delvigne (2000) argue
that the alluvial plain was given to the Setia colonists to cultivate in the
early fourth century BC, a period when parts of the Pontine landscape
appear to have been organised, with the construction of the Via Appia
and possibly the digging of the Decennovium Canal. This zone could
have furnished agricultural plots for up to 46,000 colonists, suggesting
the development of intensive cultivation. During the early Republican
period, the Pontine area was identified as a potentially productive land-
scape, and settlement was encouraged. As with most ecologically het-
erogeneous landscapes, certain areas were recognised as being useful for
specific activities. The central part of this zone, where there is a dearth of
evidence for settlement activity, probably functioned as common grazing
(de Haas 2011: 167).
A dramatic decline in settlement on the wetland area took place at
some point during the Imperial period with 50% of sites having disap-
peared by the first century AD, earlier than the decline seen along the
adjacent coastal zone (Attema, de Haas, & Gol 2011). The contraction
of activity on the Pontine Marshes might in part have been the conse-
quence of wider changes in economic and concomitant settlement pat-
terns. The possibility that changes in tax regimes and the economic costs
associated with ‘marginal’ landscapes rendered certain areas economi-
cally unattractive is something that we should consider as part of the
assessment of interaction with any environment. There is, however, lit-
tle doubt that the Pontine area did witness environmental problems. It
has been argued that the attempt at draining the area by M. Cornelius
Cethegus in 160 BC was not successful in ridding the area of malaria
(or other diseases) (Sallares 2002: 185). After the widespread settle-
ment activity across the Pontine Marshes during the Republican period,
it appears that settlement was subsequently concentrated along the via
Appia and on the higher grounds. Drainage was clearly an unrelenting
problem, and the area was described as unhealthy by Vitruvius in the late
first century, and the term Pomptinae Paludes was used for the first time
in the mid-first century by Lucanus (de Haas 2011: 207). After the first
century AD, only the higher areas to the south-west were settled. This
appears to support the notion that deteriorating environmental condi-
tions were a serious influence on settlement activity. On one site in the
south-western area, there is evidence that soil brought in from elsewhere
was added before construction (ibid.: 10). Therefore, even on higher
zones, there may have been difficulties in draining the land.
One important management technique that may have been applied
by the Romans is that of colmatage (the controlled flow of water con-
taining sediment which would fill and render a wetland exploitable). It
The Pontine Marshes: Roman ‘Relationships’ with a W
etland 113
has been assumed that colmatage was understood by Roman engineers,

but there is no clear evidence of its employment on the Pontine wet-
lands (de Haas 2011: 214). Accepting that there were periods when the
Pontine Marshes, or parts of them, were successfully managed, settled,
and exploited, we should not forget that life in such an environment can
be rendered difficult by disease. Even if malaria is not present, the con-
tinual discomfort caused by insects, especially mosquitoes, is something
that we should not make light of.
Wetlands and Disease

There is little doubt that if there is one characteristic of Mediterranean
wetlands that constituted a foundation for the construction of environ-
mental knowledge, then it must be the potential for disease in these areas.
The integration of landscape ecology and epidemiology recognises that
human activity has often been associated with the development of dis-
eases (Balée 2006: 87). Societies develop responses to disease at the local
level as well as at the macro-political level. Sometimes these responses
actually help generate or propagate the disease.
Mosquito infestation and the possibility of malaria are of course
considered the most repellent characteristic of wetlands. A virulent form
of malaria may well have been present in Greece by the eighth century
BC (Sallares 2002: 22). Whilst Sallares (2002: 33) considers that the
three species of malaria that pose a real threat to humans could have
been brought into Europe by Neolithic farmers, there is no incontro-
vertible proof of this, and Packard (2007: 35) considers that its arrival
in southern Europe may not have occurred before the first millennium
BC. The paucity of direct evidence for malaria in ancient populations is
problematic, although David Soren (2003) has argued for the impact
of malaria on a population at the site of Lugnano in Teverina, Umbria.
There are a number of diseases, such as leptospirosis, that are difficult
to distinguish, and malaria per se might not have been the principal
threat.
Despite the problems differentiating between different diseases
common in wetlands, we should accept that any such illnesses would
have radically altered the nature of human lifeways in and perceptions
of these environments, with direct consequences for the development
of environmental knowledge in these landscapes. Pliny the Elder rec-
ommended that farms should not be located close to marshes nor rivers
(Plin. Nat. 18.7, Pliny the Elder 2009). Relatively minor changes in
climate or even in interannual weather patterns will have altered the
potential for malaria or other diseases such as leptospirosis, where the
frequency of infected bodies of water increases. The warm period dur-
ing the first centuries BC and AD (along with the evidence for flood-
ing in a number of Mediterranean wetland zones) does suggest that
conditions for disease vectors would have been ideal. Research on
contemporary risks as they relate to changes in climate demonstrates
how an increase in temperature along with presence of standing water
increases the season for mosquito activity and thereby the transmission
of the malaria parasite (Sainz-Elipe et al. 2010). Deforestation does of
course increase a landscape’s exposure to insolation, and thereby over-
all ground temperature will increase, thus enhancing the vectors for
these diseases. Moreover, deforestation can change hydrological sys-
tems, even increasing the presence of run-off and resultant zones of
standing water. Consequently, these changes in the environment caused
by deforestation may well have provided new niches for mosquitoes
and bacteria (YaSuouka & Levins 2007). The palynological evidence
for the Tiber delta implies that such processes were common in the
wider region, with specific evidence from Stagno di Maccarese suggest-
ing that the lake level here was lower during the Iron Age to Roman
period, and a marshland environment developed (Di Rita, Celant, &
Magri 2009).
Whereas archaeologists tend to identify cultural processes that relate
to settlement and economic trajectories in landscapes, and concomitant
management practices associated with these activities, we should also con-
sider how environmental knowledge relating to disease mitigation devel-
oped over time. For example, most adaptations to malaria appear to have
been biological and genetic (Sabbatani, Manfredi, & Fiorino 2010: 78).
Changes in diet, which could have included the increased consumption
of broad beans, may also have helped. In addition, increased consump-
tion of milk and milk-based products may have been beneficial. There
may even have been some understanding of the difference between the
two principal forms of malaria present during the Roman period. Celsus
(a physician who lived during the first half of the first century AD) may
even have diagnosed the different clinical courses of Plasmodium falci-
parum-related fever and the Plasmodium vivax-related disease, which is
less threatening (ibid.: 70).
Discussion: Responses to Hydrological Variability 115
Discussion: Responses to Hydrological Variability
The response of different societies to fluvial risk allows us to consider the

fundamental issue of instrumental versus intrinsic values in that society’s
attitude towards nature. Unsurprisingly, Roman urban planning and rural
engineering often managed fluvial risk through the location of buildings
away from hazardous areas, or the protection of these areas with banks
and dykes, or the management of wetlands via drainage programmes.
However, there were periods (especially towards the end of the Roman
period) when these management systems either failed or were not main-
tained. For example, the sewers in Arles (south of France) appear not to
have been cleared out from the third century AD onwards, a time when
the city fell into political decline (Bruneton et al. 2001). Recent excava-
tions in this city have revealed arches parallel with Rhône, built on the
bank of the river that had not been settled prior to Rome’s arrival in the
region (Isoardi 2010a, 2010b). These arches increase in size towards
the interior of the town. The river flowed through the arches from the
river towards the town, and the arches thus reduced the velocity and
strength of flood waters. There is a possibility that flood waters dam-
aged arches during one flood event. What is more interesting is the fact
that the different flood deposits do not seem to have been cleared away
after each event, thus gradually reducing the efficiency of the arch system
(Fig. 4.18). We have to ask why these deposits were not cleared away. In
some ways, this lack of maintenance and mitigation is uncharacteristic of
Roman engineers and managers.
Evidence from rural wetland sites also reveals histories of flooding
or even destruction. The remedial construction of flood defences is not
just a spontaneous response to an environmental event, but demon-
strates awareness that an event will/can repeat itself. For example, at
Le Carrelet (a vicus on the Camargue, south of France), there is evi-
dence for the placement of boulders along the riverbank which should
be seen as part of a response to the threat of flooding (Arnaud-Fassetta
& Landuré 2003). This may not have been enough to protect the site,
as the presence of silt layers attest to frequent flooding across the settle-
ment (Excoffon et al. 2004: 218). This site did, however, experience a
serious flood at some point during the second half of the first century
BC, and was then abandoned up until the first century AD. Such actions
represent forms of environmental prediction that ultimately contribute

4.18. Excavation of riverbank arches on the Rhône in the City of Arles (south of France) (excavation: D.
Isoardi). Note the build-up of sedimentary units comprising Roman archaeological material. The flooding
events are dated by archaeological material from the latter half of the first century AD onwards. N.B. This site
is now situated within the cellars of a riverside building in the modern centre of Arles (photo: author).
to socio-environmental resilience (Westley et al. 2002). Whilst we might

assume that the responses and application of environmental knowledge
in the urban milieu would have been quite different to those developed
in rural areas, there is evidence for flood defences being constructed in
the rural landscape, perhaps mirroring a ‘Roman’ way of doing things.
The transfer of urban environmental knowledge to rural areas, and vice
versa, is a characteristic of certain complex societies, and perhaps nowhere
more so than in the Roman world. However, such responses may well
be maladaptive, and there are certain environments where technological
solutions are doomed to failure.
The application of environmental knowledge to a landscape problem
is in part constituted by the assessment of risk, that is, the likelihood of
a project’s success. Such an assessment is not purely instrumental, but
also comprises intrinsic cultural, religious, and ideological notions. The
Discussion: Responses to Hydrological Variability 117
transformation of a wetland may render it more meaningful. This may

be true if we are merely concerned with the productive value of a place.
However, landscapes, especially wetlands, can be meaningful and be
characterised by a form of environmental knowledge that is constituted
by mythological and ideological notions (Baylis-Smith & Golson 2004).
Whilst the draining of an area or the construction of flood defences
clearly have an instrumentally rational aspect, the manner in which they
were conceived of and enacted would have undoubtedly included plead-
ings and offerings to the relevant deities. For example, one Roman rite
comprised pre-construction auguring that included the examination of
livestock livers for discolouration (i.e. liver fluke). If such discoloura-
tion was present, this indicated poor-quality wet pasture, and therefore
building should not take place (Polio 2005). As such, this constitutes
the combination of pertinent environmental knowledge and ritual with a
view to making a judgement vis-à-vis economic potential and quality of
life within a specific landscape type.
The investigation of fluvial and alluvial systems, and an understand-
ing of humanity’s engagement with those systems, has to consider the
complexity of tectonic processes, local geomorphology, climate change
(and weather cycles), and then the range of human uses and manage-
ment of these varied environments. Moreover, the cultural and mytho-
logical dimensions associated with water cannot be forgotten, as these
constitute part of the environmental knowledge that underpins human
understanding of and engagements with bodies of water, whether these
be the mysterious vanishing and re-emerging of springs or wetlands char-
acterised by their almost incongruous mix of bountiful resources and
dormant disease.
The mutability of alluvial and fluvial systems in terms of their changing
courses and extent is an ever-present characteristic that Mediterranean
peoples have always had to deal with, with such changes engendering
variations in human movement and mobility, activities, practices, and
cultic rites.
Although much alluvial geoarchaeology does present human activity
as a process that contributes to changes in alluvial systems, some have
developed specifically archaeological approaches, where archaeolog-
ical and geoarchaeological evidence are combined in an assessment of
human engagements with the landscape. Zangger’s (2001: ch. 6) study
of technological solutions to alluvial sedimentation problems on the
Argive plain contributes to the discussion of people/landscape dynamics.

Even though such research addresses human responses to environmental
change within a framework of means-ends rationale, where the success of
an economic system is the paramount aim, it does allow us to move away
from the traditional geoarchaeological analysis of cause and effect within
the sedimentary system. We should also consider how changes in river
regime would have affected other resources. Something that is rarely con-
sidered is the importance of freshwater fish. As Sternberg demonstrates,
this lack of interest in freshwater fish is also reflected in the classical texts.
It is only really from the fourth century onwards that freshwater fish
are regularly mentioned in economic records as commercialised food-
stuffs (Sternberg 2004: 185). Whilst fish were kept in artificial fishponds
(Higginbotham 1997), it is also interesting to consider how variations
in fluvial action, water flow, oxygenation, temperature, and so on have
a profound influence on the potential range of species found in a river.
The hydrological regime at the river mouth (salinity levels) also have a
profound influence on the species present. These changes in hydrological
regime and their influence on fish in the past is clearly another area of
study that could develop in the future.
As archaeologists, we must explicitly adopt a nested temporal and
spatial framework in the assessment of human interaction with alluvial
and fluvial environments. Long-term, landscape-wide alluvial histories
clearly have their place, but at the same time, we should strategically
adopt methodologies that allow the identification of alluvial units on
or abutting archaeological sites. This then facilitates the identification
of any possible human response to these environmental events and thus
enhances our ability to discuss the range of possible relationships between
society and alluvial processes through time.
5
Environmental Change: Degradation and Resilience
One abiding theme running through Mediterranean landscape research

is that of environmental degradation (Conacher & Sala 1998). This and
the following chapter consider a range of data from across the region
in an assessment of evidence for landscape change, degradation, and
resilience. This chapter presents some syntheses of vegetation and ero-
sional histories within a perspective that questions the notion of land-
scape deterioration, but at the same time considers the nature of human
engagements with the landscape (see Fig. 5.1 for sites referred to in this
chapter).
The examples and case studies take us across the Mediterranean in a
broadly east to west direction, with the final case study from Spain pre-
senting a more detailed analysis of human–environmental interactions in
that area.
Approaches and Research Questions
As we have seen already, a fundamental debate concerns the evidence

for climate versus human impact on Mediterranean landscapes. The
variability of climate change across space and time is the only constant;
phases of stability exist but are often interrupted or punctuated. Perhaps
most importantly, we cannot easily tease apart the relative importance of
anthropogenic impact upon the landscape versus that caused by climate
(Allen 2001: 39–40).
For a number of decades, palaeoenvironmental research has empha-
sised the identification of periods of environmental stability and instability
that take place within an underlying trend of environmental deteriora-
tion, with the term ‘environmental crisis’ liberally employed. Whilst
119
Sea ofAzov
Bay
of
Biscay 26
30
25
BlackSea
31
35 33 36 32
34 Gulf of
Lions
37 38
27 Adriatic
24 Sea 15
22 29
Balearic
9
Sea
Gulf of
23 Aegean Sea
8
Tyrrhenian Sea Taranto 14 21
4539 40
43 41 11 12
4244 Ionian Sea
17
10
Strait of Gilbraltar 28 19
18
Sea of Crete
20
1 53
Mediterranean Sea 2
6 7
4
Gulf of Sidra
5.1. Map of sites referred to in this chapter. 1: Middle Atlas Mountains, 2: Sea of Galilee, 3: Golan Heights, 4: Negev, 5: Hula Valley and at Tel Yosef in the Harod
Valley, 6: Southern Shefela/Tell es-Safi/Gathin in Israel, 7: Dead Sea, 8: The Troad, 9: Kumptepe/Troy, 10: Amuq plain, 11: Gravgaz/Çanakh Lake, 12: Çatalhöyük,
13: Franchthi Cave, 14: Larissa basin/Sesklo, 15: Philippi plain, 16: Argolid, 17: Phlious basin/Nemea, 18: Pylos/Palace of Nestor, 19: Loúsios gorge at Górtys
(about 40 km to the south-east of Olympia, on the Peloponnese), 20: Siteia mountains/Karphi/Kavousi, 21: Gobekli Tepe, 22: Laghi di Monticchio, 23: Lago
Almini Piccolo, Apulia, 24: Tavoliere plain, 25: Terramare, on the Po plain, 26: Lago Lucone, 27: Biferno Valley, 28: Troina, 29: Basilicata, 30: Palù di Livenza,
31: Languedoc, 32: Etoile, 33: Lattes/Marsillargues, 34: Balma Margineda, 35: Grotte Camprafaud, 36: Glanum, 37: Llobregat River, 38: Banyoles, 39: Segura
Mountains, 40: Alicante Province/Polop Alto Valley, 41: Rambla Guadalentín River/Campico De Labor and La Bastida/El Culantrillo, 42: Almería/Los Millares,
43: Alpujarra, 44: Gatas, 45: Castellon Alto.
Approaches and Research Questions 121
this approach no longer underpins modern ecological thinking, many

archaeologists and some palaeoenvironmental scientists have perpetu-
ated a discourse that emphasises notions of degradation and marginality.
Despite this, notions of ‘resilience’ and ‘persistence’ have come to the
fore in recent years (e.g. Redman 2005), and some archaeologists have
developed narratives that move beyond the archetypal scientific assess-
ment of human interaction with landscapes and their sediments (e.g.
Jusseret 2010).
The Phytological Context

Sclerophyllous vegetation (plants with tough, leathery leaves on ever-
green trees/plants) is dominant across the Mediterranean (Blondel et al.
2010: ch. 6). Whilst the distribution and characteristics of this vegeta-
tion today are a product of Holocene climate change and human impact
(partly a consequence of early farmers clearing the existing forest) (Pons
& Quézel 1998), sclerophyllous vegetation has existed in parts of the
Mediterranean throughout the Pleistocene. These plants are consid-
ered to have adapted to the long (arid) summers that characterise the
Mediterranean, their leaves possessing low surface-to-volume ratios, thus
conserving water (Blondel et al. 2010: ch. 8). Many of the sclerophyl-
lous species occupy thermo-Mediterranean zones, that is, areas that are
below 700 m and generally towards the Southern Mediterranean. These
areas are characterised by hot summers and clement winters. Vegetation
histories obviously vary across the region and are the product of changes
in climate and human activity, as well as underlying geological and topo-
graphical contexts. Figure 5.2 gives one example of such geographical
variation, comparing Spain and the south of France (de Beaulieu et al.
2005).
Vegetation degradation is not a linear process but is characterised
by fluctuations, with vegetation cover waxing and waning as part of
managed ecosystems that are occasionally pushed into disequilibrium
(Blumler 2007). In a similar vein, a focus on the identification of the
start of severe erosion is in some ways problematic, as many badlands
existed well before the development of agriculture in the Mediterranean
(Mather 2009: 18), though the contribution of people in the initiation
of erosion in places around the Mediterranean is undeniable, albeit his-
torically contingent.
122 Environmental Change: Degradation and Resilience
(a) (b)
SW NE
SPAIN SOUTH-FRANCE
Climate Vegetation Climate Vegetation
1000 1000
Aridification trend
Drier and warmer

Hi
Hi General increase of
3000 3000
sclerophyllous taxa Hi
(19) 3rd expansion of
Hi
Holocene
Mediterranean taxa
(Phyllirea, Quercus ilex) (24, 26, 30) Development of
5000 occurrences of 5000 Fagus and Quercus ilex
Hi Hi
(19) Chamaerops
} “Mid-Holocene “General climatic trend”

Increased Moister
W
optimum”
(2) Optimum of Hi
(according to Magny et al, 2002) Hi
7000 Abies, Quercus 7000 (24, 26, 30) Development of Abies
(3-6)
Cooler and moister

C “8200 yr. B.P. (18) onset of Olea (22, 26, 29) Expansion of
Hi ?
9000 (15,17, 19) Event” 9000 Quercus cf pubescens and other
Mediterranean taxa
2nd expansion of (including Quercus ilex)
}
Mediterranean taxa
11000 11000
Y.D. C (Phyllirea, Quercus ilex) (29) Occurrences of herbaceous
taxa related to Mediterranean
dry meadows (Thymus, Echinops,
(18) 1st expansion of Quercus Fabaceae)
LGI
13000 13000
ilex; occurrences of Olea
(24, 26, 27, 29) Occurrences of
deciduous Quercus and Quercus ilex
(18) expansion of
15000 15000
}
O.D. deciduous Quercus
Upper Pleniglacial
17000 17000
(18) occurrences
19000 of Arbutus,Quercus ilex, 19000
Buxus, Pistacia,
Olea, Phyllirea Proximity of
glacial refuges
Cal yr. B.P. Cal yr. B.P.
Quercus ilex W Warm conditions Hi Human impact

Period characterized by a drought
Abies C Cool conditions
Period characterized by wet conditions
Fagus

5.2. The glacial and Holocene development of vegetation in Spain and the south of France. The diagram
represents the variations in influence of climate and human activity upon the vegetation of these two regions.
(By permission of Taylor & Francis: Alpes méridonales françaises, Université Aix-Marseille III: de Beaulieu,
J. L., Y. Miras, V. Andrieu-Ponel, & F. Guiter (2005), Vegetation dynamics in north-western Mediterranean
regions: Instability of the Mediterranean bioclimate. Plant Biosystems – An International Journal Dealing with
all Aspects of Plant Biology, 139(2), 114–26. Figure 2. Instability of the Mediterranean bioclimate: overview
proposed for Spain and France (a) and Italy (b). p. 117.)
The Fall from Eden
The evolution of Mediterranean forests is characterised by continual

change, even without the presence of sustained and profound human
impact. For this reason, the first part of this chapter considers evidence
for both vegetation change and periods of geomorphic stability (soil
development or absence of evidence for erosion) and instability (phases
of erosion). This is followed by an assessment of how people and climate
have influenced these developments.
Landscape Change Around the Mediterranean

Landscape degradation normally comprises two key stages: vegetation
loss and then concomitant exposure of soil to potential erosion via the
impact of people, rainfall, and gravity. The importance of vegetation
cover in controlling erosion cannot be understated, and has been the

5.3. An eroded zone where vegetation loss has exposed soil and sediment to rainfall and resulted in colluviation
(the movement of a soil and sediment due to gravity and the movement of water across a land surface) (photo:
author).
subject of much research (Kosmas et al. 2002) (Fig. 5.3). Vegetation,

including stem density and leaf litter, protects the soil in a number ways:
it prevents rain splash and shade from vegetation reduces evapotrans-
piration, and it contributes to organic matter (Thornes 1987: 43). It
is important to note that the presence of high levels of tree pollen in a
pollen diagram necessarily represents extensive tree cover and thereby
protection for the underlying soil. Many Mediterranean landscapes will
comprise patches of shrubby trees within areas of relatively exposed soil.
Grassland or dense growth of weeds and shrubs offer more protection to
the soil than sclerophyllous woodland.
Grove and Rackham (2001: 8–9) argued that the contemporary view
of a degraded Mediterranean was in essence an eighteenth-century myth,
based on a romantic view of pristine ancient landscapes. Forbes (2000:
98–9) demonstrates how, in modern times, some northern and west-
ern observers have tended to criticise the purportedly ill-considered
landscape-management strategies developed by supposedly unthinking,
exploitative Mediterranean peasants. There have, in fact, been successful

management strategies during the recent historical period, and ‘…over-
exploitation and degradation of grazing resources have simply not been
an option’ (ibid.: 107). Soil erosion and related environmental stresses
can limit the economic potential of certain landscapes (or parts of them),
but any limitations are dependent on the level of technology that a soci-
ety can employ in the improvement of land or mitigation of environmen-
tal problems.
The first key stage in the development of what we consider today
as typical Mediterranean landscapes, and thus a precursor to extensive
erosion, was the development of shrublands. These are essentially scrub
formations and possess different names across the region, such as maquis
in France (and Israel), mattoral and tomillares in Spain, and macchia in
Italy (Di Castri 1981: 3). The emergence of such areas of ‘degraded’
vegetation is an important part of the story of Mediterranean vege-
tation, as is the nature of human impact on plants and, in particular,
how people have exploited these zones once the pre-existing woodland
was replaced by scrub formations or shrublands (Blondel et al. 2010:
122–3).
Shrublands are a testament to human use of the Mediterranean land-
scape more than any other element in the region, although some types
of high maquis might be considered as climax vegetation (Tomaselli
1981: 119). However, the fact that similar vegetation communities are
found in other Mediterranean climate areas (Chaparral in California
and Fynbos in South Africa) where there is less human impact suggests
that some of these communities represent deflected climax vegetation.
Sclerophyllous vegetation developed along the Mediterranean coast
about 6,000 years ago, but the presence of xeric vegetation is not nec-
essarily a direct result of human-induced degradation (Collins, Davis, &
Kaplan 2012). Moreover, the arrival of this vegetation is not a precursor
to irreversible degradation. In fact, there is a case for the strong influ-
ence of human-management strategies actually contributing to vegeta-
tion resilience – a relationship that built up during the Holocene. In
similar Mediterranean-type landscapes, such as California, this resilience
is not apparent, as strong human–ecological relations did not develop
(Naveh 2004).
The basic evolution of Holocene Mediterranean vegetation starts with
certain plants spreading from local refugia in response to warmer and
Type Potential
climax
P
Per-humid
Coniferous Deciduous Evergreen Evergreen
mountain forest forest sclerophyllous thermophilous
Pinus nigra Quercus pubescens forest forest
Cedrus Q. faginea, Q. cerns Quercus ilex Olea, Ceratonia,
Abies Ostrya Q. calliprinos, Pistacia lentiscus
(Fagus) Q. suber
Humid
Forest
1200
Sub-humid
800
Pre-forest
600 Semi-arid
Pinus halepensis, P. brutia
J. thurifera, J. excelsa
Tetraclinis, Argania
Pre-steppic forest
400
Juniperus phoenicea
Arid
Steppeland
100 Desert
Per-arid
-7 -3 0 3 7 m
Oro-Med. Mont.-Med. Supra-Med. Meso-Med. Thermo-Med. Vegetation
“etages”
5.4. The distribution of vegetation communities across different Mediterranean topographies (By per-
mission of J. Wiley, Quézel, P. (2005), Large-scale post-glacial distribution of vegetation structures in the
Mediterranean region, in Recent Dynamics of the Mediterranean Vegetation and Landscape, fig. 1.1.)
more humid conditions right from the start of the late glacial interstadial
(Bølling) about 14,000 years ago. Evidence from fossil records, DNA,
and ecological modelling suggest that all the southern peninsulas were
refuge areas for temperate trees during the last glacial period (López de
Heredia et al. 2007; Tzedakis 2009). From about 9,000 years ago, com-
munities that spread from the south-east of the Mediterranean across the
region then complemented these initial vegetation structures. These new
vegetation structures varied across the region and changed with altitude
within relatively small subregions.
We should not forget that the Mediterranean is bounded by moun-
tains and that vegetation characteristics change radically with altitude
(discussed in more detail in Chapter 8) (Fig. 5.4). Three main types
of Mediterranean forest can be identified: sclerophyllous, broadleaved,
and coniferous (largely present in mountainous areas) (see Table 5.1
for the chronology of some of these developments). In the Western
Mediterranean, these forests tend to comprise Quercus ilex (holm oak),
Table 5.1. Key climatic and vegetation phases from around the Mediterranean.
Broad chronological c. 6200 BC (‘Rapid 4500–3000 BC c. 2000 BC c. 1500 BC c. 0 AD

phase Climate Change Event’)
North Africa Middle Atlas Mountains Decline in Quercus, Relative aridity. Possible arid
dominated by Quercus increase in conditions
canariensis (Early Poaceae. 1200–700 BC
Holocene). Appearance of (Linstädter &
General aridification Cedrus. Evidence Zielhofer 2010).
of North Africa and for human
Middle East c. 6000 BC impact from c.
(Dusar et al. 2011). 3000 BC.
Libyan Sahara: wet Relative aridity
conditions 4900–4400 between 4500–
BC (Cremaschi & Di 4000 (parts of
Lernia 1999). Morocco).
Libyan Sahara
aridity from c.
3000 BC.
Eastern Med lake Early Holocene until c. c. 3200 BC: aridity. c. 2300 BC aridity. Relatively wet.
levels (after 5000 BC generally Phases of aridity
Roberts, Brayshaw, wetter. contemporary
et al. 2011; (Dusar with Chalcolithic
et al. 2011)) to Early Bronze
Age periods
(Roberts,
Eastwood, et al.
2011).
Levant PPN: relatively wet c. 4500 BC – Golan Relative aridity. 1500–1000 BC,
conditions (Goldberg Heights: some humid phase,
1994). evidence for followed by
human impact aridity.
on vegetation
(Schwab et al.
2004)
Soreq: 3500–2800
BC: wet–dry
oscillations.
Wider Mediterranean Rapid climate
change event.
Eastern Anatolia Lake Climate moister from Pistacia, deciduous Human impact on the Pinus dominant. Establishment
Van (1,648 m), about 6200 BC: oak Quercus, max. vegetation important of Juglans
Eski Acigöl forest expansion. Quercus, Corylus from about 1800 BC. cultivation.
c. 3300 BC. Onset of arid conditions
(increase in the oxygen
18 curve, plus an increase
in charcoal).
Central/western Aridity: abandonment From 4500 BC Decline in oak (poss. caused Drier phase Increased moisture
Anatolia/Greece of Catalhöyük-East, decline in mesic by people). indicated in parts and temperature
movement to western species. Fagus of Anatolia and in some areas.
site due to irregular increase. Greece (Roberts, Olea above
water supply. Lower lake levels. Eastwood, et al. 1,400 m in
Acer, Ulmus, Quercus all 2011). certain places.
present.
(continued)
Table 5.1 (continued)
Broad chronological c. 6200 BC (‘Rapid
phase Climate Change Event’) 4500–3000 BC c. 2000 BC c. 1500 BC c. 0 AD
Oak-terebinth juniper- In Greece, diversity Northern Turkey, certain “Sediment pulse” Economic species
grass parkland dominant of mountain indicator species such first half of first such as Olea and
in some areas. species was as Planatgo lanceolata, millennium Juglans appeared
reduced due to Rumex and Polygonum BC in settled from c. 500 BC.
a warming up of aviculare-type appear or areas such as
the climate. increase at c. 2200 BC. Berket basin
Peloponnese – Isotopes from Gölhisar and Gravgaz
between 5200 Gölü generally drier (Vermoere et al.
and 3600 BC, conditions than during 2000; Kaniewski
oak woodland the period prior to 3100 et al. 2007).
became denser. BC.
Peloponnese Overall reduction of forest
maquis in parts of Greece.
expansion and 2500–2000 BC Konya
economic species plain – reduction in
such as Cistus swampy conditions &
salvifolius and development of topsoil
Olea. (Jahns (Boyer et al. 2006).
1993). Southern Argolid erosion
c. 3200 BC rapid from c. 2500 BC (van
opening up Andel et al. 1986).
of forest in Greek soil erosion identified
northern Greece. at third millennium BC
5000–3200 BC in places (van Andel &
declining lake Zangger 1990).
levels.
Crete Cretan lowland was quite Ericacae expansion After the Santorini eruption
open with relatively few in the mountains (c. 1600 BC) the
trees. at about 4600 Kournas diagram shows
After c. 6500 BC, oak BC (poss. low Cerealia-type, plus
forest spread to the area burning). reduction in olive.
around the Delphinos
River, probably due to
an increase in rainfall.
Italy Expansion of deciduous Forest becoming Abies, Carpinus betulus Abies and Taxus
oak. more open. (hornbeam) and Taxus became extinct in
Quercus forest and Ulmus. Decreasing appearing. some areas.
Northern, coastal zones: summer Juglans and Castanea in Many areas saw
vegetation on this humidity from some places. the development
coastal area comprised c. 4000 BC Alnus expanded just prior of complex
of woodlands – Abies (Magny et al. to this period – indicative agricultural
was abundant, together 2011). of an increase in rainfall landscapes
with Pinus, deciduous Between 3500 and or increase in Laghi di with the usual
Quercus, Ulmus, Tilia, 1500 BC the Monticchio. economic trees
Alnus (Mariotti Lippi maximum levels Apulia – sudden decrease in such as olive and
et al. 2007). of arboreal with tree species, poss. drier walnut.
Sicily: aridification at the distributions climate. ‘Climate reversal’
around 5000 BC (Sadori of Ulmus and Increased variation in
& Narcisi 2001). Fagus. precipitation levels from
Cereals appear. c. 2000 BC (Magny et al.
2011).
(continued)
Southern France Anthracalogical work: Nice: Deciduous Nice: Pinus. Montpellier:

Pinus sylvestris, Quercus. Montpellier: deciduous ‘Climate
Juniperus sp., Prunus Decline in lake peak at 2200 BC with becomes
amygdalus and some levels at c. Quercus ilex increasing Mediterranean at
deciduous oak. 3800–2200 BC. between 2000 and 400 c. 300 BC’ (Jalut
5400 to 5200 BC evidence Anthracological: BC. et al. 2000).
for high levels of erosion decrease in Anthracological evidence
in some areas (Rhône juniper and for 5000–3000 BC
Valley) (Berger 2005). an increase indicates anthropogenic
in deciduous impact on the vegetation;
oak along with increase of Buxus
sclerophyllous sempervirens and
oak. sclerophyllous oak.
Cooler, wetter Erica important in
summers in some places, implying
some parts of the spread of meso-
the Northern Mediterranean
Mediterranean vegetation.
implied by Replacement of Quercus ilex
charcoal by Quercus suber.
evidence. Cooler, wetter summers
in some parts of the
Northern Mediterranean
implied by charcoal
evidence.
Iberia (high altitude) Early Holocene: Quercus Corylus and Evergreen species Grassland, heath
peak continues for Betula became dominate & juniper and land and
almost two millennia. important. scelerophylous species shrublands
In another diagram, later on. developed.
Quercus peak continued
until c. 5000 BC.
Northern Iberia Abies colonisation in Decrease in
places. Corylus and
Abies; maximum
of Erica at
Banyoles. Mid-
Holocene cooler
summers and
winters.
Coastal zone: Q.
suber, Fagus,
Alnus and Q.
ilex-coccifera.
Eastern Iberia From 5000 BC evergreen Thermophilous Expansion of Buxus due to In some areas 1000–500 BC:
Quercus. maquis. human activity. Corylus, Alnus, decrease in
C. 5400 BC important 4400 BC: Fraxinus deciduous
climatic change. minimum disappeared c. Quercus.
arboreal pollen 2000 BC.
= climatic
deterioration.
Then, increase in
xeric species, e.g.
Erica
(continued)
Southern-central Towards end of humid Aridification phase. Evergreen forest maximum. From c. 600 AD:
Spain phase. Pinus and Olea, Pistacia,
Pinus & Juniperus cf. evergreen Cistus, Juglans,
thurifera. Quercus. Prunus type,
Fourth millennium Vitis.
BC – oak, with
some hazel,
birch, ash and
alder.
Then, evergreen
oak (including
Quercus
rotundifolia.
Semi-arid zone of Decrease in pine, increase Pistacia, deciduous Steppe: aridification from
Almería in evergreen trees. oak. 2500 BC.
From 5000 BC, Drier summers in Southern
expansion of Mediterranean.
Olea.
Balearic Islands Buxus/corylus (no Disappearance
longer present) & of Buxus &
Juniperus/Quercus Corylus – maquis
dominant. dominated by
Olea.
Anthropogenic and Climatic Impact 133
Q. rotundifolia (subspecies, holm oak), and Q. suber (cork oak) whilst

in the east Q. calliprinos (Palestine oak) is common. Broadleaved forests
often comprise oaks, such as Q. pubescens (pubescent oak) and Q. cerris
(turkey oak), whilst true mountainous forests will include species such
as Cedrus (cedar), Abies sp. (fir), and Pinus nigra (European black pine)
(Quézel 2004: 8–9).
In the Western Mediterranean, maquis, garrigue, and steppe can
individually constitute extensive areas, whilst in the east, the situation
is often different, with patches of mixed maquis, phrygana, and steppe.
One question is whether these plant communities represent degraded
forest. Maquis is normally considered to be degraded forest, and phry-
gana degraded maquis and steppe (Grove & Rackham 2001: 57). These
phases of degradation were supposedly caused by retrodictable stepped
phases of overexploitation and/or changes in climate.
Anthropogenic and Climatic Impact: Views from Around the

Mediterranean
The Mediterranean climate has changed continuously over time: Jalut

et al. (2000) identified six dry phases during the Holocene. The arid
phases seem to be contemporary with positive 14C anomalies (where
changes in solar activity affect 14C production). The changes in species
composition and the ratios of deciduous to sclerophyllous trees allow us
to track the evolution and migration of the Mediterranean climate across
Spain up in to southern France. The six changes in climate inferred from
pollen sequences across the Western Mediterranean are dated to the fol-
lowing broad chronological phases: 8900–7700 BC, 6400–5600 BC,
3300–2200 BC, 2300–1400 BC, 850 BC–AD 270, and AD 700–1250.
The late fourth through the third millennia BC aridification is partly
supported by Magny, Miramont, and Sivan’s identification of a general
decline in lake levels at c. 3000–2000 BC. Concurrently, the Eastern
Mediterranean saw a mid-Holocene wet phase, followed by a transi-
tion during the period 3500–2500 BC to drier conditions (Magny et al.
2002: 49–53; Roberts, Eastwood, et al. 2011). Recent work has also
demonstrated north–south contrasts developing at c. 2500–2000 BC,
with wetter conditions above 40°S and drier conditions to the south
(Magny et al. 2012). The broader trend of a transition to a drier climate

5.5. Mosaic landforms and concomitant variations in vegetation characterise many Mediterranean landscapes –
Landscape around Sikyon (Peloponnese, Greece) (photo: author).
during the Middle Holocene led to the replacement of deciduous trees

by sclerophyllous evergreen trees (Magny et al. 2002: 54).
Palynology and anthracology (the identification of carbonised wood
macrofossils) permit the identification of the presence and absence of
different plant and tree species, and the relative intensity of human activ-
ity can be inferred from these data as well as from more conventional
archaeological evidence. Moreover, the definition of possible boundaries
and patches within a mosaic can be postulated. In particular, this can be
done where topographical variation (i.e. a break in slope) could plausi-
bly have influenced the potential for the growth of different species and
could also have influenced the nature of economic activity within the
landscape (Fig. 5.5).
The fact that some fundamental changes, in particular the emergence
of sclerophyllous vegetation, is seemingly synchronous across large parts
of the Mediterranean, at least at the subregional scale, implies that such
changes were caused by climate change (see Table 5.1 for an overview
of climatic and vegetation phases). The same can be argued for erosion
in that phases of erosion that appear to be contemporaneous across large
parts of the Mediterranean could be considered a product of climate
change. Localised events that do not fit with a pattern of regional syn-
chronicity can be considered the product of specifically local, possibly
anthropogenic, processes. However, if we accept that certain cultural
and economic changes took place relatively quickly across a number of
regions, and that the chronological resolution of many pollen diagrams
and sedimentary records is such that climate-controlled synchronicity is
not always easy to identify, then the role of people in certain landscape
changes could have been significant and ‘synchronous’. It is only once
we move into the Little Ice Age (beyond the chronological scope of this
volume) where we can identify well-dated and geographically dispersed,
correlated erosion events (often referred to as the ‘Younger Fill’; Vita-
Finzi 1969) that climatically induced erosion might be inferred (Grove
2001).
Southern Mediterranean
Some landscapes are characterised as degraded due to the consequences
of aridification (climate change) and resultant desertification – the pro-
cess whereby land degrades in arid zones due to the loss of vegetation,
whether this is caused by climate change and/or human activity. Areas
where this process is considered to be marked include southern Spain,
the Near East, and parts of Mediterranean Africa.
Despite the evidence for soil depletion and landscape degradation,
there have been periods when stable soils have developed. For exam-
ple, studies of Holocene soil development in Libya have contributed
important evidence for the existence of relatively humid conditions ideal
for pedogenesis (and vegetation growth) during the Early and Middle
Holocene (Cremaschi & Trombino 1998). However, much of the
Southern Mediterranean has been more susceptible to the consequences
of climate change than northern areas, since even subtle changes can
mean the difference between vegetation growth and the development
of desert. Under an arid, warm climate, these areas can develop pre-
desert or desert-like characteristics without passing through a degraded
landscape phase. Adequate precipitation, vegetation cover, and ground
moisture are crucial for soil development. During arid phases, soils do
not develop so readily or, if they do, they will probably have low nutrient
levels, and will be susceptible to depletion.
A pollen core from the Middle Atlas Mountains (1700 m) pro-
vides a rare example of a North African vegetation history. Here, the
Early Holocene vegetation was dominated by Q. canariensis (a decidu-
ous to semi-evergreen oak that would have been present on the moist
valley floor) with Q. rotundifolia (technically the subspecies of the
Mediterranean evergreen oak, Q. ilex, which is characteristic of the
Western Mediterranean) present on the drier slopes. From the Middle
Holocene (about 4500 BC), these oaks declined, with more open, grass-
dominated areas developing. Cedrus (cedar) then appeared for the first
time (Lamb, Damblon, & Maxted 1991: 404; Lamb & Van der Kaars
1995: 404). In some areas, there is evidence for human impact on veg-
etation from c. 3000 BC (Lamb et al. 1991), with a new expansion of
Q. canariensis coming slightly later.
Relative environmental stability during the Neolithic is also seen in
the Moroccan drylands (the south-west and north-east areas of the coun-
try). A phase of aridity at c. 4500–4000 BC interrupted an otherwise
relatively humid Neolithic (Zielhofer & Linstädter 2006). In Eastern
Morocco, the Late Neolithic saw the decline of Tetraclinis (Sictus tree)
forest, replaced to some extent by Stipa tenacissima (halfah or esparto
grass) – a sign of probable human activity (Wengler & Vernet 1992). In
the Ghardimaou Basin, three phases of soil development have been iden-
tified, amongst them a phase of Neolithic soil development that probably
occurred before 3500 BC (Zielhofer et al. 2002: 121).
During a period of increased precipitation, a fluvial system in an arid
or semi-arid landscape might be expected to provoke extreme erosion.
This does not appear to have been the case during the Roman period,
where management techniques seemingly mitigated erosion problems
(Faust et al. 2004: 1771). Forest clearance occurred during the fourth
century AD, with the impact of pastoralism becoming more marked
from this period onwards, ultimately creating an open scrub landscape
(Lamb et al. 1991). Towards the end of the Roman period, climate and
human impact enhanced geomorphic activity (ibid.). At the end of the
Roman period, alluvial sediments covered a Roman settlement zone that
had never been flooded before (Zielhofer et al. 2002: 122).
In the Libyan Sahara, wet conditions prevailed from 4900–4400 BC,
then ‘severe’ dry conditions developed from c. 3000 BC (Cremaschi &
Di Lernia 1999: 216). Unsurprisingly, the mountainous areas probably

received most of the rainfall, and the inner mountainous zones appear to
have been the preferred landscape for human habitation, with relatively
high levels of pastoral activity recorded. One Algerian study has yielded
evidence for soil development on the dunes at 4350 BC. This Neolithic
landscape was quite different to that of the present day, with the pos-
sibility of a wooded steppe covering this early soil, thus protecting the
land surface from aeolian erosion (Ballais 1994: 184). Despite the possi-
bility of a viable environment, Neolithic through to Roman archaeolog-
ical sites are rare, and transhumant pastoralism may well have been the
dominant activity. During the Roman period, limes (a system of roads,
towns, fortifications at the ‘edge’ of the Empire) was established. This
also included the development of irrigated agriculture. It appears that
‘The post-Roman, linear, incision of the wadis has, at the regional level,
been linked to the hydrological effects that resulted from the extension
of farming’ (Ballais 1994: 191).
The Near East

The analysis of pollen from a series of marine (sapropel) and land cores
from the central and Eastern Mediterranean supports the widely accepted
notion of a climatic optimum (with enhanced humidity) during the Early
Holocene. However, the suggestion that this environment provided
the ideal context for the transition to farming (Rossignol-Strick 1999:
528) should not be considered an explanation for the transition to farm-
ing, as such. Rather, environmental instability during the Younger Dryas
resulted in vegetation change that probably tested environmental knowl-
edge due to the unpredictability of certain resources. These processes
might have prompted the initial steps towards agriculture (Moore &
Hillman 1992). The later spread of farming outwards from the Near East
might also, in part, have been initiated by climatic instability (Weninger
et al. 2006; Weninger et al. 2009). There is little doubt that the period
centred on 6200 BC saw shifts in climate across the Mediterranean; for
example, a transition from wet winters and dry summers, to wet winters
and summers in parts of the Central Mediterranean (Peyron et al. 2011),
and a sudden increase in Quercus due to climatic deterioration in the
Aegean (Kotthoff et al. 2008; Pross et al. 2009). However, some have
argued that the 6200 BC ‘event’ should be seen as part of a broader
chronological trend of hemispheric cooling (Robinson et al. 2006).
In parts of the Near East, some settlements could have been situated
within areas of dense vegetation or woodland. Stands of wheat and other
domesticated plants would therefore have been limited in extent. During
this period, the vegetation mosaic would have been the inverse of today’s
relatively open landscape. Epipalaeolithic and Early Neolithic openings
within woodland would have constituted the minor patches. These
groups of people could have been more reliant on forest food such as
the acorn. However, it is clear that in the Southern Levant, Pre-Pottery
Neolithic groups involved in pre-domestication agriculture exploited
diverse ranges of plants, applying local environmental knowledge to local
niches (Asouti & Fuller 2012). Whatever scenario characterised these
food-procurement strategies, we should remember that in some regions,
many early farmers would have engaged with relatively dense vegetation
or wooded landscapes. Neolithic ‘impact’ on these milieus would have
taken time and would have been patchy (Olszewski 1993). The use of
fire to burn vegetation and create patches of exploitable land might have
been quite common (Turner et al. 2010).
Evidence from certain areas, including part of northern Israel, sug-
gests low levels of deforestation even by the Early Bronze Age (Rosen
2007: 132). The expansion of evergreen oak during the latter half of
the Holocene probably does attest to the increase in human impact
across many parts of the region. Deciduous oak forests were ‘destroyed’
over much of the Mediterranean zone of the Southern Levant (Baruch
1999: 22), whilst anthropogenic indicator species, such as Plantago
lagopus (hare’s-foot plantain) and Sarcopoterium spinosum (thorny bur-
net) expanded. An ‘intact’ forest comprising Quercus species was present
around the Sea of Galilee until the end of the third millennium BC.
Human impact then led to the demise of forest cover as olive (Olea)
production developed. As human activity declined after AD 550, ‘natu-
ral’ forest re-established itself (Baruch 1986: 45). In other parts of the
Southern Levant, the Early Bronze Age saw the development of olive,
vine, and cereal production, activities that extended cultural and man-
aged landscapes (Rosen 2007: 138). Whilst the gradual evolution of
managed agricultural landscapes characterised some areas, there are oth-
ers where there is clear evidence for fluctuations in activity, and a con-
comitant waxing and waning of forest. For example, at Birkat Ram in the
Golan Heights, a core covering the last 6,500 years suggests that human
impact occurred at c. 4500 BC, when there was a decrease in deciduous
oak (Q. ithaburensis type, Tabor oak) as well as heliophilous species, with
Olea cultivation developing during the Chalcolithic. This is supported
by local archaeological evidence for olive use. During the Middle/Late
Bronze Age, and through the Iron Age, there was regeneration of decid-
uous oak forest, probably a result of a decline in settlement activity in the
region. An increase in settlement and lead mining during the Hellenistic
period unsurprisingly led to a new reduction in forest cover (Schwab
et al. 2004: 1730). Running up to the Roman and Byzantine periods,
a relatively open environment, with areas of cultivated (arboreal) trees/
shrubs comprising Olea, Vitis (vine), and Juglans (walnut) developed.
The Evidence for Soil Erosion

With changes in vegetation, phases of soil erosion are to be expected.
However, the earliest erosion, contemporary with the Pre-Pottery
Neolithic, correlates with a period of relatively wet conditions (Goldberg
1994), and there is little to suggest that this erosion was caused by farming
practices. In the Negev region, there is evidence for erosion after the Pre-
Pottery Neolithic B period (Goldberg & Bar-Yosef 1990: 73). Despite
this, there is regional evidence for a phase of stability and soil develop-
ment between 8000 and 4000 BC, a period when lakes in North Africa
and Arabia expanded. This relatively humid period also saw the develop-
ment of the so-called Hamra soils on the coastal zone of Israel, dated to
between 8000 and 5000 BC (Gvirtzman & Wieder 2001), whilst in the
Hula Valley and at Tel Yosef in the Harod Valley, soil horizons within
wetland deposits have been dated to c. 6600–6250 BC. These units were
overlain by Pottery Neolithic colluvial sediments at c. 5800 BC (Rosen
2007: 77–8). Erosion did take place here during the Chalcolithic period,
while in the Southern Shefela area, a Middle Bronze Age to Late Bronze
Age phase of erosion has been interpreted as the consequence of a drier
climate and consequently a lower water table (Goldberg & Bar-Yosef
1990: 76). A subfossil of a Tamarix (tamarisk) tree near the Dead Sea
offers a tantalisingly well-dated glimpse of an arid phase that took place
during the Intermediate Bronze Age, the tree having died in 1930 BC
(Frumkin 2009). However, despite (overall) drier conditions during the
Bronze Age through to the end of the Roman period, it is difficult to
argue for a direct relationship between climate change and societal stress.
The most effective approach to assessing human–landscape engagements
is via local, specific studies (Finné et al. 2011).
One example of a site-based assessment is the geoarchaeological work

at Tell es-Safi/Gathin in Israel. Here, we see how erosion processes
are often local and caused by specific forms of human activity. Here, a
2 km long, 8 m wide, 5–6 m deep ancient, human-dug trench was stud-
ied. Two phases of filling were identified, separated by a so-called phase
of landscape stability. The first filling is dated to the Iron Age IIA (late
nineth century BC) whilst the second is post-Byzantine (Ackermann,
Bruins, & Maeir 2005: 323). Most importantly, the erosion phase (as
represented by the ditch fills) does not appear to correlate with climatic
events such as increased rainfall but rather with phases of human activity.
On the other hand, phases of aridity (rather than wetness) and conse-
quent reduced vegetation may have provided the conditions for erosion
(ibid.). Whatever the causes, there is no reason to suggest that such ero-
sion had an impact on human activity at the site.
In certain situations, we see how human activity can contribute to ero-
sion, whilst in others, specific farming practices can conserve soil. Such a
process has been demonstrated in the Negev Highlands, where a valley
with no evidence of human impact had clearly suffered erosion through
the Holocene, whilst in another valley, where farming has a long history,
including the Roman and Islamic periods, it was revealed that run-off-
harvesting agriculture had prevented desertification (Avni et al. 2006).
It is apparent that the evidence for land degradation, or more specifi-
cally, erosion, in the Levant is patchy. Whilst we have seen some areas
where Late Neolithic and Bronze Age groups might have contributed
to erosional processes, there are other areas, such as the Jebel al-Aqra
region of the Northern Levant where erosion does not seem to have
started before the second century AD (Casana 2008).
Anatolia and Greece
Some Anatolian Trends

As Early Neolithic (c. 6500 BC) peoples moved out from the Near East,
they would have continued to apply the environmental knowledge asso-
ciated with the exploitation of forest resources and the development
of small arable zones either within or on the edges of such woodlands.
Logically, as Neolithic populations expanded and grew in size, vegetation
manipulation intensified. Research in the Troad area (Western Anatolia)
shows how the composition of woodland did change, with deciduous
oak replaced by evergreen oak from about 3300 BC. In some places, in
particular the site of Kumptepe, a break in settlement (c. 4600–3500
BC) saw the development of a soil horizon and open vegetation, with
woodland then recovering, thus demonstrating environmental resilience.
This phase of woodland recovery probably took place quite quickly. In
many locations, Mediterranean landscapes do not take longer than a gen-
eration to recover. Once the new phase of activity was underway, from
3500 BC, maquis and shrub vegetation developed. Whilst, to the east, at
Troy, intensive human impact on the vegetation was not apparent before
c. 1300 BC (Riehl & Marinova 2008).
During the third millennium BC (the Early Bronze Age) in north-
ern Syria and southern Turkey, settlement nucleation and population
growth occurred as the environment appears to have become drier, with
concomitant reductions in vegetation cover and soil erosion. The fluvial
regime on the Amuq Plain (southern Turkey) was probably stable during
this period, with a change towards an unstable regime taking place dur-
ing the late first millennium BC (Wilkinson 2005: 182). The manner in
which societies ‘manage’ the stresses caused by aridity is obviously a key
area of discussion. For example, between c. 3500 and 2800 BC there is
evidence from Soreq for three wet–dry oscillations that lasted between
one-and-a-half and two-and-a-half centuries each (Bar-Matthews &
Ayalon 2011). The Bronze Age in parts of the Eastern Mediterranean
saw the development of complex, urbanised societies that had to engage
with these phases of climatic instability (Roberts, Eastwood, et al. 2011).
The evidence for deforestation and erosion attest to some of the envir
onmental problems. The early second millennium BC saw an increase in
humidity, followed by an arid phase. This latter period corresponds with
a dramatic change in the archaeological record where we see evidence for
the collapse of the existing cultural structures. In situations such as this,
we have to ask if the emergent social structures facilitated mitigation of
these problems or whether they were in fact the root of new rigid systems
of environmental management that suffocated local forms of environ-
mental knowledge.
A number of palynological projects in south-west Turkey have iden-
tified what is commonly referred to as the Beysehir Occupation Phase,
named after the place where human impact on the environment was first
identified (Bottema, Woldring, & Aytug 1986; van Zeist, Woldring,
& Stapert 1975). The Gravgaz study, some 100 km north of the
Mediterranean coastline, investigated two pollen cores: one at 1,215 m,

and the other from the Çanakh Lake at 1,030 m. In this area, defores-
tation started between 800 and 510 BC and the arboriculture phase,
in which Olea europaea (olive), Juglans regia (walnut), Fraxinus ornus
(manna ash), Castanea sativa (sweet chestnut), and Vitis vinifera (vine)
were grown, between 400 and 210 BC (Vermoere et al. 2002: 581).
Once again, this is an area that did not witness intensive impact on the
vegetation until quite late.
In some parts of the Eastern Mediterranean, the day-to-day experi-
ences of many people would have involved movement between relatively
small zones of patches of different types of woodland and open spaces
(Forman 1995). For example, some of the work at Çatalhöyük demon-
strates how the landscape around that site might have been organised
and exploited in response to the very particular environmental context
within which it sat (Shillito 2011). The archaeobotanical evidence clearly
supports the image of heterogeneous patterns of vegetation (Asouti
& Hather 2001). Quercus and juniper would probably have been col-
lected in ‘park woodland’ areas, at least 1–12 km from the site, whilst
the Salicaceae (willows and poplars) and Ulmus spp. (elms) would have
come from the nearby alluvial zones (Fairbairn et al. 2002).
The image of a complex mosaic of vegetation is also supported further
south at Pınarbaşı, south-central Anatolia. The study of charcoal here
suggests that the most common taxa were Pistacia and Amygdalus, with
Celtis also common in most assemblages, whilst Quercus and Juniperus
were rare in all deposits. This implies a local woodland-steppe environ-
ment with widely spaced trees of species that were resistant to drought.
Other species are indicative of marshes and riparian woodland (Asouti
2005). Lakeside species were not chosen, implying that people were
making choices: almond and terebinth wood burn well, and almond pro-
duces a pleasant smell (Asouti 2003b: 1200). A mosaic landscape, with
people making niche-specific choices regarding environmental exploi-
tation and management, was probably typical of many sites across the
Mediterranean.
Variability Across Greek Neolithic, Bronze Age,

and Classical Landscapes
Bellwood (2005: 73–4) argues for a precocious Neolithic impact on
the Greek landscape that may have acted as motivation for population
dispersal. There is a possibility that some Neolithic peoples did discover,

after a relatively short period, that they needed to move, as their environ-
mental knowledge was not adapted to a specific locale, and agricultural
production became unviable at quite an early stage. However, there is
little evidence to support widespread Neolithic degradation. It is once
we move into the Bronze Age that the evidence implies more extensive
manipulation of the landscape.
Despite a great deal of archaeological evidence for early agriculture
in Greece, few data allow us to suggest that the first farmers had a pro-
found impact on the vegetation (van Andel et al. 1995: 131). However,
our knowledge of Greek vegetation is often founded on pollen diagrams
from relatively high altitudes, and these demonstrate that during the ear-
lier part of the Holocene, land at around 400–800 m on the Greek main-
land was covered with an open deciduous oak forest (Bottema 1994b:
53), with more diverse vegetation developing with the appearance of
Pistacia from about 7000 BC (Ntinou 2002). It is possible that there
was an early, if limited, impact of early farmers from about 6000 BC
(Bottema 1994b: 55).
Thessalian vegetation remained relatively stable until c. 4500 BC,
with the spread of the hop hornbeam (Ostrya). From 2000 BC, beech
expanded in some of the mountain zones. At the key site of Franchthi
Cave, it is difficult to discern any obvious Neolithic impact on the open
deciduous oak woodland (Bottema 1994b: 59). Results from many paly-
nological sites in Greece suggest that there was a reduction in the diver-
sity of tree taxa between 4000 and 2000 BC. Such a reduction may have
been a result of ‘selective felling’ or non-regeneration after clearance by
early farming communities (Willis 1992: 152). Human impact on vege-
tation was apparent in many areas by the third millennium BC, although
some zones, such as northern Greece, saw the survival of relatively dense
vegetation. Nevertheless, there was an early phase of Neolithic erosion
recorded at the Thessalian site of Sesklo. Such a phase of erosion is per-
haps unsurprising adjacent to a major site. Whether the site’s inhabitants
caused or responded to the erosion is another question.
On the Philippi Plain area of Macedonia, where there are a number of
Tell sites and, one might assume, an area where there would have been
relatively high levels of human activity, there is little evidence for erosion
(Neboit-Guilhot & Lespez 2006: 336). Important changes in the veg-
etation started at c. 3000 BC, and from c. 2000 BC, human impact in
some parts of Greece led to an overall reduction of forest cover and the
emergence of herbaceous vegetation. However, in parts of Greece (in
this instance, the north-west), there is evidence for a small re-expansion
in woodland species between 1500 and 1000 BC, possibly a consequence
of population decline (Willis 1992). We once again see how resilient veg-
etation does recover during periods of population decline, as was proba-
bly the case during this period.
As with other areas of the Mediterranean, the Greek story of Middle
Holocene environmental change is one of emerging mosaic vegetation,
with many low-lying areas of woodland gradually opened up by farmers,
with climate, especially aridification, influencing the species composition
in this mosaic. There is no doubt that some areas of the landscape would
have comprised zones of exposed sediment and soil, these landscapes
being especially susceptible to erosion at the end of long hot summers.
We cannot demonstrate that soil loss or sedimentary erosion were cata-
strophic across large temporal and spatial scales. Studies of contemporary
soil–vegetation relationships demonstrate how abandoned cultivated
soils soon recover, regaining organic content and becoming stabilised
within 10 years (Lopez-Bermudez et al. 1998; Martinez-Fernandez,
Lopez-Bermudez, & Romero-Diaz 1995). Many areas are in fact ‘meta-
stable’, where landscapes are characterised by a series of cycles which
might comprise a phase of degradation (loss of vegetation and erosion),
followed by stability (woodland regeneration, geomorphological stabil-
ity) (Fig. 5.6). However, as Roberts (1990: 64) argues for south and
south-west Turkey, whilst vegetation may regenerate, soil losses are rarely
reversed to the same extent.
In the Argolid, there is little evidence to suggest that erosion coin-
cided with the development of farming. In fact, erosion existed prior to
the arrival of the first farmers (Fuchs, Lang, & Wagner 2004: 335) and
in many ways is a natural and ever-present phenomenon. However,
we should not understate the impact of people, nor the fundamental
changes that have taken place in landscapes since the Middle Holocene
− changes that have no comparators in previous interglacials. The first
clearly identifiable phase occurred during the third millennium BC (van
Andel, Runnels, & Pope 1986; van Andel & Zangger 1990: 382) which
could well be linked with a phase of hemispheric aridity at c. 2200 BC.
Palaeoenvironmental evidence from the Eastern Mediterranean clearly
suggests drought and probable physical stress experienced by certain

5.6. The same spot in the Vallée des Baux (south of France) eight years apart. This dem-
onstrates how vegetation can re-establish itself (bottom image) on a poor-quality soil in
less than 10 years (photo: author).
populations at the end of the third millennium and start of the second
millennium BC (Wossink 2009: 25–6). This contrasts with the history
of erosion in the Larissa basin (Thessaly, north-eastern Greece) which is
slightly more complex, with the first erosive phases occurring during the
Early Neolithic (van Andel & Zangger 1990). Then, even during the
period of increased Mycenaean activity, there is little evidence for wide-
spread soil erosion. Studies of alluvial sediments imply that soil conserva-
tion schemes may have been in place by this time (see the discussion of
terracing in the following chapter). Towards the end of the first millen-
nium BC, a rural ‘depression’ is correlated with an increase in soil erosion
(ibid.: 383). During such periods, farmers may have concentrated efforts
on the best-quality soils, thus abandoning any conservation scheme on
those soils that were in fact susceptible to erosion.
Around Nemea, there is evidence for at least three phases of post-
Early Neolithic erosion on the hill slopes (Wright et al. 1990: 587). In
the adjacent Phlious basin, OSL dating of sediments demonstrates a
phase of erosion contemporary with the onset of farming in the north-
east Peloponnese during the seventh millennium BC (Fuchs et al. 2004).
The first phase of erosion in the Nemea area probably occurred during
the Middle to Later Neolithic. The presence of early Neolithic material
in the sediment and then an Early Bronze Age site on the surface of
this sediment act respectively as termini post quem and ante quem. The
fact that the valley seems to have been abandoned during the Middle
Neolithic (Wright et al. 1990: 640) implies that Early Neolithic activity
may well have had some impact on the geomorphic system. A renewed
phase of human activity from the end of the Middle Bronze Age does not
appear to have had an impact.
At Pylos (south-west Peloponnese, an area that includes the Palace
of Nestor, some 7 km to the north of the Bay of Navarino), the pres-
ence of Late Helladic IIIA and IIIB (c. 1400–1200 BC) pottery within
a debris flow deposit is clear evidence of an erosion phase that must just
postdate this period (Zangger et al. 1997: 566). Erosion did increase
from the Late Bronze Age onwards, with the highest rates of sedimen-
tation occurring during the Roman and Early Byzantine periods. The
absence of evidence for extensive erosion associated with Neolithic and
Early Bronze activity, despite a high settlement density, may well reflect
settlement and agricultural exploitation patterns, with little activity hav-
ing taken place on the slopes of the foothills within the catchment of the
streams studied in this research (Lespez 2003).
All of this work demonstrates that the erosion histories appear to fol-
low a broad pattern, in that Bronze Age landscapes appear to have wit-
nessed phases of erosion. However, within this, there is clear variation.
The availability of precisely dated units employing OSL dating bodes
well for the development of geoarchaeology, although it is always
essential that the specific location and process being dated is critically
assessed. For example, Fuchs et al. (2004) sampled from a foot slope in
the north-east Peloponnese, a context where sedimentation was almost

guaranteed. We need to consider how representative such a location is of
the human landscape. If we choose situations where we know sedimen-
tation occurred, and moreover in a zone where ‘No settlement remnants
are directly located’ (ibid.: 337), we have to ask to what extent such a
study is relevant to the investigation of human experiences of erosion.
Sedimentary histories from different cores taken close to one another
can vary (as seen in Sagalassos area; Dusar et al. 2012). This does not
negate the geoarchaeological interest of such research, especially because
people engage with the wider landscape not just the space constituted
by the archaeological site. However, archaeologists must consider the
nature of sampling situations and the relevance of these to the study of
human lifeways.
If we accept that erosion events were not an everyday occurrence
but that direct engagement with stable soil surfaces was the normal
experience, we should consider phases of stability as represented in the
archaeological record by evidence for soil development. The mapping of
modern soils can provide information on land/soil quality (Davidson &
Theocharopoulos 1992), although we should avoid transposing mod-
ern models of soil quality and erosion patterns onto interpretations of
past landscape use and landscape change. One interesting and original
approach to the investigation of the impact of people on the Cretan land-
scape is the pedological research carried out by Morris (2002). Studies
took place across a number of areas, including the Siteia mountain range
in the north-eastern corner of Crete. The research also comprised the
analysis of a sediment catchment basin near three Late Minoan IIIc sites
and on a Late Minoan IIIc to sub-Minoan site in east-central Crete, and
also a final Neolithic to Late Minoan site on eastern Crete. The principal
aim of the study was to assess the level of human impact on the soil sys-
tem (Morris 2002: xvii). The studies of these (and all) soils are informed
by models of pedogenesis, where depth and characteristics are dependent
on climate, parent material, vegetation relief, and time (Jenny 1994).
A phase of Minoan soil stability was inferred from an argillic (clayey)
horizon that contained Minoan artefacts. A pedon from an agricultural
terrace at the Karphi site revealed the surface of a residual palaeosol also
containing Minoan artefacts, thus demonstrating that the terraces were
of Minoan origin (Morris 2002: 37–8), a date confirmed by a radiocar-
bon estimation. This evidence implies a change in climate during the late
Bronze Age, ‘…with moister conditions prevalent prior to the Minoan

occupation around 3000 years ago, followed by drier conditions after
the Karphi settlement was abandoned’ (ibid.: 43). Terraces then col-
lapsed and were buried by erosion as a result of their abandonment and/
or the development of a drier climate. The relatively poor quality of this
soil may also imply that terraced hill slopes did not necessarily support
good quality soils, and were brought into the agricultural system once
all of the prime land had been exploited. At Kavousi, there was a phase
of aggradation prior to 1000 BC, with a soil then developing on top of
this deposit. It is possible that the erosion phase was triggered by desta-
bilisation upslope, perhaps as the result of Late Minoan IIIC activity.
The Minoan soil was also buried (possibly during the Geometric period)
(ibid.: 75). Consequently, this and other research from Greece presents
us with an image of a landscape characterised by a waxing and waning of
erosion and soil stability – products of a complex, undulating topography
with people adapting to this mosaic. There is no doubt that there were
moments when soils and associated crops were lost, but the identifica-
tion of specific events is problematic, and we should perhaps balance our
quest for the identification of erosion with assessments of soil develop-
ment, and even the anthropogenic enhancement of soils, as suggested for
the Epipalaeolithic (possibly Pre-Pottery Neolithic) site Göbekli Tepe,
south-eastern Turkey (Pustovoytov 2006).
The Development of the Anthropic-Climatic Regime

The downturns in settlement and economic activity during the so-called
Bronze Age Dark Ages are recorded across much of the Mediterranean
(see the following for a range of perspectives: Bachhuber & Roberts
2008; Drews 1993; Mathers & Stoddart 1994) and are considered in a
number of places in this volume.
Changes in climate might have tested Late Bronze Age palace econo-
mies, although the peak of the arid phase occurred well into the Dark
Ages period (L. D. Brandon 2012) when we see the development of
new and burgeoning urban-centred economies in many parts of the
Mediterranean (Osborne & Cunliffe 2005).
The relationship between changes in the environment and their
potential impact upon settlement, economy, political structures, and
wider culture requires nuanced approaches that comprise assessments of
a wide range of processes that must ultimately consider the ways in which
different layers within hierarchical societies develop and modify environ-

mental knowledge. Analyses of environmental, political, or sociocultural
resilience provide a framework for such analyses (Butzer 2012). Some
would argue that the classical period saw the emergence of an anthropic-
climatic regime (as opposed to an earlier climatic-anthropic regime)
where a certain threshold is crossed, and human activity becomes the
dominant factor in causing erosion (Neboit-Guilhot & Lespez 2006:
344).
The Greek countryside was most densely settled during the classical
and Early Hellenistic periods (c. 350–250 BC). Whilst some areas saw an
appreciable reduction in activity during the Hellenistic (c. 323–31 BC)
and Early Roman periods (c. 31 BC–AD 300) (Alcock 2007: 135–6),
other regions witnessed an increase in activity, with a wave of popula-
tion growth around certain ‘peripheral’ areas of Greece (such as parts of
Macedonia) during the Late Roman period (Bintliff 1997). There is no
doubt that erosion during the classical period was common, although
we need to consider the specific relationships between our sedimentary
facies and the archaeological sites, and the responses that erosion would
have engendered. For example, in the Loúsios gorge at Górtys (about 40
km to the south-east of Olympia, on the Peloponnese), the fourth cen-
tury BC baths had been covered by several metres of sediment. Despite
the fact that there is no doubt that substantial erosion has taken place
here (sediments had also built up against a Byzantine chapel, indicating
a later phase of erosion), the deposition event or events are not precisely
dated, and we do not know if erosion actually affected the site whilst it
was in use. It is safe to assume that erosion would have been an issue in
this steep-sided valley (Fig. 5.7), but it was managed (Dufaure 1976).
Here, we should assess the importance of the specific geological charac-
teristics of this area. The eroded material at Górtys is derived from flysch
(Fouache 1999: 162). Such material is easily erodible, and we have to
consider how an establishment such as a baths would have employed, or
forced people to keep the environs clear of, eroded material – an issue
that is dealt with in more detail in the final part of this chapter.
People and Environment in Italian Landscapes

As we move to the Northern Mediterranean, we come to areas that tend
to suffer less from long periods of aridity, although the archetypal long,
hot Mediterranean summer is still a defining characteristic. Consequently,

5.7. The site at Górtys (Peloponnese) – Fourth century BC baths subject to severe ero-
sion at certain points in the past. Note the build up of sediments against the external walls
(photo: author).
the development of vegetation follows slightly different trajectories to the

Southern and Eastern Mediterranean regions. Moreover, wetland zones
are more common in Italy and the south of France than further south.
Whilst there are long-term changes in vegetation (often characterised by
woodland reduction), and specific species or groups of species do not
demonstrate resilience, human societies in many instances do both cause
disturbance and absorb that disturbance (Walker & Salt 2006).
One characteristic central-Italian vegetation history is provided by the
pollen diagram from Valle di Castiglione near Rome (Follieri, Magri,
& Sadori 1989). Here, we see the expansion of deciduous trees at the
start of the Holocene, with the emergence of deciduous oak and Corylus
(hazel) as dominant species (Magri 1995: 356), a process that is broadly
contemporary with those across the Eastern Mediterranean. Although
the forest reached its maximum extent at about 3100 BC, the first indi-
cations of human activity, in the form of cereal pollen, appeared about
6,000 years ago, with Juglans (walnut) and Castanea (chestnut) then
appearing about 4,200 and 3,300 years ago respectively (Russo Ermolli
& di Pasquale 2002: 217). Oak appears to have been important in this
area up until about 4,300 years ago. Here, we have to imagine a human-
influenced but stable woodland environment that showed little signs of
early degradation but rather manipulation and management. By c. 1700
BC, the woodland had contracted, and economically important species
such as Castanea and Olea were present. These species expanded by c.
600 BC, with little change in the other woodland vegetation (Magri &
Sadori 1999: 254). Despite a decrease in relative proportions, species such
as Quercus and Corylus may well have been economically valuable, possi-
bly providing fodder but also fuel and building material. Consequently,
these species could have been ‘curated’.
In southern Italy, at the start of the Holocene, birch and steppe spe-
cies were replaced by Quercus and Ulmus (elm). Again, the relatively
high altitude of the pollen core sites means that strong evidence for
human impact appears quite late in the sequences, with the first sig-
nificant change at the the Laghi di Monticchio site (656 m) not taking
place before c. 2500 BC (Early Bronze Age). Quercus ilex (holm oak)
also colonised at this time, and may be indicative of warmer winter tem-
peratures. Abies spp. (fir) and Taxus (yew) disappeared c. 500 BC, with
climate change or the impact of agriculture cited as the causes. It is per-
haps unusual for Abies spp. to disappear entirely, and we might assume
that people would usually make some effort to conserve or manage such
a valuable resource. It is possible that this tree was killed by disease or a
change in climate (Watts et al. 1996: 124). In Apulia, the period between
3900 and 2200 BC saw the presence of mixed oak woodland, followed
by a steep decrease in woodland species. The shift towards a drier cli-
mate, rather than human impact, might explain this change (Caroli &
Caldara 2007). The coastal site at Lago Alimini Piccolo, Apulia, saw a
similar pattern, with deciduous oak dominating the landscape between
3200 and 2350 BC, and a reduction in forest cover after this period.
However, the presence of extensive Neolithic activity to the west on the
Tavoliere plain (Pessina & Tin‚ 2008: 167–70) would have had some
impact on vegetation across the region. Later, during the Bronze Age, an
initial, albeit temporary, deforestation phase at around 2000 BC might
have been caused by drought. The predictable, anthropogenic woodland

mosaic with Olea then emerged during the period 1900–100 BC, with
the Roman period characterised by the decline in woodland (Di Rita &
Magri 2009).
In some parts of Italy, there is relatively little evidence for Early
Neolithic activity or impact on the landscape. For example, there is
an Early Neolithic ‘gap’ on the central Adriatic coast (Skeates 1999).
During the later Neolithic and into the Bronze Age, arboreal pollen lev-
els were quite high, with Ulmus and Fagus at their maximum. Other
Mediterranean diagrams also fail to demonstrate substantial prehistoric
impact on tree taxa. Magri (1995: 357) argues for a ‘natural’ state where
mosaic cycles are dominant, with continuous change characterising for-
est development.
In the Terramare, on the Po plain, cereal field pastures and meadows
characterised the landscape during the Middle–Late Bronze Age (1650–
1200 BC). Some consider that climatic deterioration from around 1300
BC may have contributed to the decline of the Terramare di Montale
(Mercuri et al. 2006). At 249 m, the Lago Lucone in northern Italy
attracted Early–Middle Bronze Age settlers who appear to have cleared
the mixed oak forest that had been established during the Early Holocene.
From about 1100 BC, the palaeobotanical evidence suggests a reduction
in human activity in this area. Here, the decline in activity cannot be
explained by climatic deterioration alone (Valsecchi et al. 2006), and may
well be part of a broader trend of Late Bronze Age decline.
As in other regions across the Mediterranean, Bronze Age erosion has
been identified in a number of areas in Italy, including the Biferno Valley
(east-central Italy) (Hunt 1995). Some areas did see earlier erosion, such
as the Marche region of central Italy where Neolithic forest clearance
might have caused erosion (Coltorti 1993). In Sicily, late fourth-millen-
nium (Sicilian Copper Age) hill-slope erosion in the Troina territory has
been identified (Ayala & French 2003), and a similar date range is sug-
gested for erosion at Basilicata in southern Italy. However, this erosive
phase has been given two quite different termini ante quem through the
radiocarbon dating of a buried soil on top of the eroded sediment: one
estimation was c. 2100 BC (Neboit 1984), whilst the second estimation
gives the buried soil a date of the early to mid-fifth millennium.
Although wetland zones are often considered marginal (a topic con-
sidered in Chapter 4), it is possible that malaria was not a threat in the
Neolithic, and the advantage of having access to a constant water sup-

ply and aquatic food resources was the fundamental attraction. Palù di
Livenza (30 m above sea level in the Friulian foothills in north-east Italy)
was located within or close to dense forest (Pini 2004: 773), and the
presence of pollen of Cyperacae (sedges) and Sparganium (bur-reed)
implies the existence of wetland zones in this area. The village com-
prised pile dwellings, and when the settlement expanded, Corylus (hazel)
and Alnus glutinosa (European alder) were still present, and species such
as Hedera (ivy) appeared, the latter possibly employed as winter fodder
for cattle (ibid.: 774). Wetland species actually decreased, and the for-
est contracted − a major impact on the woodland would have been the
use of timber for the pile-dwelling structures that covered an area of
about 60,000 m2. Three phases of construction have been identified: the
first 4325–4715 BC, the second 3910–4170 BC, and the third phase
3305–3840 BC (ibid.: 771). Although the earliest dates for the settle-
ment indicate a human presence in the area at c. 4495 BC, the pollen
diagram does not indicate any agricultural activity prior to 3960 BC.
Therefore, it is possible that the earliest occupants on the site were not
involved in agriculture or, more likely, that woodland density and con-
comitant pollen production may have masked the agricultural pollen
signal. The pollen and other archaeobotanical evidence (charcoal, fossil
fruits, and seeds) suggest that the wood used for construction and for
burning came from around the mire, with cereal cultivation taking place
close to the pile dwellings. This emphasises the importance of forest or
woodland agriculture, where vegetation management is dominated by
woodland-management strategies. The same was true to the west in the
Languedoc (south of France) where the first evidence for forest transfor-
mation is dated to the Middle Neolithic, with increases in Buxus (box)
and Quercus ilex (holm oak). Chabal (1998: 77) is quite clear in her def-
inition of the differences between deforestation and clearing, essentially
arguing that, during the Neolithic, clearance − the preparation of land
for agricultural activity rather than the complete suppression of trees −
would have been the norm.
People and Environment in Southern French

Landscapes
A strong case for Early Neolithic erosion is made for the lower Rhône
Valley where we see an increase in the flooding of low-lying levels as
part of a wetter climatic phase c. 5300–5200 BC. This may well explain
the subsequent increase in the number of settlements on higher ground
(Berger 2005: 164) (Fig. 5.8). High levels of erosion took place dur-
ing the period centred on 5400 to 5200 BC. The causes are unclear,
as population levels would surely have been quite low. However, if we
accept that, in some environments, early farmers may have had a dispro-
portionate impact on the landscape if they initially cleared zones situated
on lower valley sides and alluvial terraces, when combined with climatic
deterioration, this could have led to increased erosion. A newly cleared
land surface devoid of grass or weeds would be quite fragile. This might
constitute a form of low resilience at a specific time and place. However,
as with other regions discussed above, it is unlikely that erosion per se
resulted in economic and cultural stress. Moreover, this period was very
much a precursor to the ultimate development of what we accept as
the Mediterranean climate in this subregion. Many now agree that the
transition to a Mediterranean-type climate in the Western Mediterranean
occurred about 4,000 years ago, whilst in south-east Spain, these condi-
tions existed more than 10,000 years ago and gradually moved towards
south-eastern France, although the growing evidence for spatial and
temporal variations in climate across the Mediterranean throughout the
Holocene demonstrates that the identification of a single regional trend
is impossible (de Beaulieu et al. 2005). For example, as mentioned above,
Jalut et al. (2000) have identified six dry phases during the Holocene
contemporary with positive 14C anomalies (when sunspot activity is low,
the 14C/12C ratio is high – a positive 14C anomaly). This can result in a
cooler climate and aridity in the Eastern Mediterranean. We also know
that there was a general decline in lake levels during the third millen-
nium BC, whilst in the Eastern Mediterranean, a Middle Holocene wet
phase was followed by a transition to drier conditions during the period
3500–2500 BC (Magny et al. 2002: 49–53).
In terms of vegetation development in the south of France, we see
at the Etoile site near Nice (Dubar et al. 1986) a familiar pattern of
Alnus, Corylus, and deciduous Quercus dominating the landscape during
the Early to Middle Holocene, then, from c. 2000 BC, Pinus becom-
ing more prominent. Further west, in the Languedoc, at Marsillargues
near Montpellier (Planchais 1982) and Lattes (or Lattara – the Iron Age
and Roman settlement near modern Montpellier) (Planchais, Duzer,
& Fontugne 1991), deciduous trees reached their peak at the start of

Late Neolithic Early Bronze Age Late Bronze Age Iron Age Roman Middle Age Modern period
Erosion/degrad ation of soil and degree of soil protect ion

400
Middle Bronze Age
100
50
20
Number of settlements by half-century

closed forest Water meadows ? 15
humid zones Closed
water meadows
Very Strong riparian forest forest
fields
10
Meadows 9
open forest ? 8
drier drier drier fields, vines drier 7

Strong grasslands grasslands grasslands or open forest grasslands 6
fields 5
4
drier Fields and
water meadows 3
grasslands
2
Week
1
? 0
3000 2000
2000 1000 0 1000
BC. AD
1
2
Dominant vegetation cover inferred from
Time lag between peaks of settlement and erosion
anthracology and molluscan analyses
Settlement density inferred from Lowering of residual atmospheric 14C
archaeological fieldwork 1 (after Stuiver et Braziunas 1993)
Degradation/erosion of soils on slopes Lake levels in the Jura and Northern Alps
and alluviation on lower plains
2 (Magny 2004)
5.8. Erosion and stability in the Rhône Valley. The relationship between settlement activity, certain changes in climate, and erosion. (By permission of Berger, J.-F.
(2003), Les étapes de la morphogenèse holocène dans le sud de la France, in Archéologie et Systèmes Socio-environnementaux: Etudes Multiscalaires sur la Vallée du
Rhône dans le Programme ARCHEOMEDES, fig. 148, pp. 87–167.)
the second millennia BC, after which evergreen Quercus ilex (holm oak)
increased.
The broader, regional picture of the Neolithic decline in deciduous oak
and the creation of open spaces is supported by anthracological research
(Thiebault 1997). Anthracology also allows the identification of more
nuanced processes. The appearance and structure of oak-dominated veg-
etation varied enormously, depending on the composition of specific spe-
cies: Quercus – pubescens, robur, ilex, or suber (cork oak). Studies of the
changes in wood anatomy permit the identification of changes in climate,
as certain characteristics are determined by temperature and precipitation
(Terral & Mengual 1999). One study suggests that there was an increase
of 2–2.5°C in mean annual temperature between the Mesolithic and the
Bronze Age, the greatest increase taking place from the Middle Neolithic
to the Late Bronze Age (ibid.: 84). The Bronze Age climate could have
been quite similar to that of the present day (warm and sub-humid),
whilst an increase in rainfall may have characterised the transition from
the Bronze Age to the Iron Age (ibid.: 88). From c. 2000 BC onwards,
the replacement of deciduous woodland with open, sclerophyllous veg-
etation was a product of both increased human impact on these land-
scapes and climate change (Heinz & Thiebault 1998). At the relatively
high altitude of 970 m, the Balma Margineda record in Andorra covers
the periods from the Azilien (12,000–9000 BC) to the Early Neolithic.
By the Early Neolithic, there was forest dominated by Quercus robur
(pedunculate oak) (Vernet 1997: 93). During this period, we appear to
see what Vernet refers to as a balanced exploitation of the vegetation,
implying little or no impact from farming (ibid.: 97). However, certain
periods do seem to have seen some erosion. A similar situation may well
have existed in Lower Provence where the absence of palaeoecologi-
cal evidence prevents a clear assessment of vegetation change, but geo
archaeological data suggest possible impact on the geomorphic system
by Late Neolithic farmers (Jorda & Provansal 1996).
Although Neolithic societies would have experienced – and perhaps
even caused – landscape degradation at the local level, there is no evi-
dence to suggest that such environmental changes were insurmountable,
causing changes in Neolithic societies or their economic practices.
Moving into the Bronze Age, some have blamed erosion for induc-
ing serious economic and thus cultural problems. For example, a reduc-
tion in activity during the Early Bronze Age in the Languedoc might be
explained by one of the following: (1) the cultural sequences employed

in the definition of the Bronze Age in the Languedoc are imperfect,
and sites from earlier or later periods may in fact be Bronze Age; (2) the
Early Bronze Age was characterised by economic collapse, and popu-
lations were dispersed and/or situated on marginal sites; or (3) Early
Bronze Age sites have been removed or masked by certain erosive pro-
cesses (Wainwright 1994: 287–8). There is little doubt that any such
erosion was related to changes in vegetation cover. Here, the develop-
ment of garrigue would have been important. Garrigue vegetation did
develop during this period, but its evolution was not synchronous across
the region and was not therefore exclusively caused by a change in cli-
mate. Consequently, it is likely that Bronze Age land degradation was
caused by overexploitation of environments in upland areas combined
with erosion aggravated by extreme storms (ibid.: 300). Gasco (1994:
100) considers that the oak forest in the south of France had already
been degraded by the start of the Bronze Age. A wider range of cul-
tural and economic processes, rather than environmental changes, may
well have influenced modifications in Early Bronze Age settlement and
economy. The possible arrival Beaker (pottery) using peoples could quite
easily have led to new forms of engagement with these landscapes that
resulted in new influences on vegetation and sediments.
During the Late Bronze Age, deforestation around densely populated
areas was marked. However, certain trees such as walnut and hazel may
well have been protected, along with oaks, as food sources. At the Grotte
Camprafaud (Hérault), wild chestnuts almost disappeared between 2770
and 2115 BC, being replaced by the domesticated variety. Evergreen oak
then became dominant from 1990 BC (Gasco 1994: 102). By the end
of the Bronze Age, large areas of previously uncultivated land had been
cleared. The increase in the area of land under cultivation is a profoundly
noteworthy phenomenon when considering how people interacted with
the farmed landscape. During the Early Bronze Age, with fields close to
the settlement, people did not have to travel any significant distance to
work their holdings. By the Late Bronze Age and Early Iron Age, the
evidence from storage pits and pollen diagrams suggests extended zones
of agricultural activity under the control of individual settlements. Thus,
people were engaged with an extensive landscape and experienced a differ-
ent intensity of engagement with that landscape, as they may have moved
through ‘corridors’, along paths, and so on to go and work patches of land
some distance from their settlements. Livestock, especially pigs and cattle,
were probably reared in wooded areas, whilst sheep may well have been
kept on the edge of the forest, thus degrading it further (ibid.: 103). This
pattern of landscape organisation and concomitant impact on vegetation
is reflected on the étangs (coastal mixed saline/freshwater bodies) sites in
Provence and the Languedoc, where woodland clearance took place from
c. 3000 BC onwards. Typical woodland species were replaced by Juglans,
Olea, and cereals, indicating agricultural clearance (Laval, Medus, &
Roux 1991: 270). Unsurprisingly, as time moved on, a more extensive
range of plants was exploited within an increasingly open landscape. In
the Languedoc, Papaver somniferum (opium poppy) and flax were grown
around the Bronze Age lagoon sites, but wild fruits, including acorns,
were still gathered. One interesting feature of the Languedocian settle-
ment pattern is the increase in the number of sites towards the end of the
Bronze Age (850–700 BC), with coastal settlements occupied during the
summer and hilltop sites providing permanent settlement locations. The
two zones share similar sets of archaeobotanical remains, comprising cere-
als and pulses, implying no profound production differences between the
two zones (Bouby, Leroy, & Carozza 1999).
When we visit certain Mediterranean landscapes, we might be sur-
prised by the emphasis placed on discussions of degradation. For exam-
ple, today, the lower Rhône Valley is quite fertile and green. So to what
extent can we demonstrate that degradation was ever a real problem
(van der Leeuw, Favory, & Fiches 2003)? The evidence attests to sta-
bility from the Late Neolithic through to the Chalcolithic (a period of
relative climatic warmth) (Berger 2003: 113), and then erosion phases
developed during the Early Bronze Age. To the south-east of this area,
at Glanum in the Alpilles (a low, short, limestone mountainous chain in
western Provence), quite severe erosion appears to have occurred during
the Chalcolithic. It is unclear, however, whether this directly affected
people as such (Provansal 1995b). Across the lower Rhône Valley area,
the archaeological evidence points to continuous settlement during these
periods, such that cycles of stability and erosion did not result in the
abandonment of this landscape. Whilst the Early Iron Age landscape
appears to have been subject to quite severe environmental degradation
(Berger 2003: 121–2), settlement activity, albeit restructured, continued
to take place. A newly stable geomorphic milieu provided the context for
later Iron Age and Early Roman activity.
People and Environment in Spanish Landscapes

Some Iberian research highlights the differences between the Northern
and Southern Mediterranean zones in terms of the relationship between
climate changes and vegetation development, especially for the Early to
Middle Holocene. Most lake and geochemical records indicate that the
Early Holocene across much of the Mediterranean was wetter. In the
southernmost areas (including southern Iberia) where water is a limit-
ing factor, evergreen taxa flourished during the Early Holocene. Pistacia,
for example, is characteristic of the Early Holocene in the Eastern
Mediterranean, and evergreen oaks appear to have flourished in southern
Iberia and north-west Africa (Fletcher, Boski, & Moura 2007; Fletcher
& Sánchez Goñi 2008; Rossignol-Strick 1999). In the northernmost
areas where winter temperatures are a limiting factor, evergreen taxa col-
onised later in the Holocene.
Variations in Vegetation Change Across Spain

One long-term trend in vegetation change across Spain, and indeed
most Mediterranean regions, is an ever-decreasing biomass (Pèrez-
Obiol et al. 2011). Within Iberia, there are significant variations in vege-
tation, ranging from the Pyrenees in the north, via the central plains and
then central mountains, across to the semi-arid landscapes of south-east
Spain. This variation is a consequence of geographical location, topo-
graphical position (i.e. altitude), proximity to the sea, wider changes in
climate, and, of course, human activity. The variation across Iberia also
includes what Carrión et al. (2010) characterise as ‘surprises’ in the bio-
geography of Mediterranean vegetation development; that is, Holocene
vegetation patterns that seem unusual, in that the species composition
is not always what we might expect given the geographical location
and concomitant climatic characteristics. This reinforces the need for
archaeologists to incorporate local vegetation histories wherever possi-
ble, and to avoid making generalising inferences from regional vegeta-
tion histories.
As already noted, the Iberian Peninsula covers a range of
biogeographical zones. If we can identify one useful geographical
boundary from the palaeoecological evidence, it is the Llobregat River
(Catalonia) (Fig. 5.1). This serves as a margin between evergreen and
deciduous communities (Riera Mora, Léspez, & Argilagés 2004). Close
to this boundary, in the Catalan interior areas (Banyoles), deciduous
oak forests were dominant until 5500/5000 BC, when pine colonised
(Pèrez-Obiol & Julià 1994).
With the Neolithic, a new engagement with woodland emerged,
where trees were not only cleared for arable agriculture and the creation
of pasture, but wood was also exploited as part of the production process
for ceramics and the building of farms and villages. Even if Neolithic
populations were comprised of incoming peoples with different attitudes
towards the natural world, we should not forget the problems, even dan-
gers, of living and working in a forested landscape, and those early farm-
ers would probably have sought areas that were naturally open or easy to
clear. As thermo-Mediterranean vegetation emerged, the opportunistic
expansion into zones of open woodland with smaller trees makes per-
fect sense, especially if incoming agriculturalists brought environmental
knowledge from a more arid, Eastern Mediterranean landscape.
In the Segura Mountains (Murcia, south-eastern Spain, about 1,100
m), human impact on the vegetation did not occur until quite late in the
Holocene. Pine, deciduous oaks, and other species, including hazel and
birch, were dominant during the period 4600–2800 BC, but after this
point, there was a ‘dramatic’ decline in deciduous oak as it was replaced
by evergreen oak and other evergreen species. The abrupt shift to an
evergreen oak-dominated forest appears to have occurred within a period
of 10–30 years (based on interpolated radiocarbon dates) (Carrión et al.
2001). Such a process is too quick to have been a response to climate
change. Senescence, nutrient deficiency, or diseases are possible explana-
tions. Whatever the reason, such a change would have been observable
within a human generation. After this period, up until c. 2120 BC, ever-
green oak – including Quercus rotundifolia (holm oak), the acorns of
which are used today as fodder – replaced deciduous oak and sclerophyl-
lous (hard-leaved) vegetation, with this woodland achieving its maxi-
mum between 2120 and 1630 BC. Survey work in the northern Alicante
Province demonstrated how the later Neolithic period saw a concentra-
tion of activities within a relatively restricted area (Barton et al. 1999).
Based on the isotopic analyses of human remains from sites spanning
the Late Neolithic to the Bronze Age in the Alicante region, it seems
that there was little change in diet over this period, and, as with other
studies in the Mediterranean, there is evidence of nutritional deficiencies
(McClure et al. 2011). Such evidence implies a certain level of stability
or perhaps inertia in environmental knowledge, as these groups do not
appear to have changed their food-production strategies. Neolithic and

Bronze Age settlement is sparse in this area, and the palynological evi-
dence corroborates this picture of limited human impact.
In most parts of Spain, there is little evidence of Early Neolithic
(5600–4500 BC) influence upon the vegetation, although vegetation
patterns may well have changed quite radically prior to this due to the
c. 6200 BC rapid climate-change event (González-Sampériz et al. 2009).
As we move into the Middle and Later Neolithic (4500–2400 BC), the
combination of aridification and probable human impact on vegetation
becomes apparent. For example, in the Navarrés region, increased clear-
ance took place between 4000 and 2700 BC. The presence of macro-
charcoal (evidence for localised burning events) could represent natural
or human-induced fire. An increase in Plantago (plantain) pollen just
after a decrease in Quercus in pollen diagrams certainly supports the
case for anthropogenic impact on this landscape (Carrión & Van Geel
1999: 231). Archaeological evidence indicates that the first domesticated
cereals (einkorn wheat and barley), legumes (peas, fava beans, and len-
tils), and animals (cow, sheep, goat, and pig) were present by 5600 BC
(Bernabeu et al. 2001; Bernabeu & Bernabeu 1993). Settlements were
small and dispersed across the area around Valencia during the Early
Neolithic. The settlement pattern appears to shift from dispersed, rela-
tively ephemeral settlements in the Early Neolithic to aggregated villages
during the Middle and Late Neolithic (McClure et al. 2006). During the
Early Neolithic, it is possible that the rotation of cereals and legumes was
frequent under the traditional Mediterranean el huerto system of garden-
plot cultivation (ibid.: 9). If this were the case, then extensive woodland
clearance would not have been necessary, and this should be reflected in
pollen diagrams, with tree-pollen values remaining relatively stable for
much of the Neolithic.
As discussed in the introductory chapter, different forms of cultural
and human ecology have been important in the assessment of human–
environment interactions around the world, and certain forms of cul-
tural ecology do underpin some of the discursive sections in this book.
However, we should be aware of the limitations of Human Behavioural
Ecology (HBE), and in particular, one of its foundational tenets, Ideal
Free Distribution (IFD) (McClure et al. 2006). Essentially, IFD posits
that animals in a landscape will be distributed across a range of resources,
and that the number of animals present within an area will be a direct
function of the resource level within that area. This model was initially
developed to explain habitat distribution in birds (Fretwell & Lucas
1969). Its application to the study of the agricultural and pastoral poten-
tial of Neolithic landscape in Spain allows us to assess and characterise the
nature of early farming economies within specific landscapes. However,
whilst such a model might be applicable to non-human animals, its
application to the study of humans (or, more specifically, ancient human
awareness of carrying capacity) is problematic. IFD models implic-
itly deny the probability of environmental knowledge being culturally,
ideologically, and religiously structured: all environmental knowledge is
mediated by cultural processes. Despite these issues, the IFD model for
the Neolithic in eastern Spain does articulate some interesting notions.
Relatively small changes in population density can result in settlement
shift, and population changes do not need to be substantial in order for
land-use changes to take place. A small increase in population density
on Early Neolithic settlements on the valley floors may have encouraged
movement towards the valley edges, where the growing population used
ox-drawn ploughs for the first time and may have constructed terraces,
thus improving a landscape that might have been initially considered
of secondary value. This change would have been increasingly impor-
tant as people and settlements moved away from the riverine margins
(McClure et al. 2006: 25). The expansion into new wooded areas may
also have been facilitated by transformations of the vegetation caused
by climate change. Such processes could have invited or encouraged the
expansion of certain agricultural activities. As the configuration of wood-
land evolved into ‘Mediterranean’ vegetation, with evergreen oaks and
other, smaller sclerophyllous species, this more open landscape would
have facilitated mobility and new intensified agricultural practices during
the later Neolithic.
During the later Neolithic, we see the extraction of surplus products
from animals − the defining characteristic of the ‘secondary products revo-
lution’ (Greenfield 2010; Sherratt 1981). Whilst this might have developed
initially during the early stages of the Neolithic (Evershed et al. 2008),
the intensification of these processes occurred during the later Neolithic
(4500–2400 BC) (Bintliff 2012: 51–2) with greater numbers of cows,
sheep, and goats reaching maturity, thus increasing grazing/browsing
pressure and concomitant impact on vegetation. In some areas, charcoal
analyses suggest that forest was restricted to higher elevations. Therefore,
it seems likely that animal husbandry around later Neolithic settlements

was responsible for much of the local impact on vegetation at this time.
One of the limits to vegetation reconstruction based on palynolog-
ical work is that, more often than not, the results can only be used to
suggest variations in the spatial distribution of different types of vegeta-
tion within a landscape. Palynology can only hint at the development of
vegetation mosaics in the past. This is a problem in any region, and for
the Mediterranean, the history of complex vegetation mosaics and the
impact of people upon these is a particularly important issue. Here, vege-
tation mosaics and patches are vital to our understanding of past human–
environment interactions. Mosaics and patches are easily observable in
the modern landscape, but cannot be reconstructed with accuracy for
past landscapes. However, using palynological and anthracological infor-
mation in conjunction with geological, geomorphological, and hydro-
logical evidence, we can suggest how different mosaic communities may
have been distributed across the landscape. We might place the village
at the centre of a postulated mosaic, and consider how the inhabitants
of Early to Middle Neolithic villages would logically have opened the
vegetation closest to them. Moreover, we should not forget that arable
agriculture in open plots was just one of many activities that would have
taken place around a Neolithic settlement. The possible variations in the
combinations of pastoral, arable, and wild-resource areas and the distri-
bution and relative proportions of these different land-use categories can
only be estimated and perhaps modelled via geographic information sys-
tems (GIS) (Robb & Van Hove 2003). One process is clear – that of an
ever-increasing opening of areas close to sites during the Late Neolithic
as new farming strategies emerged in certain parts of the Mediterranean.
Considered by McClure to be similar to the traditional Mediterranean
farming system known as secano or dry-land farming (Bernabeu 1995),
it comprised a system where crop rotation was longer, and the distances
between newly farmed areas and the village increased. These changes also
included the use of cave sites as enclosures for sheep and goats (McClure
et al. 2006: 23). Consequently, we would expect to see these changes
reflected in pollen diagrams.
Late Neolithic–Bronze Age Landscapes in Spain

Parts of Spain saw the development of complex and successful soci-
eties during the later prehistoric periods. The combination of
palaeoenvironmental and archaeological research here allows us to assess

some of the relationships that developed between people and landscapes
during these periods.
Despite evidence for erosion during the Neolithic, in areas such
as the Polop Alto Valley (eastern Spain) (Barton et al. 2002: 165), it
seems unlikely that erosion posed a serious threat to Iberian Neolithic
farmers. It was during the Bronze Age that potential problems devel-
oped − not simply environmental problems, but stresses exacerbated by
a complex social hierarchy within a landscape that contained patches
of degraded land that may well have become unusable at certain times.
Such a scenario seems quite likely for the east and south-east areas of
Spain, where a period of change saw a large number of settlements
abandoned and new ones created. Thus, a series of well-dated sites
show that settlements in eastern Iberia had shorter occupation phases
and were abandoned earlier than sites in the west, with many aban-
doned c. 2200 BC. That environmental processes may well have caused
this seems plausible, especially as the east is more arid than the west
(Lillios 1997: 173). Specific geoarchaeological and archaeological
evidence comes from the Librilla rambla, a torrential tributary of the
Guadalentín River in south-east Spain, in whose valley are impressive
sedimentary units, one sequence of which, dated to 5409–5149 BC to
3070–2711 BC, is 7–8 m thick. Neolithic sites were found within this
sequence, whose accumulation represents a low-energy semi-endorheic
environment (a basin that is closed in the sense that there is little dis-
charge to external bodies of water) (see Cano Gomariz 1993 cited in
Calmel-Avila 2002: 105). The rambla then cut into these earlier layers,
depositing 8 m of sediment (sand and pebble bars). Chalcolithic shards
and traces of a settlement site dated c. 2360 BC were found within this
layer. A third layer, some 11 m thick, cuts and covers the previous units.
Such sediments indicate that there were moments of powerful water
discharge (Calmel-Avila 2002: 106), most likely storm events, dislodg-
ing sediment from exposed areas across a catchment. Consequently, it
is important to consider the nature of human activity both on and off
site, and thus to assess the possible reasons for sediment removal and
then its subsequent redeposition. The study of the relationships of sites
with their catchments is crucial, as human experience of activity in the
landscape and any environmental processes is not fixed to one particu-
lar point in space.
Thus far, we have seen similar patterns in the evolution of vegetation

during the first part of the Holocene across the Northern Mediterranean.
However, there were some zones that were never covered by dense for-
est. One such region is the semi-arid zone of Almería near the south-
eastern coast of Spain. Vegetation here has oscillated between steppe and
shrub communities. Between 5000 and 2500 BC, the landscape com-
prised extensive vegetation cover. Shrubs such as Pistacia expanded, and
deciduous oak was present (Pantaleon-Cano et al. 2003: 115). This was
followed by the establishment of steppe, the disappearance of deciduous
oak, and the development of a so-called marginal landscape − a landscape
which is considered degraded as a consequence of human activity. One of
the interesting aspects of some of the work in the Almería area is the pres-
ence of Olea pollen from the Early Holocene onwards. The expansion of
this tree was tied to the development of shrub communities in general,
and its history is not exclusively tied to anthropogenic intervention or
exploitation. Moreover, the absence of obvious anthropogenic indicators
in these pollen diagrams is also important (ibid.: 115–6). Whilst there is
no doubt that this area was economically important from the middle of
the third millennium BC onwards, with agricultural and metallurgical
activities taking place, these activities were not registered in the palyno-
logical record. Any species that might be associated with anthropogenic
impact may also have existed under natural conditions (ibid.: 117).
One infamous landscape type in this part of the world is the so-called
badlands – zones that do not necessarily represent the inevitable product
of extreme erosion processes but rather a specific landform type (Grove
& Rackham 2001: ch. 15). They tend to exist in areas characterised by
relatively fragile, superficial geological deposits (usually soft marls) and
where tectonic activity is important. Quite often, eroded material (via
gullying and slumping) is removed from the basal areas of these land-
forms by rivers or streams. Eastern Spain is one of the most arid zones
in Europe and the Northern Mediterranean (Fig. 5.9). Today, parts of
this area are considered to be ‘Euro-Desert’ and, with an annual rainfall
of less than 300 mm, there is little chance of vegetation taking hold.
Almería and the eastern Alpujarra are the most arid zones in Spain today.
This part of Spain is where the Los Millares culture developed, and there
is little doubt that this zone was not as arid during the Chalcolithic. The
changes in vegetation (as seen in pollen diagrams) imply that this area
became more arid at a time (between 1300 and 1000 BC) when many

5.9. Badlands topography (south-east Spain) (photo: author).
of the Argaric settlements appear to have been abandoned (Castro et al.

1999).
Site Catchments in Proto-Historic Spain

So far, no reference has been made to site catchment analysis (SCA)
(Vita-Finzi & Higgs 1970), partly because this approach has not been
explicitly employed in recent years, although much GIS-based work
does incorporate some form of SCA. The assessment of catchment areas
around sites is important, and is in fact an integral part of modern land-
scape archaeology, albeit implicit in many approaches. Characterising
Mediterranean landscapes as complex mosaics where patchiness increases
with time in many parts of the Mediterranean does mean that we need
to consider the range of probable patch types across our study areas. The
use of GIS and the development of models is one approach that can help
in this endeavour (Robb & Van Hove 2003). Traditionally, SCA involves
the characterisation of resources available within a given distance that
can be walked within a certain time, often between two and four hours.
Sometimes, these zones are subdivided into shorter time–distance zones
as well. More often than not, the potential of a landscape is based upon
modern vegetation, soils, and agricultural use. More recent develop-
ments, employing cost–distances analyses, incorporate characteristics of
topography and, when possible, what is known about political or cultural
districts (e.g. Farinetti 2009), thus facilitating more ‘realistic’ assessments
of how people engaged with landscapes.
Gilman and Thornes’s (1985: 33) pioneering use of site catchment
analyses in south-east Spain presented direct and indirect evidence for
the development of Bronze Age irrigation. The existence of sites in areas
that currently receive less than 250 mm rainfall in a year is indirect evi-
dence, whilst the presence of ditches on some sites might be consid-
ered direct evidence. The information on land quality and water supply
was then used to inform the assessment of site territories and the evi-
dence for land-use preferences associated with specific sites. The rela-
tive proportions of the different land-use categories within the 12- and
30-minute zones were compared with the categories in the two-hour
zone (the peripheral zone). It is interesting to note that two sites close
to one another (the Campico De Labor and La Bastida sites) do not
have direct access to good-quality land suitable for irrigation (according
to the land-potential characterisation model developed by Gilman and
Thornes). This situation raises an important issue. Can these types of
territorial analysis work when several sites occupy the same territories,
or when their territories overlap? Whilst many of the areas within which
archaeological sites were situated appear to have been stable, it is quite
probable that other parts of the landscape (off-site zones), such as fields,
were subject to erosion. It is likely that settlement sites were purposely
located on stable areas. This does not mean to say that people, as part of
their working lives across other parts of the landscape, did not have to
deal with erosion.
In many parts of the Iberian Peninsula, the Bronze Age witnessed the
emergence of a managed and more open landscape, with arable and pas-
toral agriculture spreading out over much wider areas, along with exten-
sive exploitation of woodland for timber and fuel, and the use of these
materials in processes ranging from construction through to mining.
However, the story is not one of ever-increasing human expansion and
concomitant impact on the landscape and vegetation, but rather a wax-
ing and waning of activity, where archaeologists must carefully assess the
intersection of cultural and environmental processes, where vegetation
depletion can be a response to climate change, economic changes, or

both. Palaeoecological evidence elucidates these processes in a number
of ways.
In south-east Spain, the evidence supports the possibility of dry, rain-fed
polyculture farming not being practiced until the second millennium BC.
If the climate was stable (i.e. arid), it has been argued that any early settle-
ment in the south-east of Spain must have included the development of
irrigation technology, otherwise the arid conditions would have impeded
activity here (see Gilman 1976: 316). As Ruiz et al. (1992) observe, evi-
dence for climate change often lacks precision when attempting to under-
stand variations between different chrono-cultural periods. The problem
when talking about climate is that we can rarely measure what weather
was like within a given climatic period. For example, the overall levels of
precipitation are less important than the seasonal variation in rainfall; rain
needs to fall at the right time of year for successful farming.
The first ‘peak’ in population was the Argaric Bronze Age, with the
population growing during the third millennium BC (Los Millares
Copper Age). It is during this period that we see the development of
social inequalities, differences in access to exotic materials, and the devel-
opment of fortifications around settlements. The landscape around these
settlements was increasingly opened up and woodland disappeared. ‘The
environmental impact of the Argaric political system was clearly felt in
the second half of the second millennium BC’ (Castro et al. 2000: 155).
During the later Argaric period, slow colluviation and possible slope
management in the form of small terraces might represent a response
to these environmental characteristics (Wainwright 2004). A non-des-
ert landscape normally passes through a phase of ‘degradation’ before
‘desertification’, so certain regions may experience several phases of deg-
radation but recover from these. South-eastern Spain may have gone
through at least four cycles of degradation (van der Leeuw 2003: 12),
and each time a reassessment of the landscape and the identification of
new economically viable niches may well have allowed a group of people
to remain active within a given area. The landscape around the enig-
matic Argaric site of Castellon Alto provides an excellent example of a
complex mosaic landscape comprising a river with its terraces, largely
suitable for arable agriculture, then small niches within the surrounding
undulating topography that would have provided grazing and zones for
arboriculture (Fig. 5.10) (Cortés et al. 1997; Rodriguez-Ariza & Ruiz

5.10. The landscape around Castellon Alto (south-east Spain). Note the variation in landform types, with rich
vegetation and agricultural activity concentrated along the valley bottom (photo: author).
1996). The Copper and Bronze Age site of Gatas, which is 13 km away
from El Argar and 2 km south of the River Aguas, provides an example
for discussion. The botanical remains from the site do not permit the
inference of local climatic fluctuations (Ruiz et al. 1992: 20), but they
do present a clear image of the natural resources exploited at the site.
For example, it is apparent that 90% of the plant resources came from
the immediate area.
It is possible that under conditions where the environment came under
‘stress’, elite groups in Argaric society could not incorporate more exten-
sive areas into their production system. Another possibility is that lower-
status groups resisted the developing social inequalities (ibid.: 32–3):
an increase in Bronze Age violence is attested to by palaeopathological
research (Jiménez-Brobeil, du Souich, & Al Oumaoui 2009). During the
post-Argaric phase (1500–1300 BC), the range of crop species grown
was greater than the preceding phase, coinciding with the appearance of
flax, vine, and olive (Castro et al. 1999: 851–2). The charcoal evidence
implies further clearance, as well as a reduction in the use or availability
of timber. The presence of Tamarix implies the development of saline

soils. Between 1300 and 1000 BC, many of the settlements that were
established during the Argaric period, including that at Gatas, were aban-
doned. The range of plant species grown on or near the site decreased,
whilst the exploitation of domestic and wild animals increased (ibid.:
853). Deciduous trees also disappeared entirely during this period, as
environmental conditions in the area became more arid (ibid.).
Many of these landscapes would have comprised a patchwork of het-
erogeneous farming practices – systems that may have been more akin
to the modern concept of ‘permaculture’ (sustainable agricultural prac-
tices that are successfully adapted to the range of niches within a given
area), rather than open landscapes dominated by a restricted number of
crops and practices, as often implied by crude percentages of woodland,
grassland, arable, and pastoral indicators drawn from pollen diagrams.
Agricultural regimes developed strategies for the exploitation of specific
niches across valleys, with the agricultural potential of each part of the
landscape successfully harnessed, and in particular, the river terrace soils
used for arable production, especially for legume crops (Chapman 2008:
202). Despite this legitimately optimistic assessment of environmental
potential, we now need to consider a range of specific engagements with
these environment types and the evidence for human response to envir
onmental potential and environmental problems. The following chapter
therefore assesses a range of specific processes and human engagements
with vegetation and geomorphic processes within a perspective that con-
siders temporally and spatially specific examples of human interaction
with and impact upon these landscape elements.
6
Working and Managing Mediterranean Environments
Lifeways in Mediterranean Environments
As discussed in the preceding chapter, much palaeoenvironmental work

has traditionally been concerned with the description of the history of
geomorphological and palaeoecological processes – ascribing natural
and/or anthropogenic causes to phenomena observed in the environ-
mental record within a framework that emphasises chronological phases
of ‘stability’ and ‘instability’, but with an overall trend of landscape deg-
radation. The fundamental questions are when did degradation start, and
who or what were responsible (e.g. Delano-Smith 1996)? For example,
the vegetation changes that we see across the Mediterranean through
the Holocene might be considered as the product of climate change
(and therefore ‘natural’), whilst the development of agricultural environ-
ments, whether for crops or pasture, is clearly an anthropogenic pro-
cess. However, those large areas of rarely exploited scrubby woodland
which cover large parts of the Mediterranean are in some ways a natural
response to changes in environmental conditions. One argument is that
this secondary succession vegetation (the sclerophyllous scrubby-wood-
land) is a direct consequence of both Middle Holocene changes in cli-
mate and human impact. This is an example of resilience, and we should
consider that the degradation narrative is therefore culturally defined,
where researchers imply or assert that cultural determinants were more
influential than climatic, with climate change creating conditions that
predisposed vegetation to degradation. What we have to consider is the
extent to which people and/or climate have rendered landscapes unus-
able or unproductive in the past.
171
172 Working and Managing Mediterranean Environments
Unsurprisingly, different researchers involved in the study of

Mediterranean landscape change have varied approaches to the analy-
sis of these phenomena (e.g. Bintliff 2002; Thornes 1987; van Andel
et al. 1986; van Andel & Zangger 1990). Some scholars have criticised
the degradationist paradigm (notably Horden & Purcell 2000; Grove &
Rackham 2001), and, to some degree, they have questioned the extent
and impact of erosional processes within the Mediterranean. At one level,
these narratives accentuate the long and medium durée, and articulate
models that present humans as an anonymous, nebulous group operat-
ing within long timescales (Walsh 2004).
In this chapter, we build upon the processes outlined in the preceding
chapter through an analysis of possible scenarios where we can talk about
direct engagement with certain environmental processes, and consider
practices and lifeways that contributed to the construction of new cul-
tural ecologies. The final section presents specific case studies where we
can identify various forms of interaction with landscape processes, and in
the case of the Sainte Victoire Mountain (south of France), an example
where the assessment of diachronic changes in engagements with the
same landscape can be inferred.
Clearance, Terracing, and the Creation of the

Sustainable Mediterranean Landscape
As we have already seen, Neolithic populations had some impact on
parts of the landscape, but it is difficult to argue for intensive and exten-
sive management of the vast areas of the Mediterranean prior to the
Bronze Age. One of the important changes to Bronze Age landscapes
was the impact on vegetation. For example, a low-altitude pollen dia-
gram from Lake Lerna on the Argive plain makes a valuable contribu-
tion to understanding of the vegetation changes close to a number of
important Bronze Age sites: Mycenae, Tiryns, and Lerna (Jahns 1993:
187) (Fig. 6.1). From c. 5700 BC, deciduous species including Quercus
(oak), Corylus (hazel), and Carpinus (hornbeam) were common, along
with evergreen species such as Phillyrea, Pistacia (cashew family, incl.
pistachio), and evergreen oaks. Later, a changing vegetation pattern does
point to some human impact on the environment, but such changes
could well be part of a ‘normal’ successional change, and do not sug-
gest profound anthropogenic influences (Jahns 1993). In the Nemea
Valley, some 20 km to the north of Argos, Ostrya (hophornbeam), Tilia

Sea of Azov
Bay
of
Biscay
Black Sea
1522 2426
23 Gulf of
Lions
14 8 10
13 9 Adriatic
16 Sea
Balearic
Sea
Gulf of Aegean Sea
25
11 Ionian Sea 21
32 7
Strait of Gilbraltar 1 4
17
18 Sea of Crete
5
6
Mediterranean Sea
12
20
Gulf of Sidra 19
6.1. Map of sites referred to in this chapter. 1: Lake Lerna, 2: Mycenae/Tiyrns, 3: Nemea Valley/Kleonai, 4: Methena, 5: Kalavasos Tenta/Mitsinjites, 6: Pseira,
7: Delos, 8: Ebro Valley, 9: Barcelona plain, 10: River Llobregat, 11: Sierra de Gádor, 12: Kfar Samir, 13: Eastern Cossetània, 14: Lozoya Valley, 15: Gasquinoy
(Béziers, Hérault), 16: Lago Albao and Lago di Nemi, 17: Medjerda floodplain (northern Tunisia), 18: Segermes valley (north-east Tunisia), 19: At Khirbet Faynan,
20: Tripolitania, 21: Sagalassos, 22: Lattara, 23: Montou, 24: The Crau/Barbegal/Arles, 25: El Acebrón/Las Madres, 26: Sainte Victoire/ Domaine Richeaume/
Bramefan/Roque Vaoutade.
(lime), and Q. pubescens (pubescent oak) were present during the Early
Neolithic (6300–5100 BC), but it seems unlikely that dense woodland
was ever present here. During the Bronze Age, there was an expan-
sion of a maquis vegetation (Q. coccifera (kermes oak), Arbutus (incl.
strawberry tree), Phillyrea, and Pistacia), with deciduous species such
as Castanea (chestnut), Juglans, and Platanus (plane tree), along with
Olea emerging. From the Late Bronze Age through to the classical/
Hellenistic periods, we see an increase in open-country species, especially
Poaceae (grasses), taken to imply an increase in agricultural activity. Any
‘semi-natural’ woodland had disappeared by the Roman period, when
an open arable landscape with a dispersed settlement pattern had devel-
oped (Atherden, Hall, & Wright 1993: 354–5). Parts of lowland Greece
would have comprised open (i.e. unwooded) areas from relatively early
on, but there is little evidence for either extensive or intensive woodland
clearance.
The creation of relatively open landscape increases the possibility for
erosion. However, we cannot escape the fact that many sedimentary
sequences will have been truncated, and portions of the erosional record
lost (Delano-Smith 1996: 161). For example, Moody’s (2000: 58) con-
structive assessment of the Cretan record contends that certain sedi-
ments deposited by intense storms in the past may have been removed
by subsequent events. Storms and their associated flash floods and ero-
sion episodes can be localised, especially across landscapes comprised of
extreme topography. Therefore, it would require a series of spatially dis-
parate storms at a particular time of year (harvest or sowing time) for
such events to have an effect on the economy and society of an island
such as Crete (Moody 1997). There is evidence for Minoan flash-flood
deposits across Crete, in particular dated to the Middle and Late Minoan
periods (broadly the second millennium BC). The fact that there are few
flash-flood deposits dated to the Archaic–Hellenistic periods, and that
these Minoan deposits are as well preserved as those from the Little Ice
Age, suggests that the Late Bronze Age on Crete may have experienced
conditions – and, in particular, unpredictable, extreme weather – similar
to that of the Little Ice Age (Moody 2000: 58). However, as is argued
later, it is quite likely that extreme precipitation events and concomitant
erosion are more likely during warmer climates as evapotranspiration
and atmospheric instability are enhanced (Diodato et al. 2011), and this
Early to Middle Bronze Age phase is characterised by a complex climatic
Lifeways in Mediterranean E
nvironments 175
oscillation, with an arid phase bracketed by two wetter periods (Magny,

Vanniere, et al. 2009).
Although there is little doubt that deforestation enhances the poten-
tial for erosion, it is ploughing followed by heavy rainfall that results in
the most serious erosion events (Grove & Rackham 2001: 268). These
often-annual events are the most relevant to the study of human engage-
ments with erosion. Modern research into the relationships between
ploughing and erosion suggest that once animal-driven ploughs and ards
were introduced (during the later proto-historic periods), these may well
have led to increased erosion (Wainwright & Thornes 2004: 254). A
common response to erosion was terracing (Frederick & Krahtopoulou
2000) or trenching (Foxhall 1996). Despite the fact that ethnographic
work has shown that terrace construction is often seen as an aid to tilling
rather than a mechanism for controlling erosion (Green & Lemon 1996:
185), there is no doubt that terracing does help mitigate erosion.
Research has been carried out in various regions around the
Mediterranean to characterise and date terrace systems. The majority
of those apparent today are post-medieval (Blanchemanche 1990). It is
possible that terracing was initially developed during the Bronze Age or
perhaps earlier. In some areas of the Mediterranean, these systems are,
and have been, extremely extensive (Whitelaw 1991: 405). There is con-
vincing evidence for the use of terraces during antiquity (L. Nixon &
Price 2005). As Foxhall (1996) suggests, the absence of unambiguous
archaeological evidence is problematic, and we have to acknowledge that
trenching (which would not be as visible in the archaeological record
as terraces) was possibly used on Greek estates. Whatever method was
employed, the construction of terraces, or the digging of trenches, is
more than an instrumental and economically rational response to hill-
slope erosion. The terraces that were constructed around the Aegean
during the Bronze Age represent the organisation and control of the
landscape by dominant social groups: terraces were as much an articu-
lation of political power in the landscape as they were forms of erosion
management designed to enhance food production (Fig. 6.2). Whilst
unequivocal evidence for terraced Bronze Age landscapes is rare, there
is little doubt that later periods did see extensive and intensive forms of
landscape management. Terracing is but one response to erosion. Forbes
(2000) demonstrates how place-specific control mechanisms, such as the
regulation of the number of grazing animals in a given area, contribute
6.2. Terraces near to Stymphalos. Note Roman tile in section (lower right of section face). Note also the
terrace walls visible on the surface that are of course relatively modern. (b) Watercolour reconstruction of the
early terraces on Delos (island, top-left; terrace plan, top right) (by permission of J.-M. Gassend).
nvironments 177
to the successful management of areas which might otherwise evolve as

degraded landscapes.
Archaeological evidence for terraces (in the form of datable lengths
of walls that run along slope contours) reveals how erosion and potential
instability might have been understood and responded to. Butzer’s non-
equilibrium model is a useful tool for appreciating the non-linear nature
of landscape change, in particular, erosion processes in the Mediterranean.
Essentially, non-equilibrium models suggest that repeated disequilibrium
(phases of landscape change, such as severe erosion and soil depletion)
results in progressively different landscape characteristics that develop
over a number of centuries which can lead to complex ecological prob-
lems (Butzer 2005: 1784–5). The fact that significant erosion events
occurred during the Bronze Age, and that limited erosion appears to
characterise the classical period (it is estimated population levels were
four times greater during the Roman period than during the Bronze
Age) demonstrates that high population levels are not the principal cause
of erosion. Whilst one has to agree with Butzer’s (2005: 1786) observa-
tion that we cannot prove or disprove adaptive responses to prehistoric
destructive erosion, it is likely that terraces in some areas did constitute
a response.
An attempt to date field terraces in Methena demonstrates how the
precise dating of these all-important features is difficult (James, Mee,
& Taylor 1994). Links have been posited between the Neolithic site
of Kalavasos Tenta and nearby terraces in the Vasilikos Valley, south-
ern-central Cyprus. One notable aspect of this work was the attempt
to identify links between terrace formations and property boundaries.
Here, many terraces appear to coincide with property boundaries, and
they are thus interpreted as possible elements of individual landholdings
(Wagstaff 1992: 156–8). An adjacent terrace system at Mitsinjites dem-
onstrates no such relationship with property divisions. As noted before,
terrace systems are difficult to date, and the few sherds retrieved from
these terraces only allow us to suggest a possible Bronze Age start for
their development (ibid.: 160). An early example comes from Pseira,
where terraces are dated to 2200–1700 BC (Vogiatzakis & Rackham
2008: 248), and a Bronze Age origin for terraces in Lebanon and Delos
is possible (Harfouche 2007: 155, 170), whilst a twelfth to eleventh
centuries BC origin for Levantine terraces has been suggested (Gophna
1979). Over time, environmental knowledge and the nature of work in
the landscape would have become increasingly socially structured, and

the experience of erosion mitigation, via the construction of terraces,
would have emerged as an important element in the hierarchical organi-
sation and control of landscape.
As described above, in the Argolid, it was during the middle of the
third millennium BC that the first phase of severe soil erosion occurred,
whilst the end of the third millennium saw a substantial reduction in the
number of sites, probably caused by social and political processes beyond
the Argolid (van Andel & Runnels 1987: 93). However, certain parts
of the region, and the wider Mediterranean, would have been affected
by the c. 2200 BC arid phase, although, as explained earlier, citing this
event as a cause of economic and settlement contraction is problematic
(L. D. Brandon 2012; Roberts, Eastwood, et al. 2011).
During the Late Bronze Age (in particular, during the Mycenaean
period), a structured settlement hierarchy emerged (represented by, for
example, the palace sites of Mycenae and Tiryns). The Early Bronze
Age erosion phase might be seen as a short-term effect of a change in
settlement pattern – an increase in activity on coastal zones and a con-
comitant increase in the use of surrounding hill slopes (Weiburg et al.
2010). Towards the end of the thirteenth century BC, many sites were
destroyed. In the Argolid, there is a gap in the settlement record, with
evidence for a markedly reduced human presence until the mid-ninth
century BC. As settlement expanded once again, van Andel and Runnels
(1987: 102–3) contend (based on a hypothesised soil quality) that the
areas of deep soil were chosen, but that poor-quality zones were also
exploited – areas that had previously witnessed little activity. Olives will
grow on relatively poor-quality land, and will easily grow on hill slopes,
especially if the slopes are managed through terracing. Thus, the expan-
sion of cash crop agriculture, with its emphasis on olive production,
should be seen as an important thread in a network of processes related
to erosion management on terraced hill slopes characterised by poor-
quality soils and erosion.
The soil erosion that occurred between 300 and 50 BC took place
during a period of settlement/activity decline. This erosion may have
been caused by grazing on the areas of poorer-quality land, whilst the
good-quality land closer to the settlements was maintained as arable.
As van Andel and Runnels (1987: 147–8) note, terracing is an excellent
soil conservation mechanism, as long as terraces are maintained, but if
nvironments 179
SETTLEMENT HISTORY
Slaviv Migration
& POPULATION ( 3 x ) OF
Martitime commerce
Black Death
SOUTHERN ARGOLID
“Dark Age” migration

General destruction,
Breakdown of trade
Mycenanean Era
10,000
Disintensification,
Intensification,
Warfar
Destruction
Metallurgy,
Expansion of Olive Cultivation

5000
Late Roman agricultural

Threshold for
Overseas Colonization,
3000
hierarchical
2000 settlement
networks
1000
500 ?
City States
Commerce
Expansion
Terracing
200
Olives
100
?
BRONZE
MIDDLE
EB
SLAVIC
EARLY BRONZE HELLENISTIC
ROMAN
LATE MEDIEVAL
NEOLITHIC ?
GEO
LATE
BRONZE METRIC A C
TO MIDDLE-
I & II III ROMAN TO MODERN
6000 BCE 5000 4000 3000 2000 1000 0 1000 CE
VEGETATION & LAND USE, ARGIVE PLAIN

WOODLAND FOREST WOODLAND MAQUIS PHRYGANA MAQUIS
DECIDUOUS OAK AGROPASTORALISM MED. POLYCULTURE AGROPASTORALISM MP
5200 3500 1600 300 650 1830

SOIL EROSION
Sheet Erosion Gullying Local
Soil
Soil
Soil
(Southern Argolid)
Debris Flows
?
Flood Debris Gullies
Silts Flows
Aluvial Strong Floods
? ? ?
Soil
Fans
? Alluviation
? Colluvium Colluvium
Alluviation
(Argive Plain)
Sheet Erosion
Soil
(Messenia) ? ? Flood Silts

?
Debris Flows
Strong
Floods Lateral Braided
Soil
(Olympia)
Channel Colluvial Fans Channel
Gravels
6.3. Settlement history, land-use change, and soil erosion in the Argolid and other locations of the Peloponnese.
(By permission of Elsevier, Butzer, K. W. (2005), Environmental history in the Mediterranean world: Cross-
disciplinary investigation of cause-and-effect for degradation and soil erosion. Journal of Archaeological Science
32(12), fig. 2.)
this does not happen, they may well collapse, and the soil stored behind
them will be removed. If an area is depopulated, and terraces are left for
some time, it is quite likely that naturally occurring maquis vegetation
will colonise and prevent erosion. It is temporary, or partial, abandon-
ment that subjects the soil system to the greatest risk, especially if the
terraced slopes are occasionally exploited as pasture. One correlation
that is apparent when looking at Figure 6.3 is the evidence for collu-
viation that occurs after peaks in settlement activity. Therefore, in many
instances, environmental resilience in a cultural–ecological system might

be the norm until the economic system collapses. As it stands, there is no
evidence for erosion having been the cause of an economic downturn.
Therefore, mitigation, including terracing along with the management
of fields, may well have been quite successful.
The creation of fields is an integral element in the creation of artificial
niches for food production. Their management is particularly important
within landscapes that are susceptible to relatively minor changes in cli-
mate, and leaving fields fallow is an obvious soil conservation technique.
If, for some reason, it becomes difficult to maintain adequate periods
of fallow (due to increased food requirements, perhaps through crop
failure in preceding years), then the probability of erosion occurring
may increase. As with terraces, the presence of field systems immediately
implies a series of management and political/cultural processes within a
landscape.
Whilst we know that Bronze Age field systems and terraces existed in
different forms across Europe, research on Iron Age systems across the
Mediterranean has suffered to a certain extent because of an emphasis
on the study of Roman field systems (centuriation), and the fact that
many Iron Age systems were undoubtedly masked by this very process
(Buchsenschutz 2004). There is, for example, inconclusive evidence for
land division in Provence from the fourth and third centuries BC onwards
(Boissinot 2000: 29). Clearly, with time, the evidence for complex field
systems and ownership is more abundant.
There is no doubt that the utility of manuring was understood by most
Mediterranean societies. There is evidence for knowledge of manuring,
crop rotation and tillage in classical Greek societies (Burford 1993: ch.
3). Whereas the management of fields includes rotation systems, manur-
ing, and possibly even the creation of boundary structures that reduce
erosion, fields can also contribute to environmental problems, especially
when they are applied uniformly to the landscape at the behest of a
centralised authority. In classical Greece, many tenancies, both public
and sacred (plots ‘owned’ by the Gods), were subject to strict control
regarding farming methods, crops grown, and the exact timing of when
certain activities should be undertaken (ibid.: 23). The presence of such
systems across parts of the Mediterranean during different periods rep-
resents forms of environmental management that might have prevented
flexible responses to environmental problems. In the Roman world, as
nvironments 181
the Late Empire increased state involvement in rural production, there

was probably an increase in land division, thus demonstrating the criti-
cal link between land demarcation and economic control (Leveau et al.
1993). Some have argued that, at its height, the increased allocation of
parcels of land to centuriation during the first to third centuries AD did
engender severe soil loss (Marchetti 2002). Despite the importance of
centralised land allocation and large latifundia, smallholding – or sub-
sistence farming – did exist (Frayn 1979), and there is no doubt that
farmers were able to respond to environmental problems in an effective
manner.
The Role of Fire

Forest and brush fires are a common occurrence across the Mediterranean,
and even take place in southern alpine valleys during especially dry sum-
mers. Whereas fires have immediate and dramatic consequences for the
populations that live in the affected area, we should not see these events
as inherently destructive. Rather, they may be part of the normal envir
onmental round or anthropic actions that have a clearly defined and
rational purpose vis-à-vis environmental management.
Palynological studies often include the analysis of micro-charcoal, and
unsurprisingly, there is often some debate as to whether the charcoal
represents natural fires or anthropogenic burning. Fire is, however, the
most basic and common management tool in Mediterranean manipula-
tion of the forest (di Pasquale, di Martino, & Mazzoleni 2004: 18). One
of the biggest problems in palaeoecology is differentiating natural fires
from those started by people as part of a landscape-management strategy
(or through accidents and acts of arson).
During the Early Holocene, evidence from the Eastern Mediterranean
(in particular, south-west Asia) suggests that (natural) summer fires actu-
ally delayed the colonisation of woodland by two or three millennia
(Turner et al. 2010). Whilst, during the Neolithic, it is possible that people
started to employ fire as part of landscape management and maintenance
of agricultural zones (Roberts 2002). However, we should not assume
that the control of Mediterranean vegetation was always uncomplicated.
Some Mediterranean vegetation is partially resistant to fire. Many spe-
cies will re-establish themselves quickly after a fire has taken place. These
include Q. coccifera (prickly or kermes oak), Q. suber (cork oak), and
Olea (wild olive) (Grove & Rackham 2001: 48–9). The deciduous oaks
(including Q. pyrenaica, Q. cerris, Q. pubescens, and Q. brachyphylla) are

also quite resistant to fire (Grove & Rackham 2001: 54).
During the Middle Holocene ‘Thermal Maximum’ between 5500
and 2500 BC, charcoal records indicate there was an overall increase
in fires in the Northern Mediterranean. Between 3500 and 3000 BC,
fire activity abruptly changed across the Mediterranean, as fire decreased
in northern and eastern regions (because of cooler wetter summers),
it progressively increased in southern regions. These southern regions,
experiencing drier summers at this time, would not have been affected by
increased annual precipitation (Vannière et al. 2011). During the period
c. 4500–3000 BC, there was an increase in fire activity in some parts of
the Mediterranean (Magny et al. 2011; Vannière et al. 2011). In cer-
tain zones, the transition to a more arid environment during the Middle
Holocene saw reductions in woodland vegetation and potentially com-
bustible material. In some areas, it seems that more fires occurred during
the relatively wet climatic optimum. In certain arid regions of Spain, such
as the Ebro Valley or Almeria, the Holocene optimum saw the expansion
of scrub and pine forests, thus providing more fuel. Prior to 7500 BC,
and after 2500 Cal BC, it has been argued that the amount of biomass
was below the required threshold for extensive and regular fires (Riera
Mora et al. 2004).
On the Barcelona plain, a number of fires took place near the coast
between 6520 and 6400 BC, followed by increases in the pollen record
in Cistus (rockrose family), Buxus, Plantago lanceolata (plantain) type,
and Asteraceae (daisy/sunflower family). Similar processes also took
place to the south of the River Llobregat between 7850 and 6200 BC.
Later on, a possible phase with frequent fires occurred on the Castellón
coast, where Plantago sp. increased between 4850 and 3980 BC.
Between 4350 and 3775 BC, Ericaceae and Pistacia lentiscus increased.
Such a sequence, with the development of the vegetation outlined here,
strongly suggests that these fires may well have been part of a manage-
ment system, especially as high percentages of Cerealia type and Vitis
also appear. The radical change in economic strategies that occurred
around 3775 BC, when fire use was patchy, can be explained by the
development of a more sedentary lifestyle adopted by groups during
the Middle Neolithic (Riera Mora et al. 2004). Other fire events during
the Middle Holocene had obvious impacts on the forest, and the last
of these fires (c. 2000 BC) was followed by a phase of ‘instability’, and
nvironments 183
the first clear evidence for anthropogenic impact in the area (Stevenson
2000: 607).
Despite the evidence for biomass availability being the fundamental
control on fire occurrence in arid environments and the consequent rela-
tively low frequency of fires during the Later Neolithic and Bronze Age,
there are of course exceptions to this trend. The pollen site at Sierra de
Gádor in southern Spain (1530 m) is a rare example of a pollen record
from a truly Mediterranean arid zone (Carrión et al. 2003: 839). A series
of major fires had a significant effect on the vegetation from c. 2200 BC.
Bearing in mind the altitude of this site, the increase in fire activity may
represent Bronze Age expansion into a new upland zone that had not
been cleared prior to the Argaric period.
Each vegetation-burning event has to be studied in context, and a
clear assessment of natural or anthropogenic origins has to be made. Fire
has always been an essential tool in humankind’s suite of environmental
management options, and should be studied as the archetypal form of
hybrid cultural–ecological practice – one which will be considered again
in Chapter 8 as part of the discussion on Mediterranean mountains.
Agricultural and Productive Vegetation

More than most other regions around the world, the Mediterranean
agricultural landscape is defined by its arboriculture, in particular the
ubiquitous vines and olive trees, with stands of other tree crops such
as pistachio, fig, and other fruits that inhabit the terraced landscapes.
The range of crops grown across the Mediterranean has varied through
time, and is the subject of books on agriculture and archaeobotany (e.g.
Zohary & Hopf 2001). We know, for example, that Roman agriculture
exploited a wide range of crops and trees well beyond the renowned
Mediterranean triad of wheat, olives, and vines (Leveau et al. 1993).
The complexity of knowledge associated with olive propagation, from
the use of cuttings and planted ovules to the range of ground preparation
techniques, is a research subject in itself, and Foxhall’s (2007) definitive
analysis of olive cultivation is essential reading. Despite the emphasis on
the olive, this tree may well have been grown with other tree crops, such
as almonds. The combination of relatively fast-growing crops with olive,
a tree that requires longer cultivation, served as a form of risk buffering,
with the fast-growing tree crops being removed once the olive trees pro-
duced their full potential yield (ibid.: 113). Successful olive production
also involves complex soil-management techniques, including trenching

and a continuous maintenance of the soil system (ibid.: 124).
The identification of the moment when these species were first domes-
ticated is one of the significant questions in Mediterranean palaeoecol-
ogy, and some work has seen contributions from genetics (Guillaume
et al. 2011), with one study demonstrating that genuinely wild olive
varieties still exist in a number of Mediterranean forests (Lumaret &
Ouazzani 2001). One way of assessing evidence for human cultivation of
olive is the identification of olive pollen values that are higher than ‘natu-
ral’ values, that is, where pollen levels indicate that olive trees constitute a
higher than ‘natural’ proportion of the local vegetation. In the Southern
Levant, olive pollen levels fluctuated throughout the Holocene, with
peaks achieving levels higher than expected during a number of peri-
ods, but in particular during the Hellenistic–Roman–Byzantine periods
(Baruch 1999: 20). Prior to this, Neolithic evidence for olive exploi-
tation comes from the now-submerged Kfar Samir site to the south of
Haifa, where a date in the fifth millennium BC is given. Here, archaeo-
botanical evidence implies the large-scale use of olives. Intensive olive
use is also implied archaeobotanically for the Bronze Age, although there
is no direct archaeological evidence for intensive olive exploitation at this
time (ibid.: 21).
Evidence from the Eastern Pyrenees for the selective exploitation and
management of olive trees reveals that, from the Late Neolithic to the
Late Bronze Age, only younger branches were cut. Such a practice might
imply that olive-tree management started during the Late Neolithic.
Increases in these thermophilous tree crops such as Olea europaea and
Pistacia lentiscus may have been a result of anthropogenic manipulation
rather than an increase in temperature (Heinz et al. 2004: 625). Increases
in olive are seen during the proto-historic periods in eastern Cossetània
(Spain), this being accompanied by a concomitant decrease in Quercus
sp. (Burjachs & Schulte 2003).
The combination of botanical remains (seeds and wood) of vine,
along with the archaeological evidence for the technological apparatus
required for grape pressing and suitable fields for vine growing, allows
the unequivocal identification of certain areas as wine-producing zones.
Whilst the classical periods witnessed the development of intensive and
extensive wine production (Brun 2004), increasingly we see evidence for
prehistoric viniculture, with much of the evidence generated by traditional
nvironments 185
archaeological as well as bioarchaeological research (McGovern 2003;

Schlumbaum, Tensen, & Jaenicke-Després 2008). Research into the
DNA of Vitis vinifera suggests domestication c. 8,000 years ago (Arroyo-
García et al. 2006). However, incontrovertible evidence for landscape
management associated with wine production has a much later date. For
example, vineyards of some description, as represented by vine ditches,
existed in the Marseille area from the fourth century BC (Boissinot 2001)
(Fig. 6.4), and evidence from the Gallo-Roman site Gasquinoy (Béziers,
Hérault) in the south of France comprises vine plantation marks delim-
ited by drainage ditches. Perhaps one of the most interesting aspects
of this work is the study of plant macrofossils, and the evidence for a
complex mosaic landscape around the actual site. The following environ-
ments or vegetation zones were identified: cereal production areas, cul-
tivated trees, forest edges, dry areas, grasslands/meadows, and wetlands
(Figueiral et al. 2010: 144).
Other than vine and olive, there are other species that emerged as
significant Mediterranean tree crops, in particular nuts such as Castanea
(sweet chestnut) and Juglans (walnut). Whilst Castanea makes an early
(c. late eighth millennium BC) appearance in some pollen diagrams, such
as that from the Kleonai core on the Peloponnese (collected as part of
the Nemea Valley Archaeological Project) (Wright et al. 1990: 593),
there is no clear evidence for its intensive exploitation until much later.
Chestnut trees, which are useful for both their nuts and timber, have
quite stringent moisture requirements.
The development of landscapes with patches of managed tree crops
occurs across the Mediterranean, with many zones seeing an intensifica-
tion of arboriculture during the Iron Age and Roman period. Some spe-
cies, such as Juglans, are even considered by palynologists as chronological
markers in this instance, marking the impact of Rome on the landscape.
At the central Italian pollen sites of Lago Albano and Lago di Nemi,
there is evidence for Iron Age Cannabis (hemp), followed by increases in
Castanea, Juglans, and Olea (Mercuri et al. 2002). The introduction and
exploitation of Castanea sativa is considered to be a very Roman affair,
although it is likely that the intensive propagation of sweet chestnut was
not that common (Conedera et al. 2004). Castanea was also naturally
present in parts of the Mediterranean. Palynological evidence from the
Lozoya Valley, Sierra de Guadarrama (1,113 m), central Spain, aimed
to answer the question as to whether Castanea was introduced by the
6.4. Traces of vine plantations. Top: Saint Jean du désert; bottom: Les Girardes Lapalud (photos: Philippe
Boissinot).
nvironments 187
Romans or was in fact native (Franco Mugica, Garcia Anton, & Sainz
1998: 70). Castanea and Olea appear early in the Holocene within a
landscape dominated by Pinus. It is likely that an opportunistic use was
made of chestnut, but intensified exploitation did then develop in this
area during the Roman period.
Come the Greek and Roman periods, we have to imagine the devel-
opment of a complex mosaic of plants, and geomorphic processes, per-
haps even more heterogeneous than today. We should not forget how,
in recent years, EU quotas and financial support have resulted in the
relative homogenisation of certain productive strategies, and this, com-
bined with relatively low rural populations, means that many areas of the
modern Mediterranean countryside are once again covered in woodland
(Obando 2002).
Olive production, more than any other agricultural activity, gave
rise to the creation (albeit relatively temporary) of intensively managed
agricultural systems in landscapes that had otherwise seen relatively
low levels of agricultural production. The infamous incorporation
of North African semi-arid and arid landscapes into the Roman pro-
ductive zone is perhaps the best example of the successful applica-
tion of environmental knowledge by a hegemonic political structure.
However, it is likely that this knowledge was founded on an adapta-
tion of indigenous practices, which certainly seems to be the case in
Tripolitania (Mattingly 1996). There is little doubt that the indige-
nous Garamantes exploited the hydrological sources along cliff base
lines in the area (Leveau 2009). Despite variations in the climate, it
is likely that for about 3,000 years, the Libyan pre-desert climate was
relatively stable, with some phases of increased precipitation. One
of these phases was contemporary with the Roman Empire. At the
Wadi Tanezzuft, a period of increased precipitation rendered the area
economically viable during the Roman period (Cremaschi 2003). At
a broader regional level, evidence from the mid-Medjerda flood plain
(northern Tunisia) indicates the stabilisation of forest cover during
the Roman occupation, a period preceded by phases of environmental
instability (Zielhofer & Faust 2003: 211). The Roman environmental
optimum in this area may have been characterised by the develop-
ment of more humid conditions. This may have supported vegeta-
tion persistence, as seen in the Segermes valley (north-east Tunisia),
where Roman impact on the vegetation was no more intense than in
preceding periods. The most important change was the introduction

of olive production (Kolstrup 1994).
Responses to climate change (and day-to-day environmental processes)
are influenced by the politics pursued by ruling elites (see McIntosh,
Tainter, & McIntosh 2000; Rosen 2007: 2). Classical period exploitation
of desert areas occurred during a period of improved climatic conditions,
which may have included some phases of greater precipitation (Rosen
2007: 3). Although we cannot generalise and state that this period was
noticeably wetter than the preceding periods, some minor changes in
rainfall patterns may have been enough to allow movement into arid
zones – although it was the political and economic will to exploit those
areas that was of paramount importance. Technological innovation (or
adaptation of existing practices) appears to have helped render parts of
this region profitable. Romano-Libyan farmers practised flood-water
farming, building walls to divert surface run-off after seasonal storms
(Gilbertson, Hunt, & Smithson 1996). Cereals were grown, including
six-row hulled barley, durum or hard wheat, and bread wheat, plus len-
tils, peas and grass peas, olive, linseed, vine, almonds, and peaches (Van
Der Veen 1996). At Khirbet Faynan (southern Jordan), olives and vines
did grow in the area, and charcoal from mining areas show how Roman
miners had to bring in wood from the plateau – a different experience
to that of Nabataean people who were using local wood (Barker 2002;
Engel 1993).
During the first to the fifth centuries AD, flood-water farming in cer-
tain pre-desert zones was probably established by Rome within a climatic
context perceived as being stable, and identified as suitable for agricultural
production given the right technological and management structures.
Eventually, landscape degradation led to a loss of efficiency in this flood-
water farming system, and pollution may have become a serious problem
(Barker 2002). One possible effect of intense agriculture and grazing in
an arid zone is an increase in dust storms (Gilbertson 1996: 299). Also,
after some time, soil exhaustion as a result of short fallow periods may
have posed problems for pre-desert farming systems (ibid.: 304–5). Soil
erosion in arid environments is potentially disastrous. Gullying, in par-
ticular, is a problem on wadi floors, and there is no doubt that periods
of soil loss did occur during antiquity. Gilbertson (1996: 314) suggests
that neglect and inadequate maintenance of landscape management fea-
tures, rather than the intensification of agriculture, may well have led to
nvironments 189
increased erosion. It could be that erosion was restricted because of an

emphasis on tree crops, and this may have helped maintain a relatively
stable landscape for a certain time (Barker et al. 1996: 278). However,
evidence for interbedded aeolian sands and flood-loam deposits do indi-
cate important, perhaps short-lived, changes in the wadi environment
during the Romano-Libyan and later periods (Anketell et al. 1995: 239).
Consequently, whilst Rome managed the successful optimisation of the
environmental potential of these landscapes, there would undoubtedly
have been years, even decades, when environmental conditions were dif-
ficult. In particular, annual rainfall would have fluctuated and inhibited
production. Despite the potential for erosion, and even periods of crop
failure or reduced yields, there is no doubt that this arid landscape was
productive, and represents the successful application of environmental
knowledge and concomitant technologies within such areas. However,
we should not lose sight of the potential hardship for those who worked
in these landscapes. There is evidence for an awareness of the impor-
tance of conservation strategies amongst Romano-Libyan desert farmers
in Tripolitania. Sluices were blocked, possibly to restrict flood-water flow
into the fields (Barker 2002: 496). Despite the evidence for regional dry-
ing towards the end of the Roman period in the Eastern Mediterranean
(Orland et al. 2009), and probably the South as well, the demise of this
successful agricultural exploitation cannot be put down to environmental
change (Gilbertson 1996). The contraction of the Roman Empire and
its markets, and the possible loss of pertinent environmental knowledge,
may well have contributed to the eventual decline of this productive
system.
Woodland and Landscape Management (Dehesa and

Other Systems)
The ‘Fall from Eden’ discourse is partly founded on the notion of
humanity’s removal and mismanagement of Mediterranean vegetation.
Whilst there is no doubt that people have cleared vast swathes of wood-
land and forest, Mediterranean vegetation has often demonstrated nat-
ural resilience, secondary woodland recolonising areas with surprising
ease. There is a long history of woodland management, with economic
needs balanced with specific woodland ecologies. A complex multi-proxy
study (using pollen, charcoal, phytolith, and sedimentary analyses) of the
Middle Neolithic landscape in the Middle Rhone Valley demonstrates
the emergence of a mosaic of agro-sylvo-pastoral management that

allowed the emergence of grasslands under relatively open woodland of
deciduous oaks. This configuration seems similar to the dehesa system
(discussed below), although it appears that this landscape was sensitive
to either environmental change or cultural change (or a combination of
the two), as it had disappeared by the end of the fourth millennium BC
(Delhon, Thiébault, & Berger 2008).
There are some parts of the Mediterranean where there is no doubt
that woodland never managed to recover from early human impact.
For example, in Lebanon, cedar was present during the first part of the
Holocene. However, by 5900 BC, these trees were subject to clearance,
and by 3700 BC, much of the forest had totally disappeared as olive
groves expanded. In the epic of Gilgamesh (written some 4,600 years
ago), Lebanese cedar is mentioned, as the King of Uruk required timber
from this tree in order to construct his city. However, if the King had
come to this area in search of cedar, he would not have found any (Yasuda,
Kitagawa, & Nakagawa 2000: 133). Levantine peoples were obliged to
trade for timber during the Bronze Age, with societies in this region
looking to maritime trade as the only means for economic advancement
(Akar 2009). Moving into the Iron Age, we see how Phoenicia, delim-
ited and protected to the east by the Lebanese mountains, lacked access
to an extensive range of arable land and woodland resources. The nar-
row band of land suitable for agriculture between the mountains and the
coast meant that Phoenicians looked to the sea for economic expansion
and the supply of essential materials (Aubet & Turton 2001).
The notion that early farmers lived in harmony with the environment
and cared for the natural world (Hughes 2005: 21) requires careful
assessment. We could consider that if human impact does not change the
ecology of a certain area to the extent that any reversion to a previous
state becomes impossible, then a form of metastability is possible and
quite common in many landscapes. As presented in the previous chap-
ter, there is limited evidence for extensive impact on woodland and soils
during the Early Neolithic. We can identify some human impact on the
vegetation, especially from the Late Neolithic onwards (Reille & Pons
1992; Vernet 1997), and some have suggested that shortages of tim-
ber affected societies as far back as the Bronze Age. As Meiggs (1982:
98) notes, Arthur Evans felt that the Minoans suffered a shortage of tim-
ber, not being able to fulfil their requirements from within Crete. Evans
nvironments 191
had considered that one of the causes of the downfall of the Minoans was
the destruction of timber resources (Evans quoted in Meiggs 1982: 98).
This is unlikely, since there are no signs of timber shortage on some of
the palace sites (Meiggs 1982: 99).
Some have argued that human impact on vegetation only becomes
apparent during the Roman period, and that all preceding changes can
be explained by climate change (e.g. Magri 1995; Yll et al. 1997). The
project directed by Frenzel and Reisch (1994) provides one of the most
useful contributions to this debate. Arboreal pollen percentages for the
Roman period across Greece vary quite substantially. Some have argued
that the proportion of open landscape in southern lowland Greece was
very high and comparable to recent levels before reafforestation com-
menced (Bintliff 1993). Others have suggested that the notion that low-
altitude sites would have been subject to intense deforestation whilst
higher altitudes were relatively untouched cannot be demonstrated:
There are low-altitude and high-altitude sites with high arboreal pollen
percentages, and some high-altitude sites with low arboreal pollen per-
centages (Bottema 1994a: 70–1).
Pliny the Elder stated that, during the fourth century BC, the beech
tree grew towards sea level along the Tiber. However, during his life-
time (first century AD), the beech was considered to be a mountain
tree (Reale & Dirmeyer 2000: 167). This type of documentary evi-
dence, in combination with archaeological site evidence (the presence
of large settlements in now semi-arid, arid, or desert-like environments)
and vegetation models based on palynological evidence from around the
Mediterranean, demonstrates a substantial difference between the den-
sity and distribution of modern-day vegetation compared with that from
2,000 years ago (Reale & Dirmeyer 2000; Reale & Shukla 2000). At
Sagalassos in the western Taurus mountains in Turkey (situated between
1,490 and 1,600 m), there is little evidence that the Roman forest was
depleted. It seems that this area was comprised of woods and open areas
(Waelkens et al. 1999: 702). The charcoal evidence shows that most of
the wood employed in smelting came from pine and juniper, with some
oak also being employed (Schoch 1995 cited in Waelkens et al. 1999).
Moreover, the presence of olive wood in these assemblages and stone
weights that may well have been parts of olive presses, combined with
evidence for slightly warmer winters during the Graeco-Roman period,
implies that olive was probably grown in the vicinity, whereas today it
cannot grow (Waelkens et al. 1999: 705). The development of today’s

open landscape did not actually occur until after the Roman period.
The primordial use of wood was for making fire. The use of fire
ranged from the obvious functional roles of providing heat for people
and for cooking through to warding off wild animals and producing aes-
thetic experiences that would have had a range of social and ritual func-
tions. Wood for fire would normally have been gathered from the forest
floor rather than procured through the wholesale felling or destruction
of trees; the gathering of wood was probably a task carried out as part
of other agricultural activities, with wood collected from extended geo-
graphical areas. This is implied in the charcoal assemblages that often
comprise a wide range of species present across a set of varied niches that
must have been visited during the collection of wood. This was especially
true for the Roman period, when timber and wood exploitation was
intensive. In some areas, such as the plain around Lattara (the Iron Age
and Roman settlement near modern Montpellier), we see an absence of
species from the wider landscape, implying that, by the end of the Iron
Age, this area was completely opened up for agriculture (Chabal 1998:
134–5). Charcoal data do not merely confirm or complement palynolog-
ical histories, but they provide specific information on the species of tree
exploited by people in the past. The charcoal evidence from the Middle
Neolithic to the Late Bronze Age Montou site (270 m above sea level)
shows that firewood was collected from upland areas as well close to the
site. It is also possible that this wood, comprising Pinus sylvestris type
(Scots pine) and Abies alba (silver fir), could have floated downriver and
thus represents opportunistic exploitation (Heinz et al. 2004: 624).
Flexible forms of environmental knowledge allow people to practice
the exploitation of natural resources without causing degradation and
instability. Woodland management, and in particular the dehesa system
(and regional variations of that system), are one example of such a prac-
tice. One of the most important management strategies within woodland
would have been the creation of openings for grazing for both wild and
domesticated animals. Grazing animals have had to shoulder the blame
for the ‘degraded’ Mediterranean environment (more specifically, vege-
tation) more than any other agent. In some landscapes, there is evidence
for severe Roman impact on local ecologies through livestock pastur-
ing. For example, focussed ecological research on arid grasslands dem-
onstrates the importance of the Roman ‘footprint’ upon certain types of
nvironments 193

6.5. Roman sheep enclosure on the Crau (Provence): The site of Négreiron 6 (photo:
Gaëtan Congès). Note the impoverished vegetation.
vegetation. The Crau, an extensive semi-arid plain in the south of France,

witnessed intensive pastoral activity during the Roman period (Fig. 6.5).
The soil seedbank (the viable stored seeds within a specific soil unit) in
the areas exploited during the Roman period show how these grasslands
have failed to recover, both in terms of plant species richness and abun-
dance, as well as in terms of soil quality (F. Henry 2009).
Dehesas are perhaps the best example of a persistent landscape type
that demonstrates successful management. These areas are open park-
lands in which the woodland has been managed with a view to pro-
viding a whole series of resources: fuel, fruits, cork, wood, and pasture
(Fig. 6.6). They are predominantly found in the south and south-west
of Portugal and Spain. If these areas are overgrazed, they often develop
into scrub. It is argued that such a system of dehesa would have existed

6.6. Dehesas in the distance with pastoral structures in the foreground (photo: Santiago Riera-Mora).
during the second millennium. This is reflected in the palynological evi-

dence. Harrison (1994: 85) considers that Mediterranean Spain did not
adopt the ‘typical’ Mediterranean triad of wheat, olives, and vines due
to the threat of drought. Instead, a system which made efficient use of
the secondary products from animals provided the foundations for the
economy in this part of the Mediterranean. Palynological work from El
Acebrón and Las Madres in south-west Spain demonstrates a pattern of
vegetation change that is quite different from the so-called natural suc-
cession that one would expect in this area, with the first simple dehesas
appearing during the Copper and Bronze Ages. The first phase (4000–
2500 BC) from the pollen diagrams from Las Madres indicates the pres-
ence of large amounts of Vitis and low values for trees. Such a signal may
represent viticulture or the management of wild vine in order to collect
the autumn fruits. At the same time, there were increases in Quercus
along with Plantago, Anthemis type (dog fennel chamomile), Rumex type
(docks), and Artemisia (wormwoods). The second phase (2500–1600
BC) saw the re-emergence of oak and pine forest with some high values
for ruderal indicators, and this may be the first phase of true oak dehesa
Environmental Change and Social Geoarchaeology 195
(Stevenson & Harrison 1992: 241–2). This is contemporary with the

climax of the Millaran societies. Phase III (1600 BC–AD 500) saw the
destruction of the forest cover in the Las Madres area. Despite the fact
that this destruction is put down to anthropogenic activities, there is little
evidence for settlement before the end of this phase. Olive pollen appears
in the Las Madres core for the first time during this phase. During phase
IV (500 BC–1200 AD), there is a clear peak in oak values along with
pollen from ruderal indicators, a combination typical of a classic dehesa.
The dehesa system was ‘fully defined’ by 500 BC, and is thus considered
a Phoenician product (ibid.: 243). Research demonstrates that dehesas
actually mimic ecosystems, in that these human-made agro-ecosystems
appear stable and have certainly been resilient whilst providing a range
of useful resources (Joffre, Rambal, & Ratte 1999). Dehesa-montado
systems have existed for many centuries. Humans manage these hybrid
systems, enhancing the available ecological characteristics in the produc-
tion of a particularly successful socio-environmental niche (Pereira & da
Fonseca 2003). The essential element in the notion of hybridity is that
all manifestations of activity in the landscape, whether terracing or forest
management, are the product of a complex intersection of a range of cul-
tural processes, ranging from the economically rational through to ritual
(Latour 1997; Whatmore 2002).
Environmental management often manifests itself through forms of
proscription. We know that some woodlands were sacred. One such possi-
ble sacred ‘deciduous-oakery’ – Skotitas – was named after Zeus Skotitas.
This was situated on the borders of Lakonia, and still exists as a substan-
tial coppice (Rackham 2001: 16). There is little doubt that certain places
at certain points in the past have witnessed vegetation exhaustion, but
relatively few palaeoecological studies imply long-term, extensive vegeta-
tion loss. There are many examples of persistent woodland, and whilst
the dehesa system might represent an unusually successful example of
sustainable woodland management, it is quite likely that similar forms
of environmental knowledge have been successfully applied across other
parts of the Mediterranean during the Middle to Late Holocene.
Environmental Change and Social Geoarchaeology
In order to develop some of the notions and themes presented in this and
the preceding chapter, two specific examples of landscape archaeological

6.7. The Sainte Victoire landscape with the c. 1,000 m high mountain ridge (running west–east) and the plain
in front (to the south of the mountain) (photo: author).
research and environmental reconstruction work are considered below.

The aim is to provide an example of how we might make inferences in
situations where archaeological and environmental evidence intersect.
The Sainte Victoire: Changing Patterns of Interaction

with Environment
One example of a notionally ‘degraded’ (comprised of steep slopes and
relatively poor soils) Mediterranean landscape is the Sainte Victoire mas-
sif (south-east France). This area was the object of detailed landscape
survey and excavations, designed to investigate the Iron Age and Roman
settlement and economy of this area, as well as the changing nature of
human engagements with landscape processes, including hydrology and
erosion (Walsh & Mocci 2003). This landscape comprises a 1,000 m
high mountain that dominates a relatively flat plain to the south, and
undulating topography to the north, with steep, heavily eroded zones
and a more stable, agriculturally ‘rich’ plain to the south (Fig. 6.7). This
landscape witnessed a waxing and waning of activity during the Iron Age
and Roman periods.
The Early Iron Age is represented by only six sites over an 8,500 ha
area: four oppida and two low-lying sites. It appears that, during the
Early Iron Age, the Sainte Victoire was not considered as economically
valuable vis-à-vis neighbouring areas. Other areas in eastern Provence
appear to have been relatively highly populated, whilst the Sainte Victoire
was neglected (Bérato 1995; Trément 1993; Walsh & Mocci 2003). The
perception of peoples from beyond the Sainte Victoire may well have
been one that saw this micro-region as too risky and difficult to man-
age, and this might explain the relatively low levels of settlement during
this period. The Middle and Late Iron Ages (which correspond to La
Tène I, II, and III) saw an important increase in settlement activity. Fifty
sites were active across the late third and second centuries BC, and a
recognisable settlement hierarchy emerged, with 10 oppida, at least 35
minor sites, as well as 30 indications of other smaller sites (Fig: 6.8).
The Roman period saw no settlement on the mountain itself, with
sites now located at the foot of the mountain and across the plain. Many
of these low-lying sites were often established on or close to their Late
Iron Age precursors. The first villas were located adjacent to low alluvial
terraces with direct access to the best agricultural land. During the first
and second centuries AD, there was a gradual intensification of settle-
ment on the massif, with a total of 35 sites dated to this period. The
third century AD saw a decline in activity, followed by a re-emergence of
many settlements during the fourth century. The final decline then took
hold during the sixth century, and this situation continued into the early
medieval period.
The combination of landscape survey and excavation of specific sites
permits the assessment of two spatial scales – that of the wider land-
scape and changes in engagements with topography and location vis-à-
vis resources, and then the scale of the site. Here, we can make inferences
relating to specific responses to and engagements with localised environ-
mental processes.
Excavations on the Domaine Richeaume, at the foot of the Cengle
(a limestone bar that delimits the southern edge of the Sainte Victoire
mountain), provided stratigraphic and geoarchaeological detail for
this landscape research. The excavation of one Early Iron Age site
(Richeaume III) and a Roman villa (Richeaume I) allowed the compar-
ison of two decidedly different sites, with very different environmental
characteristics.
N
Le Grand Sambuc
MONTAGNE DES UBACS
500
VAUVENARGUES
Col des Portes
Les Bonfillons
ST-MARC
JAUMEGARDE
Le Pic des Mouches

MONTAGNE STE-VICTOIRE
Le Pas du Berger
Les Blaireaux
1000 500
Le Mitronet
L'Infernet Bramefan
Baume Vaoutade
Baume Vaoutade Roque Vaoutade PUYLOUBIER
Le Bayon Clément
LE THOLONET
Parret
L'Avocat
Calandre
ST-ANTONIN
/BAYON
500 Richeaume III - IV

BEAURECUEIL
Les Masques
Prébois La Tour 1
BARRE DU CENGLE Le Pas de Magnan
300
CHATEAUNEUF-
0 1 2 km
LE-ROUGE
M. Borély - V. Dumas - F. Mocci (CNRS-CCJ 1997)
Archaeological site distribution on the Sainte-Victoire: Late Iron Age
Major settlement Tumulus Minor upland settlement Minor settlement Cave site Isolated find or indication of site
MONTAGNE DES UBACS
500
VAUVENARGUES
Puits d'Auzon
ST-MARC
JAUMEGARDE
MONTAGNE STE-VICTOIRE
1000 500
L'Infernet
Baume Vaoutade PUYLOUBIER
LE THOLONET
Parret I-II
Le Général Calandre
L'Avocat
ST-ANTONIN
? Maurély /BAYON
500
BEAURECUEIL
? Bayle
Richeaume I
La Meironette
BARRE DU CENGLE St Pancrace
300
Le Jasmin
CHATEAUNEUF-
0 1 2 km
LE-ROUGE
M. Borély - V. Dumas - F. Mocci (CNRS-CCJ 1997)
Archaeological site distribution on the Sainte-V ictoire: Romano-gallic sites (Ist-IIIrd c AD)
Villa Minor settlement Pottery production Oil production Burial site Isolated find or indication of site
6.8. The Sainte Victoire landscape with sites referred to in the text plotted. Note the changes in settle-
ment density and location across the Iron Age and Roman periods (figure: F. Mocci, V. Dumas, & K.
Walsh).
Richeaume III was located on a talus of eroding Pleistocene clays at

the foot of the Sainte Victoire. More than 500 sherds of hand-thrown
pottery, along with an example of an imported kylix from Marseille,
dated this site to the beginning of the sixth century BC. There was no
convincing evidence for any kind of structure on the site. An in-filled
palaeo-ravine contained sherds of Early Iron Age pottery at the base
of the fill. The basal layer of this ravine was the local red clay sub-
strate. The cutting and the filling of the ravine must have taken place
at the end of the Early Iron Age. This process might have taken place
once the site was abandoned, although it is possible that this erosion
phase forced people to abandon the site (Fig. 6.9). This provides an
example of the interface between human activity and a sedimentary
process, where we can demonstrate via (geo)archaeological stratigra-
phy that people must have had direct experience of the sedimentary
environment. At the regional level, there are a number of other sites
across Provence that also demonstrate phases of erosion during this
period (Provansal 1995b). The evidence from several sites in Provence
for an Early/Middle Iron Age phase of erosion is convincing, although
one cannot argue for exact synchronicity. Locally, just 2 km from
Richeaume, the Bramefan oppidum also saw a phase of erosion which
has been dated to the end of the first occupation phase (sixth century
BC) (Jorda & Mocci 1997). This was followed by a second occupation
phase dated to the La Tène III period (Bofinger, Schweizer, & Strobel
1996). Here, it seems that the erosion marks a hiatus between the two
phases of settlement, but not necessarily the reason for the end of the
first settlement.
As there were remarkably few Early Iron Age sites on the Sainte
Victoire, we can be quite sure that there was little pressure on land
because of a low population density. In such circumstances, even dra-
matic erosional events may not have been perceived as a problem. This
would be especially true for semi-sedentary populations who had yet to
establish permanent substantial settlements, which may well have been
the case in parts of Early Iron Age Provence (Garcia 2002: 90–2). In
addition, if an agricultural system allows the avoidance of the hazard
posed by degradation, that is, through mobility, then people may just
move to a new zone where erosion is not a problem. However, degrada-
tion does have an impact in stratified societies with rigid land-ownership
mechanisms that prevent low-status groups from taking up new land.

6.9. Richeaume III – An Iron Age site comprised of a dense pottery scatter located on an
eroding slope. It seems likely that the response to erosion in this landscape was a decision
to relocate close by (photo: author).
Consequently, the agency of mundane environmental processes is con-

tingent upon class and proprietorial institutions within a landscape.
The Roman villa (Richeaume I) is situated just 500 m to the south
of Richeaume III. However, its topographic and geomorphic situation
is quite different, located on a flat area, the eastern extremity of the
site abutting a stream terrace. The site comprises a range of typical villa
buildings, including the owners’ domestic areas, plus an ornamental gar-
den and some agricultural buildings. Here, we are more interested in the
zone abutting the stream where hydraulic structures and a palaeochannel
were discovered. One aqueduct transported water away from the site –
this element runs from north to south, and was built during the first
half of the second century AD. The structure turns east and would have
emptied into the stream channel (now an in-filled palaeochannel). The
agricultural buildings, located just to the west of the aqueduct junction,
were subjected to flooding, as alluvial sediments abut all of the buildings
in this area. Four distinct phases of channel activity were identified, cov-
ering the second through to sixth centuries AD. These phases reflect the
principal periods of flood-plain development and management. The first
phase of alluviation is dated by ceramic material that provides a terminus
ante quem of the fifth century AD. However, the date proposed for the
construction of the aqueduct demonstrates that the channel was active
from the second century AD. On or after this date, overbank deposits
were laid down either side of the channel and threatened some of the
villa structures. As a consequence, a dyke was constructed, undoubtedly
as a response to the flooding. The dyke is dated by a second-century,
upturned, votive, sealed jar containing burnt grain. The next phase com-
prised the cutting of the flood-plain silts by a new channel with almost
vertical sides. Coarse colluvial material entered this channel, probably
during the sixth or seventh centuries. These sediments were washed
down from the surrounding hill-slopes – areas that had undoubtedly
been managed (possibly terraced) during the Roman period. There is
other evidence for a late/post-Roman phase of erosion in the area. At
Roque Vaoutade, 2 km upstream from Richeaume, results from geo-
archaeological work also suggest a Late or post-Roman phase of sedi-
mentation; one radiocarbon-dated sequence yielded a date centred on
AD 566 (Ballais & Crambes 1993: 472). At this Roman site, erosion
contemporary with the villa’s use was managed via technological inter-
vention, seen in the construction of protective walls and the maintenance
of the aqueduct.
The Roman Watermill at Barbegal

Another example of how we might directly assess the interaction between
archaeological sites (more specifically, the people associated with the
site) and the geomorphic system comes from the Roman watermill at
Barbegal (in the Vallée des Baux, 7 km east of the city of Arles in west-
ern Provence) (Fig. 6.10). This mill is one of the most impressive pieces
of hydrological engineering in the Roman world. It was supplied with
water by an extensive aqueduct system that also supplied Arles. This
type of site undeniably represents a very different form of interaction
Wetland
Barbegal
Limestone ridge

6.10. Barbegal in its landscape setting. Left – Aerial view of Barbegal situated on a lime-
stone bar (photo: by permission of Centre Camille Jullian). Right – View of the Barbegal
mill from the south (photo: author).
with the landscape and natural process, as its function is almost entirely
economic.
The mill is located on the south-facing slope of a limestone bar (part
of the Alpilles) (Leveau 1995; Leveau et al. 2000). The wetland in
front of the mill was relatively dry during the Roman period (Bruneton
1999, 2000), probably managed and drained by Roman engineers.
Palynological research reveals the development of a typical Middle
Holocene, Mediterranean, mixed-oak forest in this area. Moving into
the Iron Age and Roman period, there was a notable increase in arable
farming, especially during the Roman period when we also see the pres-
ence of Juglans in the pollen diagram (Andrieu-Ponel et al. 2000).
Geoarchaeological work at the mill identified a series of 11 colluvial
units, the earliest of which were hydromorphic clays and the limestone
slope on which the Roman activities took place. The subsequent units
directly lain upon these clays were coarse and stony, and contained
Roman archaeological material (Walsh 2004). The inception of hill-slope
erosion is therefore dated to the post-Roman period. However, the hill
slope would have witnessed earlier phases of colluvial activity through-
out the Holocene. Consequently, the construction of the mill undoubt-
edly included the removal of pre-existing sediments in order to facilitate
construction. This removal of sediments extended down as far as the
hydromorphic clays that define the northern edge of the wetland, and
the limestone slope that dips down under these clays. Immediately to the
east of the mill, four Late Roman burials were discovered (the pottery
directly associated with one of these burials is dated between the end of
the third and the start of the fourth century AD). The trenches for these
burials were dug into the hydromorphic clays. The different events rep-
resented by the construction of the mill and by the burials demonstrate
that the area around the mill was managed, and any colluvial depos-
its present around the site were removed, or the slope was managed to
such an extent that the potential for colluviation was obviated during
the Roman Period, with colluviation starting again, and no longer sub-
ject to clearance, once the mill went out of use. A similar sequence was
also recorded at the Pont Simian Roman Bridge, less than 2 km from
Barbegal (Ballais & Bellamy 2000). At these sites, the geoarchaeological
work demonstrated that sediments were removed as part of construction
and maintenance practices.
Discussion: The Human Scale of Interaction with Past

Environmental Processes
At both Barbegal and Richeaume I, anthropogenic manipulation of
environmental features and processes is a critical issue. Anthropogenic
truncation of sedimentary records should not be seen as a problem but
rather a phenomenon with a cultural aspect that archaeologists should
attempt to understand and interpret. The colluvial facies at Barbegal
(and the nearby Pont Simian) reveal periods of direct intervention in
the landscape. Such endeavours were designed to facilitate engineer-
ing projects that would contribute to agricultural production, as well
as the supply of water to the city of Arles. This intervention in the
geomorphic system may seem banal, but it should be studied in the
same way as the architecture and statuary that represent economic and
political power within urban and rural imperial landscapes. The manip-
ulation of the environment is very much associated with a certain atti-
tude towards nature, and an appreciation of what can and cannot be
controlled within the natural world. We know that in Pliny the Elder’s
Natural History, ‘…man was at the centre of the story: nature had
made all things for him, and Pliny’s book was partly a survey of what
was available’ (R. French 1994: 207). Economically valuable landscapes
were controlled and managed. Barbegal and Richeaume I represent two
decidedly different types of site, but the geoarchaeological work in both
places reveals evidence for direct management of the landscape. When
considered in conjunction with other types of evidence for landscape
management (in particular, centuriation), this type of geoarchaeologi-
cal sequence is unsurprising. Such sequences, which often represent
phases of human truncation of the geoarchaeological record, may mean
that strata representing earlier erosional events were in fact removed
by Roman intervention and land-management practices. We need to
ask if these landscapes were characterised by geomorphological stabil-
ity, or whether the Roman management of the geosystem disguised or
masked some phases of erosional activity. Stability is often assigned to
periods where facies are either absent or comprised of fine sedimen-
tary material. This relative dearth of evidence for Roman erosion in
this region does not necessarily imply that erosion was not a problem.
As Beagon (1996: 293) demonstrates, Pliny observed the destructive
effect of certain types of activity on the landscape. Whilst one would
have to disagree with the general tone of Hughes’s (1994b) pessimistic
account of the impact of Greek and Roman civilisations on the natural

environment, his assessment of attitudes to the environment is never-
theless useful. It is apparent that classical philosophers questioned the
notion that the gods were active in every facet of environmental, and
in particular, agricultural processes. Consequently, we see the devel-
opment of an ethical system that allowed more pragmatic approaches
to the management of the environment and that involved direct, and
sometimes significant, intervention.
At Richeaume, we certainly see the intersection of pragmatic and spir-
itual engagements with the landscape. As well as the votive jar beneath
the dyke (designed to protect the site from flooding), the anthracologi-
cal evidence from recent excavations at the Richeaume Necropolis, 200
m to the east of the villa, suggests quite specific choices of wood for
cremation rites (Cenzon-Salvayre & Durand 2011). The anthracologi-
cal assemblages comprise Pinus halepensis (Aleppo pine) along with the
Q. coccifera and Q. ilex (evergreen oaks), as well as examples of vitis
and beech. It would seem that the Roman vegetation mosaic in this
area was not that different to that present today. Whilst it is unsurpris-
ing that the wood for cremation pyres was gathered from around the
immediate area, specific choices were made regarding the employment
of different species in the staged construction of the pyre. In one crema-
tion, two different layers of carbonised wood have been identified, the
bottom layer comprised almost entirely Q. coccifera-ilex, and the upper
layer, Pinus halepensis-pinea. When we look at the combined archaeolog-
ical, palaeoenvironmental data and assess topographical characteristics
on specific sites, we can start to evaluate the full range of instrumental
(technological) engagements with the landscape, as well as more spiri-
tual relationships with nature. This complex intersection of technological
(pragmatic) responses to landscape processes with spiritual notions reit-
erates the point that the range of societies that occupied that space over
time will have perceived the same landscape with its associated natural
processes differently.
One way to assess how perceptions of the geomorphic system have
evolved over time is to consider how notions of risk and hazard change
with specific spatial and chronological contexts, ideally through the dia-
chronic study of particular micro-regions. On the Sainte Victoire, the
change in settlement organisation (density and location of sites; Fig. 6.8)
between the Late Iron Age and Roman period would have comprised
substantial changes in the ways in which people engaged with the land-
scape (in terms of movement and work) across this incredibly varied
topography. Changes in settlement patterns and associated economic
practices expose sites and people to different environmental processes –
a site on a hill slope (or mountain side) would be subject to colluvial
processes (soil erosion through to rock slides); a site on the plain would
probably have increased interaction with alluvial processes (from the
deposition of useful silty sediments through to the dangers posed by
flooding). The possible flexibility of the Early Iron Age settlement sys-
tem on the Sainte Victoire demonstrates how mobility may well have
been the answer to erosional events. As we move into the Roman period,
we see how a society attempts to control the same landscape through
engineering and complex landscape management.
The apparent success of the Roman system in controlling erosion
belies the probability that small, independent peasant farmers and slaves
working on the villas or on processing-sites such as Barbegal would have
been obliged to deal directly with any hazards on behalf of the villa or
mill owners. Storm erosion and flood damage would have been repaired
by these workers, and even the mundane, annual, small-scale hazards
would have had a direct impact on their lives. Erosion is a process, an
agency that people have to engage with, and in the case of Barbegal, it
was not merely an inconvenience that had to be removed, but a phenom-
enon that impinged upon the workers who had to remove the sediments.
Erosion, and thereby soil quality, are therefore culturally transformative
in that their characteristics affect people. It is the relationship between
the human and non-human agents that is absent from the macro-scale
discourses promoted by many archaeologists and ancient historians. If
we accept this, then archaeological sites (human technologies) consti-
tute ‘hybrids’, phenomena that develop in a network where distinctions
between the social and the natural might not exist. Sustainability of
any activity is the product of human management of natural processes,
whereby successful forms of environmental knowledge contribute to the
maintenance of resilience. The watermill at Barbegal was the product
of a very specific socio-economic system – the Roman Empire, with its
desire to manipulate and control nature through complex forms of engi-
neering. The decline of the mill was the product of changes in wider
economic and political processes, whilst the associated re-emergence of
the unmanaged wetland at the foot of the watermill was not necessarily
caused by environmental change, but was another consequence of the

demise of the Empire.
As discussed in Chapter 5, we have to consider how forms of environ-
mental knowledge varied across time, space, culture, and class. Greek,
Roman, and perhaps even earlier urban groups and elites possessed dif-
ferent attitudes to environmental processes. Roman managers and elite
classes may have developed an asymmetric view of the world, in which
humans could control and manage the environment as part of endeavours
to maximise economic output. In Critias, Plato describes the environ-
ment of Athens, and in particular, he comments on the excellent quality
of the soil. Plato’s discussion of deforestation and soil erosion thus implies
that he was aware that humanity depended on a certain awareness of the
impact that different actions could have on the environment (Goldin
1997). However, we can only postulate the extent to which this form
of environmental understanding was shared across society, and if under-
stood, how this knowledge was then applied in the landscape.
The issue of temporal scale is also important, along with the notion of
the palimpsest (G. Bailey 2007). Erosional events occurred over periods
that we cannot precisely date. The seven phases of sedimentation identi-
fied in the Argolid ‘…were brief, lasting from a few thousand to as little
as a few hundred years, whereas the intervals when erosion was negli-
gible…’ were longer (van Andel & Runnels 1987: 138). This notion of
‘briefness’ is problematic if we reconsider the relevance of these phases
and their chronologies at a resolution that correlates with that of human
experience. The sediments examined in the field may well cumulatively
represent processes that took place over the ‘long term’, but can we be
sure that the processes responsible for the deposition of these sediments
were experienced by people who lived and worked on or near these sites?
It is their direct relationships with these non-human agents that are rel-
evant. As with many environmental processes, there would also have
been a temporal aspect to the relationship between people and erosion.
Colluviation in a Mediterranean environment is invariably cyclical, often
associated with late summer and autumn storms. Thus, the sedimenta-
tion and its removal may well have acted as a temporal marker within the
annual round of tasks. As societies become more complex, the decisions
regarding how such environmental processes should be understood and
responded to may be made by people who do not actually work the
land. Thus, the perceived temporality and transformative characteristics
of the environmental processes are contingent upon how the process is

experienced: the landowner may associate erosion with a fall in produc-
tion and a loss of profit, while the peasant or slave would see a purely
physical task – sediment that had to be removed. As such, sedimentation
is transformative but heterogeneous in its influences on different actors
within and beyond the actual environmental niche. The heterogeneity of
different peoples’ engagements with the landscape would also have been
class and gender-based. Actions and being within any landscape become
more heterogeneous as social hierarchies become more complex. Labour
is increasingly divided and subdivided between different groups, and is
often controlled/influenced/manipulated by social elites, who in turn
influence the understandings and perceptions that workers would have
had of non-human agents, including erosion.
Most societies demonstrate resilience in the face of menacing environ-
mental processes. High-frequency events, such as soil erosion induced by
annual storms, are often ‘shifted’ to occur at a lower frequency (Redman
& Kinzig 2003). This ‘shifting’ transpires through technological and
cultural management practices that are specific to each society; for exam-
ple, terraces built to control erosion, or defensive walls and drainage
channels to protect sites from flooding. Essentially, we should be asking
for whom were these processes relevant in the past? It is perhaps useful to
think of sediments as an integral agent within the cultural world. People
would have possessed the environmental knowledge vis-à-vis the local
causes of erosion. The removal of vegetation upslope of a site followed
by storms is in some ways an uncomplicated set of events that anyone liv-
ing in such a locale would understand. People do often contribute to the
causes of erosion. In some instances, erosion can be beneficial, as much
re-deposition is comprised of sediments that have been removed from
unusable mountain slopes down to readily exploitable flood plains. We
know that some Nepalese farmers actually cause landslides themselves
in order to bring down more fertile soil from the upper slopes (Forsyth
1998: 109).
In Mycenaean landscapes, palace-based elites commissioned terrace
building as part of a wider socio-economic-ideological strategy – an
example of where erosion mitigation possibly constituted the imposition
of forms of environmental knowledge upon low-ranking groups. At the
same time, local farmers might have applied their own forms of envir
onmental knowledge and developed specifically local forms of terrace
construction, choosing specific points on a hill slope, repairing terraces

at specific times of the year, and so on. For the workers at Barbegal
and other landscapes around the Mediterranean, the sediment would
have created a task that had to be completed under coercion from land
owners or managers, or possibly even ‘the gods’. Soil quality (and prob-
lems with the soil) is a phenomenon that would have been understood
via discourses that are quite different to today’s post-enlightenment sci-
entific descriptions of the physical world. The intersection with other
worlds, those of the gods, would have been an integral part of the envir
onmental knowledge of soils and sediments (Retallack 2008). It is inter-
esting to consider the forms of environmental knowledge that relate to
human engagements with erosion. In particular, we should aim to assess
not only evidence for erosion caused by past societies, but also their
responses to erosion. Do periods of erosion that follow on from the col-
lapse of mitigation strategies, such as terracing, represent a loss of envir
onmental knowledge, or inability to maintain a sustainable landscape
and/or incongruent forms of environmental knowledge where there is
a mismatch of environmental management strategies promoted by social
elites and local environmental knowledge held by those who work the
land (Redman 1999: 212)? Or are we looking at phases when erosion
was an irrelevance, as eroding areas were not economically valuable? It
is more useful from an archaeological perspective to assess the human–
environment dynamic as one characterised by transformation rather than
destruction (Butzer 2005: 1773), although there have been and still are
moments when people have created environmental domains that could
or cannot support economic activity.
7
Islands: Biogeography, Settlement, and Interaction
All Mediterranean Islands resemble each other; each island is different in its own
way. (Vogiatzakis, Pungetti, & Mannion 2008: 3)
Introduction
There are about 5,000 islands and islets across the Mediterranean,
although the number of larger, inhabited islands is closer to 100. This
statistic in itself demonstrates the extent to which many islands have not
supported long-term human activity. Sicily, Sardinia, Cyprus, Corsica, and
Crete are the largest Mediterranean islands, with Mallorca, Minorca, and
Malta falling into the medium-sized category. This chapter will only con-
sider one of the larger islands, Corsica, with some material from Cyprus,
Crete, and Sicily considered in other chapters. As the environments on
these larger islands possess a potential similar to that of continental zones,
the island biogeographical notions considered in this chapter do not apply
in the same way. There are several groups of small islands: the Aeolian
Islands, the Ionian Islands, the Sporades, the north Aegean Islands, the
Dodecanese Islands, and the Cycladic Islands (Fig. 7.1). Many of the
islands possess historical significance that is disproportionately greater
than their physical size. This chapter will start with an assessment of
island biogeography and its application in Mediterranean archaeology.
Then, an overview of how the study of island environments has been
employed by archaeologists will provide the core of this chapter.
Environmental characteristics across the various groups of islands in
the Mediterranean are of course largely influenced by their respective
positions across the various biogeographic zones outlined in Chapter 2.
Two of the larger islands in the northern-central Mediterranean – Corsica
210
Sea of Azov
Bay
of
Biscay
Black Sea
Gulf of
Lions
Corsica
Adriatic
Sea
d s
lan Sardinia Thasos
Is
ric
Balearic
Sea
lea 4 3 Minorca
Ba Gulf of Aegean Sea
Ibiza Majorca Tyrrhenian Sea Taranto
Sporades
Formentera Ionian Islands
Aeolian Islands Lefkada Chios
Volcano Andros Samos
Kefalonia Tenos
Strait of Gilbraltar Naxos
Melos 2 Amorgos
Rhodes
Pantelleria Kythera Santorini Dodecanese Islands
1 Antikythera Thera Karpathos Cyprus
5
Crete
Mediterranean Sea
Gulf of Sidra
7.1. Mediterranean island groups and islands mentioned in the text. Islands named on map: Sicily, Sardinia, Cyprus, Corsica, Crete, Mallorca, Minorca, Malta,
Sporades, Aeolian Islands (Vulcano), Dodecanese Islands (Rhodes, Karpathos), Cycladic Islands (Thasos, Andros, Naxos, Tenos, Santorini/Thera, Melos, Amorgos,
Chios, Samos), Pantelleria, Kythera, Antikythera, Ionian Islands (Kefalonia and Lefkada), Ibiza, Formentera, Karpathos. 1: Skorba, 2: Daskaleio-Kavos on Keros/
Heracleia (Irakleia), 3: Cova des Moro/Cova des Mussol (Minorca), 4: Alcudia (Majorca), 5: Pseira.
212 Islands: Biogeography, Settlement, and Interaction
and Sardinia – share similar geologies and environmental characteristics,

whilst at the western end of the Mediterranean, the Balearics are situated
within a biogeographical zone which shares traits with Spain and North
Africa. Moreover, this archipelago is unusual in that only 4 of the 151
islands and islets are settled today (Vogiatzakis et al. 2008: 12). Whilst
island size is obviously important, the presence of island groups, or archi-
pelagos such as in the Aegean, has significant biological economic and
cultural ramifications. Even though UNESCO defines ‘small’ as 10,000
km2 or less (ibid.), within the context of this chapter, where the assess-
ment of small islands is prioritised, ‘small’ is taken to be islands of about
5,000 km2 or less (the Balearics as group constituting 5,014 km2, and
Malta 316 km2).
If there is one aspect of Mediterranean archaeology that has been
dominated, or at least heavily influenced, by environmental models, it is
the study of islands (see Van Dommelen 1999: 247). Due to the bounded
nature of islands (especially small islands), they offer a unique perspective
on environmental processes and their history. In some ways, human set-
tlement on islands is concerned with managing the fragility and restric-
tions of island resources (soils, vegetation) that in some instances degrade
more quickly than they might do on continental zones. The fact that some
researchers do group mammals and people when assessing the settlement
of Mediterranean islands has led to some biogeographic models ignoring
the importance of cultural processes in influencing human decisions to
explore and settle. The discourses employed by some are unsubtle, and
refer to the invasion of islands by people and the consequent ‘extermina-
tion’ of fauna by these people (Schuele 1993). Therefore, an assessment
of island biogeography and its application in Mediterranean archaeology
is one of the themes running through this chapter. However, the relative
dearth of palaeoenvironmental landscape reconstruction work on many
of these islands is problematic. Obviously, those environmental processes
that have been dealt with in the preceding chapters are all pertinent to
the study of islands. The difference, however, is that islands, especially
the smaller ones, might show less resilience, and could therefore have
been subject to precocious and irreversible environmental degradation.
The Mediterranean is in many ways defined by its islands and moun-
tains. These zones constitute the quintessential domain for the intersec-
tion of local, specific environmental knowledge and imposed structures
of environmental knowledge. These environmentally complex zones have
Mediterranean Islands and Island Biogeography 213
always demanded local, specific forms of environmental understanding

and management. When systemised forms of environmental knowledge,
imposed by social elites, and local understanding of the environment
diverge, these zones may well witness periods of socio-economic and
environmental volatility – forms of environmental instability that are not
controlled by changes in climate, but rather by changes in the relation-
ships between local and elite forms of knowledge. The success or failure
of many islands, especially the smaller ones, is in part related to their posi-
tion within wider economic and cultural networks (Knappett, Evans, &
Rivers 2011). During the Bronze Age in particular, one might argue that
the environmental, economic, and cultural trajectories of islands were
linked to the development of networks of storage and exchange managed
by social elites, this being particularly true in Eastern Mediterranean,
where maritime economies were a defining characteristic of complex
societies (S. W. Manning 2008).
An overview of the great variety of island types and their physical
characteristics forms the introduction to this chapter. Whilst there has
been much discussion of the idea that insularity leads to the develop-
ment of cultural characteristics that are peculiar to islands (Patton 1996),
this chapter also focuses on the environmental trajectories that some
Mediterranean islands have taken, and the nature of human interaction
with these particular environment types. Where possible, archaeological
and environmental data are employed in an analysis of how island popu-
lations responded to environmental and landscape changes, and are used
to consider whether these responses are specific to island environments.
As with all of the chapters, a diachronic approach (from the Neolithic to
the Late Roman period) employs archaeological evidence in an assess-
ment of how different societies responded to similar sets of environmen-
tal problems.
Mediterranean Islands and Island Biogeography
Without a doubt, MacArthur and Wilson’s (1967) The Theory of Island

Biogeography has been of fundamental importance in the development
of ecologically informed island archaeological projects. Then, Evans’s
(1973) early paper on the topic established the archaeological adaptation
of these ideas. The most significant variables that affect the develop-
ment of island ecologies are the distance of an island from a continental
landmass, and the size of the island. Therefore, a small island should only
be able to support a reduced number of plant and animal species, and
the greater the distance of an island from a continental land mass, the
more difficult it will be for species to colonise and recolonise that island
(MacArthur & Wilson 1967). One consequence is that many islands pos-
sess a limited range of floral and faunal resources (Patton 1996: 2). The
island laboratory supposedly presents archaeologists with an opportunity
to study a (partially) closed environment, and therefore assess the man-
ner in which a given island community interacts with a bounded environ-
ment. Of course, such definitions are spurious, as the sea, despite its own
set of risks, is not necessarily a barrier to communication and the transfer
of people, ideas, plants, and animals. Island trajectories may be different
to those on the continent, but then so is the development of a continen-
tal mountainous landscape, or even a wetland. They are all the product of
an infinite network of influences that are both ‘natural’ and cultural.
According to Simmons (1998: 232), human impacts on Mediterranean
island environments ‘…have been especially harsh from Neolithic times
up to the present’. There is no doubt that, in some instances, the effects
of human activity on an island ecosystem can be disastrous, the most
infamous example being Easter Island (Rainbird 2002). As Patton (1996:
7) observes, Mediterranean islands tend to be less isolated than their
Pacific counterparts. Consequently, if we wish to test the notion that the
cultural and environmental trajectories followed by island societies are in
some way different to those followed in continental zones, we have to
demonstrate first that there are environmental processes or outcomes of
these processes that are peculiar to islands, and that, in turn, there are
cultural responses to these environments which are also quite original:
are there facets of past island cultures (including landscapes) which are a
consequence of insularity?
In order to identify changes in island environments and the nature of
the relationships that people had with the natural aspects of their land-
scape, we need to consider some of the evidence for settlement, econ-
omy, and cultural aspects of island life during the past. As noted above,
one aim is to identify processes that are peculiar to islands that did not
develop on the continental Mediterranean zones.
Islands do possess environmental characteristics that differ from con-
tinental zones. The relatively low number of plant and animal species,
along with a high number of endemic species, are the defining ecological
characteristics of many islands. Though a plant or animal species is more

likely to arrive on a large island close to a continental zone, we also need
to consider the ‘stepping stone’ effect: animals and plants may move
across a sea from one island to another. Such a process would certainly
have been common in the Aegean. Such island biogeographical models
cannot be applied to the study of human colonisation of islands: people
do not behave like animals and plants. Decisions to explore and colonise
are cultural decisions, although movement across the sea is constrained
by maritime technology, and the awareness of currents and weather
cycles (ibid.: 24). In some ways, one might argue that the very idea of
applying biogeographical principles to human behaviour is problematic
and results in studies of humanity that are not particularly enlightening.
At the same time, we must accept that the availability of island resources
and carrying capacity are a matter of biogeography. What interests us
here is the story of how island environments develop over time, and how
people interact with these environments. We then need to consider if the
characteristics of human/landscape dynamics on islands are in any way
different from those that occur in continental zones. Islands might be
considered as ‘habitats surrounded by radical shifts in habitat’, or even as
places where the task of controlling one’s fate is rendered more difficult
than on a continental area (Terrell 1999: 2040–2).
Colonisation of Mediterranean Islands

Whilst the chronological remit of this book does not include the
Mesolithic, a brief overview of the evidence for the earliest human arriv-
als on the Mediterranean islands is useful. Whereas people were present
on the Mediterranean continental zones from the Lower Palaeolithic
onwards, and there is growing evidence for pre-Neolithic activity on a
number of Mediterranean islands (Broodbank 2006; Cherry 1990), it
appears that many islands, especially the smaller ones, were not settled
until the Mesolithic or later. Moreover, the biogeographical character-
istics of many islands would have been quite different prior to and at
the start of the Holocene. Some islands would have been connected
or closer to the continental zones (van Andel & Shackleton 1982).
Biogeographical variables are still pertinent, and the most recent
research still suggests that it was the Neolithic period that saw the set-
tlement of many islands across the Mediterranean (Dawson 2011),
whilst initial visits, such as those to the Cyclades, took place during the
Upper Palaeolithic (Phoca-Cosmetatou 2011b). In fact, many islands

are not suited to hunting and gathering. Smaller islands would proba-
bly not have possessed the wide range of resources required by hunter-
gatherers, and this might explain the dearth of Mesolithic evidence
on most Mediterranean islands. However, herders and farmers with
focussed food-production strategies could successfully explore with-
out colonising these islands (Patton 1996: 139). Some islands, such
as Melos, may have been discovered at quite an early stage, as access
would have largely been across dry land (van Andel & Shackleton 1982).
Whereas resource availability would have influenced cultural processes
in the Adriatic during the Early Neolithic, it is quite likely that dis-
tances between the continent and islands were not considered signif-
icant (Bass 1998: 183). A Late Epipalaeolithic occupation of Cyprus
has been recognised for some time (Peltenburg 2003; Peltenburg et al.
2000; Simmons 1991), along with the early development of Early
Aceramic Neolithic communities (Knapp 2010). This is an example of
early exploitation. Moreover, it is possible that the transition to farm-
ing on Cyprus was forager led, with peoples who had been present
on Cyprus from the Early Holocene developing new food-produc
tion strategies (S. Manning et al. 2010). This does of course mean
that domesticates would have been brought in subsequently by other
people. Alternatively, some fisher-foragers continued to exploit coastal
resources for a few thousand years more at least (see Ammerman 2010;
Knapp 2010), and these groups developed strategies that were differ-
ent to the typical ‘Neolithic’ farming communities. Recent research
has demonstrated that many of the larger islands did see Palaeolithic
and Mesolithic activity, for example Lower Palaeolithic and Mesolithic
activity on Crete (Strasser et al. 2010, 2011). Mesolithic occupation
is also known on Sardinia (Sondaar et al. 1984), Corsica (Costa 2004;
Lanfranchi & Weiss 1973; Lanfranchi & Weiss 1997), and Mallorca
(Ramis et al. 2002). It is possible that migrant farmers were the first to
colonise Malta fully, with the earliest settlement dating to the transition
of the sixth and fifth millennia based on the dates from Skorba, perhaps
colonised by groups from Sicily (Guilaine 2003: 231).
Patton notes that the chronology of colonisation does not fit the vis-
ibility-based predictive models. A number of medium-sized, less visible
islands were discovered or colonised at a relatively early date (Patton
1996: 55). Although the biogeographic ranking models provide a
2.6
Asia Minor Mediterranean area

2.4
Balkans Iberia-France-Italy
North Africa
2.2
Sardinia
Corsica Peloponnese
SPECIES
Sicily
2.0
Balearic Islands Crete
Cyprus
1.8 Rhodes
Cyrenaica
Formentera
1.6
Malta
1.4
Cabrera Island
1.2
Log A
1.0
2 3 4 5 6 7
AREA
7.2. Double-logarithmic, species/area relationship of birds in some islands and mainland regions of the
Mediterranean (Blondel, J. (2010), The Mediterranean Region Biological Diversity in Space and Time,
p. 142, fig. 7.3; originally published in Blondel, J. (1986), Biogeographie Evolutive, Paris: Masson.)
reasonable prediction of island colonisation, it should be noted (as per

Fig. 7.2) that biogeographic ranking models work particularly well for
non-human animals.
It is possible that dry-shod colonisation of some islands took place
where people settled prior to sea-level rise (Broodbank 1999: 22–3). The
Aegean islands of Thasos and Chios were probably examples of islands
where dry-shod colonisation occurred, followed by insularity (ibid.: 30).
In such cases, farmers would have had to make decisions as to whether
it was ‘worth’ remaining in an area where rising sea levels possibly con-
tributed to an enhanced feeling of insularity – island life imposed by the
natural world. From the perspective of this book, we can take this to
imply that at the time of insularisation, these island environments were
perfectly capable of supporting their farming communities, and there

was clearly no need to abandon these zones. Moreover, the process of
insularisation would have been a slow one, allowing time for people pro-
gressively to adapt their lifeways.
Settlement, Economy, and Insularity

With the right maritime technology, the sea is not a barrier and, in cer-
tain cases, may actually prove to be a more effective medium for move-
ment and communication than dry land. Consequently, the notion of
insular carrying capacity might be considered increasingly irrelevant,
as trading and communication networks became reliable and complex.
However, if a region enters a period of reduced economic activity,
islands might be the first to suffer, especially if their productive envir
onment is already fragile. Our task as archaeologists is to assess how
the natural environment was exploited, and of particular relevance to
island archaeology, we have to consider if carrying capacity for people
was reduced by human impact. People can reduce the carrying capacity
of an island to sustain certain species, in particular, domesticates, thus
enhancing the capacity to sustain people. Perhaps more intriguing is
the issue of whether the unique cultural trajectories often found on
islands reflect the development of cultural ecologies that coped with
the exigencies of island environments. For example, it seems that Later
Neolithic peoples on certain Cycladic islands emphasised sheep and
goat rearing, as opposed to cattle, suggesting that there was an adap-
tion of strategies to the ecological exigencies of the island environ-
ments (Phoca-Cosmetatou 2011b: 89).
In many ways, islands are far from being the ideal ‘laboratories’ for
the investigation of well-defined relationships between people and their
environments. They are in fact nodes in the most complex web of his-
torical ecological processes. At one level, they provide an insurmount-
able set of fascinating questions rather than answers. Even without
considering cultural processes, biologists have noted the almost infi-
nite variability that characterises Mediterranean islands. The variation
in ecology and environments across Mediterranean islands is such that
it led Greuter (quoted in Vogiatzakis & Griffiths 2008: 66) to state, ‘In
the Mediterranean the choice is yours: there are about 5000 of them …
Just define your problem and choose the island or islet tailored to your
needs’.
Aspects of Insular Environmental and Cultural C
hange 219
Aspects of Insular Environmental and Cultural Change
In the following section, a discussion of the principal environmental and

cultural developments on a series of Mediterranean islands is consid-
ered. Where available, evidence for the impact of climate change along
with evidence for vegetation change and so-called phases of sedimentary
instability are considered in conjunction with pertinent archaeological
evidence.
Malta
Malta is a relatively small island (300 km2) situated between Sicily and
Tunisia. One notable and potentially challenging environmental char-
acteristic is that it is largely dependent on meteoric water. There are no
lakes or permanent rivers on the island. Today, Malta is heavily reliant
on desalinised water. In recent years, palaeoenvironmental research has
developed on the island (Caroll 2004; Carroll et al. 2012; Hunt 1997;
Marriner et al. 2012; Schembri 2009), and some key results are pre-
sented below.
Early colonisation on Malta probably started c. 5000 BC. Evidence
for Neolithic impact on the landscape comes from molluscan analysis
and palynology which implies that the areas around the Neolithic monu-
ments on Malta were cleared quite quickly, and this could have produced
‘the potential for serious ecological stress, provided population levels
were sufficiently high’ (Bonanno et al. 1990: 195). The sedimentolog-
ical and malacological analyses imply increased aridity, and the conse-
quential reduction in plant cover is broadly contemporary with the first
colonisation of Malta (Fenech 2007: 106). Furthermore, a significant
erosive event is contemporary with the construction of the first mega-
lithic sites. Of course, the relationship between possible environmental
destabilisation and temple building is almost impossible to articulate,
but this is a process and relationship that we should consider. As there is
nothing to suggest that the Maltese population grew beyond the carry-
ing capacity of the island, Evans (1977) argued that some kind of mech-
anism must have been developed in order to limit population growth.
Evans thus developed a model for small island development that tried
to draw a link between island biogeographical characteristics and cul-
tural development. There is no denying that the evidence for prehis-
toric ‘difference’ on Malta vis-à-vis other areas and islands. However,
this type of difference is in many ways similar to that seen in other well-
known or enigmatic prehistoric landscapes such as the Stonehenge envi-
rons or Val Camonica (Robb 2001). The building of Maltese temples
started c. 3600 BC. There are 30 temples (plus numerous other ruins)
on Malta and Gozo. Some argue that the temples on Malta developed
during the fourth to third millennia as an ideological response to the
stresses produced by biogeographical isolation (Stoddart et al. 1993).
Recent environmental research clearly suggests early clearance of vegeta-
tion and later soil erosion that was a product of human activity and also
climatic aridification at c. 2300 BC (Carroll et al. 2012). One important
change to the island’s environment was the loss of the Burmarrad Ria (a
coastal embayment/floodplain zone) which proffered a range of useful
resources. Sedimentary inputs (probably a consequence of environmen-
tal degradation referred to above) in-filled much of the bay (Marriner
et al. 2012). We must accept that it is highly probable that all small to
medium-sized island environments are susceptible to quite rapid envir
onmental deterioration, with changes across certain small niches having
had important consequences for an entire island.
Islands like Malta had strong links with other parts of the
Mediterranean – in Malta’s case, Sicily. However, this does not have to
mean that the island’s natural environment did not have an active role to
play in constituting insular culture. In fact, the possible ‘fragile’ nature of
the island environment and its limits on certain types of economic pro-
duction would have encouraged, even necessitated, contact with the out-
side world. One consequence of such regular contact with other groups
might have been the emergence of a perceived need to articulate an orig-
inal and overt identity vis-à-vis this world. It is possible that the temples
and the art within them represented components of the islanders’ cos-
mology – the land and the sea. Therefore, these temples might be seen
as metaphors for the island (Grima 2008). Robb (2001: 177) considers
that ‘islands did not fashion Maltese temple society but that, Maltese
temple society created cultural islands in the process of forming a local
identity’. Whilst this is an appealing idea, it denies agency for the island
itself, as constituted by all of its environmental and geographical char-
acteristics. Another important characteristic of any place is the manner
in which the strength and extent of contact with the wider world varies
over time. Therefore, levels of insularity on the same island must also
vary. After a possible period of reduced contact with wider networks,
hange 221
Stoddart (1999: 141) considers that during the later third millennium
BC, Malta was reincorporated into the Mediterranean system (see also
Malone et al. 2009).
As noted above, one of Malta’s environmental weaknesses is the rel-
ative dearth of water. On parts of the island, there would have been
an adequate water supply, with aquifers supplying natural reservoirs
(Fig. 7.3). However, part of the distribution of temples covers an area
where there may not have been adequate water for farming. Yet, Bonanno
(2005) argues that the situation would have been quite different during
the temple-building period, and it is quite likely that water capture and
storage mechanisms guaranteed the maintenance of a flourishing cul-
ture. However, the only dated evidence for water storage is constituted
by later Bronze Age cisterns. It is only in recent centuries that human
activity, and perhaps climate change, have contributed to the develop-
ment of a dry Maltese landscape. Moreover, research from eastern Spain
suggests that the all-important levels of summer rainfall may have been
higher in parts of the Western Mediterranean at certain points during
the Holocene (Aguilera et al. 2012). The famous, ubiquitous Maltese
‘cart ruts’ may have been field boundaries and/or irrigation channels.
It is also feasible that this system of ruts was related to the develop-
ment of ridges and furrows and the creation of artificial soils (Fig. 7.4).
The fact that some of the areas that comprise these ruts are now barren
implies that the system ultimately failed. Sagona (2004: 54–6) argues
that despite the difficulty of dating these ruts, it would be reasonable to
assume that a Temple Period date (c. 4000–2500 BC) could be possi-
ble, as this period saw a high level of landscape management and ‘social
cohesion’. Moreover, this complex culture would have probably tried to
expand its agricultural basis to the absolute limits, exploiting land that
may once have been considered marginal. It is of course possible that the
sudden end of the Temple Period at c. 2500 BC was a result of environ-
mental mismanagement and the inability of a small island with a limited
biodiversity to recover from such a process (Sagona 2004). However,
this is difficult to demonstrate without well-dated palaeoenvironmental
evidence (which could take the form of well-dated pollen diagrams, as
well as OSL-dated sedimentary sequences that demonstrate soil deple-
tion or exhaustion). As we have seen above, some new evidence does
seem to support the notion that climatic and anthropogenic processes
did combine, with some areas experiencing degradation.

7.3. Malta’s hydrological characteristics – schematic cross-section of Malta. It is important to note that sup-
ply from a perched aquifer is limited and that ground water on an island like Malta can be affected by saline
inputs from the sea. (By permission of Marianne Stuart, BGS/NERC – http://www.bgs.ac.uk/research/
groundwater/quality/Malta.html.)

7.4. Maltese cart ruts (photo: Shirley Cefai).
Tilley (2004) notes the main differences in the two limestone types –
the (harder) Coralline limestone and the Globigerina limestone – that
exist on Malta and Gozo. He argues that the very different characteristics
(hardness, colour, surface appearance, tactility) result in the monuments
hange 223
made from these two different rocks being perceived/experienced in

different ways. It is argued that the relationship of the temples to the
landscape was mimetic (ibid.: 97). Temples, such as Ggantija supposedly
mirror features of the surrounding landscape (ibid.: 109). Whilst we are
unable to present a clear image of the nature of the landscape within
which these temples were situated, it is fair to assume that unlike north-
ern Europe, these temples were probably situated within a relatively open
environment. At the very most, woodland would have comprised low-
maquis or mattoral formations, probably not affecting temple-landscape
inter-visibility. Despite these shortcomings in our data, it seems plausi-
ble that these temple complexes are one articulation of a form of social
storage that relates to human–environment relationships. As Broodbank
(2000) observes in his discussions of the Cyclades, communities in more
favoured environments would not need to involve themselves in net-
works of social storage. It might therefore be interesting to consider the
impact of climatic variability between different islands and attempt to
identify if the relative intensity and extension of social storage networks
was greater in more marginal zones; that is, small islands at relatively
great distances from their nearest neighbours (ibid.: 176). In addition,
we have to ask if these temple complexes developed as more strain was
placed on natural resources.
As it stands, there is limited evidence for Phoenician and Punic rural
settlement sites, whilst Roman rural activity is clearly attested to (Spanò
Giammellaro, Spatafora, & Van Dommelen 2008: 153). Pollen analysis
of a Punic ash pit implied a mosaic of grassland with areas of cereals
and olives plus small trees and scrubland. The number of Roman villas
(Fenech 2007: 17), and inferences made by reference to Ceres (goddess
of abundance), might suggest that Malta was a relatively fertile island dur-
ing this period. Ovid states that compared with sterile Pantelleria, Malta
was a fertile island (ibid.: 55–6). The pollen evidence for the Phoenician
and Punic periods implies relatively intensive arable agriculture. Arable
activity was also high during the period between the second century BC
and first century AD. This was followed by a relatively slow recovery of
the vegetation roughly between the first and second centuries AD. Cereal
pollen was at its lowest during the period between the third and fifth
centuries AD before recovering during the sixth century AD. Whilst the
island or the area around Marsa where these cores were obtained appears
quite resilient, it is interesting to note that Juglans’ (walnut) success was
short-lived, only appearing briefly during the Phoenician/Punic phase

(ibid.: 111–14). Here, we see how a possible attempt to introduce a cer-
tain species fails, as environmental knowledge did not encompass a full
understanding of the environmental limits of the island.
Malta offers one of the most enigmatic examples of the evolution of
an insular culture that may represent the evolutions of different popu-
lations’ relationships with their island and its natural environment. We
should never lose site of the fact that, for most pre-industrial people,
culture was in many ways an articulation of relationships and understand-
ings with the natural world. Environmental stress or restrictions would
have made profound contributions to these cultural configurations.
Smaller Islands
If biogeographical models were to have any value, it would be in their
application to the study of smaller islands – taken in this section to be
smaller than 500 km2 and including islands such as Antikythera (20.8
km2) (Bevan & Conolly 2013). There are a number of islands that clearly
have seen levels of environmental degradation greater than those seen
on most continental zones. It is on these smaller islands that we see the
development of the most enigmatic manifestations of cultural activity.
These cultures, which include the temple builders of Malta, are not the
product of insular environments per se, but as with any culture, they are
without a doubt the product of the complex interplay of socio-economic
processes and environmental agents, where the perception of the envi-
ronment has a role to play in the construction of the day-to-day lives of
all people.
As stated above, there are a number of small Mediterranean islands
whose cultural/historical importance belies their diminutive size. These
‘successful’ islands tended to have important but relatively short-lived
roles (within the context of the Holocene), and exploited a specific
resource and/or occupied a fortuitous geographical position within the
Mediterranean, having access to important maritime routeways. Clearly,
if small islands were exploited for traditional arable and pastoral agri-
culture, then in many instances their inherent environmental fragility
would eventually have curtailed this. However, small islands do have
some important environmental advantages. They present the inhabitants
with easy access to marine resources, even if they live towards the inte-
rior. However, as Broodbank (2000: 85) argues, marine resources in the
hange 225
Mediterranean are less abundant and predictable than in other oceans.

Activity on small islands will be influenced by environmental character-
istics that are localised and specific to that island. An essential resource
is water, and there are important inter-annual variations in rainfall over
relatively small groups of islands such as the Cyclades. For example, the
geological structure of Naxos (one of the ‘large’ small islands) means
that there are plenty of natural springs across the island (Dalongeville
& Rénault-Miskovsky 1993); not all small islands will possess natu-
ral springs. There is also plentiful groundwater on Andros and Tenos
(Broodbank 2000: 78). Islands like Santorini, with little topographic
variation and that are quite low, tend to be drier than continental areas,
although they will experience more fog (Rackham 2008: 37).
Although identified as possessing poor environments (Broodbank
2008: 47), life on the Cyclades was certainly not one characterised by a
struggle for survival. Many Saliagos culture sites do demonstrate a prefer-
ence for ‘environmentally clement niches’ (Broodbank 2000: 147). This
seems to mean that people made seemingly logical choices regarding
easy access to water and topographical situations with access to flat val-
ley lands. Moreover, Saliagos settlement seems to have taken place only
on the medium and large islands. Therefore, there may have been some
notion or understanding of how large an island should be in order to
minimise risk, possibly a ‘list’ of attributes including fresh water, arable
land, and a safe landing for boats. Homer refers to these very characteris-
tics in Book 9 of the Odyssey where he describes the land of the Cyclopes
(Butler 2006).
Whilst agriculture was practised on many smaller islands, some of these
places were exploited for specific resources from at least the Neolithic
onwards, the most famous example being obsidian extraction on Melos
(Renfrew & Wagstaff 2005; Tykot 1997, 2004) where the earliest
obsidian exploitation is now dated to the period between 13,000 and
10,000 years ago (Laskaris et al. 2011). During the late fourth to third
millennium BC, the dispersed settlement pattern of Melos characterised
an island which at this point was probably relatively independent from
the rest of the world (Wagstaff & Cherry 1982b: 251). Throughout its
history, the waxing and waning of the island’s prosperity was more influ-
enced by its economic and political relationships with the wider world
than any ‘constraints’ placed upon it by the environment. Despite eco-
nomic specialisation, at the very least, subsistence farming would have
been necessary in order to support the population. As it stands, the only

dated evidence for landscape degradation is derived from alluvial units
that contain pottery sherds with dates between c. 600 BC and c. 500 AD.
In landscapes such as those on Melos, it is difficult for soils to redevelop
once the bare bedrock is exposed, and there is little hope of vegetation
colonisation from abutting zones. On a small island, the possibility of
movement to new ‘soil-rich’ areas is reduced (Davidson & Tasker 1982).
This erosion may have been caused by demographic growth during the
late eighth century BC (Wagstaff & Cherry 1982a). Despite these envir
onmental problems, Melos continued and continues to produce a wide
range of agricultural products successfully, and in recent centuries has
managed to respond to changes in external economic demands (Wagstaff
& Augustson 1982: 131). The vegetation history of Melos is one char-
acterised by gradual degradation, although sudden degradation due to
long-term drought or human impact has to be considered a possibility.
Away from Melos, Herodotus referred to a seven-year drought on Thera
(fifth century BC) during which nearly all of the trees withered (Wagstaff
& Gamble 1982: 96).
The transition towards a drier climate during the Neolithic and Bronze
Age is perhaps one process that needs to be assessed more carefully. On
the Cyclades, it is clear that there was a move towards dairying subsis-
tence economies and an emphasis on small farmstead-size settlements.
This is taken to represent a more flexible economic pattern. The ‘Saliagos
settlements represent, in effect, a small number of precious eggs placed
in large baskets chosen for their safety, whilst Grotta-Pelos settlements
represent a great number of individually expendable eggs distributed in
small baskets of variable quality’ (Broodbank 2000: 155). The emergence
of the late Neolithic and early Bronze Age Grotta-Pelos culture repre-
sents a move into the smaller islands and also the adoption of niches that
had hitherto been unsettled (ibid.: 151). Whilst this macro-economic
perspective is interesting, and undoubtedly provides a useful resume of
two different situations, it does not explain how human–landscape inter-
action actually operated on the smaller scale.
Broodbank (2008: 53) considers that the plough, in comparison with
continental areas, would have had less impact on the broken terrain that
characterised these islands. In explaining the changes in the late-third
millennium Cycladic society, Broodbank (2000: 338) suggests that over-
exploitation through farming practices may have led to some desertion
hange 227
episodes, and as a result some settlements may have moved towards

lowland areas that ‘benefited from hill slope soil loss’. Such a process may
be represented in the soil profiles at Markiani on Amorgos. However, it
is unlikely that such a process would have been uniform throughout the
Cyclades. It is difficult to demonstrate that sudden climate change could
have been responsible for these late third millennium changes. A new
arid phase should have registered settlement shift towards areas with a
reliable water supply.
As the economies of small islands developed, their integration into
wider Bronze Age economic networks clearly influenced the manner in
which their landscapes evolved. For example, it is argued that changes in
the settlement patterns on Naxos were linked to the island’s integration
into the Minoan and Mycenaen exchange systems. One might argue that
for a relatively small island such as Naxos, the development of links with
other islands and continental zones was necessary. During the period
of close integration (LH IIIA1–IIIA2), the settlement pattern became
nucleated, and small settlements were abandoned. After this period, and
the reduction in levels of Mycenaean trade, the settlement pattern takes
on a dispersed structure once again (Cosmopoulos 1998).
The Ionian island of Kythera is just under 300 km2 and was clearly
large enough to support a relatively dense settlement pattern. For exam-
ple, Bevan (2002: 225) refers to the Neopalatial (c. 1700–1440 BC) as
‘…a brief pulse of settlement activity and an example of pioneer efforts
at colonizing new terrain’. Holdings may have been quite small and con-
centrated around the farm building. In the Kytheran environment, it has
been suggested that ecological risk was probably managed in different
ways, via a diversification of crops and animals within intensely farmed
plots rather than over a dispersed set of plots. Alternatively, there may
have been a form of social storage, with neighbouring farms assisting
one another in times of need (ibid.: 237–8). What is interesting from a
human and island-landscape dynamics perspective is that this level of dis-
persed settlement was quite short-lived on Kythera. It is possible that an
island such as this could only support such a system for a relatively short
period, and even then, the success of the system was due to a high level of
mutual support amongst the peoples of Kythera during the Neopalatial
period. Small islands such as Antikythera (about 40 km to the south-
east of Kythera) may well have witnessed quite dramatic changes in pop-
ulation (Bevan et al. 2007) as economic networks, and possibly even
forms of environmental knowledge fluctuated. In some islands, includ-

ing Kythera, a number of farmsteads with ‘a Minoanising material cul-
ture’ spread into zones that had seen relatively little activity prior to the
Neopalatial period (Broodbank 2004: 69).
Antikythera provides an example of a diachronic landscape survey of
a small-island where the historical ecology of the landscape was assessed
(Bevan & Conolly 2013). As with many small islands, it seems likely that
early exploitation comprised short-term hunting and/or gathering trips,
which could have included the exploitation of terrestrial fauna such as
hare, deer, or goat, as well as migratory birds (ibid.: 139). Even dur-
ing the later Neolithic to Early Bronze Age, permanent, all-year-round
settlement might not have been established. The evidence suggests that
these sites did not necessarily prioritise the best agricultural land. Sites
were often situated in zones that offered ‘good points of departure for
hunting and gathering expeditions around the island’ (ibid.: 147) – an
interpretation supported by the presence of projectile points on these
sites. The seasonal (spring–autumn) occupation on the island once again
fits with the notion of mobility across groups of islands, with larger
neighbouring islands or mainland acting as the core zones. As there are
no known obsidian projectile points from Crete, it is possible that the
Late–Final Neolithic phase of exploitation was undertaken by Cyclades
or Argo-Saronic Gulf communities (Bevan, pers. comm.). However, by
the third millennium BC, Cretan communities are responsible for the
settlement of the island (the shortest crossing between the two islands is
35 km) (Fig. 7.5). At the latest, the late third and second millennia saw
some permanent settlement located in zones more suited to arable agri-
culture. As with other small islands, it seems unlikely that continuously
high levels of population were maintained. The Hellenistic name for the
island – ‘Aegila’ (Goat Island) – possibly signifies the nature and limited
extent of activity (or even abandonment) during this period. The Early
Roman period was also characterised by low levels of activity (Bevan &
Conolly 2013: 164). However, an increase in population characterised
the later Roman period.
This waxing and waning of activity is typical of certain landscapes
(islands, wetlands, mountains, and arid zones) that receive people for
different reasons at different times. In these landscapes in particular,
the notion of anthropogenic niche creation (ibid.: 132) is important,
as such processes can have long-term benefits (whether purposeful or
hange 229

7.5. Photo taken from the far south of Antikythera with north-west coast of Crete visible 35 km away. (By
permission of Cambridge University Press, Bevan, A., and Conolly, J. (2013), Mediterranean Islands, Fragile
Communities and Persistent Landscapes: Antikythera in Long-term Perspective, Plate 22. 6.4.)
accidental) for settlers who return to an island in that the legacy of

certain forms of landscape management constitute in situ markers of
environmental knowledge (Fig. 7.6). Evidence suggests that vegetation
would have recolonised agricultural zones quite quickly (possibly within
20–60 years) (Palmer et al. 2010). Therefore, in many instances, periods
of abandonment allowed recovery. Phases of abandonment or reduced
human pressure and concomitant recovery might be seen as ‘natural’
historical-ecological cycles, or as changes imposed upon island dwellers
by a demanding and marginal landscape.
Another example of a ‘small-island adventure’ is represented by the
site at Daskaleio-Kavos, situated on the ‘small and environmentally mar-
ginal’ island of Keros (Broodbank 2000: 223). With no known earlier or
later settlement evidence, this site provides us with an interesting exam-
ple of how people took risks at certain times in the past in what appear to
be environmentally marginal niches.
On some medium-small Ionian Islands, there is evidence of
environmental degradation. Parts of Kefalonia (781 km2) and Lefkada

7.6. Antikythera – Zones of more continuous or repeatedly episodic use (‘anthropogenic
niches’), where relatively stable and resilient systems of landscape exploitation devel-
oped. (By permission of Cambridge University Press, Bevan, A., and Conolly, J. (2013),
Mediterranean Islands, Fragile Communities and Persistent Landscapes: Antikythera in
Long-term Perspective, fig. 9.1.)
(336 km2) possess evidence for sheet-wash fans and hill-wash mounds
and a number of abandoned terraces on now denuded slopes, all of
which indicate an ongoing cycle of erosion/deposition (Souyoudzoglu-
Haywood 1999: 5). Even relatively minor tectonic events can have a
particular relevance for small islands and their people. On Samos in the
Aegean, sudden coastal uplift has been identified from a number of dated
wave notches (Stiros et al. 2000). A sudden relative drop in sea level
c. 3,900–3,600 years ago and possibly 1,500 years ago could have had
implications for coastal installations such as harbours, as well as the water
hange 231
table, wells, and springs. However, there is no direct evidence for such
a problem on Samos. The lack of wells and springs may explain the dif-
ficulty of maintaining permanent settlement on the volcanic island of
Pantelleria in the Sicilian Channel. Obsidian was exploited during the
Neolithic, and the island was abandoned from the Late Bronze Age until
the arrival of the Phoenicians when rural activities did take place and con-
tinued into the Roman period. Almost 590 cisterns have been found on
the island; half these are clearly Punic (Spanò Giammellaro et al. 2008:
150–1).
The Balearics
Constituted by 151 islands and islets, the Balearics provide an interesting
case study, largely because research on the principal islands – Mallorca,
Minorca, Ibiza, and Formentera (which comprise around 99% of the total
land surface; Morey & Ruiz-Perez 2008: 272) – includes palaeoecologi-
cal work. Biogeographical frameworks have been useful and important in
certain aspects of the study of these islands. For example, the extinction
of certain fauna has been employed as an indicator of when people may
have arrived, based on the notion that people and any fauna that arrived
with them contributed to changes in Mallorca’s fauna and vegetation.
Taking into account the various dates of faunal extinctions and the chro-
nology of the first archaeological sites, it is inferred that people had colo-
nised the island by 2350–2150 BC. The use of extinction dates for the
identification of human arrivals is an approach deeply informed by island
biogeography, and is implicitly founded on the idea that such environ-
mental impact might be peculiar to islands. However, new populations
with new practices might have caused extinctions on the continent as
well. Despite this, it is apparent that the various forms of evidence that
date the arrival of people on some islands do correlate. Shrews disap-
peared on the Gymnesic Islands (the eastern Balearics) sometime after
3030 BC, and Myotragus balearicus, a dwarf caprine, became extinct.
The reasons for its extinction have been debated, and some are now
convinced that the chronological pattern of extinction and human colo-
nisation suggests that extinction was caused by people (Bover & Alcover
2003). However, if Ramis and Bover (2001) are correct, the correlation
between settlement and extinction is much less obvious.
When archaeologists (and biologists) discuss extinction of a species,
they rarely move beyond a discourse that states that the loss of the animal
or plant was probably due to human impact or climate change. The dis-
cussion of the consequences for local ecologies caused by the extinction
is not discussed, let alone the probable change in the wider landscape and
the way in which the perception of that landscape may have changed due
to the absence of that animal. Such a phenomenon would have been even
more important on a small island, where large animals were either rare
or absent. Moreover, if Myotragus had been domesticated, which seems
unlikely (ibid.), the impact for islanders would have been even greater.
As we know that the preferred food of Myotragus was box (Buxus bala-
erica) (Alcover et al. 1999), there may have been an increase in this tree
once the animal died out. Perhaps this is an event that might one day be
identifiable in high-resolution pollen diagrams.
The earliest unequivocal archaeological evidence for settlement indi-
cates permanent stable colonisation from 3000 BC for Mallorca and late
in the third millennium for Minorca. Between 2600 and 2500 BC, it
appears that ‘adaptation to the island environment on Mallorca is fully
completed’ (Guerrero-Ayuso 2001: 148). However, some consider that
the archaeological data are at best ambiguous. Ramis et al. (2002) posit
that the arrival of humans on Mallorca is registered in the sedimentary
record at the site of Cova des Moro where there is a ‘sedimentary discon-
tinuity’ characterised by an increase in charcoal. This process has been
dated to c. 2400 BC (ibid.: 17). The review of the reliability of dates
from the Balearics carried out by Ramis et al. suggests that people did
not settle on Mallorca prior to the third millennium BC. Whilst colonisa-
tion of Minorca may not have occurred prior to 1930 BC, there is some
possibility that humans did arrive prior to this period (ibid.: 18). These
islands receive irregular amounts of rainfall. Whilst the average on Ibiza
is 400 mm per annum, only 50 mm fell between April and September
1971 (Gómez Bellard 1995: 445), and people may well have had to
negotiate similar fluctuations in the past. An initial phase of occupation
on Ibiza and Formentera, two of the smaller Balearic Islands (570 km2
and 83 km2 respectively), is dated to c. 2000 BC. Though there is a pos-
sible Neolithic site (c. 4500 BC) on Ibiza, this phase is followed by what
appears to be a hiatus lasting about 600 years. It seems likely that the
limited natural resources, in particular water, may have rendered these
islands less attractive to colonisers (Gómez Bellard 1995). Again, the
development of an island-specific form of architecture (the navetas) is
another manifestation of enigmatic insularity (Ramis 2010).
hange 233
The availability of pollen diagrams renders the assessment of envir

onmental change on these islands more effective. There is clear evidence
for human impact on the vegetation on Mallorca and Minorca from
c. 2300 BC. It seems that the initial human impact on some of the
Balearics was quite profound. Moreover, deterioration in certain habitats
may well have been brought about by introduced rodents that arrived
with people (Bover & Alcover 2008). Pollen diagrams from Mallorca
and Minorca imply impact on these islands’ ecosystems after 4500 BC
(Yll et al. 1997). Yll et al. consider that the changes in vegetation that
occurred on Mallorca and Minorca could have been caused by a change
in climate and/or the impact of people (1997). In Mallorca, there was
a clear reduction of boxwood (Buxus balaerica) and juniper (Juniperus)
and a marked increase in wild olive (Olea europea). Moreover, evergreen
oak increased, replacing deciduous oak (Pèrez-Obiol et al. 2011). The
substitution of mesophilous plants by maquis scrub could have been due
to the development of a relatively dry climatic phase (Pèrez-Obiol &
Sadori 2007). The fact that this same process occurs on Minorca about
one millennium later leads Guerrero-Ayusos (2001: 145) to question
the hypothesis that climate change was responsible for the evolution of
these new vegetation patterns. Whilst the demise of tree cover does not
necessarily constitute a ‘fragile’ environment and concomitant economic
problems, it is a fundamental change in the nature of that environment
and landscape, and will therefore affect the way in which that landscape
is perceived and engaged with. Such a change would have been a con-
stitutive element in the construction of an ‘insular’ culture. However,
the range of cereals present during the Bronze Age was probably the
same as that on the mainland (Moffett 1992). Therefore, despite the
evidence for an initial impact on the environments of these islands, it
seems that a period of relative resilience followed. The finds from the
enigmatic Cova des Mussol burial site (situated on a cliff face), per-
haps literally defining the edge of life, include wooden carved figures
made from Olea europaea and Buxus balearica. The objects dated to
1393–1295 BC and 1192–1027 BC (Pèrez 2005) were probably carved
from local wood – wood that possibly took on more value as scarcity
increased. Despite the undoubted changes in the later prehistoric envir
onment, the absence of new technologies with a drastic impact on the
environment and the possibility that an insular form of environmen-
tal knowledge emerged engendered equilibrium. Consequently, these
islands did not really witness any profound changes before the arrival of
Rome (Alcover 2008).
An extensive Beaker phase of settlement with concomitant agricul-
tural and other economic activities is attested to on the islands (Waldren
1997). Ramis (2010) argues convincingly for a late third-millennium
colonisation with evidence for cultural (Beaker) links with the south of
France.
It is interesting to note that people on a relatively small island did
not appear overly reliant on marine food sources, despite the obvious
benefits of replacing terrestrial-based proteins with marine proteins as
part of curation of the terrestrial environment. There is little evidence
for consumption of marine resources by prehistoric communities on the
Balearic Islands (Van Strydonck et al. 2009). Even in the Roman period,
marine proteins appear to have been minimal. The low biotic productiv-
ity, touched upon in Chapter 3, might go some way in explaining this.
However, there was also the notion that a marine diet was considered to
be a sign of poverty (Purcell 1995), although the upper classes did con-
sume garum (B. T. Fuller Márquez-Grant, & Richards 2010).
The proto-historic and Roman periods are not always covered in the
Balearic pollen diagrams. However, the diagram from Alcudia in Mallorca
does demonstrate that woodland vegetation did not recover (Burjachs
et al. 1994). Important changes in the culture of the Balearic Islands
occurred during the ninth century BC, including the development of
the Talayotic culture with its enigmatic stone-built towers. This period
saw the demise of the navetas on Minorca and a change in burial rite also
appearing on Mallorca. The absence of evidence for new arrivals on the
Balearic Islands between 1500 BC and the Roman period might imply
that people were aware that emigration from the continent was not via-
ble (Van Strydonck 2004). One interesting Late Holocene process is the
evidence for the probable impact of alien rodents (which may well have
arrived during the Roman period) on the islands’ ecosystems (Traveset
et al. 2009). The potential impact of rodents on island environments
should not be underestimated (Drake & Hunt 2009).
Unsurprisingly, the trajectories of small islands are often different to
their continental neighbours. There is evidence that small islands are
more susceptible to environmental degradation. However, such processes
do not necessarily lead to abandonment, as islands emerge as important
nodes in trade networks and often possess raw materials that are valued
Discussion: Are Islands Different? 235
elsewhere. Some small islands were settled quite late on, and may have
witnessed periods of abandonment. For example, Ibiza and Formentera
were probably not settled until the end of the third millennium BC, and
were then abandoned between the thirteenth and seventh centuries BC.
The Phoenicians then settled these islands – probably the only islands
settled by the Phoenicians not to have an indigenous population (Gómez
Bellard 2008: 46). Once settled, some small islands developed intensive
forms of production concentrating on the export of a limited range of
goods or materials. Rural activity on parts of Ibiza and Formentera was
quite intensive, with evidence for olive oil and wine production, espe-
cially from the third century BC onwards (ibid.: 67). This specialised
agriculture undoubtedly suited the relatively poor-quality soils on these
islands.
Discussion: Are Islands Different?
There is no doubt that the environmental development (including the

waxing and waning of flora and faunal diversity and species numbers)
of islands is often different to that on continental areas. Whilst islands
and continental landscapes share common niche types, the issue is that
the same habitat on an island may not recover from degradation in the
same way as that on the continent. Islands do possess a series of envir
onmental characteristics that differentiate them from continents, and
these characteristics will have influenced the development of insular cul-
tures. Species richness and diversity are often such that recovery from a
phase of environmental degradation is more problematic. Mediterranean
island vegetation is especially important, and plant communities can be
quite different to their continental counterparts. For example, Crete pos-
sesses about 200 endemic species and subspecies (Rackham 2008: 37),
thus demonstrating difference vis-à-vis continental landscapes but not
necessarily environmental resilience. Environmental change on islands
might accelerate at greater rates than on continental areas. For example,
on Minorca, box-juniper was quickly replaced by olive, phrygana, and
steppe (ibid.: 49) – a development that is not always mirrored in conti-
nental landscapes.
From an environmental perspective, one key question in the study
of people and islands is the issue of human impact on endemic species
(both plants and animals) and the importation of alien species. Whereas
most plants and animals on continental Mediterranean areas migrate

naturally (there are of course exceptions), some species are either pur-
posefully or accidentally brought onto islands. In some instances, cer-
tain domesticated animals may have been brought onto islands but for
some reason became feral (Groves 1990). The impact of such animals on
the environment can be quite devastating, as was the case with certain
rodents. Moreover, on islands, certain unusual processes can occur. For
example, in the absence of a competitor, some animals on islands (that
also exist on the neighbouring continent) may extend to their full fun-
damental niche, rather than the realised niche occupied in the presence
of that competitor or other constraint (Blondel 2008). Such changes
not only have ecological repercussions, but also demand the develop-
ment of new forms of environmental knowledge. This intersection of
cultural and environmental processes once again articulates a historical
ecology where the boundaries between culture and nature are neces-
sarily blurred. The very act of environmental manipulation is an eco-
logical-cultural process. Successful colonisation demands processes of
‘landscape learning’ (Phoca-Cosmetatou 2011a: 20; Rockman 2003).
As people adapt to the specificities of their environment, environmental
knowledge manifests itself via economic and cultural practices. An eth-
nographic study of crop processing on the Aegean islands of Karpathos
and Amorgos demonstrated the ‘time-stress’ associated with harvesting
and crop processing. The level of labour required within a short period
places extreme demands on peasant farmers. This research also demon-
strated that farmers were flexible in their day-to-day decision making
regarding the effects of weather and other commitments (to family and
other more economically attractive tasks). The dispersal of fields across
the island also appeared as an advantage, as rainfall patterns can vary
across an island quite substantially, thus spreading the risk of crop failure
should there be low rainfall on one particular part of the island. Variation
in the type of crops grown is also an important risk buffer (Halstead &
Jones 1989: 50–1). Perhaps the one key environmental characteristic
that renders certain small islands marginal is the fact that crop growing
on or near exposed windy coasts is extremely difficult (Halstead 2008:
234).
We have also seen how some small islands were exploited for a spe-
cific ‘niche’-based activity, such as grazing. The potential of some small
islands seems surprising. Published in 1910, Seager’s (2000: 6) account

7.7. Mochlos – a small island off the north-east of Crete with Minoan house tombs (photo: Berengere Perez).
of his work on the Island of Pseira (2 km off the north-east coast of

Crete) starts with the observation that this barren rock of an island
could have offered little to even Minoan settlers, other than its har-
bour. However, despite its small size, the site at Pseira boasted a series
of Minoan buildings, including a shrine and a cemetery (Betancourt &
Davaras 1988). The cemetery constructs a specific identity for some-
thing as environmentally insignificant as an islet, but at the same time,
demonstrates that 2 km is not far enough to render an islet insular.
We see a similar situation on the small island of Mochlos, also off the
north-east coast of Crete (Fig. 7.7) with its rich Minoan house tombs
(Soles 1992). Clusters of islands that have easy sea access between them
might well be considered as a single entity. This serves as an example
of how finely honed environmental knowledge vis-à-vis insular climate,
weather, and topography characterises certain stages of island histories.
For example, it seems that early metalworkers on eastern Crete chose
topographical situations where the wind was funnelled, thus improving
the action of furnaces (Betancourt & Farrand 2006: 20). However, with
the resilience of an island’s environment damaged, life can become more
complicated for the island’s inhabitants in a relatively short period. For

example, small changes in weather patterns, such as those that may have
constituted the more arid Bronze Age climate, may have made some
agricultural activities more difficult (ibid.). In addition, the impact of
sea-level change on island communities may be greater than for those
living on the continent, especially on smaller islands, as such changes in
coastal configuration have a proportionally greater effect on the land-
scape available to islanders.
If simple biogeographical models were to work, then we would expect
the cultural ‘richness’ and variety to observe some basic rules. Larger
islands close to continents should develop stable, rich cultures (as rep-
resented by material culture, settlement evidence, etc.). There should
then be a reduction in richness on smaller islands that are further away
from core, continental cultures. The development of influential cul-
tures on Crete, then, should not surprise us. In turn, large, ‘success-
ful islands’ such as Crete should then influence adjacent smaller islands.
We do see such trends at certain points during the past. Consequently,
some islands are often subservient to an external (continental) power.
Rich, albeit temporary, cultures on smaller islands, such as the Cyclades,
Melos, or Santorini, do not necessarily observe the rules articulated by
biogeographical models. However, the often-temporary nature of the
cultural and/or economic influence of these smaller islands is some-
thing that we must consider. Moreover, we should assess the influence
of the biogeographical/environmental characteristics and changes on
such islands. If a small island is exploited for a specific resource, then the
environmental impact associated with that exploitation will be intense.
For example, some particularly small islands were sometimes used for
pasture. Therefore, the impact on vegetation would have been intense,
and we know that grazing and burning on some Greek islands have
adversely affected woodland conservation (Thornes 2009: 572). Goats
were in fact proscribed on some islands. As Constantakopoulou (2007:
210) observes, the decree forbidding the import and feeding of goats on
Heracleia must be due to the problems caused by large herds of unsu-
pervised goats destroying crops and threatening agricultural production
in a sensitive environment.
A most useful definition of insularity is that employed by Knapp (2008:
18): ‘The quality of being isolated as a result of living on islands, or being
somewhat detached in outlook and experience. Insularity can result from
personal, historical or social contingency’. The limitations posed by insu-

lar resources are fundamental attributes of island life. Nevertheless, it
is the attitudes of each society to insularity and the sea that truly influ-
ence how insular any given society becomes. Important changes in the
Chalcolithic economy and society on Cyprus occurred as numbers and
the ‘stress’ on resources increased. One hypothesis that will need contin-
uous testing is the idea that increased competition on islands (with their
inherently reduced carrying capacities) led to an increased investment
in monuments (Renfrew 1976). At the same time, we should also con-
sider the ‘cultural isolation model’, where island populations may follow
different trajectories to those on continental areas in terms of landscape
manipulation and management. Patton (1996: 136) also argues that we
should consider the relevance of the sociogeographic model, in particu-
lar, the idea that certain ritual practices and the authority associated with
such practices is exaggerated on island communities. Patton feels that
the sociogeographic model is useful. However, it seems that we should
examine how the various threads from each of these models interconnect
and have led to specific and original human–landscape relationships on
some Mediterranean islands. Costa (2008: 7) implies that insularity itself
could explain why there is a certain similarity in megalithic forms across
various Mediterranean islands. The southern part of Corsica alone has
25 stone alignments (800 stones in total) (ibid.: 92). In a discussion of
the third millennium in France, Lichardus et al. (1985: 557) comment
that ‘Corsica, as in all periods, presents cultural facies that are different
to those found on the continent’ (authors’ translation). Also, as Guilaine
(2003: 308) notes, despite their proximity, Corsica and Sardinia develop
quite different cultural trajectories.
The presence of several megalithic tombs on Pantelleria (a small island
about 70 km to the east of Tunisian coast and about 200 km to the
north-west of Malta) is interesting, as the principal economic activity
here seems to have been the mining of obsidian (one of the few sources
present in the Western Mediterranean). As with many Bronze Age soci-
eties around Europe, what we see across much of the Mediterranean is
the removal of the domestic habitats away from agricultural land. The
emergence of complex funerary monuments across these landscapes is
unsurprising as different groups, now some distance from their land,
ensure that their ownership is clearly articulated. These processes should
also be understood as part of a wider network of changes in relationships
with the natural world, a world which in some places, such as on smaller
islands, may have become more ecologically fragile (Costa 2008). It is
wrong to claim that the monument building on these islands was a prod-
uct of complete isolation from the continent. Rather it was a process
of island societies remembering links with continental areas, and then
moulding and developing new specific monument forms and related cul-
tural practices (Gili et al. 2006).
What we have seen in this chapter is a notable variation in the approaches
to the study of human–island interactions across the Mediterranean.
Archaeologists do critically assess approaches underpinned by bio-
geographical notions. Each island has its own set of specific contingent
historical and environmental histories (Cherry 2004). The changes that
we see on some small islands do demonstrate how insularity can prevent
environmental recovery. However, the environmental and landscape his-
tories of many larger islands are not so different from their continen-
tal counterparts, although the impact of a small number of alien (floral
and faunal) species has had important consequences for some insular
landscapes. The study of endemic floras and faunas – such as the work
carried on Corsica (Vigne & Valladas 1996) and the Balearic Islands – is
one important and productive research strategy. However, many islands
lack integrated palaeoecological and geoarchaeological research, and it
is often difficult to assess the extent to which human impact on island
environments has diverged from that on continental zones. One mech-
anism for assessing the historical fragility of insular environments vis-
à-vis similar continental analogues is to carry out vegetation and soil
surveys and compare these with one another, as ultimately island land-
scapes should be more degraded than their continental counterparts if
both have experienced similar histories of human settlement and activity.
We cannot equate islands with isolation; the complex set of metaphors
associated with insularity has hindered the development of a nissology
that avoids the wholesale application of a rigid biogeographical para-
digm (Eriksen 1993; Hay 2006). Early islanders did not necessarily live
in harmony with their environments, and the notion of endemic equi-
librium is clearly unfounded (Broodbank 2000: 7). However, environ-
mental knowledge that was specific to island life did develop. Degraded
island environments may well constitute the quintessential example of
diverging forms of environmental knowledge. Those people who directly
engaged with the natural world would have been influenced by elite
socio-economic structures and imposed forms of environmental knowl-

edge. The attraction of certain islands as nodes with a cultural economic
network may have increased demographic pressure to the point where
the normal carrying capacity of these islands was exceeded, and the now
infamous island cultures with their enigmatic monuments and material
culture must be seen as being partly the product of island environmental
trajectories.
8
Mountain Economies and Environmental Change
Introduction: Vertical Spaces, Cyclical Time
Extreme topographical variation within Mediterranean subregions has

led to complex understandings of and engagements with these landscape
mosaics, thus engendering intricate forms of environmental knowledge.
For these reasons, mountainous landscapes are the focus of the final the-
matic chapter. Here, we draw together the themes and environmental
processes discussed in the preceding chapters. Mountain environments
comprise most of the landscape characteristics and processes discussed
in the previous chapters. Issues of human engagements with vegetation,
fluvial and alluvial systems, and terrestrial (colluvial) processes are con-
sidered here within a framework which assesses the chronological and
spatial variation of different societies’ relationships with the most enig-
matic of all landscape types. In some ways, even the question of insularity
is relevant. Mountain peoples have often been viewed as ‘alien’ vis-à-
vis their lowland counterparts. The topographical relief (the peaks and
ridges) are to mountain peoples what the sea is to islanders. As McNeil
(1992: 12) suggested, the Mediterranean is not so much the sea between
the lands, but rather the sea amongst the mountains. The changes in the
patterns of exploitation of these different horizontal zones are the fasci-
nating phenomena here. It is rare for any permanent settlement to exist
above 2,000 m. In the Alps, the highest parish or communes are Trepalle
(Italy) at 2,209 m and Saint Verran (France) at 2,040 m – places at the
very edge of the Mediterranean world (Fig. 8.1 – map of places referred
to in this chapter). Mountains protect the Mediterranean lowlands from
almost guaranteed desertification and insurmountable aridity. The ‘relief’
proffered by mountains traps passing air masses, and ensures orographic
242

Sea of Azov
Bay 14
of 15 1
Biscay
Alps 13
26
17
2
Li pen
21 24 16
gu n
A
Black Sea
ria ine
18 12
n s
19 11
20 Gulf of
Lions
Ap
Pyrenees
9
en
ni
25 Adriatic
22
ne
Sea
s
27
Pin
Balearic
Sea
6 5
du
Gulf of Aegean Sea
s
23
Rif
7 8
Middle Atlas
3 Mediterranean Sea
High Atlas
Gulf of Sidra
8.1. Mountain ranges of the Mediterranean, and places referred to in the chapter. 1: Trepalle, 2: Saint Verran, 3: Tigalmamine, 4: Samarinia, 5: Grevena highland
zones, 6: Rezina Marsh, 7: Sphakia/Cretan White Mountains, 8: Lasithi plain Dikataian Mountain range, 9: Calderone Glacier, 10: Prato Mollo, 11: Pian del Lago,
12: Prato Spilla (Emilia-Romagna), 13: Palughetto basin, 14: Forcellin-Foscagno glacier/Val Febbraro, upper Valle di Spluga, 15: Lago Basso/Val Vidröla/Borghetto
Sotto/Lago Grande west, 16: Ubaye Valley/Col de Roburent, 17: Puy-St-Vincent/Rama/Fangeas, 18: Mercantour/Mont Bego, 19: Vicdessos, 20: Cerdagne/
Donezan/Enveig Mountain – Madriu valley (Andorra), 21: Asturia, 22: Sierra de Gredos, 23: Segura Mountains, 24: Lac des Lauzons, 25: Aniene Valley/Simbruni
Mountains, 26: Petit Col Saint Bernard, 27: Lake Ohrid.
244 Mountain Economies and Environmental C
hange
precipitation that feeds the streams and ultimately the rivers that run
into the lowlands. This process is as true on those islands with mountain
ranges as it is of the continental areas with the Alps, the Pyrenees, and
Atlas Mountains.
This chapter will start with an overview of archaeological and envir
onmental characterisations of Mediterranean mountains, followed by a
Mediterranean-wide review of environmental research, and, where possi-
ble, associated archaeological work. The final sections comprise a synthe-
sis of recent integrated environmental and landscape archaeological work
in the Pyrenees and the Southern French Alps. These latter examples are
situated within a historical and cultural ecological framework that once
again considers notions of environmental knowledge. The key precept
is that, as with small islands, life in the mountains demands varied sets
of environmental knowledge. This knowledge is constituted locally, and
periods when rigid, hierarchical socio-economic systems (such as empires)
attempt to impose incongruous forms of institutionalised environmental
knowledge may often see the emergence of environmental problems or
even the withdrawal of people from certain landscape types.
Defining Mountain Landscapes

The defining characteristic of mountainous areas is vertical zonation –
the relationship between environments (climate, vegetation, soil, etc.)
across the vertical zones, the types of activity that can take place therein,
and when (in which season) these activities can take place. These notions
underpin the framework and models for much research in mountains
(Fig. 8.2). The reduction in temperature of 1°C for every 300 m climbed
is one of the key defining environmental variables in mountainous areas.
Consequently, as with the study of islands, biogeographical models
have influenced the ways in which we interrogate and interpret human–
environment engagements in mountains. One important observation in
mountain ecology is the idea that the high altitudes are ‘buffered’ by
lowland forests, grasslands, or deserts in the same way that islands are
buffered by seas. Moreover, the fact that high-altitude zones tend to be
cold adds another form of buffer (Billings 1979: 97). As with discus-
sions of island biogeography, we must avoid the uncritical application of
such models when assessing human colonisation and life in high-altitude
zones. There are, however, simple, unavoidable biogeographical rules:
the level of pedogenesis decreases with altitude, and unsurprisingly, the
Alpine
2100-3000 m
Summer pasture
hunting, limit of forest
in many mountain ranges
Subalpine
1600-2100 m
Summer pasture, hay making,

some arable at lower limit,
forestry, mining
Montane
1000-1600 m
Arable, permanent settlement

8.2. Vertical zonation in mountain landscapes: ecological characteristics and principal human activities. Note certain activities such as mining are not constrained by
ecological zonation. Also, the vertical zonation heights (montane, subalpine, and alpine) vary across the Mediterranean (figure: F. Mocci and K. Walsh).
hange
number of different species of plants and animals also decreases. These

characteristics are directly influenced by the amount of exposure to sun-
light, wind, slope, geology, and so on (ibid.: 105). It is the exposure to
sunlight – the different orientations of mountain slopes – that is possibly
the defining characteristic in many mountainous zones, and thus leads to
specific designations for north- and south-facing slopes, such as ubac and
adret in French. The orientation of the slope is a fundamental control on
vegetation growth and the upper limit of the forest. The natural limit of
the forest is the timberline (the limit of the closed canopy), and the tree
line is the limit of tree growth. Those limits both advance and retreat,
being influenced by changes in climate and human exploitation of the
forest. Whilst the primary productivity of an alpine zone will influence
what people do at high altitude in terms of farming and hunting, there
are certain resources, especially minerals and ores, that attract people
to live and work in these zones no matter what the primary ecologi-
cal productivity. However, humans are usually absent from high-altitude
zones during periods when plants are dormant – a situation that is not
true of other environments. Many alpine zones are only productive for
10–25% of the year (ibid.: 113), although longer periods of productivity
characterise some Mediterranean high-altitude zones. Using the Andes
as an example, Winterhalder (1994: 35) observes that the predictability
of monthly precipitation is reduced at higher altitudes, implying a con-
comitant increase in subsistence risk. However, risk is entirely depen-
dent on environmental knowledge, and the effectiveness of any culture
in successfully applying that knowledge in the mitigation of unpredict-
able events.
One of the fundamental problems in assessing the impact of climate
change on mountain communities is the fact that each valley, and even dif-
ferent zones within a valley, will experience differing consequences from
any change in climate. Valley weather systems are exceptionally complex,
and modelling such processes is difficult. Mountains and uplands are
sensitive to environmental stresses and environmental change. We tend
to assume that environmental stress is a corollary of increased environ-
mental change, but this depends entirely on the nature of the change.
Climatic cooling can actually engender the conservation of certain
environment types, whilst warming may lead to an increase in extreme
weather events in certain landscape types, particularly in mountainous
zones. Therefore, climatic cooling should not necessarily be considered
Human–Landscape Engagements across Mediterranean Mountains 247
as causing environmental stress in mountainous environments, whilst

climatic warming might well lead to new stresses via extreme weather
events such as summer storms (Beniston 2000).
Although it is partly true that the dearth of evidence for early human
activity in areas above 2,500 m around the world is explained by the
relative remoteness and ecological poverty of such zones (Aldenderfer
2006), we should not forget that the difficulties in accessing high-alti-
tude areas varies from region to region, and, moreover, that the levels
of ecological productivity vary. Many Mediterranean high-altitude zones
are relatively easy to access, and the extreme variation in Mediterranean
topography results in landscapes where vertical height can be attained
over relatively short horizontal distances; mountains are not impenetra-
ble (Horden & Purcell 2000: 131). One problem in demonstrating that
people have always incorporated these zones into their lifeways is the fact
that the archaeological signals tend to be ephemeral, as mobile peoples
rarely leave substantial built remains (Cribb 1991). Moreover, detailed
research in many mountainous areas is limited.
Several questions are paramount in the study of mountainous land-
scapes. In particular, we need to assess whether human exploitation of
mountains is continuous. Can we identify certain periods when there was
an expansion into higher altitudes, and conversely, were there periods
when mountainous areas were avoided? An important theme in moun-
tain archaeology and history is the idea that mountainous areas act as
marginal zones that ‘receive’ people at times of climatic amelioration.
However, the examples cited in the latter part of this chapter question
this sentiment, and comprise an assessment of sociocultural reasons for
the waxing and waning human activity in mountainous zones.
Human–Landscape Engagements Across Mediterranean

Mountains
This section comprises an assessment of some of the work that has taken
place across Mediterranean mountains. As ever, emphasis is placed on
research that combines palaeoenvironmental data with archaeolog-
ical evidence. Archaeological survey in high-altitude zones is either
rare or demands a great deal of time and effort for low ‘archaeological’
rewards (i.e. sites and artefacts). One approach to the study of moun-
tainous archaeological landscapes and their environments is the use of
hange
palynology and anthracology as proxies for human activity. Whilst some

palaeoenvironmental research has been carried out in North African
highland areas, such work is still relatively uncommon.
Prehistoric settlement appears to have been sparse in the Rif range
(about 350 km in length and 80–100 km wide) in northern Morocco
(McNeill 1992: 83). Today, long-distance transhumance does not exist
in the Rif, perhaps due to intertribal conflict. However, short-distance
movement of flocks does exist (ibid.: 48). The nature of seasonal move-
ment of people and animals obviously structures the nature of the rela-
tionships that people have with mountainous landscapes. In the Middle
Atlas Mountains (1,700 m), the palynological evidence does not suggest
an early anthropogenic influence on the vegetation, although there is
some evidence for Neolithic impact, and tentative evidence for brows-
ing during the first millennia BC and AD. Pastoral activity was certainly
taking place from the fourth century AD onwards (Lamb et al. 1991;
Lamb & Van der Kaars 1995). Pine was also cleared periodically about
2,000 years ago, and then about 500 AD, as pastoralism was replaced
by arable agriculture (Lamb et al. 1991). It seems likely that pre- and
proto-historic peoples were active in North African mountain ranges,
but the archaeological and palaeoecological evidence rarely implies high
levels of activity, or these were forms of activity that left ephemeral traces
in both the archaeological and palaeoecological records. However, as we
move into the Roman period, the nature of society’s relationships with
some of these areas did change. Palynological work from Tigalmamine
(1,626 m) in the Middle Atlas reveals human impact in this area from
the third century BC when there was a decline in Fraxinus (ash) (Lamb
et al. 1989: 72).
One important advantage of settlement in some North African moun-
tain ranges is the relatively large quantities of water present therein. In
the Libyan Sahara, where wet conditions prevailed from 4900–4400 BC,
the inner mountainous areas appear to have been the preferred landscape,
where relatively high levels of pastoral activity took place. Then, ‘severe’
dry conditions developed from c. 3000 BC, and this appears to be cor-
related with a reduction of activity in the mountains (Cremaschi & Di
Lernia 1999: 216). During the Roman period, the capture and control
of waters emanating from the mountains was essential for the successful
development of economies in lower-lying areas (Trousset 1986).
The impact of Imperial Rome on mountainous zones across the

Mediterranean is of fundamental interest, as such zones, which are
often on the edges of empires, may have developed characteristics that
are quite different to the Romanised lowlands. In some instances, they
maintained a level of cultural and economic independence, with environ-
mental trajectories that were in many ways a continuation of Iron Age
processes. In other areas, mountainous zones emerged as fully incor-
porated landscapes. As Roman agriculture and landscape organisation
was imbued with many different systems of technological and ideological
control (the two being inherently intertwined), we need to consider the
extent to which these systems (initially developed in the more hospitable
landscapes of Italy) were successfully transposed onto these decidedly
different mountainous zones across the Empire. For some time, ancient
historians and archaeologists considered mountain societies as detached
from those on the plains. This notion was particularly prevalent amongst
French geographers and historians in the French North African colonies,
an attitude which in many ways merely reflected the reality of a French
colonial economy that itself avoided the mountainous zones, consider-
ing them marginal (Leveau 1977, 1986). In this view, Roman civilisation
spread like water; it invaded the plains without covering the mountains
(Courtois 1955: 121).
In the low mountainous zone, or high steppe (800 m and above)
around Kasserine (Tunisia), a large number of Roman sites were estab-
lished. Based on the palaeobotanical work that has been carried out in
this area, it seems that this zone has been best suited to nomadic pasto-
ralism. Even though this landscape may well have been slightly different
to the present steppe environment, it certainly appears that pastoralism
was the dominant activity prior to the arrival of Rome (Hitchner 1988:
8). Essentially agricultural in function, many of these Roman sites were
involved in oleiculture as well cereal and pulse production. The amount
of olive oil potentially produced in this area would have exceeded local
needs, whilst crop production was probably for local consumption.
Animal husbandry was also a significant activity. One important element
in the development of this landscape would have been the construc-
tion and/or maintenance of terrace systems, essential for enhancing
and protecting the soil required for crop production in semi-arid zones
(Fig. 8.3) that are highly susceptible to erosion during extreme pre-
cipitation events. A number of cemeteries formed part of this Roman
hange

8.3. Djebel Chambi near Kasserine (photo: Bruce Hitchner).
landscape, whilst zones that included earlier burials under cairns appear
to have been what Hitchner (1990: 244–5) refers to as ‘negative space’ –
areas that remained ‘marginal’ to the core Empire. Although research
into North African mountainous areas has not been as intensive as other
parts of the Mediterranean, what we do see is a heterogeneous pattern
of activity during the Holocene, with some evidence for Rome in par-
ticular adapting indigenous environmental knowledge and linking core
economic and social processes in these areas, but not necessarily inte-
grating them.
Greece and Anatolia

The Pindus Mountains comprise parallel limestone ridges that extend
from Albania down to the Gulf of Corinth. This part of the Mediterranean
is characterised by complex, extremely varied topography, from the hin-
terland right down to the coast (Fig. 8.4). Much of this landscape is
seemingly barren, but people have settled in those ‘ecological islands’

8.4. The coastal zone before the Pindus Mountains, north-west Greece. In many Mediterranean regions, the
mountains are never far from the coast, creating complex topographies and niches that are often difficult to
exploit (photo: author).
nestled between the ridges and peaks of the mountains. Whilst geological
formations do not dictate behaviour, the topographical characteristics
of a mountainous region do considerably influence the ways in which
people move across a landscape (Llobera & Sluckin 2007). We have to
consider if geological and topographical configurations have any spe-
cific cultural ramifications. Whilst the structure of valley systems, passes,
and ridges do influence mobility and links that people have with differ-
ent places (towns, villages, farms, pastures), this spatial organisation of
the mountainous zone only becomes relevant if we can truly demon-
strate that the day-to-day human experience of life in this zone would
have been different had the geological and topographical characteristics
been different. There is no doubt that topographical configuration does
influence settlement and the ways in which spaces are used and engaged
with. Ecological zonation, and therefore the limit for vegetation growth,
obviously varies across Mediterranean mountain ranges. The altitude
at which good quality pasture will grow also differs. These and other
hange
factors affect the maximum height at which village or hamlet settlement

occurred. In the Pindus Mountains, the highest village (Samarinia) is sit-
uated at 1,650 m. In the Grevena highland zones (the Pindus chain of
western Macedonia, Greece), regular use of these upland areas may well
have started at the end of the Neolithic, with pastoral activity probably
practised during the Middle to Late Bronze Age. This is supported by
some now rather dated palynological evidence that suggests a change in
vegetation during the Bronze Age (Higgs 1978). However, more recent
work from the northern Pindus Mountains (the Rezina Marsh site) indi-
cates human impact from as early as 6,000 years ago, when there was a
rapid reduction in tree diversity and density, although Acer (maple) and
Carpinus betulus (common hornbeam) increased (proportionately). A
thousand years later, some woodland species did increase. Then from
about 4,000 years ago, forest clearance took place, and the increase in
Rumex (docks) implies that this later phase was one characterised by
anthropogenic impact (Willis 1994). Human activity appears to have
taken place at a range of altitudes, between 285 m and 1,800 m (and
probably above and below these heights, as allocthonous pollen would
have been present in these cores). The study of the soils in the area sug-
gests that the weakly developed profiles are a product of human distur-
bance of vegetation, and consequent churning and soil creep. Despite
this, there is no evidence to suggest that this has prevented pastoral activ-
ity, although it may well have reduced flock capacity in the area. The con-
stant cycle of activity in this fragile environment is thus deemed to have
put pressure on the landscape (Efstratiou et al. 2006: 430). When Rome
imposed itself upon the area, the enslavement of a large proportion of
the population seems to have led to a long-term decline in activity in the
Pindus (McNeill 1992: 78). Despite the fact that there is little archaeo-
logical evidence that allows us to test this hypothesis, such a process does
correspond with what we see in other mountain ranges.
Other evidence from the northern mountainous zones of Greece
shows how deciduous oaks and lime dominated the Middle Holocene
woodland. Between c. 2130 and 1320 BC, Abies (fir) and Pinus gradually
replaced these trees, whilst Fagus (beech) started to expand. Eventually,
from about 800 BC at the lowest sample site, and later in the higher
altitude sites, Abies became almost extinct and Fagus became dominant.
Juglans (walnut) and Castanea (chestnut) also appear during this period
(Gerasimidis & Athanasiadis 1995: 113).
On the extreme northern edge of this subregion, the Galičica National

park (on the border of Macedonia and Albania), with mountains as high
as 2,250 m, includes Lake Ohrid. This lake possesses some of the most
important palaeoenvironmental archives in Europe, and they comprise
a range of proxy indicators for both climate change and human impact
on the surrounding landscape. Whilst specific archaeological evidence
for the exact nature of human activity in the mountains around the lake
might be absent, or as yet unrecorded, pollen, fungal spores, and sed-
iment chemistry, including lipid biomarkers, provide a broad image of
the development in human activity around Lake Ohrid. Coprostanols
(a lipid biomarker that is indicative of faecal matter) suggest the pres-
ence of people and animals around the lake from more than 8,000 years
ago (Holtvoeth et al. 2010). Incontrovertible evidence for the impact
of people in this area is attested to (by pollen and fungal spores) from
c. 3000 BC, with the landscape undergoing a significant transforma-
tion (probable opening up for agro-pastoral activities) from c. 400 BC
(Wagner et al. 2009). Whilst it is possible that these signals represent
lakeside activity, it may be that they are the consequence of increased
human activity in the upland areas of the lake catchment. As discussed
later in this chapter, an increase in upland activity is a phenomenon that
we see in many mountainous areas around the Mediterranean during the
Chalcolithic and Bronze Age.
The Cretan Mountains

Archaeological research is rare in the mountains of the major
Mediterranean islands, although research on Crete provides a useful
exception. Today, the upper limit of permanent settlement on Crete is
800 m (Rackham & Moody 1996: 93). The Sphakia survey covered a
district in south-western Crete, an area which covers a number of differ-
ent zones from the coast up to the White Mountains. The highest point
in the Cretan White Mountains (Lefka Ori) is 2,453 m. What is more
striking is that this peak is only 16 km from the coast, and therefore
presents people with an incredible variation in topography over a very
short distance.
Prehistoric pottery was found over all of Sphakia, suggesting an
‘extensive, not intensive, exploitation of the entire area during this
period’ (L. Nixon et al. 1990). Whilst there are a large number of pre-
historic sites comprising artefact scatters and structures such as cisterns,
hange
precise dates and the specific function of this upland area are difficult to
ascertain without excavation, although the role of herding in the later
proto-historic economy appears to have increased in the Cretan moun-
tains (Nowicki 1999). A possible reduction in activity at higher altitudes
occurred during the Graeco-Roman phase, a period when settlement
was concentrated down towards the coast (Moody & Nixon 2006).
Although documentary sources suggest an increase in pastoral and
associated upland activities during the Classical and Hellenistic period
(Chaniotis 1999), we do not know where and at what altitudes. Despite
the archaeological evidence for high-altitude human activity, the paly-
nological evidence that exists implies no significant vegetation change
in the White Mountains prior to the second Byzantine period (eleventh
to twelfth centuries AD) (Atherden & Hall 1999). The apparent low
level of impact on vegetation may be due to the catchment from which
the pollen core was taken, or human activities were such that impact was
low.
At an average altitude of 840 m, the Lasithi plain (eastern central
Crete) in the Dikataian Mountain range is encircled by peaks that reach
an altitude of almost 2,150 m. One difference between this area and
other mountainous zones discussed in this chapter is that arable agricul-
ture has probably been a key element in the local economy due to the
relatively low mean altitude. The earliest sites date to the Late Neolithic
or Early Minoan periods, and probably represent seasonal pastoral activ-
ity (Watrous 1982: 10). Activity in the plain continued to develop during
the Minoan period. However, the evidence for the Late Minoan I period
is quite ephemeral, and may be explained by emigration towards the prin-
cipal Neopalatial sites (ibid.: 15). Population and activity levels seemed
to contract throughout the rest of the Late Minoan periods. During
the Geometric and Archaic periods, settlements re-emerged around the
edge of the plain. However, the plain was largely depopulated during the
Classical and Hellenistic periods, perhaps due to a perception that such
upland zones were unproductive. There may well have been other polit-
ical and cultural reasons for this reduction in activity. During the Roman
period, the plain was repopulated, with some sites situated on the allu-
vium rather than on the slopes (ibid.: 24).
In Crete, the work that has been done on mountainous landscapes does
reveal patterns that we will see repeated as we move into the Alps and the
Iberian mountains. Cretan societies made use of the high-altitude zones
Italy, France, and Spain 255
during the later prehistoric periods. However, impact on the mountains

was not continuous, as there seems to be a withdrawal from the highest
altitudes during the Roman period, or a change in the way these zones
were exploited.
Italy, France, and Spain (the Apennines, Alps, and Pyrenees)
As we move into the Euro-Mediterranean mountain ranges, the amount

of palaeoenvironmental evidence and integrated archaeological research
increases. This alone is the rationale for concentrating on projects from
this part of the Mediterranean.
There is no doubt that parts of the Northern Mediterranean wit-
nessed a change in climate c. 4,000 years ago. There are variations in
the characteristics of this climate change, and ostensibly subtle differ-
ences between the Apennines and the Alps to the north. The Calderone
Glacier is Europe’s southernmost glacier (situated in the Apennines).
This re-formed during the latter half of the third millennium BC, and
subsequently expanded during the eighth century BC and then again
during the seventh century AD (Giraudi et al. 2011: 107). The Bronze
Age is when many mountainous zones were incorporated into struc-
tured, economically productive, vertical landscapes – a later phase within
the Mediterranean secondary products revolution (Greenfield 2010).
As recently observed, this phase, which marks the transition between
the Middle and Late Holocene, is characterised by a complex pattern of
wet/dry oscillations (Magny, Galop, et al. 2009). The key issue for the
Apennines, and other mountainous areas, is that different palaeoenvir
onmental records support the inference of a short cool period (glacial
advance and lake-level increase) that was followed by cool and arid condi-
tions (glacial advance and lake-level decrease) (Giraudi et al. 2011: 112).
Moreover, the climatic oscillations that characterise this period some-
times took place over relatively short periods of time and would have had
significant consequences for high-altitude landscapes and the resource
base available to the people who were starting to live and work at these
altitudes in greater numbers (Giraudi et al. 2011). These relatively quick
changes in climate are important, as they, as much as anything, would
have tested the ability of mountain communities to adjust and develop
their environmental knowledge vis-à-vis changes in precipitation, the
length of growing seasons, and associated environmental hazards.
hange
In the Ligurian Apennines (north-western Italy), a complex series

of human–environment relationships developed during the fourth
and third millennia BC, with the establishment of the earliest western
European copper mines in this area. Here, we see evidence for a practice
that became widely used across mountain ranges – the use of fire in order
to clear woodland for pastures, plus the use of wood in the mining pro-
cess. These initial clearances of woodland, where there was little or no
grass or understory to protect the soil, could have resulted in substantial
erosion. Such early phases of woodland clearance may well have created
some of the most unstable or sensitive environmental conditions during
the entire Holocene. Extensive pastoralism developed throughout the
Copper and Bronze Ages. At Prato Mollo (1,480 m), tree cover declined
as a result of fire between 3079 and 2642 BC, and between 2889 and
2472 BC. Lower down towards the coast, at 830 m, the site of Pian
del Lago produced evidence of local cereal production prior to 5000
BC (De Pascale et al. 2006: 116–17). If people were using fire to man-
age vegetation, were the Copper and Bronze Age uses of fire at higher
altitudes the result of differing attitudes towards different altitudes, or
due to the application of a more extreme and haphazard measure in the
higher altitudes, where the potential risk to settlement and arable lands
was minimal (ibid.)?
Palynological work at Prato Spilla (Emilia-Romagna), at an alti-
tude of 1,500 m, indicates that, at the start of the Holocene, this area
was covered with an Abies–Pinus (fir–pine) woodland that was soon
replaced with a Quercus–Abies (oak–fir) woodland with some Corylus
(hazel). Human impact may have started during the late fifth millen-
nium BC, when tree pollen totals start to decrease with reductions of
Ulmus (elm) and Fraxinus (ash). This period may reflect forest clear-
ance that was taking place lower down in the valley towards 700 m.
The next ‘anthropogenic event’, during the early fourth millennium BC,
comprised a change in the composition of the woodland, in particular
the establishment of Fagus (beach); the extreme reduction in Abies; the
disappearance of Fraxinus, Tilia (lime), and Ulmus; the re-emergence of
Betula (birch); and the first appearance of Olea (Lowe et al. 1994: 157).
The end of the Neolithic and the start of the Copper Age saw burning
and soil disturbance across Liguria, and it seems plausible that this type
of disturbance was taking place at Prato Spilla as well. Moreover, the
appearance of some nitrophilous herbs at this time is indicative of human

activity. About 4,000 years ago, changes in the composition of the wood-
land continued apace (Lowe et al. 1994), and may have been related
to human activity which is known to have taken place up to altitudes
of about 1,500 m from the Neolithic into the Bronze Age in parts of
Liguria (Maggi 1999: 30).
The Alps
Even though the alpine arc covers countries other than Italy and France,
these nations (Switzerland, Austria, and Germany) are not defined as
Mediterranean. Therefore, the examples presented here are taken from
the two principal Mediterranean Alpine countries: Italy and (southern)
France. Ecologically (based on climate and vegetation patterns), some of
these French and Italian alpine areas would be defined as Supra- to Oro-
Mediterranean (Quézel 2004) (see Fig. 5.4).
As already stated, the interpretation of human activity in mountains
has often cited environmental change as a driver of developments in
human activities. Although there have been periods of relative cooling
and warming, the Holocene has seen a steady overall increase in tempera-
ture, but human activity in the high-altitude zones has waxed and waned
rather than continuously increased. The analysis of stable isotopes from
stalagmites from the south-east Italian Alps demonstrates this increase in
temperature, with warmer periods during the Roman and high medieval
periods (Frisia et al. 2005). However, the relatively cool early stages of the
Holocene in certain valleys did not deter activity. There is now no doubt
that people moved into the high-altitude zones for hunting early on dur-
ing the Holocene. In the Venetian Pre-Alps, a palynological core from
the Palughetto basin at about 1,000 m and archaeological sites between
1,900 m and 2,400 m demonstrate that late Early Mesolithic peoples
were in this area, with hunting activity occurring in a newly afforested
environment. Despite evidence for Palaeolithic and Mesolithic activity,
there is no archaeological evidence for a Neolithic presence (Avigliano
et al. 2000: 798).
The upper limit of activity in the Alps is not only influenced by the
waxing and the waning of the tree line, but the ebb and flow of glaciers is
also important. Activity at the edge of a glacier is possible, but an advanc-
ing glacier destroys pasture and, as we know from the Little Ice Age, can
hange
even destroy villages, although this was less true in the more southern
Alpine areas. As far as the Neolithic period is concerned, a phase of gla-
cier advance occurred in the eastern Alps between 3350 and 2700 BC,
and readvanced during the Bronze Age, possibly centred on the period
1930 to 1250 BC. This was followed by an Iron Age readvance (which is
difficult to date precisely due to the flat zone on the 14C calibration curve
for this period) between 980 and 370 BC (Grove & Rackham 2001:
147). Quite early in the Holocene, the Forcellin–Foscagno glacier in the
Italian Central Alps (between 2,800 and 2,500 m) completely retreated,
with evidence for a glacial advance about 3,000 years ago. The glacier
once again retreated during the mid-first millennium BC, and permafrost
formed (Guglielmin, Cannone, & Dramis 2001).
In the Val Febbraro, upper Valle di Spluga (Italian central Alps),
Neolithic to Roman activity is attested to at altitudes between 1,830
and 2,304 m, inferred from both archaeological and palaeoecological
evidence. The upper limit of this activity probably existed at the extreme
edge of viable vegetation. The presence of charcoal dated to the early
Neolithic at the timberline at Lago Basso may represent human impact
on the forest (Wick 1994). Similar evidence is present in neighbour-
ing valleys (Fedele 1992). There was general reduction in forest cover
during the Neolithic and into the Bronze Age, along with a lowering
of the tree line. Species such as Rumex acetosa (common sorel or spin-
ach dock) increased. Whilst archaeological evidence is quite common
below 2,000 m, the research in this area has produced four finds between
2,000 and 2,500 m. During the Middle Bronze Age, there were notable
changes in the high-altitude vegetation, with the development of grass
meadow. There are some archaeological sites, but their precise function
is unclear. These later Bronze Age people either collected or burnt wood,
and used the area for grazing. Whilst the period before 1000 BC is con-
sidered to have been warm and dry, climatic deterioration followed on
from this period. Despite this, human activity in the high-altitude zone
continued. The Iron Age in particular saw activity at the full range of alti-
tudinal zones within this area. However, the Roman period saw a recov-
ery of the woodland (at Val Vidröla, 2,235 m, and at Borghetto Sotto,
1,897 m). This trend is seen in other areas, such as at Lago Grande west
and Val di Starleggia, about 1,830 m (Moe & Fedele 1997; Moe et al.
2007). This seemingly surprising trend of relatively low Roman impact
on some high-altitude areas is considered in detail later.
Southern French Alps

The Southern French Alps is one of the few mountainous zones where
extensive palaeoecological research – with a pedigree dating back to the
1970s (de Beaulieu 1977) – can now be incorporated with archaeologi-
cal evidence. Unsurprisingly, the palaeoecological work is dominated by
palynology, and provides us with a detailed history of vegetation in these
supra-Mediterranean valleys. Palynological reconstructions of changes in
the different vegetation belts at different altitudes are problematic, as
the wind can easily transport pollen up to mountain lakes from the valley
bottom, bringing up pollen from all of the vertical zones (Ortu, Brewer,
& Peyron 2006: 622–3). However, when combined with anthracology,
an assessment of timberline movement is possible. Moreover, these spa-
tially specific data allow the identification of choices of wood made by
people for their fires, and provide evidence of localised burning and man-
agement activities.
The composition and vertical extent of the forests and woodlands do
not just provide the floral context for human activity, but allow us to
characterise one of the key environmental elements with which people
have engaged during the Holocene. Trees are not only a valuable eco-
nomic resource, but they also influence the way in which the landscape is
perceived. Woodland affects movement through the mountains and the
nature of relationships with other animals, including wolves and bears –
animals that pose a threat to humans and their livestock. Miners from
the Copper Age onwards would have used wood and timber in a whole
range of mining-related practices, including fire setting for the extraction
of copper ore (Barge & Talon 2012; Bourgarit et al. 2008). Ultimately,
the earliest pastoralists managed the woodland and created openings for
pasture. Although open pasture might have been preferred, access to this
would have been controlled by the natural limit of the forest – the tim-
berline and the tree line. The limits of the forest are an ecotone, not only
beneficial to pastoralists but also to hunters, the edge being the zone
where prey becomes clearly visible and easier to kill.
The composition of the Middle Holocene forest in the Southern Alps
changed quite dramatically: Abies alba (silver fir) suffered two phases
of decline, the first at c. 5000 BC and the second at c. 4000 BC. This
second phase was succeeded by localised extinctions of fir trees in areas
below 1,000 m. The probable primary cause of Abies decline was the
reduction in moisture supply due to climate change, human impact, and
hange
fire, which presumably could have had both natural and anthropogenic
causes (Wick & Möhl 2006). However, it is unlikely that Mesolithic
hunters created clearances via burning. Early Mesolithic charcoal within
travertine deposits at 1,750 m in the southern French pre-Alps is unlikely
to be the product of fire setting by hunters (Roiron et al. 2006).
Pinus cembra is one key species that has gradually retreated from its
highest altitudes. When it was first present in the Southern French Alps
some 9,000 years ago, it attained altitudes 375 m beyond the present-
day tree line. In any valley, the tree line is directly influenced by glacier
configuration, and in the Ubaye, glaciers may well have totally melted
between 8,000 and 6,000 years ago (Assier 1996: 4). Fire events dated
to the mid-sixth and late fifth millennia were probably started by light-
ning, as there is no evidence for human manipulation of the forest at this
time (Touflan, Talon, & Walsh 2010). It is assumed that human activity
from the Neolithic onwards contributed to the contraction of the for-
est (Ali, Carcaillet, et al. 2005). However, the possibility that Neolithic
hunters and pastoralists had such an impact on the forest in the high-
altitude zone is a hypothesis that requires testing. In the Ubaye Valley
in the Southern French Alps, the tree line (dominated by Pinus cembra
(Swiss pine)) was at close to 2,400 m during the Neolithic, descended
to 2,200 m during the Bronze Age (the altitude at which the Bronze
Age structures, discussed below, are often found), and then returned to
2,400 m during the Iron Age (Ali, Roiron, et al. 2005).
The spatial and chronological heterogeneity of human activity
and vegetation change is rarely inferred from single pollen diagrams.
Fortunately, research in the Southern French Alps (centred on the Ecrins
National Park) demonstrates how the analysis of records from a range
of altitudes permits the diachronic assessment of human activity across
the full range of altitudinal zones (Court-Picon 2007; Richer 2009). It
is apparent that Neolithic populations were actively farming and man-
aging the valley bottoms, and that human impact on the high-altitude
(or subalpine and alpine zones; those areas above 2,000 m) landscapes
did not develop until the very end of the Neolithic and Chalcolithic. As
soon as the glaciers retreated at the start of the Holocene, early hunt-
ers followed their prey up into these areas, probably during the summer
months. Hunting continued in these areas even after the establishment
of agricultural groups at lower altitudes (Walsh & Mocci 2011). The
highest find in this area is a flint arrowhead at 2,510 m on the Col de
Roburent: located on a large plateau dominated by two lakes, an area
Puy-St Vincent Grand Founze VI

>3360 (2614 m)
>3120 Grand Founze I La Folie IV
>2880 (2210 m) (1880 m)
rnel
>2520 Fou
Col du l’Aulp Maurin Haute Vallée du Fournel - ent du
>2280 (2761 m) torr
Lac des Lauzons Serre de l’Homme
>2040 Serre de l’Homme II
>2040 (2190m)
(2208m)
>1800 Col du Pas de la Cavale L‘Argentière-la-Bessée
>1560 (2735 m)
>1200 Serre de l’Homme XI
(2252-2262 m)
Serre de l’Homme X
(2220 m)
Freissinières
Haute Vallée de Chichin
chin
Col d’Orcières torrent de Chi
(2782 m) Dormillouse
Champoléon Chichin III
Chichin II
(2080 m)
(2230 m)
Parc national des Ecrins
Fangeas Faravel VIIId
a Plateau
Dr
Col d’Orcières
(2120 m)
c
(2782 m) et montagne
de Faravel Faravel XVIII
(2130 m)
Jujal Faravel XIV
(2140m) (2450 m) Faravel XXXIII
(2133 m)
Faravel XIX
(2303 m)
Col du Palastre Faravel XIII
isil
(2200m) Prapic (2110 m)
Bla
Col des Terres Blanches
(2721 m)
Orcières
Dr
ac
ac
Dr
St-Jean St Nicolas 0 1,5 km

V. Dumas, Fl. Mocci, K. Walsh Centre Camille Jullian CNRS-Univ. de York 2012
Sites or indications of sites with Pastoral structures and dated by carbone 14 Pollen cores
individual lithic flint (Late third millenium BC to 4th c. AD) Lac des Lauzons (2190 m)
Tourbière de Fangeas (1990 m)
Late Néolithic/Early Bronze Age Middle/Late Bronze Age
Prehistoric sites or indications of human activity Lac-tourbière du Serre de l’Homme (2234 m)
Early Bronze Age (hunting) Iron Age
Late Paléolithic Mésolithic
Late Bronze Age Roman Period

Middle Néolithic Late Néolithic
8.5. Distribution of sites across the high altitude zone of the Ecrins national park (figure: F. Mocci and K.
Walsh).
that would have been a rich hunting zone, partly as a higher tree line
would have provided the ideal niche for game. This situation is quite
similar to that at the Grand Founze VII site, Puy-St-Vincent at 2,600 m.
In the Ecrins, there are about 10 Neolithic sites that comprise flint scat-
ters; some are quite ephemeral. A Late Neolithic arrowhead at 2,450 m
in the Ecrins once again implies the continuation of hunting at high
altitudes during this period. The nature of human engagement with the
high altitudes changed dramatically during the late third millennium BC.
From c. 2500 BC, the first built structures appeared at altitudes between
2,100 and 2,400 m. Most of these sites comprise a similar architectural
form, often exploiting the presence of naturally occurring erratics and
large boulders. These sites have been interpreted as animal enclosures
(Tzortzis et al. 2008; Walsh 2005). There are now 12 well-dated sites
of this type, plus at least 30 other typologically similar sites from across
the Ecrins National Park (Fig. 8.5). In the Ubaye Valley, 23 prehistoric
hange
sites have been found, all above 1,200 m (up to 2,509 m). The majority
of the 23 prehistoric sites tend to be located on one of the five axes that
lead to and from the Mercantour area which includes Mont Bego. The
prehistoric sites tend to be situated on moraine, plateaux, or at the con-
fluence of water sources. Palynological evidence for the third and second
millennia BC indicates an intensification of pastoral activities (Court-
Picon et al. 2007; Mocci et al. 2009; Walsh et al. 2005). During the
Neolithic, such activities were concentrated towards the valley bottoms
and lower slopes. Then, during the third millennium, these activities
expanded, moving into the high-altitude areas. On certain high-altitude
sites, fungal spores and nitrophilous plant pollen indicate Bronze Age
pastoral activity in the immediate vicinities of these sites (Walsh & Richer
2006). Moreover, charcoal evidence from the pollen cores, along with
charcoal from sediments found within and around the archaeological
sites, is indicative of woodland clearance designed to create more suitable
pasture. Charcoal and palynological evidence from one series of Bronze
Age sites demonstrates how the low-lying area at the foot of a moraine
would have seen standing water during extreme rainfall events. All the
structures were thus located on the moraine, above this intermittently
wet zone (Fig. 8.6). Moving into the Iron Age, there appears to have
been a reduction in high-altitude activity. Although some palynologi-
cal signals imply continued human impact on these landscapes during
the first millennium BC and into the Roman period, there is a strange
absence of archaeological sites in the high-altitude zone, but a radical
reconfiguration of the valley bottoms, with the development of towns
and associated communications. Decisions to change the nature of high-
altitude pastoral activity may have been influenced by climatic deteri-
oration during the Middle Iron Age, although we must also consider
the importance of changing cultural perceptions of the mountains. The
sheer range of data types that indicate a relatively warm climate during
the Roman period should leave us in no doubt as to the general char-
acteristics of climate at this time (Frisia et al. 2005; Mangini, Spötl, &
Verdes 2005; Schmidt, Kamenik, & Roth 2007). It is also quite possible
that the Roman period was more moist than other periods, which not
only has consequences for vegetation but, more importantly, for weather
and in particular the probability of summer storms (Reale & Dirmeyer
2000; Reale & Shukla 2000). The Roman climatic ‘amelioration’ could
have comprised an increase in extreme precipitation events and storms
X=1000
X=1100
X=1200
X=900
Y=2300
Y=2300
torrent de Narreyroux
2225.00
Y=2250
Y=2250
Y=2200
Y=2200
2234.00
tourbière
Y=2150 Y=2150
5.00
223
Y=2100
Y=2100
SDH X
2250.00
Y=2050
Y=2050
SDH XXI
SDHXIX
SDHXIh
SDH XIII
.00
50
22
SDHXIa
SDHXIc
SDHXIg
SDHXIf
.00
SDHXIh 40
0
SDHXIb Y=2000
.0
Y=2000 22
0
3
2
2
SDHXId
0
.0
0
2
2
CT1
2
0
.0
60
22
Y=1950
Y=1950
SDH XX
0 25 50 75 100 m
V. Dumas, CCJ-CNRS/AMU, 2009
X=1000
X=1100
X=1200
X=900
8.6. The Serre de l’Homme area in the Ecrins National Park (Southern French Alps) at about 2,200 m. A
complex of Bronze Age enclosures located along the entirety of the moraine. Despite being more exposed to
winds, this location avoided the wet zone at the foot of the morraine (figure: V.Dumas, F. Mocci, and K. Walsh;
photo: author).
Alpes du Sud - Ubaye
Indetermined
Alpine
M e d ieval / Modern
G a l l o R o m an
Protohistori c
2300-4000 m
Prehistori c
Indetermined
Sub-Alpine
M edieval / Modern
G a l l o R o m an
1700-2300 m Protohistori c
Prehistori c
Montagnard
G a l l o R o m an
Protohistori c
Prehistoric
1000 -1700 m
G a l l o R o m an
Colinnéen
Protohistori c
500 -1000 m
5 10 15 20 2 5%
V. Dumas, Fl. Mocci, Centre Camille Jullian UMR CNRS-AMU, 2012
2 5 3 6m 2529 m 2561m
2 485m
2359m
M oy :
2278m M oy :
M oy : 2264m
32.08%
2238 m
1955m
Indetermine d
( a gro -p a s t ora l sites)
2 5 .4 8 % M oy :
M oy :
1508m
1423m
1245m
20.75%
Prehistoric
( set t lemen ts, 1210m
12.26% 9,43%
rock shelters,
in d iv id u al fi nds) M e d ieval/ M o d ern
790m 790m ( a gro -p a st oral sites)
Protohistoric G a l l o R o m an
(agr o- pastoral si tes, (si tes,
bur nt mound, rock shelters,
i ndi vi dual fi nds) i ndi vi dual f i n ds)
V. Dumas, Fl. Mocci, Centre Camille Jullian UMR CNRS-AMU, 2012
8.7. Altitudinal zonation of sites by period in the Ubaye Valley – alt. 700–2,650 m, Alpes de Haute
Provence, France. (Programme collectif de Recherche 2001–2006, dir. D. Garcia and Fl. Mocci, Centre
Camille Jullian). Prehistoric includes Mesolithic and Neolithic flint; proto-historic tends to be Iron Age
pottery, with some possible Bronze Age pottery along with enclosures which typologically appear to be
Bronze Age. Note that the chronology of these sites is usually determined by the nature of the mate-
rial culture found on the soil surface. At lower altitudes, architectural remains of Roman sites are also
present.
at high altitude caused by increases in convective currents – a product

of a warmer climate (Giguet-Covex et al. 2012). Such weather patterns
may have also conditioned certain responses to life at high altitude, with
people avoiding these zones at certain times of the year. The relative
absence of Roman archaeological sites at high altitude is repeated further
south in the Ubaye Valley (Garcia, Mocci, & Walsh 2007) (Fig. 8.7).
Non-Romanised people might have continued to use the higher altitude
areas at this time, whilst the main focus of activity shifted to the lower
altitudes and urban centres. These changes include the development of
towns, communication networks, and ‘stations’ that provide interme-
diate resting places for travellers moving around the Empire. One such
station is that at Rama, at the foot of one of the valleys studied as part of
research in the Ecrins National Park. The presence of spores associated
with soil erosion, dated in a pollen diagram from a lake next to this site,
also suggests phases of erosion during the Early Roman period (Farjon
2007; Richer 2009), this erosion probably being caused by one or more
extreme precipitation events.
The Pyrenees
The results of recent research in the Pyrenees mirrors some of the gen-
eral processes described above for the Southern French Alps. Research
on both the French, Spanish, and Catalan areas of the Pyrenees has
incorporated detailed palaeoenvironmental work, combined with archae-
ological survey and excavations. As with the Alpine research, understand-
ing of environmental change and human activities during the Holocene
has dramatically improved during the last 15 years (Ejarque et al. 2010;
Mazier et al. 2009; Palet et al. 2007; Rendu 2003).
In the French Pyrenees, the earliest evidence for agro-pastoral activ-
ity dates to the end of the Early Neolithic (c. 4900 BC) through to the
first half of the Middle Neolithic (4400–3800 BC). During this period,
people were exploiting the lower and mid-range altitudes, and whilst
impact on the subalpine zone was minimal, it seems that there was some
activity taking place there as well. During this period, high-altitude veg-
etation would have adjusted to changes in climate. We know that in the
eastern Pyrenees, thermomediterranean-type vegetation developed dur-
ing the Middle Neolithic and comprised Olea europaea, Pistacia lentiscus,
Phillyrea sp., Cistus sp., Rhamnus alaternus (Mediterranean buckthorn),
and Quercus ilex/coccifera (Heinz et al. 2004). As with the Southern
hange
French Alps, it was the late third and second millennia that saw the devel-
opment of human activity at higher altitudes. This is when anthropogen-
ically modified mountainous landscapes started to develop, when forest
was cleared through a process of slash and burn (Galop 1998: 254–5).
The later Iron Age and the Roman period are interesting phases in the
Pyrenees, as there was probably some continuity in the exploitation pat-
terns established during the Bronze Age, but the dramatic changes wit-
nessed in other landscapes across the Roman Empire did not manifest
themselves in the Pyrenees. In some areas, such as Vicdessos, Cerdagne,
and Donezan, there appears to have been a reduction in human activity
during the Iron Age and perhaps at the start of the Roman period. A sim-
ilar sequence of events has also been identified on the Enveig Mountain.
With massifs above 2,500 m, its highest pastures are situated between
2,200 and 2,600 m. The earliest activity here is dated to c. 3000 BC,
implied by the appearance of Plantago lanceolata (ribwort plantain), as
well as evidence for the opening up of the forest by burning as high as
2,100–2,300 m (Davasse, Galop, & Rendu, 1997: 587). Between the
fourth century BC and the first century AD, there appears to have been a
hiatus or at least a reduction in high-altitude pastoral activity in this area.
The anthracological evidence indicates localised opening of the forest
between the first and eighth centuries AD. Pastoral activity was proba-
bly reoriented towards the low-altitude mountainous zones during the
Roman period (ibid.: 591).
On the southern-central (Andorran) area of the Pyrenees, evidence
for early high-altitude pastoralism is dated to c. 5000 BC onwards
(Ejarque 2010: 241; Ejarque et al. 2009). The Neolithic pastoral hut
in the Madriu Valley (2,530 m), dated to the mid-fifth millennium BC,
provides supporting archaeological evidence. Slightly later, the relatively
cool period c. 5,500 years ago (as seen by Magny & Haas 2004) would
have resulted in a more open forest suitable for grazing, and may well
have attracted Neolithic pastoralists to this area (Ejarque et al. 2010).
Pastoral activity appears to have been taking place between 4300 and
3500 BC at around 2,500 m, as indicated by coprophilous spores and
apophytes. Moreover, the earliest stone-built structures in this area are
dated to the late fourth to early third millennia BC (Orengo 2010).
As in the Ecrins, there is also evidence for burning in some Pyrenean
Neolithic landscapes, with consequent erosion caused by this defores-
tation. A substantial increase in pastoral indicator species at 2350 BC,
along with a number of Late Neolithic and Early Bronze Age high-alti-
tude sites in the Vallé de Madriu, suggests a significant increase in pasto-
ral activity (Ejarque 2010: 66). In some areas, the Bronze Age saw lower
grazing pressure than during the Middle Neolithic, and this waxing and
waning of activity is reflected by woodland recovery. More specifically,
we see a reduction in pastoral activity in some areas during the Middle
Bronze Age, this having been preceded by the most intensive phase of
human activity between 1850 and 1750 BC. This phase was followed by
renewed activity in some areas between 1650 and 1050 BC. Here, we
see how problematic the traditional ‘evolutionary’ model is; we cannot
assume that human impact on a landscape increases continuously from
the Neolithic onwards (Ejarque et al. 2010).
In the French Pyrenees, there is evidence for Chasséen activity at alti-
tudes of between 1,600 and 1,800 m (Rendu 2003: 420), with more
intensive activity developing towards the end of the Neolithic and into
the Chalcolithic. However, there is relatively little archaeological evi-
dence in the high-altitude zones above 2,000 m. The Bronze Age wit-
nessed the first obvious structuring of some Pyrenean mountainsides
(ibid.: 513), with grazing emerging at the end of this period as an impor-
tant practice between 1050 and 700 BC. Between 700 and 400 BC, an
increase in Pinus is inferred as evidence for reduction in pastoral activity
at high altitudes. In the eastern Pyrenees (the Vallée de Madriu-Perafita-
Claror), several pastoral structures in the subalpine zone have been dated
to the first and second centuries AD. What we do not appear to have is
a continuous record of activity in these areas during the Roman period,
nor an intensity of activity that we might associate with the levels of eco-
nomic activity seen elsewhere in the Empire. However, six pitch-ovens
do represent a highly specific and important activity in this area (Leveau
& Palet-Martinez 2010: 180–1). The intensification of charcoal burn-
ing during the Roman period (especially during the third century AD)
is attested to by a number kilns and corroborative palaeoecological evi-
dence across a range of altitudes (Pèlachs et al. 2009). The evidence
implies that the Roman period was not that different to the Bronze Age
in terms of the relative intensity of pastoral activity, with low grazing
pressure attested to on some sites. Here, the Roman period is charac-
terised by the development of a complex mosaic of grazing, burning,
and forest management activities, in some ways different to the evidence
from the alpine zones in the Southern French Alps (Ejarque et al. 2009).
hange
In particular, the presence of resin ovens dated to the first to second

centuries AD is interesting, as this demonstrates the direct use of a forest
resource. In Asturia (the north-western Iberian Peninsula), the moun-
tainous zone appears to have been integrated into the Roman economic
system. Here, Roman mining activity had a noticeable impact on the
vegetation (Ruiz del Árbol et al. 2003). Mining, like any activity, is one
element within a network of activities. In Asturia, agricultural terraces
and terraces on settlement sites were created, this practice being com-
mon on many managed Roman landscapes. When the mines and adja-
cent zones were abandoned at the end of the second century AD, the
forest re-established itself to a certain extent, and eroded soil covered the
Roman horizons that had once formed the terraces in this area. Here,
we see how this upland area was fully integrated into the Roman econ-
omy, and provides a contrast with other mountainous zones, such as the
Sierra de Gredos in central Spain. This area does not appear to have been
fully incorporated into the Roman system, with little evidence for human
activity between c. 15 BC and the fifth century AD (López-Merino &
López-Sáez 2009: 48). Low levels of impact also characterise this period
in the Segura Mountains (Murcia, south-eastern Spain), where signifi-
cant human impact on the vegetation did not occur before the seventh
century AD (Carrión et al. 2004). Once again, this palaeoenvironmental
evidence demonstrates how some areas remained relatively unaffected,
although probably not untouched, by people until the historic periods.
Even areas that saw precocious Neolithic activity were not necessarily the
domain of continuous human impact.
Patchy Porosity: Mediterranean Mountains and Variable

Integration
What we see now in the archaeological and palaeoenvironmental data

from European, Mediterranean mountains are clear demonstrations of
how traditional, environmentally determined hypotheses are untenable.
Some researchers argued that cold, wet periods should lead to lower
levels of human activity in mountainous areas, and clement periods to an
increase in activity (Bocquet 1997). Moreover, the fact that there has not
been a continuous, steady exploitation of the high-altitude zone during
the last 6,000 years implies that knowledge of, and attitudes towards,
these landscapes has fluctuated over time. Furthermore, there is little
Patchy Porosity: Mediterranean Mountains and Variable Integration 269
evidence to suggest a diachronic, incremental evolution of such envir

onmental knowledge. At a very basic level, ecological decisions require
knowledge of the seasons, weather, hydrology, vegetation, geology,
as well as domestic and wild animal behaviour. More specifically, such
knowledge also includes an awareness of hazards associated with activ-
ity in the subalpine and alpine zones. This environmental knowledge
changed with each period; Bronze Age environmental knowledge was
probably different to that of the Roman Period. During the Holocene,
the manner in which such environmental knowledge was mediated, con-
trolled, and transmitted changed continuously.
If there is one activity that has been continuously linked to life in the
mountainous zone, it is pastoralism – more specifically, transhumance
(Fig. 8.8). Although there is no doubt that pastoralism has been an
important activity in many mountainous areas across the Mediterranean,
we must also consider the nature of other engagements with moun-
tainous landscapes. One of the problems with the mountain-pastoralist
discourse is its inherent ecological foundations; that is, the notion that
transhumance in particular is an economic strategy that best exploits the
seasonal and vertical changes that characterise mountainous zones. In
some ways, the pastoralist/transhumant discourse is one dominated by a
rational instrumental view of humanity’s relationship with the landscape,
or perhaps more subtly ‘transhumance reveals the existence, via animals,
a way of living that and the appropriation of space that conditions the
ensemble of a social organisation’ (Duclos 2006: 17; author’s transla-
tion). When compared with arable agriculturalists, pastoralists have a
decidedly different relationship with the land. Arable farmers rely upon
the earth/soil itself, and must invest much of their time safeguarding or
improving this essential resource. Pastoralists, however, simply rely upon
the seasonal growth of pasture, and rarely invest much time and effort
in its maintenance, let alone improvement. Although, as we have seen,
there is evidence to suggest that areas of forest were opened/burnt in
order to create a clearing for grazing, this may well have been the extent
of pastoral land management in the high-altitude zones.
From the middle of the fifth millennium BC, archaeological and
palaeoenvironmental evidence suggests the emergence of a pastoral sys-
tem where individual (or very small numbers) of shepherds worked in
the lower-altitude zones in some mountain ranges. These shepherds were
probably detached from the larger community (Brochier & Beeching
Marne
Meuse
Dn
Lo Tisa
iste
ir e
r
Inn
Danube Somes
Sea of Azov
Bay Mur
Allier
of Drava
Biscay Olt
Tisa
Po Sava
Ga
ro
Rhone
nn
e
Danub
e BlackSea
Ebro
Duero Gulf of
Lions
Adriatic
TagusTajo
Sea Mu
rat
Guadiana
Balearic
Sea
Gulf of Aegean Sea
E uphrates
Ionian Sea
Sea of Crete
Mediterranean Sea
Dead Sea
Gulf of Sidra
Nile
8.8. Principal long-distance transhumant routes – most involve movement from the lowlands to the mountains during the summer. (After Braudel, F. (1985), La
Méditerrannée et Le Monde Méditerrannée.) One of the key research questions in Mediterranean landscape archaeology is when did this model of transhumance
develop.
2006: 151). In fact, prior to the Late Bronze Age, the high-altitude zone
might have been perceived as a remote landscape, rarely witnessed by the
vast majority of people. The most enigmatic evidence for this ‘individu-
alistic’ landscape comes in the form of ‘Otzi’, although he, like all moun-
tain dwellers, would have been a member of valley-based community. The
archaeological and archaeozoological evidence points to small groups of
specialised Neolithic pastoralists carrying out specific activities across a
range of different sites – activities and sites that would have been exploited
at different times of the year as a part of the pastoral round (Bréhard,
Beeching, & Vigne 2010). As time progressed, we can see that one char-
acteristic shared by both the Alps and the Pyrenees is the evidence for
prehistoric burning of the woodland. Possibly this clearance was a form
of fire setting, which might be classified as a form of landscape ‘manage-
ment’, implying controlled engagement with the natural world where
outcomes are usually predictable. This may well have been the case, but
it is an important notion that requires some reflection. Management sug-
gests planning and control, which might mean that exploitation of high-
altitude landscapes was part of a wider socio-economic strategy, perhaps
manipulated by an elite. Were these pioneer shepherds entirely aware of
the environmental consequences of their actions? To what extent did the
initial impact of woodland clearance induce a phase of relative instability,
where soil erosion was in fact quite extreme, and decreased once new
agricultural practices were established and grasslands stabilised (Jacob
et al. 2009). Whilst a form of fire setting appears to be common to both
the Pyrenees and the Southern French Alps, there is a significant differ-
ence in terms of evidence for built structures. There appear to be more
structures in the Alps than in the Pyrenees, and in turn, fewer sites on the
French side of the Pyrenees than on the Catalan/Spanish flanks. Again,
this implies variation in the Bronze Age forms of environmental know
ledge and day-to-day ways of life between the two mountain ranges and
even within ranges. The construction of animal enclosures represents a
specific and innovative engagement with the mountains, and implies a
development of a new form of environmental knowledge related to the
pastoral round (Walsh & Mocci 2011). Despite the relatively low num-
bers of people active in the high-altitude zones, it is likely that the initial
impact of this activity on the environment was disproportionately high
due to the fragility and low sustainability of certain activities in the alpine
zone.
hange
In the Pyrenees, there were two phases of human expansion during

the late third and second millennia BC: the first just prior to the start of
the second millennium BC, and the second at the start of the first millen-
nium BC. From 2000 BC, the reduction of heliophilous species (birch
and hazel) implies a shortening of the forest fallow period (Carozza et al.
2005: 9). The study of lead isotopes implies the development of min-
ing during the Late Bronze Age, and come the end of the Bronze Age,
other indicator species emerged in grassland fallow zones, revealing the
development of prairies and pasture. This transition from an agro-forest
to an agro-pastoral system is a fundamental change in the landscape. In
the Basque mountains, pollen cores from three different altitudes (500
m, 910 m, and 1,300 m) spanning the period between 2100 and 1700
BC show how beech expanded as pastoral and arable activity became
prominent at most altitudes with little substantial deforestation taking
place. Then, from 1700 to 1300 BC, pastoral activity increased at the
expense of arable agriculture. During this period, there was still little
destruction of overall forest cover. The archaeological evidence (in the
form of sites with clear evidence for a pastoral economy) supports the
interpretation of the palynological evidence (ibid.: 17). From 1300 to
1000 BC, arable agriculture re-emerged as an important activity, with
pastoral activity decreasing in some areas except in the altitudes centred
on 1,100 m. Between 1000 and 750 BC, an increase in agro-pastoral
activity and deforestation occurred at all altitudes. In both the Pyrenees
and the Southern French Alps, the end of the Bronze Age was character-
ised by an overall reduction in activity at high altitude (around 2,000 m
and above). However, some palynological signals do imply continued
pastoral activity. Does this combination of evidence (a lack of structures
but a continued presence of pastoralism) indicate a new form of engage-
ment with the subalpine zone? Was it perceived and managed in a dif-
ferent way?
There is no doubt that there was a complex relationship between rit-
ual and landscape, along with the associated cycles of seasons, activities,
and ritual events. A fundamental part of daily life in the high-altitude
zone includes dealing with radical changes in weather, even during the
summer. These changes in weather affect both people and animals, and
are something that we must take into consideration (Strauss & Orlove
2003). What we can be sure of is that the high altitudes are character-
ised by geomorphic and meteorological processes that are so much more
dramatic than the lowlands – areas with which mountain peoples suppos-
edly shared cultural traits. Human interaction with the environment is a
constitutive element in the development of culture, and mountain cul-
tures were (and are) different from their lowland counterparts. This is not
to say that mountain peoples were completely isolated from or entirely
different to their lowland counterparts, but that there were undeniable
differences in their everyday experiences of landscape and environment
and therefore in their cultural ecologies. Bronze Age shepherds in the
high-altitude zones were probably influenced by a hegemonic system
and associated ideological-religious actions. Environmental knowledge
via a cosmology can sustain equilibrium within an environment, whereas
imperfect knowledge – where key knowledge-bearers disappear or where
a change in an environment is misread – can lead to environmental and
societal problems. A change in the quality of environmental knowledge
(a cosmological imbalance, if you like) can lead to phases of instability in
the environment. Consequently, we have to ask how unstable or unpre-
dictable the environments were that would have characterised these
landscapes, and to what extent the waning of activity can actually be
explained by instances of poor knowledge and maladjusted responses to
changes in the environment.
The Middle Iron Age did undoubtedly witness a phase of climatic
deterioration that affected both the Alps and the Pyrenees (Van Geel &
Berglund 1997), but this alone cannot explain the apparent reduction in
pastoral activity in the subalpine zone. As well as an obvious reduction in
the number of sites, pollen diagrams such as that from Lac des Lauzons
in the Southern French Alps (Richer 2009) and the eastern Pyrenees
(Ejarque et al. 2010) reveal woodland regeneration. Changes in weather
and climate, along with associated developments in vegetation, would
have engendered developments in environmental knowledge. Whilst the
high and middle altitudes may have held some ritual significance in the
Early Iron Age, most day-to-day activities were at lower altitudes. As
Iron Age societies developed stronger economic ties with the Roman
world, there would have been a concomitant development of new proto-
urban centres in the valley bottoms.
We might expect that the arrival of Rome in the Alps and the Pyrenees
(and the other mountain ranges mentioned in this chapter) to have her-
alded dramatic changes in the cultural ecologies and associated forms
of environmental knowledge that operated in these mountainous
hange
landscapes. There is no doubt that such changes did occur at lower alti-
tudes – the development of urban centres and roads being the obvious
developments. During the Roman warm period (centred on the first cen-
turies BC and AD), there is relatively little archaeological evidence for
activity in the subalpine and alpine zones of the French Alps, despite the
climatic amelioration. This reflects trends in the Italian Alps described
earlier (Moe et al. 2007). Meanwhile, in the Pyrenees, there was a radical
change in the manner in which (productive) landscapes were organised,
but this was not a decisive moment in the history of human impact on
the subalpine and alpine milieus. There is little archaeological or palaeo-
ecological evidence to suggest dramatic change in the ‘natural’ environ-
ment at this time. Certain areas were heavily exploited, whilst other areas
saw the reestablishment of woodland, with the concentration of pastoral
activity towards lower altitudes (Davasse et al. 1997). In some places,
such as Vicdessos, Cerdagne, and Donezan, there seems to have been
a reduction in human activity during the Iron Age, and perhaps even at
the start of the Roman period (Galop 1998), with other areas showing
a decrease from the third to sixth centuries AD (Mazier et al. 2009). A
change in dietary preferences might be one explanation for a change in
the pastoral regime, with a new Roman emphasis on forest-based hus-
bandry, where pigs were the principal domesticate. Thus, there may not
have been the same requirement for high-altitude pasture. This cannot
be the sole explanation for the relative withdrawal from the high-altitude
zone, and may have been quite specific to the Pyrenees. We know that
Strabo referred to the excellence of the hams from the Kerretani terri-
tory. As Rendu (2003) observes, there may well be two (non-contra-
dictory) explanations for this change during the Roman period: a clean
break vis-à-vis earlier periods in the use of different parts of the landscape
(plains, piedmonts, and slopes), which led to a gradual marginalisation
of the higher reliefs; whilst the zones along the plain edges were more
integrated as part of an increase in agricultural production. This lowered
the ‘centre of socio-economic gravity’ towards the lower altitudes, and
this included pastoral activity that had once exploited the higher zones
(ibid.: 520–1).
What we see in the Alps and the Pyrenees is an adaption of the ‘tra-
ditional’ Roman rural economy to these mountainous areas. The appear-
ance of cultivated trees within a mosaic landscape, where some zones are
employed for arable agriculture whilst others continued to be used for
pasture, is in part a chronological extension of some pre-existing ‘Iron

Age’ activities, with the development of certain Roman elements, per-
haps the most noteworthy being the development of mining. Mining
activity (with concomitant deforestation) was clearly a core activity dur-
ing the Roman period (Monna et al. 2004). The fact that there is little
evidence for an increase in agricultural activities suggests that mining
was the core activity. The Roman period unsurprisingly saw the appear-
ance of certain cultivated trees (walnut, chestnut, and even olive, at some
distance from the high altitudes). At Fangeas in the Southern French
Alps, the reduction in woodland may well have been related to min-
ing indirectly inferred from a recorded increase in lead isotopes (Segard
2009: 197–200). However, the nature of the subalpine and alpine zones
meant that there was also a continuation of the heterogeneous Alpine
landscape, with certain zones going in and out of use depending on local
cultural and socio-economic factors. In the Pyrenees, incontrovertible
archaeological and environmental evidence for specialised activities, such
as mining, resin collection, and exploitation, are all indicative of complex
suites of environmental knowledge, where the influence of a diversified
and multifaceted imperial economy affected the manner in which human
ecologies developed in these mountains. The pre-existing forms of envir
onmental knowledge were probably adapted. Awareness of the dangers
and the resources present in the mountains continued, but new eco-
nomic prospects resulted in the opportunistic intensification of certain
‘niche’ activities, such as charcoal burning and pitch production. It is the
combination of a suite of activities and practices that constitute a specific
cultural ecology – a way of living and working in the mountains. These
forms of environmental knowledge probably existed prior to the arrival
of Rome, but their intensity and configurations evolved as different val-
leys developed economic ties with the imperial network.
If we accept that some evidence does reflect a reduction in activity
in the high-altitude zone during the Roman period, we have to explain
this. At one level, we might consider the nature of taboo and appre-
hensions that existed in various societies (and classes) regarding certain
types of environment and animals and plants therein. Whilst the start
of the Roman Period in the Alps is supposedly characterised by a cli-
matic amelioration, it is unclear to what extent that this had an influ-
ence on people’s lives in the Alps. Perhaps the most salient characteristic
of environmental knowledge is its contingency vis-à-vis social groups,
hange
whether these are defined by class or the geographical origins of the

group. Research into population mobility should allow us to consider
how changes in the forms of environmental knowledge and cultural ecol-
ogies are influenced by immigration and emigration. People arrive and
leave with environmental knowledge. Mountainous landscapes are cer-
tainly not closed landscapes, but there is evidence to suggest that some
areas have been relatively isolated for quite long periods of time. Physical
geography can have an important influence on genetic diversity (Rosser
et al. 2000). Some studies have revealed genetic isolation, with one com-
munity from the South Tyrol possessing a higher Palaeolithic genetic
component than their neighbours (M. G. Thomas et al. 2007). Even the
relatively accessible mountains of central Italy (the Aniene Valley and the
Simbruni Mountains not far from Rome) include populations that are
genetically isolated (Messina et al. 2010). Consequently, we are forced to
revisit well-rehearsed discussions relating to acculturation for the prehis-
toric periods, and Romanisation for later chronological phases.
The verticality of mountains and understanding of the environmen-
tal characteristics of ecological zones are fundamental for any activity
in these areas. There has been much discussion of Roman society’s
attitudes towards mountains and what might be described as a Roman
anguish regarding such landscapes. First, we must accept that there is no
one, single Roman society, and that environmental knowledge within
the Roman Empire would have been socially and culturally contingent.
Moreover, the discussion of Roman angst vis-à-vis the high altitudes
is misplaced and far more complicated than stating that the presence
of Roman activity at high altitude in one place demonstrates that the
trepidation that certain parts of Roman society had towards mountains
did not exist. For example, an altitude of 2,200 m in one valley might
be easily accessible, such as over certain passes (e.g. the Petit Col Saint
Bernard), whereas the same altitude in a topographically diverse zone,
such as many of the valleys in the Ecrins National Park, is in fact dif-
ficult to reach. A simple cost–surface analysis would demonstrate this.
We cannot demonstrate that Rome tried to ‘manage’ all mountainous
areas in the same manner. The Pyrenees may well have been perceived a
distant frontier, whilst the Alps were possibly considered as a more sensi-
tive, core frontier, defining the edge of the Roman homeland. Moreover,
practices specific to these different mountain ranges would have devel-
oped as a consequence of regional economic processes, but also because
Conclusio 277
Rome may have encouraged certain forms of management. For example,

as Leveau (2008a) argues, Rome did raise awareness about and respond
to hydrological problems, in particular, increases in flooding intensity
due to deforestation in the Appenines. Such increases in flood-related
problems may also have been a consequence of increased storm intensity
during periods of relative climatic warming. Consequently, we see how
the nature of human engagements is characterised by a complex interplay
of environmental processes, socio-economic patterns, myth, and reli-
gion, all intersecting to produce historical ecologies that are specific to
each valley and even different zones within that valley due to the radical
changes in environmental characteristics which are directly influenced by
vertical zonation. The history of human–environment interaction in the
mountains is characterised by variations in the emphasis placed on the
incorporation and exploitation of each vertical zone. It is this complex
vertical ecology that demands an environmental perspective if we are to
develop a clear understanding of how these landscapes have changed.
Conclusion
The study of mountainous landscapes has been dominated by ecological

models. These have operated at two levels: first, the underlying notion
that climate has directly influenced the waxing and waning of human
activity in mountainous areas; second, that when people do live and work
in mountains, their activities are structured by vertical ecological zona-
tion. Transhumance, with the seasonal movement up through different
areas of pasture during the summer, is the most obvious example of such
an ecologically driven process. Whilst it is untrue that the presence and
absence of people in mountainous zones has been directly influenced
by climate, there is a certain amount of truth in the nature of activities
being influenced by the ecological potential of a given altitudinal zone.
However, in the past, we have placed too much emphasis on two or
three essential forms of activity: pastoralism, mining, and sylviculture.
There were clearly many other important activities which varied in their
extent and intensity, depending on the socio-economic context and
forms of environmental knowledge present within each valley or set of
valleys. As the transfer of culture, ideas, and environmental knowledge
has varied across time, we have to accept that valleys or groups of val-
leys will have experienced variations in ‘permeability’; the strength of
hange
links and connections with other places varies with each period. Activities
and practices, settlement density, and human impact on mountains vary
across time and space. There are no homogenous trends. Not all val-
leys saw the construction of enclosures during the Bronze Age, not all
valleys witnessed a Roman absence or decline in activity. The location
of individual sites was influenced by comprehension of potential haz-
ards (flooding, rock slides, and avalanches). Recent ethnoarchaeological
research also suggests that shepherds avoid zones where there is a high
probability of lightning strikes (Carrer 2012: chs 5 & 6). It is apparent
that certain zones within valleys have a higher probability of being hit
by lightning, and bearing in mind that storm intensity could increase
during periods of climatic warming, we have to consider the impor-
tance of probable trends in weather events and the influence of these
on settlement, economy, and culture in Mediterranean landscapes. The
combined effects of these environmental variables engender enormous
variation in the choices made for site location in mountainous zones,
as past peoples applied locally specific environmental knowledge when
making choices about site location and the patterning of economic activ-
ities. The notion that we can develop generalised arguments regarding
the nature of human–mountain relationships based on the commonality
of environmental characteristics across vertical mountain zones is clearly
problematic. This form of cultural ecological approach possesses all the
dangers of any normative analysis. So many descriptions of mountain
economic systems are founded on fundamental precepts that then lead
us unwittingly into an environmentally deterministic discourse. Moving
on, some have then argued that success in mountainous zones is based
upon the flexible responses of humans to the demanding environmental
conditions. The response types defined by R. B. Thomas (1979: 161–7)
were rotation, regulation, cooperation, mobility, and storage – indeed,
response mechanisms that one might expect to find in any landscape. As
Guillet (1983: 562) remarks, ‘adaptiveness’ can become a tautology for
any situation. What we should not forget is that the study of human–
environment relationships often overlooks the people behind the ‘anthro-
pogenic processes’. Human activity in mountains, and, indeed, in any
environment, operates on many different scales and is a class-based expe-
rience. By their very nature, the vertical stages within mountains may
well have reflected the structure of some societies in the past: the urban
rich at low altitudes, and the rural poor in the most elevated and harshest
Conclusio 279
areas. This may appear a simplistic and rather crass form of determinism,
but such a structure is in fact quite common. These different groups
have left different environmental signals and produced different histori-
cal ecologies. Mountains, more than any of the landscapes discussed in
this book, proffer a myriad of resources, but are at the same time harsh,
difficult, and often deadly environments. The environmental processes
that constitute these landscapes are a part of everyday life and contribute
to the construction of Mediterranean cultures.
9
Conclusions – The Mediterranean Mosaic: Persistent

and Incongruent Environmental Knowledge
The preceding chapters have assessed the evidence for changes in human
engagements with a range of Mediterranean landscapes – landscapes
constituted by dynamic mosaics that shift with time and can only be
understood via the analysis of spatially specific palaeoenvironmental and
archaeological records. In many instances, the essence of Mediterranean
environmental management does not lie with the imposed, macro-eco-
nomic structures and concomitant networks of trade and commerce, but
with the environmental knowledge developed by those who worked the
land. Whilst we can never identify specific stories of human engagement
and concomitant social relationships, we can imagine scenarios not so
dissimilar from those articulated in modern literature. Conflicts over
land, and especially water, may well have been common in the pre- and
proto-historic pasts. The human experience and the development of
social relations are formed via interactions with the natural world. Novels
such as Jean de Florette (Pagnol 1988) or Batailles dans la Montagne
(Giono 1979) paint images of human relationships with the environ-
ment and the manner in which these contribute to the development of
interpersonal interactions within specific landscapes. Such relationships
are of course contingent upon a number of specific historical and geo-
graphical phenomena, and, as archaeologists, we can only ever propose
possible scenarios. An integrated landscape archaeology, which operates
within historical and cultural ecological frameworks, does enhance the
elucidation of the lifeways of those peoples who engaged with the envir
onmental processes discussed in this volume.
We know that those who worked the land were often influenced or
even controlled by macro-economic structures and political hegemonies –
structures that would have passively (via ideological and belief systems)
280
Conclusions – The Mediterranean M
osaic 281
and aggressively imposed certain forms of environmental knowledge

and concomitant management strategies. The periods of environmental
stability and instability that we identify were not necessarily the conse-
quence of natural changes in the environment (Kuniholm 1990), and
phases of instability in particular were not the result of unthinking,
unfettered anthropic exploitation of these landscapes. Any changes in the
environment were the product of intersecting socio-economic, ideologi-
cal, and environmental processes. This is perhaps best characterised by
the notion of ‘Panarchy’ by resilience theorists (Gunderson & Holling
2002). Panarchy describes cross-scale interactions, and ‘the resilience of
a system at a particular focal scale will depend on the influences from
states and dynamics at scales above and below. For example, external
oppressive politics, invasions, market shifts, or global climate…’ (Walker
et al. 2004: 5).
It is important that we investigate the two dominant paradigms
in Mediterranean environmental history. The first is the notion that
Mediterranean landscapes can be characterised as ruined, and that their
history is one of a downward spiral of degradation, with humans and
climate working together to ensure their blight (see Hughes 1994b esp.
ch. 11; 2005 esp. ch. 7). The other position in this dichotomised debate
is represented by the more recent and fashionable model, which consid-
ers that the above process of degradation is overstated, and that peo-
ple have continuously managed to exploit the various niches across the
Mediterranean through adaptive strategies which have guaranteed suc-
cess and avoided disaster (see Horden & Purcell 2000 esp. ch. 8).
One phenomenon that we cannot directly identify as archaeologists
but needs to be considered is the variation in perceptions and understand-
ings of landscapes and their natural processes. During those periods when
activity, especially political and economic control, was centred on fixed
settlements (from the Neolithic onwards), the agricultural world would
have been centred on the village and the abutting areas (Robb 2007).
Here, the view of nature might have been one centred on the produc-
tive zones close to the settlement, whilst more distant activities, includ-
ing transhumance, wood, and wild-produce collection, might have been
detached experiences for certain people. Consequently, if reconstruc-
tions of past environmental processes are inferred from sample locations
that were not adjacent to activity zones, then the inferred environmental
processes may not have been perceived as important by most people.
282 Conclusions – The Mediterranean Mosaic
Moreover, we have to consider explicitly how the environmental pro-

cesses that we characterise would have been spread across different scales
of time, even if this means making inferences not directly supported by
the data. One of the problems with archaeological conceptions of time is
that – as with the traditional, Western view of nature – we place ourselves
outside of these different realms of time, and consider processes and the
associated experiences of time as remote phenomena. The longue durée
is normally defined as being that category of time associated with envir
onmental processes (Braudel 1985). However, it is the day-to-day (évé-
nementielle) experience of environmental processes (such as weather and
erosion) combined with the annual cycle of seasonal activities that actu-
ally define human experience of the natural world. Some have argued for
two forms of time in agricultural societies: cyclical, relating to the annual
round of tasks in the landscape, and then a longer-term, linear form of
time as represented by monuments and important natural places (D. W.
Bailey 1993). In fact, there is no need to oppose cyclical time and linear
time (Lucas 2005: 93), and the existence of such discrete chronological
frameworks should also be questioned (G. Bailey 2007). Landscape fea-
tures are often hybrids, where natural and cultural characteristics inter-
sect. These imbue landscapes with temporal character, where a defined
notion of linear time is replaced by a notion of ‘pedigree’ or ancestry.
People would have discovered traces of the past when working the land.
They would also have come across evidence of past environmental pro-
cesses (Bradley 2002: 13–14), and environmental knowledge of these
processes contributed to landscape resilience. In societies where notions
of linear time did not exist, an appreciation of long-term landscape deg-
radation was probably absent, and if any temporal notion was notewor-
thy, it would have been that association with seasonal-cyclical time.
Retrodicting Human Engagement with the Landscape
Each moment of human–landscape interaction is ‘perfect’ in the sense

that the nature and characteristics of cultural environmental processes
are peculiar to each moment in time and space (Phillips 2007). We will
rarely identify specific moments of human–landscape interaction, some-
times characterised as the ‘Pompeii Premise’ (Binford 1981; Schiffer
1985). Perhaps we should accept that we can only ever infer ‘fuzzy’
landscapes, where space is measurable but time is blurred, where we
Retrodicting Human Engagement with the L
andscape 283
can never have a direct handle on the time at which an environmental

event took place. We can know that certain processes took place within
a certain time frame, and can assess the extent to which these processes
were experienced and the consequences acted upon. This ‘fuzzy’ process
can be identified when palaeoenvironmental evidence and contemporary
archaeological evidence intersect with one another in the same landscape
(e.g. Pucci et al. 2011). Environmental processes have a past (the event),
present (the state of the environment as perceived at a point in time), and
future (an appreciation of how and when this process might reoccur and
its implications for landscape, economy, etc.). Environmental manage-
ment, in all its forms, implies not only a comprehension of past events,
but also an awareness of the possibility/probability of an event reoccur-
ring. Such management practices can include terracing, woodland man-
agement, flood prevention and mitigation systems (such as at Tiryns), or
site movement or relocation. The length of time (i.e. the start and end
of an event or process) is less important than the frequency of human
engagement with that process (or space). It is this frequency that gives
environmental processes their level of relative importance.
Horden and Purcell (2000 ch. VII.2) rightly identify the potential
for local, community-level manipulation of landscapes, but at the same
time question the somewhat romantic notion of cooperation, especially
in hydraulic landscapes. Here, their examples tend to be medieval, with
some reference to earlier periods. If we are to assess such processes for
the proto- and prehistoric pasts, we require well-dated, overlapping
palaeoenvironmental and archaeological records. Where sedimentary
sequences need to be dated and the waxing and waning of hydrological
processes identified, future research may well need to call upon well-
dated sequences of OSL dates that have been subject to Bayesian analysis
(Blaise & Balasse 2011). Only then can we hope to develop narratives
of past, local-scale human–environment engagements. What we might
aim for is evidence of predictive or anticipatory (ex ante) actions vis-
à-vis the environment, or reactive (ex post) responses to an environ-
mental process. The identification of these two very different actions
would allow a clearer characterisation and assessment of environmen-
tal knowledge in the past. Once we see the emergence of similar (tech-
nological) responses to comparable environmental processes across the
Mediterranean, we might be able to consider a Mediterranean-wide
cultural ecology. However, this can never be a full human ecology, as
responses to any environmental process will include cultural filters and

religious attitudes which will have varied across space. For example, it is
apparent that from the Bronze Age onwards, different societies across
the Mediterranean were adopting similar strategies across certain land-
scape types: the development of ports, terracing, dams, irrigation, and so
on. If we see a homogenisation of technological responses to environ-
mental problems, then this might be due to the presence of an imperial
power or a particularly strong political force within a region. However,
certain categories of environmental knowledge and associated forms of
landscape management would have been transmitted between and across
societies via a number of mechanisms, including elite interactions, the
movement of traders (Kristiansen & Larsson 2005), and perhaps even
people with specialist environmental knowledge.
Revisiting Mediterranean Environmental Problems
One environmental constant across Mediterranean landscapes is the pres-

ence of tectonic structures and associated hazards, such as volcanic erup-
tions and earthquakes. These do not merely provide the physical backdrop
for the development of human landscapes across the Mediterranean, but
these processes are active agents in the construction of Mediterranean
culture, belief systems, and socio-economic processes. Whereas certain
catastrophic events clearly have immediate consequences, tectonic activ-
ity also contributes to the structuring of more mundane processes, in
particular, erosion. The complex mosaic of undulating topography also
influences the composition and distribution of vegetation. The cultural
importance of the catastrophe narrative is perhaps nowhere more appar-
ent than with the story of the universal flood myth (Wilson 2001), pos-
sibly grounded in actual events early on in the Holocene. We should
of course be aware of the dangers of placing too much emphasis on
catastrophism (Morhange et al. 2013). Catastrophes, such as tsunamis,
provide a stark contrast with the more mundane but significant low-mag-
nitude processes, such as sea-level changes and the threats often posed
to coastal sites by the salinisation of water and sedimentation leading to
coastal advance and the silting-up of harbours. These latter processes tie
the coastal zone to the hinterland, with its constant but variable scales
of erosion, sediment transport, and (re)deposition. The potential risks
posed by sea-level rise today are widely acknowledged. The combination
Revisiting Mediterranean Environmental P
roblems 285
of subsidence (in north-east Italy, for example), thermal seawater expan-

sion, and glacier and ice-sheet melting could result in a relative sea-level
rise of between 180 mm and 1,400 mm by AD 2100 (Lambeck et al.
2011). It is this combination of low-magnitude, seemingly mundane
processes that pose some of the greatest threats.
Alluvial records are often used in the assessment of erosional histories;
that is, sediments removed and transported due to increased precipita-
tion and/or human disturbance. However, it is human engagement with
and responses to the floods that are just as interesting. Moreover, we can
identify society’s responses to floods in the archaeological record. It is the
explicit amalgamation of archaeological and palaeoenvironmental data
that permit the development of the richest stories of the development of
environmental knowledge across different times and spaces. For exam-
ple, there are times when people appear to have abandoned wetlands,
possibly due to disease or simply because inundation became a com-
mon occurrence. It is difficult to demonstrate that enhanced flooding or
wetland encroachment can be correlated with periods of climatic dete-
rioration (i.e. increased precipitation). It is likely that flooding occurred
under warmer climatic conditions, with concomitant concentrated sea-
sonal storminess – an all too common process in the Mediterranean. Just
as important are the changes in river-channel configurations – changes in
size and depth to the migration of channels across a flood plain. Rivers
and their flood plains clearly provide rich opportunities, the most impor-
tant being the deposition of silts in relatively moist zones that permit the
development of arable agriculture. This continual process of sediment
removal and transportation means that environmental productivity is in
some ways moved from one part of the landscape to another (Bintliff
2002). Eroded material from one (now partially, pedologically impover-
ished) area, redeposited within a landscape, can provide a new ecological
potential. Over the long term, relatively few landscapes have witnessed
degradation that has ultimately rendered them permanently unusable,
although many places have witnessed phases of environmental instability
to which specific populations or groups of people have had to respond.
Responses to environmental features and changes in the landscape
were not merely economically rational and instrumental; they were also
cultural, ideological, and religious. Some religious systems, such as the
Minoan religion, possessed distinctive features, including large quanti-
ties of ‘religious’ material culture, a relative dearth of temples, and a
large number of nature sanctuaries that constituted the key component

in this society’s relationship with nature and landscape. As Herva (2006:
593) suggests for the Minoan religion, the emphasis on a form of ani-
mism, rather than an importance of divinities, might be explained by the
incredible variation of environments and niches across Crete, which is in
turn a product of its tectonic characteristics and its insularity. In the con-
tinental Greek context, the presence of so many deities and their sanctu-
aries across the landscape (Cole 1996; Jost 1996) might imply a level of
environmental insecurity. The need to imbue the productive rural world
with sanctuaries and offerings suggests that the natural world needed to
be coaxed into guaranteeing arable and pastoral success.
The greatest threat to agricultural production and, indeed, all plant
and animal life in the Mediterranean is aridity and drought. Whereas the
development of the arid, notionally archetypal Mediterranean environ-
ment might be seen by some as the friable foundations for subsequent land-
scape degradation, the stabilisation of eustatic sea level, the emergence of
sclerophyllous vegetation, and the other characteristics that define many
Mediterranean zones today could be considered as the framework within
which relatively stable and complex forms of environmental knowledge
could develop. However, continued changes in climate and associated
transformations in weather patterns with their concomitant effects on
crops, pasture, and soil were never predictable. Long-term climatic sta-
bility does not explain the success of civilisations (Crumley 1994a), but
unpredictable weather will test socio-economic resilience. The likeli-
hood of environmental degradation may well have increased where local,
community-based responses to environmental problems broke down.
Therefore, periods of environmental instability may not necessarily corre-
spond with demographic pressure and the zenith of a given civilisation’s
power, but rather with periods of civilisation collapse or phases of inter-
nal conflict. Such phases of internal conflict can of course occur when a
civilisation is at its most powerful, and certain environmental initiatives
imposed by hegemonic structures may cause environmental stresses due
to incongruous hegemonic and local forms of environmental knowledge
(as discussed in Chapter 4). A period when aridity might have had a dra-
matic effect on societies is the Bronze Age (at c. 2200 BC and c. 1000
BC (Finné & Holmgren 2010), when some consider that deterioration
in climate might have been responsible for some socio-economic crises
(Kaniewski et al. 2010). Rather than continuous aridity, the probability
Revisiting Mediterranean Environmental P
roblems 287
of climatic instability – of a drying trend with shorter phases of humidity,

as seen in Central Anatolia (Kuzucuoglu et al. 2011) – could have tested
the flexibility of environmental knowledge. Moving on to the Iron Age,
climatic variability is underpinned by the emergence of increased aridity
in some areas (Roberts, Brayshaw, et al. 2011), whilst some regions were
wetter and cooler than the preceding and subsequent periods. It seems
likely that the Northern Mediterranean did witness similar changes to
those seen across north-west Europe (Provansal 1995a; Van Geel &
Berglund 1997). Increased flooding (because of enhanced precipitation)
may well have caused problems for some Iron Age settlements (Simeoni
& Corbau 2009; Veggiani 1994). The Iron Age site at El Gallo, near
Cadiz in south-west Spain, saw the development of an enhanced fresh-
water marsh and an increase in riparian vegetation (Saez et al. 2002),
whilst changes in the anatomy of olive-wood structure in parts of Spain
and the south of France suggest an increase in rainfall during the tran-
sition from the Bronze Age to the Iron Age (Terral & Mengual 1999).
Despite evidence for increased precipitation during the Iron Age – possi-
bly solar-influenced changes (Biserni & Van Geel 2005; Mauquoy et al.
2002) – we see little evidence in the Mediterranean for such changes
in climate having a negative effect on these emerging civilisations. The
development of the Greek and Etruscan civilisations, and ultimately the
Roman Republic, took place during a period that some geomorphologists
would characterise as being one of environmental deterioration – essen-
tially defined by increased precipitation and lower average temperatures.
In fact, enhanced rainfall would have rendered many marginal arid zones
more productive. The earlier phase of the Iron Age in the Near East may
have witnessed slightly more arid conditions, whilst during the Persian,
Hellenistic, and the Early Roman periods, rainfall may have been slightly
higher than the preceding periods (Rosen 2007: 42–3). What is appar-
ent is the fluctuation in climatic patterns and, in particular, precipitation.
Until we have clear evidence for the seasonal (weather) precipitation,
assessing the possible role of aridity in causing socio-economic stress will
be problematic.
The first centuries BC and AD are documented as being a period of
climatic amelioration. Climatic amelioration, of course, means different
things in different parts of the Mediterranean. Some areas saw increased
average temperatures (Chen, Zonneveld, & Versteegh 2012), which as
noted elsewhere in this volume, should not be interpreted as an entirely
positive development, as such a change can enhance seasonal evapotrans-

piration in the Mediterranean and result in extreme storm events. The
opportunities proffered by this period of climatic amelioration varied
from one part of the Mediterranean to the other. For example, more
alpine passes were open for longer periods during the Roman climatic
optimum. The high-altitude pastoral season would have been longer as
well, although this was not necessarily taken advantage of by all Roman
or Romanised societies. This phase of climatic amelioration might have
provided the ideal conditions for intensive arable and pastoral activity if
these landscapes were well managed. However, zones which were sub-
ject to several seasons of aridity and poor plant growth may well have
witnessed seasonal erosion. This is why this volume has included spe-
cific landscape and site-based examples, as no general rules or trends can
be inferred. Only trajectories can be considered, characterised by the
notion that a trajectory comprises a predictable starting condition for a
landscape – its geology, hydrology, probable soils – but that its develop-
ment will vary depending on a number of external influences (climate,
weather, human activities, with all their abundant socio-economic and
cultural elements).
Some have argued that Mediterranean societies and, in particular, the
Greeks and Romans mistreated the natural environment (Hughes 2005:
197). There is limited evidence for this. As already demonstrated, certain
places have witnessed environmental degradation at certain times in the
past. However, we can infer management practices that were designed
to mitigate degradation, undoubtedly with a view to maintaining agri-
cultural production. Some classical thinkers did articulate a philosophy
which might be considered ‘conservationist’. There is an argument that
some of Plato’s ideas were akin to modern ‘deep ecology’ (the belief that
humanity should respect the natural world) because Plato rejected ego-
ism and anthropocentrism (Mahoney 1997: 48), although some argue
that Plato’s ethics were essentially egocentric. Moreover, it seems that
Plato saw the natural world as a connected organism of which people
were an integral part. For the Roman period, some of the writings of the
agronomists might also be considered conservationist, although these
accounts were blatantly concerned with an economically rational per-
spective vis-à-vis productivity. Nevertheless, such institutionalised and
systematised forms of environmental knowledge, compounded with
Mediterranean Landscapes; Past, Present, and Future… 289
an unbalanced market, may well have contributed to certain types of

environmental problems.
Horden and Purcell’s integrated Mediterranean serves as a useful
framework for the assessment of different levels of environmental knowl-
edge. However, their integrated network implies the existence of some
forms of universal environmental knowledge. In fact, environmental
knowledge is socially structured and specific to each group’s form of
engagement with the environment. There are undoubtedly certain forms
of ancient landscape management that have demonstrated both cultural
and ecological resilience. For example, centuriated fields appear to have
supported the ecological structure and resilience of some agricultural
landscapes (Caravello & Giacomin 1993), whilst there are some niches
that appear to have never fully recovered from Roman exploitation,
including the Crau in the south of France (F. Henry 2009; F. Henry,
Talon, & Dutoit 2010).
Despite the emphasis on correlating culture and environmental change
over the long and medium durée (e.g. using climate change to explain
culture collapse or settlement abandonment within a given region), our
data are actually more suited to the assessment of specific local events
(the événementielle). The combination of archaeological and palaeoenvir
onmental evidence do allow the identification of site-based responses to
environmental events or specific examples of human manipulation of the
environment. Once we have established a series of such studies within a
micro-region, we then have the foundations for the analysis of longer-
term responses to environmental changes in the past.
Mediterranean Landscapes: Past, Present, and Future…
Whereas the causes and trajectory of twenty-first century climate change

are in many ways different to past phases of climate change, there is no
doubt that similar conditions were witnessed in parts of the region at
different times during the Middle and Late Holocene. Local forms of
environmental knowledge that are not driven by homogenising political
and hegemonic structures are fundamental to the mitigation of environ-
mental problems across landscapes with multiple niches and economic
potentials. Increased aridity and possible desertification are the greatest
problems facing some parts of the Mediterranean (de Wrachien, Ragab,
& Giordano 2006). Future climate change will probably exacerbate the
seasonal characteristics of Mediterranean weather, with longer, hotter

summers characterised by extreme orographic precipitation (Diodato
et al. 2011). Such events undoubtedly took place in the past, such as
the catastrophic flooding event at the end of the fifth century AD at
Mahuzah d’Iamnin (associated with Yavneh-Yam) on the coast 15 km
south of Jaffa-Tel Aviv (Fischer 2005). Climate change is producing sub-
stantial variations in inter- and intra-annual precipitation and thereby
freshwater supply across the Mediterranean. Overexploitation of aquifers
and the potential for salinisation is another twenty-first century problem
that has been with Mediterranean societies for millennia. Today, many of
the nations with the highest water footprint (cubic metres per year per
capita) globally are Mediterranean, including Portugal, Spain, Greece,
Italy, Israel, Tunisia, and Libya (Hoekstra & Mekonnen 2012).
One important change that has taken place over the last 50 years or so
is the enormous decline in the rural population, especially on the north-
ern half of the Mediterranean, from 46 million in 1960 to 12 million in
2000. To a certain extent, if we accept that vegetation will protect soils
(as it has done in the past), then these depopulated zones may well wit-
ness a phase of unprecedented environmental stability, even if there is a
swing towards potentially damaging, intense, and concentrated storm-
related precipitation. However, falling water tables and increased levels
of combustible vegetation will probably lead to more forest fires and
consequent erosion. Nevertheless, some Mediterranean rural zones are
now witnessing a stabilisation and small growth in populations (Benoit
& Comeau 2005: 250–3). This is part of newly diversifying rural econ-
omies, and the impact on environmental resources will not be the same
as during the nineteenth century (when the rural population was at its
highest). There is no doubt that good-quality agricultural land is being
lost across the Mediterranean. Perhaps the most dramatic statistic is the
37% loss of good-quality land on Malta in 90 years (ibid.: 267). Bringing
us back to the question of insular biodiversity, this statistic alone demon-
strates how, up until recent years, Maltese environmental management
was in many ways successful. Even with today’s models, ex ante or predic-
tive responses to environmental change, especially regarding hydrology,
are extremely difficult (Arnold, Walsh, & Hollimon 2004), and there is
no doubt that prediction in the past of many environmental processes
was even more problematic. Therefore, flexibility and adaptability under-
pinned successful mitigation. As we have seen, certain societies in the past
Mediterranean Landscapes; Past, Present, and Future… 291
that were situated within semi-arid, degraded landscapes still successfully

exploited specific niches – an obvious example being the Argaric culture
in south-east Spain. However, flexibility is often stifled when hegemonic
political structures control the use and management of landscapes. In
recent decades, Northern Mediterranean agriculture has adopted a
‘delocalised’ form of production that adapts to large-scale markets rather
than adapting to the ecological characteristics of the locales within which
production takes place (Benoit & Comeau 2005: 272). Although such
homogenisation would never have been as intensive and extensive in
the past, there were undoubtedly periods when local landscapes were
obliged to respond to regional, hegemonic imperatives. Despite these
problems, there are many examples of successful anthropogenic niches.
Possibly the best example is the cork oak savannas, where diversity and
resilience are only guaranteed by active management (Bugalho, Caldeira,
et al. 2011). Here, management practices sustain grazing and augment
habitat heterogeneity. More specifically, the diversity of herbaceous
plants and invertebrates is promoted (Bugalho, Lecomte, et al. 2011).
As with many regions around the world, Mediterranean cultures’ inter-
actions with the environment in the past have often been influenced by
overarching ideological processes. The geological and ecological com-
plexity of Mediterranean landscapes has always demanded complex and
adaptable forms of environmental knowledge. Environmental problems
can include drought, flooding, coastal change, forest fires, and erosion,
and because of some (or all) of these, the results may include downturns
in agricultural production and even injury and death. Insurmountable
problems have rarely been caused by the environment alone, but have
rather been the product of intersecting social and environmental issues,
where rigid forms of environmental knowledge (often overly structured
by hegemonic forces) have prevented individuals or small communities
from mitigating these threats.
Bibliography
Abbott, J. T., & S. Valastro, 1995. The Holocene alluvial records of the chorai
of Metapontum, Basilicata, and Croton, Calabria, Italy, in Mediterranean
Quaternary River Environments: Refereed Proceedings of an International
Conference University of Cambridge/United Kingdom/28–29 september, eds.
J. Lewin, M. G. Macklin, & J. C. Woodward. Rotterdam: AA Balkema,
pp. 195–205.
Ackermann, O., H. J. Bruins, & A. M. Maeir, 2005. A unique human-made
trench at Tell es-Sâfi/Gath, Israel: Anthropogenic impact and landscape
response. Geoarchaeology, 20(3), 303–27.
AEMET, Valores climatológicos normales – Agencia Estatal de Meteorología –
AEMET. Gobierno de España.
Aguilera, M., J. P. Ferrio, G. Pérez, J. L. Araus, & J. Voltas, 2012. Holocene
changes in precipitation seasonality in the Western Mediterranean Basin: A
multi-species approach using δ13C of archaeobotanical remains. Journal of
Quaternary Science, 27(2), 192–202.
Agusta-Boularot, S., M. Christol, M. Gazenbeek, Y. Marcadal, V. Mathieu, J. L.
Paillet, A. Rapin, A. Roth-Conges, J. C. Sourisseau, & H. Treziny, 2004.
Dix ans de fouilles et recherches à Glanum (Saint-Rémy-de-Provence):
1992–2002. Journal of Roman Archaeology, 17, 27–56.
Agusta-Boularot, S., & M. Gazenbeek, 2003. Glanum (Saint-Rémy-de-Provence,
Bouches-du-Rhône): les installations hydrauliques du vallon Saint-Clerg de
la fin de l’âge du Fer à nos jours. Revue d’Archéologie Narbonnaise, 36,
93–132.
Akar, M., 2009. The Role of Harbour Towns in the Re-Urbanization of the
Levant in the Middle Bronze Age (1800–1600 BC). Perspectives from
Eastern Cilicia and the Amuq Plain of Hatay, PhD. dissertation, Università
degli Studi di Firenze, Italy.
Aksu, A. E., R. N. Hiscott, P. J. Mudie, A. Rochon, M. A. Kaminski, T. Abrajano,
& D. Yasar, 2002. Persistent Holocene outflow from the Black Sea to the
Eastern Mediterranean contradicts Noah’s flood hypothesis. GSA TODAY,
12(5), 4–9.
293
294 Bibliograph
Albore Livadie, C., 2003. Piano di Sorrento (Naples): évidence d’une

modification du réseau hydrique sur le site archéologique de la Trinité,
in Variazioni Climatico-Ambientali e Impatto sull’uomo nell’area circum-
Mediterranean durante l’Olocene, eds. C. A. Livadie & F. Ortolani. Bari:
Edipuglia, pp. 359–64.
Alcock, S. E., 2007. The essential countryside: The Greek world, in Classical
Archaeology, eds. S. E. Alcock & R. Osborne. Oxford: Wiley Blackwell,
pp. 120–38.
Alcover, J. A., 2008. The first Mallorcans: Prehistoric colonization in the Western
Mediterranean. Journal of World Prehistory, 21(1), 19–84.
Alcover, J. A., R. A. M. O. Perez-Obiol, E. I. Yll, & P. E. R. E. Bover, 1999.
The diet of Myotragus balearicus Bate 1909 (Artiodactyla: Caprinae), an
extinct bovid from the Balearic Islands: Evidence from coprolites. Biological
Journal of the Linnean Society, 66(1), 57–74.
Aldenderfer, M., 2006. Modelling plateau peoples: The early human use of the
world’s high plateaux. World Archaeology, 38(3), 357–70.
Aldrete, G. S., 2007. Floods of the Tiber in Ancient Rome. Baltimore: Johns
Hopkins University Press.
Ali, A. A., C. Carcaillet, B. Talon, P. Roiron, & J. F. Terral, 2005. Pinus cembra
L. (arolla pine), a common tree in the inner French Alps since the Early
Holocene and above the present tree line: A synthesis based on charcoal
data from soils and travertines. Journal of Biogeography, 32(9), 1659–69.
Ali, A. A., P. Roiron, J. L. Guendon, P. Poirier, & J. F. Terral, 2005. Holocene
vegetation responses to fire events in the inner French Alps (Queyras Massif):
Data from charcoal and geomorphological analysis of travertine sequences.
The Holocene, 15(1), 149–55.
Allen, H. D., 2001. Mediterranean Ecogeography. Harlow: Prentice Hall/
Pearson Education.
Allen, J. R., W. A. Watts, E. McGee, & B. Huntley, 2002. Holocene environ-
mental variability – The record from Lago Grande di Monticchio, Italy.
Quaternary International, 88(1), 69–80.
Altunel, E., I. S. Stewart, A. Barka, & L. Piccardi, 2003. Earthquake faulting at
ancient Cnidus, SW Turkey. Turkish Journal of Earth Sciences, 12, 137–51.
Ambraseys, N., 2009. Earthquakes in the Mediterranean and Middle East: A
Multidisciplinary Study of Seismicity Upto 1900. Cambridge: Cambridge
University Press.
Ammerman, A. J., 2010. The first argonauts: Towards the study of the earliest
seafaring in the Mediterranean, in The Global Origins (and Development) of
Seafaring, eds. A. Anderson, J. Barrett, & K. Boyle. Cambridge: McDonald
Institute For Archaeological Research, pp. 81–92.
Ammerman, A. J., N. Efstratiou, M. Ntinou, K. Pavlopoulos, R. Gabrielli, K. D.
Thomas, & M. A. Mannino, 2008. Finding the Early Neolithic in Aegean
Thrace: The use of cores. Antiquity, 82(315), 139–50.
Bibliograph 295
Andreou, S., M. Fotiadis, & K. Kotsakis, 1996. Review of Aegean prehistory V:

The Neolithic and the Bronze Age of northern Greece. American Journal
of Archaeology, 100(3), 537–97.
Andrieu-Ponel, V., P. Ponel, A. J. T. Jull, J. L. Beaulieu, H. Bruneton, &
P. Leveau, 2000. Towards the reconstruction of the Holocene vegetation
history of Lower Provence: Two new pollen profiles from Marais des Baux.
Vegetation History and Archaeobotany, 9(2), 71–84.
Anketell, J. M., S. M. Ghellali, D. Gilbertson, & C. O. Hunt, 1995. Quaternary
wadi and floodplain sequences of Tripolitania, northwest Libya: A synthe-
sis of their stratigraphic relationships, and their implications in landscape
evolution, in Mediterranean Quaternary River Environments: Refereed
Proceedings of an International Conference University of Cambridge,
United Kingdom, 28–29 September, eds. J. Lewin, M. G. Macklin, & J. C.
Woodward. Rotterdam: AA Balkema, pp. 231–45.
Ånstöm, P., & K. Demakopoulou, 1996. Signs of an earthquake at Midea?, in
Archaeoseismology, eds. S. C. Stiros & R. E. Jones. Athens: Institute of Geology
and Mineral Exploration & The British School at Athens, pp. 37–40.
Antonioli, F., M. Anzidei, K. Lambeck, R. Auriemma, D. Gaddi, S. Furlani,
P. Orrù, E. Solinas, A. Gaspari, S. Karinja, V. Kovačić, & L. Surace, 2007.
Sea-level change during the Holocene in Sardinia and in the northeastern
Adriatic (central Mediterranean Sea) from archaeological and geomorpho-
logical data. Quaternary Science Reviews, 26(19–21), 2463–86.
Anzidei, M., F. Antonioli, K. Lambeck, A. Benini, M. Soussi, & R. Lakhdar,
2011. New insights on the relative sea level change during Holocene along
the coasts of Tunisia and western Libya from archaeological and geomor-
phological markers. Quaternary International, 232(1–2), 5–12.
Armijo, R., H. Lyon-Caen, & D. Papanastassiou, 1991. A possible normal-fault
rupture for the 464 BC Sparta earthquake. Nature, 351(6322), 137–9.
Arnaud-Fassetta, G., & C. Landuré, 2003. Hydroclimatic hazards, vulnerability
of societies and fluvial risk in the Rhone delta (Mediterranean France) from
the Greek period to the Early Medieval Ages, in The Mediterranean World
Environment and History, ed. E. Fouache. Paris: Elsevier, pp. 51–76.
Arnold, J. E., M. R. Walsh, & S. E. Hollimon, 2004. The archaeology of
California. Journal of Archaeological Research, 12, 1–73.
Arroyo-García, R., L. Ruiz-García, L. Bolling, R. Ocete, M. A. López,
C. Arnold, A. Ergul, G. Söylemezoğlu, H. I. Uzun, F. Cabello, J. Ibáñez,
M. K. Aradhya, A. Atanassov, I. Atanassov, S. Balint, J. L. Cenis, L. Costantini,
S. Gorislavets, M. S. Grando, B. Y. Klein, P. E. McGovern, D. Merdinoglu,
I. Pejic, F. Pelsy, N. Primikirios, V. Risovannaya, K. A. Roubelakis-Angelakis,
H. Snoussi, P. Sotiri, S. Tamhankar, P. This, L. Troshin, J. M. Malpica,
F. Lefort, & J. M. Martinez-Zapater, 2006. Multiple origins of cultivated
grapevine (Vitis vinifera L. ssp. sativa) based on chloroplast DNA polymor-
phisms. Molecular Ecology, 15(12), 3707–14.
296 Bibliograph
Asouti, E., 2003a. Wood charcoal from Santorini (Thera): New evidence for cli-
mate, vegetation and timber imports in the Aegean Bronze Age. Antiquity,
77(297), 471–84.
Asouti, E., 2003b. Woodland vegetation and fuel exploitation at the prehistoric
campsite of Pınarbaşı, south-central Anatolia, Turkey: The evidence from
the wood charcoal macro-remains. Journal of Archaeological Science, 30(9),
1185–201.
2005. Woodland vegetation and the exploitation of fuel and timber at
Neolithic Çatalhöyük: Report on wood-charcoal macro-remains, in
Inhabiting Çatalhöyük: Reports from the 1995–99 Seasons (Çatalhöyük
Research Project), ed. I. Hodder. Cambridge: MacDonald Institute for
Archaeological Research.
Asouti, E., & D. Fuller, 2012. From foraging to farming in the Southern Levant:
The development of Epipalaeolithic and Pre-pottery Neolithic plant man-
agement strategies. Vegetation History and Archaeobotany, 21(2), 149–62.
Asouti, E., & J. Hather, 2001. Charcoal analysis and the reconstruction of ancient
woodland vegetation in the Konya Basin, south-central Anatolia, Turkey:
Results from the Neolithic site of Çatalhöyük East. Vegetation History and
Archaeobotany, 10(1), 23–32.
Assier, A., 1996. Glaciers et Glaciers Rocheux de l’Ubaye, Barcelonnette: Sabenca
de la Valeia.
Atherden, M. A., & J. A. Hall, 1999. Human impact on vegetation in the White
Mountains of Crete since AD 500. The Holocene, 9(2), 183–94.
Atherden, M. A., J. Hall, & J. C. Wright, 1993. A pollen diagram from the
northeast Peloponnese, Greece: Implications for vegetation history and
archaeology. The Holocene, 3(4), 351–6.
Attema, P., T. de Haas, & G. W. Gol, 2011. Between Satricum and Antium:
Settlement dynamics in a coastal landscape in Latium Vetus. Leuven:
Peeters.
Attema, P., & J. Delvigne, 2000. Settlement dynamics and alluvial sedimen-
tation in the Pontine region, central Italy: A complex relationship, in
Geoarchaeology of the Landscapes of Classical Antiquity, eds. F. Vermeulen &
M. De Dapper. Leiden: Stichting Babesch, pp. 35–47.
Attema, P., G.-J. Burgers, & P. M. van Leusen, 2010. Regional Pathways to
Complexity: Settlement and Land-Use Dynamics in Early Italy from the
Bronze Age to the Republican Period. Amsterdam: Amsterdam University
Press.
Aubet, M. E., & M. Turton, 2001. The Phoenicians and the West: Politics, Colonies
and Trade. Cambridge: Cambridge University Press.
Auriemma, R., & E. Solinas, 2009. Archaeological remains as sea level change
markers: A review. Quaternary International, 206(1–2), 134–46.
Avigliano, R., G. Di Anastasio, S. Improta, M. Peresani, & C. Ravazzi, 2000.
A new late glacial to Early Holocene palaeobotanical and archaeological
Bibliograph 297
record in the Eastern Pre-Alps: The Palughetto basin (Cansiglio Plateau,

Italy). Journal of Quaternary Science, 15(8), 789–804.
Avni, Y., N. Porat, J. Plakht, & G. Avni, 2006. Geomorphic changes leading
to natural desertification versus anthropogenic land conservation in an
arid environment, the Negev Highlands, Israel. Geomorphology, 82(3–4),
177–200.
Ayala, G., & C. French, 2003. Holocene landscape dynamics in a Sicilian upland
river valley, in Alluvial Archaeology in Europe, eds. A. J. Howard, M. G.
Macklin, & D. G. Passmore. Lisse: Swets & Zeitlinger, pp. 229–35.
Ayala, G., & C. French, 2005. Erosion modeling of past land-use practices in the
Fiume di Sotto di Troina river valley, north-central Sicily. Geoarchaeology,
20(2), 149–68.
Bachhuber, C., & R. G. Roberts, 2008. Forces of Transformation: The End of the
Bronze Age in the Mediterranean. Oxford: Oxbow Books.
Bailey, D. W., 1993. Chronotypic tension in Bulgarian prehistory: 6500–3500
BC. World Archaeology, 25(2), 204–22.
Bailey, G., 2007. Time perspectives, palimpsests and the archaeology of time.
Journal of Anthropological Archaeology, 26(2), 198–223.
Balée, W., 2006. The research program of historical ecology. Annual Review of
Anthropology, 35(1), 75–98.
Ballais, J. L., 1994. Aeolian activity, desertification and the ‘green dam’ in the
Ziban range, Algeria, in Environmental Change in Drylands: Biogeographical
and Geomorphological Perspectives, eds. A. C. Millington & K. Pye. London:
Wiley, pp. 177–98.
Ballais, J. L., & A. Crambes, 1993. Morphogenèse Holocène et anthropisa-
tion sur la Montagne Sainte Victoire, in Archéologie et Environnement de
la Sainte-Victoire aux Alpilles, eds. P. Leveau & M. Provansal. Aix-en-
Provence: Publications Université de Provence, pp. 467–83.
Ballais, J. L., & P. Bellamy, 2000. Le pont Simian à Fontvieille: Étude géo-
archéologique d’un pont-aqueduc, in Milieux et Sociétés dans la Vallée des
Baux, eds. P. Leveau & J. P. Saquet. Montpellier: Revue archéologique de
Narbonnaise, pp. 25–38.
Balmuth, M. S., D. K. Chester, & P. A. Johnston, 2003. Cultural Responses to
the Volcanic Landscape: The Mediterranean and Beyond (AIA Colloquia and
Conference Papers). Boston: Archaeological Institute of America.
Bar-Matthews, M., & A. Ayalon, 2011. Mid-Holocene climate variations revealed
by high-resolution speleothem records from Soreq Cave, Israel and their
correlation with cultural changes. The Holocene, 21(1), 163–71.
Barge, H., & B. Talon, 2012. Attaque au feu au Bronzeancien sur le gîte de
cuivre de Saint-Véran (Hautes-Alpes). Bulletin de la Société Préhistorique
Française, 109(1), 145–54.
Barker, G., 2002. A tale of two deserts: Contrasting desertification histories on
Rome’s desert frontiers. World Archaeology, 33(3), 488–507.
298 Bibliograph
Barker, G., & C. Hunt, 2003. The role of climate and human settlement in
the evolution of the Biferno Valley (Molise, Central-Southern Italy), in
Variazioni Climatico-Ambientali e Impatto sull’uomo nell’area circum-
Mediterranean durante l’Olocene, eds. C. A. Livadie & F. Ortolani. Bari:
Edipuglia, pp. 183–91.
Barker, G., D. Gilbertson, C. O. Hunt, & D. Mattingly, 1996. Roman-Libyan
agriculture: Integrated models, in Farming the Desert: The UNESCO Libyan
Valleys Archaeological Survey, ed. G. Barker. London: Society for Libyan
Studies, pp. 265–90.
Barriendos, M., 1997. Climatic variations in the Iberian Peninsula during the
late Maunder Minimum (AD 1675–1715): An analysis of data from roga-
tion ceremonies. The Holocene, 7(1), 105–12.
Barriendos, M., & M. C. Llasat, 2003. The case of the ‘Malda’ anomaly in the
Western Mediterranean Basin (AD 1760–1800): An example of a strong
climatic variability. Climatic Change, 61(1/2), 191–216.
Barton, C. M., J. Bernabeu, J. E. Aura, & O. Garcia, 1999. Land-use dynam-
ics and socioeconomic change: An example from the Polop Alto Valley.
American Antiquity, 64(4), 609–34.
Barton, C. M., J. Bernabeu, J. E. Aura, O. Garcia, & N. La Roca, 2002. Dynamic
landscapes, artifact taphonomy, and landuse modeling in the Western
Mediterranean. Geoarchaeology, 17(2), 155–90.
Baruch, U., 1986. The Late Holocene vegetational history of Lake Kinneret
(Sea of Galilee), Israel. Paléorient, 12(2), 37–48.
1999. The Contribution of Palynology and Anthracology to Archaeological
Research in the Southern Levant, eds. S. Pike & S. Gitin. London: Archetype,
pp. 17–28.
Bass, B., 1998. Early Neolithic offshore accounts: Remote islands, mari-
time exploitations, and the trans-Adriatic cultural network. Journal of
Mediterranean Archaeology, 11(2), 165–90.
Baylis-Smith, T., & J. Golson, 1999. The meaning of ditches: Deconstructing
the social landscapes of New Guinea, Kuk, phase, in The Prehistory of Food:
Appetites for Change, ed. J. G. Hather. London: Routledge, pp. 199–231.
Beagon, M., 1996. Nature and views of her landscapes in Pliny the Elder, in
Human Landscapes in Classical Antiquity: Environment and Culture, eds.
G. Shipley & J. Salmon. London: Routledge, pp. 284–309.
Bellotti, P., F. L. Chiocci, S. Milli, & P. Tortora, 1994. Sequence stratigraphy
and depositional setting of the Tiber delta: Integration of high-resolution
seismics, well logs, and archeological data. Journal of Sedimentary Research
Section B, 64(3), 416.
Benedetti, L., R. Finkel, D. Papanastassiou, G. King, R. Armijo, F. Ryerson,
D. Farber, & F. Flerit, 2002. Post-glacial slip history of the Sparta fault
(Greece) determined by 36 cl cosmogenic dating: Evidence for non-periodic
earthquakes. Geophysical Research Letters, 29(8), 1–87.
Bibliograph 299
Beniston, M., 2000. Environmental Change in Mountains and Uplands. London:

Arnold.
Benito, G., A. Sopena, Y. Sanchez-Moya, M. J. Machado, & A. Perez-Gonzalez,
2003. Palaeoflood record of the Tagus River (central Spain) during the
Late Pleistocene and Holocene. Quaternary Science Reviews, 22(15–17),
1737–56.
Benito, G., V. R. Thorndycraft, M. Rico, Y. Sánchez-Moya, & A. Sopeña, 2008.
Palaeoflood and floodplain records from Spain: Evidence for long-term
climate variability and environmental changes. Geomorphology, 101(1–2),
68–77.
Benoit, G. & A. Comeau, 2005. Méditerranée: Les Perspectives du Plan Bleu sur
l’Environnement et le Développement. Sophia Antipolis: Éditions de l’Aube.
Benoit, G., A. Comeau, L. Chabason, 2005. Méditerranée: Les Perspectives du
Plan Bleu sur l’Environnement et le Développement. Valbonne: Éditions de
l’Aube.
Benouar, D., 2004. Materials for the investigation of historical seismicity in
Algeria from the records of past earthquakes. Annals of Geophysics, 47(2/3),
555–60.
Benvenuti, M., M. Bonini, G. Moratti, M. Ricci, & C. Tanini, 2008. Tectonic
and climatic controls on historical landscape modifications: The avulsion of
the lower Cecina River (Tuscany, central Italy). Geomorphology, 100(3–4),
269–84.
Benvenuti, M., M. Mariotti-Lippi, P. Pallecchi, & M. Sagri, 2006. Late-Holocene
catastrophic floods in the terminal Arno River (Pisa, central Italy) from the
story of a Roman riverine harbour. The Holocene, 16(6), 863–76.
Bérato, J., 1995. L’Âge du Fer dans la dépression permienne, et dans les massifs
des Maures et de l’Esétrel (Var). Documents d’Archéologie Méridionale, 18,
45–77.
Berger, J. F., 2003. Les étapes de la morphogenèse Holocène dans le sud
de la France, in Archéologie et Systèmes Socio-environnementaux: Études
Multiscalaires sur la Vallée du Rhône dans le Programme ARCHEOMEDES,
eds. S. van der Leeuw, F. Favory, & J. L. Fiches. Paris: CNRS, pp. 87–167.
2005. Sédiments, dynamique du peuplement et climat au Néolithique ancien,
in Populations Néolithiques et Environnements, eds. J. Guilaine, P. Chambon,
J. Zammit, J. D. Vigne, & Collectif. Paris: Éditions Errance, pp. 155–212.
Bernabeu, J., 1995. Origen y consolidación de las sociedades agrícolas. El País
Valenciano entre el Neolítico y la Edad del Bronce. Actes de les Segones
Jornades d’Arqueologia, 37–60.
Bernabeu, J., C. M. Barton, & M. Perez Ripoll, 2001. A taphonomic perspective
on Neolithic beginnings: Theory, interpretation, and empirical data in the
Western Mediterranean. Journal of Archaeological Science, 28, 597–612.
Bernabeu, J., & J. Bernabeu, 1993. Al Oeste del Eden. Valencia: Sintesis
Editorial.
300 Bibliograph
Betancourt, P. P., & K. Davaras, 1988. Excavations at Pseira, 1985 and 1986.
Hesperia, 57, 207–25.
Betancourt, P. P., & W. R. Farrand, 2006. The natural environment. Hesperia
Supplements, 36, 19–44.
Bettinger, R., P. Richerson, & R. Boyd, 2009. Constraints on the development
of agriculture. Current Anthropology, 50(5), 627–31.
Bettinger, R., 1991. Hunter-Gatherers: Archaeological and Evolutionary Theory.
New York: Plenum Press.
Bevan, A., 2002. The rural landscape of neopalatial Kythera: A GIS perspective.
Journal of Mediterranean Archaeology, 15(2), 217–55.
Bevan, A., J. Conolly & A. Tsaravopoulos, 2007. The fragile communities of
Antikythera. Archaeology International, 10, 32–6
Bevan, A., & J. Conolly, 2013. Mediterranean Islands, Fragile Communities and
Persistent Landscapes: Antikythera in Long-term Perspective. Cambridge:
Cambridge University Press.
Bevan, A., J. Conolly, & A. Tsaravopoulos, 2007. The fragile communities of
Antikythera. Archaeology International, 10, 32–6.
Bicket, A. R., H. M. Rendell, A. Claridge, P. Rose, J. Andrews, & F. S. J. Brown,
2010. A multiscale geoarchaeological approach from the Laurentine
Shore (Castelporziano, Lazio, Italy). Géomorphologie: Relief, Processus,
Environnement, 4(2009), 241–56.
Bicknell, P., 2000. Late Minoan IB ware, the marine environment of the Aegean,
and the Bronze Age eruption of the Thera volcano, in The Archaeology
of Geological Catastrophes (Geological Society Special Publication), eds.
D. McGuire, D. Griffiths, P. L. Hancock, & I. Stewart. London: Geological
Society Publishing House, pp. 95–103.
Billings, W. D., 1979. High mountain ecosystems, in High-altitude Geoecology,
ed. P. J. Webber. Boulder: Westview Press, pp. 97–125.
Binford, L. R., 1981. Behavioral archaeology and the ‘Pompeii Premise’. Journal
of Anthropological Research, 37(3), 195–208.
Bini, M., H. Brückner, A. Chelli, M. Pappalardo, S. Da Prato, & L. Gervasini,
2012. Palaeogeographies of the Magra Valley coastal plain to constrain
the location of the Roman harbour of Luna (NW Italy). Palaeogeography,
Palaeoclimatology, Palaeoecology, 37–51, 337–8.
Bintliff, J., 1993. Forest cover, agricultural intensity and population density in
Roman imperial Boeotia, central Greece, in Evaluation of Land Surfaces
Cleared from Forests in the Mediterranean Region During the Time of the
Roman Empire. Stuttgart: Fischer Verlag, pp. 133–43.
1997. Regional survey, demography, and the rise of complex societies in the
ancient Aegean: Core-periphery, meo-Malthusian, and other interpretive
models. Journal of Field Archaeology, 24(1), 1–38.
2002. Time, process and catastrophism in the study of Mediterranean alluvial
history: A review. World Archaeology, 33(3), 417–35.
Bibliograph 301
2009. The implications of a phenomenology of landscape, in Die Landschaft

und die Religion. Stuttgarter Kolloquium zur Historischen Geographie des
Altertums 9, 2005, ed. E. Olshausen & V. Saue. Stuttgart: Franz Steiner
Verlag, pp. 27–45.
2012. The Complete Archaeology of Greece. New York: John Wiley.
Bintliff, J., E. Farinetti, K. Sarri, & R. Sebastiani, 2006. Landscape and early
farming settlement dynamics in central Greece. Geoarchaeology, 21(7),
665–74.
Biserni, G., & B. Van Geel, 2005. Reconstruction of Holocene palaeoenvi-
ronment and sedimentation history of the Ombrone alluvial plain (South
Tuscany, Italy). Review of Palaeobotany and Palynology, 136 (1–2), 16–28.
Blackman, D., 1982a. Ancient harbours in the Mediterranean, Part 1.
International Journal of Nautical Archaeology and Underwater Archaeology,
11(2), 79–104.
1982b. Ancient harbours in the Mediterranean, Part 2. International Journal
of Nautical Archaeology and Underwater Archaeology, 11(3), 185–211.
Blaise, E., & M. Balasse, 2011. Seasonality and season of birth of modern and
late Neolithic sheep from south-eastern France using tooth enamel δ18O
analysis. Journal of Archaeological Science, 38(11), 3085–93.
Blanchemanche, P., 1990. Bâtisseurs de Paysages: Terrassement, Pierrement et
Petite Hydraulique Agricoles en Europe (XVIIe–XIXe Siècles). Paris: Maison
des Sciences de l’Homme.
Blanchemanche, P., L. Chabal, C. Jorda, & C. Jung, 2004. Le Delta du Lez
dans tous ses états: quels langages pour quel dialogue?, in Fleuves et Marais,
Une Histoire au Croisement de la Nature et la Culture, eds. J. Burnouf &
P. Leveau. Paris: CTHS, pp. 157–74.
Blondel, J., 1986. Biogéographie Évolutive. Paris: Masson.
2008. On humans and wildlife in Mediterranean islands. Journal of
Biogeography, 35(3), 509–18.
Blondel, J., & J. Aronson, 1999. Biology and Wildlife of the Mediterranean
Region. Oxford: Oxford University Press.
Blondel, J., J. Aronson, J. Y. Bodiou, & G. Boeuf, 2010. The Mediterranean
Region: Biological Diversity in Space and Time. Oxford: Oxford University
Press.
Blumler, M. A., 2007. Near Eastern pollen diagrams and ‘deforestation’. Middle
States Geographer, 40, 150–7.
Bocquet, A., 1997. Archéologie et peuplement des Alpes françaises du Nord au
Néolithique et aux âges des Métaux. L’Anthropologie, 101(2), 291–393.
Bofinger, J. P., P. Schweizer, & M. Strobel, 1996. Das Oppidum Bramefan (com.
Puyloubier, dép. Bouches-du-Rhône) und die südfranzösiche Eisenzeit: Ein
Beitrag zum Stand de Forschung (Seminar Marburg, Sonderban 10, 1996),
in Europe Celtique, Untersuchungen zur Hallstatt-und Latsnekultur, ed.
T. Stillner. Marburg: Verlag Marie Leidorf, pp. 55–84.
302 Bibliograph
Boissinot, P., 2000. L’environnement at la construction des paysages, in Les

Temps des Gaulois en Provence, ed. J. Chausserie-Laprée. Martigues: Musée
Ziem, pp. 26–30.
2001. Archéologie des vignobles antiques du sud de la Gaule. Gallia, 45–68.
Bonanno, A., 2005. L’acqua presso le comunità preistoriche di Malta, in Il Culto
Mediterraneo delle Acque, Sito archeologico di S. Cristina, 26–28 settembre
2003.
Bonanno, A., T. Gouder, C. Malone, & S. Stoddart, 1990. Monuments
in an island society: The Maltese context. World Archaeology, 22(2),
190–205.
Bottari, C., M. D’Amico, M. Maugeri, A. Bottari, G. D’Addezio, B. Privitera,
& G. Tigano, 2009. Location of the ancient Tindari harbour from geoar-
chaeological investigations (NE Sicily). Environmental Archaeology, 14(1),
37–49.
Bottari, C., & P. Carveni, 2009. Archaeological and historiographical impli-
cations of recent uplift of the Peloro Peninsula, NE Sicily. Quaternary
Research, 72(1), 38–46.
Bottema, S., 1994a. Forest, forest clearance and open land at the time of the
Roman Empire in Greece: The palynological record, in Evaluation of Land
Surfaces Cleared from Forests in the Mediterranean Region During the Time
of the Roman Empire, eds. B. Frenzel & L. Reisch. Stuttgart: Gustav Fischer
Verlag GmbH, pp. 59–72.
1994b. The prehistoric environment of Greece: A review of the palynological
record, in Beyond the Site – Regional Studies in the Aegean Area, ed. P. N.
Kardulias. New York: University Press of America, pp. 45–68.
Bottema, S., & A. Sarpaki. 2003. Environmental change in Crete: A 9,000-year
record of Holocene vegetation history and the effect of the Santorini erup-
tion. The Holocene, 13(5), 733–50.
Bottema, S., H. Woldring, & B. Aytug, 1986. Palynological investigations on the
relation between prehistoric man and vegetation in Turkey: The Beysehir
occupation phase, in 5th OPTIMA Meeting, Istanbul, pp. 315–28.
Bouby, L., F. Leroy, & L. Carozza, 1999. Food plants from late Bronze Age
lagoon sites in Languedoc, southern France: Reconstruction of farming
economy and environment. Vegetation History and Archaeobotany, 8(1/2),
53–70.
Bourgarit, D., P. Rostan, E. Burger, L. Carozza, & G. Artioli, 2008. The begin-
ning of copper mass production in the western Alps: The Saint Véran min-
ing area reconsidered. Historical Metallurgy, 42(1), 1–11.
Bover, P., & J. A. Alcover, 2003. Understanding Late Quaternary extinctions:
The case of Myotragus balearicus (Bate, 1909). Journal of Biogeography,
30(5), 771–81.
Bover, P., & J. A. Alcover, 2008. Extinction of the autochthonous small mammals
of Mallorca (Gymnesic Islands, Western Mediterranean) and its ecological
consequences. Journal of Biogeography, 35(6), 1112–22.
Bibliograph 303
Boyer, P., N. Roberts, & D. Baird, 2006. Holocene environment and settlement
on the Carsamba alluvial fan, south-central Turkey: Integrating geoarchae-
ology and archaeological field survey. Geoarchaeology, 21(7), 675–98.
Bradley, R., 2002. The Past in Prehistoric Societies. London: Routledge.
Brandon, C. J., L. H. Robert, & P. O. John, 2008. The concrete construction of
the Roman harbours of Baiae and Portus Iulius, Italy: The ROMACONS 2006
field season. International Journal of Nautical Archaeology, 37(2), 374–9.
Brandon, L. D., 2012. The influence of climatic change on the Late Bronze Age
Collapse and the Greek Dark Ages. Journal of Archaeological Science, 39(6),
1862–70.
Braudel, F., 1985. La Méditerranée et le Monde Méditerrannéen à l’Époque de
Philippe II. Paris: Arnaud Colin.
Brayshaw, D. J., C. M. C. Rambeau, & S. J. Smith, 2011. Changes in
Mediterranean climate during the Holocene: Insights from global and
regional climate modelling. The Holocene, 21(1), 15–31.
Bréhard, S., A. Beeching, & J. D. Vigne, 2010. Shepherds, cowherds and site
function on middle Neolithic sites of the Rhône valley: An archaeozoo-
logical approach to the organization of territories and societies. Journal of
Anthropological Archaeology, 29(2), 179–88.
Bresson, A., 2005. Ecology and beyond: The Mediterranean paradigm, in
Rethinking the Mediterranean, ed. W. V. Harris. Oxford: Oxford University
Press, pp. 94–114.
Brewster, H., 1997. The River Gods of Greece: Myths and Mountain Waters in the
Hellenic World. London: I. B. Tauris.
Brochier, J. L., & A. Beeching, 2006. Grottes bergeries, pastoralisme et mobilit‚
dans les Alpes au Néolithique, in Aux Origines de la Transhumance: Les
Alpes et la Vie Pastorale d’Hier à Aujourd’hui, eds. C. Jourdain-Annequin
& J. C. Duclos. Paris: Picard, pp. 131–57.
Broodbank, C., 1999. Colonization and configuration in the insular Neolithic
of the Aegean, in Neolithic Society in Greece, ed. P. Halstead. Sheffield:
Sheffield Academic Press, pp. 15–41.
2000. An Island Archaeology of the Early Cyclades. New York: Cambridge
University Press.
2004. Minoanisation. Proceedings of the Cambridge Philological Society, 50,
46–91.
2006. The origins and early development of Mediterranean maritime activity.
Journal of Mediterranean Archaeology, 19(2), 199–230.
2008. The Early Bronze Age in Cyclades, in The Aegean Bronze Age, ed. C. W.
Shelmerdine. Cambridge: Cambridge University Press, pp. 47–76.
Brown, A. G., 1997. Alluvial Geoearchaeology. Cambridge: Cambridge University
Press.
Brown, A. G., & C. Ellis, 1996. People, climate and alluviation: Theory, research
design and new sedimentological and stratigraphic data from Etruria. Papers
of the British School at Rome, 63, 45–73.
304 Bibliograph
Brown, A.G., & K. Walsh, forthcoming. Archaeology and hydro-environ-

ment in the Stymphalos Valley from the prehistoric to classical periods.
Geoarchaeology.
Brückner, H., 1997. Coastal changes in western Turkey: Rapid delta prograda-
tion in historical times. Bulletin de l’Institut oc‚ Anographique (Monaco),
18, 63–74.
Brückner, H., 2001. Delta Evolution and Culture – Aspects of Geoarchaeological
Research in Miletos and Priene, eds. G. A. Wagner, E. Pernicka, & H. P.
Uerpmann. Berlin: Springer, pp. 121–42.
Bruins, H. J., J. A. MacGillivray, C. Synolakis, C. Benjamini, J. Keller, H. J. Kisch,
A. Klugel, & J. van der Plicht, 2008. Geoarchaeological tsunami deposits at
Palaikastro (Crete) and the Late Minoan IA eruption of Santorini. Journal
of Archaeological Science, 35(1), 191–212.
Brun, J. P., 2004. Archéologie du Vin et de l’Huile, de la Préhistoire à l’Époque
Hellenistique. Paris: Éditions Errance.
Bruneton, H., 1999. Evolution Holocène d’un hydrosystème nord méditerranéen et
son environnement géomorphologique. Les plaines d’Arles à l’interface entre le
massif des Alpilles et le Rhône. Unpublished PhD thesis.
2000. La dynamique Holocène des paysages du Marais des Baux: Une pre-
mière approche morpho-sédimentaire, in Milieux et Sociétés dans la Vallée des
Baux, eds. P. Leveau & J. P. Saquet. Montpellier: Éditions de l’Association
de la Revue archéologique de Narbonnaise, pp. 15–24.
Bruneton, H., G. Arnaud-Fassetta, M. Provansal, & D. Sistach, 2001.
Geomorphological evidence for fluvial change during the Roman period in
the lower Rhone valley (southern France). Catena, 45 (4), 287–312.
Bruni, S., & M. Cosci, 2003. ‘Alpheae veterem contemptlor originis urbem
quam cingunt geminis Arnus et Ausur aquis’. Il paesaggio di Pisa etrusca
e romana: materiali e problemi, in Il Porto Urbano Di Pisa Antica: La
Fase Etrusca il Contesto e il Relitto Ellenostico, ed. S. Bruni. Milan: Silvana
Editoriale, pp. 29–43.
Buchsenschutz, O., 2004. Réflexion sur les parcellaires de L’âge du Fer en France
septentrionale, in La Dynamique des Paysages Protohistoriques, Antiques,
Médiévaux et Modernes, eds. J. Burnouff, J. P. Bravard, & G. Chouquer.
Sophia Antipolis: Éditions APDCA, pp. 13–20.
Bugalho, M. N., M. C. Caldeira, J. S. Pereira, J. Aronson, & J. G. Pausas, 2011.
Mediterranean cork oak savannas require human use to sustain biodiver-
sity and ecosystem services. Frontiers in Ecology and the Environment, 9(5)
278–86.
Bugalho, M. N., X. Lecomte, M. Gonçalves, M. C. Caldeira, & M. Branco,
2011. Establishing grazing and grazing-excluded patches increases plant
and invertebrate diversity in a Mediterranean oak woodland. Forest Ecology
and Management, 261(11), 2133–9.
Burford, A. 1993. Land and Labor in the Greek World. Baltimore: Johns Hopkins
University Press.
Bibliograph 305
Burjachs, F., R. Perez-Obiol, J. M. Roure, & R. Julia, 1994. Dinaïmica de la

vegetacioïn durante el Holoceno en la isla de Mallorca, in Trabajos de pal-
inología Básica y Aplicada, eds. I. Mateu, M. Dupré, J. Güemes, & M. E.
Burgaz. Valencia: Universitat de València, pp. 199–210.
Burjachs, F., & L. Schulte, 2003. El paistage vegetal del Penedès entre la prehisto-
ria i el mon antic, in Territoris antics a la Mediterrània i a la Cossetània orien-
tal: actes del Simposi internacional d’arqueologia del Baix Penedès, El Vendrell,
del 8 al 10 de novembre de 2001, eds. J. Guitart, J. M. Palet, & M. Prevosti.
Barcelona: Generalitat de Catalunya Departament de Cultura, pp. 249–54.
Burroughs, W. J., 2005. Climate Change in Prehistory: The End of the Reign of
Chaos. Cambridge: Cambridge University Press.
Butler, S., 2006. The Odyssey: Rendered Into English Prose for the Use of Those Who
Cannot Read the Original. Whitefish: Kessinger.
Butzer, K. W., 2005. Environmental history in the Mediterranean world: Cross-
disciplinary investigation of cause-and-effect for degradation and soil ero-
sion. Journal of Archaeological Science, 32(12), 1773–800.
2012. Collapse, environment, and society. Proceedings of the National Academy
of Sciences, 109(10), 3632–9.
Buxton, R., 1994. Imaginary Greece: The Contexts of Mythology. Cambridge:
Cagatay, M. N., N. Gorur, O. Algan, C. Eastoe, A. Tchapalyga, D. Ongan, T.
Kuhn, & I. Kuscu, 2000. Late Glacial-Holocene palaeoceanography of the
Sea of Marmara: Timing of connections with the Mediterranean and the
Black Seas. Marine Geology, 167(3–4), 191–206.
Calmel-Avila, M., 2002. The Librilla ‘rambla’, an example of morphogenetic
crisis in the Holocene (Murcia, Spain). Quaternary International, 93–94,
101–8.
Camin, L., & W. McCall, 2002. Settlement patterns and rural habitation in the
middle Cecina Valley between the Hellenistic to Roman age: The case of
Podere Cosciano. Etruscan Studies, 9(1), 19–27.
Caputo, R., J. P. Bravard, & B. Helly, 1994. The Pliocene-Quaternary tecto-
sedimentary evolution of the Larissa Plain (Eastern Thessaly, Greece).
Geodinamica Acta, 7(4), 219–31.
Caravello, G. U., & F. Giacomin, 1993. Landscape ecology aspects in a terri-
tory centuriated in Roman times. Landscape and Urban Planning, 24(1–4),
77–85.
Carayon, N., 2005. Le cothon ou port artificiel creusé. Essai de définition.
Méditerranée, 104(1.2), 5–13.
Carmona, P., & J. M. Ruiz, 2008. Geoarchaeological study of the Phoenician
cemetery of Tyre-Al Bass (Lebanon) and geomorphological evolution of a
tombolo. Geoarchaeology, 23(3), 334–50.
Caroli, I., & M. Caldara, 2007. Vegetation history of Lago Battaglia (eastern
Gargano coast, Apulia, Italy) during the Middle–Late Holocene. Vegetation
History and Archaeobotany, 16(4), 317–27.
306 Bibliograph
Carozza, L., D. Galop, F. Marembert, & F. Monna, 2005. Quel statut pour
les espaces de montagne durant l’âge du Bronze? Regards croisés sur les
approches société-environnement dans les Pyrénées occidentales. Documents
d’Archéologie Méridionale, 28, 7–23.
Carrer, F., 2012. Etnoarcheologia dei Paesaggi Pastorali nelle Alpi Strategie
Insediative Stagionali d’Alta Quota in Trentino, Università degli Studi di
Trento. Unpublished PhD thesis.
Carrión, J. S., A. Andrade, K. D. Bennett, C. Navarro, & M. Munuera, 2001.
Crossing forest thresholds: Inertia and collapse in a Holocene sequence
from south-central Spain. The Holocene, 11(6), 635–54.
Carrión, J. S., & B. Van Geel, 1999. Fine-resolution Upper Weichselian and
Holocene palynological record from Navarrés (Valencia, Spain) and a
discussion about factors of Mediterranean forest succession. Review of
Palaeobotany and Palynology, 106(3–4), 209–36.
Carrión, J. S., E. I. Yll, K. J. Willis, & P. Sánchez, 2004. Holocene forest his-
tory of the eastern plateaux in the Segura Mountains (Murcia, southeastern
Spain). Review of Palaeobotany and Palynology, 132(3–4), 219–36.
Carrión, J. S., P. Sánchez-Gomez, J. F. Mota, R. Yll, & C. Chain, 2003.
Holocene vegetation dynamics, fire and grazing in the Sierra de Gador,
southern Spain. The Holocene, 13(6), 839–50.
Carrión, J. S., S. Fernández, P. González-Sampériz, G. Gil-Romera, E. Badal,
Y. Carrión-Marco, L. López-Merino, J. A. López-Sáez, E. Fierro, &
F. Burjachs, 2010. Expected trends and surprises in the Lateglacial and
Holocene vegetation history of the Iberian Peninsula and Balearic Islands.
Review of Palaeobotany and Palynology, 162(3), 458–75.
Casana, J., 2008. Mediterranean valleys revisited: Linking soil erosion, land use
and climate variability in the Northern Levant. Geomorphology, 101(3),
429–42.
Castro, P., S. Gili, R. W. Chapman, E. I. Yll, V. Lull, R. Mico & R. Risch,
2000. Archaeology and desertification in the Vera basin (Almería, south-
east Spain). European Journal of Archaeology, 3(2), 147–66.
Castro, P., R. W. Chapman, S. Gili, V. Lull, R. Mico, C. Rihuete, R. Risch, &
M. E. Sanahuja, 1999. Agricultural production and social change in the Bronze
Age of southeast Spain: The Gatas Project. Antiquity, 73(282), 846–56.
Cenzon-Salvayre, C., & A. Durand, 2011. The cremation structures of the Roman
Empire: Anthracological data versus historical sources, in 5th International
Meeting of Charcoal Analysis: Charcoal as Cultural and Biological Heritage,
Saguntum, 11, pp. 191–2.
Chabal, L., 1998. Forêts et Sociétés en Languedoc (Néolithique Final, Antiquité
Tardive). Paris: Documents d’Archéologie Française.
Chaniotis, A., 1999. Milking the mountains: Economic activities on the Cretan
Uplands in the classical and Hellenistic period, in From Minoan Farmers to
Roman Traders: Sidelights on the Economy of Ancient Crete, ed. E. Chaniotis.
Stuttgart: Franz Steiner Verlag, pp. 181–219.
Bibliograph 307
Chapman, R., 2008. Producing inequalities: Regional sequences in later prehis-

toric southern Spain. Journal of World Prehistory, 21(3–4), 195–260.
Châtelain, T., 2007. La Grèce antique et ses marais: Perception des milieux palus-
tres chez les Anciens. Unpublished PhD thesis, Université de Neuchâtel and
Université de Paris IV – Sorbonne.
Chen, L., K. A. F. Zonneveld, & G. J. M. Versteegh, 2012. Short term climate
variability during ‘Roman Classical Period’ in the eastern Mediterranean.
Quaternary Science Reviews, 30(27–8), 3880–91.
Cherkauer, D. S., 1976. The stratigraphy and chronology of the River Treia
alluvial deposits, in A Faliscan Town in South Etruria. Excavations at Narce
1966–71, ed. T. W. Potter. London: British School at Rome, pp. 106–20.
Cherry, J. F., 1990. The first colonisation of the Mediterranean Islands: A review
of recent research. Journal of Mediterranean Archaeology, 3(2), 145–221.
2004. Mediterranean island prehistory: What’s different and what’s new?,
in Voyages of Discovery: The Archaeology of Islands, ed. S. M. Fitzpatrick.
Westport: Greenwood, pp. 234–48.
Cita, M. B., & G. Aloisi, 2000. Deep-sea tsunami deposits triggered by the
explosion of Santorini (3,500y BP), eastern Mediterranean. Sedimentary
Geology, 135(1–4), 181–203.
Cita, M. B., A. Camerlenghi, & B. Rimoldi, 1996. Deep-sea tsunami depos-
its in the Eastern Mediterranean: New evidence and depositional models.
Sedimentary Geology, 104(1/4), 155–74.
Claridge, A. J., 2007. Interim report on archaeological fieldwork at Castelporziano
April–May 2007.Unpublished report. Accessed online (10 April 2013) from
http://www.rhul.ac.uk/classics/laurentineshore/ASSETS/PDF-files/
Excavation%20report%20April%202008%20website%20version.pdf
Clark, G., 1968. Prehistoric Europe: The Economic Basis. London: Taylor & Francis.
Clendenon, C., 2009. Hydromythology and the Ancient Greek World: An Earth
Science Perspective Emphasizing Karst Hydrology. Michigan: Fineline Science
Press.
Clevis, Q., G. E. Tucker, G. Lock, S. T. Lancaster, N. Gasparini, A. Desitter, &
R. L. Bras, 2006. Geoarchaeological simulation of meandering river deposits
and settlement distributions: A three-dimensional approach. Geoarchaeology,
21(8), 843–74.
Cole, S. G., 1996. Demeter in the ancient Greek city and its countryside, in
Placing the Gods: Sanctuaries and Sacred Space in Ancient Greece, eds. S.
Alcock & R. Osborne. Oxford: Clarendon Press, pp. 199–216.
Collins, P. M., B. A. S. Davis, & J. O. Kaplan, 2012. The mid-Holocene vege-
tation of the Mediterranean region and southern Europe, and comparison
with the present day. Journal of Biogeography, 39(10), 1848–61.
Collombier, A. M., 1988. Harbour or harbours of Kition on southeastern
coastal Cyprus, in Archeology of Coastal Changes. Proceedings of the First
International Symposium ‘Cities on the Sea. Past and Present’, ed. A. Raban.
Oxford: British Archaeological Reports, pp. 35–46.
308 Bibliograph
Coltorti, M., 1993. Human impact in the Holocene fluvial and coastal evolution
of the Marche region, Central Italy. Catena, 30(4), 311–36.
Conacher, A. J., & M. Sala, 1998. Land Degradation in the Mediterranean
Environments of the World: Nature and Extent, Causes and Solutions.
Chichester: John Wiley.
Conedera, M., P. Krebs, W. Tinner, M. Pradella, & D. Torriani, 2004. The culti-
vation of Castanea sativa (Mill.) in Europe, from its origin to its diffusion on
a continental scale. Vegetation History and Archaeobotany, 13(3), 161–80.
Constantakopoulou, C., 2007. The Dance of the Islands: Insularity, Networks, the
Athenian Empire, and the Aegean World. Oxford: Oxford University Press.
Cortés, F. C., O. R. Ariza, J. A. C. Serrano, & A. M. Onorato, 1997. Hace 4000
Años … Vida y Muerte en dos Poblados de la Alta Andalucía. Grenada: Junta
de Andalucia.
Cosmopoulos, M. B., 1998. Reconstructing Cycladic prehistory: Naxos in the
Early and Middle Late Bronze Age. Oxford Journal of Archaeology, 17(2),
127–48.
Costa, L. J., 2004. Corse Préhistorique: Peuplement d’une Île et Modes de Vie des
Sociétés Insulaires (IXe–IIe Millénaires av. J.-C.). Paris: Éditions Errance.
2008. Mégalithismes Insulaires en Méditerranée. Paris: Éditions Errance.
Court-Picon, M., 2007. Mise en place du paysage dans un milieu de moyenne et
haute montagne du tardiglaciaire à l’époque actuelle, L’U.F.R. des Sciences et
Techniques de l’Université de Franche Comté. Unpublished PhD thesis.
Court-Picon, M., K. Walsh, F. Mocci, M. Segard, & J. M. Palet-Martinez, 2007.
Occupation de la montagne et transformation des milieux dans les Alpes
méridionales au cours de l’âge du Bronze: Approche croisée des données
palynologiques et archéologiques en Champsaur et Argentiérois (Hautes-
Alpes, France), in Environnements et Cultures à l’Âge du Bronze en Europe
Occidentale, eds. C. Mordant, H. Richard, & M. Magny. Aris: CTHS,
pp. 89–106.
Courtois, C., 1955. Les Vandales et l’Afrique. Paris: Arts et Métiers Graphiques.
Cremaschi, M., 2003. Steps and timing of the desertification during Late Antiquity.
The case study of the Tanezzuft Oasis (Libyan Shara), in Arid Lands in Roman
Times, ed. M. Liverani. Florence: All’Insegna del Giglio, pp. 1–14.
Cremaschi, M., & L. Trombino, 1998. The palaeoclimatic significance of paleo-
sols in Southern Fezzan (Libyan Sahara): Morphological and micromor-
phological aspects. Catena, 34(1/2), 131–56.
Cremaschi, M., & S. Di Lernia, 1999. Holocene climatic changes and cultural
dynamics in the Libyan Sahara. African Archaeological Review, 16(4),
211–38.
Cremaschi, M., C. Pizzi & V. Valsecchi, 2006. Water management and land use
in the terramare and a possible climatic co-factor in their abandonment:
The case study of the terramara of Poviglio Santa Rosa (northern Italy).
Quaternary International, 151(1), 87–98.
Bibliograph 309
Cribb, R., 1991. Nomads in Archaeology. Cambridge: Cambridge University

Press.
Crouch, D. P., 2004. Geology and Settlement: Greco-Roman Patterns. New York:
Oxford University Press.
Crumley, C. L., 1994a. The ecology of conquest: Contrasting agropastoral and
agricultural societies’ adaptation to climatic change, in Historical Ecology:
Cultural Knowledge and Changing Landscapes, ed. C. L. Crumley. Santa
Fe: School for Advanced Research Press, pp. 183–201.
1994b. Historical Ecology: Cultural Knowledge and Changing Landscapes.
Santa Fe: School for Advanced Research Press.
Dakoronia, P., 1996. Earthquakes of the Late Helladic III period (12th century
BC) at Kynos (Livantates, Central Greece), in Archaeoseismology, eds. S. C.
Stiros & R. E. Jones. Athens: Institute of Geology and Mineral Exploration
& The British School at Athens, pp. 41–4.
Dalongeville, R., & J. Rénault-Miskovsky, 1993. Paysages de l’île de Naxos,
in Recherches dans les Cycaldes, Résultats des Travaux de la RCP 583, eds.
R. Dalongeville & G. Rougemont. Lyon: Maison de l’Orient Méditerranéen,
pp. 21–32.
Davasse, B., D. Galop, & C. Rendu, 1997. Paysages du Néolithique à nos jours
dans les Pyrénées de l’Est d’après l’écologie historique et l’archéologie pas-
torale, in La Dynamique des Paysages Protohistoriques, Antiques, Médiévaux et
Modernes, eds. J. Burnouf, J. P. Bravard, & G. Chouquer. Sophia Antipolis:
Éditions APDCA, pp. 577–99.
Davidson, D., & C. Tasker, 1982. Geomorphological evolution during the Late
Holocene, in An Island Polity, The Archaeology of Exploitation in Melos,
eds. C. Renfrew & M. Wagstaff. Cambridge: Cambridge University Press,
pp. 82–94.
Davidson, D. A., & S. P. Theocharopoulos, 1992. A survey of soil erosion
in Viotia, Greece, in Past and Present Soil Erosion, Archaeological and
Geographical Perspectives, eds. M. Bell & J. Boardman. Oxford: Oxbow,
pp. 149–54.
Davis, J. L. & S. E. Alcock 1998. Sandy Pylos: An Archaeological History from
Nestor to Navarino. Austin: University of Texas Press.
Dawson, H., 2011. Island colonisation: settling the Neolithic question, in The
First Mediterranean Islanders: Initial Occupation and Survival Strategies, ed.
N. Phoca-Cosmetatou. Oxford: David Brown Book Company, pp. 31–53.
de Beaulieu, J.-L., 1977. Contribution pollenanalytique à l’histoire tardiglaci-
aire et holocène des Alpes méridonales françaises. Unpublished PhD thesis,
Université Aix-Marseille III, France.
de Beaulieu, J.-L., Y. Miras, V. Andrieu-Ponel, & F. Guiter, 2005. Vegetation
dynamics in north-western Mediterranean regions: Instability of the
Mediterranean bioclimate. Plant Biosystems – An International Journal
Dealing with all Aspects of Plant Biology, 139(2), 114–26.
310 Bibliograph
De Boer, J. Z., & J. R. Hale, 2000. The geological origins of the oracle at Delphi,
Greece. Special Publication – Geological Society of London, 171, 399–412.
de Haas, T., 2011. Fields, farms and colonists: Intensive field survey and early
Roman colonization in the Pontine region, central Italy. Unpublished PhD
thesis, Rijksuniversiteit Groningen.
De Lange, G. J., J. Thomson, A. Reitz, C. P. Slomp, M. Speranza Principato,
E. Erba, & C. Corselli, 2008. Synchronous basin-wide formation and
redox-controlled preservation of a Mediterranean sapropel. Nature Geosci,
1(9), 606–10.
De Meo, A., G. Espa, S. Espa, A. Pifferi, & U. Ricci, 2003. A GIS for the study
of the mid-Tiber valley. Comparisons between archaeological settlements of
the Sabine Tiberine area. Journal of Cultural Heritage, 4(3), 169–73.
De Pascale, A., R. Maggi, C. Montanari, & D. Moreno, 2006. Pollen, herds, jas-
per and copper mines: Economic and environmental changes during the 4th
and 3rd millennia BC in Liguria (NW Italy). Environmental Archaeology,
11(1), 115–24.
de Wrachien, D., R. Ragab, & A. Giordano, 2006. Climate change, land degrada-
tion, and desertification in the Mediterranean environment. Desertification
in the Mediterranean Region. A Security Issue, 3, 353–71.
Delano-Smith, C., 1996. Where was the ‘wilderness’ in Roman times, in Human
Landscapes in Classical Antiquity, eds. G. Shipley & J. Salmon. London:
Routledge, 154–79.
Delhon, C., S. Thiébault, & J.-F. Berger, 2009. Environment and landscape
management during the Middle Neolithic in southern France: Evidence
for agro-sylvo-pastoral systems in the Middle Rhone Valley. Quaternary
International, 200(1–2), 50–65.
Di Castri, F., 1981. Mediterranean-type Shrublands of the World, in
Mediterranean-Type Shrublands, eds. F. Di Castri, D. W. Goodall, & R. L.
Specht. Amsterdam: Elsevier, pp. 1–52.
di Pasquale, G., P. di Martino, & S. Mazzoleni, 2004. Forest history in the
Mediterranean region, in Recent Dynamics of the Mediterranean Vegetation
and Landscape, eds. S. Mazzoleni, G. di Pasquale, M. Mulligan, P. di
Martino, & F. Rego. Chichester: John Wiley, pp. 13–20.
Di Rita, F., A. Celant, & D. Magri, 2009. Holocene environmental instability in
the wetland north of the Tiber delta (Rome, Italy): Sea–lake–man interac-
tions. Journal of Paleoliminology, 44(1), 51–67.
Di Rita, F., & D. Magri, 2009. Holocene drought, deforestation and evergreen
vegetation development in the central Mediterranean: A 5,500 year record
from Lago Alimini Piccolo, Apulia, southeast Italy. The Holocene, 19(2),
295–306.
Diodato, N., G. Bellocchi, N. Romano, & G. B. Chirico, 2011. How the
aggressiveness of rainfalls in the Mediterranean lands is enhanced by climate
change. Climatic Change, 108(3), 591–9.
Bibliograph 311
Dixon, J. E., & A. H. F. Robertson, 1996. The Geological Evolution of the

Eastern Mediterranean (Geological Society Special Publication). London:
Geological Society of London.
Drake, D., & T. Hunt, 2009. Invasive rodents on islands: Integrating historical
and contemporary ecology. Biological Invasions, 11(7), 1483–7.
Drews, R., 1993. The End of the Bronze Age: Changes in Warfare and the
Catastrophe ca 1200 BC. Princeton: Princeton University Press.
Driessen, J., 1998. The Troubled Island: Minoan Crete Before and After the
Santorini Eruption. Liège: Université de Liège.
2002. Towards an archaeology of crisis: Defining the long-term impact of the
Bronze Age Santorini eruption, in Natural Disasters and Cultural Change,
eds. R. Torrence & J. Grattan. London: Routledge, pp. 250–63.
Dubar, M., F. Damblon, S. Nichol-Pichard, J. L. Vernet, L. Chaix, J. Cataliotti,
F. Irr, & J. F. Babinot, 1986. L’environnment côtier des Alpes Maritimes
à la fin de la transgression versilienne d’après l’étude biostratigraphique du
site de l’Etoile à Nice (France). Revue de Paléobiologie, 5(2), 289–310.
Duchêne, H., R. Dalongeville, & P. Bernier, 2001. Transformations du Paysage
Naturel et évolution du Littoral dans l’Archipel délien, in Le Paysage
Portuaire de la Délos Antique: Recherches sur les Installations Maritimes,
Commerciales et Urbaines du Littoral Délien, eds. H. Duchêne, P. Fraisse,
R. Dalongeville, & P. Bernier. Athens/Paris De Boccard: École Française
d’Athènes.
Duclos, J. C., 2006. La Pratique de la transhumance d’hier à aujourd’hui, in
Aux Origines de la Transhumance: Les Alpes et la Vie Pastorale d’Hier à
Aujourd’hui, eds. C. Jourdain-Annequin & J. C. Duclos. Paris: Picard,
pp. 17–23.
Dufaure, J. J., 1976. Le terrasse Holocène d’Olympie et ses équivalents méditer-
ranéens. Bulletin de l’Association Géographique Française, 433, 85–94.
Dusar, B., G. Verstraeten, B. Notebaert, & J. Bakker, 2011. Holocene
environmental change and its impact on sediment dynamics in the Eastern
Mediterranean. Earth-Science Reviews, 108(3–4), 137–57.
Dusar, B., G. Verstraeten, K. D’Haen, J. Bakker, E. Kaptijn, & M. Waelkens,
2012. Sensitivity of the Eastern Mediterranean geomorphic system towards
environmental change during the Late Holocene: A chronological perspec-
tive. Journal of Quaternary Science, 27(4), 371–82.
Eastwood, W. J., J. Tibby, N. Roberts, H. Birks, & H. F. Lamb, 2002. The envi-
ronmental impact of the Minoan eruption of Santorini (Thera): Statistical
analysis of palaeoecological data from Golhisar, southwest Turkey. The
Holocene, 12(4), 431–44.
Efstratiou, N., P. Biagi, P. Elefanti, P. Karkanas, & M. Ntinou, 2006. Prehistoric
exploitation of Grevena highland zones: Hunters and herders along the
Pindus chain of western Macedonia (Greece). World Archaeology, 38(3),
415–35.
312 Bibliograph
Ejarque, A., 2010. Génesis y configuracion microregional de un paisaje cul-

tural pirenaico de alta monta a durante el Holoceno: estudio polínico y de
otros indicadores paleoambientales en el valle del Madriu-Perafita-Claror
(Andorra), Institut Catalan d’Arqueologia Clàssica. Unpublished PhD the-
sis, Universitat Rovira i Virgili.
Ejarque, A., R. Julia, S. Riera, J. M. Palet, H. A. Orengo, Y. Miras, & C. Gascon,
2009. Tracing the history of highland human management in the eastern
Pre-Pyrenees: An interdisciplinary palaeoenvironmental study at the Pradell
fen, Spain. The Holocene, 19(8), 1241–55.
Ejarque, A., Y. Miras, S. Riera, J. M. Palet, & H. Orengo, 2010. Testing micro-
regional variability in the Holocene shaping of high mountain cultural land-
scapes: A palaeoenvironmental case-study in the eastern Pyrenees. Journal
Engel, T., 1993. Charcoal remains from an Iron Age copper smelting slag heap
at Feinan, Wadi Arabah (Jordan). Vegetation History and Archaeobotany,
2(4), 205–11.
Eriksen, T. H., 1993. In which sense do cultural islands exist? Social Anthropology,
1, 133–47.
Erlandson, J. M., & S. M. Fitzpatrick, 2006. Oceans, islands, and coasts: Current
perspectives on the role of the sea in human prehistory. Journal of Island
and Coastal Archaeology, 1, 5–32.
Evans, J. D., 1973. Islands as laboratories for the study of culture process, in
The Explanation of Culture Change: Models in Prehistory, ed. C. Renfrew.
London: Duckworth, pp. 517–20.
1977. Island archaeology in the Mediterranean: Problems and opportunities.
World Archaeology, 9(1), 12–26.
Evelpidou, N., P. Pirazzoli, A. Vassilopoulos, G. Spada, G. Ruggieri, & A.
Tomasin, 2012. Late Holocene sea level reconstructions based on observa-
tions of Roman fish tanks, Tyrrhenian coast of Italy. Geoarchaeology, 27(3),
259–77.
Evershed, R. P., S. Payne, A. G. Sherratt, M. S. Copley, J. Coolidge, D. Urem-
Kotsu, K. Kotsakis, M. Ozdogan, A. E. Ozdogan, O. Nieuwenhuyse,
P. M. M. G. Akkermans, D. Bailey, R.-R. Andeescu, S. Campbell, S. Farid,
I. Hodder, N. Yalman, M. Ozbasaran, E. Bicakci, Y. Garfinkel, T. Levy, &
M. M. Burton, 2008. Earliest date for milk use in the Near East and south-
eastern Europe linked to cattle herding. Nature, 455(7212), 528–31.
Excoffon, P., C. Landuré, & M. Pasqualini, 2004. Habitat et risque fluvial dans
le delta du Rhône au Ier s. av. J.- C. Les habitats de la Capelière et du Grand
Parc en Camargue, in Fleuves et Marais, une Histoire au Croisement de la
Nature et la Culture: Sociétés Préindustrielles et Milieux Fluviaux, Lacustres
et Palustres: Pratiques Sociales et Hydrosystèmes, eds. J. Burnouf & P. Leveau.
Paris: CTHS, pp. 213–34.
Bibliograph 313
Fairbairn, A., E. Asouti, J. Near, & D. Martinoli, 2002. Macro-botanical evi-

dence for plant use at Neolithic Catalhoyuk, southcentral Anatolia, Turkey.
Vegetation History and Archaeobotany, 11(1/2), 41–54.
Farinetti, E., 2009. Boeotian landscapes. A GIS-based study for the reconstruc-
tion and interpretation of the archaeological datasets of ancient Boeotia.
Unpublished PhD thesis, Leiden University.
Farjon, P., 2007. Enregistrements sédimentaires des changements climatiques et
des activités anthropiques depuis 3000BP dans le lac de la Roche-de-Rame,
Université Paul Cézanne, master’s SET.
Faust, D., C. Zielhofer, R. Baena Escudero, & D. del Olmo, 2004. High-
resolution fluvial record of late Holocene geomorphic change in northern
Tunisia: Climatic or human impact? Quaternary Science Reviews, 23(16–
17), 1757–75.
Fedele, F., 1992. Steinzeitliche Jäger in den Zentralalpen: Piano dei Cavalli
(Splügenpass). Helvetia Archaeologica, 89, 2–22.
Fenech, K., 2007. Human-Induced Changes in the Environment and Landscape
of the Maltese Islands from the Neolithic to the 15th Century AD. Oxford:
Archaeopress.
Figueiral, I., L. Bouby, L. Buffat, H. Petitot, & J. F. Terral, 2010. Archaeobotany,
vine growing and wine producing in Roman southern France: The site of
Gasquinoy (Béziers, Hérault). Journal of Archaeological Science, 37(1),
139–49.
Finné, M., & K. Holmgren, 2010. Climate variability in the Eastern Mediterranean
and the Middle East during the Holocene, in The Urban Mind: Cultural
and Environmental Dynamics, eds. P. Sinclair, G. Nordquist, F. Herschend,
& C. Isendahl. Uppsala: African and Comparative Archaeology, Department
of Archaeology and Ancient History, pp. 29–60.
Finné, M., K. Holmgren, H. S. Sundqvist, E. Weiberg, & M. Lindblom, 2011.
Climate in the Eastern Mediterranean, and adjacent regions, during the
past 6,000 years – A review. Journal of Archaeological Science, 38(12),
3153–73.
Fischer, M., 2005. Peter the Iberian at Yavneh-Yam: Calamities in context, in
Concepts, Pratiques et Enjeux Environnemetaux dans l’Empire Romain, eds.
R. Bedon & E. Hermon. Limoges: Pulim, pp. 267–84.
Flemming, A., 2006. Post-processual landscape archaeology: A critique.
Cambridge Archaeological Journal, 16(3), 267–80.
Fletcher, W. J., T. Boski, & D. Moura, 2007. Palynological evidence for environ-
mental and climatic change in the lower Guadiana valley, Portugal, during
the last 13,000 years. The Holocene, 17(4), 481–94.
Fletcher, W. J., & M. F. Sánchez Goñi, 2008. Orbital- and sub-orbital-scale cli-
mate impacts on vegetation of the Western Mediterranean basin over the
last 48,000 yr. Quaternary Research, 70(3), 451–64.
314 Bibliograph
Florido, E., R. Auriemma, S. Faivre, I. Radić Rossi, F. Antonioli, S. Furlani,

& G. Spada, 2011. Istrian and Dalmatian fishtanks as sea-level markers.
Quaternary International, 232(1–2), 105–13.
Follieri, M., D. Magri, & L. Sadori, 1989. Pollen stratigraphical synthesis from
Valle di Castiglione (Roma). Quaternary International, 3–4, 81–4.
Forbes, H., 2000. Landscape exploitation via pastoralism, in Landscape and
Land Use in Postglacial Greece, eds. P. Halstead & C. Frederick. Sheffield:
Forenbaher, S., & P. Miracle, 2005. The spread of farming in the Eastern
Adriatic. Antiquity, 79, 514–28.
Forman, R. T. T., 1995. Land Mosaics: The Ecology of Landscapes and Regions.
Cambridge: Cambridge University Press.
Forsyth, T., 1998. Mountain myths revisited: Integrating natural and social envi-
ronmental science. Mountain Research and Development, 18(2), 107–16.
Fouache, E., 1999. L’Alluvionnement Historique en Grèce Occidentale et au
Péloponnèse: Géomorphologie: Géomorphologie, Archéologie, Histoire. Athens:
École Francaise d’Athènes.
Fouache, E., & R. Dalongeville, 2003. Recent relative variations in the shore-
lines: A contrastive approach of the shores of Croatia and southern Turkey,
in The Mediterranean World Environment and History, ed. E. Fouache.
Paris: Elsevier, pp. 467–78.
Foxhall, L., 1996. Feeling the earth move: Cultivation techniques on steep
slopes in classical antiquity, in Human Landscapes in Classical Antiquity:
Environment and Culture, eds. G. Shipley & J. Salmon. London: Routledge,
pp. 44–68.
2007. Olive Cultivation in Ancient Greece: Seeking the Ancient Economy.
Oxford: Oxford University Press.
Franco Mugica, F., M. Garcia Anton, & O. Sainz, 1998. Vegetation dynamics
and human impact in the Sierra de Guadarrama, Central System, Spain. The
Holocene, 8(1), 69–82.
Frayn, J. M., 1979. Subsistence Farming in Roman Italy. London: Centaur Press.
Frederick, C., & A. Krahtopoulou, 2000. Deconstructing agricultural terraces, in
Landscape and Land use in Postglacial Greece. Sheffield: Sheffield Academic
Press, pp. 79–94.
French, E. B., 1996. Evidence for an earthquake at Mycenae, in Archaeoseismology,
eds. S. C. Stiros & R. E. Jones. Athens: Institute of Geology and Mineral
Exploration & The British School at Athens, pp. 51–4.
French, R., 1994. Ancient Natural History. London: Routledge.
Frenzel, B., & L. Reisch, 1994. Evaluation of Land Surfaces Cleared from
Forests in the Mediterranean Region During the Time of the Roman Empire.
Stuttgart: Gustav Fischer Verlag.
Fretwell, S. D., & H. L. Lucas, 1969. On territorial behavior and other factors
influencing habitat distribution in birds. Acta biotheoretica, 19(1), 16–36.
Bibliograph 315
Friedrich, W. L., 2004. Fire in the Sea: The Santorini Volcano: Natural History
and the Legend of Atlantis. Cambridge: Cambridge University Press.
Frisia, S., A. Borsato, C. Spotl, I. Villa, & F. Cucchi, 2005. Climate variability in
the SE Alps of Italy over the past 17,000 years reconstructed from a stalag-
mite record. Boreas, 34(4), 445–55.
Frisone, F., 2012. Rivers, land organization, and identity in Greek Western
Apoikíai. Mediterranean Historical Review, 27(1), 87–115.
Frost, H., 1995. Harbours and Proto-harbours: Early Levantine Engineering, eds.
V. Karageorghis & D. Michaelides. Nicosia: University of Cypurs/Cyprus
Ports Authority, pp. 1–22.
Frumkin, A., 2009. Stable isotopes of a subfossil Tamarix tree from the Dead
Sea region, Israel, and their implications for the Intermediate Bronze Age
environmental crisis. Quaternary Research, 71(3), 319–28.
Fuchs, M., A. Lang, & G. A. Wagner, 2004. The history of Holocene soil ero-
sion in the Phlious Basin, NE Peloponnese, Greece, based on optical dating.
The Holocene, 14(3), 334–45.
Fuller, B. T., N. Márquez-Grant, & M. P. Richards, 2010. Investigation of
diachronic dietary patterns on the islands of Ibiza and formentera, Spain:
Evidence from carbon and nitrogen stable isotope ratio analysis. American
Journal of Physical Anthropology, 143(4), 512–22.
Fuller, I. C., M. G. Macklin, D. G. Passmore, P. A. Brewer, J. Lewin, & A. G.
Wintle, 1996. Geochronologies and environmental records of Quaternary
fluvial sequences in the Guadalope basin, northeast Spain, based on lumi-
nescence dating. Geological Society, London, Special Publications, 115(1),
99–120.
Fuller, I. C., M. G. Macklin, J. Lewin, D. G. Passmore, & A. G. Wintle, 1998.
River response to high-frequency climate oscillations in southern Europe
over the past 200 k.y. Geology, 26(3), 275–8.
Galili, E., B. Rosen, A. Gopher, & H. Kolska, 2002. The emergence and
dispersion of the Eastern Mediterranean fishing village: Evidence from
submerged Neolithic settlements off the Carmel Coast, Israel. Journal of
Galili, E., D. Zviely, & M. Weinstein-Evron, 2005. Holocene sea-level changes
and landscape evolution on the northern Carmel coast (Israel). Méditerranée,
104(1.2), 79–86.
Galili, E., M. Weinstein-Evron, I. Hershkovitz, A. Gopher, M. E. Kislev,
O. Lernau, L. Kolska-Horwitz, & H. Lernau, 1993. Atlit-Yam: A prehis-
toric site on the sea floor off the Israeli coast. Journal of Field Archaeology,
20 (2), 133–57.
Galop, D., 1998. La Foret, L’Homme et le Troupeau dans les Pyrénées Toulouse.
Toulouse: GEODE – Université de Toulouse–Le Mirail.
Garcia, D., 2002. Dynamique territoriales en Gaule M‚ridionale durant l’âge du
Fer, in Territoires Celtiques: Espaces Ethniques et Territoires des Agglomerations
316 Bibliograph
Protohistoriques d’Europe Occidentale, eds. D. Garcia & F. Verdin. Paris:

Éditions Errances, pp. 88–102.
Garcia, D., F. Mocci, & K. Walsh, 2007. Archéologie de la vallée de l’Ubaye
(Alpes-de-Haute-Provence, France): Présentation des premiers résultats
d’un Projet collectif de Recherche. Preistoria Alpina, 42, 23–48.
Gasco, J., 1994. Development and decline in the Bronze Age of southern France:
Languedoc and Provence, in Development and Decline in the Mediterranean
Bronze Age, eds. C. Mathers & S. Stoddart. Sheffield: Continuum – Sheffield
Academic, pp. 99–128.
Gassends, J. M., & B. Maillet, 2004. Structures immergées dans l’anse Saint-
Gervais (Fos-sur-Mer, Bouches-du-Rhône), in Delta du Rhône Camargue
Antique, Médiévale et Moderne, eds. C. Landur, M. Pasqualini, & A. Guilcher.
Aix-en-Provence: Service Régional de l’Archéologie, pp. 151–63.
Gerasimidis, A., & N. Athanasiadis, 1995. Woodland history of northern Greece
from the mid Holocene to recent time based on evidence from peat pollen
profiles. Vegetation History and Archaeobotany, 4(2), 109–16.
Ghilardi, M., A. Genç, G. Syrides, J. Bloemendal, D. Psomiadis, T. Paraschou,
S. Kunesch, & E. Fouache, 2010. Reconstruction of the landscape his-
tory around the remnant arch of the Klidhi Roman Bridge, Thessaloniki
Plain, North Central Greece. Journal of Archaeological Science, 37(1),
178–91.
Gifford, J. A., 1995. The physical geology of the Western Mesara and Kommos,
in Kommos: An Excavation on the South Coast of Crete. Volume I: The
Kommos Region and Houses of the Minoan Town, eds. J. W. Shaw & M. C.
Shaw. Princeton: Princeton University Press, pp. 30–90.
Gifford, J. A., G. Rapp, & V. Vitali, 1992. Paleogeography of Carthage (Tunisia):
Coastal change during the first millennium BC. Journal of Archaeological
Science, 19, 575–96.
Giguet-Covex, C., F. Arnaud, D. Enters, J. Poulenard, L. Millet, P. Francus,
F. David, P.-J. Rey, B. Wilhelm, & J.-J. Delannoy, 2012. Frequency and
intensity of high-altitude floods over the last 3.5 ka in northwestern French
Alps (Lake Anterne). Quaternary Research, 77(1), 12–22.
Gilbertson, D., 1996. Explanations: Environment as agency, in Farming the
Desert: The UNESCO Libyan Valleys Archaeological Survey, ed. G. Barker.
London: Society for Libyan Studies, pp. 291–317.
Gilbertson, D., C. O. Hunt, & P. A. Smithson, 1996. Quaternary geomorphol-
ogy and palaeoecology, in Farming the Desert: The UNESCO Libyan Valleys
Archaeological Survey, ed. G. Barker. London: Society for Libyan Studies,
pp. 49–82.
Gili, S., V. Lull, R. Mico, C. Rihuete, & R. Risch, 2006. An island decides:
Megalithic burial rites on Menorca. Antiquity, 80(310), 829–42.
Gilman, A., 1976. Bronze Age dynamics in southeast Spain. Dialectical
Anthropology, 1, 307–19.
Bibliograph 317
Gilman, A., & J. Thornes, 1985. Land-Use and Prehistory in South-East Spain.
London: G. Allen and Unwin.
Giono, J., 1979. Batailles dans la Montagne. Paris: Gallimard.
Giraudi, C., 1998. Late Pleistocene and Holocene lake level variations in Fucino
Lake (Abruzzo – Central Italy) inferred from geological, archaeological and
historical data. Paläoklimaforschung, Palaeoclimate Research, 25, 1–17.
2004. Le oscillazioni di livello del lago di Mezzano (Valentano-VT): variazioni
climatiche e interventi antropici. Il Quaternario, 17, 221–30.
2005. Late-Holocene alluvial events in the Central Apennines, Italy. The
Holocene, 15(5), 768–73.
Giraudi, C., C. Tata, & L. Paroli, 2009. Late Holocene evolution of Tiber
river delta and geoarchaeology of Claudius and Trajan Harbor, Rome.
Geoarchaeology, 24(3), 371–82.
Giraudi, C., M. Magny, G. Zanchetta, & R. N. Drysdale, 2011. The Holocene
climatic evolution of Mediterranean Italy: A review of the continental geo-
logical data. The Holocene, 21(1), 105–15.
Gkiasta, M., 2010. Social identities, materiality and connectivity, in Material
Connections in the Ancient Mediterranean, eds. P. Van Dommelen &
B. Knapp. London: Routledge, pp. 85–105.
Goalen, M., & D. Fortenberry, 2002. The Villa del Discobolo at Castelporziano
on the Tyrrhenian coast of central Italy. Antiquity, 76(291), 29–30.
Goiran, J.-P., U. Collalelli, F. Salomon, H. Djerbi, & H. Tronchère, 2009.
Découverte d’un niveau marin biologique sur les quais de Portus: Le port
antique de Rome. Méditerranée. Revue Géographique des Pays Méditerranéens,
112, 59–67.
Goldberg, P., 1994. Interpreting late Quaternary continental sequences in
Israel, in Late Quaternary Chronology and Paleoclimates of the Eastern
Mediterranean, eds. O. Bar-Yosef & R. S. Kra. Arizona: University of
Arizona Press: Radiocarbon, pp. 89–102.
Goldberg, P., & O. Bar-Yosef, 1990. The effect of man on geomorphologi-
cal processes based upon evidence from the Levant and adjacent areas, in
Man’s Role in the Shaping of the Eastern Mediterranean Landscape, eds.
S. Bottema, G. Entjes Nieborg, & W. van Zeist. Rotterdam: AA Balkema,
pp. 71–85.
Goldin, O., 1997. Ecology of Critias and Platonic metaphysics, in The Greeks
and the Environment, eds. L. Westra & T. M. Robinson. Oxford: Rowman
& Littlefield, pp. 73–80.
Gómez Bellard, C., 1995. The first colonization of Ibiza and Formentera
(Balearic Islands, Spain): Some more islands out of the stream? World
Archaeology, 26(3), 442–55.
2008. Ibiza: The making of new landscapes, in Rural Landscapes of the
Punic World, eds. P. Van Dommelen & C. G. Bellard. Sheffield: Equinox,
pp. 44–75.
318 Bibliograph
González-Sampériz, P., P. Utrilla, C. Mazo, B. Valero-Garcés, M. C. Sopena,

M. Morellón, M. Sebastián, A. Moreno, & M. Martínez-Bea, 2009. Patterns
of human occupation during the Early Holocene in the Central Ebro Basin
(NE Spain) in response to the 8.2 ka climatic event. Quaternary Research,
71(2), 121–32.
Gopher, A., 1993. Sixth–fifth millennia BC settlements in the coastal plain in
Israel. Paléorient, 19(1), 55–63.
Gophna, R., 1979. Post-Neolithic settlement patterns, in The Quaternary of
Israel, ed. A. Horowitz. London: Academic Press, pp. 319–21.
Gourley, B., & H. Williams, 2005. The fortifications of ancient Stymphalos.
Mouseion, 49, 213–59.
Gorokhovich, Y., 2005. Abandonment of Minoan palaces on Crete in rela-
tion to the earthquake induced changes in groundwater supply. Journal of
Archaeological Science, 32(2), 217–22.
Goy, J. L., C. Zazo, & C. J. Dabrio, 2003. A beach-ridge progradation complex
reflecting periodical sea-level and climate variability during the Holocene
(Gulf of Almeria, Western Mediterranean). Geomorphology, 50(1–3),
251–68.
Grattan, J. P., & D. D. Gilbertson, 2000. Prehistoric ‘settlement crisis’, environ-
mental changes in the British Isles, and volcanic eruptions in Iceland: An
exploration of plausible linkages. Special Publication – Geological Society of
America, 345, 33–42.
Greaves, A. M., 2000. Miletos and the sea: A stormy relationship, in The Sea
in Antiquity, eds. G. J. Oliver, R. Brock, T. J. Cornell, & S. Hodkinson.
Oxford: Archaeopress, pp. 39–61.
Green, S., & M. Lemon, 1996. Perceptual landscapes in agrarian systems:
Degradation processes in north-western Epirus and the Argolid Valley,
Greece. Ecumene, 3(2), 181–99.
Greenfield, H. J., 2010. The secondary products revolution: The past, the pre-
sent and the future. World Archaeology, 42(1), 29–54.
Grima, R., 2008. Landscape, territories, and the life-histories of monuments in
temple period Malta. Journal of Mediterranean Archaeology, 21(1), 35–56.
Grmek, M., & D. Gourevitch, 1998. Les Maladies dans l’Art Antique. Paris:
Fayard.
Grove, A. T., 2001. The ‘Little Ice Age’ and its geomorphological consequences
in Mediterranean Europe. Climatic Change, 48(1), 121–36.
Grove, A. T., & O. Rackham, 2001. The Nature of Mediterranean Europe: An
Ecological History. Yale: Yale University Press.
Groves, C. P., 1990. Feral mammals of the Mediterranean islands: Documents
of early domestication, in The Walking Larder: Patterns of Domestication,
Pastoralism, and Predation, ed. J. Clutton-Brock. London: Unwin Hyman,
pp. 46–58.
Bibliograph 319
Guerrero-Ayuso, V. M. G., 2001. The Balearic Islands: Prehistoric coloniza-

tion of the furthest Mediterranean islands from the mainland. Journal of
Guglielmin, M., N. Cannone, & F. Dramis, 2001. Permafrost-glacial evolution
during the Holocene in the Italian central alps. Permafrost and Periglacial
Processes, 12(111), 124.
Guidoboni, E., & J. E. Ebel, 2009. Earthquakes and Tsunamis in the Past.
A Guide to Techniques in Historical Seismology. Cambridge: Cambridge
University Press.
Guilaine, J., 1994. La Mer Partagée: La Méditerranée avant l’Écriture, 7000–
2000 avant Jésus-Christ. Paris: Hachette.
Guilaine, J., 2003. De la Vague à la Tombe: La Conquéte Néolithique de la
Méditerranée. Paris: Seuil.
Guillaume, B., H. Pilar, K. Bouchaib, D. Gabriel, & S. Vincent, 2011. Genomic
profiling of plastid DNA variation in the Mediterranean olive tree. BMC
Plant Biology, 11.
Guillet, D., 1983. Toward a cultural ecology of mountains: The central Andes
and the Himalayas compared. Current Anthropology, 24(5), 561–74.
Gulchard, F., S. Carey, M. A. Arthur, H. Sigurdsson, & M. Arnold, 1996.
Tephra from the Minoan eruptions in sediments in the Black Sea. Nature,
363, 610–12.
Gunderson, L. H., & C. S. Holling, 2002. Panarchy: Understanding
Transformations in Human and Natural Systems. Washington: Island
Press.
Gvirtzman, G., & M. Wieder, 2001. Climate of the last 53,000 years in the
Eastern Mediterranean, based on soil-sequence stratigraphy in the coastal
plain of Israel. Quaternary Science Reviews, 20(18), 1827–49.
Hadjidaki, E., 1988. Preliminary report of excavations at the harbour of Phalsarna
in west Crete. American Journal of Archaeology, 92(4), 463–79.
Halstead, P., 2008. Between a rock and a hard place: Coping with marginal
colonisation in the Later Neolithic and Early Bronze Age of Crete and the
Aegean, in Escaping the Labyrinth: The Cretan Neolithic in Context, eds.
V. Isaakidou & P. Tomkins. Oxford: Oxbow Books, 229–57.
Halstead, P., & G. Jones, 1989. Agrarian ecology in the Greek islands: Time
stress, scale and risk. Journal of Hellenic Studies, 109, 41–55.
Hamilakis, Y., 1996. Wine, oil and the dialectics of power in Bronze Age Crete:
A review of the evidence. Oxford Journal of Archaeology, 15(1), 1–32.
Hamilton, S., R. Whitehouse, K. Brown, P. Combes, E. Herring, & M. S.
Thomas, 2006. Phenomenology in practice: Towards a methodology for a
subjective approach. European Journal of Archaeology, 9(1), 31–71.
Hancock, P. L., & E. Altunel, 1997. Faulted archaeological relics at Hierapolis
(Pamukkale), Turkey. Journal of Geodynamics, 24(1–4), 21–36.
320 Bibliograph
Harding, A., J. Palutikof, & T. Holt, 2009. The climate system, in The Physical
Geography of the Mediterranean, ed. J. C. Woodward. Oxford: Oxford
University Press, pp. 69–88.
Harfouche, R., 2007. Histoire des Paysages Méditerranéens Terrassés: aménage-
ments et Agriculture. Oxford: Archaeopress.
Harris, W. V., 2005a. The Mediterranean and ancient history, in Rethinking
the Mediterranean, ed. W. V. Harris. Oxford: Oxford University Press,
pp. 1–38.
2005b. Rethinking the Mediterranean. Oxford: Oxford University Press.
Harrison, R. J., 1994. The Bronze Age in northern and northeastern Spain, in
Development and Decline in the Mediterranean Bronze Age, eds. C. Mathers
& S. Stoddart. Sheffield: Continuum – Sheffield Academic, pp. 73–97.
Hay, P., 2006. A phenomenology of islands. Island Studies Journal, 1(1),
19–42.
Heiken, G., F. McCoy, & M. Sheridan, 1990. Palaeotopographic and palaeogeo-
logic reconstruction of Minoan Thera, in Thera and the Aegean World III.
Vol. 2: Earth Sciences, eds. D. A. Hardy, J. Keller, V. P. Galanopoulos, N. C.
Flemming, & T. H. Druitt. London: Thera Foundation, pp. 370–6.
Heiken, G., R. Funiciello, & D. De Rita, 2005. The Seven Hills of Rome.
Princeton: Princeton University Press.
Heinz, C., I. Figueiral, J. F. Terral, & F. Claustre, 2004. Holocene vegetation
changes in the northwestern Mediterranean: New palaeoecological data
from charcoal analysis and quantitative eco-anatomy. The Holocene, 14(4),
621–7.
Heinz, C., & S. Thiebault, 1998. Characterization and palaeoecological
significance of archaeological charcoal assemblages during late and post-
glacial phases in southern France. Quaternary Research, 50(1), 56–68.
Henry, D. O., 1994. Prehistoric cultural ecology in southern Jordan. Science,
265, 336–41.
Henry, F., 2009. Origine et dynamique long terme d’un écosystème herbacé‚
pseudo steppique, le cas de la plaine de La Crau (Bouches du Rhône, France).
Unpublished PhD thesis, Université Paul Cézanne Aix-Marseille III.
Henry, F., B. Talon, & T. Dutoit, 2010. The age and history of the French
Mediterranean steppe revisited by soil wood charcoal analysis. The Holocene,
20(1), 25–34.
Herva, V. P., 2006. Flower lovers, after all? Rethinking religion and human–
environment relations in Minoan Crete. World Archaeology, 38(4), 586–98.
Hieke, W., & F. Werner, 2000. The Augias megaturbidite in the central Ionian
Sea (Central Mediterranean) and its relation to the Holocene Santorini
event. Sedimentary Geology, 135(1–4), 205–18.
Higginbotham, J. A., 1997. Piscinae: Artificial Fishponds in Roman Italy. Chapel
Hill: University of North Carolina Press.
Higgins, C., 1966. Possible disappearance of Mycenean coastal settlements of
the Messenian Peninsula. American Journal of Archaeology, 70(1), 23–9.
Bibliograph 321
Higgs, E., 1978. Environmental changes in northern Greece, in Environmental

History of the Near and Middle East, ed. W. G. Brice. London: Academic
Press, pp. 41–9.
Hitchner, R. B., 1988. The Kasserine archaeological survey, 1982–1986.
Antiquités Africaines, 24, 7–41.
1990. The Kasserine archaeological survey, 1987. Antiquités Africaines, 26,
231–60.
Hodge, A. T., 1995. Roman Aqueducts and Water Supply. London: Gerald
Duckworth.
Hoekstra, A. Y., & M. M. Mekonnen, 2012. The water footprint of humanity.
Proceedings of the National Academy of Sciences, 109(9), 3232–7.
Hoffman, G., & H. D. Schulz, 1988. Coastline shifts and Holocene stratigraphy
on the Mediterranean coast of Andalucia (southeastern Spain), in Archaeology
of Coastal Changes. Proceedings of the First International Symposium ‘Cities
on the Sea. Past and Present’, ed. A. Raban. Oxford: B.A.R., pp. 53–70.
Höghammar, K., 2010. Long-term resilience: The reconstruction of the ancient
Greek polis of Kos after earthquakes in the period c. 200 BCE to c. 200
CE, in The Urban Mind: Cultural and Environmental Dynamics, eds.
P. Sinclair, G. Nordquist, F. Herschend, & C. Isendahl. Uppsala: African and
Comparative Archaeology, Department of Archaeology and Ancient History,
pp. 261–75.
Hohlfelder, R. L., 1976. The ports of Roman Baetica: A preliminary
reconnaissance. Journal of Field Archaeology, 3, 465–8.
Holtvoeth, J., H. Vogel, B. Wagner, & G. Wolff, 2010. Lipid biomarkers in
Holocene and glacial sediments from ancient Lake Ohrid (Macedonia,
Albania). Biogeosciences, 7(11), 3473–89.
Hooke, J. M., 2006. Human impacts on fluvial systems in the Mediterranean
region. Geomorphology, 79(3–4), 311–35.
Horden, P., & N. Purcell, 2000. The Corrupting Sea: A Study of Mediterranean
History. Oxford: Blackwell.
Howard, A. J., & M. G. Macklin, 1999. A guide for archaeologists investigating
Holocene landscapes. Antiquity, 73(281), 527–41.
Hughes, J. D., 1994a. Ecology in ancient Mesopotamia, in Green History: A
Reader in Environmental Literature, Philosophy and Politics, ed. D. Wall.
London: Routledge, pp. 33–5.
1994b. Pan’s Travail: Environmental Problems of the Ancient Greeks and
Romans. Baltimore: Johns Hopkins University Press.
2005. The Mediterranean: An Environmental History. Santa Barbara: ABC-
CLIO.
Hunt, C., 1995. The natural landscape and its evolution, in A Mediterranean
Valley: Landscape Archaeology and Annales History in the Biferno Valley, ed.
G. Barker. London: Leicester University Press, pp. 62–83.
Iakovidis, S., 2001. Gla and the Kopais in the 13th century BC. Library of the
Archaeological Society at Athens, 221.
322 Bibliograph
Isoardi, D., 2010a. Découverte récente à Arles d’amènagements de berge du

Rhône datés du Haut-Empire, in Seminaire d’Antiquités Nationales, Centre
Camille Jullian.
2010b. Rapport de Fouille de Sauvetage au Lieu-dit 5, Place Jean-
Baptiste Massillon (Arles) – 2009. Rapport de Fouille. Aix-en-Provence:
S.R.A. – P.A.C.A.
Issar, A. S., Y. Govrin, M. A. Geyh, E. Wakshal, & M. Wolf, 1992. Climate
changes during the Upper Holocene in Israel. Israel Journal of Earth-
Sciences, 40(1–4), 219–23.
Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J.
Bourque, R. H. Bradbury, R. Cooke, J. Erlandson, J. A. Estes, T. P. Hughes,
S. Kidwell, C. B. Lange, H. S. Lenihan, J. M. Pandolfi, C. H. Peterson,
R. S. Steneck, M. J. Tegner, & R. R. Warner, 2001. Historical overfishing
and the recent collapse of coastal ecosystems. Science, 293(5530), 629–37.
Jacob, J., J. R. Disnar, F. Arnaud, E. Gauthier, Y. Billaud, E. Chapron, &
G. Bardoux, 2009. Impacts of new agricultural practices on soil erosion dur-
ing the Bronze Age in the French Prealps. The Holocene, 19(2), 241–9.
Jahns, S., 1993. On the Holocene vegetation history of the Argive Plain
(Peloponnese, southern Greece). Vegetation History and Archaeobotany, 2,
187–203.
Jalut, G., A. Esteban Amat, L. Bonnet, T. Gauquelin, & M. Fontugne, 2000.
Holocene climatic changes in the Western Mediterranean, from south-east
France to south-east Spain. Palaeogeography Palaeoclimatology Palaeoecology,
160(3–4), 255–90.
James, P., C. Mee, & J. G. Taylor, 1994. Soil erosion and the archaeological
landscape of Methana, Greece. The Journal of Field Archaeology, 21(4),
395–416.
Jeftic, L., S. Keckes, & J. C. Pernetta, 1996. Climatic Change and the
Mediterranean: Environmental and Societal Impacts of Climate Change and
Sea Level Rise in the Mediterranean Region. London: Edward Arnold.
Jenny, H., 1994. Factors of Soil Formation: A System of Quantitative Pedology.
New York: Dover.
Jiménez-Brobeil, S. A., P. du Souich, & I. Al Oumaoui, 2009. Possible rela-
tionship of cranial traumatic injuries with violence in the south-east Iberian
Peninsula from the Neolithic to the Bronze Age. American Journal of
Physical Anthropology, 140(3), 465–75.
Joffre, R., S. Rambal, & J. Ratte, 1999. The dehesa system of southern Spain and
Portugal as a natural ecosystem mimic. Agroforestry Systems, 45(1), 57–79.
Johnson, M., 2007. Ideas of Landscape. Oxford: Blackwell.
Jolivet, L., J.-P. Brun, B. Meyer, G. Prouteau, & Collectif, 2008. Géodynamique
Méditerranéenne. Paris: Vuibert.
Jones, R. E., & S. C. Stiros, 2000. The advent of archaeoseismology in the
Mediterranean. Special Publication – Geological Society of London, 171,
25–32.
Bibliograph 323
Jorda, M., & F. Mocci, 1997. Sites protohistoriques et gallo-romains du massif

Sainte-Victoire dans leur contexte morphodynamique, in La Dynamique des
Paysages Protohistoriques, Antiques, Médiévaux et Modernes, eds. J. Burnouf,
J. P. Bravard, & G. Chouquer. Sophia Antipolis: APDCA, pp. 211–30.
Jorda, M., & M. Provansal, 1989. Le site de Glanum et l’histoire de
l’environnement. Les Dossiers de l’Archéologie, 140(juillet–août), 2–7.
Jorda, M., & M. Provansal, 1996. Impact de l’anthropisation et du climat sur
le detritisme en France du sud-est (Alpes du sud et Provence). Bulletin –
Société Géologique de France, 167(1), 159–68.
Jost, M., 1996. The distribution of sanctuaries in civic space in Arkadia, in
Placing the Gods: Sanctuaries and Sacred Space in Ancient Greece, eds.
S. Alcock & R. Osborne. Oxford: Clarendon Press, pp. 217–30.
Judson, S., 1963. Stream changes during historic time in east-central Sicily.
American Journal of Archaeology, 67(3), 287–9.
Judson, S., & A. Kahane, 1963. Underground drainageways in southern Etruria
and northern Latium. Papers of the British School at Rome, 31, 74–99.
Jusseret, S., 2010. Socializing geoarchaeology: Insights from Bourdieu’s the-
ory of practice applied to Neolithic and Bronze Age Crete. Geoarchaeology,
25(6), 675–708.
Kaniewski, D., V. De Laet, E. Paulissen, & M. Waelkens, 2007. Long-term effects
of human impact on mountainous ecosystems, western Taurus Mountains,
Turkey. Journal of Biogeography, 34(11), 1975–97.
Kaniewski, D., E. Paulissen, E. Van Campo, H. Weiss, T. Otto, J. Bretschneider,
& K. Van Lerberghe, 2010. Late second–early first millennium BC abrupt
climate changes in coastal Syria and their possible significance for the history
of the Eastern Mediterranean. Quaternary Research, 74(2), 207–15.
Katsouras, G., A. Gogou, I. Bouloubassi, K.-C. Emeis, M. Triantaphyllou,
G. Roussakis, & V. Lykousis, 2010. Organic carbon distribution and isoto-
pic composition in three records from the Eastern Mediterranean Sea dur-
ing the Holocene. Organic Geochemistry, 41(9), 935–9.
Keay, S. J., M. Millett, L. Paroli, & K. Strutt, 2006. Portus: An Archaeological
Survey of the Port of Imperial Rome. Rome: British School at Rome.
Keller, J., T. Rehren, & E. Stadlbauer, 1990. Explosive volcanism in the Hellenic
arc: A summary and review, in Thera and the Aegean World III. Vol. 2: Earth
Sciences, eds. D. A. Hardy, J. Keller, V. P. Galanopoulos, N. C. Flemming,
& T. H. Druitt. London: Thera Foundation.
Keys, D., 1999. Catastrophe: An Investigation into the Origins of the Modern
World. London: Arrow.
Kislev, M. E., A. Hartmann, & E. Galili, 2004. Archaeobotanical and archae-
oentomological evidence from a well at Atlit-Yam indicates colder, more
humid climate on the Israeli coast during the PPNC period. Journal of
Archaeological Science, 31(9), 1301–10.
Knapp, A. B., 2008. Prehistoric and Protohistoric Cyprus: Identity, Insularity, and
Connectivity. Oxford: Oxford University Press.
324 Bibliograph
2010. Cyprus’s earliest prehistory: Seafarers, foragers and settlers. Journal of

World Prehistory, 23(2), 79–120.
Knappett, C., T. Evans, & R. Rivers, 2011. The Theran eruption and Minoan
palatial collapse: New interpretations gained from modelling the maritime
network. Antiquity, 85(229), 1008–23.
Knauss, J., 1985. Die Melioration des Kopais-Beckens im Altertum und heute.
The reclamation of the Copais basin in the antiquity and today. Wasser und
Boden, 37(12), 611–14.
Kolstrup, E., 1994. Late Holocene pollen records from the Segermes valley, NE
Tunisia. Historical Biology: An International Journal of Paleobiology, 9(1),
131–6.
Kontogianni, V. A., N. Tsoulos, & S. C. Stiros, 2002. Coastal uplift, earthquakes
and active faulting of Rhodes Island (Aegean Arc): Modeling based on geo-
detic inversion. Marine Geology, 186(3–4), 299–317.
Kosmas, C., N. Danalatos, F. Lopez-Bermedez, & M. A. Romero Diz, 2002. The
effect of land use on soil erosion and land degradation under Mediterranean
conditions, in Mediterranean Desertification: A Mosaic of Processes and
Responses, eds. N. Geeson, C. J. Brandt, & J. B. Thornes. New York: John
Wiley, pp. 58–70.
Kotthoff, U., U. C. Muller, J. Pross, G. Schmiedl, I. T. Lawson, B. van de
Schootbrugge, & H. Schulz, 2008. Lateglacial and Holocene vegetation
dynamics in the Aegean region: An integrated view based on pollen data
from marine and terrestrial archives. The Holocene, 18(7), 1019–32.
Krahtopoulou, A., 2000. Holocene alluvial history of northern Pieria, Macedonia,
Greece, in Landscape and Land Use in Postglacial Greece, eds. P. Halstead &
C. Frederick. Sheffield: Sheffield Academic Press, pp. 15–27.
Kristiansen, K., & T. B. Larsson, 2005. The Rise of Bronze Age Society: Travels,
Transmissions and Transformations. Cambridge: Cambridge University Press.
Kuniholm, P. A., 1990. Archaeological evidence and non-evidence for climatic
change. Philosophical Transactions of the Royal Society of London. Series A,
Mathematical and Physical Sciences, 330(1615), 645–55.
1996. Anatolian tree rings and the absolute chronology of the Eastern
Mediterranean 2220–718 BC. Nature, 381(27), 780–3.
Kuzucuoglu, C., W. Dörfler, S. Kunesch, & F. Goupille, 2011. Mid- to Late-
Holocene climate change in central Turkey: The Tecer Lake record. The
Holocene, 21(1), 173–88.
Laborel, J., 1986. Vermetid gastropods as sea-level indicators, in Sea Level
Research, A Manual for Collection and Evaluation of Data, ed. O. Van de
Plassche. Norwich: Geo Books, pp. 28–310.
Laborel, J., & F. Laborel-Deguen, 1994. Biological indicators of relative sea-
level variations and of co-seismic displacements in the Mediterranean region.
Journal of Coastal Research, 10(2), 395–415.
Bibliograph 325
Lamb, H. F., F. Damblon, & R. W. Maxted, 1991. Human impact on the vege-
tation of the Middle Atlas, Morocco, during the last 5,000 years. Journal of
Biogeography, 18, 519–32.
Lamb, H. F., & S. Van der Kaars, 1995. Vegetational response to Holocene cli-
matic change: Pollen and palaeolimnological data from the Middle Atlas,
Morocco. The Holocene, 5(4), 400–8.
Lamb, H. F., U. Eicher, & V. R. Switsur, 1989. An 18,000-year record of veg-
etation, lake-level and climatic change from Tigalmamine, Middle Atlas,
Morocco. Journal of Biogeography, 16(1), 65–74.
Lambeck, K., F. Antonioli, A. Purcell, & S. Silenzi, 2004. Sea-level change along
the Italian coast for the past 10,000yr. Quaternary Science Reviews, 23(14–
15), 1567–98.
Lambeck, K., F. Antonioli, M. Anzidei, L. Ferranti, G. Leoni, G. Scicchitano,
& S. Silenzi, 2011. Sea level change along the Italian coast during the
Holocene and projections for the future. Quaternary International, 232(1–
2), 250–7.
LaMoreaux, P. E., 1995. Worldwide environmental impacts from the eruption of
Thera. Environmental Geology, 26(3), 172–81.
Lanfranchi, F. d., & M. C. Weiss, 1973. La Civilisation des Corses: Les Origines.
Ajaccio: Éditions Cyrnos.
Lanfranchi, F. d., & M. C. Weiss, 1997. L’Aventure Humaine Préhistorique en
Corse. Ajaccio: Éditions Albiana.
Laskaris, N., A. Sampson, F. Mavridis, & I. Liritzis, 2011. Late Pleistocene/
Early Holocene seafaring in the Aegean: New obsidian hydration dates with
the SIMS-SS method. Journal of Archaeological Science, 38(9), 2475–9.
Latour, B., 1997. Nous N’avons Jamais Été Modernes. Paris: La Découverte.
Laurens, H. 2010. Le Rêve Méditerranéen: Grandeurs et Avatars. Paris: CNRS.
Laval, H., J. Medus, & M. Roux, 1991. Palynological and sedimentological
records of Holocene human impact for the Etang de Berre, southeastern
France. The Holocene, 1(3), 269–72.
Le Gall, J., 1953. Le Tibre: Fleuve de Rome dans l’Antiquité. Paris: Presses
Universitaires de France.
Leighton, R., 1999. Sicily Before History: An Archaeological Survey from the
Palaeolithic to the Iron Age. London: Duckworth.
Leroy, S. A. G., S. Marco, R. Bookman, & C. S. Miller, 2010. Impact of earth-
quakes on agriculture during the Roman-Byzantine period from pollen
records of the Dead Sea laminated sediment. Quaternary Research, 73(2),
191–200.
Lespez, L., 2003. Geomorphic responses to long-term land use changes in east-
ern Macedonia (Greece). Catena, 51(3–4), 181–208.
Leveau, P., 1977. L’opposition de la montagne et de la plaine dans l’historiographie
de l’Afrique du Nord antique. Annales de Géographie, 86, 201–5.
326 Bibliograph
1986. Occupation du sol, géosystème et systèmes sociaux. Rome et ses ene-

mis des montagnes et du desert dans le Maghreb antique. Annales ESC, 6,
1345–58.
1995. Les moulins romains de Barbegal, les ponts-aqueducs du vallon des Arcs
et l’histoire naturelle de la vallé des Baux (Bilan de six ans de fouilles pro-
grammées). Paris: Comptes Rendus de l’Académie des Inscriptions et Belles
Lettres, pp. 115–44.
Leveau, P., 2008a. La ville romaine alpine dans son contexte environnemental:
Géoarchéologie et histoire du climat dans les Alpes, in La Ville des Alpes
Occidentales à l’Époque Romaine, eds. P. Leveau & B. Rémy. Grenoble: Les
Cahiers du CRHIPA, pp. 47–99.
Leveau, P., 2008b. Les inondations du Tibre à Rome: Politiques publiques et
variations climatiques à l’époque romaine, in Vers une Gestion Intégrée de
l’Eau dans l’Empire Romain. Actes du Colloque International Université
Laval Octobre 2006–290 p, ed. E. Hermon. Rome: L’Erma di Bretschneider,
pp. 137–46.
2009. Les conditions environnementales dans le nord de l’Afrique à l’époque
romaine. Contribution historiographique à l’histoire du climat et des rela-
tions homme/milieu, in Sociétés et Climats dans l’Empire Romain. Pour
une Perspective Historique et Systémique de la Gestion des Ressources en Eau
dans l’Empire Romain, ed. E. Hermon. Naples: Editoriale Scientifica,
pp. 308–48.
Leveau, P., & J. M. Palet-Martinez, 2010. Les Pyrénées romaines, la frontière,
la ville et la montagne. L’apport de l’archéologie du paysage. Pallas, 82,
171–98.
Leveau, P., K. Walsh, G. Bertucchi, H. Bruneton, & B. Tremmel, 2000. Le
Troisieme siècle dans la Valle des Baux: Les fouilles de la partie basse et
de lemissaire oriental des moulins de Barbegal. Revue d’Archéologie
Narbonnaise, 33, 387–439.
Leveau, P., P. Sillières, & J. P. Vallat, 1993. Campagnes de la Méditerranée
Romaine. Paris: Hachette.
Lewin, J., & J. C. Woodward, 2009. Karst geomorphology and environmental
change, in The Physical Geography of the Mediterranean, ed. J. C. Woodward.
Oxford: Oxford University Press, pp. 287–317.
Lichardus, J., M. Lichardus-Itten, G. Bailloud, & J. Cauvin, 1985. La Protohistoire
de l’Europe: Le Néolithique et le Chalcolithique. Paris: Presses Universitaires
de France.
Lillios, K. T., 1997. The third millennium BC in Iberia: Chronometric evidence
for settlement histories and socio-cultural change, in Third Millenium BC
Climate Change and Old World Collapse, eds. H. N. Dalfes, G. Kukla, &
H. Weiss. Berlin: Springer, pp. 173–91.
Lippi, M. M., C. Bellini, C. Trinci, M. Benvenuti, P. Pallecchi, & M. Sagri,
2007. Pollen analysis of the ship site of Pisa San Rossore, Tuscany, Italy: The
Bibliograph 327
implications for catastrophic hydrological events and climatic change during

the late Holocene. Vegetation History and Archaeobotany, 16(6), 453–65.
Llobera, M., & T. J. Sluckin, 2007. Zigzagging: Theoretical insights on climb-
ing strategies. Journal of Theoretical Biology, 249(2), 206–17.
Lolos, Y. A., 1997. The Hadrianic aqueduct of Corinth (with an Appendix on
the Roman aqueducts in Greece). Hesperia, 66(2) 271–314.
López-Bermudez, F., A. Romero-Diaz, & J. Martinez-Fernandez, 1998.
Vegetation and soil erosion under a semi-arid Mediterranean climate: A case
study from Murcia (Spain). Geomorphology, 24(1), 51–8.
López de Heredia, U., J. S. Carrión, P. Jiménez, C. Collada, & L. Gil, 2007.
Molecular and palaeoecological evidence for multiple glacial refugia for
evergreen oaks on the Iberian Peninsula. Journal of Biogeography, 34(9),
1505–17.
López-Merino, L., & J. A. López-Sáez, 2009. 2,000 years of pastoralism and
fire shaping high-altitude vegetation of Sierra de Gredos in central Spain.
Review of Palaeobotany and Palynology, 158(1–2), 42–51.
Lowe, J. J., C. Davite, D. Moreno, & R. Maggi, 1994. Holocene pollen
stratigraphy and human interference in the woodlands of the northern
Appennines, Italy. The Holocene, 4, 153–64.
Lucas, G., 2005. The Archaeology of Time. London: Routledge.
Lumaret, R., & N. Ouazzani, 2001. Plant genetics: Ancient wild olives in
Mediterranean forests. Nature, 413(6857), 700.
MacArthur, R. H., & E. O. Wilson, 1967. Island Biogeography. Princeton:
Princeton University Press.
Macklin, M. G., I. C. Fuller, J. Lewin, G. S. Maas, D. G. Passmore, J. Rose, J. C.
Woodward, S. Black, R. H. B. Hamlin, & J. S. Rowan, 2002. Correlation of
fluvial sequences in the Mediterranean basin over the last 200 ka and their
relationship to climate change. Quaternary Science Reviews, 21(14–15),
1633–41.
Macklin, M. G., & J. C. Woodward, 2009. River systems and environmental
change, in The Physical Geography of the Mediterranean, ed. J. C. Woodward.
Oxford: Oxford University Press, pp. 319–52.
Macklin, M. G., S. Tooth, P. A. Brewer, P. Noble, & G. A. Duller, 2010.
Holocene flooding and river development in a Mediterranean steepland
catchment: The Anapodaris Gorge, south central Crete, Greece. Global and
Planetary Change, 70(1), 35–52.
Maggi, R., 1999. Coast and Uplands in Liguria and Northern Tuscany from
the Mesolithic to the Bronze Age, eds. R. H. Tykot, J. Morter, & J. Robb.
London: Accordia Research Institute University of London, pp. 48–65.
Magny, M., B. Vanniere, G. Zanchetta, E. Fouache, G. Touchais, L. Petrika,
C. Coussot, A. V. Walter-Simonnet, & F. Arnaud, 2009. Possible complex-
ity of the climatic event around 4300–3800 Cal. BP in the Central and
Western Mediterranean. The Holocene, 19(6), 823–33.
328 Bibliograph
Magny, M., C. Miramont, & O. Sivan, 2002. Assessment of the impact of cli-
mate and anthropogenic factors on Holocene Mediterranean vegetation
in Europe on the basis of palaeohydrological records. Palaeogeography
Palaeoclimatology Palaeoecology, 186(1–2), 47–59.
Magny, M., D. Galop, P. Bellintani, M. Desmet, J. Didier, J. N. Haas,
N. Martinelli, A. Pedrotti, R. Scandolari, A. Stock, & B. Vanniere, 2009a.
Late-Holocene climatic variability south of the Alps as recorded by lake-level
fluctuations at Lake Ledro, Trentino, Italy. The Holocene, 19(4), 575–89.
Magny, M., J.-L. de Beaulieu, R. Drescher-Schneider, B. Vannière, A.-V. Walter-
Simonnet, Y. Miras, L. Millet, G. Bossuet, O. Peyron, & E. Brugiapaglia,
2007. Holocene climate changes in the central Mediterranean as recorded
by lake-level fluctuations at Lake Accesa (Tuscany, Italy). Quaternary Science
Reviews, 26(13), 1736–58.
Magny, M., & J. N. Haas, 2004. A major widespread climatic change around
5300 Cal. yr BP at the time of the Alpine Iceman. Journal of Quaternary
Science, 19(5), 423–30.
Magny, M., O. Peyron, L. Sadori, E. Ortu, G. Zanchetta, B. Vannière, &
W. Tinner, 2012. Contrasting patterns of precipitation seasonality during
the Holocene in the South- and North-Central Mediterranean. Journal of
Magri, D., 1995. Some questions on the Late-Holocene vegetation of Europe.
The Holocene, 5(3), 354–60.
Magri, D., & L. Sadori, 1999. Late Pleistocene and Holocene pollen stratigra-
phy at Lago di Vico, central Italy. Vegetation History and Archaeobotany,
8(4), 247–60.
Mahoney, T. A., 1997. Platonic ecology, deep ecology, in The Greeks and the
Environment, eds. L. Westra & T. M. Robinson. Oxford: Rowman &
Littlefield, pp. 45–54.
Malone, C., 2003. The Italian Neolithic: A synthesis of research. Journal of
World Prehistory, 17(3), 235–312.
Malone, C., S. Stoddart, & A. Bonanno, 2009. Mortuary customs in prehistoric
Malta: Excavations at the Brochtorff circle at Xagħra (1987–94). Cambridge:
McDonald Institute for Archaelogical Research.
Malone, C., S. Stoddart, G. Barker, G. Clark, M. Coltorti, L. Costantini,
J. Giorgi, J. Harding, C. Hunt, T. Reynold, & R. Skeates, 1992. The neo-
lithic site of San Marco, Gubbio (Perugia), Umbria: Survey and excavation
1985–87. Papers of the British School at Rome, LX, 1–69.
Mangini, A., C. Spötl, & P. Verdes, 2005. Reconstruction of temperature in the
Central Alps during the past 2000 yr from a δ18O stalagmite record. Earth
and Planetary Science Letters, 235(3–4), 741–51.
Manning, J. G., & I. Morris, 2007. Introduction, in The Ancient Economy:
Evidence and Models, eds. J. G. Manning & I. Morris. Stanford: Stanford
Bibliograph 329
Manning, S. W., 1999. A Test of Time: The Volcano of Thera and the Chronology
and History of the Aegean and East Mediterranean in the Mid Second
Millennium BC. Oxford: Oxbow.
2008. Protopalatial Crete and formation of the palaces, in The Aegean Bronze
Age, ed. C. W. Shelmerdine. Cambridge: Cambridge University Press,
pp. 105–20.
Manning, S. W., C. B. Ramsey, C. Doumas, T. Marketou, G. Cadogan, & C. L.
Pearson, 2002. New evidence for an early date for the Aegean Late Bronze
Age and Thera eruption. Antiquity, 76(293), 733–44.
Manning, S. W., C. McCartney, B. Kromer, & S. Stewart, 2010. The earlier
Neolithic in Cyprus: Recognition and dating of a Pre-Pottery Neolithic A
occupation. Antiquity, 84(325), 693–706.
Marchetti, M., 2002. Environmental changes in the central Po plain (north-
ern Italy) due to fluvial modifications and anthropogenic activities.
Geomorphology, 44(3–4), 361–73.
Mariotti Lippi, M., M. Guido, B. I. Menozzi, C. Bellini, & C. Montanari, 2007.
The Massaciuccoli Holocene pollen sequence and the vegetation history of
the coastal plains by the Mar Ligure (Tuscany and Liguria, Italy). Vegetation
History and Archaeobotany, 16(4), 267–77
Marriner, N., & C. Morhange, 2006. Geoarchaeological evidence for dredging
in Tyre’s ancient harbour, Levant. Quaternary Research, 65(1), 164–71.
Marriner, N., & C. Morhange, 2007. Geoscience of ancient Mediterranean har-
bours. Earth Science Reviews, 80(3–4), 137–94.
Marriner, N., C. Morhange, & M. Saghieh-Beydoun, 2008. Geoarchaeology
of Beirut’s ancient harbour, Phoenicia. Journal of Archaeological Science,
35(9), 2495–516.
Marriner, N., C. Morhange, & N. Carayon, 2008. Ancient Tyre and its harbours:
5,000 years of human–environment interactions. Journal of Archaeological
Science, 35(5), 1281–310.
Marriner, N., J. P. Goiran, & C. Morhange, 2008. Alexander the Great’s tom-
bolos at Tyre and Alexandria, eastern Mediterranean. Geomorphology, 100
(3–4), 377–400.
Marsh, B., 1999. Alluvial burial of Gordion, an Iron-Age City in Anatolia.
Journal of Field Archaeology, 26, part 2, 163–76.
Martinez-Fernandez, J., F. Lopez-Bermudez, & A. Romero-Diaz, 1995. Land
use and soil–vegetation relationships in a Mediterranean ecosystem: El
Ardal, Murcia, Spain. Catena, 25(1/4), 153.
Mastronuzzi, G., & P. Sansò, 2000. Boulders transport by catastrophic waves
along the Ionian coast of Apulia (southern Italy). Marine Geology, 170(1–
2), 93–103.
Mastronuzzi, G., & P. Sansò, 2004. Large boulder accumulations by extreme
waves along the Adriatic coast of southern Apulia (Italy). Quaternary
International, 120(1), 173–84.
330 Bibliograph
Mather, A. E., 2009. Tectonic setting and landscape development, in The Physical
Geography of the Mediterranean, ed. J. C. Woodward. Oxford: Oxford
Mathers, C., & S. Stoddart, 1994. Development and Decline in the Mediterranean
Bronze Age. Sheffield: Continuum – Sheffield Academic.
Mattingly, D., 1996. Explanations: People as agency, in Farming the Desert: The
UNESCO Libyan Valleys Archaeological Survey, ed. G. Barker. London:
Society for Libyan Studies, pp. 319–42.
Mauquoy, D., B. Van Geel, M. Blaauw, & J. van der Plicht, 2002. Evidence from
northwest European bogs shows ‘Little Ice Age’ climatic changes driven by
variations in solar activity. The Holocene, 12(1), 1–6.
Mazier, F., D. Galop, M. J. Gaillard, C. Rendu, C. Cugny, A. Legaz, O. Peyron,
& A. Buttler, 2009. Multidisciplinary approach to reconstructing local pas-
toral activities: An example from the Pyrenean Mountains (Pays Basque).
The Holocene, 19(2), 171–88.
McClure, S. B., M. A. Jochim, & C. M. Barton, 2006. Human behavioral
ecology, domestic animals, and land use during the transition to agricul-
ture in Valencia, eastern Spain, in Behavioral Ecology and the Transition to
Agriculture, eds. D. J. Kennett & B. Winterhalder. Berkeley: University of
California Press, 197–216.
McClure, S. B., O. Garcia, C. Roca de Togores, B. J. Culleton, & D. J. Kennett,
2011. Osteological and paleodietary investigation of burials from Cova de la
Pastora, Alicante, Spain. Journal of Archaeological Science, 38(2), 420–8.
McCoy, F., & G. Heiken, 2000a. The Late-Bronze Age explosive eruption of
Thera (Santorini), Greece: Regional and local effects, in Volcanic Hazards
and Disasters in Human Antiquity, eds. F. McCoy & G. Heiken. Boulder:
Geological Society of America, pp. 43–70.
McCoy, F., & G. Heiken, 2000b. Volcanic Hazards and Disasters in Human
Antiquity. Boulder: Geological Society of America.
McGovern, P. E., 2003. Ancient Wine: The Search for the Origins of Viniculture.
Oxford: Princeton University Press.
McIntosh, R. J., J. A. Tainter, & S. K. McIntosh, 2000. Climate, history, and
human action, in The Way the Wind Blows: Climate, History, and Human
Action. New York: Columbia University Press, pp. 1–42.
McNeill, J. R., 1992. The Mountains of the Mediterranean World. Cambridge:
Meiggs, R., 1982. Trees and Timber in the Ancient Mediterranean World. Oxford:
Oxford University Press
Mercuri, A. M., M. Mazzanti, P. Torri, L. Vigliotti, G. Bosi, A. Florenzano,
L. Olmi, & I. N’siala, 2012. A marine/terrestrial integration for mid–late
Holocene vegetation history and the development of the cultural landscape
in the Po valley as a result of human impact and climate change. Vegetation
History and Archaeobotany, 1–20.
Bibliograph 331
Mercuri, A. M., C. A. Accorsi, & M. Bandini Mazzanti, 2002. The long history
of cannabis and its cultivation by the Romans in central Italy, shown by pol-
len records from Lago Albano and Lago di Nemi. Vegetation History and
Archaeobotany, 11(4), 263–76.
Mercuri, A. M., C. A. Accorsi, M. B. Mazzanti, G. Bosi, A. Cardarelli,
D. Labate, M. Marchesini, & G. T. Grandi, 2006. Economy and environ-
ment of Bronze Age settlements – Terramaras – on the Po plain (Northern
Italy): First results from the archaeobotanical research at the Terramara di
Montale. Vegetation History and Archaeobotany, 16(1), 43–60.
Messina, F., G. Scorrano, C. M. Labarga, M. F. Rolfo, & O. Rickards, 2010.
Mitochondrial DNA variation in an isolated area of central Italy. Annals of
Human Biology, 37(3), 385–402.
Meyer, W. J., & C. L. Crumley, 2012. Historical ecology: Using what works to
cross the divide, in Atlantic Europe in the First Milennium BC: Crossing the
Divide, ed. T. Moore & X.-L. Armada. Oxford: Oxford University Press,
pp. 109–34.
Mikhailova, M. V., 2002. The hydrological regime and the peculiarities of for-
mation of the Po river delta. Water Resources, 29(4), 370–80.
Mikhailova, M. V., P. Bellotti, P. Valeri, & P. Tortora, 1998. The Tiber river delta
and the hydrological and morphological features of its formation. Water
Resources, 25(5), 572–82.
Minoura, K., F. Imamura, U. Kuran, T. Nakamura, G. A. Papadopoulos,
T. Takahashi, & A. C. Yalciner, 2000. Discovery of Minoan tsunami depos-
its. Geology, 28(1), 59–62.
Mitsch, W. J., & J. G. Gosselink, 1993. Wetlands. New York: Van Nostrand
Reinhold.
Mocci, F., K. Walsh, S. Richer, M. Court-Picon, B. Talon, S. Tzortzis, J. M.
Palet-Martinez, & C. Bressy, 2009. Archéologie et paléoenvironnement
dans les Alpes méridionales françaises. Hauts massifs de l’Argentièrois,
du Champsaur et de l’Ubaye (Hautes-Alpes et Alpes de Haute Provence)
(Néolithique final – début de l’Antiquité). Cahiers de Paléoenvironnement,
6, 235–54.
Moe, D., & F. G. Fedele, 1997. Pollen analytical study on the early use of a
present-day trail in the Central Alps. Preistoria Alpina, (33), 171–5.
Moe, D., F. G. Fedele, A. E. Maude, & M. Kvamme, 2007. Vegetational changes
and human presence in the low-alpine and subalpine zone in Val Febbraro,
upper Valle di Spluga (Italian central Alps), from the Neolithic to the Roman
period. Vegetation History and Archaeobotany, 16(6), 431–51.
Moffett, L., 1992. Cereals from a Bronze Age storage vessel at Torralba D’En
Salort, Menorca, Spain. Vegetation History and Archaeobotany, 1(2),
87–91.
Monaghan, J. J., P. J. Bicknell, & R. J. Humble, 1994. Volcanoes, tsunamis and
the demise of the Minoans. Physica D, 77, 217–28.
332 Bibliograph
Monna, F., D. Galop, L. Carozza, M. Tual, A. Beyrie, F. Marembert, C. Chateau,

J. Dominik, & F. E. Grousset, 2004. Environmental impact of early Basque
mining and smelting recorded in a high ash minerogenic peat deposit.
Science of the Total Environment, 327(1–3), 197–214.
Moody, J., 1997. The Cretan environment: Abused or just misunderstood?,
in Aegean Strategies: Studies of Culture and Environment on the European
Fringe, eds. P. N. Kardulias & M. T. Shutes. Lanham: Rowman & Littlefield,
pp. 61–77.
2000. Holocene climate change in Crete: An archaeologist’s view, in Landscape
and Landuse in Postglacial Greece, eds. P. Halstead & C. Frederick. Sheffield:
Moody, J., & L. Nixon, 2006. The Sphakia Survey: Internet Edition. Oxford:
University of Oxford.
Moore, A. M. T., & G. C. Hillman, 1992. The Pleistocene to Holocene transition
and human economy in southwest Asia: The impact of the younger Dryas.
American Antiquity, 57(3), 482–94.
Morales, J. A., J. Borrego, E. G. San Miguel, N. Lopez-Gonzalez, & B. Carro,
2008. Sedimentary record of recent tsunamis in the Huelva Estuary (south-
western Spain). Quaternary Science Reviews, 27(7–8), 734–46.
Morelli, C., A. Carbonara, V. Forte, R. Giudice, & P. Manacorda, 2008. The
landscape of the Ager Portuensis, Rome: Some new discoveries, 2000–
2002, in Archaeology and Landscape in Central Italy. Papers in Memory of
John Lloyd, eds. J. A. Lloyd, G. R. Lock, & A. Faustoferri. Oxford: School
of Archaeology, pp. 213–31.
Morelli, C., L. Paroli, & P. A. Verduchi, 2006. Summary of other recent field-
work at Portus, in Portus: An Archaeological Survey of the Port of Imperial
Rome, eds. S. J. Keay, M. Millett, L. Paroli, & K. Strutt. Rome: British
School at Rome, pp. 241–68.
Morey, M., & M. Ruiz-Perez, 2008. The Balearic Islands, in Mediterranean
Island Landscapes: Natural and Cultural Approaches, eds. I. N. Vogiatzakis,
G. Pungetti, & A. M. Mannion. Dordrecht: Springer, pp. 271–96.
Morfis, A., H. Zojer, T. Harum, J. Zötl, & International Working Group on
Tracer Methods in Hydrology, 1985. Karst hydrogeology of the Central
and Eastern Peloponnesus (Greece), in Steirische Beiträge zur Hydrogeologie.
Berlin: Kommission Springer-Verlag, pp. 1–301.
Morhange, C., C. Vella, M. Provansal, M. Hesnard, & J. Laborel, 1999.
Human impact and natural characteristics of the ancient ports of Marseille
and Fos, in Provence, southern France, in Environmental Reconstruction
in Mediterranean Landscape Archaeology, eds. P. Leveau, F. Tremant,
K. Walsh, & G. Barker. Oxford: Oxbow, pp. 145–53.
Morhange, C., F. Blanc, S. Schmitt-Mercury, M. Bourcier, P. Carbonel,
C. Oberlin, A. Prone, D. Vivent, & A. Hesnard, 2003. Stratigraphy of
Bibliograph 333
Late-Holocene deposits of the ancient harbour of Marseilles, southern

France. The Holocene, 13(4), 593–604.
Morhange, C., J. P. Goiran, M. Bourcier, P. Carbonel, J. Le Campion, J. M.
Rouchy, & M. Yon, 2000. Recent Holocene paleo-environmental evolu-
tion and coastline changes of Kition, Larnaca, Cyprus, Mediterranean Sea.
Marine Geology, 170(1–2), 205–30.
Morhange, C., N. Marriner, G. Bony, N. Carayon, C. Flaux, & M. Shah-Hosseini,
2013. Coastal geoarchaeology and neocatastrophism: A dangerous liaison?
BYZAS, DAI, Istanbul.
Morhange, C., P. A. Pirazzoli, N. Evelpidou, & N. Marriner, 2012. Late
Holocene tectonic uplift and the silting up of Lechaion, the western harbor
of ancient Corinth, Greece. Geoarchaeology, 27(3), 278–83.
Morris, M. W., 2002. Soil Science and Archaeology: Three Test Cases from Minoan
Crete. Philadelphia: Institute for Aegean Prehistory Academic Press.
Morton, J., 2001. The Role of the Physical Environment in Ancient Greek
Seafaring. Leiden: Brill.
Naveh, Z., 2004. The evolution of the cultural Mediterranean landscape in Israel
as affected by fire, grazing and human activities, in Evolutionary Theory and
Processes: Modern Horizons: Papers in Honour of Eviatar Nevo, eds. E. Nevo
& S. P. Vasser. Dordrecht: Kluwer Academic, pp. 337–409.
Neboit, R., 1984. Érosion des sols et colonisation grecque en Sicile et en Grande-
Grèce. Bulletin de l’Association des Géographes Français, 499, 5–21.
Neboit-Guilhot, R., & L. Lespez, 2006. Alluvionnement et creusement sur
la rive nord de la Méditerranée: Vers une lecture systémique des rythmes
historiques de la morphogenèse, in L’Érosion entre Soci té, Climat et
Paléoenvironnement, eds. P. Alle & L. Lespez. Clermont-Ferrand: Presses
Universitaires Blaise-Pascal, pp. 335–52.
Newell, W. L., B. Stone, & R. Harrison, 2004. Holocene alluvium around
Lefkosia (Nicosia), Cyprus: An archive of land-use, tectonic processes, and
climate change, in Geo-Environment, eds. J. F. Martin-Duque, C. A. Brebbia,
A. E. Godfrey, & J. R. Diaz de Teran. Southampton: WIT, pp. 71–80.
Nir, Y., 1997. Middle and Late Holocene sea levels along the Israel Mediterranean
coast – Evidence from ancient water wells. Journal of Quaternary Science,
12(2), 143–52.
Nixon, I. G., 1985. The volcanic eruption of Thera and its effect on the Mycenean
and Minoan civilisations. Journal of Archaeological Science, 12, 9–24.
Nixon, L., J. Moody, V. Niniou-Kindelis, S. Price & O. Rackham, 1990.
Archaeological survey in Sphakia, Crete. Classical Views, XXXIV(9),
213–20.
Nixon, L., & S. Price, 2005. Ancient agricultural terraces: Evidence from
texts and archaeological survey. American Journal of Archaeology, 109(4),
665–94.
334 Bibliograph
Nowicki, K., 1999. Economy of refugees: Life in the Cretan mountains at the turn
of the Bronze and Iron Ages, in From Minoan Farmers to Roman Traders:
Sidelights on the Economy of Ancient Crete, ed. E. Chaniotis. Stuttgart: Franz
Steiner Verlag, pp. 145–71.
Ntinou, M., 2002. Vegetation and human communities in prehistoric Greece, in El
paisaje en el Neolitic mediterráneo: Neolithic Landscapes of the Mediterranean,
eds. E. Badal, J. Bernabeu, & B. Martí. Valencia: Universitat de Valencia,
Papeles del Laboratoria de Arqueolgia de Valencia, Saguntum, pp. 91–103.
Nur, A., & D. Burgess, 2008. Apocalypse: Earthquakes, Archaeology, and the
Wrath of God. Princeton: Princeton University Press.
Nur, A., & E. H. Cline, 2000. Poseidon’s horses: Plate tectonics and earthquake
storms in the Late Bronze Age Aegean and Eastern Mediterranean. Journal
Obando, J. A., 2002. The impact of land abandonment on regeneration of
semi-natural vegetation: A case study from the Guadalentín, in Mediterranean
Desertification: A Mosaic of Processes and Responses, eds. N. Geeson, C. J.
Brandt, & J. B. Thornes, London: John Wiley, pp. 58–70.
Oleson, J. P., C. Brandon, S. M. Cramer, R. Cucitore, E. Gotti, & R. L.
Hohlfelder, 2004. The ROMACONS project: A contribution to the histori-
cal and engineering analysis of hydraulic concrete in Roman maritime struc-
tures. International Journal of Nautical Archaeology, 33(2), 199–229.
Ollivier, V., J. L. Guendon, A. Ali, P. Roiron, & P. Ambert, 2006. Postglacial
evolution of travertine environments in the French Alps and Provence: New
chronology, facies, and morphosedimentary dynamics. Quaternaire, 17(2),
51–68.
Olszewski, D. I., 1993. Subsistence ecology in the Mediterranean forest:
Implications for the origins of cultivation in the Epipaleolithic Southern
Levant. American Anthropologist, 95(2), 420–35.
Orengo, H., 2010. Impacto humano en un medio altimontano pre-pirenaico
durante los últimos 1500 años. Análisis paleoambiental de una turbera alcal-
ina. Unpublished PhD thesis, Institut catalan d’arqueologia classica.
Orland, I. J., M. Bar-Matthews, N. T. Kita, A. Ayalon, A. Matthews, & J. W.
Valley, 2009. Climate deterioration in the Eastern Mediterranean as revealed
by ion microprobe analysis of a speleothem that grew from 2.2 to 0.9 ka in
Soreq Cave, Israel. Quaternary Research, 71(1), 27–35.
Ortu, E., S. Brewer, & O. Peyron, 2006. Pollen-inferred palaeoclimate recon-
structions in mountain areas: Problems and perspectives. Journal of
Osborne, R., & B. Cunliffe, 2005. Mediterranean Urbanization 800–600 BC.
Oxford: Oxford University Press/British Academy.
Packard, R. M., 2007. The Making of a Tropical Disease: A Short History of
Malaria. Baltimore: Johns Hopkins University Press.
Bibliograph 335
Paget, R., 1968. The ancient ports of Cumae. The Journal of Roman Studies,
58, 152–69.
Pagnol, M., 1988. Jean de Florette. New York: Andre Deutsch, North Point.
Palet, J. M., A. Ejarque, Y. Miras, S. Riera, I. Euba, & H. Orengo, 2007. Formes
d’ocupació d’alta muntanya a la vall de la Vansa (Serra del Cadí – Alt Urgell)
i a la vall del Madriu-Perafita-Claror (Andorra): estudi diacrònic de paisat-
ges culturals pirinencs. Barcelona: Generalitat de Catalunya, pp. 229–53.
Palmer, C., S. Colledge, A. Bevan, & J. Conolly, 2010. Vegetation recolonisation
of abandoned agricultural terraces on Antikythera, Greece. Environmental
Pantaleon-Cano, J., E. I. Yll, R. Perez-Obiol, & J. M. Roure, 2003. Palynological
evidence for vegetational history in semi-arid areas of the Western
Mediterranean (Almeria, Spain). The Holocene, 13(1), 109–20.
Papanastassiou, D., K. Gaki-Papanastassiou, & H. Maroukian, 2005. Recognition
of past earthquakes along the Sparta fault (Peloponnesus, southern Greece)
during the Holocene, by combining results of different dating techniques.
Journal of Geodynamics, 40(2–3), 189–99.
Pareschi, M. T., M. Favalli, & E. Boschi, 2006. Impact of the Minoan tsunami
of Santorini: Simulated scenarios in the Eastern Mediterranean. Geophysical
Research Letters, 33(18), L18607-L.
Pasquinucci, M., & G. Rossetti, 1988. The harbour infrastructure at Pisa and
Porto Pisano from ancient times until the Middle Ages, in Archeology of
Coastal Changes. Proceedings of the First International Symposium ‘Cities
on the Sea. Past and present’, ed. A. Raban. Oxford: B.A.R., pp. 137–55.
Pasquinucci, M., & S. Menchelli, 1999. The landscape and economy of the ter-
ritories of Pisae and Volaterrae (coastal North Etruria). Journal of Roman
Patterson, H., F. di Gennaro, H. di Giuseppe, S. Fontana, V. Gaffney, A. Harrison,
S. J. Keay, M. Millett, M. Rendeli, & P. Roberts, 2000. The Tiber Valley
project: The Tiber and Rome through two millennia. Antiquity, 74(284),
395–403.
Patterson, H., H. di Giuseppe, & R. Witcher, 2004. Three south Etrurian ‘cri-
ses’: First results of the Tiber Valley project. Papers of the British School at
Rome, 1–36.
Patton, M., 1996. Islands in Time. London: Routledge.
Pedley, M. A. R. T., 2009. Tufas and travertines of the Mediterranean region:
A testing ground for freshwater carbonate concepts and developments.
Sedimentology, 56(1), 221–46.
Pèlachs, A., J. Nadal, J. Soriano, D. Molina, & R. Cunill, 2009. Changes in
Pyrenean woodlands as a result of the intensity of human exploitation:
2,000 years of metallurgy in Vallferrera, northeast Iberian Peninsula.
336 Bibliograph
Peltenburg, E. J., 2003. Conclusions: Mylouthkia 1 and the early colonists

of Cyprus, in The Colonisation and Settlement of Cyprus: Investigations
at Kissonerga-Mylouthkia, 1976–1996, ed. E. J. Peltenburg. Uppsala:
P. Åström’s Vörlag, pp. 83–103.
Peltenburg, E. J., S. Colledge, P. Croft, A. Jackson, C. McCartney, & M. A.
Murray, 2000. Agro-pastoralist colonization of Cyprus in the 10th millen-
nium BP: Initial assessments. Antiquity, 74(286), 844–53.
Pereira, P. M., & M. P. da Fonseca, 2003. Nature vs. nurture: The making of the
montado ecosystem. Conservation Ecology, 7(3).
Pèrez, R. M., 2005. Towards a definition of politico-ideological practices in the
prehistory of Minorca (the Balearic Islands): The wooden carvings from
Mussol Cave. Journal of Social Archaeology, 5(2), 276–99.
Pèrez-Obiol, R., G. Jalut, R. Julià, A. Pèlachs, M. J. Iriarte, T. Otto, &
B. Hernández-Beloqui, 2011. Mid-Holocene vegetation and climatic his-
tory of the Iberian Peninsula. The Holocene, 21(1), 75–93.
Pèrez-Obiol, R., & L. Sadori, 2007. Similarities and dissimilarities, synchronisms
and diachronisms in the Holocene vegetation history of the Balearic Islands
and Sicily. Vegetation History and Archaeobotany, 16(4), 259–65.
Pèrez-Obiol, R., & R. Julià, 1994. Climatic Change on the Iberian Peninsula
recorded in a 30,000-yr pollen record from Lake Banyoles. Quaternary
Research, 41(1), 91–8.
Perlès, C., 1999. The distribution of Magoules in eastern Thessaly, in Neolithic
Society in Greece, ed. P. Halstead. Sheffield: Sheffield Academic Press,
pp. 42–56.
2001. The Early Neolithic in Greece: The First Farming Communities in Europe.
Cambridge: Cambridge University Press.
Pescatore, T., M. R. Senatore, G. Capretto, & G. Lerro, 2001. Holocene coastal
environments near Pompeii before the AD 79 eruption of Mount Vesuvius,
Italy. Quaternary Research, 55(1), 77–85.
Pessina, A. & V. Tin, 2008. Archeologia del Neolitico: l’Italia tra VI e IV secolo
a. C. Rome: Carocci.
Peyron, O., S. Goring, I. Dormoy, U. Kotthoff, J. Ç. Pross, J. L. De Beaulieu,
R. Drescher-Schneider, B. Vannière, & M. Magny, 2011. Holocene season-
ality changes in the Central Mediterranean region reconstructed from the
pollen sequences of Lake Accesa (Italy) and Tenaghi Philippon (Greece).
The Holocene, 21(1), 131–46.
Phillips, J. D., 2007. The perfect landscape. Geomorphology, 84(3–4), 159–69.
Phoca-Cosmetatou, N., 2011a. The First Mediterranean Islanders: Initial
Occupation and Survival Strategies. Oxford: David Brown Book Company.
2011b. Island occupation of the Cycladic islands in the Neolithic: Strategies
for survival, in The First Mediterranean Islanders: Initial Occupation and
Survival Strategies, ed. N. Phoca-Cosmetatou. Oxford: David Brown Book
Company, pp. 77–97.
Bibliograph 337
Piccarreta, M., M. Caldara, D. Capolongo, & F. Boenzi, 2011. Holocene geo-

morphic activity related to climatic change and human impact in Basilicata,
southern Italy. Geomorphology, 128(3–4), 137–47.
Pini, R., 2004. Late Neolithic vegetation history at the pile-dwelling site of
Palu di Livenza (northeastern Italy). Journal of Quaternary Science, 19(8),
769–82.
Pirazzoli, P. A., 1976. Sea level variations in the Northwest Mediterranean dur-
ing Roman times. Science, 194(4264), 519–21.
Pirazzoli, P. A., J. Ausseil-Badie, P. Giresse, E. Hadjidaki, & M. Arnold, 1992.
Historical environmental changes at Phalasarna Harbor, west Crete.
Geoarchaeology, 7(4), 371–92.
Piva, A., A. Asioli, F. Trincardi, R. R. Schneider, & L. Vigliotti, 2008.
Late-Holocene climate variability in the Adriatic Sea (Central Mediterranean).
The Holocene, 18(1), 153–68.
Planchais, N., 1982. Palynologie lagunaire de l’étang de Maugio.
Paléoenvironnement végétal et évolution anthropique. Pollen et Spores,
24(1), 93–118.
Planchais, N., D. Duzer, & M. Fontugne, 1991. Palynologie de dépôts holocènes
de Lattes (Hérault). Comptes Rendus de l’Academie des Sciences Paris,
313(2), 1357–60.
Pliny the Elder, 2009. The Natural History of Pliny, trans. J. Bostock & H. T.
Riley. Charleston: BiblioBazaar.
Polio, M. V., 2005. The Ten Books on Architecture. Boston: Adamant Media
Corporation.
Pons, A., & P. Quézel, 1998. A propos de la mise en place du climat méditer-
ranéen. Comptes. Rendus de l’Academie des Sciences Paris – Series IIA – Earth
and Planetary Science, 327(11), 755–60.
Powell, J., 1996. Fishing in the Prehistoric Aegean. Jonsered: Åström Editions.
Primavera, M., O. Simone, G. Fiorentino, & M. Caldara, 2011. The palae-
oenvironmental study of the Alimini Piccolo lake enables a reconstruc-
tion of Holocene sea-level changes in southeast Italy. The Holocene, 21(4),
553–63.
Pross, J., U. Kotthoff, U. C. Müller, O. Peyron, I. Dormoy, G. Schmiedl, S.
Kalaitzidis, & A. M. Smith, 2009. Massive perturbation in terrestrial eco-
systems of the Eastern Mediterranean region associated with the 8.2 kyr BP
climatic event. Geology, 37(10), 887–90.
Provansal, M., 1995a. Holocene sedimentary sequences in the Arc River delta and
the Etang de Berre in Provence, southern France, in Mediterranean Quaternary
River Environments: Refereed Proceedings of an International Conference
University of Cambridge/United Kingdom/28–29 September, eds. J. Lewin,
M. G. Macklin, & J. C. Woodward. Rotterdam: AA Balkema, pp. 159–65.
1995b. The role of climate in landscape morphogenesis since the Bronze Age
in Provence, southeastern France. The Holocene, 5(3), 348–53.
338 Bibliograph
2006. Seminaire – Hommages … Philippe Leveau, Centre Camille Jullian, Aix-

en-Provence, France.
Pucci, S., D. Pantosti, P. M. De Martini, A. Smedile, M. Munzi, E. Cirelli,
M. Pentiricci, & L. Musso, 2011. Environment–human relationships in his-
torical times: The balance between urban development and natural forces at
Leptis Magna (Libya). Quaternary International, 242(1), 171–84.
Purcell, N., 1995. Eating fish: The paradoxes of seafood, in Food in Antiquity,
ed. J. Wilkins, D. Harvey, & M. Dobson. Exeter: University of Exeter Press,
pp. 132–49.
1998. Alla scoperta di una costa residenziale romana: Il litus Laurentinum e
l’archeologia dell’ otium. Castelporziano, III, 11–32.
2005. Statics and dynamics: Ancient Mediterranean urbanism, in Mediterranean
Urbanization 800–600 BC, eds. B. Cunliffe & R. Osborne. Oxford: Oxford
University Press/British Academy, pp. 249–72.
Pustovoytov, K., 2006. Soils and soil sediments at Gobekli Tepe, southeastern
Turkey: A preliminary report. Geoarchaeology, 21(7), 699–719.
Quézel, P., 2004. Large-scale post-glacial distribution of vegetation structures
in the Mediterranean region, in Recent Dynamics of the Mediterranean
Vegetation and Landscape, eds. S. Mazzoleni, G. di Pasquale, M. Mulligan,
P. di Martino, & F. Rego. Chichester: John Wiley, pp. 3–12.
Raban, A., 1988. Coastal processes and ancient harbour engineering, in
Archaeology of Coastal Changes. Proceedings of the First International
Symposium ‘Cities on the Sea. Past and Present’, ed. A. Raban. Oxford:
B.A.R., pp. 185–208.
1992. Sebastos: The royal harbour at Caesarea Maritima – A short-lived giant.
International Journal of Nautical Archaeology, 21(2), 111–24.
Rackham, O., 2001. Trees, Wood and Timber in Greek History. Oxford: Leopard’s
Head Press.
2008. Holocene history of Mediterranean island landscapes, in Mediterranean
Island Landscapes: Natural and Cultural Approaches, eds. I. N. Vogiatzakis,
Rackham, O., & J. Moody, 1996. The Making of the Cretan Landscape.
Manchester: Manchester University Press.
Raible, C., M. Yoshimori, T. Stocker, & C. Casty, 2007. Extreme midlatitude
cyclones and their implications for precipitation and wind speed extremes
in simulations of the Maunder Minimum versus present day conditions.
Climate Dynamics, 28(4), 409–23.
Rainbird, P., 2002. A message for our future? The Rapa Nui (Easter Island)
ecodisaster and Pacific island environments. World Archaeology, 33(3),
436–51.
Ramis, D., 2010. From colonisation to habitation: Early cultural adapta-
tions in the Balaeric Bronze Age, in Material Connections in the Ancient
Bibliograph 339
Mediterranean: Mobility, Materiality and Identity, ed. P. v. Dommelen.

Abingdon: Taylor & Francis.
Ramis, D., J. A. Alcover, J. Coll, & M. Trias, 2002. The chronology of the first
settlement of the Balearic Islands. Journal of Mediterranean Archaeology,
15(1), 3–24.
Ramis, D., & P. Bover, 2001. A review of the evidence for domestication of
Myotragus balearicus Bate 1909 (Artiodactyla, Caprinae) in the Balearic
Islands. Journal of Archaeological Science, 28(3), 265–82.
Reale, O., & J. Shukla, 2000. Modeling the effects of vegetation on Mediterranean
climate during the Roman classical period: Part II. Model simulation. Global
and Planetary Change, 25(3–4), 185–214.
Reale, O., & P. Dirmeyer, 2000. Modeling the effects of vegetation on
Mediterranean climate during the Roman classical period. Part I: Climate
history and model sensitivity. Global and Planetary Change, 25(3–4),
163–84.
Redman, C. L., 1999. Human Impact on Ancient Environments. Tucson:
University of Arizona Press.
2005. Resilience theory in archaeology. American Anthropologist, 107(1),
70–7.
Redman, C. L., & A. P. Kinzig, 2003. Resilience of past landscapes: Resilience
theory, society, and the longue durée. Conservation Ecology, 7(1), 14.
Reille, M., & A. Pons, 1992. The ecological significance of sclerophyllous oak
forest in the western part of the Mediterranean basin: A note on pollen ana-
lytical data. Vegetatio, 99–100, 13–17.
Rendell, H. M., A. J. Claridge, & M. L. Clarke, 2007. Late Holocene
Mediterranean coastal change along the Tiber delta and Roman occupation
of the Laurentine Shore, central Italy. Quaternary Geochronology, 2(1–4),
83–8.
Rendu, C., 2003. La Montagne d’Enveig. Canet: Éditions Trabucaire.
Renfrew, A. C., 1976. Megaliths, territories and populations, in Acculturation
and Continuity in Atlantic Europe, ed. S. DeLaet. Ghen: De Tempel,
pp. 198–220.
Renfrew, C., & M. Wagstaff, 2005. An Island Polity: The Archaeology of
Exploitation in Melos. Cambridge: Cambridge University Press.
Retallack, G. J., 2008. Rocks, views, soils and plants at the temples of ancient
Greece. Antiquity, 82(317), 640–57.
Richer, S., 2009. From pollen to people: The interaction between people and their
environment in the mid- to high- altitudes of the Southern French Alps.
Unpublished PhD thesis, University of York.
Riehl, S., & E. Marinova, 2008. Mid-Holocene vegetation change in the Troad
(W Anatolia): Man-made or natural? Vegetation History and Archaeobotany,
17(3), 297–312.
340 Bibliograph
Riera-Mora, S., J. A. Léspez, & J. B. Argilagés, 2004. Premières traces

d’anthropisation à l’est de la péninsule ibérique et les Îles baléars, in
Néolithisation Précoce. Premières Traces d’Anthropisation du Couvert
Végétal à Partir des Données Polliniques, ed. H. Richard. Besançon: Presses
Universitaires Franche-Comté, pp. 195–219.
Robb, J., 2001. Island identities: Ritual, travel and the creation of difference in
Neolithic Malta. European Journal of Archaeology, 4(2), 175–202.
2007. The Early Mediterranean Village: Agency, Material Culture and Social
Change in Neolithic Italy. Cambridge: Cambridge University Press.
Robb, J., & D. Van Hove, 2003. Gardening, foraging and herding: Neolithic land
use and social territories in Southern Italy. Antiquity, 77(296), 241–54.
Roberts, N., 1990. Human-induced landscape change in south and south-
west Turkey during the later Holocene, in Man’s Role in the Shaping of the
Eastern Mediterranean Landscape, eds. S. Bottema, G. EntjesNieborg, &
W. van Zeist. Rotterdam: AA Balkema, pp. 53–67.
2002. Did prehistoric landscape management retard the post-glacial spread of
woodland in Southwest Asia? Antiquity, 76(294), 1002–10.
Roberts, N., & A. Rosen, 2009. Diversity and complexity in early farming com-
munities of Southwest Asia: New insights into the economic and envi-
ronmental basis of Neolithic Çatalhöyük. Current Anthropology, 50(3),
393–402.
Roberts, N., D. Brayshaw, C. Kuzucuoğlu, R. Perez, & L. Sadori, 2011. The
mid-Holocene climatic transition in the Mediterranean: Causes and conse-
quences. The Holocene, 21(1), 3–13.
Roberts, N., P. Boyer, & R. Parish, 1996. Preliminary results of geoarchaeologi-
cal investigations at Çatalhöyük, in On the Surface: Çatalhöyük 1993–95, ed.
I. Hodder. Cambridge: McDonald Institute for Archaeological Research,
pp. 19–40.
Roberts, N., W. J. Eastwood, C. Kuzucuoğlu, G. Fiorentino, & V. Caracuta,
2011. Climatic, vegetation and cultural change in the Eastern Mediterranean
during the mid-Holocene environmental transition. The Holocene, 21(1),
147–62.
Robinson, S. A., S. Black, B. W. Sellwood, & P. J. Valdes, 2006. A review of palae-
oclimates and palaeoenvironments in the Levant and Eastern Mediterranean
from 25,000 to 5000 years BP: Setting the environmental background for
the evolution of human civilisation. Quaternary Science Reviews, 25(13–
14), 1517–41.
Rockman, M., 2003. Knowledge and learning in the archaeology of colonization,
in Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation,
eds. M. Rockman & J. Steele. London: Routledge, pp. 3–24.
Rodriguez-Ariza, M. O., & S. Ruiz, 1996. Paleobotany of a Bronze Age commu-
nity, Castellon Alto (Galera, Granada, Spain). Revue d’Archéométrie, 191–6.
Bibliograph 341
Rodriguez-Vidal, J., L. M. Caceres, J. C. Finlayson, F. J. Gracia, & A. Martinez-

Aguirre, 2004. Neotectonics and shoreline history of the Rock of Gibraltar,
southern Iberia. Quaternary Science Reviews, 23(18–19), 2017–29.
Rohling, E. J., R. H. Abu-Zied, J. S. L. Casford, A. Hayes, & B. A. A. Hoogakker,
2009. The marine environment: Present and past, in The Physical Geography
of the Mediterranean, ed. J. C. Woodward. Oxford: Oxford University Press,
pp. 33–67.
Roiron, P., A. A. Ali, J. L. Guendon, M. E. Migueres, S. D. Muller, & V. Ollivier,
2006. New data on the Holocene vegetation in the southern Alps: The fos-
sil flora of the Serre de Montdenier travertine (Alpes de Haute Provence,
France). Quaternaire, 17(2), 69–78.
Rosen, A. M., 1995. The social response to environmental change in Early
Bronze Age Canaan. Journal of Anthropological Archaeology, 14(1), 26.
2005. Phytolith indicators of plant and land use at Çatalhöyük, in Inhabiting
Çatalhöyük: Reports from the 1995–99 Seasons (Çatalhöyük Research Project
S.), ed. I. Hodder. Cambridge: MacDonald Institute for Archaeological
Research, pp. 203–12.
2007. Civilizing Climate: Social Responses to Climate Change in the Ancient
Near East. Plymouth: AltaMira.
Rosser, Z. H., T. Zerjal, M. E. Hurles, M. Adojaan, D. Alavantic, A. Amorim,
W. Amos, M. Armenteros, E. Arroyo, G. Barbujani, G. Beckman, L.
Beckman, J. Bertranpetit, E. Bosch, D. G. Bradley, G. Brede, G. Cooper,
H. B. S. M. Côrte-Real, P. de Knijff, R. Decorte, Y. E. Dubrova, O.
Evgrafov, A. Gilissen, S. Glisic, M. Gölge, E. W. Hill, A. Jeziorowska,
L. Kalaydjieva, M. Kayser, T. Kivisild, S. A. Kravchenko, A. Krumina,
V. Kucinskas, J. Lavinha, L. A. Livshits, P. Malaspina, S. Maria, K.
McElreavey, T. A. Meitinger, A. V. Mikelsaar, R. J. Mitchell, K. Nafa, J.
Nicholson, S. Nørby, A. Pandya, J. Ç. Parik, P. C. Patsalis, L. Pereira, B.
Peterlin, G. Pielberg, M. J. Prata, C. Previderé, L. Roewer, S. Rootsi, D. C.
Rubinsztein, J. Saillard, F. R. Santos, G. Stefanescu, B. C. Sykes, A. Tolun,
R. Villems, C. Tyler-Smith, & M. A. Jobling, 2000. Y-chromosomal diver-
sity in Europe is clinal and influenced primarily by geography, rather than by
language. The American Journal of Human Genetics, 67(6), 1526–43.
Rossignol-Strick, M., 1999. The Holocene climatic optimum and pollen records
of sapropel in the Eastern Mediterranean, 9000–6000 BP. Quaternary
Science Reviews, 18(4/5), 515–30.
Ruiz, M., R. Risch, M. Gonzalez, P. Castro, V. Lull & R. W. Chapman, 1992.
Environmental exploitation and social structure in prehistoric southeast
Spain. Journal of Mediterranean Archaeology, 5(1), 3–38.
Ruiz del Árbol, M., F. J. Sánchez-Palencia, J. A. López-Sáez, P. López,
R. Macías, & O. López, 2003. A geoarchaeological approach to the study
of Roman terraces: Landscape transformations in a mining area in the
342 Bibliograph
north-western Iberian Peninsula, in The Mediterranean World Environment

and History, ed. E. Fouache. Paris: Elsevier, pp. 331–9.
Russo Ermolli, E., & G. di Pasquale, 2002. Vegetation dynamics of south-west-
ern Italy in the last 28 kyr inferred from pollen analysis of a Tyrrhenian Sea
core. Vegetation History and Archaeobotany, 11(3), 211–20.
Ryan, W., & W. C. Pitman, 2000. Noah’s Flood: The New Scientific Discoveries
about the Event that Changed History. London: Touchstone.
Ryan, W. B. F., W. C. Pitman, C. O. Major, K. Shimkus, V. Moskalenko, G. A.
Jones, P. Dimitrov, N. Goruer, M. Sakinc, & H. Yuece, 1997. An abrupt
drowning of the Black Sea shelf. Marine Geology, 138(1/2), 119–26.
Sabbatani, S., R. Manfredi, & S. Fiorino, 2010. Malaria infection and the
anthropological evolution. Sao Paulo Medical Journal, 19(1), 64–83.
Sadori, L., & B. Narcisi, 2001. The postglacial record of environmental history
from Lago di Pergusa, Sicily. The Holocene, 11(6), 655–72.
Saez, J. A. L., P. L. Garcia, & M. M. Sanchez, 2002. Palaeoecology and Holocene
environmental change from a saline lake in south-west Spain: Protohistorical
and prehistorical vegetation in Cadiz Bay. Quaternary International, 93/94,
197–206.
Sagona, C., 2004. Land use in prehistoric Malta. A re-examination of the Maltese
‘cart ruts’. Oxford Journal of Archaeology, 23(1), 45–60.
Sainz-Elipe, S., J. M. Latorre, R. Escosa, M. Masià, M. V. Fuentes, S. Mas-
Coma, & M. D. Bargues, 2010. Malaria resurgence risk in southern
Europe: Climate assessment in an historically endemic area of rice fields at
the Mediterranean shore of Spain. Malaria Journal, 9.
Sallares, R., 2002. Malaria and Rome: A History of Malaria in Ancient Italy.
Oxford: Oxford University Press.
Salowey, C. A., 1994. Herakles and the waterworks: Mycenean dams, classi-
cal fountains, Roman aqueducts, in Archaeology in the Peloponnese: New
Excavations and Research, ed. K. A. Sheedy. Oxford: Oxbow, pp. 77–94.
Sampson, A., 1996. Causes of earthquakes at Mycenean and pre-Mycenean
Thebes, in Archaeoseismology, eds. S. C. Stiros & R. E. Jones. Athens:
Institute of Geology and Mineral Exploration & The British School at
Athens, pp. 113–17.
Sangiorgi, F., L. Capotondi, N. C. Nebout, L. Vigliotti, H. Brinkhuis, S. Giunta,
A. F. Lotter, C. Morigi, A. Negri, & G. J. Reichart, 2003. Holocene seasonal
sea-surface temperature variations in the southern Adriatic Sea inferred from
a multiproxy approach. Journal of Quaternary Science, 18(8), 723–32.
Schiffer, M. B., 1985. Is there a ‘Pompeii Premise’ in Archaeology? Journal of
Anthropological Research, 41(1), 18–41.
Schlumbaum, A., M. Tensen, & V. Jaenicke-Després, 2008. Ancient plant DNA
in archaeobotany. Vegetation History and Archaeobotany, 17(2), 233–44.
Schmidt, R., C. Kamenik, & M. Roth, 2007. Siliceous algae-based seasonal
temperature inference and indicator pollen tracking ca. 4,000 years of
Bibliograph 343
climate/land use dependency in the southern Austrian Alps. Journal of

Paleoliminology, 38(4), 541–54.
Schonfeld, J., & R. Zahn, 2000. Late glacial to Holocene history of the
Mediterranean outflow. Evidence from benthic foraminiferal assemblages and
stable isotopes at the Portuguese margin. Palaeogeography Palaeoclimatology
Palaeoecology, 159(1–2), 85–111.
Schuele, W., 1993. Mammals, vegetation and the initial human settlement of the
Mediterranean islands: A palaeoecological approach. Journal of Biogeography,
20(4), 399.
Schulte, L., 2002. Climatic and human influence on river systems and glacier
fluctuations in southeast Spain since the Last Glacial Maximum. Quaternary
International, 93/94, 85–100.
Schwab, M. J., F. Neumann, T. Litt, J. F. Negendank, & M. Stein, 2004.
Holocene palaeoecology of the Golan Heights (Near East): Investigation
of lacustrine sediments from Birkat Ram crater lake. Quaternary Science
Reviews, 23(16–17), 1723–31.
Seager, R. B., 2000. Excavations on the Island of Pseira. Crete: Borgo Press.
Segard, M., 2009. Les Alpes Occidentales à l’Époque Romaine: Développement
Urbain et Exploitation des Ressources des Régions de Montagne. (Italie, Gaule
Narbonnaise, Provinces Alpines). Aix-en-Provence: Éditions Errance: Centre
Camille Jullian.
Shackleton, J. C., & T. H. van Andel, 1986. Prehistoric shore environments, shell-
fish availability, and shellfish gathering at Franchthi, Greece. Geoarchaeology,
1(2), 127–43.
Shaw, B. D., 2006a. At the Edge of the Corrupting Sea. Oxford: University of
Oxford.
Shaw, J. W., 2006b. Kommos: A Minoan Harbor Town and Greek Sanctuary in
Southern Crete. Princeton: American School of Classical Studies at Athens.
Shaw, J. W., & M. C. Shaw, 1993. Excavations at Kommos (Crete) during 1986–
1992. Hesperia, 62, 129–90.
Sheffer, N. A., M. Rico, Y. Enzel, G. Benito, & T. Grodek, 2008. The palaeo-
flood record of the Gardon River, France: A comparison with the extreme
2002 flood event. Geomorphology, 98(1–2), 71–83.
Sherratt, A., 1981. Plough and pastoralism: Aspects of the secondary products
revolution. Pattern of the Past Studies in Honour of David Clarke, eds.
I. Hodder, G. L. Isaac, & N. Hammond. Cambridge: Cambridge University
Press, pp. 261–305.
Shillito, L. M., 2011. Simultaneous thin section and phytolith observations of
finely stratified deposits from Neolithic Çatalhöyük, Turkey: Implications
for paleoeconomy and Early Holocene paleoenvironment. Journal of
Sigurdsson, H., S. Carey, M. Alexandri, G. Vougioukalakis, K. Croff, C. Roman,
D. Sakellariou, C. Anagnostou, G. Rousakis, C. Ioakim, A. Gogou,
344 Bibliograph
D. Ballas, T. Misaridis, & P. Nomikou, 2006. Marine investigations of

Greece’s Santorini volcanic field. EOS, Transactions, American Geophysical
Union, 87(34), 337–42.
Silva, P. G., M. Sintubin, & K. Reicherter, 2011. New advances in studies of
earthquake archaeology and palaeoseismology. Quaternary International,
242(1), 1–3.
Simeoni, U., & C. Corbau, 2009. A review of the Delta Po evolution (Italy)
related to climatic changes and human impacts. Geomorphology, 107(1–2),
64–71.
Simmons, A. H., 1991. Humans, island colonisation and Pleistocene extinctions
in the Mediterranean: The view from Akrotiri Aetokremnos, Cyprus.
Antiquity, 65, 857–69.
1998. Of tiny hippos, large cows and early colonists in Cyprus. Journal of
Sivan, D., S. Wdowinski, K. Lambeck, E. Galili, & A. Raban, 2001. Holocene sea-
level changes along the Mediterranean coast of Israel, based on archaeolog-
ical observations and numerical model. Palaeogeography Palaeoclimatology
Palaeoecology, 167(1–2), 101–17.
Skeates, R., 1999. Unveiling Inequality: Social Life and Social Change in the
Mesolithic and Early Neolithic of East Central Italy, eds. R. H. Tykot,
J. Morter, & J. Robb. London: Accordia Research Institute University of
London, pp. 15–45.
Soles, J. S., 1992. The Prepalatial Cemeteries at Mochlos and Gournia and the
House Tombs of Bronze Age Crete. Princeton: American School of Classical
Studies at Athens.
Soloviev, S. L., 2000. Tsunamis in the Mediterranean Sea 2000 BC–2000 AD.
Dordrecht: Kluwer Academic.
Sondaar, P. Y., P. L. DeBoer, M. Sanges, T. Kotsakis, & D. Esu, 1984. First
report on a Palaeolithic culture in Sardinia, in Early Settlement in the West
Mediterranean Islands and the Peripheral Areas, eds. W. H. Waldren, R. W.
Chapman, J. Lewthwaite, & R. C. Kennard. Oxford: B.A.R., pp. 29–59.
Soren, D., 2003. Can archaeologists excavate evidence of malaria? World
Archaeology, 35(2), 193–209.
Souyoudzoglu-Haywood, C., 1999. The Ionian Islands in the Bronze Age and
Early Iron Age. Liverpool: Liverpool University Press.
Spanò Giammellaro, A., F. Spatafora, & P. Van Dommelen, 2008. Sicily and
Malta: Between sea and countryside, in Rural Landscapes of the Punic
World, eds. P. Van Dommelen & C. G. Bellard. Sheffield: Equinox,
pp. 129–58.
Spondilis, I., 1996. Contribution to a study of the configuration of the coast of
Pylia, based on the location of new archaeological sites, in Archaeoseismology,
Bibliograph 345
Stanley, D. J., & F. C. Wezel, 1985. Geological Evolution of the Mediterranean

Basin: Raimondo Selli Commemorative Volume. New York: Springer.
Stefaniuk, L., C. Morhange, P. F. Blanc, S. Francou, & J. P. Goiran, 2005.
Évolution des paysages littoraux dans la dépression sud-ouest de Cumes
depuis 4,000 ans. La question du port antique. Méditerranée, 104(1.2),
49–59.
Sternberg, M., 2004. Le rôle des fleuves dans la pêche du Ier au VIe siècle:
État des connaissances, in Fleuves et Marais, Une Histoire au Croisement
de la Nature et la Culture, eds. J. Burnouf & P. Leveau. Paris: CTHS,
pp. 185–99.
Stevenson, A. C., 2000. The Holocene forest history of the Montes Universales,
Teruel, Spain. The Holocene, 10(5), 603–10.
Stevenson, A. C., & R. J. Harrison, 1992. Ancient forests of Spain. A model for
land-use and dry forest management in south west Spain from 4000 AD to
1900 AD. Proceedings of the Prehistoric Society, 58, 227–47.
Steward, J. H., 1955. The concept and method of cultural ecology, in Theory
of Culture Change: The Methodology of Multilinear Evolution, ed. J. H.
Steward. Urbana: University of Illinois Press.
2005. The concept and method of cultural ecology, in Anthropology in Theory:
Issues in Epistemology, eds. H. L. Moore & T. Sanders. Oxford: Wiley
Blackwell, pp. 100–6.
Stewart, I., & C. Morhange, 2009. Coastal geomorphology and sea-level change,
in The Physical Geography of the Mediterranean, ed. J. C. Woodward. Oxford:
Oxford University Press, pp. 385–413.
Stiros, S. C., J. Laborel, F. Laborel-Deguen, S. Papageorgiou, J. Evin, & P. A.
Pirazzoli, 2000. Seismic coastal uplift in a region of subsidence: Holocene
raised shorelines of Samos Island, Aegean Sea, Greece. Marine Geology,
170(1–2), 41–58.
Stoddart, S., 1999. Long-term dynamics of an island community: Malta 5500
BC–2000 AD, in Social Dynamics of the Prehistoric Central Mediterranean,
eds. R. H. Tykot, J. Morter, & J. Robb. London: Accordia Research
Institute, University of London, pp. 137–47.
Stoddart, S., A. Bonanno, T. Gouder, C. Malone, & D. Trump, 1993. Cult
in an island society: Prehistoric Malta in the Tarxien period. Cambridge
Archaeological Journal, 3(1), 3–19.
Strasser, T. F., E. Panagopoulou, C. N. Runnels, P. M. Murray, N. Thompson,
P. Karkanas, F. W. McCoy, & K. W. Wegmann, 2010. Stone Age seafaring in
the Mediterranean: Evidence from the Plakias region for Lower Palaeolithic
and Mesolithic habitation of Crete. Hesperia: The Journal of the American
School of Classical Studies at Athens, 79(2), 145–90.
Strasser, T. F., C. Runnels, K. Wegmann, E. Panagopoulou, F. McCoy,
C. Digregorio, P. Karkanas, & N. Thompson, 2011. Dating Palaeolithic sites in
southwestern Crete, Greece. Journal of Quaternary Science, 26(5), 553–60.
346 Bibliograph
Strauss, S., & B. Orlove, 2003. Up in the air: The anthroplogy of weather and
climate, in Weather, Climate, Culture, ed. S. Strauss & B. Orlove. Oxford:
Berg, pp. 3–14.
Sutton, M. O., & E. N. Anderson, 2004. Introduction to Cultural Ecology.
Oxford: Berg.
Terral, J. F., & X. Mengual, 1999. Reconstruction of Holocene climate in south-
ern France and eastern Spain using quantitative anatomy of olive wood and
archaeological charcoal. Palaeogeography Palaeoclimatology Palaeoecology,
153(1–4), 71–92.
Terrell, J. E., 1999. Comment on Paul Rainbird, ‘Islands Out of Time: Towards
a Critique of Island Archaeology’. Journal of Mediterranean Archaeology,
12(2), 240–5.
Terrenato, N., 1998. Tam Firmum Municipium: The romanization of Volaterrae
and its cultural implications. The Journal of Roman Studies, 88, 94–114.
Terrenato, N., & A. J. Ammerman, 1996. Visibility and site recovery in the
Cecina Valley survey, Italy. Journal of Field Archaeology, 23(1), 91–110.
Thiebault, S., 1997. Early-Holocene vegetation and the human impact in central
Provence (Var, France): Charcoal analysis of the Baume de Fontbrégoua.
The Holocene, 7(3), 343.
Thomas, M. G., I. Barnes, M. E. Weale, A. L. Jones, P. Forster, N. Bradman, &
P. P. Pramstaller, 2007. New genetic evidence supports isolation and drift in
the Ladin communities of the South Tyrolean Alps but not an ancient origin
in the Middle East. European Journal of Human Genetics, 16(1), 124–34.
Thomas, R. B., 1979. Effects of change on high mountain human-adaptive pat-
terns, in High-Altitude Geoecology, ed. P. J. Webber. Boulder: Westview
Press, pp. 139–88.
Thorndycraft, V., & G. Benito, 2006. The Holocene fluvial chronology of Spain:
Evidence from a newly compiled radiocarbon database. Quaternary Science
Reviews, 25(3), 223–34.
Thornes, J. B., 1987. The palaeoecology of erosion, in Landscape and Culture,
ed. J. M. Wagstaff. Oxford: Blackwell, pp. 37–55.
2009. Land degradation, in The Physical Geography of the Mediterranean, ed.
J. C. Woodward. Oxford: Oxford University Press, pp. 563–81.
Tilley, C., 1994. A Phenomenology of Landscape: Places, Paths and Monuments.
Oxford: Berg.
2004. The Materiality of Stone: Explorations in Landscape Phenomenology.
Oxford: Berg.
Tomaselli, R., 1981. Main physiognomic types and geographic distribution of
shrub systems related to Mediterranean climates, in Mediterranean-Type
Shrublands, eds. F. Di Castri, D. W. Goodall, & R. L. Specht. Amsterdam:
Elsevier, pp. 95–106.
Touflan, P., B. Talon, & K. Walsh, 2010. Soil charcoal analysis: A reliable tool for
spatially precise studies of past forest dynamics: A case study in the French
southern Alps. The Holocene, 20(1), 45–52.
Bibliograph 347
Tozzi, P., 1993. Il libro del Po: Storie di acque, di terre, di uomini. New York:
New Press.
Traveset, A., M. Nogales, J. Alcover, J. Delgado, M. López-Darias, D. Godoy,
J. Igual, & P. Bover, 2009. A review on the effects of alien rodents in the
Balearic (Western Mediterranean Sea) and Canary Islands (Eastern Atlantic
Ocean). Biological Invasions, 11(7), 1653–70.
Trément, F., 1993. Le secteur des etangs de Saint-Blaise: Pour une approche
archéologique et paléoecologique d’un milieu de vie, in Archéologie et
Environnement de la Sainte-Victoire aux Alpilles, eds. P. Leveau & M. Provansal.
Aix-en-Provence: Publications Université de Provence, pp. 83–108.
Trifonov, V. G., & A. S. Karakhanian, 2004. Active faulting and human environ-
ment. Tectonophysics, 380(3–4), 287–94.
Trousset, P., 1986. De la montagne au désert. Limes et maitrise de l’eau. Revue
de l’Occident Musulman et de la Méditerranée, 41(1), 90–115.
Turner, R., N. Roberts, W. J. Eastwood, E. Jenkins, & A. Rosen, 2010. Fire,
climate and the origins of agriculture: Micro-charcoal records of biomass
burning during the last glacial–interglacial transition in Southwest Asia.
Journal of Quaternary Science, 25(3), 371–86.
Tykot, R. H., 1997. Characterization of the Monte Arci (Sardinia) obsidian
sources. Journal of Archaeological Science, 24(5), 467–80.
2004. Neolithic exploitation and trade of obsidian in the Central Mediterranean:
New results and implications for cultural interaction. BAR International
Series, 1303, 25–36.
Tzedakis, P. C., 2009. Museums and cradles of Mediterranean biodiversity.
Journal of Biogeography, 36(6), 1033–4.
Tzortzis, S., & F. Mocci, K. Walsh, B. Talon, M. Court-Picon, V. Dumas, V. Py,
& S. Richer, 2008. Les massifs de l’Argentièrois du Mésolithique au début
de l’Antiquité: Au croisement des données archéologiques et paléoenviron-
nementales en haute montagne (Hautes-Alpes, parc national des Ecrins), in
Le Peuplement de l’Arc Alpin, eds. D. Garcia & H. Richard. Paris: CTHS,
pp. 123–48.
Unkel, I., C. Heymann, O. Nelle, & E. Zagana, 2011. Climatic influence on
Lake Stymphalia during the last 15,000 years, in Advances in the Research
of Aquatic Environment, eds. N. Lambrakis, G. Stournaras, & K. Katsanou.
Berlin: Springer, pp. 75–82.
Urbainczyk, T., 2008. Slave Revolts in Antiquity. Berkeley: University of
California Press.
Vallianou, D., 1996. New evidence of earthquake destruction in Late Minoan
Crete, in Archaeoseismology, eds. S. C. Stiros & R. E. Jones. Athens: Institute
of Geology and Mineral Exploration & The British School at Athens,
pp. 153–67.
Valsecchi, V., W. Tinner, W. Finsinger, & B. Ammann, 2006. Human impact
during the Bronze Age on the vegetation at Lago Lucone (northern Italy).
348 Bibliograph
van Andel, T. H., 1987. The Landscape, in Landscape and People of the Franchthi
Region, eds. T. H. van Andel & S. B. Sutton. Bloomington: Indiana
van Andel, T. H., & C. Runnels, 1987. Beyond the Acropolis: A Rural Greek Past.
Stanford: Stanford University Press.
van Andel, T. H., C. Runnels, & K. O. Pope, 1986. Five thousand years of land
use and abuse in the Southern Argolid, Greece. Hesperia, 55, 103–28.
van Andel, T. H., & E. Zangger, 1990. Land use and soil erosion in prehistoric
and historical Greece. Journal of Field Archaeology, 17, 379–96.
van Andel, T. H., & J. C. Shackleton, 1982. Late Paleolithic and Mesolithic
coastlines of Greece and the Aegean. Journal of Field Archaeology, 9(4),
445–54.
van Andel, T. H., K. Gallis, & G. Toufexis, 1992. Early Neolithic Farming in a
Thessalian River Landscape, Greece, eds. J. Lewin, M. G. Macklin, & J. C.
Woodward. The Netherlands: AA Balkema, pp. 131–44.
van Andel, T. H., K. Gallis, & G. Toufexis, 1995. Early Neolithic farming
in a Thessalian river landscape, Greece, in Mediterranean Quaternary
River Environments: Refereed Proceedings of an International Conference
University of Cambridge/United Kingdom/28–29 September, eds. J. Lewin,
M. G. Macklin, & J. C. Woodward. Rotterdam: AA Balkema, pp. 131–43.
van der Leeuw, S., 2003. Une étude des causes de la dégradation des terres et
la désertification, in Archéologie et Systèmes Socio-environnementaux: Études
Multiscalaires sur la Vallée du Rhône dans le Programme ARCHEOMEDES,
eds. S. van der Leeuw, F. Favory, & J. L. Fiches. Paris: CNRS, pp. 11–19.
van der Leeuw, S., F. Favory, & J. L. Fiches, eds., 2003. Archéologie et Systèmes
Socio-environnementaux: Études Multiscalaires sur la Vallée du Rhône dans
le Programme ARCHEOMEDES. Paris: CNRS.
Van Der Veen, M., 1996. Romano-Libyan agriculture: Crops and animals, in
Farming the Desert: The UNESCO Libyan Valleys Archaeological Survey, ed.
G. Barker. London: Society for Libyan Studies, pp. 227–63.
Van Dommelen, P., 1999. Islands in history. Journal of Mediterranean
Van Geel, B., & B. E. Berglund, 1997. A causal link between a climatic deteriora-
tion around 850 Cal BC and a subsequent rise in human population density
in NW-Europe? Alfred-Wegener-Stiftung, pp. 126–30.
Van Strydonck, M., 2004. Radiocarbon and archaeological evidence for a pos-
sible climate-induced cultural change on the Balearic Islands around 2700
BP, in Radiocarbon and Archaeology, eds. T. Higham, C. Bronk Ramsey, &
C. Owen. Oxford: Oxford University School of Archaeology, pp. 247–62.
Van Strydonck, M., M. Boudin, E. Ervynck, J. Orvay, & H. Borms, 2005. Spatial
and temporal variation of dietary habits during the prehistory of the Balearic
Islands as reflected by 14C, d15N and d13C analyses on human and animal
bones. Mayurqa, 30(2), 525–41.
Bibliograph 349
van Zeist, W., H. Woldring, & D. Stapert, 1975. Late Quaternary vegetation and
climate of southwestern Turkey. Palaeohistoria, 17, 53–143.
Vannière, B., M. J. Power, N. Roberts, W. Tinner, J. Carrión, M. Magny,
P. Bartlein, D. Colombaroli, A. L. Daniau, W. Finsinger, G. Gil-Romera,
P. Kaltenrieder, R. Pini, L. Sadori, R. Turner, V. Valsecchi, & E. Vescovi,
2011. Circum-Mediterranean fire activity and climate changes during
the mid-Holocene environmental transition (8500–2500 Cal. BP). The
Holocene, 21(1), 53–73.
Vaudour, J., 1994. Évolution Holocène des travertins de vallée dans le Midi
Méditerranéan français. Géographie Physique Quaternaire, 48, 315–26.
Veggiani, A., 1994. I deterioramenti climatici dell’età del ferro e dell’alto medio-
evo. Bollettino della Società Torricelliana di Scienze e Lettere, Faenza, 45,
3–80.
Vermoere, M., E. Smets, M. Waelkens, H. Vanhaverbeke, I. Librecht,
E. Paulissen, & L. Vanhecke, 2000. Late Holocene environmental change
and the record of human impact at Gravgaz near Sagalassos, southwest
Turkey. Journal of Archaeological Science, 27(7), 571–96.
Vermoere, M., S. Bottema, L. Vanhecke, M. Waelkens, E. Paulissen, &
E. Smets, 2002. Palynological evidence for Late-Holocene human occupa-
tion recorded in two wetlands in SW Turkey. The Holocene, 12(5), 569–84.
Vernet, J.-L., 1997. L’Homme et la Forêt Méditerranéenne de la Préhistoire à nos
Jours. Paris: Éditions Errance.
Vigne, J. D., & H. Valladas, 1996. Small mammal fossil assemblages as indicators
of environmental change in northern Corsica during the last 2,500 years.
Journal of Archaeological Science, 23(2), 199–216.
Vita-Finzi, C., 1969. The Mediterranean Valleys: Geological Changes in Historical
Times. Cambridge: Cambridge University Press.
Vita-Finzi, C., & E. Higgs, 1970. Prehistoric economy in the Mount Carmel
area of Palestine. Site catchment analysis. Proceedings of the Prehistoric
Society, 36(1), 37.
Vitali, V., J. A. Gifford, F. Djindjian, & G. R. Rapp, Jr., 1992. A formalized
approach to analysis of geoarchaeological sediment samples: The loca-
tion of the early punic Harbor at Carthage, Tunisia. Geoarchaeology, 7(6),
545–81.
Vogel, S., & M. Märker, 2010. Reconstructing the Roman topography and envi-
ronmental features of the Sarno River Plain (Italy) before the AD 79 erup-
tion of Somma-Vesuvius. Geomorphology, 115(1–2), 67–77.
Vogel, S., M. Märker, & F. Seiler, 2011. Revised modelling of the post-AD
79 volcanic deposits of Somma-Vesuvius to reconstruct the pre-AD 79
topography of the Sarno River plain (Italy). Geologica Carpathica, 62(1),
5–16.
Vogiatzakis, I. N., & G. H. Griffiths, 2008. Island biogeography and land-
scape ecology, in Mediterranean Island Landscapes: Natural and Cultural
350 Bibliograph
Approaches, eds. I. N. Vogiatzakis, G. Pungetti, & A. M. Mannion.

Dordrecht: Springer, pp. 61–81.
Vogiatzakis, I. N., G. Pungetti, & A. M. Mannion, 2008. Introduction to the
Mediterranean island landscapes, in Mediterranean Island Landscapes:
Natural and Cultural Approaches, eds. I. N. Vogiatzakis, G. Pungetti, &
A. M. Mannion. Dordrecht: Springer, pp. 3–14.
Vogiatzakis, I. N., & O. Rackham, 2008. Crete, in Mediterranean Island
Landscapes: Natural and Cultural Approaches, eds. I. N. Vogiatzakis,
Waelkens, M., E. Paulissen, M. Vermoere, P. Degryse, D. Celis, K. Schroyen,
B. De Cupere, I. Librecht, K. Nackaerts, & H. Vanhaverbeke, 1999. Man
and environment in the territory of Sagalassos, a classical city in SW Turkey.
Quaternary Science Reviews, 18(4/5), 697–710.
Wagner, B., A. Lotter, N. Nowaczyk, J. Reed, A. Schwalb, R. Sulpizio,
V. Valsecchi, M. Wessels, & G. Zanchetta, 2009. A 40,000-year record of
environmental change from ancient Lake Ohrid (Albania and Macedonia).
Journal of Paleoliminology, 41(3), 407–30.
Wagstaff, J. M., 1992. Agricultural terraces: The Vasilikos Valley, Cyprus, in Past
and Present Soil Erosion, Archaeological and Geographical Perspectives, eds.
M. Bell & J. Boardman. Oxford: Oxbow, pp. 155–61.
Wagstaff, M., & C. Gamble, 1982. Island resources and their limitations, in An
Island Polity: The Archaeology of Exploitation in Melos, eds. C. Renfrew &
M. Wagstaff. Cambridge: Cambridge University Press, pp. 95–105.
Wagstaff, M., & J. F. Cherry, 1982a. Settlement and population change, in An
Island Polity: The Archaeology of Exploitation in Melos, eds. C. Renfrew &
Wagstaff, M., & J. F. Cherry, 1982b. Settlement and resources, in An Island
Polity: The Archaeology of Exploitation in Melos, eds. C. Renfrew &
Wagstaff, M., & S. Augustson, 1982. Traditional land use, in An Island Polity:
The Archaeology of Exploitation in Melos, eds. C. Renfrew & M. Wagstaff.
Cambridge: Cambridge University Press, pp. 106–33.
Wainwright, J., 1994. Anthropogenic factors in the degradatation of semi-arid
regions: A prehistoric case study in Southern France, in Environmental
Change in Drylands: Biogeographical and Geomorphological Perspectives, eds.
A. C. Millington & K. Pye. London: John Wiley, pp. 285–304.
2004. A review of European Union funded research into the history and
evolution of Mediterranean desertification. Advances in Environmental
Monitoring and Modelling, 1(4), 1–87.
Wainwright, J., & J. B. Thornes, 2004. Environmental Issues in the Mediterranean:
Processes and Perspectives from the Past and Present. London: Routledge.
Waldren, W., 1997. The definition and duration of the Beaker Culture in
the Spanish Balearic Islands: A radiocarbon survey. Oxford Journal of
Bibliograph 351
Walker, B. H., & D. A. Salt, 2006. Resilience Thinking: Sustaining Ecosystems

and People in a Changing World. Washington: Island Press.
Walker, B. H., C. S. Holling, S. R. Carpenter, & A. Kinzig, 2004. Resilience,
adaptability and transformability in social-ecological systems. Ecology and
Society, 9(2).
Wallace, S., 2010. Ancient Crete: From Successful Collapse to Democracy’s
Alternatives, Twelfth to Fifth Centuries BC. Cambridge: Cambridge
University Press.
Walsh, K., 2004. Caring about sediments: The role of cultural geoarchaeology
in Mediterranean landscapes. Journal of Mediterranean Archaeology, 17(2),
223–45.
2005. Risk and marginality at high altitudes: New interpretations from field-
work on the Faravel Plateau, Hautes-Alpes. Antiquity, 79, 289–305.
Walsh, K., & F. Mocci, 2003. Fame and marginality: The archaeology of
Montagne Sainte Victoire (Provence, France). American Journal of
Walsh, K., & F. Mocci, 2011. Mobility in the mountains: Late third and second
millennia alpine societies’ engagements with the high-altitude zones in the
Southern French Alps. European Journal of Archaeology, 14(1–2), 88–115.
Walsh, K., F. Mocci, J. Court-Picon, J. M. Palet-Martinez, & S. Tzortzis, 2005.
Dynamique du peuplement et activités agro-pastorales durant l’âge du
Bronze dans les massifs du haut Champsaur et de l’Argentièrois (Hautes-
Alpes). Documents d’Archéologie Méridionale, 28, 25–44.
Walsh, K., & S. Richer, 2006. Attitudes to altitude: Changing meanings and per-
ceptions within a ‘marginal’ Alpine landscape – The integration of palaeo-
ecological and archaeological data in a high altitude landscape in the French
Alps. World Archaeology, 38(3), 436–54.
Walters, B. B., & A. P. Vayda, 2009. Event ecology, causal historical analysis,
and human–environment research. Annals of the Association of American
Geographers, 99(3), 534–53.
Warren, P., 1996. The Aegean and the limits of radiocarbon dating. Acta
Archaeologica, 67, 283–90.
Watrous, L. V., 1982. Lasithi. A History of Settlement on a Highland Mountain
Plain in Crete. Princeton: American School of Classical Studies at Athens.
Watrous, L. V., D. Chatzi-Vallianou, K. O. Pope, N. Mourtzas, J. Shay, C. T.
Shay, J. Bennet, D. Tsoungarakis, E. Angelomati-Tsoungarakis, C. Vallianos,
& H. Blitzer, 1993. A survey of the Western Mesara Plain in Crete:
Preliminary report of the 1984, 1986, and 1987 field seasons. Hesperia, 62
(2), 191–248.
Watts, W. A., J. R. M. Allen, B. Huntley, & S. C. Fritz, 1996. Vegetation history
and climate of the last 15,000 years at Laghi di Monticchio, Southern Italy.
Quaternary Science Reviews, 15(2/3), 113–32.
Weiburg, E., M. Lindblom, B. Leppänen Sjöberg, & G. Nordquist, 2010. Social
and environmental dynamics in Bronze and Iron Age Greece, in The Urban
352 Bibliograph
Mind: Cultural and Environmental Dynamics, eds. P. Sinclair, G. Nordquist,

F. Herschend, & C. Isendahl. Uppsala: African and Comparative Archaeology,
Department of Archaeology and Ancient History, pp. 149–94.
Wengler, L., & J. L. Vernet, 1992. Vegetation, sedimentary deposits and
climates during the Late Pleistocene and Holocene in eastern Morocco
Palaeogeography Palaeoclimatology Palaeoecology, 94 (1–4), 141–67.
Weninger, B., E. Alram-Stern, E. Bauer, L. Clare, U. Danzeglocke, O. Jöris,
C. Kubatzki, G. Rollefson, H. Todorova, & T. van Andel, 2006. Climate
forcing due to the 8200 Cal yr BP event observed at Early Neolithic sites in
the Eastern Mediterranean. Quaternary Research, 66(3), 401–20.
Weninger, B., L. Clare, E. Rohling, O. Bar-Yosef, U. Böhner, M. Budja, M.
Bundschuh, A. Feurdean, H. G. Gebe, O. Jöris, J. Ç. Linstädter, P. Mayewski,
T. Mühlenbruch, A. Reingruber, G. Rollefson, D. Schyle, L. Thissen,
H. Todorova, & C. Zielhofer, 2009. The impact of rapid climate change
on prehistoric societies during the Holocene in the Eastern Mediterranean.
Documenta Praehistorica, 36, 7–59.
Westley, F., S. Carpenter, R. W. A. Brock, C. S. Holling, & L. H. Gunderson,
2002. Why systems of people and nature are not just social and ecologi-
cal systems, in Panarchy: Understanding Transformations in Human and
Natural Systems, eds. L. H. Gunderson & C. S. Holling. Washington: Island
Press, pp. 103–19.
Whatmore, S., 2002. Hybrid Geographies: Natures, Cultures, Spaces. London: Sage.
Whitelaw, T. M., 1991. The ethnoarchaeology of recent rural settlement
and land use in northwest Keos, in Landscape Archaeology as Long-Term
History: Northern Keos in the Cycladic Islands from the Earliest Settlement
until Modern Times, eds. J. F. Cherry, J. L. Davies, & E. Mantzourani. Los
Angeles: Institute of Archaeology, University of California, pp. 403–54.
Whittle, A., 2002. Houses in context: Buildings as process, in Neolithic Houses in
Northwest Europe and Beyond, eds. T. Darvill & J. Thomas. Oxford: Oxbow,
pp. 13–26.
Wick, L., 1994. Vegetation development and human impact at the forest limit:
Palaeoecological studies in the Splügen Pass area (north Italy), in Highland
Zone Exploitation in Southern Europe, eds. P. Biagi & J. Nandris. Brescia:
Museo civico di scienze naturali di Brescia, Monographie di ‘Natura
Bresciana’ 20, pp. 123–32.
Wick, L., & A. Möhl, 2006. The mid-Holocene extinction of silver fir (Abies
alba) in the Southern Alps: A consequence of forest fires? Palaeobotanical
records and forest simulations. Vegetation History and Archaeobotany, 15(4),
435–44.
Wilkinson, T. J., 2005. Soil erosion and valley fills in the Yemen highlands and
southern Turkey: Integrating settlement, geoarchaeology, and climate
change. Geoarchaeology, 20 (2), 169–92.
Wilkinson, T. J., & S. T. Duhon, 1990. Franchthi Paralia: The Sediments,
Stratigraphy, and Offshore Investigations, eds. T. J. Wilkinson & S. T.
Duhon. Bloomington: Indiana University Press.
Bibliograph 353
Willis, K. J., 1992. The Late Quaternary vegetational history of northwest

Greece. III. A comparative study of two contrasting sites. New Phytologist,
121, 139–55.
1994. Altitudinal variation in the Late Quaternary vegetational history
of northwest Greece. Historical Biology: An International Journal of
Paleobiology, 9(1), 103–16.
Wilson, I., 2001. Before the Flood. London: Orion.
Winterhalder, B., 1994. Concepts in historical ecology: The view from evolutionary
ecology, in Historical Ecology: Cultural Knowledge and Changing Landscapes,
ed. C. L. Crumley. Santa Fe: School of American Research Press, pp. 17–41.
Wittenberg, L., H. Kutiel, N. Greenbaum, & M. Inbar, 2007. Short-term
changes in the magnitude, frequency and temporal distribution of floods
in the Eastern Mediterranean region during the last 45 years – Nahal Oren,
Mt. Carmel, Israel. Geomorphology, 84(3–4), 181–91.
Wossink, A., 2009. Challenging Climate Change: Competition and Cooperation
Among Pastoralists and Agriculturalists in Northern Mesopotamia (c. 3000–
1600 BC). Leiden: Sidestone Press.
Wright, J. C., J. F. Cherry, J. L. Davis, E. Mantzourani, S. B. Sutton, &
R. F. Sutton, 1990. The Nemea Valley archaeological project: A preliminary
report. Hesperia, 59(4), 579–659.
Yalcner, A. C., B. Alpar, Y. Altnok, I. Ozbay, & F. Imamura, 2002. Tsunamis in
the Sea of Marmara. Marine Geology, 190(1–2), 445–63.
Yaltrak, C., B. Alpar, M. Saknc, & H. Yuce, 2000. Origin of the Strait of Canakkale
(Dardanelles): Regional tectonics and the Mediterranean-Marmara incur-
sion. Marine Geology, 164(3–4), 139–56.
Yanko-Hombach, V., P. Mudie, & A. S. Gilbert, 2011. Was the Black Sea cata-
strophically flooded during the Holocene? – Geological evidence and archae-
ological impacts, in Submerged Prehistory, eds. J. Benjamin, C. Bonsall, &
A. Fisher. Oxford: Oxbow Books, pp. 245–62.
Yasuda, Y., H. Kitagawa, & T. Nakagawa, 2000. The earliest record of major
anthropogenic deforestation in the Ghab Valley, northwest Syria: A palyno-
logical study. Quaternary International, 73–4, 127–36.
YaSuouka, J. U. N. K., & R. I. C. H. Levins, 2007. Impact of deforestation and
agricultural development on Anopheline ecology and malaria epidemiology.
American Journal of Tropical Medicine and Hygiene, 76(3), 450–60.
Yener, K. A., C. Edens, T. P. Harrison, J. Verstraete, & T. J. Wilkinson, 2000.
The Amuq Valley regional project, 1995–1998. American Journal of
Archaeology, 104, PART 2, 163–220.
Yll, E. I., R. Perez-Obiol, J. Pantaleon-Cano, & J. M. Roure, 1997. Palynological
evidence for climatic change and human activity during the Holocene on
Minorca (Balearic Islands). Quaternary Research, 48(3), 339–47.
Zachos, K., 1996. Tracing a destructive earthquake in the southwestern
Peloponnese (Greece) during the Early Bronze Age, in Archaeoseismology,
354 Bibliograph
Zangger, E., 1991. Prehistoric coastal environments in Greece: The vanished

landscapes of Dimini Bay and Lake Lerna. Journal of Field Archaeology, 18,
1–15.
Zangger, E., 1994. Landscape changes around Tiryns during the Bronze Age.
American Journal of Archaeology, 98(2), 189–212.
Zangger, E., 2001. The Future of the Past. London: Weidenfeld and Nicholson.
Zangger, E., M. E. Timpson, S. B. Yazvenko, F. Kuhnke, & J. Knauss, 1997.
The Pylos regional archaeological project: Part II: Landscape evolution and
site preservation. Hesperia, 66 (4), 549–641.
Zazo, C., J. L. Goy, L. Somoza, & C. J. Dabrio, 1994. Holocene sequence of sea-
level fluctuations in relation to climatic trends in the Atlantic–Mediterranean
linkage coast. Journal of Coastal Research, 10(4), 933–45.
Zielhofer, C., & D. Faust, 2003. Palaeoenvironment and morphodynamics in the
mid-Medjerda floodplain (northern Tunisia) between 12,000 and 2000 BP:
Geoarchaeological and geomorphological findings, in Alluvial Archaeology
in Europe, eds. A. J. Howard, M. G. Macklin, & D. G. Passmore. Lisse:
Swets & Zeitlinger, pp. 203–16.
Zielhofer, C., & D. Faust, 2008. Mid- and Late Holocene fluvial chronology of
Tunisia. Quaternary Science Reviews, 27(5), 580–8.
Zielhofer, C., D. Faust, F. az del Olmo, & R. Escudero, 2002. Sedimentation
and soil formation phases in the Ghardimaou Basin (northern Tunisia) dur-
ing the Holocene. Quaternary International, 93–4, 109–25.
Zielhofer, C., D. Faust, & J. Linstädter, 2008. Late Pleistocene and Holocene
alluvial archives in the Southwestern Mediterranean: Changes in fluvial
dynamics and past human response. Quaternary International, 181(1),
39–54.
Zielhofer, C., D. Faust, R. B. Escudero, F. D. Del Olmo, A. Kadereit, K. M.
Moldenhauer, & A. Porras, 2004. Centennial-scale late-Pleistocene to mid-
Holocene synthetic profile of the Medjerda Valley, northern Tunisia. The
Holocene, 14(6), 851–61.
Zielhofer, C., & J. Linstädter, 2006. Short-term mid-Holocene climatic dete-
rioration in the West Mediterranean region: Climatic impact on Neolithic
settlement pattern. Zeitschrift für Geomorphologie NF, 142, 1–17.
Zielinski, G. A., & M. S. Germani, 1998. New Ice-Core Evidence Challenges the
1620s BC age for the Santorini (Minoan) Eruption. Journal of Archaeological
Science, 25, 279–89.
Zohary, D., & M. Hopf, 2001. Domestication of Plants in the Old World. Oxford:
Oxford University Press.
Index
Acarnania, Greece, 77 Andros, island, 225

Acheloös, river, Greece, 77 Aniene Valley, Italy, 276
Actor Network theory (ANT) anthracology, 21, 134, 192, 205, 259
hybrids, 206 Alps, 258
Adriatic, 33, 39, 40, 52, 152, 216 cultural choices of wood/plants,
Aegean, 22, 25, 33, 38, 39, 41, 56, 142, 205
65, 137, 175, 210, 215, 217, Anthropic-Climatic Regime, 148
230, 236 Antikythera, island, 224, 228–9
Aegila (Goat Island), 228 Apennines, Italy, 96, 97, 255–7
Aeolian Islands, 210 Apulia, Italy, 38, 129, 151
aeolian processes, 45, 49, 137, 189 aqueducts, 108, 109, 201
Aetolia, Greece, 77 Aquileia, Italy, 53, 58
Agios Dimitrios (stream), 83 arboriculture, 142, 168, 183, 185–7,
agro-sylvo-pastoral management, 190 252, 275
Aguas river, Spain, 84 Arcadia, Greece, 109
Albania, 250, 253 archaeobotanical remains, 42
Algeria, 19, 137 Argaric culture, Spain, 166, 168–70,
alluvial geoarchaeology, 69, 85, 117 183, 291
alluvial ‘stability’ and ‘instability’, Argive, Greece, 47, 105, 118, 172
69 Argolid, Greece
Almería, Spain, 132, 165, 182 erosion and time, 207
Alpilles, France, 158, 203 evidence for erosion, 178–80
Alps, 257–65 Argos, Greece, 103, 172
Alpujarra, Spain, 165 arid landscapes, 136, 189
Amorgos, island, 227, 236 aridification, 5, 94, 126, 129, 132,
Amuq Plain, Turkey, 141 133, 135, 144, 161
Amuq Valley, Turkey, 81 Arles, France, 115, 201, 204
Anatolia, 24, 85, 127, 140, 142, 250, Arno river, Italy, 57
287 Artemis (goddess), 109
Andalucia, 60 Athens, Greece, 207
Andorra, 266 Atlantis, 15
355
356 Index
Atlas Mountains, Morocco, 126, 136, impacts on environment (Near

248 East), 138–9
Atlit-Yam, Israel (site), 41, 64 impacts on environment (South of
Auser river, Italy, 57 France), 156–8
Ayios Dhimitrios, Greece (site), 19 impacts on environment (Spain),
163–6, 167–70
badlands, 121, 165 interactions with (small) islands,
Baetica, Spain (site), 60 226–8
Baiae, Italy (site), 61 interactions with alpine
Balearics, 212, 231–5 environments, 262
Balma Margineda, Andorra (pollen interactions with mountains, 266,
core), 156 267, 271–3
Bamboula, Cyprus (site), 50 use of timber/wood, 191
Banyoles, Spain, 131 Bronze Age aridity, 141
Banyoles, Spain (pollen core), 159 Bronze Age Dark Ages, 22, 148
Barbegal, France (Roman watermill), Butzer, K.
201–3, 206 non-equilibrium model, 177
Barcelona, Spain, 182
Basilicata, Italy, 69, 152 Cacchiavia Valley, Italy, 84
Beirut, Lebanon, 56 Caesarea, Israel (site), 57
Belo, Spain (site), 60 Calabria, Italy, 84
Beqa’a Valley, Lebanon, 14 Calabrian Arc, 11
Beysehir Occupation Phase, 141 Calderone Glacier, Italy, 255
Biferno Valley Project, Italy, 89 Camargue, France, 14, 79, 115
Biferno Valley, Italy, 152 Campania, Italy, 92
biogeographical characteristics of the Campo Imperatore, Italy, 96, 97
Mediterranean, 11–14 Campus Martius, Rome, 97
biogeography, 238 Çanakh Lake, Turkey, 142
islands, 218 canals, 47, 105, 112
mountains, 244–6 Cardial Neolithic, 39, 40
Birkat Ram, Golan Heights (pollen Carmel, Israel, 41, 42
core), 138 Çarşamba river (Turkey), 85
Black Sea, 20, 31 Carthage, Tunisia, 53, 55
Blondel, J., 14 cash crops, 183–8
Borghetto Sotto, Italy (pollen core), Castellón Alto, Spain (site), 168
258 Castellón, Spain, 182
Bramefan, France (site), 199 Castelporziano, Italy (site), 62
Braudel, F., 29, 282 Çatalhöyük, 85, 86, 142, 296, 340,
Briatico, Italy, 34 343
bridges, 102–3, 203 Çatalhöyük, Turkey (site), 85–7, 142
Bronze Age catastrophies, 284
development of terraces, 177 catastrophism, 15–27
impacts on environment (Greece), Cecina river, Italy, 73
146–8 Celsus (ancient physician), 114
Index 357
centuriation, 111, 180, 204 Coto Doñana, Spain, 79

Cerdagne, France, 266, 274 Crau plain, France, 193, 289
Cesano, river, Italy, 93 Crete, 20, 21, 22, 34, 35, 39, 49, 53,
Chios, island, 217 129, 147, 174, 190, 210, 216,
Claudian harbour (Rome), 61 228, 229, 235, 237, 238, 253,
climate, 159 254, 286
14C anomaly, 154 mountains, 255
6200 BC (8200 BP) ‘event’, 137 Croatia, 35
alpine climate change, 257 Cultural Ecology, 7–9
Bronze Age, 255 Cultural Ecology & Historical
Bronze Age aridity, 139, 148–9, Ecology, 280
164, 168–70, 178, 226–7, environmental knowledge –
286–7 definition, 7
climate and vegetation change, Human Behavioural Ecology, 161
133–4 Cumae, Italy (site), 56
Holocene changes South of France, Cuniculi, 97
156 Cyclades, 210, 215, 218, 223, 224–7,
identifying weather patterns, 228, 238
103–4, 168 Cyprus, 41, 50, 74, 177, 210, 216,
impact on landscapes, 119–21 239
Iron Age, 273, 287
Iron Age ‘deterioration’, 262 Dalmatia, 39
rainfall patterns, Holocene (Spain), dams, 47, 96, 105
221 Daskaleio-Kavos, Keros (site), 229
Roman ‘warm period’, 262, 275 Dead Sea, 26, 139
Roman ‘climatic optimum’, 288 degradation
transition to a Mediterranean-type degradation hypothesis, 6, 165,
climate, 154 171–2, 240, 281
Cnidus, Turkey (site), 24 degradation myth (‘Fall from
coastlines Eden’), 3, 123–4
archaeological sites and sea levels, Greek/Roman culpability, 288–9
35–6 landscape degradation (evidence for
biological zonation and sea levels, and against), 135, 142–8, 154,
35 188, 226, 285, 288–9
Col de Roburent, France, 260 pristine landscapes myth, 6
colmatage (use of sediments to Dehesa, 192, 193–5
improve wetlands), 112 deities
Copais, Greece, 105 and landscape/environmental
copper mines, 256 characteristics, 109
Corinth, Greece (site), 65, 108, 109, Delos, Greece (site), 52, 177
250 Delphi, Greece (site), 27
Corsica, 40, 73, 210, 216, 239, 240 Oracle, 27
Cossetània, Spain, 184, 305 Delphinos, Crete (pollen core), 21,
cothon, 54–5 129
358 Index
deltas, 47, 57, 60, 77, 90, 97 Campania, Italy, 92

desert/pre-desert exploitation, 189 Classical periods, 149
desertification, 135, 140, 168, 242, climatically induced, 135
289 deforestation as cause, 93, 141,
Dikataian Mountains, Crete, 254 174–5, 207, 256, 266
Dimini, Greece (site), 44 Eastern Mediterranean examples,
disease, 80, 85, 105, 112, 113, 114, 139–40
117, 151, 160, 285 erosion and labour, 207–9
Dodecanese Islands, 210 erosion mitigation, 136
Domaine Richeaume, France (group geology and climate, 11
of sites), 197 Greece, Neolithic Bronze Age,
Donezan, France, 266, 274 143–7
drainage, 117 human responses to, 199–200
drainage systems, 99, 105 islands, 226, 230
Cloaca Maxima, 97 Italy, 152
dredging, 53, 55, 59, 65 Malta, 219
dykes, 106, 115, 201, 205 North Africa, 188–9
ports & harbours, 57–60
earthquakes, 19, 23, 24, 25, 26, 37, Pyrenees, 266
49, 50 Roman control/mitigation, 206–7
Bronze Age ‘earthquake storm’, South of France, 153–4, 158, 199,
22–4 231–2
earthquakes – impact on society, 24–7 Spain, 164
Ebro Delta, Spain, 79 Thessaly, 87
Ebro Valley, Spain, 182 Tiber, 97
Ecrins National Park, France, 260, 276 truncation of units, 204
El Acebrón, Spain (pollen core), 194 estuaries, 46
El Gallo, Spain (pollen core), 287 etesians (winds), 67
el huerto (garden plot cultivation), 161 Etruscans, 97
Enveig Mountain, France, 266 eustasy, 31, 34, 35, 36, 37, 40, 45,
environmental determinism 46, 55, 65, 286
problems with, 268 extinctions (faunal), 231–2
Ephesus, Turkey (site), 58
Epipalaeolithic, 138, 148, 216 Fangeas, France (site), 275
erosion fields, 180–1, 184, 236
alluvial, 69–70, 85 Fiora, river, Italy, 93
Alps, 265 fire
and landscape degradation fire (natural events), 161, 181–3,
hypothesis, 122–4 290
and seismic causes, 25 fire (use of for landscape
and weather events, 103 management), 138, 161, 181–3,
Bronze Age (Argolid), 178 271
Bronze Age to Classical, 174–81 fire making, 192
Index 359
fish Gouves, Crete (site), 22, 39

fishponds, 118 Gozo, Malta (site), 220, 222
fish/fishing, 33, 41, 49, 62, 81, 91, Gravgaz, Turkey (cores), 128, 141
118 grazing (impacts), 162–3, 178, 188,
Fiume di Sotto di Troina River Valley, 192, 238, 267
Sicily, 70 Greece, 191
flood mitigation, 93 evidence for Late Neolithic/Bronze
flood myth, 284 Age impact, 143–8
floods, 68, 277 evidence for vegetation change,
flash-floods, 92 172–4
mitigation, 99, 101 Grotta-Pelos
Rome, 96–100 culture, 226
floodwater farming, 87, 188 settlement, 226
fluvial, definition, 69 Guadalentín River, Spain, 164
Fontaine du Vaucluse, France, 78 Gulf of Cadiz, Spain, 38
foraminifera, 33, 36 Gulf of Genoa, 11
Formentera, island, 231, 232, 235 Gymnesic Islands, 231
Fos, France (site), 53, 58, 59
Franchthi Cave, Greece (site), 45–6, harbours. See ports & harbours
143 hemp (Cannabis)
Fréjus, 53 production/propogation, 185
Fréjus, France (site), 53 Heracleia, island, 238
Heracles, 77, 106, 110
Galičica National park, Macedonia/ Stymphalian birds, 106
Albania, 253 Herodotus, 110, 226
Garamantes (people), 187 Hierapolis, Turkey (site), 24
Gasquinoy, France (site), 185 Homer, 225
Gatas, Spain (site), 170 Odyssey, 225
geology Horden, P. & Purcell, N., 15, 27,
of coasts, 35 283, 289
structure of region, 10–11 Huelva Estuary, Spain, 38
Gerakaris Valley, Greece, 83 Hula Valley, Israel, 139
Ggantija, Malta (site), 223 hydraulic systems/engineering,
Ghardimaou Basin, Tunisia, 136 101–2, 105–9
Gilgamesh, epic, 190 Roman, 248
Gla, Greece (site), 106 hydrology
glaciers environmental knowledge, 110
and climate change, 258 subterranean channels, 106
Glanum, France, 94, 95, 96, 158
Göbekli Tepe, Turkey (site), 148 Ibiza, 231, 232, 235
Gordion, Turkey (site), 100–1 IFD (Ideal Free Distribution)
Gornalunga Valley, Sicily, 69 ecological model, 162
Górtys, Greece (site), 149 Inachos, river, Greece, 106
360 Index
Ionian Sea, 21, 35, 210, 227, 229 Khirbet Faynan, Jordan, 188
Iron Age Kition, Cyprus (site), 50
interactions with mountains, 273 Kleonai, Greece (pollen core), 185
responses to erosion, 199 Klidhi bridge, Greece, 102
irrigation, 24, 26, 167, 168, 221, 284 Knossos, Crete (site), 39
islands, 235 Kommos, Crete (site), 49
as ‘laboratories’ for the study of Kos, Aegean, island, 25
culture, 214, 218 Kumptepe, Turkey (site), 141
biogeographic ranking models, 217 Kythera, island, 227
biogeography, 212, 219
carrying capacity, 218, 219, 239 Lac des Lauzons, France (pollen
colonisation, 215–18, 219, 228, core), 273
231–2 Laghi di Monticchio, Italy (pollen
communication/connectedness, core), 129, 151
218, 224, 225–7 Lago Albano, Italy (pollen core), 185
consequences of sea-level change, Lago di Fogliano, Italy (pollen core),
215, 217, 238 111
economic specialisation, 224–6 Lago di Nemi, Italy (pollen core),
relative levels of human impact, 185
235–6 Lago Grande, Italy (pollen core), 258
size range definitions, 212 Lago Lucone, Italy (site), 152
isostasy, 31, 34 lagoons, 46, 50, 54, 55, 56, 58, 59,
Israel, 43, 51, 87, 124, 138, 139, 61, 158
140, 290 Lake Copais, Greece, 105
Italy Lake Gölhisar, Turkey (pollen core),
evidence for impacts on 21
environment (Neolithic/Bronze Lake Lerna, Greece (pollen core), 172
Age), 149–53 Lake Ohrid, Macedonia/Albania, 253
vegetation development, 150–2 Lakonia, Greece, 195
Languedoc, France, 153, 154, 156,
Jebel al-Aqra, Syria, 140 157, 158
Larnaca, Cyprus, 50
Kalavasos Tenta, Cyprus (site), 177 Las Madres, Spain (pollen core), 194
Karpathos, island, 236 Lattara, France (site), 77, 154, 192
karst, 78–9 Laurentine Shore, Italy, 62
karstification, 78 Lazio, Italy, 62, 93
Kasserine, Tunisia, 249, 250 Le Carrelet, France (site), 115
katavothres (sinkholes), 106 Le Gall, J. (Ancient Historian), 97
Kavousi, Greece, 148 Lebanon, 14, 177, 190
Kefalonia, island, 229 Lechaion, Greece (site), 65
Kephissos Valley, Greece, 105 Lefka Ori. See White Mountains,
Keros, island, 229 Crete
Kfar Samir, Israel (site), 184 Lefkada, island, 229
Index 361
Lefkosia, Cyprus, 74 as a discourse, 121

Leptospirosis, 113 geological underpinnings, 11
Lerna, Greece (site), 172 of mountains, 247, 249, 250
Lerna, marshes, Greece, 110 wetlands as marginal, 80, 152
Levant maritime technology, 218
evidence for Neolithic/Bronze Age boats/ships, 215
impacts on environment, 140 Marseille, France, 52, 57, 58, 59,
Lez delta, France, 77 185, 199
Libya, 46, 126, 135, 136–7, 187, Marsillargues, France (pollen core),
188, 189, 248, 290 154
Lichines, Cyprus, 50 Marta, river, Italy, 93
Liguria, Italy, 256 Masara Plain, Crete, 49
Livorno, Italy, 57 Matala Bay, Crete, 49
Llobregat river, Spain, 159, 182 Mediterranean Sea
Lolkos, Greece (site), 44 characteristics, 31–3
Los Millares, Spain, 76, 165, 168 ecological productivity, 33
culture, 195 Mediterranean Outflow Water, 31
Lozoya Valley, Spain (pollen core), mediterraneanism, 2–4
185 Medjerda, river, Tunisia, 94
Lugnano, Italy (site), 113 Medjerda, Tunisia, 94, 102, 187
Luna, Italy (site), 65 Melos, 216, 225–6, 238
Menderes Delta, Turkey, 56
MacArthur, R.H. & Wilson, E.O. Mercantour, Alps, France, 262
(island biogeography theory), 213 Mesolithic
Macedonia, 83, 143, 149, 252, 253 colonisation of islands, 215, 216
Madrid, Spain, 14 continuation of food gathering
Madriu Valley, Pyrenees, 266 strategies, 91
Madriu-Perafita-Claror, valley, in Alps, 257, 260
Pyrenees, 267 mesolithic/epipalaeolithic
Mahdia, Tunisia (site), 55 exploration/colonisation of islands,
malaria, 80, 112, 113–14 215–16
problems with identification, 113 Messenia, Greece, 23, 47
Mallorca, 216, 231–4 Mezzaluna, Italy (pollen core),
Malta, 212, 216, 219–24, 290 111
‘cart ruts’, 221 Miletos, Turkey (site), 56
temples, 220–3 mining, 259, 268, 272, 275, 277
maquis, 124, 128, 131, 132, 133, use of wood, 256
141, 174, 179, 223, 233 Minoan, 21, 22, 39, 49, 147, 174,
Marche region, Italy, 93, 152 190–1, 227, 237, 254, 285
marginality religion, 286
and islands, 236 society, 21
and landscape degradation, 165 Minorca, 232–4
and Maltese temple building, 223 Misa, river, Italy, 93
362 Index
Mitsinjites, Cyprus (terraces), 177 impacts on environment (South of

Mochlos, Crete (site), 21, 237 France), 156
Mont Bego, France (site/s), 262 impacts on environment (Spain),
Mont Lepini, Italy, 111 159–63
Montou, France (site), 192 interactions with alpine
Montpellier, France, 77, 130, 154, environments, 260–2
192 interactions with coasts, 39–46
Morocco, 81, 126, 136, 248 interactions with islands, 215–18
mountains interactions with mountains, 252,
and marginality, 247 265, 266, 269–71
as barriers, 247 interactions with rivers and
biogeography, 244 wetlands, 85–8
Bronze Age ‘incorporation’ of Nile, Egypt, 20, 74
mountains, 255
climate change, 247 obsidian, 225
North Africa, 250 Olive (Olea), 151, 158, 165, 174, 178
Roman interaction with, 258 domestication, 184
timberline, 259 production, 138, 139, 191
transhumance, 248–9 production North Africa, 188
vertical (ecological) zonation, 246, production/propogation, 183–4
251 Ostia, Italy (site), 53, 58, 60, 63
Mycenae, Greece (site), 23, 172, 178 Ostra, Italy (site), 93
Mycenaean, 22, 47, 106, 145, 178, ostracods, 36
208, 227 Otzi, 271
hydraulic engineering, 105–6 Ovid, 223
Myotragus balearicus, 231
myths, 4, 15, 51, 77, 79, 110, 111 Palace of Nestor, Greece (site), 23,
rivers, 110 47, 48, 146
Palughetto basin, Italy (sites), 257
Nabataean (society), 188 palynology
Narbo, Italy, 58 taphonomic issues, 259
Narce, Italy (site), 93 Pamukkale, Turkey (site), 24
Navarrés region, Spain, 161 Panarchy, 105, 281
Naxos, 227 Pantano Piccolo, Sicily, 51
Naxos, island, 225 Pantelleria, island, 223, 231, 239
Negev, Israel, 139, 140 pastoralism, , 4, 70, 86, 136, 137,
Nemea Valley, Greece, 92 193, 248, 249, 252, 254, 256,
Nemea, Greece, 92, 146, 172, 185 262, 265, 266, 267, 269, 272,
Neolithic 273, 274, 277, 288
early impacts on environment, 138, Alpine pastoralism, 258
142–3, 153–8 mountain-pastoralist/transhumance
impacts on environment (Pre- discourse, 269
Pottery Neolthic), 139 Pefkakia, Greece (site), 44
Index 363
Peloponnese, 23, 26, 47, 105, 128, Portugal, 193

134, 146, 147, 149, 150, 179, Poseidon (god), 109
185 postprocessualism, 6
Peloro Peninsula, Sicily, 50 phenomenology, 6
permaculture, 170 Po-Venetian Plain. See Po Plain
Petit Col Saint Bernard, France/Italy, pozzolana, 56
276 Prato Mollo, Italy (pollen core), 256
Petromagoula, Greece (site), 44 Prato Spilla, Italy (pollen core), 256
Pevkakia-Magoula, Greece (site), 44 Pre-Pottery Neolithic, 41, 138, 139,
Phaistos, Crete (site), 22 148
Phalasarna, Crete (site), 52 Priene, Turkey (site), 56
Philippi Plain, Greece, 143 pristine landscapes myth.
Phocean, 58 See degradation
Phoenicia, 190 progradation, 30, 35, 36, 47, 49, 57,
Phoenician (period), 50, 53, 55, 60, 62, 65, 66, 85, 90, 91
195, 223 Pseira, island, 177, 237
phreatic systems, 78 Punic, 55, 223, 231
Phrygian period, 100 Puy-St-Vincent, France, 261
Piano di Sorrento, Italy, 92 Pylos, Greece (site), 23, 47, 146
Piera, Greece, 83 Pyrenees, 184, 265–7
Pindus Mountains, 253 Pyrgos, Greece (site), 23
Pisa, Italy, 57, 100
Pistachio (Pistacia), 26, 127, 132, Rama, France (site), 265
142, 143, 159, 165, 172, 174, Ravenna, Italy, 53, 58
182, 184, 265 resilience, 15, 83, 85, 86, 121, 124,
Pitdiai, Crete (site), 22 180, 281, 282
Plato, 207, 288 and scale, 92
Pliny the Elder, 204 environmental, 141, 189
Pliny the Younger, 20, 64 islands, 212, 233, 237
ploughing (impact on landscapes), of agricultural landscape, 289
175, 226 of vegetation, 150, 154, 171
Po Plain, Italy, 89–92, 152 of vegetation after fires, 181
poljes, 106 protected by management, 291
pollution, 22, 59, 188 socio-cultural, 24, 25, 31, 116,
Polop Alto Valley, Spain, 164 149, 206, 208
Pompeii Premise, 282 tested by weather, 286
Pomptinae Paludes. See Pontine theory, 5, 8, 105
Marshes resin collection, 275
Pont Simian, France (bridge), 203 Rezina Marsh, Pindus Mountains
Pontine Marshes, Italy, 79, 111–13 (pollen core), 252
Roman Management, 111–13 Rhodes, 21, 37, 38
ports & harbours, 49–61, 100 Rhône Valley, France, 153, 155, 158,
silting, 47, 53, 55, 57, 58, 61, 65 189
364 Index
Rhône, river, France, 115, 130, 153, erosion, 201–7

155, 158, 293, 299, 301, 303, warm period. See Roman ‘climatic
304, 312, 316, 320, 322 optimum’
Rif, Mountains (Morocco), 248 Roman ‘climatic optimum’, 114, 262,
risk (definitions), 246 287–8
river Acheloös, 77 Roman villa, 197, 200, 201, 206
rivers Rome
anastomosing, 75, 82 floods, 96–100
braided rivers, 75, 83, 92 Roque Vaoutade, France, 201
channel migration, 92–3 ruined landscape theory.
characteristics of Mediterranean See degradation
rivers, 70–7
flash-floods, 74, 94, 174 Sainte Victoire, France, 196–201
floods, deforestation as cause, 277 Saliagos culture, 225
incision, 84 salinisation, 44, 284, 290
influence of glaciers, 97 salt production, 61
influence of tectonics and Samos, island, 230
topography, 70, 74, 84 San Piero a Grado, Italy, 57
influence on coastal sanctuaries, 49, 65, 109, 286
geomorphology, 100 as hybrids of cultural and
myths and religion, 110–11 environmental processes, 109
neolithic societies and alluvial sand dunes, 14, 49, 55, 62, 63
systems, 88 sandbars, 56, 59
responses to human activity and/or Santorini, 19, 20, 21, 39, 129, 225,
climate, 70, 85, 94, 104 238
Roman ash-fallouts, 20
attitudes to mountains, 277 chronology for eruption, 20
augering (‘environmental ritual’), eruption impact, 20–2
117 tsunami, 21
environmental optimum, 187 Sapropel, 33, 137
impact on woodland, 191–2 Sardinia, 40, 52, 210, 212, 216, 239
interactions with wetlands, 111–13 SCA. See site catchment analysis
interactions with alpine Scilla, Italy, 34
environments, 262–5 sclerophyllous vegetation, 13, 121,
interactions with desert/pre-desert, 123, 125, 130, 133, 134, 156,
188–9 160, 162, 171, 286
interactions with islands (Malta), scrub vegetation
223 chaparral, 124
interactions with mountains, fynbos, 124
248–50, 252, 266, 268 macchia, 111, 124
interactions with river systems, maquis. See maquis
96–101, 115 mattoral, 124, 223
responses to/management of phrygana, 133, 235
Index 365
sea currents, 65 Sparta, 26

sea level change, 230, 284–5 Sphakia, Crete, 253
impact on islands, 217 Sporades, islands, 210
sea levels (mean changes over time), springs, 78–9, 106, 109
37, 52 Stagno di Maccarese, Italy (pollen
Sea of Galilee, 138 core), 114
Sea of Marmara, 31, 38 storms (importance of in erosion),
sea temperature change, 40 174
Sebkhet Kelbia, Tunisia, 79 Strabo, 105, 274
secano (dry-land farming), 163 Straits of Gibraltar, 38
secondary products revolution, 162, Stymphalos (Stymfalia), Greece,
255 106–9
Segermes valley, Tunisia, 187 Suasa, Italy, 93
Segura Mountains, Spain, 160, 268 sweet chestnut (Castenea)
seismic, 22, 24, 25, 26, 27, 37, 38 production/propogation, 185
Sesklo, Greece (site), 44, 143 sylviculture, 277
Setia, Pontine, Italy, 111 Syria, 141
Sicily, 11, 50, 69, 70, 93, 129, 152,
210, 216, 219, 220 Tagus, river, Spain, 104
Sierra de Gádor, Spain (pollen core), Taormina, Italy, 34
183 Taurus mountains, Turkey, 191
Sierra de Gredos, Spain, 268 Tavoliere, Italy, 151
Sierra de Guadarrama, Spain, 185 tectonics, 14, 24, 31, 49, 65
Simbruni Mountains, Italy, 276 basic structures of the region,
Simitthus, bridge, Tunisia, 102 10–11
site catchment analysis, 166–7 Tel Dor, Israel (site), 51
Skorba, Malta (site), 216 Tel Nami, Israel (site), 42
Smyrna, Turkey (site), 58 Tel Yosef, Israel, 139
soils Tell es-Safi/Gathin, Israel (site), 140
development and stability, 147–8 Tells (Neolithic site-type), 143
ideal conditions for development, temporality
135 time and environmental knowledge,
stability promoted by vegetation, 207, 209, 282–3
123 Tenos, island, 225
solution notches, 53 terraces (agricultural), 175–80, 268
Soreq Cave, Israel, 141 social/political organisation, 178
South of France trenching, 175
evidence for impacts on Terramare, Italy, 91, 152
environment, 153–8 Thalassocracy, 63
Spain, 60, 103, 159–70, 193, 265, Thasos, island, 217
271 Thebes, Greece (site), 23
impacts on environment Thera. See Santorini
(Neolithic), 163 Thessaloniki, Greece, 102
366 Index
Thessaly, 87, 88, 145 research in the Argolid, 178–80

Neolithic settlement, 87–8 Vasilikos Valley, Cyprus, 177
Thrace, 39 Vauclusian springs, 78
Tiber, river, Italy, 58, 96–100, 114, vegetation
191 changes in mountains, 252, 256–7,
tides, 30, 46 258, 259–60
Tigalmamine, Atlas Mountains changes linked to climate, 133–4,
(pollen core), 248 135
timberline, 246 changes linked to climate or people,
Tindari, Sicily (site), 51 171
Tiryns, Greece (site), 23, 47, 105, development (Early Holocene),
172, 178, 283 124–5
Toscanos, Spain (site), 60 development of Mediterranean-type
Trajan, 61, 62, 63 vegetation, 165, 265
transhumance, 269–73, 277 development of Mediterranean-type
travertine, 78, 260 woodland, 123–33, 140, 152,
tree line, 246 154–6, 161
Treia, river, Italy, 93 general characteristics, 121
Tripolitania, 187, 189 island characteristics, 235
Troad, Turkey, 140 island vegetation changes, 233
Troy, Turkey (site), 53, 141 mountains – movements of
tsunamis, 21, 23, 38–9 treeline/timberline, 246
tufa, 78 North Africa, 136
Tunisia, 46, 53, 79, 81, 94, 102, 187, protection of soil, 123
219, 249, 290 relationships with erosion, 121–4,
Turkey, 35, 56, 85–7, 128, 141, 191, 175
See also Anatolia, 21, 24 resilience, 189, 229
Tyre, Lebanon (site), 55–6 resilience vis-à-vis fire, 181
Tyrrhenian Sea, 39, 51 reslience as a consequence of
human management, 124
Ubaye Valley, France, 260, 261, 265 Roman impact on vegetation,
Uruk, King of, 190 191–2
sacred woodland, 195
Val di Starleggia, Italy (pollen core), timber, 153, 167, 170, 185, 190,
258 191, 192, 259
Val Febbraro, Italy, 258 woodland use and management,
Val Vidröla, Italy (pollen core), 258 189–95, 259
Valencia, Spain, 14, 161 Vera basin, Spain, 83, 84
Valle di Castiglione, Rome, 150 Vesuvius, Italy, 11, 15, 18, 19, 20,
Vallée des Baux, France, 79, 145, 35, 92
201, 297, 304 Via Appia, Italy, 111
Vallon Saint-Clerg, France, 95 Vicdessos, France, 266, 274
van Andel, T.H. & C. Runnels Vine (Vitis), 194
Index 367
domestication, 185 weather systems (and the sea), 66–7

production/vineyards, 184–5 wells (water), 41, 64, 231
Voidokilia, Greece (site), 19 wetlands, 104–14, 152–3
Volaterrae region, Italy, 73 characteristics, 80
Volos Bay, Greece, 44 drainage, 112
environmental knowledge,
Wadi Tanezzuft, Libya, 187 110–11
wadis, 87, 137, 187, 188 management techniques,
walnut (Juglans), 223 112
as a marker of Roman impact, 185, White Mountains, Crete, 253
203 winds, 65–7
production/propogation, 185 woodlands. See vegetation
water shortages, 221, 231
water storage, 95, 231 Xerias, river, Greece, 103
water table, 42, 91, 139, 231 xeric. See sclerophyllous
watermills. See Barbegal, France
wave notches, 35, 36, 230 Yavneh-Yam, Israel (site), 290
weather, 247 Younger Dryas, 137
importance of considering possible Younger Fill (erosion phase), 135
weather patterns, 238, 265
importance of vis-à-vis climate, Zangger, E., 44
277, 278, 290 Zeus, 195

Walsh The Archaeology of Mediterranean Landscapes

Uploaded by

Copyright:

Available Formats

Walsh The Archaeology of Mediterranean Landscapes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Walsh The Archaeology of Mediterranean Landscapes

Uploaded by

Copyright:

Available Formats

The Archaeology of Mediterranean

This volume presents a comprehensive review of palaeoenvironmental evi­

Kevin Walsh is senior lecturer in the Department of Archaeology at the

Cambridge University Press is part of the University of Cambridge.

Cambridge University Press has no responsibility for the persistence or accuracy

Characteristics of Mediterranean rivers 70

The Roman watermill at Barbegal 201

to say that there is a dearth of medieval evidence. In fact, another rea­

defines the Mediterranean. Most of these environments are dealt with

have experienced the same processes that have profoundly transformed

Frameworks for the Assessment of Human–Environment

When we think about how people have interacted with an environ­

geomorphic system and vegetation), choice of crops, extraction of rocks

Environmental Knowledge and Cultural E

Rather than provide a comprehensive overview of the development of

management that are successful and enthusiastically adopted by local

From Geology to Biology: Defining the Mediterranean

This chapter comprises a brief overview of the geological and biogeograph-

Fundamental Geological and Biological Characteristics

The Mediterranean is defined in part by its geology. The Iberian, Eurasian,

volcanoes (Mather 2009) (Fig. 2.1). This half of the Mediterranean is

Basic Climatic and Biogeographical Characteristics

climate change comprise adjustments in average seasonal temperature

Tectonics and the Creation and Destruction of ­Niches

Although tectonic processes and associated volcanic activity can damage

evaluate possible human engagements with various environmental pro-

The precise definition of the chronological framework is essential for

earthquake ‘storm’ – a series of earthquakes in relatively short succes-

of catastrophic natural events, and whilst citing geological catastrophe

Hellenistic and Roman Catastrophes

We should not forget the other environmental consequences of seis-

In most instances, individual earthquakes, even if they are very pow-

Sea-Level Change and Coastal Settlement – Human

Characterising the Mediterranean

The virtual absence of tides across most of the Mediterranean results

(land advance caused by sediment deposition on river deltas) from about

The Mediterranean is a small ocean that has a direct connection with

Mediterranean. More specifically, plankton and foraminifera produc-

Many processes and configurations of the different seas are directly

2009). Coastline development is the result of the complex interplay of

Whereas eustatic sea-level rise has followed a relatively steady trend,

Examples of Mediterranean Coastal Change

One common feature of many reports on sea-level change is the use of

notch elevation above

main seismic uplift

AD event is the earliest in the sequence of tsunamis identified here

Pre- and Proto-Historic Coastal Exploitation

Mediterranean coasts could be considered as the most significant envi-

1: Mersin; 2: Parekklisha-Shillourokambos; 3: Akrotiri-Aetokremnos; 4: Kissonerga-Mylouthkia; 5: Ilipinar; 6: Yarimburgaz; 7: Nessonis; 8: Achilleion; 9: Theopetra;

t­ ransgression about 2,900 years ago, and consequently, any Neolithic

if the benefits of locating in such a zone are outweighed by the threats

4 G.W.L. Ground water level Water B 4

who undoubtedly continued to exploit marine resources. We know that

Coastal Exploitation: The Development of Ports and Harbours

Whilst there is plenty of evidence for mixed coastal economies, evidence

However, one fundamental problem was sedimentation and the infilling

management, which endeavoured to reduce the ever-present problem of

Early Ports and Harbours

The development of ports on Mediterranean islands during the Bronze

This volume presents a comprehensive review of palaeoenvironmental evi

to say that there is a dearth of medieval evidence. In fact, another rea

When we think about how people have interacted with an environ

Tectonics and the Creation and Destruction of Niches

t ransgression about 2,900 years ago, and consequently, any Neolithic