THE FLORIDA STATE UNIVERSITY
COLLEGE OF ARTS AND SCIENCES
PLACES, POTS, AND KURGANS: LATE COPPER AGE PATTERNS OF SETTLEMENT
AND MATERIAL CULTURE ON THE GREAT HUNGARIAN PLAIN
By
TIMOTHY A. PARSONS
A dissertation submitted to the
Department of Anthropology
in partial fulfillment of the
requirements for the degree of
Doctor of Philosophy
Degree Awarded:
Spring Semester, 2011
The members of the committee approve the dissertation of Timothy A. Parsons defended on
November 17, 2010.
_______________________________________
William A. Parkinson
Professor Directing Dissertation
_______________________________________
Lynne Schepartz
Co-Chair/Committee Member
_______________________________________
Daniel Pullen
University Representative
_______________________________________
Joseph Hellweg
Committee Member
Approved:
_____________________________________
Glen Doran, Chair, Department of Anthropology
_____________________________________
Joseph Travis, Dean, College of Arts and Sciences
The Graduate School has verified and approved the above-named committee members.
ii
© 2011 Timothy A. Parsons
iii
Dedicated to my family and friends. Life is grand, love is real, and beauty is everywhere.
iv
ACKNOWLEDGEMENTS
My interest in issues of migration and material culture started in 2004 on the Körös
Regional Archaeological Project, when, while driving along a two-lane road on a lazy Sunday
morning with Bill Parkinson, he pointed out the numerous small burial mounds along the road to
Gyula in Békés County, and he said, “Timmy, you should do kurgans!” This led me into the
works of Andrew Sherratt and David Anthony, whose perspectives on social, economic, and
migratory change greatly influenced my thinking and research trajectory in planning an
archaeological study of the Late Copper Age on the Hungarian Plain. Though kurgans aren’t the
only focus of this work, they were the beginning. And, I hope that this dissertation makes a
modest contribution to understanding them.
Faculty, friends, and family members have helped me complete this dissertation. It is an
impossible and unthinkable task to fit so many acknowledgements onto this regrettably short
space, and I apologize in advance for forgetting anyone who has contributed to this project along
the way. That said, at the very least I can do my best to highlight the individuals whose
encouragement and support have helped me so much over the years.
The faculty of this department has my utmost respect. In the most challenging of times,
they have steadfastly remained dedicated to the remaining graduate students. For this, we all
owe a debt of gratitude. Thank you to my committee members, Lynne Schepartz, Joseph
Hellweg, and Daniel Pullen. They have all been supportive and helpful during the production of
my dissertation.
Mike Galaty in the Department of Sociology and Anthropology at Millsaps College
became my undergraduate advisor in the spring of 1999 and gave me my first opportunities at
archaeological field experience, in addition to challenging me in numerous classroom settings
during my four years in Jackson, Mississippi. He taught me how to use a trowel, dig a shovel
test, draw profile and plan maps, use GPS and total station equipment, and introduced me in both
theory and practice to petrographic ceramic analysis. Most importantly, Mike taught me not to
limit myself to one dogmatic way of approaching solutions to archaeological problems, and to be
a “jack-of-all-trades” when it comes to theoretical and methodological considerations. He is a
great teacher, and I hope he keeps educating and encouraging young anthropologists and
archaeologists for years to come. The discipline will be better for it.
v
Bill Parkinson has been a strong and supportive advisor and mentor to me throughout my
graduate career. He provided me with both the firm guidance and great freedom to pursue the
work presented here. In addition to his unbounded archaeological knowledge, Bill successfully
saw me through the administrative, cultural, linguistic, and economic challenges of conducting
research in a foreign country. It is his faith in my ability to “just get it done” that gave me the
confidence to dive into the world of Hungarian archaeology. Billy was once described to me as a
“real archaeologist.” His students and colleagues all know that this is true – and we don’t need
the Indiana Jones ringtone to be sure of it.
I can’t thank Attila Gyucha, formerly of the Kulturális Örökségvédelmi Szakszolgálat and
now with the Hungarian National Museum, enough. I stayed at his home, ate his food, drank his
vodka, and took much more of his time than I deserved. He introduced me to important people,
prepared and submitted permit applications, and supplied every bit of his support that he could
without once complaining (at least to me), all while working day and night to finish his own
dissertation. He is a dedicated colleague, a talented researcher, and above all, he is a good
friend.
Paul Duffy let me tag along with him around Békés County as a wet-behind-the-ears
graduate student. He taught me how to use GIS, and how to quickly and effectively collect sites.
Paul was ceaselessly supportive of this research, and is a great (but tough) example to follow.
He also taught me how to get cars out of the mud equipped only with a tractor, the assistance of
two farmers, and pink twine.
Gábor “Baxi” Bácsmegi made sure that my project was never without the supplies and
knowledge necessary to succeed. Baxi provided the positive attitude and optimism that I
sometimes lacked, and encouraged me to stop and smell the roses. His mother and stepfather
welcomed me warmly into their home, and I will always cherish the time spent around fires in
the backyard cooking sausages and drinking beer. My time in Hungary would not have been the
same without Baxi, and it certainly would not have been as fun.
Over the years, many participants in Billy and Attila’s Körös Regional Archaeological
Project served as sounding boards and sources of advice and information. In no particular order
they are Meg Morris, Rod Salisbury, Hanneke Hoekman-Sites, Daniel Sosna, Rick Yerkes, Sam
Duwe, Nisha Patel, Julia Giblin, Smiti Nathan, Walt Warner, Abby Smith, Amy Nicodemus,
Michelle Markovics, and of course Billy, Attila, Paul and Baxi. Meg provided invaluable GIS
vi
help, and Rod helped me hash out site collection methodology as we visited sites in the Skoda in
the summer of 2009.
The many people who made my time in Hungary brilliant and made the country one of
my favorite places are too many to mention, and I regret that I’ve forgotten or never knew the
names many people who helped me. Again in no particular order, I thank Dori Kékegyi, Gergő
Bóka, Ottó Fogas, Ferenc Horváth, Pál Megyesi, Veronika Csik, the Csóti family, and the
women who run the Arany Oroszlán Panzió in Békéscsaba and provided me with many morning
cappuchinos and many evening Sopronis. I also thank all the farmers who allowed a foreigner
speaking poor Hungarian to tromp through their fields. Archaeology is everywhere in Békés
County, and farmers are used to archaeologists walking along with their faces toward the ground.
An archaeologist in a baseball cap with a funny accent tethered with twine to a pin flag walking
in circles is a somewhat less common sight. Nonetheless, I was consistently met with inquiring
minds and friendly (if bemused) faces.
I give special thanks to Attila Krieter and his colleagues at the Kulturális Örökségvédelmi
Szakszolgálat in Budapest. Attila taught me how to thin section ceramics and allowed me access
to his lab and equipment despite his busy schedule. Director Imre Szátmári, Anita Vári, Adrienn
Szanda, and the staff of the Munkácsy Mihály Múzeum in Békéscsaba were helpful and friendly
in the summer of 2009, and welcomed me into their museum to analyze and photograph
ceramics.
Virginia Carr, Alex Parsons, Annalee Shum, and Nicole DeFrancisco flew long distances
to photograph ceramics, collect sites, wash and label artifacts, sort lots, and endure more than a
few rainy, cold, and occasionally snowy days. Rumor has it that they also ate a lot of good food,
met great people, and had a pretty good time – and enjoyed a great number of British soaps on
cable. I thank them very much. Adam Kereki deserves thanks for his support and
encouragement.
I thank the Eisele Foundation and the National Science Foundation for their financial
support (Dissertation Improvement Grant Award Number BCS-0910071), especially program
administrator John Yellen and the anonymous reviewers that provided thoughtful and
constructive criticisms of my research plans.
My colleagues at the National Park Service Southeast Archeological Center deserve
thanks both for their continuous support of, and dedicated, good-natured belligerence toward my
vii
specialization in European prehistory and dissertation research in Hungary. A great deal of this
dissertation was written while on Section 106 and 110 compliance projects and ARPA
investigations in hotel rooms in Lake City and Jacksonville, Florida, Natchez, Mississippi,
Chattanooga, Tennessee, and Murray, Kentucky, as well as in hotel locations all along the Gulf
Coast (Mobile, Alabama, Ocean Springs and Pascagoula, Mississippi, and Perdido Key in
Pensacola, Florida) while deployed on resource advising assignments during the BP Gulf of
Mexico oil spill incident.
Many thanks to Dr. William Parker in the Florida State University Department of
Geology for allowing me access to both his laboratory and a petrographic microscope.
My fellow graduate students and friends in the Anthropology Department deserve thanks
for the support and encouragement that only good friends can give. Thank you to Alex Parsons,
Katie O’Donnell, Dan Seinfeld, Cyndi Bellacero, Josh Englehart, Guy and Ivy Hepp, Hanneke
Hoekman-Sites, Michelle Markovics, Sarah Moore, Ian Pawn, Collete Berbesque, Julie Byrd,
Ryan Duggins, Geoff Thomas, and all of my friends in Tallahassee. Extra special thanks to
Shannon Tucker and Malinda Carlisle, who made sure I was registered for hours and safe in the
embrace of bureaucratic morass when I was too far away or too absent-minded to take care of it
myself.
Aaron Head, Brandi Head, David Morreale, and Alison Morreale are all fine human
beings. Aaron and Brandi wouldn’t have let me quit had I tried, while David and Ali encouraged
me to take the first step and get started. David, especially, taught me that we grow our wings as
we fall. Thanks for being alive.
My family, especially my parents Paul and Carmen Parsons, has always encouraged me
to relentlessly pursue my goals. My father still serves as my paradigm of persistence and
stubbornness. Despite great adversity, he solders on through the nonsense of life and reminds
me that there is always another challenge to overcome and project to embrace. We could all
learn a lesson from my dad, and we would all be better people for it.
Most of all, I thank my wife. Alex is an inspiration to me, and has never let me fall into
the trap of frustration and discouragement that often accompanies work on dissertations. I can
only hope to live up to her example, and she is more than I deserve.
viii
TABLE OF CONTENTS
List of Tables ............................................................................................................................... xiii
List of Figures ............................................................................................................................... xv
Abstract ........................................................................................................................................ xix
CHAPTER ONE: INTRODUCTION ............................................................................................. 1
The Research Agenda ............................................................................................................. 1
Archaeological Models ................................................................................................. 2
Research Questions ....................................................................................................... 5
Overview of Methods ................................................................................................... 6
Settlement Pattern Analysis .............................................................................. 6
Current Archaeological Data ............................................................................ 7
Ceramic Analysis .............................................................................................. 7
Overview of Results and Implications .......................................................................... 9
Structure of the Dissertation ....................................................................................... 11
CHAPTER TWO: THEORETICAL BACKGROUND ............................................................... 14
Migration and Archaeology: Beyond the Normative Approach .......................................... 15
Migration as an Explanation: Sufficient Models for Material Culture Change? ........ 20
Interaction Spheres, Egalitarianism, and Social Change ............................................ 22
Beyond Migration: Interaction Spheres and the Spread of Material Culture ............. 22
Social Trajectory, Interaction, and Change................................................................. 24
Models for the Emergence of Ranked Societies ......................................................... 24
Approaching Migration, Archaeology, and Regional Models of Material Culture Change 27
Migration as an Explanation for the Appearance of Homogeneous Material Cultures24
Clovis in North America ................................................................................. 27
The Archaic North American Borderlands ..................................................... 28
The European Magdalenian ............................................................................ 29
The Neolithization of Europe.......................................................................... 30
The Iron Age Celtic Migrations ...................................................................... 32
Migration, Materially Homogeneous Material Cultures, and Social Change ............. 32
Summary............................................................................................................................... 33
CHAPTER THREE: THE ARCHAEOLOGICAL, GEOGRAPHIC, AND GEOLOGICAL
SETTING ...................................................................................................................................... 34
Introduction .......................................................................................................................... 34
The Geographic and Geological Setting of the Great Hungarian Plain ............................... 34
The Geological Setting of the Great Hungarian Plain ................................................ 34
The Geological Setting of the Körös-Berretyó Region............................................... 37
General Soil and Environmental Characteristics ........................................................ 39
The Archaeological Setting .................................................................................................. 39
The Neolithic .............................................................................................................. 40
The Late Neolithic Tisza-Herpály-Csőszhalom complex ............................... 40
The Copper Age .......................................................................................................... 43
The Early Copper Age Tiszapolgár Culture ................................................... 43
ix
The Middle Copper Age Bodrogkeresztúr Culture ......................................... 45
The Late Copper Age Boleráz-Baden Culture ................................................ 46
The Late Copper Age Kurgan Culture ............................................................ 52
The Early and Middle Bronze Age ................................................................. 53
The Developmental Trajectory of the Great Hungarian Plain .................................... 54
Invasion vs. Economy: A Tale of Two Models on the Great Hungarian Plain .................... 57
The Invasion/Migration Hypothesis ........................................................................... 57
The Environmental/Economic Model ......................................................................... 59
Baden Pots with Local Roots? Defining the Late Copper Age on the Plain ........................ 60
Summary............................................................................................................................... 61
CHAPTER FOUR: THEORETICAL EXPECTATIONS AND RESEARCH DESIGN ............. 62
Introduction .......................................................................................................................... 62
Temporal and Geographic Scales of Analysis...................................................................... 62
Analyzing Social Change through Settlement Patterns and Regional Analysis................... 64
Previous Regional Analysis Projects on the Great Hungarian Plain .......................... 69
A Problem with Regional Studies, and how to Approach it in the Future.................. 73
Social Change, Ceramics, and Technology .......................................................................... 74
A Technological Approach to Ceramic Analysis ................................................................. 76
Macroscopic Analysis ................................................................................................. 76
Microscopic Analysis.................................................................................................. 78
Interpretive Framework ........................................................................................................ 82
Spatial Analysis: Observing Nucleation, Dispersal, and Association through Time . 83
Ceramic Analysis: Measuring and Observing Technological Change through Time 85
Summary............................................................................................................................... 87
CHAPTER FIVE: METHODOLOGY ......................................................................................... 88
Introduction .......................................................................................................................... 88
Selection of the Study Area .................................................................................................. 88
Geographic and Archaeological Location .................................................................. 89
Foundation of Recent and Previous Research in the Region ...................................... 89
Availability of Materials ............................................................................................. 90
Spatial Analysis .................................................................................................................... 91
Fieldwork and Site Revisits .................................................................................................. 93
Overview of Site Revisits and Collection ................................................................... 93
Site Selection .............................................................................................................. 94
Site Collection ............................................................................................................. 95
Description and Documentation of Finds ............................................................................. 97
Ceramic Coding .......................................................................................................... 98
Other Materials ........................................................................................................... 98
Photography ................................................................................................................ 98
Current Location of the Material .......................................................................................... 99
Ceramic Analysis.................................................................................................................. 99
Macroscopic Ceramic Analysis .................................................................................. 99
Microscopic Ceramic Analysis ................................................................................. 100
Summary............................................................................................................................. 101
x
CHAPTER SIX: ARCHAEOLOGICAL SITES AND ASSEMBLAGES ................................. 104
Introduction ........................................................................................................................ 104
MRT (Hungarian Archaeological Topography) Sites ........................................................ 104
Sites Revisited During Fieldwork ...................................................................................... 105
Sites Collected and Intensively Surveyed During Fieldwork............................................. 106
Békés 26 .................................................................................................................... 107
Békés 178 .................................................................................................................. 108
Bélmegyer 82 ............................................................................................................ 108
Biharugra 33.............................................................................................................. 109
Bucsa 13 .................................................................................................................... 110
Füzesgyarmat 97 ....................................................................................................... 111
Gerla 64 ..................................................................................................................... 111
Körösladány 21 ......................................................................................................... 112
Mezőberény 34.......................................................................................................... 113
Szeghalom 80 ............................................................................................................ 114
Tarhos 67 .................................................................................................................. 115
Previously Excavated Sites................................................................................................. 116
Doboz Homokgödöri Tablá ...................................................................................... 116
Hódmezővásárhely-Kopáncs I., Olasz-tanya ............................................................ 116
Summary............................................................................................................................. 117
CHAPTER SEVEN: RESULTS OF THE SPATIAL ANALYSIS ............................................ 134
Introduction ........................................................................................................................ 134
A Reassessment of Settlement in the Körös River Study Area .......................................... 136
Average Nearest Neighbor and Density Analysis .................................................... 138
A Reevaluation of Late Copper Age Settlement Location in the Körös Region ...... 145
Discussion and Conclusions of the Settlement Pattern Research ............................. 148
Summary............................................................................................................................. 151
CHAPTER EIGHT: RESULTS OF CERAMIC ANALYSIS .................................................... 152
Introduction ........................................................................................................................ 152
Description of Variables ........................................................................................... 153
Results of the Macroscopic Analysis ................................................................................. 154
Diachronic Ceramic Variability: Middle Copper Age vs. Late Copper Age ............ 154
Spatial Ceramic Variability: Late Copper Age Inter-site Variability in the Study
Region ....................................................................................................................... 157
Inter-Regional Variability of Baden Ceramics from the Körös and Maros Regions 158
Conclusions and Interpretations of the Macroscopic Ceramic Data ......................... 163
Results of the Petrographic Analysis .................................................................................. 164
General Petrographic Characteristics ........................................................................ 165
Diachronic Petrographic Variability ......................................................................... 169
The Middle Copper Age ..................................................................................... 169
The Late Copper Age .......................................................................................... 171
The Early Bronze Age ........................................................................................ 175
The Middle Bronze Age...................................................................................... 178
Summary of Diachronic Variability.................................................................... 182
xi
Spatial Variability in Late Copper Age Ceramics .................................................... 185
Petrographic Variability in the Körös Region .................................................... 185
Inter-regional Petrographic Variability ............................................................... 187
Discussion of the Petrographic Data ......................................................................... 189
Summary ......................................................................................................................... 190
CHAPTER NINE: DISCUSSION .............................................................................................. 191
Introduction ........................................................................................................................ 191
Modeling Change on the Great Hungarian Plain ............................................................... 192
Kurgan Builders, Migration, and the Late Copper Age ............................................ 192
Discussion of the spatial results .......................................................................... 192
Discussion of the ceramic results ........................................................................ 197
A Revisited and Expanded Model of Late Copper Age Settlement and Economy .. 199
Kurgans and cultural context .............................................................................. 201
Archaeological and modern examples of emulation ........................................... 202
Summary ............................................................................................................. 204
Long-term Population and Economic Continuity on the Great Hungarian Plain ..... 205
Implications for Anthropological Models of Homogeneous Material Cultures ................. 209
Summary............................................................................................................................. 211
CHAPTER TEN: CONCLUSION AND FUTURE RESEARCH DIRECTIONS ..................... 212
Conclusions ........................................................................................................................ 212
Future Research .................................................................................................................. 214
APPENDIX A: SITE COLLECTION SUMMARIES ............................................................... 217
APPENDIX B: PETROGRAPHIC DATA ................................................................................. 230
WORKS CITED ......................................................................................................................... 237
xii
LIST OF TABLES
4.1 Table depicting the interpretive framework for data sets utilized in this research ............... 83
6.1 Summary of analyzed ceramics from the MRT collection and Doboz H. tábla by site and
cultural period ..................................................................................................................... 129
6.2 Summary of analyzed ceramics from the collected sites and Hódmezővásárhely by site and
cultural period ..................................................................................................................... 130
6.3 Summary of sites described in MRT volumes as containing Late Copper Age surface
material ............................................................................................................................... 130
6.4 Summary of sites collected during the fall 2009 field season, as described in the MRT
volumes ............................................................................................................................... 133
7.1 All nearest neighbor calculations for the three analytical regions in the Körös study region,
organized by cultural phase ................................................................................................ 141
7.2 Average nearest neighbor calculations for Early Copper Age, Middle Copper Age, and Late
Copper Age archaeological sites in the Körös study region ............................................... 142
8.1 Sorting of visible inclusions in Middle and Late Copper Age ceramics from archaeological
sites in the Körös River watershed ..................................................................................... 155
8.2 Kneading of raw material (clay) in Middle and Late Copper Age ceramics from
archaeological sites in the Körös River watershed ............................................................. 155
8.3 Texture of a fresh break in Middle and Late Copper Age ceramics from archaeological sites
in the Körös River watershed. ............................................................................................ 155
8.4 Firing characteristics in Middle and Late Copper Age ceramics from archaeological sites in
the Körös River watershed ................................................................................................. 155
8.5 Firing characteristics in Late Copper Age ceramics from archaeological sites in the Körös
River watershed .................................................................................................................. 159
8.6 Sorting of visible inclusions in Late Copper Age ceramics from archaeological sites in the
Körös River watershed ....................................................................................................... 159
8.7 Sorting of visible inclusions in Late Copper Age ceramics from HódmezővásárhelyKopáncs I., Olasz-tanya in the Maros River watershed and Late Copper Age ceramics from
the Körös region ................................................................................................................. 159
8.8 Firing characteristics in Late Copper Age ceramics from archaeological sites in the Maros
region and from the Körös region....................................................................................... 162
xiii
8.9 Summary statistics of Middle Copper Age, Late Copper Age, Early Bronze Age, and
Middle Bronze Age petrographic point counts from sites in the Körös study region ........ 170
8.10 Summary statistics of Late Copper Age point counts from the Körös and Maros regions. 189
xiv
LIST OF FIGURES
3.1 The Carpathian Basin ........................................................................................................... 35
3.2 Rivers in the Körös-Berettyó study region after 19th century regulation ............................. 37
3.3 Prehistoric hydrology of the Körös region ........................................................................... 38
3.4 The Late Neolithic Tisza-Hérpály-Csőszhalom complex .................................................... 42
3.5 Extent of the Early Copper Age Tiszapolgár culture ........................................................... 44
3.6 Extent of the Middle Copper Age Bodrogkeresztúr culture ................................................. 46
3.7 Approximate extent of the Late Copper Age Baden culture ................................................ 48
4.1 The Körös River basin study area, including modern cities and MRT parish boundaries as
discussed in the text .............................................................................................................. 64
5.1 The Körös River basin study area, including modern cities and MRT parish boundaries ... 89
5.2 Systematic “dog-leash” collection unit site collection strategy............................................ 96
5.3 Coding sheet for quantitative ceramic petrography............................................................ 102
5.4 Coding sheet for qualitative ceramic petrography.............................................................. 103
6.1 Sites recorded in the MRT revisited during the fall 2009 field season .............................. 105
6.2 Sites systematically collected during the fall 2009 field season ........................................ 106
6.3 Transects and surface find locations at the site of Békés 26 .............................................. 118
6.4 Transects and surface find locations at the site of Békés 178 ............................................ 119
6.5 Collection units at the site of Bélmegyer 82 ...................................................................... 120
6.6 Collection units at the site of Biharugra 33 ........................................................................ 121
6.7 Transects and surface find locations at the site of Bucsa 13 .............................................. 122
6.8 Transects and surface find locations at the site of Füzesgyarmat 97.................................. 123
6.9 Collection unit locations at the site of Gerla 64. ................................................................ 124
6.10 Transects and surface find locations at the site of Körösladány 21.................................... 125
xv
6.11 Collection unit locations at the site of Mezőberény 34 ...................................................... 126
6.12 Collection unit locations at the site of Szeghalom 80 ........................................................ 127
6.13 Collection unit locations at the site of Tarhos 67 ............................................................... 128
7.1 Andrew Sherratt’s Dévaványa Plain study region in northern Békés County (outlined in
red), in contrast with the present study area ....................................................................... 135
7.2 Kurgan locations overlaid with kernel density map of kurgans per square kilometer ....... 136
7.3 Late Neolithic, Early Copper Age, Middle Copper Age, and Late Copper Age sites in the
Sherratt study area .............................................................................................................. 137
7.4 The three average nearest neighbor analytical zones in the study region .......................... 138
7.5 The Körös study region and all Early, Middle, and Late Copper Age sites ....................... 139
7.6 Early and Middle Copper Age site distribution in Sherratt’s study region ........................ 143
7.7 Kurgan locations overlaying a map of kurgan density to illustrate kurgan “clusters,” and
Late Copper Age site distribution in Sherratt’s study region ............................................. 144
7.8 Middle Copper Age Bodrogkeresztúr site distribution ...................................................... 145
7.9 Kurgans, density of kurgans per square kilometer, and Late Copper Age site distribution in
the Körös River study region .............................................................................................. 146
7.10 Late Copper Age sites located within kurgan clusters near Körösújfalu ........................... 147
7.11: Late Copper Age sites in close association with kurgans south of Gyomaendrőd ........... 148
7.12 Kurgan clusters near a concentration of Late Copper Age sites near Bélmegyer .............. 149
8.1 Firing conditions of Middle and Late Copper Age ceramics ............................................. 157
8.2 Firing condition of Late Copper Age ceramics from sites in the Körös Region ................ 160
8.3 Sorting of visible inclusions in Late Copper Age ceramics from sites in Körös Region ... 160
8.4 Kneading in Late Copper Age Ceramics conditions from Hódmezővásárhely and in the
Körös region ....................................................................................................................... 161
8.5 Paste texture in Late Copper Age ceramics from Hódmezővásárhely and in the Körös
region .................................................................................................................................. 161
xvi
8.6 Comparison of Late Copper Age ceramic firing conditions from Hódmezővásárhely and in
the Körös region ................................................................................................................. 163
8.7 Late Copper Age from Tarhos 67 ....................................................................................... 165
8.8 Early Bronze Age from Békés 26 ....................................................................................... 166
8.9 Neolithic from Gerla 64...................................................................................................... 167
8.10 Early Bronze Age from Szeghalom 80 ............................................................................... 167
8.11 Middle Copper Age from Szeghalom 80............................................................................ 171
8.12 Ternary plot of Middle Copper Age and Late Copper Age ceramic paste composition .... 172
8.13 Ternary plot of Middle Copper Age and Late Copper Age ceramic body composition .... 172
8.14 Middle Copper Age from Szeghalom 168.......................................................................... 173
8.15 Late Copper Age from Tarhos 67 ....................................................................................... 174
8.16 Late Copper Age from Mezőgyán 2 ................................................................................... 174
8.17 Early Bronze Age from Szeghalom 80 ............................................................................... 175
8.18 Ternary plot of Late Copper Age and Early Bronze Age paste composition ..................... 176
8.19 Ternary plot of Late Copper Age and Early Bronze Age body composition ..................... 176
8.20 Early Bronze Age from Szeghalom 80 ............................................................................... 177
8.21 Middle Bronze Age from Békés 26 .................................................................................... 178
8.22 Middle Bronze Age from Békés 178 .................................................................................. 179
8.23 Ternary plot of Early Bronze Age and Middle Bronze Age paste composition................. 180
8.24 Ternary plot of Early Bronze Age and Middle Bronze age body composition .................. 180
8.25 Middle Bronze Age from Békés 26 .................................................................................... 181
8.26 Ternary plot of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle
Bronze Age paste compositional variability ....................................................................... 183
8.27 Ternary plot of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle
Bronze Age body compositional variability ....................................................................... 183
xvii
8.28 Body composition of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle
Bronze Age ceramic samples ............................................................................................. 184
8.29 Percentage of temper and sand observed in point counted Middle Copper Age, Late Copper
Age, Early Bronze Age, and Middle Bronze Age ceramic samples .................................. 184
8.30 Ternary plot of Late Copper Age paste compositional variability in the Körös Region .... 186
8.31 Ternary plot of Late Copper Age body compositional variability in the Körös Region .... 186
8.32 Ternary plot of average paste composition of Late Copper Age ceramics from the Körös
region and from the site of Hódmezővásárhely-Kopáncs I., ............................................. 188
8.33 Ternary plot of average body composition of Late Copper Age ceramics from the Körös
region and from the site of Hódmezővásárhely-Kopáncs I., ............................................. 188
9.1 Monument to a modern Hungarian political movement placed atop a kurgan in Békés
County ................................................................................................................................ 201
xviii
ABSTRACT
This dissertation examines how patterns of regional homogeneity in material culture
develop on the local level. Archaeologists have long been concerned with how large, materially
homogeneous culture groups develop over large regions relatively quickly. Often, this
phenomenon has been associated with migration. In many cases, such as the Linearbandkeramik
(LBK) group in Europe and Clovis in North America, migration models are the best explanation.
However, in other cases such as the Early Copper Age Tiszapolgár culture on the Great
Hungarian Plain, local models of indigenous better fit the patterns of settlement and material
culture.
This project focuses on changes at the beginning of the Late Copper Age on the Great
Hungarian Plain at around 3,500 B.C. At this time, the relatively homogeneous Baden culture
became the dominant material culture group on the Plain. Two models of changes are tested
here: 1) that the Late Copper Age Baden culture developed out of local populations’ intensified
involvement in an interregional interaction sphere; and 2) change occurred through migration or
migrations onto the plain or a diffusion of material culture and other behaviors that drastically
affected settlement and social organization. In this vein, the presumably intrusive kurgan burial
tumuli that appeared in the region at about this time are of special interest.
These models are tested in two primary ways: 1) a multi-scalar settlement spatial analysis
of known archaeological sites; and 2) macroscopic and petrographic ceramic analysis aimed at
identifying technological and manufacturing changes over time that might point to either the
arrival of new people in the region (migration) or diachronic population continuity.
Insufficient evidence exists to support a migration catalyzing the social and settlement
changes observed at the beginning of the Late Copper Age. Although a migration scenario
cannot be ruled out definitively, settlement pattern analysis supports a model of internal social
trajectories leading to the changes, while macroscopic and petrographic ceramic analyses do not
reveal any changes in technological preparation or manufacturing methods indicating the arrival
of new people or pottery technologies in the region. Ultimately, the results of this research
suggest that even dramatic shifts in material culture and incorporation into wider material culture
groups can occur in times of population continuity through a combination of social and economic
processes.
xix
CHAPTER ONE
INTRODUCTION
The Research Agenda
This study addresses anthropological and archaeological questions about why societies at
certain times throughout human history exhibit material culture similarities over very large
geographic regions. Archaeologists have long been concerned with how and why cultures
change, and with how material culture traits spread and sometimes become ubiquitous across a
large region. The subject has roots in the earliest days of anthropology and has been an ongoing
topic of discussion and debate. Boas (1889, 1940), in the absence of a framework for crosscultural comparison, considered diffusion and migration as the primary methods by which
material culture spread and by which regionally homogeneous material cultures developed.
Much later, Steward’s (1955) multilinear evolution allowed for change from within. That is,
different populations could independently develop parallel features, without the necessity of
diffusion or migration. Both models leave something to be desired. Migration and diffusion
models are unsatisfying as they leave little leeway for independent cultural development, while
models of indigenous change often fail to account for remarkable material culture similarity over
large geographic areas. This dichotomy set the theoretical stage for the two competing models
addressed in this study: the necessity of migration for sudden, widespread material culture
change, or the possibility of change emerging from within a population.
The competing models of migration and continuity are best tested through the collection
and analysis of archaeological data. Archaeological research is unique in that data can be
analyzed quickly over large geographic areas, and across long periods of time. Such a regional
approach is key to identifying markers of migration in the archaeological record. As Binford
(1980) stated, subtleties in material culture – such as form, design, and manufacture – can reveal
evidence of migration not necessarily observable through other lines of evidence. These slight
differences marking migration in the archaeological record may also mark other human
behaviors on the landscape. For example, Stark (1988) argued that social boundaries can be
subtly marked by small technological and manufacturing indicators, and that even subconscious
technological choices can indicate conservative social boundaries. The operational sequences
that create these indicators are surprisingly resistant to change (Leroi-Gourhan 1993:305, 319),
1
and can remain reflected in material culture despite the long-term flow of personnel across
boundaries (Barth 1969).
The development of widespread, materially homogeneous cultures in relatively short
amounts of time has long intrigued archaeologists. Examples exist throughout prehistory and in
multiple locations throughout the world. Clovis in North America, Linearbandkeramik (LBK) in
Europe, the Körös-Starčevo-Criş culture of the northern Balkans, and the Early Copper Age
Tiszapolgár culture on the Great Hungarian Plain are all examples of material culture horizons
that are homogeneous over wide geographic areas. Often –as with Clovis and Körös – this
phenomenon is associated with social processes such as migration and diffusion. On the other
hand, in cases like the Tiszapolgár, indigenous change provides a better explanation. This
research addresses how such wide-ranging transitions play out at multiple analytical scales and,
ultimately, on the local level. This concern is driven by the question, “how do local populations
react, adapt, and change in response to the development and local adoption of regional-scale
material culture systems?” I seek to answer this question by investigating how local populations
on the Hungarian Plain became incorporated into the wider Baden material culture at the
beginning of the Late Copper Age (ca. 3,500 B.C.). In the past, researchers have proposed two
general models for explaining this change: 1) large-scale migration and diffusion from outside
the region (Anthony 1990; Gimbutas 1977); and 2) local change driven by patterns of regional
social and economic integration and organization (Sherratt 1997a, 1997b).
The Middle and Late Copper Age on the Great Hungarian Plain provides an excellent test
case for an anthropological discussion of migration and population continuity, as the periods in
question have been the focus of a migration/invasion controversy for decades. Additionally the
study of subtle material culture differences is possible through the examination of previously
collected materials and through systematic site surface collection. As such, it is possible to
identify local markers of manufacture that have persisted despite the region’s incorporation into
a geographically large, homogeneous material culture group, if such markers exist.
Archaeological Models
The fundamental way in which people organized their settlements on a regional scale,
buried their dead, and interacted socially changed dramatically during the Late Copper Age
(3,500-3,000 B.C.) on the Great Hungarian Plain (Anthony 1990; Childe 1930; Gimbutas 1979,
2
1997). During previous phases of the Copper Age, the eastern Hungarian cultural landscape
consisted of a largely homogeneous material culture assemblage that nonetheless varied greatly
from those of their contemporary neighbors in Transdanubia and across the Carpathian
Mountains to the north and northeast. During the latter part of the Copper Age, the expansive
Baden material culture group, which extended throughout northern Serbia, southern Poland,
southern Germany, and parts of Switzerland, became ubiquitous on the Hungarian Plain.
Gimbutas (1977) and Makkay (1986) suggested that the process of trans-regional
homogenization might have been influenced by the arrival of migratory kurgan (burial mound)
builders from the Eurasian steppe. At about the time of the beginning of the Late Copper Age,
thousands of kurgan burial tumuli in the style of Yamnaya tumuli of the Steppe (e.g., earthen
mound covering a burial chamber containing a body in the supine position with raised knees)
appeared on the eastern Hungarian Plain, and Gimbutas (1963, 1979) argued for the kurgan
builders’ influence on not only material culture change, but also a fundamental set of shifts in
religion and language as the first speakers of the Indo-European language family entered the
continent. Anthony (1990) agreed that migration should not be ruled out as a possibility.
Sherratt (1997a, 1997b) and Parkinson (2006) stressed the importance of structured, diachronic
change in the region. They argued that local patterns of settlement nucleation and diffusion
resulted from social leveling mechanisms and various levels of economic interaction with
populations outside of the Plain. However, Anthony, Sherratt, Parkinson, and Gyucha (2010) all
suggested that one prime mover was likely insufficient to explain the changes on the Plain at the
end of the Copper Age. The homogeneity and ubiquity of Baden during this time period
complicate the problem of Baden origins on the Plain and the role that the “Kurgan Culture”
(Gimbutas 1977) might have played in its appearance.
This research examines how local populations on the Great Hungarian Plain were
affected by incorporation into the wider Baden material culture complex. The research goals
include: 1) to identify changes in Middle Copper Age, Late Copper Age, and Early Bronze Age
ceramic manufacturing technology that may indicate the presence of an outside population’s
influence on manufacture; 2) to specify how settlement patterns changed during the period of
time leading up to Baden; and 3) to use ceramic and settlement data to understand how
populations on the Hungarian Plain reacted to Baden influence. It is hoped that anthropologists
3
and archaeologists facing similar situations throughout the world can use the methods and
framework presented here.
In order to test the competing models introduced above, this study examines settlement
patterns at multiple scales, focusing on the wider Körös River basin study area. This study area
composes the western two-thirds of Békés County in southeastern Hungary. The goal is to
determine if micro-regional patterns mirror those on the scale of the entire study area, and on the
Hungarian Plain generally, by declining in number and density toward the end of the Copper
Age. Such a pattern will mirror Sherratt’s (1997a, 1997b) observation of a decline in site
number and size at the scales of the entire Hungarian Plain, and a smaller study area in northern
Békés County. This project also examines less visible and conservative local indicators of
continuity and change – specifically, ceramic manufacturing technology – that would have been
less susceptible to change over time (see Lemmonier 1992). Although ceramic form may have
changed as Baden became prominent, subtle macroscopic and microscopic characteristics (such
as paste constituency, firing method, level of kneading, and presence, size, and frequency of
specific mineral inclusions) would likely have remained the same if no migration and
replacement scenario took place.
On the Great Hungarian Plain, the two competing models to be tested are best described
as 1) a model of indigenous change; and 2) a model of migratory change. Under the indigenous
change model, the Late Copper Age Baden culture on the Plain developed out of an intensified
involvement of local populations in an interregional interaction sphere. Under the migration
model, a series of migrations onto the Plain beginning as early as the Middle Copper Age
drastically affected both material culture and settlement patterns. These models are probabilistic,
and it is entirely possible that a combination of local development and migratory populations
contributed to culture change on the Plain at this time. The primary variables employed in this
study to test the models are 1) settlement patterns over time; and 2) ceramic manufacturing
techniques and technology over time. In order to understand how these variables differ at
varying geographic resolutions, each will be observed at multiple scales – e.g., the site level, the
micro-region, and the regional level.
4
Research Questions
Based on previous research, there are two models for social change at the end of the
Copper Age on the Great Hungarian Plain: 1) a migratory population of pastoralists from the east
arrived on the Plain sometime around 3,500 B.C., catalyzing social change and fundamentally
altering the linguistic, cultural, burial, and settlement characteristics of the Hungarian Plain; and
2) change at the end of the Copper Age was part of an indigenous long-term cyclical process of
integration and regional differentiation, greatly influenced by advances in transportation
technology (see Anthony 2007). The following research questions are used as a guiding
framework for understanding if Baden on the Hungarian Plain developed out of local
populations’ intensified involvement in an interregional interaction sphere, or through a series of
migrations onto the Plain.
1) Do the sudden changes in material culture on the Hungarian Plain during the Late Copper Age
represent a demic migration into the region, or the adoption of a regional style by indigenous
populations? If a demic migration onto the Plain drastically affecting populations in the region
occurred during this time period, then both the technological and design elements of ceramic
manufacture should reflect this change. However, if no migration occurred or if the arrival of
kurgan builders was an epiphenomenal occurrence, then elements of ceramic manufacture should
remain the same throughout the Copper Age and into the Bronze Age, despite changes in form
and decoration. Such continuity might involve similarities in paste constituency, similar firing
characteristics (e.g., level of reduction/oxidization), and similar patterns of mineral inclusions
and void space ratios. If such similarities between ceramics of different periods in the Copper
Age are observed, it will support a model of long-term developmental processes affecting change
in the Late Copper Age.
2) Why do social groups choose to do things in a similar manner over wide geographic areas, and
how do local groups change their behavior when incorporated into regionally homogeneous
material culture groups? This question is relevant to addressing both of the competing models
presented above, simply because changes in material culture are often the most easily observable
variables in the archaeological record. However, other relevant patterns must also be addressed.
For example, long-term trajectories in shifting settlement patterns have been observed
5
throughout the Neolithic, Copper Age, and Early Bronze Age on the scale of the Hungarian Plain
(Sherratt 1997a, 1997b); however, few studies have addressed how these settlement patterns
changed at the local level. Do they reflect wider regional trends, or is there local variability?
Overview of Methods
The research methods employed in this study consist of two analytical components: 1)
settlement pattern analysis at multiple resolutions (the level of the Hungarian Plain, the Körös
River basin, Sherratt’s [1997a, 1997b] previous study area, and micro-levels within Békés
County), and 2) petrographic and macroscopic analysis of Middle Copper Age, Late Copper
Age, Early Bronze Age, and Middle Bronze Age ceramics from the Hungarian Archaeological
Topography surveys (MRT, see Chapter 4, pages 69-70) and from recent systematic collection in
the Körös River study area. Along with prehistoric settlement, paleoenvironmental, and
topographic data already available, data collected as part of this project include information
gathered during site revisits and systematic collection throughout the Körös region and Békés
County. Each of the models for social change being tested here has discrete patterning that is
observable archaeologically (see Table 4.1). These models are not mutually exclusive. Some
patterns of material culture and settlement change may be best interpreted as resulting from the
effects of migration and long term, patterned internal change.
Settlement Pattern Analysis. Since the 1970s, archaeologists have been concerned with
systematizing different aspects of archaeological spatial analysis, including settlement analysis,
site system analyses, regional studies, territorial analyses, locational analyses, catchment area
studies, distribution mapping, density studies, and ultimately the integration of these types of
information into single large databases (Clarke 1977; Galaty 2005). Each of these forms of
spatial study can be used at particular scales and in particular contexts to answer specific
archaeological questions (Clarke 1972a:47, 1977). This study is concerned with multiple scales
of analysis, and one of the questions framing the research is: how do settlement patterns, or
changes in settlement patterns over time, serve as an indicator of social change?
Parkinson’s (2006) and Gyucha’s (Gyucha et al. 2004; Gyucha 2010) research has
reaffirmed continuity between the Neolithic, Early Copper Age, and Middle Copper Age on the
eastern Great Hungarian Plain. They have noted, along with others (see Bankoff and Winter
1990; Sherratt 1997a, 1997b), a break in this continuity between the Bodrogkeresztúr (Middle
6
Copper Age) and Boleráz/Baden (Late Copper Age) periods based primarily on differences in
ceramic form and decoration. Stark (1998a:1, 1998b) argued that social groups and their
boundaries are marked by observable patterns in the archaeological record, meaning that such
breaks in ceramic continuity could indicate a different social group (e.g., different people) in the
region. Therefore, a study of formal variation in settlement type, location, and degree of
nucleation or dispersal is useful in determining the degree of continuity or change in the region.
Current archaeological data. Unlike most of Europe, the Körös River basin of the
eastern Great Hungarian Plain has three decades of archaeological survey available for spatial
analysis, encompassing an area of over 3,000 km2 (Escedy et al. 1982; Jankovich et al. 1989;
Jankovich et al. 1998). This multi-volume effort – the Hungarian Archaeological Topography,
or MRT, provides an advantage over the regional study of prehistoric societies elsewhere. As
such, the survey data for most of Békés County serves as the largest analytical scale for the
original research presented in this dissertation.
Almost 600 presumably invasive kurgan burial mounds are recorded in the published
MRT volumes for the study area, along with 70 Middle Copper Age sites, and 105 Late Copper
Age sites. This provides a statistically significant sampling universe in terms of qualitative
settlement pattern analysis, nearest neighbor analysis, and density analysis. The principle units
of analysis, in addition to relevant geographic and cartographic data (rivers, modern cities, etc.)
include settlement data on Early, Middle, and Late Copper Age sites, kurgan burial mound
locations, as well as locations of Late Neolithic, and Early and Middle Bronze Age
archaeological sites. The specific spatial methodology employed in this research project is
described completely in Chapter Five.
Ceramic Analysis. In addition to MRT materials, 11 Late Copper Age sites (singlecomponent and multi-component) were systematically collected in the Körös region of Békés
County, in order to obtain more ceramic samples of various time periods throughout the region,
and to field-test the accuracy of the MRT survey data. Additionally, ceramics from the
excavated Late Copper Age site of Doboz Homokgödöri-tablá in the Körös region were included
not only to augment the sample, but also to serve as a chronological and stylistic control.
Samples from the excavated site of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the adjacent
Maros River watershed were analyzed as a counterpoint to the Körös region samples, and to
7
account for any possible intra-regional variability not ascribable to the effects of migration or
invasion.
To address the research questions and the testable models, ceramics were analyzed and
coded macroscopically according to a battery of descriptive variables (see Chapter Five). The
relative frequencies of these variables, and how they change diachronically and according to
archaeological culture, illustrate how attributes of preparation and manufacture changed over
time. A sub-sample of these materials was then selected for microscopic petrographic analysis.
Using Whitbread’s (1995) and Stoltman’s (1989) methodologies, ceramic fabrics were classified
qualitatively and quantitatively based on frequency and size of natural mineral inclusions,
intentionally added tempering materials, void space characteristics, and other variables of the
ceramic paste and body. Each sample was thin-sectioned, and point counting was conducted
with a polarizing microscope under both plane-polarized light and under crossed polars. A
mechanical click-stop stage was utilized for precise counting at 2mm intervals. Multiple
transects per slide were counted, and point counting was performed blind in order to ensure
objectivity. Following Stoltman’s (1989) methodology for point count analysis, the proportions
of matrix, temper, and sand were calculated for a vessel’s clay body and matrix, silt, and sand for
a vessel’s paste. These proportions, when tri-graphed, allow different ceramic fabrics to be
delineated and compared, and illustrate changes in paste and body over time.
Whitbread’s (1995) and Stoltman’s (1989) methods of ceramic analysis allow for
qualitative, quantitative, macroscopic, and microscopic description of change over time. Indeed,
a recent dissertation by Kreiter (2005) illustrated the utility of using petrographic methods to
analyze technological change in Early and Middle Bronze Age ceramics from Transdanubia.
Change in finishing surface treatments and/or design and decoration are to be expected, since
these characteristics often change on a generational basis. However, culturally embedded
techniques for preparation and manufacture tend to be conservative and resistant to change
(Lemmonier 1992; Michelaki 1999). Therefore, if distinctive changes in the manufacturing
process of paste composition are observed over time, and especially if they correspond
temporally with the drastic form and decoration changes at the end of the Middle Copper Age
and beginning of the Late Copper Age, it would indicate the possibility of an outside cultural
influence. In conjunction with settlement data indicating the culmination of a long-term
8
organizational trajectory, the results of a detailed ceramic analysis could support either an
internal process of change, or migration and diffusion leading up to the Early Bronze Age.
Overview of Results and Implications
The results of the spatial and ceramic analyses do not support a model of settlement and
material culture change, and material culture homogenization, catalyzed by a prime-mover event
such as migration. Conversely, nothing emerging from the analyses rules out a migration of
kurgan-building pastoralists onto the Great Hungarian Plain at 3,500 B.C.; however, the impact
of such a migration on the indigenous populations and their developmental and economic
trajectories appears to have been minimal.
The spatial analysis supports the conclusions of Sherratt’s (1997a, 1997b) analysis of
settlement patterns on the Dévaványa Plain in northern Békés County, Hungary. At county-level
resolution, kurgans and Late Copper Age archaeological sites appear to exist in a complementary
distribution accross the landscape. However, at a higher resolution in multiple locations
throughout study region, kurgans and Late Copper Age sites are quite close to one another. This
indicates that a model employing a strategy of avoidance between two populations cannot be
safely assumed, or at the very least that kurgans and Late Copper Age settlements were not
constructed or utilized contemporaneously. Furthermore, the results presented in this
dissertation concur with Sherratt’s model of a continuing dispersal and perhaps depopulation
during the Middle and Late Copper Age. As in the Dévaványa Plain study region, Middle and
Late Copper Age settlement occurs less frequently, at lower density, and more ephemerally than
in previous and subsequent periods. This supports a model of a long-term social and settlement
involving cycles of population nucleation and dispersal rather than a sudden migration event
leading to population and settlement changes.
Similarly, the macroscopic and petrographic ceramic analyses support a model of
indigenous change rather than a sudden shift that one would expect under a migration model.
Macroscopic analysis indicates similar ceramic preparation and manufacturing techniques over
space within and beyond the study region during the Late Copper Age, and diachronically within
the study region. Petrographic analysis reveales the same pattern, as well as a moderated, longterm shift in the intentional addition of grog temper to the ceramic paste. Such patterns of longterm, measured change do not support a migration hypothesis for the social, settlement, and
9
material culture changes witnessed in the latter half of the Copper Age on the Great Hungarian
Plain.
The results of this study speak to more than just regional and local issues of material
culture and economic change. Most importantly, this dissertation argues that migration
scenarios, the spread of interaction spheres, and models of economic interaction and integration,
when applied individually, often fail to satisfactorally account for the development of regionally
homogeneous material cultures. In the case of the Great Hungarian Plain at the end of the
Copper Age, a complex interaction of factors created the cultural tapestry visible in the
archaeological record. A modest migration of kurgan builders into the region might account for
the appearance of burial tumuli across the landscape, as the migrants were quickly incorporated
into the indigenous economic and social structures. Kurgans quickly became dominant features
and landmarks in the region, and their emulation and reuse throughout later periods is testament
to their importance through time. At about the same time, the Baden ceramic tradition become
prevelent on the Hungarian Plain. This occurred as economic ties with populations beyond the
Carpathian Mountains solidified after several centuries of settlement intensification on the
margins of the Plain, near access to trade routes and raw material sources.
The Hungarian Plain at the end of the Copper Age is a place and time demonstrative of
what occurs when interaction spheres converge and overlap, in terms of both material culture and
settlement. Although the addition of kurgans to the fabric of culture on the Plain contributed to
the region’s uniqueness, Baden material culture appeared on the Plain primarily as the result of
integration into a wider economic structure. It was the economic ties developed by the end of the
Late Copper Age allowed for intensified acquisition and use of bronze and metal objects. This
contributed to a return to settlement nucleation and tell-centered economy and settlement on the
Plain during the Early and Middle Bronze Age, and untimately the appearance of
institutionalized hierarchy as higher value items moved through the economic network. The
regional material culture homogeneity observed on the Great Hungarian Plain should therefore
be viewed as an indicator of economic integration rather than as evidence of change catalyzed by
migration or invasion.
The implications of these results will hopefully shape future research into the Late
Copper Age in the Carpathian Basin. Most importantly, it is suggested that the process of
material culture homogenization and incorporation into the wider Baden material culture horizon
10
during this time period was not the result of a demic migration and resulting population
replacement or demographic shift. Second, it is indicated that dramatic changes in material
culture and incorporation into geographically wider material and economic horizons can occur
swiftly, but concurrently with long-term indigenous social and economic trajectories. Finally,
the implications for the prehistory of the Körös region of the Great Hungarian Plain are wide
ranging, but boil down to the indication of a long-term process of population nucleation and
dispersal cycles from the Neolithic to the Bronze Age shaping social trajectories and economic
interaction, eventually culminating in a settlement nucleation and return to a tell-centered
economy and settlement system during the Middle Bronze Age. The Late Copper Age period of
dispersal and settlement intensification at the margins of the Plain played a key role in this
diachronic process.
Structure of the Dissertation
The purpose of this dissertation is to clearly and succinctly describe unresolved questions
in the prehistory of the Great Hungarian Plain, address them through rigorous analysis, and draw
supportable, but testable, conclusions from the results. Ultimately, the results speak to the larger
anthropological issues of migration, diffusion, and large-scale regional material culture
homogenization.
Given the range of material covered in the study, this dissertation is organized in a
straightforward, traditional manner in order to present the anthropological and archaeological
background, theoretical expectations, research design and methodology, and results of the study
as clearly and logically as possible. In this chapter, I provide a brief outline and summary of the
argument that follows. Chapter Two presents the theoretical expectations underpinning the
research, focusing on issues related to the development of regionally homogeneous material
culture groups. I frame the phenomenon of regionally homogeneous material culture in the
context of migration models, which are the framework that this research study tests most
directly. I also discuss issues of migration and the development of regionally homogeneous
material culture in a cross-cultural and anthropological context by summarizing the theoretical
bases of migration, diffusion, and indigenous development models as explanations for change in
settlement patterns and material culture. Overviews of other anthropological models for the
development of regionally homogeneous material culture groups are also presented, including
11
interaction spheres and the theoretical links between the appearance of materially homogeneous
regional cultures and the development of institutionalized hierarchy.
Chapter Three provides the relevant background to the prehistory of the Great Hungarian
Plain. I place the region into its archaeological context, and integrate the theoretical foundation
presented in Chapter Two into the specific archaeological context of the Plain. Geographic and
geological summaries of the Plain and the Körös River watershed study area are also provided.
A period-by-period archaeological overview of social development and change is provided, with
a detailed focus on Early and Middle Copper Age trends toward local and regional style groups,
followed by the Late Copper Age incorporation of the Plain into the wider Baden material
culture group.
Chapter Four establishes the methodological links between the archaeological and
anthropological concepts introduced in Chapter Two and the archaeological test case of this
dissertation (the material culture and settlement changes at the beginning of the Late Copper
Age). I present the methodological principles that guided the research design and provide an
interpretive framework for the results of the study. The second half of the chapter explicitly
discusses historical approaches to social change through ceramic analysis, and how technological
analysis of ceramics has developed over the last century.
Chapter Five discusses the analytical methods used as part of this study to examine
changes in settlement patterns between the Late Neolithic and Early Bronze Age in the Körös
River study region, the field methods used for site visitations, mapping, measurement, and
systematic surface collection, and macroscopic and microscopic ceramic analysis. The specific
methodologies pertinent to ceramic analysis and ceramic petrography are outlined, and the
selection criteria and preparation of ceramic samples is discussed.
Chapter Six contains descriptions of the different sources of ceramic material analyzed as
part of the research project and provides descriptions and maps of all visited and collected
archaeological sites.
The results of the spatial analysis of prehistoric site distribution in the Körös River study
area are presented in Chapter Seven, and compare, contrast, and combine the recently collected
data with Sherratt’s (1997a, 1997b) data. Chapter Eight presents the results of both the
macroscopic and petrographic ceramic analyses and discusses the patterns of variability
12
identified over time between the Middle Copper Age, Late Copper Age, and Early Bronze Age
and variability over space at different geographic resolutions.
The discussion in Chapter Nine addresses the implications of the results presented in
Chapters Seven and Eight and for the regionally specific archaeological models presented in
Chapter Three. Furthermore, the implications of the results for wider anthropological theories of
the development of regionally homogeneous material culture groups and their local signatures
are discussed, as are the potential roles of migration, diffusion, invasion, and indigenous
development in the appearance of such ubiquitous culture groups.
Chapter Ten concludes the dissertation, and future research directions for the topic,
region, and time period under study are discussed.
13
CHAPTER TWO
THEORETICAL BACKGROUND
Introduction
Throughout prehistory there are numerous examples of ceramic types that appear
relatively quickly and are remarkably homogeneous across a macro-regional landscape. Cultural
activities such as agriculture, animal domestication, and settlement patterns also have been
shown to appear suddenly across large areas. This phenomenon of regional homogenization is
by no means monolithic, however, and occurs as part of a variety of processes and at different
scales (e.g., continentally, regionally, or sub-regionally). Over the last century, archaeologists
have investigated material culture and cultural homogenization in many locations employing
various explanatory models. The application of such models, and their archaeological correlates,
is especially appropriate in an archaeological setting such as the Körös region of the Great
Hungarian Plain, where settlement and material culture shifts occurred dramatically and
sometimes quite suddenly.
In this chapter, I frame the phenomenon of regionally homogeneous material culture in
the context of migration models. These have served as a primary explanation for the appearance
of homogeneous material cultures, and it is migration that this dissertation most explicitly aims
to test. I discuss issues of migration and the development of regionally homogeneous material
culture in a cross-cultural and anthropological context. I first summarize the theoretical
underpinnings of migration, diffusion, and indigenous development models for explaining
changes in settlement patterns and material culture. This foundation is necessary for
understanding the role that migration may have played in prehistory, as people or populations
moved from one region to another carrying both material and ideological templates for things
such as pottery, settlement structure, households, and religion. An event of this nature would
have caused fundamental social change throughout an arrival region, and can account for both
material culture change and the rapid expansion of elements of material culture such as pottery
decoration and form.
I also present overviews of two other anthropological processes and their archaeological
correlates that have been used to explain the appearance and development of regionally
homogeneous material culture groups. First, I discuss the paradigm of interaction spheres,
framed in the contexts of the appearance of the Hopewell elite material culture in North America
14
and in the development of lowland Maya culture in Mesoamerica. This paradigm has
implications not just for the interpretation of migration models, but also for the development and
application of migration and diffusion models, as well as the appearance, development, and
spread of institutionalized hierarchy. I subsequently provide several archaeological and
historical case studies from around the world in order to illustrate various explanations and
models for migration and material culture homogenization.
In addition to regionally homogeneous material cultures, migration and diffusion models
have been linked to the appearance of institutionalized hierarchy in some areas of the world.
This is true of the late prehistoric period on the Hungarian Plain, when the lengthy trend of
egalitarian chiefdom level societies gave way to an institutionalized hierarchical social system
during the Bronze Age, that some have linked temporally to the arrival of migratory populations
during the Late Copper Age (see Gimbutas 1977, 1979, 1980; Sherratt 1997b). As such, I
provide a background of social development and temporal durability of egalitarian societies,
since the application of these anthropological principles in this dissertation relates directly to
societies prior to, and on the cusp of, the development of institutionalized hierarchy. I close with
a summary discussion of how the topics discussed in this chapter provide a solid theoretical
foundation for the archaeological background in Chapter Three, and the research presented in
subsequent chapters.
Migration and Archaeology: Beyond the Normative Approach
Migration has long been used as an explanation for the appearance of new material
cultures across regions, both occupied and previously unoccupied. It also has been one of the
primary models for explaining material culture change in Central and Eastern Europe as a whole,
and specifically the Great Hungarian Plain, at least since the early 20th century (Childe 1950,
1959). In terms of this dissertation project, migration is one of several possibilities put forth by
researchers in various contexts to explain the settlement and material culture changes observed at
the end of the Middle Copper Age and beginning of the Late Copper Age on the Great
Hungarian Plain (Anthony 1990; Gimbutas 1963, 1977; Sherratt 1997a, 1997b).
The question “what causes migrations?” is perhaps one of the most vexing that
researchers face on this subject. Anthony (1990:898) stated that the causes of migratory
movements are often so complex that in many cases the proximate causes of migrations can no
15
longer be identified by what remains in the archaeological record, and that finding an explicit
cause for a migration should not be the primary focus for archaeologists. Nevertheless,
numerous models for identifying the ultimate causes of migration have been discussed over the
last century. Negative push and positive pull factors at both the point of origin and the
destination are often cited as causes for migration, and resource supply and demand in both
locations can be included in models that attempt to explain migratory behavior. Lee (1966)
stated that migration is most likely to occur when there are negative stresses (such as warfare) in
the home region and positive pulls (such as an abundance of natural resources) in the destination
region. Lewis (1982) suggested that the push factors most often associated with long-distance
migration are primarily economic. Specifically, differences in economic opportunities (such as
trade or agricultural production) between regions are a predictable antecedent to migratory
movement. Kearney (1986) addressed a specific situation in which the push/pull model may be
applicable. Migration may occur, he suggested, when a dominant population exploits a
dependent population. In this case, the consequences of travel would, in almost all
circumstances, be fewer than remaining in the home region. However, reliable data to perform
in-depth analysis addressing all of these factors is often unavailable to archaeologists.
The early models of social change in central and eastern Europe were firmly rooted in the
normative theory of the day. This approach eschewed much in the way of linking method and
theory, and instead relied on what has become known in simple terms as “pots equals people”;
or, more descriptively, that objects recovered from the archaeological record directly reflect the
behavior and culture of the people that created them. Spatial and temporal divisions of material
culture were the primary means of determining cultural boundaries and change; therefore, any
similarities in material culture over space or time implied a cultural relationship or cultural
homogeneity. This perspective was summarized well by Willey and Phillips (1958:18)
In strictly archaeological terms, the locality is a geographical space small
enough to permit the working assumption of complete cultural
homogeneity at any given time.
In an excellent criticism of the normative approach, Binford (1965:204) summarized how
anthropologists and archaeologists of the time approached the issues of formal variation in
material culture, material culture homogeneity, social boundaries, and migration and diffusion:
16
Spatial discontinuities in the distribution of similar formal characteristics
are perceived as either the result of (1) natural barriers to social
intercourse, or (2) the presence of a value system which provides a
conservative psychological matrix that inhibits the acceptance of foreign
traits, or (3) the migration or intrusion into the area of new peoples who
disrupt the previous pattern of social intercourse.
In essence, any model of culture change from a normative perspective usually requires an
external catalyst, be it the removal of natural barriers (usually by technological advancement),
the adoption of a new value system, or the arrival of a foreign population. Unfortunately, these
kinds of changes are often difficult to directly observe in the archaeological record. Caldwell
(1958:1) overcame this problem by suggesting that, “other things being equal, changes in
material culture through time and space will tend to be regular,” thus implying that any sudden
or dramatic change in observable material culture indicates an invasion, migration, or other
process that cannot be directly observed in the archaeological record.
Although the above perspectives were straightforward and logical, a problem with these
early and mid-20th century approaches to diffusion and migration was, itself, the normative
approach to model building and a lack of integration between method and theory (Anthony
1990). Archaeological evidence for migration hypotheses was often elusive (Parkinson 2006b),
and usually relied on the assumption that any new patterns in the settlement system of a region or
sudden change in material culture indicated migration or invasion rather than indigenous
development or the result of some other social process or trajectory. Binford criticized the
normative approach and addressed the need to integrate method and theory when dealing with
migration and diffusion models. He stated that the influences and relationships that composed
the vocabulary of migration studies were analytically inadequate and lacked a methodology
amenable to testable hypotheses. Binford instead insisted that archaeologists must adopt a
“multivariate” approach to the study of change, rejecting “assumptions about units or the natural
‘packages’ in which culture occurs” (1965:204). These “natural packages” could include any
number of artifacts commonly found in association with one another in a specific region, but in
Central and Eastern Europe this concept can be most directly applied to ceramic types that are
often used to construct a timeframe for culture change and population arrivals.
Although migration models on the Great Hungarian Plain, for example, have hardly
ignored ceramic variability, one of the primary lines of evidence for migration in the region’s
17
archaeological record has traditionally been the rather sudden, widespread appearance of kurgan
burial tumuli around the end of the Middle Copper Age Bodrogkeresztúr period and the
appearance of the Late Copper Age Baden complex. Gimbutas (1963) most directly suggested
this link, and it has remained a model for change in the region for decades. Recent criticisms of
such normative models, however, have suggested that a more rigorous archaeological approach
to migration should rely less on potentially diffused cultural elements – such as kurgan burial
tumuli – and more on changes in observable characteristics of material culture (see Anthony
1990; Binford 1965).
Anthony (1990) addressed the need outlined by Binford for an integration of method and
theory in migration studies. A problem with earlier perspectives on migration and diffusion was
that, rather than being incorporated into models of change and social evolution, migration was
viewed as an external phenomenon that was a catalyst of, but not a part of, culture change.
Anthony provided several reasons why migration models have been unsatisfactory for explaining
culture change. First, migration is often incorrectly characterized as a one-way event. For
example, Rouse (1958, 1986) described migration as a linear process by which a migrating
population invades an inhabited territory and through violence or assimilation establishes
permanent residence. According to Anthony, this kind of event is extremely rare. He agreed
with Gmelch (1980) who described migration as a two-way process involving return migrations
and the exchange of information between individuals in the frontier and home areas. As such,
this “counterstream” of return migration should have archaeological and material consequences
in the homeland (Lee 1966).
Anthony then addressed the issue that many previous and contemporary researchers (e.g.,
Rouse 1958, 1986) have ignored modern migration studies or taken the perspective that modern
migrations are irrelevant to the study of prehistoric migration. Anthony stated that both
archaeologists and others interested in the study of migration have no reason to assume that
migrations in prehistory operated differently from recent and modern migrations (1990:898).
Indeed, for over a century researchers (see Ravenstein 1885, 1889) have observed regular,
repeated patterns in recent migrations with consequences for material culture that could be of
great interest to archaeologists modeling the process of prehistoric migrations.
Most significantly, though, Anthony (1990:898, 1992:174) illustrated the need to
examine structure before cause in regards to prehistoric migration. He stated:
18
Proximate causes of prehistoric migrations are probably lost forever – we
can only hope to identify structural conditions that made migration more
or less likely to occur. In addition, one cannot begin an analysis of
migration by attempting to identify the archaeological signature of a
migration event. It is only after the structure of the migration process is
understood that appropriate methods can be identified or developed to
detect its archaeological signature (1992:174).
Under this premise, Anthony constructed the most plausible reexamination of Marija Gimbutas’
migration model of the Yamnaya (a Eurasian Steppe culture group largely defined by the
construction of burial tumuli, or kurgans) into the Great Hungarian Plain since Andrew Sherratt’s
treatments in the previous decade (Sherratt 1983, 1984, 1997a, 1997b). Implicitly, Anthony
refuted Gimbutas’ Kurgan invasion model into the Plain, and her contention that the invasion of
warlike horse-riders from the Pontic Steppe effectively ended the Copper Age cultures in the
region (see Anthony 1990; Gimbutas 1970, 1977). Though he noted the coincidal disruption of
cultural trajectories and the appearance of kurgan burial tumuli at about 3,500 B.C. on the
Hungarian Plain, he explicitly stated that the nature of the interaction between intrusive kurgan
builders and the local population remains largely un-discussed and almost entirely
misunderstood, given the lack of archaeological evidence for interaction between the groups
(1990:908).
Anthony’s measured approach to the consideration and archaeological study of migration
is not without its critics, however. Chapman and Dolukhanov (1992) took issue with Anthony’s
premise that archaeologists often focus on the wrong questions regarding migration, especially
Anthony’s contention that archaeologists should identify the structure of a migration before they
try to identify its cause (Anthony 1992:174; Chapman and Dolukhanov 1992:170). The
identification of a migratory process would, of course, be ideal, but our knowledge of the
structure of prehistoric migrations is limited in the same sense as all archaeological research –
the majority of evidence has degraded or disappeared over time. Only very rarely do
archaeologists encounter a situation where the form of a migration, society, village, or hunting
encampment is accessible at the outset of research.
Interestingly, the expansion of the Yamnaya kurgan burial tradition and their appearance
on the Hungarian Plain may be one of the best examples of visible migration structure.
Ironically, though, an in-depth chronological understanding of how and when the kurgans came
to be across the landscape is still lacking, preventing a detailed understanding of the impact of
19
kurgan builders on the Plain. Without an understanding how the kurgans appeared across the
landscape, an understanding of the structure of the migration is precluded. Until the time and
resources can be devoted to understanding the development of the kurgans over time and space,
the precise nature of their builders’ effect on social organization across Europe will remain
misunderstood. And, the effect of the migration on the kurgan builders and on indigenous
populations will remain unclear. To apply this perspective widely, the understanding of a
migration’s structure and progression may well be one of the most complex cultural events to
approach archaeologically. Even when clear evidence of a migration is present – as in the
kurgan example – parsing the structure of a migration from its social and cultural effects is
difficult given the ever-present limitations of time and resources.
Migration as an Explanation: Sufficient Models for Material Culture Change?
The spread of different forms of material culture, and often the development of largescale regional homogeneity, tie each of the following case studies together. As shown by the
brief summaries and critiques of migration models above, migration as a solid explanation for
the appearance of new personnel in a previously unoccupied region, the spread of domesticated
plants and animals, or the appearance of a new material culture tradition or artifact type in a
region remains a debated topic in archaeology. In all cases presented here, a debate has ensued
regarding the indigenous or foreign nature of dramatic changes observed in the archaeological
record, and whether or not a migration scenario satisfactorily accounts for these changes.
Though bolstered by archaeological evidence, migration models in and of themselves are rarely
sufficient for explaining such changes, as they tend to minimize or altogether ignore the
contribution of indigenous populations to the development of (or integration into) wider material
culture complexes. For example, Anthony (1986) soundly criticized Gimbutas for the blurring of
variability in the Yamnaya horizon (which Gimbutas called Kurgan I-II in her 1970 publication).
Nonetheless, he emphasized that migration must be dealt with effectively, since they are known
historically, and should be carefully modeled rather than used as a simple explanatory
mechanism.
Anthony’s (1992:174) basic theoretical premise, that uniformitarian models are necessary
to understand the archaeological record, applies as much to hypotheses of migration as to any
other archaeological problem. Uniformitarian models of migration are possible, though difficult,
20
to create. Migration’s tendency to be a patterned, structured behavior that reacts to localized
patterns both in the home region and the region of destination makes it difficult to both recognize
and model accurately. As such, archaeologists must identify and understand the structure of a
migration before the creation or identification of methods to detect its archaeological structure.
An underlying principle of this perspective is that cultures do not migrate – people migrate in
defined subgroups with specific goals (Anthony 1990, 1992). One can therefore hypothetically
observe the structured, patterned movement of these groups in the archaeological record. Long
distance migration, for example, should result in the development of long-distance networks set
up in order to gather information regarding scattered resources of the type exploited by the
migratory population along the migration route and at the destination. These networks would
form along migration streams, or essentially established migratory highways. These streams
should result in artifact distributions that follow a specific line of movement. Unfortunately,
these streams would still be ephemeral and therefore difficult to observe archaeologically. In the
destination region, change and innovation may lead to a sort of founders effect, resulting in rapid
stylistic change from what may have initially been a restricted pool of variability (Anthony
1990). The heart of this approach to migration is a relatively simple idea that contradicts many
earlier approaches to the topic: migration is a process, not an event. As the migration process
unfolds, it creates its own unique patterns and dynamics.
There is no doubt that Anthony’s approach to the study of migration in the archaeological
record provides archaeologists with a new and innovative framework for the study of migration
and culture change. However, there are practical problems with Anthony’s framework, some of
which he addressed himself (see Chapman and Dolukhanov 1992). For example, for one to
understand the structure of a migration one must rely on a combination of two lines of evidence:
the archaeological record as it is, which is often inadequate for understanding the structure and
causes of migration, and historical or ethno-historical records that may have little or no
relationship to the migration under study. However, in the case of Eastern Europe – where
migration models have never been fully eclipsed – new approaches to old migration ideas can
provide insight to the processes that underlie social and cultural change.
21
Interaction Spheres, Egalitarianism, and Social Change
Though the research presented in this dissertation is primarily concerned with testing the
development of regionally homogeneous material cultures within the contexts of migration and
indigenous development models, it should be recognized that social and economic interaction
often play a direct role in material culture and settlement changes. Especially deserving of
consideration in this dissertation are the incorporation of an indigenous population into a wider
material culture group through involvement in spheres of interaction and exchange, and a
discussion of how interaction models have been applied to their archaeological correlates to
explain the transition from egalitarian to hierarchical social organization. According to the
models presented below, the emergence of social complexity is a gruadual process, with
interaction solidifying economic and social relationships, leading to the appearance of regionally
homogeneous material culture areas.
Beyond Migration: Interaction Spheres and the Spread of Material Culture
Caldwell (1966) initially developed the concept of the interaction sphere in order to deal
with the wide geographic distribution of the North American Hopewell complex, which
consisted of a regionally homogeneous complex of elite material culture. The Hopewell
complex crosscut the traditionally defined culture area boundaries, including discernable
archaeological cultures delineated by non-elite material culture (such as pottery and stone tools)
(Freidel 1979). In contrast to other models of the homogeneous elite Hopewell material culture,
Caldwell (1964:141) interpreted interaction between diverse sociocultural and sociopolitical
groups as both beneficial and as formative in the development of elite institutions and an
overarching institutionalized hierarchy amongst the participating groups. Under the interaction
sphere model, the development of large-scale, homogeneous elite social institutions is caused by
a network of information and exchange among participating elites, rather than by localized
conditions as modeled under the culture area paradigm. Binford enthusiastically approved of
this approach to cultural development and change in the early 1970s (1972:204). Freidel
summarized the interaction sphere paradigm by stating:
Theoretically, the interaction sphere concept envisions the initial emergence of
elites as a means of distributing scarce and vital resources between local areas. In
contrast to [cultural ecological models], the interaction sphere postulates that the
initial economic monopolization by elites was over the distribution of raw
materials and finished products rather than over the actual means of production
22
(such as arable land). In short, this concept postulates that elite institutions arose
largely through the interaction of local communities rather than as an adaptation
by them to local conditions (1979:50).
Freidel (1979) used the concept of the interaction sphere to explain the origin and
evolution of lowland Maya civilization in Mesoamerica. He expressed skepticism of previous
models of Maya development based upon the culture area concept that presumed that
sociocultural innovation occurred as a localized response to local natural and social conditions.
Rather, much like Caldwell (1964), Freidel sought to emphasize interaction and exchange
between groups of elites across large geographic regions. These regional networks would have
developed from small-scale local networks and ultimately through down-the-line long-distance
exchange networks (Freidel 1979; Renfrew 1975).
Blanton (1976), working from Caldwell (1964), stated that the development of an
extended interaction sphere requires only a degree of sedentism amongst the participating
groups, and Flannery and Schoenwetter (1970) stressed the importance of economic
interdependence between the participants. However, Freidel (1979:51) suggested that these
criteria are not sufficient conditions for the development of interaction spheres, and that a
systemic change in the use of nonlocal materials involving their use as prestige items is the
crucial element. So, although Caldwell initially conceptualized the interaction sphere model as
more ideological than economic, the model has been utilized successfully as a framework for
both economy and political economies.
More recently, other researchers have operationalized the interaction sphere concept in
different contexts around the world to explain cultural change and to link such changes to the
development of hierarchical social organization and the appearance of state level societies
(Trigger 1989:331). In Mesopotamia, civilization was modeled as a large zone in which many
cultures influenced each other’s development through political and social interaction (see Alden
1982; Kohl 1978; Lamberg-Karlovsky 1975). Renfrew and Shennan (1982), Cherry (1984), and
Renfrew and Cherry (1986) have discussed “peer-polity interaction” in the Aegean region of
prehistoric Europe. Recently, the peer-polity model has been expanded to include models of
interaction between smaller polities and larger, established primary states (Parkinson and Galaty
2007). Blanton et al. (1981) and Trigger (1989:331) have suggested that the development of a
singular region cannot be understood without considering the developmental trajectories of its
23
neighbors. Flannery (2002) adopted this perspective. He proposed a model by which extended
households developed in competition with one another in the social context of large villages and
competition over natural resources.
Social Trajectory, Interaction, and Change
In addition to migration, material culture changes, and shifts in settlement patterns,
sudden breaks in continuity can also be explained through social and economic models focusing
on social complexity and factors like hereditary inequality, division of labor, subsistence, trade,
wealth, and other aspects of economy. Speaking generally, archaeologists have tended to model
such changes on a linear scale ranging from less socially complex egalitarian societies to more
complex, hierarchical and ranked state-level societies. However, in some regions, prehistoric
societies did not progress according to such models.
The prehistory of the Great Hungarian Plain, and of Central and Eastern Europe, for
example, is exceptional for the temporal durability exhibited by tribal societies during the
Neolithic, Copper Age, and Bronze Age. In contrast to other parts of the Old World, such as the
Eastern Mediterranean where institutionalized hereditary inequality and ascribed ranking
emerged during the Early and Middle Bronze Age, the cyclical nature of settlement variability on
the Great Hungarian Plain suggests that basically egalitarian societies existed in the region until
well into the Bronze Age (Galaty and Parkinson 1999). By that time, socially stratified states
had emerged throughout the Eastern Mediterranean (Renfrew 1975).
The Emergence of Ranked Societies and Ranking’s Relationship to Regional Homogeneity
Implicit in the migration model of Gimbutas (1977, 1980) is the arrival of Indo-European
populations on the Hungarian Plain. Essentially, she considered the migratory arrival of IndoEuropean peoples to be the catalyst that destroyed the cyclical social structure of “Old Europe”
on the Plain and installed the patriarchal hierarchy that she considered typical of later periods.
Importantly, Gimbutas considered the whole of the regionally homogeneous Baden culture to
embody Indo-European characteristics, including the social and political structures that
contributed to the development of institutionalized inequality on the Great Hungarian Plain
(Gimbutas 1977, 1980). As such, an overview of theoretical perspectives on the emergence of
ranked society and its signature in the archaeological record is useful. It is presented below in an
effort to clarify possible processes present on the prehistoric Hungarian Plain.
24
The emergence of ranked societies is a topic widely considered in archaeology since the
social evolutionists of the late 19th century first proposed levels of human civilization (e.g.,
Morgan 1877; Tylor 1871). Since that time, archaeologists have worked primarily from two
perspectives in attempting to understand and describe the development of social inequality:
1. Social anthropological models that focus on how an egalitarian system changes into a nonegalitarian system, either suddenly or over time. These models describe how societies move
from a “created equal” social reality to one where certain individuals are born into exclusive
social ranks.
2. Archaeological models, built upon social anthropological models, that draw upon data
collected from the archaeological contexts to determine where a society fits in a developmental
model, or try to describe what a certain level of development (e.g., band, tribe, chiefdom) should
look like archaeologically.
After the advent of the New Archaeology in the 1950s and 1960s and the subsequent
development of processualist archaeological theory and methodological approaches, numerous
models for the development of inequality were proposed. Ultimately, all of these models were
built upon the “levels of integration” initially proposed by Steward (1955) and developed further
by others. Freid (1967) wrote that inequality developed as the result of unequal access to roles
and resources, and that it was ultimately concerned with political organization. He therefore
divided social organization into three main categories:
1. Egalitarian societies, which have as many roles to fill as there are individuals.
2. Ranked societies, which may contain achieved and ascribed positions, with limited access to
certain social roles.
3. Stratified societies, where an entire subsection of society has no ability to participate in
certain roles.
Although they were also concerned with the hierarchical roles of individuals in societies,
Sahlins and Service (Sahlins and Service 1960; Service 1971) were more concerned with
functional organization. They differentiated non-egalitarian chiefdoms from egalitarian tribes
based on the level of centralized coordination of economic, religious, and social activities. They
assigned these organizational characteristics to specific social integration levels, creating the
25
well-known scale of bands, tribes, chiefdoms, and states that has remained a popular framework
for archaeologists concerned with levels of social integration as well as the development of
institutionalized hierarchy.
Though useful, all of these models are problematic in that they do not account for
variability at levels of social integration geographically or diachronically. Along with their
tendency to pigeonhole societies into one of several normalist categories, the models fail to
describe how a society moves from one level to another on the scale. Regardless of these
shortcomings, such models retain utility in cross-cultural comparative studies (as the authors
intended), though they remain troublesome for archaeologists concerned with development over
the long-term.
More recent models (Clark and Blake 1994; Hayden 1995) emphasized human agency as
a tool in social change, and have described Service’s “tribe” and “chiefdom” levels of social
integration as “transegalitarian” societies. That is, they are neither egalitarian nor politically
stratified. This term seems particularly appropriate for the Great Hungarian Plain in the
Neolithic and Copper Age (Bogucki 1999), as the roots of inequality had been present on the
Plain and in the region for quite some time. Full development of ranked societies may have been
held in check by the cycles of population nucleation and dispersal (Parkinson 2002), which often
serve as markers between culture periods on the Plain throughout this period.
Parkinson and Guycha (Parkinson 2002:8; Parkinson and Gyucha 2007), Gyucha et al.
(2004), and Fowles (2002) argued that, in order to understand the long-term nature of change in
egailitarian and transegalitarian societies and to model integration over the long-term,
archaeologists should focus on segmentation in tribal systems. Essentially, Parkinson and
Gyucha suggested that it is most productive to envision different economic, environmental, and
social mechanisms that encourage fusion among tribal societies at some times, and fission at
others. This speaks to the concept of tribal disunity, and the idea that tribes remained segmented
with clear social boundaries because they were always fissioning (Morgan 1885; Parkinson
2002:8). Parkinson stated:
While some tribal societies certainly do exhibit clear boundaries, others appear as
smears across the archaeological landscape, with few discernible internal or
external boundaries. The segmented nature of tribal systems, combined with their
tendency to fission and fuse given different social and environmental conditions,
26
results in a social picture that assumes discreet boundaries at only isolated
moments in time (2002:8).
Such an approach to understanding the behavior of tribal societies over the long-term allows
archaeologists to speak on how non-ranked, egalitarian societies such as those that existed on the
Great Hungarian Plain from the Neolithic to the Bronze Age maintained tribal structures long
after contemporary and comparable societies in other parts of the world developed political and
social ranking and institutionalized hierarchy, up to, and including, the emergence of state level
societies.
Approaching Migration, Archaeology, and Regional Models of Material Culture Change
The linking of material culture to specific populations – especially migratory ones – is
difficult archaeologically, and as such more general anthropological models are relied upon to
bridge the gaps between migration, models of cultural change, and the archaeological evidence.
Migration as an Explanation for the Appearance of Homogeneous Regional Cultures
Despite the fact that direct archaeological evidence supporting migration explanations is
usually elusive, research across space and time has often attempted – with varying degrees of
success – to apply migration models to explain instances of new materials or products appearing
in a region. What follows are summaries of several scenarios at varying spatial scales where
migration has been used as a framework for explaining archaeological evidence. These
archaeological and historical examples are included in order to provide specific outcomes of the
anthropological processes presented above.
Clovis in North America. Clovis finds in North America represent the first clear evidence
of late Pleistocene people on the continent around 11,500 years ago (Anderson 1990, Haynes
2002:52). Sites with characteristic fluted points have been discovered across the continent, from
Central America to the Maritime Provinces of Canada on the Northeast Atlantic coast. Although
the highly contested issue of Clovis vs. pre-Clovis, and the possibility of human occupation in
North America prior to the appearance of the Clovis assemblage, is far beyond the scope of this
dissertation, what is pertinent is the relatively rapid appearance of a materially homogeneous
culture across a large geographic area.
27
Numerous general models have been proposed for the migration of Beringian groups into
North America. Typically, these have considered north-south migrations in single or multi-wave
events either internally or along the Pacific Coast, ultimately culminating in Clovis either as they
initially entered North America or after a period of change within the continent (Barton et al.
2004; Faught 2008; Goebel et al. 2008; Madsen 2004). Other researchers, drawing on
discoveries of pre-Clovis assemblages and their interpretations, have argued for a purely in situ
spread of Clovis based on diversity in pre-Clovis assemblages (see Adovasio et al. 1983). Under
this model, it is presumed that the Clovis assemblage and characteristic fluting technique
developed within North America and spread organically as other groups rapidly adopted the tool
making tradition.
Recently, Faught (2008) argued that Clovis might have spread throughout North America
as a ménage of human groups, possibly migrating onto the continent via different routes, arrived
in seperate regions almost contemporaneously. He cited as evidence a large number of Clovis
radiocarbon dates from across both North and South America, with the earliest populations
having appeared just before and after 12,000 years ago in four different regions almost
simultaneously. Ultimately, he argued that there is no clear north-south migratory trajectory, and
indeed, no clear trajectory is discernable at all based on the analyzed radiocarbon dates. This
interpretation dovetails well with Anthony’s (1990) insistence that archaeologists must integrate
method and theory in order to understand migration as a patterned human behavior, often with
multiple points of origin and destinations.
The Archaic North American Borderlands. At the regional scale in North America, the
spread of various cultigens to the North American Desert Borderlands during the Archaic Period
(3,000-1,500 B.C.) serves as an example of a previously unknown technology appearing
relatively rapidly across a region. The “Desert Borderlands,” as described by Minnis (1992),
consist of the Mogollon Highlands and the Colorado Plateau to the northern border of the
Sonoran Desert of the southwestern United States and northern Mexico. Western North America
was beyond the area of natural distribution of classic Mesoamerican domesticates such as maize
and beans. Maize had to be brought from its subtropical origins to a more temperate climate,
where it eventually became the cornerstone of a field crop subsistence system (Bogucki 1999).
Some archaeologists (Berry 1992; Matson 1991) have argued for a northward migration of
peoples from Mesoamerica, using the appearance of maize in the region of the Desert
28
Borderlands as evidence for human movement. As part of a migration model, new arrivals from
the south would have brought with them their agricultural technology, and with it social and
settlement changes that often accompany agricultural development – such as sedentism and large
field crop systems, that appear to have developed at around the same time as the appearance of
Mesoamerican domesticates.
However, arguments for the indigenous adoption of maize, squash, and beans, and the
contention that the spread of cultigens can spread across long distances without the direct
migration of people are also compelling (see Minnis 1992; Wills 1995). Minnis (1992) noted
that Mesoamerican domesticates appear to have been integrated effortlessly into the preexisting
subsistence structure of the Archaic people of the region. He characterized the period between
the first appearance of maize and the emergence of communities fully based on agricultural
subsistence as a time of “casual agriculture,” when a semi-reliance on domesticates was a low
cost and low effort way in which to increase economic stability. Sedentism would have
developed slowly as a logical extension of the incipient agricultural system, rather than suddenly
as the migration models imply. This scenario does not imply or require migration, and it
suggests that the inherent social organization in the region was maintained rather than
fundamentally changed; in this model, casual agriculture simply complemented rather than coopted the system already in place. Willis (1992, 1995) also noted increased sedentism following
the initial adoption of maize and other domesticates. He similarly suggested that a desire for
predictability and control of subsistence systems – rather than an influx of new personnel into the
region – led to a gradual shift in settlement and economic system.
The European Magdalenian. In Europe, the migration concept has also been widely used
to explain the appearance of new material cultures and, presumably, the arrival of new
populations on the continent. For example, the arrival of new personnel is a logical framework
for explaining the dramatic increase in Magdalenian settlement in regards to an expansion of
settlement to the north and east of Europe over time (Jochim et al. 1999; Otte 1998). In the
terminal period of the last glacial maximum in France, Magdalenian site numbers increased
nearly four-fold, and the number of open-air sites (as opposed to rock shelter sites) increased
dramatically. It appears that the dispersal of peoples into this region began before 15,000 B.P.,
just as climatic conditions began to improve in the northern latitudes of Europe. Numerous
small, seasonal camps that exploited seasonal reindeer migration patterns appeared. By the
29
Bölling warm period (around 13,000 B.P.) just before the last glacial maximum, sites begin to
appear further north. After the glacial maximum between 13,000 and 11,000 B.P., the
Magdalenian had spread throughout much of continental Europe, and occupation was year-round
(Jochim 2002). In this case, improving environmental and climatic conditions opened a niche
and afforded Paleolithic populations the opportunity to expand into a previously unoccupied
region, possibly following faunal resources as they migrated into recently habitable areas.
The Neolithization of Europe. The Neolithic period is Europe is typically described as
the period when farming first replaced hunting and foraging as the primary method of
subsistence. The period is also often described by other specific characteristics, such as
permanent settlements and structures, the first widespread creation and use of pottery, and the
first domestication of various animal species. Throughout southeastern and central Europe, a
common set of subsistence, settlement, and material culture attributes existed at this time,
making the period an excellent case study for the examination of migration and the appearance
of regionally homogeneous cultural and social characteristics.
Numerous models have been proposed and debated regarding the first appearance of
agriculture in Europe. Childe (1958a, 1958b) established the tradition of diffusionist arguments
to explain the transition from hunting and foraging subsistence strategies to agricultural ones.
Many models have evolved from Childe’s original assumptions, perhaps most notably Renfrew’s
(1987) assertion that the spread of framing from Anatolia to Europe accompanied the spread of
the Indo-European language family. Geneticits researching modern European populations have
in part substantiated this model (Ammerman and Cavailli-Sforza 1984; van Andel and Runnels
1995). However, data regarding the spread of the Indo-European language family widely across
the European continent are still contested (see Forster and Renfrew 2006), and other researchers
(Anthony 2007; Kristiansen 2005) have supported a much later entrance of the Indo-European
family to the continent based on both archaeological and linguistic evidence. Many
archaeologists have also emphasized the importance of indigenous Mesolithic developments of
farming in Europe (see Chapman 1994), while Bogucki (1996) took a more inclusive approach
by insisting that no single method or mechanism can explain the appearance and development of
the Neolithic in Europe.
Regardless of the plurality of mechanisms for the spread of the Neolithic in Europe, a
rapid increase in site number can be observed in Greece around 7,000 B.C. that corresponds to
30
the first definite appearance of agriculture on the European continent (Whittle 1996:22-23, 4041). Based on radiocarbon evidence, a Neolithic presence existed in Macedonia and Crete from
ca. 7,000 B.C. and in Thessaly by 6,500 B.C. Sites like Karanovo in Bulgaria existed by 6,000
B.C., and there is an early Neolithic presence in Hungary on the Great Hungarian Plain, in
another area with comparatively little Mesolithic human occupation. However, the period of
time immediately preceding the Neolithic on the Hungarian Plain still requires much more
research to clarify the relationship between the Mesolithic and the Early Neolithic (Kertéz 1996;
Makkay 1996).
It is difficult to imagine a scenario where a migration explanation is more fitting than the
spread of the Neolithic package into Europe, at least in terms of the spread of agriculture from
western Anatolia into Greece and north to the Carpathian Basin. Aside from the first appearance
of ceramics at Neolithic sites in Europe, nearly all Neolithic European sites contain plants and
animals first domesticated in the Levantine region of the Near East, including barley, emmer,
einkorn, lentil, pigs, goats, and sheep. Furthermore, recent genetic data suggests that no
independent domestication of animals took place in Neolithic Europe (Bellwood 2004:68-69).
Renfrew (1987) initially characterized this process as a “wave of advance” by demic diffusion
from Anatolia into southern and central/eastern Europe, creating a spread of Neolithic
settlements engaged in agricultural subsistence and the production of domesticated animals.
Renfrew tied his migration model directly to the spread of the Proto-Indo-European language
family (PIE), which led to criticisms from those supporting a much later entrance of PIE
concurrent with the arrival of kurgan builders from the Pontic Steppe (see Anthony 1990;
Kristiansen 2005). Renfrew later revised his position on the arrival of PIE in Europe, and
suggested that migratory populations from Anatolia into Europe around 7,000 BC spoke an even
more ancient version of PIE (2003). However, the wave of advance migration scenario remained
relatively unchanged.
Recent publications have followed Bogucki’s (1996) lead by de-emphasizing migration
as the single explanation for the appearance of agriculture and domesticated animals in Europe.
Séfériadès (2007), for example, emphasized the indigenous adoption of Neolithic lifeways. He
suggested that the Aegean area was densely populated in the Mesolithic, and that the diffusionist
understanding of the region prior to the Neolithic is shaped by a lack of exploration and research
into the region’s Mesolithic roots. Though the issue remains unsettled and open to discussion, it
31
serves as an excellent example of how the discussion of how material cultures spread and
become dominant in a region or continent is still framed within paradigms of migration and
indigenous development.
The Iron Age Celtic Migrations. During the Iron Age, the period of the great Celtic
migrations is documented in texts by Greek and Roman authors (Kristiansen 1998; Moscati et al.
1991). They describe movements of peoples (called Celts by the Greeks and Gauls by Romans)
from north of the Alps. As might be expected, much discussion and debate has taken place
regarding the authenticity of the written accounts, and whether or not archaeological evidence
supports these accounts (Wells 2002). Roman tradition describes migration of the Gauls
southward across the Alps into Italy between the 6th and 4th centuries B.C. According to some
historic sources, Gauls passed through the alpine passes and descended into the Po River plain.
Some remained in the Po region as agriculturalists, while others continued southward, ultimately
defeating a Roman army in 387 B.C. (Frey 1995; Wells 2002).
The archaeological evidence in Italy supports the idea of movements from the north of
the Alps into Italy. However, migrations on the scale described by Classical writers are not
supported (Frey 1995). Archaeological evidence for the migrations includes new burial practices
at this time along the Adriatic coast of Italy, and the presence and style of certain grave goods
that suggest connections with eastern France and southern Germany. One problem has been that
the texts of Classical writers typically describe large scale, one-way migrations. This is
problematic, as there is no historical evidence, textural or otherwise, to suggest the true scale of
the migrations. Anthony (1990, 1992) and others have recently indicated that return migration is
a common and perhaps necessary part of large-scale migrations. The combination of
archaeological and historical evidence of settlement in the Po valley and archaeological evidence
contained in graves supports the idea that return migration and continued contact with the
homeland through maintained economic ties is an important and perhaps necessary part of
human migration.
Migration, Regionally Homogeneous Material Cultures, and Social Change
The mosaic of anthropological models and their archaeological correlates for explaining
the development of regionally homogeneous material culture groups presented here is complex,
both in terms of theoretical approaches and the models developed to explain archaeological
32
phenomena. However, each of the models and theoretical perspectives presented above –
migration and diffusion, interaction spheres, and the development of hierarchical social structure,
are intertwined both anthropologically and theoretically. It is fitting, then, that the research
presented in this dissertation draws upon all of the material presented in this chapter in order to
frame the discussion in the following chapters.
To conclude this chapter, it is important to consider the multiple models described above
as contributing factors to the development of regionally homogeneous material cultures. Even
more, it is imperative to recognize them as not mutually exclusive. Indeed, these models often
reference one another even unintentionally – migration has been said to usher in the appearance
of hierarchy (see Gimbutas 1970), and interaction spheres have contributed to the appearance of
institutionalized hierarchy (see Sherratt 1997a, 1997b) as well as the development of state level
society (see Renfrew and Shennan 1982; Renfrew and Cherry 1986). It is perhaps ineffective to
discuss one without incorporating other models into the fold, and it is the aim of the remainder of
this volume to present research that cohesively integrates these multiple theoretical paradigms.
Summary
This chapter has outlined the general anthropological theoretical considerations
underpinning the archaeological research in subsequent chapters that focuses more specifically
on issues of material culture homogenization, migration, interaction spheres, and social evolution
on the eastern Great Hungarian Plain. The chapter has discussed various models used by
archaeologists to explain migration and the formation of homogeneous regional material
cultures, and provided descriptions and case studies to illustrate these models and place them in
their proper theoretical contexts. Chapter Three will apply these theoretical underpinnings to
the archaeological background of the Great Hungarian Plain, paying special attention to how
models of migration, diffusion, indigenous development, and settlement and material culture
change have contributed to our understanding of the region’s prehistory.
33
CHAPTER THREE
THE ARCHAEOLOGICAL, GEOGRAPHIC, AND GEOLOGICAL SETTING
Introduction
In this chapter, I provide the relevant background to the prehistory of the Great
Hungarian Plain by placing the region into its archaeological context and integrating the
theoretical foundation presented in Chapter Two into the specific archaeological context of the
Plain. I also provide a geographic and geological summary of the Plain in general and of the
Körös River watershed study area more specifically.
The chapter begins with a geographic, geological, and geomorphological overview of the
study region, and a summary discussion of how ecological conditions affected prehistoric
settlement patterns in the area. I continue with a period-by-period overview of social
development and change and settlement organization, beginning with the material culture and
settlement organization shifts of the Late Neolithic/Early Copper Age transition. I then provide a
more detailed description of Early and Middle Copper Age trends toward local and regional
homogenization in terms of settlement and material culture. This is followed by a discussion of
the Plain’s Late Copper Age incorporation into a much wider material culture group and
economic system. I conclude with a brief description of the region’s trend toward regional
differentiation during the Bronze Age. I pay special attention to and describe in more detail the
development of the Late Copper Age Baden culture in a wider European context, and then
discuss practical limitations of characterizing Baden development in the Körös region as
compared to other areas in central and eastern Europe and the Balkans.
The Geographical and Geological Setting of the Great Hungarian Plain
The Geological Setting of the Great Hungarian Plain
The geographical, geological, and geomorphological setting of the Great Hungarian
Plain, and its relationship to prehistoric settlement, has been analyzed and described in depth in
several other publications (see Frolking, unpublished manuscript; Gyucha 2010; Parkinson 1999,
2002; Sherratt 1983, 1984, 1997a, 1997b). As such, only an overview will be provided here in
order to provide a general understanding of how the local geology and hydrology developed
34
Tisza River
Danube
River
Transdanubia
Körös-Berettyó
River System
Carpathian
Mountains
Great Hungarian Plain
100 km
Figure 3.1. The Carpathian Basin. Map used with permission of Dr. László Zentai, Eötvös Loránd University,
Budapest.
and subsequently influenced prehistoric settlement and changes over time in the Körös Basin
study region.
The modern country of Hungary and the Carpathian Basin are bisected by the Danube
River. This forms two generalized and distinct landscapes in the Basin (see Figure 3.1).
To the west, the area known as Transdanubia (Dunantúl) is characterized by hills and mountains.
East of the Danube, the Great and Little Plains (Nagy and Kis Alföld, respectively), are
characterized by flat, alluvial Quaternary deposits. During the Pliocene, up to 3,000 meters of
sandy-clay sediments were deposited as the Pannonian Sea, an inland body of water that once
covered the entirety of the Carpathian Basin, subsided, to form a series of freshwater lakes and
eventually rivers that continued to deposit more sediment above the Pannonian deposits.
Moreover, the rivers that became the Danube and Tisza incised and filled the Carpathian Basin
with fluvial sediment eroded from the uplands surrounding the Nagy and Kis Alföld (Pécsi 1964).
Ultimately, the landscape encountered by prehistoric inhabitants of the Great Hungarian Plain –
and indeed, modern people as well – was an essentially flat landscape dominated by alluvial clay
35
and loess with seasonal flooding, slow-moving and temporary waterways. It is a land entirely
devoid of naturally occurring stone.
To the west of the Körös region lies a sandy interfluve between the Tisza and Danube
Rivers (the Duna-Tisza Köze). This feature of the Great Hungarian Plain was formed as the
alluvial fan of the Danube was blown southeast, forming a series of dunes stretching to the Tisza
floodplain (Parkinson 2006b:97; Sherratt 1997a:274). This interfluvial area remained relatively
unoccupied for much of prehistory. This created a geographic and social boundary between the
Great Hungarian Plain to the east and Transdanubia to the west. However, by the Middle Copper
Age Bodrogkeresztúr period on the Plain this region was settled. This settlement may indicate a
widening economic and social reach of the eastern Plain’s people, and increased contact with the
Transdanubian Lengyel culture (Makkay 2007).
A series of alluvial fans, formed by smaller streams as they flowed from the mountains to
the Tisza during the Pleistocene and Holocene, separates the Tisza floodplain from the northern
mountains (see Sherratt 1997a:275, figure 11.3). The northeastern Nyirség Pleistocene alluvial
fan and the southeastern Maros alluvial fan formed in the Pleistocene and Holocene when a
series of braided rivers deposited thick layers of sand and sandy gravel in the region (Nádor et al.
2005; Parkinson 1999:97). Parkinson and others (Pécsi and Sárfalvi 1964; Pécsi 1970; Gyucha
2010) summarized the Körös-Berettyó region – an area of the study region in this research
project – that lies between these Pleistocene formations, as dominated by the Tisza River and its
thick fluviatile sediments. In prehistory, the Körös Region was dominated by the Tisza,
Berettyó, and Körös Rivers. These rivers were responsible for regional inundations throughout
much of the Holocene. Prior to the river system’s regulation in the 19th century, this part of the
country was a complex series of swamps and oxbows interspersed with slightly higher areas of
relief composed of alluvial silts and clays, and pockets of redeposited loess.
36
Berretyó River
Sebes -Körös R.
Hármas-Körös R.
Kettős –Körös R.
Figure 3.2. Rivers in the Körös-Berettyó study region after 19th century regulation.
The Geological Setting of the Körös-Berretyó Region
The Körös River valley in the Tisza drainage (also called the Körös-Berettyó
region), consists of four separate channels (The Berettyó, Sebes Körös, Fekete Körös, and
Feher Körös). The four rivers ultimately converge approximately 20 kilometers west of
Vésztő to form the Hármas-Körös (Triple Körös) River that flows into the Tisza near the
modern city of Szeged. This portion of the Tisza drainage was similarly drained and
canalized in the 19th century (see Figure 3.2). The triangular depression drained by the
Körös and Berettyó Rivers is a region of extremely low relief, and is generally defined by
three Pleistocene loess fans to the north, west, and south (Sherratt 1997b:295-296, figures
11.10-13). It must be noted that this slope is not noticeable at the micro level, as it
37
Figure 3.3. Prehistoric hydrology of the Körös region (recreated after Gyucha 2010).
changes in elevation only approximately 20 meters over 90 kilometers (Sherratt
1997b:295, figure 11.11). Given the lack of slope in the region, much of the Körös basin
consists of poorly drained floodplain and backswamp with a tangle of discontinuous
levee-like soil features (Frolking, n.d.; Pécsi 1970; Sherratt 1983, 1997b).
Prior to the regulation and canalization of the rivers in the Körös drainage in the
19th century, seasonal flooding left much of the region inundated for large periods of the
year rendering it unsuitable for permanent habitation (see Figure 3.3). Settlements
throughout prehistory, therefore, were limited to islands of higher ground within the
Körös depression, and to the tops of natural levees formed by alluvial materials deposited
by the spring and summer floods. Sherratt (1997b:297) stated that occasional exceptional
flooding reached even the highest levels of Holocene deposition, so that occupants of the
38
area sought the highest elevations available, even when the additional elevation offered
even less than a meter of extra protection.
Perhaps due to the frequent course changes as late as the 18th and 19th centuries, it
had previously been accepted that rivers in the Körös-Berettyó drainage frequently
changed beds in prehistory, in addition to flooding seasonally. Recent geomorphological
research in the region by Frolking (n.d.) and Gyucha (2010), however, suggest that the
river channels were remarkably stable for the vast majority of the Holocene, travelling
only three meters per 100 years with no meanders or cutoffs in some areas.
General Soil and Environmental Characteristics
Parkinson (1999:98; 2006b) described the soil formation in the Great Hungarian Plain as
very slow due to the frequent flooding of rivers in the Tisza and Körös drainages. Sherratt
(1997a:276-277) also described the soils and vegetation of the Great Hungarian Plain in more
detail. Although presently devoted to industrial agriculture, including the production of
sunflower, wheat, barley, corn, and rice, as well as the continuing tradition of stockbreeding
(Pécsi and Sárfalvi 1964), the region is technically the westernmost extension of the Eurasian
Steppes. Most of the Plain is naturally classified as forest steppe, though large areas are likely to
have remained unforested during the Holocene. In wet areas and depressions as are common in
the Körös region, however, riparian forests are likely to have been characteristic (Pécsi and
Jakucs 1971).
A synthesis of pollen core data (Gyulai 1993) that built upon the work of Szujkó-Lacza
(1991) suggested that the early Neolithic was characterized by mixed oak forests on loess soils.
After 5,000 B.C. at the Neolithic/Early Copper Age transition, a general cooling trend ushered in
a period of beech dominance, culminating in the creation of large deciduous forests as well as
parkland steppe areas toward the later phases of the Copper Age. This steppe was gradually
replaced during an extended cool period with a mixture of beech and oak forests during the
Bronze Age (Gyulai 1993:13-18; Parkinson 1999:99).
The Archaeological Setting
The prehistory of Central and Eastern Europe is exceptional for the temporal durability
exhibited by egalitarian tribal societies during the Neolithic, Copper Age, and Bronze Age. In
39
contrast to other parts of the Old World, such as the Eastern Mediterranean, where
institutionalized hereditary inequality and scribed ranking emerged during the Early and Middle
Bronze Age, the cycling and fission/fusion nature of settlement variability on the Great
Hungarian Plain suggests that basically egalitarian societies existed in the region until well into
the Bronze Age (Galaty and Parkinson 1999; Makkay 1982). By that time, socially stratified
states had emerged throughout the Eastern Mediterranean (Renfrew 1975).
On the Hungarian Plain, the process of cycling is most observable in the nucleation into
centralized tell-based society and economy in the Middle/Late Neolithic and Middle Bronze Age
and the dispersal into less centralized systems during the Early Copper Age. Although this
pattern was the dominant one on the Plain during this time period, another trajectory appears to
have taken hold in the Middle Copper Age Bodrogkeresztúr period and culminated in the Late
Copper Age Baden phase in what Sherratt (1984, 1997b) characterized as an abandonment of the
central Hungarian Plain and an emerging focus on settlement at the margins of the Plain near
access points to raw material sources and trade routes that provided prestige goods. This process
may have played a more fundamental role in leading to institutionalized and hereditary
inequality in the region in the second half of the Bronze Age than diffusion or migration. The
following sections explore long-term and short-term trajectories of change on the Hungarian
Plain from the Neolithic to the Middle Bronze Age.
The Neolithic
The Neolithic period on the Great Hungarian Plain is characterized by a trend toward an
increase in regional differentiation in ceramic styles, settlement patterning, and resource
exploitation. The trend began with the earliest arrival of farmers on the Plain in the Early
Neolithic – the Körös Culture (see Bökönyi 1988; Comşa 1974; Kalicz and Makkay 1977;
Kertész 1996; Kutzián 1944; Makkay 1996; Tringham 1971:91-96). It continued in the Middle
Neolithic with the distinction of Dunatúl Vonaldiszes Kerámia (DVK) in Transdanubia and
Alföldi Vonaldiszes Kerámia (AVK) on the Plain (Bognár-Kutzián 1966; Kalicz and Makkay
1977; Kosse 1979; Makkay 1982), and culminated in the highly differentiated Tisza-HerpályCsőszhalom complex in the Late Neolithic (Parkinson and Galaty 2007).
The Late Neolithic Tisza-Herpály-Csőszhalom complex. The trend of regional
differentiation that began with the subdivision of the Carpathian Basin into two discreet cultural
40
complexes – the AVK on the Plain and the DVK in Transdanubia – continued into the Middle
Neolithic on the Great Hungarian Plain with ceramic culture groups such as Szakálhát, Esztár,
and Tiszadob. This general trend ultimately resulted in the division of the region into three
discreet groups during the Late Neolithic. Known as the Tisza-Herpály-Csőszhalom complex, it
is roughly contemporary with Lengyel I-II in Transdanubia and the Petreşi culture in Romania
(Bognár-Kutzián 1966; Kalicz and Raczky 1987a; Sherratt 1997a; Parkinson 2006a).
Traditionally, the three ‘cultures’ are differentiated through distinctive ceramic assemblages, but
differences in settlement patterns and other material culture have also been observed:
Decorated Tisza fineware, with its intricate incised, textile-like patterns is
characteristic of the southern area and shows continuity from the Szakálhát group.
The northern part of the Plain – with the exception of the mountain fringes, where
settlement was abandoned – was occupied by the Herpály-Csőszhalom group,
with painted finewares. This replaced the multitude of smaller groups of the
northern edge of the Plain, and is found in the former Esztár area. Small
differences in the color-combinations distinguish Herpály from Csőszhalom, but
together they differ from Tisza not only in their pottery but also in their
settlement-pattern and in the occurrence of small numbers of simple copper
objects, for instance at Herpály. Since both these groups shared a common set of
domestic pottery, the sharp frontier between Tisza and Herpály finewares may
indicate a genuine social boundary… (Sherratt 1987a:280-281).
Additionally, researchers noted that true tells – which were likely founded near the end of the
Middle Neolithic Szakálhát phase – occur more frequently in the southern area of the Plain and
are associated with the Tisza group in the Late Neolithic (Kalicz and Raczky 1987a; Makkay
1982; Sherratt 1987a; Parkinson 2006). This settlement pattern consisted of the occupation of
the large tells up to six hectares in area associated with large horizontal settlements of up to 11
hectares, and numerous outlying flat sites. Together these constituted three basic settlement
types: genuine tell settlements, tell-like settlements with a more modest height and occupation
duration, and single layer horizontal (flat) settlements (Kalicz and Raczky 1987a:15; Parkinson
2006). These site types are often found in close spatial association with one another, and seem to
indicate a social and settlement pattern focused around occupation of tell sites at this time
(Kalicz and Raczky 1987a:17; Parkinson 2006; Sherratt 1987a:280).
The subdivision of the Plain into three distinct groups during the Late Neolithic appears
to have happened gradually, and Kalicz and Raczky (1984:131) argued that the Herpály and
Csőszhalom complexes did not develop directly from local Middle Neolithic groups, but from a
41
Csőszhalom
Tisza
Herpály
100 km
Figure 3.4. The Late Neolithic Tisza-Hérpály-Csőszhalom complex. Map used with permission of Dr. László
Zentai, Eötvös Loránd University, Budapest.
“remarkably uniform” Tisza culture that occupied the Hungarian Plain at the beginning of the
Late Neolithic. Besides the clear settlement differentiation that occurred during the period,
Parkinson (1999:111) noted “the various differences between the three groups extended to other
aspects of social organization as well…more subtle patterns in subsistence patterns, and
settlement organization are beginning to further distinguish one from the other” (Figure 3.4).
In additional to dramatic shifts in settlement form and pattern of distribution, ceramic
manufacture also underwent many changes during the Late Neolithic:
One conspicuous feature of this period is a basic change in pottery technology as
compared to preceding periods. Chaff was no longer used for tempering, and the
qualitative differences between coarser and finer wares practically
disappeared…The new pottery forms making their appearance in this period
include various amphora-shaped vessels and high pedestalled bowls (Kalicz and
Raczky 1987a:19).
42
Although the shift in pottery manufacturing techniques is associated with all three cultures of the
Tisza-Herpály-Csőszhalom complex, Kalicz and Raczky (1987a:20) also identified differences in
pottery design and ornamentation that are culturally diagnostic between the three groups.
Parkinson (1999:112) summarized these differences:
Tisza ceramics are characterized by deeply-incised meandric patterns on
unpolished, usually open-mouthed vessels. The incised meanders normally occur
in decorative panels, and the vessels are normally fired to a pale orange or brightred color. In contrast to ceramics of the Herpály and Csőszhalom groups, which
are typically painted with various colors, Tisza ceramics are painted only
occasionally with wide black bands.
This overall trend toward regional differentiation came to a halt around 4,500 B.C., at the
beginning of the Copper Age on the Great Hungarian Plain.
The Copper Age
The Early Copper Age Tiszapolgár Culture. In stark contrast to the trend of discrete
cultural differentiation throughout the Neolithic, the Early Copper Age Tiszapolgár culture (ca.
4,500-4,000 B.C.) exhibited a return to pottery and settlement types that were homogeneous
across the Great Hungarian Plain (Bognár-Kutzián 1966; Parkinson 2006b). The Tiszapolgár
area extended across the entire Plain, south into the Banat of northern Serbia, into the foothills of
Romanian Transylvania, and north into the mountains of southern Slovakia. The area of
Tiszapolgár distribution roughly corresponded to the previous Late Neolithic Tisza-HerpályCsőszhalom complex; however, the Tiszapolgár occupation extended to higher elevations in the
east and the north (Parkinson 1999:126) (see Figure 3.5). Since Bognár-Kutzián’s (1963, 1972)
analysis, most researchers have considered the Tiszapolgár to be a direct extension of their Late
Neolithic predecessors on the Plain. The fact that the Early Copper Age population on the Plain
is considered an extension of Neolithic forebears is especially interesting given the dramatic
changes in house form, settlement type, and settlement location that occurred during the
transition between the Late Neolithic and Early Copper Age.
The nucleated settlement pattern of large tell sites surrounded by flat sites of various sizes
predominant in the Late Neolithic was replaced by less nucleated, more evenly dispersed
settlement (Bognár-Kutzián 1972; Makkay 1982; Parkinson 2006b). Additionally, houses in the
Early Copper Age were much smaller than those in the Late Neolithic, possibly indicating a
trend toward dispersal of residential or family groups (Makkey 1982; Parkinson 2002, 2006b).
43
100 km
Figure 3.5. Extent of the Early Copper Age Tiszapolgár culture. Map used with permission of Dr. László Zentai,
Eötvös Loránd University, Budapest.
The long-distance trade networks that may have been associated with regional
differentiation in the Late Neolithic developed into networks bringing copper, gold, and chert
onto the Plain (Sherratt 1987a, 1987b). Additionally, the intramural burials common throughout
the Neolithic period were largely replaced by burial in highly organized, extensive cemeteries
such as Tiszapolgár-Basatanya (Bognár-Kutzián 1963).
In contrast to the three distinct ceramic groups of the Late Neolithic, Early Copper Age
Tiszapolgár pottery is essentially homogeneous across the entire Plain. The vessels of this
period are often decorated with lugs and knobs that themselves are sometimes pierced or semipierced (Bognár-Kutzián 1963; Parkinson 2006). Mcroscopic as well as a macroscopic
homogeneity is indicated by petrographic analysis of vessel fragments from this period (Parsons
2005); however, trace element analysis revealed regional variability in paste composition, and
non-locally produced pottery was present at some sites (Hoekman-Sites et. al 2007). Therefore,
it appears that pottery was locally produced and widely exchanged, suggesting a high degree of
44
interaction and low boundary maintenance that concurs with the results of Parkinson’s (2006a)
stylistic analysis of Tiszapolgár ceramics in the Körös region of the Great Hungarian Plain.
The Middle Copper Age Bodrogkeresztúr Culture. During the Middle Copper Age (ca.
4,000-3,500 B.C.) the Early Copper Age Tiszapolgár pattern continued on the eastern Hungarian
Plain. As such, Bodrogkeresztúr is considered a direct temporal extension of the Tiszapolgár.
Parkinson (2006) and others (e.g., Sherratt 1997a, 1997b) have noted that continuity between the
periods is supported by continuous use of settlement sites (e.g., Vésztő-Mágor) and cemeteries
(e.g., Tiszapolgár-Basatanya), and by the considerable overlap of radiocarbon dates in some
areas (Bognár-Kutzián 1972; Forenbaher 1993; Makkay 2007). As a result, the break between
Tiszapolgár and Bodrogkeresztúr is rather arbitrary and is marked by minor changes in ceramic
form and decoration, especially the development of closed vessel types referred to as “milk jars”
and incised square decorations with bands. Other Tiszapolgár decorative characteristics
continued into the Bodrogkeresztúr period (Bognár-Kutzián 1963, 1972).
Settlement organization during the Bodrogkeresztúr period remained largely unchanged
from the Early Copper Age. Habitations were small and dispersed, although fewer sites existed
in total. Sherratt (1997b) argued for a depopulation of the central Plain during this period based
on the decrease in number of sites from the previous period and the ephemeral nature of
Bodrogkeresztúr sites in the area.
Others have noted finds that seem to contradict Sherratt’s position. Makkay (2007)
described a multi-component Middle Copper Age site featuring what might have been a
subterranean pit-house – a feature previously unobserved at any other Bodrogkeresztúr sites.
Such dwellings are associated with the Yamnaya culture of the Eurasian steppe, which
presumably constructed the pit-grave kurgans throughout the plain beginning in the Middle
Copper Age (Ecsedy 1979; Makkay 1986). Makkay (1983) also discussed an anomalous
settlement (Szarvas 38) that contained a large roundel resembling those at Late Neolithic and
Early Lengyel sites west of the Danube River (ca. 4,700-4,000 B.C.). This may indicate a link
between the Bodrogkeresztúr culture on the eastern Hungarian Plain and various Transdanubian
cultures (for example, the Balaton-Lasinja culture). This is supported by the presence of
Bodrogkeresztúr sites in the previously uninhabited area between the Danube and Tisza Rivers
45
Bodrogkeresztúr
100 km
Figure 3.6. Extent of the Middle Copper Age Bodrogkeresztúr culture. Map used with permission of Dr. László
Zentai, Eötvös Loránd University, Budapest.
(Sherratt 1987a, 1987b) (Figure 3.6). Ultimately, such evidence supports an argument for
intensifying interaction between the people of the Plain and those outside of it as the Copper Age
progressed, or at the very least it suggests a higher level of social or economic integration across
the region at this time.
The Late Copper Age Boleráz-Baden Culture. The Late Copper Age on the eastern
Hungarian Plain marked a change in how cultures interacted within and outside of the Plain. It is
around 3,500 B.C. that an apparent discontinuity appears in the archaeological record in Eastern
Europe. Anthropomorphic clay figurines disappear, many large settlements were abandoned,
and thousands of burial mounds appear across the landscape (Milisauskas and Kruk 2002:247).
Additionally, three large homogeneous “style groups” appeared across Europe at this time: the
Corded Ware, Globular Amphora, and Baden groups. The people of the Hungarian Plain
became incorporated into the Baden group, a large material culture sphere that involved much of
Central and Eastern Europe beginning around 3,500 B.C. Boleráz (early Baden) and Baden sites
occur throughout Hungary, Austria, southern Slovakia, and western Romania. This represents a
46
regional material culture homogeneity not seen in the region since the Neolithic Alföld Linear
Pottery culture (AVK), if even then (Figure 3.7).
Comparisons between Boleráz and Baden, and contemporary southeastern European
cultures (especially in the Balkans) have long appeared in literature. Banner (1956) first
presented Baden finds from the culture’s central region of distribution, while Kalicz (1958) first
assigned Baden to the Late Copper Age, immediately following the Middle Copper Age
Bodrogkeresztúr phase. Kalicz (1963) proposed a link between Baden and Troy based on
ceramic similarities and cross-dating in the absence of radiometric dates, but admitted that such
an interpretation is faulty in light of radiocarbon evidence (Kalicz 2001). Němejcová-Pavuková
(1984) proposed a polygenetic scenario of Baden development. Under this model, BolerázBaden would have developed out of the Copper Age Lengyel culture of Transdanubia, with
elements of the Early Bronze Age Bulgarian Ezero culture and the Cernavoda III/Coţofeni group.
Design elements of these Balkan ceramic cultures (said to mimic bronze pots, see below) are
especially visible in Baden ceramic form. Němejcová-Pavuková’s analyses were based almost
entirely on selected elements of pottery styles that, while compelling, ignore a wider range of
variability in both Boleráz/Baden and comparative assemblages (see Němejcová-Pavuková 1981,
1983; Sochacki 1985). Unfortunately, Němejcová-Pavuková’s research was conducted without
rigorous chronological control, as well into the later 20th century (and indeed, until the present
day) very little, if any, radiocarbon data were available for the period and cultures in question.
Indeed, contemporaneous researchers compared similar data and reached different conclusions
(see Geogiev et al. 1979, who placed the development of Baden much earlier than Kalicz or
Němejcová-Pavuková, and not contemporary with the Ezero culture in Bulgaria).
Sochacki (1980a, 1984), rather than arguing for a patterned, diffusionist development of
Baden, suggested that parallels between Baden and southeastern European complexes were less
pronounced than presumed by other researchers. He extensively discussed the development of
Baden in the Balkans, and the subdivision of Baden into numerous cultural phases based on
ceramic typology. He concluded that Baden, Cernavoda III, and Ezero were roughly
contemporary and developed as the result of economic interaction with common outside entities,
potentially from Anatolia (as Kalicz [1963] had earlier suggested). Thus, rather than drawing a
47
100 km
Figure 3.7. Approximate extent of the Late Copper Age Baden culture. After Horváth 2008 and Sherratt 1997a,
1997b. Map used with permission of Dr. Zentai László, Eötvös Loránd University, Budapest.
direct link between Cernavoda III, Ezero, and Baden, he suggested a scenario of parallel
evolution.
In response to, and in stark contrast to, this position, as recently as 1998 NěmejcováPavuková focused on establishing a chronological link between Cernavoda III and Boleráz, and
Roman and Diamandi (2001) pointed out that Boleráz pottery of essentially one style was used
along nearly the entire course of the Danube, while regional differential between Baden
assemblages developed later and have more limited distributions. Ultimately, the question arises
of whether the early Boleráz-Cernavoda III pottery and the later Baden pottery represent two
separate cultures, or different phases of the same group (Furholt 2008; Roman and Diamandi
2001).
Research has since focused on discerning the archaeological reality of a Baden “culture,”
and recent emphasis has been placed on the identification of regional differentiation and
48
expression within the wider Baden horizon (Furholt 2008). The traditional concept of a Baden
culture has been called into question, and the model that the pottery style associated with Baden
equates to a distinct social group and a homogeneous culture is in doubt. Furholt (2008) soundly
rejected the concept of a unified Baden culture, and instead drew upon a wide range of evidence
to suggest that Baden was actually a diverse assemblage of regionalized groups unified by
similarities in ceramic form and decoration. In terms of other elements of material culture, little
congruence exists between other archaeologically recovered materials and Baden pottery.
Kaczanowka (1982/1983) and Pelisiak (1991) have shown a lack of geographic uniformity in
Baden flint industries. Human figurines are often mentioned as characteristic finds associated
with Baden pottery on the Hungarian Plain and in southwestern Slovakia (Kalicz 2002; Novotný
1981), while animal figurines are more common in the Austrian region (Furholt 2008; Pavelčík
1982, 1992; Ruttkay 1995:154). Furholt also discussed examples of regional variability in burial
customs associated with Baden pottery (see Sachße 2005), and significant regional variability in
faunal assemblages (see Benecke 1994:89). Ultimately, Furholt suggested that Baden pottery
has no equivalency with other cultural practices, and that Baden pottery itself – though certainly
similar throughout the region – exhibits striking variability:
The so-called Baden culture does not embrace a consistent cultural package, and
even if expressed by pottery along has been shown here to be a course
approximation of a number of ceramic subsystems. What is more, fine and
courseware pottery show different developments. In the early phase (3,650-3,350
B.C.) course fabrics are regionally diverse and local in their context and meaning.
The earliest fine wares, the Boleráz wares, have their first use in Austria (and the
adjacent region), but then spread over a short time span to north and west mixing
with other cultural attributes (2001:627).
A great deal has been written over the last 30 years in regards to the development of the
Late Copper Age Boleráz and Baden material cultures and their relationships to Cernavoda III
and other Bronze Age groups in Bulgaria and the Balkans. As has been presented above, the
relationship between Baden and contemporary cultures, and sub-groups within the Baden
tradition, remains debated after more than 30 years of discussion. This topic alone is worthy of a
dissertation, and as such a complete treatment of the related literature is beyond the scope of the
current project’s focus on the appearance and development of Baden on the eastern Great
Hungarian Plain. Only a brief review of this literature has been presented here. However, it
should be noted that the precise relationship between Boleráz/Baden and cultures to the south
49
remains unclear, and the precise cultural origins of the Late Copper Age Boleráz/Baden complex
remain a topic of discussion.
Banner (1956) presented the first treatment of the Late Copper Age Baden culture on the
Hungarian Plain, and produced a map of sites that was published in his monograph Die Peceler
Kultur. As Sherratt (1997a:291) pointed out, however, lowland eastern Hungary, including the
Körös-Berretyó drainage, has a notable gap in coverage. Roman and Németi (1978) filled in
some of this conspicuous gap, while the publication of the Hungarian Archaeological
Topography series (Escedy et al. 1982; Jankovich et al. 1989; Jankovich et al. 1998) provided a
much more complete picture of Late Copper Age occupation in the southeastern Plain. In terms
of animal use, Bökönyi (1974) suggested an increasing emphasis on cattle in upland valleys and
a preponderance of sheep in the lowlands; however, excavated evidence from the Körös region is
still lacking, and a sufficient characterization of animal use is not possible at this time.
Interestingly, the trend toward settlement dispersal that began during the latest phase of
the Neolithic and continued into the Early and Middle Copper Age (Gyucha 2010; Parkinson
2006b; Sherratt 1997b) appears to have continued into the Late Copper Age despite widespread
material culture change. Many fewer Late Copper Age sites exist in the study region in the
center of the Plain, and the very few excavations in the Körös region consisted only of a few pits
(Megyesi 1982, 1983, personal communication 2009). Indeed, data from the surface survey and
collection of Late Copper Age Boleráz and Baden sites in the region are consistent with this
pattern (see Chapter Five), and lend support to Sherratt (1997b) and Gyucha (2010), who saw
widespread dispersal and perhaps depopulation of the central Hungarian Plain at this time,
despite a continued involvement in import and the external trade.
Settlement locations suggest increased penetration into the upland regions surrounding
the Hungarian Plain, as well as the previously mentioned region between the Danube and Tisza
Rivers. The Maros Fan (Maros River watershed) just to the south of the Körös watershed study
area remained almost completely bereft of settlement during this time, in spite of a slightly
increased settlement density in the Tisza-Danube interfluve. A lack of settlement in the Maros
region – which had no active river channels at this point in the Holocene (Sherratt 1983, 1997a)
– may indicate that while settlement and population density had been on the decline, access to
the active transport waterways of the Körös region and, thus, raw and finished materials from the
50
margins of the Plain and further afield, may have remained an important aspect of Late Copper
Age society.
Boleráz and Baden ceramic design in the Körös Region exhibits characteristics quite
different from Early and Middle Copper Age types, though the differentiation between Boleráz
and Baden on the Hungarian Plain is largely based on the presence or absence of certain
decoration and design features rather than rigorous chronological or stratigraphic evidence.
Baden ceramics are characterized by handled jugs and cups, which stylistically suggest affinities
with Bulgaria, northern Greece, and the Aegean region (Banner 1956; Kalicz 1963). The people
of the Baden period, therefore, may have been more closely linked with neighboring
archaeological cultures than in previous times (Whittle 1996).
Late Copper Age ceramics on the Hungarian Plain are often burnished and decorated
with incised stacked chevron and incised obtuse and acute cross-hatching not seen in Early or
Middle Copper Age assemblages. Additionally, these incised patterns are often observed in
concert with linear and rectilinear patterns of punctuations, and occasionally with pinched rims
or linear impressions across rims.
The appearance of wheeled vehicles in Europe occurred around this time, and likely
represents a major socioeconomic development (Bakker et al. 1999; Maran 2001, 2004). Clay
wagon models belonging to the Baden culture have been found at Budakalász and
Szigetszentmárton (Banner 1956; Kalicz 1976). Such technological developments would have
facilitated easier transportation and movement of goods (Anthony 1995; 2007), implies the use
of traction/draft animals, and could help account for the rapid development of large, relatively
homogeneous material culture groups across Europe.
Despite the list of discontinuities in the archaeological record at the onset of the Late
Copper Age, some characteristics, in addition to the continuing trend of settlement on the
margins of the Plain indicate a measure of continuity between Boleráz, Baden, and earlier culture
phases of the Copper Age. One such practice, the tradition of placing the dead in large, formal
cemeteries, continued at places such as Alsónémedi in Pest County (see Bökönyi 1951; Korek
1951; Némeskeri 1951). However, no such cemeteries have up to this point been discovered in
southeastern Hungary, and Sherratt described most known Baden cemeteries as “quite small”
(1997b:309). He did not rule out the possibility that some of the smaller find spots studied
51
during his Szeghalom survey on the Dévaványa Plain are in fact cemeteries, but no evidence has
been collected to substantiate this.
Unfortunately, the vast majority of research on the Late Copper Age Boleráz and Baden
culture group in Hungary has taken place to the west of the Körös basin study region, primarily
in Transdanubia. It is here where radiocarbon dates firmly illustrate the existence of the ProtoBoleráz, Boleráz, Early Classical Baden, and Classical Baden (see Horváth et al. 2008). Such
rigorous chronological control in the southeastern Plain is not yet possible given the lack of
investigated sites and radiocarbon samples gathered from stratigraphically controlled
excavations. Despite the lack of excavated and dated materials in the region, evidence further
afield has noted regional variability in Late Copper Age Baden assemblages (see Furholt 2008,
as discussed above), making this period in a relatively restricted geographic area an excellent test
case for modeling the development and adoption of geographically extensive material cultures.
The Late Copper Age Kurgan Culture. The most unusual features of the Late Copper
Age landscape are the numerous pit-grave kurgans (mounded tumuli) that dot the flat landscape
of the Great Hungarian Plain (see Ecsedy 1979). Burials under the kurgans are laid in the supine
position with the knees raised, are often covered with red ochre, and are sometimes found with
the remains of textiles and a wooden cover or container. Although usually sparse, grave goods
include pottery resembling that of the Yamnaya of the Russian Steppe. This has led some to
support a migratory explanation for their appearance on the Plain (Gimbutas 1963, 1977, 1979,
1980). Importantly, Anthony (1986) noted that no Yamnaya horizon homeland truly exists,
contrary to Gimbutas’ eastern origins hypothesis. He contended that the so-called kurgan
hypothesis is disputed by stratigraphic evidence and a lack of supporting radiometric dates.
Rather, he suggested that the “homeland” is a broad region of steppe environment stretching
from the Dnieper to the Volga, likely based on arguments put forth by Merpert (1968, 1974) and
Mallory (1989), who envisioned the Yamnaya as a collection of widespread cultural traits rather
than a single ethnic group. The kurgan tumulus burials – the hallmark of the Yamnaya – may
have served as clan territorial markers under this model (Anthony 1986:297).
Although some stratigraphic data exist (Ecsedy 1973, 1979: 47-52, 1981), the exact
chronological relationships between kurgans and the Baden and Bodrogkeresztúr cultures
remains unclear. Sherratt (1997b:310) noted, however, that the distribution of kurgans in the
Körös River basin is spatially distinct from that of Baden sites. On this evidence, he suggested
52
that the kurgan builders were an intrusive pastoral group from the east that intentionally placed
their mortuary structures in areas unoccupied by Baden agriculturalists. Since the majority of
Late Copper Age sites in this region consist only of a few pits (see Megyesi 1982, 1983), it is
difficult to assess the relationship between Baden sites and kurgan pit burials. Therefore, the
impetus of social change on the Plain associated with the Baden culture requires further
clarification (see O’Shea 1996).
As an alternative to a migratory explanation for the Yamnaya appearance east of the
Hungarian Plain, Telegin (1973) suggested that kurgans developed directly from the earlier
Copper Age Stredni Stog culture already present in the region north of the Black Sea. Although
the immediate origins of the Yamnaya culture are complicated and remain disputed (Anthony
1986), for the purposes of this research it suffices to say that the Stredni Stog and other cultures
further to the east were the predecessors of Yamnaya groups. Such a large territory (over 3,000
kilometers from west to east) could be considered a Yamnaya cultural-historical area rather than
a single archaeological culture (Mallory 1989; Merpert 1968, 1974). Gimbutas (1965, 1977,
1980) followed this in her later publications, describing the “kurgan culture” not as a cultural
unit, but rather as a collection of cultural traits unified by a common symbolic burial tradition.
Anthony (1986) specifically spoke out against a specific Yamnaya homeland or point of origin,
though he soundly criticized Gimbutas’ interpretations of culture change and migration on the
Eurasian steppe.
The existence of a homeland for migrating kurgan builders has been debated, though
most researchers currently accept the idea of a migration onto the Hungarian Plain at the end of
the Middle or beginning of the Late Copper Age. A much smaller migration at the end of the
Early Copper Age represented by the Marodécse-type graves – although considered unrelated to
the migrations of kurgan-builders – is also recognized (Ecsedy 1981:78-79). Some (see Kulcsár
2003:141; O’Shea 1996:361) have suggested that a migration onto the Plain occurred at the
beginning of the Early Bronze Age. Regardless of the chronology, the extent to which the arrival
of the kurgan people affected life and material culture for the indigenous people of the Plain
remains uncertain.
The Early and Middle Bronze Age (ca. 3,000-1,600 B.C.). During the Early Bronze Age,
a trend toward regional differentiation had once again emerged as the procurement of raw and
finished bronze materials from outside the Plain became increasingly important. O’Shea (1996)
53
argued that the existing social order of the Copper Age disintegrated and smaller, regional
culture groups developed during the Bronze Age as manufacture and control of valuable items
became specialized. In the Körös region of the Plain, this correlates with the Makó, Nyírség, and
Ottomány cultures. Although the overall settlement pattern footprint remained similar to that of
the Late Copper Age, Sherratt (1997b) noted another increased settlement focus on the margins
of the Great Hungarian Plain and a return to the occupation of tell sites in key central locations
on the landscape. He attributed this to an increased focus on procuring raw materials for bronze
production from the mineral-rich foothills of the Carpathians, and escalated trade for finished
goods throughout the Plain. In economic terms, this is a logical conclusion to the long-term
trend of increasing focus on the edges of the Plain as the production, use, and ownership of
bronze and other metals became more important.
The beginning of the Bronze Age in Hungary is linked to sudden and dramatic cultural
changes following the decline of the Baden complex at the end of the Copper Age (see Kulcsár
2003:141). Settlement patterns exhibited a return to the use of central tell sites in key locations
across the landscape (Sherratt 1997b), and Bóna (1965) and O’Shea (1996:361) suggested that a
migration of the kurgan-builders of the Pitváros group might account for the notable change in
material culture. These changes included not only the widespread trade of raw and finished
bronze items (Sherratt 1997a), but also distinct changes in ceramic form and decoration, and
potentially manufacturing techniques (Kulcsár 2003:141-142).
The Developmental Trajectory of the Great Hungarian Plain
Clearly, the developmental trajectory of the Great Hungarian Plain does not fit easily into
any of the models summarized in the previous chapter. A primary factor affecting social
development in the region involved a long-term pattern of nucleation and dispersal beginning in
the Middle and Late Neolithic, and extending into the Early and Middle Bronze Age. It
culminated with the appearance of ranked societies on the Great Hungarian Plain – much later
than in many other parts of the world. Why did it take so long for ranked societies to emerge on
the Hungarian Plain? Why did ranking appear when it did, and what trajectory or trajectories led
to its appearance?
The question of why the development of ranked societies on the Great Hungarian Plain
lagged in comparison to other places in Europe and the Eastern Mediterranean is not a question
54
best approached from a single line of evidence. However, it can be argued that the late
development of ranking can be attributed to two primary factors:
1. The cyclical nature of tribal society and settlement variability on the Plain
2. The relative geographic isolation afforded by the Carpathian Basin and the Hungarian
Plain’s lack of valuable raw resources.
Parkinson (2002, 2006a) described the nature of social integration and interaction on the
Hungarian Plain during the Copper Age as “tribal.” Although the term is loaded with the
baggage of decades of discussion (see Sahlins and Service 1960; Service 1971), Parkinson’s
primary concern was with the inherently flexible, pre-existing social structure inherent in tribal
society that allows for fission/fusion type cycling over long periods of time. On a small scale,
Parkinson and colleagues (Parkinson et al. 2004:118) and Gyucha et al. (2004) discussed this
process through analysis of spatial segregation of craft activities at the site of Vésztő-Bikeri on
the eastern Plain. The segregated activity areas in different structures suggest that the Early
Copper Age village functioned as an “integrated economic unit.” The high density, multi-family
households and sites of the Late Neolithic diffused at the end of the Neolithic and beginning of
the Copper Age. An individual household would have branched off from the wider settlement,
becoming a seperate settlement and functioning as a discreet economic unit. The ability for
villages to disperse into smaller, more independent settlements would have restricted household
competition, effectively prohibiting the development of institutionalized hierarchy during the
Late Neolithic and into the Copper Age.
As illustrated by the Hungarian Archaeological Topography surveys (Escedy 1982;
Jankovich et al. 1989; Jankovich et al. 1998), later by Sherratt (1997a, 1997b), and discussed in
even greater detail by Parkinson (1999, 2002), a clear pattern of increases and decreases in site
number and size occurred between the Middle/Late Neolithic and Middle Bronze Age on the
eastern Great Hungarian Plain. As previously noted elsewhere, this has often been interpreted as
a process or cycle of population nucleation and dispersal in the region. Parkinson (1999, 2002)
noted a correlation between regional material culture differentiation and degree of integration
(e.g., nucleated or dispersed settlement structure). In periods of greater nucleation, such as the
Late Neolithic period with a tell-centered settlement system, material culture expressed more
regional differentiation as seen in ceramic design and decoration and house construction. On the
55
other hand, in periods characterized by dispersal, such as the Early Copper Age Tiszapolgár
phase, material culture was more homogeneous over a wider geographical area.
The Late Copper Age Baden phase falls rather neatly into Parkinson’s and Makkay’s
models of nucleation, dispersal, and material culture regional homogeneity and differentiation.
At least on the Central Plain and in the Körös Region, the dispersal that originated in the Early
Copper Age continued into the Late Copper Age, and the distribution of Baden sites in the Great
Hungarian Plain as a whole indicates the growing importance of the areas on the edges of the
Plain (Banner 1956; Roman and Németi 1978). This is especially true of intermontane valleys,
resource-rich areas that then supported a substantial population in contrast to earlier periods
(Sherratt 1997a:291).
The pattern established by 3,500 B.C. in the Late Copper Age largely continued for the
next thousand years, and the general pattern of the Early Bronze Age on the Plain follows from
Baden – with a settlement focus near the edges of the Plain, as opposed to dense settlement in the
central Plain or Körös Region. Moreover, many researchers (see Bóna 1975; O’Shea 1978;
Sherratt 1997b) have noted the significant role of trade in explaining settlement near major rivers
and on the edges of the Plain during the Late Copper Age and Early Bronze Age. Sherratt
(1997a:291) argued that the Tisza and Maros Rivers took on an especially important role at this
time, due to their potential for long-distance trade by canoe. O’Shea (1978, 1996) noted
imported items such as copper ornaments and shell beads in the Bronze Age graves of the Maros
Region, indicating that the procurement of finished foreign materials increased in importance
and in volume over this time period.
Although the importance of trade and economy cannot be overlooked, the possibility
must be considered that a migratory population of kurgan builders appeared on the Hungarian
Plain and in the Körös Region sometime around 3,500 B.C. This population often has been
associated with the material changes witnessed at the beginning of the Late Copper Age, and
with the economic changes that followed during the Early and Middle Bronze Age (Anthony
1990; Gimbutas 1977, 1980; Milisauskas and Kruk 1989, 2002:247). So, while this burgeoning
economic pattern and pattern of nucleation and dispersal may have respectively contributed to
and hindered the development of regional political systems with a tributary economy, craft
specialization, and institutionalized hierarchy on the Plain (Earle 2002; O’Shea 1996), the
considered.
56
Invasion vs. Economy: A Tale of Two Models on the Great Hungarian Plain
The migration discussion within archaeology has shaped interpretations of the
archaeological record throughout the world, and the eastern Great Hungarian Plain is no
exception in this regard. Indeed, two primary models have served as lenses through which to
observe the Neolithic, Copper Age, and Bronze Age on the Plain and the Körös River study
region. Divided roughly into invasion/migration and economic/environmental models, the two
perspectives are associated most strongly with the archaeologists Marija Gimbutas and Andrew
Sherratt, respectively. As the heart of this research, these models frame the research questions
approached in this dissertation and, as such, deserve specific treatment and criticism here.
The models of Gimbutas (1970, 1977, 1979) and Sherratt (1983, 1984, 1997a, 1997b)
have been two of the primary frameworks for interpreting change on the prehistoric Great
Hungarian Plain for years. Both Sherratt and Gimbutas, speaking generally, painted a picture of
long-term diachronic change punctuated by “prime-mover” type events for the Neolithic and
Copper Age. This process eventually contributed to the appearance of institutionalized social
inequality at the beginning of the Bronze Age at around 3,000 B.C. These models are in many
ways exclusive of one another, though they both have much to offer in terms of understanding
prehistoric social change on the Plain. I suggest here that the relatively sudden, prime-mover
events are actually part of the long-term trajectory that began in the Neolithic period. The nature
of the change became easier to observe in the archaeological record near the end of the Copper
Age, at which time the regional Baden material culture group became the dominant presence on
the Hungarian Plain. The following sections explore each of the models in turn, and present
support and criticism for portions of each.
The Invasion/Migration Hypothesis
Gimbutas’ (1963, 1970, 1977, 1979) continental-scale model for social change on the
Great Hungarian Plain focused primarily on the sudden appearance of thousands of
homogeneous burial mounds in the region during the Middle Copper Age, around 3,500 B.C. It
can be summarized as follows:
1. Yamnaya pastoralists (referred to as “kurgan people”) from the Pontic Steppe arrived
on the Hungarian Plain around 3,500 B.C.
57
2. The kurgan peoples’ entrance marked the arrival of the Indo-European language,
culture, and religion in Europe.
3. The arrival brought about sudden, drastic culture change, with the Late Copper Age
Boleráz and Baden culture complexes becoming completely “Indo-Europeanized,”
thus marking the end of “Old Europe.”
Gimbutas used several lines of evidence to support her hypothesis. First, and most striking, are
the kurgan burial mounds themselves. They are essentially identical in construction and contents
to Yamnaya kurgans to the east. They consist of a mound of earth containing a supine skeleton
with legs slightly raised, with the body usually oriented on an east-west axis (Escedy 1979;
Gazdapusztai 1967; Gimbutas 1977). Additionally, their distribution across the eastern portion
of Europe and into western Europe look, essentially, like what one would expect from a
migration (Anthony 1990, see above). Under Gimbutas’ model, this arrival of mound builders
would have ushered in a higher level of chiefly complexity that was shaped during the Boleráz
and Baden periods of the Late Copper Age and fully realized in the Early and Middle Bronze
Age.
In terms of the Great Hungarian Plain during the Middle and Late Copper Age (ca. 4,0003,000 B.C.), a migratory population is most clearly observed in the kurgan burial mounds that
dot the landscape of the Körös region (Escedy 1979). As described in more detail in the
archaeological background section, kurgans on the eastern Hungarian Plain are essentially
identical to Yamnaya burials on the Eurasian Steppe, and one can observe a “stream” of kurgan
burials stretching from the Steppe near the Dneiper River west into the Hungarian Plain. Indeed,
these kurgans may be the most solid archaeological evidence for migration in the region.
However, it remains unclear exactly how the kurgans developed across the landscape despite the
many kurgans that have been systematically excavated in the last half-century. Until an in-depth
program involving the chronological characterization of these mounds over a very wide
geographic region is undertaken, the questions of the spread of the so-called kurgan people will
remain unanswered. Additionally, until a more solid kurgan chronology is developed for the
Körös region and the Hungarian Plain more generally, their assignment to particular indigenous
cultural phases remains unclear.
58
The Environmental/Economic Model
Sherratt’s model differs from that of Gimbutas in many ways, most noticeably in his
multi-scalar, regional approach. Primarily, Sherratt analyzed settlement patterns on the scale of
the Great Hungarian Plain in order to observe general patterns of settlement development, and
then analyzed a much smaller micro-region – a part of the MRT survey on the Dévaványa Plain
near the towns of Szeghalom and Dévaványa – to comment more specifically on those general
trends (1984, 1997b).
Sherratt emphasized “a considerable measure of continuity” from the Neolithic to the
Early Bronze Age (Sherratt 1997b:292). To generalize, his model can be distilled into several
main points regarding settlement patterns:
1. Rivers, waterways, and seasonal flooding helped shape settlement patterns throughout
the Neolithic and Copper Age. Sites in all periods tended to occur either on raised
“islands” amid areas of predictable seasonal flooding, or atop natural riverbank
levees.
2. During the Bodrogkeresztúr Period in the Middle Copper Age (ca. 3,900-3,500 B.C.),
the Plain underwent depopulation; fewer Middle Copper Age sites existed than earlier
Tiszapolgár Early Copper Age sites, and they occured in previously unoccupied areas
of the Plain. Parts of the eastern Plain are subsequently reoccupied in the Late
Copper Age.
3. Middle/Late Copper Age kurgan burial mounds are spatially complementary with
Late Copper Age Boleráz and Baden settlements, possibly indicating an intentional
avoidance of specific areas occupied by migratory pastoralist groups.
Sherratt used a large-scale, eastern Hungarian Plain site distribution map to argue for a
depopulation event in the central Plain during the Middle Copper Age (1997b:308). This
depopulation – due to local environmental factors and a shifting emphasis on the importance of
goods and raw materials from outside of the Plain – may have made room for the pastoral kurgan
builders to move into the now less densely occupied Körös River basin. He supported this
hypothesis with data from the Szeghalom survey in northern Békés County.
Sherratt’s two analytical resolutions were a weakness in his otherwise extensive analysis.
His wider scale, on the level of the entire Hungarian Plain, and his smaller scale, a small portion
of northern Békés County in the southeastern Plain, provide us with a wealth of settlement and
59
interpreted economic information. However, a more moderate scale at the county level would
provide intermediate level data, and an information bridge between Sherratt’s two scales to either
bolster his conclusions or provide a more detailed interpretation. This level of analysis is
included in the current research. Sherratt’s model will be tested and discussed more extensively
in Chapter Six.
Baden Pots with Local Roots? Defining the Late Copper Age on the Great Hungarian Plain
On the eastern Great Hungarian Plain, including the Körös region, the Baden material
culture complex and the people who created it remain even less understood in terms of their
development and external relationships than in other regions that have seen years of research and
many publications contributing to the discussion. However, given the homogeneity seen in the
Boleráz phase over a large geographic scale and the regional variability noted in Baden
assemblages over the same area in Eastern and Central Europe, the eastern Great Hungarian
Plain provides an excellent test case for characterizing the development of Baden as a local
phenomenon or one catalyzed by the movement of personnel from one region to another.
One potentially confounding problem of studying the Late Copper Age in the Körös
region, however, is the lack of systematically and stratigraphically excavated sites. As
previously mentioned, almost all of the Baden excavations from the eastern Plain have consisted
only of a few pits, and are typically considered intrusive features at other sites (see Megyesi
1983). With very few exceptions, the Late Copper Age material from the Körös region available
for study comes from surface contexts. The lack of chronological control makes establishment
of a Late Copper Age ceramic typology, as has been done in other regions, virtually impossible.
Additionally, the distinction between Boleráz and Baden material – which has been solidly
demonstrated in Transdanubia and in areas beyond Hungary (Furholt 2008; Horváth et al. 2008)
– has not been established in the Körös region. Many stylistic elements between Boleráz and
Baden are shared, and the combination of a lack of chronological control and inconsistent
description of surface finds in the MRT series (Ecsedy et al. 1982; Jankovich et al. 1989;
Jankovich et al. 1998) makes the establishment of a similar typology on the Plain difficult.
However, that is not to say that approaching the topic of Late Copper Age Development
on the Plain is impossible. Although the establishment of a definite ceramic typology may have
to wait for further investigation and radiocarbon data, a portion of this research project
60
approaches the problem by grouping all Late Copper Age materials, and comparing them with
previous periods in order to identify differences in manufacturing technology that may indicate
an internal or migratory nature of the material culture shift.
Summary
This chapter has presented the geographic, geological, geomorphological, and
archaeological background to the prehistory of the eastern Great Hungarian Plain. It has
presented information regarding the appearance of the Late Copper Age Boleráz and Baden
complexes that will form the backdrop for the research presented in the following chapters. By
presenting an archaeological background beyond the Late Copper Age focus of this dissertation,
I have placed the Baden material culture complex into a wider context, which is an important and
necessary requirement for both conducting and understanding the research presented in
subsequent chapters. Additionally, this chapter has integrated the wider anthropological
considerations presented in Chapter Two with the specific archaeological context of the
prehistoric Great Hungarian Plain.
61
CHAPTER FOUR
THEORETICAL EXPECTATIONS AND RESEARCH DESIGN
Introduction
In this chapter I provide the methodological links between the archaeological and
anthropological concepts discussed in Chapter Two and the archaeological case of interest in this
dissertation – the social, settlement, and material culture changes that took place at the end of the
Copper Age on the Great Hungarian Plain (see Chapter Three). This chapter explains the
methodological principles that have guided the current research design, and presents the
interpretive framework by which the results are judged. First, I discuss why the temporal and
geographic scales of analysis used in this research project are appropriate for investigating the
research questions. I then provide an overview of regional analysis research and previous
regional analysis on the Great Hungarian Plain, and how these projects have shaped the research
in this volume.
In the second half of the chapter, I discuss how archaeologists have approached social
change through ceramic analysis, and how ceramic research design developed and changed over
the last century. I first discuss traditional approaches to ceramic analysis, primarily in terms of
pottery form and decoration. I then discuss the technological approach to ceramic analysis
employed in this dissertation, and describe how both macroscopic and macroscopic analyses
developed and how they have been utilized in other archaeological projects. Special attention is
given to petrographic ceramic analysis, as petrographic research is particularly suited for
approaching questions of technological and production variability over space and time. In this
vein, I then outline the interpretive frameworks for testing the models in Chapter Three against
the results of the present research.
Temporal and Geographic Scales of Analysis
Both temporal and geographic scales of analysis must be considered when modeling
long-term prehistoric social, settlement, and geographic change. Although the Late Copper Age
(corresponding to the Boleráz and Baden material culture traditions) lasted roughly 500 years on
the Great Hungarian Plain, placing this period of time within its proper context requires the
consideration of periods that preceded and came after it. So, although the primary questions
associated with the research at hand are concerned with the Late Copper Age, a methodological
62
approach considering the Neolithic, Early and Middle Copper Age, and Early and Middle Bronze
Age is useful and necessary for understanding the development of the Late Copper Age in the
region. Additionally, a wider geographic scope is required for assessing the unique nature of the
changes in the study region, or for observing regional settlement changes that took place during
this time. Although for decades a distinction has been made between two primary components
of the Late Copper Age in the region – Boleráz and Baden – this distinction is based primarily on
the presence or absence of particular ceramic designs, and has yet to be solidly demonstrated
chronologically in the study region. Even more, the Boleráz-Baden classification scheme is used
inconsistently in the Körös region, with different archaeologists placing similar or identical
design elements into different periods (see Escedy et al. 1981; Jankovich et al. 1989; Jankovich
et al. 1998). Since the majority of the material analyzed as part of the present investigation was
collected from surface contexts, and is therefore not rigorously chronologically controlled, Late
Copper Age Boleráz and Baden material will be treated as a single temporal unit that lasted
approximately 500 years, from ca. 3,500 B.C. to ca. 3,000 B.C.
The most obvious geographic scale of analysis for understanding changes at the end of
the Copper Age in the Carpathian Basin is the Great Hungarian Plain in its entirety. Generally
speaking, the Plain is a geomorphic and topographic unit approximately 50,000 square
kilometers in area. Some researchers, especially Sherratt (1997a, 1997b) have used the entire
Plain as an analytical unit in terms of generalized material culture distribution. Indeed, part of
the site distribution spatial analysis in the present research discusses large-scale patterns at this
analytical scale. However, the intensive analysis of such a large area precludes the present
investigation; such an analysis is far beyond the scope of this project. Even more, such a large
scale analysis would prevent the detailed analysis of sub-regional and local areas that will aid in
modeling the changes that took place during the Late Copper Age on a local level (for example,
analysis on sub-regional and local scales allows one to ask questions relating to how the area
became incorporated into the wider Baden archaeological and anthropological homogeneous
material culture group; see Chapter Two and Chapter Three). Additionally, it is important to
assess detailed local information such as soil type and river location when conducting a site
spatial analysis. This kind of specific local data would be unmanageable over an area as large as
the entire Hungarian Plain. A more restrictive geographic scale of analysis is a more manageable
63
MRT Volume 8
MRT Volume 6
MRT Volume 10
Figure 4.1. The Körös River basin study area, including modern cities and MRT parish boundaries as discussed in
the text. Located in southeastern Békés County, Hungary (inset).
sample, and allows local data to be contextualized within the general area of the Great Hungarian
Plain. For the purposes of the present study, the geographic unit of analysis (the study area)
consists of a roughly 3,000 square kilometer area in the Körös River valley drainage in Békés
County (Figure 4.1).
Analyzing Social Change through Settlement Patterns and Regional Analysis
Regional settlement pattern research has a rich history in many parts of the world (for
recent review articles and multiple examples see Billman 1999; Galaty 2005). However, it was
only as late as the second half of the 20th century that an emphasis was placed on creating a more
detailed and systematic study of spatial patterning (Hodder and Orton 1976). After initially
adopting some forms of spatial analysis from geography and plant ecology (see Haggett 1966),
64
archaeologists such as Hodder and Orton (1976) and Clarke (1972, 1977) recognized the need
for a more rigorous approach. Although site distribution maps were often employed and usually
quite useful, such visual methods of analysis can often seem uncritical and unreliable (Hodder
and Orton 1976). In the present study region on the southeastern Hungarian Plain, this is
beginning to come to light. For example, Sherratt (1997b:310) noted that Late Copper Age
kurgans and Baden sites are geographically exclusive. Although this is true in a very general
sense, it now appears that this is not always the case at high geographical resolutions (see
Chapter Six).
Much of modern regional analysis stems from Binford’s (1964) influential article, “A
Consideration of Archaeological Research Design.” Binford argued that the task of isolating and
studying processes of cultural and social change is best approached regionally and through
sampling techniques at multiple scales. He drew ideologically from Steward (1960) and White
(1959) in saying that artifacts, over geographic areas, can be classified into distinct groups. Then
an archaeological typology can be developed in terms of artifact form and complexity. One can
then make the jump from the static artifacts to culture types, getting at the form, organization,
and complexity of a society based on the geographic distribution of complexity and form in a
society’s material culture. Such groupings based on technology and design can be used to define
regions of analysis.
Hodder and Orton (1976) similarly emphasized the need for rigorous, systematic analysis
of patterns across an archaeological landscape and discussed the difficulty of inferring process
from form. They suggested a greater emphasis on reorganizing the process – or signature – that
led to the creation of an archaeological distribution. Subsequently, Anthony (1986, 1990) once
again emphasized the importance of identifying the signatures of archaeological phenomena in
regards to migration patterns (see Chapter Three). It could be inferred that he intentionally chose
to address an element of Hodder and Orton’s argument that receives little attention: when dealing
with social and material culture change, archaeologists often rely on invasion and migration
hypotheses regardless of the difficulty of demonstrating such a process archaeologically
(1976:3). Similarly, Clarke (1977) suggested that elements such as raw materials, artifacts,
features, structures, sites, and routes should be used in conjunction to identify archaeological
signatures. Clarke characterized regional archaeology as concerned with not only the retrieval of
information from archaeological spatial relationships, but with the study of the flow and
65
integration of activities within and between structures, sites, and resources spaces on scales
ranging from the inter-site level to the large-scale geographic level of analysis.
Many years after his initial article on the subject, Binford (1982) once again approached
the topic of why the region is the most appropriate unit of analysis for understanding variability
in the archaeological record. He emphasized that the explanation is relatively straightforward:
different assemblages at different site-types do not necessarily indicate the presence of multiple
archaeological cultures. Behaviors at various locations utilized for discrete purposes can vary
seasonally, annually, or at other temporal scales, thereby leaving distinct assemblages and
patterns that at first glance may appear to have been created by different peoples or cultures.
By associating region with material culture, Binford (1982) essentially defined a “region”
as coterminous with the geographic extent of a cultural system. However, Haggett (1966) noted
much earlier that most researchers have tended to use ad hoc geographical divisions to meet
specific research needs or to address specific questions. Indeed, Foley (1981) stated that
numerous factors come into play when identifying proper regions at different analytical scales in
the archaeological record. These facts include topography, environmental productivity, climate,
habitat, and cultural characteristics such as a group’s particular diet and subsistence strategy
(1981:4). Although Foley primarily oriented his methodology toward highly mobile, huntergatherer societies, his factors for identifying appropriate regional boundaries remain important
for the regional study of any society or material culture, including the present study.
In terms of variability within material culture regions, Stark (1998b:3, see 1998a) stated
that “a primary goal in studying formal variation across space is to identify social groups, whose
boundaries are marked by distinctive patterns in the archaeological record”. However even the
identification of formal variation within a material culture group is problematic. For example,
within the Late Copper Age Baden material culture group on the Great Hungarian Plain, formal
variation across space – even between large regions – is not extensively documented, and as
discussed above even chronological variability between formative and later phases of the Late
Copper Age can be called into question when closely scrutinized. This is what Wandsnider
(1988) called the “methodological double bind”; that is, in order to measure something we have
to know what it is like, but in order to know what it is like we must measure it. One way in
which to overcome this problem is to conduct detailed analyses of technological aspects of
material culture, such as production or preparation methods, in order to identify subtle regional
66
or temporal similarities or differences (Lemmonier 1992; Stark 1998a, 1998b). Since
technological characteristics are often conservative (as opposed to design and form, which can
change drastically over fairly short periods of time by comparison), technological similarities
may indicate cultural affiliations or anthropological processes not always visible in the
archaeological record, such as migration or invasion. These concepts will be discussed in more
detail below, in the section dealing with ceramic analysis methodology.
Hodges (1987) addressed the most important element of regional spatial analysis in his
discussion of space through time. He stated that archaeological methodologies must take into
account the proper scales at which particular research questions should be approached.
Essentially, in order to move beyond spatial patterns of settlement, production, and distribution
to modeling the events that created such a pattern, one must examine a space diachronically at
several scales of analysis. Such an approach accounts for the problem that patterns that hold true
in a certain area at a certain scale may not hold true in the same area at a different scale, or in a
different area at the same scale. If any particular lesson from decades of discussion of regional
analysis should be specifically applied to the Great Hungarian Plain, it is that a diachronic, multiscalar analysis is absolutely necessary in order to accurately observe patterned variability in
settlement location over hundreds and thousands of years.
For several decades, archaeologists working in eastern Europe in the field of regional
analysis have refined the gamut of regional analysis methodologies. Galaty (2005:291) has
recently argued that that “steady investments in the technology, methods, and theory of regional
archaeological analysis and surface survey have stimulated advances in the study of settlement
patterns and settlement pattern change through time in many parts of Europe.” He also noted a
challenge in that Europe is difficult to define in terms of archaeological method and theory due
to the wide variety of research questions being asked in various regions, as well as the different
theoretical and methodological approaches taken by archaeologists in different countries across
the continent. However, the rich nature of settlement studies in Europe – especially in areas such
as Britain, Denmark, and Greece – also means that numerous methodological innovations have
advanced regional studies in Europe over the last several decades and have contributed to the
advance of archaeological science as a whole.
At the beginning of the 20th century, many European archaeologists focused on defining
the material characteristics of the artifacts within the boundaries of their own countries and
67
functioned within a patriotic or nationalistic framework, with a priority of constructing a
particular culture history to suit individual home countries (Trigger 1989). It was from within
this context that settlement archaeology first developed (Galaty 2005), especially by the
Germans who aimed to establish the German people as the original Europeans (Chapman 1997;
Trigger 1989). Later in the century, archaeologists like V. Gordon Childe (1930, 1958b)
continued to build on earlier culture history frameworks, and used careful studies of material
culture in order to map culture groups and trace their interactions (Galaty 2005; Sherratt 1997a).
By the middle of the 20th century, European archaeologists began to develop explicitly
regional approaches to the past. The primary data collection method for these initial regional
scale projects was via surface survey (Galaty 2005). Most of these projects, many of them in
Greece, focused on topographically interesting areas based on the kind of sites under
investigation and the kind of research questions being asked (Cherry 2003). The University of
Minnesota Messenia Expedition (UMME) is an example of this kind of extensive topographic
survey (McDonald and Rapp 1972). Later surveys, like the Pylos Regional Archaeological
Project (PRAP) in Messenia and the Mallakastra Regional Archaeological Project (MRAP) in
Albania were intensive surveys with the goal of surveying wide swaths of land of variable
topography. In doing so, these studies attempt to eliminate the sampling bias inherent in
extensive, more topographically selective surveys (Davis 1998, 2008).
In recent years, archaeologists working in Europe have focused on landscape studies that
treat the entire regional landscape as if “it were one large, ever-changing artifact” (Galaty
2005:296). Under this approach, the landscape concept unifies many variables of regional
analysis, including ecological, social, and geographic considerations. These studies suggested
that settlements are only one component of a wider landscape that consists of any number of
geographic and geological elements such as water bodies and forests. Although the landscape
study approach explicitly brings together ecological, social, and geographic approaches, it should
be remembered that such methodologies are not truly limited to the last two decades. As
discussed below, large-scale surface surveys in Hungary led to Sherratt’s (1983, 1984, 1997a,
1997b) detailed analysis of Neolithic and Copper Age settlements on the Great Hungarian Plain.
Although not specifically presented as a landscape study, the project did consist of all of the
necessary elements of the landscape-oriented approach.
68
A question raised in regards to the effectiveness of regional settlement analysis is
whether data collected for projects typically focused on relatively small, well-bounded
landscapes have utility in addressing general, comparative questions about human behavior in
the past (Galaty 2005). Some scholars have especially been concerned with the scale of many
European projects, suggesting that settlement data at these small scales are not useful for
understanding patterns of political, social, and cultural change (Blanton 2001, Cherry 2003). I
disagree, and suggest that multi-scalar regional analysis projects concerned with settlement
patterns has, in fact, been highly successful in identifying and modeling past social and cultural
change in prehistoric Europe. As evidence, I present below an overview of several regional
studies that have contributed to the understanding of prehistory on the Great Hungarian Plain.
Previous Regional Analysis Projects on the Great Hungarian Plain
Unlike most of Europe, the Körös River drainage in Hungary has three decades of
archaeological survey available for spatial analysis, encompassing an area of over 3,000 km2
(Escedy et al. 1982; Jankovich et al. 1989; Jankovich et al. 1998). Beginning in the 1960s the
Institute of Archaeology of the Hungarian Academy of Sciences, in collaboration with
archaeologists at regional museums across Hungary, began conducting extensive field-walking
survey projects in order to identify all of the archaeological sites across the landscape. What
emerged from these projects was a multi-volume effort – the Magyarország Régészeti
Topográfiája (MRT). Each MRT volume combines all known documentation (at the time) of
known sites and private collections within parishes (administrative districts below the county
level). Each volume corresponds to several of the administrative districts within the county.
Each parish was intensely surface surveyed with the goal of identifying all of the unknown and
previously discovered sites in the area. To date, ten volumes have been published. Of these ten,
volumes 6, 8, and 10 correspond to the study area relevant to this dissertation.
Each MRT volume documents the location of all previously known sites and sites located
as part of the survey. All areas free of impassable obstructions were surveyed with the aid of
1:10,000 scale topographic maps. Field walkers walked transects at 15-20 meter intervals. The
eastern Hungarian Plain is ideal for this kind of research, as the surface geomorphology of the
region has remained relatively stable since the Neolithic period or the beginning of the Holocene
(ca. 10,000 years ago). Additionally, most of the region is under intensive industrial agriculture
69
with large, open, plowed areas well suited to the surface identification of archaeological sites
(Parkinson 1999, 2006b). When a new site was identified, site size was roughly estimated based
on the surface scatter of artifacts, and a non-systematic collection of diagnostic sherds was
conducted in order to determine which period or periods was represented. The sample sizes were
intentionally kept small so as not to affects the results of any future surveys or excavations
(Gyucha 2007, personal communication). The 1:10,000 topographic maps marked with the
location of all identified sites are kept at the Institute of Archaeology of the Hungarian National
Academy of Sciences in Budapest, and the artifacts collected during fieldwork are stored in the
county and city museums in Békés County, the Hungarian National Museum, and the Institute of
Archaeology.
Over the last decade, most of the topographic maps used for the surveys in Békés County
have been photocopied and subsequently electronically scanned and georeferenced for use in
Geographic Information Systems. Although the MRT studies have been an invaluable tool for
Hungarian and international archaeologists working on the eastern Hungarian Plain, very little
effort was made for interpretation of the data. There have been some notable exceptions, as
Makkay (1981, 1986a, 2007) excavated and published on some sites shortly after their
identification and publication in the MRT. Recently, foreign researchers have expanded upon
the initial MRT research, and Gyucha (2010) published an excellent analysis of Early Copper
Age trends on the eastern Plain.
In addition to the Hungarian MRT research in the region, a collaborative
British/Hungarian research project (Sherratt 1983, 1984, 1997a, 1997b) was undertaken in the
northern part of Békés County in the late 1970s and early 1980s. Sherratt originally published
the results of this research in the Oxford Journal of Archaeology (1983, 1984). They have
subsequently been reprinted in a more recent volume:
Fieldwork in Hungary was designed to track changes in object-distributions and
settlement-patterns over thousands of years, from c. 6,000 to 2,000 B.C., in order
to see how long-term social processes work out on the ground in a nodal area of
central Europe. The contrasting patterns of site-distribution in the survey area are
a unique record of fluctuating (and slowly evolving) forms of spatial organization
over long periods of time, which a comprehensiveness and degree of resolution
unequalled anywhere (Sherratt 1997:270).
70
Sherratt’s Körös-Berettyó project described the long-term patterns of settlement and settlement
change that characterized 4,000 years of agricultural life on the eastern Hungarian Plain. His
analysis and interpretation consisted of two parts: a regional description (1983, 1997a) that
provided a detailed description of the current and prehistoric geomorphology of the region and
how Neolithic and Copper Age site distribution correlated with geological features such as soil
type, hydrology, and elevation; and a detailed discussion of patterns of site location based on
previous survey work in the Körös Basin, including the MRT. In fact, most of the data for
Sherratt’s analysis came from the initial stages of survey for the first MRT volume published on
Békés County.
Sherratt noted a “considerable measure of continuity” from the Neolithic to Early Bronze
Age on the eastern Plain in his high-resolution study region, which was limited to a section of
northern Békés County covered in volume 6 of the MRT near the modern towns of Szeghalom
and Dévaványa (1997a:292). However, on a wider scale he also noted variations in and between
the central area of the Plain and on the margins of the occupied areas. Most notably, at the scale
of the entire Hungarian Plain he documented a trend toward the focus on the resource-rich edges
of the Plain in the foothills through time, culminating in a depopulation of the center of the Plain
in the Middle Copper Age Bodrogkeresztúr phase. Although the central area of the region had
been re-occupied by the Early Bronze Age, he observed a focus of settlement in the foothills,
presumably the result of an emphasis on procuring raw and finished materials, and access to
trade routes along which finished, perhaps high-prestige, materials were traded (Sherratt 1997b).
At a much smaller scale on the Dévaványa Plain of northern Békés County, this same
pattern could not be observed (probably due to its position in the center of the Plain). Sherratt
(1997b:311) did note, however, the cyclical nature of settlement variability that oscillated
between low-density, ephemeral sites and high clustered, high density occupation. The lowdensity early Neolithic sites began a period of 700 years of aggregation into larger, higher
density, more stable settlements. By the Early Copper Age, smaller more ephemeral settlements
were dispersed more widely across the landscape, possibly partly as a result of the emerging
trend of settlement focus on the margins of the Plain nearer to copper and other raw material
sources. The economic pull of the foothills, potential local ecological changes, and the arrival of
migratory pastoral populations from the east during the Middle Copper Age all may have played
a part leading up to a period of relative abandonment in this area of the Plain. According to
71
Sherrat, by the Late Copper Age small enclaves of Baden agriculturalists lived amongst the
kurgan-building populations, and the focus of settlement continued to shift into the surrounding
uplands. By ca. 3,000 B.C. and the Early Bronze Age, the agricultural areas in the center of the
Plain were characterized by small defended villages in an area not marginal to the main zone of
occupation to the east (Sherratt 1997b:311-312).
The work of Sherratt in the late 1970s and early 1980s and the work of the Hungarian
archaeologists who compiled the MRT volumes constituted a major achievement in regional
methodology and the analysis and interpretation of regional data not only in Hungary, but in
Eastern Europe and the Balkans as a whole. In many ways, the Körös-Berettyó project laid the
foundation for later regional projects on the Hungarian Plain.
The Körös Regional Archaeological Project (KRAP), which focuses on a region of the
eastern Plain characterized by the slow-flowing channels of the Körös and Berettyó Rivers, was
started in the late 1990s. Parkinson (1999, 2004, 2006b) used such a methodology as advocated
by Binford in his stylistic analysis of the Late Neolithic Tisza-Herpály-Csőszhalom complex and
the Early Copper Age Tiszapolgár culture group. The initial goal was to develop a more
complete understanding of the social changes affecting the nature of social organization at the
beginning of the Copper Age on the eastern Hungarian Plain. Concerned with the cyclical nature
of settlement organization that has been discussed previously, Parkinson (1999:355-428, 2006b)
analyzed a battery of 20 stylistic ceramic attributes in order to measure the degree of interaction
between Early Copper Age settlements in the study region. The results of the stylistic analysis
were used along with a cyclical model of social interaction derived from diachronic settlement
pattern data analyzed by KRAP and first collected in the MRT volumes. Parkinson determined
that the overall pattern was one of uniformity, which suggests a high degree of continuous
interaction between the sites in the study area. However, the distribution of incised ceramics
suggests the marking and maintenance of social boundaries within the study area. Interestingly,
the same area identified by Parkinson as a boundary previously marked a more discreet boundary
zone between the Herpály and Tisza groups during the Late Neolithic. This suggests that
boundary maintenance, though present, was less actively maintained during the Early Copper
Age.
At the site level, KRAP has focused primarily on two adjacent Early Copper Age sites on
the Great Hungarian Plain. Excavation at Vésztő-20 and Körösladány-14 has shed much light on
72
the household organization and economy of the period. Research conducted by KRAP modeled
social organization during the Late Neolithic and Early Copper Age across the Körös River study
region, and suggested that household competition was restricted by patterns of nucleation and
dispersal (see Chapter Three). This leveling mechanism effectively prohibited the development
of social ranking on the Plain during the Neolithic and the Copper Age (Gyucha et al. 2004;
Parkinson et al. 2004:118), to the point where it did not appear until other economic and material
factors took hold during the Middle Bronze Age, over 1,000 years later. The regional, multiscalar, diachronic nature of research at KRAP allows for the observation of changes not only in
settlement patterns over time, but when modeled effectively also sheds light on how settlement
and social systems change over the long term.
A Problem with Regional Studies, and how to Approach it in the Future
Regional multi-scalar research does have a major problem relating to research location
and scale. To illustrate this problem, I refer to a segment of the research conducted as part of
Andrew Sherratt’s Körös-Berettyó project.
Sherratt analyzed settlement patterns on the scale of the Great Hungarian Plain in order to
observe general patterns of settlement development, and then analyzed a much smaller microregion – a part of the MRT survey on the Dévaványa Plain in northern Békés County near the
towns of Szeghalom and Dévaványa – to comment more specifically on those general trends
(Sherratt 1984, 1997b). He argued for cultural continuity from the Neolithic to the Early Bronze
Age in the area.
Though Sherratt’s analysis shed a great deal of light on prehistoric settlement on the
Great Hungarian Plain, his project also illustrates the problem of geographic location and scale in
many regional projects. Although general patterns may exist at very large scales – such as the
spatially complementary distribution of Late Copper Age settlements and kurgan burial tumuli –
and may hold true in some smaller analytical regions, the same patterns may not hold true in
other nearby micro-regions. Depending on a researcher’s sampling strategy and how one defines
the study region, patterns could be falsely identified or missed altogether. This will be discussed
further in Chapter Six.
A problem with the identification of variability also exists in regional analysis when
working at very large geographic scales. A major point in Marija Gimbutas’ model for the
73
arrival of kurgan building people in eastern Hungary in the Middle and Late Copper Age is the
homogeneous nature of the Yamnaya culture of the Eurasian steppe, and their emigration from
their homeland (Gimbutas 1952, 1979, 1989). However, it has been noted that a high degree of
variability exists in the archaeological record for kurgan builders at this time, to the point that the
term “Yamnaya” may be best considered a package of shared behaviors and material culture
rather than a unified social or cultural group (Merpert 1968, 1974). Anthony (1986, 1990) has
been critical of Gimbutas’ homogeneous characterization of the Yamnaya, and proposed a
different model for Yamnaya migration less dependent on the movement of “cultures” and more
concerned with the movement of people acting within the framework of a cultural system
(1990:895). One can more readily observe such cultural variability at numerous resolutions and
different geographic scales, rather than examining primarily large or small scales.
Social Change, Ceramics, and Technology
Ceramic analysis has, for decades, been the lynchpin in establishing chronological
cultural sequences archaeologically and, to a very large extent, defining culture groups across not
only time but also space. Although archaeologists have mostly moved past the “pots equals
people” mentality of previous generations, ceramic assemblages across the world are still often
assumed implicitly to represent both the geographic extent and ethnicity of culture groups. This
is problematic when it comes to large, regionally homogeneous material cultures like the Baden
horizon, since such material representations of human activity do not necessarily represent
ethnicity or, necessarily, even cultural affiliation.
A great deal of attention has been devoted to the study of pottery in modern archaeology.
Rice (1987:24-25) listed a number of reasons for this disproportionate amount of study,
including that pottery has a long history and is present in almost all parts of the world, it is
essentially non-perishable and preserves well in the archaeological record, and that unlike other
artifacts such as projectile points, pottery sherds are not particularly attractive for looters and
collectors, and it is in general not an exotic or highly valued good and not usually confined to
specific sectors of a population. Most importantly, however, is its manufacturing method:
74
Pottery is formed and informed: pottery making is an additive process in which
the successive steps are recorded in the final product. The shape, decoration,
composition, and manufacturing methods of pottery thus reveal insights – lowly
and lofty, sacred and profane – into human behavior and the history of
civilizations. Potters’ choices of raw materials, shapes to be constructed, kinds of
decoration, and location of ornamentation all stand revealed, as do cooking
methods, refuse disposal patterns, and occasional evidence of clumsiness and
errors in judgment. The sensitivity, spatial as well as temporal, of pottery to
changes in such culturally conditioned decisions has fed archaeologists’
traditional dependency on this material for defining prehistoric cultures and their
interrelations (Rice 1987:25).
Rice (1987:25) went on to say that most modern archaeological pottery studies are based on one
or more of three approaches: classification, decorative analyses, and compositional studies.
Classification studies focus primarily on the grouping of sherds or vessels representative of a
particular material culture or phase of a material culture, and sometimes the subsequent
comparative analysis of groupings over space or time. Such studies formed the basis of
archaeological chronology, especially in the 19th and most of the 20th centuries. Decorative
analyses focus on painting, surface treatment, and plastic decoration of vessels, and in addition to
offering insight on esthetics and ideological systems, stylistic variability can also deliniate social
boundaries across space (Parkinson 2006a; Stark 1998a, 1998b).
This dissertation embraces Rice’s belief that pottery manufacturing methods are just as
important, if not more important, than ceramic form and design. However, I further incorporate
more recent perspectives on technology in anthropology and ceramic analysis in order to refine
this perspective and to make it directly applicable to the anthropological and archaeological
problem at hand. Lemmonier (1992) followed Mauss (1955) in suggesting that even our most
basic acts – sitting down, standing up, scratching one’s nose – are culturally determined, and
included more complex modes of technology (such as building a jumbo jet or creating a pot from
clay) as culturally structured and influenced. As such, he identified two types of technological
traits: stylistic traits and functional traits. Stylistic traits, such as color, design, decoration, and
sometimes form, are often correlated with social identities based on characteristics such as
gender, age, hierarchy, or membership in a particular society or group. Such characteristics
convey conscious and unconscious messages that provide information regarding the object and
its provenance and/or provenience to the observer to construct a social identity. On the other
hand, functional characteristics –the selection of certain materials for certain objects or purposes,
75
or for a particular technological action, or a particular chain of events by which an object is
created – often unintentionally convey social or cultural messages (Lemmonier 1992:85-86).
A Technological Approach to Ceramic Analysis
A theme running throughout Lemmonier’s work – that intentional, stylistic elements of
technology can substantially change arbitrarily and relatively quickly while functional
technological elements are of a more conservative nature and resistant to change – underpins
much of the present research (see Lemmonier 1992:51-77). Lemmonier’s perspective on
technological change is nowhere more appropriate than in the study of archaeological ceramics,
where design and decoration often appear to change arbitrarily and suddenly. Early and middle
20th century archaeologists often explained such changes in terms of invasions, migrations, or
other direct movement of people (see Adams et al. 1978; Anthony 1990; Childe 1958a, 1969;
Gimbutas 1963, 1977; and Chapter Two of the present work for thorough discussions of material
culture change and migration).
While more antiquated ceramic studies equated material culture changes (e.g., changes in
form, design, and/or decoration) with ethnic changes or the arrival of new cultures into a
previously occupied region, more recent studies have emphasized human continuity during
periods of material culture change, often over thousands of years. On the Great Hungarian Plain,
arguments for continuity from the Neolithic through the Early and Middle Copper Age have been
successfully supported for years. Sherratt (1997a, 1997b), Parkinson et al. (2004, 2010), Gyucha
et al. (2004), and Parkinson and Gyucha (2007) have supported models of long-term human
continuity over the course of systemic, cyclical social and economic changes in the region. They
cite not only direct evidence from the archaeological record as support for these models (for
example, continuity in burial practices; see Bognar-Kutzian 1963), but also patterns of economic
specialization as part of nucleation and dispersal cycles across the Carpathian Basin (Makkay
1982; Parkinson et al. 2004).
Macroscopic Analysis
The majority of research conducted on archaeological ceramics over the past century has
been without the aid of a microscope, using only characteristics visible to the naked eye (design,
form, some kinds of temper, and firing conditions) or with low-power hand lenses. Although
76
space and the focus of the present research prevent a comprehensive review of such a large
portion of the archaeological publication and research record, I do present an overview of
literature that has contributed methodologically and theoretically to macroscopic methods, and I
briefly discuss some of the weaknesses of this method of research.
Macroscopic analysis of ceramics initially focused on the creation of detailed typologies
of design and form, and delineating geographically-based culture-groups on continental or
regional scales (Childe 1958a; see Trigger 1989:122-123). In North America, this manifested in
the systematic study of variation in the archaeological record oriented toward defining
geographic rather than chronological patterns, and followed the tendency of American
ethnologists in the late 19th century to organize the study of material culture similarities and
differences into culture areas. For example, McGuire (1899) created fifteen geographic cultural
divisions based on the distribution of different types of Native American pipes. In terms of
ceramic analysis, Holmes (1903) used stylistic analysis as well as some technological variables
(such as temper and firing conditions) to define pottery regions for the eastern United States.
In Europe during the late 19th and early 20th century, the culture-historical perspective in
archaeology both relied on and shaped the study of material culture, including ceramics. Unlike
in North America, however, European antiquarians and archaeologists were more concerned
with establishing chronological sequences based on how design, form, and the use of materials
changed over time. The primary European objectives of this time were nationalist in nature, as
researchers associated artifact (e.g., ceramic) types with ethnic groups in order to learn more
about their specific national heritage and how their ancestors lived (Sklenář 1983:91; Trigger
1989:149). In central and eastern Europe especially, an orientation toward nationalistic study led
to an emphasis on the Neolithic and later periods (Trigger 1989:149) (which not only focused on
the establishment of ethnic continuity over long periods of time, but also on the development of
material culture groups spreading across both time and space).
Orton et al. (1993) divided this general process into several chronological phases of
ceramic analysis: the art historical phase, the typological phase, and the contextual phase. Most
of the previous analyses of Late Copper Age Hungarian ceramics (and indeed, most ceramic
studies in the region) can roughly be said to belong to the typological phase. This phase has
roots in the 1880s with Pitt-Rivers’ (1906) development of typological classes for many
categories of artifacts. Subsequently, this set a trend for the development of type-series (see
77
Déchelette 1904; Dragendorff 1895; Knorr 1906; Ludowici 1904; Orton et al. 1993; Walters
1908). Kidder’s (1924, 1931) integration of stratigraphy, survey, and ceramics set the stage for
decades of research to come. This method of ceramic study is integral to Childe’s general
approach to material culture and its representation of cultures, and though the focus was
chronological, the typological approach was crucial to the development of geographic cultural
groupings. Ultimately, it was the typological approach that was initially used to develop
chronological sequences and define culture areas throughout North America, Europe, and the
Great Hungarian Plain.
Anna O. Shepard’s (1956) work, as part of what Orton et al. (1993:13) called the
“contextual phase,” emphasized a holistic approach to ceramic studies that emphasized
chronology, distribution, and technology. She was also one of the first to directly highlight the
importance of studying technological aspects of ceramic form and manufacture. Technologically
focused studies are largely indebted to her pioneering and comprehensive work, and scientific
methods became increasingly utilized in the study of archaeological ceramics. Many of these
scientific methods came in the form of chemical and, more important for the present research,
microscopic analysis.
Microscopic Analysis
Unlike macroscopic ceramic analysis, microscopic analysis and the observation and
classification of ceramic inclusions under high magnification does not have a theoretical history
stretching to the nationalist and evolutionist perspectives of the late 19th and early 20th centuries.
Scientific, microscopic approaches to ceramic analysis have most notably been useful in dating,
sourcing, and function studies (Orton et al. 1993:18). Of most importance for the research
presented as part of this study, the thin sectioning and microscopic analytical techniques were
first explored quite early (Bamps 1883), and much later refined and used for investigating more
specific questions (Peacock 1967). Shepard (1942) applied thin-sectioning techniques to show
how Rio Grande glaze-paint pottery was traded over very long distances, and by the 1930s thinsectioning and petrographic analysis had gained popularity on both sides of the Atlantic Ocean
(Orton et al. 1993; see Gladwin 1937; Liddell 1932; Obenaur 1936).
Ceramic petrography, of the many various methods of ceramic analysis, may be uniquely
able to best address questions of technology, manufacture, and variability in production
78
techniques over time. Most often, the goal of ceramic petrography is to classify sherds or wares
according to material or technology, or to identify categories such as wares, series, or types
(Reedy 2008:151). Numerous studies have shown that precise information gathered from
petrographic analysis can establish sets of standards that allow ceramic characterization using
only low-powered microscopy (see Chandler 2001; Freestone 1995; Reedy 2008; Shepard 1939,
1942).
Petrographic analysis of ceramics can address many questions of archaeological
significance. Peacock (1968) examined questions of provenance in Iron Age British ceramics,
while Beynon et al. (1986) and Kreiter (2005) have used petrographic methods to investigate
questions of manufacture, technology, and cultural implications for various methods of vessel
construction. They also noted functional reasons for the inclusion of different tempers, and
Kreiter (2005) discussed grog as a non-functional inclusion with imbedded cultural meaning,
linking potters with their ancestors through the reuse of crushed ceramic material. Although a
complete review of the utilization of ceramic petrography is beyond the scope of this study (for a
recent survey of ceramic petrography see Reedy 2008), it should be remembered that
petrography can address numerous archaeological questions via various nuanced methodologies.
Most modern thin section studies involve analysis, identification, and characterization of
non-plastic inclusions (Reedy 2008). These studies may involve determining provenance
through the comparison of ceramic mineral inclusions with clay samples gathered from the field,
and they may be complemented by the observation of other features not observable
macroscopically that are useful in describing manufacturing techniques or grouping fabric types
(Whitbread 1995). Some clay minerals, referred to as “plastics” because of their malleable
nature when wet, usually compose more than 50% of a ceramic sample’s volume, but they
cannot be characterized by thin section analysis due to their small size (approximately 1µm
thick) (Reedy 2008:124). The analysis and characterization of this amorphous groundmass, or
matrix (see Stoltman 1989, 2001) can often prove just as useful as the study of non-plastic
mineral inclusions, especially when identifiable minerals are homogeneous over time or space, or
when studying technological characteristics of the ceramic samples. Reedy (2008:124)
characterized the most important contributions of thin-section petrography:
The most important focus of thin-section petrography of pottery is on
identifying, quantifying, and interpreting nonclay inclusions. These
inclusions improve working, drying, and firing properties of the clay, and
79
can also affect the use properties of the finished ceramic material. Beyond
identifying the inclusions present, useful interpretations may require
identifying a variety of characteristics such as abundance (frequency of
inclusions), size (average or modal, or most common, as well as the grainsize distribution), internal distribution of particles, morphology, and
alteration mechanisms.
Shape, abundance, and distribution of non-plastic particles can prove useful; for example,
rounder quartz inclusions suggest a longer period of alluvial transport and weathering
(Reedy 2008:124).
Speaking generally, two types of petrographic microscopic analysis have been applied to
the description of archaeological ceramics (see Middleton and Freestone 1991; Reedy 2008).
The first, qualitative petrography, identifies minerals present in each sherd, roughly characterizes
the size and shape of mineral inclusions, and describes paste characteristics that provide details
about production and construction (see Whitbread 1989, 1995). The mineralogical differences
identified by qualitative analysis may serve to delineate ceramic types and establish provenance
(Galaty 1999). Other characteristics, especially in terms of the appearance of the clay matrix,
can also be used to bolster qualitative research.
The second petrographic analytical technique, quantitative petrography, applies a more
rigorous and objective methodology for collecting petrographic data. Typically, mineral
inclusions and void space are systematically counted and recorded according to size and shape in
a process called “point-counting.” Many modern petrographic researchers utilize the
methodology developed by Stoltman (1989, 1991) that emphasizes a distinction between clay
body and clay paste. Clay “body” is defined as “the bulk composition of a ceramic vessel,
including clays, larger mineral inclusions in the silt, sand, and gravel ranges, and temper”
(Stoltman 1991:109). “Paste” is defined as the aggregate of natural materials, i.e., clays and
larger mineral inclusions, to which temper was later added to produce the body from which a
vessel is made (Stoltman 1991:109-110). This distinction is important because it recognizes the
independent origins of the artificial temper and natural paste collected and worked by a potter
(Galaty 1999; Stoltman 1989, 1991).
Because quantitative petrography under Stoltman’s methodology accounts for both
natural and artificial inclusions present in the composition of a ceramic material, the combination
of a systematic quantitative analysis with a detailed qualitative analysis can be especially useful.
80
Both natural and added components of a potter’s mixture can be identified, and the
technological, functional, and design priorities of a pot’s creator can be inferred. For example,
design and production characteristics identified in paste characteristics qualitatively, such as
coiling versus slab forming methods, can be observed. Or, ratios of certain mineral grits
identified in quantitative analysis may be added only to cookpots, for example, in order to
improve resistance to thermal stress (Arnold 1985:25; Galaty 1999:39; Rice 1987:228-229).
Stoltman’s (1989, 1991) methodology requires the petrographer to accurately distinguish
between natural and intentional (temper) inclusions. Galaty (1999) and others (see Whitbread
1995; Zubrow 1988) noted guidelines for making these distinctions. Generally, rock temper
grains are larger, polyminerallic, and more angular than natural inclusions. As a result, a
bimodal size distribution of mineral inclusions may indicate the intentional addition of temper to
a clay’s paste (Whitbread 1995; Kreiter 2005).
In addition to the identification and characterization of non-plastic inclusions in a paste, a
key element of quantitative petrography involves point counting the slide in order to determine
inclusion percentages. Stoltman (2001) described point counting as superimposing a grid over
the thin section and recording all observations at the cross hairs of set intervals. These
observations may be mineral inclusion grains, clay matrix, void space, or other features. So long
as rigorous point counting procedures are followed, the basic theory of the method is that the
counts made in this section analysis reliably estimate the actual volumetric proportions of each
constituent feature (Reedy 2008). Experimental work by geologists (Chayes 1954a, 1954b) has
validated this relationship.
Some studies have attempted to add a more rigorous component to qualitative analysis by
utilizing estimation charts to estimate variables such as mineral inclusion abundance (Orton et al.
1993; Reedy 2008; Rice 1987; Stoltman 1989). Although it is possible to criticize semiquantitative methodologies, it has been shown that such an approach can quickly and accurately
characterize ceramic fabrics both within and between chronological periods (see Kreiter 2005).
Additionally, archaeologists have used combinations of petrographic and chemical or traceelement analyses to characterize the nature of ceramic production and exchange. Roper et al.
(2010) utilized a dual methodology of petrographic analysis and oxidization analysis on order to
demonstrate the local production of shell tempered pottery in the North American Central Plains,
while simultaneously describing regional variability in firing techniques. Galaty (1999) used
81
petrographic analysis and chemical analysis (a combination of weak acid extraction and
inductively coupled plasma spectroscopy) in order to characterize Mycenaean coursewares and
finewares, and to demonstrate different administrative implementation strategies for mobilizing
utilitarian goods.
While many archaeologists have used ceramic petrography to characterize fabric types
within periods or between contemporaneous geographic or geological zones in order to
understand trade, economic systems, and manufacturing techniques (see Galaty 1999; Parsons
2005), the utility of microscopic ceramic analysis goes beyond the establishment and discussion
of fabrics. Utilizing the history of petrographic research as a foundation, the present research
projects examines ceramics and ceramic fabrics over a long period of time (approximately 1,500
years) in order to identify changes in manufacturing or production techniques that could indicate
an influx of people into the Körös region during the transition between the Middle and Late
Copper Age, and into the Bronze Age.
Interpretive Framework: Linking Settlement Patterns and Pottery in the Körös Region
Late Copper Age
The framework for interpreting the data and applying them to the tested models in this
research is not a complicated one. However, imperative to the interpretation of the spatial and
ceramic results is the understanding that the lines of evidence presented here are related analyses
that must be interpreted together. Despite the various methodologies employed as part of this
research, the results speak to the same issues and must be considered as a package. Indeed, it is
hoped that the results and interpretations presented in this volume will serve as a foundation for
the establishment of future research questions. As such, a simplified interpretive framework of
possible outcomes of the spatial and ceramic analyses is presented in Table 4.1.
82
Table 4.1. Table depicting the interpretive framework for data sets utilized in this research. Note that change in
ceramic design has previously been established and well documented, and while not directly tested as part of this
project is considered as part of the interpretive framework.
Model of Change
Data Sets
Expected Patterning
Migration
Ceramic Design
Distinct change in form and decoration
Ceramic
High degree of variability in microscopic and macroscopic characteristics
Paste/Body
between cultural phases; indicators of manufacturing and/or raw material
preparation changes, and/or changes in firing technology (examples include
changes in oxidization of paste, changes in b-fabrics, preferred orientation,
changes in natural mineral inclusion or intentional temper ratios)
Settlement Patterns
High degree of spatial exclusivity between Late Copper Age archaeological sites
and kurgans. Possible observable and quantifiable reduction in Middle and Late
Copper Age site frequency and density
Adoption
Ceramic Design
Possible change in form and decoration, but retention of design characteristics
(e.g., surface treatment) throughout the study between cultural phases
Ceramic
Little variability in microscopic and macroscopic characteristics within the study
Paste/Body
area between cultural phases; no indication of changes in manufacturing
technology, or observable, quantifiable change over time (through cultural
phases)
Settlement Patterns
Little or no spatial exclusivity between Late Copper Age sites and kurgans. Any
changes in site frequency and density attributable to other factors
Combination
Ceramic Design
Change in form and decoration between cultural periods with possible retention of
earlier characteristics
Ceramic
Diachronic variability observed in some manufacturing/technological features and
Paste/Body
not others (e.g., change in mineral inclusions or tempering, especially sudden
change, but no change in other fabric characteristics)
Settlement Patterns
Little or no qualifiable or quantifiable spatial exclusivity between Late Copper
Age sites and kurgans; or, spatial exclusivity between sites of different cultural
phases attributable to one phase's spatial relationship to kurgans
Spatial Analysis: Observing Nucleation, Dispersal, and Spatial Association through Time
The analysis of site distribution and settlement location change over time is a primary
concern of this study. As such, a spatial analysis research design must account not only for how
sites appear and contrast across a large geographic region but also at multiple smaller units. This
is an important step in site spatial analysis, as patterns that hold true at one resolution do not
necessarily hold true at others. In order to account for scalar variability, and to observe the longterm patterns of nucleation and dispersal discussed elsewhere (see Sherratt 1997a, 1997b;
Parkinson 2006), two broad perspectives are utilized in the spatial analysis conducted as part of
83
this research project. First, a traditional multi-scalar approach generally observes long-term
diachronic changes in settlement patterns over several cultural phases at several scales in the
Körös region. This largely expands geographically on Sherratt’s (1997a, 1997b) mid-1980s
study in northern Békés Country in an effort to determine if his observed patterning at two scales
(the Dévaványa Plain and the Hungarian Plain more generally) holds true elsewhere in the
research area. This line of evidence in the spatial analysis focuses heavily on the spatial
relationship between kurgans and Late Copper Age Baden archaeological sites, as a major
observation of Sherratt’s research (1997b) was the spatial exclusivity of these sites at the
resolution of his study region and the Hungarian Plain. This led him to conclude that two
separate populations – the indigenous Late Copper Age people and the migratory kurgan builders
– intentionally avoided contact and settled in different areas of the Plain. The multi-scalar
approach applied to a wider geographic region in the present research aims to test and expand
upon Sherratt’s conclusions.
Second, known Early, Middle, and Late Copper Age sites are subjected to average
nearest neighbor statistical analysis, in order to quantify clusters of sites (nucleation) and
ascertain the statistical significance of qualitatively observed nucleation and dispersal over the
time period covered by the research. This tests the conclusions of other researchers (Gyucha
2010; Makkay 1982; Parkinson 2006; Sherratt 1997a, 1997b) who have observed nucleations
and dispersals over time, and serves as a supporting line of evidence for the multi-scalar study
undertaken as part of the present research.
The two approaches aimed at qualifying and quantifying changing settlement patterns
over time in the Körös region of the Great Hungarian Plain will speak to the nature of social
change during the Late Copper Age. A qualifiable and quantifiable spatial exclusivity between
kurgans and Late Copper Age Baden sites, for example, supports a model that includes the
migration of kurgan builders onto the Plain. If associated with a reduction in density and
frequency of Middle and Late Copper Age sites, such a pattern may suggest a disruption in local
economic and social processes. On the other hand, the presence of Late Copper Age sites near
kurgans or kurgan clusters would make a model of intentional avoidance untenable, and would
not support a model by which major changes on the Plain at this time were catalyzed by
migrating kurgan builders (see Table 4.1). It is imperative to remember, however, that the
arrival of new people on the Plain would not necessarily directly cause major material culture
84
and settlement change in the study region. As discussed elsewhere (see Chapter Three), factors
other than migration may account for both changes in material culture and settlement patterns,
though a migration as a separate event may have occurred at around this time.
Ceramic Analysis: Measuring and Observing Technological Change through Time
The methods of ceramic analysis described in the research examples above allow for
qualitative and quantitative data to be gathered macroscopically and microscopically. While
such a dual approach to addressing the same research questions in a single research project is
unusual, in this case it addresses specific variables best approached using a separate analytical
methods. For example, macroscopic analysis of a fresh break in a vessel fragment allows for
rapid description of features such as firing environment, completeness of raw clay kneading, and
sorting of visible mineral and intentional inclusions. A large number of samples can be quickly
and accurately described, making for a large sample size and large amount of data to interpret.
Petrographic analysis, on the other hand, is better suited for accurate groundmass descriptions of
the ceramic paste. This may include identification of birefringent (a condition in which minerals
refract light in multiple directions, causing the mineral’s color to appear differently when viewed
at different polarizations) fabrics, the identification of specific minerals or mineral classes, the
identification of specific intentional tempers difficult to identify in hand sample, the
classification of void space, and specific qualtitative descriptions of fabrics between samples and
groups. Quantitatively point-counting mineral inclusions for fabric characterization and size
estimates, on the other hand, produces specific ratios of paste and body inclusions. Point
counting requires a great deal more time in terms of both sample preparation and analysis. As
such, a combination of both approaches allows for the characterization and description of
manufacturing techniques and technological approaches of vessel manufacture in a timely,
accurate, and efficient manner.
A ceramic analysis research design making use of macroscopic and petrographic
techniques is also useful in this case due to the unusual design of the petrographic study,
involving the study of change over a long period of time. The majority of petrographic studies
have focused on the identification of variability over a relatively restricted timeframe and
attempted to identify specific fabrics within that timeframe (see Krieter 2005; Reedy 2008). This
allows the researcher to discuss implications for the existence of different fabrics,
85
technologically and in terms of vessel form, design, and function. In this research project, fabric
types are also identified but are used as general representations of fabrics over time, rather than
the basis for comparison. Here, cultural phases (e.g., Middle Copper Age, Late Copper Age) are
the default groupings for comparative purposes. By grouping samples in this way, it is possible
to identify variability over long periods of time, as well as within specific cultural phases. As
such, the results of the macroscopic and petrographic analyses augment one other, and act as a
control to subsequently identify any potential problems with the methodology.
Following collection, data can be analyzed diachronically in order to qualify and quantify
how ceramic technology changed over various cultural phases, beyond the most obvious
characteristics of form and design. Indeed, Krieter (2005) demonstrated the utility of using
petrographic methods to analyze technological change in Early and Middle Bronze Age ceramics
from Transdanubia. Such a technique can also be applied to a research design constructed to
identify changes in manufacturing techniques over the long-term.
Change in form, design, surface treatment, and other seemingly superficial elements of
ceramic manufacture has been well documented in the Körös region during the Middle and Late
Copper Age (see Jankovich et al. 1989; Jankovich et al. 1998), at which point the regionally
homogeneous Baden material culture tradition became dominant on the Hungarian Plain.
Though often attributed to the influence of a foreign population present in the region, such
explanations are not sufficient for explaining material culture change, as the characteristics often
change over a short-term, even generational, basis, and can be attributed to other factors such as
economic integration into wider interaction spheres. However, culturally embedded techniques
for preparation and manufacture tend to be conservative and resistant to change (Lemmonier
1992; Michelaki 1999; see Chapter Three). Therefore, if distinctive changes in the
manufacturing process indicated in paste composition and groundmass appearance are observed
between cultural phases (specifically between the Middle and Late Copper Age), it may indicate
an outside influence on pottery technology, and would lend support to a migration or invasion
model. On the other hand, if little or no change is observed in the technological aspects of
ceramic manufacture over time, a migration model would not be supported. In such a case, a
model of population continuity and incorporation into a wider economic and interaction sphere
would better explain the material culture changes observed at this time (Table 4.1).
86
Summary
In this chapter, I have presented the links between research design and research
methodology and the anthropological and archaeological information presented in Chapters Two
and Three. I discussed the temporal and geographic appropriateness for the research in this
volume, and provided brief research histories of research veins important to the study at hand.
Finally, the framework by which the collected data will be interpreted was discussed and
presented. The following chapter will apply these design and methodological links directly to
the present research by discussing the methodologies utilized for the spatial and ceramic analyses
presented in this dissertation.
87
CHAPTER FIVE
METHODOLOGY
Introduction
In this chapter, I first present a rationale for conducting the present research in the Körös
River study area. I then present the analytical methods used to examine changes in settlement
patterns between the Late Neolithic, Early and Middle Copper Age, Late Copper Age, and Early
Bronze Age, the field methods used for site visitations, mapping, measurement, and systematic
collection, and macroscopic and microscopic ceramic analysis. I then discuss the methods and
purpose of long-term settlement analysis in this project, and explain the principles of
macroscopic and microscopic ceramic analysis and ceramic petrography. The process for
description and documentation of the finds from the site collections and surveys is included.
The selection and preparation of ceramic samples are also discussed.
Selection of the Study Area
The foundations for this research were developed over the course of three seasons of
work with the Körös Regional Archaeological Project (Parkinson 2002, 2006b) and the Bronze
Age Körös Off-Tell Archaeological Project (Duffy 2010). While both of these projects provide
excellent insight into their respective periods of study, the intermediate cultural phase in the
region – the Late Copper Age Baden phase – now remains the least understood in terms of how it
fits into the long-term patterns of social and settlement change on the southeastern Hungarian
Plain.
The area of the Körös River system drainage is an ideal location for archaeological
research in many respects. The location and general characteristics of the landscape are
convenient and accessible, a foundation of previous research is available in a number of
languages with high-quality publications available in serial, monograph, and book forms, and
materials from previous projects are available for those conducting continuing research in the
area.
88
MRT Volume 8
MRT Volume 6
MRT Volume 10
Figure 5.1. The Körös River basin study area, including modern cities and MRT parish boundaries. Located in
southeastern Békés County, Hungary (inset).
Geographic and Archaeological Location
The study area selected for the research includes over 3,000 km2 in the western half of
Békés County in the eastern portion of the Great Hungarian Plain. Known as the Körös-Berretyó
Region or the Körös River Valley, the region is named for, and characterized by, the
river system that dominates the local topography and has produced the complex cultural history
and geomorphology of the region (see Figure 5.1).
Foundation of Recent and Previous Research in the Region
The Körös-Berretyó Region has been the focus of extensive and intensive archaeological
research for decades, and a solid foundation of previous research exists to guide and inform
current researchers. This body of literature, produced by both Hungarian and foreign
researchers, is discussed at length in Chapters Three and Four and will not be given a full
89
treatment here. However, a brief overview is useful in understanding how previous research
contributed to both the corpus of archaeological knowledge in the region, as well as the research
topics addressed in the present research.
Using the Hungarian MRT research, as well as other settlement data on the Hungarian
Plain as a platform, Andrew Sherratt’s Körös-Berretyó project on the Dévaványa Plain in the
1970s and 1980s synthesized decades of pervious research and conducted more intensive
settlement investigation in the extreme northern parishes of Békés County (1983, 1984, 1987a,
1987b). Sherratt’s economic and settlement models, along with the pioneering studies of
Gimbutas (see Gimbutas 1997), helped shape much of the tone of the present research.
Most recently, Parkinson and Gyucha’s Körös Regional Archaeological Project (KRAP)
has focused a critical mass of Hungarian and foreign archaeological researchers on the eastern
Hungarian Plain and the Körös-Berretyó study region in particular. Though Parkinson and
Gyucha’s project began as a vehicle by which to understand the social and cultural mechanisms
that shaped the Early Copper Age apart from the earlier Neolithic, it now includes the study of
virtually all aspects of the transition between the Late Neolithic and Early Copper Age, including
household and settlement organization, settlement patterns, social boundaries, and geological and
geomorphological studies. As more researchers have collaborated with Parkinson and Gyucha,
KRAP has produced several related projects that cover much of the Holocene. These include
studies of the Neolithic (Salisbury 2010), the Early and Middle Bronze Ages (Duffy 2010), and
now the Late Copper Age. In addition to research carried out by local county museums and
other Hungarian researchers, this modern foundation of research offers a rich tapestry of material
from which to formulate new research questions and develop new projects.
Availability of Materials
As a result of the extensive surface survey carried out as part of the production of the
MRT volumes, a great deal of material from the surface of identified sites has been curated at
national and county museums throughout the country. In Békés County, the Munkácsy Mihály
Múzeum stores the majority of items collected as part of the survey in Békés County. I was
granted the opportunity to analyze the ceramic material from all of the collected sites in the study
region, which served as the initial foundation for both the petrographic study and site revisits.
Additionally, colleagues with the Kulturális Örökségvédelmi Szakszolgálat (the Field Service for
90
Cultural Heritage, or KÖSZ), allowed me to both visit and photograph a large Baden site
currently being excavated named Hódmezővásárhely-Kopáncs I., Olasz-tanya. I was also
permitted to take a sample of ceramics from this site to include in my study as a control for the
stratigraphically ambiguous collected surface material. All of this material was incorporated into
the current research project.
Spatial Analysis
Since the 1970s, archaeologists have been concerned with systematizing different
manifestations of archaeological spatial analysis, including settlement analysis, site system
analysis, regional studies, territorial analyses, locational analyses, catchment area studies,
distribution mapping, density studies, and ultimately the integration of these types of information
into large databases (Clarke 1977; Galaty 2005). Each of these forms of spatial study can be
used at particular scales at in particular ways to answer specific archaeological questions (Clarke
1972:47, 1977). This research project is concerned with multiple scales of analysis, and one of
the questions framing the project as a whole is: how do settlement patterns, or changes in
settlement patterns, serve as an indicator of social change? This spatial study aimed to test
models of change involving kurgans in the Körös region of the Great Hungarian Plain during the
Middle and Late Copper Age.
Previous research has reaffirmed continuity between the Neolithic, Early, and Middle
Copper Age on the eastern Hungarian Plain. Parkinson (1999, 2005) noted, along with others
(see Bankoff and Winter 1990), a break in this continuity between the Middle Copper Age
Bodrogkeresztúr and Late Copper Age Baden periods. Stark (1998a:1, 1998b) argued that social
groups and their boundaries are marked by observable patterns in the archaeological record.
Therefore, a study of formal variation in settlement type, location, and/or degree of centralization
is useful to determine the degree of continuity or change in the region.
The goal of the spatial analysis component of this research is to expand upon Sherratt’s
(1983, 1984, 1997a, 1997b, see Chapter Three for a detailed discussion of Sherratt’s research in
the region) previous spatial study in northern Békés County. Since the 1980s, more data has
become available, including the publication of additional MRT data, such as site locations and
site descriptions. Including such an analysis as part of this research project allows for an update
of Sherratt’s research, while at the same time expanding the research both in terms of geography
91
and scales of analysis. Ultimately, the spatial analysis conducted here broadly tested two models
of change during the Late Copper Age: Andrew Sherratt’s environmental and economic model
and the migration and invasion model as popularized by Marija Gimbutas.
Three primary questions were asked in conduction the spatial analysis. 1) Did the arrival
of kurgan people cause dramatic, sudden change during the latter half of the Copper Age? 2)
Are kurgans and Late Copper Age sites spatially complementary, as Sherratt stated? And, 3)
what implications does the spatial relationship between kurgans and Late Copper Age sites have
for understanding social and settlement changes at the end of the Late Copper Age?
Additionally, the spatial analysis broadly examined changes in settlement patterns over time,
from the Late Neolithic through the Late Copper Age, with an emphasis on how settlement
evolved between the Early, Middle, and Late Copper Ages. The questions were approached
from three scales of analysis: 1) The eastern Great Hungarian Plain (using Sherratt’s maps and
data); 2) the Körös River Basin study region (see Figure 5.1); and, 3) several smaller microregions throughout the study region.
A total of 588 presumably invasive kurgans are recorded and published in the MRT
volumes for the study area, along with 70 Middle Copper Age sites and 105 Late Copper Age
sites. This provides a statistically significant sampling universe in terms of visual settlement
pattern analysis, average nearest neighbor analysis, and density analysis.
The creation of a detailed geographic information system (GIS) was the first
methodological step in conducting the settlement analysis. ArcMap 9.3 was used for this
purpose. The principle units of analysis, along with relevant geographic and cartographic data
(rivers, maps, borders, cities) include settlement data on Middle and Late Copper Age sites,
kurgan locations, and information on Late Neolithic, Early Copper Age, and Early Bronze Age
sites. All site locations were recorded in the GIS using georeferenced and rectified digital copies
of the original 1:10,000 survey maps.
Within the Körös River study region, average nearest neighbor analysis was used to
determine level of clustering and randomness within cultural periods, based on the nearest
neighbor index. The nearest neighbor index is the ratio of the actual distance between sites
divided by the expected difference based on the area of study. The expected difference is the
average distance between neighbors in a hypothetical random distribution. If the index is less
than 1, the pattern exhibits clustering. If the index is greater than 1, the trend is toward
92
dispersion. The Z score in nearest neighbor analysis is a measure of statistical significance that
indicates whether or not to reject the null hypothesis, which in this case is that all points are
randomly distributed across the landscape. At a 95% confidence level, a Z score between -1.96
and 1.96 means that the null hypothesis cannot be rejected (Ebdon 1985). The Körös region was
divided into three zones for analytical purposes, and the mean nearest neighbor index and Z
score was used.
Using the nearest neighbor data as a guide, density maps of kurgans were constructed
(based on number per square kilometer) in order to clearly define kurgan clusters, and to
determine if kurgan clusters throughout the Körös region were spatially correlated with Late
Copper Age archaeological sites. Additionally, the creation of site distribution maps of various
cultural periods across the Körös region allows for a detailed analysis of settlement shifts over
long periods of time, at a larger scale (but at an equal resolution) as Sherratt’s previous
settlement analysis.
Fieldwork and Site Revisits
As part of the field component of the research, site revisits and systematic collection were
undertaken in order to address several issues. First, it was necessary to field-check the MRT
assignments of periods represented at each site and to assess, at multi-component sites, the
approximate areas of Late Copper Age archaeological sites since the MRT records only the
distribution of all material at each site, rather than site size by cultural phase. Second, a
systematic collection served the purposes of both establishing an approximate site size and
collecting diagnostic ceramic samples for subsequent macroscopic and petrographic analysis.
Finally, collecting ceramic materials from the surface of many sites throughout the county was
necessary to compile a sample size large enough to effectively identify any local variability in
ceramic production.
Overview of Site Revisits and Collection
Since an extensive amount of previous research had been conducted in the study area
over the last four decades, only a sample of Late Copper Age sites were revisited with the
following goals in mind:
93
1. Field check MRT period data, in order to determine if maps based on MRT settlement
data are an accurate representation of site distribution in the region.
2. Approximate size of Late Copper Age settlement, based upon the distribution of ceramic
material and other artifacts in the plowzone, and the presence of surface features.
3. Collect a representative sample of ceramic samples from single and multi-component
sites in order to macroscopically and microscopically test for change in ceramic
production over time.
Site Selection
According to the MRT, many recorded Late Copper Age sites were not occupied in
previous or earlier periods; however, the vast majority have at least one earlier or later material
culture component. Unfortunately, Late Copper Age sites are often described as thin scatters of
material on the surface, with little diagnostic material present. Using the MRT descriptions as a
guide, all 105 Late Copper Age Baden and Boleráz sites were placed in a database and coded
according to size and cultural components.
The following criteria were used to determine which sites to revisit and potentially
collect:
1. Cultural phases represented, as documented in the MRT site descriptions.
Single component Late Copper Age sites were given priority for re-visitation
and collection in order to maintain chronological control. However, collection
of sites exhibiting only Late Copper Age material was uncommon.
2. Surface representation of ceramic materials, as documented in the MRT site
descriptions. An attempt was made to focus on Late Copper Age sites
exhibiting large quantities of Boleráz or Baden material, in order to measure
relative site size as accurately as possible, and to ensure an adequate
collection for further ceramic analysis.
3. Previous research conducted at the site, as documented in the MRT site
descriptions and other sources. An attempt was made to avoid sites where
previous excavation or systematic collection had taken place.
With these criteria in mind, the ultimate goal was to revisit and collect as many Late Copper Age
sites as possible given the constraints of time, weather, site accessibility, and site condition.
94
Site Collection
During the re-visitation of Late Copper Age sites, it was first determined whether or not
the surface finds warranted a systematic surface collection or survey. The final decision to
systematically collect samples from the surface was made on the basis of: 1) quantity of material
on the surface; 2) culturally identifiable material on the surface, including a significant Late
Copper Age assemblage; and 3) surface visibility, including vegetation and soil condition
(plowing). Many sites described as single component Late Copper Age sites in the MRT were
devoid of surface artifacts upon re-visitation, while several multi-component sites produced very
few or zero Boleráz of Baden artifacts on initial investigation. Additionally, much of the autumn
field season was plagued by long stretches of wet, inclement weather. As a result, a number of
sites were virtually submerged, and several were unreachable due to severe road deterioration.
Once the decision to systematically collect samples from the surface of the site was
made, the extent of the surface scatter was determined by placing fieldwalkers at 15-20 meter
intervals and walking parallel transects, marking diagnostic ceramic material, surface features,
and other artifacts or artifact clusters of interest with pinflags. Since collection of multicomponent sites was necessary, special effort was made to establish the density and extent of the
Late Copper Age assemblage. Once density and size was roughly established, a site datum was
placed in the center of the densest area of Late Copper Age material and its coordinates recorded
with a hand-held GPS unit.
Systematic collection was based on one of two strategies: dog-leash collection units, or
systematic surface survey and collection. Dog-leash collection followed Parkinson’s (1999:184)
methodology. Surface samples were collected using 5 meter radius circular collection units
(“dog-leashes”) at 30 meter intervals (Figure 5.2, see Figures 6.3-6.13). Each 78m2 unit was
named according to its relative position to the established datum (such as, Center, North 30,
North 60) and its coordinates recorded with a hand-held GPS unit. Collection continued at 30
meter intervals in the four cardinal directions until the number of surface finds dropped below
five artifacts for two consecutive collection units.
In many cases, the placement of collection units extending in four cardinal directions
would not accurately define the size or shape of the site, or would be limited by topographic
features (canals, roads). As such, the units were slightly skewed in order to more efficiently
95
Figure 5.2. Systematic “dog-leash” collection unit site collection strategy.
collect a site. In some of these cases, or when the number of collected diagnostic ceramics was
insufficient for further analysis, additional collection units were conducted at 45-degree angles
from the site datum (unit Center), beginning at 15 meters from the datum and subsequently
conducted at 30 meter intervals.
Every artifact was collected within each unit and sorted into different categories
(including pottery, daub, lithics, bone, metal, and shell). Each category was counted, weighed,
and recorded in the field. Diagnostic ceramic material (including rims, bases, and incised or
decorated body sherds) were sorted separately, counted, bagged, and saved for further analysis.
Lithic materials, daub samples, faunal samples, and human remains were also saved for potential
further analysis.
96
In cases where significant artifacts (e.g., large diagnostic ceramics), surface features (e.g.,
houses, plowed-out graves), or significant clusters were discovered during collection but not
included in the systematic collection units, additional collection of features or clusters was
conducted. Individual artifacts were piece-plotted on maps, their location recorded with a
handheld GPS unit, and documented as “special finds” (SF). Artifact clusters were collected,
their location recorded, and documented as “extra units” (XU). Specific features were collected,
their location recorded, and documented as “feature” (F).
At particularly large sites, or when large sites exhibited a low surface density but unusual
numbers of diagnostic ceramics, a systematic linear survey was conducted in order to
approximately measure the site and collect a representative ceramic assemblage for further
analysis. Fieldwalkers were placed at 15-20 meter intervals and walked parallel transects until
no more artifacts were seen on the surface. Pin flags were placed at the location of diagnostic
ceramics. The ceramics were later collected. Their location was documented with a handheld
GPS unit, and recorded as special finds.
This combination of collection strategies proved efficient and effective at determining the
approximate size of sites and collecting a systematic sample of archaeological material from the
surface. Unfortunately, since most of the collected and survey sites were multi-component and
had two or more periods represented on the surface, assessments of site size were difficult and,
while occasionally useful, should be used with caution. The general impression, however, is that
Late Copper Age sites, with rare exceptions, are relatively small and ephemeral in comparison
with sites of other cultural affiliation in the study region.
Description and Documentation of Finds
The collected finds from each site were washed, assigned field specimen (FS) numbers,
and cataloged upon returning from the field. All material was measured, weighed and described.
A site collection report, diagnostic ceramic sort by unit, unit collection forms, and FS logs were
entered into a collection database. With the exception of diagnostic ceramics, which were
needed for further analysis, artifacts were organized for curation by site, collection unit, and FS
number.
97
Ceramic Coding
Every diagnostic sherd was described and recorded in the ceramic analysis database in
order to record variables useful for identifying characteristics of production and manufacture.
Certain stylistic attributes were also recorded in order to identify any variability or trends within
Late Copper Age pottery. The variables coded for each sample include descriptive variables
such as cultural affiliation, weight, thickness, and length, as well as variables designed to record
characteristics of production and manufacture. These include fabric characteristics (hardness,
feel, texture, grain size, kneading, identifiable natural mineral inclusions and/or artificial
inclusions, inclusions sorting of those inclusions, and color and structure of the firing core),
surface finish (exterior surface, interior surface, decoration), and firing characteristics (paste
color and surface color).
Special attention was paid to variables and variable sets that indicate raw material
preparation, production, or vessel formation techniques (i.e. technological or functional
variables). These variables, such as data on firing characteristics and fabric descriptions, record
embedded information on practices that, unlike decorative techniques, are resistant to change
over time. These variables in particular are useful for analyzing change over time, and to
determine if changes in form and design indicate the arrival of new personnel bringing their own
techniques with them, or if production and manufacture remained unchanged and decorative
changes resulted from other processes.
Other Materials
Though potsherds comprised the majority of material collected from each site, other
materials were often encountered. These included chipped stone (lithics), grinding stone
fragments, animal bones, and rarely human bones and metals. With the exception of faunal
material (which was sampled), all of this material was collected, washed, documented, and
curated.
Photography
Every artifact collected from the field was photographed obverse and reverse using a
Nikon D60 digital SLR camera with an 18-55mm lens. Photographs were imported onto a
computer in .tiff format and maintained at the highest possible resolution.
98
Current Location of the Material
Material systematically surface collected from sites during the fall of 2009 is curated at
the Munkácsy Mihály Múzeum in Békéscsaba, along with the previously collected MRT
material included in this study and the pottery from Doboz Homokgödöri-tablá. The material
from Hódmezővásárhely-Kopáncs I., Olasz-tanya is curated by the KÖSZ.
Ceramic Analysis
Macroscopic Ceramic Analysis
As described in detail above, every potsherd collected during survey was subject to
coding of numerous descriptive and functional variables, with a focus on those that provide
information on production technology. The macroscopic research portion of this study aimed to
observe any changes in ceramic production between the Middle Copper Age Bodrogkeresztúr
and Late Copper Age Baden phases, and to identify any interregional or within-region inter-site
variability within the Late Copper Age. Material for this study included ceramic sherds collected
as part of the MRT survey, material collected as part of the field component of this dissertation
research, described above, ceramics from the excavated context of the Doboz Homokgödöritablá Late Copper Age site, and ceramics from excavated contexts from HódmezővásárhelyKopáncs I., Olasz-tanya in the adjacent Maros River watershed. All MRT materials were coded
at the Munkácsy Mihály Múzeum in Békéscsaba, Hungary during the summer of 2009, while the
ceramics collected during the field component of this research were collected and coded in the
fall of the same year.
Sherds were coded according to multiple variables, including diagnostic type, thickness,
hardness, feel, texture, grain size, completeness of kneading, natural and intentional inclusions
(e.g., clay nodules or grog), inclusion sorting, color, surface treatments, and decoration. In total,
352 of 509 total analyzed and coded sherds were assigned to the Middle and Late Copper Age.
Of these, 303 sherds came from surface contexts at sites in the Körös study region, 21 from the
excavated site of Doboz Homokgödöri-tablá in the Körös Region, and 28 from the ongoing
excavation of a Baden settlement near Hódmezővásárhely in the adjacent Tisza River watershed.
The Hódmezővásárhely ceramics are included as a counterpoint and control for the Körös region
ceramics, and in order to observe any interregional variability (see Tables 6.1 and 6.2 for a
summary of collected ceramic material).
99
Microscopic Ceramic Analysis
Of the 509 sherds coded during the fall 2009 field season, 147 were selected for
microscopic petrographic analysis. Sherds were selected based on diagnostic characteristics –
primarily incised decoration and patterns of punctation – that unambiguously assigned them to
specific known chronological phases. This allowed variability between the samples to be
accurately described over both space and time.
Prior to analysis, every sherd was thin sectioned with the assistance of Dr. Attila Kreiter
and his colleagues at the Ceramics Laboratory of the KÖSZ in Budapast, Hungary. Thin
sectioning first involved removing a small sample from the sherd, polishing the fresh break, and
mounting the sample onto a glass slide using an impregnative epoxy. The sample was then
ground to a thickness of .02-.03 mm, the point at which light characteristically passes through
clay groundmass and mineral inclusions and allows for specific description and identification.
Thin sections were then analyzed under a polarizing light microscope in Tallahassee,
Florida. A mechanical “click-stop” stage (a stage that moves a slide a set distance and direction
with the turn of a knob) facilitated point counting. Counts were made at 2 mm intervals as per
Stoltman’s methodology (1989, 1991). As many as 150, but never fewer than 100, points were
counted per slide. Qualitative description followed Whitbread’s methodology and classification
scheme (1995). Data were recorded on quantitative and qualitative collection sheets and entered
into a Microsoft Access database for sorting and further analysis (Figures 5.3 and 5.4). Recorded
quantitative data included point counts for matrix, voids, natural inclusions, temper inclusions
(including clay nodules), and ratios of matrix, sand, silt, and temper for body and paste. Natural
and temper inclusions were tallied by size, ranging from silt at .25 mm to gravel at 8 mm. Also
calculated were sand size index and temper size index, which are ordinal numerical indexes
between 1 and 5 that represent a general size assessment of natural and intentional inclusions
(Stoltman 1989). General qualitative description included description of void size and shape,
kneading, natural inclusion grain distribution, sorting of natural mineral inclusions, groundmass
description (including crystallitic birefringence and birefringent fabric description, or inactive
groundmass description), and color. Although samples were not analytically divided into fabric
types within cultural periods, samples were assigned to fabric types according to Riederer’s
(2004) classification scheme. Microphotographs were taken of every slide’s fabric, and of
important features of a ceramic sample (such as evidence of clay mixing, rare minerals, lithic
100
inclusions, distinctive grog, and so on). To ensure objectivity, slides were analyzed blind.
Additionally, slides were counted in groups of mixed cultural period and provenience, and
results were tabulated upon completion of the petrographic analysis.
Summary
In this chapter, I outlined the specific methodology used in the field and in the laboratory
to address the research questions at hand in this project. The methods applied to this research
and outlined in this chapter function to observe any variability, macroscopically and/or
microscopically, in prehistoric ceramics from the Körös Region of the Great Hungarian Plain.
Any distinctions delineated between cultural phases, especially those related to manufacturing
and production technology, may indicate the presence of new personnel in the region and support
a hypothesis of migration, invasion, or intrusion onto the Plain to account for social, settlement,
and material culture changes. On the other hand, absence of marked variability, or continuity in
production methodology over time, may indicate that local populations adopted foreign traditions
as part of a wider shift in economy and social interaction in the Carpathian Basin at this time.
101
Figure 5.3. Coding sheet for quantitative ceramic petrography.
102
Figure 5.4. Coding sheet for qualitative ceramic petrography.
103
CHAPTER SIX
ARCHAEOLOGICAL SITES AND ASSEMBLAGES
Introduction
In this chapter, I present and discuss the four different sources of material analyzed as part of
this research project:
1) The Magyarország Régészeti Topográfiája (MRT) series (see Ecsedy et al. 1982;
Jankovich et al. 1989; Jankovich et al. 1998) and their associated illustrated and 1:10,000
topographic maps.
2) A sample of the ceramic material gathered from archaeological sites in the Körös River
Basin study region during the MRT surveys (Table 6.1);
3) Fieldwork conducted in the autumn of 2009 at a sample of archaeological sites in Békés
County in the Körös River Basin study region and in the Maros alluvial fan, and the
ceramic samples collected during this fieldwork (Table 6.2);
4) Ceramic material from excavated contexts at the archaeological sites of Doboz
Homokgödöri-tábla and Hódmezővásárhely-Kopáncs I., Olasz-tanya (Tables 6.1 and
6.2).
Each source will be discussed in a separate section. Descriptions of all visited sites are
provided, and detailed descriptions of surveyed and collected sites, including maps, are included.
MRT Sites
The MRT volumes 6, 8, and 10 (Ecsedy et al. 1982; Jankovich et al. 1989; Jankovich et
al. 1998) list a total of 105 sites containing Late Copper Age material. Of these sites, five are
listed as single component sites, while the remaining sites also contain material dating to other
periods. Table 6.3 provides information on all relevant sites as described in the MRT, and Table
6.4, and Figures 6.2-6.13 provide an overview of the relevant sites included in this analysis.
They are listed alphabetically by parish name (e.g., Tarhos) and site number (e.g., Tarhos 67 =
Kifli Domb). The various periods documented at each site by the MRT are included in Tables
6.1 and 6.2.
Site sizes listed in the table are based on site sizes as measured by the MRT fieldwalkers
and as they are indicated on the original 1:10,000 topographic field maps used during the survey.
As such, they indicate the site size as a whole, meaning that all periods are included in the
104
Figure 6.1. Sites recorded in the MRT revisited during the fall 2009 field season (n=49).
estimation. This is problematic, as material from the various periods is often clustered in certain
areas of the site. As such, when multi-component sites were collected, an effort was made to
delineate the Late Copper Age artifact scatter and estimate site size based on collection rather
than site size as indicated by the MRT.
Sites Revisited During Fieldwork
A sample of Late Copper Age sites described in the MRT were revisited during fieldwork
during the fall of 2009. Single component Baden sites were given priority for revisitation.
However, low surface densities or an absence of surface representation required the visitation
and collection of multi-component sites. The specific criteria used to select sites for revisitation
and collection are given in Chapter Four. A total of 49 sites were revisited between October and
December 2009 (Figure 6.1).
105
Bucsa 13
Füzesgyarmat 97
Szeghalom 80
Biharugra 33
Körösladány 21
Bélmegyer 82
Mezőberény 34
Bélmegyer 32
Békés 26
Békés 178
Gerla 64
Figure 6.2. Sites systematically collected during the fall 2009 field season (n=11).
Sites Collected and Intensively Surveyed During Fieldwork
Of the 49 sites revisited during the fall 2009 field season, 11 featured enough Late
Copper Age material visible on the surface to warrant either collection or transect survey. Site
summaries for these sites are provided in Table 6.4, and more detailed description of collected
material is found in Appendix A. The specific criteria for selecting sites to collect and
determining which method to use for collection are outlined in Chapter Four. Collection
methods and analytical techniques are also outlined in Chapter Four. The following section
gives a brief geographic, geomorphological, and archaeological description of each collected
sites. A summary of materials collected from each site is also provided.
106
Békés 26 – Kászmánkert I
Békés 26 (see Figure 6.3) is located just to the south of the town of Békés in the eastern
part of the parish. The site is located in the center of a small settlement just outside of town, and
is surrounded by houses. At least a few of the structures appear to have been built in the last 1015 years and have destroyed part of the northern section of the site. Jankovich et al. (1998:5960) described the area as a large, multi-component site containing ceramics corresponding to the
Early Neolithic Körös and Middle Neolithic AVK, Early Copper Age Tiszapolgár, Late Copper
Age Baden, Middle Bronze Age Ottomány, Szarmation, and Avar periods.
At the time of research, the site was divided into several fields with different vegetation
and soil conditions. The easternmost fields were covered in vegetation with very little surface
visibility, and a rectangular field in the western portion of the site containing a small structure
was completely overgrown. The four fields that were surveyed had recently been plowed, with a
small amount of re-growth making for visibility of between 75-85% on average. Based on
artifact density on the surface, it was decided to conduct a transect survey before attempting a
systematic dog-leash collection of such a large, complex multi-component site. In total, 12
transects were walked with field-walkers spaced at 20 meter intervals. Prehistoric diagnostic
sherds were flagged, while non-prehistoric sherds were noted and not flagged, collected, or piece
plotted. In addition to prehistoric ceramic fragments, animal bone (including cow), and a small
amount of chipped stone (obsidian) were observed on the surface. Small amounts of glazed
modern pottery were present on the surface, and Szarmation material was thinly scattered across
the site, but not flagged or piece plotted. A total of 11 surface finds were collected and pieceplotted with GPS. One sherd dated to the Early Copper Age Tiszapolgár phase, five to the Late
Copper Age Baden period, three to the Early Bronze Age, and two dated to the Middle Bronze
Age. An accurate site size estimate is difficult due to recent construction and fields that have
been left to grow fallow since Jankovich et al.’s (1998) estimate of approximately five hectares;
however, artifact distribution was roughly congruent with the site boundaries indicated on the
1:10,000 MRT topographic map.
Békés 178 – Lápos-Domb
The site of Békés 178 (see Figure 6.4) is located on a low 1.5 meter rise along the bank of
an irrigation canal, formerly a streambed in prehistory. The site is located approximately 300
107
meters to the northeast of the modern Kettős Körös River. Jankovich et al. (1998:115) described
the site as a scatter of prehistoric material running north-northeast along the natural levy above
the channel. At the time of MRT research, field walkers observed Middle Neolithic Szakálhát,
Late Copper Age Baden, and Middle Bronze Age Ottomány ceramic material.
At the time of the present research, the conditions were very wet, the site had been
plowed and disked, and a large amount of grassy overgrowth covered the surface, making for a
surface visibility of only 40%. A transect survey was conducted in order to gather more
information about the site. Six transects were walked with the contour of the landscape and field
in a southeast-northwest orientation with field walkers placed at 20 meter intervals. Prehistoric
diagnostic sherds were flagged, and a total of 14 diagnostic ceramic surface finds were collected
and piece plotted with GPS. Three sherds could not be classified according to period, one sherd
dated to the Early Copper Age Tiszapolgár period, four to the Late Copper Age Baden phase,
four to the Early Bronze Age Makó culture, and two to later periods. A site size estimate based
on finds of 1.5 hectares is roughly comperable to Jankovich et al.’s (1998) estimate as drawn on
the MRT 1:10,000 topographic map.
Bélmegyer 82 – Cserszád I
Located near the center of the parish and just north of the town of Bélmegyer, the site
(see Figure 6.5) consists of approximates 1.6 hectares and runs roughly 200 meters northeast to
southwest across three agricultural fields, and is between 50 and 100 meters wide along this
distance (Jankovich et al. 1998: 365). The site sits atop a small loess rise and is bordered to the
north by a defunct streambed. Jankovich et al. (1998:365) found some AVK material on the
surface of the site, as well as classic Late Copper Age Baden material and material with incised
and punctated design that corresponded to the Early Late Copper Age Boleráz period. They also
located material that dated to the Middle Ages.
At the time of the research, the three fields in which the site is located were in various
states of production. The north 1/3 of the site was under clover and fodder and had not been
plowed in at least two years. The central field was under thick layer of clover and had not been
plowed in at least one year. The southernmost 1/3 of the site had recently been plowed, with very
little overgrowth. Collection was done using 78.5m2 collection units (5 meter radius “dog-leash
units”). A total of 12 units were collected in this manner, for a total of 942m2. Visibility was
108
quite poor across most of the site (30% on average), but a total of 137 prehistoric sherds were
collected and sorted. A total of 11 diagnostic prehistoric sherds were recovered from three of the
12 units, with all of the diagnostic sherds coming from within 30 meters of unit N60. However,
the scatter of material including ceramics, animal bone, and daub, is roughly consistent with
Jankovich et al.’s (1999) assessment of 1.6 hectares of area. Two of the diagnostic ceramics
could not be identified to prehistoric material culture period, and two Neolithic sherds (one of
them AVK) were recovered. Three Boleráz sherds were recovered, as well as four Late Copper
Age sherds that could not be distinguished between Boleráz and Baden. One Szarmation sherd
was collected, and several modern ceramics were noted on the surface. Additionally, 700 grams
of daub were collected from the surface, as well as five yellow chert chipped stone fragments. A
total of 196 grams of faunal material was collected from the surface of the site.
Biharugra 33 – Kincses Tanya
Located in the far southwestern corner of Biharugra parish, the site (see Figure 6.6) is
located in a small agricultural field approximately two hectares in area, and approximately 100
meters north of an agricultural processing center. The field is bordered on the north by a levy
approximately 2m high, and is bordered on the south by a shallow irrigation ditch. Beyond the
ditch to the south was a grassy, unplowed field. Escedy et al. (1982: 21) described the site as
having prehistoric ceramics on the surface, specifically sherds that corresponded to the Early
Copper Age Tiszapolgár phase, and some sherds that corresponded stylistically to the Late
Copper Age Baden phase.
At the time of research, the site had recently been harvested of a cereal crop and plowed.
The site was also extremely muddy, which made initial field sorting of diagnostic ceramics
difficult. On the other hand, very little new growth existed on the surface, making for an
excellent average surface visibility of 92.5%. Initial field walking revealed Tiszapolgár lugs,
Baden sherds, and several Early Bronze Age Ottomány decorated body sherds. Subsequently,
collection was done using 78.5 m2 collection units (5 meter radius “dog-leash” collection units).
A total of seven units were collected in this manner, for a total intensively collected area of 453
m2. One Special Find (SF1) of two Baden sherds was also collected.
A total of 229 body sherds (6,400 grams) were collected, along with 80 pieces of daub
(2,250 grams). The vast majority of the daub came from collection unit S60, where a dense daub
109
concentration covered an approximately 16m2 area of the surface (Feature 1). A total of 43
diagnostic sherds were collected, including 16 rims, one base, 22 decorated body sherds, two
handles, and two undecorated body sherds. Eight of the sherds were clearly prehistoric but could
not be assigned to cultural period. Interestingly, we collected no Tiszapolgár sherds from the
collection units. Three sherds were assigned to the Middle Copper Age Bodrogkeresztúr period,
23 sherds to the Baden period, two to the Early Bronze Age Makó period, and seven to later
periods. Most of the later material could be assigned to the Middle Bronze Age Ottomány phase.
Although no faunal material was observed on the surface, two small pieces of chipped stone (one
chert and one obsidian) were collected.
Bucsa 13 – Kis Kecskés
The site of Bucsa 13 (see Figure 6.7) is located on the western bank of a defunct
streambed, just to the north of the multi-component archaeological site of Bucsa 12, which
shares the same geographic name. Escedy et al. (1982:32) described the site as containing a
small Late Copper Age Baden component as well as Szarmation and Árpádian material.
At the time of research, the site was extremely wet, and had been replanted with a small
amount of new crop growth. Average visibility was moderate at approximately 75-80%.
Unfortunately, a large amount of standing water was present on the northern and western
portions of the site, which made the inundated areas unwalkable. At this site, a total of six
transects were spaced at 20 meter intervals and walked in a northeast to southwest direction
along the natural topography of the paleochannel. Despite a relatively steady distribution of
prehistoric ceramic sherds across the surface of the site, only three diagnostic samples were
collected during the investigation (57 grams). These samples were collected and piece plotted
with GPS. All three collected samples were decorated body sherds, two of which were assigned
to the Late Copper Age Baden material culture phase and one to the Middle Bronze Age
Ottomány phase.
Füzesgyarmat 97 – Pázmán
The site of Füzesgyarmat 97 (see Figure 6.8) is located to the northeast of the town of
Füzesgyarmat in the northeastern region of the parish. It is located just to the south of and across
a small irrigation canal from Füzesgyarmat 96. Escedy et al. (1982:96) described the site as
110
containing a Late Copper Age Baden component based on the collection of punctated and incised
ceramic body sherds, as well as characteristically thick-walled rim fragments. Additionally, the
MRT surface survey observed Szarmation and Árpád ceramics.
At the time of research the site had recently been plowed and disked, with very little
recent re-growth on the surface, making for visibility of 85%. The site is bordered on the northnortheast and east by irrigation canals, but neither of the canals appears to truncate the site. An
initial survey revealed a thin but consistent scatter of prehistoric material across the surface of
the site, including prehistoric ceramic fragments and a small amount of animal bone. Few
diagnostics were observed, but it was decided to conduct a transect survey in order to gather
more information about the site, including site size and distribution of artifacts of different
periods. Eight transects at 20 meter intervals were walked in a northeast to southwest direction,
with the orientation of the field. Flags were placed at the location of each diagnostic ceramic,
and every collected ceramic was piece-plotted with GPS. A total of seven diagnostic ceramics
were collected from the site, including four decorated body sherds and three rims, for a total of
94g of collected ceramic material. Of the seven diagnostic sherds, five were assigned to the Late
Copper Age Baden material culture group based on design characteristics, while two sherds
dated to the later Szarmation period. Based on artifact distribution across the surface of the site,
a site size estimate of one hectare is roughly congruent with Escedy et al.’s (1982:96)
assessment.
Gerla 64 - Veres Gyűrűs
The site of Gerla 64 (see Figure 6.9) is located in the northeastern portion of the parish, to
the northeast of the village of Gerla. It is located on the west bank of a defunct prehistoric
streambed atop a low redeposited loess rise. Jankovich et al. (1998:451) noted the presence of
Middle Neolithic Szakalhát ceramic fragments as well as Late Copper Age Boleráz material
scattered across the surface of the site.
At the time of research, the surface had been plowed and disked within a few months, and
a large amount of regrowth obscured the surface, making for an average surface visibility of only
28.6%. Additionally, conditions were very wet. Collection was done using 78.5 m2 circular
collection units (5 meter radius “dog-leash” units), and a total of seven units were collected for a
total of 549.5 m2 intensively collected. Due to the very small size of the site, intervals between
111
collection units were placed at 15 meters rather than the normal 30 meters. In total, 127 body
sherds were collected, with the vast majority of material coming from within 15m of the site
center. Twenty-nine miscellaneous fragments of faunal material were collected (77 grams), and
127 pieces of daub were collected and weighed in the field (2,600 grams). Approximately 80%
of the daub came from Unit C, where a dense daub scatter covered the surface over an area of
approximately 15m2 (Feature 1).
A total of 18 diagnostic ceramics were collected for further analysis, with 11 of those
coming from Unit C, spatially associated with Feature 1. Two prehistoric sherds were collected
that could not be assigned to cultural period. Seven sherds were assigned to the Middle
Neolithic Szakalhát phase, eight to the Late Copper Age Boleráz phase, and one to the Early
Bronze Age. The very small size of the site (less than .5 hectares) is roughly consistent with
Jankovich et al.’s (1998) assessment of the site.
Körösladány 21 – Tekerő
Located in the southeastern area of the parish near its eastern border with Szeghalom
parish, the site (see Figure 6.10) proved difficult to locate due to its small size and modest scatter
of prehistoric material. Escedy et al. (1982:107) described the site as containing a scatter of
prehistoric ceramics, some of which could be assigned to the Baden period of the Late Copper
Age based on punctated and incised decoration. Only Baden diagnostic ceramics were located as
part of the MRT surface collection.
At the time of research, the hayfield in which the site was located had recently been
plowed and disked, with some chaff remaining on the surface. After walking the site and
attempting to locate concentrations of prehistoric material, we decided that although the material
was likely Late Copper Age, not enough material was present on the surface to warrant a
collection. However, in order to collect a small sample of diagnostic material from the site, we
conducted a transect survey with field walkers walking transects across the site at 20 meter
intervals. Diagnostic ceramics were flagged, piece plotted in GPS, and subsequently collected
for further analysis. A total of eight transects were walked in this manner. Only a light scatter of
prehistoric material over an area of less than one hectare was observed, and only two Late
Copper Age Baden diagnostic ceramics were recovered from the surface of the site. One small
112
flake of obsidian was also recovered; very little daub and faunal material was seen on the surface
of the site.
Mezőberény 34 – Frei-Tanya
Mezőberény 34 (see Figure 6.11) is a relatively small site in the northeast portion of the
parish, just to the north of the Kettős-Körös River and adjacent to the river’s levy and an
unpaved service road. Jankovich et al. (1999:561) described the site as a dense artifact scatter on
the surface approximately 200x100 meters in area. The MRT description lists finds as Middle
Neolithic AVK ceramics, prehistoric sherds resembling Late Copper Age Classic Baden
ceramics, and a number of rim sherds assigned to the Árpád, or Hungarian Conquest, period.
The Baden finds were described as straight rims, multiple lines of linear punctations, and
crosshatched incisions common to the period.
At the time of research, the sunflower field in which the site is located had recently been
plowed and disked, making for an average surface visibility of 76.6%. Located on a small
crescent-shaped rise approximately .5 meter in elevation, we located Neolithic, Late Copper Age,
Baden, Árpádian, and a small number of modern sherds after walking systematic transects across
the site. Unfortunately, it appears that one of the areas of the site densest with cultural material
may have been truncated and destroyed during the construction of the levy, road, and ditch at the
south end of the site. Collection was done using 78.5 m2 collection units (5 meter radius “dogleash” collection units). A total of six units were collected in this manner, for a total of 453 m2
intensively collected.
A total of 270 body sherds (3,000 grams) were collected, along with 8 small pieces of
daub (200 grams) and five unidentifiable animal bone fragments (<100 grams). An
unfortunately small number of 13 diagnostic sherds were collected, including five rims, two
bases, five decorated body sherds, and one lug. Four of the sherds could not be definitely
assigned to a cultural period. We collected two Neolithic sherds, and two sherds assigned to the
Late Copper Age Classic Baden phase. Based on unit collection, my site size estimate of less
than 1 hectare is not consistent with Jankovich et al.’s estimate (1999:561). However, a thin,
inconsistent scatter of material continued beyond the area intensively collected, and it is likely
that their estimate includes this area.
113
Szeghalom 80 – Dió-Ér
The site of Szeghalom 80 (see Figure 6.12) is located in the southwestern portion of
Szeghalom parish adjacent to the Dió-Ér irrigation canal. This portion of the canal was a natural
channel in prehistory, and the site itself sits atop an alluvial deposit approximately 1m in
elevation above this streambed. Escedy et al. (1982:154) described MRT finds as large amounts
of animal bone and daub across the surface of the site, as well as a number of decorated body
sherds corresponding to the Neolithic period, the Early Copper Age Tiszapolgár phase, the Late
Copper Age Baden phase, the Early Bronze Age Makó culture, a relatively small number of
Szarmation sherds, and some large diameter sherds described as large Árpád cooking vessels.
Between 1971 and 1973, István Escedy excavated 400 square meters of the site in an
effort to investigate the site’s Baden and Neolithic components (Escedy 1973, 1973a).
Ultimately, a Szakálhát component of the site was discovered, and Bodrogkeresztúr and Makó
sherds were also uncovered during the excavation. However, the primary focus was a Baden
settlement (including a small Boleráz component). Located 40-50 cm below the modern ground
surface, the Late Copper Age layers consisted of a large amount of daub interpreted as a house
structure, several areas of burned material, and 14 middens/pits.
At the time of research as part of this project the site had recently been plowed and
disked, making for an average surface visibility of 87.5%. An initial survey of the site revealed
Neolithic, Early and Middle Copper Age, and Late Copper Age ceramic material. As a result of
the variety and density of material on the surface, a decision was made to collect the site
intensively despite the previous excavations. Collection was done using 78.5 m2 collection units
(5 meter radius “dog-leash” collection units), and a total of 16 units were collected for a total of
1,256 m2 intensively collected.
A total of 587 body sherds were collected (9,300 grams), along with 341 pieces of daub
(7,000 grams). The majority of the daub on the surface of the site came from Unit C and two
units just to the east, E30 and E60. However, notable amounts of daub were also collected in
units W30, W60, S30, N30, NE15, and NE45. Most daub, therefore, was concentrated within
about 60 meters of the site center at the top of the loess deposit, and tapered off downslope. A
total of 78 diagnostic sherds were collected from the surface of the site, including 26 rims, 46
decorated body fragments, three lugs and handles, and three other sherds. Of those, 24 could not
be identified to cultural period and were classified as general prehistoric sherds. Six Neolithic
114
sherds were collected, including one Early Neolithic Körös sherd, four Middle Neolithic AVK
sherds, and one sherd classified as general Neolithic. Seven Middle Copper Age
Bodrogkeresztúr sherds were collected, as well as 20 Baden sherds and four Boleráz fragments.
Eleven Early Bronze Age sherds were recovered, and six sherds dating to later periods were also
found. In all, the finds closely correspond to Escedy et al.’s (1982) initial assessment of the site.
Additionally, the site size estimate based on unit collection matches the MRT estimate of 2.73
hectares, as measured from the site boundary drawn on the 1:10,000 topographic map.
Tarhos 67 – Kifli Domb
Located in the northwestern part of Tarhos parish, northwest of the village of Tarhos and
just to the east of the sites Békés 34 and 35, the site (see Figure 6.13) is located just to the east of
a 19th century levy that serves as the boundary between Békés and Tarhos parishes. The site
itself sits on a one meter high crescent-shaped redeposited loess rise on the bank of a small
defunct fluvial streambed. There is a small irrigation canal that bisects the site running roughly
east to west, and acts as a field boundary. Jankovich et al. (1998:666) recorded large numbers of
Late Copper Age Baden diagnostic sherds with punctated linear decoration, as well as incised
decoration typical of the Classic Baden phase. They also noted some sherds identified as
belonging to the Middle Neolithic AVK period.
The half of the site south of the canal had recently been plowed at the time of research
with very little overgrowth, while the half north of the canal was still under corn and overgrown
with thistles. The southern portion was therefore ideal for surface collection, while the northern
half was not collected due to a zero visibility condition. Collection was done using 78.5 m2
circular collection units (5 m radius “dog-leash units”) at 30 meter intervals. A total of 13 units
were collected in this manner, for a total intensively surveyed area of 1,020.5 m2. Nondiagnostic body sherds were scattered across the site, and 469 total sherds (15,950g) were
collected and sorted. A total of 63 diagnostic sherds were recovered in the 12 units, with ten
sherds (“Special Finds”) collected and piece-plotted onto the map (see Figure 6.13). Seven of
the diagnostic sherds could not be assigned to a specific cultural phase, and six sherds belonging
to the Middle Neolithic AVK material culture were collected. One Avar (Middle Ages) sherd
was collected. The majority of diagnostic sherds recovered from this site correspond with Late
Copper Age design and manufacture, including 48 Classic Baden sherds and one probable
115
Boleráz sherd. Additionally, we recovered four fragments of chipped stone, one ground stone
axe or hammerhead, a polished canine tooth, and 297 grams of miscellaneous faunal material.
Interestingly, we observed and collected very little daub (<100 grams) at the site, though a thick
daub scatter (Feature 2) was located in the northwest area of the site. A number of modern
ceramic sherds were present on the site. Additionally, a sub-adult human maxilla fragment and
phalange were collected from the site surface (Feature 1).
The majority of the collected material came from the seven units nearest the center of the
site, within an area of 0.6 hectares. This is not consistent with the six hectare area derived from
the 1:10,000 topographic MRT map created by Jankovich et al. (1998); however, since much of
the site was not collectable at the time of research, the entire area of the site remains uncertain.
Previously Excavated Sites
Doboz Homokgödöri-tablá
The site of Doboz Homokgödöri-tablá, located near the town of Doboz in central Békés
County, lies just outside of the study area proper, but within the Körös River watershed. The site
was excavated by Megyesi (1982, 1983) between 1980 and 1982, and consisted primarily of a
Szarmation settlement. However, several Late Copper Age Baden pits were excavated at the
site. Though the Baden pits themselves were unsubstantial and yielded little in the way of
artifacts, a number of ceramic fragments were recovered. Twenty-six diagnostic ceramics from
this site were included in the ceramic analysis.
Hódmezővásárhely-Kopáncs I., Olasz-tanya
At the time of writing, this site near the town of Hódmezővásárhely in Csongrád County
in southern Hungary continues to be excavated by archaeologists from the KÖSZ. Though the
work is ongoing and the results will not be published for some time, the KÖSZ allowed me to
take a sample of 28 diagnostic ceramics to include in this analysis. The ceramics came from a
variety of feature types, including trash pits, pits inside of houses, and house floors. Though they
originate from the adjacent Maros River watershed, these samples are included as a counterpoint
to the Körös material, and to observe any potentially inter-regional variability in production
technology.
116
Summary
In this chapter, I have outlined the sources of the data analyzed in Chapter Eight. The
data analyzed in this dissertation originates from four sources. This data originates from
multiple sources, and each source has its own benefits, drawbacks, and makes separate
contributions to the research project.
The spatial data derived from the MRT volumes (Escedy et al. 1982; Jankovich et al.
1989; Jankovich et al. 1998) allow synchronic and diachronic patterns to be studied at the
regional level, while measurements and observations made during site revisitations allows for a
refinement of the MRT data and analysis at a finer spatial resolution. Analysis of the ceramic
material collected during the MRT surveys allowed for the rapid collection of a large amount of
macroscopic ceramic data, while also ensuring that ceramic samples for both the macroscopic
and microscopic portions of the analysis were included from a large number of sites across the
entire study region. Fieldwork and site revisits in the fall of 2009 allowed me to field-check the
MRT spatial data, while simultaneously refining site descriptions and the spatial distribution of
sites of different periods across the study region landscape. The diagnostic ceramics collected
during this fieldwork bolstered the sample number for ceramic analysis, while also providing site
level spatial data based on the distribution of surface ceramics. Finally, the ceramics from
excavated contexts provide excellent comparative data, in terms of inter-site, inter-regional, and
intra- regional comparison.
As a whole, this collection of data from various sources allows for a more complete and
integrated analysis of changes in settlement and material culture during the Late Copper Age on
the Plain. The goal of the following chapters, therefore, is to apply the information derived from
the combination of these datasets to interpret these changes. Chapter Seven focuses on the
results of the spatial analysis, and Chapter Eight presents the results of the macroscopic and
microscopic ceramic analysis. Ultimately, the results of the analysis will focus specifically on
understanding settlement and material culture change as the Hungarian Plain became
incorporated into the regionally materially homogeneous Baden material culture group.
117
Figure 6.3. Transects and surface find locations at the site of Békés 26.
118
Figure 6.4. Transects and surface find locations at the site of Békés 178.
119
Figure 6.5. Collection units at the site of Bélmegyer 82.
120
Figure 6.6. Collection units at the site of Biharugra 33.
121
Figure 6.7. Transects and surface find locations at the site of Bucsa 13.
122
Figure 6.8. Transects and surface find locations at the site of Füzesgyarmat 97.
123
Figure 6.9. Collection unit locations at the site of Gerla 64.
124
Figure 6.10. Transects and surface find locations at the site of Körösladány 21.
125
Figure 6.11. Collection unit locations at the site of Mezőberény 34.
126
Figure 6.12. Collection unit locations at the site of Szeghalom 80.
127
Figure 6.13. Collection unit locations at the site of Tarhos 67.
128
Table 6.1. Summary of analyzed ceramics from the MRT collection and Doboz H. tábla by site and cultural period.
MCA=Middle Copper Age, LCA=Late Copper Age, EBA=Early Bronze Age, MBA=Middle Bronze Age.
Site
Békés39
Békés75
Bélmegyer56
Biharúgra53
Bucsa12
Dévaványa166
Doboz H. tábla
Füzesgyarmat18
Füzesgyarmat97
Körösladány16
Körösladány21
Körösladány33
Mezőgyán2
Okány43
Szeghalom112
Szeghalom168
Szeghalom177
Szeghalom194
Szeghalom49
Szeghalom58
Szeghalom60
Szeghalom80
Szeghalom89
Vésztő119
Vésztő17
Vésztő4
Vésztő49
Vésztő65
TOTAL:
MCA
2
8
3
2
2
17
LCA
24
1
46
6
1
26
1
2
1
4
3
7
14
8
5
5
1
1
7
4
7
3
1
2
10
190
129
EBA MBA Other
1
1
1
1
1
1
2
Table 6.2. Summary of analyzed ceramics from the collected sites and Hódmezővásárhely by site and cultural
period. MCA=Middle Copper Age, LCA=Late Copper Age, EBA=Early Bronze Age, MBA=Middle Bronze Age.
Site
MCA LCA EBA MBA
Békés178
4
4
4
Békés26
5
3
2
Bélmegyer82
7
Biharúgra33
3
23
2
6
Bucsa13
2
1
Füzesgyarmat97
5
Gerla64
8
Hódmezővásárhely
28
Körösladány21
2
Mezőberény34
2
Szeghalom80
7
24
11
4
Tarhos67
49
TOTAL:
10
159
20
17
Other
4
1
4
9
2
10
11
32
14
87
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
130
*
*
*
*
*
*
LATER
*
*
*
SAR
EBA
BA
Kurgan
LCA-BAD
LCA-BOL
MCA
ECA
L-NEO
Size
(HA)
8.37
6.02
8.69
4.86
1.84
3.43
1.36
2.42
4.71
7.38
8.89
5.75
2.17
5.96
2.11
2.87
E-M-NEO
Site Name
Békés 103
Békés 109
Békés 110
Békés 112
Békés 117
Békés 124
Békés 125
Békés 131
Békés 134
Békés 135
Békés 161
Békés 169
Békés 171
Békés 178
Békés 200
Békés 203
NEO
Table 6.3. Summary of sites described in MRT volumes as containing Late Copper Age surface material.
NEO=Neolithic, E-M-NEO=Early/Middle Neolithic, L-NEO=Late Neolithic, ECA=Early Copper Age,
MCA=Middle Copper Age, LCA-BOL=Late Copper Age Boleráz, LCA-BAD=Late Copper Age Baden,
BA=Bronze Age, EBA=Early Bronze Age, SAR=Szarmation.
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
131
LATER
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
BA
Kurgan
LCA-BAD
LCA-BOL
*
*
*
*
*
MCA
*
*
SAR
*
*
*
*
ECA
*
*
L-NEO
*
*
EBA
1
1.03
1.06
7.17
1.2
2.53
5.51
3.19
4.12
4.15
7
5.47
n/a
1.11
5.08
2.17
4.12
2.41
1.48
4.81
4.2
6.04
2.26
3.97
1.94
6.78
1.63
8.76
5.51
1.56
n/a
n/a
n/a
n/a
1.44
n/a
n/a
E-M-NEO
Site Name
Békés 26
Békés 35
Békés 38
Békés 39
Békés 51
Békés 52
Békés 55
Békés 58
Békés 75
Békés 88
Békéscsaba 332
Békéscsaba 443
Békésszentandras 12
Bélmegyer 13
Bélmegyer 14
Bélmegyer 15
Bélmegyer 16
Bélmegyer 17
Bélmegyer 2
Bélmegyer 22
Bélmegyer 41
Bélmegyer 53
Bélmegyer 56
Bélmegyer 65
Bélmegyer 66
Bélmegyer 81
Bélmegyer 82
Bélmegyer 87
Biharugra 1
Biharugra 33
Biharugra 53
Bucsa 12
Bucsa 13
Busca 11
Ecsegfalva 1
Endrőd 101
Endrőd 149
Size
(HA)
NEO
Table 6.3 - continued
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
132
*
*
*
*
*
*
*
EBA
*
*
BA
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Kurgan
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
LCA-BAD
*
*
*
*
*
*
LCA-BOL
MCA
ECA
*
*
*
LATER
*
*
*
*
*
*
L-NEO
*
*
SAR
*
*
*
E-M-NEO
Site Name
Endrőd 6
Endrőd 89
Füzesgyarmat 18
Füzesgyarmat 69
Füzesgyarmat 97
Gerla 30
Gerla 33
Gerla 53
Gerla 54
Gerla 55
Gerla 63
Gerla 64
Gyoma 121
Gyoma 125
Gyoma 15
Gyoma 170
Gyoma 176
Gyoma 19
Gyoma 196
Gyoma 7
Körösladány 21
Körösladány 32
Körösladány 33
Körösújfalu 12
Körösújfalu 15
Körösújfalu 4
Mezőberény 2
Mezőberény 23
Mezőberény 32
Mezőberény 34
Mezőberény 35
Örménykút 78
Szarvas 29
Szarvas 93
Szeghalom 229
Szeghalom 234
Szeghalom 58
Size
(HA)
n/a
n/a
5.56
8.47
7.43
3.57
2
3.37
1.85
1.68
4.68
7.49
1.65
7.24
n/a
n/a
n/a
n/a
n/a
n/a
6.89
5.54
9.47
7.59
2.66
2.33
n/a
8.34
6.36
9.64
2.98
n/a
n/a
n/a
n/a
n/a
6.84
NEO
Table 6.3 – continued
*
*
*
*
*
EBA
*
*
*
*
*
*
*
*
*
*
*
*
*
LATER
*
*
BA
Kurgan
LCA-BAD
*
*
*
*
SAR
*
*
LCA-BOL
MCA
ECA
L-NEO
NEO
Site Name
Szeghalom 59
Szeghalom 80
Szeghalom 84
Szehalom 97
Tarhos 67
Telekgerendás 142
Telekgerendás 65
Vésztő 17
Vésztő 49
Size
(HA)
2.29
2.6
6.97
n/a
6.29
n/a
n/a
9.25
2.7
E-M-NEO
Table 6.3 – continued
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
133
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
LATER
*
*
*
*
*
*
Kurgan
LCA-BAD
LCA-BOL
MCA
ECA
*
SAR
*
*
EBA
*
*
*
BA
1
5.96
1.63
1.56
n/a
7.43
7.49
6.89
9.64
2.6
6.29
L-NEO
Size
(HA)
E-M-NEO
Site Name
Békés 26 – Kászmánkert I
Békés 178 – Lápos-Domb
Bélmegyer 82 – Cserszád I
Biharugra 33 – Kincses T.
Bucsa 13 – Kis Kecskés
Füzesgyarmat 97 –Pázmán
Gerla 64 - Veres Gyűrűs
Körösladány 21 – Tekerő
Mezőberény 34 – Frei-T.
Szeghalom 80 – Dió-Ér
Tarhos 67 – Kifli Domb
NEO
Table 6.4. Summary of sites collected during the fall 2009 field season, as described in the MRT volumes.
NEO=Neolithic, E-M-NEO=Early/Middle Neolithic, L-NEO=Late Neolithic, ECA=Early Copper Age,
MCA=Middle Copper Age, LCA-BOL=Late Copper Age Boleráz, LCA-BAD=Late Copper Age Baden,
BA=Bronze Age, EBA=Early Bronze Age, SAR=Szarmation.
*
*
*
*
CHAPTER SEVEN
RESULTS OF THE SPATIAL ANALAYSIS
Introduction
In this chapter, I provide the results of the spatial analysis of prehistoric site distribution
in the Körös River study area, and how this distribution changed over time. It revisits the
research conducted in the 1980s by Andrew Sherratt (1997a, 1997b) and reinterprets the data
with the inclusion of site information not available during the previous analysis. Additionally,
the results presented here provide information on a wider geographic scale at multiple spatial
resolutions than previous studies. This allows for a more thorough interpretation of changing
settlement patterns over time. The chapter essentially reports a test of Sherratt’s
environmental/economic model by statistically and qualitatively observing the nature and degree
of association between the culture groups present on the Plain during the Middle and Late
Copper Age.
The transition between the Middle Copper Age Bodrogkeresztúr phase and Late Copper
Age Baden phase coincides with a dramatic change in material culture (especially ceramic form
and decoration), burial practices, the more intensive use of beasts of burden, and the first
appearance of wheeled vehicles (Anthony 1996, 1990; Banner 1956; Kalicsz 1998; Whittle
1996). Additionally, it has been argued that a migratory population of burial mound (kurgan)
builders appeared on the Great Hungarian Plain at this time. This arrival has often been
associated with these material changes, and with the economic changes that followed during the
Early and Middle Bronze Age (Anthony 1990; Gimbutas 1977, 1980; Milisauskas and Kruk
1989, 2002:247). The potential social, settlement, and material culture effects of the arrival of
such a migratory population are tested as part of the present research. It has been argued that this
transition and burgeoning economic pattern may have contributed to the development of regional
political systems with a tributary economy, craft specialization, and institutionalized hierarchy
(Earle 2002; O’Shea 1996). Andrew Sherratt’s spatial study in the Körös region of the Great
Hungarian Plan dovetails with other researchers’ economic models, and as such is deserving of
reevaluation and expansion.
Sherratt’s settlement research conducted almost 30 years ago focused on a study area in
northern Békés County (Figure 7.1). Within this relatively restricted area, he observed a
dispersal and an increase in site number between the Late Neolithic and Early Copper Age, a
134
Figure 7.1: Andrew Sherratt’s Dévaványa Plain study region in northern Békés County (outlined in red), in contrast
with the present study area.
reduction in site number and further dispersal in the Middle Copper Age (interpreted as a
population decline), an almost total abandonment of the area in the Late Copper Age and the
intrusion of kurgans (Figure 7.2), and a subsequent resettlement and nucleation by the Middle
Bronze Age (Figure 7.3). Site data at this resolution correlated well with his lower resolution
assessment of site distribution and density of the entire Hungarian Plain at this time (see Sherratt
1997b:304, figure 11.17). His general conclusions also fit well into other economic and
social/settlement interpretations of the Carpathian Basin during the Copper Age and Bronze Age
(Childe 1930; Pare 2000; Parkinson 2002; Sherratt 1993). However, the study was based on a
limited data set of site locations available at the time of research. The following results
incorporate more recent site and settlement data (see Jankovich et al. 1989; Jankovich et al.
1998) into the analysis.
135
Figure 7.2: Kurgan locations overlaid with kernel density map of kurgans per square kilometer. Darker blue shading
indicates a higher density of kurgans per square kilometer.
A Reassessment of Settlement in the Körös River Study Region
Study area boundaries – both geographic and political – have a profound impact on the
archaeological observation of patterns, since modern boundaries rarely correspond to prehistoric
cultural distributions. This is true both descriptively and statistically, as site locations and
clusters of sites can appear more closely related in one area, or at one resolution, than another. It
is therefore important for archaeologists interested in observing distributions and patterns in a
geographic region to operate within a multi-scalar framework. The present research takes a
multi-scalar approach by describing changes in settlement patterns over time at three resolutions
(Sherratt’s study region, the Körös River drainage study region, and micro-regions within the
Körös drainage), in addition to quantitatively describing settlement change in the region utilizing
average nearest neighbor analysis.
136
Figure 7.3: Late Neolithic, Early Copper Age, Middle Copper Age, and Late Copper Age sites in the Sherratt study
area. Site locations are laid over a map of kurgan density. Darker blue shading indicates a higher density of kurgans
per square kilometer. The density map is intended to illustrate kurgan clusters in contrast with site locations.
Three specific questions related to the migration of kurgan builders into the study region
were approached in this spatial analysis, which adds an intermediate level of resolution (the area
of the Körös River drainage system) to Sherratt’s analysis: 1) Did kurgan builders cause dramatic
change during the latter half of the Copper Age; 2) Are kurgans and Late Copper Age
archaeological sites spatially complementary throughout the entire region, as per Sherratt’s
conclusions? And 3) what implications does the spatial relationship between kurgans and Late
Copper Age sites have for understanding material culture and settlement changes at the end of
the Copper Age? More generally, the analysis evaluates settlement pattern change in the Körös
region over time in light of Sherratt’s diachronic study in his higher resolution study area. The
results presented here largely supports Sherratt’s earlier conclusions. Observable as the Middle
137
North
East
South
Figure 7.4: The three average nearest neighbor analytical zones in the study region.
and Late Neolithic progressed, a nucleation gave way to a dispersal and increase in site number
during the Early Copper Age, and a progressively less intense occupation during the Middle and
Late Copper Age. The average nearest neighbor analysis, general spatial analysis, as well as data
gathered during the visitation of 49 prehistoric archaeological sites in the Körös region during
the fall of 2009, approach these questions and considerations, and test Sherratt’s (1997a, 1997b)
conclusions against a wider data set.
Average Nearest Neighbor and Density Analysis
Within the Körös study region, the average nearest neighbor statistic was used to determine level
of clustering or randomness within cultural periods, based on the nearest neighbor index. The
nearest neighbor index is the ratio of the actual distance between sites divided by the expected
difference based on the area of study, with a range of 1-2.15. It is calculated as described
138
Figure 7.5: The Körös study region and all Early, Middle, and Late Copper Age sites.
below, with “Rn” as nearest neighbor value, “D(Obs)” as mean observed nearest neighbor
distance, “a” as area under study, and “n” as total number of points:
The expected difference is the average difference between neighbors in a hypothetical random
distribution. If the index is less than 1, the pattern exhibits clustering. If the index is greater than
1, then the trend is toward dispersion. The Z score in nearest neighbor analysis is a measure of
statistical significance that indicates whether or not to reject the null hypothesis, which in this
139
case is that all points are randomly distributed across the landscape. At a 95% confidence level,
a Z score between -1.96 and 1.96 means that the null hypothesis cannot be rejected (Ebdon
1985).
The Körös Region was divided into three zones for analytical purposes (Figure 7.4), as
the non-symmetrical boundary of the study area, as well as distribution of sites throughout the
county on a general level, confounded nearest neighbor results when calculated as a whole. The
mean nearest neighbor index and Z score of the three zones was used for this analysis. All
nearest neighbor calculations for all regions and all cultural phases are provided in Table 7.1.
Using the nearest neighbor data as a guide, density maps of kurgans (based on number per square
kilometer) were constructed on order to clearly identify kurgan clusters, as have been described
in the literature (see Gimbutas 1997). The goal within the scales of resolution was to determine
if Late Copper Age Boleráz and Baden sites, as identified in the MRT survey (Escedy 1979;
Jankovich et al. 1989; Jankovich et al. 1998) exist within these clusters and, by proxy, if they are
spatial correlated rather than spatially complementary, as Sherratt suggested (1987b).
Unfortunately, the majority of the nearest neighbor calculations produced Z values of
greater than 1.96 (Tables 7.1 and 7.2). This means that the results are not statistically significant
at the 95% confidence level. Two primary factors, among others, might account for these
results. First, the irregular boundaries of the analytical zones (and of the study area as a whole)
likely confounded the nearest neighbor calculations, as the statistic is most accurate in discretely,
evenly bounded areas. Second, dense clusters of sites in certain areas of the study region and an
almost complete lack of sites in others may have generated results indicating clustering, while
not producing a statisitically significant result. This is especially true of Late Copper Age sites
and site clusters. Despite the lack of statistical significance for most of the calulations, the
results are still useful in conjunction with the density analysis discussed above. The results are
also useful in a general sense for measuring settlement nucleation and dispersal, but the
interpreations should be carefully considered.
With the realities of statistical significance in mind, the average nearest neighbor
analysis, density analysis, and observations and data on spatial relationships of sites of different
periods largely support Sherratt’s earlier conclusions. At the scale of eastern Hungary, his model
holds generally true, as seen in his distribution maps and in maps presented in this chapter (see
Figures 7.5 and 7.6) that incorporate additional site data culled from the more recently published
140
Table 7.1. All nearest neighbor calculations for the three analytical regions in the Körös study region, organized by
cultural phase.
North Group
East Group
South Group
Average
Average Nearest Neighbor for Kurgans in Study Region
NN
Z
Index
Value Average NN Distance (m) Expected NN Distance if Random (m)
0.8
6.91
674
871
0.69 11.31
524
1076
0.9
6.7
1025
1047
0.8
8.3
741
998
North Group
East Group
South Group
Average
Average Nearest Neighbor for Boleráz-Baden Sites in Study Region
NN
Z
Index Value Average NN Distance (m) Expected NN Distance if Random (m)
0.84
1.11
2889
3451
0.62
5.11
1191
1911
0.87
1.09
2889
3667
0.78
2.43
2323
3010
North Group
East Group
South Group
Average
Average Nearest Neighbor for Bodrogkeresztúr Sites in Study Region
NN
Z
Index Value Average NN Distance (m) Expected NN Distance if Random (m)
1.39
2.9
3623
2605
1.01
0.37
3153
3289
0.96
0.68
2409
2390
1.12
1.32
3062
2761
Körös Region
Average Nearest Neighbor for Tiszapolgár Sites in Study Region
NN
Z
Index Value Average NN Distance (m) Expected NN Distance if Random (m)
0.61 14.05
985.14
1604.63
141
Table 7.2. Average nearest neighbor calculations for Early Copper Age, Middle Copper Age, and Late
Copper Age archaeological sites in the Körös study region.
Average Nearest Neighbor for ECA-LCA Sites in Study Region
NN
Z
Index
Value Average NN Distance (m)
Expected NN Distance if Random (m)
ECA
0.61
14.05
985
1604
MCA
1.12
1.32
3062
2761
LCA
0.78
2.43
2323
3010
MRT volumes. At this large scale, the relative density of sites in the Middle Copper Age
decreases when compared to that of the periods immediately before and after. This is supported
by the nearest neighbor statistic, which suggests that density and tendency toward clustering
decreased in the Middle Copper Age and increased in the Late Copper Age (Table 2).
Additionally, Late Copper Age Boleráz and Baden sites do, in a very general sense, exist
spatially exclusively of kurgans.
Sherratt’s model also holds true at a more detailed resolution in his select micro-region
on the Dévaványa Plain. A decrease in settlement density is observed during the Middle Copper
Age (Figure 7.6), for a total of 15 Bodrogkeresztúr sites. Only nine Late Copper Age sites are
documented by the MRT within the boundaries of this study region. Most of these sites are
between two kurgan groups, and are no farther than 1 km from each other (Figure 7.7).
Additionally, kurgan clusters, as observed by Sherratt (1997b), are clearly visible. As he might
have suspected, these clusters are statistically demonstrable (though not significant) in the
northern part of Békés County, with a nearest neighbor index of 0.8 (Z=6.91).
However, at the scale of the Körös River Basin study region Middle Copper Age
settlement is not as sparse, and Late Copper Age sites do correlate spatially with kurgans and
kurgan clusters in some cases. Early Copper Age settlements occur densely and in clusters
throughout the region. Although there is a significant reduction in number of Middle
Copper Age sites – 394 Early Copper Age Tiszapolgár sites vs. 70 Bodrogkeresztúr sites – small
clusters of Middle Copper Age sites do occur in several locations on the Plain; for example, just
north and south of the modern city of Gyomaendrőd and near the towns of Békésszentandrás and
Szarvas (Figure 7.8). However, these clusters are hardly notable in comparison to the clustering
exhibited by Early Copper Age sites and kurgan burial mounds.
142
Figure 7.6: Early (n=97) and Middle Copper Age (n=15) site distribution in Sherratt’s study region. Note decrease
in site number and density.
A density map of kurgan tumuli emphasizing the footprint of statistically significant
kurgan clusters overlaid with the locations of Late Copper Age Boleráz and Baden sites shows
that, in some areas in the study region, kurgans and Late Copper Age sites are not always
spatially exclusive and exist quite close to one another (Figure 7.9). Interestingly, Boleráz and
Baden sites were over 5 km on average from their nearest Late Copper Age neighbor, while all
are located on average only 2.3 km from the nearest kurgan. More than 50% of all Late Copper
Age sites are within 1.5 km of a kurgan. This pattern could partially be attributed to the sheer
number of kurgans in the study area (n=591). But, more than 33% (n=36) of Late Copper Age
sites occur adjacent to or within zones with the highest concentrations of kurgans.
143
Figure 7.7: Kurgan locations overlaying a map of kurgan density to illustrate kurgan “clusters,” and Late Copper
Age site distribution in Sherratt’s study region. Note the general spatial exclusivity between kurgan clusters and
Late Copper Age sites. Darker blue shading indicates a higher density of kurgans per square kilometer.
At an even finer resolution, one can see that Late Copper Age sites exist within kurgan
clusters in some areas of the study region. For example, two significant kurgan clusters near
Körösújfalu have a Late Copper Age presence (Figure 7.10). Two Baden sites exist in the
middle of two significant kurgan clusters in the eastern part of the study region near the town of
Biharugra (Figure 7.10). Near Gyomaendrőd, a Baden site exists within 200 meters of two
kurgans (Figure 7.11). Near the town of Bélmegyer, a large cluster of Late Copper Age sites sits
within 200 meters of several kurgans (Figure 7.1). Given the proximity of numerous kurgans
and Late Copper Age settlements throughout the study area, a practice of avoidance between two
populations does not seem tenable. However, this assumes contemporaneity between the
kurgans, settlements, and their builders. This sticking point has, unfortunately, not been
addressed satisfactorily.
144
Figure 7.8: Middle Copper Age Bodrogkeresztúr site distribution. Potential “clusters” are circled in red. Though
the clusters themselves are not significant in comparison to the dense cluster of Early Copper Age sites and kurgans,
their presence in the center of the Hungarian Plain is intriguing in terms of Sherratt’s argument for depopulation and
shifting economic focus away from the region. Darker blue shading indicates a higher density of kurgans per square
kilometer.
A Reevaluation of Late Copper Age Settlement Location in the Körös Region
Field research conducted in the fall of 2009 involved the visitation, collection, and rough
measurement of Late Copper Age archaeological sites. Initially, single component
archaeological sites were given visitation priority under the assumption that the collection of
multi-component sites introduces a lack of temporal control and causes difficulty in determining
size of individual period occupations. Unfortunately, all Late Copper Age sites described as
single component in the MRT lacked identifiable material on the surface, or more often were
completely devoid of surface material. This is notably the case at the sites in the previously
mentioned cluster of Late Copper Age settlements in the center of the county near the city of
Békés. Even more, when Late Copper Age material was encountered and systematically
145
Figure 7.9: Kurgans, density of kurgans per square kilometer, and Late Copper Age site distribution in the Körös
River study region. Darker blue shading indicates a higher density of kurgans per square kilometer.
collected at multi-component sites, Baden artifact density was quite low and estimated site size
was very small, never exceeding 2.7 hectares (see Chapter Six for more detailed descriptions of
archaeological sites and collected assemblages).
Although these descriptive observations do not call into serious question the accuracy of
the site descriptions and maps created by the MRT researchers (Escedy 1982; Jankovich et al.
1989; Jankovich et al. 1998), as surface assemblages on regularly plowed archaeological sites
change constantly and rapidly, it does require a reevaluation of how archaeologists working the
Körös region assess site representation, frequency, and density. This is especially true of cultural
phases such as the Late Copper Age, when artifact density and surface representation tends to be
low, even at large sites. Variability in how sites were assessed and recorded during the MRT
146
Figure 7.10: Late Copper Age sites located within kurgan clusters near Körösújfalu. Darker blue shading indicates
a higher density of kurgans per square kilometer.
surveys – which took place many years apart, and involved many different researchers – may
have affected maps produced during different periods. This could explain the extremely high
density of Late Copper Age sites recorded in the in the center of the county (MRT Volume 10,
published in 1998), as opposed to the much lower frequency, and almost total absence of single
component, Late Copper Age sites recorded in previous volumes. Even more, the Late Copper
Age Boleráz/Baden ceramic phase distinction is used inconsistently between MRT versions, with
discrepancies between the assignment of sherds to the respective phases based on decorative
characteristics. When taken into consideration as a whole, it is possible that Late Copper Age
occupation and site distribution may, in reality, more closely resemble the pattern observed in the
rest of the county, including Sherratt’s Dévaványa Plain study region.
147
Figure 7.11: Late Copper Age sites in close association with kurgans south of Gyomaendrőd. Darker blue shading
indicates a higher density of kurgans per square kilometer.
Discussion and Conclusions of the Settlement Pattern Research
Following the Middle Copper Age, the Baden material culture tradition had become
ubiquitous on the Hungarian Plain. A similar level of homogeneity had not been seen since the
Middle Neolithic AVK period. However, even this dramatic material culture change does not
necessarily support a migration model such as the one formulated by Gimbutas. Indeed, material
excavated from kurgans suggests interaction and perhaps trade – not just abandonment –was
happening at this time. The presence of Bodrogkeresztúr sites in a previously unoccupied area of
the Plain – the Tisza and Danube interfluve – points to an increase in interaction with
Transdanubian populations. Additionally, the possibility of a large Lengyel- type earthen
roundel at the Middle Copper Age site Szarvas 38, in the center of the Plain, lends support to this
assertion (Makkay 1983), though whether or not this site actually consisted of an empty earthen
148
Figure 7.12: Kurgan clusters near a concentration of Late Copper Age sites near Bélmegyer. Darker blue shading
indicates a higher density of kurgans per square kilometer.
roundel, or was a more typical Middle Copper Age settlement, remains debatable (Parkinson
2010, personal communication). At Kétegyháza, a kurgan within a Bodrogkeresztúr settlement
contained a great deal of Middle Copper Age Material (Escedy 1979). At DebrecenDunahálom, numerous sherds of typical Baden form were recovered from a kurgan burial
(Escedy 1979).
Ultimately, the Middle and Late Copper Age settlement evidence in the Körös region
suggests that long-term cultural processes on the Plain were responsible for changes in both
material culture and settlement behavior, and any migratory populations on the Plain at this time
acted within this context, rather than having catalyzed it. Indeed, based on the present evidence,
it is unlikely that the arrival of migratory kurgan builders instigated a sudden change in material
culture and settlement patterns in the study region. Rather, Sherratt’s model of diachronic
149
economic change is much more tenable – the trend toward settlement on the margins of the
Plain, perhaps to exploit raw material sources and participate in trade networks, developed over
the long-term with roots in the Neolithic at the breakup of AVK, the subsequent development of
the Lengyel interaction sphere in Transdanubia, and the Tisza-Herpály-Csőszhalom complex on
the Eastern Plain. The development was more obviously stated during the Tiszapolgár and
Bodrogkeresztúr phases of the Copper Age, and by the Late Copper Age was fully realized as
populations became incorporated into a regionalized, more homogeneous material culture group
with strong economic and social ties beyond the edges of the Great Hungarian Plain.
Despite these conclusions, the kurgans themselves and their appearance across the
landscape must be accounted for. Due to the lack of direct kurgan evidence, save for limited
excavation and stratigraphic evidence (see Escedy 1979), this remains difficult. Although long
presumed that Yamnaya migrants from the east built the kurgans, it has not been demonstrated
that all of the tumuli were constructed by the same population, or even contemporaneously.
Other possibilities for the development of kurgans across the landscape must therefore be
considered. For example, it is possible that a migration from the east occurred late in the Middle
Copper Age, but later kurgans are an adopted practice emulated by indigenous inhabitants of the
Hungarian Plain. This may explain why Yamnaya material was not present in all excavated
kurgans, and why Middle Copper Age material was located at some kurgan excavations such as
at Kétegyháza (Escedy 1979). Even more, kurgans were often reused as burial locations and
sites of later construction well into the historic and modern periods, and more kurgans were
constructed in the region later in the prehistoric period (Escedy 1979). This suggests that they
were frequently emulated over time, and may not have simply been a Middle/Late Copper Age
Yamnaya phenomenon. This line of reasoning may also explain why some Late Copper Age
sites do exist near kurgans and within kurgan clusters, though this relationship will remain
unclear until more kurgans and Late Copper Age sites are accurately dated in the Körös region.
Given the uncertainties of kurgan cultural affiliation in the Körös region and across most
of the Great Hungarian Plain, it must be considered that even small emigrations of people can
have considerable impact on location culture through emulation over the long-term. In other
words, large-scale, demic migrations such as those discussed by Anthony (1990) are not the sole,
or even the best, explanation for a phenomenon such as kurgan appearance. Nor is a diffusionist
model based solely on adoption sufficient to explain their appearance and spread on the
150
Hungarian Plain. A model of small-scale migration and subsequent emulation of kurgan
building accounts not only for the initial appearance and spread of kurgans across the landscape
of the Körös region, it also accounts for the lack of Yamnaya-like settlement sites in the region.
Importantly, the problem of establishing contemporaneity between kurgans, Middle
Copper Age sites, and Late Copper Age sites continues to confound efforts to understand the
nature of the kurgan builders’ influence on the settlement patterns, material culture, economy,
and society of the latter half of the Copper Age on the Great Hungarian Plain. Until more is
understood about how the kurgans came to be across the region – in terms of both chronology
and developmental pattern – models of the exact nature of their interaction with the indigenous
population of the Plain will remain incomplete.
Summary
In this chapter, I have presented the results of the spatial analysis conducted to build upon
the body of spatial knowledge established by Andrew Sherratt’s research on the Hungarian Plain
(1997a, 1997b), and the work of Hungarian archaeologists in developing and publishing the
volumes in the MRT series. Overall, the results presented here at the scale of the Körös River
study region support Andrew Sherratt’s results at the scale of the Dévaványa Plain study region
and the Great Hungarian Plain as a whole. However, at the highest resolution analyzed in
specific areas of Békés County, the general spatial correlations between Middle Copper Age,
Late Copper Age, and kurgan archaeological sites are not as strong. This suggests that the
kurgan builders and inhabitants of Late Copper Age Baden settlements in the region may not
have avoided each other at the level proposed by Sherratt. Based on the evidence presented here,
I do not argue in favor of a migration or invasion explanation for material culture and settlement
change during the Late Copper Age ca. 3,500 B.C., and instead support a model of long-term
change with social and economic implication on and beyond the eastern Hungarian Plain. The
results of the ceramic analysis presented in Chapter Eight will bolster this model, and its
implications for understanding the nature of Late Copper Age settlement and culture on the
Hungarian Plain will be discussed in Chapter Nine.
151
CHAPTER 8
RESULTS OF THE CERAMIC ANALYSIS
Introduction
In this chapter, I present the results of the macroscopic and microscopic analyses of
ceramics from the Körös study region, as well as a small number of samples from the adjacent
Tisza River watershed as a control and comparison. Material for the study was obtained in three
ways: 1) material collected as part of the MRT survey was analyzed at the Munkácsy Mihály
Múzeum in Békéscsaba, Hungary, during the summer of 2009; 2) ceramics collected as part of
the field component of the Late Copper Age Körös Archaeological Project in the fall of 2009;
and, 3) samples from the recent excavations at the Late Copper Age Baden site of
Hódmezővásárhely-Kopáncs I., Olasz-tanya.
Ceramic form and decoration can change quickly during times of population continuity as
well as during times of rapid social, economic, or political change (Kreiter 2003; Lemmonier
1992). However, production technology – including raw material preparation, the addition of
tempers, and firing techniques – is more conservative, even when form and decoration undergo
marked changes. Therefore, approaching culture change through pottery typologies is not the
most reliable method by which to model migration or other demographic shifts. This is
especially pertinent in the eastern part of the Hungarian Plain, which experienced a discontinuity
in ceramic form and decoration as well as settlement patterns at the beginning of the Late Copper
Age, concurrent with a possible migration of new people into the region.
In order to observe differences in ceramic production technology between cultural
phases, and to identify any regional variability within the Late Copper Age phase, two ceramic
analysis strategies were employed: 1) a macroscopic study, in which whole sherds were analyzed
and coded in hand sample; and 2) a petrographic study, in which microscopic characteristics of
ceramic paste, temper, and mineral inclusions were analyzed. The two methods provide different
data sets and different perspectives on understanding prehistoric pottery, and when used together
provide a clearer picture of ceramic manufacture in the past than the use of either method
exclusively.
152
Description of Variables
Many of the variables discussed in this chapter are measured in both macroscopic and
microscopic analyses, albeit at different resolutions. As such, an overview of variables and their
various indications is provided here, while their more specific applications will be discussed in
the sections dealing with macroscopic and microscopic results, respectively.
Sorting describes the distribution of natural and intentional mineral inclusions in the
ceramic paste, and is measured along a range of poorly, fair, good, and very good. A poorly
sorted sample will contain unevenly distributed mineral inclusions of unequal size, and is
especially used to describe a bimodal distribution of natural inclusions such as quartz. A sample
described as “very good” will have evenly distributed inclusions of equal size. Good sorting of
natural inclusions often indicates a high degree of raw material preparation through techniques
such as levigation. Poor sorting indicates relatively little raw material preparation, or in the case
of a bimodal distribution the intentional addition of crushed minerals as temper (Galaty 1999;
Shepard 1956; Orton et al. 1993; Whitbread 1989).
Kneading is a measure of the amount of void space present within the body of a sherd,
and is an indicator of how well and for how long the potter folded, pressed, and kneaded the clay
(Orton et al. 1993). Kneading is measured along a scale of poorly (many large voids),
moderately (relatively few voids), and well (few small voids). In hand sample (the unaided eye),
kneading is measured by observing the texture of the paste in a freshly broken sherd, while void
space is qualitatively and quantitatively measured directly in petrographic analysis (Reedy 2008;
Whitbread 1989).
Like kneading, the texture of a fresh break in hand sample indicates the thoroughness of a
potter’s processing of raw clay before the material was shaped into final vessel form. Texture is
measured along a scale of hackly (very irregular and angular break), irregular (irregular break),
fine (slightly angular break), and smooth (smooth break).
The measurement and description of a sherd’s firing condition is largely based on color,
with darker color brown, grey, and black paste colors indicating a reducing (oxygen poor) firing
environment, and reds and brighter browns indicating an oxidizing (oxygen rich) firing
environment. Reducing environments are often indicative of long firing times, the use of closed
pits and coals as firing environments, or both. An oxidizing environment may involve open
flame, and usually short firing times. Of course, a range of variability exists within these
153
extremes, and so qualitative descriptions of firing environments are usually generalized (Orton et
al. 1993). In this research project, firing conditions of complete sherds are described
macroscopically as reduced exterior/oxidized interior, reduced sandwich (oxidized on interior
and exterior surfaces with a reduced core), oxidized exterior/reduced interior, oxidized sandwich
(reduced interior and exterior surfaces with an oxidized core), all reduced, and all oxidized.
Results of the Macroscopic Analysis
The macroscopic ceramic study aimed to measure any changes in ceramic production
technology between the Middle Copper Age Bodrogkeresztúr and Late Copper Age
Boleráz/Baden phases, and to measure any variability within the Late Copper Age assemblage.
Sherds were coded according to multiple variables aimed at measuring elements of production,
including diagnostic type (rim, decorated body, base), thickness (mm), hardness, grain size,
quality of kneading, natural and intentional inclusions (such as clay nodules or grog), inclusion
sorting, color, surface treatments, and decoration.
In all, 352 Middle and Late Copper Age sherds were analyzed. Three hundred three of
these sherds came from surface contexts in the Körös study region, 21 from the excavated site of
Doboz Homokgödör-tablá in the Körös Region, and 28 from excavated feature contexts of the
Baden settlement of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the adjacent Tisza River
watershed. The Hódmezővásárhely ceramics are included as a counterpoint and control for the
Körös region ceramics, and to measure any interregional variability.
Diachronic Ceramic Variability: Middle Copper Age vs. Late Copper Age
Middle Copper Age and Late Copper Age ceramics from sites in the Körös River
watershed exhibit a great deal of similarity, though some variability over time was measured.
The visible inclusions in hand sample are sorted similarly, with the majority of samples from
both phases classified as sorted “good” or “very good” (Table 8.1). Likewise, Middle Copper
Age and Late Copper Age ceramics exhibit similarity both in terms of kneading (Table 8.2) and
texture of a fresh break (Table 8.3). Even more, the Middle and Late Copper Age assemblages
exhibit general similarities in firing conditions (Table 8.4).
Kneading and texture of a fresh break both indicate the thoroughness of processing
before the raw clay material was shaped into its final vessel form by the potter. As might be
154
Table 8.1. Sorting of visible inclusions in Middle and Late Copper Age ceramics from archaeological sites in the
Körös River watershed.
Period
MCA (n=31)
LCA (n=321)
Poorly
1
3%
25
8%
Fair
2
6%
45
14%
Good
12
39%
125
39%
Very Good
16
52%
126
39%
Table 8.2. Kneading of raw material (clay) in Middle and Late Copper Age ceramics from archaeological sites in
the Körös River watershed.
Period
MCA (n=31)
Well
18
58%
130
41%
LCA (n=321)
Moderately
11
36%
133
41%
Poorly
2
6%
58
18%
Table 8.3. Texture of a fresh break in Middle and Late Copper Age ceramics from archaeological sites in the Körös
River watershed.
Period
MCA (n=31)
LCA (n=321)
Smooth
2
6%
2
1%
Fine
8
26%
19
6%
Irregular
16
52%
200
62%
Hackly
5
16%
100
31%
Table 8.4. Firing characteristics in Middle and Late Copper Age ceramics from archaeological sites in the Körös
River watershed.
Period
MCA (n=31)
LCA (n=321)
All
Oxidized
1
3%
11
4%
All
Reduced
21
68%
235
73%
Oxidized
Sandwich
1
3%
1
0%
Reduced
OxiEx/ReduIn Sandwich
4
4
13%
13%
47
23
15%
7%
ReduEx/OxiIn
0
0%
4
1%
expected the observed level of kneading and paste texture are closely correlated in this analysis.
A very low percentage of sherds from both the Middle and Late Copper Age have a smooth paste
texture, while the majority of samples from both phases have an irregular, or rough and angular,
paste texture. While the overall pattern in raw material preparation of Middle and Late Copper
Age ceramic samples is one of continuity, some diachronic variability does exist. Notably, a
higher proportion of Late Copper Age sherds exhibits only moderately or poorly kneaded paste,
and an irregular and hackly paste texture, while simultaneously exhibiting a tendency toward
155
better sorting of visible inclusions. This suggests that less time was being spent in the
manipulation of raw clay during processing in the Late Copper Age, while more effort was
expended ensuring that intentional and unintentional inclusions were of equal size and
distribution. This slightly different approach to raw material processing was likely an effort to
prevent unintentionally added large inclusions from causing cracking while drying and/or
structural failure of the pot while firing.
Interestingly, despite a modest shift in raw clay processing between the Middle and Late
Copper Age, firing conditions went essentially unchanged between the cultural phases (Table
8.4, Figure 8.1). The majority of sherds are reduced throughout, indicating a firing environment
low in oxygen. The vast majority of remaining samples in both periods are reduced on either
interior or exterior surfaces, while less than 4% of sherds in both the Middle and Late Copper
Age are oxidized throughout the sample (indicating a firing environment high in oxygen). Most
telling, only two sherds (one Middle and one Late Copper Age) exhibit an oxidized core,
suggesting that the overall firing environment was a reducing one, and that interior and exterior
color variation are likely the result of differential exposure to heat, flame, and coal as the result
of a relatively uncontrolled firing process, compared to later techniques. The overall pattern
suggests that firing conditions did not undergo substantial change between the Middle and Late
Copper Age, and probably consisted of open pit-firing techniques utilizing smothered coals and
relatively long firing times with little or no direct exposure to the flame. Fired pots were cooled
slowly, and were probably removed from the pit several hours to days after firing (Orton et al.
1993). Additionally, the similarity in firing conditions also suggests a measure of continuity in
raw material collection and preparation. The pattern of reduction suggests that organic material
present in the raw clay remained in the paste throughout the entire manufacturing process, and
was not systematically removed during the raw material processing stages.
Such a pattern indicates significant continuity in the production of ceramics during the
Middle and Late Copper Age. The differences should not be ignored, however. For example, a
higher incidence of poorly sorted samples in Late Copper Age materials suggests either a less
rigorous approach to raw material processing, or possibly a bimodal distribution of visible
inclusions – which points toward intentional addition of mineral or grog temper.
156
100%
90%
80%
ReduEx/OxiIn
70%
Reduced Sandwich
60%
50%
OxiEx/ReduIn
40%
Oxidized Sandwich
30%
All Reduced
20%
All Oxidized
10%
0%
MCA
LCA
Figure 8.1. Firing conditions of Middle and Late Copper Age ceramics. Note the almost identical distribution of
measured firing characteristics between the Middle and Late Copper Age, particularly the significant majority of
samples reduced throughout the firing core.
Additionally, more frequent observation of poorly kneaded and hackneyed textured Late
Copper Age ceramics points to slight changes in production and, again, may signify less focus on
raw material processing and the increasing use of mineral or temper to stabilize the clay paste
during the drying and firing phases of production. The addition of tempers, especially grog, in
Late Copper Age ceramics is discussed more extensively in the section below detailing the
results of the petrographic analysis that focused more explicitly on the identification and
classification of intentional and unintentional inclusions.
Spatial Ceramic Variability: Late Copper Age Inter-site Variability in the Study Region
A total of 191 Late Copper Age sherds from six sites within the Körös River study region
were included in the inter-site analysis. This includes 26 samples from the excavated site of
Doboz Homokgödöri-tablá.
The ceramics from Late Copper Age sites in the study region exhibit characteristics along
a common range of variability, though notable (if not dramatic) differences exist between sites in
terms of both firing conditions and sorting of visible inclusions (Tables 8.5 and 8.6). The vast
majority of the sherds from all sites were reduced and probably open-fired at low temperatures,
as described above. Variables intended to observe firing characteristics reveal very little
indication of oxygen-rich reducing environments, as completely reduced samples and samples
157
with reduced cores were predominant (see Table 8.5 and Figure 8.2). This pattern suggests that
Late Copper Age inhabitants of the region utilized very similar techniques for firing their pots,
resulting in similar proportions of oxidized and reduced ceramics at settlements throughout the
region. It seems likely given the similarities in firing condition between Middle Copper Age and
Late Copper Age ceramic samples, and similarities in firing condition between samples from
Late Copper Age sites throughout the Körös region, that the firing process remained relatively
unchanged during the approximately 1,000 year period encompassing the two cultural phases.
Interestingly in terms of intra-regional variability, one site – Bélmegyer 56 – did exhibit a
different set of characteristics from other Late Copper Age sites in the region, especially in
regard to the sorting of visible inclusions (Table 8.6, Figure 8.3). The ceramic material from
Bélmegyer 56 is typically very well sorted, consisting of evenly spaced and relatively uniformly
sized inclusions. On the whole, Late Copper Age ceramics are not very well sorted, often
containing inclusions of various sizes unevenly spaced. Multiple sets of behaviors could account
for the aberrant characterization of ceramics from this site. However, it is likely that the raw
clay materially was more thoroughly processed and, perhaps intentionally cleaned of visible
inclusions either for aesthetic or practical reasons (i.e., to prevent cracking during drying).
Regardless of the purpose of the behavioral variability, the variability in sorting indicates that a
completely uniform set of pottery production practices was not always used at settlement sites in
the Körös region during the Late Copper Age.
Inter-Regional Variability of Baden Ceramics from the Körös and Maros Watersheds
In order to measure possible variability of Baden ceramic production between regions on
the Great Hungarian Plain, 26 sherds from the site of Hódmezővásárhely-Kopáncs I., Olasztanya in the Maros River watershed were obtained and analyzed. Additionally, the inclusion of
ceramics from the Maros watershed in this analysis serves as a control for the interpretation of
diachronic ceramic variability in the Körös region. If significant diachronic variability had been
observed in the Körös watershed, and if significant inter-regional variability had been observed
from in the Maros watershed, it could have indicated an indigenous increase in spatial variability
158
Table 8.5. Firing characteristics in Late Copper Age ceramics from archaeological sites in the Körös River
watershed.
Site
Tarhos67 (n=48)
Bél.56 (n=43)
Biharugra33 (n=22)
Szeghlm80 (n=28)
Békés39 (n=24)
D. H. tábla (n=26)
All
Oxidized
2
4%
3
7%
0
0%
1
3%
1
4%
0
0%
All
Reduced
37
77%
27
63%
18
78%
26
93%
14
59%
20
77%
Oxidized
Sandwich
0
0%
1
2%
0
0%
0
0%
0
0%
0
0%
OxiEx/ReduIn
5
11%
6
14%
4
18%
1
4%
7
29%
5
19%
Reduced
Sandwich
4
8%
5
12%
0
0%
0
0%
2
8%
1
4%
ReduEx/OxiIn
0
0%
1
2%
1
4%
0
0%
0
0%
0
0%
Table 8.6. Sorting of visible inclusions in Late Copper Age ceramics from archaeological sites in the Körös River
watershed.
Site
Tarhos67 (n=48)
Bélmegyer56 (n=43)
Biharugra33 (n=22)
Szeghalom80 (n=28)
Békés39 (n=24)
Doboz H. tábla (n=26)
Very Poorly
1
2%
0
0%
0
0%
0
0%
0
0%
0
0%
Poorly
3
6%
0
0%
2
9%
4
14%
4
17%
3
11%
Fair
9
19%
1
8%
8
36%
5
18%
7
29%
0
0%
Good
17
35%
7
41%
10
46%
13
47%
13
54%
14
54%
Very Good
18
38%
35
50%
2
9%
6
21%
0
0%
9
35%
Table 8.7. Sorting of visible inclusions in Late Copper Age ceramics from Hódmezővásárhely-Kopáncs I., Olasztanya in the Maros River watershed and Late Copper Age ceramics from the Körös region.
Region
Maros (n=28)
Körös (n=321)
Poorly
3
11%
25
8%
Fair
4
14%
45
14%
159
Good
8
29%
125
39%
Very Good
13
46%
126
39%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
ReduEx/OxiIn
Reduced Sandwich
OxiEx/ReduIn
Oxidized Sandwich
All Reduced
All Oxidized
Figure 8.2. Firing condition of Late Copper Age ceramics from sites in the Körös Region.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Very Good
Good
Fair
Poorly
Very Poorly
Figure8.3. Sorting of visible inclusions in Late Copper Age ceramics from sites in the Körös Region
of Late Copper Age ceramics on the Plain and not necessarily an invasion, migration, or
diffusion scenario, as might be initially assumed were diachronic variability observed in the
Körös region. The fact that this interregional pattern is observable despite the dramatically
different sample sizes illustrates the strength of the relationship.
160
100%
90%
80%
70%
60%
Poorly
50%
Moderately
40%
Well
30%
20%
10%
0%
HMVH
Körös
Figure 8.4. Kneading in Late Copper Age Ceramics conditions from Hódmezővásárhely and in the Körös region.
100%
90%
80%
70%
60%
Hackly
50%
Irregular
40%
Fine
30%
Smooth
20%
10%
0%
HMVH
Körös
Figure 8.5. Paste texture in Late Copper Age ceramics from Hódmezővásárhely and in the Körös region.
Speaking generally, the 28 samples from Hódmezővásárhely-Kopáncs I., Olasz-tanya in
the Maros region resemble the Late Copper Age sherds collected from the Körös region, though
some regional variability was detected. For example, the sorting of visible inclusions in pottery
from the Körös and Maros regions are quite similar, indicating similar techniques for the
collection and processing of raw clay materials (Table 8.7). However, samples from
Hódmezővásárhely are collectively more poorly kneaded than the Körös material (Figure 8.4).
161
Table 8.8. Firing characteristics in Late Copper Age ceramics from archaeological sites in the Maros region and
from the Körös region. The regional difference in firing characteristics observable despite radically different sample
sizes illustrates the strength the pattern.
Region
HMVH (n=28)
Körös (n=321)
All
Oxidized
1
3%
11
4%
All
Reduced
26
93%
235
73%
Oxidized
Sandwich
0
0%
1
<1%
OxiEx/ReduIn
0
0%
47
15%
Reduced
Sandwich
1
4%
23
7%
ReduEx/OxiIn
0
0%
4
1%
Additionally, 62% of Late Copper Age sherds from the Körös region are irregular in crosssection, while 50% of the sherds from Hódmezővásárhely are hackly in cross-section, correlating
with the kneading measurements (Figure 8.5). This indicates that while the removal of
inclusions from the raw clay material was conducted similarly in both regions, the process of
preparing the material for shaping did vary.
A significant difference in indicators of firing condition was observed between Late
Copper Age samples from the Körös region and samples from the Maros region (Table 8.8).
The vast majority of samples in both cases were reduced throughout the interior and exterior
surfaces, and through the cores of the sherds. This again indicates a low oxygen, relatively slow
firing environment with little or no direct contact with flame. However, 93% of
Hódmezővásárhely Baden samples exhibited total reduction, while Körös region Late Copper
Age sherds were thoroughly reduced in only 73% of the sample. More variability in terms of
indicators of firing conditions is visible in the materials from the Körös region (Figure 8.6). This
difference may represent differences raw material preparation, the heterogeneity of raw materials
between the watersheds, or slightly different firing techniques that resulted in more variable
indicators of firing conditions in the Körös region. Although these possibilities are worth
considering, it is important to note the sample sizes from the region (n=28, Maros; n=321,
Körös). It is possible that the much larger samples size from the Körös region accounts for the
greater variability in firing condition indicators. However, the fact that the differences appear
despite the different sample sizes may also indicate the strength of the relationship.
162
100%
90%
80%
ReduEx/OxiIn
70%
Reduced Sandwich
60%
50%
OxiEx/ReduIn
40%
Oxidized Sandwich
30%
All Reduced
20%
All Oxidized
10%
0%
HMVH
Körös
Figure 8.6. Comparison of Late Copper Age ceramic firing conditions from Hódmezővásárhely and in the Körös
region.
Conclusions and Interpretations of the Macroscopic Ceramic Data
All of the materials evaluated as part of this study exhibited a similar range of
characteristics over time and space. Notably, however, inter-site variability within the Körös
region and interregional variability between the Körös and Maros regions was observed in Late
Copper Age ceramics. Although slight, given the relative homogeneity of the Hungarian Plain’s
geology and, thus of raw material sources, even subtle differences between ceramic assemblages
should be discussed. Slight but observable variability in the sorting of visible inclusions at the
site of Bélmegyer 56 suggests that, although generally similar throughout the region, some
variability in ceramic processing and production did occur on the site level – or, perhaps even on
the level of individual, highly productive potters within a particular village. Although more
samples from different culture phases are necessary to make such a determination, it is possible
that Bélmegyer 56 and perhaps other sites in the region maintained independent local processing
and production methods that result in slightly different signatures when submitted to the battery
of measurements used here. Interregionally, the observed differences in indicators of production
and firing techniques are relatively subtle, but have significant consequences for understanding
the nature of the Plain’s incorporation into the Baden material culture during the Late Copper
Age. Since variability exists between two adjacent regions, both part of the Baden material
culture complex, a cautious approach must be taken when interpreting slight variability over time
163
as evidence of a break in population continuity between ceramic phases in the Körös region.
Indeed, when viewed as a whole, the macroscopic data suggest a great deal of continuity in raw
material preparation and ceramic manufacturing techniques between the Middle and Late Copper
Age. When considered alongside the spatial data suggesting a marked level of continuity, and
the presence of an internal cultural trajectory consisting of cycles of population nucleation and
dispersal, the data support an overall trend of continuity in the Körös region between the Late
Neolithic and the Early Bronze Age.
Results of the Petrographic Analysis
Of the samples collected and analyzed from museum, site surface, and excavation
contexts, a total of 147 were selected for thin sectioning and point counting under the
microscope. The samples were chosen due to their clear association with a cultural period,
typically based on surface decoration such as patterns of incised design. Three forms of
information were collected during microscopic analysis that contributed to the description of
change in prehistoric ceramics on the Hungarian Plain between the Copper Age and Bronze Age:
1) point-counts of natural and intentional temper inclusions; 2) rock and mineral identifications;
and, 3) general groundmass descriptions. As discussed in Chapter Five, Stoltman’s (1989, 1991)
point counting method and Whitbread’s (1989, 1995) system of petrographic description was
used. This point-count analysis does not aim to produce a catalog of fabric classes within
cultural periods as is common in many petrographic studies (see Galaty 1999; Kreiter 2005),
even though samples are divided into fabrics according to Riederer’s (2004) classification
scheme as part of general groundmass description. Fabrics specific to material culture phases
were not developed as part of this project, as previous petrographic studies (Hoekman-Sites et al.
2007; Parsons 2005) as well as the current macroscopic analysis have observed little
compositional variability over individual cultural phases. Rather, these results focus on the
identification of diachronic changes in composition and manufacture.
Data pertaining to characteristics of the different ceramic material culture groups are
presented in Appendix II, and in summary form in Tables 8.9 and 8.10. A number of
photomicrographs of samples illustrating key petrographic characteristics, including mineral
identifications, elements of technology, indicators of manufacturing technique, and different
forms of temper, are presented throughout the chapter in Figures 8.7-8.18. The field of view in
164
Figure 8.7. Sample 001, Fabric E1. Late Copper Age from Tarhos 67. Crossed polars. Natural mineral inclusions.
The bright sub-angular inclusions are monocrystalline quartz in a birefringent groundmass. The very small streaks
in groundmass are muscovite mica lathes.
the microphotographs is approximately 2 mm. Taken together, the data from Appendix II and
the tables and figures in this chapter allow for both the general and specific characterization of
ceramics in the Körös Region of the Great Hungarian Plain.
The petrographic results correlate well with the macroscopic ceramic results presented
earlier in this chapter. The overall pattern is one of homogeneity across both space and time,
with samples falling within a common range of variability. However, notable variability is
observable at the inter-site level in the Körös region, and a clear discontinuity in fabric
characterization exists between the Early and Middle Bronze Age. Although not likely
substantial enough to indicate sudden production changes vis-à-vis the introduction of a new
population into the region, the variability over space and time is indicative of social processes
occurring in the region during the time period covered by the study.
General Petrographic Characteristics
Almost all of the ceramic samples from the Middle Copper Age, Late Copper Age, and
Early and Middle Bronze Age fall into a common range of variability. Nearly all thin-sections
165
Figure 8.8. Sample 015, FabricF2. Early Bronze Age from Békés 26. Crossed polars. Caliche is present around the
edges of the large vughs in the image. Natural mineral inclusions, predominately silt sized quartz inclusions, are
present in a mosaic speckled b-fabric.
across all cultural phases contained monocrystalline quartz, muscovite mica, and potassium
feldspar (Figure 8.7). Other common mineral inclusions present in the majority of samples
include olivine, pyroxene, plagioclase feldspar, amphibole, and calcite. To a lesser extent,
epidote, chalcedony, apatite, serpentine, and sandstone were found in samples across all periods
as small detrital grains. All of these inclusions are naturally occurring minerals found in the raw
clay paste. Virtually all of the natural inclusions in the paste fall into the silt size class, though a
small percentage is classified as fine sand. Most of the samples contained opaque minerals that
could not be definitively identified. Using Whitbread’s (1989, 1995:386-387) methodology,
opaques could be categorized as amorphous concentration features (ACFs) given their
unidentifiable nature. In the present research, they are referred to simply as “opaques.”
Carothers (1992:310) and Galaty (1999:50) both suggested that opaque minerals in thin section
are most likely hematite and less often magnetite. Calcite is found in two forms in the samples.
Occasionally, large angular grits are observed, while less common were amorphous crystallized
grains of calcite that resulted from the transformation of calcite at high temperatures. Calcium
carbonate is visible in nearly all of the samples in both hand sample and microscopically as
caliche. Caliche is easily distinguished from the other forms of calcite described here, and is
166
Figure 8.9. Sample 007, Fabric E2. Neolithic from Gerla 64. Crossed polars. Unprocessed clay nodule inclusion.
Note sub-angularity of inclusion and inclusion boundary incorporated into surrounding matrix near top-left of
image.
Figure 8.10. Sample 008, Fabric D1. Early Bronze Age from Szeghalom 80. Crossed polars. Grog temper. Note
angularity of inclusion and distinctive void space around edges. Fire-clouding is visible (inclusion is darker at
bottom), indicating a previous firing episode.
167
formed as the result of secondary crystallization of calcite in and along the edges of voids as
water percolated through the porous ceramic as a post-depositional process (Figure 8.8).
Another inclusion found in clay pastes on the Great Hungarian Plain – amorphous clay
inclusions resembling intentional temper – are almost ubiquitous in the samples (Figure 8.9).
Under Whitbread’s (1989, 1995:386-387) classification system, these clay inclusions could be
either amorphous concentration features (ACFs), similarly to opaques, or as textural
concentration features (TCFs). In this research, they are categorized as TCFs, but are referred to
simply as nodules or clay nodules. In thin-section, clay nodules are similar in appearance to grog
– bits of crushed fired ceramic material intentionally added by the potter to the clay paste before
forming the vessel. However, the nodules characteristically differ from grog in subtle ways.
Grog is typically highly angular and usually surrounded by void space resulting from water loss
and contraction during firing (Figure 8.10). Conversely, clay nodules are typically sub-angular
(often completely round), and commonly have little or no void space separating them from the
surrounding paste. This difference occurs due to natural clay inclusions contracting during firing
along with the surrounding paste, while the previously fired grog temper does not contract during
the firing process. Additionally, grog may exhibit signs of fire-clouding, or color differences
creating during the initial firing episode. Such color differences are not present on naturally
occurring clay nodules.
It is therefore likely that the sub-angular clay nodules found almost universally in
samples of prehistoric ceramics on the Hungarian Plain are not intentionally added temper.
Rather, it is almost certain that they are unprocessed remnants of the base clay material, left over
from the preparation process (Whitbread 1995:387). As small portions of the raw clay material
from which the paste was derived, the number and size of the nodules is dependent on the degree
to which the raw material was crushed before wetting and forming, or possibly crushed or
separated during levigation. Resulting from this process, the clay nodules contain silt and fine
sand inclusions and occasionally void space, in almost identical proportions to the surrounding
clay matrix. On the other hand, due to their ubiquitous nature whether or not the presence of the
clay nodules in the paste is due to the actions of the potter is difficult to ascertain. The
possibility exists that the nodules occur as an intentional byproduct of the production sequence,
or they were processed separately from the same source material and added during the mixing or
folding of the clay before vessel formation. Given these possibilities and the uncertainty of their
168
provenance, clay nodules were counted as temper during point counting, although the presence
of nodules, grog, and other potentially intentional temper were recorded separately.
Only occasionally present in samples from all time periods, the occurrence of grog
increases slightly in the Late Copper Age and in all subsequent cultural phases. However, point
counted samples from every period contained grog (Table 8.9), though not nearly as frequently
as clay nodules. Few samples contained grog in proportions higher than 5% of the clay matrix,
and in the vast majority of cases only a single grog fragment was identified in the sample. This
suggests that the grog played no functional role in preventing cracking while drying or firing
(Kreiter 2005). Although it is not impossible to rule out the intentional or ritualistic addition of
grog temper, especially beginning late in the Copper Age (see Kreiter 2005), it seems more
likely that these rare additions were accidental on the part of the potter, and are small fragments
of broken sherds that unintentionally found their way into the clay paste during the processing
and preparation stages of ceramic production.
Although they are a common temper used throughout the world, no prehistoric ceramic
samples from the Hungarian Plain included in this analysis contained sedimentary or
metamorphic rock fragments that were beyond a doubt intentional temper. The appearance of
crushed fragments of polycrystalline quartz is often indicative of intentional tempering, and
polycrystalline quartz did appear in numerous samples from multiple periods. However, their
sporadic appearance and low inclusion percentage, as well as their sub-angular, weathered
appearance, suggest that they, too, are naturally occurring minerals in the clay source. Indeed,
the ceramic petrographic samples from the study region throughout the periods sampled can be
described as generally homogeneous, with occasional localized or exceptional variability. The
possible exception to this trend is the increased presence of grog temper in the Late Copper Age,
and an even greater increase in the Early and Middle Bronze Age samples (grog is present in
greater than 60% of both Early and Middle Bronze Age samples, Table 8.9).
Diachronic Petrographic Variability: The Middle and Late Copper Ages, and Early and Middle
Bronze Ages
The Middle Copper Age. As was the case with the macroscopic analysis, the Middle and
Late Copper Age petrographic samples exhibit a great deal of similarity in petrographic
169
Table 8.9. Summary statistics of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle Bronze Age
petrographic point counts from sites in the Körös study region.
PASTE
% Matrix
% Sand
BODY
% Silt
%Mat/Silt
% Sand
% Temp
Period
MCA
Average
85.42
0
14.58
98.4
0
2.6
(n=14)
Std. Dev.
6.16
0
6.16
2.65
0
2.75
LCA
Average
83.14
0.54
16.32
94.21
0.53
4.1
(n=89)
Std. Dev.
8.67
1.67
8.55
11
4.01
1.65
EBA
Average
82.12
0.13
18.75
91.57
0.11
6.52
(n=15)
Std. Dev.
6.42
0.35
6.42
8.33
0.29
4.21
MBA
Average
84.2
0.61
15.19
89.29
1.61
9.1
(n=10)
Std. Dev.
2.92
1.29
2.86
5.63
3.26
6.12
description and point count analysis (Table 8.9). Middle Copper Age Bodrogkeresztúr sherds
exhibit remarkably similar fabric composition, consisting almost entirely of silt-sized naturally
occurring mineral inclusions. Inclusions are usually single or double-spaced (Figure 8.11), with
one open-spaced exception. When averaged over all samples, paste consists of 85% clay matrix
and 15% silt (Figure 8.12), while the body consists of 97% matrix and silt, and 3% temper
(temper consists primarily of clay nodules, which were likely unintentional on the part of the
potter) (Figure 8.13).
Natural mineral inclusions consist predominantly of quartz, potassium feldspar, and
muscovite mica lathes. Samples are otherwise undifferentiated in terms of natural inclusions,
which consist of occasional, rare, or very rare amphibole, pyroxene, and calcite. All samples are
optically active under crossed polars, with a mosaic speckled birefringent groundmass.
Half of the samples exhibit a slight preferred orientation of muscovite mica at an
approximate 45° angle toward the surface of the sherd, while the others had no clearly
identifiable preferred orientation of natural mineral inclusions. Each of the samples contains
unprocessed clay nodules with diffuse and merging inclusions boundaries, and two of the
samples contain one piece each of angular grog. In both cases, the matrix of the grog is sorted
differently and contains a different ratio of natural mineral inclusions than the paste that
surrounds them. The fabric of one sample (Figure 8.14) contains a mixture of clays, evidenced
by slight differences in paste color, a folded-over appearance, and a slightly different pattern of
170
Figure 8.11. Sample 030, Fabric E1. Middle Copper Age from Szeghalom 80. Crossed polars. Typical example of
MCA mosaic speckled b-fabric, with a slight striated fabric area near the center of the image. Muscovite mica
displays an angular preferred orientation. Other natural inclusions are primarily silt-sized monocrystalline quartz.
Note the uncommon channel void near the center of the image.
sorting and spacing between the clay types. Voids are characterized in all samples as either
planar voids – long, thin void spaces that develop as a formed pot dries prior to firing – or as
vughs – large, irregular voids that remain following the kneading and shaping process.
All Middle Copper Age ceramics fell into three fabric categories – E1, E2, and F1
according to Riederer’s (2004) classification system. This indicates an extremely fine, silty
paste, with occasional larger quartz and potassium feldspar inclusions. All three fabrics are
similar, with fabric F1 containing slightly smaller silt-sized natural mineral inclusions than the E
fabrics. The fabrics are undifferentiated in terms of natural mineral inclusions, optical
birefringence, and characterization of void space.
The Late Copper Age. Late Copper Age Boleráz and Baden petrographic samples are
also similar in terms of paste, body, and natural and intentional inclusions (Figure 8.15). Paste
consists of 83% matrix, less than 1% sand, and 16% silt-sized natural inclusions (Table 8.9,
Figure 8.12). Body consists of 94% matrix and silt, less than 1% sand, and 4% clay nodules and
grog temper (Figure 8.13). Grain distribution varies from single to double-spaced, with very rare
171
% Silt
MCA
LCA
60
90
10
% Sand
% Matrix
40
10
40
Figure 8.12. Ternary plot of Middle Copper Age and Late Copper Age ceramic paste composition. Though
compositionally similar, Late Copper Age samples are often slightly sandier than Middle Copper Age samples.
% Sand
60
40
90
10
%
40
% Matrix + Silt
MCA
LCA
10
Temper
Figure 8.13. Ternary plot of
Middle Copper Age and Late Copper Age ceramic body composition. Note higher
proportions of temper and sand in the Late Copper Age.
172
Figure 8.14. Sample 137, Fabric E1. Middle Copper Age from Szeghalom 168. Crossed polars. Mixing of
different clays is evident in slight color variation and inclusion spacing visible between left and right side of image.
open-spaced exceptions. Natural mineral inclusions are similar to those in Middle Copper Age
samples, consisting predominately of monocrystalline quartz, potassium feldspar, and muscovite
mica lathes. Also present are occasional calcite, olivine, and pyroxene, and rare or very rare
instances of amphibole, chlorite, chalcedony, biotite mica, epidote, apatite, and serpentine.
Hematite staining is occasionally present in the samples, and three unintentional lithic inclusions
are present in three separate samples. All samples exhibit an optically birefringent fabric,
characterized predominately as mosaic speckled with occasionally striated areas. All samples
contain large and small vughs, most contain planar voids, with vesicles and channel voids only
rarely observed. Muscovite mica expresses a slightly diagonal preferred orientation to the
surface of the sherd in the majority of samples, though samples with no obvious preferred
orientation of naturally occurring minerals are common. Like Middle Copper Age samples, the
majority of Late Copper Age samples fall into fabric categories E1, E2, and F1 under Riederer’s
(2004) fabric classification system. Fabrics with slightly larger natural mineral inclusions and
slightly sandier fabrics are rare. The fabrics are undifferentiated in terms of natural mineral
inclusions, optical birefringence of groundmass (all active), or characterization of void space.
173
Figure 8.15. Sample 54, Fabric E2. Late Copper Age from Tarhos 67. Crossed polars. Typical mosaic speckled bfabric of Late Copper Age samples. Natural inclusions include monocrystalline quartz and muscovite mica lathes.
Note clay nodules (dark areas in top right corner of image) with diffuse inclusion boundaries.
Figure 8.16. Sample 125, Fabric E1. Late Copper Age from Mezőgyán 2. Crossed polars. Typical mosaic
speckled b-fabric of Late Copper Age samples. Large angular grog inclusion with evidence of previous smoothing
or burnishing (light grey color at bottom of grog). Large vugh in bottom of picture.
174
Figure 8.17. Sample 42, Fabric F1. Early Bronze Age from Szeghalom 80. Crossed polars. Typical mosaic
speckled b-fabric of Early Bronze Age samples, including unintentional clay nodules. Natural inclusions include
monocrystalline quartz and potassium feldspar.
Interestingly, 25 of 89 (30%) of Late Copper Age samples contain grog in addition to
naturally occurring clay nodules. This is in contrast to the rare occurrence of grog in Middle
Copper Age samples. Grog is differentiated from clay nodules by its angular shape, the presence
of sharp boundaries between the grog and surrounding paste, and commonly the presence of void
space around the grog separating it from the paste (Figure 8.16). This difference is reflected in
the percent of the body consisting of temper – less than 3% in the Middle Copper Age samples,
and just over 4% in the Late Copper Age samples.
The Early Bronze Age. Early Bronze Age ceramic petrographic samples from the Körös
region (n=15) are similar to ceramics of other cultural phases included in this analysis. Early
Bronze Age paste, on average, consists of 82% matrix, less than 1% sand, and 18% silt, mostly
in the form of naturally occurring mineral inclusions (Table 8.9, Figures 8.17 and 8.18). Body
consists on average of 92% matrix and silt, less than 1% sand, and 7% temper (Figure 8.19).
Temper occurs in the forms of unprocessed clay nodules and intentional grog inclusions.
175
% Silt
60
LCA
EBA
90
10
% Matrix
40
10
% Sand 40
Figure 8.18. Ternary plot of Late Copper Age and Early Bronze Age paste composition. Both exist within a
common range of variability, though the Late Copper Age group contains sandier outliers.
% Sand
LCA
EBA
90
10
%
40
% Matrix + Silt
60
40
10
Figure 8.19. Ternary plot ofTemper
Late Copper Age and Early Bronze Age body composition. Note an increased number
of tempered samples in comparison to the Middle Copper Age. More frequent grog temper accounts for the increase
in the proportion of temper in the body. Additionally, note the proportionally infrequent occurrence of samples
falling into the arbitrary “heavy fraction” containing greater than 10% temper inclusions.
176
Figure 8.20. Sample 8, Fabric D1. Early Bronze Age from Szeghalom 80. Crossed polars. Example of a striated
active b-fabric.
Grog continues to increase in frequency in Early Bronze Age ceramic material, occurring
in 73% (n=11) of the samples. Once again, an increase in the percentage of temper pointcounted in the ceramic body indicates the more frequent use of temper. However, grog in Early
Bronze Age ceramics still does not occur in significant enough proportion to affect the drying or
firing of the vessel.
Naturally occurring minerals in the clay paste consist almost exclusively of single and
double-spaced silt-sized particles. Predominate minerals include monocrystalline quartz,
potassium feldspar, and muscovite mica. Plagioclase feldspar and calcite are common, with
pyroxene, amphibole, and unidentifiable opaque mineral inclusions occurring less frequently.
Epidote, olivine, chlorite, apatite, and serpentine are observed only rarely or very rarely. As in
other periods, several Early Bronze Age samples contain hematite staining.
All of the samples feature an optically active birefringent groundmass. Most b-fabrics
are characterized as mosaic speckled, with striated b-fabrics observed slightly more often than in
the Middle or Late Copper Age samples (Figure 8.20). A preferred orientation of muscovite
mica lathes is observable at an approximately 45° angle to the surface of the sherd in most
177
Figure 8.21. Sample 16, Fabric E2. Middle Bronze Age from Békés 26. Crossed polars. Mosaic speckled and
striated b-fabric containing naturally occurring monocrystalline quartz and muscovite mica inclusions.
samples, though some samples have no clear preferred orientation of natural mineral inclusions.
Like samples from previous cultural phases, Early Bronze Age samples fall into fabric types E1.
E2, and F1 in Riederer’s (2004) classification system. However, one sample fell into the slightly
sandier fabric D1, and one falls under the extremely silty F1 fabric. Void space consists
primarily of irregular vughs, and less commonly planar voids that formed during the vessels’
drying phase. Vesicles are not common, and channels are extremely rare. Within the period, the
fabrics are undifferentiated in terms of groundmass description, natural mineral inclusions, and
characterizations of void space.
The Middle Bronze Age. Middle Bronze Age ceramic petrographic samples from sites in
the Körös region (n=10) exhibit homogeneity within the cultural phase (Figure 8.21). On
average, Middle Bronze Age paste consists of 84% matrix, less than 1% sand, and 15% silt
(Table 8.9, Figure 8.23). The paste exhibits a significantly decreased range of compositional
variability compared to earlier periods (Figure 8.23), which points toward increasing
specialization of pottery manufacture (Budden and Sofaer 2009; see discussion below). The
body consists of 89% matrix and silt, 2% sand, and 9% temper inclusions (Table 8.9, Figure
8.24). Temper inclusions include naturally occurring clay nodules and intentionally added grog
178
Figure 8.22. Sample 80, Fabric F2. Middle Bronze Age from Békés 178. Crossed polars. Medium-sized
polycrystalline quartz inclusions in a mosaic speckled b-fabric.
temper. One sample contains frequent sand-sized polycrystalline quartz inclusions (Figure 8.24).
Abundant, angular inclusions of polycrystalline quartz often indicate intentional tempering of the
paste; however, since only one sample exhibits this unusual characteristic and no pattern can be
established, no definite conclusions can be drawn regarding the intentionality or function of large
quartz inclusions in this sample.
Naturally occurring mineral inclusions consist predominantly of single- and doublespaced monocrystaline quartz, potassium feldspar, and muscovite mica lathes. Polycrystalline
quartz, pyroxene, calcite, and amphibole inclusions are not as common, but still frequently
observed. Less common mineral inclusions include olivine, pyroxene, and plagioclase feldspar,
with serpentine, chalcedony, biotite, and unidentifiable opaque minerals are rarely observed.
Grog occurs in 60% of the samples (see Figure 8.25). Though this is a reduction in
frequency compared to the Early Bronze Age, grog remains more common than in either the
Middle or Late Copper Age. Interestingly, the average percentage of temper in the body of
Middle Bronze Age (n=15) samples increases from 7% in the Early Bronze Age (n=10) to 9%,
despite a reduction in grog frequency. This is due in part to a general increase in size of both
intentional grog temper and unprocessed clay nodules.
179
% Silt
40
90
10
% Matrix
EBA
MBA
60
10
% Sand 40
Figure 8.23. Ternary plot of Early Bronze Age and Middle Bronze Age paste composition. Note a decreased range
of variability in Middle Bronze Age samples compared to all previous cultural phases.
% Sand
EBA
MBA
90
10
%
40
% Matrix + Silt
60
40
10
Figure 8.24. Ternary plot of Early Bronze Age and Middle Bronze age body composition. Though compositionally
Temper
similar to previous periods, a higher proportion of Middle Bronze Age samples fall into the arbitrary “heavy
fraction” containing greater than 10% temper inclusions.
180
Figure 8.25. Sample 17, Fabric F2. Middle Bronze Age from Békés 26. Large grog inclusion with fire clouding,
in speckled and striated b-fabric.
All of the samples possess an optically active birefringent groundmass. Most b-fabrics
are mosaic speckled, with striated b-fabrics observed roughly as equally as in Early Bronze Age
samples, but more frequently than in Middle or Late Copper Age samples (Figure 8.25). A
preferred orientation of muscovite mica lathes exists at an angle to the surface of the sherd in the
majority of samples, though some samples have no clear preferred orientation of natural mineral
inclusions. Middle Bronze Age samples fall into a wider variety of fabric types than previous
periods, including fabrics E1, E2, E3, F1, and F2 according to Riederer’s (2004) fabric
classification system. This wider variety of fabric types suggests a slight trend toward sandier
fabrics, including frequent grog inclusions, and a higher frequency of polycrystalline quartz
when compared to other periods. Additionally, sand-sized monocrystalline quartz, which occurs
naturally in the clay parent material, occurs more frequently in Middle Bronze Age samples.
Voids are characterized similarly to earlier periods, consisting primarily of irregular
vughs with very common planar voids, caused by vessel shrinkage during the drying process.
Channels and vesicles occur only very rarely. The fabrics within the Middle Bronze Age are
undifferentiated in terms of groundmass description, mineral inclusions, and characteristics of
void space.
181
Summary of Diachronic Variability. In a general sense, Middle Copper Age, Late
Copper Age, Early Bronze Age, and Middle Bronze Age petrographic samples fall into a
common range of variability (see Figures 8.26 and 8.27). Despite this homogeneity, however,
subtle differences and consistent changes over time are present, and must be discussed. Most
notably, an increase in the frequency of intentionally tempered ceramics is visible between the
Middle Copper Age and the Middle Bronze Age. Although the first significant appearance of
tempered samples occurred after the Middle Copper Age, the general trend for the approximately
2,000 year time span encompassing the Middle and Late Copper Age and Early and Middle
Bronze Age is a subtle, but steady and marked, increase in the frequency of temper, especially
grog, observed petrographically (see Figures 8.28 and 8.29). Accounting for this increase is
important, as it indicates a continuous trend rather than an abrupt change in vessel manufacturing
technology.
The increasing temper trend is observable by establishing a set of arbitrary heavy and
light fractions, with samples containing more than 10% temper (grog, clay nodules, and
intentional mineral inclusions) in the body designated the heavy fraction, and samples consisting
of less than 10% temper in the body as light fraction. Very few samples contained greater than
10% sand in the paste, therefore naturally occurring sand is not considered in this example.
All Middle Copper Age samples fall into the light fraction, with none of the samples
containing more than 10% temper in the body (see Figures 8.13and 8.26). In fact, sand-size
particles were not observed in any Middle Copper Age samples in this study. The paste in these
samples is extremely silty, containing primarily of a very fine, highly processed, clay matrix.
This concurs with previous studies (see Hoekman-Sites et al. 2007; Parsons 2005) on Early
Copper Age Tiszapolgár pottery, which also contained very little sand. Conversely,
approximately 12% of Late Copper Age samples fall into the heavy fraction. This change is
explained by the more frequent appearance of grog in thin-section; though, it should be
emphasized that grog temper never occurs in a proportion great enough to affect vessel stability
during the drying or firing process. Though unusual, Kreiter (2005) noted a similar phenomenon
in Transdanubian Bronze Age ceramic assemblages. The addition of grog in such small amounts
could be an accidental byproduct of the production sequence, or as Kreiter suggested, it could be
an intentional, non-discursive action not related to vessel structure.
182
% Silt
MCA
LCA
EBA
MBA
60
90
10
% Sand
% Matrix
40
10
40
Figure 8.26. Ternary plot of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle Bronze Age
paste compositional variability. A trend toward the inclusion of more grog temper is observed over time (see text,
Table 8.1).
% Sand
60
40
90
10
%
40
% Matrix + Silt
MCA
LCA
EBA
MBA
10
Figure 8.27 Ternary plot of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle Bronze Age body
compositional variability.Temper
Note higher percentage of temper in Early Bronze Age samples.
183
100%
90%
80%
70%
60%
%Mat/Silt
50%
% Sand
40%
% Temp
30%
20%
10%
0%
MCA
LCA
EBA
MBA
Figure 8.28. Body composition of Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle Bronze
Age ceramic samples. A steady shift in the ratio of temper (including grog) to sand and matrix plus silt is
observable over time. This shift is subtle, but important, as it indicates minor, continuous change in vessel
manufacturing rather than an abrupt change. Such a trend indicates continuity in manufacturing technology over the
long term.
10%
9%
8%
7%
6%
5%
% Temp
4%
% Sand
3%
2%
1%
0%
MCA
LCA
EBA
MBA
Figure 8.29. Percentage of temper and sand observed in point counted Middle Copper Age, Late Copper Age, Early
Bronze Age, and Middle Bronze Age ceramic samples. Note the marked trend of an increased appearance of temper
(especially grog) over time.
184
Early Bronze Age samples fall into the heavy fraction at a rate of approximately 16%
(see Figures 8.19 and 8.26), a 5% increase over the Late Copper Age. Again, this is due to an
increase in the frequency of grog temper observed petrographically, with grog recorded in more
than 70% of Early Bronze Age samples. Moreover, 50% of Middle Bronze Age samples fall into
the heavy fraction. As with Early Bronze Age material, a very high percentage (60%) of the
samples contains grog. Ultimately, this long-term, measured increase in average percent of
temper in ceramic material does not point to dramatic changes in ceramic manufacturing
technology at any point during the time period covered by this study.
Spatial Variability in Late Copper Age Ceramics
Ceramic variability between sites in the Körös region is measured as a control for the
observation of change over time in the region. In other words, marked heterogeneity in ceramic
assemblages between sites within the region would confound results focused on identifying
change over time based on averages and standard deviations of materials from multiple sites.
Additionally, the identification of subtle differences in assemblages from different sites would
indicate the maintenance of localized manufacturing techniques, rather than a replacement of
local practices with outside methods.
The observation of inter-regional petrographic variability also serves as a control
measurement for the diachronic ceramic data. Additionally, it allows for speculation on the roles
of regional ceramic traditions and different regional geology on changes in material culture at the
end of the Copper Age.
Petrographic variability in the Körös Region. Very little variability was measured in the average
paste and body composition ratios of Late Copper Age ceramics in the Körös River study region
(Figure 8.30). When averaged by site, paste composition never exceeds 4% sand, and is
described as silty clay. Silt content exists in a wide range between ceramic samples from sites
across the region, ranging from 5% to 23% silt. This does not indicate the intentional addition of
temper or removal of naturally occurring mineral inclusions. Rather, this range suggests slight
differences in mineral composition of raw clay sources, perhaps weathered from different parent
materials upriver (Frolking 2009).
185
% Silt
40
90
10
% Sand
% Matrix
60
10
40
Figure 8.30. Ternary plot of Late Copper Age paste compositional variability in the Körös Region. Each colored
square represents the average paste value of one site.
% Sand
40
90
10
%
% Matrix + Silt
60
10
40
Figure 8.31. Ternary plot ofTemper
Late Copper Age body compositional variability in the Körös Region. Each colored
square represents the average paste value of one site.
186
A similar homogeneity exists in the average body composition of ceramics from sites in
the Körös region (Figure 8.31). A consistent range of variability was measured between the
sites, with no site average consisting of more than 10% temper and 4% sand. This indicates a
standardized methodology for raw clay gathering, processing, and vessel production in the Körös
region during the Late Copper Age.
Inter-regional petrographic variability. The petrographic results of ceramics from the
site of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the Maros region (n=5) match the results
of the macroscopic analysis, but they should be considered cautiously due to the small sample
size. Natural mineral inclusions are similar to samples from the Körös region and consist
primarily of monocrystalline quartz, potassium feldspar, and muscovite mica lathes. Less
common minerals include pyroxene, amphibole, and opaque minerals. Serpentine, apatite,
chlorite, calcite, and chalcedony are observed only rarely.
The paste composition of ceramics from the two regions is virtually identical and is
statistically indistinguishable (Table 8.10, Figure 8.32). This similarity in paste composition
suggests similar choices in raw material procurement and clay processing in the regions during
the Late Copper Age. On the other hand, differences in ceramic composition between the Körös
and Maros regions are evident in the ceramic body composition (Table 8.10, Figure 8.33).
Matrix and silt compose 94% and 88% of the fabric in the Körös and Maros regions respectively,
and temper composes 4% and 9% respectively. Two factors explain the difference. First, grog
temper is present in 60% of the Maros samples – a much higher frequency than in Late Copper
Age samples from the Körös region. Second, the Maros samples fall into fabrics D1, D2, E1,
and E2 according to Riederer’s (2004) fabric classification system. The D fabric classifications
are slightly sandier than fabric classes E and F; and indeed, large monocrystalline and
polycrystalline quartz inclusions are present in one of the Maros samples. The high frequency of
polycrystalline quartz suggests that the mineral may have been crushed and intentionally added
to the raw clay paste during the manufacturing process. However, the large quartz inclusions
also exist within unintentional clay nodules in the sample, suggesting that the inclusions are
naturally present in the source material. It is therefore difficult to say with certainty that the
quartz is intentional temper, though it cannot be ruled out as a possibility.
187
Körös
% Silt
60
Maros
90
10
% Matrix
40
10
% Sand 40
Figure 8.32. Ternary plot of average paste composition of Late Copper Age ceramics from the Körös region and
from the site of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the adjacent Maros region.
% Sand
Körös
60
40
90
10
%
40
% Matrix + Silt
Maros
10
Figure 8.33. Ternary plot ofTemper
average body composition of Late Copper Age ceramics from the Körös region and
from the site of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the adjacent Maros region. Note the higher average
ratio of temper to sand and matrix plus silt in the Maros samples.
188
Table 8.10. Summary statistics of Late Copper Age point counts from the Körös and Maros regions..
PASTE
BODY
% Matrix
% Sand
% Silt
%Mat/Silt
% Sand
% Temp
Average
83.14
0.54
16.32
94.21
0.53
4.1
Std. Dev.
8.67
1.67
8.55
11
4.01
1.65
Average
82.22
0.7
18.08
88.92
0.6
8.69
Std. Dev.
8.69
1.57
8.47
5.25
5.05
1.32
Region
Körös
Maros
Discussion of the Petrographic Data
As with the macroscopic data, the materials evaluated as part of the petrographic study
exhibit a common range of characteristics over time and space. Ceramic fabrics from all periods
tend to be very silty and have similar compositional signatures. Although all samples fall into a
shared range of variability diachronically, subtle but consistent change in ceramic body
composition is observable over time, especially in terms of the temper to sand ratio, and the ratio
of matrix to silt. The addition of grog temper and its increasing frequency in the Late Copper
Age and Early Bronze Age does not happen abruptly, however, and grog does not occur in
amounts significant enough to mitigate undesirable effects – such as cracking – in the drying and
firing stages of the production sequence. Homogeneity of body and paste composition, natural
mineral inclusions, preferred orientation of muscovite mica lathes, optically active groundmass,
and mosaic speckled b-fabric across time and space also suggests continuity in the collection of
raw materials, preparation of clay, vessel forming techniques, and firing process through the time
periods covered in this study.
Despite significant changes in vessel form and decoration during the transitions between
the Middle Copper Age, Late Copper Age, Early Bronze Age, and Middle Bronze Age, vessel
production technology and the firing process underwent few significant modifications during this
time period. When observed, changes in body composition occurred gradually – not indicative
of significant change in production technology. Ultimately, the petrographic data do not support
a migration or invasion hypothesis for material culture change at the beginning of the Late
Copper Age.
189
Summary
In this chapter, I presented the data and the results of the macroscopic and microscopic
petrographic ceramic analysis. The macroscopic analysis, petrographic point-counting, and
petrographic general description all suggest continuity in ceramic production technology during
the approximately 1,500 year span covering the Middle Copper Age, Late Copper Age, Early
Bronze Age, and Middle Bronze Age. This includes similar choices in raw material collection
and common behaviors in processing raw materials, techniques of vessel production, and firing
processes. Interestingly, a long-term change in body composition to include a higher percentage
of grog by the latter periods is observed, as is a more restricted range of compositional variability
in Middle Bronze Age materials. This trajectory points to an evolving non-discursive knowledge
in ceramic production over time, as well as specialization in ceramic production by the Middle
Bronze Age.
Importantly, the changes described in this chapter over the long-term do not indicate an
influx of a migratory population producing vessels under a different discursive or non-discursive
template, nor do they indicate the penetration of foreign ideas of pottery production into the
practices of the local population. In short, both the macroscopic and petrographic data support a
model of population continuity in the Körös River study region, as ceramic production methods
and non-discursive knowledge of potting did not change following the appearance of kurgans on
the Great Hungarian Plain. A migration of peoples into the region who drastically altered
production methods, material culture, and settlement patterns is not supported by this data.
190
CHAPTER NINE
DISCUSSION: MODELS OF CHANGE ON THE GREAT HUNGARIAN PLAIN DURING
THE LATE COPPER AGE
Introduction
In this chapter, I discuss the implications of the results presented in Chapters Seven and
Eight for the general anthropological models outlined in Chapter Two. I also discuss how these
interpretations affect our understanding of the specific archaeological models presented in
Chapter Three. This chapter integrates the spatial and ceramic data generated as part of this
research project with those general and regional models.
The multiple-resolution spatial analysis presented in this dissertation demonstrates
patterns of nucleation and dispersal and a tendency toward settlement dispersal in the Körös
region and the central Hungarian Plain from the Late Neolithic period through the Late Copper
Age. The broad pattern observed by Sherratt (1997a, 1997b) in his study region in northern
Békés County holds true at the larger scale, lower resolution Körös River watershed. His
observation that kurgan burial mounds constructed by a presumably migratory population on the
eastern Plain during the Late Copper Age are spatially discrete from Late Copper Age Boleráz
and Baden archaeological sites also holds true at the scale of the Körös region. However,
exceptions on the micro-level at high resolution where kurgans and Late Copper Age sites are
within less than 1,000 meters of each other are common. Ultimately, the patterns observed in this
study concur with Sherratt’s model, rather than with models of migratory change.
Similarly, the results of the macroscopic ceramic analysis presented in Chapter Eight
support a model of long-term population continuity rather than a model of invasion or migration.
Data regarding raw material processing, vessel forming, and surface treatment do not suggest the
introduction of foreign pottery production methods. In fact, the opposite is indicated. More
variability in production methodology was observed between sites within the Körös region and
than between cultural phases. The results of the petrographic data also support a model of
production technology continuity throughout the Middle and Late Copper Ages and Early and
Middle Bronze Ages. Samples from all periods fall within a common range of variability.
Importantly, however, changes in production methods over the long-term were noted. A subtle
but steady increase in the use of grog temper was measured during the approximately 2,000 year
time span between the Middle Copper Age and Middle Bronze Age. This does not indicate the
191
appearance of new production methods, and supports a model of population continuity rather
than a migration model.
The sections in this chapter address the models and cultural trajectories discussed in
Chapters Two and Three. They incorporate the results of the present study into our current
archaeological understanding of social and settlement change on the Great Hungarian Plain
during the Late Copper Age, and how a refined understanding of the nature of changes during
this period affects our understanding of later prehistory in the region.
Modeling Change on the Great Hungarian Plain
Kurgan Builders, Migration, and the Late Copper Age
Discussion of the spatial results. As Parkinson (2006b) noted, and as is indicated in this
dissertation, the social and cultural tapestry of the Late Copper Age on the Great Hungarian Plain
cannot be fully understood until the appearance of the kurgan burial mounds is explained, and
they are aligned definitively with specific cultural phases. This, of course, has not yet been
achieved. Unfortunately, the exact nature of the relationship between the kurgans and the people
of the Baden material culture in the Körös region remains unclear.
Some of the results presented in this research project focus on reevaluating
archaeological perspectives on migration, and migration’s impact on society and settlement
patterns. This speaks directly to the appearance of kurgans on the Hungarian Plain, and although
the results of this project do not resolve the issue of relationship between kurgans and Late
Copper Age settlements, it does contribute to the discussion – both in terms of the spatial
analysis results and the results of the ceramic analyses.
This study determined that kurgans and Late Copper Age Boleráz and Baden settlements
exist in a complementary distribution at the scale of the Körös River watershed study area. At
higher resolutions throughout the region, though, Baden archaeological sites and kurgans exist
quite close to one another. Furthermore, average nearest neighbor data are at best inconclusive
in statistically demonstrating Sherratt’s (1997b) claim of ongoing settlement dispersal during the
Middle and Late Copper Ages (on the other hand, distribution maps of Middle and Late Copper
Age sites do show a decrease in site number during this time period). Given these data,
Sherratt’s conclusions are supported. Unfortunately, many questions remain regarding the
relationship between kurgans, their builders, and the indigenous people of the Körös region.
192
Sherratt’s (1997a) analysis of the spatial relationship between kurgans and Late Copper
Age sites concluded that, for one reason or another, kurgans were constructed away from Baden
settlement locations, or Baden settlements were intentionally placed at a distance from kurgans.
The results of this study confirm this observation in Sherratt’s study area in northern Békés
County, at a low resolution across the entire study area. Kurgans and Late Copper Age sites are
located quite close to one another (sometimes within a few hundred meters) at multiple locations
throughout the study area when observed at higher resolution. This calls into question portions
of Sherratt’s conclusions. Specifically, a strict strategy of avoidance between contemporaneous
kurgan builders and an indigenous Late Copper Age population is untenable given the proximity
of Baden settlements and kurgans in the study region. Unfortunately, lacking temporal control
for the construction of the kurgans it remains impossible to describe what, if any, implication this
has for understanding the relationship between kurgans and Baden.
The implications of the kurgan density analysis go beyond providing new data on the
spatial relationship between kurgans and Late Copper Age archaeological sites in the Körös
region. The differences in results and interpretation of spatial analyses in Sherratt’s (1997b)
study and the present study illustrate the importance of measuring regional and local patterns at
multiple analytical scales. The use of multiple resolutions in this study has redefined the
archaeological understanding of the relationship between Baden settlement and kurgan building
on the Great Hungarian Plain. Though much remains to be learned, this is a small but important
step toward understanding when and by whom the kurgans were constructed.
What is clear about the kurgans on the eastern Hungarian Plain at this time is their
resemblance to the monumental burial architecture of the Yamnaya of the Eurasian Steppe.
Given the archaeological evidence (see Escedy 1979), a model of independent development for
kurgans on the Plain, or even their appearance through diffusion without a migration, seems
untenable. Indeed, in his discussion of the kurgan phenomenon, Anthony (1990) suggested that
if a migration were to have taken place, the patterned kurgan distribution across eastern Europe
into the Great Hungarian Plain is what one would expect a migration to look like. However,
several vexing issues complicate the migration scenario. First, and perhaps most importantly,
where are the settlements of the kurgan builders? Up to this point, no convincing evidence of
kurgan builders exists in the region except the tumuli themselves, and the skeletons and pottery
deposited below them. It is of course possible that that they constructed temporary shelters that
193
would leave little or no trace in the archaeological record. As such, expecting to find
architecture as a mark of their presence may be unreasonable. Second, what of the subsistence
and economy of the kurgan builders? Given the pastoral subsistence economy of the Yamnaya,
evidence for a similar strategy would be expected on the Plain, or in any other destination
regions. Yet, no archaeological evidence has yet been found that indicates an increase in
pastoralism or transhumance during the Middle/Late Copper Age transition, though such a
scenario has been suggested for both the appearance of Baden in the region and for the kurgan
builders (see Escedy 1979; Gimbutas 1977; Mallory 1989).
The geology of the Körös watershed itself raises questions regarding Baden and kurganbuilder pastoralism. Although the Körös region in prehistory was ideal for agriculture due to
predictable seasonal flooding and the regular deposition of nutrient-rich fluvial soils (Frolking
2009; Gyucha 2010), the region was not amenable to a pastoral lifestyle. The mosaic of
marshes, inundated areas, large and small rivers, and relatively dry land led prehistoric occupants
of the region to engage in a subsistence strategy featuring the exploitation of domesticated plants
and animals that, through variable over time, remained the predominate subsistence strategy for
thousands of years. On the other hand, the hydrologically more balanced and much drier Maros
River alluvial fan sits immediately adjacent to the Körös River study region, and its grasslands
offered more favorable conditions for large-scale animal keeping and herding (Gyucha 2010).
As such, one would expect to find evidence of pastoral kurgan builders in this region only tens of
kilometers from kurgan “cemeteries” (Escedy 1979; Gimbutas 1979). This is not the case,
however, as no archaeological sites attributed to the kurgan builders (and very few kurgans) have
been located in the Maros region at the present time. Furthermore, if Baden economy had relied
heavily on pastoralism, one would expect to find Baden settlements in the region, and
architectural elements (such as apsidal houses) found at Baden sites in regions outside of the
Carpathian Basin that have been associated with animal husbandry (Mallory 1989). Such
evidence, however, has not been located. Indeed, the Maros region does not come under
substantial settlement until much later during the Iron Age, at which point transhumant
pastoralists occupied the area (Jankovich et al. 1998).
As a result of the lack of archaeological data pertaining to the kurgan builders, a model of
Yamnaya migration onto the Great Hungarian Plain during the Middle/Late Copper Age
transition is easy to envision, but difficult to support concretely. As such, two general
194
possibilities exist for the appearance of kurgans on the Plain at this time: 1) local adoption of a
foreign practice through diffusion and acculturation; or 2) a migration onto the Plain from the
east that left few traces in the archaeological record. Snow (2009) recently addressed similar
migration issues. He stated that a key element of understanding the nature of a prehistoric
migration involves not only having knowledge of its structure, but also estimating (or accurately
determining) the size of the movement. As such, a large-scale migration may occur by which a
large migratory population subjugates a subordinate population in the destination region, with
possible consequences for the subjugated population including displacement, absorption, and
annihilation. Conversely, a relatively small migration or a steady flow of multiple small groups
or bands of people over a long period of time can result in a non-violent arrival in a previously
occupied region, where migratory populations receive subordinate status (Snow 2009:11).
Possible long-term consequences for such a population include absorption into the local
population, isolation (insulation or marginalization), annihilation of the group by locals,
expulsion, or return migration.
In addition to outlining long-term consequences of migration for local and migrant
populations, Snow listed four classes of evidence that one would expect to find preserved from
the donor culture in the destination. In order of significance, they are burials, architecture,
ceramics, and economy (Snow 2009:13). Snow’s justification of selecting these particular
culture characteristics for preservation in the destination region is similar to Lemmonier’s (1992)
description of conservative technologies and Budden and Sofaer’s (2009) illustration of the
importance of non-discursive knowledge and technology in pottery manufacture. To distill their
main ideas into a single thought, certain cultural and technological characteristics are subject to
change relatively rapidly, while other more conservative technological attributes are deeply
imbedded non-discursive knowledge, and are therefore highly resistant to change.
In applying some of Snow’s (2009) argument to the appearance of kurgans of the Great
Hungarian Plain, this project’s spatial analysis tested absorption of a migratory population vs.
isolation of an invasive population by local groups. Sherratt (1997b) argued for the isolation of
the migratory kurgan builders based upon the spatial exclusivity of kurgan burial tumuli and Late
Copper Age Baden archaeological sites. The results of this dissertation’s spatial analysis support
Sherratt’s conclusions that kurgans and Late Copper Age settlements exhibit a complementary
distribution at a low resolution, though a number of exceptions to this rule exist at higher
195
resolutions in the Körös River study area. Ultimately, Sherratt argued for a case of isolation
(1997b:310), by which a relatively small population of kurgan builders arrived in the Körös
region around 3,500 B.C. and became insulated against the surrounding Baden population. This
isolation occurred either through social pressures, or more likely due to the kurgan builders’
selection of previously deforested open areas for stockbreeding. These areas were unpopulated
after half a millennium of thinning Middle Copper Age populations due either to environmental
or economic factors (Sherratt 1997a:281-282, 1997b:309:310). Although such a model is
satisfying, it fails to account for the lack of other archaeological evidence that could explain the
presence of a foreign population in the Körös region at this time (for example, increased
evidence of stockbreeding, campsites, or stronger evidence of interaction with the indigenous
population), or why they apparently failed to exploit the nearby Maros fan that would have been
ideal environment for herding.
As an alternative to Sherratt’s model, I propose a scenario under which the migratory
population of kurgan builders was absorbed into the local population of the Körös region
relatively rapidly after their arrival. This differs from Sherratt’s model, which proposed a
prolonged period of avoidance (perhaps territoriality) between the Körös region’s indigenous
population and the kurgan builders. The results presented in earlier chapters of this dissertation
do not conclusively rule out a migration into the region around the time of the Middle/Late
Copper Age transition, and indeed could be used to support a model of migration with an
extremely limited effect on the cultural or economic trajectory of the Körös region’s indigenous
population. Such a migration scenario fits into the frameworks set forth in this volume,
including those of Sherratt (1997a, 1997b) and Snow (1999), while simultaneously taking into
account Anthony’s (1990) contention that migrating culture groups tend to replicate the parent
culture in simpler form. For the present discussion, Anthony’s replication hypothesis explains
the retention of conservative behaviors – such as burial traditions – that are extremely resistant to
change.
A key component to an absorption scenario is Snow’s belief that demic expansions of
even dominant culture groups like the Yamnaya can result in their absorption by societies in the
destination region if they spread themselves too thinly, or if their numbers are too few to
maintain a dominant presence on the landscape (2009:19). The Huns are an excellent example of
this scenario. Despite their continuous expansion and eventual ubiquity between the Caspian Sea
196
and France, the nature of their society was such that few traces of them are found in the
archaeological record (Thompson 1996). Indeed, by the time the Huns had spread across
Europe, they had all but abandoned pastoralism and were almost entirely dependent on local
subject populations for food and shelter, having almost no means of production. After A.D. 450,
the thinly spread Hun Empire lost their military and strategic edge against the Romans, and Hun
leaders began to look to their own local interests. This led to the dissolution of the empire
without a decisive military battle. Although many Huns may have been killed by local subject
populations or fled east in a return migration, many thousands of Huns were absorbed into local
populations and dissolved into the societies that they once had conquered (Snow 2009:17;
Thompson 1996).
Although no solid archaeological evidence exists to label the Yamnaya or the migrant
kurgan builders as conquerors, the signature of kurgan builders in the archaeological record of
the Great Hungarian Plain is comparable to that of the Huns. The presently available evidence
suggests that a migration, even of modest size, did indeed occur around 3,500 B.C. onto the
Great Hungarian Plain. The migrants constructed burial monuments in their own tradition, but
they were most likely absorbed into local populations relatively quickly. This scenario is
plausible given not only the lack of an archaeological signature beyond monumental burial
architecture, but also the lack of evidence for a significant increase in a pastoral subsistence
economy in the Körös region, or in the adjacent Maros fan that was not intensively exploited
during this time period despite an environment ideal for stockbreeding and herding (Gyucha
2010).
Discussion of the ceramic results. Ceramic data analyzed as part of this study support a
scenario of population continuity between the Middle Copper Age, Late Copper Age, and Early
Bronze Age. Both macroscopic and petrographic data suggest relatively homogeneity in
production methods, firing technology, and the addition of temper over time. Variability does
exist, however. Most notably, the addition of grog became more frequent between the Middle
Copper Age and Middle Bronze Age, and subtle but measurable change in paste and body ratios
took place during the same time period. Importantly, this change is measured over a long period
of time, rather than as a disjunction between two distinct cultural phases. In fact, the first
notably grog tempered ceramics occurred after the Middle Copper Age, and the general trend of
the approximately 2,000 year period of the Middle and Late Copper Age and Early and Middle
197
Bronze Age is a subtle, but steady increase in the frequency of grog temper. Thus, changes in
body composition occurred gradually and in a way not indicative of an abrupt shift in production
technology. Such a change is not the archaeological signature expected in the case of a
migration or invasion scenario, where the more likely pattern is one of change. Indeed, even
ceramics from the Middle Copper Age and Early Bronze Age exhibit measured, rather than
abrupt changes in body and paste composition at a time of drastic change in ceramic form and
decoration (Jankovich et al. 1989; Jankovich et al. 1998).
Similar to the settlement pattern data, the ceramic data do not point to a significant
cultural impact caused by a migratory population entering the Great Hungarian Plain at this time.
In fact, a recent elemental study of white encrustation on ceramics from the Hungarian Plain by
Parkinson et al. (2010) determined that bone paste was applied to incised decoration on Copper
Age and Bronze Age ceramics. The authors applied the results to the question of material
culture impact at the time of the appearance of kurgans, determining that the influence of the
“Kurgan invasion” did not affect subtle aspects of ceramic decorative tradition (2010:69). As a
decorative technique highly susceptible to change in the case of cultural contact through
migration or invasion, such a characteristic persisting over a long period of time contributes to an
argument of population continuity and little or no indigenous change resulting from the arrival of
a population of kurgan builders.
Despite the ceramic data presented in this dissertatation and in Parkinson et al.’s (2010)
pilot study, it remains impossible to rule out an immigration event based on the archaeological
data, as limited as they may be, and the settlement data relating to kurgan burial tumuli.
Stratigraphic data indicate that at least some of the kurgans were constructed at around the time
of the Middle/Late Copper Age transition (Escedy 1979), and although the relationship between
kurgans and settlements remains difficult to interpret without greater chronological control, the
complementary distribution of kurgans and Late Copper Age sites is compelling in regards to a
migration scenario. As discussed above, migratory populations can be and have been
incorporated into local populations relatively quickly and without a significant signature in the
archaeological record (Thompson 1996). The conclusions presented here largely support
Sherratt’s (1997a, 1997b) contention that a population of Yamnaya kurgan builders co-existed
with the indigenous Late Copper Age (and possibly Middle Copper Age) population in the
region, and that the populations tended to place settlements and burial monuments in separate
198
locations across the landscape. This may have stemmed from a desire for areas of higher or
lower relief, access to fertile agricultural soils, a need for soils more suited for pastoralism, or
perhaps just a general strategy of avoidance. This is in contrast to the scenario of invasion,
subjugation, and replacement in indigenous Hungarian Plain groups by conquerors from the east
proposed by Gimbutas (1970).
A Revised and Expanded Model of Late Copper Age Settlement and Economy
The set of settlement and material culture changes experienced on the Great Hungarian
Plain during the Late Copper Age was multi-faceted and more complicated than most previous
models have indicated. Although the period was characterized by the Plain’s incorporation into
the wider, materially homogeneous Baden material culture horizon, multiple internal and
external factors were in play that made this time period exceptional. Sochacki (1990:99-101) has
previously discussed the complexity of Baden at the scale of eastern Europe, and the regional
range of diversity in economic strategies and burial customs. Furholt (2008) also discussed
variability of social and cultural practices between regional populations of the Baden material
culture group in his analysis of stylistic variability in pottery. The results presented as part of
this research project also point toward modest compositional variability between pottery of the
Körös and Maros regions (though a more complete comparison of a battery of stylistic attributes
would be useful for comparative purposes). The variability at multiple scales throughout the
Baden material culture area does not resemble a pattern one might expect in a destination region
with a large demic migration ushering in substantive material culture change. Indeed, the
presence of thousands of kurgans scattered across the Hungarian Plain is the only convincing
tangible evidence of migration from the east at this time. This renders a migration model in and
of itself insufficient to explain the changes on the Plain at the beginning of the Copper Age, and
of the development of the materially homogeneous Baden material culture as a whole. Rather, a
complex interaction of multiple factors contributed to changes at the local level in the Körös
region that in many ways mirrored changes throughout the Carpathian Basin and eastern Europe.
The changes observed in ceramic form and design late in the Copper Age may partially
reflect an attempt on the part of the Plain’s indigenous population to display their economic and
social identities in the face of an intrusive migration. Bourdieu (1984:281) noted the symbolic
value of ceramic display, and the adoption of the wider Baden material culture style may have
199
been a conscious or unconscious move on the part of the people of the Hungarian Plain to align
themselves economically and socially with their trading partners beyond the Carpathian
Mountains. Bowser (2000) discussed pottery decoration as a signaling process for the expression
of social identity, and such behavior has analogues during the Neolithic on the Great Hungarian
Plain when social boundaries were more rigorously maintained between groups in the region
despite general similarities in economy, subsistence, and settlement organization (see Parkinson
2006a; Chapter Three, this volume).
A population of migrants likely entered the Carpathian Basin from the east sometime
around 3,500 B.C., at or near the time of the Middle/Late Copper Age transition. Currently,
archaeologists lack the evidence necessary to estimate the size of this migration, or even the path
of a migration stream or multiple, smaller migrations. However, the population must have been
sizeable enough to create an initial signature of kurgan burial mounds across the landscape of the
Körös River watershed, though the duration of the initial occupation remains uncertain given the
absence of definitive stratigraphic, settlement, or radiometric data. The issue then becomes not
one of the existence of a migration, but the impact of such an event on the indigenous people of
the Great Hungarian Plain. Did the arrival of the kurgan builders significantly impact the
lifeways of the native population or substantially affect the social and cultural trajectories
already in play? The results of this study do not support such a scenario. The ceramic and
spatial data together support a model of long-term population continuity, characterized by an
extended pattern of population nucleation and dispersal closely related to social and economic
factors. This interpretation supports the conclusions of researchers such as Sherratt (1997a,
1997b) and Gyucha (2010), and the results of recent studies pointing toward long-term
population continuity in the region (Parkinson et al. 2010).
The contribution of the kurgan builders to the cultural tapestry of the Hungarian Plain
may not have been the introduction of specific cultural characteristics, but may have been the
construction of the kurgans themselves. Following their construction, the kurgans became
dominant landmarks on a landscape otherwise characterized by its flatness. Aside from large tell
settlements, larger kurgans would have been the most visible landscape features in the Körös
region, and even the smaller kurgans would have been noteworthy monuments on the Plain.
This implies that a relatively small migration of kurgan builders into the region late in the Middle
200
Figure 9.1. Monument to a modern Hungarian political movement placed atop a kurgan in Békés County. A
cartographic survey marker is visible in the background, at the apex of the mound. Even in the modern era, kurgans
remain important landmarks and prominent features on the Great Hungarian Plain.
Copper Age or early in the Late Copper Age would have had an impact on the visual landscape
of the Körös region. Previous models of Yamnaya migration onto the Hungarian Plain (see
Gimbuas 1977, 1979, 1980; Anthony 1990) have described migrations occurring in multiple
waves over an extended period of time, or at least one large migration into the region. However,
it is possible that a modestly sized migration event could have resulted in the archaeological
signature seen in the Körös region, if the tradition of kurgan building was adopted by local
populations and carried on into later periods. The social and cultural impact of this signature –
the kurgans – may have had a greater impact over time than in their initial appearance.
Kurgans and cultural context. Although kurgans have been excavated in the Körös
region and across the Great Hungarian Plain (Escedy 1979), it has not been firmly established
that Yamnaya immigrants from the east constructed all of the tumuli. The possibility exists that
people indigenous to the region who emulated the burials of Yamnaya migrants constructed
many kurgans on the Hungarian Plain later, or perhaps contemporaneously. Such a scenario
201
seems plausible, since not all kurgans contained Yamnaya materials (see Chapter Three; Escedy
1979). Subsequent inhabitants of the region reused kurgans as burial locations, and Scythians
and Sarmations constructed burial mounds as late as the Iron Age (Jankovich et al. 1989;
Jankovich et al. 1998). In the Middle Ages and later, churches and other buildings were
constructed atop kurgans much like monks between the 11th and 13th centuries placed the Csolt
Monastary atop the Vésztő-Mágor tell near the city of Vésztő in the Körös region. Even now,
kurgans retain place names that are hundreds of years old and serve as landmarks for farmers,
geographers, and even as monuments for modern political movements (Figure 9.1). Given their
significance after the Copper Age, the continuing construction and reuse of burial mounds later
in prehistory, emulating the earliest kurgans on the Hungarian Plain, may represent the long-term
cultural contribution of the Middle/Late Copper Age Yamnaya migrants into the region.
Higgenbotham (2000:7) noted the importance of not only the construction and use of
emulated aspects of material culture, but also the necessity of modification of borrowed features
in order to be legitimized by local populations. She stated that one of the best indicators of
cultural emulation is the modification or hybridization of features of the emulated object to
integrate them into the local cultural context. The placement of some kurgans near Late Copper
Age settlements (or vice-versa), as well as the presence of Middle and Late Copper Age ceramic
material in kurgan burials and construction layers, serve as indications of modification,
hybridization, and legitimization. Further archaeological research is necessary to bolster these
claims, though the questions of hybridization and emulation remain intriguing and important.
Archaeological and modern examples of emulation. Many examples of emulation,
modification, and hybridization can be found in the archaeological, ethnohistoric, and historical
records. The examples presented here are meant to briefly illustrate this phenomenon through
various lines of archaeological evidence at different times throughout history.
Pottery and other ceramic materials were often emulated in prehistory; or, as Dickinson
(1994:131) noted, vessels of wood and stone were often produced alongside pottery throughout
the Neolithic, though this practice subsided as ceramic technology improved. Even more, nonceramic materials of the Bronze Age Aegean were primarily produced for display, made of
attractive materials (usually bronze), were difficult to work, and were often decorated more
elaborately then their clay counterparts. These metal vessels were associated with status – the
majority of examples have been found in contexts such as palaces, large buildings, religious
202
sites, and elaborate graves. As such, it is not surprising that the emulation of metal vessel forms
in clay took place as far away as the Carpathian Basin. During the Late Copper Age, some
vessel handles emulated the handles of bronze and ceramic vessels from the Balkans (Kalicz
1963). Although the emulation of metal vessel forms did not necessarily accompany dramatic
change in society of economy in the Carpathian Basin at this time, it does illustrate how the
introduction of foreign ideas can have lasting effects on the material culture of a region, despite
little direct interaction with foreign populations.
From a much different perspective, emulation has taken place in contact and colonization
scenarios, where subjugated populations incorporated certain characteristics of material culture
from the dominant population. For example, Colono ware is low-fired, unglazed earthenware
produced from the east coast of North America and the Caribbean in the 18th and 19th centuries.
As one of the best examples of emulation and hybridization found in the archaeological record,
Colono ware represents a combination of features of European refined earthenwares, Native
American course earthenwares, and traditions in West African pot-making brought to North
America by African slaves (see Noël Hume 1962). Quite often, Colono ware emulated form and
function of European finewares, while utilizing the less sophisticated production technology
available to African slaves. Especially common was the emulation of specific design features
such as rim shape and footrings, in addition to overall form.
Colono wares are particularly common in archaeological contexts in Spanish colonial
regions of North America, including the Caribbean and parts of Florida. Such pottery frequently
reproduced Spanish forms of refined earthenwares, but sometimes retained local forms,
manufacturing techniques, and materials. Local aboriginal groups as well as Africans produced
Colono wares, and they incorporated a wide variety of local ceramic making traditions into the
largely homogeneous Spanish colonial material culture (Deagan 1987; see also Noël Hume 1962;
Fairbanks 1962; Ferguson 1978). Deagan (1987:104) stated that the incorporation of local New
World ceramic characteristics and the manipulation of local materials to accommodate Hispanic
preferences was an intentional adaptive strategy on the part of aboriginal and African
populations.
Although the emulation of ceramic form and design can only be applied broadly to the
emulation of kurgans on the Hungarian Plain, examples depicting the emulation of burial
mounds and monumental burial architecture also exist in the archaeological record. For
203
example, Williams (2006) described changes in existing mortuary traditions that occurred over a
very short time period at Sutton Hoo in medieval England. Variability is observed over time in
terms of kind and number of grave goods. Despite this variability, however, large earthen
mounds were continuously constructed atop graves throughout the early medieval period. The
mound burials were not reused, and appear to have been singular events commemorating the
death and remembrance of individuals rather than groups or lineages (Williams 2006:159-162).
Interestingly, prehistoric barrow cemeteries exist in the region surrounding Sutton Hoo (Carver
1998), and Williams (2006:161) suggested that the medieval mounds at the site may have evoked
associations with prehistoric barrow cemeteries in the region rather than, or in addition to,
association with other medieval mounds at and around the site. In addition to its cemeteries of
burial mounds, Sutton Hoo is well known for its rich assortment of grave goods and its famous
ship burial, and thus its description as a “burial ground of kings” (Carver 1998). Although in
many ways exceptional in comparison to the kurgan tumuli of the Great Hungarian Plain at the
end of the Copper Age, the mounds at Sutton Hoo may illustrate a similar phenomenon. As on
the Hungarian Plain, barrows and barrow cemeteries built prehistorically remained dominant
features on the landscape for many centuries, into the early medieval period in Britain. The
mounds at Sutton Hoo may have referenced the prehistoric barrows as powerful or important
places on the landscape, and the individuals interred beneath the mounds may have sought an
association with the ancestors who constructed other important and dominant monuments on the
landscape.
Summary. Ultimately, there is not sufficient evidence to state that kurgan builders
dramatically affected the established social structure and cultural trajectory of the Hungarian
Plain prior to their arrival. The results of this study suggest a period of long-term population and
economic continuity stretching from the Neolithic to the Early Bronze Age. Internal economic
and social trajectories led to a decrease in occupation of the central Plain by the Late Copper
Age, tighter integration with economies outside of the Carpathian Basin, and incorporation into
the materially homogeneous Baden material culture horizon.
However, such a model does not mean that even a modest migration of kurgan builders
onto the Plain at around 3,500 B.C. did not have a lasting impact. The initial constructions of
kurgans in the region may have resulted in the continued emulation and construction of kurgans
well after the initial Yamnaya migrants were absorbed into the local population. One premise
204
regarding migration implied by the model described above – that even small emigrations of
people can have a dramatic impact on local culture through emulation over the long-term – is
supported by other archaeological evidence and by the results of the archaeological investigation
presented in this dissertation. Unfortunately, until we better understand how kurgans developed
across the landscape, and until we develop a detailed chronological sequence for their
appearance, the true impact of their construction on the Great Hungarian Plain will remain
shrouded in uncertainty.
Long-term Population and Economic Continuity on the Great Hungarian Plain
The results of this research project support a model of settlement and material culture
change based on a foundation of long-term population continuity. Based upon the settlement and
ceramic evidence presented in Chapters Seven and Eight, the model described in this chapter
resembles and expands upon the settlement, economic, and environmental model developed by
Sherratt (1997a, 1997b) as part of his work in on the Dévaványa Plain area of Békés County.
The overall pattern in the Körös region and of the central Plain moves away from settlement in
the center of the Plain, and shows an increase in site number and density on the edges of the
Plain. This intensification represents a movement closer to raw materials, finished goods, and
access to trade and exchange routes. Punctuated by internal cycles of population nucleation and
dispersal, this pattern persisted throughout the Neolithic, Copper Age, and into the Early and
Middle Bronze Age.
As discussed in Chapter Three, the developmental trajectory of the Great Hungarian Plain
does not fit easily into traditional models of social development. Sherratt (1997a, 1997b) most
directly approached the long-term movement of settlements away from the center of the Plain
and, essentially, out of the Körös region. This is especially true during the Middle Copper Age,
when fewer Bodrogkeresztúr settlements are known than in the previous Early Copper Age
Tiszapolgár phase. The settlement data presented as part of this research mirror Sherratt’s
assessment of his study region on the Dévaványa Plain at the scale of the entire Körös region.
This continued dispersal from the central tell sites of the Late Neolithic resulted both from a
cyclic social leveling mechanism (e.g. voting with one’s feet; see Gyucha et al. 2004; Parkinson
2002, 2006b), and from the wider pull of large-scale trade and exchange systems operating
primarily outside of the Plain. During the Late Neolithic, Early Copper Age, and Middle Copper
205
Age, Parkinson (2002) noted a correlation in material culture differentiation between groups in
the Körös region, and degree of nucleation or dispersal (termed “integration”). In other words, in
periods characterized by nucleation, material culture expressed more regional differentiation.
This is clear during periods such as the Late Neolithic and Early Bronze Age in the Körös River
watershed, where multiple material culture groups occupy relatively distinct geographic areas.
The Late Copper Age Baden material culture phase adheres to this pattern. Previous assessments
of stylistic variability (see Escedy et al. 1982; Jankovich et al. 1989; Jankovich 1998) and the
macroscopic and petrographic analyses of visible features of ceramic manufacturing technology
presented here support a model of broadly homogeneous ceramic material within the Körös
region and between the Körös and Maros regions. In this period of settlement dispersal, almost
no material culture differentiation is observed.
On the other hand, for the first time since the Neolithic, the Körös region and the Great
Hungarian Plain became incorporated into a homogeneous material culture horizon that
developed and existed well beyond the Carpathian Basin. At around 3,500 B.C., settlement
focused on the margins of the Plain (Banner 1956; Roman and Németi 1978; Sherratt 1997a,
1997b), and the pattern of exchange with populations outside of the Plain witnessed in the Early
and Middle Copper Age continued to intensify (Gyucha 2010; Parkinson 2006b). The resourcerich intermountain valleys that previously supported only modest populations became the focus
of Baden settlement in the region (Sherratt 1997a:291). Although the overall pattern during the
Late Copper Age was one of dispersal from the Körös region and the central Hungarian Plain, it
is incorrect to suggest that these areas became a cultural backwater. In fact, the archaeological
evidence points to the contrary. Late Copper Age ceramic assemblages from both surface and
excavated contexts across the central Plain are clearly related to the wider Baden cultural horizon
based on vessel form and decoration (Jankovich et al. 1989; Jankovich 1998; Megyesi 1983).
Furholt (2008) demonstrated that regional variability does exist amongst various Late Copper
Age Baden assemblages, but this variability pales in comparison to the regional variability
observed during the Late Neolithic and Early Bronze Age. Even more, as previously discussed,
regional Baden horizons maintained a variety of local practices (burial, house design) in spite of
the generally homogeneous ceramic characteristics (Sochacki 1985).
The retention of some local and regional cultural characteristics and the adoption of
visible, easily transferrable characteristics like ceramic form and design do not point to a sudden
206
invasion or migration, even throughout the entire Baden region. Rather, the pattern suggests
population continuity, with the adoption of certain cultural characteristics to indicate and
legitimize participation in a wider economic system. Examples of the adoption and/or emulation
of certain highly visible elements of material culture exist throughout the prehistoric
archaeological record around the world. The Southeast Ceremonial Complex of the Mississippi
Period (ca. 800-1500 C.E.) in North America is perhaps one of the geographically largest
examples of this phenomenon. Although local populations retained some elements of material
culture and cultural practices, specific highly visible characteristics were adopted that emulated
the material culture and monumental architecture of larger Mississippi period settlements. Some
examples of these characteristics include the construction of large platform mounds at major
centers, repousse copper breastplates, and the use of shell temper in most Mississippi ceramics
(Bense 1994; Pauketat 1994). In this way, authority was maintained by elites, and participation
in a much wider cultural system was indicated and legitimized.
The incorporation of the Hungarian Plain’s inhabitants into the wider Baden material
culture and economic system significantly impacted the internal cycle of nucleation and dispersal
– the leveling mechanism that held in check social momentum toward the development of
institutionalized hierarchy. While levels of social complexity beyond egalitarianism did not
develop fully at the end of the Neolithic, and the population subsequently dispersed across the
landscape, the involvement of the Plain’s people in a wider economic system allowed for the
procurement and utilization of prestige items – especially those made from bronze – near the end
of the Late Copper Age and beginning of the Early Bronze Age. This is especially true of the
central Plain, where both raw and finished materials were scarce and likely held a special
significance. Corellates for the significance of rare or important objects exist in Middle Copper
Age Bodrogkeresztúr burials, both intramurally and in cemeteries such as TiszapolgárBasatanya, where copper axe-heads were occasionally interred with male bodies (BognárKutzián 1963). Unfortunately, not enough burial data exists in the Körös region, or on the
eastern Hungarian Plain in general, to definitely observe similar behavior in Late Copper Age
contexts.
The general pattern established by the Late Copper Age in the study region continued for
the next millennium. This lends support to a scenario by which the material culture
homogenization observed during the Baden horizon had economic and social, rather than
207
migratory, origins. Although settlement nucleation and a return to a tell-based settlement system
(as well as the appearance of institutionalized hierarchy and craft specialization) occurred on the
Plain by the Middle Bronze Age, a focus on settlement near the edges of the Plain continued for
the next 1,000 years. Bóna (1975), O’Shea (1978), and Sherratt (1997b) noted the role of trade
in explaining intensified settlement near major rivers on the edges of the Hungarian Plain during
the Late Copper Age and Early Bronze Age. Due to their potential for linking long-distance
trading partners and serving as established transportation routes, the Tisza and Maros Rivers took
on important roles at this time (Sherratt 1997a:291). O’Shea noted imported items such as
copper ornaments and shell beads in the Bronze Age graves of the Maros Region (1978, 1996),
indicating that the procurement of finished foreign materials increased in importance and in
volume over this time period.
Ultimately, incorporation into a wider economic system led to the establishment and
ubiquity of the Late Copper Age Baden material culture horizon in the Körös River watershed
and across the Great Hungarian Plain. The migratory arrival of kurgan-building invaders, or a
fully realized, internal cultural trajectory, are both insufficient explanations for the changes seen
at this point in the region’s prehistory. The more frequent appearance of metal objects in burial
contexts, the increasing focus of settlement on the resource-rich margins of the Plain, and the
increasing importance of metal and prestige goods at this time have been well documented.
Many archaeologists have noted the fundamental shifts in social interaction brought about by the
mining, trading, and production of bronze during the Early and Middle Bronze Age, from some
of the earliest studies of the period to more recent investigations (Childe 1930; Pare 2000;
Raczky et al. 1995; Shennan 1986; Sherratt 1993). Although the use of prestige objects to
express status in the social hierarchy had not been fully realized as early as the Middle and Late
Copper Age on the Great Hungarian Plain, the economic engine that funneled such goods into
the region had become well established.
By the Late Copper Age, the Great Hungarian Plain was incorporated into a pan-central
and -eastern European exchange system. The Baden material culture complex, most especially
an extremely similar set of ceramic assemblages, was ubiquitous across the region, though a
homogeneous ethnic, religious, or otherwise culturally related population did not exist at this
time. Baden should not be conceptualized as a homogeneous “culture.” Rather, Baden existed
as a cultural complex linked by a wide-ranging economic system and extremely similar
208
assemblages of ceramic materials. This model of Baden exchange and interaction concurs with
Furholt’s (2008) recent assessment of heterogeneity in Boleráz and Baden materials on a
continental scale, while simultaneously accounting for the general homogeneity of Late Copper
Age ceramic material on the Plain, and in the Körös River watershed study region.
The incorporation of people on the Hungarian Plain into a wider, more materially diverse
exchange system during the Late Copper Age fundamentally affected the internal cycle of
nucleation and dispersal, and the social leveling mechanism by which a relatively egalitarian
social structure was maintained for thousands of years. The arrival and production of bronze
prestige goods, coinciding with the beginning of a period of nucleation during the Early Bronze
Age, contributed to nucleation during the Middle Bronze Age. This period experienced a
renewed emphasis on tell settlements and the first appearance of craft specialization and
institutionalized hierarchy on the Great Hungarian Plain. Far from resulting from the arrival of
an invasive and dominating migratory population, the social structure consisting of population
nucleation and dispersal, in combination with stronger economic and social ties with economic
partners outside of the Plain, led to elements of societal organization never previously observed
in the region. The changes documented at the end of the Copper Age did not result from a fully
realized internal social trajectory. In fact, the internal trajectory likely delayed the initial
development of specialization and institutionalized hierarchy in the region during the Late
Neolithic/Early Copper Age transition (Parkinson 2002). So, the Late Copper Age Baden culture
on the Great Hungarian Plain is not an anomalous phenomenon, but is instead a vital link, both in
terms of economy and settlement, between the largely egalitarian social structure of earlier
periods and the hierarchical social structures that appear later during the Bronze Age.
Implications for Anthropological Models of Homogeneous Material Cultures
One purpose of this research project – to link the specific regional archaeological case
study of the Baden Culture with wider anthropological models explaining the development of
homogeneous material cultures – has framed the discussion of migration models, interaction
spheres, and economic interaction as they pertain to the Late Copper Age on the Great Hungarian
Plain. Each of the models for homogeneous material culture change discussed in Chapter Two
are directly or indirectly tested in this dissertation through the application of the results of the
209
settlement and ceramic analyses presented in Chapters Seven and Eight to the archaeological and
social models of Marija Gimbutas and Andrew Sherratt (see Chapter Two).
In terms of migration as an explanatory force for the appearance of regionally
homogeneous material cultures, archaeological, historical, and ethnohistorical examples exist
that show the role that migration can play in such scenarios (see Chapter Two). Many
archaeologists, from the earliest days of the discipline to the present, have alternately
emphasized or questioned the role of migration in shaping cultural history. Less frequently
discussed, however, is how relatively small immigrations of migratory people can fundamentally
alter the physical signature of people occupying a region over the long-term. As suggested
above, a modest arrival of kurgan-building migrants in the Great Hungarian Plain can account
for the thousands of tumuli across the landscape, not only through primary construction but also
through emulation and reuse of the mounds by indigenous people. The ephemeral signature of
those who initially constructed kurgans in the region is explained by a rapid acculturation and an
integration of tumulus use by the subsequent culture groups on the Plain.
Although the Late Copper Age Baden material culture does not correspond to the precise
use of Caldwell’s (1966) concept of an interaction sphere as initially developed, the appearance
and ubiquity of Baden ceramic material on the Plain was due primarily to its incorporation into a
wider economic interaction sphere. Caldwell did state, however, that interaction between
sociocultural and socio political groups is amenable to and likely formative in the development
of elite institutions and an overarching institutionalized hierarchy amongst participating groups
(1944:141). When accounting for the influence of economy in the interaction sphere model, one
can argue that the economic and information exchange networks, even amongst culture groups
generally considered egalitarian and without ruling elites, can contribute to the development of
institutionalized hierarchy depending on the material and cultural value of the items moving
through the network. In the case of the Great Hungarian Plain, the economic ties developed and
solidified by the Late Copper Age allowed the frequent acquisition and use of bronze and metal
objects. As such, the regional material culture homogeneity seen on the Great Hungarian Plain
should be viewed as an indicator of economic integration, and not as evidence of change
catalyzed by migration or invasion.
210
Summary
The results of this research support Sherratt’s model of economic integration and its
consequences for settlement and material culture in the Körös region of the Plain. However,
migration cannot, and should not, be eliminated as a contributing factor to the local
archaeological signature over the long-term, especially when dealing with kurgans, their
construction, and their subsequent use and reuse by populations indigenous to the study region.
This is especially important to consider given the highly visible consequences of even relatively
small migrations, such as the emulation of ceramic forms or, in this case, burial tumuli.
The long-term prehistoric trajectory of settlement on the Great Hungarian Plain consisted
of a regional trend toward economic and social interaction, supported by a local foundation of
population and settlement nucleation and dispersal cycles. This research supports a model of the
Late Copper Age on the Great Hungarian Plain not as an anomalous end to the social and
settlement pattern visible in the Late Neolithic and Early and Middle Copper Age, but rather as
an economically and culturally integrated part of the region’s prehistory that set the stage for
social structures in the Bronze Age previously unseen on the Plain.
The material culture homogenization on the Hungarian Plain during the Late Copper Age
marked by pottery characteristic of the Boleráz and Baden material culture groups can be best
explained by a model combining increased integration into an outside economic system and a
social trajectory internal to the Plain. In terms of wider anthropological models, this research has
demonstrated that material culture homogenization can be closely related to economy and
exchange, and migration, invasion, or elimination scenarios are not the most likely or even most
convenient explanations.
211
CHAPTER TEN
CONCLUSION AND FUTURE RESEARCH DIRECTIONS
Conclusions
In closing, this research provides a glimpse of how regionally homogeneous material
cultures can develop across a landscape over time, and of how such material culture horizons
play out on the local level in terms of economy and settlement. It is hoped that long term social
and economic trajectories, often spanning hundreds or thousands of years, will begin to gain
more consideration for modeling material culture and settlement change, along with the more
traditional explanatory mechanisms of migration and invasion. A focus on identifying continuity
in the more subtle aspects of the archaeological record is indeed emerging, and though such
studies often present more practical and theoretical challenges than the more accessible study of
discontinuity and sudden change, they can often be just as, if not more, effective in developing
and testing models of change. Recently, archaeologists working on the Great Hungarian Plain
have greatly redefined our understanding of the region’s prehistory, especially in terms of
modeling transitions between cultural phases possessing different settlement pattern and material
culture characteristics. The results presented as part of this research project simply add to this
ongoing effort, and contribute to the ever growing archaeological and anthropological knowledge
of the region’s prehistory.
The research described in this volume contributes to the wider body of archaeological
and anthropological methods and theory in several ways. First and foremost, the results
presented here contribute to our understanding of how regionally homogeneous material cultures
develop across a landscape over time. Multiple trajectories can contribute to an artifact type’s
ubiquity over a region or continent. As discussed in Chapter Two, human behaviors including
migration, invasion, interaction spheres, and economic trends can contribute to regional
homogeneity. Less often, however, a combination of multiple behaviors is used to explain the
phenomenon. This is in spite of the likelihood that multiple factors can contribute to material
culture spread, adoption, and ubiquity. This dissertation emphasizes the role in material culture
homogenization played by both internal social trajectories and forces that involve interaction
with outside populations, such as the development and intensification of economic ties and the
duel dynamic of migration and integration. Such an approach can help explain not only how a
particular aspect of material culture became extremely common over a wide area, but it can also
212
explain how and why other aspects of culture and societal structure, such as settlement patterns,
changed as they did.
Given the approach taken in this project, the research secondly contributes to modeling
social, settlement, and material culture change on the Great Hungarian Plain at the end of the
Copper Age. It speaks directly to the nature of the region’s incorporation into the wider Baden
material culture complex by illustrating the relative compositional, technological, and
manufacturing homogeneity of ceramic materials from the Middle Copper Age, Late Copper
Age, and Early Bronze Age in the Körös region of the Plain. Variability among the ceramics of
different periods, when it does exist, tends to occur steadily and slowly, rather than suddenly.
This suggests that the culturally embedded framework and methodology for creating pots
changed little over the approximately two millennia covered by this study, and indicates
population continuity rather than a large migration of new people into the region. Additionally,
the results contribute to the overall understanding of the social trajectory at play between the
Neolithic and the Bronze Age on the Great Hungarian Plain. The model presented here suggests
that the internal process of population nucleation and dispersal actually acted in concert with an
intensifying settlement emphasis on the margins of the Plain as access to raw and finished
materials, especially bronze goods, became more important at the end of the Copper Age and
beginning of the Bronze Age. Ultimately, these two combined processes contributed to the
development of institutionalized hierarchy and craft specialization on the Hungarian Plain during
the nucleation phase of the Middle Bronze Age. In this context, the Late Copper Age was
actually a pivitol period of time for the development of social complexity on the Plain.
Third, from a methodological standpoint this research contributes to the understanding
that multiple lines of evidence are required for the testing and refining of specific archaeological
and anthropological models. Furthermore, this study has illustrated the importance of measuring
local and regional patterns at multiple analytical scales since variables such as site distribution
and density can dramatically differ from one portion of a study area to the next. And, the fact
that patterns can appear quite different on local, regional, and continental scales demonstrates the
value in observing patterns at multiple resolutions. For the present research, the development of
the model described in Chapter Nine describing the behaviors that led to the study region’s
incorporation into the Baden material culture horizon and economic system, the changing
settlement pattern of the Copper Age, the integration of a migratory population into indigenous
213
society, and the social trajectory culminating with the nucleation of the Middle Bronze Age
would not have been possible without multi-scalar spatial and ceramic analyses.
While an expanded and updated archaeological site spatial analysis or macroscopic and
petrographic ceramic analyses would have been interesting of their own merits, the use of just
one line of evidence would have been insufficient for approaching the set of changes at the end
of the Copper Age on the Hungarian Plain. Furthermore, a multi-scalar spatial study (e.g. site
level, sub-region level, and regional level) allows for a controlled analysis of variability in
material culture that is not possible at one scale along. For example, this study determined that
Late Copper Age ceramics remained similar in terms of design and production throughout the
region and between regions, though slight inter-site and inter-regional variability did, in fact
exist. While this variability is not significant enough to suggest that an appearance of new
manufacturing techniques influenced one particular group over another, it does suggest that
regional variation in production and manufacture were maintained despite the Great Hungarian
Plain’s incorporation into the Baden horizon, and as such it dovetails with Furholt’s (2008)
results that detailed local stylistic variability in Boleráz and Baden materials over a wide area.
Finally, the data and interpretations presented here can be made available to a wider
archaeological and anthropological audience. Although the tested case studies are specific to the
prehistoric Great Hungarian Plain, the overarching anthropological issues involved – including
the development of regionally homogeneous material culture groups, and how local populations
are incorporated into wider economic structures – are relevant in multiple time periods in many
areas across the globe. Even more, the methodologies employed in both the settlement analysis
and ceramic analyses may be useful for other archaeologists interested in the same types of
questions in other regions and during other time periods. As such, this volume provides data
useful for broader comparison, not just on the Great Hungarian Plain but in other regions where
similar phenomena are being studied.
Future Research
Although the results of this research project and the model developed from them stand on
their own as a cohesive piece of work, numerous possibilities exist for the expansion of the
research presented here and for future research directions. First and foremost, it is possible to
expand upon the work presented here by incorporating petrographic data from different Baden
214
regions within and beyond Hungary in order to establish if local production traditions maintained
themselves elsewhere. Although the material on the Great Hungarian Plain indicates population
continuity over the long-term, this may not be true in other regions incorporated into the Baden
material culture horizon, and other social processes may have been at work in the appearance of
Baden elsewhere. Similarly, a multi-scalar study of settlement location and settlement change
over time may reveal different processes at work before and during the Late Copper Age in
different places.
Second, further systematic excavation of intact Late Copper Age sites in the Körös region
is necessary in order to firmly establish a ceramic chronology and to determine the relationship
between Boleráz and Baden materials in the region. A major hindrance of the present research –
a lack of chronological control between ceramic samples described variously as Boleráz and
Baden – prevented a high-resolution comparative macroscopic and petrographic analysis of these
phases of the Late Copper Age. Furthermore, excavation of Late Copper Age settlement sites in
the region would bring to light other elements of material culture – such as house structure,
settlement organization, burial traditions, and so on – that are largely lacking in the Körös region
for the time period. Currently, the site of Hódmezővásárhely-Kopáncs I., Olasz-tanya in the
Maros watershed to the southwest is under investigation, and it should produce a wealth of useful
data of this type of analysis and interpretation. The excavation of such a site in the Körös
watershed will provide an opportunity to continue and expand upon the comparative research
presented in this volume. During the process of site collection in the fall of 2009, the site of
Tarhos 67 was identified as a candidate for further testing and possible investigation. In the
coming months and years, I hope to conduct an in-depth investigation of this site (and ideally
other sites in southeastern Hungary) in order to compare the results with other geographic
regions and countries.
The biggest question remaining unanswered at the end of this research relates to the
kurgans and how they came to be across the landscape of the Great Hungarian Plain. Given the
approach of the present research, and the resources necessary to directly investigate the kurgans,
a more in-depth study was beyond the scope of this dissertation. Unfortunately, without further
research directly addressing the question of the kurgans their exact nature will remain uncertain.
However, it may be possible to understand their relationship to the sites and people of the Late
Copper Age on the Hungarian Plain by employing a number of different research strategies. A
215
direct study of a large number of kurgans will be a lengthy and expensive endeavor. But, by
sampling a large number of kurgans across the landscape of the Hungarian Plain, it may be
possible to establish and refine a chronology for their appearance and spread. First and foremost,
the direct radiometric dating of skeletal material excavated from the primary burials of kurgans
would be useful. Isotopic studies (especially strontium) of skeletal material could shed light on
the provenance of the interred individuals, thereby directly approaching the question of
migration. Nitrogen and oxygen isotopic study could shed light on the unanswered question of
pastoralism on the Plain during the Late Copper Age. Accelerator mass spectrometry dating may
be useful in conjunction with the collection of a large number of core samples from kurgans;
however, this line of research would only prove useful if datable material was retrieved in the
cores. Ultimately, a research project consisting of a battery of lines of evidence will be
necessary for fully exploring and perhaps answering the question of the kurgans.
The results of further research into the nature of social and settlement changes on the
Great Hungarian Plain during the Late Copper Age will shed light not only on the region and
time period under study. Like the research presented in this dissertation, it has the potential to
speak to numerous other archaeological and anthropological issues. Although this dissertation is
but a small contribution to the excellent body of scholarship available on Eastern European and
Hungarian prehistory, I hope to make further contributions by continuing the research with
similar archaeological and anthropological goals in mind. And, though this study has addressed
a specific set of research questions and provided interpretation and discussion of them, future
research will expand upon these questions and continue to examine broader anthropological
phenomena with the support of archaeological lines of evidence.
216
APPENDIX A
SITE COLLECTION SUMMARIES
217
Site: Békés 26
Transects Walked: 12
Average Visibility: 80%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
0
11
0
0
0
Diagnostic Sherds
Rims
Bases
3
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
0
157
0
0
0
Dec Body
Lugs/Handles
0
0
0
0
0
0
0
8
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N) BodyWt (g)
Other
0
Total
0
0
1
0
5
0
3
2
Pre Diag
Later
SF1
1
9
1
SF2
1
52
1
SF3
1
12
1
SF4
1
17
1
SF5
1
7
1
SF6
1
10
1
SF7
1
4
1
SF8
1
6
1
SF9
1
11
1
SF10
1
7
1
SF11
1
22
1
218
Mod
Daub
Bone
Other
Site: Békés 178
Transects Walked: 6
Average Visibility: 40%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
0
14
0
0
0
Diagnostic Sherds
Rims
0
233
0
0
0
Bases
1
Dec Body
Lugs/Handles
1
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
3
0
0
0
0
0
11
Other
1
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Total
0
14
1
0
4
0
4
2
Distribution of Material (Rough Sort)
Unit
Body (N)
BodyWt (g)
Pre Diag
Later
SF1
3
29
3
SF2
1
7
1
SF3
1
23
1
SF4
1
8
1
SF5
2
29
2
SF6
1
18
1
SF7
1
17
1
SF8
2
28
2
SF9
1
68
1
SF10
1
6
1
219
Mod
Daub
Bone
Other
Site: Belmegyer 82
Area Intensively Collected: 942 m2
Average Visibility: 30
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
64
137
37
5
0
Diagnostic Sherds
Rims
1700
3050
196
20
Bases
6
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Dec Body
Other
2
0
2
1
0
1
0
0
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
0
0
4
3
0
0
Distribution of Material (Rough
Sort)
Unit Body (N) BodyWt (g)
C
Lugs/Handles
1
Pre Diag
Later
Mod
Total
2
Daub
11
Bone
Other
13
150
0
0
0
11
9
1
S30
1
<100
0
0
2
0
5
1
W30
6
100
0
0
3
6
6
0
E30
1
<100
0
0
1
2
1
3
N30
6
100
0
0
0
3
0
0
N60
36
600
4
0
0
16
7
0
N90
8
100
0
0
0
5
1
0
XU1
18
600
0
0
0
8
0
0
XU2
6
100
0
0
1
0
0
0
XU3
24
650
3
1
0
8
3
0
XU4
17
650
4
0
0
3
3
0
XU5
1
<100
0
0
0
2
2
0
220
Site: Biharugra33
Area Intensively Collected: 453 m2
Average Visibility: 92.5%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
80
229
0
2
0
Diagnostic Sherds
Rims
16
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Bases
2250
6400
0
2
0
Dec Body
Lugs/Handles
1
8
0
0
0
0
0
22
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough
Sort)
Unit
Body (N)
BodyWt (g)
Pre Diag
Other
2
Total
2
43
0
3
23
0
2
7
Later
Mod
Daub
Bone
Other
C
90
2650
22
0
0
5
0
1
S30
46
2000
7
0
0
6
0
0
S60
36
800
3
0
0
57
0
0
N30
35
800
4
0
0
2
0
1
N60
5
<100
1
0
0
5
0
0
W30
11
150
2
0
0
3
0
0
E15
6
<100
1
0
0
2
0
0
SF1
0
0
2
0
0
0
0
0
221
Site: Bucsa 13
Transects Walked: 6
Average Visibility: 70%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
0
3
0
0
0
Diagnostic Sherds
Rims
Bases
0
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
0
57
0
0
0
Dec Body
Lugs/Handles
0
0
0
0
0
0
0
3
0
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough
Sort)
Unit Body (N)
BodyWt (g)
Pre Diag
1
27
1
SF2
1
14
1
SF3
1
16
1
Total
0
3
0
0
2
0
0
1
Later
SF1
Other
222
Mod
Daub
Bone
Other
Site: Fuzesgyarmat 97
Transects Walked: 8
Average Visibility: 85%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
0
7
0
0
0
Diagnostic Sherds
Rims
Bases
3
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
0
94
0
0
0
Dec Body
Lugs/Handles
0
0
0
0
0
0
0
4
Other
Total
0
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
0
0
0
5
0
0
2
Distribution of Material (Rough Sort)
Unit
Body
(N)
Pre
Dia
g
BodyWt (g)
Late
r
Mo
d
Dau
b
Bone
Other
SF1
1
6
1
0
0
0
0
0
SF2
1
19
1
0
0
0
0
0
SF3
1
7
1
0
0
0
0
0
SF4
1
3
1
0
0
0
0
0
SF5
1
26
1
0
0
0
0
0
SF6
1
9
0
1
0
0
0
0
SF7
1
24
0
1
0
0
0
0
223
7
Gerla 64
Area Intensively Collected: 549.5 m2
Average Visibility: 28.6%
‘
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
128
127
29
0
0
Diagnostic Sherds
Rims
10
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Bases
2600
5150
77
0
0
Dec Body
Lugs/Handles
3
2
0
0
7
0
0
41
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N) BodyWt (g)
Other
Total
9
0
63
0
0
0
8
1
0
Pre Diag
Later
Mod
Daub
Bone
Other
C
50
2200
11
0
0
90
13
0
W15
16
600
3
0
0
11
7
0
W30
1
100
0
0
0
6
1
0
E15
16
800
1
0
0
4
3
0
E30
0
0
0
0
0
0
0
0
S15
9
700
1
0
0
1
0
0
N15
17
750
2
0
0
16
5
0
224
Site: Korosladany21
Area Intensively Collected: n/a
Average Visibility: 85
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
0
2
0
1
0
Diagnostic Sherds
Rims
0
10
0
1
0
Bases
1
Periods Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Dec Body/Lugs
0
0
0
0
0
0
0
0
1
0
5
Total
0
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N) BodyWt (g)
SF1
1
5
SF2
SF3
Other
1
Pre Diag
2
0
0
2
0
0
0
1
Later
0
0
Daub
0
0
1
0
0
0
0
0
0
225
Mod
Bone
0
Other
0
0
0
1
0
Site: Mezobereny34
Area Intensively Collected: 453 m2
Average Visibility: 76.6%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
8
270
5
0
0
Diagnostic Sherds
Rims
Bases
5
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
200
3300
<100
0
0
Dec Body
Lugs/Handles
2
4
2
0
0
0
0
5
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N) BodyWt (g)
Other
1
Total
0
13
0
0
2
0
0
5
Pre Diag
Later
Mod
Daub
Bone
Other
C
74
1000
2
0
0
4
5
0
N30
41
100
2
0
0
2
0
0
N60
1
<100
0
0
0
0
0
0
S30
57
1200
3
0
0
1
0
0
W30
90
1000
5
1
1
1
0
0
E30
7
<100
0
0
0
0
0
0
226
Site: Szeghalom 80
Area Intensively Collected: 1256 m2
Average Visibility: 87.5
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
341
587
37
18
0
Diagnostic Sherds
Rims
26
Periods Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Bases
7000
9300
149
80
0
Dec Body
Lugs/Handles
0
24
1
1
4
0
0
46
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N)
BodyWt (g)
Pre Diag
Other
Total
3
3
78
0
7
20
4
11
6
Later
Mod
Daub
Bone
Other
C
69
900
13
0
0
105
0
2
E30
96
1200
16
0
0
38
7
2
E60
84
1000
10
4
1
19
0
1
E90
20
200
0
0
0
9
0
1
E120
7
<100
1
0
0
3
0
0
E150
1
<100
0
0
0
1
0
0
W30
29
600
14
0
0
39
5
0
W60
53
400
0
0
1
18
2
5
W90
23
200
0
0
0
3
0
0
W120
3
<100
0
0
0
1
0
0
S30
55
1000
7
0
0
35
16
4
N30
58
2000
7
0
1
22
0
0
N60
6
<100
0
0
0
3
0
0
NE15
58
1200
7
0
0
35
7
2
NE45
22
600
0
0
0
10
0
0
NE75
3
<100
0
0
0
0
0
0
227
Site: Tarhos 67
Area Intensively Collected: 1020.5 m2
Average Visibility: 95%
Overview
Material
Daub
Total Body
Total Bone
Chipped Stone
Ground Stone
N
Wt (g)
7
468
11
4
1
Diagnostic Sherds
Rims
10
Periods
Represented
Prehistoric General
Neolithic General
Körös
AVK/Szakalhat
Tisza
Csoszhalom
Bases
<100
15950
297
4
66
Dec Body
Lugs/Handles
3
7
0
0
6
0
0
41
Tiszapolgar
Bodgrogkeresztur
Baden
Boleraz
EBA
Later
Distribution of Material (Rough Sort)
Unit
Body (N) BodyWt (g)
Other
9
Total
0
63
0
0
48
1
0
1
Pre Diag
Later
Mod
Daub
Bone
Other
C
98
2000
13
0
4
4
25
1
N30
59
1900
8
0
3
0
14
2
S30
73
1700
5
1
7
1
5
2
E30
60
3000
8
0
2
1
4
1
E60
0
0
0
0
0
0
1
1
S60
31
1750
1
1
4
0
0
0
S90
3
<100
0
0
2
0
0
0
W30
2
<100
0
0
2
0
0
0
SE15
96
3000
15
0
0
0
5
0
SE45
25
800
2
1
6
0
9
1
SE75
1
<100
0
0
0
0
1
0
NW15
9
800
0
0
3
1
4
2
NW45
11
1000
1
0
0
0
0
1
SF1
1
31
1
0
0
0
0
0
SF2
1
15
1
0
0
0
0
0
228
Site: Tarhos 67, Distribution of Material, continued.
Unit
Body (N)
BodyWt (g)
Pre Diag
Later
Mod
Daub
Bone
Other
SF2
1
15
1
0
0
0
0
0
SF3
3
62
3
0
0
0
0
0
SF4
1
16
1
0
0
0
0
0
SF5
1
28
1
0
0
0
0
0
SF6
3
67
3
0
0
0
0
0
229
APPENDIX B
PETROGRAPHIC DATA
230
231
% Voids
%
Matrix
% Sand
% Silt
% Matrix + Silt
% Temper
% Sand in
Body
E1
4.3
66.7
2.4
30.9
93.9
3.8
2.3
E1
11.1
75.4
0
24.6
95
5
0
LCA
E1
8.8
73.8
0
26.2
91
9
0
LCA
C2
8.2
76.3
3.7
20
96.3
0
3.7
Szeghalom80
LCA
C2
7.9
80.6
0
19.4
93.1
6.9
0
Tarhos67
AVK
C2
8.3
81
0.8
18.2
98.4
0.8
0.8
007
Gerla64
Neolithic
E2
2.5
94.2
0
5.8
88.4
11.6
0
008
Szeghalom80
EBA
D1
2.7
94.3
0
5.7
81.3
18.7
0
009
HMVH
LCA
D1
8
93
0
7
83.5
16.5
0
010
HMVH
LCA
D2
11
90
0
10
85.7
14.3
0
011
HMVH
LCA
E2
2.5
74.1
0
25.9
94.1
5.9
0
012
HMVH
LCA
E2
9
78.6
3.5
17.9
86.3
10.7
3
013
Békés178
MBA
F3
4.9
89.2
0
10.8
84.7
5.1
10.2
014
Békés26
EBA
F2
12.3
82.6
0
17.4
92
8
0
015
Békés26
EBA
F2
16.7
72.1
0
27.9
95.6
4.4
0
016
Békés26
MBA
E2
11.7
84
0
16
82.7
17.3
0
017
Békés26
MBA
F2
10.3
81.7
0
18.3
89.4
10.6
0
018
Békés26
EBA
F3
7.2
78.1
0
21.9
93.2
6.8
0
019
Békés26
LCA
E2
7.6
86.9
0
13.1
90
10
0
020
Békés26
LCA
E2
4.3
81
0
19
95.5
4.5
0
021
Gerla64
LCA
E2
11.5
81.6
0
18.4
98
2
0
022
Gerla64
LCA
D2
7
77.1
0
22.9
99
1
0
023
Gerla64
Neolithic
D3
4.1
77.7
4.5
17.8
92.3
3.4
4.3
024
Gerla64
EBA
E2
8.8
80
1
19
91.2
8
0.8
025
Gerla64
Neolithic
E3
8.8
81.9
2.1
16
89.3
8.8
1.9
026
Gerla64
Neolithic
E2
3.8
74.4
1.6
24
93.7
4.7
1.6
027
Szeghalom80
AVK
F1
2.4
81
0
19
89.4
10.6
0
028
Szeghalom80
LCA
E1
4.5
76.5
0
23.5
96.2
3.8
0
029
Szeghalom80
LCA
E2
2
77.3
1.1
21.6
87
12
1
Sample Number
Site
Culture
Fabric Class
001
Tarhos67
LCA
002
HMVH
LCA
003
Tarhos67
004
Tarhos67
005
006
Sample Number
Site
Culture
Fabric Class
% Voids
%
Matrix
% Sand
% Silt
% Matrix + Silt
232
% Temper
% Sand in
Body
030
Szeghalom80
MCA
E1
3.3
71.4
0
28.6
031
Szeghalom80
LCA
F1
6.5
85.9
0
14.1
100
0
0
80
20
0
032
Szeghalom80
LCA
E1
7
89.7
0
10.3
95.5
4.5
0
033
Szeghalom80
MCA
F1
8
78.8
0
21.2
96.1
3.9
0
034
Szeghalom80
MBA
F1
5.5
035
Szeghalom80
EBA
E1
2.8
79.8
0
20.2
96.1
3.9
0
83
1
16
96.1
3
0.9
036
Szeghalom80
MCA
E2
3.7
81.8
0
18.2
94.3
6.7
0
037
Szeghalom80
MBA
E3
8.3
84.4
3.1
12.5
93
4
3
038
Szeghalom80
EBA
039
Szeghalom80
AVK
E1
13
75.3
0
24.7
97.6
2.4
0
E3
9.3
76.3
0
23.7
100
0
0
040
Szeghalom80
EBA
F1
15
87.2
0
12.8
92.5
7.5
0
041
Szeghalom80
LCA
F1
5.2
71.8
0
28.2
94.5
5.5
0
042
Szeghalom80
EBA
F1
4.9
69.9
0
30.1
69.6
3.4
0
043
Szeghalom89
LCA
E1
5.5
62.4
0
37.6
89.4
10.6
0
044
Szeghalom80
EBA
E1
5.5
84.2
0
15.8
92.2
7.8
0
045
Szeghalom80
LCA
E1
12.4
77.8
0
22.2
93.4
5.6
0
046
Tarhos67
LCA
E1
4.9
60.9
0
39.1
94.8
5.2
0
047
Szeghalom80
LCA
F1
4.8
84.5
0
15.5
98.3
1.7
0
048
Tarhos67
LCA
F2
3.1
64.3
0
35.7
99.2
0.8
0
049
Tarhos67
AVK
E3
13.9
74.7
0
25.3
84.9
15.1
0
050
Tarhos67
LCA
E1
4.8
69
0
31
100
0
0
051
Tarhos67
AVK
D1
5.2
77.1
0
22.9
99
1
0
052
Tarhos67
LCA
F1
2.6
76.4
0
23.6
98.2
1.8
0
053
Tarhos67
LCA
F1
8.7
71.6
0
28.4
97.1
2.9
0
054
Tarhos67
LCA
E2
3.6
81.4
0
18.6
90.7
9.3
0
055
Tarhos67
LCA
D3
8.4
78.9
5.3
15.8
91.8
3.1
5.1
056
Mez!berény34
Neolithic
D3
5.9
79.5
4.5
16
95.5
0
4.5
057
Tarhos67
AVK
E1
0.9
89.6
0
10.4
86.5
13.5
0
058
Tarhos67
LCA
E1
1
86.5
0
13.5
96
4
0
Site
Culture
Fabric Class
% Voids
%
Matrix
% Sand
% Silt
% Matrix + Silt
% Temper
% Sand in
Body
059
Tarhos67
LCA
E1
16.4
91.9
060
Mez!berény34
LCA
E1
6.3
77.7
0
8.1
93.5
6.5
0
0
22.3
99
1
0
061
Bélmegyer82
LCA
F1
9.7
82.7
1.2
16.1
86
12.9
1.1
062
Bélmegyer82
Neolithic
E1
9.2
85.3
0
14.7
94.4
5.56
0
063
Bélmegyer82
LCA
064
Füzesgyarmat97
LCA
E1
5.1
79.4
0
20.6
95.5
4.5
0
F1
4.6
81.7
0
18.3
100
0
0
065
Füzesgyarmat97
LCA
F1
5.3
81.7
0
18.3
97.2
2.8
0
066
Füzesgyarmat97
Szarmation
F2
5.9
84.1
0
15.9
99.2
0.8
0
067
Füzesgyarmat97
Szarmation
E1
0.9
78
0
22
100
0
0
068
Füzesgyarmat97
LCA
E1
6.4
79.6
0
20.4
100
0
0
069
Füzesgyarmat97
LCA
F2
9.7
88.4
0
11.6
93.1
6.9
0
070
Busca13
LCA
F1
6.3
88.8
0
11.2
94.2
5.8
0
071
Busca13
LCA
F1
7.3
86.6
0
13.4
97.4
2.6
0
072
Busca13
MBA
E3
4.6
81
3
16
94.2
2.9
2.9
073
Békés178
EBA
E2
4.3
89.4
0
10.6
93.7
6.3
0
074
Békés178
EBA
F2
5
85.8
0
14.2
93.8
6.2
0
075
Békés178
LCA
E3
8.3
74.5
0
25.5
99
1
0
076
Békés178
LCA
E2
9.8
80
3
17
96
1
3
077
Békés178
EBA
E1
14
83.2
0
16.8
96.9
3.1
0
078
Békés178
ECA
E1
1.8
83.6
0
16.4
100
0
0
079
Békés178
Prehistoric
E1
12
81
0
19
95.5
4.5
0
080
Békés178
MBA
F2
14.2
85.7
0
14.3
79.4
20.6
0
081
Békés178
Prehistoric
F1
10.7
76.5
0
23.5
98
2
0
082
Békés178
EBA
E1
6.2
81.1
0
18.9
89.6
10.4
0
083
Békés178
LCA
F1
15.2
79.8
0
20.2
94.4
5.6
0
084
Békés178
LCA
F1
6.3
88.8
0
11.2
84.8
15.2
0
085
Biharugra33
MBA
E1
9.6
85.9
0
14.1
90.4
9.6
0
086
Biharugra33
LCA
F1
17.3
88.9
0
11.1
89
11
0
087
Biharugra33
LCA
F1
6.6
86.7
0
13.3
90.9
9.1
0
Sample Number
233
Culture
Fabric Class
% Voids
%
Matrix
% Sand
% Silt
234
% Temper
% Sand in
Body
Biharugra33
LCA
F1
11.4
82.1
0
17.9
Biharugra33
Prehistoric
F1
8.5
90.2
0
10.8
100
0
0
94.8
5.2
0
090
Biharugra33
LCA
F1
10.3
86.2
0
13.8
90.4
9.6
0
091
Biharugra33
MBA
E1
10.3
83
0
17
95.2
4.8
0
092
Biharugra33
MCA
E1
10.9
91.9
093
Biharugra33
MBA
E2
12.6
87.3
0
8.1
95.6
4.4
0
0
12.7
87.8
12.2
0
094
Biharugra33
MCA
F1
7.9
85.2
0
14.8
92.3
7.7
0
095
Biharugra33
LCA
F1
5.8
91.6
0
8.4
97.9
2.1
0
096
Biharugra33
MCA
097
Biharugra33
EBA
E1
16.7
85.1
0
14.9
98.9
1.1
0
F1
11
85.6
0
14.4
98.2
1.8
0
101
Vészt!49
LCA
E2
5.9
75.2
0
24.8
93.8
6.2
0
102
Körösladány16
MCA
F1
0.9
85.6
0
14.4
93.7
6.3
0
103
Bélmegyer56
LCA
E1
3.3
77.6
0.8
21.6
97.5
1.7
0.8
104
Doboz H. tábla
LCA
E1
3
79.2
0
20.8
99
1
0
105
Bélmegyer56
LCA
E2
6.3
88.3
0
11.7
86.7
10.3
0
106
Bélmegyer56
LCA
E2
10.7
89.1
0
10.9
100
0
0
107
Vészt!119
LCA
E2
3.9
92.1
0
7.9
93.4
6.6
0
108
Biharugra53
LCA
F1
14.4
84
0
16
99
1
0
109
Vészt!65
MCA
E1
9.5
86.4
0
13.6
98.1
1.9
0
110
Bélmegyer56
LCA
E3
3.9
77.8
5.1
17.1
94.9
0
5.1
111
Szeghalom60
LCA
F1
6.1
83.2
0
16.8
96.7
3.3
0
112
Bélmegyer56
LCA
E3
4.4
85.2
5.6
9.2
94.4
0
5.6
113
Békés39
LCA
E1
5.5
89.3
0
10.7
99
1
0
114
Szeghalom89
LCA
E1
1.9
89.4
0.9
9.7
99.1
0
0.9
115
Szeghalom80
MCA
E1
7.9
90.5
0
9.5
100
0
0
116
Doboz H. tábla
LCA
E1
3.3
85.7
0
14.3
96.7
3.3
0
117
Doboz H. tábla
LCA
E1
6.6
84
0
16
100
0
0
118
Doboz H. tábla
LCA
F1
10.9
88.9
0
11.1
100
0
0
119
Békés39
LCA
E2
4.5
80.2
0
19.8
100
0
0
Sample Number
Site
088
089
% Matrix + Silt
Sample Number
Site
Culture
Fabric Class
% Voids
%
Matrix
% Sand
% Silt
% Matrix + Silt
% Temper
% Sand in
Body
120
Körösladány33
LCA
F1
7.8
94.3
0
5.7
92.6
7.4
0
121
Vészt!49
LCA
F1
4.1
96.3
0
3.7
92.2
7.8
0
122
Szeghalom89
LCA
F1
8.5
85.6
0
14.4
100
0
0
123
Szeghalom112
LCA
E1
8.3
80.6
0
19.4
99
1
0
124
Békés39
LCA
E2
13
95.1
0
4.9
95.4
4.6
0
125
Mez!gyán2
LCA
E1
13
77.1
0
22.9
96.5
3.5
0
126
Doboz H. tábla
LCA
E2
13
88.5
1.1
10.4
98.9
0
1.1
127
Doboz H. tábla
LCA
F1
7.3
92.1
0
7.9
100
0
0
128
Bélmegyer56
LCA
E1
2.9
87
0
13
99
1
0
129
Doboz H. tábla
LCA
E3
0.09
76.8
10.7
12.5
89.3
0
10.7
130
Doboz H. tábla
LCA
E3
2.8
80.4
4.9
14.7
92.4
2.8
4.8
131
Körösladány16
MCA
F1
1.9
96.1
0
3.9
100
0
0
235
132
Okány43
LCA
F1
1.9
96
0
4
96.2
3.8
0
133
Dévaványa166
MCA
F1
6.2
81.1
0
18.9
100
0
0
134
Okány43
Neolithic
F1
2.9
91.6
0
8.4
95
5
0
135
Szeghalom168
MCA
E2
1.9
90.1
0
9.9
98.1
1.9
0
136
Szeghalom168
MCA
F1
2.5
88.7
0
11.3
97.5
2.5
0
137
Szeghalom168
MCA
E1
5.9
83.2
0
16.8
100
0
0
138
Szeghalom60
LCA
E1
2.8
90.1
0
9.9
96.2
3.8
0
139
Szeghalom168
LCA
E1
8.8
90.1
0
9.9
97.8
2.2
0
140
Szeghalom60
LCA
F1
4.6
82.8
0
17.2
96.1
3.9
0
141
Szeghalom168
LCA
F1
3
81.2
0
18.8
98
2
0
142
Szeghalom168
LCA
F1
7.9
95.2
0
4.8
90.3
9.7
0
143
Vészt!17
LCA
E1
4.8
92.7
0
7.3
96.3
3.1
0
144
Békés39
LCA
E1
2.9
86.7
0
13.3
98
2
0
145
Doboz H. tábla
LCA
E3
14.9
92.6
0
7.4
92.2
7.8
0
146
Békés75
LCA
E2
6.6
87.6
0
12.4
97.9
2.1
0
147
Békés39
LCA
E1
11.1
90
0
10
93.8
6.2
0
148
Körösladány21
LCA
F1
10.9
90.9
0
9.1
97.8
2.2
0
Sample Number
Site
Culture
Fabric Class
% Voids
%
Matrix
% Sand
% Silt
% Matrix + Silt
% Temper
% Sand in
Body
149
Füzesgyarmat97
LCA
F1
10.1
80.9
0
19.1
95.9
4.1
0
150
Szeghalom60
LCA
F1
8.6
95.8
0
4.2
100
0
0
236
WORKS CITED
Adams, W. Y., D. P. Van Gerven and R. S. Levy
1978 The Retreat from Migrationism. Annual Review of Anthropology 7:483-532.
Adovasio, J. M., J. D. Gunn and J. Donahue
1983 Evidence from Meadowcroft Rockshelter. In Early Man in the New World, edited
by R. Shutler. Sage, Beverly Hills, CA.
Ammerman, A. J. and L. L. Cavalli-Sforza
1984 The Neolithic Transition and the Genetics of Populations in Europe. Princeton
University Press, Princeton.
Anderson, D. G.
1990 The Paleoindian Colonization of Eastern North America: A View from the
Southeastern United States. In Early Paleoindian Economies of Eastern North America,
Research in Economic Anthropology, Supplement 3, edited by K. B. Tankersly and B. L.
Isaac, pp. 163-190. JAI Press, Greenwich, CT.
Anthony, D. W.
1986 The "Kurgan Culture," Indo-European Origins, and the Domestication of the
Horse: A Reconsideration. Current Anthropology 27(4):291-313.
1990 Migration in Archeology: The Baby and the Bathwater. American Anthropologist
92(4):895-914.
1992 The Bath Refilled: Migration in Archaeology again. American Anthropologist
94(1):174-176.
1995 Horse, Wagon, and Chariot: Indo-European Languages and Archaeology.
Antiquity 69:554-565.
2007 The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian
Steppes Shaped the Modern World. Princeton University Press, Princeton, NJ.
Arnold, D.
1985
Ceramic Theory and Cultural Process. Cambridge University Press, Cambridge.
Bakker, J. A., J. Kruk, A. E. Lanting and S. Milisauskas
1999 Bronocice, Flintbek, Uruk, and Jebel Aruda: The Earliest Evidence of Wheeled
Vehicles in Europe and the Near East. Antiquity 73:778-790.
Bamps, A.
1883 La Céramique Americaine au Point de vue des Éléments Constitutifs de la Pâte et
de sa Fabrication. Congrès des Americanistes Cinquième Session.
237
Bankoff, H. A. and F. A. Winter
1990 The Later Aeneolithic in Southeastern Europe. American Journal of Archaeology
94(2):175-191.
Banner, J.
1956
Die Péceler Kultur. Akadémiai Kiadó, Budapest.
Barton, C., G. A. Clark, D. R. Yesner and G. S. Pearson
2004 The Settlement of the American Continents: A Multidisciplinary Approach to
Human Biogeography. University of Arizona Press, Tucson.
Bellwood, P.
2004 First Farmers: The Origins of Agricultural Societies. John Wiley and Sons,
Hoboken, NJ.
Benecke, N.
1994 Archäozoologische Studein zur Entwicklung der Haustierhaltung in Mitteleuropa
und Südskandinavien von den Anfangen bis zun ausgehenden Mittelalter. Akademie
Verlag, Berlin.
Bense, J. A.
1994 Archaeology of the Southeastern United States: Paleoindian to World War I.
Academic Press, New York.
Berry, M. S.
1992 Time, Space, and Transition in Anasazi Prehistory. University of Utah Press, Salt
Lake City.
Beynon, D. W., J. Donahue, R. T. Shaub and R. A. Johnston
1986 Tempering Types and Sources for Early Bronze Age Ceramics from Bab edhDhra' and Numeria, Jordon. Journal of Field Archaeology 13(3):297-305.
Billman, B.
1999 Settlement Pattern Research in the Americas: Past, Present, and Future. In
Settlement Pattern Studies in the Americas: Fifty Years Since Virú, edited by B. Billman
and G. M. Feinman, pp. 1-5. Smishsonian Institution Press, Washington, D.C.
Binford, L.
1963
1964
441.
Archaeology as Anthropology. American Antiquity 28(2):217-225.
A Consideration of Archaeological Research Design. American Antiquity 29:425-
1965 Archaeological Semantics and the Study of Culture Process. American Antiquity
31:203-210.
238
1982
The Archaeology of Place. Journal of Anthropological Archaeology 1:1-31.
Blanton, R. E.
2001 Mediterranean Myopia. Antiquity 75:627-629.
Boas, F.
1889
1940
The Aims of Ethnology. Hermann Bartsch, New York.
Race, Language, and Culture. The Macmillan Company, New York.
Bognár-Kutzián, I.
1963 The Copper Age Cemetery of Tiszapolgár-Basatanya. Archaeologica Hungarica.
Akadémiai Kiadó, Budapest.
1966
Das Neolithikum in Ungarn. Archaeologia Austriaca 40:249-280.
1972 The Early Copper Age Tiszapolgár Culture in the Carpathian Basin.
Archaeologica Hungarica. Akadémiai Kiadó, Budapest.
Bogucki, P.
1996
1999
The Spread of Early Farming in Europe. American Scientist 84:242-253.
The Origins of Human Society. Blackwell, Oxford.
Bökönyi, S.
1951 Untersuchung der Haustierfunde aus dem Gräberfeld von Alsónémedi Acta
Archaeologica Academiae Scientarum Hungaricae 1:72-79.
1974 History of Domestic Mammals in Eastern and Central Europe. Akadémiai Kiadó,
Budapest.
1988 History of Domestic Mammals in Central and Eastern Europe. Akadémiai Kiadó,
Budapest.
Bóna, I.
1965 The Peoples of Southern Origin of the Early Bronze Age in Hungary. I. The
Pitvaros Group. Alba Regia 1:17-63.
1975 Die Mittlere Bronzezeit Ungarns und ihre südöstlichen Beziehungen. Akadémiai
Kiadó, Budapest.
Bourdieu, P.
1984 Distinction: A Social Critique of the Judgement of Taste. Harvard University
Press, Cambridge.
239
Bowser, B.
2000 From Pottery to Politics: An Ethnoarchaeological Study of Political Factionalism,
Ethnicity, and Domestic Pottery Style in the Ecuadorian Amazon
Journal of Archaeological Method and Theory 7:219-248.
Budden, S. and J. Sofaer
2009 Non-discursive Knowledge and the Construction of Identity: Potters, POtting, and
Performance at the Bronze Age Tell of Százhalombatta, Hungary. Cambridge
Archaeological Journal 19(2):203-220.
Caldwell, J. R.
1958 Trend and Tradition in the Prehistory of the Eastern United States. Memoirs of
the American Anthropological Association 88. Menasha.
Carver, M. O. H.
1998 Sutton Hoo: Burial Ground of Kings? University of Pennsylvania Press,
Philadelphia.
Chandler, G.
2001 Development of a Field Petrographic Analysis System and its Application to the
Study of Socioeconomic Interaction Networks of the Early Harappan Northwestern Indus
Valley of Pakistan. British Archaeological Reports 995, Oxford.
Chapman, J.
1994
The Origins of Farming in South East Europe Préhistorie Européenne 6:133-156.
1997 Landscapes in Flux and the Colonization of Time. In Landscapes in Flux: Central
and Eastern Europe in Antiquity, edited by J. Chapman and P. M. Dolukhanov, pp. 1-21.
Oxbow Books, Oxford.
Chapman, J. and P. M. Dolukhanov
1992 The Baby and the Bathwater: Pulling the Plug on Migration. American
Anthropologist 94(1):169-174.
Chayes, F.
1954a Petrographic Modal Analysis. Willey, New York.
1954b The Theory of Thin Section Analysis. Journal of Geology 62:92-101.
Cherry, J. F.
1984 Emergence of the State in the Prehistoric Aegean. Proceedings of the Cambridge
Philological Society 210:18-48.
240
2003 Archaeology Beyond the Site: Regional Survey and its Future. In Theory and
Practice in Mediterranean Archaeology: Old World and New World Perspectives, edited
by J. Papadopoulos and R. Leventhal. Cotsen Institute of Archaeology at UCLA, Los
Angeles.
Childe, V. G.
1930 The Bronze Age. The Macmillan Company, New York.
1950 Prehistoric Migrations in Europe. Instituttet for Sammenlignende
Kulturforskning, Oosterhout N.B., The Netherlands.
1958a New Light on the Most Ancient East. Routelege, London.
1958b The Prehistory of European Society. Penguin, London.
1969
Prehistoric Migrations in Europe. Oosterhout N.B., The Netherlands.
Clark, D. L.
1972 Models and Paradigms in Archaeology. In Models in Archaeology, edited by D.
L. Clark. Metheun, London.
1977 Spatial Information in Archaeology. In Spatial Archaeology, edited by D. L.
Clark. Academic Press, London.
Clark, J. E. and M. Blake
1994 The Power of Prestige: Competative Generosity and the Emergence of Rank
Societies in Lowland Mesoamerica. In Factional Competition and Political Development
in the New World, edited by E. M. Brumfiel and J. W. Fox, pp. 17-30. Cambridge
University Press, Cambridge.
Comşa, E.
1974 Die Entwicklung, Periodisierung, und relative Chronologie der jungsteinzeitlichen
Kulturen Rumäniens Zeitschrift für Archäologie 8:1-44.
Davis, J. L. (editor)
1998 Sandy Pylos: An Archaeological History from Nestor to Navarino. The University
of Texas Press, Austin.
2008
The Mallakastra Regional Archaeological Project: Internet Edition. vol. 2008.
Deagan, K.
1987 Artifacts of the Spanish Colonies of Florida and the Caribbean, 1500-1800.
Volume I: Ceramics, Glassware, and Beads. Smithsonian Institution Press, Washington,
D.C.
241
Déchelette, J.
1904 Les Vases Céramiques de la Gaule Romaine, Paris.
Dragendorff, H.
1895 Terra Sigillata. Bonner Jahrbücher 96:18-155.
Earle, T. K.
2002 Bronze Age Economics: The Beginnings of Political Economies. Westview Press,
Boulder, CO.
Ebdon, D.
1985
Statistics in Geography: A Practical Approach. . Blackwell.
Ecsedy, I.
1973
Szeghalom 80 Dió-Ér. Régészeti Füszetek 1973:20-21.
1979 The People of the Pit-grave Kurgans in Eastern Hungary. Akadémiai Kiadó,
Budapest.
1981 A Keletmagyarországi Rézkor Fejlődésének Fontosabb Tényezői (On the factors
of the Copper Age development in Eastern Hungary). Janus Pannonius Múzeum
Évkönyve 26:73-95.
Ecsedy, I. and S. Bèokèonyi
1979 The People of the Pit Grave Kurgans in Eastern Hungary. Fontes archaelogici
Hungariae. Akadâemiai Kiadâo, Budapest.
Ecsedy, I., L. Kovács, B. Maraz and I. Torma (editors)
1982 Magyarország Régészeti Topográfiája. VI. Békés Megye Régészeti Topográfiaje:
A Szeghalmi Jaras (IV/1). Akadémiai Kiadó, Budapest.
Fairbanks, C.
1962 A Colono-Inidian Ware Milk Pitcher. Florida Anthropologist 15:103-106.
Faught, M. K.
2008 Archaeological Roots of Human Diversity in the New World: A Comilation of
Accurate and Precise Radiocarbon Ages from Earliest Sites. American Antiquity
73(4):670-698.
Ferguson, L.
1978 Looking for the "Afro" in Colono-Indian Pottery. Conference on HIstoric Site
Archaeology Papers 12:68-86.
Flannery, K. V.
2002 The Origins of the Village Revisited: From Nuclear to Extended Households.
American Antiquity 67(3):417-433.
242
Foley, R.
1981 A Model of Regional Archaeological Structure. Proceedings of the Historical
Society 47:1-17.
Forenbaher, S.
1993 Radiocarbon Dates and Absolute Chronology of the Central European Early
Bronze Age. Antiquity 67:218-256.
Forster, P. and C. Renfrew (editors)
2006 Phylogenic Methods and the Prehistory of Languages. McDonald Institute for
Archaeological Research, Cambridge.
Fowles, S. M.
2002 From Social Type to Social Process: Placing "Tribe" in a Historical Framework.
In The Archaeology of Tribal Societies, edited by W. A. Parkinson, pp. 13-33.
International Monographs in Prehistory, Ann Arbor, MI.
Freestone, I. C.
1995 Ceramic Petrography. American Journal of Archaeology 99(1):111-117.
Frey, O. H.
1995
The Celts in Italy. In The Celtic World, edited by M. Green. Routlege, London.
Fried, M. H.
1967 The Evolution of Political Society: An Essay in Political Anthropology. Random
House, New York.
Frolking, T.
2009 Title. Holocene Hydrology, Soil Development and Landscape Evolution of the
Bikeri Area in the Körös Basin of Eastern Hungary, in press.
Furholt, M.
2008 Pottery, Cultures, People? The European Baden Material Re-examined. Antiquity
82:617-628.
Galaty, M. L.
1999 Nestor's Wine Cups: Investigating Ceramic Manufacture and Exchange in a Late
Bronze Age "Mycenaean" State. British Archaeological Reports International Series 776.
British Archaeological Reports, Oxford.
2005 European Regional Studies: A Coming of Age? Journal of Archaeological
Research 13(4):291-336.
243
Galaty, M. L. and W. A. Parkinson
1999 Putting Mycenean Palaces in Their Place. In Rethinking Mycenaean Palaces: New
Interpretations of an Old Idea, edited by M. L. Galaty and W. A. Parkinson, pp. 1-8. The
Cotsen Institute of Archaeology, Los Angeles.
Gazdapusztai, G.
1967 Chronologische Fragen in der Alfölder Gruppe der Kurgan-Kultur. Móra Ferenz
Múzeum Évkonyve 1966-67:91-100.
Georgiev, G. I., N. Y. Merpert, R. W. Katinsharov and D. G. Dinitrov
1979 Ezero, Rannobronzovoto Selišče (Ezero, eine Siedlung aus der Frübronzezeit),
Sofia.
Gimbutas, M.
1952 On The Origin of North Indo-Europeans. American Anthropologist 54(4):602611.
1963 The Indo-Europeans: Archaeological Problems. American Anthropologist
54(4):602-611.
1965
Bronze Age Culture in Central and Eastern Europe. Mouton, The Hague.
1970 Proto-Indo-European Culture: The Kurgan Culture during the Fifth, Fourth, and
Third Millenia B.C. In Indo-European and Indo-Europeans, edited by G. Cardona, H. M.
Hoeningswald and A. Senn. University of Pennsylvania Press, Philadelphia.
1977 The first wave of Eurasian Steppe Pastoralists into Copper Age Europe. Journal
of Indo-European Studies 5:277-331.
1979 The three Waves of the Kurgan People into Old Europe, 4500-2500 B.C. Archives
Suisses d'anthropologie genéral 43(2):113-137.
1980 The Kurgan Wave #2 (c. 3400-3200 B.C.) into Europe and the Following
Transformation of Culture. Journal of Indo-European Studies 8(3&4):273-315.
1997 The Kurgan Culture and the Indo-Europeanization of Europe, edited by M.
Robbins Dexter and K. Jones-Blet. Journal of Indo-European Studies Monograph No. 18.
Institute for the Study of Man, Washington, D.C.
Gladwin, N.
1937 Petrography of Snaketown Pottery. In Excavations at Snaketown, edited by H. S.
Gladwin. Medallion Papers, Gila Pueblo, Arizona.
Gmelch, G.
1980
Return Migration. Annual Review of Anthropology 9:135-159.
244
Goebel, T., M. R. Waters and D. H. O'Rourke
2008 The Late Pleistocene Dispersal of Modern Humans in the Americas. Science
319:1497-1502.
Gyucha, A.
2007 Personal Communication. Magyarország Régészeti Topográfiája Collection
Methodology.
2010 A Körös-vidék Kora Rézkora (The Early Copper Age in the Körös Region).
Dissertation, Eötvös Loránd Science University, Budapest.
Gyucha, A., W. A. Parkinson and R. W. Yerkes
2004 El″ozetes Jelentés A Körös Regionális Régészeti Program 1998-2002 Között
Végzett Munkájáról (Preleminary Results of The Körös Regional Archaeological Project,
1998-2002). Studia Archaeologica X(2004):25-52.
Gyulai, F.
1993
Environment and Agriculture in Bronze Age Hungary. Archaeolingua, Budapest.
Haggett, P.
1966
Location Analysis in Human Geography. St. Martin's Press, New York.
Hayden, B.
1995 Pathways to Power: Priciples for Creating Socioeconomic Inequalities. In
Foundations of Social Inequality, edited by T. D. Price and G. M. Feinman, pp. 15-86.
Plenum Press, New York.
Haynes, G.
2002 The Early Settlement of North America: The Clovis Era. Cambridge University
Press, New York.
Higgenbotham, C. R.
2000 Egyptianization and Elite Emulation in Ramesside Palestine: Governance and
Accomodation on the Imperial Periphery. Culture and History of the Ancient Near East
2. BRILL.
Hodder, I. and C. Orton
1976 Spatial Analysis in Archaeology. Cambridge University Press, Cambridge.
Hodges, R.
1987 Spatial Models, Anthropology, and Archaeology. In Landscape and Culture:
Geographical and Archaeological Perspectives, edited by J. M. Wagstaff. Basil
Blackwell, New York.
245
Hoekman-Sites, H., T. Parsons and S. Duwe
2007 Elemental Interaction: Stylistic, Composition, and Residue Analyses of Copper
Age Ceramics on the Great Hungarian Plain. SAS Bulletin 29(4).
Holmes, W. H.
1903 Aboriginal Pottery of the Eastern United States. Bureau of American Ethnology,
Annual Report 20:1-237.
Horváth, T., S. Svingor and M. Molnár
2008 New Radiocarbon Dates for the Baden Culture. Radiocarbon 50(3):447-458.
Jankovich, D., J. Makkay and M. Szoke (editors)
1989 Magyarország Régészeti Topográfiája. VIII. Békés Megye Régészeti
Topográfiaja: A Szarvasi Járás (IV/2). Akadémiai Kiadó, Budapest.
Jankovich, D., P. Medgyesi, E. Nikolin, I. Szatmári and I. Torma (editors)
1998 Magyarország Régészeti Topográfiája. VIII. Békés Megye Régészeti
Topográfiaja: Békés es Békéscaba Környéke. Akadémiai Kiadó, Budapest.
Jochim, M.
2002 The Upper Paleolithic. In European Prehistory: A Survey, edited by S.
Milisauskas. Plenum Publishers, New York.
Jochim, M., C. Herhahn and H. Starr
1999 The Magdalenian Colonization of Southern Germany. American Anthropologist
101(129-142).
Kaczanowska, M.
1982/1983
Z Badań nad przemysł Krzemiennym Kultury Ceramiki Promienisej. Acta
Archaeologica Carpatica 22:65-96.
Kalicz, N.
1958 Rézkori Sztratagráfia Székelu Közég Határában (Copper Age Stratigraphy Near
the Village of Székelu) ArchÉrtesítő 85:3-6.
1963
Die Péceler (Badener) Kultur und Alatolien Studia Archaeologica, Budapest.
1976 Ein neues kupferzeitliches Wagenmodell aus der Umgebung von Budapest.
Festschrift für Richard Pirrioni sum siebzigsten Gubertstag. Deuticke, Vienna.
1998 Figürliche Darstellung und Bemalte Keramik aus dem Neolithikum Westungarns.
Archaeolingua, Budapest.
246
2001 Die Protoboleráz phase an der Grenze von zwei Epochen. In Ein
Vorgeschichtliches Phänomen zwischen dem Oberrhein und der Unteren Donai, edited
by P. Roman, pp. 385-422. Studia Danubiana ser. Cernavoda III-Boleráz Symposium
2001. Institude of Thracology, Bucharest.
2002 Eigenartige anthropomorphe Plastik der kupferzeitlichen Badener Kultur im
Karpatenbecken. Budapest Régiségei 36:11-53.
Kalicz, N. and J. Makkay
1977 Die Linienbandkeramik in der Grossen Ungarischen Tiefebene. Studia
Archaeologica. Akadémiai Kiadó, Budapest.
Kalicz, N. and J. Makkey
1997 Die Linienbandkeramik in der Drossen Ungarischen Tiefebene. Studia
Archaeologica. Akadémiai Kiadó, Budapest.
Kalicz, N. and P. Raczky
1984 Preliminary Report on the 1977-82 Excavations at the Neolithic and Bronze Age
Tell Settlement of Berettyóújfalu Herpály. Part I. Neolithic. Acta Archaeologica
Hungarica 36:85-136.
1987a The Late Neolithic of the Tisza Region: A Survey of Recent Archaeological
Research. In The Late Neolithic of the Tisza Region, edited by P. Raczky. Szolnok
County Museums, Budapest-Szolnok.
Kearney, M.
1986 From the Invisible Hand to Visible Feet: Anthropological Studies of Migration
and Development Annual Review of Anthropology 15:331-361.
Kertész, R.
1996 The Mesolithic in the Great Hungarian Plain: A survey of the evidence. In At the
Fringes of Three Worlds: Hunter-Gatherers and Farmers in the Middle Tisza Valley, pp.
5-34. Damjanich Museum Press, Szolnok, Hungary.
Kertéz, R.
1996 The Mesolithic in the Great Hungarian Plain: A Survey of the Evidence. In At the
Fringes of Three Worlds: Hunter-Gatherers and Farmers in the Middle Tisza Valley, pp.
5-34. Damjanich Museum Press, Szolnok, Hungary.
Kidder, A. V.
1924 An Introduction to the Study of Southwestern archaeology. Papers of the
Southwestern Expedition, Department of Anthropology 1. Yale University Press, New
Haven.
1931 The Pottery of Pecos. Papers of the Southwestern Expedition, Department of
Anthropology 5. Yale University Press, New Haven.
247
Knorr, K.
1906
Die Verzierten Terra Sigillata von Cannstatt, Stuttgart.
Korek, J.
1951 Ein Gräberfeld der Badener Kultur bei Alsónémedi. Acta Archaeologica
Academiae Scientarum Hungaricae 1(35-51).
Kosse, K.
1979 Settlement Ecology of the Early and Middle Körös and Linear Pottery Cultures in
Hungary. BAR International Series No. 64. British Archaeological Reports, Oxford.
Kreiter, A.
2005 Technological Choices and Material Meanings: Analyses of Early and Middle
Bronze Age Ceramics from Hungary. Thesis for the Degree of Doctor of Philosophy,
School of Humanities, University of Southamptom, Southampton.
Kristiansen, K.
1998 Europe Before Prehistory. Cambridge University Press, Cambridge.
2005 What Language did Neolithic Pots Speak? Colin Renfrew’s European FarmingLanguage-Dispersal Model Challenged. Antiquity 79:679-691.
Kulcsár, G.
2003 The Early Bronze Age. In Hungarian Archaeology at the Turn of the Millennium,
edited by Z. Visy and M. Nagy. Ministry of National Cultural Heritage, Budapest.
Kutzián , I.
1944
A Körös-kultura. Akadémiai Kiadó, Budapest.
Lee, E. S.
1966
A Theory of Migration. Demography 3:47-57.
Lemmonier, P.
1992 Elements for an Anthropology of Technology. Anthropological Papers No. 88.
University of Michigan Museum of Anthropology, Ann Arbor, MI.
Lewis, G. J.
1982
Human Migration: A Geographical Perspective. St. Martin's Press, New York.
Liddell, D.
1932 Report on the Excavation at Hembury Fort, Third Season, 1932. Proceedings of
the Devon Archaeological Exploration Society 1:162-183.
Ludowici, W.
1904 Stempel-Namen Römischer Töpfer von meinen Ausgrabungen im Rheinzabern,
Munich.
248
Madsen, D. B.
2004 Entering America: Northeast Asia and Beringia Before the Last Glacial
Maximum. University of Utah Press, Salt Lake city.
Makkay, J.
1981 Eine kultstätt de Bodrogkeresztúr-kultur in Szarvas und fragen der sakralen hügel.
Mittelilungen des Archäologisches Institut der Ungarischen Akademie der
Wissenschaften 10/11:45-57.
1982 A magyarországi neolitikum kutatásának új eredményei (New Issues in
Hungarian Neolithic Research). Akadémiai Kiadó, Budapest.
1983 Eine Kultstätte der Bodrogkeresztúr-Kultur in Szarvas und Fragen der sakralen
Hügel. Mitteilungen des archaeologischen Instituts der Ungarischen Academie der
Wissenschaften 10-11:45-47.
1986
Gyomaendrőd-6 lelőhely. Régészeti Füszetek 40(13).
1996 Theories about the Origin, the Distribution, and the End of the Körös Culture In
At the Fringes of Three Worlds: Hunter-Gatherers and Farmers in the Middle Tisza
Valley, pp. 35-49. Damjanaich Museum Press, Szolnok, Hungary.
2007 The Excavations of the Early Neolithic Sites of the Körös Culture in the Körös
Valley, Hungary: The Final Report. In The Excavations of the Early Neolithic Sites of the
Körös Culture in the Körös Valley, Hungary: The Final Report, edited by E. Starnini and
P. Biagi. Società per la Preistoria e Prorostoria dell Regione Friulli-Venezia, Guilia.
Makkey, J.
1996 Theories about the Origin, the Distribution, and the End of the Körös Culture. In
At the Fringes of Three Worlds: Hunter-Gatherers and Farmers in the Middle Tisza
Valley, pp. 35-49. Damjanich Museum Press, Szolnok, Hungary.
Mallory, J. P.
1989 In Search of the Indo-Europeans: Language, Archaeology, and Myth. Thames and
Hudson, London.
Maran, J.
2001 Zur Westausbreitung von Boleráz--Elementen in Mitteleuropa. In Cernavodă IIIBoleráz. Ein vorgeschichtliches Phänomen zwischen dem Oberrhein und der unteren
Donau, edited by P. Roman and S. Diamandi. Vavila, Bucharest.
2004 Die Badener Kultur und ihre Räderfahrzeuge. In Rad und Wagen. Der Ursprund
einer Innovation. Wagen Orient und Europa, edited by S. Burmeister and M. Fansa, pp.
265-282. Philipp von Zabern, Mainz.
249
Matson, R. G.
1991 The Origins of Southwestern Agriculture. University of Arizona Press, Tucson.
Mauss, M.
1955 The Gift: Forms and Functions of Exchange in Archaic Societies. Routledge,
London.
McDonald, W. A. and G. R. Rapp (editors)
1972 The Minnesota Messenia Expedition: Reconstructing a Bronze Age Regional
Environment. University of Minnesota Press, Minneapolis.
McGuire, J. D.
1899 Pipes and Smoking Customs of the American Aborigines, Based on Material in
the U.S. National Museum. . United States National Museum, Annual Report
1897(1):361-645.
Megyesi, P.
1982
Jelentég a Doboz – Homokdörögdi táblán 1982. Unpublished field notes.
1983
Doboz-Homokgördöri táblá. Régészeti Füzetek.
2009
Personal Communication, Békéscsaba, Hungary
Merpert, N. Y.
1968 Drevenishaia Istoriia Naselnia Stepnoi Polos Vostochnoi Evropy (IIIrd-Nachalo
Ilnd Tys. Do N.E.). Akademiia Nauk, Moscow.
1974
Drevnejshie skotovody volzhsko-ural'skogo mezhdurech'ja, Moscow.
Middleton, A. and I. C. Freestone
1991 Recent Developments in Ceramic Petrology. British Museum, London.
Milisauskas, S. and J. Kruk
1989 Economy, Migration, Settlement Organization, and Warfare During the Late
Neolithic in Southeastern Poland. Germania 67(1):77-96.
2002 Late Neolithic Crises, Collapse, New Ideologies, and Economies. In European
Prehistory: A Survey, edited by S. Milisauskas. Kluwer Academic/Plenum Publishers,
New York.
Minnis, P. E.
1992 Earliest Plant Cultivation in the Desert Borderlands of North America. In Origins
of Agriculture: an International Persoective, edited by C. W. Cowan and P. J. Watson,
pp. 121-141. Smithsonian Institution Press, Washington.
250
Morgan, L. H.
1885 Ancient Society. University of Arizona Press, Tucson.
Moscati, O. H., V. Kruta, B. Rafferty and M. Szabó (editors)
1991 The Celts and their Movements in the Third Century BC. Rizzoli, New York.
Nádor, A., Á. Magyari, E. Thamó-Bozsó, E. Babinszki and Z. Kercsmár
2005 The role of the Érmellék-Berettyó-Körös depression in the river course
development of the Great Hungarian Plain. Geophysical Research Abstracts 7.
Němejcová-Pavuková, V.
1981 Náčrt Periodizácie Badenskej Kultúry a jej chronologických vztahov k juhový
chodeney Europe (An Outline of the Periodical System of Baden Culture and its
Chronological Relations to Southeast Europe). Slovenská Archeológia 29(2):261-296.
1984 K problematike trvania a konca boleazkej skupiny na Slovensku. Slovenská
Archeológia 34:133-176.
Némeskeri, J.
1951 Anthropologische Untersuchund der Skelettfunde von Alsónémedi Acta
Archaeologica Academiae Scientarum Hungaricae 1:55-72.
Noël Hume, I.
1962 An Indian Ware of the Colonial Period. Quarterly Bulletin of the Archeological
Society of Virginia 17(1):2-14.
Novotný, B.
1981 Zur Idolatrie der Badener Kultur in der Slowakei. Slovenská Archeológia 29(131138).
Obenauer, K.
1936 Petrographische Untersuchung der Keramik. In Die Bandkeramische Ansiedelung
bei Köln-Lindenthal, edited by W. Buttler and W. Haberey, pp. 123-129. de Gruyter,
Berlin.
Orton, C., P. Tyers and A. Vince
1993 Pottery in Aarchaeology. Cambridge Manuals in Archaeology. Cambridge
University Press, Cambridge.
O'Shea, J. M.
1978 Mortuary Variability: an Archaeological Investigation with Case Studies from the
Nineteenth Century Central Plain of North America and the Early Bronze Age of
Southern Hungary, University of Cambridge, Cambridge.
251
1996 Villagers of the Maros: A Portrait of an Early Bronze Age Society. Plenum Press,
New York.
Otte, M.
1988 Processus de Diffusion a Long Terme au Magdalénian. In Le Peuplement
Magdalénian. Éditions du CTHS, Paris.
Parkinson, W. A.
1999 The Social Organization of Early Copper Age Tribes on the Great Hungarian
Plain. Dissertation, Department of Anthropology, The University of Michigan, Ann
Arbor.
2002 Integration, Interaction, and Tribal 'Cycling:' The Transition to the Copper Age on
the Great Hungarian Plain. In The Archaeology of Tribal Societies, edited by W. A.
Parkinson. Archaeological Series. International Monographs in Prehistory, Ann Arbor.
2006a Tribal Boundaries: Stylistic variability and boundary maintenance during the
transition to the Copper Age on the Great Hungarian Plain. Journal of Anthropological
Archaeology 25:33-58.
2006b The Social Organization of Early Copper Age Tribes on the Great Hungarian
Plain. BAR International Series. Archaeopress, Oxford.
Parkinson, W. A. and M. L. Galaty
2007 Secondary States in Perspective: An Integrated Approach to State Formation in
the Prehistoric Aegean. American Anthropologist 109(1):113-129.
Parkinson, W. A. and A. Gyucha
2007 A Késő Neolitikum-Kora Rézkor Átmeneti Időszakának Társadalomszerkezeti
Változásai az Alföldön. Rekonstrukciós Kísérlet (Changes in Social Organization During
the Transition from the Late Neolithic to the Early Copper Age Periods on the Great
Hungarian Plain: An Attempt at a Reconstruction). Archaeológiai Értesítő 132(1):37-81.
Parkinson, W. A., A. Gyucha, R. W. Yerkes, M. Hardy and M. Morris
2004 Settlement Reorganization at the End of the Neolithic in Central Europe: Recent
Research in the Körös River Valley, Southeastern Hungary. Journal of Eurasian
Prehistory 2(2):57-73.
Parkinson, W. A., E. Peacock, R. A. Palmer, Y. Xia, B. Carlock, A. Gyucha, R. W. Yerkes and
M. L. Galaty
2010 Elemental Analysis of Ceramic Encrustation Indicates Long-Term Cultural
Continuity in the Prehistoric Carpathian Basin. Archaeology, Ethnology, and
Anthropology of Eurasia 38(2):64-70.
252
Parsons, T.
2005 A Petrographic Analysis of Early Copper Age Ceramics from Vésztö 20,
Körösladány 14, Vésztö-Mágor, and Örménykút 13, in Bekés County, Hungary. La
Tinaja (May 2005).
Pauketat, T. R.
1994 The Ascent of Cheifs: Cahokia and Mississippian Politics in Native North
America. University of Alabama Press.
Pavelčík, J.
1982 Drobné terrakoty z Hlinska u Lipníku (okr. Přerov). Památky Archeologické
73:261-292.
1992 Nové nálezy anthropomorfních a zoomorfních idolůna Moravé. Pravěk Nova
Řada (2):205-214.
Peacock, D. P. S.
1967 The Heavy Mineral Analysis of Pottery: A Preliminary Report. Archaeometry
10:97-100.
1968 A Petrological Study of Certain Iron Age Pottery from Western England.
Proceedings of the Prehistoric Society 34(414-427).
Pécsi, M.
1964 Part One: The Physical Geography of Hungary. In The Geography of hungary,
edited by M. Pécsi and B. Sárfalvi. Collets, London.
1970
Geomorphological Regions of Hungary. Akadémiai Kiadó, Budapest.
Pécsi, M. and P. Jakucs
1971 The Natural Vegetation of Hungary. In Hungary: Geographical Studies, edited by
M. Pécsi, G. Enyedi and S. Marosi, pp. 11-64. Akadémiai Kiadó, Budapest.
Pécsi, M. and B. Sárfalvi
1964 The Geography of Hungary. Collets, London.
Pitt-Rivers, A. H. L. F.
1906 The Principles of Classificiation. In The Evolution of Culture and Other Essays,
edited by J. L. Myers. Clarenden Press, Oxford.
Ravenstein, E. G.
1885 The Laws of Migration. Journal of the Royal Statistical Society 48:167-235.
1889
The Laws of Migration. Journal of the Royal Statistical Society 52:241-305.
253
Reedy, C.
2008 Thin-Section Petrography of Stone and Ceramic Cultural Materials. Archetype
Publications, London.
Renfrew, C.
1975 Trade as Action at a Distance: Questions of Integration and Communication. In
Ancient Civilization and Trade, edited by J. Sabloff and C. Lamberg-Karlovsky, pp. 3-59.
University of New Mexico Press, Albuquerque.
1987 Archaeology and Language: The Puzzle of Indo-European Origins. Jonathan
Cape Ltd., London.
2003 Time Depth, Convergence Theory, and Innovation in Proto-Indo-European. Paper
presented at the Languages in Prehistoric Europe, Eichstätt University.
Renfrew, C. and J. F. Cherry (editors)
1986 Peer Polity Interaction and Socio-political Change. Cambridge University Press,
Cambridge.
Rice, P.
1987
Pottery Analysis: A Source Book. University of Chicago Press, Chicago.
Riederer
2004 Thin Section Microscopy Applied to the Study of Archaeological Ceramics.
Hyperfine Interactions 154(1-4):143-158.
Roman, P. and S. Diamandi
2001 Cernavodă III-Boleráz. Ein Vorgeschichliches Phänomen zwischen dem
Oberrhein und der unteren Donau. Vavila, Bucharest.
Roman, P. and I. Németi
1978 Cultura Baden în România. Institute of Archaeology, Bucharest.
Roper, D. C., R. L. Josephs and M. E. Beck
2010 Determining Provenance of Shell-Tempered Pottery from the Central Plains
Using Petrography and Oxidation Analysis. American Antiquity 75(1):134-157.
Rouse, I.
1958 The Inference of Migrations from Anthropological Evidence. In Migrations in
New World Culture History, edited by R. Thompson. University of Arizona Bulletin.
University of Arizona Press, Tucson.
1986 Migrations in Prehistory: Inferring Population Movement from Cultural Remains.
Yale University Press, New Haven, CT.
254
Ruttkay, E.
1995 Spätneolithikum. In Jungsteinzeit im Osten Österreichs, edited by E. Lenneis, C.
Neugebaur-Maresch and E. Ruttkay, pp. 108-209. Niederösterreichisches Pressehaus,
Vienna.
Sachße, C.
2005 Untersuchungen zu den Bestattungssitten der Badener Kultur in ihrem räumlichen
und zeitlichen Umfeld., University of Heidelberg, Heidelberg, Germany.
Sahlins, M. and E. Service
1960 Evolution and Culture. University of Michigan Press, Ann Arbor, MI.
Salisbury, Roderick B.
2010 Sediments, Settlements, and Space: A Practice Approach to Late Neolithic
Communities on the Great Hungarian Plain. Unpublished Doctoral dissertation, State
University of New York at Buffalo.
Séfériadès, M.
2007 Complexity of the Processes of Neolithization: Tradition and Modernity of the
Aegean world at the dawn of the Holocene Period. Quaternary International 167168:177-185.
Service, E.
1971 Primitive Social Organization: An Evolutionary Perspective. Second ed. Random
House, New York.
Shepard, A. O.
1939 Appendix A: Technology of La Plata Pottery. In Archaeological Studies in the La
Plata District, edited by E. Morris. vol. 519. Carnegie Institution of Washington
Publication, Washington, D.C.
1942 Rio Grande Glaze Paint Ware: A Study Illustrating the Place of Ceramic
Technological Analysis in Archaeological Research. Carnegie Institute of Washington,
Washington.
1956
Ceramics for the Archaeologists. Carnegie Institute of Washington, Washington.
Sherratt, A.
1983 The Development of Neolithic and Copper Age Settlement in the Great
Hungarian Plain. Part I: The Regional Setting. Oxford Journal of Archaeology 1(1):287316.
1984 The Development of Neolithic and Copper Age Settlement in the Great
Hungarian plain. Part II: Site Survey and Settlement Dynamics. Oxford Journal of
Archaeology 2(1):13-41.
255
1997a The Development of Neolithic and Copper Age Settlement in the Great
Hungarian Plain I: The Regional Setting. In Economy and Society in Prehistoric Europe:
Changing Perspectives, pp. 270-292. Princeton University Press, Princeton, New Jersey.
1997b The Development of Neolithic and Copper Age Settlement in the Great
Hungarian Plain II: Site Survey and Settlement Dynamics. In Economy and Society in
Prehistoric Europe: Changing Perspectives. Princeton University Press, Princeton, NJ.
Sklenář, K.
1983 Archaeology in Central Europe: The First 500 Years. Leicester University Press,
Leicester.
Snow, D.
2009 The Multidisciplinary Study of Human Migration: Problems and Principles. In
Ancient Human Migrations: A Multidisciplinary Approach, edited by P. N. Peregrine, I.
Peiros and M. Feldman. Foundations of Archaeological Inquiry, J. M. Skibo, general
editor. The University of Utah Press, Salt Lake City.
Sochacki, Z.
1980a Kultura Ceramiki Promienistej w Europie (Kultur der Radialverzierten Keramik
in Europa), Warsaw.
1985 The Relationship Between the Baden Culture and the South-East European
Cultures. Journal of Indo-European Studies 13(3 & 4):257-268.
Stark, M.
1998a The Archaeology of Social Boundaries. Smithsonian Institution Press,
Washington, D.C.
1998b Technical Choices and Social Boundaries in Material Culture Patterning: An
Introduction. In The Archaeology of Social Boundaries, edited by M. Stark. Smithsonian
Institution Press, Washington, D.C.
Steward, J.
1955 Theory of Culture Change: The Methodology of Multilinier Evolution. University
of Illinois Press, Urbana, IL.
1960 Evolutionary Principles and Social Types. In The Evolution of Man: Mind,
Culture, and Society, edited by S. Tax. University of Chicago Press, Chicago.
Stoltman, J.
1989 A Quantitative Approach to the Petrographic Analysis of Ceramic Thin Section.
American Antiquity 54(1):147-160.
1991 Ceramic Petrography as a Technique for Documenting Cultural Interaction: An
Example from the Upper Mississippi Valley. American Antiquity 56(1):103-120.
256
2001 The Role of Petrography in the Study of Archaeological Ceramics. In Earth
Sciences and Archaeology, edited by P. Goldberg, V. T. Holliday and C. R. Ferring, pp.
297-326. Kluwer/Plenum, New York.
Szujkó-Lacza, J.
1991 Reconstruction of the History of the Postglacial Flora and Vegetation in the
Region Between the Danube and Tisza and Tiszantul, Kiev.
Telegin, D. I.
1973 Seredno-Stogivska Kul'tura Epokhi Midi. Naukova Dumka, Kiev.
Thompson, E.
1996 The Huns. Blackwell, Oxford.
Trigger, B.
1989
A History of Archaeological Thought. Cambridge University Press, Cambridge.
Tringham, R.
1971 Hunters, Fishers, and Farmers of Eastern Europe: 6000-3000 B.C. Huchinson
Univerisity Library, London.
Tylor, S. E. B.
1871 Primitive Culture. John Murray, London.
van Andel, T. H. and C. H. Runnels
1995 The Earliest Farmers in Europe. Antiquity 69:380-390.
Walters, H. B.
1908 Catalogue of the Roman Pottery in the Department of Antiquities of the British
Museum. British Museum, London.
Wandsnider, L.
1988 Regional Scale Processes and Archaeological Landscape Units. In Unit Issues in
Archaeology, edited by A. F. Ramenofsky and A. Steffen. University of Utah Press, Salt
Lake City.
Wells, P.
2002 The Iron Age. In European Prehistory: A Survey, edited by S. Milisauskas, pp.
335-383. Plenum Publishers, New York.
Whitbred, I.
1989 A Proposal for the Systematic Description of This Sections Towards the Study of
Ancient Ceramic Technology. In Archaeometry: Proceedings of the 25th International
Symposium, edited by Y. Maniatis. Elsevier, Amsterdam.
257
1995 Greek Transport Amphoraw: A Petrological and Archaeological Study. Fitch
Laboratory Occasional Paper 4. British School at Athens, Athens.
White, L. A.
1959 The Evolution of Culture. McGraw-Hill Book Company, New York.
Whittle, A.
1996 Europe in the Neolithic: The Creation of New Worlds. Cambridge University
Press, Cambridge.
Willey, G. R. and P. Phillips
1958 Method and Theory in American Archaeology. University of Chicago Press,
Chicago.
Williams, H.
2006 Death and Memory in Early Medieval Britain. Cambridge Studies in
Archaeology. Cambridge University Press, Cambridge.
Willis, W. H.
1992 Foraging Systems and Plant Cultivation During the Emergence of Agricultural
Economies in the Prehistoric American Southwest. In Transitions to Agriculture in
Prehistory, edited by A. B. Gebaur and T. D. Price. Prehistory Press, Madison
1995 Archaic Foraging and the Beginnings of Food Production in the American
Southwest. In Last Hunters, First Farmers: New Perspectives on the Prehistoric
Transition to Agriculture, edited by T. D. Price and A. B. Gebaur. School of American
Research, Santa Fe.
Zubrow, E., J. Fountain, A. Pelcin, K. Allen, P. Bush, K. Bartolotta and L. Hunt
1988 Prehistoric Distribution Networks of Domestic Ceramics in Sosogon, Phillipines:
An Archaeometric Study. In Ceramic Ecology Revisisted 1987: The Technology and
Socioeconomics of Pottery. , edited by C. Kolb, pp. 33-56. British Archaeological
Reports International Series. vol. 436(i). British Archaeological Reports, Oxford.
258