University of Alberta
Dental Disease in Roman Period Individuals from the Sodo and Terontola,
in the Territory of Cortona, Italy
by
Erin Lindsay Jessup
A thesis submitted to the Faculty of Graduate Studies and Research
in partial fulfillment of the requirements for the degree of
Master of Arts
Anthropology
©Erin Lindsay Jessup
Fall 2012
Edmonton, Alberta
Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis
and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is
converted to, or otherwise made available in digital form, the University of Alberta will advise potential users
of the thesis of these terms.
The author reserves all other publication and other rights in association with the copyright in the thesis and,
except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or
otherwise reproduced in any material form whatsoever without the author's prior written permission.
ABSTRACT
This study presents the results of a pathological examination of the dental
remains of nine individuals from the territory of Cortona (Tuscany), dated to the
Late Republican period (first century BC to first century AD). Eight of the
individuals were excavated from the area of the Sodo tumuli, Etruscan funerary
mounds dating to the sixth century BC, and a single individual from the nearby
modern city of Terontola. The sample was examined for caries, abscesses,
antemortem tooth loss, calculus, and enamel hypoplasia. Analysis reveals a high
prevalence of caries, consistent with an agrarian diet, and enamel hypoplasia,
indicative of poor health during childhood. Comparison of these data with those
from other studies on Etruscan and Roman populations suggests that although the
“typical” Mediterranean regimen of grains, legumes, and olives was consumed
throughout the region, significant differences existed between populations in
terms of the day-to-day diet.
ACKNOWLEDGEMENTS
I would like to thank Dott. Paolo Guillerini and the rest of the staff at the
Museo dell‟Accademia Etrusca e della Città di Cortona, as well as the
Soprintendenza per i Beni Archeologici della Toscana, for their assistance and
support of this project. I especially thank Dr. Helena Fracchia, for facilitating this
project from the very beginning and for her never-ending willingness to share her
extensive knowledge of the Val di Chiana.
I extend my deepest gratitude to my supervisor, Dr. Nancy Lovell, without
whose guidance, patience, and sense of humour this thesis would not exist. I am
also indebted to Dr. Lovell for the countless research, presentation, and travel
opportunities she has awarded me during my graduate and undergraduate studies.
I would like to thank Hillary Sparkes and Katie Waterhouse, not only for
their encouragement and advice, but for all of the lunches and coffee breaks that
made up such an invaluable part of my graduate school experience. I would also
like to thank Christopher White who, in addition to serving as a sounding board
and providing much-appreciated help with the figures that appear in this thesis,
ensured there was never a dull moment in TB-50.
Most importantly, I would like to thank my family. I thank my parents,
Lloyd and Kathy Jessup, and my sisters, Kelly and Lindsay Jessup, for their
unwavering support and love. And I thank my fiancé, Chase Gordon: for keeping
me fed and motivated, of course, but mostly for everything else.
Funding for this project was provided by the University of Alberta and the
Social Sciences and Humanities Research Council.
TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION ........................................................................... 1
WORKS CITED ............................................................................. 6
CHAPTER 2: THE SPATIAL, CULTURAL, AND HISTORICAL CONTEXT
OF THE VAL DI CHIANA .................................................................................... 8
INTRODUCTION .......................................................................... 8
ETRURIA ....................................................................................... 8
THE VAL DI CHIANA AND THE TERRITORY OF CORTONA
IN THE ETRUSCAN PERIOD .................................................... 11
ROMAN CONQUEST OF ETRURIA ......................................... 14
THE INHABITANTS OF ETRURIA BETWEEN THE FIRST
CENTURY BC AND THE FIRST CENTURY AD .................... 17
THE ROMAN VAL DI CHIANA AND THE ROMANIZATION
OF CORTONA ............................................................................. 20
CONCLUSION ............................................................................. 23
WORKS CITED ........................................................................... 26
CHAPTER 3: THE SITE AT SODO AND ITS SETTING ................................. 28
INTRODUCTION ........................................................................ 28
ETRUSCO-ROMAN BURIAL PRACTICES AND FUNERARY
STRUCTURES IN THE TERRITORY OF CORTONA ............. 28
THE TUMULI DEL SODO AND THE ROMAN PERIOD
BURIALS ..................................................................................... 31
Sodo Tumulus I ................................................................. 32
Sodo Tumulus II ............................................................... 34
The Circoli ........................................................................ 36
RE-USE OF PREHISTORIC BURIAL SITES IN THE ROMAN
WORLD AND POSSIBLE IMPLICATIONS FOR THE SITE AT
SODO ............................................................................................ 37
CONCLUSION ............................................................................. 44
WORKS CITED ........................................................................... 49
CHAPTER 4: DENTAL DATA ........................................................................... 52
INTRODUCTION ........................................................................ 52
SKELETAL MATERIALS .......................................................... 53
METHODS OF PALAEOPATHOLOGICAL IDENTIFICATION
AND ANALYSIS ......................................................................... 58
RESULTS ..................................................................................... 62
Individual Count ............................................................... 62
Tooth Count ...................................................................... 63
Caries .................................................................... 63
Antemortem Tooth Loss ....................................... 63
Enamel Hypoplasia ............................................... 64
Calculus................................................................. 64
DISCUSSION ............................................................................... 65
Caries ................................................................................ 67
Antemortem Tooth Loss ................................................... 69
Enamel Hypoplasia ........................................................... 70
Calculus............................................................................. 71
The Role of Dental Data in Understanding Diet ............... 73
Diet and Dental Health in the Population at Sodo ............ 75
CONCLUSION ............................................................................. 80
WORKS CITED ........................................................................... 90
CHAPTER 5: TERONTOLA ............................................................................... 95
INTRODUCTION ........................................................................ 95
BURIAL DETAILS ...................................................................... 95
DENTAL PATHOLOGY ............................................................. 96
TOOTH WEAR ANALYSIS ....................................................... 98
CONCLUSION ........................................................................... 101
WORKS CITED ......................................................................... 110
CHAPTER 6: CONCLUSIONS ......................................................................... 112
APPENDIX A - SUMMARY OF BURIAL INFORMATION .......................... 114
APPENDIX B - THE DIAGENESIS OF BONE IN GROUNDWATER .......... 115
INTRODUCTION ...................................................................... 115
GROUNDWATER ..................................................................... 117
STRUCTURE AND COMPOSITION OF BONES AND TEETH
..................................................................................................... 119
DIAGENESIS OF BONES AND TEETH IN GROUNDWATER
..................................................................................................... 123
LIMITING FACTORS CONTROLLING BONE DIAGENESIS
..................................................................................................... 126
Hydrology of the Burial Environment ............................ 127
Soil Composition and pH ................................................ 130
Surface Area and Porosity of Bone ................................. 132
CONCLUSION ........................................................................... 135
WORKS CITED ......................................................................... 136
APPENDIX C – SAMPLE DENTAL DATA COLLECTION FORM .............. 141
LIST OF TABLES
Table 4.1. Individual count comparison of dental disease frequencies in the Sodo
sample. .................................................................................................................. 82
Table 4.2. Tooth count prevalence of caries in the Sodo sample. ........................ 82
Table 4.3. Tooth count prevalence of antemortem tooth loss in the Sodo sample.
............................................................................................................................... 83
Table 4.4. Tooth count prevalence of enamel hypoplasia in the Sodo sample. .... 83
Table 4.5. Tooth count prevalence of calculus in the Sodo sample. ..................... 84
Table 5.1. Tooth count prevalence of caries in the individual from Terontola. . 102
Table 5.2. Tooth count prevalence of antemortem tooth loss in the individual from
Terontola. ............................................................................................................ 102
Table 5.3. Tooth count prevalence of calculus in the individual from Terontola.
............................................................................................................................. 103
Table 5.4. Tooth wear analysis in the individual from Terontola. ...................... 103
LIST OF FIGURES
Figure 2.1. Map of central Italy showing Pisa, Rome, and significant Etruscan
settlements. (Adapted from Barker and Rasmussen 1998, Fig. 4). ....................... 24
Figure 2.2. Val di Chiana in the territory of Cortona with important sites
indicated. The modern SS71/SR71 highway is indicated (arrows). (©2012
Google). ................................................................................................................ 25
Figure 3.1. A satellite image showing the location of the Sodo tumuli. The city of
Cortona is visible in the bottom right while the modern highway SS71/SR71 is
indicated by white arrows. (©2012 Google). ........................................................ 45
Figure 3.2. A plan of the Sodo tumuli and Circoli showing the locations of the
Roman period burials. (Adapted from a display plaque at the MAEC). ............... 45
Figure 3.3. Plan of Sodo Tumulus I showing the location of SI-1, SI-2, and SI-3 in
the dromos. (Adapted from Vallone 1995, Fig. 18). ............................................. 46
Figure 3.4. A drawing of Sodo tumulus II revealing the plans of Tomb 1, Tomb 2,
and the terrace altar. (Adapted from a display plaque at the MAEC). .................. 46
Figure 3.5. The reconstructed stone terrace altar at Sodo tumulus II. .................. 47
Figure 3.6. Plan of the tombs surrounding the terrace altar at Sodo II, showing the
locations of burials SII-2, SII-3, SII-4, SII-5, SII-8, and SII-9. The precise
location of SII-2 could not be determined from available records. (Adapted from a
display plaque at the MAEC). ............................................................................... 47
Figure 3.7. Plan of Circolo I showing the location of C-1, an intrusive inhumation
disrupting an earlier cremation burial. (Adapted from a display plaque at the
MAEC). ................................................................................................................. 48
Figure 3.8. Photo of C-1 in situ, during excavation. (Credit: MAEC). ................ 48
Figure 4.1. The cranial remains from SII-2 after cleaning. Remains were highly
fragmentary. .......................................................................................................... 85
Figure 4.2. A long bone fragment from SII-1. The fragile nature of the bone is
visible, as is the prolific infiltration by plant roots (arrows). ................................ 85
Figure 4.3. A long bone cross section from SI-3 under 2.5x magnification. The
invasive organic matter appears black. ................................................................. 86
Figure 4.4. The same long bone cross section from SI-3 under 20x magnification.
At this level, the globular appearance of the organic matter is visible, as is the
extent of the infiltration. ....................................................................................... 87
Figure 4.5. A gross carious lesion in the right first molar of SI-3. ....................... 88
Figure 4.6. Antemortem loss of the lower left first and second molars in SII-3.
The rounded cavity of the abscessed second molar is visible (arrow). ................. 88
Figure 4.7. Linear enamel hypoplasia in the left mandibular canine of SII-1
(arrows). ................................................................................................................ 89
Figure 4.8. Slight calculus on the lingual surface of the left mandibular teeth in
SI-3 (arrows). ........................................................................................................ 89
Figure 5.1. Map of Terontola showing the location of the burial under the current
carabinieri station. Also pictured is the location of the borgata rurale, a source of
several contemporaneous burials. (©2012 Google). ........................................... 104
Figure 5.2. T-1 as received at the MAEC, in extended supine position on two
terracotta roof tiles. ............................................................................................. 105
Figure 5.3. The cranium and mandible of T-1. The remains were highly
fragmentary. ........................................................................................................ 106
Figure 5.4. Left femur of T-1. The femoral shafts of T-1 were severed postmortem
and the medullary cavities infiltrated by mud (pictured: left). ........................... 106
Figure 5.5. Interproximal carious lesions in the upper right second and third
premolars (arrows). ............................................................................................. 107
Figure 5.6. Antemortem loss of the lower right second premolar and molars. Bone
remodelling is incomplete but well advanced, indicating that tooth loss occurred
significantly before death. ................................................................................... 107
Figure 5.7. Slight calculus on the lingual surface of the lower right lateral incisor
(arrow). Note also the minimal wear on the right incisors, canine, and first
premolar. ............................................................................................................. 108
Figure 5.8. Smith‟s system for scoring stages of tooth wear. Reprinted from Smith
(1984). ................................................................................................................. 108
Figure 5.9. Severe wear on the lower left second molar. .................................... 109
Figure 5.10. The wear on this upper right first molar is severe, scored as a 7 on
Smith‟s scale. Additionally, the tooth was worn obliquely, so that the chewing
surface extended past the cemento-enamel junction to include part of the root
(arrow)................................................................................................................. 109
1
CHAPTER 1: INTRODUCTION
The last centuries BC and first centuries AD saw the expansion of Roman
control across the Italian peninsula, and throughout Europe, the Near East and
North Africa. This was a period of cultural change during which many aspects of
Roman culture were adopted by local populations to serve specific ends. Other
aspects were rejected or were somehow altered or combined with existing cultural
strategies to create new, unique traditions (Wallace-Hadrill 2008: 10).
Conventionally, these processes have been obscured by the blanket term
„Romanization‟, which has increasingly been criticized for implying a passive
adoption of homogeneous traditions across a wide geographic area (Mattingly
2011, Terrenato 1998, Wallace-Hadrill 2008).
In an attempt to bypass these problems, Wallace-Hadrill (2008: 13)
suggests thinking about Roman expansion in terms of the coexistence of “multiple
identities” in the colonized regions. He suggests that a single individual could be
bilingual, embody a unique combination of Roman and indigenous characteristics,
and consider him- or herself to be both Roman and non-Roman. Additionally, a
number of authors have emphasized the importance of a regional approach, where
individual areas are examined for their own unique combination of Roman and
non-Roman cultural characteristics (Mattingly 2011, Terrenato 1998, Witcher
2006). In this paradigm, overarching models and generalizations are abandoned,
at least temporarily, for local explanations based on the archaeological and
historical evidence from a given area.
2
The territory of Cortona, located in northern inland Etruria‟s Val di
Chiana, is a prime candidate for a regional approach to Romanization.
Throughout their histories, Etruscan and Roman cultures impacted and influenced
one another through trade, warfare, and various other forms of contact (Barker
and Rasmussen 1998, Johnson 1996, Toynbee 1971). Although the diffusion of
cultural attributes between Etruria and Rome existed for centuries, this process
intensified and became decidedly uneven in the fourth century BC as the Romans
began to take control of the region. Between the fourth and second centuries BC
the various city-states of Etruria succumbed to Roman rule and were subjected to
varying amounts of castigation, dependent largely on the degree to which they
resisted Roman invasion (Barker and Rasmussen 1998, Harris 1971, Scullard
1967, Torelli 1986).
Scholars agree that, relative to many of the other Etruscan city-states, the
territory of Cortona had a nonviolent incorporation into the expanding Roman
Empire. The new relationship was predominantly characterized by the signing of
truces and the stability of the existing social hierarchy (Fracchia 2006, Harris
1971). Unfortunately, although great strides have been made in the last several
decades, many of the details of this process remain unknown. The degree to
which various aspects of the pre-existing Etruscan culture were kept, lost, or
transformed in the Roman period is unclear. Additionally, the movement of
people as well as ideas has complicated the issue, as scholars strive to interpret
the cultural and biological make-up of the region throughout this time period (H.
3
Fracchia, pers. comm.). More information is needed to truly understand how the
region shaped a place for itself after Roman conquest.
Nearly all of our knowledge of the Etruscans comes to us from the texts of
Greek and Roman writers, who began documenting the Etruscans long before,
and continued long after, their assimilation (Harris 1971, Spivey and Stoddart
1990). The Etruscans themselves are viewed as a „proto-historic‟ society by
historians: they developed writing some time before 700 BC but the surviving
texts are few and restricted in content (Barker and Rasmussen 1998).
Archaeological and other “on the ground” studies have been an invaluable source
of information on the Etruscans, particularly in the sphere of Etruscan tombs and
other monuments, material culture, and art (e.g. Dennis 1883, Izzet 2007, Oleson
1982, Spivey and Stoddart 1990, Zamarchi Grassi 1992). Etruscan skeletal studies
have traditionally focused on cranial metrics (e.g. Claassen and Wree 2004, Sergi
1900-01) and establishing the Etruscan typology, with an increasing emphasis on
variations within and between populations (e.g. Moggi-Cecchi et al. 1997, Rubini
et al. 1997). Recently, ancient DNA studies have proliferated, often with the goal
of determining the biological and geographic origin of the Etruscans (see Perkins
2009 for a review of recent Etruscan DNA studies).
The site at Sodo, in the territory of Cortona, consists of two Etruscan
funerary mounds dated to the sixth century BC. Throughout the sixth, fifth and
fourth centuries BC the tumuli were used and re-used for burial by the local elite
before they fell into disuse in the third century BC (Vallone 1995). In the Roman
period, the tumuli once again became a burial location when a number of
4
individuals were interred in and around the earthen mounds. These individuals
were identified as low status from the paucity of grave goods and the burial form,
a cappuccina, which was common for low status individuals (Barker and
Rasmussen 1998: 234, Toynbee 1971: 101). Fourteen individuals from the first
century BC to the first century AD were recently made available for study by the
Museo dell‟Accademia Etrusca e della Città di Cortona (MAEC) and eight of
these were examined for pathological conditions of the teeth. Additionally, the
first century BC to first century AD grave of one individual from Terontola,
located approximately 9 kilometres south-east of the Sodo tumuli, was made
available. This individual was also evaluated for dental disease. Limited skeletal
analysis was completed on the remains from both the Sodo and Terontola,
although data collection was hampered by the poor preservation of the remains.
The value of the Sodo and Terontola samples in contributing to a better
understanding of the Romanization of the territory of Cortona is clear. Not only
do they provide skeletal data from a region and time period of which there is very
little but they also represent a population that is both temporally and spatially
defined. Being such an obviously small portion of the local population, the
remains unfortunately do not provide an accurate sample for palaeodemography
studies. They do, however, provide an interesting glimpse into the inhabitants of a
specific territory in the Late Republican period and are valuable in attempts to
understand how the Etruscan Val di Chiana was reconstructed culturally,
biologically and physically in the Roman period.
5
The primary purpose of this thesis, then, is to contribute to the discussion
of the Romanization of the Val di Chiana by providing information on the
individuals inhabiting the territory of Cortona between the first century BC and
the first century AD. To achieve this goal, the specific research objectives are: (1)
to identify pathological lesions in the dentition of the skeletons from Sodo and
Terontola; (2) to explore these data in the context of Romanization of the Val di
Chiana, particularly with respect to diet and health.
This study will evaluate the dental data obtained from the Sodo and
Terontola in light of existing information on the Romanization of the region. As
such, Chapter 2 provides the cultural, historical and spatial background necessary
to interpret the skeletal material. Chapter 3 discusses the site at Sodo, chronicling
its use as a burial location and its excavation history, and provides a framework
for interpreting the re-use of burial spaces, examining similar instances
throughout the Roman world. Chapter 4 first outlines the preservation issues with
the Sodo sample and the materials and methods used to collect and evaluate data
from the skeletal remains. Chapter 4 then goes on to provide an examination of
the pathological conditions identified in the teeth of the Sodo sample, including
broad links to diet and behaviour. Chapter 5 examines the individual from
Terontola independently and evaluates his relevance to the study of the Sodo
material. Finally, Chapter 6 provides a summary and discusses conclusions.1
1 Elements of this thesis have already been published in a report submitted to the Museo
dell‟Accademia Etrusca e della Città di Cortona, and the Soprintendenza per i Beni Archeologici
della Toscana.
6
WORKS CITED
Barker G, Rasmussen T. 1998. The Etruscans. Oxford: Blackwell Publishers Ltd.
Claassen A, Wree H. 2004. The Etruscan skulls of the Rostock anatomical
collection – how do they compare with the skeletal findings of the first
thousand years BC.? Annals of Anatomy 186: 157–63.
Dennis G. 1883. The Cities and Cemeteries of Etruria vol. 2. 3rd
edition. London:
John Murray.
Fracchia H. 2006. The Villa at Ossaia and the Territory of Cortona in the Roman
Period. Siracusa: Lombardi editori.
Harris WV. 1971. Rome in Etruria and Umbria. Oxford: Clarendon Press.
Izzet V. 2007. The Archaeology of Etruscan Society. Cambridge: Cambridge
University Press.
Johnson MJ. 1996. The Mausoleum of Augustus: Etruscan and other influences
on its design. In: Hall JF, editor. Etruscan Italy. Provo (UT): Brigham
Young University, pp217-239.
Mattingly DJ. 2011. Imperialism, Power, and Identity: Experiencing the Roman
Empire. Princeton (NJ): Princeton University Press.
Moggi-Cecchi J, Pacciani E, Chiarelli B, D‟Amore G, Brown K. 1997. The
Anthropological Study of Etruscan Populations. Etruscan Studies 4(1): 73-
86.
Oleson JP. 1982. The Sources of Innovation in Later Etruscan Tomb Design (ca.
350-100 BC). Rome: Giorgio Bretschneider Editore.
Perkins P. 2009. DNA and Etruscan Identity. In: Swaddling J, Perkins P, editors.
Etruscan by Definition; The Cultural, Regional and Personal Identity of
the Etruscans. The British Museum Research Publications (173). London:
The British Museum Press, pp95-111.
Rubini M, Bonafede E, Mogliazza S, Moreschini L. 1997. Etruscan biology: The
Tarquinian population, seventh to second century BC (Southern Etruria,
Italy). International Journal of Osteoarchaeology 7(3): 202-211.
Scullard HH. 1967. The Etruscan Cities and Rome. Ithaca (NY): Cornell
University Press.
7
Sergi G. 1900-01. Studi di crani antichi. Atti della Società Romana di
Antropologia 7: 162-174.
Spivey N, Stoddart S. 1990. Etruscan Italy. London: B.T. Batsford Ltd.
Terrenato N. 1998. Tam Firmum Municipium: The Romanization of Volaterrae
and its cultural implications. The Journal of Roman Studies 88: 94-114.
Torelli M. 1986. History: Land and people. In: Bonfante L, editor. Etruscan Life
and Afterlife. Detroit (MI): Wayne State University Press, pp47-65.
Toynbee JMC. 1971. Death and Burial in the Roman World. London: Thames and
Hudson.
Vallone MT. 1995. Cortona of the principes. Etruscan Studies 2: 109-139.
Wallace-Hadrill A. 2008. Rome‟s Cultural Revolution. Cambridge: Cambridge
University Press.
Witcher R. 2006. Settlement and society in early imperial Etruria. The Journal of
Roman Studies 96: 88-123.
Zamarchi Grassi P, editor. 1992. La Cortona dei principes. Cortona: Calosci –
Cortona.
8
CHAPTER 2: THE SPATIAL, CULTURAL, AND
HISTORICAL CONTEXT OF THE VAL DI CHIANA
INTRODUCTION
This study examines available skeletal data from the Etruscan Sodo tumuli
and Terontola in order to contribute to a fuller picture of the „Romanization‟ of
the Italian Val di Chiana. Such an examination cannot be completed outside a
comprehensive, historical context. This chapter outlines the temporal, spatial and
cultural framework within which the analysis should be considered.
ETRURIA
Etruria was located in central Italy and was comprised of portions of what
are now Tuscany, Umbria and Latium. Etruria-proper was bordered by the
Tyrrhenian Sea to the west, the Arno River to the north, the Tiber River to the
south and east, and the Apennine Mountains to the north-east (Figure 2.1). Today,
the climate of Etruria is typically Mediterranean with summer temperatures
varying between 30 and 35°C and winter temperatures between 2 and 7°C. The
average yearly rainfall in the region varies between 700 and 1,000mm, with most
of the precipitation occurring in autumn and winter (Barker and Rasmussen 1998:
25).
Etruria‟s natural boundaries may have been the Arno and the Tiber, but
Etruscan expansion took the culture well beyond these limits: north into the Po
Valley and south into Campania (Nielsen 1984: 255, Torelli 1975). Moreover, the
Etruscan “Tarquin” dynasty ruled Rome for roughly a century until they were
9
expelled in 509 BC and the Etruscans have been credited with having a profound
influence on the development of the early Roman Republic (Barker and
Rasmussen 1998: 139-40).
The Etruscan culture first appeared in the seventh century BC, when the
Iron Age Villanovan traditions of central Italy gave way to the Orientalizing
period, heavily influenced by the Greek and Phoenician cultures (Ibid.: 1-6).1 The
term „Etruscan‟ refers to people associated with what is perceived to be a different
material culture than their Iron Age predecessors, although it is notable that the
inland cities of northern Etruria were not as subject to Eastern influence and
remained tied to their Villanovan traditions much more than the coastal and
southern cities (Torelli 1986).
Etruria was divided into a series of autonomous settlements.2 Initially
these settlements were monarchies, each ruled by its own king (Holder 1999: 33).
Later, these kings were superseded by aristocratic families, the principes, and
Etruria became a collective of oligarchical city-states. Despite civil and
diplomatic ties between the various sovereign territories, specifically the famous
League of the Twelve (Etruscan) Peoples, the city-states remained independent
1 There has been much debate in the literature as to whether this change involved the immigration
of new people, or whether it was a progression of material culture within a continuous occupation.
“Virtually all archaeologists now agree that the evidence is overwhelmingly in favour of the
„indigenous‟ theory of Etruscan origins: the development of Etruscan culture has to be understood
within an evolutionary sequence of social elaboration in Etruria…” (Barker and Rasmussen 1998:
44). 2 As with Etruria itself, Etruscan settlements and the individual properties within them were
usually delineated and defined by natural boundaries such as rivers, valleys and mountains.
Division of land was fundamental to the Etruscan world view and boundaries were viewed as
immutable; handed down from the gods (Edlund-Berry 2006, Jannot 2005).
10
and there is little evidence that they presented a clear, unified political force
(Harris 1971: 96).
Throughout Etruria social stratification intensified and a division is
discernible between the aristocracy and what Terrenato (1998) refers to as the
“dependent” class.3 The power of the principes over this dependent class extended
in part from their control of the prophets or haruspices. Religion permeated all
aspects of Etruscan life and this close relationship with divinity appears to have
reinforced the authority of the ruling class (Jannot 2005).
The Etruscans understood their civilization to be divided into a series of
periods of varying lengths called saecula, the details of which were predetermined
by the gods but unknown to the people. According to this system, the Etruscan
civilization began in approximately 970 BC with the beginning of the first
saeculum and would have a finite end with the close of the tenth saeculum (Harris
1971).
Historians mark the Etruscan Age as beginning slightly later and divide it
into the following periods (after Barker and Rasmussen 1998: 5-6):
700 BC-570 BC – the Orientalizing period; Eastern influences brought about
changes in material culture that coincided with the rise of
city-states and an increase in ostentation
570 BC-470 BC – the Archaic period; the height of Etruscan power and influence
3 While many labels, including “serfs” and “slaves” have been utilized in attempts to describe the
dependent class, the actual nature of the relationship is unclear and likely varied between
territories (Harris 1971, Terrenato 1998, Torelli 1986).
11
470 BC-300 BC – the Classical period, sometimes included within the Archaic
period; saw the gradual retreat of Etruscan power
300 BC-31 BC – the Hellenistic period; characterized by the Roman conquest and
acculturation of the Etruscan peoples
THE VAL DI CHIANA AND THE TERRITORY OF CORTONA IN THE
ETRUSCAN PERIOD
The Val di Chiana4 is a 1,367 square kilometre watershed between the
Tiber Valley and the Ombrone Valley (Alexander 1984: 528). Sitting at the
boundary between the „anti-Apennine‟ and the „pre-Apennine‟ hills and
intersecting with a number of routes to the Tiber Valley, the Val di Chiana is an
important transit corridor through northern inland Etruria (Figure 2.2). To the
west, the sandstone, clay and limestone anti-Apennines reach several hundred
metres above sea level (asl). Beginning in the Val di Chiana and continuing to the
east, the predominantly limestone pre-Apennines rise irregularly from the valley
floor (Barker and Rasmussen 1998: 16-19). Throughout the Val di Chiana these
hills separate a western string of hot springs from the eastern cold springs
(Fracchia 2006: 20).
Today, the valley floor descends from 260 metres asl at Chiusi to 243
metres asl at Arezzo and the mountainous boundary ranges from altitudes of 600
metres to 1,148 metres asl (Alexander 1984: 528). Once a shallow lake as far as
Arezzo, the Val di Chiana was drained by the Etruscans in the late seventh or
early sixth century BC so they could take advantage of the highly fertile alluvial
4 Also called the Valdichiana and the Chiana, or Clanis, River Valley
12
deposits (Fracchia 2006: 20). Despite the slow moving waters of the Chiana River
and the subsequent siltation issues, the Val di Chiana possessed “an agricultural
potential unequalled in central Italy” (Vallone 1995: 109, for siltation see
Alexander 1984: 531-2). Even by the fourth century BC, Chiana grain was
exported as far away as Rome, where it was prized for its high gluten content and
enormous yields (Fracchia, pers. comm.).
In the Etruscan period, the Val di Chiana would have likely supported
deciduous forest, with a variety of trees including willow, alder, oak, ash, and elm
extending up the slopes to about 1,000 metres asl (Barker and Rasmussen 1998:
38). Fracchia (1996) argues that the archaeological evidence suggests that the
lower slopes of the south-eastern Val di Chiana were densely populated in this
period, in the form of small settlements and isolated establishments. Due largely
to farming and herding practices, the original tree cover gradually opened up so
that, by the Roman period, the valley was a predominantly open landscape
(Barker and Rasmussen 1998: 40).
The city of Cortona was one of twelve major Etruscan centres located
between the Arno and the Tiber, and one of three (along with Arezzo and Chiusi)
found within the Val di Chiana (Ibid.: 143).5 The city sits in a saddle on Mt.
Sant‟Egidio, between 450 metres and 550 metres asl, while the sixteenth century
fortress of Girifalco sits on the summit at 650 metres asl (Cataldi and Lavagnino
1990: 49, Holder 1999: 20). Continuous occupation of Cortona into modern times
5 While Cortona was certainly a dodecapolis (one of the twelve ruling city states of Etruria) by the
third century BC, there is some disagreement as to whether it was a founding member of the
original Etruscan league (cf. Fracchia 1996: 198, Holder 1999: 33, Torelli 1975)
13
has obliterated most traces of the Etruscan and Roman settlement, but it is
generally agreed to have most likely run down the western slope of the hill, in a
similar manner to the modern settlement (Scullard 1967: 156-7).
Below the city, the slopes of Mt. Sant‟Egidio provided forage for grazing
animals, workable sandstone and timber for building, and an ample supply of
copper, iron and other minerals (Holder 1999: 21). In the Orientalizing and
Archaic periods particularly, Cortona was a centre for decorative bronzework and
other forms of metallurgy (Boitani et al. 1975: 50, Holder 1999: 30). The valleys
around Cortona were also home to naturally occurring clay beds that provided the
material for brick, tile and other ceramics (Fracchia 2006: 20; Holder 1999: 22).
The wealth of the city, however, was primarily agricultural: the floor of the Val di
Chiana provided extensive, fertile agricultural land and Etruscan crop husbandry
involved the expanded cultivation of a wide range of cereals, legumes and tree
crops (Barker and Rasmussen 1998: 182-184, Fracchia 2006: 12).
Animal husbandry in the Etruscan era included animals such as horses,
donkeys, chickens, cattle, sheep, goats, pigs, and dogs, and the period saw an
increased emphasis on producing secondary products from these domesticates,
such as cheese, milk, and wool (Barker and Rasmussen 1998: 185). While hunting
was not a vital subsistence practice, it may have been an important contributor to
the elite diet, and the detection of wild foods such as fruit, nuts and large and
small game at archaeological sites suggests local use of seasonal resources to
compliment crops and domesticates (Ibid.: 200). Forniciari and Mallegni‟s (1987)
study of strontium and zinc levels in two Etruscan populations indicate that
14
generally, the diet came to rely increasingly on vegetable and plant-related foods
and less on animal derived protein.
The wealth of goods and foodstuffs in the Val di Chiana stemmed not just
from local resources but also from the valley‟s location in a trade network that
extended across the Italian peninsula and beyond. Through the numerous routes
that passed through the valley, the territory of Cortona was well-connected to the
upper Tiber Valley to the east, the Ombrone River Valley to the west and, by
extension, the Adriatic and Tyrrhenian coasts (Fracchia 2006: 21-2). Additionally,
at the centre of a triangle formed by Arezzo, Perugia, and Chiusi, Cortona is less
than a day‟s journey on horseback from all three (Holder 1999: 23, see also
Cataldi and Lavagnino 1990: 49). This proximity was crucial, as economic
exchange at the local and regional scale was integral to the Etruscan economy
(Barker and Rasmussen 1998: 210-2).
ROMAN CONQUEST OF ETRURIA
Roman expansion across the lands of Etruria was lengthy process, marked
by a series of major battles, small skirmishes and discrete treaties that bound
Rome to the individual city-states one at a time. While the bulk of the fighting
took place in the fourth and third centuries BC, total „conquest‟ was not truly
completed until several centuries later. But as the increasing pressure of Roman
military and political forces became inescapable, the Etruscan cities ultimately
succumbed and subsequently struggled to adapt to their new reality.
15
The Romans established a frontier in Southern Etruria by the early fourth
century BC with the fall of Veii in 396. The town‟s territory was subsequently
divided among Roman settlers and the Veientane survivors were likely enslaved.
The defeat of Veii provided the Romans with a foothold in Etruria and left the
land beyond exposed to Roman advancement. After this initial establishment of
Roman settlers in the South and the subsequent establishment of Latin colonies
nearby at Sutri and Nepi, the Roman conquest of Etruria progressed northward.
The final series of Etruscan wars with Rome began in 311 BC and included most
of the important towns of northern inland Etruria, including Arezzo, Chiusi,
Cortona, Perugia, Volterra, and Volsinii (Harris 1971: 47, 96).
In 310 BC indutiae, or truces, of thirty years were signed between Rome
and Perugia, Arezzo, and Cortona (Harris 1971: 54, Scullard 1967: 273).
Cortona‟s initial status (i.e. municipium, colonia or praefectum) after the truce is
unknown and, as a result, its level of self-governance and submission to Rome is
unclear. Although, through the conditions of the truce, life in Cortona was to
continue relatively uninterrupted, the city would have certainly been forced to pay
taxes to the Roman government and supply soldiers to the Roman army (Holder
1999: 43-5). The resulting depleted labour force, in addition to the loss of capital,
would have had serious repercussions on the local economy and, potentially,
social structure. Like other city-states, however, Cortona would have received
little by way of economic or political benefits in return (Harris 1977: 58).
The Social War of 91-88 BC was relatively uneventful for Cortona, as it
remained allied with Rome against the other tribes of the Italian peninsula. While
16
the general Sulla‟s passage through the Val di Chiana wreaked havoc on the
territories of Chiusi and Arezzo, Cortona appears to have survived unscathed
(Fracchia 2006: 34). By the end of 89 BC, the last of the Etruscan families
received Roman citizenship (Harris 1971: 231). The Etruscans were also
registered into Roman tribes for voting and bureaucratic purposes and Cortona
was registered into the tribe Stellatina, along with Tarquinia, Tuscania, Nepi and
others (Barker and Rasmussen 1998: 292, Harris 1971: 244, Torelli 1986).
With the Roman civil wars between Marius and Sulla, beginning in the
second decade of the first century BC, the Etruscans‟ recent good fortune
floundered. Almost unanimously, the city-states of Etruria sided with Marius and,
subsequently, were punished by the victorious Sulla (Barker and Rasmussen
1998: 266, Harris 1971: 251, Torelli 1986). Centres of resistance to Sulla included
Chiusi, Arezzo, Populonia and Volterra; fierce battles were fought at all of these
places (Torelli 1986). After the defeat of Marius, Sulla severely limited the
privileges enjoyed by the Etruscan people and seized huge tracts of land to
penalize them for their disloyalty. Particularly in the territories of Fiesole, Arezzo,
Volterra and Chiusi, land was confiscated and redistributed among Sulla‟s
veterans as reward for active service. These veteran colonies had a devastating
and demoralizing effect on the people of northern inland Etruria. Many, now
landless, emigrated to Africa and Spain; those that stayed behind never fully
recovered from the aggressive and devastating invasion (Barker and Rasmussen
1998: 266, Torelli 1986).
17
THE INHABITANTS OF ETRURIA BETWEEN THE FIRST CENTURY
BC AND THE FIRST CENTURY AD
Although the separation of „Etruscan‟ and „Roman‟ becomes an
increasingly problematic distinction to maintain into the second and first centuries
BC (e.g. Fracchia 2006: 14, Holder 1999: 46), there remains a rationale for
attempting to determine the ancestry, or ancestries, of the inhabitants of a given
area. „Romanization‟, the process of acculturation in which local language,
religion, dress and other cultural aspects are replaced by Roman traits, has been
criticized in recent years for the implication of Roman cultural dominance over
conquered peoples and for conjuring images of uniformity between the Roman
provinces (e.g. Wallace-Hadrill 2008: 9-14). For the purposes of this paper,
„Romanization‟ will be employed as defined by Harris (1971: 147): “the process
by which the inhabitants came to be and to think of themselves as Romans.” It is
important to bear in mind, however, that the population of Etruria was
heterogeneous and the degree to which the different inhabitants self-identified as
Roman would have varied across families, territories and social statuses.
Determining the biological and cultural ancestries of the inhabitants of Etruria can
help provide clues as to the true composition of „Roman Etruria‟.
In the third and second centuries BC Roman colonization of Etruria was
limited and despite several Latin colonies in the South, the majority of Etruria had
yet to be occupied by Roman colonists (Harris 1971: 97). The Romans likely
chose not to colonize extensively in Etruria because they were able to ensure the
security of their interests through the preservation of the existing social structure
and the local principes (Harris 1971: 202, Torelli 1986). With Etruscan society
18
sharply divided between the aristocratic principes and the dependent class, Rome
was able to enlist the help of the former against the latter: providing a method of
maintaining security that made colonization largely unnecessary. As such, despite
their numerous military victories the Romans founded no Latin colonies in
northern Etruria before the first century BC (Barker and Rasmussen 1998: 266).
Roman colonies in the South were more prolific although initially they were
predominantly military bases, essential for further expansion, rather than civil
colonies (Harris 1971).
Harris (1977: 56) argues that despite a lack of evidence, it is probable that
Rome confiscated land from the territories of Perugia, Chiusi, Arezzo, Cortona,
Roselle, and Volterra and converted it into ager publicus for settlement and
agriculture by 270 BC. How much of this land actually accommodated Roman
settlers as opposed to simply providing revenue for the Roman treasury is
uncertain, however, as is whether such seizures of land affected all of these city-
states. Furthermore, recent archaeological evidence from the territory of Cortona
suggests that the distribution of Etruscan villages along the valley floor remained
unchanged from the sixth century BC to the first century AD (H. Fracchia, pers.
comm.).
Latin settlement of the territories of Arezzo and Tarquinia certainly
occurred between 133 and 123 BC, with the land reform laws of the Gracchi
(Torelli 1986).6 Less than half a century later, the Sullan colonies at Fiesole,
6 Also associated with the establishment of Roman colonies is the relocation of the various Italic
peoples. As part of their land reorganization strategies, the Romans habitually forced Etruscans,
19
Arezzo, Chiusi and Volterra significantly increased the number of Roman settlers
in northern inland Etruria (Barker and Rasmussen 1998: 266, Harris 1971: 261-3).
Cortona has been put forward as another potential Sullan colony site but there is
no clear evidence to support this claim and Harris (1971: 264 n.9) suggests that
this assertion is due to the confusion of Cortona with Croton, the Hellenistic
settlement in modern Calabria.
In the last decades leading up to the turn of the millennium, Arezzo again
saw the establishment of Roman colonies, along with Perugia, Siena and several
other city-states (Harris 1971: 306-10, Torelli 1986). Unlike Sulla‟s veteran
colonies, however, these new colonies contributed to the “economic and social
stabilization and reconciliation” of these areas (Torelli 1986: 62). This
“stabilization and reconciliation” was, in Perugia, undoubtedly hampered by the
violent nature of the settlement: in addition to redistributing much of the land
surrounding the city, Octavian‟s orders called for the slaughter of most of the
wealthy Perugini (Harris 1971: 309, Torelli 1986). Regardless, the Augustan
settlements are attributed with accomplishing a great deal in terms of the
„Romanization‟ of Etruria (Harris 1971: 313-4).
By the first century BC, then, there were a number of Roman colonies in
northern inland Etruria, specifically in the territories adjacent to Cortona. Yet
none of the evidence argues for a significant number of Roman immigrants within
the territory of Cortona itself despite the specific mention of other city-states,
including nearby Arezzo. One can hardly be expected to believe that despite all
Umbrians, Lucanians and other Italic tribes to move around the peninsula (H. Fracchia, pers.
comm.).
20
types of contact between Cortona and the surrounding areas, as well as Cortona‟s
military defeat and subsequent submissiveness to Rome, there was a complete
absence of Latin colonists in her territories. The evidence seems, however, to
indicate that Cortona was not overrun with immigrants in the same way as many
of the other Etruscan city-states.
THE ROMAN VAL DI CHIANA AND THE ROMANIZATION OF
CORTONA
Throughout the Late Republican and Early Imperial periods, the cities of
Southern Etruria began to fade in importance while the cities of the North,
including Cortona, remained vital economic power centres (Bonfante 1986: 11).
A major contributor to the survival of the Northern cities was the Roman road
system, which came to consist of four principal roads in Etruria: the Via Aurelia
which ran parallel to the Tyrrhenian coast, the Via Flaminia which ran across
Southern Etruria and Umbria to the Adriatic coast, and the Vie Clodia and Cassia,
both of which ran through central Etruria (Harris 1971: 161). The Via Cassia,
dated to the first half of the second century BC, ran through the Val di Chiana
past the base of Mt. Sant‟Egidio, following the route of the earlier Etruscan road
almost exactly as it linked Cortona to Arezzo, Chiusi and beyond (Harris 1971:
161-8). Such proximity to the principal overland route through Etruria surely
facilitated the exportation of grain and other commodities from the lands around
Cortona (Barker and Rasmussen 1998: 23, Harris 1971: 161-8, Harris 1977: 57).
Fracchia (2006: 30) describes the third and second centuries BC as
belonging to a “general trend of architectural monumentalization and spatial
21
definition” as Cortona vied for position in the new Roman reality. Although
construction within the walls would have been limited by the existing city,
Cortona certainly saw the erection of Roman baths. The presence of other typical
hallmarks of a Roman town, such as a theatre and basilica, is also likely, though
no traces remain (Fracchia 2006: 34-5, Holder 1999: 45, Torelli 1986). Through
the adoption of such significant Roman social institutions, the principes of
Cortona displayed a willingness to Romanize in the public sphere and this „self-
Romanization‟ guaranteed minimum social upheaval (Fracchia 2006: 35).
Consequently, the traditional class system persisted in northern inland Etruria and,
despite a severe crisis in the first half of the second century BC, endured into the
first century BC (Harris 1971: 202-12, Torelli 1986)7.
In the territory around Cortona, a number of small rural holdings,
associated with a rapid population expansion, began to appear in close proximity
throughout the valley in the early third century BC. After the Battle of Trasimeno
and the associated devastation of 217 BC, the entire countryside around Cortona
was deserted, although by the late third and early second centuries BC there was
intense repopulation of specific areas, with concentrations detectable in the
vicinity of the modern municipalities of Camucia and Terontola (Fracchia 2006:
11). While there is evidence for the concentration of land (and, resultantly, power)
at the turn of the second century BC, in general the rural settlement pattern in the
Val di Chiana appears to have remained largely unchanged (Fracchia 1996,
7 The system was not without changes, however, and the dependent class was able to partially
integrate into the Etruscan political system. Throughout the second century BC, a number of
subordinate farmers came to own and work small tracts of land around Perugia, Chiusi, Volterra
and Arezzo (Torelli 1986).
22
Fracchia 2006: 25-7). Terrenato (1998) has argued convincingly that in Volterra,
the endurance of settlement pattern was related, at least in part, to the interwoven
nature of social order, landholding practices, religion and ideology. Whatever the
reason, evidence for a number of villas and rural hamlets has been discovered
throughout the valley, indicating a continuation of Etruscan habitation sites
(Fracchia 1996).8
The late second century and early first century also saw the function of
locations that had been important to the Etruscan population “transformed” to
have a different importance in the new Roman reality (Fracchia 2006: 14).
Etruscan shrines and springs associated with healing waters became sites for
Roman baths and, throughout the valley, a number of villas have been identified
on the sites of Etruscan tombs, sanctuaries, or other “cult” places (Fracchia 2006).
This was a period of true absorption: Latinization, followed subsequently by
Romanization, occurred to the extent that by the third decade of the first century
BC, “Etruria no longer existed as a separate entity” (Barker and Rasmussen 1998:
292, see also Bruun 1975: 496, Harris 1971: 318).
Throughout this period, however, there is evidence that many of the
powerful Etruscan families continued to own their land and to fill politically
important positions in the running of Cortona. At the villa at Ossaia, located
approximately halfway between Camucia and Terontola on the Roman Via Cassia
and dated to the late second or early first century BC, the earliest owners were of
8 Although Witcher (2006) notes that the lack of large-scale and systematic surface surveys limits
our ability to draw conclusions about the settlement pattern in the Val di Chiana in the Early
Imperial period.
23
Etruscan descent, though they displayed significant „Romanization‟ in terms of
language and fashions (Fracchia 2006). The famous first century BC
L’Arringatore, or The Orator, statue found near the shores of Lake Trasimeno
depicts Aulus Metellus: a member of a wealthy Etruscan family. This statue
displays an interesting combination of Roman and Etruscan characteristics, as it
portrays an individual from an Etruscan family in a Roman-appointed government
position, in Roman dress with the Latinized version of his name inscribed on his
toga in Etruscan (H. Fracchia, pers. comm.). This continuity of people and place,
with the gradual adoption of Roman attributes, characterizes the Romanization of
the Val di Chiana as a whole and the territory of Cortona in particular.
CONCLUSION
Analysis of the inhumations from the Sodo tumuli and Terontola is only
valid within a thorough understanding of the cultural context presented above.
With the necessary background information established, it is now possible to
move forward into an examination of the archaeological context of the burials
themselves.
24
Figure 2.1. Map of central Italy showing Pisa, Rome, and significant Etruscan
settlements. (Adapted from Barker and Rasmussen 1998, Fig. 4).
25
Fig
ure
2.2
. V
al d
i C
hia
na
in t
he
terr
itory
of
Cort
ona
wit
h i
mport
ant
site
s in
dic
ated
. T
he
moder
n S
S71/S
R71 h
ighw
ay i
s
indic
ated
(ar
row
s). (©
2012 G
oogle
).
26
WORKS CITED
Alexander D. 1984. The reclamation of Val-di-Chiana (Tuscany). Annals of the
Association of American Geographers 74(4): 527-550.
Banti L. 1973. Etruscan Cities and their Culture. Bizzarri E, translator. Berkeley
(CA): University of California Press.
Barker G, Rasmussen T. 1998. The Etruscans. Oxford: Blackwell Publishers Ltd.
Boitani F, Cataldi M, Pasquinucci M. 1975. Etruscan Cities. Atthill C, Evans E,
Pleasance S, Swinglehurst P, translators. London: Cassell.
Bonfante L. 1986. Introduction/Etruscan studies today. In: Bonfante L, editor.
Etruscan Life and Afterlife. Detroit (MI): Wayne State University Press,
pp1-17.
Bruun P. 1975. Conclusion: The Roman census and Romanization under
Augustus. In: Bruun P, Hohti P, Kaimio J, Michelsen E, Nielsen M,
Ruoff-Väänänen E. Studies in the Romanization of Etruria. Roma: Bardi
Editore, pp435-505.
Cataldi G, Lavagnino E. 1990. La pianificazione antica della Valdichiana: un
piano da venticinque secoli. In: Cataldi G, Cherici A, Gialluca B,
Lavagnino E, Maffei G, Orgera V, Vaccaro P, editors. Cortona: Struttura e
Storia, pp33-138.
Edlund-Berry, IEM. 2006. Ritual space and boundaries in Etruscan religion. In:
de Grummond NT, Simon E, editors. The Religion of the Etruscans.
Austin (TX): University of Texas Press, pp116-131.
Forniciari G, Mallegni F. 1987. Indagini paleonutrizionali su campioni di
popolazioni a cultura etrusca. In: Barbieri, G, editor. L‟alimentazione nel
mondo antico: gli etruschi. Rome: Istituto Poligrafico e Zecca dello Stato,
pp135-139.
Fracchia H. 1996. Etruscan and Roman Cortona: New evidence from the
southeastern Val di Chiana. In: Hall JF, editor. Etruscan Italy. Provo (UT):
Brigham Young University, pp191-215.
Fracchia H. 2006. The Villa at Ossaia and the Territory of Cortona in the Roman
Period. Siracusa: Lombardi editori.
Harris WV. 1971. Rome in Etruria and Umbria. Oxford: Clarendon Press.
27
Harris WV. 1977. Economic conditions in northern Etruria in the second century
BC. In: Cristofani M, Martelli M, editors. Caratteri dell‟Ellenismo nelle
Urne Etrusche. Florence: Centro Di Firenze, pp56-63.
Holder PN. 1999. Cortona in Context: The History and Architecture of an Italian
Hill Town to the 17th
Century. Cortona: Arti Tipografiche Toscane.
Jannot JR. 2005. Religion in Ancient Etruria. Whitehead J, translator. Madison
(WI): University of Wisconsin Press.
Nielsen E. 1984. Some observations on early Etruria. In: Hackens T, Holloway
ND, Holloway RR, editors. Crossroads of the Mediterranean.
Archaeologica Transatlantica 2. Louvain-La-Neuve, Belgium: Université
Catholique de Louvain/Providence (RI): Brown University, pp255-260.
Scullard HH. 1967. The Etruscan Cities and Rome. Ithaca (NY): Cornell
University Press.
Spivey N, Stoddart S. 1990. Etruscan Italy. London: B.T. Batsford Ltd.
Terrenato N. 1998. Tam Firmum Municipium: The Romanization of Volaterrae
and its cultural implications. The Journal of Roman Studies 88: 94-114.
Torelli M. 1975. Introduction. In: Boitani F, Cataldi M, Pasquinucci M. Etruscan
Cities. Atthill C, Evans E, Pleasance S, Swinglehurst P, translators.
London: Cassell, pp9-28.
Torelli M. 1986. History: Land and people. In: Bonfante L, editor. Etruscan Life
and Afterlife. Detroit (MI): Wayne State University Press, pp47-65.
Vallone MT. 1995. Cortona of the principes. Etruscan Studies 2: 109-139.
Wallace-Hadrill A. 2008. Rome‟s Cultural Revolution. Cambridge: Cambridge
University Press.
Witcher R. 2006. Settlement and society in early imperial Etruria. The Journal of
Roman Studies 96: 88-123.
28
CHAPTER 3: THE SITE AT SODO AND ITS SETTING
INTRODUCTION
This chapter provides a description of: 1) Etrusco-Roman burial practices;
2) funerary structures in the territory of Cortona, including the Sodo tumuli; 3) the
Roman period burials from the Sodo; and 4) the literature exploring the locations
and rationalizations of Roman-period burials in association with prehistoric
funerary structures.
ETRUSCO-ROMAN BURIAL PRACTICES AND FUNERARY
STRUCTURES IN THE TERRITORY OF CORTONA
Both the Etruscans and the Romans varied between cremation and
inhumation throughout their long histories, with different styles gaining and
losing popularity at different times (Davies 1977). In Rome, both customs were
practiced side by side until the beginning of the fourth century BC, when
cremation became the standard and remained so until the second century AD
(Davies 1977, Morris 1992: 41, Toynbee 1971: 39-40). Cremation was gradually
adopted by the Etruscans throughout the Late Bronze Age and Iron Age, although
it did not become prevalent until the fourth to second centuries BC. Even then,
cremation was most common in central and northern Etruria while inhumation
remained more frequent in the South and along the coast (Barker and Rasmussen
1998: 58, Bietti Sestieri 1984: 57, Toynbee 1971: 15). Barker and Rasmussen
(1998: 122) suggest that in some cases, the Etruscans may have cremated the
wealthy “as a mark of special status” but this practice was certainly not restricted
29
to the upper class, as cremation burials of low-status individuals have been found
throughout Etruria.
In the seventh and sixth centuries BC, the custom among the principes
changed and increasingly elaborate chamber tombs began to appear throughout
Etruria. The chamber-tombs varied according to terrain: some were rock-cut,
others were built out of blocks and many were a combination of the two
techniques (Barker and Rasmussen 1998: 120, Toynbee 1971: 18). Within these
chamber-tombs the aristocratic dead were housed: either interred in sarcophagi or
cremated and placed in urns. The Etruscan tumulus, with its circular retaining
wall and earthen mound, is generally considered to be the source of inspiration
behind the Roman mounds of similar design, including the Mausoleum of
Augustus (Johnson 1996, Toynbee 1971, but cf. Holloway 1966). Around the
tombs, it was common to bury the remains of poorer individuals. Cremations were
often buried a pozzo, in cylindrical wells cut or dug into the earth, or a casetta, in
stone-lined trenches. Inhumations were generally in simple fossae, trench graves,
or free-standing stone sarcophagi (Barker and Rasmussen 1998: 234, Toynbee
1971: 18). All burials, whether cremations or inhumations, were generally
accompanied by pottery, metalware and other grave goods (Barker and
Rasmussen 1998: 239).
In the seventh century BC, the elaborate chamber-tombs that had become
common elsewhere in Etruria spread to the North (Torelli 1975: 18). While in the
southern territories a large number of elite families were burying their dead in the
necropoles, in Cortona the political dominance of a few powerful families
30
remained fixed, restricting the number of sixth century tumuli (Barker and
Rasmussen 1998: 131, Spivey and Stoddart 1990: 143). Initially, the use of each
tumulus was limited to the powerful family that ordered its construction (Holder
1999: 40). The tumuli were re-used over the centuries, however, and burials from
the subsequent phases of use cannot be assumed to belong to the same lineage.
The earliest of these tumuli in the Val di Chiana, the Tumulus of Camucia,
was built around 600 BC, followed by the two Sodo tumuli, built after 580 BC
(Vallone 1995). These structures, referred to as meloni (melons) in the local idiom
because of their domed appearance, are essentially burial chambers (tombs)
surrounded by stone drums and covered in earthen mounds that closely resemble
other Etruscan circular tumuli, such as those at Populonia and Cerveteri. A single
tumulus can contain more than one tomb: the Tumulus of Camucia, for example,
surrounds Tomb A and Tomb B, both of which contain funerary gifts datable to
between the end of the seventh and the beginning of the fourth centuries BC
(Vallone 1995).
Banti (1973: 172) has argued that the Sodo and Camucia tumuli are
located too far from Cortona to belong to the city‟s elite and that they are more
likely the tombs of rich valley landowners. Indeed there is a great deal of debate
in the literature as to the date of Cortona‟s emergence as an Etruscan „urban
centre‟. Some researchers (e.g. Bruschetti 1992: 183, Fracchia 1996: 194) argue
that the locations of the tumuli suggest the presence of small habitation sites
nearby. It is possible that the Val di Chiana was populated primarily by such sites
scattered across the valley floor and the lower slopes of the surrounding hills
31
rather than by a large settlement on Mt. Sant‟Egidio. What is relevant for the
purposes of this study, however, is that the ruling principes of the Val di Chiana,
wherever they were settled, built the enormous tumuli “to bury their dead in the
trappings of economic and social power” (Vallone 1995: 109).
In the third and second centuries BC, a number of new monumental tombs
were constructed in the territory of Cortona. Several of these tombs, such as the
Tanella Angori and the Tanella di Pitagora are closer to Cortona than the sixth
century tumuli and Fracchia (2006: 27) suggests that their positions on the main
roads to the city underscore the increasing importance of „urban identity‟ in the
Val di Chiana. The Tanella di Pitagora, now missing its protective mound, is the
best preserved of these later tombs and contains ten niches designed to
accommodate cinerary urns (Barker and Rasmussen 1998: 284, Oleson 1982: 87).
THE TUMULI DEL SODO AND THE ROMAN PERIOD BURIALS
The Sodo tumuli are located slightly over 2 kilometres west of Cortona,
just across the modern highway SS71/SR71, which corresponds to the Roman Via
Cassia and the earlier Etruscan road (Fracchia 2006: 24, Vallone 1995). The two
structures sit approximately 300 metres apart on either side of the Loreto stream
(Figures 3.1 and 3.2). In view of the Etruscan predilection for natural boundaries,
researchers have suggested that the corresponding ancient stream could have
served as a border between properties. If the tumuli were indeed located along
such a boundary, their original purpose can be understood, in part, as
32
monumentum: conspicuous symbols of the power and dominance of the
landholding families, as well as tombs to shelter the dead (Vallone 1995).1
Sodo Tumulus I
Tumulus I, located on the south bank of the Loreto, was built in the first
half of the sixth century BC, although possibly as late as 550 BC (Vallone 1995).
After falling into disuse, the tomb was utilized as a source of building materials
and sacked for valuable grave goods (Ibid.). The mound had been discovered by
1909, but “secure reports on the discovery, on the dates and on the methods of the
discovery, on what the grave goods consisted of and what happened to them” are
lacking (Ibid.: 119). The artificial mound is approximately 10 metres high and
over 50 metres in diameter (Holder 1999: 36, Vallone 1995). The tumulus
contains a single five-chambered tomb with a southeast-northwest floor plan
(Figure 3.3). The covered dromos leads to three antechambers, before terminating
in a larger, rear chamber. A total of four inter-connecting lateral rooms project off
the second and third antechambers. The pavement of the chamber-tomb is
comprised of local grey sandstone, while the wall structures are made of the
yellowish sandstone of the Val di Chiana (Vallone 1995).
The date of the first phase of Tumulus I indicates that the property owner
was one of the local principes (Ibid.). Re-use of the tomb in the Hellenistic Age is
documented by inscriptions and grave goods datable to the fourth century BC
(Ibid.). Specifically, an inscription cut into the main door of one of the lateral
1 The use of tumuli to indicate and reinforce control of a territory by family groups elsewhere in
Etruria has been widely discussed (e.g. Riva 2010, Zifferero 1991).
33
chambers indicates that the tomb was re-occupied by Arnt Mefanate, of Umbrian
origin, and his local wife Velia Hapisnei.
In the Late Republican and Early Imperial eras, Tumulus I was again
reused for burials. A number of burials dating from the first century BC to the
first century AD were excavated from within the drum of the tumulus, although
none were found in the tomb chambers themselves. The Roman period burials
were identified as low status individuals based on the paucity of grave goods,
which consisted exclusively of simple artefacts: commonplace items of everyday
use (Ibid.). The form of the burials, the a cappuccina tomb type, also indicates
individuals of lower status. This unpretentious design, in which the deceased is
laid out under a structure of gabled roof tiles, had become the norm for low-status
individuals throughout Roman Etruria. The design owed its popularity to the
inexpensive and uncomplicated nature of its construction (Barker and Rasmussen
1998: 234, Toynbee 1971: 101).
The precise number of Roman period burials at Tumulus I is unclear, as is
the exact provenance of the interments, but three of the burials: SI-1, SI-2, and SI-
3 were recently made available for study by the Museo dell‟Accademia Etrusca e
della Città di Cortona (MAEC)2. All three of the burials in question were found in
the dromos of the tomb.
2 Burial numbers were assigned by the author to facilitate data organization; they are not the
original burial numbers used by the excavators. For detailed burial information see Appendix A.
34
Sodo Tumulus II
Tumulus II is located on the north bank of the Loreto and is datable to
between 575 and 550 BC (Vallone 1995). While the construction of Tumulus I at
Sodo was more refined than that of Tumulus II, the second mound is larger in
size, measuring 64 metres in diameter. Like the Tumulus of Camucia, Tumulus II
contains two discrete tombs: Tomb 1, which was excavated by Antonio Minto in
1928 and 1929, and Tomb 2, which remained undiscovered until 1992 (Figure
3.4). Research at the tumulus has been hampered by the sacking of both tombs in
antiquity, as well as the infiltration of groundwater into the excavation area
(Ibid.).
On the southwest side of the tumulus, a long dromos leads to Tomb 1,
constructed at the same time as the tumulus itself. The seven-chambered tomb is
generally oriented east-west, with the entrance facing west. Past the dromos, two
antechambers, with a total of three cubicles per side, lead to a final room. That
this tomb was re-used in the Hellenistic period is clear, despite the destruction
caused by vandalism in antiquity (Ibid.).
Tomb 2 was constructed between 480 and 460 BC and belongs to the
second phase of the tumulus‟ use (Ibid.). Simpler than Tomb 1, it consists of a
short dromos and two consecutive chambers separated by parapets. Upon
discovery, only the lower 180 centimetres of the walls remained and the roof was
missing entirely, yet Tomb 2 managed to preserve six sarcophagi in pietra fetida
from the fifth century BC and five cinerary urns of sandstone and terracotta
indicating re-use of the tomb in the Hellenistic period (Ibid.). Tomb 2 also yielded
35
more than one hundred pieces of Etruscan goldwork, seemingly overlooked by the
ancient grave robbers (Ibid.).
On the east side of Tumulus II, facing Mt. Sant‟Egidio and the city of
Cortona, a monumental stone terrace-altar (Figure 3.5) was discovered in 1990
(Holder 1999: 39). Excavators suggest that the altar, which is 5 metres wide and 9
metres long, can be compared to the „bridges‟ or access ramps of Southern Etruria
and was likely used to obtain access to the top of the tumulus mound (Vallone
1995, Zamarchi Grassi 1992: 124-31). Further excavation has revealed signs of a
temple or sacellum higher up the mound, that was likely associated with the
terrace-altar (Zamarchi Grassi 2000, Zamarchi Grassi 2006). Zamarchi Grassi
(1992: 124-31) argues it is also likely that the altar itself was a space for cultic
activities, due in large part to the immense size of the terrace platform.
Irrespective of whether mourners actually proceeded to the top of the tumulus
mound, the placement of the terrace altar was likely connected with religious
ceremonies and processions honouring the deceased (Barker and Rasmussen
1998: 238, Fracchia 1996, Vallone 1995).
In the Roman period, the stone altar collapsed and the tombs fell into
disuse. A number of inhumations dating from the first century BC to the first
century AD have been found at the tumulus, however, particularly around the
terrace altar (Figure 3.6). According to Zamarchi Grassi (2010) seventeen tombs
were discovered in the vicinity of the altar but the total number of Roman period
inhumations for the tumulus in unclear. These burials consisted mostly of a
cappuccina tombs in wood or bricks; many of the burials that were interred after
36
the altar collapsed incorporated some of the remaining stone. The grave goods of
these burials, too, were unexceptional: cooking vessels and tableware.
Ten inhumations from the area of Tumulus II were available for study
although for most, information on the precise context is unavailable. Inhumation
SII-1 was likely found in the dromos of Tomb 2 (H. Fracchia, pers. comm.). SII-2
was most likely found in the vicinity of the terrace-altar, although its exact
provenience is unclear (H. Fracchia, pers. comm.). SII-3, SII-4, SII-5, SII-8, and
SII-9 were excavated from the area immediately surrounding the terrace altar.
Based on the MAEC display plaques, SII-3, SII-4, and SII-8, from “graves” D, G,
and L respectively, appear to have lacked any sort of stone construction and could
possibly be included in the a cappuccina type tombs in wood mentioned above,
although this is uncertain. SII-5, from “grave” M, and SII-9, from “grave” E,
appear to have had substantial stone, brick or tile coverings. All five burials date
to the period after the altar collapse. While SII-6 certainly comes from the area of
Tumulus II, more information is lacking. It is possible that the remains were
found in layers above those of the altar burials or that they came from closer to
the top of the tumulus but these speculations are unsubstantiated at this point (H.
Fracchia, pers. comm.). The commingled remains of two other individuals, SII-7a
and SII-7b, were found in association with Tomb 2: most likely in the
entranceway to the first chamber (Pacciani and Mainardi 2006).
The Circoli
Approximately 40 metres north of Sodo Tumulus II, the Circoli necropolis
was revealed in July of 2005, during a salvage excavation carried out in
37
conjunction with work to shift the course of the Loreto stream. The necropolis is
composed of two circoli, or circular areas, and is essentially an urn-field, much
smaller but not unlike those of the Villanovan period (Torelli 1975). The circles
contain cremation burials dated to the early Orientalizing period, between the
mid-seventh and mid-sixth centuries BC. Circolo I, between 7.5 and 8 metres in
diameter, contains six tombs a cassetta, in stone-lined trenches. Circolo II, 8
metres in diameter and outlined in roughly squared sandstone slabs, has thus far
yielded fifteen burials. Numerous Hellenistic and Roman period burials have been
discovered in the necropolis, several of which disturbed older deposits. To date,
only one of these intrusive burials is available for study: C-1, a Roman period
inhumation cut into Tomb 2 of Circolo I (Figures 3.7 and 3.8).
The foundations of a substantial building were also discovered
approximately 200 metres west of the Circoli. Although the structure‟s
association with the necropolis and Sodo tumuli is unknown, a preliminary
assessment suggests it was likely a Roman farmhouse, dated to the first century
BC. Given the location and date of the structure, as well as the conspicuous
absence of individual farmhouses in the valley between the Sodo and Lake
Trasimeno, it is probably that the farmhouse represents the southernmost of the
Arezzo land allotments allocated by Sulla (H. Fracchia, pers. comm.).
RE-USE OF PREHISTORIC BURIAL SITES IN THE ROMAN WORLD
AND POSSIBLE IMPLICATIONS FOR THE SITE AT SODO
The relationship between burial and cultural identity “dominates” research
into burials in the Roman provinces (Pearce 2000: 2) but this line of questioning
38
has been relatively overlooked in Etruria. Before we progress to an examination
of the dental data, it is worth entering into a brief discussion of the issue of
interpretation of the Sodo burials. According to available information, the Roman
period individuals from the Sodo were all found in association with the drums of
the tumuli but outside the actual tomb chambers. As stated above, poor
individuals were frequently buried around tumuli throughout Etruria but the late
date of the Roman period burials, as well as the extreme proximity to the tumuli,
is unusual. This positioning raises interesting questions regarding the identity of
the individuals and their motivation for choosing, if they did indeed choose, to be
buried at the feet of these ancient funerary mounds. Vallone (1995) mentions that
the beginning of the second century BC saw the occupation of all the cultivable
space in the Val di Chiana and it could be argued that the choice in burial site was
simply due to a lack of alternative locations. However despite the fact that the
total number of Roman period burials at the Sodo is unclear, it is certainly not
representative of the number of people living in the Val di Chiana at that time.
Any theory on the re-use of the Sodo as a burial space must account for the fact
that other inhabitants of the area were most certainly burying their dead
elsewhere.
Torelli (1986: 62) says that at the turn of the millennium there developed
among the Etruscans, “a thirst to rediscover the institutions and symbols of a past
glory.” At this time the „League of the Twelve Peoples‟ reappeared with fifteen
members and memorials were erected to celebrate past achievements. This period
also saw the beginning of a trend among Etruscan scholars to create a
39
comprehensive documentation of Etruscan histories, genealogies, and religious
texts. Zamarchi Grassi (1992) suggests that re-use of the Sodo tumulus II could
have been an attempt on the part of the Roman period inhabitants to share in the
prestige of the deceased principes. Fracchia (1996) adds that there were, at this
time, hundreds of years of precedent for the sharing of myths, origin stories and
other legends between the Etruscan principes and Rome. It is possible, she argues,
that the re-use of the Sodo II tumulus is analogous to the way the aristocracy at
this time was re-using Etruscan monuments and sacred places for their own
purposes (Fracchia 2006: 33-4). There is a crucial difference, however, between
the Roman period re-use of the Sodo and the re-use of other local Etruscan sites
discussed in Chapter 2: at the Sodo it is not the aristocracy but low-status
individuals that are burying their dead around the tumuli. This burial at the Sodo
of poor individuals should be distinguished from the occupation of important
Etruscan sites by the powerful inhabitants of the Val di Chiana.
Roman period re-use of prehistoric burial sites is not uncommon,
particularly in the provinces (e.g. Galliou 1989, Jessup 1962, Vermeulen and
Bourgeois 2000). Morris (1992: 51) argues that it is “tempting” to see this revival
of the use of tumuli as “working at one level as a form of symbolic resistance to
imperialism” by descendants of the indigenous population. Indeed, the argument
that changes in funerary custom were often related to conflict between Roman
forces and the local populations is a convincing one. Although dealing only
minimally with the re-use of Etruscan tumuli, Thoden Van Velzen (1992)
explores how funerary practices figured prominently in the struggle between
40
Etruscans and Romans in the second and first centuries BC at Chiusi. She
observes that there, the indigenous Etruscans legitimized their inheritance claims
and stressed a shared identity through the use of funerary inscriptions.
Conversely, she notes multiple cases of Roman immigrants appropriating
Etruscan tombs and deposing the original inhabitants: an act, she suggests, that
was intended to establish and enforce their own territorial rights (Ibid.).
Other researchers have suggested that the re-use of traditional tumuli is
less a political statement than a statement on the mystical nature of the tumuli
themselves. When discussing the burial of Roman period individuals in and
around prehistoric monuments in Brittany, Galliou (1989: 31) suggests that the
size of these monuments as well as the darkness and mystery of their interiors led
people to associate them with magical protection. He gives, as evidence, examples
of monuments that have been found to contain Roman goods, such as coins and
votive statues, apparently not associated with burials and suggests that these sites
were understood to possess a power that extended past their purpose as funeral
locales.
Vermeulen and Bourgeois (2000) explore the idea of how people in the
past decided where to bury their dead and whether the presence of considerably
older graves affected this decision. They discuss five potential explanations for
why Roman graves, whether isolated or in cemeteries, are often found in
association with prehistoric cemeteries:
First, there was a Roman law that stated that the presence of a grave
transformed any land, public or private, into a locus religiosus. Lands with this
41
status were respected and were considered legally protected against grave robbing
and other forms of desecration. It is possible then, that Roman-period individuals
in the Val di Chiana would choose to bury their dead at the Sodo if, as the authors
suggest, indigenous burial grounds retained some level of protection from
disturbance well into the Roman period.
Second, prominent funerary monuments are easy to identify in the
landscape. Even if the burials themselves become overgrown, the prehistoric
tombs can be used as a kind of grave marker, to help the living locate their dead.
The effectiveness of prehistoric monuments as grave markers is heightened by
close proximity to a road, such as at Sodo, which facilitates access to the site.
Third, burial monuments were frequently used as property markers in
antiquity. The potential role of the Sodo tumuli, for example, as markers of two
properties separated by the Loreto stream has already been discussed. Vermeulen
and Bourgeois (2000) argue that in similar situations, the re-use of archaic burial
spaces would grant them de novo the status of “structuring elements”: reinforcing
boundaries and ownership of the land. Riva (2010: 126) suggests that the
conspicuous and extravagant nature of the funerary processions enhanced political
authority and reaffirmed ownership of the land. In both scenarios, the Roman
period use of prehistoric monuments serves a judicial function, endowing the
tombs with a renewed relevance in the political as well as topographical and
funerary landscapes.
Fourth, the spiritual or religious nature of prehistoric funerary monuments
was still honoured in the Roman period. Similar to Galliou‟s conclusions in
42
Brittany, in this model a prehistoric site retains its importance in the landscape not
because of pragmatic or logistical issues but due to its persistence as a sacred
space. The relevance of this theory in interpreting the Romanization of the Val di
Chiana has already been discussed in Chapter 2 with regards to the transformation
of Etruscan springs, shrines and sanctuaries into important Roman sites. Although
the purpose or ownership of the land may have changed, it continued to be
regarded as an important, powerful or even sacred place. In the case of the Sodo
tumuli, then, it is possible that the sacred or mystical aspects of the sites were
considered to endure well into the Roman period.
Finally, the continued use of ancient cemeteries could have reinforced the
belief that supernatural elements were “permanently present” which, in turn,
stimulated further use of the cemeteries (Vermeulen and Bourgeois 2000: 145).
The sites of prehistoric burials were then considered to be a kind of liminal zone:
locations for ritual activities relating to the cult of the deceased. Izzet (2007: 88)
argues that in the Orientalizing and Archaic periods, tumulus mounds throughout
Etruria formed an “interface” between the living and the dead, and were important
for “materialising social and cultural attitudes” towards the deceased. Riva (2010:
124) suggests that the construction of Tomb 2 at Sodo Tumulus II in the fifth
century BC was a “physical alteration” of the tumulus‟ boundaries to suit the new
reality. It is possible that similarly, the Roman period burials at the Sodo altered
the boundaries of the tumulus and thus, the “interface” to which Izzet refers. The
newly redefined burial zone could then, once again, have served as a locus for
cultic activities focusing on the deceased. Vermeulen and Bourgeois (2000) apply
43
this explanation to the continuous and prolonged use of prehistoric cemeteries,
rather than individual funerary mounds. However, the authors credit such
“indigenous traditions” (Vermeulen and Bourgeois 2000: 146) as responsible for
the lasting importance of prehistoric tombs in the Roman landscape and the
potential importance of ancestor worship at the Sodo should not be overlooked.
In creating their explanations, Vermeulen and Bourgeois borrow heavily
from the phenomenological perspective outlined by Tilley (1994). Such an
approach emphasizes how people in the past interacted with, interpreted and
related to their environment. Parker Pearson (1993), too, emphasizes the
importance of a “symbolic landscape” when attempting to understand the
relationship between the living and the dead in archaeological contexts. At the
Danish Iron-Age sites of southern Jutland, Parker Pearson identifies a number of
general themes when interpreting the relationship between the living and the dead
including: the use of the dead for political legitimation, the identification of the
dead with a distant past, the naturalization of social order by reference to a
supernatural or ancestral hierarchy and the emulation of restricted practices by
inferior social groups.
Certainly, the strongest interpretations for the re-use of prehistoric
funerary monuments involve a combination of explanations. Although these
scenarios are helpful for interpreting the Roman period use of the Sodo tumuli,
direct comparisons between sites are ill-advised. A regional approach is essential
and only more information on the Sodo material can improve our understanding
of who was buried at Sodo during the Roman period, and why.
44
CONCLUSION
This chapter outlines the archaeological context for the Roman period
Sodo material, and briefly reviews the literature on Roman period burials in
prehistoric funerary monuments. Although potential reasons for the Roman period
re-use of the Sodo site is a thought-provoking topic, at this time there is
insufficient information to draw any firm conclusions about motivation or
intention. The remainder of this thesis will focus on the available dental data from
the Sodo and Terontola.
45
Figure 3.1. A satellite image showing the location of the Sodo tumuli. The city of
Cortona is visible in the bottom right while the modern highway SS71/SR71 is
indicated by white arrows. (©2012 Google).
Figure 3.2. A plan of the Sodo tumuli and Circoli showing the locations of the
Roman period burials. (Adapted from a display plaque at the MAEC).
46
Figure 3.3. Plan of Sodo Tumulus I showing the location of SI-1, SI-2, and SI-3
in the dromos. (Adapted from Vallone 1995, Fig. 18).
Figure 3.4. A drawing of Sodo tumulus II revealing the plans of Tomb 1, Tomb
2, and the terrace altar. (Adapted from a display plaque at the MAEC).
47
Figure 3.5. The reconstructed stone terrace altar at Sodo tumulus II.
Figure 3.6. Plan of the tombs surrounding the terrace altar at Sodo II, showing the
locations of burials SII-2, SII-3, SII-4, SII-5, SII-8, and SII-9. The precise
location of SII-2 could not be determined from available records. (Adapted from a
display plaque at the MAEC).
48
Figure 3.7. Plan of Circolo I showing the location of C-1, an intrusive inhumation
disrupting an earlier cremation burial. (Adapted from a display plaque at the
MAEC).
Figure 3.8. Photo of C-1 in situ, during excavation. (Credit: MAEC).
49
WORKS CITED
Banti L. 1973. Etruscan Cities and their Culture. Bizzarri E, translator. Berkeley
(CA): University of California Press.
Barker G, Rasmussen T. 1998. The Etruscans. Oxford: Blackwell Publishers Ltd.
Bietti Sestieri AM. 1984. Central and southern Italy in the Late Bronze Age. In:
Hackens T, Holloway ND, Holloway RR, editors. Crossroads of the
Mediterranean. Archaeologica Transatlantica 2. Louvain-La-Neuve,
Belgium: Université Catholique de Louvain/Providence (RI): Brown
University, pp55-122.
Bruschetti P. 1992. Il Sodo: Il tumulo I. In: Zamarchi Grassi P, editor. La Cortona
dei principes. Cortona: Calosci-Cortona, pp170-186.
Davies G. 1977. Burial in Italy up to Augustus. In: Reece R, editor. Burial in the
Roman World. London: Council for British Archaeology Research Report
22, pp13-19.
Fracchia H. 1996. Etruscan and Roman Cortona: New evidence from the
southeastern Val di Chiana. In: Hall JF, editor. Etruscan Italy. Provo (UT):
Brigham Young University, pp191-215.
Fracchia H. 2006. The Villa at Ossaia and the Territory of Cortona in the Roman
Period. Siracusa: Lombardi editori.
Galliou P. 1989. Les Tombes Romaines d‟Armorique: Essai de Sociologie et
d‟Économie de la Mort. Documents d‟Archéologie Française No 17. Paris:
Éditions de la Maison des Sciences de l‟Homme.
Holder PN. 1999. Cortona in Context: The History and Architecture of an Italian
Hill Town to the 17th
Century. Cortona: Arti Tipografiche Toscane.
Holloway RR. 1966. The tomb of Augustus and the princes of Troy. American
Journal of Archaeology 70(2): 171-173.
Izzet V. 2007. The Archaeology of Etruscan Society. Cambridge: Cambridge
University Press.
Jessup RF. 1962. Roman barrows in Britain. In: Hommages à Albert Grenier II
(Collections Latomus v. 58), pp853-867.
Johnson MJ. 1996. The Mausoleum of Augustus: Etruscan and other influences
on its design. In: Hall JF, editor. Etruscan Italy. Provo (UT): Brigham
Young University, pp217-239.
50
Morris I. 1992. Death-ritual and Social Structure in Classical Antiquity.
Cambridge: Cambridge University Press.
Oleson JP. 1982. The Sources of Innovation in Later Etruscan Tomb Design (ca.
350-100 BC). Rome: Giorgio Bretschneider Editore.
Pacciani E, Mainardi S. 2006. I reperti scheletrici umani della tomba tardo arcaica
del tumulo II del Sodo. In: Zamarchi Grassi P, editor. Il Tumulo II del
Sodo di Cortona: La tomba di età tardo arcaica. Cortona: Accademia
Etrusca, pp105-111.
Parker Pearson M. 1993. The powerful dead: Archaeological relationships
between the living and the dead. Cambridge Archaeological Journal 3(2):
203-229.
Pearce J. 2000. Burial, society and context in the provincial Roman world. In:
Pearce J, Millett M, Struck M, editors. Burial, Society and Context in the
Roman World. Oxford: Oxbow Books, pp1-12.
Riva C. 2010. The Urbanisation of Etruria: Funerary Practices and Social Change,
700-600 BC. Cambridge: Cambridge University Press.
Spivey N, Stoddart S. 1990. Etruscan Italy. London: B.T. Batsford Ltd.
Thoden Van Velzen D. 1992. A game of tombs: The use of funerary practices in
the conflict between Etruscans and Romans in the second and first
centuries BC in Chiusi, Tuscany. Archaeological Review from Cambridge
11(1): 65-76.
Tilley C. 1994. A Phenomenology of Landscape. Oxford: Berg Publishers.
Torelli M. 1975. Introduction. In: Boitani F, Cataldi M, Pasquinucci M. Etruscan
Cities. Atthill C, Evans E, Pleasance S, Swinglehurst P, translators.
London: Cassell, pp9-28.
Torelli M. 1986. History: Land and people. In: Bonfante L, editor. Etruscan Life
and Afterlife. Detroit (MI): Wayne State University Press, pp47-65.
Toynbee JMC. 1971. Death and Burial in the Roman World. London: Thames and
Hudson.
Vallone MT. 1995. Cortona of the principes. Etruscan Studies 2: 109-139.
51
Vermeulen F, Bourgeois J. 2000. Continuity of prehistoric burial sites in the
Roman landscape of Sandy Flanders. In: Pearce J, Millett M, Struck M,
editors. Burial, Society and Context in the Roman World. Oxford: Oxbow
Books, pp143-161.
Zamarchi Grassi P. 1992. Il Sodo: Il tumulo II. In: Zamarchi Grassi P, editor. La
Cortona dei principes. Cortona: Calosci-Cortona, pp120-138.
Zamarchi Grassi P. 2000. Il tumulo II del Sodo di Cortona (Arezzo). In: Comune
di Bologna Museo Civico Archeologico. Principi etruschi tra
Mediterraneo ed Europa. Venice: Marsilio, pp140-142.
Zamarchi Grassi P. 2006. Storia delle ricerche. In: Zamarchi Grassi P, editor. Il
Tumulo II del Sodo di Cortona: La tomba di età tardo arcaica. Cortona:
Accademia Etrusca, pp15-21.
Zamarchi Grassi P, editor. 2010. The II° Tumulus at Sodo. Ministero per I Beni
Culturali e Ambientali. Soprintendenza ai Beni Archeologici Della
Toscana. Sezione Didattica.
Zifferero A. 1991. Forme di possesso della terra e tumuli orientalizzanti nell‟Italia
central tirrenica. In: Herring E, Whitehouse R, Wilkins J, editors. Papers
of the Fourth Conference of Italian Archaeology I. The Archaeology of
Power Part 1. London: Accordia Research Centre, pp107-134.
52
CHAPTER 4: DENTAL DATA
INTRODUCTION
The Etruscans inhabited central Italy from the seventh century BC to the
first century BC. Their land, known as Etruria, was comprised of portions of the
modern regions of Tuscany, Umbria, and Latium. For most of its history, Etruria
was divided into a series of oligarchical city-states ruled by families called
principes. These powerful families expanded Etruscan control north into the Po
Valley and south into Campania (Barker and Rasmussen 1998: 139-40, Nielsen
1984: 255, Torelli 1975: 11).
Over the course of the fourth and third centuries BC, Rome conquered
Etruria through a series of military battles, truces, and treaties (Harris 1971). The
ensuing centuries saw the forfeiture of goods, the reallocation of lands, and the
movement of peoples throughout the region. By the first century BC, Etruria no
longer existed as a separate entity from the Roman Empire and the Etruscans, as a
distinct and recognizable cultural group, were gone (Barker and Rasmussen 1998:
292, Harris 1971).
The hilltop city of Cortona was one of twelve major Etruscan centres
located between the Arno and the Tiber (Barker and Rasmussen 1998: 143).
Following its absorption into the Roman Empire, Cortona remained an important
economic hub for the exportation of grain and other commodities (Bonfante 1986:
11, Harris 1971: 161-8). Unfortunately, despite Cortona‟s role as a vital economic
power centre in the Val di Chiana, relatively little is understood about the people
who lived there during the various phases of Etruscan and Roman occupation,
53
particularly at the turn of the millennium. As such, the details of Cortona‟s
transition from an Etruscan city-state to a Roman economic centre are unclear.
This report presents the results of an examination of the dental remains of
eight individuals from the territory of Cortona, dated to between the first century
BC and the first century AD. It is intended as a contribution to the study of
Cortona‟s population biology in the Late Republican period.
SKELETAL MATERIALS
The skeletal material was excavated from the vicinity of two large
Etruscan funerary mounds, approximately 2 kilometres west of the city of
Cortona. The mounds, or tumuli, are referred to as Sodo I and Sodo II and date to
the sixth century BC; they sit on either side of the Rio Loreto, or Loreto stream,
thought to have originally functioned as a boundary between lands (Vallone
1995). Originally used by wealthy Etruscan landowners, the tumuli were reused in
the fifth and fourth centuries BC before falling into disuse and, ultimately, being
partially dismantled and ransacked for grave goods. In addition to the funerary
mounds, a small urn-field is located approximately 40 metres north of Sodo II,
with burials dating from the mid-seventh to mid-sixth centuries BC.
The Roman-period burials, excavated in the 1990s and early 2000s, were
found in and around the tumuli and urn-field. Despite their proximity to the
Etruscan burial chambers, however, the Roman-period inhumations were not
found within the tombs but rather in the dromos or entranceway, around the
outside of the tumuli, and in the urn-field (Vallone 1995, Zamarchi Grassi 2010).
54
The burials were a cappuccina, that is, under a structure of gabled roof tiles or
wood: the standard in Roman period Etruria for low status individuals (Barker and
Rasmussen 1998: 234, Toynbee 1971: 102). The grave goods were a modest
selection of common objects of everyday use such as cooking vessels and
tableware, consistent with an interpretation of low status.
The total number of Roman-period inhumations from the Sodo tumuli is
unclear. Although the discovery of seventeen tombs from the vicinity of the
terrace altar at Sodo II is confirmed (Zamarchi Grassi 2010), at least eight other
individuals were excavated from the territory of Sodo I, Sodo II and the urn-field.
Containers holding the remains of fourteen individuals were brought to the Museo
dell‟Accademia Etrusca e della Città di Cortona (MAEC) from the Centro
Restauri (Sodo Restoration Workshop) where they are curated. Additional
material included nonhuman skeletal fragments as well as some terracotta and
brick.
Preservation was a major issue with the Sodo material, due in large part to
the location of the site on the valley floor and the resulting high water table. The
obstructive and invasive nature of the surrounding groundwater during excavation
is evident from the photographs and notes documenting the process (e.g.
Bruschetti and Zamarchi Grassi 1999, Vallone 1995). Unfortunately, the negative
effects of groundwater on the Sodo material started significantly before the
55
twentieth century, as the two had been interacting for approximately 2,000 years
by the time of excavation.1
Diagenetic changes to archaeological bone include changes in both the
organic and inorganic components, and involve hydrolysis, dissolution, mineral
replacement, and recrystallization (Hare 1980, Von Endt and Ortner 1984). These
changes can be affected by a number of factors, including duration of interment,
hydrology of the burial environment, and sediment pH and composition
(Henderson 1987, Nicholson 2001, Sandford and Weaver 2000).
Generally, the longer the burial interval, the more advanced the diagenesis
but the degree of diagenetic change does not necessarily bear a linear relationship
to time (Sillen 1989). The movement of water in the burial environment is one of
the most important factors affecting diagenesis, as water acts as the medium in
which most forms of change take place (Millard 2001, Nielsen-Marsh et al. 2000).
Water determines the movement of solutes into and out of the bone; it enables the
leaching of peptides and amino acids and promotes recrystallization (Hare 1980,
Nielsen-Marsh et al. 2000, Von Endt and Ortner 1984). The characteristics of the
surrounding soil also play an important role in regulating the rate of diagenesis:
grain size affects the flow of groundwater, with coarser soils allowing for more
movement in the space between particles, and soil pH heavily influences the
effect of water on archaeological bone, with acidic soils intensifying diagenetic
processes (Gordon and Buikstra 1981, Henderson 1987, Nielsen-Marsh et al.
2000).
1 What follows is a brief outline of bone diagenesis as it pertains to the Sodo material. For a more
comprehensive review of the literature on bone diagenesis, see Appendix B.
56
Although the hydrology of the Sodo site has not remained unchanged over
the last 2,000 years, due to the location of the site on a valley floor and the
saturation and subsequent draining of the Val di Chiana, we can assume that
groundwater infiltration has been a continual problem (for archaeological and
historical evidence of flooding in the Val di Chiana, see Alexander 1984, Spivey
and Stoddart 1990: 30). Soil samples from four different burials (SI-2, SII-1, SII-
2, and SII-3) representing both tumuli were analyzed for acidity and particle size,
in the hopes of better understanding the effects of the burial environment on the
Sodo material. Tests indicate a range in pH values from 7.22 to 7.51, which
should have retarded the degradation of the mineral phase, if only slightly.
Samples were found to consist predominantly of silt and sand (2-50 micrometres
and greater than 50 micrometres, respectively). Only a small percent of the soil
samples were clay (under 2 micrometres): less than 20 per cent in all samples and
as low as 13 per cent in SII-2.2 Despite the slightly basic nature of the burial
environment, the relatively large grain size of the soil surrounding the Sodo
burials would have accelerated diagenesis by facilitating the movement of water
around and through the skeletal material.
The Roman-period material from the Sodo displayed an extreme degree of
degradation; exhibiting a fragility characteristic of diagenetically altered bone
(Figures 4.1 and 4.2). Preliminary x-ray fluorescence analysis of long bone shaft
fragments from two individuals, SI-3 and SII-2, has revealed abnormally high
2 The soil from burials SI-2 and SII-3 meets the criteria for a medium loam, according to the
United States Department of Agriculture soil texture guidelines, while the soil from SII-1 and SII-
2 is a sandy loam.
57
levels of silicates and iron, which are among those elements most frequently taken
up by skeletal remains in archaeological contexts (see Parker and Toots 1980,
Sandford and Weaver 2000). Additionally, collapse of the tomb walls and
ceilings, as well as the a cappuccina tiles, has rendered the remains highly
fragmentary: most likely exacerbated by the diagenetically weakened structure of
the bone (e.g. see Pacciani and Mainardi 2006).
Light microscopy (LM) of thin sectioned long bone fragments from SI-3
and SII-2 revealed extensive microbial infiltration. Microbes such as bacteria and
fungi are known to infiltrate bone by way of the lacunae and canaliculi, before
spreading by way of the „path of least resistance‟ until all areas are infiltrated
(Pitre 2011, Turner-Walker and Jans 2008)3. Once microbial infiltration is
complete, the organic material develops a mottled appearance and begins to break
up and recede, leaving behind tunnels in the bone created through
demineralisation and redeposition of the bone matter. The appearance under LM
of the bone from SI-3 and SII-2 was consistent with that described by Pitre
(2011), in that both the receding organic matter and some tunnels were visible
(Figures 4.3 and 4.4). The Haversion canals were the only identifiable structures
remaining in the histology of the Sodo bone: the remaining bone had been
demineralised or hypermineralised by the microorganisms to such an extent that
the structures were rendered unrecognizable.
3 Pitre (2011) found that biodeterioration was guided by the preservation of the bone‟s mineral and
collagen components, with microorganisms showing a preference for less mineralised areas
whenever available. In the Sodo sample, infiltration would have certainly been facilitated by the
degree to which the mineral component had been compromised by diagenesis.
58
Due to the degree of taphonomic change in the Sodo material, postcranial
analysis was largely impossible. Dental enamel‟s low solubility, low porosity and
low organic content, however, means that teeth are more likely to survive burial
conditions that destroy bone (Carlson 1990, Parker and Toots 1980). This study
has therefore been necessarily restricted to dental pathology, for which the
dentitions of only eight individuals were available.
The study sample consists of eight skulls yielding 127 permanent teeth: 79
in situ and 48 loose, and 159 observable alveoli. Given that the normal human
dentition has 32 teeth, for eight individuals we would expect a study sample of
256 teeth. Thus, slightly less than half the expected number of teeth are present.
While most of the missing teeth were likely lost postmortem, the 97 unobservable
alveoli means that the frequency of antemortem tooth loss must be considered a
minimum.
METHODS OF PALAEOPATHOLOGICAL IDENTIFICATION AND
ANALYSIS
Data collection for this study was conducted by the author under the
supervision of Dr. Nancy Lovell in June of 2010, at the Museo dell‟Accademia
Etrusca e della Città di Cortona in Cortona, Italy. Detailed written descriptions
were recorded for each of the Sodo skulls and extensive photographs were taken4.
Postcranial remains were examined and notes were taken on features that
contributed to age and sex determinations for the skeletal remains.
4 A sample data collection sheet is included in Appendix C.
59
In spite of the poor degree of preservation of the skeletal material, sex
could be determined by a number of features of the skeleton, including degree of
robusticity, size of teeth, and morphology of the skull and pelvis, where available
(following the standards described by Bass 2005; Buikstra and Ubelaker 1994).
Female skeletons were identified by a broad sciatic notch, small mastoid
processes, sharp supra-orbital margins, a gracile mandible and smaller tooth
crowns. Male skeletons were characterized by a narrow sciatic notch, large
mastoid processes, dull supra-orbital margins, a robust mandible with a
pronounced chin and larger tooth crowns. Sex was determinable for all eight
individuals in this study: three females and five males. In the other five
individuals from the Sodo, sex could not be determined due to poor preservation
of the skeletal material or ambiguous characteristics: these individuals were
labeled as “Unknown”.
There were no juveniles in the Sodo remains. Adult age at death was
estimated in broad terms and individuals were grouped into four categories:
Young Adults (20-34 years), Middle Adults (35-49 years), Old Adults (50+ years)
and Unknown. Use of age categories such as these has been deemed helpful in
bioarchaeology because they allow for the detection of age related trends.
Techniques for specific age estimation in adults can be problematic even in ideal
situations; in the Sodo sample, many of the features recommended for age
estimation in adults, such as the pubic symphysis or the auricular surface of the
pelvis, were not preserved. As a result, age estimation was based primarily on
relative degree of tooth wear (Brothwell 1981, Smith 1984) and cranial suture
60
closure (Meindl and Lovejoy 1985). One has to be careful when using tooth wear
as an indicator of age because it can be affected by a number of other factors
including diet and cultural practices. Within a given population where a similar
diet can be assumed, however, tooth wear can be broadly understood to increase
with age5. Cranial suture closure rates also see some variability but generally,
cranial sutures fuse with age so that in juveniles, the cranium is composed of a
number of completely separate bones that fuse over time and in Old Adults, the
sutures are not only fused but largely obliterated. One burial, SI-2, was very
incomplete and was identified as a Middle Adult from the presence of
osteoarthritis on the patellae. Osteoarthritis is a degenerative condition and as
such, typically more common in middle-aged and older adults (Ortner 2003). In
all, data on age was obtainable from the eight individuals in the study: six Young
Adults and two Middle Adults, and a ninth individual without a preserved
dentition: a Middle Adult.
The identification and documentation of pathological features of the teeth
followed the macroscopic standards outlined by Buikstra and Ubelaker (1994) and
Hillson (1996). Although other forms of analysis, including x-rays and extensive
microstructural analysis can provide valuable information, these were not applied
to the Sodo material due to logistical constraints. That such forms of analysis are
time consuming, require specialized equipment and are often destructive in nature
placed them outside the scope of this study.
5 The wear observed on the first molars of the Sodo sample averaged a high 3/low 4 on the Smith
scoring scale, as compared to the much higher wear observed in the individual from Terontola (see
Chapter 5).
61
Caries was documented in terms of the affected tooth type and surface.
Only carious lesions that penetrated the surface enamel were documented.
Abscesses were evaluated based on the presence of a channel in the alveolar bone
for pus drainage. Dental calculus was recorded as absent, minimal, moderate, or
severe and its position on the buccal/labial or lingual aspect of the tooth was
noted. Enamel hypoplasia was recorded in terms of the affected teeth and the type
of deformity. Forms of expression include linear horizontal grooves, linear
vertical grooves and nonlinear arrays of pits. Where tooth loss was recorded for
the Sodo skulls, attempts were made to differentiate between antemortem tooth
loss (AMTL) and postmortem tooth loss (PMTL), based on the appearance of the
alveolar bone. When a tooth is lost antemortem, the body fills the alveolus in with
bone until it is obliterated completely and the surface is smooth and level with the
surrounding bone. In postmortem tooth loss, the alveolus does not fill with bone
and so the tooth socket is preserved. When teeth were missing but it was not
possible to distinguish between antemortem and postmortem tooth loss, usually
due to a lack of preservation of the alveolar bone, they were recorded as
“unknown.” Despite extensive efforts to prepare the Sodo material for
observation, it was difficult in some cases to identify pathological conditions with
certainty, due to the preservation issues discussed above. In these instances, the
lesions were not recorded and thus, certain pathological conditions may be
underrepresented in this study.
As a result of the small sample size in this study, statistical analysis was
deemed largely unhelpful. Still, attempts were made to identify the most prevalent
62
conditions and any strong differences in distribution between sex or age
categories. Results are presented in both the individual count (number of affected
individuals relative to the number of observable individuals) and tooth count
(number of affected teeth/alveoli relative to the number of observable
teeth/alveoli) formats. Despite the small sample size, the individual count method
permits an understanding of the prevalence of a given disease across the sample.
The tooth count method is useful in comparison with the individual count results
and also allows for an evaluation of disease frequencies in different teeth.
RESULTS
Individual Count
The prevalence of dental disease by individual count is presented in Table
4.1. Caries was the most common dental affliction, with carious lesions observed
in all eight of the individuals. Antemortem tooth loss was observed only in a
single middle adult male. Enamel hypoplasia was also relatively common, with 63
per cent of the individuals exhibiting at least one affected tooth. The young age of
the individuals is likely responsible for the low frequency of calculus (38 per cent
of individuals) and the complete lack of peri-apical abscesses, although the poor
preservation of the bone certainly would have contributed to these results. Due to
the small sample size of the individual count method, statistical analysis was
deemed unhelpful. However, there appears to be no marked differences in the
frequencies of any of these diseases between the sexes.
63
Tooth Count
The results of the tooth count evaluation of dental disease for the skeletal
sample are presented in Tables 4.2-4.5. Due to the poor state of preservation, the
total number of teeth and alveoli were not observable for every condition: thus the
number of observable teeth or alveoli varies between tables. 6
Caries
The sample exhibits a carious tooth frequency of 14 per cent (Table 4.2).
The largest number of affected teeth in a single individual was four: found in SI-
3, a young adult male (Figure 4.5). The absence of caries in the anterior teeth
(incisors and canines) and the high frequency of caries in the molars appear
consistent with the common explanation that carious lesions frequently originate
in the pits and fissures of the posterior teeth (Hillson 1996: 272). However, the
most common type of carious lesion was interproximal, accounting for ten of the
observed lesions. The close proximity of the premolars and molars likely accounts
for the prevalence of interproximal caries in these teeth. Pit and fissure cavities
were the second most common type, accounting for seven of the observed lesions.
Antemortem Tooth Loss
Antemortem tooth loss was not a common condition in this sample,
affecting only three teeth in a single individual: SII-3, a middle adult male (Table
4.3). The exclusive occurrence of antemortem tooth loss in the molars (both M2s
and a left M1) is again consistent with the tooth morphology and could indicate
6 The presence of isolated teeth was taken as clear indication that they were lost postmortem and
so, for the purposes of evaluating antemortem tooth loss, contributed to the number of observable
tooth sockets. Thus, in this study, the number of observable tooth sockets referenced in Table 4
includes both observable alveoli and isolated teeth.
64
that caries was the most important factor in the aetiology of AMTL, with
subsequent infection and, ultimately, loss of the tooth. The M2s were clearly lost
due to abscessing, but the reason for the loss of the M1 is unclear (Figure 4.6).
Caries is, of course, not the exclusive cause of AMTL; extreme wear and the loss
of alveolar bone due to periodontal disease are other potential factors.
Enamel Hypoplasia
Linear enamel hypoplasia was observed in 25 teeth (20 per cent), making
it the most common dental affliction by tooth count (Table 4.4), however a single
individual (SI-3, a Young Adult male) accounted for 16 of those 25 affected teeth.
The condition appears to affect males more than females but it is difficult to know
if this trend would hold in a larger sample size. Enamel hypoplasia represents
disruption of the enamel matrix secretion during the crown formation stage of
tooth development (Figure 4.7). Hypoplastic lesions are nonspecific indicators of
stress to the body and can result from malnutrition, disease or other factors
(Hillson 1996: 165-167).
Calculus
Supra-gingival calculus was observed on 15 per cent of the teeth (Table
4.5). In all cases, the calculus depositions were slight (after Brothwell 1981) and
no associated resorption of the alveolar bone was observed (Figure 4.8). The
mandibular teeth appear to be more affected than the maxillary teeth and while it
is difficult to pronounce with certainty, it is likely a larger sample size would
support this trend. The lingual surface of the mandibular incisors tends to be the
area most susceptible to calculus formation, and only 11 of a potential 32 lower
65
incisors were observable. Males also appear to be more affected than females in
this sample, a trend that has been observed in modern populations (Beiswanger et
al. 1989).
DISCUSSION
Several factors make comparison of the Sodo dental data with those of
other samples problematic. First and foremost is the small sample size. Only eight
individuals are represented by the dental data above and none of these individuals
possessed a complete dentition, unaffected by postmortem loss. Additionally, the
average age of the study sample is relatively young, with six Young Adults and
two Middle Adults. Many of the disease processes discussed in this paper are
progressive, and the frequencies observed in a young sample cannot be
generalized across a normal population age distribution.
Second, at the turn of the millennium, the various territories of what had
been Etruria were inhabited by a mixture of local Etruscan descendants as well as
immigrants: predominantly Roman war veterans and members of the other Italic
tribes (H. Fracchia, pers. comm., Harris 1971). The composition of the population
living in the territory of Cortona in this period is difficult to establish, let alone the
cultural and biological identities of the individuals buried at the Sodo. Thus, with
the population to which the Sodo individuals belonged unknown, available
Etruscan and Roman reports can be compared and contrasted with the data from
this study, but conclusive assertions cannot be made.
66
Third, there is a shortage of skeletal material from this area that can be
used for comparative purposes. While this scarcity is due in part to the rite of
cremation that waxed and waned in popularity throughout the Italian peninsula, it
is also attributable to improper excavation and storage techniques that pervaded
Etrusco-Roman archaeology for most of its history (e.g. see Pacciani et al. 1996).
This dearth of skeletal material has led many researchers to group together
remains from various sites and time periods for the purposes of examination (e.g.
Brothwell and Carr 1962, Moggi-Cecchi et al. 1997). In addition to providing the
sample size necessary for statistical significance, such grouping together of
material from discrete sites enables the establishment of skeletal characteristics
typical of the “Etruscan group” for the purposes of biological relatedness studies.
Generalizations regarding “typically Etruscan” skeletal features can be useful for
comparison with other populations but they do little to increase the understanding
of population biology in a specific area and/or time period. In assuming that the
Etruscan people, whose lands ranged from Rome to Pisa over hundreds of years,
were a consistent and biologically homogeneous population, regional and
temporal differences between populations are glossed over.
Despite these difficulties, an examination of the dental data from the Sodo
sample is presented here, along with available data on comparable Etruscan and
Roman populations. A brief discussion of Etruscan and Roman diet is also
undertaken, with attempts to draw correlations between eating patterns and
observed dental disease frequencies.
67
Caries
Caries occurs when the bacteria in the mouth break down the sugars and
carbohydrates found in food and in the saliva. Upon metabolizing these sugars,
the bacteria produce organic acids which are excreted into the plaque fluid and
come into contact with the surfaces of the teeth. The metabolism of proteins,
peptides, and amino acids produces alkaline waste products, which help to
balance the plaque pH (Hillson 1996: 254-79). The frequent presence of dietary
sugars in the oral cavity, i.e. in a diet high in carbohydrates and sugars but low in
protein, results in a consistent decrease in plaque pH and subsequent
demineralisation of the tooth mineral that is symptomatic of caries. Carious
lesions, or cavities, are holes in the tooth that develop when demineralised
patches, referred to as “incipient decay,” progress to the point of irreversible
destruction of the enamel or cementum; these lesions are pathognomonic of the
disease caries.
In the Sodo sample, all of the individuals displayed at least one carious
lesion, with an average of two affected teeth per individual (14 per cent of all
teeth were affected). There is no apparent difference in the frequency of carious
lesions between the Young Adults (15 per cent of teeth were affected) and the
Middle Adults (13 per cent of teeth were affected), though it is difficult to know if
larger samples would provide a different outcome.
Other studies of Etruscan and Roman dental health have found lower rates
of caries than exhibited in this sample. In Capasso‟s (1987) study of 119 Etruscan
crania from the territory of Chiusi, only 28 per cent of individuals and 7 per cent
68
of teeth displayed signs of caries. In the population from the Monterozzi
necropolis at Tarquinia, 25 per cent of individuals and only 3 per cent of teeth
were affected (Bartoli et al. 1991). Brothwell and Carr (1962) found that 4 per
cent of teeth were affected in their study of a number of skulls from all over
Etruria. The Roman Imperial Age population from Isola Sacra was reported as
exhibiting caries in 36 per cent of individuals and 4 per cent of teeth, while the
population from Lucus Feroniae exhibited the disease in 52 per cent of individuals
and 6 per cent of teeth (Manzi et al. 1999). Higher values were found in a sample
of 67 adults of Roman Imperial Age from Quadrella, southeast of Rome: 72 per
cent of individuals showed signs of the disease, and 15 per cent of teeth were
affected (Bonfiglioli et al. 2003). While, at Quadrella, the tooth count frequency
was slightly higher than that in the Sodo sample, the individual count frequency
was still much lower.
The extremely high frequency of caries in the Sodo sample is likely at
least partially attributable to the relatively young age of the sample. Caries is a
progressive disease than can result in abscess and/or antemortem tooth loss when
untreated. In young adults, caries has rarely had the time to progress to these
stages but in middle and older adults, carious teeth are frequently lost
antemortem, eliminating them from the count7.
7 Unless an attempt is made to compensate for the AMTL through the use of a caries correction
factor (see Chapter 5, Footnote 4)
69
Antemortem Tooth Loss
Although it can have several aetiologies, antemortem tooth loss is
generally caused by carious infection of the pulp cavity (a process that can be
accelerated by extreme wear or dental trauma). It represents a gradual
deterioration that begins with cavities or severe wear and progresses to infection,
abscess and, ultimately, loss. As such, AMTL is uncommon in young adults and
generally becomes more frequent with age.
As expected, none of the Young Adults in the Sodo sample exhibited
AMTL but a single Middle Adult male, SII-3, lost three teeth antemortem: both
M2s and the left M1. The fact that the Sodo sample is predominantly composed of
young adults, combined with the progressive nature of AMTL, makes comparison
with other samples particularly difficult but the findings of several studies are
presented here for informative purposes.
Capasso (1987) observed AMTL in 50 per cent of individuals and 12 per
cent of alveoli in his sample from Chiusi. In this sample, the primary identified
causes were dental attrition and caries but 24 per cent of AMTL could not be
definitively linked with either condition and the author suggested dental trauma as
a potential cause. Bonfiglioli et al. (2003) had similar numbers with 60 per cent of
individuals exhibiting at least one case of AMTL, and 13 per cent of teeth being
lost antemortem. The populations at Isola Sacra and Lucus Feroniae, with 44 per
cent and 48 per cent of individuals affected respectively, appear to have similar
values (Manzi et al. 1999). However the tooth count reveals a disparity between
the populations: almost twice the number of alveoli were affected at Lucus
70
Feroniae (12 per cent) than at Isola Sacra (7 per cent). In all of these cases,
posterior teeth were more affected by AMTL than the anterior teeth and at Isola
Sacra and Lucus Feroniae, the maxillary teeth were more commonly affected than
the mandibular teeth.
Enamel Hypoplasia
Enamel hypoplasia (EH) describes an episodic deficiency in enamel
thickness that occurs during tooth formation. Hypoplastic enamel defects can be
related to localized disturbances during tooth formation (e.g. trauma, osteitis) or
to a more generalized disturbance such as systemic disease or chronic
malnutrition (Hillson 1996: 165-7). Localized disturbances tend to cause enamel
hypoplasia on an isolated tooth or group of teeth, normally in a single area of the
mouth. Defects on a number of teeth throughout the mouth are generally
indicative of a systemic aetiology (Hillson 1996: 278-9). Hypoplastic defects can
have a number of forms but are generally grouped into three categories: Furrow-
type defects (also referred to as Linear Enamel Hypoplasia), pit-type defects, and
plane-type defects (Hillson 1996: 166-7).
Of the eight individuals in the Sodo sample, five (63 per cent) exhibited
Linear Enamel Hypoplasia (or LEH) in at least one tooth, with a total of 20 per
cent of all teeth affected. The largest number of teeth affected in a single
individual was 16 (found in SI-3, a Young Adult male), while all other individuals
had no more than three affected teeth. Due to the large number of hypoplastic
teeth, the defects observed in the individual SI-3 can most certainly be said to
have a systemic cause. For the other individuals in the sample it is difficult to be
71
certain due to the incomplete nature of the dentitions. LEH was observed in both
the Young Adults and the Middle Adults. Because enamel hypoplasia occurs
during tooth formation, in a larger sample we would expect no significant change
in the frequency of occurrence between Young Adults and Middle or Older
Adults.
The morbidity of EH varies highly between comparative populations, even
within the Etruscan and Roman studies already discussed. Almost 42 per cent of
the individuals from the Tarquinia had EH, and 18 per cent of teeth were affected.
Brothwell and Carr (1962) recorded that approximately one third of the
individuals in their sample displayed some degree of EH, although they
acknowledged that accurate numbers were difficult due to postmortem loss.
Bonfiglioli et al. (2003) found LEH in 95 per cent of the individuals in their
sample, and 59 per cent of teeth. The authors conclude that such a high frequency
of the condition suggests rampant metabolic stress during growth in the Quadrella
population. Although not as high as the Quadrella sample, Manzi et al. (1999) list
relatively high individual count values (81 per cent and 82 per cent of individuals)
and tooth count values (36 per cent and 46 per cent of teeth) for LEH at Isola
Sacra and Lucus Feroniae, respectively.
Calculus
Dental calculus is the mineralized form of plaque and therefore is linked to
extensive plaque build-up which, in turn, is linked to poor oral hygiene and
significant carbohydrate consumption (Hillson 1996: 254-60). Plaque is
composed of a number of micro-organisms, including caries-causing bacteria,
72
which live in the oral cavity. Tooth fissures, interproximal areas and gingival
crevices are particularly susceptible to plaque build-up, protected as they are from
the cleaning actions of the tongue (Hillson 1996: 254). While the progression
from plaque to calculus is not well understood, it is known that the minerals
involved in calculus formation derive originally from saliva and so tooth sites
closest to salivary glands are particularly susceptible to mineralisation (Hillson
1996: 254-60). Major salivary glands are found under the tongue and inside the
cheeks, making the lingual surface of the anterior teeth and the buccal surface of
the molars prime locations for calculus development.
Slight calculus was observed in three (38 per cent) of the individuals in the
Sodo sample with a total of 18 (15 per cent) of the teeth. The condition was
considerably more prolific in the mandibular teeth (26 per cent) than in the
maxillary teeth (6 per cent) and slightly more common in males than in females
(16 per cent to 12 per cent respectively). The mineralization of calculus into
plaque is a progressive process and we would generally expect to see a higher
frequency of calculus in the older age groups. In this sample, no calculus was
observed on SII-3, the Middle Adult. This anomaly is most likely explained by
the high degree of antemortem and postmortem tooth loss in the individual. SII-3
lacks 20 of 32 teeth, including the lower incisors and six molars. With these key
sites for calculus build-up missing, it is impossible to determine what the actual
occurrence of calculus in this individual would have been.
As with AMTL, the young age of the individuals in the Sodo sample
makes comparison with other samples problematic and it should not be surprising
73
that the findings of other studies cite much higher frequencies of calculus.
Bonfiglioli et al. (2003) observed calculus deposits in 84 per cent of the Quadrella
individuals and 51 per cent of the teeth. In the sample from Lucus Feroniae, 67
per cent of individuals had calculus build-up on at least one tooth, but only 27 per
cent of teeth were affected. In the population from Isola Sacra, only 59 per cent of
individuals exhibited calculus but 34 per cent of teeth were affected. At Isola
Sacra, as with the Sodo sample, calculus was more common in males than in
females and at both Isola Sacra and Lucus Feroniae, the mineralised build-up was
more common in the mandibular teeth (Manzi et al. 1999).
The Role of Dental Data in Understanding Diet
Much of the evidence for reconstructing ancient diets comes from written
sources and tomb paintings (Barker and Rasmussen 1998: 180, Brothwell and
Brothwell 1998: 13). Small pieces of information are seized upon by researchers
and extrapolations are applied to “Etruscans” or “Romans” as a whole. Dental
data, however, can help clarify the details regarding diet and nutrition in a specific
sample. This information is valuable for highlighting similarities and differences
between populations and between individuals within a population. Differences in
diet, pertaining to age, sex, or class, for example, can be perceived only when the
variances within a population are thoroughly examined.
The use of dental data to interpret diet is based on a simple principle: the
chemical and physical composition of food affects the teeth that help to digest it.
Thus, diets that are high in a certain substance have a different effect on the teeth
than diets that are low in that substance but high in another. While general
74
knowledge may indicate, then, that a population‟s diet consisted of cereals, meat,
vegetables, and fruit, the dental data can help reveal the role filled by the various
foods.
The function of sugars in the aetiology of caries has already been
discussed: bacteria in the oral cavity digest the sugars, releasing acids that lower
the oral pH and can result in cavities. A diet high in sugary foods, then, can give
rise to a consistently acidic oral environment, and a relatively high frequency of
caries. The extent of cariogenicity is dependent on the frequency, degree, and
form of sugar consumption. Dried, sticky fruits such as figs carry an extremely
high risk of caries, while the situation with starches is more complex: starchy
foods tend to have a less immediate and less pronounced effect on the oral pH, but
the consequences are longer lasting (Hillson 1996: 278, Rugg-Gunn 1993).
Other conditions, such as calculus and enamel hypoplasia, cannot be as
simply interpreted as caries. Dental plaque appears to be related to the
consumption of carbohydrates but the mineral precipitation that leads to calculus
build up is linked to high-protein diets and an alkaline oral environment
(Bonfiglioli et al. 2003, Hillson 1979, Hillson 1996: 254-60). Perhaps even more
difficult to interpret, enamel hypoplasia is a nonspecific indicator of childhood
health and while this can include diet, it is useful only as corroborative evidence.
Ultimately, when using dental data to make interpretations about diet it is crucial
to consider the varying forms of evidence in conjunction with each other.
75
Diet and Dental Health in the Population at Sodo
Most information on Etruscan and Roman diet is quite generalized: subtle
differences between geographic regions and time periods are not well understood.
Certain conclusions about wide-ranging nutrition and eating habits can be drawn,
however, and these points should be kept in mind when interpreting the dental
data.
Compared with the earlier Iron Age “Villanovan” period, the Etruscan
period saw an increase in the number and variety of cereals and legumes
cultivated and also involved the introduction of “tree crops” to the regional diet
(Barker and Rasmussen 1998: 184). Generally, it can be said that Etruscan
agriculture included the cultivation of barley, wheat, millet, and other cereals, as
well as grapes, olives, and figs. The Val di Chiana, in particular, provided some of
the most extensive, fertile, agricultural land in central Italy and the territory of
Cortona derived much of its wealth from a strong agricultural industry (Barker
and Rasmussen 1998: 182, Fracchia 2006: 12, Vallone 1995: 109). Wild
foodstuffs such as fruit (e.g. cherries and pears) and nuts (including hazelnuts and
acorns) were also consumed.
Animal husbandry in the Etruscan era included horses, donkeys, cattle,
sheep, goats, pigs, dogs, and, possibly, chickens. The early (Archaic) Etruscan
period saw an increased emphasis on producing secondary products from these
domesticates, such as cheese, milk, and wool (Barker and Rasmussen 1998: 185).
From the Archaic to the Hellenistic periods, however, the Etruscan diet came to
76
rely increasingly on vegetable and plant foods and less on animal-derived protein
(Forniciari and Mallegni 1987).
By the Roman period, the Val di Chiana was a predominantly open
landscape, allowing for increased agricultural intensity (Barker and Rasmussen
1998: 40). Wheat had emerged as the favoured cereal and the Val di Chiana was a
prime region of wheat production (Barker and Rasmussen 1998: 182, Brothwell
1988). Cereals and legumes were dietary staples for the poorer classes and the
typical meal consisted of wine, root vegetables, legumes, lard, fruit, olives, and
unrefined bread or puls, a porridge-like mixture of cereals, water, salt, and oil
(Bonfiglioli et al. 2003, Brothwell 1988, Brothwell and Brothwell 1998, White
1988).
Due to the small sample size, it is difficult to say much about the diet of
the Sodo population based on dental pathology. Nevertheless, one conclusion
seems evident: the high frequency of caries (100 per cent of individuals and 14
per cent of teeth) appears to be a clear indicator of a diet high in sugars. This
finding confirms the important role of carbohydrates in the Etruscan and Roman
diets, and is consistent with the dominance of the Val di Chiana in cereal-
production throughout this period. Conclusions regarding dental calculus or
antemortem tooth loss are more difficult to interpret, due not only to the size of
the sample but also the predominantly young age of the adults studied.
The relatively high frequency of Linear Enamel Hypoplasia (63 per cent
of individuals and 20 per cent of teeth) indicates that some sort of systemic stress
was relatively prevalent in young children in the Sodo population. This finding
77
supports the commonly held attitude that childhood in antiquity was a hazardous
time, with children at a higher risk for disease and malnutrition than their adult
counterparts (e.g. FitzGerald et al. 2006, Garnsey 1991, Goodman and Armelagos
1989). Although actual numbers are difficult to establish with certainty, Garnsey
(1991) estimates that of infants born alive, 28 per cent died in the first year and 50
per cent died before they reached ten years of age (see also Dupont 1994: 220-
222).
The presence of LEH, however, indicates that the individual survived the
period(s) of stress. We are interested, then, in ailments that could have caused
periods of systemic stress in children but did not necessarily result in death. There
are a number of such illnesses that could have affected children in the Roman
period, including: infantile dysentery, chronic diarrhoea, various skin diseases,
intense fevers (often with associated weight loss and dehydration), and lead
poisoning, which can stunt growth, and lead to anaemia and kidney damage
(Dupont 1994: 255-256, Soranus 2.27-8 in Temkin 1956, Summerton 2007: 29).
One of the most serious diseases known to have been endemic to the region is
malaria, in particular, the strains caused by the parasites Plasmodium malariae
and Plasmodium vivax (Harris 1977: 61-2, Soren 2003). These forms of malaria
are rarely fatal, but can still cause high fever, respiratory distress, diarrhoea,
abdominal pain, cough, and anaemia, symptoms which are particularly severe in
children (Webb 2009, Wilks et al. 2003). Additionally, malaria brought on by P.
malariae can cause the kidneys to leach protein into the urine (nephrotic
syndrome) and, although rare, congenital malaria of the P. vivax strain can occur,
78
presenting as progressive haemolytic anaemia (Wilks et al. 2003). Many these
illnesses could have been endemic to the Sodo population and could have
functioned alone, or in consort with one another, to create the type of systemic
stress that resulted in enamel hypoplasia.
When considering potential aetiologies for LEH it is also important to
consider the role of malnutrition, particularly in relation to weaning practices.
Weaning in the Roman period tended to be associated with problems in general
health and development, as parents were unable to match the nutrition of breast
milk with supplementary foods. Children typically suffered from a shortage of
protein and numerous vitamins, both from the insubstantial nature of the new diet
and from poor food preparation techniques that lead to numerous gastrointestinal
issues (frequently grouped under “weanling diarrhoea”) and subsequent nutrient
malabsorption. Such issues were particularly common in the middle and lower
classes, which would have relied on cereal as the primary weaning food (Garnsey
1991).8 At Rome, babies were typically weaned around three years of age, but
early weaning was often a problem in poor, rural communities where the women
needed to return to work in the fields (Dupont 1994: 20, Garnsey 1991, Soranus
2.21 in Temkin 1956). In addition to causing malnutrition, early weaning or poor
dietary practices9 could impair the development of the child‟s immune system,
8 As demonstrated by the historical and dental evidence discussed above, the Val di Chiana was a
prosperous agricultural region with a diet that relied heavily on grains. Goodman and Armelagos
demonstrated that the decrease in overall health typically associated with a dietary shift to
agriculture is particularly pronounced in children, who consistently exhibit signs of severe
malnutrition in the archaeological record (1989, see also Cohen and Armelagos 1984). 9 Soranus, the Greek physician, advised against allowing newborns to consume the colostrum (the
form of milk produced in late pregnancy and just after childbirth), which is higher in protein and
antibodies, and lower in fat, than regular breast milk (2.12 in Temkin 1956). Additionally,
79
making them more susceptible to disease and infection (Garnsey 1991, Goodman
and Armelagos 1989). Clearly, although any one of these stressors could be
behind the presence of LEH in the study sample, the reality is likely more
complicated, with a combination of factors affecting the growth and overall health
of the children at Sodo.
The limited dental data (and subsequent dietary information) from Sodo is
more informative when viewed in light of other studies of comparable
populations. Although information about diet and nutrition can be generalized
across the Etruscan and Roman territories, specific populations (and groups
within populations) differ in their habitual regimes. In the population from Chiusi,
for example, researchers interpreted a low frequency of caries, high frequency of
AMTL and severe dental wear as being indicative of a diet that was low in
carbohydrates and high in fibre (Capasso 1987, Capasso and Di Tota 1993). The
population from Tarquinia were deemed to have a cereal-dominated diet based on
a high prevalence of caries, and the heavy wear observed in the sample was
attributed to poorly ground flour (Bartoli et al. 1991). The Roman population
from Quadrella was interpreted as having a rather limited diet, composed mainly
of soft carbohydrates, due to a low frequency of wear and a high frequency of
caries (Bonfiglioli et al. 2003).10
When Manzi et al. compared the population from Isola Sacra with that of
Lucus Feroniae, they determined that the Lucus Feroniae diet appeared less
numerous ancient scholars advised against allowing children to eat until full, either as a response
to illness or to encourage proper growth (e.g. Dupont 1994: 223, Soranus 2.25 in Temkin 1956). 10 In this case, however, the authors acknowledged that extensive cleaning and cooking of
typically fibrous foods (e.g. vegetables) can reduce the abrasive effects and make these foods
difficult to see in the skeletal record.
80
differentiated than that of Isola Sacra (1999). This finding was consistent with the
different nature of the populations at the two sites: Isola Sacra was populated
primarily by an urban middle class population that should have had access to a
good supply of “primary foodstuffs”, whereas Lucus Feroniae was a colony of
relatively poor war veterans, slaves and freemen that would have fed mainly on
cereals. In this study, the authors cited the importance of age at onset for
antemortem tooth loss, which tends to be lower in agricultural populations, as
central to their interpretation.
CONCLUSION
The archaeological and documentary evidence indicates that lower-class
Etruscans and Romans had diets consisting of crude bread, puls, root vegetables
and legumes, and fruit. The dental data from the Sodo sample suggests that this
population in particular had a diet high in sugars, most likely derived from starchy
carbohydrates and, potentially, dried fruit. The findings from the Sodo are most
consistent with those of the population from Quadrella, a first to fourth century
AD Roman settlement in the territory of Molise (Bonfiglioli et al. 2003).
Interestingly, the sample of fourth to second century BC Etruscan crania from the
nearby territory of Chiusi displayed an entirely different pattern of dental disease,
indicating a low carbohydrate and high fibre diet (Capasso 1987).
Many of the interpretation issues with the Sodo sample can be resolved
once more Roman burials become available for study. A larger sample size would
allow more conclusions to be drawn regarding the specifics of diet and nutrition,
81
as well as other areas of potential analysis such as dental trauma. Additionally, a
broader skeletal sample, particularly in terms of age at death, would result in a
better understanding of the distribution of the different dental diseases across the
entire population. The turn of the millennium saw the movement of great numbers
of people across the Italian peninsula. The lines between Etruscan and Roman,
and between Roman and the other Italic tribes, became increasingly blurred. The
territory of Cortona was no exception; for several hundred years the immigration
and emigration of individuals left the populace in a state of flux. Further dental
and skeletal analysis is necessary to the development of a comprehensive
understanding of this complex, heterogeneous population.
82
Table 4.1. Individual count comparison of dental disease frequencies in the Sodo
sample.
Caries
n (%)
AMTL
n (%)
Hypoplasia
n (%)
Calculus
n (%)
Sex Male 5 (100) 1 (20) 3 (60) 1 (20)
Female 3 (100) 0 2 (67) 2 (67)
Age Young Adult 6 (100) 0 4 (67) 3 (50)
Middle Adult 2 (100) 1 (50) 1 (50.0) 0
Total 8 (100) 1 (13) 5 (63) 3 (38) n = number of affected individuals; (%) = number of affected individuals/number of observable
individuals x 100, rounded to the nearest full percentage; AMTL = Antemortem tooth loss.
Table 4.2. Tooth count prevalence of caries in the Sodo sample.
na no %
Tooth class Incisor 0 17 0
Canine 0 18 0
Premolar 2 38 5
Molar 16 52 31
Sex Male 11 75 15
Female 7 50 14
Age Young Adult 16 109 15
Middle Adult 2 16 13
Jaw Maxilla 10 70 14
Mandible 8 55 15
Total 18 125 14 na = number of affected teeth; no = number of observable teeth; (%) = number of affected
teeth/number of observable teeth x 100, rounded to the nearest full percentage.
83
Table 4.3. Tooth count prevalence of antemortem tooth loss in the Sodo sample.
na no %
Tooth class Incisor 0 35 0
Canine 0 23 0
Premolar 0 43 0
Molar 3 58 5
Sex Male 3 101 3
Female 0 58 0
Age Young Adult 0 131 0
Middle Adult 3 28 11
Jaw Maxilla 0 87 0
Mandible 3 72 4
Total 3 159 2 na = number of affected tooth sockets; no = number of observable tooth sockets; (%) = number of
affected tooth sockets/number of observable tooth sockets x 100, rounded to the nearest full
percentage.
Table 4.4. Tooth count prevalence of enamel hypoplasia in the Sodo sample.
na no %
Tooth class Incisor 3 17 18
Canine 8 17 47
Premolar 2 38 5
Molar 11 51 22
Sex Male 20 73 27
Female 5 50 10
Jaw Maxilla 11 68 16
Mandible 14 55 25
Total 25 123 20 na = number of affected teeth; no = number of observable teeth; (%) = number of affected
teeth/number of observable teeth x 100, rounded to the nearest full percentage.
84
Table 4.5. Tooth count prevalence of calculus in the Sodo sample.
na no %
Tooth class Incisor 4 17 24
Canine 3 17 18
Premolar 6 38 16
Molar 5 52 10
Sex Male 12 74 16
Female 6 50 12
Age Young Adult 18 108 17
Middle Adult 0 16 0
Jaw Maxilla 4 69 6
Mandible 14 55 25
Total 18 124 15 na = number of affected teeth; no = number of observable teeth; (%) = number of affected
teeth/number of observable teeth x 100, rounded to the nearest full percentage.
85
Figure 4.1. The cranial remains from SII-2 after cleaning. Remains were highly
fragmentary.
Figure 4.2. A long bone fragment from SII-1. The fragile nature of the bone is
visible, as is the prolific infiltration by plant roots (arrows).
86
Figure 4.3. A long bone cross section from SI-3 under 2.5x magnification. The
invasive organic matter appears black.
87
Figure 4.4. The same long bone cross section from SI-3 under 20x magnification.
At this level, the globular appearance of the organic matter is visible, as is the
extent of the infiltration.
88
Figure 4.5. A gross carious lesion in the right first molar of SI-3.
Figure 4.6. Antemortem loss of the lower left first and second molars in SII-3.
The rounded cavity of the abscessed second molar is visible (arrow).
89
Figure 4.7. Linear enamel hypoplasia in the left mandibular canine of SII-1
(arrows).
Figure 4.8. Slight calculus on the lingual surface of the left mandibular teeth in
SI-3 (arrows).
90
WORKS CITED
Alexander D. 1984. The reclamation of Val-di-Chiana (Tuscany). Annals of the
Association of American Geographers 74(4): 527-550.
Barker G, Rasmussen T. 1998. The Etruscans. Oxford: Blackwell Publishers Ltd.
Bartoli F, Mallegni F, Vitello A. 1991. Indagini nutrizionali e
odontostomatologiche per una definizione della dieta alimentare in un
gruppo umano a cultura etrusca: gli inumati della necropoli dei
Monterozzi di Tarquinia (VI-II sec. A.C.). Studi Etruschi 56: 255-269.
Bass WM. 2005. Human Osteology: A Laboratory and Field Manual. 5th
edition.
Columbia (MO): Missouri Archaeological Society.
Beiswanger BB, Segreto VA, Mallatt ME, Pfeiffer HJ. 1989. The prevalence and
incidence of dental calculus in adults. Journal of Clinical Dentistry 1(3):
55-58.
Bonfante L. 1986. Introduction/Etruscan studies today. In: Bonfante L, editor.
Etruscan Life and Afterlife. Detroit (MI): Wayne State University Press,
pp1-17.
Bonfiglioli B, Brasili P, Belcastro MG. 2003. Dento-alveolar lesions and
nutritional habits of a Roman Imperial Age population (1st-4
th c. AD):
Quadrella (Molise, Italy). HOMO 54(1): 36-56.
Brothwell DR. 1981. Digging up Bones. 3rd
Edition. Ithaca (NY): Cornell
University Press.
Brothwell DR. 1988. Foodstuffs, cooking and drugs. In: Grant M, Kitzinger R,
editors. Civilization of the Ancient Mediterranean. New York (NY):
Charles Scribner‟s Sons, pp247-261.
Brothwell D, Brothwell P. 1998. Food in Antiquity: A Survey of the Diet of Early
Peoples. Expanded Edition. Baltimore (MD): Johns Hopkins University
Press.
Brothwell DR, Carr HG. 1962. The dental health of the Etruscans. British Dental
Journal 113: 207-210.
Bruschetti P, Zamarchi Grassi P. 1999. Cortona Etrusca: Esempi di architettura
funeraria. Cortona: Calosci-Cortona.
91
Buikstra JE, Ubelaker DH, editors. 1994. Standards for Data Collection From
Human Skeletal Remains. Fayetteville (AR): Arkansas Archaeological
Survey, Research Series No. 44.
Capasso L. 1987. Dental pathology and alimentary habits reconstruction of
Etruscan population. Studi Etruschi 53: 177-191.
Capasso L, Di Tota G. 1993. Etruscan teeth and odontology. Dental Anthropology
Newsletter 8(1): 4-7.
Carlson SJ. 1990. Vertebral dental structures. In: Carter JG, editor. Skeletal
Biomineralization: Patterns, Processes and Evolutionary Trends. New
York (NY): Van Nostrand Reinhold, pp531-556.
Cohen MN, Armelagos GJ, editors. 1984. Paleopathology at the Origins of
Agriculture. New York (NY): Academic Press.
Dupont F. 1994. Daily Life in Ancient Rome. Woodall C, translator. Oxford:
Blackwell Publishers Ltd.
FitzGerald C, Saunders S, Bondioli L, Macchiarelli R. 2006. Health of infants in
an Imperial Roman skeletal sample: Perspective from dental
microstructure. American Journal of Physical Anthropology 130: 179-189.
Forniciari G, Mallegni F. 1987. Indagini paleonutrizionali su campioni di
popolazioni a cultura etrusca. In Barbieri, G, editor. L‟alimentazione nel
mondo antico: gli etruschi. Rome: Istituto Poligrafico e Zecca dello Stato,
pp135-139.
Fracchia H. 2006. The Villa at Ossaia and the Territory of Cortona in the Roman
Period. Siracusa: Lombardi editori.
Garnsey P. 1991. Child rearing in ancient Italy. In: Kertzer DI, Saller RP, editors.
The Family in Italy: From Antiquity to Present. New Haven (CT): Yale
University Press, pp48-65.
Goodman AH, Armelagos GJ. 1989. Infant and childhood morbidity and
mortality risks in archaeological populations. World Archaeology 21(2):
225-243.
Gordon CC, Buikstra JE. 1981. Soil pH, bone preservation, and sampling bias at
mortuary sites. American Antiquity 46(3): 566-571.
92
Hare PE. 1980. Organic geochemistry of bone and its relation to the survival of
bone in the natural environment. In Behrensmeyer AK, Hill AP, editors.
Fossils in the Making. Chicago (IL): University of Chicago Press, pp208-
219.
Harris WV. 1971. Rome in Etruria and Umbria. Oxford: Clarendon Press.
Henderson J. 1987. Factors determining the state of preservation of human
remains. In: Boddington A, Garland AN, Janaway RC, editors. Death,
Decay and Reconstruction: Approaches to Archaeology and Forensic
Science. Manchester: Manchester University Press, pp43-54.
Hillson S. 1979. Diet and dental disease. World Archaeology 11(2): 147-162.
Hillson S. 1996. Dental Anthropology. Cambridge: Cambridge University Press.
Manzi G, Salvadei L, Vienna A, Passarello P. 1999. Discontinuity of life
conditions at the transition from the Roman Imperial Age to the Early
Middle Ages: Example from central Italy evaluated by pathological dento-
alveolar lesions. American Journal of Human Biology 11: 327-341.
Meindl RS, Lovejoy CO. 1985. Ectocranial suture closure: A revised method for
the determination of skeletal age at death based on the lateral anterior
sutures. American Journal of Physical Anthropology 68: 57-66.
Millard A. 2001. The deterioration of bone. In: Brothwell DR, Pollard AM,
editors. Handbook of Archaeological Sciences. Chichester: John Wiley &
Sons, Ltd., pp637-647.
Moggi-Cecchi J, Pacciani E, Chiarelli B, D‟Amore G, Brown K. 1997. The
anthropological study of Etruscan populations. Etruscan Studies 4(1): 73-
86.
Nicholson RA. 2001. Taphonomic investigations. In: Brothwell DR, Pollard AM,
editors. Handbook of Archaeological Sciences. Chichester: John Wiley &
Sons, Ltd., pp179-190.
Nielsen E. 1984. Some observations on early Etruria. In: Hackens T, Holloway
ND, Holloway RR, editors. Crossroads of the Mediterranean.
Archaeologica Transatlantica 2. Louvain-La-Neuve, Belgium:
UniversitéCatholique de Louvain/Providence (RI): Brown University,
pp255-260.
93
Nielsen-Marsh C, Gernaey A, Turner-Walker G, Hedges R, Pike A, Collins M.
2000. The chemical degradation of bone. In: Cox M, Mays S, editors.
Human Osteology in Archaeology and Forensic Science. London:
Greenwich Medical Media, pp439-454.
Ortner DJ. 2003. Identification of Pathological Conditions in Human Skeletal
Remains. 2nd
Edition. San Diego (CA): Academic Press.
Pacciani E, Chiarelli B, D‟Amore G, Moggi-Cecchi J. 1996. Paleobiology of
Etruscan populations. Human Evolution 11(2): 159-170.
Pacciani E, Mainardi S. 2006. I reperti scheletrici umani della tomba tardo arcaica
del tumulo II del Sodo. In: Zamarchi Grassi P, editor. Il Tumulo II del
Sodo di Cortona: La tomba di età tardo arcaica. Cortona: Accademia
Etrusca, pp105-111.
Parker RB, Toots H. 1980. Trace elements in bones as paleobiological indicators.
In: Behrensmeyer AK, Hill AP, editors. Fossils in the Making. Chicago
(IL): University of Chicago Press, pp197-207.
Pitre MC. 2011. Microbial Biodeterioration of Human Skeletal Material from Tell
Leilan, Syria (2900 – 1900 BCE) [PhD thesis]. Edmonton (AB):
University of Alberta.
Rugg-Gunn AJ. 1993. Nutrition and Dental Health. Oxford: Oxford Medical
Publications.
Sandford MK, Weaver DS. 2000. Trace element research in anthropology: New
perspectives and challenges. In: Katzenberg MA, Saunders SR, editors.
Biological Anthropology of the Human Skeleton. New York (NY): Wiley-
Liss Inc., pp329-350.
Sillen A. 1989. Diagenesis of the inorganic phase of cortical bone. In: Price TD,
editor. The Chemistry of Prehistoric Human Bone. Cambridge: Cambridge
University Press, pp211-229.
Smith BH, 1984. Patterns of molar wear in hunter-gatherers and agriculturalists.
American Journal of Physical Anthropology 63: 39-56.
Soren D. 2003. Can archaeologists excavate evidence of malaria? World
Archaeology 35(2): 193-209.
Spivey N, Stoddart S. 1990. Etruscan Italy. London: B.T. Batsford Ltd.
Summerton N. 2007. Medecine and Health Care in Roman Britain. Shire
Archaeology Series No. 87. Buckinghamshire: Shire Publications Ltd.
94
Temkin O, translator. 1956. Soranus‟ Gynecology. Baltimore (MD): The John
Hopkins University Press.
Torelli M. 1975. Introduction. In: Boitani F, Cataldi M, Pasquinucci M. Etruscan
Cities. Atthill C, Evans E, Pleasance S, Swinglehurst P (trans.) London:
Cassell.
Toynbee JMC. 1971. Death and Burial in the Roman World. London: Thames and
Hudson.
Turner-Walker G, Jans M. 2008. Reconstructing taphonomic histories using
histological analysis. Palaeogeography, Palaeoclimatology,
Palaeoecology 266: 227-235.
Vallone MT. 1995.Cortona of the principes. Etruscan Studies 2: 109-139.
Von Endt DW, Ortner DJ. 1984. Experimental effects of bone size and
temperature on bone diagenesis. Journal of Archaeological Science 11:
247-253.
Webb JLA Jr. 2009. Humanity‟s Burden: A Global History of Malaria.
Cambridge (NY): Cambridge University Press.
White KD. 1988. Farming and animal husbandry. In: Grant M, Kitzinger R,
editors. Civilization of the Ancient Mediterranean. New York (NY):
Charles Scribner‟s Sons, pp211-245.
Wilks D, Farrington M, Rubenstein D. 2003. The Infectious Diseases Manual. 2nd
Edition. Malden (MA): Blackwell Science Ltd.
Zamarchi Grassi P, editor. 2010. The II° Tumulus at Sodo. Ministero per I Beni
Culturali e Ambientali. Soprintendenza ai Beni Archeologici Della
Toscana. Sezione Didattica.
95
CHAPTER 5: TERONTOLA
INTRODUCTION
In 2003 the skeletal remains of two individuals were discovered during an
archaeological survey prior to the construction of a carabinieri station on the Via
Combattenti in Terontola. Only several hundred metres east of the SS71/SR71
that follows the course of the Roman Via Cassia, the remains were found in a
cappuccina graves, near the probable foundations of a funerary monument dating
to the Early Imperial Age1 (Gori et al. 2007). Nearby, to the west of the Terontola
train station and north of the local Landrucci road (and the Rio Cesa that runs
parallel to it), archaeological excavation has revealed the presence of a borgata
rurale (rural township), likely associated with a nearby villa (Figure 5.1).
Preliminary excavation of the borgata has revealed additional a cappuccina
tombs, also dating to the Early Imperial Age (H. Fracchia, pers. comm., Gori et
al. 2007).
In June of 2010, one of the skeletons, an Older Adult male, from the site
of the carabinieri station in Terontola was made available for study at the Museo
dell‟Accademia Etrusca e della Città di Cortona (MAEC).
BURIAL DETAILS
The individual, labeled T-1 by the author, was delivered to the MAEC in
extended, supine position on terracotta roof tiles, consistent with the a cappuccina
form of burial (Figure 5.2). The first roof tile held the skull, the upper axial
1 The relation of the funerary monument to the burials is not well understood at this time.
96
skeleton and humeri. The second roof tile held the lower axial skeleton (including
pelvis), the radii and ulnae, and upper portions of the femorae which had been
severed mid-shaft postmortem. Presumably, a third roof tile held the lower leg
and foot bones but whether this portion of the burial has been excavated is not
known. As with the Sodo material, the preservation of T-1 was poor: skeletal
elements were highly fragmentary and encased in hardened mud2 (Figures 5.3 and
5.4).
The individual T-1 was identified as an Older Adult (50+ years) by the
high degree of ectocranial suture closure (Meindl and Lovejoy 1985), the
presence of arachnoid granulations on the endocranial surface (Basmajian 1952,
Yew et al. 2011), and severe tooth wear (Brothwell 1981, Meindl and Lovejoy
1985)3. Although none of these features alone is a sure indicator of age, together
they provide strong, independent evidence for categorization of T-1 as an Older
Adult. T-1 was identified as male based on pronounced brow ridges, large
mastoid processes, a narrow and deep sciatic notch, and the overall degree of
skeletal robusticity (anatomical features were evaluated according to the standards
proposed by Buikstra and Ubelaker 1994).
DENTAL PATHOLOGY
Eight teeth from the dentition of T-1 were unobservable: the left M3 and
P2 were clearly lost postmortem, while five teeth could not be classified as neither
2 The soil encasing T-1 had a pH of 7.04 and was a clay loam, according to the United States
Department of Agriculture soil texture guidelines. 3 Use of tooth-wear to estimate age at death will be elaborated on below.
97
the teeth nor the alveoli were present for examination4. A sixth tooth, reduced by
postmortem damage to a minute root fragment, could not be identified and was
omitted from the study. From the remaining 24 teeth, T-1 was determined to have
suffered from caries, antemortem tooth loss and calculus. No enamel hypoplasia
or abscessing was observed.
The prevalence of caries by tooth count is listed in Table 5.1. All six
carious lesions were interproximal (Figure 5.5). The relatively low frequency of
caries in the molars (a common site for carious lesions) is likely explained by the
presence of only two of the twelve molars, the rest were lost either antemortem or
postmortem.5
A total of seven teeth were lost antemortem (Table 5.2). The right P2, M1,
M2, and M3 were lost significantly before death, allowing for substantial re-
modeling of the mandibular bone (Figure 5.6). The left M2 was lost more recently
than the lower right teeth, although the alveolus had still filled with woven bone
prior to death. The upper teeth: the right C and I1, and the left I
1, I
2, C, P
1, and P
2
were, likewise, all lost significantly before death. Both upper canines were clearly
lost due to abscessing, determinable from the expanded round alveoli at the points
of infection. The cause of AMTL for the other teeth is unidentifiable but
significant dental wear and the high frequency of carious lesions in the remaining
teeth indicates that caries, exacerbated by severe wear, is a likely cause.
4 It is possible that the missing M3s in this individual were congenitally absent, a trait that is not
uncommon in Etruscan populations (e.g. see Becker et al. 2009). 5 A number of attempts have been made to compensate for the effects of AMTL on caries
frequencies, most notably the caries correction factor (CCF) put forward by Lukacs (1995). The
CCF was not applied to T-1 for two reasons: 1) it was designed for application to a skeletal
population, not a single individual, and 2) the accuracy of the CCF would be compromised by the
high degree of postmortem tooth-loss and the poor preservation of T-1‟s dentition.
98
The high frequencies of caries and antemortem tooth loss in T-1 are
consistent with an agrarian diet high in starches and sugars, similar to the Sodo
sample. The considerably higher prevalence of AMTL in T-1 is likely attributable
to his advanced age. As outlined in Chapter 4, AMTL is a gradual process, and we
would expect it to be more common in Older Adults than in Middle or Younger
Adults.
Slight calculus was observed on a single tooth, the right I1, in this
individual (Figure 5.7, Table 5.3). The lingual surface of the lower incisors and
the buccal surface of the molars are the two regions typically most susceptible to
calculus formation in humans (Hillson 1996: 254-60). It is curious, given the
advanced age of the individual, that the I1 was the only tooth on which calculus
was observed, especially given the relatively good preservation of all the lower
incisors in the mandibular bone.
TOOTH WEAR ANALYSIS
Dental wear is produced by a combination of two general processes:
attrition, which is produced by tooth-on-tooth contact, and abrasion, which is
produced when the tooth comes into contact with a foreign object (frequently
food). The degree of tooth wear is progressive, and is affected by many
conditions, including the age of the individual, the eruption sequence within the
dentition, and any pathological conditions that might have affected use of the
tooth.
99
Wear due to mastication is a complex process produced by a combination
of attrition and abrasion. The nature of the food being consumed can have a
profound influence on tooth wear as it both affects the degree and type of chewing
that take place (attrition) and the food particles that come into contact with the
teeth (abrasion). Dental wear patterns can therefore be useful for interpreting the
abrasiveness of a particular diet. Detailed inferences are difficult however, as
many different types of food can wear down the tooth surfaces and severe wear
can be indicative of anything from rock granules in flour, to bone and animal
tissue (Hillson 1979). Despite these difficulties, a thorough examination of the
rate and pattern of wear can provide an understanding of diet beyond its general
abrasiveness. Smith (1984), for example, showed that hunter-gatherers tend to
display flatter molar wear due to the mastication of tough and fibrous foods, while
agriculturalists display an oblique molar wear pattern due to preparation
techniques that include grinding and cooking with water.
The use of teeth as tools and the intentional modification of teeth for
aesthetic or other purposes can also complicate interpretations of dental wear. The
anterior teeth in particular tend to be used for a variety of tasks related to food or
craft processing and leisure (pipe smoking, etc.), and are often singled out for
intentional modification (Hillson 1996). In Etruscan and Roman societies,
intentional modification often involved the creation and application of elaborate
dental appliances of gold or other materials, and false teeth (e.g. Becker 1994,
Becker 1999, Crubézy et al. 1998).
100
Due to its progressive nature, many attempts have been made to use tooth
wear as evidence in age estimation (e.g. Brothwell 1981, Brothwell 1989,
Lovejoy 1985). Many of these attempts have been problematic, however, as tooth
wear is degenerative, and is a direct indicator of usage, not age. A young
individual with an exceptionally abrasive diet will display more severe wear than
an older individual with an exceptionally soft diet. Nevertheless, when applied to
a single population with a consistent diet, tooth wear can be effectively used as an
indicator of general age (e.g. Lovejoy 1985, Mays 2002, Walker et al. 1991).
Ultimately, if the varying forms of evidence are considered in conjunction with
one another, dental wear analysis can yield important information about diet,
habitual activity, and age in archaeological populations.
The mandibular anterior teeth of T-1 showed very light wear, scoring a 3-
4 on the Smith scoring scale, suggesting early loss of the corresponding maxillary
teeth (Figures 5.7 and 5.8, Table 5.4). The wear on both the upper and lower
posterior teeth was more severe, with the premolars ranging between 5 and 7, and
the molars between 7 and 8 (Figure 5.9). The presence of secondary dentine in all
observable teeth indicates the wear occurred gradually, over a long period of time
(Hillson 1996). Although the issues with using tooth wear to determine age are
clear, the extensive wear observed in T-1 is consistent with his categorization as
an Older Adult, particularly when compared to the minimal wear observed in the
Sodo sample (see Chapter 4). Additionally, it was observed that T-1‟s maxillary
premolars and molars were worn obliquely, to the point that portions of the roots
101
functioned in occlusion (Figure 5.10). Such a wear pattern is consistent with
Smith‟s observations on agricultural societies.
CONCLUSION
The Terontola excavation is only 9 kilometres southeast of the Sodo, and
is connected to the tumuli by the modern, Roman, and Etruscan road that runs
adjacent to both sites. Dating of the Terontola and Sodo burials based on grave
goods places them all within the Early Imperial period (first century BC to first
century AD). Moreover, T-1 was buried in the same manner as the individuals
from the Sodo, a cappuccina, indicating that he, too, was likely an individual of
low status.
The individual from Terontola is an Older Adult, an age category not
represented in the Sodo collection, and thus serves as a useful compliment to the
material discussed in Chapter 4. The high frequency of caries in the dentition of
T-1 is consistent with a diet similar to that of the Sodo population: high in
starches and other sugars. The substantial number of teeth lost antemortem by T-1
likely provides an indication of what could be expected in Older Adult individuals
from the Sodo, although this hypothesis can only be tested if more individuals
become available at a future date.
102
Table 5.1. Tooth count prevalence of caries in the individual from Terontola.
na no %
Tooth class Incisor 0 4 0
Canine 2 2 100
Premolar 3 4 75
Molar 1 2 50
Jaw Maxilla 3 3 100
Mandible 3 9 33
Total 6 12 50 na = number of affected teeth; no = number of observable teeth; (%) = number of affected
teeth/number of observable teeth x 100, rounded to the nearest full percentage.
Table 5.2. Tooth count prevalence of antemortem tooth loss in the individual
from Terontola.
na no %
Tooth class Incisor 3 7 43
Canine 2 4 50
Premolar 3 8 38
Molar 4 7 57
Jaw Maxilla 7 10 70
Mandible 5 16 31
Total 12 26 46 na = number of affected tooth sockets; no = number of observable tooth sockets; (%) = number of
affected tooth sockets/number of observable tooth sockets x 100, rounded to the nearest full
percentage.
103
Table 5.3. Tooth count prevalence of calculus in the individual from Terontola.
na no %
Tooth class Incisor 1 4 25
Canine 0 2 0
Premolar 0 4 0
Molar 0 2 0
Jaw Maxilla 0 3 0
Mandible 1 9 11
Total 1 12 8 na = number of affected teeth; no = number of observable teeth; (%) = number of affected
teeth/number of observable teeth x 100, rounded to the nearest full percentage.
Table 5.4. Tooth wear analysis in the individual from Terontola.
Tooth Category Wear Score
Maxilla Incisors Missing
Canines Missing
Premolars RP1 and RP2 = 7
left Ps missing
Molars RM1 = 7
other Ms missing
Mandible Incisors 3-4
Canines 4
Premolars LP1 =5
RP1 =6
P2s missing
Molars LM2=8
other Ms missing Values assigned according to the system outlined by Smith (1984).
104
Fig
ure
5.1
. M
ap o
f T
eronto
la s
how
ing t
he
loca
tion o
f th
e buri
al u
nder
the
curr
ent
cara
bin
ieri
sta
tion. A
lso p
ictu
red i
s th
e lo
cati
on
of
the
borg
ata
rura
le, a
sourc
e of
sever
al c
onte
mpora
neo
us
buri
als.
(©
2012 G
oogle
).
105
Figure 5.2. T-1 as received at the MAEC, in extended supine position on two
terracotta roof tiles.
106
Figure 5.3. The cranium and mandible of T-1. The remains were highly
fragmentary.
Figure 5.4. Left femur of T-1. The femoral shafts of T-1 were severed
postmortem and the medullary cavities infiltrated by mud (pictured: left).
107
Figure 5.5. Interproximal carious lesions in the upper right second and third
premolars (arrows).
Figure 5.6. Antemortem loss of the lower right second premolar and molars.
Bone remodelling is incomplete but well advanced, indicating that tooth loss
occurred significantly before death.
108
Figure 5.7. Slight calculus on the lingual surface of the lower right lateral incisor
(arrow). Note also the minimal wear on the right incisors, canine, and first
premolar.
Figure 5.8. Smith‟s system for scoring stages of tooth wear. Reprinted from
Smith (1984).
109
Figure 5.9. Severe wear on the lower left second molar.
Figure 5.10. The wear on this upper right first molar is severe, scored as a 7 on
Smith‟s scale. Additionally, the tooth was worn obliquely, so that the chewing
surface extended past the cemento-enamel junction to include part of the root
(arrow).
110
WORKS CITED
Basmajian JV. 1952. The depressions for the arachnoid granulations as a criterion
of age. The Anatomical Record 112(4): 843-846.
Becker MJ. 1994. Etruscan gold dental appliances: Origins and functions as
indicated by an example from Orvieto, Italy, in the Danish National
Museum. Dental Anthropology Newsletter 8(3): 2-8.
Becker MJ. 1999. Etruscan gold dental appliances: Three newly “discovered”
examples. American Journal of Archaeology 103(1): 103-111.
Becker MJ, Turfa JM, Algee-Hewitt B. 2009. Human Remains from Etruscan and
Italic Tomb Groups in the University of Pennsylvania Museum. Biblioteca
di Studi Etruschi No. 48. Pisa: Fabrizio Serra Editore.
Brothwell DR. 1981. Digging Up Bones. 3rd
Edition. Ithaca (NY): Cornell
University Press.
Brothwell DR. 1989. Relationship of tooth wear to aging. In: Iscan MY, editor.
Age Markers in the Human Skeleton. Springfield (IL): Charles Thomas,
pp303-316.
Buikstra JE, Ubelaker DH, editors. 1994. Standards for Data Collection From
Human Skeletal Remains. Fayetteville (AR): Arkansas Archaeological
Survey, Research Series No. 44.
Crubézy E, Murail P, Girard L, Bernadou JP. 1998. False teeth of the Roman
world. Nature 391: 29.
Gori S, Guidelli F, Salvi A. 2007. Località Terontola Stazione, Cortona-proprietà
Farina. In: Notiziario della Soprintendenza per i Beni Archeologici della
Toscana 2/2006: Scavi e Ricerche sul Territorio. Firenze: All‟Insegna del
Giglio s.a.s., pp186-187.
Hillson S. 1979. Diet and dental disease. World Archaeology 11(2): 147-162.
Hillson S. 1996. Dental Anthropology. Cambridge: Cambridge University Press.
Lovejoy CO. 1985. Dental wear in the Libben population: Its functional pattern
and role in the determination of adult skeletal age at death. American
Journal of Physical Anthropology 68: 47-56.
Lukacs JR. 1995. The „caries correction factor‟: A new method of calibrating
dental caries rates to compensate for antemortem loss of teeth.
International Journal of Osteoarchaeology 5: 151-156.
111
Mays S. 2002. The relationship between molar wear and age in an early 19th
century AD archaeological human skeletal series of documented age at
death. Journal of Archaeological Science 29: 861-871.
Meindl RS, Lovejoy CO. 1985. Ectocranial suture closure: A revised method for
the determination of skeletal age at death based on the lateral anterior
sutures. American Journal of Physical Anthropology 68: 57-66.
Smith BH. 1984. Patterns of molar wear in hunter-gatherers and agriculturalists.
American Journal of Physical Anthropology 63: 39-56.
Walker PL, Dean G, Shapiro P. 1991. Estimating age from tooth wear in
archaeological populations. In: Kelley M, Larsen CS, editors. Advances in
Dental Anthropology. New York (NY): Wiley-Liss, pp169-178.
Yew M, Dubbs B, Tong O, Nager GT, Niparko JK, Tatlipinar A, Francis HW.
2011. Arachnoid granulations of the temporal bone: A histologic study of
dural and osseous penetration. Ontology & Neurontology 32: 602-609.
112
CHAPTER 6: CONCLUSIONS
This study examines pathological conditions in the dentitions of Roman
period individuals excavated from the Sodo tumuli and Terontola. The results
help to shed light on the diet and overall health of the population of the territory
of Cortona between the first century BC and the first century AD. Although the
bulk of the data come from the Sodo sample, the individual from Terontola
provides an example of an Older Adult, a demographic currently lacking in the
Sodo collection.
The individuals buried at Sodo suffered from a high prevalence of caries.
Comparison of these preliminary data with those of other, similar studies
demonstrates that despite widespread trade and shared dietary staples, food
consumption varied significantly from region to region throughout the peninsula.
The Sodo sample consumed a diet that was particularly high in carbohydrates and
other sugars, even relative to individuals from the nearby territory of Chiusi. That
two populations with such varying frequencies of dental disease could be located
in such close proximity to one another in the Val di Chiana outlines the
importance of a regional approach to understanding the Roman world.
The high frequency of enamel hypoplasia indicates a widespread
prevalence of disease and malnutrition in children from the population at Sodo,
consistent with issues faced by children elsewhere in classical antiquity. Although
enamel hypoplasia is a nonspecific indicator of systemic stress, we can
hypothesize that poor weaning practices, as well as diseases such as infantile
dysentery, chronic diarrhoea, and malaria may have been potential sources of
113
poor health. While the presence of enamel hypoplasia indicates that the
individuals at Sodo survived these periods of childhood stress, the proportion of
children that did not survive to adulthood is currently unknown.
Analysis of the Sodo material was limited by the small sample size and
poor preservation of the skeletal remains. If more of the Roman-period burials
from Sodo become available, greater conclusions may be drawn about the
frequency of dental disease, thus strengthening the interpretations of diet and
health. Additionally, future analyses could elaborate on the findings of this study
significantly by comparing them to earlier Etruscan and later Roman period data
from the territory of Cortona. Such a comparison would allow us to track changes
in diet and health in this region over the entire period of Romanization, thus
providing invaluable data on changing lifeways in the territory of Cortona
throughout this process. Finally, comparison of the Roman-period Sodo
individuals with other individuals from the territory of Cortona may reveal
information on why these particular individuals were buried at the feet of the
grand tumuli of the Etruscan principes.
114
APPENDIX A - SUMMARY OF BURIAL INFORMATION
BURIAL NUMBER
AGE
SEX
DENTAL DATA
C-1 Young Adult Female Yes
SI-1 Young Adult Female Yes
SI-2 Middle Adult Unknown No
SI-3 Young Adult Male Yes
SII-1 Young Adult Male Yes
SII-2 Young Adult Female Yes
SII-3 Middle Adult Male Yes
SII-4 Unknown Unknown No
SII-5 Unknown Unknown No
SII-6 Young Adult Male Yes
SII-7a Unknown Unknown No
SII-7b Middle Adult Male Yes
SII-8 Unknown Unknown No
SII-9 Unknown Unknown No
T-1 Older Adult Male Yes
115
APPENDIX B - THE DIAGENESIS OF BONE IN
GROUNDWATER
INTRODUCTION
The poor state of preservation of the Sodo material hampered data
collection and limited the information that was obtainable from the remains. The
myriad of issues typical of waterlogged sites is a familiar tale in the
bioarchaeological literature: prolonged contact with groundwater has been found
to compromise the integrity of skeletal remains and elaborate drying and
consolidating strategies have been employed in an effort to minimize damage
(e.g. Chaplin 1971: 18, Johnson 1994, Koob 1992, Rahtz and Hirst 1976, Stone et
al. 1990). Most of the publications that discuss the taphonomic effects of water,
however, focus on fluvial transport, diagenesis in marine environments, and the
effects of freeze/thaw cycles, rather than the diagenetic effect of groundwater on
bone chemistry and structure (but cf. Hedges and Millard 1995, Nielsen-Marsh et
al. 2000, Pike et al. 2001). As such, most diagenetic changes are not well
understood (Hedges and Millard 1995, Millard 2001: 642). This appendix will
review the basics of hydrology and bone structure, as well as the known processes
of bone diagenesis, to provide a better understanding of the factors behind the
preservation issues with Sodo material.
In biological anthropology, diagenesis typically refers to the physical and
chemical changes that affect skeletal material in a burial context. Along with
mortuary practices, weathering, modification by biological organisms and other
processes, it falls under the sphere of “taphonomy”: a term coined by the Russian
116
palaeontologist J.A. Efremov in 1940 from the Greek taphos (burial) and nomos
(laws). Taphonomy was originally defined as “…the study of the transition (in all
its details) of animal remains from the biosphere into the lithosphere…” (Efremov
1940: 85) but has since come to encompass all attempts to deal with the biased, or
incomplete, nature of any fossil, skeletal, or archaeological assemblage. With
varying emphasis on diagenesis, taphonomy has become an essential subfield of
many of disciplines, including palaeontology, forensic archaeology,
zooarchaeology, and bioarchaeology (e.g. Behrensmeyer and Hill 1980, Haglund
and Sorg 1997, Lyman 1994: 12-33, Stodder 2008). Diagenesis has received
special attention in recent decades in the literature on palaeodietary reconstruction
and palaeoecology, due to its effects on trace element and stable isotope analyses
(e.g. see Lambert et al. 1985a, Lambert et al. 1985b, Sandford and Weaver 2000,
Sillen 1989).
Diagenesis of skeletal tissue involves the hydrolysis, dissolution, mineral
replacement and recrystallization of its constituent parts.1 These processes are
highly variable and can result in major or minor changes to both the organic and
inorganic components of bone (Hare 1980, Von Endt and Ortner 1984). The
chemical reactions are influenced by factors that are intrinsic to the bone tissue
such as size, porosity, and chemical structure, and extrinsic, such as duration of
interment, sediment pH, climate, and aeration (Henderson 1987, Nicholson 2001,
Sandford and Weaver 2000: 335, Sillen 1989: 212, Von Endt and Ortner 1984).
1 In some publications, diagenesis also includes the microbial decomposition of collagen, in addition to the
abiotic processes. For the purposes of this article, however, biotic factors are not included in diagenesis and
will only be briefly considered (see Child 1995 and Sillen 1989 for a more detailed discussion of the
degradation of bone by microbial agents).
117
Generally, the deeper the burial, the better the preservation of the skeletal
remains: an increase in depth is accompanied by a reduction in insect activity,
temperature fluctuations, and other harmful effects of the burial environment
(Henderson 1987: 52). However, the advantages of a greater depth are
complicated if the deeper soil is waterlogged.
Water, soil, temperature, and many other facets of the burial environment
function together to accelerate (or limit) bone diagenesis. Focusing on a single
aspect is simplistic and can provide an inaccurate understanding of the processes
involved (Henderson 1987). Nevertheless, water is widely accepted to be one of
the most important agents of deterioration (Goffer 1980: 239-46, Millard 2001,
Nawrocki 1995, Nielsen-Marsh et al. 2000) and an analysis of diagenesis is
woefully inadequate without proper consideration of the hydrology of the burial
environment.
GROUNDWATER
Groundwater can be defined as “all the water contained in spaces within
bedrock and regolith,” where the regolith is the layer of loose rock and mineral
fragments that have separated from the original bedrock due to weathering
(Skinner and Porter 1987: 241). The total global volume of groundwater
(measured to a depth of 4 kilometres) is 8,350,000 cubic kilometres, more than
half of which can be found within 750 metres of the surface although this
abundance certainly varies by geographical location (Ibid.: 240-2). Geologists
recognize two distinct subterranean regions, defined by the incidence of
118
groundwater: the unsaturated zone, or zone of aeration, and the saturated zone
(Holmes 1962: 402, Skinner and Porter 1987: 240-2). The zone of aeration is that
in which the open spaces of the regolith and bedrock contain predominantly air,
although water (termed vadose water) is present occasionally, such as in the
period following a rain. The depth of the zone of aeration can vary greatly: from
less than a metre to hundreds of metres. Below the zone of aeration, in the
saturated zone, all the open spaces are filled with water instead of air. The upper
surface of the saturated zone is termed the “water table.”
Almost all groundwater, regardless of the zone in which it is located,
originates in rainfall and can be understood to be travelling, albeit very slowly,
back to the ocean (Holmes 1962: 403, Leopold 1974: 18-9, Skinner and Porter
1987: 240-8). Underground, the water table rises until it is exposed in a low-lying
“discharge area,” such as a valley, creating a watercourse on the surface. Rivers
and streams that continue to flow between rains are usually supplied with water
from the saturated zone: an indication that they intersect the water table.
The infiltration of groundwater is affected by the characteristics of the
soil: the larger the particles of dirt, the larger the spaces between the particles, and
the more water enters the ground instead of running off into natural channels. The
vegetation on the soil surface also plays a role by reducing surface run-off and
increasing the amount of groundwater that can be stored below (Leopold 1974:
10-2). Once it has entered the zone of aeration, the movement of ground water is
predominantly affected by topography and by the size of the pore spaces within
the regolith. The movement of vadose water is achieved by two forces: capillarity
119
and gravity. Capillary action plays a larger role in fine-grained soils, while in
coarse soils the water travels down by gravity through the spaces between
particles (Basile 1971: 18-29, Leopold 1974: 13-15).
Movement of water in the saturated zone is called percolation and is
stimulated by gravity, as well as the pressure of the ground above. Groundwater
in this zone will flow sideways until the water table becomes horizontal, at which
point the flow ceases. The general level of the water table tends to be uniform,
such that the saturated zone is located farther beneath the surface of a hill than it
is in a valley. However, as water percolates quite slowly through the pores,
cracks, and openings in the regolith,2 the water table is always undulating:
fluctuating most beneath hills and least beneath valleys (Holmes 1962: 402-3,
Leopold 1974: 20, Skinner and Porter 1987: 243-4).
Now that the foundations of groundwater hydrology have been
established, it is important to proceed to the basics of bone composition, before
the diagenetic processes that act on human bone can be explored.
STRUCTURE AND COMPOSITION OF BONES AND TEETH
There are two basic types of bone: cancellous bone and cortical bone.
Cancellous bone, also called trabecular or spongy bone, is found predominantly at
the ends of long bones and in the bodies of vertebrae. It consists of tiny spicules
of bone, called trabeculae, which are arranged in a lattice pattern and function like
scaffolding. In living bone, the spaces between trabeculae are filled with marrow.
2 The speed of groundwater is measured in centimetres per day or metres per year (Skinner and Porter 1987:
243).
120
Cancellous bone is quite light and, due to the arrangement of the trabeculae,
remarkably strong.
Cortical bone, also called compact bone, is the dense bone found in the
shafts of long bones, around the large marrow cavities, as well as in the outer
layer of most bones. Much denser than cancellous bone, it is composed of
cylindrical Haversian systems: concentric layers of ossified bone matrix around a
central canal that run parallel to the length of the bone. Haversian canals, which
contain nerves and vessels during life, are connected to each other by Volkmann‟s
canals. In both cancellous and cortical bone, bone cells (called osteocytes) are
contained in spaces throughout the ossified bone matrix, known as lacunae.
Lacunae are linked by canaliculi, miniscule connecting channels that run through
the bone matrix.
On a chemical level, bone tissue can be considered in terms of its organic
and inorganic components. The organic fraction of bone is composed largely of
collagen fibres (large protein molecules) and cell tissue. The inorganic, or
mineral, fraction is essentially composed of minute apatite crystals, which are
arranged around the bundles of collagen fibres (Chaplin 1971: 13, Schultz 1997:
188). In theory, most bone apatite is hydroxyapatite, a complex calcium
phosphate with the formula Ca10(PO4)6(OH)2.3 In actuality, during life much of a
person‟s bone apatite is not true, stoichiometric hydroxyapatite but rather a
transitory, more readily soluble phase of calcium phosphate (Pate and Brown
3 The formula for hydroxyapatite is actually Ca5(PO4)3OH but is usually written as Ca10(PO4)6(OH)2 to
indicate that a single crystal unit of hydroxyapatite contains two base units.
121
1985: 489, Sillen 1989: 213). The chemical makeup of the inorganic component
of bone fluctuates as it stores minerals and releases them into the blood stream.
In living bone the intercellular matrix is approximately 25 per cent
organic, 50 per cent inorganic, and 25 per cent water. In dry bone, almost one
third of the weight of bone matrix is collagen, while the inorganic component
provides approximately two thirds (Pritchard 1972, Schultz 1997: 189). The
mineral component of bone is traditionally associated with rigidity and the
collagen with tensile strength (e.g. Chaplin 1971: 12). Ascenzi and Bell (1972)
demonstrate that both the organic and inorganic components of bone are essential
for maintaining its physical properties. If either of the two components are
partially removed or weakened, the strength and hardness of the bone will be
compromised.
The general structure of teeth is similar to that of bone. Each of the three
mineralized dental tissues (dentine, cementum, and enamel) includes the
inorganic apatite and the organic proteins and lipids, but varies slightly in terms of
composition and organization. Additionally, dentine, cementum, and enamel are
avascular, unlike bone which relies heavily on a constant blood supply.
Dentine, which comprises the majority of a mature tooth, has a chemical
composition similar to bone but harder (Carlson 1990, Pritchard 1972, Scott and
Symons 1982: 227). The apatite crystallites in dentine are approximately the same
size as those in bone but are found in a higher concentration. The organic
component of dentine is proportionately less than in bone and the collagen fibres
are more densely packed. The structure of dentine differs from that of bone, but
122
like bone it is permeated throughout by cells called odontoblasts. The
protoplasmic processes of the odontoblasts are contained in dentinal tubules:
minute conduits that run between the outer surface of the dentine and the pulp
cavity. The presence of these tubules gives dentine a comparable porosity to bone
(Scott and Symons 1982: 227).
Cementum, found in a thin layer surrounding the dentine of tooth roots is
less hard than dentine but otherwise similar in terms of chemical composition.
There are two types of cementum: acellular, which is deposited first, and cellular,
which is found in the root apices. Cellular cementum is more permeable than the
acellular, although the permeability of both has been shown to decrease with age.
Out of the three mineralized dental tissues, cementum most closely resembles
bone (Carlson 1990, Scott and Symons 1982: 263-7).
Enamel covers the crown of the tooth, supported by the underlying
dentine. Ectodermal in origin, it is completely acellular and devoid of canals. The
minor organic portion is non-collagenous, composed of soluble and insoluble
proteins, peptides, and citric acid. The enlarged inorganic portion (approximately
96 per cent of the enamel tissue) is primarily composed of calcium phosphate
crystals that are considerably larger than those found in dentine, cementum, or
bone (Carlson 1990, Pritchard 1972, Scott and Symons 1982: 194-9). Together,
the low porosity, substantial inorganic component, and large, crystalline structure
make enamel the hardest material in the skeleton, denser and less soluble than the
other tissues (Carlson 1990, Clement 2009: 336, Pritchard 1972).
123
DIAGENESIS OF BONES AND TEETH IN GROUNDWATER
The diagenetic processes that affect bone vary greatly and, consequently,
can produce highly variable results. The various chemical reactions in bone
diagenesis impact one another: some slow one another down, others speed one
another up, some can happen simultaneously while others are sequential (Von
Endt and Ortner 1984). Diagenesis is complicated and even if a bone appears
unaltered at the macroscopic level, it can prove to be significantly altered or
damaged when examined under a microscope (Pfeiffer 2000: 291). Regardless of
the nature of the change being examined, however, water is nearly always the
medium in which diagenesis takes place.
As previously mentioned, much of living bone differs from true,
stoichiometric hydroxyapatite: existing in forms that are less thermodynamically
stable and more soluble. The diagenesis of the inorganic phase of bone “can be
seen as a correction of this aberration” (Sillen 1989: 213). Two main processes
constitute this stage of diagenesis: the loss, uptake, and exchange of minerals or
trace elements, and internal changes in crystallinity.
The process of mineral loss, uptake, and exchange is the most important
aspect of bone diagenesis, and is achieved through dissolution of minerals in the
adjoining groundwater (Von Endt and Ortner 1984). So long as the groundwater
is not already saturated with the minerals present in the skeletal apatite, ions are
leached from the bone tissue and can be replaced with soluble ions from the
surrounding soil matrix (Dodd and Stanton 1981: 128, Gill-King 1997: 105,
Henderson 1987: 44, Nielsen-Marsh et al. 2000, Pfeiffer 2000: 291, Sandford and
124
Weaver 2000: 334). In acidic soils, for example, calcium ions from the skeletal
apatite move into the groundwater and are replaced in the crystal structure by
hydrogen ions (Gill-King 1997: 105). This type of exchange is highly dependent
on the concentration of the groundwater solution, and the processes can slow, or
even reverse, if conditions change. In the above example, should the groundwater
becomes less acidic, calcium ions could move back into the bone from the soil,
replacing protons and reconstituting the original mineral (White and Hannus
1983).
Dissolution in groundwater affects most of the elements contained in
bones and teeth, with strontium being a notable exception (Parker and Toots
1980: 197). Among the trace elements typically lost in such a process are calcium,
sodium, and chlorine, while fluorine, silicon, manganese, and iron are among
those typically taken up by the skeleton (Hedges and Millard 1995, Lambert et al.
1985b, Nielsen-Marsh et al. 2000, Parker and Toots 1980, Sandford and Weaver
2000). The exchange of fluorine in groundwater for the hydroxyl in the apatite
matrix has been particularly well-documented (e.g. Lambert et al. 1979: 119).
During diagenesis, the inorganic matrix can also see the recrystallization,
growth, and “spontaneous rearrangement” of apatite crystals (Henderson 1987:
44, Sandford and Weaver 2000: 334, Sillen 1989: 221). Through this process, the
inorganic phase moves from the unstable, microcrystalline form of apatite
towards larger crystals and a more stable structure (Sillen 1989: 213, Von Endt
and Ortner 1984). Two mechanisms are credited with the crystallinity increase
(Nielsen-Marsh et al. 2000: 443-4). The first is the dissolution of the smallest
125
crystallites, where preferential loss favours retention of the larger units. The
second is dissolution and the subsequent recrystallization to larger, more
thermodynamically stable crystals. Most likely, both mechanisms play a role in
the internal changes observed during diagenesis but identification of instances of
recrystallization is particularly significant to researchers in stable isotope analysis,
as it could indicate the incorporation of exogenous ions deep into the mineral
structure of the bone (Ibid.).
Diagenesis of the organic phase of archaeological bone deals primarily
with collagen deterioration, as the cell tissue is subject to relatively rapid decay
after death (Chaplin 1971: 13, Millard 2001: 640). Deterioration of bone collagen
can be achieved by two processes: bacterial decomposition and hydrolysis of the
collagen protein. In moist environments, the presence of fungi and other micro-
organisms is prolific and the biological decomposition of bone collagen is
substantial. In addition to decomposing the organic phase of bone, these micro-
organisms excrete acids that expedite the dissolution of the apatite in the
inorganic phase, further propagating the diagenetic processes (Sillen 1989: 220).
In hydrolysis, water breaks collagen down into polypeptides and amino
acids, lowering the molecular weight of the bone protein and increasing its
solubility (Hare 1980: 212-4, Hedges and Millard 1995, Henderson 1987: 44, Von
Endt and Ortner 1984). This process does not require significant amounts of
groundwater and can be achieved with only the naturally-occurring water in bone
or atmospheric water vapour. These smaller peptides and amino acids can then be
leached from the skeletal tissue by available groundwater. Like in the process of
126
mineral exchange, the rate at and extent to which groundwater leaches the soluble
protein from bone depends on the saturation of the groundwater (Hare 1980: 212-
4). As a structural protein, collagen has a low solubility and for the leaching
process a substantial supply of groundwater is necessary, to prevent the water
from reaching the saturation point.
LIMITING FACTORS CONTROLLING BONE DIAGENESIS
While the processes of diagenesis all act within and through groundwater,
other properties of the burial environment influence its progression: limiting or
exacerbating the effects of groundwater on bone. Subsurface animal activity, for
example, is known to create all sorts of issues for buried bone and one of the
major consequences is the increased flow of groundwater into the burial and the
subsequent increase in the rate of diagenesis (Nawrocki 1995: 51). The use of
coffins (i.e. of stone, lead, or wood) can also affect decay: they have been shown
to restrict the drainage of water from the burial environment, exacerbating
diagenetic change (Pfeiffer 2000: 291). Rahtz and Hirst (1976) found a correlation
between evidence for a coffin and poorer skeletal preservation at their site at
Bordesley Abbey. Unfortunately, while the effect of coffins on soft tissue
decomposition has been relatively well-considered (e.g. see Henderson 1987: 51
for a brief overview), the impact on skeletal tissue is not conclusive.
Temperature, too, is an important determining factor for the effects of
groundwater on bone (Pate and Brown 1985, Sandford and Weaver 2000: 335,
Von Endt and Ortner 1984). The relationship between temperature and diagenesis
127
is complex but it can be generalised that cold temperatures tend to retard
diagenetic change and that, typically, a 10°C rise in temperature can double the
rate of diagenetic processes (Henderson 1987: 47).
While there are a number of limiting factors that determine the rate and
progression of bone diagenesis, some of the most important include: the
movement of water through the burial environment, the physical and chemical
properties of the soil matrix, and the intrinsic properties of the archaeological
bone. These elements will be discussed in more detail below.
Hydrology of the Burial Environment
The hydrology of the burial environment essentially determines the
movement of solutes to, from, and within the bone. The effect of groundwater on
bone is thus highly dependent on variations in flow rate (Chaplin 1971: 16, Gill-
King 1997: 94, Hedges and Millard 1995, Henderson 1987: 46, Nicholson 2001:
186, Nielsen-Marsh et al. 2000, Sandford and Weaver 2000: 335, Stodder 2008).
Average humidity, annual rainfall, and drainage all affect the degree of
fluctuation in terms of contact between bone and groundwater.
Hedges and Millard (1995) identify three extreme hydraulic regimes (later
expanded upon by Pike et al. 2001) as potential burial contexts, each primarily
dependent on a different process for affecting bone diagenesis. The first of these
is the Diffusive System, where there is no net flow of water but solutes may travel
across a diffusion gradient (e.g. in permanently waterlogged soils). This model
has the potential to allow for the most rapid progression of diagenesis, however
128
diffusion becomes limited as surrounding groundwater becomes saturated with
respect to bone apatite and collagen.
The second regime is the Hydraulic Flow System, where there is an
episodic flow of water through soil and bone, under a hydraulic gradient. In this
model, the supply of unsaturated groundwater can lead to a faster rate of
dissolution. Nevertheless, the rate will still depend on the volume of water
(mainly due to sporadic rainfall) and the relative conductivities of bone and the
surrounding soil. In many cases, water can pass through the soil more easily than
the buried bone and so it follows the path of least resistance. The more
diagenetically altered a bone becomes, however, the greater the porosity and
subsequent conductivity, so the effects of hydraulic flow become more significant
as diagenesis progresses. In the third regime, the Recharge System, the dissolution
rate of the Hydraulic Flow System is accelerated by a fluctuating hydraulic
potential. Frequent cycles between wet and dry drive water in and out of the bone.
Nielsen-Marsh et al. (2000) consider the three regimes put forward by
Hedges and Millard and reduce the hydrology of burial environments into two
basic categories: saturated or dry environments (with little or no change in
groundwater content) and fluctuating environments (where there is substantial
oscillation in the groundwater content around buried bones). In such a system of
hydrological classification, fluctuating conditions are more conducive to
diagenetic change than static environments. Completely waterlogged and
anaerobic environments (e.g. peat bogs and well deposits), with a negligible flow
of ground water, can be as beneficial as dry sites in terms of bone preservation,
129
due largely to the lower saturation point of standing water (Nicholson 2001: 181-
2, Nielsen-Marsh et al. 2000). Water fluctuations, however, result in the poor
preservation typically associated with infiltration by groundwater (Matheson and
Brian 2003: 134-6).
Cycles of wet and dry encourage bone fracturing as the bone expands and
contracts with each oscillation, accelerating the diagenetic processes (Chaplin
1971: 18, Nawrocki 1995, Nawrocki 2009). Conversely, cycles of wet and dry can
also slow down diagenesis, as the lack of diffusion matrix during dry periods
limits the amount of dissolution that takes place. Pike et al. (2001: 131), for
example, showed that an average flow rate of at least 100 litres a day was
necessary for a water fluctuation model to compete with a diffusion model in
terms of bone dissolution.
More detrimental than cycling between wet and dry is a burial
environment in which the bones are constantly exposed to fresh water. In such an
environment, the constant presence of water provides a continuous diffusion
matrix, while the regular replenishment of the surrounding water prevents it from
reaching the saturation point, the major limiting factor in the Diffusive System.
Once the point of saturation is reached, diagenesis slows considerably. If the
water is being replenished frequently, it is less likely to reach its saturation point
in terms of collagen and minerals, and the dissolution through diffusion will be a
continuous process (Hare 1980, Nielsen-Marsh et al. 2000: 446-7, Stodder 2008).
130
Soil Composition and pH
The physical and chemical characteristics of the surrounding soil matrix
also play a role in determining the effects of groundwater on archaeological bone.
One of the most important characteristics is pore-size distribution, which affects
the movement of water through the soil. Dissolution of bone‟s mineral and protein
content usually requires permeable sediment: mediums like gravel tend to
exacerbate diagenesis because they increase percolation, while soils of a finer
texture retard diagenesis, especially in a waterlogged environment (Chaplin 1971:
16, Dodd and Stanton 1981: 128).
The chemical composition of the soil matrix is also a key limiting factor in
diagenesis. The rate of bone dissolution is determined, in large part, by the
chemistry of the groundwater relative to bone (Hare 1980, Stodder 2008). The
chemistry of the groundwater, while refreshed by a high flow rate, is essentially
determined by the chemistry of the soil in which it is found (Basile 1971: 9,
Holmes 1962: 409, Skinner and Porter 1987: 253-4). As groundwater passes
through the bedrock and regolith, minerals are dissolved and pass into the water
solution. The quantity of iron, sulfates, and bicarbonates of calcium, magnesium,
and other ions in groundwater are all determined by the surrounding soil.
Groundwater that is already high in calcium ions due to the composition of the
bedrock will reach its saturation point more quickly and dissolve less calcium
from the bone apatite. Accordingly, the greater the difference between the mineral
composition of the soil and that of the bone apatite, the greater the degree of
diagenetic change (Dodd and Stanton 1981: 128).
131
Through the identical process, the pH level (an expression of the
concentration of hydrogen ions) of the groundwater is determined by the pH of
the soil (Gill-King 1997: 94). A significant correlation exists between soil pH and
bone preservation, so that as the soil becomes more acidic, the dissolution of the
bone‟s mineral content is accelerated (Chaplin 1971: 16-7, Gordon and Buikstra
1981: 569, Henderson 1987: 46, Lambert et al. 1979, Lambert et al. 1985b,
Nielsen-Marsh et al. 2000, Pate and Brown 1985, Sandford and Weaver 2000:
335, Stone et al. 1990). As stated above, acidic soils and groundwater work in
concert so that through groundwater, the acids in the soil dissolve the inorganic
matrix, leaving behind a “model” of organic matrix that‟s then more susceptible
to leaching by water. Base-rich soils lack the surplus of hydrogen ions present in
acidic soils, and so are much kinder to bone. Using a laboratory environment,
Pike et al. (2001) modeled that a 500g long bone fragment would take 0.08 years
to dissolve in calcium-free groundwater with a pH of 5, 0.64 years in groundwater
with a pH of 6, and 4.83 years at a pH of 7.
Because the composition of groundwater depends on the kind of rock in
which it occurs, it is specific to each burial environment and cannot be determined
without direct testing or analysis of the underlying geological foundations.
Nevertheless, Baas Becking et al. (1960) provide data that establish the limits of
different burial environments in terms of pH. Globally, the pH of groundwater can
vary between 2.8 and 10. Wet soils (described as soils that are dry for part of the
year but can have seasonal waterlogging) can have a pH anywhere between 3.7
132
and 8.5, while permanently waterlogged soils are restricted to a pH between 5 and
8 (Baas Becking et al. 1960: 252-6).
Surface Area and Porosity of Bone
Surface area is a crucial determining factor in the destruction of both the
organic and inorganic phases of bone because it determines the extent of the
“interface” between the bone matter and the surrounding matrix (Gill-King 1997:
105, Von Endt and Ortner 1984). An increase in surface area means an increase in
the speed at which water chemically breaks down the collagen and mineral
matrix, as water can only interact with bone at the exposed surfaces (e.g. Hedges
and Millard 1995: 156). Other than size and degree of fragmentation, the primary
determinant of surface area in bone is porosity. Pore spaces also define the
capillarity of bone. Capillary action is responsible for pulling groundwater into
bone, accelerating diagenesis.
Martill (1991: 285-7) identifies two pore-space environments within the
bone: the marrow cavities and spaces located in long bone shafts and within the
trabecular bone, and the lacunae and linking canaliculi of the bone itself. The
marrow spaces are connected to the exterior environment by foramina, holes for
blood vessels that pass through the cortical bone. The lacunae and other structural
spaces are not naturally well-connected to the exterior, although ante- or post-
mortem damage can result in their exposure (Ibid.).
Although it has been demonstrated that chemical weathering of bone is not
directly related to pore size (White and Hannus 1983), the varying porosity of
different skeletal tissues affects the progress of diagenetic change (Bell 1990: 99,
133
Hedges and Millard 1995, Lyman 1994: 418, Sandford and Weaver 2000: 335,
Von Endt and Ortner 1984). Diagenetic processes affect enamel significantly less
than dentine or bone, for example, because it is denser, less soluble, and has a
lower porosity (Carlson 1990: 545, Clement 2009, Parker and Toots 1970, Parker
and Toots 1980: 199-200). At a more minor level, dentine is also slightly less
susceptible to diagenetic change than bone, although the difference is much less
significant than that between bone or dentine and enamel. Likewise, juvenile bone
is more predisposed to diagenetic change than adult bone because of its lower
mineral content and high porosity (Gordon and Buikstra 1981, Sandford and
Weaver 2000: 335). Similar results can even be seen between differing skeletal
elements, so that small, porous bones like ribs have been shown to be more
susceptible to protein loss and mineral exchange than larger, denser,
predominantly cortical bones like femora (Hanson and Buikstra 1987: 552-3,
Hare 1980: 214, Lambert et al. 1982, Lambert et al. 1985a: 478, Nawrocki 1995,
Sandford and Weaver 2000: 335). Due, in part, to the exposure of lacunae and
canaliculi to the burial environment, crushed or broken bones and bones with
pathological lesions are also more susceptible than whole bone to these diagenetic
changes (Bell 1990, Lambert et al. 1985b).
An increase in porosity is the most common change resulting from the
mineral dissolution of bone (Nielsen-Marsh et al. 2000). Indeed, when Hedges et
al. (1995) and Nielsen-Marsh et al. (2000: 443-4) identified a series of diagenetic
parameters intended to measure the state of bone diagenesis, increased porosity
134
was identified as a vital indicator.4 Initially this increased porosity means an
increased rate of diagenesis, which in turn means an increase in porosity, etc.
Eventually, however, the overall surface area begins to decline as the pores join
together, resulting in a reduced boundary for interaction with the burial
environment (Hedges et al. 1995: 207, Hedges and Millard 1995). Still, an
increase in macroporosity at the expense of microporosity allows for increased
water movement in and out of the bone itself. Clearly, a complicated relationship
exists between diagenesis and porosity that would benefit from further study.
The result of a limiting factor is shaped by each of the other aspects of the
burial environment, and despite the generalisations discussed above, “it is not
possible to be too sweeping about the effects of a given soil/water regime because
the factors involved are numerous and complex and may have changed with time”
(Chaplin 1971: 17). For example, dissolution rates will certainly differ between a
warm, waterlogged burial environment with a low concentration of calcium ions
in the groundwater, and a cool, fluctuating burial environment with a high
concentration of calcium ions. How they differ, however, may be in complex and
perhaps unanticipated ways. Additionally, while the various chemical reactions in
bone diagenesis impact each other, the processes occur independently and a
certain degree of mineral exchange does not imply the same degree of protein
hydrolysis. Hedges et al. (1995) noted a low correlation between the different
diagenetic parameters, further suggesting that diagenesis is a complex,
multifaceted collection of processes. Any attempt to predict the precise impact of
4 Other diagenetic parameters included histological preservation, protein content, crystallinity increase, and
incorporation of exogenous ions (Hedges et al. 1995, Nielsen-Marsh et al. 2000: 443-4).
135
a specific burial environment in bone diagenesis must consider all the relevant
components.
CONCLUSION
The diagenesis of archaeological bone refers to the processes of ion
exchange, recrystallization, hydrolysis and dissolution. Together, these processes
function to weaken the protein-mineral bond which, in turn, makes the bone more
susceptible to future change. As the medium in which said processes occur, water
(particularly groundwater) is critical to diagenetic change and features such as the
water‟s pH and flow rate can largely determine the degree of deterioration found.
Even so, the multitude of factors that can affect diagenesis make the particulars
difficult to foresee, and the outcome can vary widely between contexts. This
summary has attempted to provide a framework of known information regarding
the diagenesis of bone in groundwater, to create a better understanding of the
forces that acted on the Sodo material in situ. Clearly, much research is still
required for archaeologists to understand and predict the changes that can occur to
human remains in the burial environment.
136
WORKS CITED
Ascenzi A, Bell GH. 1972. Bone as a mechanical engineering problem. In:
Bourne, GH, editor. The Biochemistry and Physiology of Bone. New York
(NY): Academic Press, pp311-352.
Baas Becking LGM, Kaplan IR, Moore D. 1960. Limits of the natural
environment in terms of pH and oxidation-reduction potentials. The
Journal of Geology. 68(3): 243-284.
Basile RM. 1971. A Geography of Soils. Dubuque (IA): WM. C. Brown
Company Publishers.
Behrensmeyer AK, Hill AP, editors. 1980. Fossils in the Making. Chicago (IL):
University of Chicago Press.
Bell LS. 1990. Palaeopathology and diagenesis: An SEM evaluation of structural
changes using backscattered electron imaging. Journal of Archaeological
Science. 17: 85-102.
Carlson SJ. 1990. Vertebral dental structures. In: Carter JG, editor. Skeletal
Biomineralization: Patterns, Processes and Evolutionary Trends. New
York (NY): Van Nostrand Reinhold, pp531-556.
Chaplin RE. 1971. The Study of Animal Bones from Archaeological Sites.
London: Seminar Press.
Child AM. 1995. Microbial taphonomy of archaeological bone. Studies in
Conservation 40(1): 19-30.
Clement JG. 2009. Forensic odontology. In: Blau S, Ubelaker DH, editors.
Handbook of Forensic Anthropology and Archaeology. Walnut Creek
(CA): Left Coast Press, pp335-347.
Dodd JR, Stanton RJ Jr. 1981. Paleoecology: Concepts and Applications. New
York (NY): John Wiley & Sons, Inc.
Efremov JA. 1940. Taphonomy: A new branch of paleontology. Pan-American
Geologist 74(2): 81-93.
Gill-King H. 1997. Chemical and ultrastructural aspects of decomposition. In:
Haglund WD, Sorg MH, editors. Forensic Taphonomy. Boca Raton (FL):
CRC Press, pp93-108.
Goffer Z. 1980. Archaeological Chemistry: A Sourcebook on the Applications of
Chemistry to Archaeology. New York (NY): John Wiley & Sons, Inc.
137
Gordon CC, Buikstra JE. 1981. Soil pH, bone preservation, and sampling bias at
mortuary sites. American Antiquity 46(3): 566-571.
Haglund WD, Sorg MH, editors. 1997. Forensic Taphonomy. Boca Raton (FL):
CRC Press.
Hanson DB, Buikstra JE. 1987. Histomorphological alteration in buried human
bone from the Lower Illinois Valley: Implications for palaeodietary
research. Journal of Archaeological Sciences 14: 549563.
Hare PE. 1980. Organic geochemistry of bone and its relation to the survival of
bone in the natural environment. In Behrensmeyer AK, Hill AP, editors.
Fossils in the Making. Chicago (IL): University of Chicago Press, pp208-
219.
Hedges REM, Millard AR. 1995. Bones and groundwater: Towards the modeling
of diagenetic processes. Journal of Archaeological Science 22: 155-164.
Hedges REM, Millard AR, Pike AWG. 1995. Measurements and relationships of
diagenetic alteration of bone from three archaeological sites. Journal of
Archaeological Science 22: 201-209.
Henderson J. 1987. Factors determining the state of preservation of human
remains. In: Boddington A, Garland AN, Janaway RC, editors. Death,
Decay and Reconstruction: Approaches to Archaeology and Forensic
Science. Manchester: Manchester University Press, pp43-54.
Holmes CD. 1962. Introduction to College Geology. New York (NY): The
Macmillan Company.
Johnson J. 1994. Consolidation of archaeological bone: a conservation
perspective. Journal of Field Archaeology 21(2): 221-233.
Koob SP. 1992. Recovery and treatment of skeletal remains at Herculaneum. In:
Payton R, editor. Retrieval of Objects from Archaeological Sites. Denbigh
(CWD): Archetype Publications, pp157-166.
Lambert JB, Szpunar CB, Buikstra JE. 1979. Chemical analysis of excavated
human bone from Middle and Late Woodland sites. Archaeometry 21(2):
115-129.
Lambert JB, Vlasak SM, Thometz AC, Buikstra JE. 1982. A comparative study of
the chemical analysis of ribs and femurs in Woodland populations.
American Journal of Physical Anthropology 59: 289-294.
138
Lambert JB, Simpson SV, Szpunar CB, Buikstra JE. 1985a. Bone diagenesis and
dietary analysis. Journal of Human Evolution 14(5): 477-482.
Lambert JB, Simpson SV, Weiner SG, Buikstra JE. 1985b. Induced metal-ion
exchange in excavated human bone. Journal of Archaeological Science
12: 85-92.
Leopold LB. 1974. Water: A Primer. San Francisco (CA): W.H. Freeman and
Company.
Lyman RL. 1994. Vertebrate Taphonomy. Cambridge: Cambridge University
Press.
Martill DM. 1991. Bones as stones: The contribution of vertebrate remains to the
lithologic record. In: Donovan SK, editor. The Processes of Fossilization.
New York (NY): Columbia University Press, pp270-292.
Matheson CD, Brian D. 2003. The molecular taphonomy of ancient biological
molecules and biomarkers of disease. In: Greenblatt CL, Spigelman M,
editors. Emerging Pathogens: The Archaeology, Ecology, and Evolution
of Infectious Disease. Oxford: Oxford University Press, pp127-142.
Millard A. 2001. The deterioration of bone. In: Brothwell DR, Pollard AM,
editors. Handbook of Archaeological Sciences. Chichester: John Wiley &
Sons, Ltd., pp637-647.
Nawrocki SP. 1995. Taphonomic processes in historic cemeteries. In: Grauer AL,
editor. Bodies of Evidence: Reconstructing History Through Skeletal
Analysis. New York (NY): Wiley-Liss Inc., pp49-66.
Nawrocki SP. 2009. Forensic taphonomy. In: Blau S, Ubelaker DH, editors.
Handbook of Forensic Anthropology and Archaeology. Walnut Creek
(CA): Left Coast Press, pp284-294.
Nicholson RA. 2001. Taphonomic investigations. In: Brothwell DR, Pollard AM,
editors. Handbook of Archaeological Sciences. Chichester: John Wiley &
Sons, Ltd., pp179-190.
Nielsen-Marsh C, Gernaey A, Turner-Walker G, Hedges R, Pike A, Collins M.
2000. The chemical degradation of bone. In: Cox M, Mays S, editors.
Human Osteology in Archaeology and Forensic Science. London:
Greenwich Medical Media, pp439-454.
Parker RB, Toots H. 1970. Minor elements in fossil bone. Geological Society of
America Bulletin 81: 925-932.
139
Parker RB, Toots H. 1980. Trace elements in bones as paleobiological indicators.
In: Behrensmeyer AK, Hill AP, editors. Fossils in the Making. Chicago
(IL): University of Chicago Press, pp197-207.
Pate D, Brown KA. 1985. The stability of bone strontium in the geochemical
environment. Journal of Human Evolution 14(5): 483-491.
Pfeiffer S. 2000. Palaeohistology: Health and disease. In: Katzenberg MA,
Saunders SR, editors. Biological Anthropology of the Human Skeleton.
New York (NY): Wiley-Liss Inc., pp287-302.
Pike AWG, Nielsen-Marsh C, Hedges REM. 2001. Modelling bone dissolution
under different hydrological regimes. In: Millard AR, editor. Proceedings
of Archaeological Sciences ‟97. British Archaeological Reports
International Series 939. Oxford: Archaeopress, pp127-132.
Pritchard JJ. 1972. General histology of bone. In: Bourne GH, editor. The
Biochemistry and Physiology of Bone. New York (NY): Academic Press,
pp1-20.
Rahtz P, Hirst S. 1976. Bordesley Abbey, Redditch, Hereford-Worcestershire:
First Report on Excavations 1969-1973. British Archaeological Reports,
23. Oxford: British Archaeological Reports.
Sandford MK, Weaver DS. 2000. Trace element research in anthropology: New
perspectives and challenges. In: Katzenberg MA, Saunders SR, editors.
Biological Anthropology of the Human Skeleton. New York (NY): Wiley-
Liss Inc., pp329-350.
Schultz M. 1997. Microscopic structure of bone. In: Haglund WD, Sorg MH,
editors. Forensic Taphonomy. Boca Raton (FL): CRC Press, pp187-199.
Scott JH, Symons NBB. 1982. Introduction to Dental Anatomy 9th
Edition.
Churchill Livingstone: Edinburgh.
Sillen A. 1989. Diagenesis of the inorganic phase of cortical bone. In: Price TD,
editor. The Chemistry of Prehistoric Human Bone. Cambridge: Cambridge
University Press, pp211-229.
Skinner BJ, Porter SC. 1987. Physical Geology. New York (NY): John Wiley &
Sons, Inc.
Stodder ALW. 2008. Taphonomy and the nature of archaeological assemblages.
In: Katzenberg MA, Saunders SR, editors. Biological Anthropology of the
Human Skeleton. 2nd
Edition. Hoboken (NJ): John Wiley & Sons, Inc.,
pp71-114.
140
Stone TT, Dickel DN, Doran GH. 1990. The preservation and conservation of
waterlogged bone from the Windover Site, Florida: A comparison of
methods. Journal of Field Archaeology 17(2): 177-186.
Von Endt DW, Ortner DJ. 1984. Experimental effects of bone size and
temperature on bone diagenesis. Journal of Archaeological Science 11:
247-253.
White EM, Hannus LA. 1983. Chemical weathering of bone in archaeological
soils. American Antiquity 48(2): 316-322.
141
APPENDIX C – SAMPLE DENTAL DATA COLLECTION
FORM
142