Eating-Drinking Roman Britain

THE SOCIAL CONTEXT OF EATING AND DRINKING
AT NATIVE
SETTLEMENTS
IN EARLY ROMAN BRITAIN
by
Karen Ingrid Meadows
A thesis submitted for the degree of Doctor of Philosophy

Department of Archaeology and Prehistory
University of Sheffield
Submitted July 2001
Table of Contents
Abstract
Acknowledgments
Definitions
Chapter 1
Introduction
1.1.
Roman Things
1.2.
The Scope of this Research
The Data
1.3.
1.4.
11
V
An Outline of the Thesis
Chapter 2
'Romanization' and the Study of Social Change
Introduction
2.1.
2.2.
The Study of Social Change in Roman Britain and the Theory of `Romanization'
2.2.1.
Romanization - defined
2.2.2.
The context of my `romanization'
2.2.3.
Roman `things'
2.2.4.
To Romanize or to not Romanize
2.3.
The study of social change in Roman Britain: negotiation and resistance
2.4.
Conclusion
Chapter 3
Towards an Archaeology of Food and Drink Consumption
3.1.
Introduction
22
3.2.
Towards a Study of Social Change at Native Settlementsin Roman Britain
3.2.1.
The household
3.2.2.
Housesand Romanimperialism - from roundedto rectangular
3.2.3.
The non-RomanRomano-Britishhousehold
3.2.4.
Diet and culinary practice
Diet, culinary practicesand imperialism
3.2.5.
`Roman' consumptionhabits
3.2.6.
3.2.7.
3.3.
3.3.1.
3.4.
Roman style goods and constructs at `non-Roman' settlements

An Approach to the Study of the Social Contexts of Eating and Drinking
Areas open to investigation
The Methodological Approach of this Thesis
A caveatto the study of the social contextsof eating and drinking

3.4.1.
Consumingstatistics
3.4.2.
The artefactsandremainsof food anddrink preparationandconsumption
3.4.3.
The social context of eating and drinking
3.4.4.
The social context of deposition
3.4.5.
The comparisonof eating and drinking practicesof households
3.4.6.
Conclusion
3.5.
The Upper Thames Valley
Chapter 4
Introduction
4.1.
The Oxford Archaeological Unit: Similarities and Contrasts with the Present Study
4.2
An Introduction to the Upper Thames Valley
4.3
The political setting in the late Iron Age and early Roman periods
4.3.1.
The nature of settlement in the late Iron Age and early Roman periods
4.3.2.
Regional discontinuity in the Upper Thames Valley
4.3.3.
50
4.3.4.
4.4.
4.4.1.
4.4.2.
Natives and Romans? in the Upper Thames Valley

The Study of Native Settlement in the Upper Thames Valley
Households without houses
A definition of Upper Thames Valley `households'
Early Romanperiod householdsin the Upper ThamesValley

4.4.3.
The Upper Thames Valley - the Sites
4.5.
Sites on the floodplain
4.5.1.
Sites on the first gravelterrace
4.5.2.
Sites on the secondgravel terrace
4.5.3.
Conclusion
4.6.
Figure 4.1
72
Barton Court Farm

Chapter 5
Table of Contents
5.1.
Introduction
The Excavation
5.2.
The site
5.2.1.
5.2.2.
5.3.
5.3.1.
Site reports, microfiche and archives

The Archaeology of Food and Drink Consumption at Barton Court Farm
The containers
5.3.2.
5.3.3.
5.3.4.
5.3.5.
The containersat the late Iron Age settlement

Other implementsat the late Iron Age settlement
The ingredients
The ingredientsat the late Iron Age settlement
5.4.
5.4.1.
Food and Drink Consumption at the Roman Period Settlement

The containers at the early Roman period settlement
5.4.2.
5.4.3.
5.5.
5.5.1.
5.5.2.
5.6.
5.6.1.
5.6.2.
5.7.
5.7.1.
5.7.2.
5.8.
5.9.
Other implementsat the early Romanperiod settlement

Ingredientsat the early Romanperiod settlement
Summary of the Artefacts and Remains of Consumption for each Settlement...

Late Iron Age settlement
The early Romanperiod settlement
The Distribution of the Remains of Eating and Drinking at Barton Court Farm
Distribution of the artefacts and remains at the late Iron Age settlement
Distribution of the artefacts and remains at the early Roman period settlement
Discussion of the Distribution of Artefacts and Remains at Barton Court Farm
The Iron Age settlement
The early Roman period settlement
The Social Contexts of Imperialism at Barton Court Farm
Conclusion
Figures 5.1-5.27
122
Roughground Farm
Chapter 6
Table of contents
6.1.
6.2.
6.2.1.
6.2.2.
Introduction
The Excavation
The site
The site reportand site archive
6.3.
6.3.1.
The Archaeology of Food and Drink Consumption at Roughground

The containers
6.3.2.
6.3.3.
The containersat the early Roman period settlement

Other implements
6.3.4.
The ingredients
6.3.5.
The ingredientsat the early Roman period settlement
6.4.
Farm
Summary of the Artefacts and Remains of Consumption at Roughground Farm
6.4.1.
6.4.2.
6.4.3.
The Containers
Other implements
The ingredients
6.5.
The Distribution of the Artefactsand Remainsof Consumptionat RoughgroundFarm
6.5.1.
The main areaof occupation
6.5.2.
6.5.3.
Pre-villa and environs

`Special deposits'
6.5.4.
Summaryof the re-contextualized
materialat the earlyRomanRoughgroundFarm
6.6.
Discussion of the distribution of artefacts and remains at Roughground Farm
6.7.
The Social Contexts of Imperialism at Roughground Farm
6.8.
Conclusion
Figures 6.1 - 6.9
Chanter 7
Old Shifford Farm
Table of Contents
7.1.
Introduction
7.2.
The Excavation
7.2.1.
The site assessment
7.2.2.
The main excavation
7.2.3.
The Site
7.2.4.
Site reports and archives
The Archaeology of Food and Drink Consumption at Old Shifford Farm
7.3.
7.3.1.
The containers
7.3.2.
The containersat the late Iron Age settlement
7.3.3.
7.3.4.
Other implements at the late Iron Age settlement - phase one and two
The ingredients
152
7.3.5.
The ingredientsat the Late Iron Age settlements
7.4.
7.4.1.
The Containersat the early Romanperiodsettlement
7.4.2.
7.4.3.
7.5.
7.5.1.
7.5.2.
7.6.
7.6.1.
7.6.2.
7.7.
7.7.1.
7.7.2.
7.8.
7.9.
Other implements at the early Roman period settlement

The ingredients at the early Roman period settlement
Summary of the Artefacts and Remains of Consumption at each Settlement
Late Iron Age settlement - phase one and two

The Distribution of the Remains of Eating and Drinking at Old Shifford Farm
Distribution of the artefacts and remains at the late Iron Age settlements
Distribution of the artefacts and remains at the early Roman period settlement
Discussion of the Distribution of Artefacts and Remains at Old Shifford Farm
The late Iron Age period settlement phasesone and two
-
The Social Contexts of Imperialism

Conclusion
at Old Shifford Farm
Figures 7.1 - 7.13

Claydon Pike
Chapter 8
Table of Contents
8.1.
8.2.
8.3.
8.3.1.
8.3.2.
8.4.
Introduction
A Note on Chronology
The Excavation
The Site
Post-excavationreports and primary records
The Archaeology of Food and Drink Consumption at Claydon Pike
195
8.4.1.
The Containers
8.4.2.
8.4.3.
The containers at the late Iron Age settlement

Other implements of consumption
8.4.4.
8.4.5.
8.4.6.
The other implementsat the late Iron Age settlement

The ingredients
The ingredientsat the Late Iron Age settlement
8.5.
8.5.1.
Food and Drink Consumption at the Early Roman Period Settlement

Containers at the early Roman period settlement
8.5.2.
8.5.3.
8.6.
8.6.1.


8.6.2.
8.7.
8.7.1.
8.7.2.
8.8.
8.8.1.
8.8.2.
8.9.
8.10.
The Distribution of the Remains of Eating and Drinking at Claydon Pike

The distribution of artefacts and remains at the late Iron Age settlement
The distribution of artefactsand remainsat the early Roman period settlement
Discussion of the Distribution of Artefacts and Remains at Claydon Pike
The late Iron Age settlement
The Social Contexts of Imperialism at Claydon Pike
Conclusion
Figures 8.1 - 8.22

Chapter 9
Food and Drink Consumption and the Study of Social Change
9.1.
Introduction
9.2.
Food and Drink Consumption at the late Iron Age Settlements
9.2.1.
Late Iron Age Barton Court Farm
9.2.2.
9.2.3.
9.3.
9.4.
9.4.1.
Late Iron Age Old Shifford Farm

Late Iron Age ClaydonPike
Late Iron Age Food and Drink consumption
Comments
Final
Food and Drink Consumption at the Early Roman Period Settlements
Early Roman period Barton Court Farm
9.4.2.
9.4.3.
9.4.4.
9.5.
9.6.
244
Early Romanperiod RoughgroundFarm

Early Romanperiod Old Shifford Farm
Early Romanperiod Claydon Pike
Early Roman Period Settlement in the Upper Thames Valley
Conclusion
Chapter 10
Bibliography
Concluding Remarks
265
270
List of Figures and Tables
Figures
Figure 4.1. Map of the Upper Thames Valley and sites discussed in thesis
Figure 5.1. Barton Court Farm
Figure 5.2. Late Iron Age pottery forms
71
112-121
Figure 5.3. and Figure 5.4. Histogramsofthe rim diametersof late Iron Age bowls andjars
Figure S.S. Histogram of late Iron Age tableware
Figure 5.6. Late Iron Age pottery fabrics
Figure 5.7. N. I. S.P. (Wilson)
Figure 5.8. N. I. S.P. (Meadows)

Figure 5.9. Late Iron Age animal bone grouping according to meat yield
Figure 5.10. Early Roman period pottery forms
Figure 5.11. Early Romanperiod pottery fabrics

Figure 5.12. and 5.13. Histogramsof early Romanperiod jars and bowls
Figure 5.14. Histogramsof early Romanperiod tableware
Figure 5.15. N. I. S.P. (Meadows)
Figure 5.16. N. I. S.P. (Wilson)
Figure 5.17. Early Romanperiod animal bone groupsaccording to meatyield
Figure5.18.andFigure5.19. Distributionof lateIronAge animalbonesandpottery
Figure 5.20. Distribution of late Iron Age tableware
Figure 5.21. Histogramsof late Iron Age jars and bowls by feature
Figure 5.22. Late Iron Age animal bone groupsby feature
Figure 5.23. Distribution of early Roman pottery
Figure 5.24. Distribution of tableware
Figure 5.25. Histogramsof early Roman periodjars and bowls by feature

Figure 5.26. Distribution of early Roman period animal bones
Figure 5.27. Distribution of early Romanperiod animal bone groups by feature
Figure 6.1. Roughlzround Farm
Figure 6.2. Early Roman period pottery forms
Figure 6.3. Histogramsof the rim diametersof different jar types
Figure 6.4. Histogramsof the rim diametersof jars
Figure 6.5. Histogramsof the rim diametersof tableware
Figure 6.6. Early Roman period N. I. S.P.
Figure 6.7. Animal bone groupsaccording to meatyield
Figure 6.8. Distribution of pottery in main occupationarea
148-151
Figure69. Comparison
of potteryformsbetweenpre-villaareaandmainoccupation
area
Figure 7.1. Old Shifford Farm
190-194
Figure 7.2. The late Iron Age pottery forms phaseone and two
Figure 7.3. Late Iron Age phaseone- N. I. S.P.
Figure 7.4. Late Iron Age phasetwo- N. I. S.P.
-
Figure 7.5.LateIron Age- bothphases

bone
animal
groupsaccordingto meatyield
Figure 7.6. Early Romanperiodpottery forms
Figure 7.7. Histogram of early Roman period pottery
Figure 7.8. Early Roman period N. I. S.P.
Figure 7.9. Early Roman period animal bone groups according to meat yield
Figure 7.10. Distribution of late Iron Age - phaseone - animal bones
Figure 7.11. Distribution of late Iron Age phasetwo animal bones
Figure 7.12. Distribution of early Roman period animal bones
Figure 7.13. Early Roman period animal bone groups by feature
Figure 8.1. Claydon Pike
Figure 8.2. Late Iron Age pottery forms
Figure 8.3. Histogram of the rim diameterof late Iron Agejars

Figure 8.4. Histogramof the rim diameterof late Iron Age bowls
Figure 8.5. Late Iron Age N. I.P.S. (Wilson)
Figure 8.6. Late Iron Age N. I. P.S. (Meadows)
Figure 8.7. Late Iron Age animal bonegroupsaccordingto meatyield

Figure 8.8. Early Romanperiod pottery forms
Figure 8.9. Histogram of the rim diameter of early Roman period jars
Figure 8.10. Histogram of the rim diameter of early Roman period bowls
Figure 8.11. Early Roman period N. I. P.S. (Meadows)
Figure 8.12. Early Roman period N. I. P.S. (Wilson)
Figure 8.13. Early Romanperiod animal bonegroupsaccordingto meatyield

Figure 8.14. Distribution of late Iron Age pottery
Figure 8.15. Distribution of late Iron Age animal bones
Figure 8.16. Distribution of late Iron Age animal groups- south
Figure 8.17. Distribution of late Iron Age animal groups- north
Figure 8.18. Distribution of late Iron Age animal groupsby feature
Figure 8.19. Distribution of early Romanperiod pottery
Figure 8.20. Distribution of early Romanperiod animalbones
Figure 8.21. Distribution of early Roman period animal groups ditches
Figure 8.22. Distribution of early Roman period animal groups - pits
237-243
Tables
Table 6.1. Early Romanperiodpotteryfabrics
130
Table62. Thedistributionof earlyRomanperiodpotteryfabrics
141
Table7.1. LateIronAge phaseoneandtwo- andearlyRomanperiodpotteryfabrics
161
Table 8.1. Late Iron Age and early Romanperiod pottery fabrics
205
Table 8.2. Late Iron Age pottery fabrics - southernand northern areas
222
Table 8.3. Distribution of late Iron Age animal bone groupsby feature
224
The Social Context of Eating and

Drinking at Native Settlements in Early Roman Britain
Karen Ingrid Meadows
Abstract
Observation of the presence and absence of Roman-style goods and structures has guided much of
the analysis of imperialism in Roman Britain and other parts of the Empire. Wealth and power have
been assumed to correlate with the extent to which a group's material culture and lifestyle appeared
`Romanized'. The concept of `Romanization' has become the primary measurement of change in the
lives of the people who were conquered: and where there was only slight evidence of Romanization,
there is an assumption that the lives of people were little changed and continued much as they did
before the conquest. Many of the signifiers used to describe `Romanization'
are tied to the
consumption of food and drink. Eating and drinking, however, is much more than the observance of
particular ingredients and containers - it is also the consideration of how and where one eats and
drinks, and with whom and why. Rarely is the totality of food and drink consumption in Roman
Britain considered. This study challenges the inventories of `Roman' and `native' material culture,
incorporate
different types of settlements and the experiences of people of different socioto
as
so
economic backgrounds into discussions of Roman Britain.
This thesis developsa methodologicalapproachto the analysisof the social contexts of the
food
drink
of
and
consumption
at `native type' settlementsduring the post-conquestperiod in an
localized
the
to
access
effects of imperialism.This approachwas realized through an inattempt
depth analysis of four sites in the Upper ThamesValley. The four sitesselectedfor analysisare:
Barton Court Farm, RoughgroundFarm, Old Shifford Farm,and ClaydonPike.
11
Acknowledgements
I would like to thank the University of Sheffield for granting me a PostgraduateResearch

Scholarshipand the Committeeof Vice-Chancellorsand Principals of the Universities of
the United Kingdom for grantingme with an OverseasResearchScholarship.Before I
Department
Archaeology
University
the
the
provided a number of
scholarship,
secured
is
There
for
I
library
department
job
the
grateful.
which am extremely
at
grantsand a great
I
like
I
thank
used
which
who gaveme a sum of money, anonymously,
also a woman would
to pay for some extra day-carefor my daughterin the final few months.Her only request
I
day
for
do
I
that
the
will.
which
same someoneelseone
was
would
There are many peoplein Sheffield that I would like to thank. Foremost,I would like to
he
in
has
been
Branigan.
He
that
Keith
gaveme
thank my supervisor
a perfect supervisor
the freedom to follow my own courseof study but provided enoughsoundadvice and
has
Pearson,
Mike
Parker
keep
my secondarysupervisor,
me on track.
critical comment to
beena continuous sourceof ideasthroughoutthe working of this thesis.I would also like to
thank John Barrett for his various insights on the subject of `Romanization'.
I am especially grateful for the encouragementand cooperationof various peopleat the

Oxford Archaeological Unit. David Miles, the former director of the Unit, for allowing me
his
for
Units
databases
the
through
commentson
to plough
records,
and archives,as well as
director
Jennings,
David
Barton
Court
Farm
the
the
now
casestudy.
an earlier rendition of
helpful
Unit,
the
particularly
was
with my queriesand made available the site archivesof
of
Old Shifford Farm and the post-excavationwork on Claydon Pike and Thornhill Farm. Tim
Allen gave me copiesof the pottery reports for Abingdon Vineyards and Gravelly Guy in
for
which I am grateful. I would like to thank Paul Booth in
advanceof publication,
for
his
for
I
comments
on
am also
my
and
my
many
questions.
work
answering
particular
his
in
providing
of
me
copies
papers
with
advanceof publication
appreciative
of various
from
`Roman
I
book
for
Oxfordshire'
he
the
chapters
me
when was
sending
co-wrote
and
in
Canada.
I
Mark
Wilson,
like
Robinson,
Bob
to
thank
to
copy
a
get
would
also
unable
111
JaneTimby and Pricilla Langewho were very accommodatingand willingly answered

many questions on the primary records.
Andrew Sherratt at the AshmoleanMuseum generouslyallowed me free reign of the

RoughgroundFarm archiveand when in a momentof self-doubt (and whilst on the wrong
had
I
boxes
Atlantic)
that
to
the
through
not
of archives ensure
side of
eleven
searched
for
Oxfordshire
Museums
Store
Debra
Keasal
Thanks
the
to
at
overlooked a notebook.
also
providing accessto the Barton Court Farm archive.
I have been inspired by many friends and colleagues over the years. In particular I would
like to thank Jane Webster for her friendship and for our many discussions on the subject of
imperialism. She also sent my a copy of her excellent paper "Creolizing the Roman
2.
The
Chapter
invaluable
in
on
comments
provinces"
advance of publication and made
helpful,
been
have
Sheffield
TRAC
enormously
various theory seminars at
and
conferences
Mark
David
Dungworth,
Clarke,
have
Chris
Cumberpatch,
Simon
as
conversations with
Edmonds, Bill Frasier, Kathryn Fewster, J.D. Hill, Chris Lemke, Matt Lemke, Colin
Merrony, Jenny Moore, Carol Palmer, Mark Pluciennik, Angela Puccini, Steve Willis, Alex
Woolf, Greg Woolf and Rob Young to name a few. Gillian Hawkes, J.D Hill, Dieke
Wesseling and Steve Willis sent me copies of papers for which I am most appreciative. I
his
for
like
Robert
Rippengal
to
thank
excellent thesis.
of
a
copy
would also
sending me
For a variety of reasons,this thesis took longer than anticipatedto complete,and I am

Ottaway
Barbara
Studies,
Department,
Graduate
Archaeology
to
the
and
sincerely grateful
Keith Branigan for allowing me to extend my studies.
There are two Kathryn's that I needto thank. Kathryn Denning, who by chancebecamemy
in
back
Toronto,
has
been
both
friend
a
colleague
since
moved
wonderful
we
neighbour
and
to Canada.She did an excellentjob editing the text andmade many important suggestions
that I have taken on board. Shehas beensuch a positive influence on my work that I will be
forever grateful. Kathryn Goldsack,the assistantto the director of the ResearchSchool, has
beenan invaluable link to the departmentand hashelpedenormouslywith many of the
issues
inevitably
that
administrative
occur when one `writes up' in anothercountry. I am
friendship,
for
her
grateful
greatunderstandingand her professionalism.
especially
1
iv
I am also hugely grateful to Silverio Ferrari at `The Computer Hospital' for services
rendered.
I now have a new family in England,and Anne and Peter Seymour,welcomed me into their
home and their lives andhaveprovided financial support and many fantastic Sunday
dinners. I would particularly like to thank Anne Seymourfor proofreadingthe text and for
her invaluable comments.
I would like to thank my parentsfor their supportand especiallytheir blind faith in my
abilities as well as their financial help over the years.I wish with all my heart that I could
thank my mum, BarbaraMeadows,in person.Shewas determinedthat I finish this thesis
and made me promise that I would and it is to her memory that this work is dedicated.
Matt Seymour,in addition to typing a large portion of the bibliography and completing
his
diagrams,
He
daughter
for
the
some of
entertainedour
countlessweekends. maintained
humour, sanity and compassionthroughoutand it was with his encouragementthat I
completed this thesis.
And finally, I thank Izabella(Izy) our wild child for keeping me groundedand for insisting
that I "stop working" and take the time to play in the sandbox.
Definitions
Given that many of the terms usedin this thesis havemultiple meanings,even within the
discipline of archaeology,below are some key definitions that will help to explain what I
mean when I use a given word. However, as is often the case in archaeology,many of
these words are highly problematic becausethe sociopolitical, temporal, or physical
phenomenathey representare not straightforward, and are not fully understood; please
seethe text for a more full discussionof such problems.
native
I use `native' in reference to the indigenous population of Roman Britain. The
term also has political connotations when it is used to characterize the social contexts of
imperialism.
native-type settlement (Chapter3 and 4)

A settlementthat is describedas `native' usually refers to a non-villa settlement.
Romanization (Chapter2)
"We must thus see Romanization as a processof dialectical change,rather than
the influence of one `pure' culture upon other. Roman culture interacted with native
cultures to producethe synthesisthat we call Romanized"(Millett 1990b:1).
household (Chapter 3 and 4)
The "group who usedand residedwithin a singlehouseor group of closely related

buildings" (Hingley 1990a:128 [my emphasis]).
native continuity (Chapter 2)
Native and Roman are generally thought of in terms of continuity and change
later
Iron Age and native traditions, and change as all
the
as
continuation
of
continuity
that which is brought about by the arrival of the Romans.
vi
consumption (Chapter3)
In this study, I use the word `consumption' as it pertains to food and drink
consumption,unlessotherwisestated.
`special deposits' (after Hill 1995)(Chapter5)
The dividing of depositsinto `special' and òrdinary' is not wholly appropriate;
however, I considered deposits as `special' when there were specific groupings of
particular types of artefactsand/or particular speciesand/or body parts, or a specific type
of butchery practice (or no butcherypractice),in particular contexts.
late Iron Age
Late first centuryBC to the mid-first centuryAD

early Roman period
Mid-first centuryAD to the mid-secondcentury AD

Upper Thames Valley (Chapter4)
The Upper ThamesValley cuts acrosssouthern-centralEngland from the source
of the River Thames in the west, one hundred kilometres eastwards.The region was
first
the
wave of conquestafter the Roman invasion of 43 AD, although evidence
within
of Roman and/or continental influence has been identified at some late Iron Age sites.
Military sites were establishedat points around the valley, at Dorchester, Gloucester
(Glevum) and at Cirencester(Corinium). A seriesof well establishedtrackwaysand roads
linking settlementswith local centres,such as Abingdon Vineyard, and the River Thames
provided communicationroutesthroughoutthe Valley and beyond.
Was there a more boring place in the world than the British Museum? If there
was, Will would not want to know about it. Pots. Coins. Jugs.Whole rooms full
of plates.There had to be a point to exhibiting things, Will decided(Nick Hornby
about a boy 1998:302).
A Chinese friend once told me that he had offered a pig to the dead. À whole
pig? ' I asked, somewhat surprised, since I knew he was far from being a rich
man. He laughed. `No. We fool them. What we do is offer the head and the tail,
maybe the feet. Then they fill in the blanks and assume we gave the rest too'
(Nigel Barley Dancing on the Grave: Encounters With Death 1995:77-8).
Roman Britain is too important to be left to the Romanists!(Barry Cunliffe 1979)
To the memoryof Barbara Meadows
Chapter 1
Introduction
Our vision of the Romano-Britishworld is expandingas surely as did the Roman Empire
itself. It is becoming more politicized through increasedconsiderationof the wide-ranging
effectsof imperialism, and more inclusive in termsof the rangeof peoplewho are considered.
The effect of the Romanconquestof Britain on nativepopulationsis increasinglyperceivedas
the lives of all who lived in Romana viable and worthwhile areaof study; as a consequence,
era Britain can be scrutinized.We now have the opportunity to explore more aspectsof the
lives of the people who lived at non-Roman-like settlementsin Roman Britain, and in the
processexamineand challengemany of the stereotypesthat haveresultedfrom the absenceof
in-depthanalysis.
This thesisis a responseto my initial queriesabouthow one might study the effectsof Roman
imperialism on the day-to-daylives of the indigenouspeople of Britain. What I hoped to
realize was somesenseof the negotiationand resistancethat inevitably must have occurredat
all levels of native society,not just amongthosewith power and wealth. There were two main
involved:
the methodological and the conceptual.Thus, a substantialportion of
challenges
this research involved the formulation and testing of a methodology for the analysis of
social change at non-Roman settlements.Only then could I undertake the conceptual and
interpretive task of addressingthe array of responsesto the Roman presence,responsesthat
touched all aspectsof life, including the mundaneor everyday. The approachthat I have
developedin this instanceis gearedtowardsthe studyof native settlementsin the early Roman
but
it
has
applicationsfor other periods and for other types of settlements,particularly
period
those that are considered more Roman-like. Many assumptions are made about the
Roman
of
style goods and structuresat these types of settlements(for example
significance
`villas') and in someinstancesthe social contextsof thesestructureshaveyet to be considered.
2
I. I.
Roman Things
Observation of the presenceand absenceof Roman-stylegoods and constructs has guided

much of the analysisof imperialism in Roman Britain and other parts of the Empire (seefor
1988).
Wealth
Trow
Woolf
1990a;
1990;
Millett
Haselgrove
1984b;
and power
example
have been assumedto correlate with the extent to which a group's material culture and
lifestyle appearedRomanized (Millett 1990a, 1990b:38). The concept of `Romanization'
has become the primary measurementof change in the lives of the people who were
is
Romanization,
there
there
an
conquered: and where
was only slight evidence of
did
lives
little
they
that
the
changedand continued much as
assumption
of people were
before the conquest (albeit with better tools and nicer pots). A number of scholarsare now
challenging this assumptionand this study contributes to the on-going exploration of the
in
Cooper
Webster
imperialism
(Hingley
1996,1997;
the
and
papers
wide-ranging effects of
1996; Barrett 1997a, 1997b;Mattingly 1997b; Webster2001). Clearly, we needto consider
the social context of the use or non-useof Roman style goods- this thesis aims, therefore,
to focus our attention on the conditions that surroundconsumptionpractices.
The omnipotence of Roman-stylegoods in Romano-British studies has nonethelesshelped

to construct this approachto the study of social change.Spurredon by the fact that many of
the signifiers usedto describe`Romanization' are tied to the consumptionof food and drink,
I decided to focus on eating and drinking habits, which are particularly sensitive to periods
food
is
The
drink
consumptionof
of social change.
much more than the observanceof
and
ingredients
how
it
is
the
eats
one
and
containers
of
and
where
consideration
particular
also
drinks,
and with whom and why. This study of consumption practices at native
and
settlements enables us to consider possible effects of imperialism that might otherwise be
ignored becausethe settlementsappearto exemplify native continuity.
1.2.
The Scope of this Research
There are a number of studiesof consumptionhabits during the Roman period in Britain but
they have a tendency to focus on individual aspectsof food and drink, such as the presence
of Roman-style pottery or animal species.Rarely is the totality of consumption considered
3
(although see Rippengal 1995; Hawkes 2001). This is partially attributable to the nature of
individual
interpretation
for
ingredients
by
and containers
site analysis, which separates
specialists.This is of coursenecessary,as analysis of excavatedmaterial requires different
is
budget
however,
the
to allow the
there
the
time
areas of expertise;
nor
often neither
artefacts and remains to be theoretically returned to their archaeological contexts for
integratedanalysis.Given the luxury of time that independentlonger-termresearchallows, I
have formulated a methodologythat attemptsto do just that.
This thesis therefore examinesthe social contexts of the consumptionof food and drink at
`native type' settlements during the post-conquest period in an attempt to access the
localized effects of imperialism. This requires consideration of ideas on the social
significance of consumption habits in terms of the theories of social change in Roman
Britain, which up until recently were containedwithin the conceptof `Romanization'. It also
Roman-style
between
the
the
goods and the
of
presence
requires assessmentof
relationship
Consideration
them.
that
of
consumption practices
are generally associated with
helped
has
to
the
consumption
and
anthropological works on
of
change
areas
social
interrogate a number of the assumptionsabout the use of goods of the dominant culture;
these works have also provided much of the stimulus to considering human responsesto
in
profound changes their social conditions.
A study of the social contextsof eating and drinking is both broad and detailed and requires
the acknowledgementof the many ways to interpret animal bones,plant remains, and their
containers,and the problems and benefits inherent in consideringthe artefacts and remains
interpret
how
is
However,
it
their
through
archaeological
contexts.
examining
we
within
archaeological remains that we can consider the social significance of the treatment of
artefacts and remains in, for example, different parts of a settlementas well as in the way
food and drink were preparedand servedover time.
My approachto the study of social changewas realized through an in-depth analysisof four
in
Upper
Thames
Valley.
The casestudy elementof this thesisconcentrates,for the
the
sites
in
both the late Iron Age and early Roman periods.
that
sites
on
were
occupied
part,
most
This not only provides a setting for the study of social changebut it situatesthe researchin
the immediate post-conquest period -a
be
is
to
time
that
pivotal
generally presumed
4
in
in
exceptionallyenigmatic the archaeologicalrecord of native settlementsexcept a very
few cases.The Upper ThamesValley was chosenas a study area for several reasons.The
the
has
site
the
settlements;
native
of
extractionof gravel
necessitated excavationof a wealth
let
Unit
to
Oxford
Archaeological
me
archiveswere accessible;and the
were very willing
in
is
for
through
their
still
plough
various records of sites
which post-excavationwork
progress.The four sitesselectedfor analysisare:Barton Court Farm,RoughgroundFarm, Old
ShiffordFarm, and ClaydonPike.
1.3.
The Data
As many published site reports do not emphasizethe archaeologicalcontexts of artefacts

have
I
for
the
on
relied
site
archive
and
primary records
each of the case
and remains,
important
is
behind
impetus
An
this
the
component
of
research understanding
studies.
interpretation.
I
has
been
can,
after
all,
and
only
comment
on
what
recorded.
excavation
Barrett statesthat "Objectivity is not a matterof 'unbiased' observation,rather it is concerned
to
the
the
evaluation
of
and
critical
way
observations
are
collected
used
constructan
self
with
(1987:
in
409);
the
the
this
thesis
past"
camepre-packagedand
sites
of
consulted
understanding
an important aspectof eachcasestudy was to examinethe natureof that package.
I do, however, believe that we can re-visit site archives and published reports and ask new
interests
My
in
bones,
the
particular
pots,
and plant remains are not necessarily
questions.
is
directors
interests
latter
The
the
the
specialists
of
the
or
affects what
of excavations.
but
is
in
does
much
of
as
what
the
contained
site
archives
not reach the published
recorded,
data
there
to
exists
an
the
opportunity
compile
on
and thereby
new observations
site report,
in
different
the
excavated
material
ways.
consider
It remainsfor me, however,to emphasizewhat this thesisis not.
This thesis is not a correction of what was donebefore. It is not meantto supersedethe work
but
it.
I thereforedo not considermy analysisof the sitesas a
to
rather
complement
of others
reinterpretation of the excavated material, but rather as a series of `newly collected
5
if
is
be
1987:
422).
1
(after
Barrett
this
thesis
would
observations'
very pleased
viewed as a
companionto the publishedreports.
1.4.
An Outline of the Thesis
The structure of this thesis follows the courseof my deliberationson how one goes about
studyingsocial changewithin an imperial context.
Chapter2-
'Romanization'and the Study of Social Change- establishesthe motivation for
this thesis, which stemsfrom my own uncertaintiesabout the conceptof `Romanization'and

the way it has beenapplied to the study of social change.The chapteressentiallyfollows the
course of my internal dialogues on imperialism (after Mattingly 1997a) and in so doing
presentsthe variousworks that haveguidedthis research.
Chapter 3-
Towards an Archaeology of Food and Drink Consumption - sets out the

food
by
how
discussing
the
this
thesis
of
consumption
methodologicalapproachof
studying
household
imperialism
drink
forge
localized
to
the
at
a
and
a way understanding
can
effectsof
level.
Chapter4- The Upper ThamesValley- providesthe settingfor the four sitesunderscrutiny.I

also presenta numberof the current interpretationson settlementin the Valley in the late Iron
Age and early Roman periods and reflect on how the characteristicsof excavation and
preservation, unique to the Valley, have helped to refine my approach to the study of
imperialism.
The four casestudiesthat constitutethe body of this thesisare outlined in Chapters5 through
8. They are presentedin the order of their analysis.The four sites are: Barton Court Farm
(Chapter 5); RoughgroundFarm (Chapter 6); Old Shifford Farm (Chapter 7) and Claydon
Pike (Chapter8). Thereis much to digestin the casestudy elementof this thesis.An approach
that considersthe social context of food and drink consumptionhas many tentaclesand the
data
in
its
the
all
of
of
various guiseshas beencentral to the developmentof my
presentation
In
approach.
an attemptto aid the reader,I provide a road map of sortsthrough
methodological
6
beginning
detailed
table
the
of each case study and through a series of
a
of contents at
summariesin the discussionof eachsite.
In Chapter9- Food and Drink Consumptionand the Studyof Social Change-I consolidate
someof the detail found in the casestudiesinto a discussionon the social context of food and
drink consumptionfor both the late Iron Age and early Romanperiods.
In Chapter 10- ConcludingRemarks-I offer somecommentsaboutwhere the study of social

change in Roman era Britain might go from here.
Chapter 2
`Romanization' and the Study of Social Change
2.1.
Introduction
In my original researchproposal,I put forth my intentions of studying the "effects of Roman

imperialism on the British ". Shortly after my arrival in Sheffield, my researchfocus evolved
into a study of the "effects of Romanizationon the British", and later to how one would go
about such as study (Meadows 1994). Discussions about the whys and wherefores of
`Romanization' aboundin the archaeologicalliterature. This chapter will set out the course
literature
deliberations
is
It
the
searchon the theoriesof
of my
on
subject.
not an exhaustive
Romanization, nor is it a complete history of the cumulative events that now dispute the
whole concept of Romanization (for discussionson the chronology of `Romanization' see
Hanson 1994; Woolf 1995; Rippengal 1995; and particularly Webster 2001); rather, it is an
account of my own dissatisfactionwith this ubiquitous term. Put another way, this chapter
setsout why `Romanization' was eventually removedfrom the title of this thesis.
2.2.
The Study of Social Change in Roman Britain and the Theory of `Romanization'
For the past one hundred years, theories of Romanization have structured the way social
in
Roman
Britain
is
described and explained. Recent discourse on the term
change
`Romanization' has highlighted many of the problemsof its usage,but the premiseon which
the concept of Romanization was basedhas been the subject of intense debatefor decades,
1970s.
A number of thesedebateshelped to structure my ideason how
the
since
particularly
to study of the effects of imperialism and they will be discussedbelow; however, before we
proceedwith a contextual analysis of the concept of Romanization, we must first define the
term.
8
Romanization- defined
The choice of definitions is of crucial importance, not becauseof what it

explains, but becauseof what it excludes - in other words, definitions
determine the scope and limits of an investigation (Gregory and Altman
1989:9).
`Romanization' was initially coined by British historian Francis Haverfield in terms of a

civilizing mission:
It has been said that Greece taught men to be human and Rome made
form
it
That
Empire;
the
took was
the
the
mankind civilized.
was
work of
Romanization(1912:11).
The civilizing mission of the Romanshas been redefined by scholars many times over the
years and a chronological selection of statementsand definitions' will help to illustrate
highlight
be
the
the
that
throughout
this
many of
points
will
chapter, as well as
made
diversity of opinion that still exists in the field.
"Romano-British culture arosefrom the impact of the civilisation of Rome upon the
Celtic people of Britain; the result, however, was not a replacementof cultures, but
rather what can broadly be describedas a synthesis"(Frere 1974:342)
"In the coloniae, civilized life was predominantly Romanized from the start, but in
the early civitas capitals it was something new, which the natives had to learn, and
the life there might at first be better describedas progresstowards Romanization"
(Wacher 1974:45).
"... Romanization is defined as being contact between...the native and the Roman,
the contact had for both cultures" (van Es 1983:5)
and all the consequences
"Romanization, the process of change through the interaction of pre- and protohistoric societieswith the Roman Empire" (Slofstra 1983:71)
"Romanization, better describedas the fusion of imperial and local institutions and
cultures, was the joint product of central governmentand local initiatives"(Garnsey
Sailer
1987:202).
and
1 It is only quite recently that scholars have felt the need to define explicitly
what they mean when using the
term `Romanization'.
9
"Romanizing -a tendency to homogenize the material culture of the island of
Britain, or parts of it with the material culture of the nearbyprovinces of the Roman
Empire" (Reece1988:11)
"[Romanization] a one-sidedform of acculturation" (Okun 1989:13)
"[Romanization] the study of the two-way process of acculturation" (Millett
1990b:37)
"... Romanization should be viewed as just one part of the broader strategies adopted
by various groups in the playing out of tensions within a complex hierarchical
structure" (Rippengal 1995)
Through the course of my preliminary reading, Romanization was to become almost

indefinable as it so clearly meant different things to different people. I had yet to encounter
or establish a definition with which I felt completely comfortable. In my initial attempts at
defining Romanization I found van Es' definition (seeabove) the most palatable becauseit
did not specify the type of contact necessaryfor Romanizationto occur; and by considering
all the consequencesfor both groups it did not place a value on the importance of one over
the other. The definition also did not contain the word that puts fear in the hearts of
Wilk
86;
1987:
its
(Moore
anthropologists - acculturation and
association with progress
1990:41; Barrett 1989a) - and it did not, I theorized, come equipped with its own
generalizedconclusionthat precludedspecific application.
It is essential to avoid judgements about the form and process of

Romanisation in Britain. Romanisation has many different forms and
encompassesall of the consequences- the occasionaland the commonplace
as well as the dramatic and the ordinary - brought about by the infiltration of
the Roman world (Meadows 1994:133).
However, in establishing a context for the study of Romanization that emphasizedthe

dynamic betweenRoman and native, I was to realize that the conceptwas not an appropriate
account of cultural change.
10
2.2.2. The context of my `Romanization'
Since the late 1970s, many Romano-British scholars have reacted against the notion that
Roman imperialism
and Romanization
were synonymous; and that the study of
Romanization should be viewed only from a Roman standpoint, starting essentially at the
it
Romano-British
point of conquest.
studies,
was argued, were preoccupied with the
assertions of contemporary sources (Caesar's The Conquest of Gaul, Cassius Dio's The
Roman History: The Reign of Augustus, Strabo's Geography etc.) and particularly with
Tacitus's account of an imperial policy of Romanization in Britain (Agricola)
1979:230; Burnham and Johnson 1979:2; Millett
(Reece
1990b:36; Branigan 1991:92; Hingley
1991:93; Scott and Gaffney 1987). It was further argued that the classical interpretation of
Roman Britain had produced a seemingly static characterization of Roman Britain that was
one-sided and effectively discounted the roles and experiences of the majority of British
people (particularly those situated in the north of the province) and their daily habits (Barrett
1989a, 235-36; Hingley 1989:3).
An emphasis on the dynamic between Roman and native, however, required a different
approachto, and new agentsof, Romanization. Garnseyand Sailer (1987), in looking at the
its
because
limiting
Romanize,
the
to
the
of
efforts of
empire
regarded
empire's role as
imperial policy. They concluded that Romanization was primarily "self-directed
after
...
Rome made its initial impact" (1987:203; for similar commentsseeMillett 1990a:7; Millett
1990b:37; Haselgrove 1990:45). Others argued that Romanization received important
direction from the pre-existing social structures that were themselves adaptable to the
Roman way (Haselgrove 1984b:6; 1990:45-6; Collis 1987:36; Millett 1990b:37). In pushing
the point of enquiry back into the Iron Age, Romanistswere introduced to the thoughts and
ideas of pre-historians,with some amusing consequences:"Roman Britain is too important
to be left to the Romanists" it was claimed (Cunliffe 1979:359; see also comments in
Bradley 1990).It was no longer appropriateto look at evidencethrough purely Roman eyes.
Native peoples were now seento have had an important role in their own Romanization,
(Millets
1990b:38). The emphasis was on continuity as well as
native
elites
especially
1981:
94-5;
(Branigan
Collis 1987:36) and many `native' and `Romanized' sites
change
for
their significance in the processof Romanization (Branigan 1981:83;
were re-examined
11
see also Slofstra 1983:97). Additionally, earlier interpretations of typically Roman
constructs were challenged,as exemplified by the reconsiderationof the origins of towns
from the native perspective(Burnham 1979:255; Todd 1985:187). Models were developed
that emphasizedthe control that native elites had in the Romanizing process.The exchange
of raw materials for prestige Roman goods (known as the `prestige goods model' core/peripherymodel), for example,was seenas the meansby which both Rome and native
increased
their power through the monopolization of the goods (Haselgrove 1984b;
elites
1989).
Romanists looked to the past to substantiatepre-existing theories. They examined the

infrastructure of Iron Age society to correlate the process of Romanization with the preexisting society. It was generally assumed that fast, effective Roman occupation was
conditional on a minimum level of social complexity (Groenman-van Waateringe
1980:1037; Roymans 1983:56-7; Hanson 1988:66; Millett 1990a:40). The two essential
prerequisiteswere: 1) an economic/agrarianinfrastructurecapableof feeding and supplying
the Roman army and 2) a social structure that could accommodate the Roman
administration (Groenman-vanWaateringe1980:1038). Discussion evolved aroundwhether
late Iron Age society measuredup to the requirementsof Rome, resulting in a tug of war
acrossthe divide of Iron Age and early Roman society. Romanists,for example,using their
definitions of urbanism,equatedurbanization in Britain with the Romans(Wacher 1974:367 although seeBurnham and Wacher 1990:8; Frere 1974:273; Branigan 1994:10-11) while
Iron Age specialists considered the oppida an urban settlement (Collis 1979; 1984:2;
Cunliffe 1976:135). Further debatealong theselines sought to determine whether Iron Age
had
it
society
what took to provision the four legion strong army (cf. Millett 1984:68-9 and
Fulford 1984:131; also Jones 1991:25). The result was that both late Iron Age and early
Roman society were being ill defined. Haselgrovespelled out the dangersof proceedingon
such as course:
both
have
archaeologists
taken
working
on
periods
up entrenchedpositions
...
in relation to the achievementsand complexity of late Iron Age
society, an
opposition which, if it is allowed to persist, can only hinder our
understandingof Roman Britain (Haselgrove 1984b:5; for similar comments
seealso Woolf 1993:213).
12
Regardlessof the intentions, how we have subsequentlydefined the structure of late Iron
Age settlementpatterns,social complexity and subsistencepotential has been in relation to
the up and coming Romanperiod.
The point at which I entered the debate on Romanization saw Richard Reece (1990) in his
article "Romanization: a point of view" attempting to uncover the best vantage point from
which to view Romanization objectively. He argued that researchers should go beyond the
amalgamation of native and Roman and look at Romanization as part of a process that went
beyond Rome (1990: 31). Haselgrove (1990) was suggesting that we stop viewing
Romanization as a single phenomenon experienced by all and to the same degree, and
instead consider it as something that was adapted and adopted in different ways, not only
from province to province, but also from group to group. Millett (1990a), probably one of
the most influential scholars on Romanization at the time, was arguing, in his influential
book The Romanization of Britain, that Romanization was a process of "dialectical change,
rather than the influence of one `pure' culture upon others" (1990a: 1). At the same time,
Hingley (1991; 1989; 1988) was stressing that current studies of Romanization, focused
only on the wealthy and powerful and contained within them notions of progress and
Romanization
(1989a)
Barrett
the
the
advancement.
concept of
was questioning
relevance of
itself, because of its static, one-sided characterization of the phenomenon of imperialism
(Barrett 1989a, 235-6). The views of Reece, Haselgrove, Millett, Hingley and Barrett depict
a general dissatisfaction, which as we have seen had been brewing for a number years (see
in particular articles in Burnham and Johnson 1979), with the study of Roman Britain. The
dissatisfaction with the theoretical underpinnings of Romanization echo, at least in part, the
concerns expressed by other disciplines (for example anthropology and sociology) with
indigenous
an
perspective to their work and perhaps more generally a rejection of
providing
western imperialism in the world (Hingley 1991:97; Hanson 1994: 149; Webster 2001: 7).
A new venue for Romaniststo vent this dissatisfactionwas establishedin the early 1990s
the annual Theoretical Roman Archaeology Conference (TRAC). TRAC provided (and
continues to provide) a much-needed arena for the integration of theory and Roman
Eleanor
Scott,
the originator of TRAC, observed in the first publication of the
archaeology.
"The
that
underlying concept of `Romanization' was ever-present,and with it
proceedings
13
was a clear understandingthat this is more thanjust a useful term: it is a processwhich must
be describedand defined" (1993:1).
The debatesabout the structure of pre- and postcolonial Britain neverthelesshelped to
reveal actually how little was known about the inner workings of Romano-British settlement
and society in general (Bradley 1984:151; Trow 1990:109; Gosden 1989:381; Haselgrove
1989:11). Scrutiny of the evidence, for instance, has shown that Roman imports were
actually quite low in number and localized in distribution (Millett 1990a:30; Woolf 1988:3;
Sharples 1990; Hill 1989). The over-estimation of the number of Roman imports and
therefore the impact of the control of trade goodshas consequentlyweakenedthe `prestige
goods model'. Additionally, it has been shown that the model failed to addressor explain
the continued use of British prestige goods well into the Roman period (Collis 1987;
Gregson 1988:22-3). One of the main problems with models such as the core/peripheryprestige goods model and other assumptionsabout the impact of Rome was that they
in
is
`sameness'
).
This
(J.
D.
Hill
to
reduced societies
especially evident the
pers. comm.
desire
Through
in
Romanization.
to
the
theory
a
assumptionsmade about native society
of
shift the balance of power from Romans onto natives, the effects of imperialism have, I
would argue, been neglected(see below). One of the aims of this thesis is to demonstrate
that in viewing native (and `Romanized') settlements as homogeneousentities we are
ignoring the diverse experiencesof imperialism (Mattingly 1997b:9).
It is also the casethat the late Iron Age and early Romanperiods in Britain have largely been
defined by the logistical requirementsof Rome. The intricacies of inter- and intra- regional
differences
for both periods have been neglectedin order to show that the
similarities and
late Iron Age and early Roman native society was not populated by the "uncivilized" as
Caesarand some Romanists would have us believe. Native and Roman are still generally
thought of in terms of continuity and change- continuity as the continuation of later Iron
Age and native traditions (which ironically have been defined by their proximity to the
Roman period), and changeas all that which is brought about by the arrival of the Romans.
This is a false dichotomy that assumesthat both `native' and `Roman' were stable and static
when of course all societies are in a state of change(Shanks and Tilley 1987:130; Jones
1997:134; Woolf 1995:346-7). Another aim of this thesis, therefore, is to chart some of the
14
include
discontinuities
to
settlements,
somewhatparadoxically,
of
non-Roman
and
changes
the social contextsof the useof Roman-stylethings.
2.2.3. Roman `things'
Goods are neutral, their usesare social; they can be used as fencesor bridges
(Douglas and Isherwood 1978:12).
is no logical reasonwhy the historical processeswhich brought the
there
...
Roman province [in Britain] into being needhave correlateddirectly with the
categories of the material evidence with which we choose to work today
(Barrett 1993:344).
It was, ironically, the consideration of Roman-style goods in the late Iron Age period that
The
decided
how
I
Romanization.
to
the
chronology of
ultimately
was
view
concept of
Romanization (and hence its association with imperialism) was obscured by the notion of
pre-Romanization'. As so much of what has been used as evidence of Romanization during
the Roman period was derived from the presence of Roman things, the debate surrounding
the concept of `pre-Romanization' was to help clarify my position on the subject. Woolf
(1988) was critical of the whole idea of pre-Romanization for the very reason that it led to a
simplistic association between the use of Roman things and Romanization. For Woolf,
Romanization was as much the "transformation of customs and values" as it was the
adoption of Roman material culture (1988: 9). He argued:
the selective adoption of some Roman goods, without the information that
governed their use and value in the classical world, does not constitute
Romanization in any meaningful sense(1988:9).
This is an important consideration in any study of Roman imperialism, but one, I would
be
impossible
that
to make at the point of defining Romanization. As Freeman
would
argue,
has observed, the current dispute with traditional Romanization studies "... questions the
very evidenceusedto characterizeand measureand quantify Romanization" (1997a:28).
15
`Roman' has come to have less connection to the city of Rome than it once had (Reece
1988: 11; Freeman 1991: 135). Reece has suggested that we view `things Roman' as
"material better known inside the Roman empire than outside" so that it is the empire and its
provinces that define what is Roman rather than the city of Rome: "Britain became more
Gaulish, more Rhinelandish, more Spanish, a little more Italian, a very little more African
and a little more Danubian" (Reece 1988:11) in the process of becoming Roman. In
response to Reece, Barrett argues "The problem, however, is that Reece simply replaces one
questionable category with a string of equally problematic ethnic labels" (1997a: 51). He
states:
I do not doubt that cultural change occurs, but I will argue that apparently
homogeneous cultural systems are in reality unstable internally and
multifaceted in terms of their meanings. We should therefore question
whether the concept `Roman' is a useful starting point in any analysis
(1997a: 51).
Reece's observation that many of the markers used to characterizea Roman lifestyle are
from the Roman provinces rather than from Rome is, nevertheless,an important one (see
also Hingley 1999:142; Jundi and Hill 1998:134; King 1999:189). Native and Roman
culture were not static and recently scholarshave been arguing that an imperialized culture
is neither native nor Roman as both were changedby conquest(Woolf 1995:341; Grahame
1998:4). Cooper (1996), for instance, arguesthat the labelling of material culture in the
Roman period as `Roman' is a mistake when what archaeologistsactually mean is "the
material culture of Britain during the time it was part of the Roman Empire" (1996:86)
adding, "the material culture of conquered populations must therefore be seen to have
remained their own even when it adoptedexternal elementsand styles" (1996:86). Reece,
Barrett and Cooper, in different ways, demonstratehow `Roman', could mean different
things to different people in different contextsand ultimately why these terms should be too
difficult to define.
My struggle with defining Romanization was directly related to my desire for some sort of
for
imperialism.
`Roman-stylethings' found in a native burial were obviously
social context
but
were the objects meaningful because they were known to be Roman,
meaningful,
becausethey were foreign, a gift, or becausethey were chosenand placed in the grave by
Conversely,
was it at times the contents of the Roman-style vessels
someonesignificant?
16
that were important rather than the container itself? Moreover, what was the actual
life?
Were
incorporated
Roman
they
the
things
context
of
within
everyday
significance of
into the daily rituals of living or were they just used on particular occasions and with
in
Roman-style
Does
the
signify
a
use of
utensils necessarily
change the
particular people?
social contexts of eating and drinking? How does the use of non-Roman things in Roman
ways fit into the current understanding of Romanization? As Webster has recently argued,
leave
interpretations
for
`native'
`Roman'
and
material culture
no room
ambiguity
current
of
(Webster 2001: 9). The notion that there was a shared understanding of Roman culture
throughout the Roman Empire only goes against the conception of alternative experiences of
imperialism (Barrett 1997b:7; Mattingly 1997b:9; Hingley 1999: 143).
Freeman (1993) and Cooper (1996) have argued that the adoption of Roman-like
accoutrements could have more to do with availability and access, than with a desire to
but
have
been
`Roman'
This
the
case
at
some
settlements
emulate
practices.
could very well
in
defining
Hingley
has
"active
the
as
role of native society
pointed out, we must not negate
the function, value and role of its own possessions" (1996: 42; for similar comments see
Grahame 1998:2). Emulation of a lifestyle considered `Roman' is one likely factor in the use
been
has
for
Roman-like
been
has
The
the
things
that
of
emulation
some people.
problem
prevailing assumption and has led to a one-dimensional view of social change. Mattingly
has called for a more introspective consideration of the material culture of the Roman
period: "... there were many divergent approaches and value-systems at work rather than a
simple pattern of emulation behaviour" (1997b: 17; for similar comments see Hingley
1999: 144; Webster 2001: 8; Miller
1995:27). Moore (in reference to the concept of
`westernization') has argued that too many assumptions are made when emulation is used as
an explanation of social change (1987: 86) adding, "similarities
in the indices of social
change should not be allowed to mask differences between the processes of change they
(1987:
85).
Foremost,
represent"
we need to consider the cultural implications and social
Roman-like
is
What
any
of
material
context
culture.
perhaps most apparent in all of the
debates about Romanized material culture is that the `shopping-list' approach to the study of
in
Roman
Britain
requires a radical rethink.
social change
17
2.2.4. To Romanizeor to not Romanize
The theory of Romanization is currently at a crossroads. Do we redefine Romanization and

make it relevant or do we decide that the term has been too used and abused to ever be
has
been
One
Romanization
it
is
the
the
that
criticisms
of
concept
relevant?
of
main
of
in
its
thinking and evaluates material culture while reducing the active role
evolutionary
played by ordinary people and their daily habits. This concept of Romanization is still alive:
"Changes in culture and everyday life still seem to be considered as a product - perhaps
even as froth on top - of large scale, impersonal, social and economic transformations (or
just another manifestation of a prestige goods system)" (Hill 1997:97; see comments by
Woolf 1998:5-7). Mattingly in the introduction of Dialogues in Roman Imperialism (1997b)
suggests that the Roman Empire is still viewed as a conglomerate; he explains, "part of the
problem lies in the monolithic nature of most visions of Roman imperialism, whether
praising it or damning it utterly" (1997b: 7). He argues that the term `Romanization'
encourages this vision through its implied "unilateral
transfer of culture"
and the
imperialism
9).
become
(1997b:
thus
experiences of
standardized
However, would the removal of the term from our vocabulary eliminate the idea of
Romanization? Perhaps, although I believe we need to dismantle the concept of
Romanization rather than presenta false senseof profundity through proscribing othersthe
use of the term. The effects of the theory of Romanizationare in all of us who study Roman
Britain, including the excavationson which our theories are currently basedthat continueto
emphasize Roman settlements and Roman material culture (Jones 1997:36-8; Hingley
1999:141; 2000:149-50).The term has neverthelessbegunto lose its applicability, and when
it is used, scholars now feel compelled to define their usageof the term (see for example
Hanson 1994:150; Rippengal 1995; Haussler 1998:10; Woolf 1998:7; Turner 2001:2; and
Hingley 2000:112 who only uses the term in the context of his critique of the concept of
Romanization). If we are to further the dialogue on imperialism we might do better to
demonstrate,rather than dictate, how the basic concept of Romanization does not fit our
imperialism
(for similar commentsseeWebster 1996:15 note 8).
models
of
various
Our goals with respect to the study of the past are changing. We are perhaps more
knowledge
is
the
there
that
with
comfortable
not one model to fit all of the experiencesof
18
Roman imperialism. "Too often, perhaps,scholarshipcreatesdichotomieswhere there are in
fact a range of possible actions, reactions,and perceptionsin between the extremesof the
by
Gaffney
1987:
64;
Scott
85).
These
(Mattingly
1996:
comments
see
also
and
are
argument"
healthy contradictions(Webster 1996:1-2), as there will alwaysbe people that see `Romans'
and others that will see`natives'.
2.3.
The Study of Social Change in Roman Britain: Negotiation and Resistance
history
like
British,
the
of
empire,
could
surely
with
a
recent
only
a
people
...
few
Pax
Romana
that
the
the
with a
noble
accept
so uncritically
construct of
exceptions the archaeology of resistance to Rome still remains outside the
mainstream of study (Webster 1996:4-5).
With the first line of my original thesis proposal, I stated my interest in studying the effects
have
Romanization
imperialism.
doubts
My
the
strengthened
of
subsequent
concept of
over
I
imperialism.
However,
helped
following
to
the
the
and
social context of
study of
refine
formed
I
bring
ideas
that
to
this
thesis
the
whilst an
also
many of
and arguments
before
in
I
in
Canada
Ontario,
undergraduate majoring
where studied societies
anthropology
English
is
Latin
As
European
America.
the
woman
an
and after
colonization of what
now
born to Welsh parents, and now a resident of a former British colony that is physically,
economically and emotionally attached to the United States, my influences are many.
Increasingly, it is being recognized that our knowledge of the Roman Empire and Roman
Britain is influenced by our own history, politics and education (see Hingley 1993; 1996;
particularly 2000; Barrett 1997b; Webster and Cooper 1996; and for archaeology in general
Gero 1996). Roman studies have lagged behind many other disciplines,
such as
anthropology, that are more willing to deconstruct their existence. Hingley has made a
strong argument that the history of the British Empire's `civilising mission', is still reflected
in our benevolent accounts of what the Romans did for us (1993; 1996:41, see also Freeman
1996).
In many respects,the preoccupation with Romanization has skewed our consideration of

did
In
Romans
incorporate
into
to
the
to
us.
the processof
our
efforts
native elites
what
Romanization, we have forgotten that the Roman conquest of Britain involved the social
19
and political control of people(Hingley 1997:82). In order to recognizeevidenceof the many
experiencesof imperialism, especially evidencefor resistance,this evidence must first be
it
in
be
`seen'
Only
then
the archaeologicalrecord:
can
acknowledged.
Perhapsthe poor and the powerlesssubtly resisted change,but how can we

challenge the progressivemodel of Romanization outlined by Haverfield,
Millett and othersif we do not excavatetheir homes(Hingley 1997:85).
Settlements,when they are inventoried primarily for their Roman-like accoutrementsare too
discarded
if
fall
they
often
outside of the boundariesof a theory of Romanization.After all,
(to
Freeman
settlements
are
settlements
paraphrase
native
native
are native settlements
1997b:9). Some of these settlements will be represented in this study. It is at these
abandoned`non-Romanized'settlements,where the majority of people lived, that daily acts
of resistance might be found (Webster 1997:180) and it will become apparentthat I seea
number of the changesin settlementpatternand consumptionpracticesat someof the sitesas
evidenceof resistanceto the Romans.Acts of resistance,however, are not only to be found
in the use of particular `native' icons (Hingley 1999:144); a resistantideology might also be
including
through
the
tactical
expressed
negotiation of a variety of cultural symbols
`Roman' ones (Webster 2001). However, it is also essential that we do not replace one
monolithic model basedon emulation with anotherbasedon opposition (Hingley 1996:44;
Kurchin
1995:124-5;
Mattingly
1996:64);
or
likewise
establish
new
`Romanization'/`resistance' dichotomy along the lines of the Roman equals change/native

equals continuity dichotomy described above (see comments in Woolf 1995:340-1;
1998:22-3; Webster 1997:167). Grahame has argued that the changes found in the
archaeological record "are not the result of Roman `domination' or native `resistance',but
document
the social politics of Roman Britain" (1998:8). The material culture of
rather
Roman Britain neverthelessneedsto be approachedfrom within the context of imperialism,
far
I
to suggestthat the considerationof the social contextsof material culture
and will go as
could potentially challengemany (if not most) of our currentperceptionsof RomanBritain.
20
2.4.
Conclusion
Many aspects of the Romanization of Britain, including the use of the term itself, are
imperialism
in early
The
the
effects of
currently under scrutiny.
scope of my. own work Roman Britain - has evolved in accordance with ' my own uncertainties regarding the
concept of Romanization. Throughout, I have been motivated by what I saw as a lack of
emphasis on the possible diversity of Romanization, on the imperial experiences of the nonelites, and on localized responses to the Roman presence. Underlying my approach have
been nagging doubts about the cultural evolutionary tone of a discourse on Romanization
that equates Romanization with `progress'. Barrett (in reference to definitions of ritual) has
also been critical of approaches to the past whereby a phenomenon is first defined and then
applied to a particular set of circumstances: "such studies give the unfortunate appearance of
knowing already what it is they are attempting to discover" (1991: 1). The current round of
dialogues on `Romanization' largely instigated by those of us who believe the concept is ill
However
forces
justify
the
term.
to
of
conceived and outmoded,
us
and explain our usage
tempting it is for me to give up the language of Romanization (because I agree with most of
the criticisms of its usage) I do believe that, unchallenged, the underlying concept of
Romanization would survive, albeit incognito. I will argue instead, that Romanization is too
simplistic an explanation for the social change of the majority of the Romano-British
population.
There is a tendency in Romano-British studies to equate `change' with Romanizing

indigenous
urban/villa-owning
elites and `continuity' with rural non-Romanized natives.
This imposed dichotomy is also used to describe the nature and extent of Romanization,
which is currently the principal model used to describe social change during the pre/post
for
(see
conquest period
example Slofstra's definition of Romanization above). The
exploration of social conditions during this period of imperial domination has consequently
been sidelined by efforts to allocate particular cultural achievements,such as urbanism, to
late
Iron
Age
Roman
the
or
either
periods. This tug of war acrossthe lines of history and
between `natives' and `Romans' has served its purpose by exposing the cultural
evolutionary tone of many of the arguments,but the time has come to move on.
21
We must challenge the inventories of `Roman' and `native' material culture, so that we can
incorporate different types of settlements and the experiences of people of different sociointo
discussions
However,
backgrounds
`Roman'
Britain.
as this study of the
of
economic
illustrate,
develop
drinking
to
also
need
will
we
paradigms for
social context of eating and
imperialism.
for
An all-encompassing concept
that
the
social change
experiences of
account
have
Romanization
of
a place in such discussions. Nevertheless, before
will ultimately not
dispose
deconstruct
it
has
term,
the
the
the
acknowledge
and
we
of
we must
concept
role
played in the construction of our Roman Britain.
In the following chapter, I will chart my approachto the study of social change in early
RomanBritain.
22
Chapter 3
Towards an Archaeology of Food and Drink Consumption
3.1.
Introduction
The Roman conquestof Britain had an impact on the daily lives of the peoplewho lived there
incorporated
into
be
but
has
to
an
obvious
statement
our accountsof
yet
perhaps, one which
the lives of the majority of the native population (Scott and Gaffney 1987:85; Branigan 1991;
Rippengal 1991:222; Hingley 1997:84). It was a generaldiscontentmentwith the polarization
of `Romanization' and `native continuity' that initially led me to consider whether the
into
insight
diet
level
household
the
subtletiesof
could provide some
and culinary practicesat
the early experiencesof imperialism. People of all backgroundseat and drink in culturally
specific ways, and many of the items usedto gaugethe impact of Rome are those which are
usedwhen consumingfood and alcohol.
The accumulation of daily life and the extent to which it is repeatedis significant to the
structure of society (Heller 1984; Conkey and Gero 1991:15-6; Johnson 1989:208). Day-today life and its socio-political-economic context are of course inseparable: "social
structures...are the medium as well as the outcomeof social practice" (Moreland 1992:116;
Barrett
1989b);
however,
the analysis of the daily habits of life is the fundamental
seealso
level at which specific and diverse aspectsof society and hence the localized effects of
imperialism can be explored.
This chapter is essentiallyan outline of my route into the study of the effects of imperialism.
It provides the framework and some of the thinking behind my methodological approachto
the study of social change.I will first explain why the household has been selectedas the
for
locus
Following
this, I will focus on the benefits of looking at the
study.
primary
dynamics of eating and drinking in providing an alternative perspectiveof imperialism. I
23
integrating
food
drink
discuss
the
then
the
and
of
with the
will
social context
practicality of
vagariesof the archaeologicalrecord.
3.2.
Towards a Study of Social Change at Native Settlements in Roman Britain
3.2.1. The household
The household has been adoptedas a primary unit of analysisbecauseit is a focal point in
the enactment of daily life. Two factors render the household an ideal concept for
archaeological interpretation. First, it immediately establishes the organizational and
conceptual level of analysis.Second,it emphasizesthe actions of people within a specific
context as opposedto just the context itself; it signifies a less contained view of domestic
life - "people do not live in, or act exclusively in, single buildings" (Rapoport 1990:12; see
Hayden
4-5;
Scott
461;
Allison
1999:
Gaffney
1987:
87;
Low
1990:
Lawrence
also
and
and
and Cannon 1983:160). The non-contained nature of the archaeological record partly
isolation
living
the
this
their
of remains
reflects
movementof people within
environmentas
is
features
in
individual
buildings
houses
time
to
their
and
space and
and
associated
160;
1983:
(Smith
Hayden
Cannon
1992;
Maltby
1985a;
1981;
Branigan
problematic
and
Hirth 1993). As a workable concept,the householdis thus particularly viable because,while
its primary locus is the houseand its associatedenvironment,it doesnot expect artefact sets
to be confined to their activity areas.This flexibility in expectationsof the archaeological
does
not of course eliminate the uncertainty behind artefacts and their contexts, an
record
issue which will
be discussed below. Nevertheless, the discordant nature of the
archaeological record never seemsto hinder the generationof syntheseson a grand scale
(R.F.J. Jones 1979:3; Conkey and Gero 1991:7).
The importance of the workings of the householdto archaeologyis now being realized with
the movement away from the view that the domestic side of life is natural, familiar and
(Tringham
1991:
100; Moore 1988:55; Carstenand Hugh-Jones 1995:4-5; Allison
constant
1999:2). Archaeologists have tended to focus on the big economic and political picture,
from
the so-called private domestic sphere into the public sphere
moving rather quickly
24
(Tringham 1991: 120; Allison 1999:2; see also Yanagisako 1979: 189). Specific studies of
households, however, have revealed that the standard distinction between public and private
life is not quite so rigid (Moore 1988:30; Yanagisako 1979: 191). An illustration of this can
be found with the study of households in the Mantaro Valley in Peru immediately before
both
Through
èlite'
Inka.
the
the
the
studying
consumption
patterns
of
and after
arrival of
in
it
households,
`commoner'
to
the shifts of power.
a
change
was
recognize
and
possible
Before the arrival of the Inka, the elites were the predominant users of highly decorated
foods
higher
such as maize, chilli
proportion of preferred
storage vessels, and consumed a
between
in
distinction
After
the
two
the
their
the
social
conquest,
groups
peppers and coca.
the use of these items was less acute. How these goods were stored also changed. The Inka
constructed large storage buildings on the outskirts of the community, which displaced the
domain of household storage of preferred goods from the local elites to the governing Inka.
It was concluded that the changes in consumption patterns reflect a shift in the control over
from
derived
longer
local
leaders
the
their
to
status
access
no
prestige goods, where
Hastorf
1990).
but
from
Inka
Earle
1989;
(Costin
the
see
also
community,
state
and
The study of where and how people live can be used to generate new ideas about how
Carsten
Samson
1990;
1994;
(e.
Parker
Richards
Pearson
and
their
people view
g.
world
and
Hugh-Jones 1995:3). For those interested in the effects of imperialism, as was seen above in
the case of the Inka, the household can provide an ideal setting for studying localized effects
identify
be
(Hastorf
1990:
262).
It
to
subtle acts of
and responses
possible
may also
resistance to outside forces from within the security of the household environment (Webster
1997: 180). In many agrarian societies, the household is the basis for production and
consumption - areas that are particularly sensitive to what is occurring in the community
beyond
(Smith
1987:297). It is an ideal unit for comparison and, as most people live in
and
form
dwelling,
is especially suited to studies interested in crossing the social
of
some
spectrum (Smith 1987:297; Hirth 1993:21). A study of households can also shed light on
held
by
the group which may not find expression elsewhere (Ardener 1993: 14).
attitudes
Their examination can help to reveal some of the traditions and customs that serve to
in
the ways of the group, community and/or region. Finally, and of
socialize members
interest
to this study, changes in the household environment often reflect changes
particular
25
in
in the political environment of a society, although not necessarily
an obvious,
1993:
30).
(Johnson
straightforwardmanner
3.2.2. Housesand Roman imperialism- from roundedto rectangular

Building - is a processthat is continually going on, for as long as people
dwell in an environment. It doesnot begin here, with a pre-formed plan, and
finished
The
form
is
but
`final'
fleeting
there,
artefact.
a
a
with
moment
end
in the life of any feature,when it is matchedto a human purpose, likewise cut
intentional
flow
(Ingold
1995:
78)
from
the
of
activity
out
Romano-British houses commonly feature in studies of Romanization. Roman and nonRoman types have been characterized,and their form and distribution have been used to
Roman
British
how
Roman
to
the
the
the
conquest
world
the
response
and
of
assess nature
by
from
houses
is
The
to
interpreted.
though
rounded rectangular
switch
characteristicwas
is
Roman
in
1997)
England
(Hingley
the
often
and
of
period
southern
universal
means
no
influence.
In
Roman
Roman
the
the
and/or
continental
period,
of
early
sign
a
as
viewed
buildings
interpretations
their
of
was
more
sporadic,
and
as
such,
of
rectangular
construction
degrees
Romanization.
`Romanization',
The
to
term
varying
of
presence often point
however, is misleading and its use as a barometerof change does not provide an adequate
in
including
those
that
the
took
changes
structural
place
at settlements,
account of many of
Roman
The
does
during
term
the
early
period.
also
not promote alternative
this study,
Roman
buildings.
for
for
It
has
been
the
of
style
presence
example,
suggested,
explanations
building
in
from
Britain
deviation
the
classical
villa
the
points to a non-Romantype of
that
by
joint
families
Rippengal
1993
based
1978;
(Smith
ownership
extended
on
see
occupation
for critique of this view). The emergence of the basic rectangular corridor house is
in
design,
its
Roman-like
but
terms
to
techniques
of
construction
according
considered
Blagg (1990:206), does not appearto facilitate the Roman custom of regular entertainment
home
(see
Black
in
feasting'
It
1994:
106-7
`Celtic
the
also
villas).
clients
at
on
of guestsor
has further been suggestedthat the interior of rectangular structures might have been
Iron
i.
in
Age
that
more
was
reminiscent
of
way
circular structures, e. central
organized a
1990a
(sleeping)
(Hingley
(cooking
spaces
and
peripheral
eating)
and
private
spaces
public
from
The
been
366).
1996:
has
Lyons
to
shift
rounded
rectangular
also
consideredas
seealso
26
in
towards
a
movement
evoking
a more permanent
change
mindset,
of
a
profound
evidence
35).
1993;
Clarke
living
(Rippengal
1998:
although
see
environment
and constructed
Conversely, the persistenceof the round house during the Roman period is increasingly
being considered as a statement of identity and resistance (Hingley 1997,1999; for
in
Roman
houses
Harding
1984:
18-20;
the
period
see
particularly
of
round
examples
Keevill and Booth 1997) and even as a form of social control (Keevill and Booth 1997:41).
Lyons (1996) in her case study of house shapesin Northern Cameroonarguesthat the shift
by some groups from rounded to rectangularstructureswas more complex than the general
belief that changesin settlement are simply the result òutside influence'. She concluded
that changesin living space were part of a conscious,outward-looking strategy to appear
both `modern' and on the side of the governing group (1996:365). Shenotesthat rectangular
front
in
lived
family
built
the
the
that
those
of
at
who
compound and
structureswere often
highway
built
(Lyons
In
1996:
364-5).
the
rectangular
structures
concordance
of
range
visual
Lyons
(above),
(1990a)
did
in
house
Hingley
that
the
argues
change
not appear
shape
with
inner
Lyons'
ìntegrity'
household
366).
(1996:
have
the
the
the
of
to
working
of
altered
importance
illustrates
from
the
the
of
considering
shape
of
a structure
within the social
work
(and
Equally,
the
the reconfiguration of a settlement
settlement
region).
whole
context of
in
house
is
shape
as significant as a settlement whose
without an attendant change
during
dramatically
this
turmoil
of
political
period
changes
or appearson the
composition
Wilk
little
As
be
"The
faces
inward
house...
both
to
affected.
reminds
us:
and
outside
household
(1990:
to
the
is
40);
the
to
and
rest
this
of
society"
equally true of the
outward,
1987:
87).
Gaffney
The
(Scott
is
and
redefinition
categories
of
perhaps an
settlement
be
households
it
is
can
through
their
useful
where
area
units
of
study,
analysis
as
additional
possibleto scrutinize conventional stereotypes.
3.2.3. The non-RomanRomano-British household
In the closing sections of Chapter 2, I discussedhow there has been a general neglect of
Romano-British housesassociatedwith the poor. This not only distorts our perceptionof the
but
Britain,
directs
in
Roman
from
the study of nonalso
attention
away
power structure
27
Romantype houses(Hingley 1989:23-4; 1991:96; Clarke 1998).The variability betweenthe
in
is
far
from
determinations
Roman
Britain
houses
of
wealth
standardized
and
types of
(Hingley 1989; Branigan 1981; Hingley 1999:145). Settlementswith both rounded houses
identified
been
have
(Keevill
Booth
1997)
and
and wealth
structures
and villa-like
differentiation on non-Roman type settlementshas been established(see Hingley 1989:3 1,
80; Leech 1982; Branigan 1981).It has also been shown that the apparenttransformationof
into
highly
Romanized
(Branigan
farmsteads
1981).
villas
was
variable
native
Parker Pearsonet al. (1996) argue that our approachto the labelling of structuresis often
basedon our own `common-sense'observationsand notions of classification: "Housesmay
live
but
to
as embodimentsof myth, places of worship, calendars
servenot simply as places
61).
1996:
(Parker
Pearson
to
the
social
and
cosmic
guides
order"
et al.
and generally
Furthermore,attitudesexpressedthrough architecturemight be quite different from attitudes
for
(Allison
Houses
households
through
consumption
practices
example.
are not
expressed
1999:4). The acknowledgementof, and study of, the full cultural range of Romano-British
householdswould help to re-focus our attentions towards the inhabitants rather than the
in
isolation.
If,
in
Inka
discussed
had
been
focus
the
the
study
the
of
on
earlier,
structures
households
distorted
been
have
Inka
the
the
alone,
power
structure
of
and
elites would
elite
the changing alliance and wealth of the elites and commoners - in relation to accessto
have
been
foods
However,
between
just
division
the
missed.
would
as neither
preferred
Roman and non-Roman or rich and poor should not be viewed as absolute, nor should
specific notions of wealth and manner of conspicuous consumption (Appadurai 1986:40;
Hingley 1989:160). The `social life of things' is determined, not by the things themselves,
but by the people who use and ascribea value to them (Appadurai 1986).
3.2.4. Diet and culinary practice
Material things - become important through their very ordinariness. They

for
the vast undersideof cultural action, for values and aspectsof their
stand
personality and world-view which men and women could not or would not
(Johnson
1993:
in
xi).
words
express
28
Forget that commoditiesare good for eating,clothing and shelter; forget their
idea
instead
that commodities are good for thinking;
try
the
and
usefulness
treat them as a non-verbal medium for the human creative faculty (Douglas
and Isherwood 1978:62).
An indiscriminate comparison of households is perhaps not the most rigorous method by
diversity.
My
diet
its
to
emphasis
uncover
on
and
culinary manifestations was
which
brought about, in part, by the simple fact that all people eat and drink, and tend to do it in a
However,
I
than
that,
more
way.
wanted to explore a theme that was not only
particular
but
households
have
been
While
transcends
to
them.
also
all
archaeologists
common
food
how
its
the
was
obtained
with
and
extent
of
relationship to the economy,
captivated
less attention has been placed on the cultural dynamics surrounding its preparation and
(Hastorf
consumption
1991:153; Hawthorne
1998: 164). Indeed, it has been argued,
believe,
I
food
that
the
we
must
understand
processes
and
consumption
of
of
successfully
drink at the household level before we dare to extrapolate on such themes as species
husbandry
(Rackham
1983:
273;
King
Branigan
42-43;
1988:
animal
and
populations
1988:52; Huelsbeck
1991:66; Reynolds
1995a:188; see also comment by Barthes
1979: 169). Food and drink undergo a variety of transformations
from production to
in
these
transformation can be significant to studies of cultural
points
consumption and
Gregory
1989b;
Altman
(Barrett
1989:
188; Messer 1984:223; Hill 1995). An
and
practices
diet
habits
the
culinary
and
of the various types of `non-Roman' Romanoemphasis on
British households while serving to redress current bias towards `Roman' households will
between
diversity
`native'
help
the
to
reveal
settlements.
also
Anthropological and historical studies of diet have shown culinary habits to be very
informative. Goody, in Cooking, Cuisine and Class (1982), and Mennell, in All Manners of
Food (1985), both illustrate how the production of food can be influenced by the nature of
historical
the
and social developmentof society. Food distribution
the political economyand
is tied up with the social politics involved in food allocation, as well as economic factors
forces,
Food
tribute
taxes.
and
market
preparation is linked to genderrelations,
surrounding
labour service and cultural and social ideasabout the way food should be flavoured, cooked,
Finally,
food
the
that
consumption
eaten.
and
of
and drink is affected by national and
served
identity,
differentiation
hospitality
and
group
competition,
notions
group
of
and sharing,
forms
the
establishment
of
specific
of etiquette, and food taboos (Goody
together with
29
interested
in
how
I
1985).
Mennell
38-40;
1982:
am especially
groups differentiate
imperial
identity
their
an
context through what and
within
own
themselvesand establish
how they eat and drink. Both Mennell and Goody essentially wanted to determine why, in
(Goody)
Britain
(Mennell),
African
and
post-medieval
societies
a
state
pre-colonial
differentiated haute cuisine did not emerge in the same way as it did in Eurasian states
(Goody) and in post-medieval France (Mennell). Both show how the emergence of a
differentiated cuisine is more complex than an association with social hierarchy. Factors
formalized
the
to
the
access
certain resources,the switching of cooking
control
over
as
such
into
domain
domain
from
the
the
of
wives
of servants or specialists with
activities
increasingstatus,cultural mores on what was consideredacceptableculinary behaviour, and
literacy
and emulation can all influence the range of culinary
the extent of cultural
differentiation (Goody 1982; Mennell 1985).While not all of thesefactors might be relevant,
Goody's
Britain,
Mennell's
in
Roman
demonstrate
the
and
observations
recognizable
or
importanceof context when determining the significance of how and why cuisines differ.
How and what people eat and drink is a form of communication (Barthes 1979; Mintz
1996:13). The developmentof a cuisine involves a whole rangeof decisions leading from its
`raw' to `cooked' state. These decisions distinguish cuisine and cultures: "specific foods,
their uses, and associations communicate, reaffirm, and aid in the construction of the
(Hastorf
1991:
135).
a
system
of
signs
as
containing
acting
social
system,
messages"
cultural
Day-to-day decisions on where, what, when, how and with whom to eat and drink are
Douglas
Isherwood
2000;
1978:
(Hamilakis
155;
Douglas
1984:
3)
and
and provide
ritualized
(Sherratt
household
1995:
11).
looking
In
to
relationships
studies
at the effects of
sustenance
imperialism, when one society is politically subsumed by another, individual and group
interpretation of the rules and histories surrounding an invasive cuisine can vary
faces
both
house
inwards
Just
as a
and outwards to the community, the
substantially.
in
drink
both
food
domain
(Wilk
1990;
the
exists
and
of
public
and
private
consumption
Smith 1987:312-13). The nature and extent of imperial contact, the status, occupation,
involved,
the
traditions
together with their propensity to
of
people
preferences
and
rituals,
follow culinary, as well as imperial, `rules' all influence the composition of cuisine (Messer
1984:222-6; Garnsey 1999:6-7). Cuisines are fluid, they are not transplanted untouchedby
30
different
it
is
in
1982:
19)
(Revel
types
their
that
various manifestations
and
time or space
be
inferred.
layers
can
communication
of
and
3.2.5. Diet, culinary practicesand imperialism

What and how we consumeis therefore socially, culturally, economically and politically
habits
imperialism
few
The
the
that
through
the
of
approach
studies
consumption
motivated.
localized
the
otherwise
neglected
conditions of conquest and
conquered emphasize
imperialism (seefor exampleHastorf 1990; Costin and Earle 1989; Brumfield 1987; 1996).
As a number of these studies have shown, analysis of consumption at the household level
1990;
imperialism
daily
habitation
how
(Hastorf
illustrate
the
affect
might
rituals
of
can
Costin and Earle 1989). Consumption's politicized dimension puts constraints, such as
how
37;
1982:
(Sherratt
Goody
1995:
12;
and
on
what
we
access,
consume
and
availability
Dietler 1990:370; Garnsey1999:5-6).
3.2.6. `Roman' consumptionhabits
Ancient texts that describethe customsand habits of conqueredpeoples have played a role
in defining the ways that Romans and non-Romans, i. e. `barbarians', approached the
drink.
The
diet
food
`barbarian'
is
depicted
the
and
of
of
against the
generally
consumption
Roman
(Reynolds
diet
1995b:
303;
the
urban
elite
of
civilized
Garnsey 1999:62).
Barbarianswere drinkers of milk and ate too much meat. They ate food without spices and
fat
instead
drank
from
they
of
olive
oil;
animal
used
alcohol made
grains rather
saucesand
barbarians
consumedwine, they did so inappropriately. Barbarians
than grapes and when
from
did
lacked
they
bone
(Tierney
table
the
and
consumed
manners
meat
as
animals
also
1959-1960; Chapman 1992:166-170; Reynolds 1995b:314; Garnsey 1999:67-8,124-7). A
few of the classical authors make specific reference to the eating and drinking habits of
British people. Caesar in Gallic Wars refers to the Britons as having a taboo on the
hare,
for
that
they
them
their
chicken
and
goose
and
states
of
raised
consumption
He
British
being
the
classes
also
as
pleasure.
consumersof milk and of
and
amusement
31
in
British
Geography
Strabo
to
the
than
as producersof grain, cattle
corn.
refers
meat rather
hunting
dogs
Ireland
texts
that
classic
cites
a
of
of
number
comment on
and exporters
British hunting dogs (1986:224-5). Strabo also commentson the lack of cheese-makingby
the British in spite of an abundanceof milk. In referenceto the `hostile' northern British
tribes, Dio Cassius in The Roman History: The Reign of Augustus remarks that the
inhabitantslived off their flocks and wild plants and did not consumefish even though it
brief
depictions
These
did
British
did
the
available.
of what
and
not
people
was readily
insightful
held
by
to
the
the authorstowards the civilized
with
attitudes
are
regard
consume
food
drink
Garnsey
(1999:
62)
has
barbarous
consumption
of
and
and,
as
pointed out,
and
interesting
fabrications
in
Gowers
(see
contradictions
with
and
comments
are replete
1993:7; Funari et al. 1999:11-12. on the contextualization of ancient sources).However,
their usefulnessin the study of the diet and customsof particular settlementsis negligible,
have
influenced
Roman
texts
the
the
classical
way
and non-Roman goods and
although
been
in
have
portrayed
academia.
customs
Many of the Roman-like goods considered symbolic of the adoption of a Romanized

lifestyle (amphorae and their edible contents; food preparation wares such as mortaria;
drinking
(Terra
such
as
samian
ware
ware
sigillata))
and
are associatedwith eating
serving
(Dannell 1979:177; Trow 1990:103; Williams and Peacock 1983).In late Iron Age Britain,
the initial appearanceof thesetypes of goods is linked to the trading practices and political
(Haselgrove
1989;
Trow
1990;
1996).
In
1990;
Dietler
the
the
elite
of
see also
ambitions
has
been
determine
Roman
their
to
the extent to which
period,
presence
used
early
indigenouselites initially emulatedand manipulatedthe customsof their Roman conquerors
(Millett 1990a, 1990b). Roman-type ingredients and methods for procuring and preparing
food have also beenidentified and used as indicators of a Romanizedlifestyle (Jones1991).
The prevalence of cattle and pigs rather than sheep at more Romanized settlements,for
is
habits
Roman
to
the
thought
the
reflect
culinary
of
example,
army, if not `Romans" (King
1991; 1999; although see Halstead 1985:224; Grant 1989:142). Particular types of dining
been
for
Romanized
have
suggested
also
and non-Romanized peoples. The
customs
entertainmentof reclining guestswithin the villa and use of many specializedvesselsstands
in contrastto outside feasting and the eating and especially drinking from large communal
Dannell
(Blagg
1990:
206;
1979;
Millett
vessels at native settlements
32
1979; Bradley
1998:49; Okun 1989;Reece1988:44). Thesecharacteristicsappearto suggestthat there was

drinking
habits
`Romanized'
the
to
eating
and
of
and native peoples,which
uniformity
some
was clearly not the case (Woolf 1995:341; Freeman 1993; Hingley 1999:143). The
Vindolanda tablets- military documentsthat list food and drink items - recoveredfrom the
Vindolanda excavation in Northern Britain serve as an example of a blending of the sodiet.
`Roman'
The
items
include:
barley
(consumption
`barbarian'
and
of which was
called
for
disgraced
beer,
Roman
(Gamsey
Celtic
Legion
1999:
120)),
a
punishment
a
considered
fish
wine,
pork-fat,
vintage
sauceand a substantialamount of meat, (Bowman and
sour and
Thomas 1983:86-94).
What is needed now is an integrated approach to the study of consumption in Roman

Britain, one that focuses on the social contexts of eating and drinking. We need to
contemplate whether there was any ambiguity in the use of so-called Roman and native
in
(after
Webster
2001:
9).
for
A
the
changes
common explanation
material culture
consumptionhabits is `Romanization' or conversely`native continuity' if the settlementhas
not changed according to expectations.Current critiques of the concept of Romanization
discussedin Chapter 2 (e.g. Freeman 1993; Webster 1996; Hingley 1996; Barrett 1997a;
Mattingly 1997b), however, argue for a more introspective analysis of the presenceor
absenceof Roman-like material culture at all levels of the social hierarchy.
3.2.7. Roman style goodsand structuresat `non-Roman' settlements
In the early Roman period, some of the Roman-stylegoods `trickled down' to the rest of the
is
It
unsatisfactoryto simply place these settlementson a sliding scale of (non)
population.
Romanization, particularly if the significance of changes in, or persistenceof, particular
been
has
from
considered
not
customs
within the overall context of the settlement
culinary
itself. Miller has established that "The point of a contextual analysis is that it relates
in
disparate
to
turn, the context for the others"
sources
of
evidence
make
each,
apparently
(1985:201). The procurement,preparation and consumption of food and drink encompass
1The `classical' diet of Rome emphasized the consumption of pork, particularly suckling pig, over the
33
dinner
far
beyond
In
table.
the
the
specializations of archaeology and reaches
most of
Britain, in spite of the repeatedrequestsover the years by bone, pot and plant specialists
(Payne 1972:80-1; Maltby 1981:193; Lambrick 1984:176; Hansen 1991; Darling 1989:98;
Hodder 1989:271; Tyers 1996:23), archaeologicalremains are rarely integrated and rarely
focus
is
instead
The
from
their
on the
excavated and social context.
within
considered
from
dissociated
ingredients
the circumstancesof
and vessels
perceivedvalue of particular
Romano-British
9).
Underscoring
1997b:
(Mattingly
the
of
eating
their use
accounts
many of
is
habits
diet
`native'
drinking
the
that
the
notion
and culinary practiceswas uncivilized
and
diet
303).
(Reynolds
1995b:
the
to
classical
when compared
Hastorf has observed that some foods may have different meanings in different contexts
135;
(1991:
been
foods
have
the
constant
may
meaning associated with certain other
whereas
has
Isherwood
hand,
65).
Douglas
1978:
Blanton,
that
the
suggested
and
on
other
see also
the usage of particular goods might be misconstrued or consumed with intent to make
`fraudulent claims' (1994: 14). As was discussed above, scholars have bestowed wealth and
been
lived
houses,
have
Roman-like
the case
this
who
at
on
people
and
although
may
power
deny
the existence of alternative expressions of
automatic
groups,
assumptions
some
with
Johnson
for
inhabited
1989;
houses
(Hingley
people who
non-Roman-like
wealth, especially
1993: 10). For example, at Watkins Farm in the Upper Thames Valley, the early Roman site
`native'
decidedly
was
particularly
in character; however, the percentage of specialty pottery,
serving-type ware, was considerably higher than for other native type
in
in
food
(Booth
how
Raven
1990:
47).
This
the
that
area
and
press;
suggests
settlements
drink was served was of particular importance to the inhabitants and quite possibly that the
inhabitants
food
drink
the
to
through
the
of
status
was
expressed
and
serving of
wealth or
the local community. It is important in this regard to acknowledge that displays of wealth
be
(Smith
1987:
317;
Dietler
is
1996:
90).
it
In
through
perishable
many societies
might also
food
the public sharing of
and drink rather than private consumption, that wealth, prestige
is
demonstrated
(Johnson
1993:
10;
Dietler
1996:
92) and that obligations are
power
and
inequalities
(Dietler
1996:
92; Grahame 1998:6). The consideration
maintained
created and
drinking
the
the
of
of
eating
and
remains
could potentially challenge many
social contexts
of
169-171).
1999:
beef
(Wilson
of
consumption
34
(if not most) of our current perceptions of `Roman' and `native' consumption habits in
RomanBritain.
3.3.
An Approach to the Study of the Social Contexts of Eating and Drinking
3.3.1. Areas opento investigation
Theuseof Romanthings
In Chapter 2, it was established that the way we use the term `Roman' is often
indiscriminate. Whether the goods that were being imported, produced and consumedat
desire
in
forces,
Britain
to
technology
a
reflect changing
and market
particular settlements
beyond
`Roman',
the simple
to
these,
go
avant-garde,
wealthy
or
of
we
none
need
appear
The
`Roman'
their
typically
of
presence
or
absence.
a
signifier, was also a
pig,
observation
309)
hunting
in
1995b:
Iron
Age
(Reynolds
and
associated
with
society
prominent animal
Griffith has found that some `poor' rural sites had higher concentrationsof samian ware 153;
1995:
Roman
(1989:
Monaghan
`rich'
76;
than
signifier some
rural sites
and
another
for
202;
is
Evans
1987:
Willis
it
1997:
42).
has
been
As
possible,
suggested
although see
feasting
have
been
for
in
that
their
certain
goods
may
embraced
use particular
example,
be
has
1990,1996;
Maltby
(Dietler
1985a:
61)
to
yet
practices
-a consideration which
is
for
Roman
Britain.
Revel
(1982)
has
that
most
closely
what
more
stressed
examined
is
herbs
importance
Apicius
the
to
the
cookbook
accredited
and
about
of
spices
and
striking
the mixing of salty and sweet ingredients in Roman cooking (1982:47; see also Veyne
1987:188). The presenceof mortaria, sweet wine and salty garum at a settlement might
in
least
in
(Gonzalez
tastes
the
of
a
a
change
particular
group
at
sphere
public
reflect
Turmo 1997:125) - or conversely,the selectionof ingredients that were approximate to, or
Anthropological
have
tastes.
the
studies
of
cuisines
existing
conquered
peoples
enhanced,
of
hand,
during
that
the
specializedevents, traditional consumption practices
one
shown, on
1984:
225),
(Messer
households,
the
that
and
on
other,
maintained
who
wealthy
are often
it
is
foods
they
to
that
traditional
wish,
often
whatever
eat
eat
afford
and
conceivably can
less wealthy householdsthat consume so-called luxury foods (Gonzalez Turmo 1997:119;
35
in
1994).
The
Barley
the
terms of the social
and
consideration
of
artefacts
remains
seealso
behaviour that surroundseating and drinking will help to situate these practiceswithin an
imperial context,rather than establishan inventory of `Roman' and `native' things.
Integration of the artefactsand remainsassociatedwith eating and drinking
A study of the diet and culinary practices of households will offer alternative accounts of the
imperialism.
integrating
Through
the artefacts and remains of eating and
culture
of
material
drinking we can look at changes in daily life, as well as possible attitudes towards, and
imperial
found
King
in
`non-Romanized'
Britain
to,
that
the
an
presence.
south
of
responses
higher
higher
have
have
`Romanized'
proportion
of
a
a
sheep
remains
whereas
sites
sites
(1984;
1991;
1999).
As
and
of
ox
pig
mentioned above, the predominance of
proportion
is
in
`Romanized'
the evaluation of a
commonly
pig
remains
used
as
and/or
signifiers
cattle
differential
The
in
different
deposition
bones
contexts
preservation
and
of
settlement.
in
(Gamble
1978;
Maltby
ideas
1985a;
Grant
1989:
136),
on
changes
any
notwithstanding
diet can only benefit from the integration of all types of remains. How, for instance, were
for
in
the
the above mentioned study of
prepared
consumption
at
sites
animals and plants
Roman pottery which found that so-called `luxury' pottery, i. e. samian ware, amphorae,
(Griffiths
Romanized
to,
etc.
were
wares
not
restricted
sites
or
concentrated
at
colour-coated
1989:69,76). The correlation between types of ingredients and types of cooking and dining
in
Roman
Britain
be
Okun's
to
settlements
could
at
prove
study
quite
enlightening.
practices
in
dining
Upper
Rhine
found
during
Roman
diet
the
the
practices
and
area
period
early
of
that despite suggestions of a Romanized diet - an increase in the use of pork, the
ingredients
(e.
food
of
new
g.
olive
oil
and
ways
of
garum) and new
preparing
consumption
(e.g. mortaria) - generally speaking, food continued to be cooked using the same type of
hearth
the
type
on
same
of
or fire as during the late La Tene period
cooking pot and
(1989: 114-122). She also shows that while Roman-like serving ware was commonly used,
the Roman practice of using individual bowls for each dish was not adopted and instead
there was a tendency towards larger - possibly communal - serving vessels (1989: 123; see
Woodward and Blinkhorn 1997 re: significance of vessel size in the Iron Age). Bakels et al.
(1997) studied the changes in diet from the Iron Age to the Roman period at Oss-Ussen, in
36
diversity
between
identified
Netherlands,
thus
the consumption
the
considerable
and
For
example, they observed that at some
a
of
settlements.
of
number
native
practices
in
diet
little
Roman
the
the
appeared
changed
period with the exception of an
settlements
increase in condiments.They in turn suggested"Although the food remained `native' in
flavourings
have
it
different
fragrance;
the
the
new
must
given
a
appearance
and
essence,
food
looked
its
impact,
important"
(1997:
209).
the
and
smelled,
public
was
obviously
way
These examples suggest that the distinction between the use of `Roman' and `non-Roman'
is
`Non-Roman'
households
be
flavouring
absolute.
culture
not
could
and serving
material
in
local
ingredients
Roman-like
households
'Roman'
ways
using
and
serving
ware.
mutton
ingredients
in
Roman
be
Roman
dishes
but
them
eating
on
noncooking and serving
could
Roman ways. Alternatively, Roman and native consumption practices could be interwoven
integration
The
of the material remains of households within the social
and reinterpreted.
drinking
and
of
eating
will
context
help to disintegrate the Roman/native divide that
Roman
Britain,
of
studies
many
and most studies of Romanization.
pervades
3.4.
The Methodological Approach of this Thesis
As was discussed in Chapter 1, one of the aims of this thesis is the formulation of a
in
Roman Britain. The approachthat
to
the
approach
study
of
social
change
methodological
I am suggestingcan be summarizedas follows:
"
Establish the dietary and culinary habits of the inhabitants of a settlementthrough reintegrating the artefacts and remains that are relevant to the preparation and
food
drink.
of
and
consumption
"
Consider these artefacts and food remains from within their excavation context and
deposition
how
their
might reflect on the social contextsof eating and drinking.
suggest
"
Analyse the material culture and contextual associations of the artefacts and food
for
from
within the context of imperialism.
particular
settlement
a
remains
"
Contrastthe types of changesat a settlementwith other settlementsin the Upper Thames

Valley to considerdiversity in the responsesto imperialism.
37
The considerationof all theseaspectsof consumptionrequires an appreciation of the many
debatesand methodologicalchallengesthat surroundthe study of the consumption of food
in
The
focus
drink
therefore
this
archaeology.
of
chapter
remainder
will
on a number of
and
the major issuesthat havehelpedto structurethis approachto the study of social change.
3.4.1. A caveatto the study of the social contextsof eating and drinking
This section is essentially a caveat to the conditions of the archaeological record. I have
immerse
to
myself
attempted
in
the debates and methodologies
of the various
specializations that encompass the acts of eating and drinking as recoverable in the
The
`jack
record.
old
adage
of all trades, master of none' has haunted my
archaeological
mind throughout the preparation of this thesis. The individual specialists might subsequently
take issue with my particular stance on their subject; however, I do appreciate that what
is
leap
faith
data'
by
àcademics'
between
`the
the
specialists
concerns
of
and
made
often
`the big idea'. I have consequently attempted to incorporate the well-known vagaries of the
archaeological record and the specific problems that plague each specialist's field of study
into my interpretation. However, before we consider many of the issues inherent in the study
food
drink,
the other implements associated with food and drink
the
of
and
containers
of
ingredients
it
is
the
their
and
and
archaeological
consumption,
and social contexts,
necessary
to comment on the use of statistics in this thesis.
3.4.2. Consuming statistics
The excavated material described in this thesis is essentially a sample of what might
have
been
Orton
has
recovered.
queriedwhether archaeologistsshould use "... the
otherwise
just
because
have,
theory
sampling
of
we
perhapsrather loosely, describedour
mathematical
is
(1978:
400).
His
query
a
sample"
relevantto all types of archaeologicalremains;
pottery as
found
in
those
this study whose sample sizes are, mathematically speaking,
particularly
interpretation
The
the
main problem with
of archaeologicalsamples seemsto
quite small.
literal
identified
taken
as
are
patterns
representationsof a population. Orton
occur when
38
(1978) has observedthat even when it is acknowledgedthat the `population' or `sample' of
is
for
jump
is
"one
tempted
to
the
than
mathematics,
make
more
casual
which we speak
from one to the other" (1978:399).
A number of specialists now consider artefacts and remains in terms of depositional

contexts and/or consumption practices (King
1991: 15; Huelsbeck 1991:70; Rackham
1983:252; Needham and Sorensen 1988; Hawthorne 1998: 164). Certain studies of animal
bones, particularly those conducted by Bob Wilson in the Upper Thames Valley, have
demonstrated how species representation is affected by excavation strategy (Wilson 1985;
1996:70-73; see also Maltby 1981; Price 1985:53). The implication is that even if you had a
huge sample, if it is taken from one feature type or from one area of a site, whether you
hand,
is
by
dry
the
or
wet sieve or
machine
sieve,
sample still not representative of
excavate
the whole site (Price 1985:53; Fisher 1985: 179). Needham and Sorensen argue that the
behaviour that modifies our samples is "more important than either the quantity of refuse or
the quality of material in use on the site" (1988: 113). The emphasis of this thesis will
be
on contextual associations of the excavated material regardless of the sample
ultimately
diameters,
(for
histograms
trends
of rim
example
size2, although certain numerical
frequencies of pot forms, animal species and their body parts) have been calculated and
for
interpretation.
basis
a
offered as
3.4.3. The artefactsand remainsof food and drink preparationand consumption
Thecontainers
it is the besetting sin of archaeologiststo write the history of containers

...
rather than that of their contents(Sherratt 1995:17-18).
The majority of the containersconsideredin this study are ceramic (although see below).
The relationship between vessel form, function and use is integral to studies of
distinction
between
The
fine
(i.
kitchenware)
coarse
current
ware
and
ware
consumption.
e.
39
(i. e. tableware) provides only part of the picture and is far from straightforward. As Booth
(in press) has stated, `fine ware' often refers to the methods of manufacturing as opposed to
the vessel's status or function. Lambrick (1984), for instance, has studied the residues of
fine
in
found
Mount
Farm
Oxfordshire
that while there was a
wares
at
and
and
coarse
distinction between the use of the two types of wares, lime scale residues on some fine ware
in
in
(1984:
169).
He
a
cooking
role
as
well
as
revealed
serving
vessels
goes on to suggest
that "evidence for how far high-quality wares were used in cooking may have a bearing on
households"
(1984: 169).
the
of
relative
wealth
or
status
of
settlements
or
assessments
Cooking with `fine wares' might also indicate the use of these vessels for communal
feasting and/or special events. Allen (1990), on the other hand, found burnt residues on
Iron
Age site at Watkins Farm, also in Oxfordshire
the
vessels
at
middle
storage-type
(1990: 39). Large vessels, for example, could be used in beer production as well as for
309),
1994:
(Vencl
or they might indicate cooking for a larger group of people. These
storage
highlight
the possibility of multiple uses for both coarse and fine wares, as well as
examples
ideas
about the value and use of containers (see Monaghan 1995: 153;
our often-ethnocentric
Hodder 1981).
How, and for what, vessels were used, is a neglectedarea of ceramic studies in Britain,
Iron
during
Age and Roman periods. An over-emphasison fabric rather than
the
especially
form, where type of use is assumed- cooking pot, serving bowl, storage vessel etc. inhibits consideration of the people who were using them (Darling 1989:99). Fulford and
Huddlestone's 1991 exposition on the state of Roman pottery studies, characterizesthe
function
"Cinderell[a]
the
as
of
pottery
of ceramic studies" (1991:51). Ethnographic
social
found
be
to question the majority of our assumptions.In his study of African
examplescan
has
documented
(1994)
Barley
`kitchen
pottery,
ware' that is better made than `religious
in
ìmported
ware'
used
everydaycontexts,and traditional `native ware' used as high
ware',
(Barley
1994:
for
120,73;
tableware
similar observations see Gonzalez Turmo
status
1997:119). These observations suggest that the association between vessel form and
function is often more complex than has been suggestedin the past. This is not to suggest
that we attempt to assign a function to every category of pot; even if this was possible,
2 This researchis prescribedby the criteria set by the variousspecialistsso, for example,at Claydon Pike some
if
by
they containedlessthan 50gmsof pottery - seecase
the
specialist
pottery
contextswere not examined
study for further commentson this particular site.
40
have
had
have
their
multiple
uses
and
significance
may
may
varied according to
vessels
(Rice
1987:232-3; Hodder 1981; Skibo 1992:33). It is more that we need
context and period
. to recognize,as Barley has done,that "pots are semanticallypromiscuous" (1994:76).
Tied to our ideas about the use of containersis the possibility that wooden vessels,which
do
in
have
in
the
survive
not
archaeological
record,
may
played a significant role
generally
the culinary process and that pottery as a result may figure too prominently in our site
in
his
(1989)
Evans
late
study
of
pre-historic wetland sites has suggestedthat
reports.
is
imagined.
be
He
"...
were
not
as
artefacts
prolific
as
often
adds
wooden
we must wary of
ìnvisible'
determine
the
to
entirely
our understanding of the `surviving"'
allowing
(1989:180). Earwood (1993), however, is critical of Evans' assertionand makes a number
important
in
She
that
to
the
points
are
relevant
study
of
excavated remains general.
of
instead
disposal
the
that
regimes of wetland sites were different from the regimes
suggests
of other types of settlementsand askswhether we can transposethe apparentlow numbers
found
artefacts
at wetland sites onto dry land sites (1993:23). Earwood admits
of wooden
that in the Roman period, the use of wooden containerswas redefined with the proliferation
of pottery, glass and metal, and that wood was used primarily for buckets, boxes and casks
(1993:90), but arguesthat "it is easyto assumethat wood has always had a lesservalue than
(1993:
229).
The
or
pottery"
significance of wooden vesselsat the settlementsin this
metal
indeterminable,
is
and this remains the casewith other consumption-relatedartefacts
study
items
food
that are not generallypreserved.
and
The pre-eminenceof pottery in our site reports,particularly its position as a luxury ware, has
Evans
been
argues that "The durability of potsherds gives a distorted
questioned.
also
prominenceto pottery as a trading commodity" (Evans 1981:519) and suggeststhat Roman
fact
in
incidental:
impression
"The
is
that it was generally regardedas cheapand
was
pottery
in
its
use polite society neededthe excuse of indigence or deliberate austerity"
common:
(Evans 1981:520-21; seealso Monaghan 1995:153). While this may have beenthe casefor
people at some settlements,our considerationof luxury and non-luxury wares is often quite
be
to
consideredwithin the context of consumption. Samian ware,
rudimentary and needs
for example, is generally used as an indicator of the status of a settlement (although see
demonstrates
(1997)
but
Willis
how
1989),
Griffiths
above
samian ware was much more
41
fascinating
distinction
has
in the ratios of plain
His
than a statusmarker.
research revealeda
between
large
Gaulish
decorated
samian
ware
military
and
south
civilian settlements,
and
favour
indigenous
favour
decorated
in
settlements,
which
ware,
and
plain
ware,
which
in
England
Wales.
Willis
and
adds:
northern
regions
various
that it was primarily the decoration of these vessels
should
not
assume
we
...
that was most significant for these consumers; it may well have been the fact
that they were fairly large bowls which made them desirable. This raises
topical questions of form and function (1997: 41; see also Rush 1997 who
considers the social context of mortaria).
Pottery is ultimately a container and can have a participatory or dominating role in the
Sherratt
If
as
suggests, the prominence of particular types of pottery
process.
consumption
itself
(Sherratt
200)
1987:
83;
J.
Evans
1987:
to
their
to
the
contents
as
pottery
as
much
owes
then in considering pottery from within the social context of consumption we might also
realize the significance of particular pottery types.
Other implementsassociatedwith food and drink
A variety of small finds recovered from the sites in this study are directly or indirectly
food
the
acquisition,
with
preparation,
and
consumption of
and drink. These
associated
include, for instance, objects associated with the processing of grains, such as quern stones;
foods
linked
to
the
acquisition
of
such as weights for fishing nets and hunting
objects
hooks
have
been used to suspend pots or
that
as
as
well
metal
chains
and
and
may
pellets,
food above fires. These types of implements help to define some of the variety of, and
foods
the
types
of,
of
consumed and the way they were prepared for
significance
possible
hunting
The
presence of
pellets and wild animals at a settlement suggests that
consumption.
these species were occasionally consumed. Chains and hooks can tell us about feasting
Gomez
1983:
147;
de
Soto
(Manning
1993:
193), and quern stones can be linked to
practices
labour and events associated with the harvest, for example. Archaeologists are now
items.
Heslop
Gwilt
(1995), for
these
the
of
contexts
of
many
social
and
considering
intra-site
focused
distribution
have
the
and
regional
on
example,
of quern stones in
42
Yorkshire to determinethe socio-economicsignificance of these items in the Iron Age and
Roman periods.
The social contexts of small finds are increasingly being explored, from brooches and
toiletry items (Jundi and Hill 1998; Hill 1997) to nails (Dungworth 1998), and while all
types of small finds are not the focus of this thesis,I will comment on their associationwith
the artefactsand remainsof the consumptionof food and drink where applicable.
The ingredients
Food and drink are perhaps the most fundamental,if short-lived, media of
material culture (Sherratt 1995:11).
The remains of eating and drinking, namely plants and animals, are more typically used to
determine the site economy than as ingredients in the more social aspects of consumption
(Butler 1995: 19; Hansen 1991:53; Sherratt 1987:83; Hastorf 1998:773; Gosden 1999:2 see
biological
is
2).
Food
1984:
Douglas
consumption
generally viewed as utilitarian and
also
).
1-2;
1999:
Gosden
2000;
Gregory
38;
Altman
1989:
(Hamilakis
and
rather than cultural
Although environmental archaeology considers behavioural practices such as butchery and
is
food
the
sporadic
consideration of
preparation, serving and consumption
crop production,
(Rackham 1995:23). The remnants of plants and animals are not generally linked with
in
is
descriptive
the
typologically
which
reflected
nature of much of the
material culture,
1995:
22-23;
despite
data
(Rackham
Butler
1995:
Hansen
53).
This,
19;
1991:
the
presented
fact that changes in diet are often tied to changes in the way food and drink was prepared
2000;
Sherratt
1995:
13).
(Hamilakis
and served
In addition to the many taphonomic factors that may affect species population and
feasting
it
is
cultural
practices
such
as
and special events,the production
abundanceratios,
direct
butchery
have
taboos
that
animal
and
practices,
etiquette
a
effect on the
of medicines,
deposition
Singer
different
(Gilbert
1982;
Grant
their
and
species
and
of
configuration
1991:111; Hayden 1990; Tambiah 1969:424; Hansen 1991; Butler 1995; Ingold 1984).
SchusterKeswani, for example,has suggestedthat herdsdominated by young animals could
43
frequency
(1994:
261)
heightened
intensification
"an
of
as
ritual consumption"
or
signify
dairying
is
(see
McCormick
1992).
to
a
common
explanation
also
which
more
opposed
Armour-Chelu in his interpretation of the partial skeletons of sheep and dog at late
deposition,
in
Castle
Dorset
Maiden
the
their
that
the
skeletons'
argued
nature of
prehistoric
lack of processingas comparedto the rest of the assemblageand the absenceof gnawing by
dogs pointed to the possibility of "`special' meals or some type of celebratory event"
(1991:151). Both Butler (1995) and Hansen (1991) emphasizethe roles that plants play in
in
decoration,
(see
to
totemism,
also
construction,
not mention
myth and ritual
medicine, as
Hastorf 1998:777 on gender roles and plant use). Hansen suggests,for example, that an
indicate
`poor
that those
that
species
are
uncommon or
producers' could
abundanceof plant
if
(1991:
important
56).
These
had
special significance
are
considerations,especially
plants
indicate
is
have
to
the
thought
abundance
of a particular species,such as cattle,
seen
as we
diet.
`Romanized'
wealth or a
The integration of artefacts and ingredientsin the Romanperiod
how
in
the
the
to
of
variety
which
we
study
much
pottery
must
relate
...
fields
food
into
kinds
the
the
were converted
and
appropriate
of
products of
drink, and how these were prepared and served (Sherratt 1987: 83).
In the late Iron Age and early Roman periods, general use vessels, namely jars, were
by
flagons,
vessels
with
more
specific
superseded
cups, platters and
uses,
such
as
gradually
dishes (Millett 1979:39; Reece 1988:44 although see Evans 1993:98,103 where in the north
jar use increased). The social link between an increase in specialized forms and the
drink
however,
food
to be established. Modes of consumption
and
needs,
consumption of
is
be
Deitler
for
(1990)
to
to
what
consumed.
related
example, has tied the storability
need
hence
the accumulation of alcohol, to the sponsorship of drinking events
of wine, and
(1990: 369; 1996). The Roman period butchery tradition of chopping the carcass into joints
by cutting through bones, as opposed to the Iron Age tradition of separating bones through
the cutting of ligaments (Grant 1989:141; Maltby 1985b:20; see Maltby 1989 who identified
knives
between
distinction
in
Roman
the
the
of
use
and
choppers
period),
a rural/urban
how
for
different
they
towards
and
animals
were prepared
attitude
consumption.
suggests a
44
In the caseof so-called `specialdeposits' involving animals, many of the Iron Age deposits
butchered,
whereasin the Roman period, animals were invariably butchered,and
were not
for
(Grant
1989:
146;
the associationbetweenfeasting and sacrifice in the
presumablyeaten
RomanEmpire seeVeyne 1987:194-8).
3.4.4. The social context of eating and drinking
The main priority must be to gain information about the inhabitants of the
site, not to study the pottery, finds or structures in isolation (Darling
1989:98).
A study of the social context of the eating and drinking practices of householdsrequires
from
features.
I am now beginning to
the
remains
of
and
artefacts
associated
analysis
looking
it
is
the
attractiveness
of
at
archaeological
understand
remains out of context, as
Several
archaeologistsand specialistshave commentedover the
much more straightforward.
dialogue
lack
between
the
of
on
specialistsstudying individual elementsof sites and
years
how
this skews interpretations of sites as a whole (Hodder 1989:271;
excavators, and
Maltby 1981:193; Mytum 1989:65; Gamble and Bailey 1994:81). If this is to be the task of
someoneoutside the realm of a particular excavation, varying emphaseson context in the
interpretationand especiallythe presentationof a site can be problematic. The site report for
Barton Court Farm, one of the sites in this study, is a good example: the report comes with
in
micro-fiche
which the pottery reports record finds in relation to individual
an extensive
interpretation
but
is
for
the most part not offered with referenceto context.
whose
contexts,
The bone report, on the other hand, records speciesand butchery practices in relation to
features,
but
does
not emphasizebutchery marks on individual animals or
groups of similar
individual contexts (Miles 1986b). This is by no meansa criticism of the report, which is
comprehensiveand together with the archival record provides a full account of individual
diameters
(rim
butchery
and
contexts
marks are recorded in the archive), but rather a
qualification that my emphasison the social context of consumption, which requires the
integration of various types of `data', doesnot necessarilyfit with the original intentions of
the excavation,the specialist analyses,and the resulting publication. The format of this and
Barrett
1987;
1997;
Gero
(Allison
1996:257) further illustrates why it
most other site reports
45
for
Fulford
deemed
the
to
to
this
archival
record
study.
and
essential
refer
was
Huddlestone's(1991) review of 85 excavationpottery reportspublished since the 1970s,for
identify
found
64%
did
the
the totals of the amount of pottery
that
of
not
reports
example,
in
74%
did
that
the
their
and
of
study
and
studied
pottery
reports
not use all of the
recovered
types of pottery in their quantification of types of fabrics (1991:9-10). The archival record is
is
by
the cultural politics of
a
of
what
only
record
was
recorded
and
conditioned
of course
data
(Gero
in
instance
in
1996:
254),
including
this
the
collection
practice
archaeological
Romano-British studiesof analysingall of the Roman-typepots but not all of the indigenous
Some
have
in
I
the
that
the course of this
of
vessels.
problems
encountered
coarse ware
be
discussed
basis
in
later
on
a
site-by-site
will
chapters.
research
Reinstatingthe remains and artefactsback into their excavatedcontext, while laborious, is
in many ways the easy part. The excavatedcontext itself also has to be culturally defined.
The majority of finds for the late Iron Age and early Roman period in the south, are
ditches,
These
from
house
1984).
(Hingley
Miles
enclosure
gullies
and
pits
and
recovered
features,
both
open
subjectto both the accumulationsof a variety of practicesover
comprise
time, and single event deposits (Hingley and Miles 1984:62; Fisher 1985:179). Though
features
the
types
these
that provide a medium for the study of the social
are
of
problematic,
led
(Moore
deposition
that
to
the
attitudes
cultural
of archaeologicalremains
practicesand
1981).As Moore states:"the placing of certain artefactsin specific positions in a housemay
function,
in
but
it
little
their
their
about
any
meaning
says
very
something
about
well say
(1981:
instance,
be
91).
For
if
to
context"
a collection of what appears
relevant social
drinking equipment,is found in the areaof a housethought to be the kitchen, what doesthis
drinking?
If
found
in
items
the
those
significance
social
of
a pit
say about
same
were
be
then
this
animal
articulated
remains,
or
a
cache
could
containing
of carbonized seeds,
interpretedas characterizingsomekind of specialevent.
Analysis of the deposition of finds and their social significance involves the examination of
the diversity and frequency of species and types of artefacts across space. Various studies have
demonstrate
bones
to
the organization of a
attempted
and
plants
primarily of
looking
location
in
the
to
through
the
stages
of
at
various
processing
of
settlement
relation
1974).
Dennell
Wilson,
1978;
has
found
for
Halstead
(see
et al.
on a
example,
refuse
46
bones
large
bones
do
be
that
the
tend
to
of
sites
of
species
and
other
coarse
not
number
domestic
bones
the
the
the
around
more
areas
of
site,
which
attract
of smaller
situated
burnt
bones.
forward
He
butchery
disposal
and
puts
species
practices, rubbish
regimes,
hygiene
interthe
of
acceptable
or
scavenging
as
explanations
possible
of
notions
repeated
(Wilson
1989,1996b).
site patterns
Most people recognize that the spatial isolation of
different types of activities from their places of discard is not straightforward. As Wilson
(1989,1996b),
Maltby (1985a), Hill (1995), Hodder (1989), Lambrick (1984), Moore
(1981), Fisher (1985: 179) and others have stressed, associations of specific remains may not
bear any relation to areas of processing or use. Rather, they may reflect the differential states
deposition
in
(for
types
and
of
preservation
of
remains
specific
contexts
example, pits
of
ditches)
from
the
or
periodic
or
symbolic
movement
of
versus
remains
one area of the site to
another.
3.4.5. The social context of deposition
Increasingly, archaeologistsare considering the relatively small amounts of artefacts and

from
recovered
archaeological sites in terms of their social context.
remains actually
Sharples (1990), for example, has suggestedthat the distribution of imported Amorican
Head
in
Hengistbury
Dorset "was controlled in a manner that reflected, on a
at
ceramics
imports
the
the
of
pattern
region:
scale,
small
were restricted to areasapart from domestic
Hill
300).
(1994,1995)
(1990:
has questionedthe notion that Iron Age pits were
activities"
for
by
ordinary
rubbish
receptacles
revealing the sequentialassociationsof certain types of
depositional
artefacts
and
with
animal remains
sequenceswhich may have taken place over
decades.
Grant
(1991,1984)
has linked distinctive animal deposits
even
of
years,
a number
and the associationof particular speciesto acts of sacrifice and/or feasting and ritualized
behaviour. Similarly, Hingley (1990b) has suggestedthat the placing of certain deposits
ditches
the
entrances
of,
at
enclosure
and
within,
servedto mark social relations within and
An
the
acknowledgementof theseentwined sequencesof social practice
settlement.
outside
is fundamentalto the significancewe place on the cataloguesof remains and artefacts.
47
3.4.6. The comparisonof eating and drinking practicesof households
Rather than regarding the data as somehowdefective and inadequatefor the

questions we might prefer to ask, we should instead ask whether these
inappropriate
(Fletcher
1992:
40).
questionsare
The uncertainty over appropriate archaeological samples has led some to place conditions
Grayson
(in
the
of
excavated
material.
argues
comparison
reference to animal
on
it
is
in
that
the
only when similar patterns occur at sites
same region
populations)
"... reasonable to conclude that changing taxonomic abundances are, in fact, being accurately
Uerpmann
111-12).
(1973)
(1984:
suggests that only similar sized samples should
measured"
be compared. Maltby (1981), on the other hand, has argued that an accurate representation
distributed,
bone
"use
butchered,
that
those
that
we
only
elements
were
of species requires
disposed of and preserved in similar ways" (1981: 170). However, while these are reasonable
is
must
not
place
an
we
unrealistic
pre-requisite
an archaeological
on
what
conditions,
for
been
has
butchery
It
that
shown,
example,
particular
practices, such as the use of
reality.
knives versus choppers, and particular cooking techniques, such as roasting and one-pot
less
butchering,
318-19;
1987:
bones
(Lyman
that
are
not
always
require
reflected on
meals
Crader 1990:708). Crabtree (1989) in her study of Anglo-Saxon animal bones has found that
in
leave
butchery
Maltby
butcher
bone
97)
(1989:
the
a
may not
any
marks on
and
a skilled
later study (1989) has revealed a contrast in butchering techniques between Roman urban
and rural sites.
This study illustrates the impossibility of standardizing archaeological settlements, as

inherently
incompatible.
I
are
samples
should add that not all specialistsare
archaeological
found
for
instance,
has
about
Evans,
the
comparison
of
excavated
circumspect
material.
as
that the various methods for quantifying pottery (sherd counts and weights) are analogous,
with the possible exception of minimum numbers of vessels (1991:72). He concludes,
"There do not seem to be good reasons for the concern expressed about comparing
by
different
data
(Evans
1991:
72).
methods"
presented
quantified
Any comparison of the diet and culinary practices of households is hindered by
between
the
and
excavation
sampling
strategies
varied
sites, and the
of
considerations
48
differing methodologiesof the specialists,not to mention the contemporaneityof the sites in
in
by
Such
deposition
the
compounded
variability
practice,
variability of
and
question3.
has
led
direct
to
that
many conclude
comparisonsof sites are not presently
preservation,
1989:
136;
Maltby
(Grant
1981:170; Gamble 1978:346; Jones 1985:114). However,
possible
this assumesthat some time in the future virginal assemblagesmight be available for
it
is
human
behaviour
though
that
clear
even
skewssamples (Price 1985:53). We
analysis,
disentangle
be
to
the layers of bias that exist at the various levels of
able
will never
interpretation.
With extremely rare exceptions, we
analysis
recovery,
and
archaeological
be
able to reconstructthe moments of deposition. In referenceto site taphonomy,
will not
Rackham(1995) adds:
Obviously it is essential to be aware and understandthe post-depositional

destructive agenciesbut it is essentially a fanciful notion to imagine that
taphonomic studies would permit the `scientific' rebuilding of the last
function of the assemblage(Rackham1995:24).
He suggestsinstead that the most we can attain is the "character of the agenciesat work"
(1995:24); I would add that it will most likely be our character that is revealed in the
interpretation.
As one of the objectives of this study is the
of
excavation
and
processes
different
types of responsesto a Roman presence,the significanceof one type
of
recognition
of response(or pattern) can only be realized when it is contrasted with another (Hingley
1984:86). It is ultimately through comparisonthat conventional, formulaic responsescan be
(Evans
1982:
175;
Webster
1996:8) and while I do not ignore the shortcomings
questioned
of the various archaeological samples,I nonethelessemphasizethe points at which they
diverge.
3I did not examinethe artefactsand remainsby layer(e.g. Hill 1995)as not of the material was recordedby
all
layer. For exampleat Barton Court Farm for eachcontext,animal boneswere recordedby layer, but pottery for
the most part was not.
49
3.5.
Conclusion
Some of the most commonly cited types of evidence for the `Romanization' of Britain are
the artefacts and remains associated with eating and drinking. A number of studies have
focused on separate aspects of diet and culinary practices, such as the introduction of wine
in
foods,
types
table
the
oil,
new
of
wares
consumption
and
changes
of
certain
olive
and
integrated
dietary
Romanization
Britain.
However,
the
the
of
context
an
study
of
of
within
habits,
from
particularly
of non-Roman-like settlements,
within the context of
and culinary
imperialism has not been conducted `Romanization' and `native continuity' are unsuitable
for
into
do
fit
Roman
the
to
the stated
the
myriad
responses
presence
and
not
explanations
aims and objectives of this thesis.
This study aims to look at the localized effects of imperialism through consideringthe social
drinking.
integrated
The
ideal
for
the
these
and
of
eating
setting
consideration
of
context
daily acts of consumption is the place where people live. The potential for this course of
be
in
best
imperialism
òrdinary'
the
effects of
on
people might
research considering
developed by concentrating on an area of Britain which is considered to be `nonRomanized' - the Upper Thames Valley, Oxfordshire; and during a period in which new
ideasabout eating and drinking appearto have beentaking form - the late Iron Age to early
Roman transition. As was demonstratedin chapter 2, the late Iron Age has primarily been
in
approached relation to the Roman period. In this study of the effects of imperialism, I will
Roman
largely
in
the
early
period
relation to the late Iron Age.
approach
50
Chapter 4
4.1.
Introduction
The majority of the settlements in the Upper Thames Valley during the early Roman period
`native'
`Roman',
because
than
rather
of the way the settlements are
considered
are
inhabitants.
Structural
the
the
the
nature
of
and
consumption
remains
practices
of
structured
fact,
floor
have
in
houses
the
not
villa-like,
a range of
are
no
obvious
survived,
and,
plans
of
habitation.
to
that
these
are
required
confirm
sites were places of
archaeological methods
Because of their classification as `native% settlements of the type described in this study are
`native
to
exemplify
continuity'
used
often
and/or low impact `Romanization'
(Raven
1990:49; Hey 1996:169; Miles 1986b).
The inconvenienceof the poor preservationof housesin the Valley during this period has
turned out to be fundamentalto my study of the households.This chapterwill thereforeboth
introduce the study region of this thesis, and argue the need to rethink a concept of
households in which the house is not the primary locus of study. It will be suggested,
little
that
settlements
paradoxically,
with
evidence of structures might provide a
perhaps
household
integrated
the
view
of
and challengemany preconceptionsof what Roman
more
imperialism looks like in the archaeologicalrecord. Before we proceed, it is necessaryto
first introduce the source of many of the excavations conducted in the Upper Thames
Valley, the Oxford ArchaeologicalUnit.
51
4.2.
The Oxford Archaeological Unit: Similarities and Contrasts with the Present Study
Many of the sites discussed in this study, including three of the four case studies were
is
U.
").
Roughground
Farm
Unit
("O.
A.
Oxford
by
Archaeological
the one
the
excavated
in
by
1960s
it
initially
Jones
1950s
in
Margaret
the
that
using
and
was
excavated
exception
funds supplied by the then Ministry of Works. The excavation was subsequently extended
by the O.A. U. in the 1980s and 1990s, and all of the post-excavation work was
later
in
1993
1993).
by
O.
U.
(Allen
A.
the
et al.
published
consolidated, re-examined and
Claydon Pike, Old Shifford Farm and Barton Court Farm were each excavated by the
O.A. U. exclusively. The post-excavation analysis for Old Shifford Farm and Barton Court
Farm is complete, and much has been published (Hey 1996; Miles 1986b). The postbeing
for
is
is
in
Claydon
Pike
largely
the
complete and
process of
excavation work
for
publication.
consolidated
The founding directives and excavation policies of the O.A. U. complement many of the
"...
O.
The
U.
in
A.
1973
the
this
to
study.
clarify
of
was
established
an
aim
with
objectives
hierarchy and function of settlements in the Iron Age and Roman period and the cultural and
influencing
factors
Excavation
(Jones
Miles
318).
1979:
was
settlement"
and
ecological
deliberately large-scale and comprehensive (Hingley and Miles 1984:52), as is necessary for
inner
the
analysis
of
workings of a settlement. Interestingly, considering the
a contextual
time,
similar
sentiments were expressed by Margaret Jones when she excavated
Roughground Farm in the 1950s and 1960s. She commented specifically on her focus on the
landscape of the settlement with a view to establishing the site plan and the rescue of "as
finds
features
(Allen
1993:
"This
their
She
4).
and
as
possible"
was not
et
al.
added,
many
quite in accord with current practice which advocated the selection of `typical' features for
intensive
excavation rather than extensive studies" (in Allen et al. 1993:4). The
more
O.A. U. places particular emphasis on the environment and ecology of a settlement using
for
the recovery and analysis of plant and animal remains; again this is of
varied strategies
importance
integration
to
the
of remains of eating, drinking with habitation. As
paramount
the O.A. U. intended at the outset to compare and contrast settlement within the Upper
Thames Valley,
they established "probabilistic
sampling strategies and standardised
Miles
data
(Hingley
1984:
53-4;
Booth in press). The approach of
and
collection"
of
methods
52
between
differences
facilitates
in
U.
O.
A.
this
my objective of exploring
the
regard also
in
Valley.
the
settlements
particular
The consistency in approach and method resulting from the fact that the O.A. U. excavated
important
it
is
benefited
has
However,
to acknowledge
this
the
sites
greatly
study.
most of
have
worked on these sites over the past three decades and that
that many people
inconsistencies exist between the various methods and approaches of the directors of
individual
and
excavations
specialists. My own objectives similarly
diverge from the
O.A. U. 's in placing particular emphasis on the social contexts of eating and drinking and
how this relates to imperialism. The four case studies themselves differ significantly. For
from
four
in
is
the
the
this
material
excavated
sites
study
not quantitatively
example,
it
is
be
illustrated
by
This
the recovery
consistently
represented
at
each
site.
can
similar, nor
forms
Old
Shifford
Farm,
identification
of
specific
pottery
at
which was poor, as
and
defined.
Roughground
At
the
and
plant
animal
samples,
with
which
were
well
contrasted
Farm the recovery and identification of pottery and querns appears quite high, whereas
butchery marks were not recorded and no plant remains were recovered at the early Roman
Court
Barton
Farm,
bones
At
sizeable
amounts
were recovered
of
pottery
and
settlement.
however,
identified
identified;
(though
for
the late Iron Age phase of the
plant
remains
and
Roman
the
not
recovered
at
early
settlement. Claydon Pike, on the other
settlement) were
hand, has considerable quantities of identified pottery, bones and plant remains, but is
insecure.
chronologically
This is the reality of archaeologyand it surely cannot meanthat no comparativework should

be undertaken.Therefore,in building on the O.A. U. 's work I have attemptedto incorporate
the processesof excavationand differencesin interpretation into this study.
4.3.
An Introduction to the Upper Thames Valley
In this section, I will discuss aspectsof the political setting of the Upper Thames Valley
during the transition from the late Iron Age to the early Roman period. I will then comment
in
Upper
Thames
Valley,
is
it
the
the
settlement
of
nature
currently understood.
on
as
Following on from this, I shall considera number the interpretationsof settlementin Valley
53
feature
in
in
how
the theories of
they
settlement
studies
of
non-Roman
and
and
Romanization and native continuity - two issuesthat have greatly influenced the course of
this research.
4.3.1. The political setting in the late Iron Age and early Roman periods
The Upper Thames Valley cuts across southern-central England from the source of the River
Thames in the west, one hundred kilometres eastwards (see Figure 4.1). The River Thames
has been described as a "channel of communication, a barrier and defence" and as both "a
into
England
into
(Allen
the
the
south-eastern
and
an
centre
et al.
artery
of
country"
gateway
1997: 114). The region was within the first wave of conquest after the Roman invasion of 43
AD (see Jones and Mattingly 1990:66-7), although evidence of Roman and/or continental
influence has been identified at some late Iron Age sites, such as Claydon Pike and
Abingdon Vineyard (Miles I986a: 50). Military sites were established at points around the
Gloucester
Dorchester,
(Glevum) and at Cirencester (Corinium)
valley, at
Mattingly
1990:88-9; McWhirr
(Jones and
1984; Burnham and Wacher 1990). A series of well
linking
track
ways
and
roads
settlements with local centres, such as Abingdon
established
Vineyard, and the River Thames provided communication routes throughout the Valley and
beyond (Miles 1986b:43).
The distribution patterns of pottery and coins appear to suggest that the Upper Thames
Valley may have provided a natural boundary for a number of late Iron Age tribal groups
(Selwood 1984): the Catuvellauni to the east,the Dobunni to the west and the Atrebates to
the south (Miles 1986a:56). There are also indications that the area was assimilated by one
Catuvellauni,
by
beginning
the
tribal
the
the
groups
of
of the first century AD (Branigan
1985; Robinson 1981:255). However, in referenceto the emphasisof this study, I would
draw attention to the argument made by Sian Jones,who has been critical of approaches
boundaries
based
ethnic
establish
purely on the distribution of material culture:
which
factors, may disrupt regular spatio-temporal
amongst
other
ethnicity,
...
in
an untidy and overlapping web of stylistic
resulting
stylistic patterning,
boundaries (in different classes of material culture and in different contexts)
in
be
discontinuous
space and time (Jones 1997: 129).
which may
54
Coins associatedwith all three tribal groups, for example, have been recovered in some
definition
boundaries
in
has
indicated
2000:
27-30).
Allen
(Allen
the
tribal
that
the
of
areas
Upper Thames valley is far from straightforward and has pointed to the possibility that
large
have
independent
tribal
three
to
the
tribes
groups
with
varying
allegiances
may
smaller
(2000:
is
indisputable
29-30).
What
is
the
that the
of
parts
valley
various
perhaps
controlled
Upper Thames Valley was not a self-contained, homogeneous landscape (Lambrick
1992:80), either physically or sociopolitically.
4.3.2. The nature of settlement in the late Iron Age and early Roman periods
Excavation and crop marks (revealed through extensive air photography of the region) have
density
Valley
high
during
throughout
that
the
this period of political
settlement
was
shown
transition (Benson and Miles 1974; Robinson 1981; Miles
1982; Allen et al. 1984;
Lambrick 1992:80), and that during the Iron Age and Roman periods, different types of sites
(Miles
1986a:
itself
Valley
52;
Lambrick
The
1992).
the
regional
zones
various
occupied
is
flanked
floodplain
by
that
a series of gravel terraces; beyond the terraces is
encompassesa
limestone
hills
downs
hinterland
slopes
and
clay
of
and
chalk
a
on the southern side of the
In
Iron
1981).
Age,
indications
in
(Robinson
the
different
there
are
site
economies
valley
of
the various regional zones: settlement on the floodplain and lower terraces was largely
higher
farming
the
terraces
on
was arable and more mixed (Robinson
pastoral, whereas
1992:56). As the water table rose during the Iron Age, settlement drifted to the terraces in
the late Iron Age and into the Roman periods, although the distinction between the site
Evidence
for
domestic occupation of the floodplain is
continued.
economies of settlements
late
late
Iron
into
Age
from
Roman
the
the
until
sporadic
period' (Lambrick and Robinson
1988:60; Lambrick 1992:82, figure 27; Allen et al. 1997: 119), though it is quite possible
that the floodplain continued to be exploited for its grasslands by pastoral settlements on the
first gravel terraces (Robinson 1992:57; Allen et al. 1997). A number of commentators have
between
there
that
cooperation
was
some settlements in the various zones,
suggested
It is possible our knowledge of the occupation of the floodplain is incomplete as sites are quite difficult to
detect (Robinson 1981:254; Allen et al. 1997: 118). Recent emphasis on the floodplain could, therefore, change
1997:
(Allen
117-8).
interpretations
et
al.
current
55
into
in
Roman
Iron
(Lambrick
Age
the
the
and
and
continuing
period
middle
starting
Robinson 1988:60; Robinson 1992:56; Allen et al. 1993:149-151;Allen 2000:11).
4.3.3. Regional discontinuity in the Upper ThamesValley

Two periods of regional `discontinuity' in the Upper Thames Valley have been identified
(Lambrick 1992). The first period occurred during the late Iron Age, when new settlements
were established and existing settlements were abandoned and re-established within the
fortified
A
of
number
same site.
settlements were erected during this time (Abingdon
Vineyard, the Big Enclosure and Dyke Hills);
they were situated alongside rivers,
floodplain.
Lambrick
Robinson
have
(1988)
the
and
suggested that construction
overlooking
indicate
"periods of stress.. not registered in the undefended
these
could
settlements
of
.
farmsteads" (1988: 73). The defensive settlements may have served to control the increasing
24-5)
floodplain
2000:
(Allen
itself.
is
It
the
trade
and/or
also in the late Iron Age that
river
the grasslands of the floodplain appear to have been managed for the first time (Lambrick
has
58).
Lambrick
1988:
Robinson
pointed to the possibility of political upheaval in the
and
late Iron Age, as reflected in changes in settlement at this time, when communal land rights
by
being
challenged
an emerging elite (1992: 105).
were possibly
The second period of regional discontinuity occurred in the late first century/early second
century AD when settlements either shifted, or were reorganized, or abandoned (Lambrick
1992:83-4). A variety of explanations has been offered to explain the changes in settlement.
These include the movement of people, particularly
poor people, from
individual
large
(Fulford
1992:32); the dislocation of people in
to
the
agricultural
estates
settlements
disintegration
to
the
of tribal groups (Lambrick 1992: 105), and a change in the
response
formation
imperial
the
with
of
estates (Miles 1984:208-9). Allen et al.
nature of settlement
(1993) have similarly suggested that by the end of the first century AD the floodplain
have
been
imperial
lands
under
might
control (1993: 196).
grazing
The casestudies in this thesis each feature settlementsthat were subject to both episodesof
fact,
In
discontinuity.
a possible third episode of change in settlement has been
regional
identified in this study, one that occurred in and around the Roman conquestof the region.
56
The possiblesignificanceof thesechangesin settlementover time will be discussedin more
detail in the casestudiesand in Chapter9.
4.3.4. Natives and Romans?in the Upper ThamesValley
Native type settlements dominated the Upper Thames Valley landscape until quite late in
the Roman period and much has been made of the absence of large villa estates. Robinson
(1992) has suggested that the combination of high population density together with an
division
land
inhibited
later
development of large villas:
tenure
the
and
system
entrenched
"there was probably little opportunity to carve out new fully Romanized estates such as
developed
in
have
the less heavily populated Cotswolds" (1992: 60). In contrast,
to
seem
Hingley has theorized that the Upper Thames Valley was occupied by more co-operative
social groups whose ties to the community at large may have inhibited the development of
Roman-type settlements after the conquest (1984; 1988:95). His conclusions have been
based
because
they
were
primarily on evidence provided by crop marks, rather than
queried
by excavation, which lack chronological distinction (Haselgrove 1984a; Lambrick 1992:7980). Recent excavations have not invalidated the premise behind Hingley's hypothesis but
the excavations do indicate that there was more of a distinction between settlements in the
Valley than his original model implies (Allen 2000: 13-4; Henig and Booth 2000: 105).
Hingley's theory nonetheless extends the idea described earlier that settlements from
different regional zones cooperated with each other. At the very least, the possibility of a
feature
in
interpretations
society
must
cooperative
of the potential
significance
of
`Romanized' material culture and structures.
Fulford (1992), on the other hand, has linked the chronology of settlement change identified
`Romanization'
in the Valley. He has observed that
to
the
subsequent
nature
of
above,
in
initially
the Iron Age, for example Ashville Trading Centre,
established
early
settlements
did not generally continue throughout the Roman period (see also Allen et al. 1993: 196),
in
late
Iron
Age, for example Barton Court Farm, tended
the
established
settlements
whereas
to be occupied throughout the Roman period. He goes on to suggest that the settlements
late
Iron
in
Age
the
are also the sites with more Romanized structural evidence,
established
have
Romanized
earlier
often
material but no structural evidence
while sites established
57
(Fulford 1992:27-9). Fulford acknowledgesthat his model of social change is a general
is
by
the nature of settlement shift, abandonment and
that
complicated
observation
in
Court
Farm,
As
Barton
that
the
shall
sites
we
see
characterizes
many
region.
resettlement
RoughgroundFarm and Old Shifford Farm - three of the four casestudiesin this thesis- do
fit
his
model.
not quite
It has further been suggested that interpretations of the Valley have been influenced by how
by
built
`Roman'
`native',
`villas'
and
and
our
emphasis
on
stone
when the most
we classify
in
Valley
the
timber
materials
construction
cob
and
plausible
- are not generally preserved
(Fulford 1992:37; Miles 1988:65; Allen et al. 1984). The implication is that we are looking
in the wrong place - structures rather than material culture - for evidence of `Romanization'
(Fulford 1992:29). This is a valid argument, particularly when considering the possibility of
less wealthy groups making an active attempt to appear `Roman'; and the possibility of
alternative expressions of wealth or status associated with Roman-like or prestigious goods
Roman-like
However,
it
be
than
to
structures.
with
remains
shown whether the
rather
forms
alternate
of Romanization remove the bias that plagues the concept
consideration of
`Romans'.
increases
I
focus
instead
just
the
that
of
on the
pool
would
suggest
we might
or
dichotomy
further
than
the
culture
of
material
and
settlement
rather
contexts
social
entrench
of Roman versus native society.
To summarize, the varying theories described in this section illustrate the dynamism of
Thames
Valley
during
Upper
in
the
this period of political transition. While each
settlement
idea may differ by degreeand emphasis,each considerssettlement from within the context
history
the
the
of settlementat each site. Notions of native continuity
and
whole region
of
in
Romanization
a number of the theories. However, what distinguishes
are
present
and
from
many of the accounts of Roman Britain is that basic
these various approaches
assumptionsthat are madegenerallyabout rural settlements(i. e. the presenceand absenceof
Romanizedstructuresand material culture) are at leastbeing challenged.
This study will continue that challenge by focusing on the nature of the impact of
imperialism on diet and culinary practices. It will becomeclear that the four casestudiesBarton Court Farm, Roughground Farm, Old Shifford Farm and Claydon Pike - do not
it
is
likely
Although
that the settlementswere occupied at
present a unified native voice.
58
different
different
housed
different
times,
they
of
status
agendas,
people
with
slightly
different appetites and consequentlydifferent experiencesof imperialism. Unfortunately,
terms such as `Romanization' and `native continuity' constrain the consideration of the
have
in
Any
to
the
which
people
might
conquest.
significance
ways
reacted
of
variety
Roman-like
the
to
or
absence
of
presence
structuresand material culture must be
attributed
contextualized.
4.4.
The Study of Native Settlement in the Upper Thames Valley
The heavy bias in favour of Romano-British villas and urban areas has had consequencesfor
our understanding of other types of settlements (Hingley
1989:4-5; Miles 1989:115;
Branigan 1991:92; Clarke 1998:28). The label `native' has obscured the diversity of
it
is
in
Roman
the
period as if by their classification as `native' we somehow
settlement
know what these types of settlements will divulge without excavation. Indeed, until recently
interest
in
the archaeology of the Upper Thames Valley focused on
the
academic
much of
the paucity of villas until late into the Roman period (see above and Young 1986:58).
However, gravel extraction in the region has resulted in the excavation of a variety of nonRoman-like settlements. Although there is now a bias towards excavated native settlements
(Miles
1986a:49; Lambrick 1992:78; though see Allen et al.
Valley's
terraces
the
gravel
on
1997 re: archaeology on the floodplain), the terraces were the most densely occupied areas
in the valley during the periods in question. As native settlements throughout Roman Britain
being
(Drury
increasingly
1982: 1; Hingley 1989:5) and published, it is
excavated
are
into
incorporating
them
to
accounts of Roman imperialism's impact.
necessary start
There are challengesand unlikely rewardsto a renewedanalysisof householdsin the Upper

ThamesValley in the late Iron Age and early Roman periods. One of the main challengesis
in
bias
houses
the
which
remains
of
a preservation
are not generally recovered.This has a
direct effect on the way `households' in the Upper Thames Valley can be defined in the
archaeologicalrecord.
59
4.4.1. Householdswithout houses
Any study of household archaeology in the Upper Thames Valley during the late Iron Age
features
is
by
Roman
the
complicated
poor preservation of structural
periods
and early
(Allen 1990:81). What we find instead are pits, gullies, enclosure ditches and occasionally
if
house-slots.
Beyond
shape
and,
conditions permit, the
and
suggestions
about
post-holes
types of materials used in their construction, the internal structure of houses in the Upper
Thames Valley (and in many other regions of Britain) is ambiguous (Allen et al. 1984). This
is in contrast with stone-built, villa-type structures found further afield where it is possible
to reconstruct floor plans and, at times, infer specialized areas of the house. The types of
innovative spatial and architectural analysis that typify the study of Romano-British villas
(see Scott 1990; Scott 1994; Smith 1978) are not always appropriate or possible for many of
the native type settlements in Roman Britain (see also discussion in Boast and Evans
1986: 194).
4.4.2. A definition of Upper ThamesValley `households'
With the settlements in the Upper Thames Valley, we need to look beyond 'the house' in
household.
Essentially,
definition
investigation
the
this
that
of
of
requires
any working
our
the term `household' has to fit the archaeology of the region. It would, for example, be
difficult to exclude non-family or extended family members from the definition, as it might
living
in
Equally,
be
the
to
the
arrangements
specify
of particular residents.
possible
not
buildings
be
it
households,
that
contain
multiple
settlements
and/or
not
might
case of
daily
interactions
disentangle
does
This
the
the
to
within
settlement.
not exclude
possible
the possibility
identifying
different
areas
of
of
or special status, of suggesting specific
family
associations or of establishing particular
gender or
practices associated with
in
Valley.
They
house
do
feature
definition.
in
For
the
the
the
simply
sites
not
particular
I
`household'
I
to
Hingley's
the
following
this
refer
when
study,
of
am
example
purposes
"group
to
the
who used and resided within a single house or group of closely
and referring
[my
1990a:
128
(Hingley
buildings"
emphasis]).
related
60
The structures that we recreate out of gullies, post-holes and foundation slots are as
interpretation
bias
in
in
(see
to
as
any
archaeological reconstruction
articles
susceptible
Drury 1982). As low-lying sites tend to have fluctuating water tables which affect the
features
has
it
been
(Pryor
1983:
191),
ideas
that
structural
argued
of
our
about the
survival
lowland
by
less
the
of
settlements
are
examples
prejudiced
of
organization
physically
(Pryor
in
1983:
190).
Reynolds,
settlements
upland
studying the archaeological
challenged
houses,
Iron
Age
has
instance
of
reconstructed
observed
one
movement
when "a post-built
structure was altered to a ring gully structure by rats tunnelling under the wall and living in
the space afforded by the wattle work" (1982: 176). He has many other such examples, as
few
instances
have
where
a
seasons
of
as
ploughing
well
removed all traces of post-holes
(1982: 190). Determining the arrangement of domestic areas at settlements in the Upper
Thames Valley requires high levels of inference and extensive excavation. The latter, I
has
benefited
the study of households in the region. At sites where
actually
would argue,
habitation
become
be
to
preserved,
all
are
poorly
aspects
of
structures
significant and need
1983:
(Pryor
197).
considered
The establishment of households without floor plans is less ambiguous if the remains
households,
and the environmental data that reflect human
commonly associated with
integrated.
House
late
Iron Age and early Roman Old Shifford Farm
sites
are
at
settlement,
(one of the sites in this study) have been proposed through the concentration of burnt pottery
impregnated
bones,
clay
with wattle, oven walls and charcoal at the terminals of
and animal
(Hey
1996: 105,111). Environmental data were
the
within
settlement
enclosures
particular
habitation
Farmoor
to
that lacked structural evidence but exhibited
of
a
site
at
used confirm
in
beetles
increase
the
number
of
associated with timber in the Roman period.
a substantial
Species of dung beetle that congregate around vegetal refuse and animal dung were also
for
habitation
at the site (Lambrick and Robinson 1979: 122,117). No
seen as evidence
structures were recovered at early Roman Watkins Farm, but, the recovery of woodworm
high
beetles
bread
and
quantities of pottery was seen to "suggest more than mucking out
and
from a settlement elsewhere" (Allen 1990:81). Artefacts and features that indicate a variety
household
done
on
a
scale, such as small scale textile production, metal
of activities
food
preparation and storage, and even the presence of human burials (Allen et al.
working,
1993: 191), contribute to the establishment of domestic occupation (Hey 1996: 138). The
hand-operated querns found at Roughground Farm, Old Shifford Farm and Claydon Pike,
61
Court
Farm,
Old
Claydon
Shifford
Farm,
Barton
hearths
the
at
and ovens recovered
and
Pike and RoughgroundFarm (Miles 1986b; Hey 1996; Allen et al. 1993:191) were all used
to substantiatethe casefor habitation at eachof the sitesfeaturedin this study.
It is of course conjectured that domestic material in circular gullies and rectangular slots
indicates house sites. If these features do not represent the location of houses, it is
interesting that domestic material accumulated at the entrances of gullies and enclosures at
in
Upper
Thames
is,
Valley.
It
the
the
sites
of
nevertheless, these apparently
many
distinctive concentrations of artefacts and remains that help to situate acts of eating and
drinking within the context of tradition, mores and outside forces that govern the whole
dietary cycle of consumption, including procurement and discard (Barrett 1989b; see Hill
1995 and Grant 1991 for the rituals of òrdinary' rubbish).
The absence of floor plans also forces us to place more importance on the deposition of
features
in
features
How
'behave'
throughout
the
culture
settlement.
surrounding
material
(Pryor 1983: 196) can reveal completely different aspects of life at a settlement than can a
house examined in isolation. According to Rapoport, "One cannot merely consider a
because
do
building
live
in,
buildings:
in,
they
people
not
or
act
exclusively
single
particular
buildings,
a variety of outdoor spaces. settlements, and whole regions"
use various
(1990: 12). This has been a criticism of excavations in general, and particularly of the
Romano-British
house
locus
has
been
the
the
villas,
where
of
primary
excavation
of study
(Gaffney
Tingle
1989:
3;
60;
Hayden
Miles
1988:
the
than
estate
and
whole
see
also
rather
160).
A
1983:
Cannon
notable consequence of the poor preservation of houses is that
and
be
have
to
the
site
plotted and excavated to determine the nature of
more parts of
distribution
Analysis
help
the
to
of
of
the
material
culture
settlement.
site may
around
house
location
the
of
possible
sites, but it also provides the means to
substantiate
household.
The
distribution of pots, bones and plants, for
the
the
of
working
contemplate
linked
be
to cuisine and rites surrounding consumption and discard. In an
example, can
integrated study of the household, all archaeological features, each artefact and biological
habitation,
fact
in
takes on importance.
trace
of
every
residue,
62
4.4.3. Early Romanperiod householdsin the Upper ThamesValley
Despite their now fragmentary appearance, pieced together from the remnants of habitation
it
Valley.
houses
in
Thames
decay,
Upper
And,
there
the
although
were
and the processes of
be
possible or even appropriate to pinpoint exact moments of change, a number of
may not
it
is
in
in
Valley
Roman
As
the
the
early
period.
were reorganized
native type settlements
likely that people who lived in these settlements also constructed them, or at least had a
hand in their construction, it might be assumed that their reconfiguration was, significant
(Rippengal 1993:93; Lawrence and Low 1990:492). Native-type settlements in the Upper
Thames Valley during the early Roman period are often identified by the presence of
in
fewer,
less
in
late
Iron
Age,
than
the
and
and
uniform, pits
rectangular enclosures, with
by
foundation
house
than
that
rather
sites
are
slots
represented
rectangular
cases,
some
is,
however,
distinction.
fine
line
divides
This
The
tenuous
that
prehistory
a
circular gullies.
firm
in
is
difficult
Romanists,
history,
the
of
minds
some
more
although
and
to isolate
fixed
).
The
is
(T.
Allen
Roman
pers. comm.
conquest
consequently not a
archaeologically
Roman
Lambrick
has
"The
in
this
the
recording
of
region;
stated,
archaeological
even
point
invisible
does
is
(1992:
105).
This
that
the
mean
on
not
gravels"
archaeologically
conquest
imperialism
by
looking
before
the
the
after
of
and
effects
at
consider
settlements
we cannot
Roman invasion, but it does require that we focus on the cumulative impact of the conquest,
fit
(Scott
historical
the
to
trying
and
than
archaeological record around specific
events
rather
Gaffney 1987:86; Jones 1997:29-30).
4.5.
The Upper Thames Valley - the Sites (seeFigure 4.1. - site number correspondsto
number on the map)
Before we commence with the case study element of this thesis, it is necessary to introduce
feature
The
throughout
that
this
the
sites
study.
of
remainder of this chapter,
a number
therefore, will seek to establish some semblance of the types of settlements identified in the
Upper Thames Valley thus far. The four sites to be examined in detail in the following
here
(signified
for
included
to
their
with
an
asterisk)
a
context
provide
chapters are
Valley.
in
is
I
the
that
this
should
settlements
not an
perhaps stress
association with other
inventory of settlement in the Upper Thames Valley during the late Iron Age and early
63
Romanperiods (seeLambrick 1992:82, figure 27 for a timeline and catalogueof settlement
in the region), and that some non-contemporaneoussites have also been included in the
form
ideas
developed
in
because
helped
they
to
this thesis.
to
give
synopsis
4.5.1. Sites on the floodplain
1. Farmoor
Farmoor is not a contemporaneous site, but it is referred to throughout this study in

terms of its environmental analysis and its characterization as a pastoral settlement, and is
thus included here. Farmoor is situated on the floodplain and the first gravel terrace. The site
in
Iron
Age
first
the
the
occupied
early
originally
on
gravel terrace and was re-occupied
was
in the middle Iron Age with a more substantial settlement encompassing both the floodplain
first
the
the
gravel terrace. It appears that the habitation area of the site was on
edge of
and
the floodplain and the first gravel terrace was used for the storage of animals. It is thought
that this settlement was entirely pastoral and quite possibly seasonally occupied as there is
for
flooding.
The settlement was soon abandoned and reseasonal
evidence
environmental
first
in
late
Roman period. As with a number of
terrace
the
the
gravel
occupied on
Farmoor
is
in
this
synopsis,
most relevant to this study because of its
settlements
environmental assessment,which was especially prescient as excavation was undertaken in
the 1970s.
References:Lambrick and Robinson 1979
2. Mingles Ditch
Mingies Ditch was situated on the floodplain. The site was occupied primarily
during the middle Iron Age c 380-1lObc. The settlement consisted of paddocks, ditched
enclosuresand a variety of gullies, which are thought to have contained structures. The
economywas primarily pastoraland evidencesuggeststhat horse rearing was an established
is
The
for
its
the
site
especially
notable
economy.
state of preservation,particularly
part of
64
best-preserved
houses:
Iron
Ditch
"Mingies
Age
the
the
was
settlement and one
of
of some
for
far
in
Upper
the
the
sites
environmental
so
most
suitable
archaeology
excavated
of
Thames Valley" (Allen and Robinson 1993:2). The site was abandoned around 50-100 years
later, possibly due to a rise in the water table, but the area was re-occupied in the early
Roman period when enclosure boundaries from an adjacent settlement cut through part of
the long abandoned Iron Age settlement.
References:Allen and Robinson 1993
4.5.2. Sites on the first gravel terrace
3. Appleford Field
Appleford Field was occupied in the Bronze Age, the early and middle Iron Ages
Roman
the
period as of the secondcentury.The early Roman settlementis characterized
and
by enclosuresand an elaborateseriesof trackways.It is thought that there were a numberof
households living at the settlement and that the settlement represents "a nucleated or
`village' settlement,as opposedto a single farmstead"(Hinchliffe and Thomas 1980:110).
References: Hinchliffe and Thomas 1980
4. *Old Shifford Farm
Old Shifford Farm lies on the first gravel terrace on the edge of the floodplain. The
from
late
Iron
Age
into
the
the early Roman period. Three phaseshave
occupied
site was
been identified for the late Iron Age/early Roman site, although occupation is thought to
have been continuous. Settlementbeganas an open group of circular gullies and a D-shape
ditch and, over time, evolved into an enclosedsettlementwith two ditched housesites. The
Roman
to
the
the
early
entrance
of
period settlement suggeststhe corralling of
structure
livestock. The economy for each phaseis thought to have been largely pastoral, although
there is evidence that plants may have been cultivated at the settlement. The site was
65
just
late
first
in
the
century;
a
established
or
early
second
new
settlement
was
abandoned
AD.
towards
the
the
third
the
settlement
earlier
end
of
century
of
north
References: site report: Hey 1996.
5. *Claydon Pike
Claydon Pike is situated on the edge of the first terrace on a series of gravel islands.
The site covers about 100 hectares and was occupied throughout the Iron Age and Roman
late
Iron
The
Age
discrete
settlement
comprised
a
number
of
zones: a trackway that
periods.
domestic
into
the
area of the settlement, a series of paddocks and an outer area of
went
The
in
first
the
settlement
enclosures.
was
completely
century AD,
reorganized
rectangular
be
to
appear
of
what
number
substantial structures were erected, and an extensive
when a
field system was established. The site is characterized by a series of platforms that enabled
(as
despite
its
low-lying
Settlement
landscape.
to
opposed
settlement
seasonal)
continuous
during the late Iron Age and early Roman period was largely confined to a single platform.
The settlement was primarily pastoral and evidence suggests that it was a part of a series of
`ranches' that specialized in the rearing of animals, in particular horses. The settlement was
first
the
the
at
end
of
completely reorganized
century AD, and expanded onto other
from
The
emphasis
economic
switched
platforms.
animal rearing to haymaking and it has
been suggested that there was some form of military presence at the settlement.
References: site report not published, for published commentary on the site see Miles and
Palmer 1983a; 1983b; Jones and Miles 1979; Wilson 1996b:52-55; Miles 1983; Miles
1984:197-203; Lambrick 1990:22-3.
6. Thornhill Farm
Thornhill Farm is located lkm west of Claydon Pike. The site was occupied during
the first century AD and was abandonedtowards the end of the century, around the time that
its neighbour Claydon Pike expanded.The site had two settlementareasthat were structured
66
differently. The first settlement area appears to have been used seasonally and comprises a
large number of animal pens and a small-scale occupation area. The second settlement area
is more substantial, and at least four possible structures have been identified. The structures
by
"The
[of
the
then
the second
animal
and
pens
enclosures.
planning
surrounded
are
dictated
by
be
constraints
more
cohesive
and
as
expected
physical
would
settlement] seems
(Palmer
Hey
33).
Evidence
1989:
occupied
site"
and
points to a specialized
permanently
a
on
`ranch'
horse
or
with
an
emphasis on
rearing. The settlement is thought
pastoral settlement
to have had a small resident population and is considered `native', ùnRomanized' and of
low status. The settlement was abandoned towards the end the first century AD and by the
Roman
former
through
the
a
road
century
cut
settlement
early second
References: post-excavation analysis is in progress; see Palmer and Hey 1989; Lambrick
1990:22-3 for discussionof site
7. WatkinsFarm
Watkins Farm lies on the first gravel terrace on an island of gravel that was higher
than the surrounding land. The site was first occupied in the middle Iron Age and again in
the Roman period beginning in the late first/early secondcenturies.The economy for both
the middle Iron Age and early Roman settlementsis thought to have beenpastoral, and there
have
been "seasonal or sporadic" (Allen
the
that
of
site
occupation
may
are suggestions
1990:81). The main circular enclosureof the middle Iron Age settlementwas re-used in the
Roman
period and a number of more rectangularenclosureditches were constructed
early
Roman
At
the
the
early
main enclosure.
period settlement,a secondhabitation area
outside
(site B) has been identified, consisting of a single `house-sized' oval enclosure with large
In
from
the
comparing
the two areasof settlement,higher
of
pottery.
pottery
concentrations
from
fine
wares were recovered
site A, which according to Allen (1990), could
numbersof
indicate that the "enclosureson Site A were being run from Site B".
References: site report: Allen 1990
67
4.5.3. Siteson the secondgravel terrace
8. Abingdon Vineyards
The site is situatedat the confluenceof the Thamesand the river Ock, under what is
in
by
Main
Iron
Abingdon.
Age
the
the middle
town
the
occupation
of
started
early
and
now
Iron Age a `village' was established, with a minimum of nine circular house sites and
in
late
had
There
Iron
Age
the
the
that
are
suggestions
structures.
settlement
other
possibly
become an oppidum as evidenced by the extent of traded goods and the construction of a
lOm wide 3m deep `defensive' ditch around the settlement. The circular houses were
in
by
Roman
The
the
square
enclosures
and
concentration
rectangular
early
period.
replaced
Roman
Abingdon,
together
the
trade
with
around
presence
of
early
sites
goods,
of rural
fine
function
high
to
that
the
ware
suggests
as
of
percentage
settlement continued
coins and
The
in
local
the mid-second century.
trading
settlement
was
abandoned
centre.
a
References:site report is not published,seeMiles 1986b:3.
9. Gravelly Guy
Gravelly Guy is situated on the edge of the secondgravel terrace and as there is no
floodplain terrace at this point in the valley, it converges with the floodplain. The site was
first settled in the early Iron Age. The Iron Age `village', consisting of as many as 30
houses, was densely packed into a 170m by 30m strip of land between a boundary of
land
land
side
and
arable
on
one
at the edge of the gravel terrace on the
common grazing
between
boundary
The
the settlement and the common land was strictly observed and
other.
is thought to have had a ritual association as it conforms to bounded space around Bronze
Age Devil's Quoits. The 'village' utilized the arable land, but the settlement was primarily
farming
is
described
the
of
as
part
co-operative
viewed
and
earlier. In the
pastoral,
regime
late Iron Age, the settlement was abandoned and a new settlement was established over the
former boundary of `common land'. This settlement, which was occupied into the Roman
68
features"
domestic
"strange
had
and
wells
structures, paddocks,
arena-like
period,
(Lambrick 1990:18).
References: site report is not published, for summaryof site see Lambrick 1990; Limbrey
138;
1988:
Allen
2000:
12-3,21.
Robinson
and
10. *Barton Court Farm
Barton Court Farm lies on the edge of the second gravel terrace. The site was
initially occupiedin the late Neolithic. The areawas re-occupiedin the late Iron Age with a
fairly substantial settlement of possibly two households and various animal pens. The
is
believed
have
been
In
Roman
the
to
the
settlement
of
early
period, a new
mixed.
economy
late
Iron
Age settlement. It consisted of a single
the
over
was
established
settlement
large
house
have
been
is
the
to
thought
and
animal
pen,
and
economy
also
rectangular
in
This
the second century and a small villa was
was
abandoned
settlement
mixed.
in
late
the
the
third century.
site
on
established
References:site report: Miles 1986b;seealso Jones1985
11.Ashville Trading Estate
The site is situated on the second gravel terrace, 2.7 km west southwestof Barton
Court Farm. The site was originally
occupied in the Bronze Age. Three phases of
for
identified
Iron
Age,
the
which roughly correspond to the early, middle
occupation were
Occupation
Age
late
Iron
periods.
was most intense during the middle Iron Age phase
and
is
there
the
evidence of a variety of post-hole structures, pits and ditches. The
where
site
of
late Iron Age phase of the site consists of a series of ditches thought to delineate a field
habitation
has
identified
been
that
suggest
which
although
and
pits,
system
on site, it
not
late
The
Iron
Age
likely
pottery at Ashville was recently re-assessedand it
nearby.
was most
is now believed that the late Iron Age settlement extended into the early Roman period
(Henig and Booth 2000: 107). The site was occupied into the -second century and was
69
followed by a hiatus in occupation until the late Roman period. The economy for the site
domestic
industries
evidence
of
small-scale
mixed
with
such as metal working,
appears
leather working and weaving. In reference to this study's emphasis and the trajectory of
Roman archaeology, the excavation is notable for its attention to specialist reports; it is
interesting that the Preface anticipates that some will view the length of these reports as
"excessive".
References: site report: Parrington 1978; Henig and Booth 2000: 107.
12. *Roughground Farm
Roughground Farm is situated on the second gravel terrace and is geographically

Court
Farm.
The
is
Barton
to
site
situated 2.5 km northwest of Claydon Pike. The
similar
in
late
Neolithic,
late
The
Bronze
Iron
Age.
Age
the
sporadically
occupied
was
site
and early
first major phase of occupation was an early Roman settlement. Although a large part of the
in
Roman
1950s,
the
the excavator's philosophy was similar
settlement
was
excavated
early
to the O.A. U., in that the aim was in widespread excavation of the settlement. Early Roman
identified
in
features
were
a number of areas of the site, although the main
period
have
been
bounded
by
to
appears
area
a rectangular ditched enclosure. Inside the
occupation
enclosure were pits, ovens and possible animal pens and a series of concentric circular
gullies which probably contained structures. A mixed farming economy has been suggested
for the settlement through the presence of querns and storage pits along with stock
The
in
the early second century when the main
settlement
was
re-organized
enclosures.
occupation area shifts and a `villa' was constructed in its place.
References: site report: Allen 1993; seealso Miles 1984:203-205
70
4.6.
Conclusion
is a region characterized by political
movement and
transformation (Miles 1982:56) and it is, for the purposes of this study, ideal for the
dynamics
imperialism
its
households.
The
the
on
of
and
effects
native
of
consideration
Oxford
large-scale
Archaeological
Unit,
the
the
excavation
of
which
emphasize
objectives
of settlements and extensive environmental assessment, together with a condition of
forces
beyond
ideal
for
`house'
the
that
the study of the
analysis
are
similarly
preservation
drinking.
the
and
artefacts
remains
associated
eating
and
of
with
contexts
social
What follows is an analysis of four sites in the Upper Thames Valley - Barton Court Farm,
Roughground Farm, Old Shifford Farm and Claydon Pike. Diversity
discontinuity'
`native
of
or
evidence
settlements
in native type
(see Meadows 1997:33; 1999) is often
better
Through
by
Roman-like.
that
are
settlements
preserved
and
ultimately
more
subsumed
during
diet
the
the
social
contexts
of
site
the consideration of
and culinary practices at each
late Iron Age and early Roman periods, I will
between
that
the
argue
relationship
discontinuity of settlement and Romanized elements is integral to questions regarding the

for
is
different
is
be
imperialism;
the
that
association
complex,
each
and one
will
effects of
become
in
Roman
It
in
that
this
the
will
clear
study.
study of social change early
settlement
Britain will not be resolved using a template of Romanization.
71
"
6
"
9f
s"5
-shames
"
kms.
20
Figure 4.1 The Upper ThamesValley and sites mentioned in this study (by C. Merrony)
(1. Farmoor 2. Mingies Ditch 3. Appleford Field 4. Old Shifford Farm 5. Claydon Pike
6. Thornhill Farm 7. Watkins Farm S. Abingdon Vineyards 9. Gravelly Guy 10. Barton
Court Farm 11. Ashville Trading Estate 12. RoughgroundFarm)
72
Chapter 5
5.1.
5.2.
5.2.1.
5.2.2.
5.3.
5.3.1.
5.3.2.
5.3.3.
5.3.4.
Table of Contents for Barton Court Farm
Introduction
The Excavation
The site
Early Roman period settlement
Site reports, microfiche and archives
The containers
The specialists' reports - the pottery
Thepottery at the late Iron Age settlement
Form and function
74
74
79
Fabrics
Rim diameters
Other implementsat the late Iron Age settlement
The ingredients
The specialists' reports - the bones
5.3.5.
Thespecialists' reports - theplant remains

The ingredientsat the late Iron Age settlement
Theanimal bonesat the Late Iron Age settlement
Species representation
Butchery practices
Meat yields
Theplant remains at the late Iron Age settlement

5.4. Food and Drink Consumption at the Roman Period Settlement
5.4.1.
90
The containersat the early Romanperiod settlement

Thepottery at the early Roman period settlement
Form, function and fabric
Rim diameters
Theearly Romanperiod glass
5.4.2.
5.4.3.
Other implements at the early Roman period settlement

Ingredients at the early Roman period settlement
Thebonesat the early Romanperiod settlement

Speciesrepresentation
Butchery practices
Meat yields
Theplant remains at the early Romanperiod settlement
5.5.
5.5.1.
Summary of the Artefacts and Remains of Consumption for each Settlement

Containers
Pottery
Other implements
Ingredients
Animal bones
Plants
5.5.2.

Containers
Pottery
Glass
96
73
Other implements
Ingredients
Animal bones
Plants
5.6.
5.6.1.
The Distribution of the Remains of Eating and Drinking at Barton Court Farm
Distribution of the artefacts and remains at the late Iron Age settlement
98
'Specialdeposits'
5.6.2.
Summary of the re-contextualized material

Distribution of the artefactsand remainsat the early Roman period settlement
'Special deposits'
Summaryof the re-contextualizedmaterial

5.7.
5.7.1.
5.7.2.
5.8.
5.9.
Discussion of the Distribution of Artefacts and Remains at Barton Court Farm

The Iron Age settlement
Conclusion
Figures 5.1- 5.27
107
109
111
112
74
Chapter 5
Barton Court Farm
5.1.
Introduction
The site of Barton Court Farm is located on the edge of the second gravel terrace, 59.5m
level,
Thames
Ock.
It
the
the
confluence
of
near
rivers
and
was a multi-period
sea
above
Neolithic,
late
Angloduring
Iron
late
Roman,
Roman
Age,
inhabited
the
and
early
site,
Saxon periods. Barton Court Farm was my introduction to `native' eating and drinking
Valley.
Upper
Thames
My
it
in
to
the
approach
structured my analysis of the other
practices
it
does
àrchetype'
in
thesis,
this
although
not
constitute
an
against which others were
sites
compared.
This chapterwill commencewith a summaryof the excavationof, and configuration of, the
late Iron Age and early Roman settlementsat Barton Court Farm. Following this, the social
drinking
first,
be
through an
around
eating
and
revolve
will
which
explored:
practices
implements
ingredients
in
both
the
the preparation and
and
of
used
examination
drink;
food
and second, through the consideration of the various
and
consumption of
in
distribution
involved
The
these
the
the
of
remains
on
chapterwill conclude
site.
processes
interpretation
discussion
of
my
of the excavated material, within the context of
with a
imperialism.
5.2.
The Excavation
The Oxford Archaeological Unit, under the direction of David Miles, excavated Barton
Court Farm in the 1970s.The site report was completedby 1980 and published in 1986.The
designed
in
three-month
as
a
originally
conducted
was
project
rescue
excavation
75
in
initial
development.
funding
The
the
aim
of
excavation,
view
of
and time
anticipation of
"reveal
the sequence of occupation and elements of settlement
to
was
constraints,
(Miles
organization"
1986b:xiii).
The site turned out to be more complex than was
because
in
late
late
Roman
Iron
Age,
to
settlement,
addition
evidence
of
projected
originally
identified.
Saxon
in
housing
As
Roman
the
the
settlement
was
of
a
slump
and
result
early
local
leading
to the establishment of an
the
government
reorganization of
market and
Oxford Archaeological Committee, the excavation was extended and "modified to take
developing
interests"
(Miles
the
the
circumstances
and
new
of
archaeological
account of
1986b:xiii). Miles provides additional insight into the excavations' rationale: "The approach
to the excavation was essentially a pragmatic one, conducted on a year-to-year basis,
by
but
by
little
thinking
amount
of
certain
wishful
governed
medium-toa
very
stimulated
long-term planning" (Miles 1986b:microfiche: 3: A3). One of the `developing archaeological
interests' at the time was the excavation and sampling of biological material. Bob Wilson
(zoologist), Mark Robinson (entomologist/botanist)
and Martin
Jones (botanist) were
be
to
an
environmental
conduct
assessment
a part of the
as
well
as
commissioned
Miles
"everyone
that
that
although
cautions
admit
process,
concerned
would
excavation
these were very much the primitive
in
days
pioneering
of environmental archaeology
Oxfordshire" (1986b: xii; see also comments by Robinson 1994: 105).
The site at Barton Court Farm was identified through a series of crop marks, which
2
hectares
and the possibility of three phasesof occupation.
of
about
suggestedcoverage
Until the 1970s, the site was contained under three agricultural fields. The site was partly
depth
but
(0.2-0.3
hand,
by
the
the
of
soil
m) mitigated the possibility of
stripped
stratigraphicevidenceand the majority of the site was stripped by JCB. It was noted in the
"Two
that
workers armed with shovelswere able to observethe topsoil stripping and
report
"This
loose
left
that
and
material"
process
a clean undisturbed gravel surface
copewith any
in which features mostly showedvery clearly" (Miles 1986b:microfiche 3:Al). Apparently,
`blank areas' were re-evaluatedwhen post-holesappearedonly "after persistentand careful
3:
A3).
1986b:
(Miles
microfiche
searching"
Approximately two-thirds of the late Iron Age enclosurewas excavatedas well as a small
trench in the northeastcorner. As there was not much re-cutting of featuresand the pottery
it
into
is
believed
types,
that the settlement
pottery
of
sequential
groups
stratified
was not
76
fewer.
began
hundred
Occupation
in
first
for
late
the
or
years
one
century
was occupied
BC/early first century AD. It seemslikely that the enclosureditches were open throughout
the occupation of the settlement,basedon their shape,the gradual accumulation of silt, the
deposition
human
burials
the
and
and
pottery,
of
within the
positioning of animal remains
ditches. The ditches at the Iron Age settlementhad silted up before the construction of the
it
is
features
Roman
that
the
so
possible
settlement,
a
number
of
were not obvious
early
The
Roman
into
divided
two
the
was
constructed.
settlement
early
settlement
was
new
when
(which
the
the
the
of
southern
area
of
settlement
contained all of the
majority
areasand
features) was excavated; thirty percent of the northern area of the site, thought to be an
The
Roman
is
have
been
believed
to
excavated.
early
also
settlement
was
also
pen,
animal
lack
features.
began
because
This
the
of
of
re-cutting
of
phase
of
occupation
short-lived
first
half
latter
the
the
century AD. (Miles 1986b:micro fiche 3)
of
towards
5.2.1.
The site (seeFigure 5.1.)
Thelate Iron Age settlement

The late Iron Age settlementappearsto have had two occupational foci, one situated inside
just
it.
Both occupation areas were defined by the
the
other
outside
the main enclosure,
irregular
two
connected
closely
structures:
a
circular
a
series
of
gully and
of
association
Miles,
has
David
the
the
author
of
site
report,
suggestedthat these roughly
post-holes.
indicate
household
the
areas
may
expansion
occupation
of
a
single
over time,
contemporary
households
differing
functional
(1986b:
There
28).
two
or
areas
possibly
of
status
separated
the
within
settlement,which separatethe two occupation areas
enclosures
of
a
number
are
define
large
that
The
area
contained
a
common
animal
pens
group
and
a
of
pits.
possibly
and
found
in
in
the
the associatedpits, as well as the
gullies
remains
and
artefactsand animal
daub
human
burials,
hearths,
the
distinguish
helped
to
and
proximity
of
the
of
presence
habitation
site.
settlementas a
77
Theearly Romanperiod settlement
In the early Roman period, the settlement was reorganized on a completely different
features
The
Roman
had
the
earlier
none
of
were
reused.
early
and
settlement
alignment
focus,
but
the size of the principal rectangular structure - "five times
occupation
only one
that of an average Iron Age house" (Miles 1986b:30) - suggests that it may have
larger
household
is
(Miles
1986b:
30).
As
there
consolidated
a
or
more
no
accommodated
break
in
the sequence of artefacts, and the early Roman period
chronological
obvious
in
location
late
it
has
been
Iron
Age
the
the
same
was
constructed
as
settlement,
settlement
(and
49;
1986b:
(Miles
that
occupation
possibly
see
ownership)
continuous
was
suggested
is
It
136).
found
in
1995:
Ferrell
therefore,
that
the
possible,
early
some of
material
also
Roman period contexts is residual. The types of artefacts and remains found itl the houseslots and surrounding site enclosure, particularly wall plaster, are again consistent with
habitation, although the nature of deposition for the two settlements was quite different.
There is only minimal evidence of manufacturing at each settlement. Storage facilities

is
lacking
for
domestic
the storage (and hence redistribution) of a
as
evidence
use,
suggest
large surplus of agricultural produce in either period (Miles 1986b:45). The biological data
from the two settlements indicates an integrated mixed economy, with production and
foods
Roman
The
cereals,
as
such
meat,
cheese
of
and
period
milk.
early
consumption
is
viewed as a non-Romanized settlement conveying a strong sense of native
settlement
50).
1986b:
(Miles
continuity
5.2.2. Site reports, microfiche and archives
My interpretation of the site is basedon the site report and accompanyingmicrofiche, and
Oxfordshire
held
Museums Store in Standlake, Oxfordshire. The
the
the site archive
at
is
a summaryof the excavation; all of the details of the excavationand
report
published site
`supporting evidence' are presented on microfiche. Miles adds, "It is hoped that by
detailed
information
in
flexibly
this
can
provide
way
we
as
as cheaply and
publishing
Together,
details
(1986b:
the
that are
published
report
and
xii).
microfiche
provide
possible"
(see
by
Huddlestone
Fulford
1991;
filtered
comments
similar
site
reports
of
out
and
often
78
Hodder 1989). The organization of the microfiche is also atypical. Rather than organizing
the finds according to material, they were organized into functional categories (contra
Wacher and McWhirr 1982; Stead and Rigby 1986; Partridge 1981). The report
format
is
that
such
a
not unproblematic:
acknowledges
There are obvious dangers in this method in that the function of objects is not
always obvious or one object may have been used for several purposes.
Nevertheless, it is hoped that this method is valid and aids the interpretation
1986b:
(Miles
5:
A1).
the
site
microfiche
of
The finds were organized in relation to "primary production activities, secondary production
1986b:
information
(Miles
5:
The
A1).
microfiche
site archive contains
and consumption"
in
in
included
and
order to clarify the contextual associations and get closer
publication
not
to the process of analysis and organization, I determined that the site archives for each case
be
In
to
my
research.
order to expand upon the work previously
essential
study would
it
database
by
into
the
to
specialists,
was
various
necessary
enter
a computerized
conducted
the original pottery and bone records for both the late Iron Age and early Roman sites, as
form
in
only.
they existed written
Hodder (1989) has been critical of the missing dialogue between specialist and
in
both
dialogue
While
the
site
published
reports
and
site
archives.
much of
excavator/editor
is missing from the reports and archives featured in this thesis, as mentioned above, for
Barton Court Farm we have the microfiche element of the site report. In addition, a number
in
interpretation
involved
in
the
Upper
Thames
the
of
excavated
the
material
specialists
of
Valley are prolific publishers and their opinions and thoughts can be found in other
Barton
In
Court
Farm,
Jones,
to
Martin
specific
reference
sources.
who analysed
published
has
commentedelsewhere(Jones 1985) on how the botanical sampling
the plant remains,
focusing
features,
hearths
the
on
site
particular
at
ovens,
employed
etc.
such as
strategy
(after Dennell 1974) - was unproductive: "Unfortunately, the resulting analysis shows no
inferred
functions
between
these
and the composition of their associated
clear relationship
111).
Comments
1985:
(Jones
by
Mark
environmental
specialist
assemblages"
plant
Robinson and animal bone specialistBob Wilson, whose works are featuredthroughout this
backdrop
humanized
to the creation of specieslists and their interpretation.
thesis,provide a
79
Robinson (1994), from his perspective as an environmental archaeologist.who was the
director of the excavationof Mingies Ditch, wryly concludes:
Perhaps one of the greatest benefits to environmental archaeology of having

directing
is
it
forces
that
an excavation
on an
an environmental archaeologist
broader
archaeologist
a
much
understanding of archaeology
environmental
(1994:
111).
the
of
excavation
process
and
Wilson (1994) argues that the "Documentation of publication is mainly a record of probable
influences upon the work of a researcher and differs from the record and intensity of
investigation undertaken in the Upper Thames Valley" (1994: 58). His frustration with the `7
to 12 year' hiatus between the completion of excavation and publication, and the effect that
this has had on particular research interests, in his case the spatial patterning of animal
bones, is clearly apparent and is probably shared by many specialists:
All this seems symptomatic of the frustrations of zooarchaeologicalwork

from which a significant proportion of my colleagues have resigned from
their contract employment during the last year or so. The longer the reports
bones
for
the
of
analysis
wait
spatial
publication the more one must
on
interests
little
from
investment
that
the
archaeological
of
wider
gain
conclude
thousandsof pounds spent by the Ancient Monuments Lab on animal bone
his
in
Wilson
65;
1995;
(1994:
1996b:77-79).
see
also
remarks
work
The underlying passion,evident in the words of many of the specialistswho publish outside
bearing
had
has
on the tone of this thesis which I hope, as is
a particular
the site report,
intended,is both appreciativeand respectfulin its re-examination.
5.3.
This section will commencewith a discussionon the nature of the remains of eating and
drinking, to include the methodologiesand conclusionsof the various specialists,as well as
between
distinction
data.
The
the work of the various specialistsand
the
my own analysisof
beginning
be
The
the
the
at
outlined
of
relevant
will
artefacts
subsections.
my own analysis
follows:
`containers'
include
òther
as
which
and
glass,
grouped
are
pottery
and remains
-
80
implements'which includes metal, stone and clay artefactsand finally 'ingredients' which
includeanimalremainsand plants.
5.3.1. The containers
Thespecialists'reports - thepottery
The pottery was analysed by David Miles with Diane Hofdahl and John Moore. A type series
for
Four
fabric
identified
types
complete
using
vessels
and
rim
sherds.
were
was established
the Iron Age pottery using criteria set out by De Roche (1978) at the late Iron Age settlement
Centre.
The
Trading
Roman
includes
Ashville
the late Iron Age
early
assemblage
period
at
fabrics and additional fabrics such as grey, white and oxidized wares that typify pottery of the
In
found
for
both
in
features
Roman
the
the
microfiche
period.
report
pottery
particular
early
listed;
in
discrepancy
listed
between
the
was
cases
of
a
on
what was
period settlements
listed
in
the primary report, the primary record was used. All of my
microfiche and what was
is
based
Miles
the
the
on
primary
pottery
of
records.
et al. also conducted some
analysis
in
in
difference
to
the
tests
there
order
pottery
establish
on
whether
pottery
statistical
was a
forms between the two periods. They established that there was `some similarity' between the
two assemblages(Miles et al. 1986:microfiche 7: C4). My findings are different because Miles
Roman
based
their
on
early
period pottery rather than the pottery recovered at
numbers
et al.
the early Roman period settlement where both native and Roman period ware was present.
The majority of the pottery came from òpen' enclosure ditches and gullies, with a small
found
in
`closed'
of
pottery
groups
primarily
pits. Miles et al. used vessel size,
amount of
form, decorationand fabric to differentiate kitchenwarefrom tableware for both the late Iron
Age and early Roman periods. This type of distinction for the Iron Age pottery, was not,
however,incorporatedinto the published report.
At Barton Court Farm, the only containersfound in stratified contexts, and that relatesto the
food
drink,
discussion
The
of
and
consumption
were
and
ceramic.
preparation
present
will
accordingly concentrateon pottery with the proviso that glass,metal and wooden containers
in
Chapter
discussion
(see
3
been
3.4.3.
have
).
section
present
may
81
5.3.2. The containersat the late Iron Age settlement
Form and function

71.6 kg of pottery was recovered and identified for the late Iron Age phase of Barton Court
Farm. As with other late Iron Age sites in the area, wheel-thrown jars and bowls were more
hand-made
The
locally
the
than
the
variety.
majority
produced,
of
vessels
were
common
beakers)
(butt
imported
late
from
bulk
The
the
the
pottery
was
some
southeast.
of
although
Iron Age assemblageconsists of a variety of jars and bowls and vessels that fall in between
the `jar' and `bowl' categorizations (see Figure 5.2.). The designation of `bowl' or `jar' is
difficult when not much of the body profile is present, and is even more problematic with the
late Iron Age pottery as much of the pottery straddles the cut-off point between the two forms
(D. Miles pers. comm.). What is particularly
noticeable in the late Iron Age pottery
high
bowls
is
in
jars
the
to
proportion
of
comparison
recovered at the
assemblage
bowls,
in
bowls,
A
these
the
of
especially
used
number
necked
were probably
settlement.
food
had
(see De Roche 1978:71)
bowls
two
the preparation and cooking of
necked
been
have
the
through
that
the
neck
which
suggests
suspended,
vessels
may
perforations
fire
below).
bowls
(see
tableware
type
nevertheless,
open
an
over
are prominent
possibly
Many of the bowls are, for example, burnished, which could have been partly for aesthetic
(Oetgen
1984:
41).
than
a
sealant
as
purposes rather
Large storage-type vessels are distinguishable, although in light of Allen's findings at

Watkins Farm (see Chapter 3 section 3.4.3.), their exclusive associationwith storage is at
least questionable.Sizeableproportions of the jars had perforated basesand are thought to
have been used for either cheesemaking, distilling, steaming (D. Miles pers. comm.), or
It
flowerpots
and
colanders.
was not possible to confirm which of the jars served
possibly as
as cooking pots since evidenceof residuesor sooting - often the only meansof specifying a
158)
in
1986:
(Woods
Woods
has
the
was
not
recorded
primary
record.
pot
pottery
cooking
Roman
Iron
Age
had
bases
(which
however,
flat
that
and
cooking
pots
generally
observed,
hard
33)).
(Barley
1994:
tables
they
that
on
surfaces
were
placed
or
were made out
suggests
had
fabric,
Woods
has
(1986:
159-163).
pronounced,
often
everted,
and
rims
also
of a coarse
82
determined that open bowls would not have been suitable for cooking (1984: 28). The two
found
late
fabrics
Iron
Age
the
at
settlement - calcareous and sandy coarse wares main
for
fit
Woods
both
(see
1986: 163); in fact,
that
these
used
vessels
characteristics
also
were
including
high
bowls,
forms,
the
proportion of
and the beakers and decorated vessels
most
both
fabrics.
discussed
As
was
out
of
above, the pottery specialists initially
were made
designated some of the late Iron Age pottery types - beakers, cordoned and necked bowls,
These
dishes
35%
identified
tableware.
types
the
as
vessel
comprise
of
pottery
and
is
if
designation
the
sizeable
which
a
percentage
at all approximates their
assemblage,
functional
distinction
The
between
is
kitchenware
tableware
usage.
and
not
suggested
is
is
What
forms
the
apparent
sheer
variety
of
and sizes of vessels, which are not
categorical.
kitchenware
does
Nor
tableware.
the pottery appear to
as
either
or
characterized
easily
conform to one particular type of consumption practice, such as large bowls and dishes
indicating communal eating. The use of beakers - whose diameters range from 14cm to
40cm - however, may indicate communal drinking. Clearly, any further attempt to refine the
distinctions between the pottery types requires consideration in terms of the ingredients and
in terms of their contextual associations.
Rim diameters
Although I initially soughtto establisha range of small, medium, and large sized vessels(so
identified
diameters
incorporate
into the analysis of function
types
to
vessel
without
rim
as
instead
found
I
its
to
what
size),
was a wide variability in size of the same
and relationship
form. For example,the rim diametersof a seriesof `large' storagetype vesselsranged from
14cm to 30cm, a group of burnished globular bowls had rim diameters which ranged from
1lcm to 30cm, and the rim diametersof a seriesof cordonedbowls (thought tentatively to
be tablewareby the pottery specialist) rangedfrom 11cm to 26cm. This obviously made any
identify
for
to
of
consumption
specific
pattern
a
particular types of vessels(such as
attempt
large or small cooking-type pots or small or large serving-type bowls) extremely difficult.
Histogramsof the rim diametersof jars and bowls revealeda range of small to large bowls,
skewedtowardsthe small to medium end of the scale(seeFigure 5.3.), and small to large
jars
had
(see
Figure
jars,
5.4.). Histograms of tableware
of
medium
sized
a
peak
which
sized
(seeFigure 5.5.) indicate a similar rangeand peak in size as was found for the bowls.
83
Fabrics
Four fabrics were identified for the late Iron Age pottery. Fabric 1, is defined by its
fabric
is
due
inclusions;
inclusions;
2,
fabric
flint
3,
has
to
a
sandy
ware
quartz
calcareous
inclusions; and the fourth fabric was usedto make butt and girth beakersand is considereda
`fine ware' fabric. The calcareous and sandy fabrics were the most common fabrics
identified and, as was stated above, most vessel forms occur in both fabrics. The only
distinction betweenthe two main fabrics was that bowls and dishes were more commonly
fabric
(see
Figure
5.6.
).
the
calcareous
of
made
It is necessaryto make one final note on the pottery recovered from the late Iron Age
layers
found
in
Two
the
top
two
samian
of
ware
and
sherds
amphorasherdswere
settlement.
dish
18
Drag
These
four
the
are
pieces
probably
residual
although
one
a
of
piecesof
pits.
of
is pre-Flavian and one of the Dressel 2-4 amphora sherds is dated to the first half of first
is
It
AD.
conceivablethat these types of vesselswere used towards the end of the
century
occupationof the settlement.
5.3.3. Other implementsat the late Iron Age settlement
Only a few artefacts were recovered that have an association with the preparation or
drink.
Fired
food
found
clay
slabs
and
were
of
at the baseof a number of pits
consumption
hearths
iron
(1986b:
6).
An
Miles
that
they
may
represent
or
parching
suggests
ovens
and
knife was recovered at the late Iron Age settlement, which may or may not have had a
culinary use.
5.3.4. The Ingredients
Thespecialists' reports - the bones

The animal bones recovered from the various phases at Barton Court Farm were well
Over
bones
fragmented.
23,000
for
but
and
shells
quite
were recovered the whole
preserved
84
handpicked
37%
bones
3%
bones
identified
the
the
to
of
and
which
of
sieved
were
site, of
bones
in
features
believed
The
butchery
the
the
to
of
are
majority
represent
and
species.
food refuse (Miles 1986b:29). The objective of the bone specialist, Bob Wilson, upon
identification of the species, was to look at the distribution of particular body parts (head,
feet, body) in particular feature types (pits and ditches) for each period (Wilson
1986:microfiche 8:B4). Wilson has conducted similar types of analysis for a variety of sites
in the Upper Thames Valley, and has observed a number of trends in the distribution of
bones around settlements (1989; see also discussion in Chapter 3 section 3.4.4.). He did not
identify any major distribution and butchery patterns at the site for the periods in question.

Martin Jonesanalysedthe plant remains at Barton Court Farm. As was mentioned above
Joneswas not completely satisfied with the sampling strategy employed,which rather than
being randomized, focused on specific features. Just short of 50% of the cereal grains
Court
Farm
identified
Barton
by
Plant
at
were
species.
remains were not
recovered
for
Roman
late
found
in
the
the
two
settlement
early
and
were only
contexts at
recovered
Iron Age settlement.The processingstagesimmediately before consumption are difficult to
identify archaeologically (Jones 1991:25) and direct evidence for preparation and
is
milling,
or
rolling,
cracking,
whether
absent. Additionally, the artefacts
consumption,
for
found
in
the
with
preparation
of
grains
associated
consumption were not
commonly
Roman
In
Iron
Age
late
this section, I will primarily comment on
contexts.
or
early
either
Jones' findings, placing particular emphasis on the wild plants that are associatedwith
consumption.
5.3.5. The ingredientsat the late Iron Age settlement
Theanimal bonesat the late Iron Age settlement
Before we proceed,it is necessaryto comment briefly on the use of M. N. I. (minimum
85
individuals)
identified
N.
I.
S.
P.
(number
or
of
of
specimens) as measurements of
number
both
frequencies.
There
N.
I.
S.
P.
types
are
problems
with
using
of
measurements:
species
favours larger species of animals whose bones are generally more fragmented and M. N. I.
tends to exaggerate the significance of infrequent species (Payne 1972; Maltby 1979:5-7).
As I do not consider species representation as necessarily a reflection of the whole animal
focus
in
I
N.
S.
for
I.
P.
the
this
thesis,
on
each species.
population,
The N. I. S.P of the four main domestics at the late Iron Age settlement as established by
Wilson, are summarized in Figure 5.7.. My calculation of N. I. S.P. (see Figure 5.8.) is
from
Wilson
because
different
is
divergence
between
that
the
there
of
some
slightly
inclusion/exclusion of a small number of contexts (my calculations do not include teeth and
bones
burials,
I
however,
)
that
to
articulated
and
are
counted
as
one.
wish
reiterate,
animal
in addition to the many taphonomic factors that may affect species population and
feasting,
butchery
cultural
practices
such
as
ratios,
animal
sacrifice
and
abundance
techniques have a direct effect on the configuration of different species.
Wilson has determinedthat cattle, and particularly sheep,both tendedto be killed at an early
bulls
killed
implies
in
life,
that
steers,
and
rams
were
off
early
and that cows and
and
age,
herds
flocks.
large
This
led
him
to suggestthat the age
of
and
a
proportion
up
ewes made
indicate
dairying
(Wilson
the
sheep
towards
cattle
and
could
a
of
propensity
sex
ratios
and
1986). This might explain the prevalenceof possible cheese-makingvessels found at the
late Iron Age settlement.Bogucki (1984) has looked at ceramics,particularly sieves,as well
importance
highlight
dairying
looking
in
kill
Neolithic.
In
the
to
possible
the
of
at
rates,
as
he
found
from
later
that
sieves
periods,
other
were
also
at
pre-historic
common
examples
important
20).
Alternatively,
(1984:
the
an
was
part
of
pastoralism
economy
as
sites where
3
(section
3.4.3.
Chapter
),
in
discussed
the prevalenceof young animals could indicate
was
that there was an intensity of ritualized consumption (Schuster Keswani 1994:261) or
feasting at the settlement.If this was indeedthe case,it is possible that the presenceof burnt
in
(Gomez
de
Solo
1993; Vencl 1994:310) and the
a
number
pits
charcoal
stones and
bases
(sieves?
)
perforated
with
vessels
of
might indicate that alcohol was
proliferation
(Gomez
de
Solo
1993:
the
191).
settlement
at
consumed
and
produced
86
Butchery practices
Wilson did not emphasizethe butchery patterns of the various species except for animal
burials. The following discussionon the butchery practicesis largely my own assessmentof
the butchery notations found in the primary bonesrecordsand as such is quite detailed so as
for
interpretation
basis
illustrate
the
my
of the consumptionpractices.
to
Most body parts of the main domesticates were represented, which suggests that whole
butchered
butchered
fragmented
The
and
on
site.
and
presence of
carcasses were present
domesticates
bones
in
indicate
jaw
domestic
deposits
is
the
that
to
of
main
seen
skulls and
food
for
(Maltby
knife
1979:
38).
Butchery
by
cuts
they were used
marks were characterized
domesticates
indicate
Marks
the
three
on
main
and chop marks.
- cattle, sheep and pigs the cutting and chopping of particular joints of meat and the stripping of meat from the
bone.
For cattle, cut marks to the mandible and hyoid suggestthat the tongue and cheekmeat were
Distal
humerus
to
the
to
the
consumed.
chops
possibly
chops
and
and
proximal
removed
joint'
dismemberment
disarticulation
(for
the
to
at
elbow
see
similar
point
points
of
radius
Crabtree 1989:101-2; Wilson 1978:120). The pelvis is extensively butchered with chops
through the ilium and occasionallythe acetabulum.Of the hind limbs, the tibia is primarily
from
less
to
the
the
the
separate
tibia
meatier
possibly
section
meaty
of
choppedmid-shaft,
lower portion (Crabtree 1989:104 and for similar points of dismembermentseepage 102).
The metacarpalsand metatarsalsappearto have been disarticulated at the ankle and wrist
joints; a number of these bones were also chopped mid-shaft. Many of the long bones of
both cattle and sheep were chopped mid-shaft, which could result from the chopping of
bones into sizes appropriatefor cooking pots. This is also suggestedby the high number of
bowls found on the site indicating the serving of stews (D. C. Wilson 1994:59). Cuts to the
both
long
Some
the
to
the
of
cattle
and
calcaneus
sheep
point
and
of
skinning.
astragalus
bones and metacarpaland metatarsusof cattle and sheepwere split which might mark the
is
There
less
bone
butchery
significantly
marrow.
of
of the skulls of sheepthan
extraction
head
horn
Sheep
chopping
primarily
associated
and
marks
with
cut
removal.
with
and
cows
87
joints,
is
disarticulated
the
tibia
the
and
at
elbow
also the most extensively
appearsimilarly
butcheredpart of the hind limb with chops occurring at both endsrather than mid-shaft.
The butchery of pigs is difficult to interpret as in most cases no distinct pattern emerges.
The most commonly butchered bones are the scapula and humerus with some butchery
disarticulation
Possible
the
the
the
and
are
at
on
pelvis
ulna.
points
of
present
marks
There
the
the
the
the
ulna,
are also
of
pelvis,
and
at
of
scapula.
end
neck
proximal
have
been
for
both
that
pigs
might
reared
consumption and sacrifice, and that
suggestions
is
deposits'
below).
It
(see
`special
that
sacrifice
were
not
consumed
unclear
were
pigs
butchery
horses
the
of
minimal
whether
-a
couple of nicks and/or cuts to two mandibles
broken
long
bones,
butchery
to
the
and cuts
pelvis - represents actual
and upper vertebrae,
for consumption or for processing. Two dog bones were butchered - cuts to one mandible
122).
is
1978:
(see
Wilson
broken
the
this
significance of
also
ulna - although
not clear
and a
It is particularly interesting that of the four deer bones recovered (excluding four antlers)
in
The
butchery
deer
have
in
is
Iron
Age
the
the
them
chop marks.
of
three of
quite rare
in
(Hingley
England
Britain,
Scotland
with
other
contrast
parts
of
such
as
south of
1995:186; Grant 1981:205).
Meat yields
In addition to the identification of possible butchery patterns I also divided butchered and
into
bones
the
groups
which
approximate
amount of meat procured, with
non-butchered
highest
5.9)
Figure
`C'
lowest
(see
À'
the
the
yield
meat
and
group
representing
group
(after Uerpmann 1973; see also Lange 1996; Maltby 1979:7). These meat groups are of
laden
and perhaps are more representativeof current westernized meat
course culturally
in
Huelsbeck
1991;
Lyman
discussions
1987:
(see
The
269).
allocation of the
preferences
bones into the various groups may not reflect actual consumption, but rather, as has been
deposition
head
feet
bones
the
of
particular
animal
ritualized
parts, such as
and
noted above,
(Schuster Keswani 1994). Nonetheless,the grouping of the bones in this manner does
facilitate the detectionof body part patterning.
1It should be noted,however,that as the proximal headof the humerusrarely survives it is perhapsnot
distal
in
is
(Maltby
butchery
Chops
the
1989:
76).
to
the
primarily
part
that
of
proximal end
evidence
surprising
88
For the two main domesticates, cattle and sheep, we can see a similar distribution of body
from
highest
bones
(B);
the
the
percentage
of
moderate
meat-yielding
group
parts, with
higher
bones
from
fall
bones
have
À'.
Pig
do
a
slightly
of.
percentage
group
primarily
cattle
in the meat yielding groups, particularly group B and a high proportion of horse bones fall
into the non-meat-yielding group of bones. These findings complement a number of the
bones
distribution
The
the
various
species
made
on
above.
cattle
of
sheep
and
observations
indicate that sheep and cattle were used for other purposes as well as for consumption, for
horse
draft
for
hides
The
as
animals
or
example.
apportionment of pig and
their wool or
bones also suggests that pigs were likely reared for consumption and possibly for sacrifice
(see below) and that horses, whose consumption is less certain, quite possibly had other
purposes.
Small quantities of fresh water fish (eel and pike), domestic fowl and wild birds were
identified. Some of these species probably made a limited, though not necessarily incidental,
diet.
to
the
contribution
The plant remains for the Iron Age farm were carbonizedand found in just two contexts-a
base
large
hearth
The
late
Iron
Age
the
two
and
at
of
a
pit.
assemblagesare
open
possible
be
by
thought
to
a
and
variety
of
weed
cereals
species,
many of which are
characterized
The
breakdown
for
the
of
amounts
chaff.
small
very
of
chaff
cereals,weed and
edible, with
the two contextsis:
"
Pit 311 - 49% cereals, 1% chaff, 50% weeds (total number of carbonized items =
5,580; volume of soil processed= 50 litres thus 111.6items per litre).
"
Hearth 749 - 58% cereals,0% chaff, 42% weeds (total number of carbonizeditems
litres
50
2.74
items
137;
litre..
thus
of
soil
processed
=
volume
=
per
The determination of the site economy based on just two plant samples is highly
have
been
The
the
of
plant
samples
that
characteristics
suggest
might
grains
problematic.
dismemberment
joint.
be
however,
that
the
point
of
suggest
might
still
the
the
elbow
radius,
of
89
lack
for
large-scale
118)
(Jones
1985:
the
the
of
evidence
settlement
although
at
produced
inhabitants
by
1986b:
(Miles
45;
the
the
that
consumed
crops
were
possibly
storage,suggests
der
Veen
1991:
357).
van
seealso
The consideration of edible weeds as part of the diet is generally overlooked (Hansen
1991:53). However, the analysis of Iron Age bread recovered from the site of Glastonbury
hulled
barley,
fragments
(Helbaek
of
wheat,
wild
oat,
chess
and
common
orache
contained
1952:212). Helbaek argues that "The inclusion of wild oat and chess makes it doubtful
in
(1952:
212).
these
the
classify
species
as
weeds
ordinary
sense"
whether one should
Indeed, of the weeds identified to species at the late Iron Age settlement, up to 48% of them
The
types of edible weed species recovered at the settlement and which therefore
edible.
are
include:
bean,
fat
hen,
knotgrass
been
(both
have
celtic
collected
sorrel
and
of which
may
dyes),
black
bindweed
Reynolds
1986a;
(Jones
be
common
orache,
and
as
used
can
1995:308; Godwin 1975:479). That these species were carbonized and found in settlement
1995b).
further
(Reynolds
the
that
various
weed
species
suggests
were
consumed
contexts
Spelt wheat, which has good milling qualities, was the most common cereal, followed by
bread
in
barley;
also
wheat
was
prominent,
and
emmer was present small quantities.
six-row
It should be noted, however, that similar types of arguments made against considering bone
have
been
of
an
animal
population
also
as
representative
made against assuming
samples
that the flora samples represent the actual order of importance of these crops (G. Jones
1991:63; Helbaek 1952:214). A similar composition of cereals was found at the late Iron
Age site at the Ashville Trading Estate, Oxfordshire, although bread wheat was barely
(Jones
late
Iron
Age
1978:
103).
in
Sites
this
Barton
Court
Farm
the
at
site
as
such
present
beginnings
[bread
in
from
"mark
the
to
of
thought
wheat's]
rise
crop
prehistory
a
minor
are
historical
in
the
period" (Jones 1986b: 120). Bread wheat is considered
to prominence
intensive
both
labour
labour
intensive.
less
The
and
whereas
rye
oats
are
much
particularly
distribution of these crops is relatively even in the Iron Age as compared to the Roman
is
irregular.
Jones
distribution
(1989)
more
suggests that the apportionment of
period where
indication
be
used
as
an
three
could
these
of economic disparity, as opposed to an
crops
increase in economic potential during the Roman period as is more commonly thought
(1989: 133).
90
5.4.
5.4.1. The containersat the early Romanperiod settlement
Thepottery at the early Romanperiod settlement

Form, function and fabric
As with the late Iron Age phase, the containers recovered for the preparation and
in
drink
food
Roman
(although
the
and
early
of
period
were
strictly
ceramic
consumption
Roman
The
late
Iron
Age
below).
`Glass'
than
the
early
pottery
assemblage
was
smaller
see
features
71.6kg),
because
fewer
(41.6kg
to
there
at
compared
as
possibly
were
assemblage
but
likely
in
due
to manuring and
could
sherds
accumulate,
which
the settlement
more
deposition
below).
)
(see
5.10.
The
Figure
(see
their
assemblage
cultural practices governing
few
imports:
Roman-like
pottery
and
a
some
small numbers of amphorae,
consisted of
likely
bowls
decorated
All
these
samian
cups,
and
plain
and
plates.
would
of
mortaria, and
have been acquired through local regional markets such as Abingdon or Dorchester (Miles
7:
B6).
The
locally
1986b:
the
majority
of
assemblage
made
microfiche
was
up
of
made
et al.
bowls and jars over half of which (56%) were of the form and fabric found at the late Iron
Age settlement (see Figure 5.11.). The distinction between jars and bowls is less blurred
than in the earlier period, which suggests a more clearly defined specialization of vessel
form and possibly function. Interestingly, there was an increase in the percentage of jars as
late
Iron
Age
is
the
to
at
was
recovered
what
settlement, which
counter to
compared
Millett's (1979) observations at other southern Roman period sites (a similar increase in the
found
Roman
jars
in
England
(Evans
at
some
was
period
sites
of
northern
proportion
1993:98,103;
1995:61). Quite large cooking type jars were identified and there was an
increase in the number of narrow-necked jars. A number of small to medium sized jars had
lid seating, suggesting that they may have been used to store liquids or dry goods. Although
forms
Roman
the
the
present
at
pottery
early
earlier
settlement are probably
some of
be
(Miles
that
these
contexts
the
and
suggested
to
quantities
wares
used
continued
residual,
B4).
Again,
7:
from
different
1986:
calculations
those of
my
are
slightly
microfiche
et al.
Miles and his colleagues who separated the native pottery from Roman pottery in the
91
function
(Miles
1986:
C3-C4)
7:
pottery
et
al.
microfiche
although as stated
analysis of
be
Roman
to
the
continued
at
pottery
used
early
settlement.
native
above,
There were, however, a number of significant differences between the two assemblages.
There was a substantial reduction over time in the number of vessels with perforated bases,
in
decline
distilling,
discussion
(see
cheese
a
making,
mark
steaming
of
or
which could
There
for
below).
between
bones
the
types
was
certainly
a contrast
of vessels used
animal
liquids.
The
the consumption of
presence of samian cups and wine amphorae, together with
beakers and narrow-necked vessels, suggests a more varied drinking practice, possibly
involving communal passing of beer-filled beakers as well as individualized
drinking of
drinking
indicate
The
tall
that people either stood or sat
of
vessels
might
presence
wine.
drinking
indicate
in
that
the
and
cups
might
a reclining
consumers were
upright while
529;
Bedoyere
1989:
1981:
88).
(Evans
increase
in
locally
An
produced shallow
position
bowls and dishes and samian bowls and plates could imply a different emphasis on the
food
Roman
decorated
Plain
the
at
of
early
period settlement.
and
serving or presentation
2:
1
(see
Willis
for
1997
at
a
of
ratio
recovered
was
ware
opposite pattern on rural
samian
Two
decorated
Britain).
29)
in
Drag
(Drag
37
types
of
were
samian
northern
and
sites
bowls
large
diameters
by
both
20cm.
Using
the
with
are
well
out
over
set
criteria
recovered,
Miles et al. the percentage of tableware is 27%, which is still high (see below) but is lower
for
late
Iron Age settlement. The ambiguity in specifying
35%
the
the
suggested
than
late
Iron
Age
however,
(in
the
Booth
at
settlement,
ware
press), using
remains.
serving-type
different criteria - fabric and specialist wares rather than form - has compared the quantities
Roman
for
Valley
in
Upper
Thames
`fine
of
early
period
a
number
the
settlements
ware'
of
including Barton Court Farm. The early Roman period pottery assemblage at Barton Court
Farm was made up of 17.1% fine and specialist wares, which is quite high compared to
in
Valley
(Booth
in
the
press).
other settlements
Rim diameters
Although the sample is small, the range in size of the different types of vessels differs
Histograms
found
diameters
jars
the
that
earlier
settlement.
to
at
the
the
of
of
rim
slightly
(see Figure 5.12.) show peaks of small, medium and large jars, reflecting the presenceof
large
The
jars
diameters
bowls
the
cooking-type
pots.
and
rim
are no
of
narrow-necked
92
longer skewed and the range is narrower; there are two peaks of small and medium sized
bowls (see Figure 5.13.). The sizes of the possible tableware (see Figure 5.14.) vessels,
however, are varied. The significance of the early Roman period histograms, given the
dramatic
indicate
in
do
debatable,
but
is
the
to
they
sizes of
a
shift
appear
not
size,
sample
differences
in
jars
than
the
mentioned above.
notable
the vesselsother
drink
indications
food
Roman
there
that
the
was
At the early
and
period settlement
are
way
dishes)
individualized
bowls
both
(cups,
involved
and
serving
small
and
practices
consumed
(beakers
large
The
consumption
events
cooking
pots).
presence of
and
more communal
food
distinctive
Roman
does
the
type
adoption
of
a
signify
of
preparation vessel mortaria
function
is
have
it
that
a non-ceramic vessel may
possible
performed a similar
although
(Evans 1993: 103-4).
Theearly Romanperiod glass

No glassvesselswere identified in early Roman period contexts. However, a few fragments
in
identified
Roman
blue/green
non-early
glass
were
period contexts:their use at the
of early
but
is
settlement conceivable uncertain.
5.4.2. Other implementsat the early Romanperiod settlement
As with the late Iron Age settlement only a few implements were recovered that have an
was
food
drink.
bronze
A
and
recovered,
netting-needle
associationwith the consumptionof
fishing
bronze
bucket
iron
bucket
in
A
handle
the
of
nets.
preparation
and
an
usedpossibly
handle mount were identified; the buckets may have been used for preparing and/or
de
Solo
1993:
191).
(Gomez
liquids
transporting
93
5.4.3. Ingredientsat the early Roman period settlement
Theanimal bonesat the early Romanperiod settlement
The animal bone sample for the early Roman period was small. Figure 5.15 represents my
from
Wilson
S.
does
(see
N.
I.
P.,
differ
that
the
which
again
significantly
of
not
calculation of
Figure 5.16). The change in the proportions of the represented main species in the early
Roman period, especially the increase in cattle, is notable, but of questionable significance
because of the small size of the sample. The bones from cows were less abundant than the
bones from bulls and steers, which together with the reduction of vessels possibly used in
in
from
dairy
to
signify
a
shift
emphasis
may
products
meat consumption, at
making,
cheese
However,
in
Roman
the
settlement.
apparent
period
changes
animal population
the early
in
ditches
deposition
found
deposits
the
their
of
of
manner
were
may partially reflect
- most
differing
the various
the
socio-cultural
practices
concerning
of
as
preparation
well
as
herds
is
for
It
in
for
that
the
the
and
possible,
example,
consumption.
change
species
indicates
in
the ritual consumption practices of the
vessels
a
change
reduction of perforated
inhabitants which might be linked to the changes in vessels used for serving. These are
important considerations, especially if the abundance of a particular species, such as cattle,
is thought to indicate a `Romanized' diet. Grant, for instance, has suggested that an increase
in cattle might indicate an increase in the amount of meat consumed which could be an
indicator of "wealth and power [more than] a desire to ape Roman habits" (1989: 142).
Butcherypractices
The existence of specific butchery patterns at the early Roman settlement is also
There
however,
in
light
between
the
sample.
the
small
of
are,
some similarities
questionable
butchery
in
types
the
techniques used (i. e. knife cuts and chop
terms
of
of
two periods
The
joints
butchered
in
the
skulls
of
meats.
and
mandibles
of
similarly
cattle are
marks) and
The
tongue
the
that
cheek
meat
and
were
consumed.
major points of
suggesting
disarticulation are not as obvious, probably owing to the small sample size. There is some
distal cutting of the humerus, which may indicate a separationpoint, although there is less
94
butchery
hind
butchery
The
limbs
the
on
radius.
of
provide single examplesof
evidenceof
joint
knee
inconclusive
The
to
the
which
are
a
as
general
pattern.
astragalus
chops
cuts and
butchered
indicates
both
trimmed
consistently
and
which
skinning and
and calcaneumwere
One
difference
in
disarticulation.
butchery
Roman
the
the
major
marks
on
early
of
points
late
bones
bones
is
long
Iron
Age
to
the
that
the
compared
of
cattle
shafts
period cattle
bones were frequently trimmed and there appearsto have been less chopping up of the
bones,which could signify a tastefor steakand/or dry roast cooking, or possibly the stewing
in
large
joints
larger
cooking pots.
of
Interestingly, sheep appear to have been more intensively processed than at the late Iron Age
indications
identified
There
(not
tongue
are
of
and
cheek
at the earlier
removal
settlement.
incidence
higher
bone
for
the
the
of
splitting,
a
especially
radius - possibly
settlement) and
in
disarticulation
The
difficult
identify
(as
to
points of
of sheep are
extraction of marrow.
broken
the
examples
of
single
particular
of
often
separation
ends
points,
e.
g.
with
cattle)
humerus and radius, proximal cuts on one radius and broken ends of the femur and tibia.
The butchery of pigs is also inconclusive. The most commonly butchered bones were the
bones,
long
indications
breakage
butchery
in
the
the
the
and
of
many
of
of
are
scapula and
in
bones
than
the cases where long bones were chopped
specific
marks,
except
the
rather
has
Crader
bones
indicate
that
the
suggested
marks
on
shaft
of
a secondary
often
mid-shaft.
implies
dismemberment
This
butchery,
(1990:
709).
the
than
of
meat
removal
rather
stage of
joints
in
large
and was then broken up for consumption, or possibly
that pork was roasted
in
in
bones
the meat falling off the bone rather than
pots
the
resulting
that
were cooked
having to be removed (see Crader 1990:708).
The evidence is more conclusive that horses were at times consumed at the early Roman
by
butchery
illustrated
bones
marks
the
on
a
greater
as
variety
more
meaty
of
settlement,
butchery
The
humerus,
tibia.
and
radius,
the pelvis,
and possible consumption of horse in
been
have
to
site-specific during the Roman period. For example, horses
the region appears
Roman
Farmoor,
1979:
131)
butchered
Oxfordshire
(Wilson
the
settlement
period
at
at
were
but not at Roman Cirencester (Thawley 1982:211; see also Maltby 1979:62 who notes that
horse hide may have been used in the Roman period). No butchery marks were found on
dogs at the early Roman period settlement - perhaps, as Thawley (1982) and Harcourt
(1974) suggest, because dogs were being used as pets (although see Garnsey 1999:84-5 who
95
do
that
not show an aversion to consuming speciesthat were also
cites classical sources
butchery
but
displayed
bones
deer
Small
any
were recovered, none
numbersof
usedas pets).
deer
both
butchery
dogs
is
It
the
that
on
and
are
of
marks
absence
conceivable
marks.
for
hunting
longer
if
dogs
wild species.
used
were no
related
Meat yields
The groupings of bones according to their meat yield (Figure 5.17.) complement a number
horse
in
increase
the
the
of
proportions
of the observations made above, most noticeably
bones from group À' and the butchery of these types of bones. Pig bones are once again
bones
bones
have
by
the
and
sheep
cattle
a very similar
more
meaty
and
represented
distribution between meat yielding and non-meat yielding bones.
The differing butcherypracticesand the consumptionof particular speciesand body parts at

(roasting
distinctive
Roman
the early
cooking preferences
period settlement may signify
delicacies.
different
food
types
of
avoidanceand
rather than one pot cooking), and possibly
The changesin cooking practicesmight have contributed to a reduction in the percentageof
bowls found at the settlement.Small numbers of domestic fowl, a wild goose and a crow
fish
bones
few
No
from
Roman
in
contexts,
contexts.
were
recovered
early
were recovered a
fish
bronze
does
that
the
this
was
not
consumed
nettingmean
at
a
not
settlement;
although
have
in
inhabitants
found
the
that
the
main enclosure, which suggests
may
needle was
fished.
Unfortunately, no plant remains or processing implements were recovered from early

Romancontexts,althoughthe presenceof two possible granariessuggeststhat some form of
limited cereal processingcontinued. Additionally, a number of millstones found in a late
Roman well may actually be from the early Roman period (Spain 1986:microfiche 5:B6)
in
late
identified
Roman
have
been
the
querns
contexts
early
also
rotary
may
and some of
Roman, as querns are typically difficult to date becausethey are often reused (Gwilt and
Heslop 1995:40).
96
Summary of the Artefacts and Remains of Consumption for each Settlement
5.5.
Before we considerthe distribution of the artefactsand the remains at the two settlements,it
is necessaryto summarizesomeof salient points madeon the containers,the other
implementsand the ingredientsfor eachphase.
5.5.1. Late Iron Age settlement

Containers
Pottery
"
"
Majority of pottery was locally produced;somebutt beakerswere imported from the

southeast.
Assemblagedominatedby jars, followed by bowls, beakersand dishes.
"
Suggestions of high proportion of `tableware' type vessels (35% of assemblage).
"
Significant number of vessels with perforated bases possibly used for cheese
making, distilling, and steamingor as flowerpots or colanders.
Histogramsof the jars indicate a wide rangeof sizeswith a peak of 14 cm to 20 cm.
Histograms of bowls also indicate a wide range of sizes with a peak of 12 cm to
18cm.
Two sherdseach of samianware and amphorawere recovered.They are believed to
have been residual although it is conceivablethat Roman-style goodswere entering
the settlementtowardsthe end of its occupation.
"
"
"
Other implements
"
"
Fired clay slabs were found at the baseof a number of pits and Miles suggeststhat
they may representhearthsor parching ovens.
An iron knife was recoveredat the settlement.
Ingredients
Animal bones
"
Sheepand cattle bonesare both prominent, sheepvery slightly more so, followed by
bones.
horse
then
pig
"
The age and sex ratios of cattle and particularly sheep could indicate a propensity
towards dairying, or an intensity of meat consumption associated with feasting.
Butchery practices appear to indicate the butchery of particular joints of meat and the
bone
for
from
the
cattle, sheep and pigs.
stripping of meat
"
"
Three of the four deer bones identified (excluding antlers) have chop marks in
England.
the
of
south
consideredquite rare
97
"
"
"
Plants
Small numbers of horse bones were butchered, including the meaty.bones, which
dog
indicate
is
that
this
they
times
consumed
although
one
were at
speculative;
could
mandible had cut marks.
A, B, C, meat yield grouping of animal bones appearsto suggestmultiple uses for
(dairying
and meat consumption),that pigs were primarily a species
sheepand cattle
that was consumedand that horses were probably used for something other than
consumption,exceptpossibly on particular occasions.
Small numbersof fresh water fish, domesticand wild birds were identified.
from
base
hearth
Plant
the
two
of a
only
contexts: a
and at
remains were recovered
"
large pit.
dominated
by
Assemblages
cereals and weeds with very small amounts of chaff "
indicating
small-scale production.
possibly
identified.
number
of
edible
weeds
were
significant
"A
"
"
Spelt wheat, was the most commonly identified cereal followed by barley, bread
wheat, and emmerwheat.
Jonesconsiderspresenceof breadwheat in sizeableamountsunusual.
5.5.2. The early Romanperiod settlement
Containers
Pottery
"
Small number of imported wares including mortaria, amphorae and samian were
identified.
"
Fabrics that typify the early Roman period such as grey wares and Black-burnished
fabrics
but
identified
in
dominate
late
Iron
Age
the
evident
also
are
ware
assemblage
the early Roman period assemblage.
"
Assemblagedominated by jars and bowls - the distinction between the two vessel
types is lessblurred than in late Iron Age.
Increasein the proportion of jars and decreasein the proportion of bowls.
Reduction in proportion of vessels with perforated bases - possibly indicating a
decline in cheesemaking or alcohol production.
.
Large cooking type vesselsare evident and there is an increase in the numbers of
jars.
narrow-necked
Histogramsof the jars reveal peaksof small, medium and large vesselsreflecting the
large
jars.
type
and
cooking
narrow-necked
of
presence
Histograms of the bowls indicate a narrower range in sizes although two peaks in
larger
20
late
bowls
Iron
14cm
Age
identified
than
the
the
cm
slightly
and
at
size at
settlement.
"
"
"
"
"
Glass
blue/green
identified
fragments
few
Roman
in
of
early
glass
were
period
non-early
"A
is
of
which
uncertain.
contexts- significance
98
Other Implements
"A
bronze netting-needle was recovered.
"A
bronzebucket handle and an iron bucket handlemount were identified.
Ingredients
Animal bones
"
"
"
"
The bone samplewas quite small in comparisonto the late Iron Age settlement.
Cattle bones were the most commonly identified bone, followed by sheep bones,
horse and pig bones.
Cows less abundant than bulls and steers, which together with reduction in vessels
for
from
indicate
dairying
to meat
cheese
used
making,
a
shift
may
possibly
consumption.
Similar butchery patterns of cattle as late Iron Age settlement, except that shafts of
long bones are frequently trimmed and less evidence of chopping which could
for
dry
taste
steak,
roast cooking or the stewing of larger joints.
signify a
intensively
bone
butchered,
Sheep
that
tongue,
more
are
suggestions
cheeks and
"
marrow were consumed.
for
butchery
indicate
Minimal
that pigs were roasted.
evidence
of
pigs
which
could
"
butchered
butchery
bones
including
Horse
the
at
settlement,
on
a
marks
variety
of
"
bones
humerus.
the
the
such
as
pelvis
and
meaty
dogs
bones
deer
butchery
butchered
No
marks
on
"
- possibly used as pets, and no
from
the settlement.
were recovered
horse
bones
higher
indicates
The
to
their
of
according
grouping
a
meat yield
"
high
bones.
of
meat-yielding
proportion
Plants
"
No plant sampleswere recoveredfor the early Roman period settlement.
"
The presence of possible granaries, contemporary querns and millstones in non-early

Roman contexts could indicate that agriculture was a part of the settlement's
economy.
5.6.
The Distribution
of the Remains of Eating and Drinking
at Barton Court Farm
The remainder of this chapter will consider all of the artefacts and remains from the two
in
from
the
preceding
section
within their archaeological contexts.
settlementspresented
General observations on the re-contextualized material will be presentedand discussed.
Following this, a number of depositsthought to representspecial mealsand/or libations will
be highlighted. A brief summaryof my findings will be made for both the late Iron Age and
Roman
period settlements.
early
99
5.6.1. Distribution of the artefactsand remainsat the late Iron Age settlement
The animal bones and pottery at the late Iron Age settlement were found in all types of
ditches
highest
in
the
the
and
gullies
pits,
with
concentrations
occurring
contexts ditches
(see
Figure
5.18.
Figure
5.19.
).
Bowls
`tableware'
in
and
and
general were
enclosure
the most dominant vessel type recovered in the ditches although interestingly both beakers
barely
(see
Figure
house
5.20.
).
Pits
dishes
represented
are
and
gullies, on the other
and
hand, each have a similar distribution of pottery types, with jars and the jar/bowl variants
being the most abundant wares followed closely by bowls and beakers and dishes. Gullies
have a slightly higher proportion of `tableware' than pits.
A variance in vessel size was also found in the different contexts (see Figure 5.21.). Small
bowls and particularly small jars were only recoveredfrom the pits and house gullies and
ditches.
Two
is
One
from
that the
the
themselves.
possible
scenarios
enclosure
suggest
not
food
different
in
types
of
preparation
and
consumption
practices and around
patternsreflect
different types of featuresat the settlement.For instance, smaller cooking pots and bowls
indicate
individualized
house
the
site
may
eating and cooking or possibly the
around
dishes
in
Allen
(see
than
rather
a
one
smaller
pot
stew
of
similar conclusions
cooking
1990:42); larger bowls and jars at the boundariesand periphery of the settlement, could
indicate more communal consumption practices.This is also suggestedby the butchery of
dearth
there
was a
of choppedbones in the gullies. A secondscenario
the various species,as
is that the patterns reflect the disposal regime of pottery, which is proportional to and
feature
(although
distribution
for
below).
the
the
see
of
size
of animal remains
appropriate
The distribution of animal bones around the settlement reveals a number of interesting
bones
Cattle
).
5.19.
in
ditches,
(Figure
sheep
and
the
are
almost
equally
represented
patterns
both
dominate
dominates
in
and
gullies
pits
where
sheep
with
contrasts
and
cattle
which
pits
in the house gullies. Distinct patternswere also observedwith some of the other speciesof
Dog
horse
bones
in
found
the
settlement.
and
at
a number of pits
were
recovered
animals
but
(This
differs
in
together.
from Grant's observation at
the
not
gully,
northern
and
Danebury where there was a statistical associationof horse bones with dog bones (1991).)
Horse and dog bones were only found together in the ditches, and it was in the main
butchered
horse
dog
(and
Dog
the
the
that
of
and
occurred.
evidence
mandibles
enclosure
100
in
in
sheep
cattle
of
and
general)
were
especially
prevalent
particular contexts
mandibles
deposition
head
that
the
the
the
which
could
suggest
settlement,
of
was particularly
within
1995:
103).
(Hill
significant
In considering the distribution of the animal bones according to their meat yield (see Figure
5.22.), it appears that group `B' bones are the most frequent in the majority of contexts
in
ditches
bones
horse
for
is
`C'
the highest and in gullies where sheep
where
group
except
bones from group À' is overwhelmingly higher; butchery marks also appear exclusively on
these bones. It is particularly noteworthy that mandibles, which are a group `B' bone and
tend to survive particularly well (see Grant 1975:385), make up 50% of group `B' cattle
bones in gullies; (see below for possible significance of mandibles around the settlement) no
from
These
findings
the
recovered
were
gullies.
mandibles
sheep
add credence to the
distribution
in
the
that
the
the gullies is linked to preparing and
of
above
remains
suggestion
in
food
a particular way.
consuming
Miles suggestedthat there might have beenhouseholdsof differing statusresiding at the late
Iron Age settlement. The contextual associationsof the pottery and animal bones indicate
that this may well have been the case. The quantity of remains in the pits, for example,
have
the
the
throughout
that
site,
although
pits
may
contained `special deposits' (see
varied
below) and all animal burials were situated within the internal enclosure. The northern
house gully contained a wide variety of vessel forms including possible tableware. Of
bases
45
interest
(21
a
cache
of
was
were perforated) that was depositedat its
particular
terminal. This could representthe deliberate deposition of these remains in an area of the
be
below,
will
shown
as
which,
was of particular significance. The gully also
settlement,
higher
À'
bones
concentration
of
group
than the southerngully and the deposit
a
contained
have
been
15
could
a ritualized deposit (see Hill 1995:103; Wait
cattle mandibles
of
1985:132,137-8).
The southern gully, on the other hand, contained a smaller and
less
varied quantity of pottery, and fewer numbers of butchered animal
substantially
in
The
types
of
pottery
recovered
various the enclosure ditches were comparable,
remains.
identify
it
in
to
a
particular
the distribution of beakers.It was
possible
was
pattern
although
found that:
101
"
"
"
Beakers were not generally situated in the main enclosure ditch (only one was
recovered).
Therewas a higher concentrationof beakersin the northern gully (six) comparedto
the southerngully (one).
The pits that containedbeakerswere situatedwithin the internal ditch.
If the beakers were indeed used for drinking, their distribution in a particular area of the
is
discussion
(see
below).
suggestive
settlement
It is difficult to comment on the significance of the distribution of the plant remains as there
largest
The
found
base
below)
311
(see
two
the
samples.
assemblage
was
at
of
pit
only
are
internal
The
the
enclosure.
second assemblage, recovered from a possible open
within
hearth within the southern gully, was very small and contained fewer wild species and no
distribution
is
it
does
The
the
the
significance
of
of
although
plant
remains
unclear,
chaff.
for
in
(if
that
the
grains
were
possibly
prepared
consumption
not
consumed)
suggest
southern occupation area.
`Specialdeposits'at the late Iron Age settlement
While it is not possible to provide a detailed description of each deposit, a comment on a

further
deposits'
help
few
illustrate
`special
to
the significance of the contextual
will
select
is
deposits.
It
but
literally,
tempting
to
take
certain
of
course
of
each
context
we
association
been
deposits
had
òbserving'
that
constructed and reconstructed by many people. The
are
dividing of deposits into `special' and òrdinary'
is also tempting, but not wholly
has
been
illustrated
by
both
This
Hill
(1995;
Grant
1994)
(1991);
and
each
appropriate.
found that so-called ordinary refuse was structured, and as such incorporated into the rituals
life
(for
Barrett
1991: 1). As was seen above, certain
similar
conclusions
see
of everyday
types of pottery and animal species and body parts were concentrated in specific areas and
features. There are occasionally, however, some deposits that appear to have distinctive
found
in
The
denotation
other
contexts.
not
of `special' meals and/or events
associations
involving food and drink is controversial (see Wilson 1992,1996a, 1996b:77-79,85-88; and
Hill 1995:13-5 for excellent discussions of this issue) and what is considered `special' to
be
òrdinary'
quite
or taphonomic to another. I considered it significant
may
one observer
when there were specific groupings of particular types of artefacts and/or particular species
102
in
butchery
butchery
(or
body
type
of
practice
no
practice),
or
a
specific
parts,
and/or
deposits'
believe
`special
I
these
types
of
a number of
are present at the
particular contexts.
late Iron Age settlement(seeFigure 5.1 for locationsof specific contexts).
Pit 379 -a possible hearth- was situatedbetweenStructure 1 and the internal enclosure.
The pit was lined with clay slabs and it contained a single butt beaker, five butchered cattle
dog
mandible.
mandibles and one
Pit 338 located on the northern side of Structure 1, contained the non-butchered (nonlined
This
)
311,
to
of
an
articulated
remains
ewe.
was
pit
was
attached
which
pit
consumed?
large
and
remains
contained
plant
storage/cooking/beer preparation type
carbonized
with
vessels.
Pit 318 locatedjust south of the northern housegully, contained a variety of jars, and two
butt beakerfragments.The pit was lined with burnt stonesand charcoal,which have been
linked to feastingactivities (Gomezde Solo 1993:191).
Pit 415, was located northeast of the northern house gully. The pit was lined with clay slabs,
burnt limestone and charcoal, contained a couple of large jars, a variety of cordoned bowls
fore-limbs
butchered
beaker,
butchered
the
together
two
articulated
with
of
sheep,
and a
burnt
bones.
pig
cattle crania and some
Each of these pits was located within the occupation area -a practice common at most Iron
Age sites (Wait 1985: 138). It is especially noteworthy that these deposits, as well as a
human burial, the burial of a non-butchered piglet and a non-butchered dog, were all
internal
in
domestic
the
the
area
within
enclosure.
situated
Summary of the re-contextualized material at the late Iron Age settlement
The distribution of the pottery and animal remains at the late Iron Age settlement reveal a
Distinctions
interesting
between
the types of artefacts
patterns.
were
observed
number of
in
features:
found
ditches,
have
Bowls
the
three
types
that
of
gullies
remains
and
pits.
and
103
been identified as possible tableware were abundant in ditches, but dishes and beakers were
is
house
There
in
the
types
also
two
gullies
and
pits.
of wares were most prevalent
not; these
jars
features
found
in
between
distinction
types
the
three
the
the
small
of
sizes of
pots
a
different
in
ditches.
bowls
Two
larger
jars
in
bowls
the
gullies, particularly pits, and
and
and
types of consumption practices are suggestedby the distribution of the vessels, one
individualized.
The
distribution
the
other
more
of the animal remains was
and
communal
in
features.
bones
Cattle
distinguished
types
the
three
and
sheep
were almost equally
of
also
in
ditches,
but
bones
in
the
cattle
were
abundant
gullies whereas sheep
more
represented
bones were more prevalent in pits.
Distinctions between the artefacts and remains were also established in different areas of the
internal
boundary,
The
the
the
northern
area
of
settlement,
appears to
within
settlement.
have been an area of specialized eating and drinking rituals. Pits containing special deposits,
beakers
in
The
burials
this
the
accumulate
and
area
of
settlement.
northern and
animal
from
distinguished
house
each other,
gullies
and
associated
structures
are
similarly
southern
but
in
deposits
in
Ritualized
the
types
also
of cattle
of
artefacts
and
size,
remains.
not only
bases
beakers
distinguish
from
the southern
the
and
and
perforated
northern gully
mandibles
been
distinctions
households
has
It
that
these
to
of
suggested
are quite possibly related
gully.
different status or to specific areasof the settlement designated as spaces for specific eating
drinking
rituals.
and
5.6.2. Distribution of the artefacts and remains at the early Roman period settlement
The pottery and animal bones recovered from the early Roman settlement revealed quite
different distributions to those of the late Iron Age settlement. Most of the pottery was
in
Figure
5.23.
)
(see
the
the
the
enclosure
at
main
periphery
with a
of
site
situated
in
dense
The
the
the
accumulation
of
vessels
small
northeast
area
enclosure.
particularly
of
found
within the settlement seem to contain distinct pottery groups: shallow
numbers of pits
bowls and dishes in one pit, narrow necked jars and cooking-type pots in another, a beaker,
in
bowls
be
As
another.
and
can
seen, a number of the pits contained pottery
cup
a samian
that is thought to represent tableware (see Figure 5.23. and Figure 5.24.). It also appearsthat
below).
distinct
butchered
The
(see
and
groups
of
animal
species
animals
pits contained
104
in
foundation
hand,
is
the
the
the
on
other
slots
of
structure,
varied.
pottery recovered
Histograms of the rim diameters of bowls and jars (Figure 5.25.) indicate that the bowls
found in pits were larger, and of the few jars recovered from pits, there are peaks of small
located
jars.
Only
living
large
three
to
the
the
were
pits
situated
area,
and
one
close
and
bowls.
In
Iron
Age
the
the
to
type
contained
only
structure
serving
contrast
within
drinking
beakers,
associated
vessels
with
samian cups and amphorae - were
settlement,
located away from the living area of the site, in the internal ditches, main enclosure and in
for
beaker
in
foundation
(except
the
the
periphery
of
slot).
on
a
site
one
pits positioned
Narrow-necked jars, mortaria and large jars were also concentrated away from the living
dense
There
in
the
the
the
and
enclosures
at
pits
periphery
of
site.
was
an
especially
area
in
types
these
the northeast section of site.
pottery
of
accumulation
Unlike the late Iron Age settlement, animal remains were not generally deposited within the
domestic area of the site. The few pits (3) situated around the structure contained small
bones
butchery
slight
with
evidence
of
as compared to the other pits situated at
amounts of
the periphery of the site. The majority of the animal bones recovered at the settlement were
deposited away from the living area, in the enclosure ditches (see Figure 5.26.). As with the
deposition of pottery, there is a concentration of butchered remains in the northeast section
distribution
The
butchered
is
the
therefore similar to
the
of
enclosure.
main
remains
of
Wilson's (1986; 1989; 1996b) observations at other sites, in terms of there being a
distinction between the deposition of bones in domestic and non-domestic areas of the
both
large
bones
from each of the meat yielding
small
and
species,
although
and
settlement,
found
in
Figure
5.27)
domestic
(see
the
were
contexts
within
area of the site.
groups,
It also appears that specific animal species and specific body parts were distributed in
For
both
in
the
two
the northeast area of
example,
of
settlement.
areas
pits,
situated
specific
the settlement, contained primarily butchered cattle and horse bones (these two contexts will
be discussed further below). This is interesting becausethe pits associated with the structure
had high percentages of pig and more particularly sheep bones; only one non-butchered
horse radius and a horse tooth were recovered (in the late Iron Age house gullies, cattle
bones were the most commonly identified bone). Sheep bones with the highest meat yield
105
found
in
foundation
The
these
pits.
slots of the structurecontainedpredominantly
also
were
the headand foot debrisof sheep,cow, and pigs, and one dog skull, but no horses.
There were no plant remainsrecoveredat the early RomanSettlement.
'Special deposits' at the early Roman period settlement
The distinctive deposits recovered at the early Roman settlement were found at the
from
living
the
the
site
away
of
area and the main entrance to the settlement.
periphery
Pit 397 was situated just outside the northeast section of the main enclosure. The pit
bowl
and the articulated lower right leg of a mature bovine and
contained a small-necked
the lower right and left leg of a mature horse. The lower legs of cattle have been found in
apparently ritual contexts at the religious complex in Uley, Gloucestershire (Ellison
1980:306). The bones in this pit were complete and displayed similar cut marks, indicating
meat and/or skin removal.
Pit 366 -a possible cesspit - was situated at the northeast section of main enclosure. The
horse
together was also encountered in this pit, where a single
cattle
and
grouping of
butchered sheep's skull and a variety of largely complete butchered horse and cattle bones
including ten cattle mandibles, were recovered. The pottery in this pit was quite distinctive,
including a samian dish, cup and bowl, four large jars, mortaria, and a beaker, as well as a
bronze coin dating to AD 10-40 and a mid-first century brooch.
It is notable that in both pits the animal bones were butchered, which contrasts with the
late
Iron
burials
Age
the
at
settlement, which were not butchered. Grant (1989) has
animal
important
"An
that
aspect of the ritual of animal sacrifice in the classical world was
stated
the butchery and division of the carcass, which was specified in sacred laws" (1989: 146).
The association of horse and ox has also been found in votive deposits at early RomanoBritish villa sites (Perring 1989:286-7; see also Moore-Colyer 1994: 11).
106
It is also possible that the concentration of head and foot remains associated with the
deliberate
deposition
body
indicates
The
the
together.
these
of
parts
association of
structure
head and feet in ritual contexts is quite common at other Roman period sites (Scott
1991:117; see Barley 1995:77-8 who gives an ethnographic example of offerings of the
head, feet and tail that were intended to fool the gods into thinking the whole pig had been
sacrificed).
Summary of the re-contextualized material at the early Roman period settlement
The distribution of the artefacts and remains at the early Roman period settlement has
is
in
different
identified
late
This
Iron
Age
to
those
the
patterns
quite
at
settlement.
revealed
but
due
Roman
to
the
the
there are
configuration
changing
of
early
period settlement,
part
in
indications
The
that
there
the
a
was
change
of
consumption
majority
practices.
also
deposited
in
dense
ditches
the
were
and
remains
enclosure
with a particularly
artefacts
in
ditches.
There
however,
the
material
northeast
of
section
a number
of
were,
accumulation
deposits
between
found
in
ditches,
distinctions
features:
the
types
the
of
pits and the
main
of
foundation slots of the structure. A high proportion of the vessels identified as tableware
from
The
in
ditches
the
the
size
pits.
of
vessels
was also quite
pits and
were recovered
different; small bowls and narrow-necked and large jars were more abundant in ditches and
large bowls were more common in the pits. At the late Iron Age settlement, the ditches
bowls
large
The
in
bones
the various early
suggestive
of
communal
eating.
animal
contained
Roman period features were also distinctive; although ritualized animal deposits were
but
butchered
in
found
they
were
pits,
while at the late Iron Age settlement such
primarily
deposits did not display obvious signs of butchery. The ritualized deposition of specific
in
identified
late
Iron Age settlement also appears to have
the
at
pits
artefacts and remains
been present at the early Roman period settlement. Different traditions were, however,
defined in the pit deposits for the two settlements - large bowls and butchered animal
Roman
bowls
in
in
the
period
and
small
early
and
non-butchered
animal
remains
remains
the late Iron Age.
There was also a distinction betweenthe distribution of artefacts and remains in different
few
The
domestic
the
the
situated
settlement.
pits
within
of
area appearto contain
areas
107
discrete groups of pottery associated with serving on the one hand and food preparation on
the other; only small amounts of animal bones were recovered from these contexts. Deposits
with a possible ritual significance were situated either within the structure itself or
from
living
is
from
distribution
different
the
this
the
away
area;
completely
of these types of
late
Iron
Age
from
late
different
Also
Iron
the
Age settlement was
the
settlement.
at
remains
the distribution of drinking vessels - at the early Roman period settlement they were largely
from
domestic
in
direct
the
the
than
part
of
away
settlement
rather
association with
situated
the structures. Ritualized events at the early Roman period settlement appear to have been
focused either within the house or on the boundary of the settlement.
5.7.
Discussion of the distribution
of artefacts and remains at Barton Court Farm
What is most interesting about the deposits and distributions of animal remains and pottery
described above is the realization that the different configurations of the settlements reflect
directly on the social contexts of eating and drinking in the late Iron Age and early Roman
periods.
5.7.1. The Iron Age settlement
The compartmentalization of the late Iron Age site is of particular interest. Miles hinted at
different
two
the possibility of
status areas within the site. As has been shown, there was
indeed a contrast between the types of assemblagesrecovered in the northern and southern
The
further
distinguished by the presence of
areas.
northern
occupation
area
was
occupation
burial
Structure
human
1, an isolated piglet burial, and the
the
to
at
eastern
entrance
a
hearths.
Beakers
of
possible
were also concentrated in this area, as were the
concentration
deposits
distinctive
animal
and the single deposit of perforated bases in the house
more
divide,
This
which accentuatesthe northern area, has been identified at a
north-south
gully.
Bronze
Iron
Age
in
Age
Britain
(Parker Pearson 1996). The main
and
settlements
of
number
located
internal
the
was
outside
of
pits
enclosure. A number of these pits formed
collection
from
line
this
area
a rectangular enclosure - thought by the excavators to have
a
separating
been an animal pen - which had a human cremation at its western entrance and an infant
108
burial on the northeastcorner.Structure 2 is situatedoutside the main enclosureand while it
hasan associatedhearth,thereare no pits within its immediatevicinity.
It could be argued that there were two late Iron Age groups, possibly related or of differing
boundaries
by
but
inclusion
living
side
within
circumscribed
side
of
or exclusion
status,
(after Hingley 1990b). The central area of the site, which separatesthe two domestic areas,
linking
is
It
interesting
two
the
that
two
area
a
common
groups.
particularly
represent
may
types of burial, two types of consumption practices and a single animal enclosure were
found in this area. It is also possible that the apparent compartmentalization of the site
from
time
the
movement
of
a
single
group
over
sequential
one area of the site to
marks
different
for
different
the
the
types of activities, such
of
areas
of
settlement
use
or
another;
its
in
In
feasting.
the
the
case,
northern
section
either
of
settlement was, at various points
as
history, a focus for specific acts of commemoration.
The early Roman period settlement, on the other hand, had one central occupation area domestic
between
late
Iron
Age
in
the
the
and
southern
areas
northern
settlement.
of
situated
This is particularly interesting as there may have been a deliberate attempt to situate the new
different
in
Iron
Age
alignment,
the
an
a
completely
area
settlement
on
of
settlement
designated as common land (see Figure 5.1.). Of the few pits discovered, some appear to be
in
the settlement: at the entrance to the main enclosure, at the point
situated at select points
ditch
between
internal
the
the
the
the
meets
main
enclosure,
and
at
passageway
where
internal ditches. The contents of the pits imply a specific association with location, for
into
the entrance of the main enclosure, close to the granaries,
cuts
example, a pit which
inside
bowls.
jars,
fine
the
situated
a
pit
structure,
whereas
contains only
ware
contain only
This, together with the generally lower numbers of artefacts and remains relative to the Iron
Age assemblages,suggestsquite a different manner of deposition to that of the late Iron Age
deposits'
`special
However,
were still contained within pits, and while situated
settlement.
late
Iron
in
Age
the
the
the
they
as
at
settlement,
settlement,
the
situated
of
were
periphery
at
northern area of the settlement.
109
The two six-poster structures interpreted by the excavators as possible. above-ground
There
leading
into
from
is also a
the
the
site
south.
granaries provide a passageway
form
that
there
was
some
of gate at the entrance, although the dating of the
possibility
Nonetheless,
is
into
domestic
from
the
the
the
secure.
entrances
not
area
site
of
postholes
The
iron
door
keys
in the pit situated
the
two
are
conspicuous.
north
recovery
and
of
south
indicates
doors
be
locked.
The eastern double ditch is thought
that
the
could
structure
within
by the excavators to have either bounded a track way or marked the construction of a
The
bank.
defined.
The
the
perimeters
around
site
are,
either
way,
well
majority
substantial
deposited
`Special
the
the
and
the
pottery
remains
were
animal
at
periphery
of
site.
of
deposits' were situated away from the main entrance and the domestic area of the
distinctive
few
The
burial,
deposits,
including
infant
pottery
and
animal
an
settlement.
from
domestic
in
its
foundation
the
the
area
were
recovered
within
slots.
structure
or
within
Enclosures and entrances to the settlement appear to have dictated the movements of people
is
It
distinction
between
(behind
that
the
there
quite
possible
site.
was
a
private
approaching
domestic
doors)
The
and
practices
open
public
ceremonies.
remains of the public
closed
however,
distance
from the entrance to the
were
situated
events;
at
a
consumption
settlement.
5.8.
The early Roman period 'native' settlement at Barton Court Farm does not fit into
traditional models of native continuity. On the other hand, the concept of Romanization. as
it is currently understood (see Chapter 2), does not account for the dramatic changes found
This
Roman
the
settlement.
period
case study illustrates how native discontinuity
early
at
fundamental
be
to considerations of social change in Roman Britain. The butchery
may
practices at the two settlements revealed clear distinctions between the mores which
butchering
the
and presumably consumption of particular species and body parts at
governed
distinctions
do
`non-Roman'
times
`Roman'
which
not
always
to
conform
particular
and
categorization. The vessels associated with cooking and containment quite possibly reflect
the changes in approach to the consumption of animals. There was also a marked contrast
identified
drinking
between
time
with
vessels
and possibly serving. One of the points at
over
110
food
is
in
domain
diverge
the
two
the
of public consumptionof
pottery assemblages
which
in
is
both
in
importantly,
Most
drink.
the
the
consumption
contrast
practices mirrored
and
distributions of the vesselsand animal remainsthroughout the two settlementsand in the
way, the two settlementswereorganized.
I was particularly struck by the contrasting arrangement of the domestic areas at the two
been
Romano-British
has
It
to
that
the
configuration
of
suggested
served
villas
settlements.
(Samson
175;
S.
The
1990:
Scott
1994).
the
or
outsiders
approach of strangers
control
domestic structure at the early Roman period settlement was not a villa. However, there is a
definite sense that the restructuring and consolidation of the early Roman settlement
inhabitants.
its
Indications
than
the
of
outsiders
and
movement
rather
approach
emphasized
late
in
Iron
identified
Age
the
contrast
with
settlement, were
of public consumption events,
domestic
from
the
there
the
although
area
of
settlement,
remains an association with
away
Indeed,
the
the size of the early Roman period structure and
settlement.
the northern area of
been
hearths
have
times
that
suggests
at
cooking and eating, may
the absence of external
doors.
behind
closed
taking place
Changes in the configuration of settlements and in social practice are often associated with
be
The
Roman
that
the
occurred
changes
at
can
change.
settlement
early
periods of societal
The
in
establishment of a new settlement, the adoption of a
of
ways.
a number
viewed
Romanized
dining
the
the
of
embracing
on
an
emphasis
practices,
structure,
rectangular
in
beef;
drinking,
the
public
and
changes
rituals
of
eating
could mark a
and
consumption of
life
Romanized
from
Romans
the
the
the
of
a
recognition
new
and
under
past
separation
-a
`native'
in
because
it
looked
the
that
as
was
viewed
primarily
past
more native than
settlement,
Roman. An alternative explanation, one which I feel emphasizes the configuration of the new
incongruities
distribution,
the
the
of
excavated
their
of
remains
some
might
and
and
settlement
inhabitants
home
felt
that
the
the
to
the
the
as
evidence
changes
protect
of
need
view many
(Ardener 1993:11). The apparentconsolidation of `the house' within a single building suggests
that interaction between household members was also consolidated. The definition of public
locked
doors,
foot
bones
head
deposits
the
of
possibility
ritualized
and
of
spaces,
and private
in the foundation slots of the house, the continued use of beakers and other specialized Iron
Age wares, and the emphasis on settlement entrances and boundaries suggests that the
household
household
being
between
those
the
the
and
outside
re-negotiated.
was
relationship
111
The apparentadoptionof public Roman-likesymbolsmay, therefore,havebeenintegral to the

in
(see
Lyons
1996
Mattingly 1997b; Hingley
the
settlement
and
comments
of
protection
1997).
5.9.
Conclusion
Barton Court Farm helped to define how I view the Romano-British world. In Chapter 2, I
for
in
first
be
before
it
be
that
the
evidence
resistance
must
envisioned
can
seen
suggested
Notions
of Romanization and native continuity can both be observed
record.
archaeological
Roman
however,
it
is
identify
the
to
settlement;
early
equally
possible
at
native
discontinuity and an alternate experience of imperialism that affected the daily habits of the
inhabitants, but is not based on emulation. This particular perspective required the
integration of the artefacts and remains that encompass eating and drinking and the
it
late
Iron
Farm
At
Barton
Court
Age
their
archaeological
contexts.
was
examination of
hearthstones,
beakers
decorated
bowls,
the
that
of
remains
particular
and
meals
established
deposited
in
in
in
house
were
animal
remains
site
pits
and around one particular
articulated
the northern section of the settlement. At the reconfigured early Roman settlement, pits were
less common and those that contained the possible remains of meals and / or libations were
from
the
the
the single house site, or within the
periphery
of
either
at
settlement
away
situated
fundamental
itself.
This
interacted
in
inhabitants
the
a
the
represents
change
way
structure
boundaries
that
suggests
and
around the settlement had become more
with each other
important than the boundaries within the settlement. At Roughground Farm, the second case
it
be
in
thesis,
this
will
possible to observe what I see as a completely different reaction
study
to the Roman presence.
112
\\J1,
RK.
&xir
awe
0W
.G
O
O
i St+naI
/7
itirIh
LIA
10
oa
loom
Figure 5.1 Late Iron Age and early Roman settlements at Barton Court Farm
(after Miles 1986b, drawn by B. Meadows)
111
Barton Court Farm late Iron Age containers
Figure 5.2 Late Iron Age potter\ forms
Figure
Histogram of late Iron Argetableware

-5.5
LIA - tableware
LIA - pots
n=306
20
50
40
30
20
10
0
10
5
vQ
Cwoo
arn
0DO
c3m
co
AONW
NNW
Cj
Rim diameter
Histograms of the rim diameters of late Iron Age bo%%Isand jars
Figures 5.3 and 5
LIA - bowls
2E5
LIA -jars
N 15
30
25
20
15
10
N
t0
10
5
E
z0
cv
co
ro
cD
Nxro
CO
Rim diameter
Figure 5.6 Late Iron Age potter
fabrics
120
100
0'
o:
03
60
04
40
20
ii
(C
N
C)
Rim diameter
LIA - pottery fabrics

n=301
80
O
NN
00
Cl)
V0
V
LC
114
Barton Court Farm late Iron Age ingredients
Figure '. )' N A.S.P. tAleadovs)
Figure 5. " N. I. S.P. (Wilson)
LIA - animal bones

=858
LIA
60
%
40 1]
- animals
= 27
60
40
20
20
0
cow
horse
sheep
dog
pig
oao a_o
p
GJ p`, ezQ Qaez o'
yr
Figure 5.9 Late Iron Ave animal bone grouping according to meat yield
LIA - animal bone grouping

70
60
50
40
30
20
QA
B
cc
10
0
cow
sheep
horse
pg
I15
Barton Court Farm early Roman period containers
Figure 5.10 Early Roman period pottery forms
ER - pottery forms
n=155
50
40
30
20
10
OOONO2
Oy3
N0O
Figure 5.1/
Early Roman period potter\ fabrics
ER - pottery fabrics
120
00
C
02
80
O3
50
s0
L
O BB
Im
sam,an
-O
ease-
:..v
ar
_. '
.....
Histogrammof early Roman period jars and how is
Figure 5.12 and Figure 5l,
ER- bowls
ER -jars
v 10
in
0
15
10
6
1-k
-fL-fl
EQ
3
ZN
Co
U1
NNWW
AO
'J1
CO
?C
E2
Z
Z0
N
Rim diameter
CO
VO
-NNMMV
CO
Rim diameter
Co
VO
LU
116
Barton Court Farm early Roman period containers and ingredients cont.
Figure 5.14 Histogram of early Roman period tableware
ER - tableware
6
d5
v4
3
z
-0
E
71m
z
0
N
VO
co
(D
00
Rim diameter
Figure 5.15 N. I. S.P. (Meadows)
Fi iure 5.1() \. I. S. P. (Wilson)
ER - animal bones
n=393
60
50
40
30
00
20
o,
Da
10
ER - animal bones
n=326
C)
OOCp
oe
a0
20
SN-G
0
(D
-0
cow
horse
Figure 5.1 " Early Roman period animal bone groups accordinsi to meat Meld
ER - animal bone grouping

80
60
CA
40
OB
Cc
20
0cow
sheep
horse
pig
sheep
pig
dog
117
of containers and inuredients at late Iron .A_e Barton Court Farn
Distribution
Figure 5.18 Distribution of late fron :age potter
LIA - pottery contexts

60
50
40
%30
20
10
0
Cl c tcnes
Octs
'
O guiles
4
0
(D
fs
Figure
5 19
o1 late
Distrihution
Iron
Ae
animal
1), me,
LIA - animal contexts

60
Qd tcnes
O oats
50
40
% 30
20
10
ill ii=mALL
cow
norse
sneer
Figure 5.20 Distribution
EL
pig
60
40
fl
0
ditches
dog
of late Iron Age tableware
LIA - tableware
n=109
20
deer
pits
gullies
118
Distribution of containers at late Iron Age Barton Court Farm cont.
Figure 5.2 1 Histograms of late fron Age bowls and jars b feature
LIA ditches - bowls
LIA ditches - jars
d 20
ai g
15
U7
10
5
Z
Qm
0
ovb
(N
(N
cz
ca
cV
tic
...,
CO
O
N
Rim diameter
(O
(, i
(N
M)
Co
O
U')
Rim diameter
LIA pits - bowls
LIA pits - jars

N
10
N
f0
U
6
4
2
ZOQQQ
(V
cc
co
NN..
c
..,
00
C')
z UUUoo
NNVODNC
NNM
Rim diameter
00
Rim diameter
LIA gullies - bowls
LIA gullies
Q, 10
g
jars
-
dg
O4
2n
ZO
rv
co
ccvc v
N)
Rim diameter
(P
co
VO
to
Rim diameter
11O
Distribution of ingredients at late fron Age Barton Court Farm cont.
Figure 5.22 Late Iron Age bone groups by feature (did not include species with less than 10 bones)
LIA - pits
LIA - ditches
LIA - gullies
60
80 70
60
50
f
30
11
50
QA
20
10
.
e eeQ ,
r
5,'
70 60 t
5C
40
0B
30
0C
20
40
3C
3C
11
2C
QA
'0
0
cow
sneep
0C
OA
10
0
cow
sheep
OB
OC
120
of containers at early Roman period Barton Court Farm
Distribution
Figure ?.23 Distribution of Roman pottery
ER - pottery
Figure 5.24 Distribution of tableware
ER - tableware
n=42
contexts
80
Cl ditches
60
60
50
40
% 30
20
10
0
Cl pits
%40
0 structure
20
0
QQdCy
nn
-_
ONNC
nNC
N7
67
(D
Figure 5.25 Histograms of jars and bowls bv feature
jars
-
ER ditches
ER ditches
bowls
-
12
10
8
N
d
U
0
6
4
2
zo
z0
N
co
'r
O
NN
(N
((
cD
CO
..
<v
00
VC
'-
Rim diameter
ER pits - jars
N
Cl)
(N
cc
CV
Rim
diameter
CO
MV
V0
Lr)
ER pits - bowls
2.5
N8
61
2
5
d1
E 0.5
E2
lul
Zo
N
Co
'V
Co
Co
VO
-0
Co
RO
(0
NN
Rim diameter
N
c')
Rim diameter
Co
C)
VO
V
tf)
121
Distribution
of ingredients at earl% Roman period Barton Court Farm
Figure 5.26 Distribution of early Roman period animal bones
ER - animal contexts
50
40
0 p,tS
Cl ditches
30
20
structure
10
n-m
0
cow
norse
sheep
pig
coo
Figure 5.27 Earls Roman period animal hone groups b,, feature (did not include species with less than 10
bone)
ER - ditches
ER - pits
80 OA
60
OB
cc
40
80
70
60
50
40
30
OA
B
0c
20
C
0
cow
Horse sneeo
pig
cow
horse
sheen
122
Chapter 6
Table of Contents for Roughground Farm

123
124
Introduction
6.1.
The Excavation
6.2.
The site
6.2.1.
The site report and site archive
6.2.2.
The Archaeology of Food and Drink Consumption at Rouphground Farm
6.3.
The containers
6.3.1.
The containers at the early Roman period settlement
6.3.2.
The early Roman period pottery
Vessel form and function
Fabrics -'fine wares'
Fabrics -'coarse wares'
Rim diameters
The early Roman period glass
Other implements
6.3.3.
The ingredients
6.3.4.
The specialists 'reports - the animal remains
The specialists' reports - the plant remains
The ingredients at the early Roman period settlement
6.3.5.
The animal bones at the early Romanperiod settlement
Butchery practices
127
Meat yields
6.4.
6.4.1.
Non-botanical evidence of agriculture and/or grain processing

Summaryof theArtefactsandRemainsof Consumption
at Eativ RomanPeriodRout
Farm
vund
137
The Containers
Pottery
Glass
6.4.2.
6.4.3.
6.5.
6.5.1.
6.5.2.
6.5.3.
6.5.4.
6.6.
6.7.
6.8.
Other implements
Ingredients
Animal bones
Non-botanical evidence of agriculture
The Distribution of the Artefacts and Remainsof Consumptionat RougharoundFarm
The main area of occupation
Pre-villa and environs
139
`Specialdeposits'
Summaryof the re-contextualizedmaterial at the early Roman RoughgroundFarm
Discussion of the distribution of artefacts and remains at Roughground Farm
The Social Contexts of Imperialism at Rouaharound Farm
Conclusion
Figures 6.1- 6.9
145
145
146
148
123
Chapter 6
Roughground Farm
6.1.
Introduction
Roughground Farm is situated on the second gravel terrace between the rivers Thames and
Leach and has a similar geography to that of Barton Court Farm. The site is thought to have
been occupied in some capacity during the Neolithic and in the late Bronze Age although clear
Iron
The
Iron
Age.
Age
is
the
of
early
settlement,
as
early
apparent
only
settlement
lived
burials,
ditches.
by
boundary
track
and
ways,
pits,
short
post-holes
and
was
characterized
during
A
is
late
Iron
Age
the
middle and
minimal.
soon abandoned; evidence of occupation
into
in
inhabited
AD
the
the
early
mid-first
century
and
established
was
was
settlement
new
in
Roman
Towards
the
the
the
the
midof
occupation
of
end
early
settlement,
century.
second
just
Roman
but
limits
the
the
south
early
of,
within
of,
second century, a villa was constructed
Roman
Thereafter
the
through
the
the
to
occupied
continuously
site was
end of
settlement.
period.
Roughground Farm is a complicated site with an excavation history that started in 1957 and
Roman
There
in
1957
1958.
Most
in
1990.
the
settlement
early
was
excavated
of
and
ended
interpretation
before
it
destroyed
the
the
of
site
with
as
were
parts of
are a number of problems
does
ideas
food
The
also
not
combine
and
material
about
perfectly with
excavated
excavation.
drink consumption in that there is a paucity of butchery notations and botanical samples.
Nevertheless, Roughground Farm is a fascinating site and suitable for inclusion in this
by
because
the
the
work,
site
archive
and
was
compiled
post-excavation
site
report
research
O.A. U.. More importantly, as I have stated previously, the study of eating and drinking
diversity
imperialism
is
the
the
to
of
on native
establishing
effects of
a means
practices
Valley.
Frankly,
become
(and
Upper
Thames
in
the
the
the
as
will
site
apparent,
settlements
124
incongruities
despite
interesting
discount
its
the
to
too
of
archaeological
was
site archive)
recording.
This chapter will commence with a summary of the excavation of, and configuration of, the
Roughground
Farm
Following
Roman
this, the social practices which
at
settlement
early
drinking
be
First,
through an examination of the
and
eating
around
will
explored.
revolve
implements and ingredients used in both the preparation and consumption of food and
drink; and second, through the consideration of the various processes involved in the
distribution of these remains on the site. The chapter will conclude with a discussion on my
interpretation of the excavated material, within the context of imperialism.
6.2.
The Excavation
Margaret Jones first excavated the site from 1957-65 and small-scale excavations were
by
O.
U.
in
A.
1981-2
in
1990.
The
the
and
excavation is recognized as "one of the
undertaken
first landscape studies undertaken in this country" (Allen et al. 1993:xxi). The site was first
identified through crop marks and through the discovery of artefacts and features by local
face
Excavation
began
in 1957 after a watching brief
the
gravel
was
exposed.
people when
during gravel extraction established the importance of the site. The excavation team was too
small to clear the whole of the site, although the removal of topsoil by gravel company
features
the
most
of
visible; this enabled their individual excavation. The
scrapers rendered
labour force was largely unskilled in archaeology, which necessitated an accessible
excavation strategy; complicated stratigraphy was consequently excavated in spits. Storage
it
highly
decided
finds
in 1961 that non-stratified pottery (see
problematic
and
was
was
of
below), fragmented animal bone and building material of no ìntrinsic interest' was to be
discarded.
Tim Allen of the O.A. U. managed and co-ordinated the post-excavation work of the early
1957-65 excavations as well as the smaller excavations by the O.A. U. in 1981-2 and 1990.
He comments on the unique task of integrating different approaches to the past and cautions
that "The partial nature of the excavation and the character and the variety of the excavation
have
left
incomplete
techniques
used
an
and recording
picture of the site" (Allen et al.
125
1993:5). I can attest to the challenge of establishing the nature of the early Roman period
is
by
its
However,
Roughground
Farm
history,
although
complicated
excavation
settlement.
I would suggest that this complication is one of degree, not kind, because most excavations
inherently
of
collecting
all
methods
and
recording
excavated
and
material
are
are partial
biased.
6.2.1. The site(seeFigure 6.1.)

The early Roman settlement is characterized by an oval house enclosure with associated pits
by
The
house
a
rectangular
enclosure.
enclosure has been re-cut many times, and it
surrounded
identify
to
a sequenceof enclosure, although the southwestern entrance was
was not possible
is
for
It
each
re-cut.
also possible that some of the enclosure ditches held a fence or
respected
inside
A
ditches
has
been identified and
the
short
series
of
post-holes
situated
one
of
slot.
wall
they may have formed part of a house. Domestic material, particularly parts of clay ovens
in
house
the
terminals
the
of
enclosure, helped to differentiate the suggestedhouse
recovered
features
from
at the settlement. Outside the entrance of the house site, a series of
other
site
contemporary pits were recovered. They are thought to have had a variety of possible functions
from
cooking to storage -a
ranging
number of the pits contained burnt stones as well as
indicate
particular cooking events. A rectangular ànimal pen'
charcoal and ash which might
has also been identified within the domestic area. The 40 metre square compound ditch which
features
is
believed
these
to have been a slightly later addition to the settlement
surrounds
òpen'
first
the
the
until
end
of
was
probably
century AD. A large areajust southwest of
which
the main occupation area, quarried away before excavation, provides an unknown quality to
the early Roman period settlement. Unlike at Barton Court Farm, there are numerous
indications that there was settlement activity beyond the ditched compound. A series of
gullies, pits and postholes west and south of the main occupation area have been identified as
Roman
the
early
with
period settlement.
contemporary
A number of disassociatedcontexts have also been assigned to this period (see Figure 6.1.).
Possible stock enclosures beneath the second century villa building had few finds but the
pottery was dated to the first century AD. It is possible that there was a second early Roman
occupation area present at the site. A series of contemporary but disconnected pits, hearths,
126
has
have
been
identified
Allen
if
theorized
that
there was a
and
shallow
gullies
and
ovens
higher
because
it
have
been
Roman
of
status,
settlement, may
of the presence of
second early
fine wares (Allen et al. 1993:181). In an area quite isolated from the main occupation area is
burial
former
Roman
that
to
situated
early
cremation
was
a
marked
next
early
an elaborately
Iron Age circular structure.
The organization of the early Roman settlement is similar to other early Roman settlements in
the Valley with rectangular enclosures surrounding house sites, animal pens, and irregularThe
Roman
house
Roughground
the
or
circular
ovoid
structure
of
pits.
early
site
at
shaped
Farm, however, differs from that of neighbouring early Roman period Claydon Pike, which
had a rectangular structure similar to that found at early Roman period Barton Court Farm
(Miles and Palmer 1983a; see also Chapter 8 on Claydon Pike). The site has been described as
"entirely native in character" (Allen et al. 1993:179).
6.2.2. The site reportand site archive
Much of the post-excavation work was compiled or at least reviewed and co-ordinated by the
O.A. U. with the co-operation of the original director, Margaret Jones. The site report was
in
Thames
Valley
1993
Landscapes series. The report comes with a
the
as
part
of
published
in
format
Barton
Court
Farm. The main text focuses on
to
that
similar
element,
of
microfiche
the summary of the site stratigraphy and finds and discussion of the site as it pertains to other
Valley
in
Upper
Thames
the
with much of the detail of the excavation and finds produced
sites
Barton
Court
Unlike
Farm, the finds are presented according to their material
on microfiche.
function.
The
bringing
than
their
together of two different recording and
composition, rather
fifteen-year
hiatus
with
a
was a challenge for the publication team. The
excavation strategies
followed
be
in
the
the site archive, which contains a fascinating
can
challenge
extent of
forty
years. As Allen states in his discussion of the format for the
correspondencespanning
publication:
be
between
full
had
description, often involving lengthy
to
made
choice
a
...
discussion of the doubtful validity of particular pieces of evidence, or more
based
description,
around whichever interpretative framework best fit
summary
the available data (Allen et al. 1993:5).
127
The second option was used with the proviso that full descriptionscould be found in the
1993:
5).
(Allen
al.
et
archive
The site report and accompanying microfiche are an invaluable resource for those studying the
Roman period. I do have one comment on the way some of the material was presented,
however. I found it at times difficult to find an early Roman period settlement in the
discussion and presentation of the excavated material. Many of the early Roman finds and
found
`villa
`pre/early
the
categories
and
within
environs'
or
villa', categories
remains were
for
(see,
iron
the
throughout
report
example,
copper
alloy
objects,
objects and animal
used
bone). This may in part be due to the small numbers of finds recovered from the early Roman
fired
in
however,
large
the
there
clay
report
where
was
a
amount of material
settlement;
Roman
(rather
from
by
the
the
settlement,
early
report was ordered
context number
recovered
than by phase) under the general category of `Romano-British'. I should add that the early
Roman occupation is specified in both the description of the excavation of the settlement and
in the general discussion element at the end of the site report.
In terms of archival research, the Roughground Farm archive, which is stored at the
Ashmolean Museum, was the most interesting archive consulted for this study. As a
follow
it
to
the thought processes of some of the archaeological
a
privilege
researcher, was
luminaries of the past, through scribbled thoughts, queries and a wealth of correspondence.
For my analysis it was necessaryto enter into a computer data base both the pottery and bone
primary records.
6.3.
The Archaeology of Food and Drink Consumption at Roughground Farm
As with the previous case study, this section will commence with a discussion on the nature of
the remains of eating and drinking, to include the methodologies and conclusions of the
data.
between
The
distinction
the
the work
my
own
well
as
analysis
of
as
various specialists,
be
beginning
the
the
own
and
my
analysis
will
specialists
outlined
at
of the relevant
various
of
follows:
The
'containers'are
grouped
as
subsections.
which include
128
includes
finally
òther
implements'
metal,
stone
and
clay
which
artefacts
and
potteryand glass,
ìngredients' which include animal remains and plants.
The pottery from the 1957-65 excavations and the trenches excavated in 1981 and 1982 was
Green,
from
Sarah
by
by
1990
Paul
the
the
and
pottery
excavation
was
analysed
analysed
Booth. Most of the early Roman pottery was recovered during the earliest excavation.
Although all of the pottery from the 1957-59 excavations was initially kept, many of the
body
in
later
the
sherds'
summarized
coarse-ware
were
pottery
notebooks
and
unstratified
discarded. A complete catalogue of the pottery was not therefore possible and Green decided
instead
her
base
to
quantification,
concentrating
analysis
on
on establishing a record and
not
forms
fabrics
(1993:
113-4).
does
As
this
and
of
pottery
study
not emphasize noncatalogue
discard
did
directly
the
practices
not
pottery,
affect the analysis and quantification of
stratified
the early Roman pottery as describedbelow.
Much of what appearsin the primary pot records can be found on microfiche (fiche 3, frame
16-91) although there was-more detail, including rim diameters, on the records themselves.
When there was an incongruity between the microfiche pottery record and the primary pottery
in
I
the primary record.
was
written
what
records, used
6.3.2. The containers at the early Roman period settlement
Theearly Romanperiod pottery

At the early Roman settlement, there was a wide variety of vessel types, both local forms and
forms brought in from further afield. The range in pottery includes variously shapedjars and
dishes, bowls, beakers, cups, flagons and a single mortarium (see Figure 6.2.). A number of
129
the bowls, dishes and particularly cups were samian. No amphorae were recovered from
both
20
(an
Dressel
47
Pelichet
(a
olive
oil
amphora)
and
although
wine
contexts,
stratified
be
identified
in
the
settlement,
were
with
residual
contemporary
amphora), which could
(from
have
been
The
Verulamium
therefore
the
may
used.
mortarium
region)
contexts - and
hand,
in
Roman
the
context,
on
other
was probably not a prominent vessel
recovered an early
type as the majority of mortaria recovered at the site were from the third and fourth centuries.
Small numbers of other non-local wares were identified, such as roughcast beakers, Blackburnished wares (BB 1) from Dorset and vessels from Wiltshire and Herefordshire. Most of the
locally
form
fabric
have
been
identified
made
and
wares
of
similar
and
was
at
pottery
Cirencester and Claydon Pike (Green 1993:134). The presence of a variety of vessel forms
implies that a number of methods were used in preparing and serving food and drink and quite
diet
(Brown
in
1997:
100).
Included
local
the
that
the
was
equally
varied
wares are a
possibly
including
fine
imitation
`Gallo-Belgic'
beakers,
bowls,
wares
and
wares,
namely
series of
dishes. Imitation wheel-thrown Black-burnished ware is also quite prominent.
Form and function

Figure 6.2. shows that of the identified forms, jars are by far the most commonly identified
by
bowls,
dishes,
beakers,
in
followed
flagons
then
cups,
with
and mortaria present
vessel,
Identified
include
beakers,
bowls,
dishes,
type
to
serving
vessels
numbers.
small
cups,
very
identified
21%
flagons
is
7%
to
the
total
of
amount
assemblage,
samian ware.
of
which
and
The distribution of plain versus decorated samian ware was similar to that found at Barton
Court Farm (contra Willis 1997). The ratio of plain to decorated ware at early Roman
Roughground Farm was 3: 1 - the only decorated saurian ware was bowls (Drag 29 and Drag
37). Jar/bowl types were not identified at Roughground Farm; it is likely that this is partly
becausethere appears to have been more of a distinction between the two forms in the early
Roman period but also becauseGreen's bowl types tend to be shallow (see further discussion
in Chapter 7 section 7.3.1.). Wide-mouth jars are the most common type of jar, though
decorated high-shouldered jars with foot-ring bases are prominent. Narrow-necked jars,
handles
jars
and the presenceof a variety of lids made out of a number of
with
straight walled
different fabrics suggestthat many of the jars had specific functions. Burnt residues were only
identified on coarsely temperedjars, and although two pieces of samian ware were burnt, it is
' It appears as if all of the Roman-style wares were kept -a common practice in Romano-British studies.
130
from
burnt
burnt
the
cooking,
or
residues
were
were
accidentallyor even
not clear whether
deliberately.
fine and specialist wares*

black burnished wares
oxidised wares
brown wares
shell tempered wares
limestone tempered wares
ro tempered wares
reduced wares
misc.
n=1072
38
95
64
63
48
75
228
447
8
%
4
9
6
6
4
7
21
42
1
Table6.1. Early Romanperiodpotteryfabrics
Fabrics-'fine wares'
Paul Booth (in press; see also Booth 1991) has assembleda list of fine ware fabrics and
in
for
Roman
Upper
Thames
include
Valley.
These
the
the
period
amphorae,
wares
specialist
Coarse
glazed
and
colour-coated
other
and
wares.
samian
wares such as 'Belgic
mortaria,
type' wares, oxidized and reduced coarse wares, Black-burnished ware and calcareous wares
high
indicator
less
Booth
òther
together
a
status
of
and
are
grouped
wares'.
as
are considered
high
fine
that
with
proportions
of
settlements
and specialistwares are possibly of
suggests
higherstatusbecause:
these wares often represent non-essential vessel types the acquisition of
...
be
determined
in
a
seen
as
matter
of
choice,
may
which
part by ability to pay
for them (economic status) and by the desire to express a particular level of
(relative) sophistication (perhapsrelated to social status) (Booth in press).
Booth calculatedthe percentagesof thesewares(basedon number of sherds)for the pottery

Roughground
1990
found
in
Farm.
during
èarly'
It
the
the
excavation
at
was
unearthed
Roman period (which actually refers mostly to the second and third centuries) that 7.5% of the
fine
(Booth
in
identified
and
specialist
ware
was
press). According to my calculations
pottery
(seeTable 6.1.), 4% of the waresat the early Romansettlementwas `fine' or specialistware,
131
higher,
for
is
the
the
as
number
of
someof
slightly
sherds
probably
percentage
although, real
found
in
i.
the primary record.
the
were
not
of
saurian
wares,
some
thesewares, e.
It is also conceivable that some of the òther wares', such as Black-burnished ware (BB1) from
Dorset and the imitation Belgic wares, were priced at a premium compared to other coarse
it
is
below).
Further,
`coarse'
(although
that
possible
see
non-local
wares were more
wares
local
`fine'
(Keith
Branigan
).
Booth's
than
comm.
ware
pers.
observations are not
expensive
based on vessel form, which may also reflect on a group's standing in the community
has
high
if
Coarse
type
a
settlement
percentage
a
of
serving
vessels.
ware serving
particularly
type vessels found at the early Roman settlement, for example, may have been prestigious held
burial
jar
that
the
cremation
the grey ware
was a coarse ware vessel. Clearly, as Booth
linking
is
far
from
the
of
particular
pots
with
settlement
status
straightforward
acknowledges,
is
there
usually an archaeological or ethnographical example that contradicts an
and
imports
decorated
(for
and
of
status
or
so
on
numbers
pots
and
status
and
with
association
Pluciennik
1997:
48-9;
Woolf
).
1993:
211;
Chapter
3
3.4.3.
discussion
see
and
section
similar
The emphasis on status and fine wares has also left so-called coarse wares, which comprise the
in
ignores
the
the
the
recovered
at
archaeological
pottery
settlements,
out
of
cold,
and
majority
food
drink
(Miller
associated
with
preparing
social
practices
and
storing
and
and
cultural
1985; Bedoyere 1989:92; Pluciennik 1997:48).
Fabrics-'coarse wares'
The coarse ware pottery recovered from early Roman Roughground Farm was made from a
The
fabrics.
fabric
followed
commonly
most
occurring
the
of
ware
variety
was
reduced
wide
by grog-tempered ware, Black-burnished ware (BB I and imitation BB 1) and fabrics with
limestone
flint
inclusions
(see
Table
fabrics).
breakdown
6.1.
for
such
as
shell,
and
of
various
There was a distinction between some of the fabrics and specific vessel forms: wide-mouthed
jars
were generally reduced, vertical wall jars had limestone inclusions and
and narrow-necked
the high-shouldered necked jars were grog-tempered. All of the non-samian cups were made
from a Severn Valley fabric. The association of specific fabrics with specific vessel types
being
for
for
indicate
these
that
were
vessels
recognized
as
appropriate
specific
uses or
could
Sherratt
1985:
152;
1987:
83;
Sillar
1997:
foods
(Miller
Evans
1987:
14;
200; Brown
specific
1997:100). 1 do not wish to imply, however, that there was only one prescribed use for each
132
been
flexible
have
(Rice
1987:
209),
The
quite
type.
and
of
some
may
vessels
use
vessel
histograms of the some of the forms with sufficient numbers of rim diameters show that there
'My
for
6.3.
).
is
Figure
(see
type
point that the purchase
each vessel
was quite a range of sizes
be
(after
Miller
1985:
162)
`familiar
of
pots might
culturally and socially prescribed
series'
of a
by
functionality
(see
for
Woods
the
the
than
purely
of
vessel
motivated
example
rather
1986:163 who observes that the shapeof the Roman period cooking pot "break[s] all the rules
for thermal shock resistance").
It is also possible that vessel fabric, in this case Black-burnished ware and brown ware, were
form
A
important
bowls,
for
bowl/dishes
jars,
the
types,
of
a
vessel.
variety
as
of
vessel
and
as
instance, were of Black-burnished ware and a series of decoratedjars, storagejars and dishes
fabric.
brown
This
factors
the
that
suggests
out
of
other
such as point of origin or
were made
(see
Cumberpatch
1997; Brown 1997:100) may
texture
the
the
and
of
vessel
colour
possibly
have influenced the selection of these particular vessels.For instance, the use of brown vessels
for
beakers
been
have
the
occasions;
particular
significant
recovered at the settlement also
may
had a brown slip and vessels made from a brown coarse fabric were decorated and/or
That
Black-burnished
in
imitated
(except
and
serving.
storing
with
ware
was also
associated
its method of manufacture - the imitated wares were wheel-turned, not hand-made) suggests
that it was a prominent import (Brown 1997:110) but also implies that the style and/or colour
least
important
the
at
as
as their point of origin and method of manufacture
vessels were
of
(see Cumberpatch 1997:126-8). Allen (2000:4). for instance, has suggested that brown and
black vesselsat Iron Age sites in Oxfordshire might be imitations of leather vessels.
Rim diameters
Histograms of the jars show a prominence of medium sized vessels as indicated by their peak
few
6.4.
).
Too
bowls
Figure
16-18
(see
the
of
cm
and dishes had definable rim diameters (3
at
2
and
rims respectively) although together with the rim diameters of the cups (as a group
rims
that might reflect serving ware) (see Figure 6.5.) it is apparent that the vessels are medium to
large sized - which may point to communal consumption, at least publicly. It is also possible
that the large open jars, which are presumably less portable, will also have played a role in
(Howard
1981:
19;
Vencl
1994:
310).
events
public consumption
133
Discussion
In contrast, the pottery recoveredat early Roman period Bartbn Court Farm (although the
be
Barton
Court
Farm
by
differing
bowls
the
at
may
slightly
exaggerated
of
percentage
forms)
higher
bowls
to
reveals
a
significantly
proportion of shallow
and
approaches classifying
beakers than was found at Roughground Farm. Roughground Farm in turn has a higher
dishes
dish/bowls.
This
different
drinking
to
and
and
cups
of
points
eating
and
proportion
In
the
two
the
settlements.
absenceof stratified amphorae,the prominenceof cups
at
practices
drinking
different
different
beverage.
indicate
than
the
custom
rather
a
consumption
of
a
may
It is also possiblethat cupswere used for particular occasions,possibly related to significant
deposits'
below),
it
is
increase
`special
in
(see
that
there
the use of
notable
was
an
events
beakerslater at the villa site, and a decline in the use of cups (Booth 1993:139).The typesof
Roman-stylevesselsidentified at the settlementappearto be associatedwith serving rather
flavouring
food
in
for
(see
Claydon
Pike,
Chapter
8
particular
and
ways
than preparing
imitation
The
Belgic-style
findings).
tablewareandother nonof
a
variety
presence
of
opposite
Roman tableware,however, suggeststhat `Roman-ness'was not necessarilya key factor in
On
Roman
Roughground
it
Farm
the
contrary,
at
events.
early
appears
public consumption
fabric
form
in
and
reflectsa complex repertoireof consumptionpractices.
that the variety
The early Roman period glass
Most of the glass recoveredfrom the site dates from the late secondcentury to the fourth
fragments
jar/flask
hexagonal
However,
late
first
bottle
date
two
and
to
the
a
century.
-a
from
later
items
Roman
but
these
recovered
were
contexts
were possibly usedat the
century;
Roman
settlement.
early
6.3.3. Other implements
A number of artefacts made out of different materials were recovered which have an
drink.
Sizeable
food
hole
amounts
and
of
clay
oven
namely
with
stoking
parts,
association
flue
fragments,
in
supports,
arches
and
plate
oven
wall
were
plates,
recovered
oven
arches,
134
been
has
It
domestic
Roman
that
the
suggested
contexts.
ovens were probably
early
several
for
both
(archive
9).
Three,
bread
for
grain,
or
report
pg.
or
roasting
possibly
making
used
four, fired clay sling pellets associated with hunting, rather than combat, were also identified
in early Roman contexts. In contrast with Barton Court Farm, numerous objects associated
the
and
whetstones
were
at
early
pestles/rubbers
recovered
querns,
preparation
grain
with
Roman settlement.
6.3.4. Ingredients
Thespecialists' reports - the animal bones

It is difficult to gaugethe significanceof the percentagesof the speciesand their body parts at
RoughgroundFarm in part becauseof the excavation and post-excavation strategies.The
Roman
is
bones
the
to
the
period
settlement
as
early
somewhat
confusing
animal
of
allocation
both
bone
focuses
the
the
published
and
report,
yearsof
on
microfiche,
on
and
archive
site
From
determine,
I
the
than
the numbers
settlement.
chronology
of
what
can
excavationrather
Roman
bones
the
there
identified
early
settlement
at
small
and
are
were no sievedsamples.
of
The collecting of the bonesduring the 1957excavation,when the extent of the early Roman
been
have
less
`may
determined
later
than
the
rigorous'
site
of
excavation
was
settlement
(Jonesand Levitan 1993:171).A Mr. Baxter, the local vet at the time, studiedthe bonesduring
identifications
his
Eric
Higgs
to
to
the
appear
stand
up
conclusions
who
and
of
the excavation
identified an additional box of bones, some of which
identifications,
Baxter's
and
reviewed
Higgs,
bones
(although
been
to
have
Roman
according
pig
may
slightly overwere early
Jones suggestedin a 1981 archival report that the small
Margaret
Baxter).
by
represented
bones
Roman
indicate
the
that
the
at
early
recovered
settlement
animal
might
of
numbers
(see
below).
`plants'
the
on
agriculture
was
settlement
emphasisat
bones
does
the
animal
The published analysisof
not refer specifically to the bonesrecovered
but
bones
Roman
dates
to
the
the
settlement,
rather,
period
refers
animal
were
at the early
bones
described
for
headings
bones
(recovered
"Cattle
Therefore,
example,
under
excavated.
between1957 and 1982)" - pertain to bonesfrom throughoutthe Roman period. My analysis
135
is largely basedon the primary bone reportsas preparedby Baxter and Higgs with additional
by
1990
Levitan.
bone
Bruce
from
the
which
were
of
excavation,
prepared
records
material

Plant remains were not recovered at the early Roman period settlement. It does appear,
however, that agriculture (or the preparation of grains) may have been important to the
inhabitants as suggestedby the non-botanical remains recovered at the settlement, which have
been assembled below.
6.3.5. The ingredientsat the early Romanperiod settlement
Based on bone counts it appearsthat cattle were the most commonly identified species,
followed by sheep,horse and pig (seeFigure 6.6.). The age of death of the cattle and sheep
including
dairy,
draft
of
indicating
uses
possible
a
range
meat,
animals.
and
and
as
wool
varied
Horses all tended to be adult, although molars from `young' horses were identified. Pigs
`young',
killed
been
have
`delicacy'
and
when
were
to
perhaps
considered
a
culinary
appear
horn
A
(Baxter: archive report). non-butchered
core of a red deer and the lower jaw of a deer
indicate
inhabitants
the
together
identified
sling
the
the
with
pellets
that
could
which
of
were
diet
hunted
the
the
contribution
to
species;
wild
of
wild
species
was probably
settlement
been
have
there
social significanceto their capture and consumption.
may
minimal although
No dog remains were identified at the settlement.In `pre/early villa' contexts from the 1990
Roman,
frequent
early
were
the
which
of
cattle
was
most
species
also
excavation,a number
identified, followed by sheep;in contrast,pigs and horsewere in the reverseorder to what was
found in the earlier excavation.
136
Butcherypractices
It was not possible to cataloguespecific butchery marks on the various speciesas they were
bones
indications
There
in
that
detailed
the
were chopped,cut and
are
primary record.
not
larger
bones
broken
"most
Baxter
the
that
of
were
up and several
observed
occasionallysawn.
it
long
(archival
In
distal
have
bones
the
the
marks"
report).
general,
was
of
ends
cross
shank
bones that were recovered suggestingpossible disarticulation of cattle and sheep at the
had
`knife
joints.
A
knee
cattle
was
phalange
or
chop
marks'
mid-shaft
which
and
shoulder
distal
horse
The
`sawn'
tibia
through;
this
end
of
a
single
skinning.
was
of
evidence
as
seen
be
horse
butchery
horse
taken
that
can
not
as
evidence
were consumed,especially
exampleof
broken
in
less
bone.
it
However,
bones
than
horse
up
cattle
was suggested an
were generally
as
bones
horse
in
butchered
bones
because
they
that
that
were
mixed
with
other
report
archive
horse
bones
found
butchered
been
the
have
species
other
a
number
of
with
consumed;
might
below).
(see
weremeat-bearing
Butcherymarks identified from `pre/earlyvilla' contextsduring the 1990excavationshow that
humerus
distal
through
that
base
the
the
chopped
of
cattle
was
and
was
scapula
end of a
of
the
had
horse
bones
butchery
No
specific
or
pig
marks, and one rib of an
sheep,
chopped.
had
`large'
the
cuts
on
shaft.
species
unidentified
Meatyields
bones
the
when
the
small,
are groupedaccordingto their potential meat
are
Although
numbers
from
bones
6.7.
),
`B'
(bones
Figure
that
(see
within
see
we
group
with moderateamounts
yield
horses,
for
bones
the
the
are
of
most
exception
commonly
each
represented
of meat), with
(minimal
`C'
from
less
Bones
amounts
group
of
meat)
are
represented,which could
species.
brought
to the settlement already `dressed',although the
indicate that some animals were
domesticates,
does
the
the
that
the
variable
ages
of
and
main
pens
suggest
animal
of
existence
is
It
bones
`C'
the
that
to
also
settlement.
possible
at
were
used
reared
group
were
animals
deposits
deposits
they
that
special
the
represent
or
tools
waste
within
or
recovered
not
make
domesticarea of the settlement.In `pre/earlyvilla' contexts, for example,the distribution of
for
bones
different
higher
À'
`C'
bones
cattle
quite
with
was
of
and
percentages
meat-bearing
bones.
The
bones
identified
`B'
in
hardly
thesecontexts were all
bones
sheep
any
and
type
137
identified
for
bones
bones,
`C'
`B'
`B'
were
and
pig and a single maxilla
meat-bearing
group
identified.
horse
humerus
was
a
of
and
The significanceof the grouping of meat-bearingbonesat the early Roman period settlement,
is
bone
bones
likely
have
the
the
the
arguable;
record,
groups
of
of
more
nature
given
individual
distribution
(see
in
below).
their
to
contexts
of
relation
remains
significance
Non-botanical evidence of agriculture and/or grain processing
Plant remains were not recoveredin the area of RoughgroundFarm where the majority of
in
is
Findings
Roman
contexts associatedwith secondcentury villa
settlement evident.
early
layers indicating small numbersof spelt wheat are of note, but are inconclusive in terms of
Plant
from
1990
being
the
the
consumed
and/or
at
settlement.
processed
samples
what was
from
features.
Certainly,
third
the
century
and
second
only
are
processingand
excavation
inhabitants
in
large
to
the
significance
the
of
particular
was
of
grain
as seen
consumption
depositof quernsin one areaof the settlement(seebelow). Furthermore,a number of the pits
directly associatedwith the houseenclosurehave been linked to `grain roasting' and possibly
Clay
been
have
that
the
to
were
recovered
around
ovens
enclosure
used
also
may
storage.
grain
low
As
bones
the
above,
noticeably
mentioned
was
numbers
of
animal
recovered
roast grain.
Roman
that
the
may
signify
settlement
emphasis at the settlement was on
the
early
at
1981
Jones
(M.
for
that
archival
report),
and
animals
were
consumption
reared
agriculture
the
economy.
settlement
than
of
part
as
rather
6.4.
Summary of the Artefacts and Remains of Consumption at Roughground Farm
Before we consider the distribution of the artefacts and the remains at the early Roman
is
it
to
summarize some of the points made above on the
necessary
period settlement,
ingredients.
implements
the
the
and
containers, other
138
6.4.1. The Containers
Thepottery
Jarsare the most dominantform, followed by bowls, cups,dishesand beakers.Flagons
in
and mortaria arepresent very small numbers.
"
jar
identified,
indicate
in
types
the methods
was
of
which
could
variety
variety
wide
"A
foods.
of preparing and cooking
fall
into
21%,
7%
that
The
the
tableware
assemblage
might
of
category
percentage
was
"
(mostly
tableware
the
was
samian
non-decorated).
of
indicate
jars
jars
Histograms
the
that
were generally medium-sizedwith a peak of
of
"
16-18cm.
indicate
Histograms
that the vessels were predominantly
serving-type
vessels
of
"
large
indicate
to
which
could
sized,
communaleating.
medium
`fine
Four
the
total
to
the
pottery
assemblage
was
of
assigned
percent
and specialty
"
imply
low
that
the
could
which
settlement
category,
was
of
status(after Booth
ware'
1991; in press).
have
been
have
that
additionally,
non-local
coarse
suggested,
ware
may
also
"I
form
importance
have
been
the
that
the
or
even
colour
and
of
vessel
may
of
prestigious
to the inhabitants.
imitation
Belgic-style
Presence
tableware
and other non-Romantablewaresuggests
of
"
that `Roman-ness'was not necessarilya factor in consumptionevents.
Theglass
"
Two fragmentsof late first century glass-a jar/flask and a hexagonalbottle - were
Roman
from
have
been
they
period
contexts;
non-early
used at the
may
recovered
settlement.
6.4.2. Other implements

in
Roman
of
clay
oven
parts
were
number
recovered
contextssignificant
early
"A
for
for
bread
making and/or roastinggrain.
usedpossibly
hunting
have
been
identified.
Sling
with
associated
pellets
"
large
collection of quernsandrubberswere recoveredat the settlement.
"A relatively
6.4.3. Ingredients
Theanimal bones
"
"
"
"
"
Cattle were the most commonly identified species,followed by sheep,horseand pig.

The ageof deathof cattle and sheepsuggestthat they had varied uses.
Pigs were primarily killed when young-a culinary delicacy?
Few wild animals were identified, althoughthe recovery of sling pellets suggeststhat
deer
bones
hunted,
two
were recovered.
and
they were
In `pre/early villa' contexts from the 1990 excavation, cattle were again the most
followed
by
identified
(rather
horse),
horses.
than
species,
sheep,
pigs
and
commonly
139
"
Butchery marks are not detailed in the primary record; Baxter observed that "most of
the larger bones were broken up and several shank bones have cross marks" (archive
report).
"
On bonesfound in `pre/earlyvilla' contextsfrom the 1990 excavation,few butchery

bones
had
large
Only
the
species
of cattle and an unidentified
marks were observed.
did
Sheep,
horse
bones
display
cut
marks.
pig
and
any specific
not
chop and/or
butcherymarks.
"
When the bones were grouped according to their meat yield, bones with moderate
(group
`B')
frequent.
the
of
meat
were
most
amounts
Non meat-bearing bones were less common at the settlement, which could indicate that
domesticate
brought
dressed,
the
to
the
were
settlement already
although the
some of
presence of possible animal pens and the variable ages of the species suggeststhat at
least some animals were reared at the settlement. It is quite possible that non meatbearing bones were disposed of elsewhere.
"
"
The animal bones at the settlementappearto indicate that animal rearing was not
by
but
for
the
the
of
a
major
part
site
economy,
was
consumption
rather
necessarily
inhabitants.
Non-botanicalevidenceof agriculture
Roman
No
the
were
recovered
at
remains
early
plant
period settlement.
"
is
by
indicate
However,
the
that
strongly
suggested
agriculture
grain
presence
of
pits
"
by
the significant numbersof ovens recoveredat
storage,
and
possibly
and
roasting
the settlement.
considerablenumber of quernsand other cerealprocessingparaphernaliawere
"A
from
the settlement.
recovered
identified
low
The
animal
of
remains
at the settlementsuggeststhat
numbers
"
have
been
important.
agriculture may
6.5.
The Distribution of the Artefacts and Remains of Consumption at Roughground Farm
Early Roman period Roughground Farm provides a unique opportunity in this study to
from
the
the
main areaof occupationwith other areasof the site.
recovered
material
contrast
Fine waresrecoveredfrom someof the featuresaway from the main areaof occupationhave
led the authors of the site report to consider whether there was a second early Roman
higher
(Allen
the
be
1993:
It
181).
status,
at
of
site
et
possibly
al.
should
noted,
settlement,
however,that excavationof the early Roman contextsbeyond the main occupationareawas a
buildings
the
the
of
villa
by-product of
un-covering
and thus the featuresdo not presenta
below
The
distribution
the
analysis
of
settlement.
of the artefactsand remains
coherentplan of
is consequentlyorganizedinto those directly associatedwith the main areaof occupationand
buildings
subsequent
villa
and their environs.
thoseareasassociatedwith
140
6.5.1 The main areaof occupation(seeFigure6.1.)
The main occupation area is characterized by a large oval house enclosure with surrounding
by
bounded
that
are
a rectangular enclosure and small groups of gullies and
pits and gullies
ditches just outside the rectangular enclosure. The distribution of the pottery types amongst the
homogeneous,
is
few
features
6.8.
Figure
):
(see
generally
with
a
notable
exceptions
various
found
in
feature
the
was
each
of
ware
groups, although the main enclosure and
serving-type
had
higher
house
slightly
percentagesof this class of wares. Samian ware was
enclosure
the
from
ditches
from
house
Pits
the
that
and
gullies
away
oval
enclosure.
only, recovered
burnt
house
late
charcoal
the
stone,
and
ash
were
also
associated
with
at
site and, as
contained
Iron Age Barton Court Farm, these features could indicate the remains of feasts (Gomez de
Solo (1993: 191). Consideration of other types of deposits reveals a marked distinction
between the various features of the settlement. The features to the west of the main enclosure
bones
in
fact
this
groups
of
animal
sizeable
area of the settlement contained the
contained
largest group of identified animal bones recovered at the early Roman settlement (see below) jars and a couple of bowls (no dishes, cups or beakers were identified) and the majority of the
found
In
house
the
paraphernalia
at
the
associated
settlement.
and
contrast,
enclosure
querns
features
hunting
burnt
bone
fragmented
contained
ovens,
pellets,
and
animal
and surrounding
it
in
of
serving-type
selection
wares:
varied
was
also this area of the settlement that
and a more
five brooches were recovered. There is clearly a distinction between the types of deposits
(see
`special
deposits'
from
below).
two
the
areas
recovered
The natureof the recoveryand recordingof the animal remains(seecommentsabove)dogged

distribution
bones.
My
the
the
of
animal
of
the examination
analysisis thus largely limited to
individual
than
contexts
rather
on
on establishing a distribution pattern for the
observations
deposits
bones
Large
from
of
animal
were
settlement.
only
recovered
one area of the
whole
just
In
the
this area meat-bearingcattle boneswere
main
occupation
of
area.
west
settlement,
dominant in the gullies and sheepwere better representedin the pits and hollows as were
few
bones.
In
identifiable
horse
bones
(10) were recovered from the
contrast,
meat-bearing
five
from
house
three
bones
identified
were
sheep,
two
enclosure;
were
cattle
and
were
main
fragmented,
the
The
burnt
enclosure
also
contained
around
pits
and generally nonas pig.
identifiablebonesthoughtto representthe remainsof meals(M. Jones1981archival report).
141
6.5.2. Pre-villa and environs
The early Romanfeaturesfound nearwhat were later villa buildings (referredto from here as
`pre-villa and environs') will be discussedprimarily in contrast to the main early Roman
because
do
they
not presenta coherentgroup of features.As was commented
occupationarea
have
the
the
of
site
authors
report
suggested that a second early Roman period
on above,
higher
have
been
status
may
of
present at the site. Figure 6.9. compares the
settlement
identified
in
differences
the
two
wares
of
areas
and
notable
percentages
are observablein
is
drinking.
Serving-type
that
associated
with
wares found in the pre-villa areas
pottery
identified
23%
in
of
that area compared to 15% of the identified
about
pottery
comprise
Roman
in
the
area
of
early
main
period occupation.The main distinction betweenthe
pottery
from
in
the
two
the percentagesof cups, particularly samian
recovered
ware
areas
was
serving
in
identified
the
in
pre-villa
than
and
environs
the percentagesof bowls and
areas
rather
cups,
disheswhich were comparablein both areas.All but one of the cups was found in what are
thought to have been stock enclosuressouth of the main areaof occupation.The other `high
in
from
the
pre/villa
area
were
recovered
wares
a pit (320), which may have functioned
status'
it
containeda samiancup and a samianbowl and two brown-slippedbeakersaswell
well;
as a
bones
(see
below).
types
of
animal
asspecific
Main occupation
area
n=548
fine and specialist wares* 13
37
Black-burnished wares
30
oxidized wares
38
brown wares
34
tempered
wares
shell
lime scale tempered wares 59
154
wares
grog-tempered
178
reduced wares
other
Villa area and environs
%
2
7
5
7
6
11
28
32
n=526
31
54
34
25
14
16
80
269
%
6
10
6
5
3
3
15
51
Table 6.2. The distribution of early Roman period pottery fabrics at RoughgroundFarm
142
A distinction was also apparent between the pottery fabrics of the two areas (see Table 6.2.)
Fine and specialist fabrics appear more prevalent in the vicinity of the villa buildings although
fabrics
in
forms.
The
is
less
difference
the
than
same coarse
overt
comparison of pottery
the
in
is
for
identified
their
the
area
and
comparable
except
case of grogeach
representation
were
in
higher
there
the prewhere
wares
was
a
of
wares
percentage reduced
tempered and reduced
in
the
than
the
site
of
areas
main area of occupation where grog-tempered
environs
and
villa
in
Grog
temper
are
more
equally
represented.
was
used
primarily
and reduced wares
jar;
it
is
form
high-shouldered
`Belgic'
the
necked
considered
and was
a
style
production of
The
(Booth
1993:
135;
Green
1993:
134).
the
sites
until
second
native
century
common at other
in
Roman
found
found
in
the villa area of the settlement.
contexts
early
was
also
mortarium
A small number of animal bones were identified in the pre-villa areas. Meat-bearing bones
from cattle, sheep and horse were identified in some of the stock enclosures, although most of
from
below
bones
buildings.
One of the pits identified
were
pits
situated
recovered
villa
the
described
the
type
320,
serving
some
of
contained
ware
as
above - also contained
which
pit
from
lower
in
teeth
the
the
the
cattle
sheep,
and
while
pig,
another
pit
area contained
primarily
jaws of sheep and cattle. The bones found in these pits are primarily from the `C' group of
bones - comprising mostly bones from the head. This contrasts with the animal remains found
in the pits in the western part of the main occupation area described above where meat-bearing
bones of the main domesticates, including horse, were identified. Again, the significance of
is
to
the
excavation and recording of the animal remains.
relative
this contrast
6.5.3. `Specialdeposits'
identified
Roman
the
the
at
early
A number of
pits
settlementand its surroundingareasmay
deposits'.
The
living
`special
functional
the
pits
west
of
main
contained
area
still
represent
(Allen et al. 1993:161) rotary querns and pestles/hammerstones (one pit, for example,has
locally
jars
bowls,
these
small
selection
a
of
objects),
made
of
and
and significant
seven
bones.
is
It
animal
possible that these pits contain the remainsof
numbersof meat-bearing
feastsor the ritualized preparationof food for feastswith an agrarian theme (see Gwilt and
Heslop 1995:4 who similarly relate the distribution of querns to social factors). An infant
143
burial was recoveredfrom the terminal of a gully in this areaand it is possiblethat the burial
was part of the ritual.
The isolated pits situated under what were to become villa buildings contained more Romanlike pottery and the remains of animal head bones which is considered a ritualized practice at
(Scott
1991:
117;
Perring
1989);
deposits
Roman
settlements
period
a
similar range of
other
deposits
in
Roman
Barton
Court
Farm
(Roman-like
identified
at
early
special
pottery and
was
The
deposits
the
types
the
contrast
with
parts).
of
special
animal
at
main settlement
non-edible
(local
Farm
bones)
features
Roughground
that are
the
pottery,
querns
and
meat-bearing
and
at
is rather interesting. The dump of samian cups in the stock
future
in
the
site
villa
situated
become
in
to
the
of
what
was
area
a villa may signify the reorganization of the
enclosures
individualized
drinking
the
and
use
of
alcohol,
cups generally associated with
settlement using
habits
is
Roman
particularly noteworthy (see below for further discussion on
consumption
the
this issue).
Situated two hundred and fifty meters away from the main area of occupation and juxtaposed
Iron
Age
Roman
an
early
circular
possibly
structure
was
was
what
a square-shapedearly
with
its
featured
At
jar
that
the
that
centre
was
a
pit
an
upright
enclosure.
contained
grey
period
human.
Within
burial
the
the
an
adult
of
enclosure
and
were
remains
surrounding
pit
cremated
have
formed
fence
they
may
a
post-holes;
around what appears to have been a
two series of
four-post structure that was situated over the top of the burial pit. It is also possible that the
four-post structure was burnt (Allen et al. 1993:53). The jar (which incidentally is technically
the
found
in
the
that
cremated
remains
contained
and
the enclosure was
pottery
coarse-ware)
This
burial
date.
is
in
type
Upper
Thames
Valley
(Allen
of
the
century
unusual
second
of early
four-post
(a
192)
1993:
burial
has
been
similar
structure,
which
contained
a cow
et al.
identified at the site of Smithsfield (Allen 2000: 20)). The burial tradition has been identified as
France and has been recognized at a small number of east
Marnian
in
the
region
of
originating
first
in
in
Britain
BC
the
AD
(Allen et al. 1993:53; Allen
centuries
and
settlements
coast
2000:20). The association of an early Roman burial enclosure with an early Iron Age structure
is also unparalleled in the area. The authors of the site report suggest that the cremation burial
immigrant
(Allen
Gallic
1993:
192).
be
et
al.
that
a
of
may
144
6.5.4. Summaryof the re-contextualizedmaterialat the early Romanperiod settlement
In spite of the incongruities in the archaeology and in the recording of the site, differences in
the distributions of the artefacts and remains associated with eating and drinking have been
in
Roman
Roughground
Farm.
In
the main
a
of
areas
at
number
early
period
established
distinction
between
living
there
the
was
a
area,
area which emphasized cooking,
occupation
feasting,
hunting
items
'
as
well
possibly
as
and
paraphernalia
and
of personal
serving
living
the
the
area
west
of
main
and
area where the contents of a group of pits and
adornment,
ingredients:
bones,
bones,
including
horse
to
emphasize
meat-bearing
animal
gullies appear
in
These
distinctions
tools
the
the
used
preparation
grains.
of
particularly
could reflect
and
different labour domains possibly associated with gender or age; it is also possible that the
different distribution patterns point to the separation of different practices linked to the
food
for
of
consumption
particular occasions.
preparation and
A contrastwas alsorevealedbetweenthe main areaof early Romanoccupationand other early

Roman period featuresthat underlay the villa buildings and their environs. The different
distributionsof the pottery in the two areascould indicate the presenceof two settlementsof
differing status. The higher percentageof samian ware, the presenceof mortaria and the
declinein the useof `native' grog-temperedwarestogetherwith the differing disposalregimes
indicators
in
the
inhabitants
pits,
are
possible
remains
of
a
second
of
with
animal
group
of
differing consumption practices. Alternatively, it is possible that the features were an
Neighbouring
Claydon
the
Pike and Thornhill Farm each
occupation
main
area.
of
extension
had stock enclosuresand featuresthat went well beyond the main occupationarea.The two
types of consumption practices could reflect variability in the way food and drink was
That
different
distinctions
between
the
found
in
events.
main
at
the
the main
consumed
pottery
from
the
in
the
and
pottery
recovered
area
pre-villa
area
occupation
was vesselsassociated
distinctly
`Belgic'
drinking
of
a
vessels
and
style could reflect the chronologicalbreadth
with
highlight
different
but
also
may
episodesof eating and drinking prior to the
of occupation
These
lie
directly
beneath secondcentury villa
enclosures
and
a
villa.
of
pits
construction
buildings and it will be suggestedbelow that the shift from `native' house to villa was a
links
to
the
strong
past.
retaining
markedevent
145
6.6.
Discussionof the Distribution of Artefacts and Remains at Roughground Farm
Similarities exist betweenthe typesof consumptionpracticesthat have been identified at early

Roman periodsBarton Court Farm and RoughgroundFarm. The range in pottery and the use
Roughground
Farm,
food,
Roman-like
to
containers
at
especially serve
are reminiscentof
of
Barton Court Farm, and both settlementsappearto have adoptedcustoms associatedwith
individualized as well as communaldrinking. The contextualassociationsof the artefactsand
drinking,
however,
between
linked
the
eating
and
with
also
revealed
clear
contrasts
remains
This
distribution
the
was
particularly
apparent
with
of cups and querns at
two settlements.
RoughgroundFarm where eachmight be associatedwith the reorganizationof the settlement.
Unlike Barton Court Farm where the reorganizationof the settlement in the early Roman
boundaries
by
households,
defined
Roman
is
the
the
and
consolidated
structure
of
early
period
from
Farm
beginning
Roughground
less
the
the
at
appears
when
confined;
and
settlement
in
the
there
early
second
reorganized
century
was
was a concomitantpresenceof
settlement
both nativeand Romanidentities.
6.7.
The Social Contexts of Imperialism at Roughground Farm

0
The limits of the early Roman settlement at Roughground Farm are difficult to define and
interpret. A rectangular ditch enclosed part of the main living area although occupation clearly
boundary ditch. The site was not occupied in the late Iron Age, which
beyond
this
existed
house
in
the early Roman period particularly
of
a
circular
the
site
construction
makes
significant
when compared to examples of
contemporary rectangular structures at
Claydon
(e.
Pike).
The
its
g.
the
settlements
shape
of
enclosure
and
southwestern
neighbouring
the
though
The
even
enclosure
times.
was
respected
was
re-cut
many
orientation of
entrance
have
had
the
to
may
enclosure
particular significance to the inhabitants (see
the entrance
Parker Pearson 1996 on the symbolism of entrance orientation). It is also apparent that the use
deposits
be
feature
for
to
daily
habitation
in
continued
a
prominent
the
ritualized
rites of
of pits
into
is
in
fourth
This
the
and
the
century
even
persisted
second
at
site
the
until
century.
well
Barton
Court
Roman
Farm,
in
Roman
period
early
the
and
with
with
period
general,
contrast
into
decline
(Allen
1993:
179).
The
went
et
al.
the
of
pits
use
positioning of an elaborate
when
burial site next to the site of an early Iron Age circular structure towards the end of the lifetime
146
Roman
links
important
that
tangible
to
the
to
the
settlement
also
suggests
early
past
were
of
the inhabitantsof the settlement.
It is particularly interesting that Roughground Farm with its numerous signposts that point to a
identity,
Barton
Roman
(unlike
throughout
the
was
of
native
occupied
period
celebration
Court Farm) and was characterized by villa-type buildings as of the second century, which is
in
Upper
Thames
Valley.
history
Indeed,
has
long
the
the
uncommon
site
of villa
a
relatively
began
demise
in
Roman
the
the
either
with
which
of,
or
conjunction
early
with,
construction
in
The
the
century.
second
second century villa occupied the same general area as
settlement
has
Allen
Roman
and
settlement
so
suggestedthat the deposits of querns might have
the early
had a "propitiatory significance" relating to the reorganization of the settlement and the
in
by
(Allen
the
1993:
161;
a
villa
of
second
century
comments
et al.
see
establishment
Ardener 1993: 13 re: the ritualized marking of significant events). The dump of samian cups in
deposits
in
other
and
special
pits situated under villa buildings could be
the stock enclosures
further examples of the importance placed on the reorganization of the settlement. That the
dating of the elaborate 'native' burial appears to coincide with the end of the early Roman
further
the
the
construction
of
and
villa
suggests that the reorganization of the
settlement
bound
identity
the
with
unabashed
commemoration
very point
was
of
at
a
native
settlement
inhabitants
Roman-like
towards
the
making
overtures
were
structures.
when
6.8.
Conclusion
Roughground
Farm involves a complex integrationof past and
the
interpretation
site
of
The
of
Emphases
have
the
to
in
past.
and
changed
priorities
excavation strategy
presentapproaches
fifty
have
through
just
been
this,
still
and
all
the
years,
past
we
someof what
to
over
able glean
is fascinatingaboutthis site. In this chapter,I haveestablishedcontrastingtraditions in the way
drink
food
was consumedthat appear to indicate celebrations that emphasize
that
and
ingredients,possibly the agricultural cycle, and the serving of food and particularly drink,
in
instance
different
Roman-style
These
one
vessels
and
in
vessels another.
using native-style
different
however,
two
living
represent
they
groups
of people
culinary stylesmay
side-by-side;
important
household.
the
by
marking
of
the
represent
events
one
at
settlement
could equally
147
While Roughground Farm is difficult to typecast, it is evident that the establishment of a villa
important
demand
to
the commemoration of particular aspectsof the
event
sufficiently
was a
from
houses
done
in
Whether
the
to
movement
rounded
was
celebration or
rectangular
past.
debate.
is
Old
Shifford
dissertation,
in
Farm,
to
this
the
trepidation
open
next case study
with
is similarly difficult to typecast - except to say that it had an entirely different approach to the
drink
food
in
Roman
the
and
of
early
period.
consumption
148
MainOccupation
Area
.\
.\
It
Figure 6.1 Early Roman period RoughgroundFarm: main area of occupation (after Allen et
Meadows)
by
B.
drawn
al.
14')
Rough Ground Farm early Roman period containers
Figure 6.2 Early Roman period potterv forms
Pottery forms
n=210
100 80
60
40
20
Cl
cN
(0
(D
Figure 6.3 Histograms of the rim diameters of different jar types
Jars - wide mouth

N
Jars - high shoulder
15 -
y6-
10
O4
E0
ZN
CO
VNC;
cc
ZN
O
NNMM
Co
Rim diameter
CD
CO
Rim diameter
Figure 6.4 Histograms of the rim diameters of Jars
Fi tere 6-5 Histograms
Jars
of tableware
Table ware
4
a
03
2.
40
30
20
10
z
0
N
Co
O
NN
cD
Rim diameter
N
c-,
cc
ci
<O
OV
co
N
CD
NNMM
Rim diameter
co
CO
150
Roughground Farm early Roman period ingredients
Figure 6.6 Early Roman period N. I. S.P.
Figure 6. - Animal bone groups according to meat yield
Animal bones
Animal bone groups
n=114
100
60 F
50
40
%30
20
10
0i
80
60
%B
40
F1
cow
sheep
pig
20
ED
horse
deer
OA
f
It
0I-1.
cow
sheep
pig
horse
Qc
ll
Distribution of containers at early Roman period Roughground farm
Figure 6.8 Distribution
in main occupation area
of potter
Main occupation
area
80
60
O ditch
pit
O gully
% 40
O environs
20
0
Qd
pC
Figure 6.9 Comparison of pottery forms between pre-villa area and main occupation area
Distribution
of pottery types
100 80 60
main
%O
40
(3 villa
20
'
0-
--
n=
-0
CD
m
mM
n3
cn
152
Chapter 7
Table of Contents for Old Shifford Farm
Introduction
The Excavation
The site assessment
The main excavation
The Site
Late Iron Age settlement - Phase One
Late Iron Age settlement - Phase Two
Early Roman settlement - Phase Three
Site reports and archives
7.2.4.
Farm
Consumption
Old
Shifford
Drink
Food
Archaeology
The
and
at
of
7.3.
The containers
7.3.1.
7.1.
7.2.
7.2.1.
7.2.2.
7.2.3.
7.3.2.
154
155
159

Thepottery at the late Iron Age settlements- phaseone and two

Form, function and fabrics- phaseone
Form, function and fabrics - phase two
Discussion
7.3.3.
7.3.4.
7.3.5.
Other implementsat the late Iron Age settlement- phaseone and two
The ingredients
Thespecialists' reports- animal bones
The specialists' reports - the plant remains
The ingredients at the Late Iron Age settlements
Theanimal bonesat the late Iron Age settlements-phase one and two
Butcherypractices
Meat yields
Theplant remainsat the late Iron Age settlement- phase one and two
7.4.
7.4.1.
7.4.2.
7.4.3.
7.5.
7.5.1.

The Containers at the early Roman period settlement
Form and Function
Fabrics
Rim diameters
The ingredientsat the early Romanperiod settlement
Butchery practices
Meat yields
Theplant remainsat the early Romanperiod settlement
Late Iron Age settlement - phase one and two
Containers
Pottery
7,5,2.
Other implements - phase one and two

Ingredients
Animal bones - phase one and two
Plant remains - phases one and two
169
174
153
Containers
The pottery
7.6.
7.6.1.
Other implements
Ingredients
Animal bones
Plant remains
The Distribution of the Remains of Eating and Drinking at Old Shifford Farm
177
Distribution of the artefactsand remainsat the late Iron Age settlements

Phase one
Phase two
7.6.2.
7.7.
7.7.1.
7.7.2.
7.8.
7.9.
'Specialdeposits'
Distributionof theartefactsandremainsat the earlyRomanperiodsettlement
'Special deposits'
Discussion of the Distribution of Artefacts and Remains at Old Shifford Farm
The late Iron Age period settlement - phasesone and two
The Social Contexts of Imperialism at Old Shifford Farm
Conclusion
Figure 7.1 - 7.13
186
187
188
190
154
Chapter 7
Old Shifford Farm
7.1.
Introduction
Old Shifford Farm is situatedat the edgeof the floodplain on the low-lying first gravel terrace
just north of the Thames.Although the site is low-lying, the land is well drained and situated
first
in
late
BC/early
The
Iron
first
level.
(late
Age
flooding
the
site
was
settled
century
above
first centuryAD) and appearsto have beencontinuouslyoccupiedthrough to the early Roman
late
in
first
The
it
AD/early
AD.
the
site
century
secondcentury
period when was abandoned
late
Roman
in
the
period.
was subsequentlyresettled
Old Shifford Farm was the most recent excavation in this study. It is also characterized by
having the smallest sample of identified pottery forms and identified animal species. It has
been necessary,therefore, to emphasize different aspects of the artefacts and remains, such as
the fabric and method of manufacture of the pottery and the relative size of the animal species.
Using the words of Orton et al. 1993: 175, "even an assemblage that is `too small' by itself
larger
is
The
larger
in
form
the
some
grouping".
this
of
part
useful
case
a
grouping
may
the
of
context
artefacts and remains associated with eating and
social
and
archaeological
drinking.
The format of this chaptermirrors that of the precedingcasesstudies,commencingwith an

brief
description
then
the
a
process,
excavation
of the site and a commenton the
overview of
Following
is
discussion
this
the
archives.
site
a
of the consumptionpracticesof
site report and
distribution
inhabitants
the
patterns of the artefacts and remains from within the
and
the
The
discussion
the
the
chapter
will
conclude
settlements.
with
possible
of
a
of
context
imperialism.
the
context
within
observations
of
my
of
significance
155
7.2.
The Excavation
Old Shifford Farm was excavated by the O.A. U. in 1988-89 under the direction of Gill Hey.
Before the farmland was designated for gravel extraction, the site was contained under
land.
identified
The
through crop marks, although the recovery
site
was
arable agricultural
field
isolated
discoveries
dates
back
Before
finds
1800's.
through
the
to
the
walking
and
of
O.
U.
Tim
Allen
A.
in
1988.
The
the
of
conducted
a
site
assessment
assessment
excavation,
by
Oxfordshire
County
Council
for
the approval of gravel
requirements
with
complied
extraction.
7.2.1. The site assessment
Ten trencheswere excavated,amounting to a 2% site sample. The topsoil and plough soil
features
75%
by
dug
by
hand.
One
the
and
of
machine
were
of the aims of the
removed
were
"blank
in
to
the cropmarks, as these
so-called
consider
areas"
was
not
apparent
assessment
designated
for
in
(Hey
98).
It
1996:
were
gravel
extraction
particular
was established
areas
beyond
indicated
by
that
went
crop marks and that these `blank' areaswere
that occupation
densely occupied. The assessmentconcluded that there were two main occupation areas,
late
Roman.
Although
in
his
Roman,
Allen
the
that
other,
evaluation
concluded
early
one,
"Neither its state of preservationnor its characterdistinguishes this site from other gravel
terrace settlementsas especially worthy of preservation" (1988:5), Allen did nevertheless
His
follows:
the
the
of
site.
the site was part of the
reasons
excavation
were
as
recommend
Windrush Valley landscapewhich was (and still is) undergoing intensive study by the
O.A. U. (e.g. Allen 1990); the site presentedan opportunity of recovering the whole plan of
the site where at other sites there was more sampling employed; the site had good
it
because
dry
features;
was
a
site
potential
which
also
contained
waterlogged
environmental
the occupational sequenceof the site - there was a hiatus between early Roman and late
Roman settlementwould for the first time enable the analysis of adjacentsettlementshift
(Allen 1988:5), a phenomenonthat was quite common in the Upper ThamesValley (Fulford
1991). The assessmentalso concluded (with reservationsin light of the small sample size)
did
have
been
to
appear
not
two
wealthy or of high status.
settlements
that the
156
7.2.2. The main excavation
Two large areas were selected for open-area excavation, a southern area (Trench L)
be
Roman
focus
Iron
Age/early
this
the
to
and
of
study - and a northern area
considered
thought to be late Roman (Trench M). The excavation concentrated on the two central
"establishing
to
the stratigraphic sequenceof the enclosure
a
view
areas,
with
settlement
ditches, obtaining datable assemblagesfrom them, examining domestic featureswithin the
locate
(Hey
1996:
101).
Restricted
to
structures"
and
attempting
resourcesmeant
enclosures,
be
by
hand
down
largely
by
JCB
to
the
could
not
excavated
site
that
whole
and was
stripped
features
Fifteen
by
hand,
percent
of
site
gravel.
were
excavated
although, as the
undisturbed
late Iron Age/early Roman site was larger, more of that settlement was hand excavated.
Some of the enclosureditches from Trench L were èmptied' using a JCB to increasethe
finds.
Only
features
bone
in
the
of
groups
groups
of
of
animal
were recovered
samplesize
by
machine, no pottery and only small amounts of fired clay were
excavated
that were
large
bones
That
is
the
animal
were
recovered
mainly
understandable
considering
retrieved.
felt
"the
have
biased
the
that
the sample
excavators
extraction;
of
process
may
method
it"
(Hey
1996:
101).
than
enhanced
rather
Trench L encompasseda large settlementareaof 4,230 squaremeters. Discrete phaseswere
difficult to identify and the caseto be made for continuous occupation before and after the
for
Old
Shifford
is
Farm.
Assigning
strong
especially
particular contexts to
conquest
difficult
there
times
for
as
the site and
at
was
was
not
a
vertical
phases
stratigraphy
specific
isolated.
Phasing
determined
by the physical association
was
generally
were
contexts
some
features
by
dating
However,
supplemented
was
and
any
as many of
material.
particular
of
diagnostic
finds
and as the length of occupation was relatively
were not particularly
the
the
the
report qualified the phasing of the settlement thus: "The
of
author
short-lived,
is,
interpretation
here
therefore,
limitations
its
an
the
presented
of
and
evidence,
scheme
1996:
101).
(Hey
in
borne
be
mind"
should
157
7.2.3. The site (seeFigure7.1.)
The late Iron Age and early Roman period settlement at Old Shifford Farm was likely
late
hundred
between
Iron
for
The
the
two
one
years.
cut-off
points
approximately
occupied
Age phases and the early Roman occupation, however, are more difficult to isolate than at
Barton Court Farm, and the development of the settlement is described as "probably more
breakdown
by
indicates"
Curvilinear
(Hey
1996:
101).
than
a
continuous
phase
organic and
by
then
that
gradually
superseded
were
were
gullies
and
angular
enclosures
enclosures
Consequently,
in
found
in
the
the
cases
some
reused.
some
various
and
of
material
extended,
is
gullies and enclosures probably residual.
Late Iron Age settlement- Phaseone

The first phaseof settlementis situatedon the easternhalf of the site and is characterizedby a
large D-shapedenclosure.At the northeastentranceof the enclosureare a numberof irregular
indicate
D-shaped
Beyond
that
the
either
a
circular
structure
or
gated
pens.
post-holes
linear
been
have
Two
house
and
curvilinear
of
series
gullies.
a
sites
possible
enclosureare
D-shaped
the
to
the north of
enclosure
and
a
semi-circular
within
gully
one
suggested:
been
designated
have
house
density
These
because
as
the
areas
possible
of
sites
of
enclosure.
burnt
fired
stone and carbonizedplant remains at the terminals of the gully
clay,
the pottery,
and enclosure.
Late Iron Age settlement- Phasetwo

The secondphaseof the Iron Age settlementextendedwest of the earlier settlement,although
from
boundaries
have
been
(see
D-shaped
to
the
phase
one
appear
of
respected
some
longer
The
the
but
of
settlement
was
structure
no
curvilinear
more rectangular,the
enclosure).
interior of the main enclosurewas compartmentedand had a southern entrance.A possible
house site has been identified in the southwestcorner of the main enclosure abutting the
features
ditches
house
The
the
surrounding
and
of
suggested
site contained
entrance.
fired
deposits
burnt
the
clay,
of
part
of
wall
of
an
charcoal,
stone and
oven,
concentrated
158
likely
is
highly
bordering
It
that
there
the main
was
a
second,
northern
enclosure
pottery.
later
ditches
its
ditches.
Southeast
although
obscure
of the main enclosureanother
enclosure
identified
(and
Dthe
which
was
aligned
was
with
possibly
parts
of)
earlier
reused
enclosure
This
linear
enclosure
contained
a number of curvilinear and
gullies that
shapedenclosure.
be
The
to
the
to
controlling
access
parts
of
enclosure.
enclosurealso containsa stack
appear
have
been
house
incorporated
features
ditches
There
that
the
also
may
a
second
site
and
ring.
Deposits
burnt
fired
the
enclosure.
main
of
charcoal;
of
stone,
clay, an oven plate,
southeast
deposit
bones
for
largest
the settlement,as well carbonizedplant remainswere
the
of
pottery,
in
this
the
area
of
settlement.
concentrated
Early Romansettlement- Phasethree

In the third phaseof settlement,the main enclosurewas greatly enlargedand it encompassed
but
Former
boundaries
the
the
of
earlier
settlements.
of
each
were respected
elements
ditches
for
The
than
the
were
much
more
substantial
main
earlier settlements.
enclosure
initially
in
but
the
the
then the
sub-rectangular
with
was
entrances
north and
east,
enclosure
The
northwards
and
a
expanded
single
was
entrance
approachto
was
established.
enclosure
have
been controlled by a seriesof shallow, linear boundary
to
the settlementalso appears
ditches that formed a "funnel-like entranceinto the main enclosure" (Hey 1996:111). Two
identified:
have
been
house
first
located
the
sites
was,
within the sub-rectangular
possible
located
the
the
(this
was
second
within
and
square
enclosure
enclosureappearsto
enclosure
haveinitially beenrectangular).As with the housesites in the earlier phases,concentrationsof
burnt
including
and
clay
stone
carbonized plant remains were recovered
charcoal,pottery,
from theseareas.Other small finds, such as broochesand worked stone, were also found in
thesetwo areasof the settlement.
One of the conclusions in the site report is that the settlements are considered to have been of
low status and hardly Romanized during the early Roman period: "This part of the Upper
Thames Valley was only slowly affected by the Roman conquest, as demonstrated by the slow
introduction of wheel-thrown pottery and Romanized wares" (Hey 1996:97). Hey also notes
that in the late Roman period the "settlement type appears little altered from the preceding
97).
(1996:
period"
159
7.2.4. Site reportsand archives
The site report was published in 1996 in the journal Oxoniensia -a different format from that
for
in
its
length.
is
less
There
therefore
the
case
studies,
and
preceding
restricted
room
of
dialogue, for speculation and much of the contextual information exists only in the site
is
The
The
typically
to
report
site
organized
site
according
composition.
material
archive.
it
locate
to
the various reports and records than
organized,
and
well
was
was
easier
archive
for
in
it.
feel
It
however,
had
it
the
this
to
the
previous
sites
case
study;
an almost edited
was
is often the `thrown in' bits that are the most interesting and informative and I must admit I
in
disorder
bit
an archive!
of
prefer a
Unfortunately,I was not able to consult all of the primary records for the pottery as they are
from
location
is
the
their
site
archive,
and
presently unknown. It was possible to
missing
details
but
from
drawings
the
about
additional
some
pottery
and other archival records
retrieve
for the most part, I had to rely on computerizedaccounts(one by SarahGreenwhich is in the
by
for
Jane
Timby),
I
the
other
provided
and
which amalgamated the purposeof
site archive
diameters
The
is consequentlyincomplete (see below), although
of
rim
this study.
record
information about burnt residuesand sootingwere detailed on the computerizedaccounts.It
bone
in
into
database,
the
to
they
enter
primary
records
existed
a computer
as
was necessary
hand-writtenform only.
7.3.
The Archaeology of Food and Drink Consumption at Old Shifford Farm
As with the previous two casestudies,this section will commencewith a discussionon the
drinking,
include
to
the
the methodologiesand conclusionsof
and
of
eating
remains
natureof
the various specialists,as well as my own analysisof the data. The distinction betweenthe
be
beginning
the
and
specialists
my
own
analysis
the
various
will
of
at
of the
outlined
work
The
follows:
`containers'
artefacts
and
remains
are
grouped
subsections.
as
relevant
- which
in the caseof Old Shifford Farm only refersto the pottery - òtherimplements'which includes
finally
'ingredients'
clay
artefacts
and
and
stone
which include animal and plant
metal,
remains.
160
Thespecialists' reports- thepottery

The pottery analysis was started by Sarah Green (who recorded and described the assemblage)
but was completed by Jane Timby, who expanded and customized the analysis according to
O.A. U. 's current fabric/typology/recording system. Timby also wrote the published site report.
The current typology used by the O.A. U. differs slightly from that originally used by Sarah
Green. The main difference lies in the creation of a category for `jarlbowl' and in the
`bowl'
According
Green's
designation,
type
to
certain
vessels.
of
which views
specification
bowls as open vessels,only one suchvesselwas identified in the whole of Trench L, whereas
Timby assigneda numberof Green'sneckedjars as bowls in her description of the illustrated
in
is
incongruity
if
The
impart
typology
to
particularly
significant
one
was
particular
sherds'.
bowls
Rice
79
(see
Orton
1993:
the
type
cultural
and
practices
onto
regimes
and
et al.
serving
1987:209-10 re: subjectivity of typology), and as such my interpretation of the typology is a
fusion
I
for
Timby
the
two
of
specialists'
accounts.
understand
why,
example,
cautious
illustrated
`bowls'
bowls,
`bowls'
the
these
as
of
although
some
were used
whether
classified
found
is
indicate
`bowls'
that some
the
since
residues
unclear
on
a
number
ware
of
as serving
for
Furthermore,
because
jar
Green's
cooking.
types,suchas
used
the
were
some
of
vessels
of
jar 4 andjar 10, are eachclassedas a bowl, a bowl/jar and a jar by Timby, I could not assume
bowl-like.
Timby's
jar
10
As
4
jar
typology is basedon the illustrated sherds
all
were
and
that
(seeTimby 1996:130-133),I have listed in the figures on pottery forms for eachphase,only
illustrated.
Similarly,
jar/bowl
bowls
the
that
were
classificationhas also beenlisted for
those
I
Green
to
these
jars.
is
tend
did,
This
view
vessels
as
although,
a rather
as
each phase,
jaribowls
bowls
be
depending
that
to
and
say
may
way
under
or over represented,
complicated
have
This
had
typology
would
you
prefer.
more of an affect on the analysisof the
on whose
identified
form
if
the
(see
below).
The
to
were
pottery
of
more
confusion over
pottery
did
however,
serveas a reminderof the often-subjectivenature of classifying pots.
typology,
What is apparentfrom the pottery assemblageat Old Shifford Farm, is that for eachphaseof
is
identification
below);
dishes,
(see
of
serving-type
the
ware
not
obvious
cups,
and
settlement
fabric
form,
focused
than
the two differing typologiesbecameapparentin the
on
vessel
rather
Timby's analysis
descriptionsof illustratedsherdsand not on Timby's computerprintout.
161
identified
beaker
bowl
at
all
only
single
were
not
and
examples
of
a
shallow
and
a
platters
in
Roman
The
the
early
period
phase.
potteryassemblageat Old Shifford Farm
were recovered
from the three other sites
in this regarddiffers markedly from eachof the pottery assemblages
in this study.
Nine hundred and sixty-five pottery sherds were recovered from Trench L and contemporary
in
Of
384
trenches.
the
these
assessment
sherds,
sherds contributed to the
contexts
identification of 71 vesselsidentified to type. Spreadover threesettlementphasesthe numbers
is
based
is
types
the
of
pottery
percentages
often quite small.
which
on
7.3.2. The containersat the late Iron Age settlement
Thelate Iron Agepottery -phase one and two

In both phaseone and phasetwo, jars are the most common vessel type (see Figure 7.2.).
Immediately apparentis the absenceof identifiable specializedforms. No dishes,cups and
identified
in
beakers,
were
either phase.This is contraryto a variety of late Iron
surprisinglyno
Age settlementsin the Upper Thames Valley, including Ashville Trading Centre,Thornhill
Farm and Barton Court Farm, which while lacking large numbers of specializedvessels,did
haveat leasta few examplesof specializedwares.
Phase 1 (LIA)
Phase2 (LIA)
Phase3 (ER)
%
%
%
n=86
n=167
n=686
10
12
70
temper
106
42
15
shell
11
13
2
limestone temper s
4
1
1
12
14
16
16
10
2
sandy wares
27
31
0
Flint temper
2
0
29
.
26
30
79
493
temper
47
72
grog
0
0
0
Malvernian
65
9
0
Table 7.1. LateIron Age- phaseoneandtwo - andearlyRomanperiodpotteryfabrics
162
Form, function and fabrics- phaseone
In phase one, of the four vessels for which manufacturing could be determined, three of the
jars
handmade
Four
the
types
and
other was wheel-finished.
and one type of
of
vessels were
bowl were identified to form in this phase (the cordoned necked bowl was wheel-finished).
The jars include three vessel types that are quite similar to each other, a vessel with upright
fourth
jar
had
known
`Belgic
the
type'
a
slack
and
profile
vessel;
a
vessel
a
as
walls, a globular
body.
As
and
a
grooved
was stated above, most of the primary pottery
shoulder
pronounced
L
for
diameters
be
for
Trench
two
are
missing,
and
only
rim
phase
could
established
records
bowl
cordoned
and a slack profile jar - had a diameter of 18 cm. A
identified
fabrics
(see
Table
7.1.
),
fabric
had
the
were
a
of
most
commonly
occurring
number
both
vessels -a
one,
fabrics
flint.
As
tempered
temper,
other
were
with
no
and
shell/limestone,
sand,
and
grog
found
for
the
manufacture
was
on
pottery
settlement, the presence of a variety of
evidence
fabrics may indicate that the inhabitants were active consumers (Lambrick 1984:170), despite
imported
lack
and
wares.
the
of specialized
Form, function and fabrics- phasetwo

In phasetwo, grog-temperedBelgic typejars with evertedrims and pronouncedshouldersare
followed
by
types,
shell/limestonetemperedvesselswith upright
vessel
the most prominent
bowls
jars.
drilled
A
holes
base
cordoned
and
two
single
post
with
walls and grog-tempered
firing was recovered.Types that are typically associatedwith cooking were identified (see
Woods 1986;Hendricksonand McDonald 1983:631). Ajar with upright walls and beadedrim
it
is
for
the
and
possible
sooting
with
absence
exterior
suspensionthat
of a means
exhibited
directly
hearth
onto a
or oven floor (Orton et al. 1993:222; Evans
this vessel was placed
1993:105). One of the vesselswith an evertedrim, rivet holes and burnishing may have been
fire
burnishing
is
increase
thought
to
open
an
the heat efficiency of vessels
over
suspended
(Orton et al. 1993:221). Burnishing is also thought to decreasethe porosity of coarse-ware
liquids
(Hendrickson
McDonald
633).
Burnished cooking
to
1983:
and
contain
used
vessels
been
have
(although
1984:
43
to
Oetgen
therefore
used
cook
stews
see
may
who considers
pots
burnishingin terms of its aesthetics).Rim diametersof six vesselswere identified; they ranged
from 13 cm. to 23 cm, with an averageof 17 cm. Six of the ten vesselswhose manufacture
handmade,
Belgic
jars
two
distinguishable
type
were
were wheel-finished and a necked
was
163
jar
include
Vessel
fabrics
bowl
Belgic
type
wheel-turned.
were
most
and a cordoned
cordoned
for
(see
Table
),
flint
fabrics
identified
7.1.
tempered
those
although
phase
one
no
were
of
identified, grog was again the most common temper followed by shell/limestone. The
fabric
form
is
be
between
there
to
an
and vessel
not absolutealthough
appears
correlation
Belgic
between
type vessels with grog temper and slack-profiled
and
necked
association
Both
for
(Evans
temper.
tempers
are
shell/limestone/sand
appropriate
cooking
with
vessels
1993:107).
Discussion
It was through the analysisof the pottery at Old Shifford that I experiencedfirst-hand how
be,
in
tableware
the
of
consideration
can
particularly the absenceof more obvious
subjective
is,
blurring
distinction
forms.
The
between
forms
late
in
Iron
Age
the
the
of
vessel
specialized
frustration,
day
by
desire
for
functional
That
perpetuated
modem
a
a
categories.
neat
of course,
did
`data'
the
not match my notions of cooking and serving forced me to considerthe pottery
in different ways and remindedme that the contentsand contextsof the vesselscould at times
be of more importancethan the shapeof the vessel or its point of origin. (see commentsin
Pluciennik(1997:48-9) who emphasizesthe importanceof `coarseware' in the rituals of daily
life). At the late Iron Age settlementat Old Shifford Farm the lack of non-local, imported and
forms
is
in
from
the
this
sets
the
what
settlement
specialized
apart
other
settlements
apparently
from
in
indeed
does
Upper
This
the
Thames
Valley.
the
many
of
settlements
not
study and
impoverished
in
definition
too
that
the
to
settlement
was
the
mean
of status
engage
necessarily
it
had
different
but
feasts
that
status
the
markers
such
possibly
or placing
as
giving of
markers
importanceon animalsand the way that they were consumed(seebelow).
7.3.3. Other implementsat the late Iron Age settlement- phaseone and two
A number of objects have been recoveredwhich have an associationwith the preparation,
food.
Oven
in
disk
walls
were
of
that
two
recovered
storage
and
as
was
clay
phase
a
cooking
been
1996:
have
to
(Barclay
138).
wooden
or
clay
cover
storage
used
containers
et
al.
may
Similar disks describedas "possibly lids for storagevessels" Sanders1979:54,53 fig. 28)
Farmoor
(Lambrick
Roman
1979:
This
Robinson
54).
period
and
at
could
were recovered
164
indicate that vesselsmade out of other materials- possibly of some value - were presentat
the settlement.A small number of metal objectswere also recovered,including a piece of an
iron knife found in a phasetwo context, which could havehad a culinary function.
7.3.4. The Ingredients
bones
Thespecialists' reports - animal
Priscilla Lange prepared the animal report. The sample size for the whole site is quite small,
949 bones. In Trench L, 508 bones were recovered. Just over 50% (255) of the sample from
Trench L was identified to species although the unidentified bone was placed in general size
Over
90%
below).
bones
The
(see
from
the
of
contexts.
were recovered
stratified
categories
but
fragmented,
butchery
Lange,
the
preserved
quite
to
well
result of, according
collection was
degree
lesser
handbone
the
The
to
a
much
excavation
process.
was primarily
practices and
bone
Additional
during
from
thirty-two soil samples that
excavation.
was
recovered
collected
it
is
floated
interesting
bones
bird
fish
and
particularly
that
and
were
or
no
were sieved
identified in any of the soil samples. Lange made a number of observations not made at the
in
is
bone
thesis,
burnt
this
the
such
as
recording
colour
of
which thought to
other case studies
indicate the method of cooking and the grouping of unidentified animal bone according to the
i.
(USM)
hare
sized
e.
small
mammals
species,
of
size
which
are
size; medium sized
relative
(UMM)
large
(ULM)
are
sheep/goat/pig
which
size
and
sized
which are
mammals
mammals
1973).
Uerpmann
(after
cattle/horse size
In her analysis,Lange mergedthe late Iron Age phasesone and two together to increasethe
for
late
Iron
the
the
Age/early Roman collection as a
considered
most
part
and
size
sample
from
light
In
late
the
in
Iron
seamless
almost
shift
to
of
the
phase
phase, especially
whole.
is
Age, the merging of phases understandable.Lange did observe that the "proportions of
different speciesvary little through time, with the exception of horse bone, which is much
(1996:
150).
Lange
3"
in
Phase
from
bones
Trench
the
also
placed
common
amalgamated
more
L into groups of body parts of speciesaccording to the amount of meat they bear (after
Uerpmann 1973) and accordingto their distribution in various featuretypes. For the purposes
165
for
in
have
I
the
analysed
each
phase
separately,
this
most
part
order to reveal
study,
of
between
distinctions
the
three
phasesof settlement.
possible
The determinationof the ageand sex ratios of the main speciesis affected by the samplesize.
Lange amalgamatedthe late Iron Age and early Roman bones in order to establish some
Lange
be
killed
between
that
tended
the
to
attributes.
concluded
various
cattle
of
repetition
four
found
in
half
there
to
although
are
years
examples
each phaseof earlier and
a
three and
later mortality rates. The mortality of sheep varied, often representedby few examples,
including foetus, under two and a half years and under three and a half to four years. Pig
indications
The
killed
be
two
to
under
years.
are that cattle, sheep and pigs were
tended
food.
for
Horses,
hand,
died
is
likely
It
the
the
that
on
other
all
over
age
of
six.
primarily used
late
Iron
Age
Roman
had
the
the
at
and
early
animals
settlements
multiple purposes:
someof
for their wool, as draughtanimals,aswell as for dairying and meat (Lange 1996:158).
Thespecialists'reports - theplant remains

Sixteen sampleswere taken from features(ditches and gullies) in Trench L. The samples
floated
0.5
"were
litres
for
12
over
and
a
mm.
mesh
and sorted
carbonizedplant
averaged
Features
less
1996:
159).
items
(Robinson
than
twenty
totals
to
the
with
were
added
remains"
for Trench L. Mark Robinson, who did the botanical analysis, concluded that the late Iron
in
"open
landscapewith little woody vegetation"(1996:
Roman
was
situated
an
site
Age/early
162), and that the site was probably surroundedby grassland.Animal husbandryis suggested
by the layout of the small enclosuresand the presenceof scarab dung beetle (see also
Robinson 1983:41) and grassyplant species.Robinson also concluded that, although many
identified
be
by-products
as
arable,
their
cultigens
only
and
could
were abundant
species
indicate
to
possiblecultivation.
enough
listed
in
the site report, although, for the most part analysiswas based
Individual contextsare
from
L.
The
Trench
dominated
by
overall
assemblage
the
assemblage
was
cereals,
entire
on
following
breakdown:
variety
of
and
wide
the
weed
chaff
species
of
with
small amounts
"51.4% grain: 2.3% chaff 46.3% non-cerealseeds"(Robinson 1996:166). Robinsonsuggests
from
best
"is
the
the parching, de-husking and fine sieving of
waste
seenas
that the sample
166
hulled cereal grain" (1996: 166). As to whether the settlement was an importer or producer of
both
(Robinson
difficult
is
Robinson
it
that
to
say
suggesting
are
possibilities
with
cereals,
1996: see also van der Veen 1991). In comparing the concentrations of carbonized plant
(which
Guy
(which
had
Gravelly
Farm
from
Thornhill
and
a pastoral economy)
remains
between
determined
lies
Old
Shifford
Farm
Robinson
that
somewhere
practiced agriculture)
by
for
is
It
the
that
the
production was small-scale
consumption
also possible
the two.
inhabitants (van der Veen 1991:357). What we can assume is that cereals and possibly other
(querns
have
been
the
site) and
were
processed
at
species
and
ovens
recovered
plant
edible
consumed on site.
7.3.5. The Ingredientsat the late Iron Age settlement
Thelate Iron Age animal remains (phaseoneand two)
The N. I. S.P. for eachspeciesin phaseone indicatesthat cattle boneswere the most frequently
bones
bones
horse
that
sheep/goat
and
pig
were
are
equally
and
species,
represented,
recovered
Figure
7.3.
No
dog,
(see
).
deer
bird
identified
in
bones
least
this phase,
or
the
represented
were
USM
identified.
is
frequency
four
The
the
of
were
species
percentage
main
of
althougha small
further suggestedby the frequency of unidentified large and medium sized species.Large
horse)
dominant
(ULM:
the
followed
by
are
and
cattle
most
medium sized
group
species
In
identified
two,
(UMM:
the
and
pig).
phase
sheep
cattle
are
again
most commonly
species
better
horse
represented
are
sheep
and
and
pig
are now equally represented
species,although
(seeFigure 7.4.). Dog and deerwere identified in this phasein small numbers.The shift in the
frequency of the four main speciesis also mirrored in the groups of unidentified species,
identification
highly
fragmented
(see
this
type
bones
to
the
of
credence
when
are
which adds
favours
700
1990:
Crader
by
identification).
this
who
also
method
of
comments
167
Butcherypractices
A variety of body parts,including feet, from the variousspeciesfrom each group of meatyield
from
late
Iron
Age
that
two
the
one
and
of
phase
suggests
settlement,
which
recovered
were
brought
`dressed'
to the settlements (Wacher and McWhirr
and
the species were not
1982:213) andwere probablybutcheredon site. The butcheredbonesindicate that the animals'
bones were both chopped and cut. As was stated above, the bone sample was highly
fragmented,many of the boneswere broken up but do not display any butchery marks. The
discussion of the butchered bones will focus on individual butchery marks found on the
butchery
than
to
attempt
establish
rather
a
patternas such.
species,
various
Of the small numbersof bonesrecoveredin phaseone, few have butchery marks. Cut marks
found
on cattle, on a mandible presumablyto remove the cheek meat and on a
were only
disarticulate
bone
from
humerus.
found
Chop
the
to
the
on two
radius
marks were
proximal
dismemberment
bone
the
tibiae
and
probably
represent
a
radius
and
of
and/or
recovery
cattle
femur
display
The
Butchery
a
sheep
and
of
mandible
of
a
marks
pig
chop
marks.
marrow.
horse
bones.
Only
found
ULM
bone
had chop marks whereasa
the
any
of
on
one
were not
bones
have
UMM
identified
butchery
the
the
that
chop
marks,
which
suggests
of
of
number
be
In
larger
two,
might
under-represented.
the
species
phase
samplesize
slightly
medium sized
bones,
in
butchery
on
a
wider
variety
of
marks
although
no cut marks were recorded
reveals
long
is
In
blade
bones
the
to
chop
marks
on
the
shafts
addition
of
this phase.
scapula
of cattle,
is
Butchery
the
the
of
sacrum
wing
chopped.
right
marks on sheepwere minimal;
choppedand
heavily
indicate
to
the
and
chops
chopped.
tibia
was
pelvis
points of
may
and
a
mandible
one
disarticulation.Pigs are only representedby a non-butcheredmandible and someteeth. Horse
display
butchery
but
do
broken
bonesare
any
not
up
marks.Interestingly,there are no obvious
butcherymarks on the phasetwo ULM bones,whereasphaseone UMM bonesappearmore
butchered.
for
The absenceof evidence the trimming of meat, togetherwith the fragmentedand chopped
indicate
bones,
in
both
that
the
have
been
of
there
two
phases
many
of
one
could
and
up nature
1990:
(Crader
710).
for
In
bones
from
cooking
the
one-pot
phase
medium
one
a preference
burnt,
in
from
large
bones
both
two
commonly
whereas
the
more
phase
were
and
sizedspecies
indicate
burnt,
differential
which
may
were
roasting of specific species,
medium sized species
168
Luff
1994).
indicate
(Pearce
do
bones
burnt
the
and
roastingof meat
not necessarily
although
had
been
differently
they
treated
the
that
is
eten.
It also possible
once
specieswere
Meat yields
The sample size of the identified to speciesbones at the two late Iron Age settlements is quite
for
bones
hesitate
than
I
the
to
their
presenting
according
meat yield other
small and
illustrative purposes (see Figure 7.5.). Some species, for example, are represented by less than
bones
horse
horse
in
in
That
(horse
the
two).
bones
five
phase one and
and pig
all of
phase
for
from
highest
À'
in
two
the
the
were
group
group with
meat yield phase
recovered
impact when you consider that we are dealing with a sample of two bones
its
loses
example,
(although see below discussion of the contexts of the bones). Suffice it to say that in terms of
`B'
bones
late
Iron
Age
the
settlements, group
recovered at
the most commonly represented
bones were the most common for each phase.
Theplant remainsat the Late Iron Age settlement(phaseone and two)

As with the animal remains, I have looked at the plant assemblageaccording to eachphase.
identified
barley
from
to
were
species, and chaff
In phase one, oats and
either spelt or
breakdown
identified.
The
deposits
the
was
of
carbonized
was: grains
emmer wheat
60%,
5%
chaff
and non-cereal seeds35% (total number of
(identified and non-identified)
43,
Very
litres
litre).
34
items
items
1.3
thus
of
soil
volume
processed
=
=
per
carbonized
few of the carbonizedweed specieswere edible,they include dock and cornsalad.
including
barley,
spelt wheat and oats were identified to species- barley
wheat,
In phasetwo,
The
breakdown
dominant
deposits
than
the
wheat.
of
was: grains
carbonized
was slightly more
51%,
1%
chaff
and non-cereal seeds48% (total number of
(identified and non-identified)
few
84,
31
litre).
items
2.7
items
Again
thus
of
soil
processed
volume
=
=
of
per
carbonized
include
identified
dock.
flax,
they
were
edible,
small
amounts
and
of orache,
the weed species
for
both
identified
phaseswas quite low, although it is possible
The number of plant remains
by
inhabitants.
for
The
barley
(although
the
consumption
of
grown
prominence
that cropswere
169
have
been
(Green
1981:
139-40)
also
used
andthe presenceof ovens,possibly
other grainsmay
for malting (van der Veen 1989:304) could indicatethat beerwas producedat the settlement.
7.4.
Food and Drink Consumption at the Early Roman period Settlement
7.4.1. The Containersat the early Romanperiod settlement
Form and function
The sample size at the early Roman period settlementwas larger than both phase one and
forty-eight
identified
beaker
identified
One
to
two
type.
vessels
were
combined;
and
was
phase
jars
jaribowls)
(and
types
the
of
are similar to thoserecoveredat the other early Roman
while
in
in
does
forms
the
include
types
this
study,
range
to
of
pots
not
appear
other than
settlements
jars and, to a much lesser extent, bowls (see Figure 7.6.). No samianware or mortaria were
is
late
from
Iron
Age
the
and
a
single
settlement,
sherd
of amphora
context
a
recoveredat
inconclusive.Indeed,comparedto other early Romanperiod sites in the area,the early Roman
Old
Shifford
had
hardly
identifiable
at
any
settlement
specialist wares (Allen 1990;
period
Booth in press).Interestingly,when settlementat Old Shifford Farm was re-establishedin the
forms
beakers,
tankards,dishesand Romanstyle wareswere
such
as
specialized
third century,
latest
in
Timby
the
phases
of
occupation.
suggeststhat the lack of Romanized
recovered
only
items,
indicate
"lack
inability
to
of
access
such
a
to purchasefine tablewareor
may
an
wares
lack
desire
for items clearly related to
oil
and
wine,
or
simply
as
olive
a
such
of
products,
Roman eating and drinking habits" (Timby 1996:129). The possible significance of the
Roman
Romanized
the
be
at
early
pottery
period
of
settlement
absence
will
examinedbelow.
As hasbeendiscussedin previouschapters,specializedforms with specific uses,especiallyfor
Roman
food,
(Millett
the
1979).However,to concludehere
of
period
characteristic
are
serving
dining
by
the
association
practices,at Barton Court and Roughground
that the pottery, and
Farms were more Romanized(in a native non-Romankind of way!) than the pottery at Old
170
Shifford, is prematureand ultimately meaninglesswithout considering other aspectsof the
.
the
the
the
the
within
context
pottery
of
other
studying
at
and
remains
recovered
pottery,
settlement.
Fabrics
Some of the observeddistinctions betweenthe jars, for example, require further comment.
Certain forms appearto have a more obvious associationwith function. We seefor the first
jars,
jars,
short-necked
wide-mouthedjars and vesselswith perforated
time, narrow-necked
bases.All of these specializedjars were temperedwith grog, as were the neckedjars and
bowls and Belgic jars. The majority of the vessels(at a ratio of almost 3: 1) are also wheelhad
burnt
burnt
the
A
vessels
sooting
or
of
exteriors
and some even contained
turned. number
(or
burning)
include
The
jar
evidence
of
cooking
with
vessels
a
short-necked and
residues.
bowl, a wide-mouthedjar and a thick-rimmed bowl. The fabrics of these vessels were
temperedwith either grog or shell/limestoneall of which are suitable for heating (Woods
1984:27).
The proportion of identified forms is, as with the late Iron Age phases,proportionally low;
identified
both
fabrics
the
and non-identified forms at the early Roman settlement
however,
of
found
late
Iron
Age
(see
In
the
that
Table
7.1.,
7.3.2).
at
settlements
as
the
same
section
are
local
In
they
Roman
exclusively
almost
are
with
the
one
exception.
words,
early
phase,
other
increasingly
handmade,
than
wheel-thrown
rather
were
the only exampleof nonwhen vessels
Malvernian
hand-made
local wares were
pots (see also Lambrick 1984:170; and Green 1980
for other examplesof the persistenceof handmadepottery in the Roman period). One of the
burnt
residues.
Malvernian pots contained
Rim Diameters
diameters
higher
than for the late Iron Age phases(see
was
slightly
The recording of rim
however,
histogram,
The
).
7.7.
should be viewed as a presentationof the rangein the
Figure
171
diametersof selectpots, rather than the identification of particular patterning for the whole
assemblage.
7.4.2. Other implementsat the early Romanperiod settlement
A number of implements were recovered at the early Roman settlement that have an
have
been
food.
include
fired
These
that
preparing
clay objects
associationwith cooking and
identified aspart of oven walls and disks that may havefunctionedas lids. Part of an iron knife
Whittle
knife
A
found
tang
complete
was
which although un-stratified
was also recovered.
has beendatedto the mid-first century AD and may have had a culinary use in either the late
few
identified
A
lead
been
Roman
have
Age
Iron
phase.
un-stratified
small
early
or
weights
for
fishing
have
been
(Allen
Glass
have
1996:
142-144);
these
nets
may
used
and
that may
been used in either the late Iron Age or early Roman period. A small number of stone
in
including
this
were
with
grinding,
recovered
phase,
quern stonesand
associated
objects
145).
Burnt
limestone,
Glass
1996:
in
(Roe
the cooking process,was only
and
used
rubbers
found in Trench L; its nearestsourceis eight kilometers away in the Witney area.
7.4.3. Ingredientsat the early Romanperiod settlement
Roman
bones
the
Theanimal
early
period settlement
at
At the early Roman period settlement, cattle was the most commonly identified species
followed by sheep/goat,and in contrastto the late Iron Age phaseof settlement,horseswere
followed
by
dominant
(see
Figure 7.8.). The increasein percentage
pigs
species,
third
the
most
by
increase
in ULM. Coincidentally or not the percentage
further
is
the
horses
suggested
of
identified
horse
is
dog
Small
ULM
cattle
the
of
and
percentage
combined!
numbers
of
and
of
deerwere identified at the settlement.
172
Butchery practices
Cattle long bones- humerus,radius - at the early Roman settlement were often chopped
longitudinally, which suggeststhat boneswere split for their marrow (Maltby 1985a:50 has
is
in
in
Iron
Age).
Roman
the
than
the
that
common
extraction
more
period
marrow
observed
Cut marks, as with the first late Iron Age phase,occurredmainly on the bones of cattle, and
horn,
indicate
the
the
the
to
of
removal
skin
and
cheek
meat
and
particularly
appear
dismembermentof body parts.There is lessevidenceof the butcheryof sheepthan in the late
Iron Age phases,the headdoesnot display any butcherymarksand neither do the long bones.
The neck of a scapulahas been choppedwhere it attachesto the humerus, and chops on the
dorsaledgeof the ischium indicatethe separationof the pelvis from the femur. Other evidence
found
butchery
fewer
Pig
bones
have
butchery
the
tarsal
on
and
metatarsals.
was
also
of
found
butchery
isolated
on a radius,an ulna and a mandible.
marks
marks;with
Horse bones, on the other hand, are clearly butcheredat the early Roman settlement.Chop
found
it
broken
the
the
to
that
on
cranium
and
appear
proximal
suggest
up,
was
were
marks
distal
femur
the
was also choppedas was
end of a metacarpal and a couple of
end of a
bones
butchered
Few
the
were
major
meat-bearing
consumptionof
which
makes
phalanges.
horsemeatat the early Roman settlementless conclusive than was the case at Barton Court
Farm. In contrastwith both phasesof late Iron Age settlement,more of the bones from the
ULM group are butcheredand only a few of the UMM bones display butchery marks. This
in
increase
butchery
found
in
decrease
horse
bones
the
the
to
marks
on
parallel
and
seems
bones.
found
butcherymarks
on sheepandpig
Meat yields
The samplesize of the identified species,though still quite small, is larger at the early Roman
bones
The
indicates
bones
to
their
of
according
that
grouping
meat yield
period settlement.
from group `B' continued to be the most commonly identified bones; once again, horsesare
7.9.
).
The
Figure
(see
varying proportionsof body partsare actually quite similar
the exception
Figure
7.5.
(see
7.3.5.
bones
Age
`C'
Iron
)
that
and
section
assemblage
the
to
except
group
for
both
horse.
identified
sheep
and
were more commonly
173
Theplant remainsat the early Romanperiod settlement
At the early Roman settlement, a number of plant species associated with middens (henbane)
is
It
identified.
(red
toad
rush) were
goosefoot and
and muddy areas such as animal corrals
further suggested,through the presence of waterlogged plant species and particular species of
Anisas
leucostoma,
Armiger
Planorbis
that
(Lymnaea
truncatula,
of
some
crista)
and
mollusc
(Robinson
below)
ditches
(see
`slow-moving'
contained
stagnant
water
or
the enclosure
1996: 162-3).
The early Roman sampleshave higher numbersof carbonizedcerealsand chaff than in the
barley
items
in
Wheat,
is
but
the
numbersof
eachsample still generally small.
earlier phases,
Age
late
Iron
identified
identified
to
the
the
species
and
at
as with
cereals
and oats were
breakdown
barley
The
the
and
seeds
outnumbered
carbonized
slightly
wheat.
of
settlement,
items
2%
47%
51%
(total
=
chaff:
grains:
non-cerealseeds
number of carbonized
chaff was
items
62
litres
135
items
6.3
litre).
However,
thus
of
390; volume of soil processed=
per
in
(measured
terms
and
variety
chaff
a
wide
of
non-cereal
seeds
waterlogged
wheat
silicified
These
deposit.
àbundant')
identified
in
Roman
`several',
`present',
and
were
also
one early
of
is
that
the
to
a
suggest
simply
samples
ratios
above
of
grain
chaff
presented
non-carbonized
breakdownof the carbonizedsamplesand may not representproduction/consumptionpatterns.
Waterloggeddepositsreveal a rich variety of plant species.Opium poppy was found in early
Roman contexts,possibly cultivated, which could have been used in the production of oil or
Mint,
Roman
identified
turnip
the
parsley
and
wild
early
were
also
at
possibly medicinally.
hen,
Other
fat
have
include
been
that
species
edible
may
consumed
phase of settlement.
This
late
Iron
Age
that
the
suggests
nettle.
and
waterlogged
with
absence
of
orache,
common
have
deposits, edible speciesmay
made more of a contribution to the diet at the earlier
settlements.
late
Iron
Age
the
settlement,the prominenceof barley could indicatethat
As was suggestedat
beer was producedat the settlement,malting being one possible function of the ovens that
der
Veen
1989:
(van
304).
identified
been
have
174
Summary of the Artefacts and Remains of Consumption at each settlement
7.5.
Before we consider the distribution of the artefacts and the remains at the late Iron Age and
it
is
to
Roman
settlements,
necessary
summarize some of the more salient
period
early
for
been
implements
ingredients
have
the
the
the
that
made
on
containers,
other
and
points
each phase.
7.5.1. Late Iron Age settlement- phaseone and two
The containers
Pottery
jars
dominant
form.
bowls,
dishes,
No
In
two,
the
and
are
one
cups
phases
shallow
"
identified
in
beakers
either phase.
were
or
determined
In
the
technique
vessels
whose
most
of
manufacturing
one,
was
phase
"
Belgic
handmade,
jar
though
type
one
was wheel-finished.
were
had
in
jars
fabrics
The
one
phase
similar
were
profiles, although a variety of
"
identified at the settlement. Grog-tempered fabrics were the most commonly
fabrics
used.
occurring
identified
jar
in
jar,
Belgic
The
two
the
type
commonly
phase
most
was
which was
"
however,
grog-tempered
and
six of the ten vessels whose
wheel-finished
identified
handmade.
were
was
manufacturing
indicate
both
the
In
two
sooting
of
exteriors
that
of
pots
phase
suspended
pots were
"
directly
fires
hearth.
the
and
placed
on
above
have
I
both
importance
For
two
that
one
and
suggested
the
phase
of the pottery may
"
in
its
been
but
have
place
of
origin
or
style
rather in the methods of preparing
not
food,
food
drink
itself
it
how
the
and
cooking
or
possibly
and where was
and
consumed.
Other implements-phase one and two
identified
in
have
disk
Clay
two
that
were
phase
walls
as
oven
well
may
clay
as a
"
been used as a lid for a container.
found
knife
in
have
had
two
was
metal
a
a
of
phase
piece
a
context, which may
"A
culinary use.
Ingredients
Animal bones- phaseone and two
"
Cattle were the most commonly identified speciesin phaseone, followed by sheep,
horse.
and
pig
175
"
"
"
"
No dog, deer or bird boneswere identified in phaseone.

In phase two cattle are again the most commonly identified species, followed by
in
better
horse
than
are equally
phase
one;
pig
and
are
represented
which
sheep,
represented.
Dog and deerboneswere identified in phasetwo in small numbers.
The frequency of the various species in both phase one and two is also suggested by
the frequency of bones that have been grouped according to animal size.
bones
distribution
too
to
The
to
the
small
size
was
give
of
great
significance
sample
"
Bones
from
`B'
(moderate
to
their
meat
yield.
group
meat-yielding) were
according
identified
bonesfor both phases.
the most commonly
did
in
bones
both
fragmented
but
two
they
The
not
phases
one
and
were quite
animal
"
display a lot of evidenceof the butcherypractice.
bones
the
In
the
the
species
only
one
with
other
cut
of
marks
was
cattle,
phase
.
displayed
chop marks.
species
identified
horse
bones
for
butchery
No
were
marks
on
either phase.
"
bones
butchery
In
two,
with
obvious
marks were chopped and no cut marks
phase
"
butchered
bones.
identified
the
on
of
any
were
for
fragmented
lack
The
trimming
the
evidence
of
of
meat
and
and chopped up
"
indicate
bones
in
for
that
the
there
could
cooking
of
was
a preference one-pot
nature
both phasesone and two.
in
from
bones
burnt;
In
the
medium sized specieswere more commonly
phaseone
"
from
bones
it
large
This
burnt.
the
two
that
was
could
species
were generally
phase
be taken as evidence of differential roasting of the species between phases of
different
killed.
treatment
the
they
of
possibly
or
of
speciesonce
were
settlement
The plant remains- phasesone and two

Small numbers of carbonized plant remains were recovered from samplestaken in
both phaseone and two.
Barley and oats were identified to species in phase one, the breakdown of the
deposits
60%,
5%
35%
grains
was
chaff
and
non-cereals
carbonized
Wheat, barley and oats were identified to speciesin phasetwo, the breakdown of the
51
deposits
%,
I%
grains
was
chaff
and non-cereals48%
carbonized
identified
in
flax
including
dock,
both
weeds
of
edible
were
number
small
phases
"A
and orache.
is
Robinson
to
late
Iron
Mark
undecided
as
whether
the
crops
at
produced
were
"
Age-early Roman period settlement. Production may have been small-scale for
inhabitants.
by
the
consumption
176
Thecontainers
Pottery
"
Jars continue to dominate the assemblage; a small number of bowls and one beaker
were identified.
"
"
No samianware, mortaria or amphoraewere recoveredfrom the settlement.

There was an apparent increase in the specialization of jars, narrow-neckedjars,
jars
identified.
jars
bases
have
been
and
with
perforated
wide-mouth
The majority of the jars are wheel-turned,although hand-madejars were still used at
the settlement.
Similar local fabrics identified in the late Iron Age phaseswere found in the early
Roman period assemblage- with one exception, the only example of non-local
hand-made
Malvernian
One
burnt
these
pots.
of
pots
contained
residues.
were
wares
"
"
Other implements
"
"
"
Clay ovens parts and fired clay disks that may have functioned as lids were
recoveredat the settlement.
Part of an iron knife was recoveredand an un-stratified Whittle tang knife dated to
have
beenused at the settlement.
AD
the mid-first century
may
Small numbers of querns and rubbers were identified in early Roman period
contexts.
Ingredients
Animal bones
identified
bones
followed
by
the
Cattle
most
commonly
were
species,
sheep, and
"
horses
Iron
Age
late
the
phases,
were more commonly identified than pigs.
unlike
in
increase
horse
is
in
in
increase
the
The
the
the
proportion
of
apparent
mirrored
"
from
large
bones
species.
of
proportion
identified
knife
bones
Both
the
the
and
chop
marks
were
cuts
on
of
various speciesat
"
the settlement,cut marks appearprimarily on cattle bonesalthough two bonesfrom a
display
evidenceof cutting.
a
pig
sheepand
longitudinally
bones
long
Cattle
chopped
are
which could mean that marrow was
"
bones.
from
the
extracted
display
butchery;
bones
bones
from
large
Horse
clear
evidence
of
the
more
of
species
"
butchered.
are also
horses
is
few
the
Whether
consumed
at
were
settlement
as
undecided
only
a
of the
"
butchered.
bones
were
more meaty
deer
bones
butchered.
dog
None
the
was
or
of
"
bones
distribution
to
their
late
Iron
The
Age
according
of
to
the
meat
yield
similar
was
"
from
`B'
Bones
identified
bones
for
the
group
were
most
commonly
each
assemblage.
bones
Group
`C'
horses.
for
both
identified
were
commonly
except
sheepand
species
horses.
particularly
177
Plantremains
"
"
"
There was an increasein the number of plant remainsper sample at the early Roman
period settlementalthoughthe numbersare still small.
Wheat,barley and oats were identified, with barley slightly outnumbering wheat.
The breakdown of the carbonized deposits was grains 51%, chaff 2% and non47%
cereals
deposit
including
revealed a variety of edible weedy species
opium
waterlogged
"A
fat
hen.
stinging
nettle,
mint
and
orache,
poppy,
for
is
late
Iron
Robinson
Age
Mark
As
the
above
mentioned
was
settlement,
"
in
late
Iron
to
the
the
crops
as
whether
were
produced
any
of
at
phases
undecided
Age-early Roman period settlement.
"
7.6.
The prevalenceof barley could indicatethat beerwas producedat the settlement.
The Distribution
of the Remains of Eating and Drinking at Old Shifford Farm
The remainder of this chapter will consider the artefacts and remains from each phase of
in
from
the
section
preceding
within their archaeological contexts.
settlement presented
General observations on the re-contextualized material will be presented and discussed.
Following this, a number of depositsthought to representspecial meals and/or libations will
be highlighted. A brief summary of my findings will be made for both the late Iron Age and
Roman
period settlements.
early
7.5.1. Distribution of the artefactsand remainsat the late Iron Age settlement
Phaseone
The first phaseof late Iron settlementis characterizedby a seriesof gullies and a small number
house
have
been
identified
Two
in
ditches.
sites
the
possible
at
settlement,one the northern
of
The
in
house
in
the
area.
eastern
eastern
the
site
other
of
gullies
and around
areaand
-a series
D-shaped
the
to
enclosure- containedthe vast majority of the pottery recovered
the entrance
burnt
bulk
The
the
clay and burnt stone assignedto phase one were also
in this phase.
of
house
The
from
these
northern
gullies.
site, a semi-circular gully, containeda small
recovered
limestone
located
A
indiscriminate
Dwares.
grog
and
the
the
pit
southern
of
arm
at
of
group
178
jar.
Other
than
three
sherds
a
single
sherd
of
a
grog-tempered
contained
enclosure
shaped
found
in
D-shaped
from
the
the
southern
arm
of
enclosure,
no
other
any
pottery was
recovered
in
D-shaped
the
the
the
the
enclosure,
or
gullies
abutting
northern
of
the
of
arc
south
gullies
of
enclosure.
bones
the
animal
Similarly,
are associated with the D-shaped enclosure. The bones
majority of
in
designated
in
two
the
the
sections
as a
concentrated
of
enclosure:
one
area
were primarily
from the
D-shaped
house
the
to
the
the
at
entrance
other,
situated
away
site
enclosure,
possible
domestic areas of the settlement at the southern-most section of the D-shaped enclosure.
found
in these two areas of the settlement. Only a small
butchery
were
also
marks
Bones with
bones
bones
house
burnt
distribution
The
the
the
are
associated
with
northern
site.
of
group of
in the ditches and gullies (see Figure 7.10.) indicates that the remains of larger species,namely
in
ditches
The
in
the
whereas
pigs
and sheep were more prevalent
gullies.
cattle, were situated
ULM bones recovered from the gullies came from the two areas thought to have been house
bones
in
is
deposit
D-shaped
The
the
the
recovered
of
southern
section
enclosure
of
sites.
butchered
ULM
butchered
bones
from
`B'
À'
and
the
cattle
and
groups of meatof
primarily
is particularly interesting that the only evidence for the cutting of bones was
It
bones.
yielding
found in this section of the D-shaped enclosure. Burnt and weathered bones were only found
in the gullies associatedwith the house sites, which suggeststhat these features may have been
have
ditches
the
contained material that was covered up.
may
òpen' whereas
It is difficult to comment on the significanceof the distribution of the plant remains at the
because,
limited
systematic
a
with
sampling
even
the
strategy,
samplesrepresenta
settlement
those
(for
a
segment
only
and
of
that
contexts,
contexts
of
similar commentssee
at
selection
1991:
359).
der
Veen
With
23;
this in mind, in phase one, the two gullies
1995:
Hill
van
location
house
the
for
Both
of
sites
to
their
were
sampled
represent
plant
remains.
thought
of
amounts
non-grain
small
seedsand primarily non-identified carbonized
comprise
samples
in
item
just
chaff
each
context.
of
one
with
grain,
Phasetwo
The distribution of the artefactsand remains in phasetwo was quite different to phaseone
been
identified
have
in
house
the southernareaof the settlement,one to the
sites
Two possible
179
in
bones
The
the
to
the
the
animal
small
were recovered
pottery and
other
west.
east and
ditches
help
define
throughout
the
that
to
the two main
and
enclosure
various gullies
quantities
largest
The
the
the
area
concentration
and
settlement.
of
pottery,
areas
at
activity/occupation
form,
house
identified
The
to
the
types
the
of vessels
was
southeast
site.
pottery
most
with
jars
from
house
this
site
were
with everted rims, which are often associated with
recovered
jars
In
jar
the
another
with
upright
rims.
area
of
settlement, a
with an upright rim
cooking, and
had sooting which suggests that these types of vessels were also used in cooking. Small
bowls
jars
from
the
and
globular
and
within
were
only
recovered
necked
of
quantities
features
and
other
around the settlement generally contained non-typed
enclosure
rectilinear
pottery.
In contrast to the first phase, cattle bones were found more commonly in the gullies, whereas
ditches
dog
in
(see
horse,
Figure
7.11.
).
Interestingly,
the
and
enclosure
sheep were prevalent
deer bones were recovered only in the enclosure ditches. The frequency of ULM bones and
UMM bones seem to reaffirm the distribution the identified species, although it is possible that
ditches
in
(as
the
cattle
well as the percentage of horses) may be slightly
the percentage of
higher. Comparably with the features in phase one, gullies contained a higher percentage of
bones
ditches
burnt
butchered.
bones
the
that
whereas
contained
and
more
were
weathered
Two different distribution patternsof animal remains were identified at the two house sites.
The southwesternhouse site containedprimarily sheepbonesand the bonesof medium sized
bones
from
forty
head,
house
from
Over
the
this
the
the
recovered
percent
of
site were
species.
from
`B'
À'
bones
bones
bones,
the
the
the
and
were
meat
yielding
groups
of
of
none
rest of
butchery.
This
burnt
there
minimal
was
and
suggeststhat not all the animal remains at
were
this housesite are related to consumptionand that possibly the cranial boneswere a ritualized
deposit(for similar conclusionsseeWilson 1996:79). The southeasternhousesite, on the other
bones
ULM
the
hand, containedprimarily
of
and UMM. The bones were primarily À' and
directly
bones
bones,
`B' type
associated
with consumption, a number of which were
butcheredand burnt. As with the first phase,a large group of boneswas depositedaway from
in
boundary
ditch. It was in this areathat the
domestic
the
a
northern
settlement,
of
areas
the
found
bones
dog
largely
the
bones
identified
the
at
settlement;
were
were
complete and
only
dog
burial
(see
Wilson
for
1990:
58-9
Allison
the
likely
of
a
single
and
represent
similar
they
distribution of dogs).
180
Five plant samples were taken from four gullies and one ditch, although only one ditch and
items
The
fewer
items
twenty
than
twenty
or
more.
samples
with
were
one gully contained
from
house
from
the stack ring and their contents are grouped
the
site
and
southwestern
taken
in with òther Trench L samples' in the site report. One of the samples with more than twenty
items was taken from the southeasternhouse site; the other was taken from a gully east of the
house site. Hey (1996: 107) has suggested that the house site may have been involved in
because
had
2:
1
items
this
the
to
sample
plants
a
ratio
of
non-cereal
whereas
cereal
processing
from
from
the
the domestic area, had a ratio of cereals to
taken
gully
situated
away
the sample
it
be
2:
1
(although
items
that
to
third
the
should
of
noted
up
a
of
non-cereal seeds,
non-cereal
is
were orache which an edible species).
`Specialdeposits'at the late Iron Age settlement(phaseoneand tivo)

The identification of `special deposits' is not obvious in either phase of the late Iron Age
Old
Shifford
Farm,
low-lying
At
there was only a scattering of pits and they
settlement.
bones
The
isolated
and
only
one
sherd
of
animal
pottery.
no
groupsof
absence
of
contained
identification
it
However,
the
of
specific
meals
and/or
events.
affects
reinforcesthe
remains
importanceof other types of featuresat the settlementwhen considering distinctive deposits.
For example,do the limited amountof cut marks found primarily on cattle bonesin phaseone
indicate
Age
Iron
late
the
before
settlement
special
these
they
the
preparation
of
animals
of
has
437)
Tambiah
(1969:
important
that
observed
to the economyof a
species
consumed?
were
The
treatment.
bones
ritualized
receive
through
may
correlation
of
separating
settlement
ligaments,
Iron
the
traditions
is,
in
the
1989:
141),
Age
(Grant
through
the
with
this
of
cutting
deposit
The
burnt
in
in
house
significant.
of
the
terminal
stones
the
particularly
of
site
context,
D-shapedenclosurein phaseone and in both of the house sites in phasetwo, could indicate
feasting
In
houses
of
activity.
areas
were
the
phasetwo there was a large depositof cranial
that
bones associatedwith one of the house sites, which could have been a deliberate and
deposition
The
bones
deposit.
butchery
of
animal
not
generally
associated
with
specialized
in
domestic
both
from
late
Iron
the
dog
burial
(a
Age
areas
phases
of
settlement
away
waste
in
in
indicate
found
two)
these
contexts
phase
may
special meals or the marking of
was also
(Hingley
1990b;
Edmonds
1993;
bounded
Moore
1981:
the
92).
settlement
that
areas
181
Summaryof the re-contextualizedmaterial at the late Iron Age settlement
In phase one, two different distribution patterns for the remains of food consumption at the
have
been identified. The artefacts and animal remains at the
house
two suggested
sites
few
house
bones
the
are
and
site
non-distinguishable
and
all
recovered are burnt. The
northern
house site situated around the entrance to the D-shaped enclosure has larger groups of
identifiable pottery and animal bones, some of the bones are burnt, and many are butchered
burnt
It
D-shaped
that
the
of
stones.
the
also
amounts
appears
southern
sizable
arm
of
and
for
bones
butchered
bones
large
the
the
receptacle
meat-bearing
was
and
species,
of
enclosure
bones
in
The
less
burnt
this
area
cattle.
were
weathered
and
not
and were probably
primarily
covered up periodically.
There are a number of possible scenariosthat might account for the distribution patterns
describedabove.It is possible,for example,that the house site associatedwith the D-shaped
house
a
principle
residence
associated
the
was
with
consuming
whereas
enclosure
northern
food
It
is
differing
functions
that
the
with
preparation.
also
associated
possible
was
of the
site
two housesites were prescribedthrough the status,age or sex of the inhabitants(seeParker
Pearson1996).The animal bonesdepositedaway from the housesites,consistingprimarily of
in
body
have
been
focal
for
either
may,
scenario,
parts,
a
area the consumptionof
consumable
beef or the repositoryfor the remainsof particular meals.
Similarly with phaseone, the two housesites and their associatedfeaturesin phasetwo have
distribution
different
from
patterns
of
artefacts
and
remains
quite
eachother.The southeastern
housesite containeda number of vesselsthat appearto be associatedwith cooking and food
The
from
the
house
is
of
the
composition
plant
sample
site also typical of the
preparation.
bones,
The
hand
the
high
animal
on
grain.
of
other
the
were
preparation
of
meat-yieldingÀ'
bones.
High
bones
`B'
meat-yielding
of
were also recoveredfrom the southwestern
group
and
housesite; however, there were also a high proportion of cranial bones,which may indicate
butchery waste or possibly ritualized deposits.Globular and cordoned neckedbowls and/or
jars, vesselspossibly associatedwith serving,were only identified in the rectangularenclosure
house
In
the
site.
southwestern
spite of the different distributionsof containers
that surrounded
for phaseone andtwo, it is possible to envision similar scenariosfor
identified
ingredients
and
the patterning.Again there are suggestionsthat there may have been a demarcationof areas
182
food
the
to
the
to
of
which
may
seasonsor
consumption
relate
and
preparation
according
inhabitants
(Schuster
Keswani
1994:
261).
As
the
of
particular
age
or
rank
gender,
possibly
deposit
been
large
from
have
to
there
a
of animal remainsaway
also appears
with phaseone
the domestic areas of the settlement, which may represent the special marking of the
boundary.
settlement
7.6.2. Distribution of the artefactsand remainsat the early Roman period settlement
is
Roman
The early
phase of settlement defined by a series of deep and quite substantial
ditches and a prominent entranceway. Both the pottery and the animal bones in this phase
distributed
in
late
Iron
Age
the
than
the
the
around
settlement
was
case
were more evenly
including
from
Each
two
the
the
the
enclosure,
main
southern
arm
away
of
arm
phases.
domesticareas,containsrelatively large groups of pottery. Pottery with evidenceof cooking
house
large
have
is
however,
to
the
two
suggested
confined
sites, which each
residues,
deposits
burnt
Large
from
that
of
are
stone
were
pottery.
of
recovered
contexts
concentrations
house
Malvernian
two
the
sites.
pottery appearsrestrictedto the northwestarea
associatedwith
in
Malvernian
is
It
the
the
that
square-shaped
around
enclosure.
and
possible
of the settlement
but
for
during
`general
the
were
pots
used
purpose'
special
occasions,
possibly
not
were
pots
ingredients
(see
in
Orton
house
225-6).
The
1993:
two
comments
et
al.
cooking of particular
house
had
types
the
although
more
of
pottery,
the
of
pottery
at
site situatedwithin
similar
sites
had
The
beaker
cooking
residues.
enclosure
only
recoveredat the settlement
the square-shaped
in
found
this
context.
also
was
in
found
bones
most ditches and each ditch displayedevidenceof
were also
Large groups of
butchery (including cut marks), weathering, gnawing and concentrations of burnt bones
ditch
This
below).
that
each
suggests
(althoughsee
was variably òpen' and `closed' at various
(ditch
C)
The
its
history.
largest
bones;
in
the
enclosure
main
contained
group
cattle
of
points
followed
by
horse
Figure
(see
7.12.
).
A
then
represented,
and
equally
are
pig
sheep
and
identified,
bones
ULM
horse
were
of
that
which
suggests
number
cattle
and
may
significant
have been even more prominent than speciesidentification indicates. In grouping the bones
7.13.
(see
Figures
),
bones
from
`C'
bones
their
the
to
yield
more
meat
group
with
according
in
deposited
the main enclosure;indeed it is the only areawhere
leastamount of meat- were
183
both
late
large
Similarly
Iron
Age
bones
`C'
with
a
settlements,
were recovered.
pig
group
from
deposited
the
the
bones
the
two
southern
of
arm
along
main
enclosure
away
was
group of
domesticareasof the settlement.This group of bonesrepresentmore of a mix of speciesand
few
display
butchery
burnt
bones
but
there
the
two
and
are
any
marks
body parts,
of
only
bonesin this group.
different
identify
bones
house
to
the
two
grouping
sites.
of
at
suggested
It was also possible
features
B)
(ditch
large
deposit
large-sized
species,
a
of
mostly
enclosure
The sub-rectangular
from
for
it
is
fact,
In
horse.
this
the
that
the
enclosure
most
convincing
evidence
of
particularly
butchery
found
horses
the
settlement
was
recovered;
at
on meatmarks were
consumptionof
horse
bones
burnt.
from
bones
`B'
Horse
the
bones
of
were
of the
number
group
bearing
a
and
here
Figure
ditches
(see
bones
than
the
prominent
more
also
at
any
were
of
other
meatyielding
house
(situated
ditch
the
At
site
the secondsuggested
within
squareshapedenclosure7.12.).
bones
fewer
horse
is
the
species
predominant
were
most
represented,
D), cattle
again
both
by
indicated
is
is
it
that
than
sheep
and
pigs were more prevalent
identified and
probable
in
Figure
7.12.
).
Group
ditch
D
bones
`B'
(see
meat yielding
are also the
identified species
horse
from
bones
in
the
this
with
exception
of
group
enclosure
of
only
which
common
most
burnt
Fewer
7.13.
).
bones
Figure
found
in
(see
this enclosurealthough
were
'C' was recovered
butchery
display
bones
marks.
the
of
a variety
for
from
four
taken
environmental
taken
assessment,
were
Nine soil samples
of which were
below).
(see
features
All
different
ditches,
in
the
features
sections
sampled
although
were
two
identified
two
be
there
that
only
gullies
were
at the settlement,andthey are not
it should noted
Five
house
items and they
the
of
samples
twenty
contained
than
sites.
more
with
associated
in the settlementthat contain the suggestedhouse sites.
from
two
the
areas
were recovered
ditch
bounded
from
little
that
the
the
enclosure
main
taken
contained
very
Samples
settlement
in
òther
from
U.
Trench
their
are
grouped
contents
with
and
material
carbonizedmaterial
between
house
in
the
two
the
composition
plant
Distinctions
sites are apparent.The two
(ditch
B) contain the highest amountsof
from
the
enclosure
sub-rectangular
taken
samples
L
(both
Trench
fragment
in
the
of
carbonized
whole
and
a
of a
silicified);
chaff recovered
from
feature.
Conversely,
this
two
the
also
recovered
three
were
a
rubber
of
and
quern
rotary
features
(ditch
D)
the
from
with
square
shaped
the
associated
enclosure
contain
taken
samples
184
identifiedin
from
L;
Trench
chaff
no
was
any of these
highest amountsof cerealrecovered
from
in
One
the
taken
the
this
of
samples
subarea.
was
recovered
rubber
a
single
and
samples
àbundant'
amounts of edible wild
and
contained
was
waterlogged
rectangular enclosure
`several'
including
specimensof opium poppy.
and
chervil,
nettle,
orache,
species
`Specialdeposits'at the early Romanperiod settlement
deposits
`special'
definition
at the early Roman period settlementwas not obvious as
The
of
identified
low-lying
had
I
Once
to
the
few
were
at
settlement.
again,
pits
non-descript
only a
In
treatment
the
an
and
placement
of
particular
species
was
specialized.
whether
consider
late
identified
Iron
Age
there
the
were
settlement,
at
phase
one
of
practice
similar
apparently
bones,
indicate
times
that
which
again
at
may
cattle
were
on
cattle
cut marks primarily
before
As
the
two
they
settlement,
consumed.
of
were
with
earlier
phases
prepared
specially
from the domestic areas,which as was theorized
deposited
bones
away
were
meat-bearing
between
boundary
indicate
the
the
the
of
consumption
association
an
of
meat
and
abovecould
human
bones
loose
deposit
A
the
the
along
square-shaped
outside
of
arm
of
settlement.
has
further
It
boundary
house
(a
the
the
settlement.
accentuates
site)
of
enclosure suggested
human
bones
boundaries
deposition
the
the
that
of
at
of settlementsrepresents
beensuggested
inside
1996:
123;
Pearson
(Parker
the
and
outside
zones
of
a
settlement
a concernwith marking
dispersal
human
bones
Iron
Age
116-8
that
the
1985:
Wait
of
who
notes
single
at
see also
bones
(Allen
is
that
the
trophies
possibly
or
quite
war
represent
of
random),
settlements not
2000:19).
deposit
is
by
the consumption of particular
suggested
a
ritualized
Another possibility of
The
horsemeat
inconclusive
both
is
the
in
settlement.
eating
of
of
at
species particular areas
in
Roman
Roman
Age
the
Iron
late
settlements,
except
early
sub-rectangular
and
early
the
bones
burnt
horse
found.
butchered
have
been
is
It
bones
in
and
meat-bearing
enclosure which
importance
they
time
the
when
were
a
gaining
at
at
settlement,
that
occasion,
on
possible
Court
Farm,
Barton
for
butchery
horses
At
the
evidence
of
was
horseswere also consumed.
but
in
Roman
butchered
late
Iron
Age
the
for
the
period,
early
period,
meatminimal
also
found
The
both
the
the
of
bones
settlement.
periphery
at
suggestion
at
early
bearing
were
in
is
times
that
and
certain
certain places the eating of, and
at
Roman period settlements
185
differentiated
(see
Dewar
194-5
horse
1969:
for
the
re:
was
of,
consumption
preparations
horse
occasions).
on
special
meat
of
consumption
Roman
the
the
material
at
early
Summaryof
period settlement
re-contextualized
from
distinct
Each
Roman
house
the
settlement
appear
quite
other.
each
early
at
The two
sites
food.
The
features
distinguishing
has
to
the
of
relating
and
consumption
preparation
house site
handmade
jar,
the
only
non-local
such
contained
pottery
one
enclosure
-a
square-shaped
beaker
identified
burnt
the
the
the
only
at
of
residues,
settlement
and
majority
contained
vessel
No
from
bore
taken
the
cooking.
chaff
that
of
was
recovered
samples
plant
residues
the vessels
had
been
for
that
the
cereals
already
consumption.
from the enclosure suggesting
processed
hand,
had
the
the
house
the
sub-rectangular
enclosure, on
other
The second
site situated within
horsemeat.
house
It
for
that
this
the
of
site
consumption
was
also
at
strongest evidence
from
The
the
were
recovered
plant
samples.
only rotary quern
chaff
of
substantial amounts
from
this context. A loom weight and a spindle
identified at the settlement was recovered
have
been
in
This
identified
the
this
that
there
area
of
settlement.
suggests
might
whorl were
including
for
either preparing or possibly consuming
particular activities
specialized areas
large
is
house
deposits
both
burnt
It
food,
contained
sites
possible
stone.
of
types
of
specific
foods
times
of
particular
consumption
was
of
and
associated
with
special
the
preparation
that
(see
Lambrick
inhabitants
1984:
for
169
or
visitors
similar observations at
the year or particular
late
Iron
Unlike
between
Age
distinctions
Ditches).
the
the
Mingies
the
settlements,
Age
Iron
house
its
do
to
the
one
emphasize
house
site
appear
and
over
not
status
possibly
sites
two
both
house
in
involved
that
is
it
the
sites
appear
noteworthy
preparation
particularly
another,
foods.
The
house
both
two
to
the
the
sites
entrance
specific
straddle
of
and consumption
in
brooch
bronze
was
recovered
each
area.
one
and
settlement
7.7.
Discussion of the Distribution
of Artefacts and Remains at Old Shifford Farm
food
preparationwere more vivid at the settlementsat Old
Imagesof cooking, roasting and
discussed
far.
Cooking
in
the
thus
Farm
case
studies
of
than
either
residuesand the
Shifford
for
focus
bones
have
the
each
the
a
of
were
specialists
and
we
the
seen
animal
of
condition
186
benefits of suchan approachto the study of consumption.The lack of imports and vesselsthat
did
have
I
fabric
`tableware'
that,
to
to
the
meant
only
also
not
pay
attention
more
and
typify
but
had
I
that
the
to
vessels,
also
consideralternativeexpressionsof eatingand
of
manufacture
drinking. The importanceof consideringthe context of the artefacts and remains was reaffirmed.
7.7.1. The late Iron Age periodsettlement- phasesone and two

The changein settlementduring the late Iron Age settlementwas an organic movementthat
define
difficult
There
in
to
to
the
the shapeof the
one
phase
or
other.
a
shift
was
was often
boundaries,which becamemore angular with time although existing boundarieshelped to
Two
house
identified
in
in
both
the
possible
shapes.
new
sites
were
each
and
phase
make up
been
have
distinction
between
food
to
there
a
consumption
appears
and
preparation,
cases,
inhabitants.
Similar
to
the
the
traditions of
status,
age
not
relate
or
sex
of
may
or
which may
bones
from
domestic
the
animal
away
meat-bearing
areas of the settlement,
positioning
families
from
the
of
settlement
and
possibly
of
continuity
one phaseto the other.
reiterate

In the early Romanperiod,the boundariesaroundthe settlementbecomesubstantial,and some
have
However,
boundaries
contained
moving
water.
even
them
once
may
are
again,
earlier
of
house sites flank what is quite a large entrance to the
The
two
suggested
perpetuated.
involved
in
both
food
in
they
appear
preparing
and
a specialized way - the
settlement;
hand
horses
the
one
and cooking using hand-madenon-local pots on the other.
on
butcheryof
The positioning of thesehousesat the entranceto the settlementimplies an engagementwith
There
is
the
a generalconsensusamong the various people who
settlement.
those who enter
in
the
involved
Roman
the
that
the
of
site
production
report
early
period settlementwas
were
lack
its
Romanization
imported
because
however,
it
is
low
of
of
and
goods;
status
possible
of
being
and
wealth
was
status
realized through the rearing of horses(see
that the settlement's
high
horses
191
1995a:
as
statusanimals)and was expressedto the community in
on
Reynolds
lessvisible ways suchas throughthe actualconsumptionof food and drink.
187
7.8.
The Social Contexts of Imperialism at Old Shifford Farm
Stepping back for a moment to consider the wider implications
of some of these
have
Roman
dietary
I
the
the
approached
study
of
early
period
and culinary
observations,
from
Farm
Old
Shifford
Roman
the
that
the
position
early
at
ultimately
settlement
practices
housed people who were subject to direct and indirect forces of imperialism. Such a stance
has had a strong bearing on my interpretation of the artefacts and remains throughout this
led
has
me to question the association of change with
thesis, and
Romanization. I instead place greater significance
the concept of
on the relationship
between the
in
drinking
the
the
settlements
and
apparent
of
changes
eating
and
customs.
restructuring
The shapeof the structuresduring the late Iron Age and early Roman occupation of Old
Shifford Farm is not clear. Wattle was recovered in each phase and judging from the
it
is
in
the
the
the
sub-enclosures
and
gullies
possible
structures
were
circular
of
contours
have
been
in
latest
Iron
the
that
the
of
structures
some
may
and
rectangular
earliest phase
Age and early Roman phase, although this is speculative. What is apparent is that Old
Shifford Farm went from a non-enclosedsettlement in the late Iron Age to an enclosed
defined
by
Parker
Roman
(see
the
and
controlled
entrance
a
with
early
period
settlement
Pearson and Richards 1994 re: entrances and the control of movement). Although the
boundaries
traditional
and
possibly
reinforced,
redefined
was
were respected.
settlement
This is the opposite of what occurred at Barton Court Farm where the early Roman
different
(see
by
Hingley
on
a
completely
alignment
constructed
was
comments
settlement
1995:189 on the significanceof the re-useof existing boundaries).
There are also suggestionsthat someof the pre-existing customsand traditions that surround
The
drinking
held
ingredients,
that
reinforced.
containers
were
the
though
and
eating
in
increasingly wheel-made,were each phase locally acquired. The few pots from outside
hand-made
The
distinction
between
kitchen and
local
and
non-Romanized.
were
area
the
isolate
difficult
to
which suggests that there was an emphasis on
serving ware was
ingredientsrather than on serving throughout the history of settlement.The dismemberment
first
in
identified
late
knives
the
Iron
the
Age
phase
of
settlement was reusing
of cattle
In
Roman
in
drinking
the
the
each
phase.
phase,
early
remains
of
eating
and
established
house
in
house
two
had
sites,
and
two
the
each
around
phase,
sites
were concentrated
188
food;
links
to
the
distinct
although at the, early Roman
preparation and consumption of
in
house
their
the
two
the
sites
appear
affiliated
positioning
on
side
either
of
settlement,
in
distribution
The
the
to
the
shared
of
and
prominent
consumption
rites.
settlement
entrance
identified
Old
Shifford
indicate
in
the
types
at
might
use
of
other
of
pottery
range
narrow
lack
39)
1982:
Dyer
(see
the
or
quite
possibly
reflect
of resources and poor status
containers
it
indicate
However,
inhabitants
different
had
inhabitants.
that
the
also
might
the
of
involved
feasts,
deposits
in
large
that
the
the
giving
of
as
status
possibly
seen
of
expressions
for
burnt
the
the
of
preparation
animals
special
consumption,
continued use of
stones,
of
hand-made vessels and possibly the drinking of beer. Viewed in the context of the changes
in settlement, the apparent indifference to, or total rejection of, Roman-like pottery and quite
is
that
them,
the
are
practices
associated
with
consumption
quite striking and
possibly
local community were strong. No people live in a vacuum regardless
the
to
ties
that
suggests
inhabitants
but
Roman
the
the
message
ever
of
what
early
period settlement
their
status,
of
be
does
that
those
to
the
there
to
to
entered
settlement,
any
not
appear
communicate
wished
Roman
to
a
presence.
attempt acknowledge
7.9.
Conclusion
it
is
Shifford
Farm,
Old
For each settlementat
perhapsthe accumulatedmaterial from the
images
ditches
both
daily
living.
In
the
that
the
the
provides
most
vivid
of
rituals
of
and
gullies
Roman
differing
distributions
Age
Iron
the
late
of
phases
settlement,
early
and
of pottery
the
bones,
burnt
bones,
burnt
animal
meat-bearing
stonesand residuesof
residues,
with cooking
differentiate
house
is
to
the
It
various
these
served
sites.
apparently
production
cereal
distinctive concentrations of artefacts and remains that help situate acts of eating and
drinking within the context of tradition, mores and outside forces that govern the whole
including
discard.
procurement
and
cycle of consumption,
in
been
has
It
established this chapter that the settlements at Old Shifford Farm are as
different as they are similar to the other settlements in this thesis. The distribution of
in
is
three
of
settlement
phases
to
all
patterned,
variable
appears
and
remains
and
artefacts
have a direct associationwith the social context of eating and drinking. What the material
in
lacks
Farm
it
Shifford
Old
in
diversity.
quantity
at
certainly
makes
up
culture recovered
189
In the next chapter,Claydon Pike - the fourth and final casestudy in this thesis- provides
Old
Shifford
Farm
interesting
to
through the presenceof significant amountsof
contrast
an
`Romanized' material recoveredat both the late Iron Age and early Roman period `native'
settlements.
190
lv
house
site
91i
phase
Pf
ER
LIA
1
-_,
Phase
r-
ase 3
house
site
features
possibly
in this phase
dark fills
charcoal
&
Roman
Age
Iron
Late
7.1
period settlements at Old Shifford Farm
early
and
Figure
(after Hey 1996, drawn by M. Seymour)
NI
Old Shifford Farm late Iron Age containers and ingredients
Figure 7.? The late iron Age pottery forms - phase one and two
LIA - pottery forms

n=23
70
60
50
i1ow20
10
0
phase 2
phase 1
Figure 7.3 Late Iron Age - phase one - N. I. S.P.
LIA (1) - animal bones

-4l
30
25
20
%15
10
5
0
n-aaCCC
ce
Figure -. 4 Late Iron Age - phase two - N. I. S.P.
Figure -. 5 Late Iron Age (1 and 2) animal bone groups
LIA (2) - animal bones
LIA
n=72
n=48
30
25
20
80
% 15
10
5QQ
0
OA
%40
fi
0C
20
O
QQ
"SN_a
NO
fD
60
CD
(ND
_
p
(D
0
cattle
sheep
pig
horse
I O?
ingredients
Roman
Farm
Shifford
containers
and
period
Old
early
Figure ,. 6 Early Roman period pottery forms
Figure 7.7 Histogram of early Roman period pottery
ER - pottery forms
n=48
cri 5
V) 4
100
80
60
2
E1
40
20
0
z
-i
0
NCNN
Z-
Ei
OA
co
C)
Rim diameter
Figure 78 Early Roman period N. I. S.P.
Figure -. 9 Early Roman period animal groups
ER - animal bones
n=388
ER
n=173
40
80 -
30
QA
60
%20
%40
10
QC
20
0
(n
Cuo
u
aaccc
uz
12
LD
0
cattle
sheep
pig
horse
I9
Old Shifford Farm distribution of late Iron Age and early Roman period containers and ingredients
Figure 7.10 Distribution of late fron Age - phase one - animal bones
LIA (1) - animal contexts

60
50
40
Q ditches
%30
gullies
20
10
0
cattle
horse
sheep
pig
ULM
UMM
USM
late
Iron
Age
unal
bones
Distribution
11
two
7
phase
of
Figure
-an
LIA (2) - animal contexts

45
40
35
30
25
oho
20
p ditches
IN gullies
Q postholes
15
10
C3
0
cattle
horse
sneep
p:9
dog
Q
ceer
ULM
UMM
USM
Roman
bones
Distribution
12
".
period
early
animal
of
Figure
ER - animal contexts
45 40
35
30
25
o'
20
Q ditch A
ditch B
Q ditch C
15
Q ditch 0
0
cattle
horse sheep
pig
aog
weer
bird
-'L %1 ',JMM
OSM
194
Old Shifford Farm distribution of early Roman period ingredients cont.
Figure -. 13 Early Roman period animal bone groups by feature
ER - ditch
ER - ditch B
120
70 60
100
80
QA
60
36
40
QC
20
50
/. 40 '
30 :.
QA
OB
QC
20
i0
0
nN
n
,no
13
1(p
ER - ditch C
ER
120
0,0
80
E] ,
135
EI c
/a 60
40
0
ditch
D
-
120
100 :n
20
I
iI
73
Qa
08
QC
% C
40
?0-
80
7D
z
0
nna
im
195
Chapter 8
Table of Contents for Claydon Pike
Introduction
The Excavation
The Site
The early Roman settlement
Post-excavation reports and primary records
8.3.2.
Drink
Consumption
Food
Archaeology
The
and
of
at Claydon Pike
g4
The containers
8.4.1.
The specialists 'reports - the pottery
The specialists' reports - the glass
late
Iron
Age
The
the
at
settlement
containers
8.4.2.
Form and function
8.1.
8.2.
8.3.
8.3.1.
8.4.3.
8.4.4.
8.4.5.
8.4.6.
R5
8.5.2.
8.5.3.
g6
8.6. I
202
Fabrics-'fine' ware
Fabrics- `coarse' ware
Theglass at the late Iron Age settlement
Other implementsof consumption
Thespecialists' reports
The other implementsat the late Iron Age settlement
The ingredients
Thespecialists' reports - the animals
Thespecialists' reports - the plants
The ingredientsat the Late Iron Age settlement
Theanimal bonesat the late Iron Age settlement
Butchery practices
Meat yields
Containers at the early Roman period settlement
J.
197
198
200
212
Thepottery at the early Romanperiod settlement

Form and Function
Fabrics
Theglass at the early Romanperiod settlement
Butchery practices
Meat yields

Containers
Pottery
Glass
Other implements
Ingredients
Animal bones
Plant remains
218
196
8.6.2.
8.7.
1,7,1.

Containers
Pottery
Glass
Other implements
Ingredients
Animal bones
Plant remains
The Distribution of the Remains of Eating and Drinking at Claydon Pike
The distribution of artefacts and remains at the late Iron Age settlement
220
'Special deposits'
8.7.2.
8.8
8,8,1,
8.8.2.
8.9.
8.10.
The distribution of artefactsand remainsat the early Roman period settlement

'Special deposits'
Discussion of the Distribution of Artefacts and Remains at Clavdon Pike
230

Conclusion
Figures 8.1- 8.22
at Clavdon Pike
233
235
237
197
Chanter 8
Clavdon Pike
8.1.
Introduction
in
is
final
My
this
thesis,
Pike,
the
study
case
not
published.
analysis of the site
Claydon
from that of the other three casestudies since I am making conclusions
differs
consequently
finalized,
than
rather
yet
on
an
individual
not
excavation
a
coherent
as
presented
reports
on
has
been
interpretation
benefits
it
has
inferred
The
challenging,
of
although
process
whole.
integration
involves.
It
the
of
should
post-excavation
of
what
reports
by giving me a sense
interpretation
in
Claydon
in
is
based
Pike
borne
be
that
my
of
mind
on
a
work
therefore,
progress.
first
Colne
bank
River
the
terrace
the
Pike
the
gravel
on
on
situated
northern
of
was
Claydon
Thames.
it
The
by
initially
identified
the
with
converges
site
was
where
just west of
hinted
of
a
succession
overlapping
at
settlements
as
as
a
of
well
number
which
cropmarks,
(Jones
Miles
led
1979:
321).
The site covers an area of
to
the
that
and
site
trackways
major
is
hectares
hundred
by
dry
ìslands'
and
characterized
a
series
of
gravel
one
approximately
or
by
held
the
and
surrounding
that
to
settlements,
that
areas
subject
were
periodic
platforms
longevity
fifth
Iron
the
Age
The
which
spans
through
settlement,
to
of
middle
the
flooding.
dry
The
the
to
the
is
management
continual
the
of
and
areas
wet
of
site.
century, attributed
it
by
lands,
covered
was
together
of
the
much
pasture
as
which
with
preserved
well
site was
deemed
Claydon
deposits
Pike
ideal
for
(Jones
waterlogged
an
site
rescue
preservationof
76).
1983:
Miles
323;
1979:
Miles
and
is
that
the
This
to
of
similar
preceding
this
format
cases
chapter
studies.
chapterwill
The
of
discussion
Claydon
be
followed
brief
the
by
Pike,
chronology
to
of
of
a
an
commencewith
description
brief
the
of
the
a
site
and
the
a
on
comment
excavation,
site
report
of
overview
198
is
discussion
Following
the
this
the
a
of
consumption
of
practices
the
archives.
site
and
inhabitantsand the distribution patternsof the artefactsand remains from within the context
discussion
The
the
conclude
with
a
will
of
chapter
possible significance
the
settlements.
of
imperialism.
the
of
context
of my observationswithin
8.2.
discussions
Roman
of
early
the
settlement at Claydon Pike identify a
Many of
published
differing
Temenos,
two
areas
occupation
possibly
of
up
of
status,
and a
made
a
settlement
Roman
fields
(Miles
Palmer
1983a:
Miles
91-92;
and
roads
and
and
series of rectangular
Hingley
1989:
80).
in
208;
However,
1984:
defined
Miles
1983b;
the
chronology as
Palmer
indicates
èarly'
that
the
the
reports
specialist
and
above
character
of
excavation
the various
hundred
to
took
one
years to manifest. Initial èarly' Roman
up
Roman settlement
ìsland'
is
in the reports as
the
to
to
of
platforms
one
and
confined
referred
was
settlement
Roman
in
it
èarly'
A
settlement
the
was
established
second
century and was
ÈR1'.
second
have
is
been
Temenos
below).
The
to
(although
thought
the
that
constructed
time
see
at this
from
late
is
Iron
Age
Roman
transition
the
the
thesis
to
the
this
on
early
period;
emphasis of
limited
is
largely
to
the
settlement
analysis
on
my
where
platform
one
consequently,
èarly'
Iron
Age
Roman
late
For
and
earlier
the
the
settlement
occurred.
of
majority
discuss
I
èarly'
Roman
the
I
this
when
to
the
chapter,
settlement,
am
of
referring
remainder
ÈRl' phase of settlement.
(section
4.2.
),
4
in
Chapter
discussed
eachof the sites in this thesis has had its own
As was
is someuncertainty over the chronology of the site.
Pike,
Claydon
At
there
set of challenges.
Claydon
has
Pike
(1995)
has
also
studied
Rippengal
who
Rob
questioned the chronology
O.A. U. and it is necessaryto comment on some of his conclusions before we
by
the
outlined
the
the
site.
of
analysis
with
proceed
looks
the
thesis
material culture of two Roman period sites, one of
at
Rippengal's
Robert
He
Pike.
Claydon
Claydon
the
Pike
an
analysis
conducted
of
chronology
of
and
which was
his
here.
The
distinction between 'ER P and
conclusions
of
some
on
it is worth commenting
Roman
the
the
which
early
at
point
settlement made a southward
ÈR' proper marks
199
125.
AD
Rippengal
from
to
the
that
areas
of
site
around
one
other
notes
platform
expansion
"... the amount of material that can be confidently assigned to one or other of these subis
falls
ER1)
broadly
(and
to
that
considerably
reduced
compared
which
particularly
phases
Rippengal
(1995).
both"
the
two
that
as
a
them
result
combined
sub-phases
noting
across
"As the alterations that took place c. 125 AD do not appear to represent a fundamental
it
is
felt
little
in
the
the
that
the
occupation
of
of
site
so
much
as
an
expansion,
nature
change
is lost in this way" (1995). I would maintain that a second domestic settlement of apparently
in
significant
expansion
and
a
lower status, a shrine,
settlement are fundamental changes to
in
the
that
the
the
of
site'
the
and
`nature
occupation
chronology
of
changes
early
of
the
important.
is
as
Roman period settlement
Rippengal's
to
of
do
approach to the chronology of settlement
I
appear critical
not want
is
have
different objectives in our analysis.
complementary
we
because although our work
have
I
tended to accept the chronology outlined by the O.A. U.
thesis
this
Throughout
familiar
but
because
because
the
they
more
with
are
material
also
we share a
primarily
development
Claydon
Pike
At
the
of
time.
the
a
settlement
with
over
concern
similar
O.A. U. 's chronology was arrived at through the combination of stratigraphic relationships
brooches.
Rippengal's
different:
dating
the
coins
and
pottery,
of
concern was quite
and the
is
identify
to
in
concern
more or less coherent groups of
"Here,
contrast, our primary
He looked at the date ranges of pottery in selected contexts and if there
(1995).
material"
date
it
in
kept,
the
75%
range
was
other more mixed contexts were
agreement
was a
dates
by
(when
provided
applicable,
coins
and samian ware were also
eliminated
When
for
LIA
LR
this
rather
produced
small
method
samples
the
and
phases of
considered).
the sample size would be larger if some of the date
that
Rippengal
noted
settlement,
he
for
increased
date
the
example,
so
brackets were combined,
range of the LIA phase to 0include
in
0-75
AD)
to
in
the
(rather
than
order
AD
pottery
the 50-100 AD range). The
100
date
by
Rippengal
the
to
follows:
according
range
the
of
pottery are as
phases established
Fifty-eight
200+.
LR:
70-250;
ER:
0-100;
contexts were retained (he does not state
LIA:
it
formed
these
that
was
contexts
and
the basis of his analysis of
how many were eliminated)
fit
does
O.
A.
U.
's
his
the
That
with
not
chronology
structural phasing was not seen
the site.
because
the
nature
multifaceted
of
of
culture
change of which structural
problematic
as
(Rippengal
In
the
1995).
conservative
is
most
as
the
contrasting
viewed
change
dates
his
fabrics
the general phases of settlement established by
and
using
of
representation
200
he
(not
define
Pottery
U.
fabrics,
in
O.
A.
to
trends
the
also
able
was
surprisingly)
sharper
the
his
(1995).
helped
to
methodology
reaffirm
which
Rippengal's chronology is not appropriate for this study for a number of reasons. As I am
interested in the conquest period up to the second century, the grouping of contexts ranging
from AD 70-250 as èarly' Roman was thus not appropriate for my purposes, particularly
in
have
been
incremental
changes
settlement
established. The elimination of contexts
when
in
of,
or
chronologically
amounts
mixed,
pottery
order to create more statistically
with small
(Orton
Tyers
1991:79; Millet and
although
an
accepted
samples,
practice
and
relevant
Graham 1986; see comments in Willis
1996:210-211), was also not appropriate for this
Green's
below
detailed
(although
see
methodology
re:
quantification) because I am
study
looking at more than one type of artefact. Some of the features at Roughground Farm, Old
Shifford Farm and Barton Court Farm, for example, are significant because of the animal
bones rather than the pottery. The elimination of these contexts would have affected the
interpretation of each site. For the purposes of this study, pottery fabrics that could not have
been used because they did not exist yet are presented but have not been included in the
discussion (for example fabric 4.1 an Oxford colour coat was not produced until the third
in
it
fabrics.
the
the
considered
analysis
was
not
of
so
century
8.3.
The Excavation
Excavation of Claydon Pike began in 1979 and continued through to 1984 in advance of the
Simon
David
Miles
Palmer
directed
and
the excavation. While still in
of
gravel.
extraction
the planning stage, the philosophy of the project was aired in a highly influential volume,
Invasion and Response: The Case of Roman Britain edited by B. C. Burnham and H. B.
Johnson (1979). It is clear that the excavation was designed to counter the perceived bias
Romano-British
date:
had
"In the past the emphasis of Romanoto
studies
that
permeated
British archaeology has been on a limited range of sites and a limited range of explanations,
Roman,
Britain"
(Jones and Miles 1979:324).
upper-class
and
military
view
of
one-sided
a
The project was to have a number of objectives that were outlined by Jones and Miles in the
investigate
integration
(a)
"to
between elements in
the
the
structure
and
examine
volume:
landscape
Romano-British
Age
Iron
and their environmental setting" and (b) "to
and
the
201
isolate the points of major change and relate them to ecological, demographic, economic
is
felt
(1979:
323).
Jones
Miles
factors"
"It
to
the
that
state:
and
go
on
only
cultural
and
for
(or
that
this
to
a
site
of
scale
approach
such
questions
on
any
scale
way
satisfactory
implementation
(1979:
323).
The
is
the
through
of
explicit
sampling
strategies"
matter)
it
facets
`probabilistic'
"...
sampling
strategy,
a
was
argued,
would
cover
all
of
enforcement
bias
(1979:
323).
the
the
personal
of
excavator"
remove
of the site and
8.3.1. The Site (seeFigure 8.1.)
Thelate Iron Age settlement

The late Iron Age settlement was occupied late in the first century BC into the early first
have
is
is
2.5
hectares.
The
AD
thought
to
covered
approximately
and
settlement
century
by
a series of oval and circular enclosures and gullies of varying sizes, and
characterized
ditches and pits, which together formed a complex, though nucleated occupation area.
Towards the end of the settlement a boundary ditch enclosed part of the settlement. As with
is
in
identification
house
late
Iron
Age
this
the
study
settlements
of possible
sites
the other
floor
lack
The
by
density
domestic
the
of
preserved
plans.
material,
of
complicated
however, points to habitation. Daub has also been identified pointing to wattle and daub
few
interestingly
tile
were
the
too
window
glass
and
recovered
although
numbers
are
walls;
in
buildings
(Wait
Hedges
indicate
their
and
unpublished small find report) The
use
to
distribution of excavated material appears to indicate the presence of domestic zones within
in
for
It
the
the
southern
section
of
was
site,
example, that items of personal
the settlement.
large
large
bones,
vessels,
of
pots,
glass
groups
groups
animal
of
oven
adornment,
fragments, loom weights, rivets and plugs for repairing pots, and whetstones were found.
Many of the other gullies and ditches around the settlement contained minimal small finds,
bones
features
been
interpreted
have
these
and
animal
of
pot
amounts
as
small
and only
field
boundaries.
It
is
small
and
possible that there was a metal working
possible paddocks
identified
in
Pits
in
this
as
smithing
slag
was
the
area
northeast
area.
also appear to
area
have been grouped in the southwest and northeast areas of the settlement. The economy is
in
because
largely
been
have
part
of the nature of the plant remains and
pastoral,
thought to
lack of evidence for storage of surplus.
202
The early Romansettlement
The
formed
top
the
the
70
AD
of
earlier
settlement.
over
Around
a new settlementwas
ignored.
The
late
Iron
Age
features
the
early
settlement were completely
layout and
of
juxtaposed
a newly constructed road and contained a rectangular
Roman settlement
is
building'.
àisled
It
is
that
there
the
to
a
second
also
possible
was
as
that
referred
structure
is
dating
A
the
this
number
of
structure
not
of
secure.
the
although
settlement
at
structure
features
fence
lines
internal
include
building
the
and
other
aisled
with
associated
were
pits
Roman
Claydon
Pike
Initial
to
the
thoughts
occupation
were
of
early
linear
as
gullies.
and
it
has
been
been
have
through
complex,
although
it
excavation
village
revealed
a
that may
fields
domestic
Only
than
the
and
paddocksrather
areas.
crop marks represent
that many of
has
led
This
identified.
to
that
the
of
domestic
suggestions
emphasis
economic
was
area
one
in
the
of
cattle
and
sheep.
rearing
was
the settlement
Post-excavationreports andprimary records

is
is
Claydon
Pike
Analysis
this
not
yet
published.
casestudy
of
As was mentioned above,
databases.
linked
The
their
together
and
reports
are
not
reports
basedon the post-excavation
found
discussion
in
that
usually
are
a site report and thus what
by the editorials and general
Claydon
Pike
interpretation
be
I
largely
is
of
and
should
entered
viewed
as
such.
my
follows
database,
in
into
form.
I
they
bone
to
as
a
exist
only
unable
written
was
records
the primary
information
the
the
amount
of
within
although
contained
the
records
primary
consult
for
is
databases
and
was
sufficient
extensive
my
purposes.
pottery
existing
8,4,
The Archaeology of Food and Drink Consumption at Claydon Pike
discussion
this
two
section
will
studies,
commence
case
a
on the
with
As with the previous
include
drinking,
to
the
and
methodologies
and
conclusions
eating
of
of
the
remains
of
nature
data.
distinction
between
The
the
analysis
the
own
my
of
as
well
as
specialists,
the various
be
beginning
the
analysis
own
will
the
and
my
at
outlined
of
specialists
the
various
work of
203
follows:
The
'containers'
are
grouped
as
which
artefacts
and
remains
relevant subsections.
include pottery and glass; 'other implements' which includes metal, stone and clay artefacts
This
include
'ingredients'
finally
section will conclude
and
plant
animal
remains.
which
and
the
settlement.
at
each
artefacts
and
remains
of
summary
a
with
8.4.1. The Containers
Thespecialists'reports- thepottery
by
Green.
Sarah
Her
define
helped
the
to
the
analysis
of
analysed
The pottery was
pottery
it
determined
both
i.
late
Iron
Age
the
that
was
at
site
where
limits of occupation
and early, e.
have
been
to
Roman
contained within Platform 1. The pottery
ÈR1',
activity appears
in comparison to the other settlements in this study, and all of
large,
is
particularly
collection
by
date
by
for
total
catalogued
general
was
pottery
weight,
context
range
and
each
the pottery
is
detailed
Quantification
form.
that
was made of all contexts that contained
fabric and
more
Contexts
help
that
to clarify the stratigraphic relationships
might
of
pottery.
50 grams or more
in
detail,
features
the
types
as
also
quantified
was
a
were
representation
of
all
of
of particular
400 kilos of the pottery was analysed in detail, although all of
Roughly
third
features.
or
one
of
below
The
the
the
at
site
samian
was
recovered
recorded'.
and
analysis of the
the amphorae
based on the detailed record of pottery, as it was not possible within the
largely
is
pottery
detailed
Where
that
the
to
to
this
pottery
was
re-examine
not
subjected
study
study.
confines of
bones
initial
the
animal
were
contained
which
cross-checked
with
applicable, contexts
for
the context.
the
pottery
to
record
establish
catalogue
Thespecialists' reports - the glass
H.
E.
by
Price
M.
Jennifer
Cool
and
who identifiedtwo peaksof use:
The glasswasanalysed
late
Flavian
in
Neroian
Roman
late
the
540
to
in
and
first
period,
the
the
second
period.
the
fifty
vessels,were recovered at the settlement.In the
of
a
minimum
fragments,representing
I
below,
findings.
the
summarize
specialists'
will
simply
sectionson glass
204
8.4.2. The Containersat the late Iron Age settlement
Imported pottery
A significant deviation from the other late Iron Age settlements in this study is the early
Roman-style
imported
The
is
pottery.
prominence
of
of
amphorae
particularly
presence
identified
in
forms
terms
bowls
of
to
was
which
equal
as the second most
noteworthy,
Figure
(see
8.2.
).
The
type
majority of the amphorae sherds were body
prominent vessel
found
in
but
it
is
they
a
wide
were
variety
of
as
contexts,
probable that their apparent
sherds,
importance is genuine. Most of the amphorae identified were the olive oil amphorae Dressel
20, although Camulodunum 186c, which contained fish-based products, was also identified.
The wine amphorae Dressel 2-4, which dates to the late first century BC and continued in
found
in
the
century
were
second
general spread contexts, which could
early
use until
indicate that wine was consumed at the settlement. Mortaria are also present in the late Iron
Age assemblage and tankards and flagons have been identified. Some pre-Flavian samian
in
identified,
been
has
although
not
context.
ware
in
the mid-first century AD, prior to the shift in settlement
that
It appears
at some point
70
AD,
the
that
goods
of
number
were arriving at the settlement are associated
a
around
Roman-type
it
it
consumption,
of
particularly
aspects
appears
certain
as pertains to the
with
flavouring and preparation of food.
Form and function

jars
dominate
Age
late
Iron
the pottery assemblage,which as we have
settlement,
At the
is
for
typical
in
the
jars
Of
identified,
quite
the
studies
case
the
previous
period.
the
seen
`jars',
30%
identified
jars,
as
simply
about
6% were
were
were
wide-mouth
majority
jars,
`Belgic'
a
small
and
number
of narrow-neck and storage type jars were
or
globular
found
jars
Burnt
on
wide-mouth
were
identified.
and to a lesserextent, on the more
residues
is
Roman-like
material
all
a commonpractice at all typesof Romano-British
of
The retentionand analysis
have
the
suggested
skews
scholars
of
material culture of the settlements(Jones
number
a
which
settlements,
205
jars
Histograms
(see
Figure
8.3.
)
`jars'.
the
of
show a peak of small to medium sized
generic
jars with small numbers of large jars. The bowls were primarily straight-sided and were
fabrics
including
local
brown
of
variety
of
a
out
and grey wares and surprisingly,
made
Black-burnishedware (BB 1), although it is likely that much of the BB 1 wares was residual
(see below). Histograms of the bowls (see Figure 8.4.) indicate that the bowls peaked at 22
large
Green
defines
bowls
`very
large'
two
(one
had a diameter of 58
quite
were
as
and
cm
-
cm, the other was not recorded).

Late Iron Ae
mortaria
am horae
British colour-coat
Forei n colour-coat
Mica-dusted ware
White ware
Parchment ware
fine and specialist wares

red/orange wares
n=3,670
12
43
41
2
2
111
2
213
379
283
Early Roman
%
33
.
1
1
06
.
06
3
06
.
n=2,429
35
34
61
4
24
155
0
%
1
1.4
3
16
.
1
6
0
6
10
8
313
424
331
13
17
14
brown wares
Calcareous ware
746
751
20
20
50
25
2
1
re /black wares
misc.
1297
1
35
03
.
1278
8
53
33
.
Table 8.1. Late Iron Age and early Roman period pottery fabrics
Fabrics- coarseware
identified
include
fabrics
both specialist and fine fabrics and
were
A variety of
which
coarse
fabrics. In Table 8.1, we can see that the grey/black wares were the most commonly
identified fabrics, followed by brown wares and calcareous wares which are equally
Black-burnished
As
is
above,
stated
ware
quite prominent, although Green
represented.
fabric
is
due
this
the
that
of
to the positioning of the sherdsin
probably
prominence
suggests
features.
late
Iron
Age
She
does
layers
the
of
the upper
add that BB I was probably arriving
first
late
AD,
by
the
the
Cirencester;
century
as
was
the
BB 1 may therefore
case
site
at
at
have been at the settlementjust prior to its reorganization.Burnt residueswere found on a
identified
Of
the
and
non-identified
vessels.
fabrics
of
various
vessel
number
with burnt
in
141;
Fulford and Huddleston(1991:9-11)
1999:
Hingley
36-8;
comments
see
1997:
206
followed
incidences
by
had
highest
(7)
the
of
grey/black
residues
wares
residues,calcareous
(2).
The
brown
(4)
brown
(5),
and
wares
and calcareouswares are
wares
red/orange
wares
both local, but the other wares are from Swindon and the Severnvalley. No burnt residues
Black-burnished
identified
the
ware.
on
were
Fabrics-'fine'
fabrics
specialist
and
fabrics
have
been
forms,
discussed
in
the
the
specialist
section
on pottery
A number of
`fine'
together
the
and
and
mortaria,
with
white
wares,
and specialty
namely amphorae
late
Iron
Age
5%
fabrics,
(One
`fine'
total
the
the
of
assemblage.
up
of
ware
wares make
included
because
Oxford
not
of its date of manufacture).As has been
4.1,
colour coat, was
however,
`fine
is
in
in
chapters,
ware'
many
respects
subjective
previous
a
expressed
does not take into account the colour of a vessel or where it was made.
that
classification
for
example,all made out of a red/orangeSwindon fabric and the black
The tankardswere,
been
have
is
is
Dorset
`fine'
from
What
bowls
tableware.
I
also
considered
may
BB
clear
late
Iron
Age
inhabitants
the
settlement were using vessels that point to
of
that the
found
tastes
at the other settlementsin this study (see also late Iron Age
Mediterranean
not
late
Iron
Age
Roman
Thornhill
A
below).
and
early
contemporary
period
settlement
at
glass
Claydon
Pike,
did
have
km
1
of
west
not
Farm, situated
any so-called luxury specialty or
Old
Shifford
Farm,
interestingly,
did
as
with
not have coarseware versions
`fine wares' and
draft
15).
(Timby
pottery
report:
unpublished
of tableware
late
Iron
Age
the
settlement
Theglass at
from
late
Iron
Age contexts and while there is
recovered
were
vessels
A number of glass
likely to be a degreeof residuality, some of the pieces were of appropriatedate. According
bottle
Hofheim
body
Cool,
fragments
blue/green
and
other
Price
cup
and
a
of
and
and
to
have
been
for
imported
to the settlementin the late Iron
may
example,
light green vessels,
Age.
207
8.4.3. Other implementsof consumption
Thespecialists' reports
by
Gerald
Wait
John
I
Hedges.
finds
computerized
and
and
will
The small
were summarized
findings
individual
the
their
the
they
conclusions
and
of
on
as
specialists
comment
primarily
food
discussion
In
look
below
the
the
on
and
consumption.
to
sections
which
present
pertain
I
the
the
contextual
settlements
and
associations
of
artefacts and remains,
at the character of
distribution
the
the
discuss
the
significance
of
patterns
possible
of
some
of
other small
will
finds such as jewellery.
late
Iron
Age
implements
the
The
at
settlement
8.4.4.
other
food
the
implements
remains of
processing and cooking equipment were
Metal
and
Iron Age contexts. A number of metal objects were identified that
late
from
recovered
function:
iron
links
iron chain possibly usedto
domestic
had
a
small
vessel,
of
an
a
probably
suspenda vessel, and an attachmentloop from a bucket shaped vessel. Whetstoneswere
The
significant
number
of
a
clay
of
remains
ovens and oven plates were
recovered.
identified.
8.4.5. The ingredients

bones
the
animal
Thespecialists' reports Bruce
Wilson
by
Bob
Levitan
and
bones
who organized the bones into
The
were analysed
`rubbish
first
be
The
all
phase
residual,
spread'
reflects
will
and
not
considered
three groups.
is
based,
bones;
this
study
were
the third group
which
on
stratified
group,
here; the second
by
Wilson
Levitan
considered
not
were
un-stratified
bones
were
and
and will not
which
of
Wilson
Levitan
In
describe
in
their
and
the phasing differently
this
report,
study.
be included
I'
is
Their
`phase
which
elsewhere `late Iron Age' is classedas
from that establishedabove.
'phase
2a'
is
ÈR1'
and
settlement"
Romano-British
native
which
elsewhere is
"early
208
bone
"`Romanization'
the
of
settlement";
and expansion
primary.
records,
classed as
is
`LIA'
ÈR1'.
My
the
analysis basedon the chronology as
and
however, record
contextsas
interpretation
late
Iron
Age and early
),
8.2.
(section
thus
contrasts
my
defined above
and
Roman
Romanized
than
native
early
and
early Roman.
Romanperiod resultsrather
is
Levitan's
comprehensive and places particular emphasis on the
analysis
Wilson and
bones.
for
Wilson,
identify
distribution
the
to
animal
example,
was
able
a
of
spatial
for
(rather
bones
by
the
than
the
whole
site
phase) that corresponds to
distribution of
animal
in
has
Upper
he
Thames
Valley:
the
at
other
sites
made
and
sheep
pig
observations
other
domestic
in
degraded
bones,
from
bones
contexts
the
and
and
bones were concentrated
(unpublished
521996:
the
at
peripheral
contexts
concentrated
report
and
larger species were
distribution
looked
hand,
in
the
the
at
Levitan,
of
species
other
particular contexts.
56).
on
late
Iron
Age
in
both
the
the
settlement
cattle
found
were
that
most
common
species
He
at
in
but
At
that
the early
sheep
and
were
equally
cattle
represented
pits.
ditches and gullies,
listed
ditches.
The
distribution
the
ER1
contexts
were
only
of the species
Roman
settlement,
below.
be
commented on
will
frequency
the
Levitan
of species,which they acknowledgedis a
Wilson and
also established
fragment
N.
M.
L.
the
that
all
of
variables
can
affect
count
and
considering
relative account
determined
butchery
discussion
also
the
were
where
possible,
and
although
Pathology, sex
the
by
to
Despite
to
over
tended
species
than
whole
refer
site
rather
phase.
of theseattributes
bias
bone
the
specialists
the
which
acknowledge
sieving
of
sample
amount
small
a
only
domesticated
bird
bones,
larger
species
sizable
number
of
some with
the
-a
towards
Claydon
Pike.
identified
at
butcherymarks, were
been
had
bone
computerizedand when I entered the record into a
not
records
The primary
between
discrepancies
few
the
phasing
there
minor
database,
were
of a
contexts. In each
in
definitive
the
the
general
outlined
the
site
I
phasing
record
as
phase,
considered
case,
difference
in
below
between
there
major
was
no
the
two
results
methods
shown
as
although
for
butchery
All
the
cattle
notions
and
sheep
the
of
not
were
recorded
on
of calculation.
illustrated
depicted
bones,
discussion
the
on
were
some
sheets,
and
of
primary recording
both
includes
types
in
this
of notations.
section
butcherypatterns
209
from
late
Roman
Iron
Age
taken
were
samples
and
early
Both carbonized and waterlogged
features at Claydon Pike. The samples were analysed by Vanessa Straker following
the
Jones
Iron
by
Martin
Anne
Perry
the
and
who
analysed
middle
methodology set out earlier
involved
feature
This
the
types and
strategy
sampling
of
a
variety
of
Age plant assemblage.
from
feature.
be
I
different
taken
were
each
will
primarily
commenting
samples
of
a number
Straker
the
that
findings
emphasis
particular
on
with
placing
plants
are
associated
of
the
on
consumption.
Iron
Late
Age
ingredients
the
The
settlement
at
8.4.6.
Iron
Age
late
bones
the
settlement
Animal
at
by
8.6.,
determined
Figure
the
8.5.
contrasts
which
percentages
of species as
In Figure
and
differences
in
does
Levitan,
the
Wilson
slight
phasing
not
of
some
contexts
and
and
myself
Cattle
identified
the
the
percentages.
are
most
overall
commonly
species
affect
significantly
Other
horse
in
dog
(10
by
species
pig.
are
and
present
small
numbers
followed
sheep,
deer
donkey
(1)
Thirty-three
bird
bones
bones)
(3
a
single
and
red
antler.
were
bones), cat
from late Iron Age contexts. The species identified include duck (3 bones), pigeon
recovered
fowl
(7),
domestic
(2),
blackbird
(1),
(2),
(1),
raven
and a number of
goose
(1), quail
birds (16). It is likely that most of the species identified, including some of the
unidentified
1988:
202)
(after
Parker
blackbird,
Some
the
were
consumed
at
settlement.
g.
e.
wild species,
had
have
inhabitants
(Wait
the
special
to
may
significance
raven,
as
such
of the species,
1988:
206-209)
Parker
125-127;
discussion
1992:
Green
the
138;
and
of their
1985:
this
in
Wilson
Levitan
later
this
consider
will
possibility.
chapter
and
make
distribution
is
species
the
of
wild
the
number
present
this
true
the
small
at
and
site
of
of
note
particular
in
this study.
Age
Iron
late
settlements
other
210
Butchery practices
domestic
Wilson
Levitan
the
discussion
the
of
various
species,
and
exploitation
In their
of
is
determine
As
to
focus
as
a
whole.
one
of
my
aims
species
whether there
on each
tended to
it
between
distinction
the
treatment
time,
and consumption of species over
was
was any
down the butchery practices for each species by phase.
break
to
necessary
four
late
Iron
Age
the
the
body
main
species
were
of
recovered
at
each
Most
of
parts
butchered
the
that
the
animals
were
probably
suggests
settlement.
at
which
settlement,
both
knives
The
indicate
bones
the
use
of
and
choppers.
cattle
with the most
Butchery marks
include
butchery
the mandible, which was extensively trimmed; the
direct evidence of
highly
butchered
is indicative of
the
were
also
tarsals
phalanges
which
and
metacarpals,
have
been
for
tarsals
to
the
also
appear
and
the
split,
possibly
removal
metacarpals
skinning;
heavily
butchered
found
The
the
also
was
bone
and
most of
vertebra
chop marks
marrow.
of
here.
There
found
butchery
bones
the
was
minimal
evidence
more
of
on
were
on cattle
bones,
late
Iron
long
Court
Farm,
i.
Age
Barton
bones,
and
scapula
pelvis,
unlike
e.
meaty
bones
bone.
the
that
stripped
the
either
right
was
off
meat
or
cooked
on
which suggests
direct
butchery.
Chop marks occur at the
displayed
bones
evidence
of
minimal
Sheep
indicating
horn;
the
the
the
possibly
removal
cranium
of
were
cut
marks
of
region
parietal
identified intermittently on a couple of metatarsal and phalanges. Wilson and Levitan
long
bones
lack
been
the
have
on
to
the
of
sheep,
of
marks
which
that
appears
suggest
indicate
left
that
bone. Pig bones also
in
the
meat
was
may
on
phases,
all
consistent
Cut
found
butchery
few
long
bones,
marks
the
were
marks.
along
a
of
displayed minimal
filleting,
suggestive of
and a chop to an upper vertebra indicates that the head was
occasionally removed.
butchered
late
been
have
Iron
the
Age
do
to
at
bones
settlement,although it is
Horse
appear
horsemeat
indicate
the
consumed,
as
was
most
of
to
appear
marks
not clear whether
long
bone
The
butchery
bones.
only
with
the
tibia,
marks
was
which
non-meaty
of
skinning
bones
butchery.
other
no
meaty
mid-shaft
cutting
show
of
evidence
of
shows evidence
butchered
been
have
identified
deer
bone
the
to
and
do
single
at the
Dogs
not appear
have
deer
been
may
antlers
the
antler;
off
worked
at
settlement
sawn
a
as
a
was
settlement
211
identified
in
late
Iron
The
Age
bone
contexts.
radius of an unidentified
were
pins
number of
bird had cut marks.
Meat yields
in
I
bones
into
this
thesis,
the
organized
studies
groups according to
As with the other case
from
1973;
Lange
1996).
Figure
bones
Uerpmann
8.7.
(after
shows cattle
their meat yield
bones
('A'
`C'
the
the
with
representing
most
and
the
meat
representing
groups
of
each
distributed.
Bones
from
higher
`C'
than
evenly
group
pretty
are
slightly
are
meat)
minimal
`B'.
Moderate
À'
bones
Groups
for
for
and
meat-yielding
are
most prevalent
was the case
high
distribution
bones.
bones,
of
meat
yielding
and
even
an
non-meat
yielding
with
sheep
higher
have
À'
bones
hand,
`B'
of
a
presence
group
and
with
particularly
Pigs, on the other
from
bones
'C'.
Horse
bones
in
distinguished
that
group
of
are
again
lower representation
bones
`C'
the
À'
are
most
prominent.
group
and
group
different
body
in
types of consumption practices
differences
suggest
part
representation
The
both
The
of
prevalence
non-meat bearing and meat bearing cattle
for the various species.
bonessupports'Wilson and Levitan's conclusionsthat cattle were used for other purposesas
bones
both
the
for
of
prominence
meat
yielding
their
meat;
sheep
and
pigs
of
as
well
The
high
consumed.
they
predominance
primarily
that
were
of
meat yielding and
suggests
bones
horse
to
the
could
point
specialized
treatment
consumption
of
and
yielding
non-meat
distribution
The
body
later
in
the
of
analysis
this chapter
of
the
parts
settlement.
horsesat
the
the
practices
of
various
consumption
species.
help
to
clarify
will
Age
late
Iron
the
settlement
Theplant remains at
At the late Iron Age settlement,thirty-six contexts were sampled for their edible contents
five
the
features
A
contexts
were
of
waterlogged.
significance;
variety
of
and environmental
ditches
Of
including
the
and
pits.
gullies,
the
carbonized
samples
most
sampled
were
barley
45%
identified taxa, followed by spelt
identified
at
was
grain
of
cereal
commonly
2%
identified
There
35%
present
at
minimally
of
were
oats
grains.
and
was a
wheat at
Straker
to
chaff.
considers the speciesassemblagesquite
higher ratio of grain and weeds
five
items
litre
had
for
than
(compare
per
of
more
the
soil
sample
example
one
only
small as
212
from
Court
Farm
(5,717)
just
items
Barton
Although
identified
two
at
contexts).
numbers of
in
late
identified
in
Iron
Age
from
increase
the
that of the
grain
settlement
there was an
it
is
latter
`consumer'
Age
thought
that
the
Iron
site
similarly
site
was
a
settlement,
middle
(although
below).
The
limited
production
see
animal
section
plant
numbers of
with possibly
include
fat
hen,
identified
also
very
small,
were
and
watercress and sorrel;
edible weeds
is
diet
the
to
unknown.
their contribution
8.5.
Roman
the
Containers
period settlement
early
at
8.5.1.
Roman
period settlement
the
early
Thepottery at
dominate
Roman
jars
less
the
8.8.
that
early
assemblage
so
Figure
although slightly
shows
(67%
Age
77%).
is
Iron
The
late
to
as
compared
settlement
the
assemblage also
than at
identified
having
higher
clearly
pottery
with
with
serving
varied,
a
slightly
slightly more
for
first
higher
identified
incidences
flagons
time
the
Platters
and
are
and
of
presence.
forms
identified
linked
drinking
7%
to
are
of
beakersare recorded;
as comparedto 4% at the
in
Mortaria,
higher
Age
are
also
present
Iron
numbers and amphoraeand
settlement.
late
this
represented;
equally
suggests that there was a continued
almost
bowls are again
foods
in
Interestingly,
particular
serving
and
ways.
no samian ware
emphasison preparing
despite
(ER1)
Roman
this
in
the
contexts,
found
presence
amount
early
of
a
sizeable
of
was
in
from
the
the
Gaulish
general
spread
recovered
site.
samian
south
diameters
function
and rim
Form,
identified
jar
is
`jar'
the
most commonly
recovered at the settlement. WideThe generic
jar
35%
jars at 6% and Belgic jars in
the
follow
assemblage
with
jars
globular
of
at
mouth
increase
is
in
2%;
there
the
jars
a
slight
percentage
at
of
narrow-necked
amounts
very small
(as
4%
to
Roman
compared
settlement
period
at the earlier settlement).Only
6% at the early
two
identified
identified
of
which
residues,
with
to form: a
were
not
four vessels were
burnt,
BB1
jar.
Histograms
small
was
a
wide-mouth
as
fabric
tankard
was
of the rim
brown
213
diameter of the jars (see Figure 8.9.) indicate a similar range in size as was determined at the
late Iron Age settlement, namely, small to medium sized jars with a peak of 18cm, and a
jars.
large
It
is
in
lids
identified
3%
increase
that
the
of
numbers
of
note
make
up
of
slight
forms whereas only one lid was recovered in late Iron Age contexts. Bowls are found in a
BB
I
imitation
BB
fabrics
I
fabric.
the
The
and
shapes;
and
are
most
common
of
variety
definable
diameters
bowls
is
histograms
bowls
with
rim
the
quite
of
small;
size
of
sample
indicate a peak of diameters at 18-20 cm (see Figure 8.10.), which is slightly smaller than
late
Iron
Age
diameters
in
the
at
recorded
settlement. A small number of smaller
the peak
bowls
large
identified
no
very
and
were defined.
bowls were
A note on samianware recoveredat the site

The majority of the early samian ware was recovered in the general spread on Platform 1,
in
Roman
that
the
(if
to
early
suggest
period
appear
not earlier see above) at
would
which
being
Gaulish
the
samian
was
used at the settlement.The extent that the
south
least some of
samian
during
being
the secondcentury and beyond, however,
or
curated
re-used
ware was
forms
identified
The
prevalent
most
is un-determinable.
were plates at 38%, followed by
dishes (5%) and plate/dishes(2%). Peter Webster, who analysed
(19%)
bowls
(36%),
cups
the
the
(96%)
that
found
majority
of
samian
notes
ware
ware
on the site was
the samian
10% of the south Gaulish samian ware recoveredfrom
to
(according
calculations
my
plain
he
decorated);
bowls
"Given
Ithat
adds
was
the chronological rangeof the
all
Platform
forms
is
decorated
dearth
unusual
the
of
and must, one presumes,reflect some
material
(un-published
the
the
He
of
site"
usage
of
samian
characteristic
report).
suggests
unusual
form
been
have
inhabitants,
religious
a
of
austerity
the
amongst
there
may
or possibly
that
inhabitants
that
the
and
cheaper
was
of the settlementsover time
that plain samian ware
The
decorated
(see
below)
ware.
prevalence
samian
of
glass
vessels,
afford
which
could not
item,
that
have
been
issue
suggests
price
may
a
cheap
not
not
necessarily
an
presumably
are
decorated
Observations
vessels.
over
in the choice of plain
presented by Willis on the
forms
forms
decorated
in
over
Yorkshire
plain
(1997)
at
of
rural
sites
predominance
in
in
made
the
previous
case
studies
this thesis, which indicate
observations
together with
identified
Claydon
that
the
is
Pike
pattern
at
suggests
pattern,
not so unusual
the opposite
determine
its
to
study
significance.
regional
comparative
requires
and
214
Fabrics
fabrics
fine,
coarse
A variety of
specialist and
were identified in early Roman period
fabrics identified at the late Iron Age settlement appear
Although
the
of
a
number
contexts.
in
in
there
the types of coarse wares (see Table
quite
significant
are
changes
use,
to continue
8.1., section 8.4.2.). The grey/black ware remains the most prevalent fabric and its presence
has increased by the early Roman period. Almost absent, however, are the brown and
late
Iron
Age
the
that
Black-burnished
were
at
prevalent
wares
settlement.
wares
calcareous
Roman
the
the
early
settlement
at
as
are
prominent
red/orange wares.
more
are
increase
in
fine
have
been
the
to
an
There appears
and specialist fabrics by the early Roman
Table 8.1., section 8.4.2.). The increase would be even more acute if the
(see
period
inhabitants of this settlement used the samian ware dated to the first century. The `fine ware'
fabrics identified at the late Iron Age settlement are present at the early Roman period
higher
The main difference is the increase in white
representation.
slightly
with
settlement,
the
Again,
wares are probably residual although the possibility of their use
of
some
wares.
Roman
is
the
this
of
the
early
sub-phase
period
of
end
more of a possibility than for
towards
Similar
Age
Iron
cautions, which were applied to the classification of `fine
settlement.
the
Age
Iron
late
fabrics
defined
to
settlement,
the
the
are
applicable
at
at the early
ware'
beakers,
Many
flagons
the
of
and tankards, for instance, were
Roman period settlement.
indicate
fabric,
fabric,
that
this
its
which
could
orange/red
an
of
possibly
colour, was
made
drinking. If we recall, at the late Iron Age settlement all of the identified
to
significant
fabric.
from
orange/red
an
tankards were made
Theglass at the early Romanperiod settlement

identified
in
Roman
early
were
vessels
contexts. These comprise various
A variety of glass
light
jugs
body
fragments,
blue/green
bottles,
and
green
blue/green
possibly cups. As was
Cool
identified
Price
Roman
a
peak
of
and
early
(22%
above
glass
use
of the
mentioned
Roman),
which sets this settlement apart from the other early
early
identified glass was
in
14%
identified
the
this
of
study;
vessels
settlements
Roman period
were of first century
215
date and 8% were of mid-first century to mid-secondcentury. Cool and Price suggestthat
from
the
that
high
recovered
such
early
of
glass
contexts
supports
notion
percentage
the
below
for
further
had
(see
have
Pike
official or military status
comment).
Claydon
may
8.5.2
Other implements at the early Roman settlement
have
had
the
that
implements
settlement
may
at
a culinary association
recovered
Other
iron
knife,
knives
identified
`general
ten
thirty-six
only
of
purpose'
were
include a single
knives
is
both the late
it
the
that
some
of
unstratified
were
used
possible
at
and
stratified,
large
iron
hook
A
have
been
Roman
settlements.
may
Age
used to suspenda
Iron
and early
Some
lead
had
from
the
of
perforated
rims.
weights
recovered
pots
early
of
number
a
as
pot
for
fishing.
been
Rivets
found
have
in
used
were
may
a variety of pottery
Roman contexts
it
is
interesting
Hedges
kitchen
i.
Wait
that
that
particularly
note
and
ware,
e.
fragments and
imported
was
and
mortaria,
repaired
as
ware
well
as
ware, such as
local grey ware, storage
the settlement,a rotary quern from Derbyshire and a
Two
at
recovered
were
querns
saurian.
Germany.
from
saddlequern
Roman
the
Ingredients
period settlement
early
at
8.5.3.
Roman
bones
the
period settlement
early
at
The animal
between
in
difference
the
of
some
contexts
phasing
myself and Wilson and
The slight
frequency
bones
the
the
has
of
affected
of particular species (see
not generally
Levitan
8.6.
8.11.
Figure
Figure
).
8.12.
and
shows that the representation of early
Figures 8.11. and
bones
in
those
late
Iron
Where
Age.
parallels
the
sheep
species
and
cattle
Roman period
bones
in
than
late
in
Iron
Age,
the
numerous
pig
bones
more
the early
slightly
were
horse
These
bones
differences
more
common.
slightly
are
subtle
pig
reflect the
Roman period
bones
the
bones
two
the
392
(excluding
at
animal
of
settlements:
sizes
differing sample
1,648
bones
Roman
late
to
Iron
compared
Age
settlement
the
period
the
at
early
teeth) at
Study of the butchery practices below, and following that, the distribution
settlement.
216
difference
help
in
there
to
the
treatment
the
whether
clarify
was
a
various
of
will
patterns,
late
Iron
few
Age
identified
As
the
settlement,
with
wild
at the early
species
were
species.
Roman period settlement. There were a slightly higher proportion of dogs identified at the
later settlement, which may indicate that they had a stronger presence at the settlement. A
identified;
bone
bones
deer
three
butchery
and
interestingly
was
cat
were
recorded,
single
bones.
found
Eleven
bird
bones
two
the
from
of
cat
on
were
were
marks
early
recovered
Roman contexts; the species identified include goose, duck and a number of unidentified
fowl.
Butchery practices
bones
The small sampleof animal
affects the determinationof the butchery practices of the
Many
found
bones
the
Roman
do
of
settlement.
marks
period
are
only
on
single
and
early
butchery
The
discussions
a
pattern
as
such.
therefore
present
represent
will emphasize
not
location
butchery
is
duplicated.
bones
the
Both cut marks and
the
of
where
marks
those
bones
detected
Butchery
the
the
on
of
were
various
species.
marks
marks on cattle
chop
lower
found
limbs, i. e. metacarpal,tarsal and phalanges,similar
the
on
bonesare primarily
late
bones.
Iron
Age
Chop
found
the
cattle
that
on
marks were also found along the
to
for
late
Iron
Age
the
the
bones;
however,
also
case
was
which
cattle
chop
marks,
vertebrate,
in
that
both
body
the
suggesting
choppers
pelvis
these
the
were
on
used
occur
areas
of
also
bones.
As
the
the
with
earlier settlement there was minimal evidence of the
to separate
bones,
long
the
which suggeststhat either meat was cooked on the bone or,
butchery of
bone.
Evidence
butchery
likely,
the
the
off
right
of
stripped
of sheepwas minimal and
more
i.
femur,
bones,
found
the
Chop
e.
meaty
more
vertebrae
the
on
and
pelvis.
marks were
was
indication
is
late
the
Iron
found
the
the
Age
and
as
pelvis
with
on
settlement,that meat
only
largely
bone.
The
the
butchery
found
consumed
on
and
cooked
was
only
mark
on
of sheep
had
the
blow.
split,
result
of
a
a
mandible
ventral
where
was
pigs
less
typically associatedwith consumption is difficult to access.The
butchery
The
of species
did
display
bones
horse
not
any obvious butchery marks. A single tibia from
few recovered
the
is
had
the
cutting
on
The
some
dog
of
shaft
purpose
evidence
of
which
unknown.
a
butchery
the
is
interesting,
on
articulated
marks
radius
and
of
ulna
a
cat
of
as cat
presence
butchered
when presentat the other sites in this study (Maltby 1979:64 also
boneswere not
217
butchered
bones
in
Roman
Green
has
domestic
that
of
cat
period).
rarity
stated
on
comments
in
identified
Iron
have
Age
first
been
deter
the
that
they
to
and
may
primarily
used
cats were
She
bones
have
from
been
that
as
also
pets.
notes
cat
or
possibly
recovered
mice and rats
(1992:
25-6,102).
Danebury
at
ritual contexts
Meat yields
The grouping of the bones of the domestic speciesaccording to their meat yield reveals
late Iron Age assemblage.As Figure 8.13. indicates, cattle bones
from
deviation
the
some
between
distributed
bones
the
three
groupings,
evenly
although
with moderate
were quite
dominant
in
late
bearing
Iron
Age
the
more
whereas
slightly
were
non-meat
meat yield
This
common.
could indicate that cattle were increasingly
bones were slightly more
distribution
The
bones
identical
the
to
that
of
sheep
settlement.
was
of
virtually
at
consumed
bones
`B'
high
Age
the
Iron
late
group
most
with
prevalent
and
meat yielding and nonthe
distribution
The
bones
late
bones
is
to
the
evenly
represented.
of
pig
similar
meat yielding
bones
that
bone
Age
except
with moderateamountsof meat (group `B') are
Iron
assemblage
difference in the representationof body parts at the early Roman
The
dominant.
main
more
in
horse
bone
found
late
Iron
At
the
Age
the
assemblage.
was
settlement
settlement
period
À'
`C'
dominant;
from
bones
and
group
were
group
horse
at the early Roman period
bones
`C'
that
`B'
bones
while
see
are
we
still
prevalent,
group
group
are now
settlement
distributions
horse
bones could very well be caused by the
different
The
of
dominant.
it
However,
is
in
horses,
could
also
reflect
the
a
change
size.
use
of
which
smaller sample
horse
bones found on the settlementand the
by
the
percentage
smaller
of
further suggested
bones.
the
butchery
on
marks
lack of
Theearly Romanplant remains

At the early Roman settlement thirty-eight contexts were sampled for their carbonized
include
features
ditches, pits and gullies. The carbonized
The
types
sampled
of
remains.
from that of the late Iron Age settlement,with wheat accounting for 51% of
differ
samples
barley
21%
by
followed
at
and oats at 1%, while the remaining cereal
identified grains,
Grains
identifiable
to
species.
and
weeds
with
amounts
not
small
of chaff
were
grains
it is thought that the settlement was also a `consumer',
the
dominate
and
sample
similarly
218
indicate
The
higher
that
the
the
of
grain
may
crops
were
site.
numbers
grown
at
although
is
items
late
identified
Iron
Age
to
the
small,
still
very
as
compared
settlement at
of
numbers
is
have
Farm,
to
Court
considered
Barton
produced its own crops. The range of weed
which
identified
for
Roman
The
low
types
is
the
settlement.
example,
early
of weeds
at
taxa
bromus
knotgrass
those
and
are
associated with cultivated or
cleavers,
sorrel,
sheep's
disturbed land. The small number of querns recovered from the settlement (two) is also
imported
Palmer
(Miles
indicate
the
to
the
that
cereals
were
to
settlement
and
thought
identified
bean
legumes,
Other
the
89).
are
species
celtic
and
and although
edible
1983a:
it
is
important
that
they
presumed
minimal,
made
a
more
contribution to
their numbers were
draft
(Straker,
2).
their
diet
suggest
report:
than
numbers
the
Summary of the Artefacts
8.6.
and Remains of consumption at each Settlement
distribution
it
the
the
the
the
two
of
artefacts
and
remains
at
settlements,
Before we consider
is necessaryto summarizesome of more salient points made thus far on the containers,the
for
ingredients
the
implements
each
phase.
and
other
8.6.1. Late Iron Age settlement
Containers
Pottery
Imported wares found at settlement including mortaria, amphorae and possibly
samianware.
bowls
Equal
and
amphorae
were
of
recovered.
numbers
.
in
layers.
1
BB
upper
presentat settlement
.
jars indicate a seriesof small and medium sizedjars.
Histograms
of
"
bowls
indicate
bowls
large.
that
Histograms
were
quite
of
.
6%
up
of
`Fine'
make
assemblage.
wares
specialist
and
.
have been important and that
that
have
coarse
non-local
ware
may
suggested
"I
have
been
`fine'
considered
also
may
wares
as
as tankards were made
red/orange
fabric.
from
this
only
Glass
"
Small amountsof early glasspresentat settlement.

bottles
blue/green
body
fragments are of appropriate
Hofheim
and
A
cup and other
date.
219
Other implements
iron
links
iron
to
of
an
chain
used
possibly
suspenda container
and
vessel
small
"A
in
late
Iron
Age
identified
contexts
were
identified
the
were
recovered
no
querns
whetstones
were
at
of
number
"A
settlement.
"
Significant numbers of pieces of clay oven and oven plates were recovered.
Ingredients
Animal bones
"
"
"
"
"
"
Cattle were the most commonly identified species followed by sheep, horse and pigs
Small numbers of dog, cat, and bird, including the raven, were identified. A single
deer antler was recovered.
Butchery marks indicate that meat from the more meaty bones of cattle, sheepand
bone.
the
off
or
cooked
on
pigs was either stripped
Minimal indications of the butchery of horse,the types of marks identified appearto
indicate the skinning of non-meatybones.
Cattle have slightly higher numbers of bonesfrom group `C' and horse have higher
bones.
`C'
À'
and group
numbersof group
Sheepand pig bone assemblagesare both dominatedby meat yielding bones.
Plant Remains
identified
followed
by
Barley
the
commonly
grain,
most
was
spelt wheat and very
"
small numbersof oats.
to
Higher
that
chaff,
small
and
weeds
numbers
grain
suggest
grains present
of
ratio
"
for
consumption.
were
8.6.2. The early Roman period settlement

Containers
Pottery
imported
including
Increase
ware
of
mortaria and amphorae.
percentage
"
Gaulish samian ware was identified in
Significant
plain
south
of
primarily
numbers
.
it
Platform
1,
is
likely
least
on
quite
that
contexts
spread
at
some of vessels
general
were usedat the settlement.
jars
identified
lower
There
of
percentage
the
a
slightly
are
settlement, and an
at
"
increasein drinking vessels.
bowls
Again,
are
equally
represented.
and
amphorae
"
jars
lids
in
Increase
of
narrow-necked
and
at settlement.
numbers
"
indicate
jars
jars
to
increase
in
Histograms
small
medium
sized
the
of
a
slight
with
"
jars.
large
numbersof
bowls
indicate
large
but
bowls
bowls
are
still
Histograms
of
small
are now evident.
"
BB1
the
imitation
at
are
common
settlement
BB1
and
"
220
Increasein percentageof `fine' and specialty ware to 13% (does not include samian
beakers,
flagons
fabric
the
tankards
of
of
which
many
and
the
out
red/orange
or
ware
formed).
are
"
Glass
light
body
fragments
jugs,
bottles
blue/green
and
green
were
and
of
variety
"A
identified at the settlement.
in
22%
identified
have
the
Cool
the
at
Price
glass
use
site,
of
glass
a
peak
early
and
"
identified at the settlement was from the early Roman period (14% of identified
8%
first
AD
from
to
the
century
and
are
of
mid-first
mid-second
are
vessels
date).
AD
centuries
Other implements
knife
Iron
recoveredat settlement
"
Large iron hook - possibly used to suspend a container was identified.

for
fishing
typically
type
the
Lead weights of
used
were found.
Two quernswere identified, one from Derbyshire the other from Germany.
Ingredients
Animal bones
followed
identified
by sheep,pig and horse.
Cattle were the most commonly
species
identified;
bird
bones
bones
dog
Higher
small
were
numbers
of
cat
and
of
numbers
"
boneswere identified. A single deerbone was recoveredat the settlement.
butchery
late
is
identified
the
type
to
the
butchery
The
similar
of
at
practice
of cattle
Iron Age settlementexcept that the pelvis was also butchered. There was minimal
butchery
that
the
and
sheep
pig,
suggests
of
which
meat was consumed
of
evidence
found
butchery
horse.
No
bone.
the
marks
were
on
obvious
off the
bones
butchered.
butchered,
bone
dog
two
and
cat
were
was
"A single
bones
from
indicate
`C'
fewer
cattle
There
of
group
an
numbers
which
may
are
"
increasein the consumptionof beef.
Sheepand pig bonesdisplay similar distribution of body parts to those speciesat the
late Iron Age settlement,i. e. higher percentageof meat-bearingbones.
dominant.
`B'
from
bones
are now
Horse
group
horses
have
the
the
is
that
of
may
It
significance
as
changed
at
settlement
possible
"
lack
identified
in
butchery
drop
the
the
by
of
species,
the
percentage
of
marks
seen
body
distribution
different
parts.
of
and
Plant Remains
followed
by
barley
by
dominated
Cereals
and
wheat,
of
oats.
very
small
numbers
are
"
indicate
that
in
may
of
grains
crops
Increase
grown
at
settlement
were
percentages
"
but small numbersare still associatedwith a `consumer' settlement.
8.7.
The Distribution
Eating
Drinking
Remains
the
of
and
of
at Claydon Pike
distribution
the
the
significance
of
consider
following
will
patterns of the
section
The
Roman
late
Iron
Age
So-called
the
and
early
settlements.
period
at
remains
and
artefacts
221
for
brief
highlighted
be
deposits'
settlement
each
and
a
summary of the
`special
will
findings will be presented.
late
Iron
Age settlement
the
distribution
The
of artefactsand remainsat
8.7.1.
in
different
be
it
that
in
zones of activity might
reflected
Early this chapter, was suggested
late
Iron
Age
the
the
and
dispersal
artefacts
remains
at
settlement.
various
of
the
density
indicate
in
to
the southern area of the
a
of
occupation
appear
Distribution patterns
distinctions
in the types of deposits in the two
determine
there
To
were
whether
settlement.
features
to
their
I
the
southern
or
northern
according
position.
organized
areas
found
in
between
the apparent concentrationof
two
the
areas
was
geographical
A contrast
brooches,
dress-fittings
bracelets
bone
copper
pins and
and
items of personaladornment
in
the
particularly
settlement,
and around enclosures six and
in the southern section of
fragments
from
the
the
two
The
glassware
were
also
of
recovered
majority
seven.
found
in
Artefacts
this
Two
that reflect
were
coins
copper
area.
enclosures.
pre-conquest
for
loom
balance
and
plugs
as
rivets
repairing
weights,
such
pots,
day-to-day activities,
fragments,
in
the
were
the
concentrated
oven
and
area
of
southern
whetstone,
a
weights,
below,
be
Large
and
animal
remains,
of
pots
which
will
commented
on
groups
settlement.
In
in
devoid
the
this
contrast,
although
area.
northern
of
area
was
not
recovered
also
were
had
debris,
domestic
contexts
such
many
small
concentrations
and
of
artefacts
and
personal
remains as to suggest different uses of the area. The excavation report, for instance,
have
functioned
is
There
the
may
gullies
of
that
a
as
paddocks.
a number
indicates
in
the
the
section
of
to
settlement
northeast
which
points
a
possible
slag,
of
concentration
However,
in
distribution
the
the
this
settlement.
quantifying
of
part
of the
industrial use of
it
in
domestic
have
that
the
two
appears
taken
areas
activity
may
remains
animal
pottery and
the settlement
of
beyond
the
area
southern
place
222
Southern area
Northern area
n=1873
n=1754
21
20
white ware
88
158
5
9
23
221
1
13
red/orange wares
Shell/limestone temper
brown wares
re /black wares
183
245
299
866
13
10
13
16
46
1
100
506
447
431
6
6
29
25
25
34
.
colour-coats
misc.
fabrics
Iron
Age
Late
8.2.
Table
pottery
- southern and northern areas
The types of pottery and animal remains recovered from the two areas are nevertheless
distinct from eachother. Table 8.2., which plots the distribution of the pottery fabrics at the
has
illustrates
higher
the
that
brown
northern
area
a
the
proportion
of
wares and
settlement,
in
late
Iron
Age
the
wares
common
calcareous
whereasthe southern area of the
the early
fewer
higher
has
these
types
examples
of
and
a
proportion of the precursorsof
settlement
dominate
to
Roman
the
go
on
ware,
which
grey
early
the sandy
period. The southern area
local
higher
the
has
of
red/orangeand white wares, which add much to the
proportions
also
in
the
wares
southern area of the settlement. In terms of the
of
represented
palate
colour
distribution of identified pottery forms it is particularly interesting that no significant
found
in terms of suggestedkitchen and tableware,
between
the
two
areas
were
distinctions
higher
deposited
a
exceptions,
percentage
in
two
notable
of
amphorae
the
were
with
dominant
the
in
and
early
settlement
mortaria
(see
were
of
the
area
southern
area
northern
households
indeed
lived at the settlement it is quite possible
If
8.14).
of
a
number
Figure
display
different
traditions
consumption
well
of
as
as
of
personal
the
that
majority of small
found
in
the southern area- have been identified at
were
adornment
finds associatedwith
the settlement.
found
bones
in
between
the
distinction
the two areasof the settlement are slight
animal
The
Figure
8.15
shows that a slightly higher proportion of animal
but are potentially significant.
in
boneswere situated the southernareaof the settlement.Cattle, followed by sheep,are the
in
both
distinction
although
the
between
areas,
clearly
species
the two
prominent
most
in
the
the
is
Pig
bones
southern
area
of
settlement.
overt
most
are slightly more
species
horse
horse
is more prominent than pig in the
in
than
the
whereas
area
southern
prominent
bones
It
is
the
higher
settlement.
at
represented
of
that
of
number
note
a
proportion
overall
223
6
7.
The
high
identified
distribution
bones
and
around enclosures
of
and
was
of cattle
higher
is
in
bones
than
the
slightly
also
cattle
northern area,whereas
yielding
meat
moderate
in
is
bones
distribution
the
two
these
types
areas comparable (see Figure
the
of sheep
of
8.16. and Figure 8.17.). Seventy-sevenpercent of the cattle bones with clear evidence of
butchery were also found in the southern area of the settlement.This may indicate that the
had
for
beef.
in
living
the
the
consumption
a
predilection
of
southernarea
people
Figure 8.18., which lists the distribution of body parts according to their meat yield, reveals
bones
large
the
of
or small speciesor meat yielding and non-meat
of
no obvious patterning
horses.
This
for
bones
that
the
as
a
whole,
suggests
a
settlement
except
perhaps
yielding
itself.
for
does
disposal
identifiable
the
regime
whole settlement
not present
consistent and
Analysis of specific features in the southern area does, however, suggest that there were
distinctions in the deposition of animal bones among the features in this area. Table 8.3.
in
from
bones)
bones
`C'
(non-meat-yielding
indicates
that
(below),
group
were prevalent
both the boundary ditches and the enclosuresand that meat-yielding bones were prominent
in the gullies and pits. Table 8.3. also highlights the scarcity of some species in certain
from
bones
from
horse
for
pits,
of
pig
and
were
recovered
example, small number
contexts;
deposition
had
is
it
ditches;
boundary
the
that
these
of
species
particular
possible
gullies and
bird
identified
bones
below).
Seventy-nine
(see
the
the
at
settlement
of
percent
significance
in
from
features
the
the
southern
area
of
settlementand sixty-eight percent
recovered
were
in
found
bones
butchered
this
area.
were
of all
As has been discussedin previous chaptersconclusions on the contextual associationsof
best
because
limits
When
tenuous
the
the
at
of
sampling
of
strategy.
are
plant remains
in
found
found
it
the
the
northern
and
southern
settlement,
of
was
samples
area
contrasting
from
however,
%
75
the
the
were
recovered
remains
plant
area;
more
that
of
northern
from
The
in
this
taken
the
the
area.
composition
samples
southern
of
actually
were
samples
indicate
higher
than
has
the
that
of
chaff
northern
area,
may
percentage
which
cereals
a
area
is
in
this
this
although
area
speculative considering the small amount of
were processed
items recovered and the lack of supporting artefactual evidence. Two whetstones were
in
Age
in
late
Iron
the
contexts,
one
each
area
of
settlement.
recovered
224
Group A
Group B
%n
21
21
8
80
130
16
16
6
62
30
enclosure
4
9
4
34
Total
Group C
%
36
21
5
89
151
24
14
3
59
34
28
10
4
117
159
18
6
3
74
36
8
18
8
67
10
20
10
94
7
15
7
70
9
12
6
27
17
22
11
50
51
21
134
56
54
23
enclosure
2
6
1
17
8
23
4
65
6
7
1
14
21
25
4
50
1
0
0
12
8
0
0
92
Total
26
39
28
42
13
19
3
2
15
10
2
2
15
15
5
0
20
0
1
14
20
5
70
34
0
9
13
0
69
22
3
17
25
12
68
43
Cow
ditch
gully
it
enclosure
Total
Sheep
ditch
gully
- it -
Pi
ditch
ull
it
Horse
ditch
gully
pit
enclosure
Total
Table 8.3. Distribution of late Iron Age animal bone groups by feature in the southernarea
'Special deposits' at the late Iron Age settlement
The identification of so-called `special deposits' is not obvious at late Iron Age Claydon
Pike. When you consider the distribution of identified pottery forms according to their
featuretype (see Figure 8.14.) we can seethat for both areasof the settlement pits and to a
lesser extent gullies are devoid of pottery associatedwith the serving of food and drink.
Thesetypes of wares are primarily found in the ditched enclosures.On the other hand, it has
been established that in the southern area the types of animal bones recovered from pits
indicate
This
that pits were the receptaclesfor particular
yielding.
could
were generally meat
It
is
that
consumption.
meat
emphasized
also possible that the deposition of certain
meals
bird
bones,
for
followed
instance,
mores,
particular
species
were not recoveredfrom pits and
found
bone
in
horse
together
the pits in the southern area. This presentsa
were
not
pig and
between
distinction
if
the
preparation,
not consumption, of types of meat and the
possible
liquids
and the use of vesselstypically associatedwith display. This is in
consumption of
225
identified
Farm
Court
Barton
late
Iron
Age
that
to
where
pits
were
as possibly
contrast
included
beakers
bones
by
their
specialized
ware
such as
and
representing special meals
bowls.
late
Iron Age settlement
the
Summary of the re-contextualizedmaterial at
It has been established that the remnants of household activities associated with food
dense
in
the
the
especially
southern
were
area
settlement,
consumption
of
and
preparation
in
Whereas
is
in
the
the
there
section.
northern
southwest
area
evidence of
particularly
smithing, ring stacks as well as a variety of enclosures and sub-enclosures with minimal
domestic debris, which may have held livestock. The types of artefacts recovered from the
include
items,
jewellery,
finds
the
personal
especially
and
small
settlement
southern area of
loom
for
household
such
as
weights,
and
production,
rivets and plugs
associated with
including
food
linked
Artefacts
to
preparation,
early mortaria and oven
repairing pottery.
fragments and large groups of kitchen and tableware, including glass vessels, as well
butchered animals, particularly cattle, and bird bones were situated in enclosures and
boundary ditches in this area of the settlement. Together with the configuration of the area
is
it
likely
houses
divisions
internal
that
its
quite
enclosures,
were situated
and rounded
with
in this section of the settlement.
in
been
the
that
the northern areaof the
has
of
enclosures
a
number
situated
It
suggested
also
The
houses.
have
co-prominence of sheep and cattle, and the
contained
settlement may
in
and
the northern
articles
glassware,
mortaria,
of
of
personal
adornment
minimal presence
distinction
The
is
late
Iron
Age
of
a
south/north
possibility
reminiscent
of
notable.
are
area
Barton Court Farm where there was a strong casefor the existenceof two domestic areasof
differing status (although in the case of Barton Court Farm the northern domestic area
(see
higher
Rippengal
been
have
1995
identified
to
also
status)
of
similarly
who
a
appears
in
Roman
distinction
Bancroft).
As
the
for
early
theorized
status
at
site
at
was
north/south
Barton Court Farm, it is possible that some of the distinctions may reflect different usesof
the settlement over time, although a number of the features in the northern area are
in
features
for
1,2,6
7
the
the
south
example,
with
and
are
enclosures
contemporary
stratigraphically similar - and northern contexts contained chronologically comparable
226
different
households
Whether
two
with
consumption practices and
we are seeing
pottery.
in
different
the
change consumption and expressionsof personal status
statusor
possibly of
it
is
Roman-like
that
that
time,
of
wares
variety
are
were
a
clear
of a single group over
in
Roman
its
the
to
the
early
period.
reorganization
settlementprior
present at
8.7.2. The distribution of artefactsand remainsat the Early Roman settlement
The distribution of artefacts and remains at the early Roman settlement are distinguished
from the late Iron Age settlement by the change in the structure of the settlement. A
ditch
domestic
defines
boundary
Roman
that
the
the
encloses
area
early
period
rectangular
found
in
direct
Rectangular
are
are
evident
and
pits
structures
association with
settlement.
The
main entrance to the settlement was positioned alongside a road.
structure.
the main
The organization of the settlement is directly comparable to early Roman Barton Court Farm
late
ignored.
be
Iron
Age
It
layout
both
the
in
the
settlements
was
will
of
cases,
and
distinctions
between
late
Iron
Age
here
the
the
that
other
and early
of
some
suggested
Roman phases at Claydon Pike represent changes in practices directly involved with
drink.
food
and
consumption of
The distribution of containersaround the early Roman settlement reveals some particularly
interesting patterns. Glass vesselswere recoveredprimarily from the boundary ditches that
flank either side of the road. That the majority of glassvesselsfound in context were bottles,
important
for
bowls,
that
glass
was
suggests
than
or
containing and pouring
say cups
rather
liquids. The majority of the beakers, flagons and tankards recovered from the settlement
ditches
boundary
(see
Figure
in
8.19.
)
(drinking
found
these
vessels were also
were also
found in pits which will be commented on below). It is quite possible that deposits of
drinking
the
Distinctions
to
the
out
singled
main
access
with
settlement.
associated
vessels
between the ditches that flank the road were also found in the deposition of the animal
remains.
The vast majority of animal bones at the early Roman settlement were recovered from
ditches (see Figure 8.20.) these findings are slightly different from that of Levitan as his
bones
in
in
distribution
list
Figure
8.21.,
did
Roman
the
the
of
pits
early
period.
not
analysis
227
distribution
bones
looks
both
to
the
their
that
according
of
at
meat
yield,
shows,
meat
which
bearing and non-meat bearing bones of each specieswere well representedin the ditches,
few
bearing
bones
The
the
non-meat
were
of
pig
where
recovered.
exception
majority
with
butchery
from
ditches.
There
bones
the
marks
were
also
recovered
actual
with
are a
of
found
in
between
bones
distinctions,
however,
ditches
flank
the
the
that
the two
of
number
bones
Almost
the
cattle
and
equal
numbers
of
sheep
were recoveredfrom the
road.
sides of
ditches on the south side of the road, which is in contrast to other areasof the settlement,
higher
dominate
bones
the
assemblages;
proportions
of
most
cattle
sheep
recovered
where
from the south side of the road were from groups À' and `B' - the edible portions bones
from
`C'.
large
42%
A
(relative
the
to the numbers
cattle
were
group
of
whereas
dog
bones,
burnt
the
group
of
some
of
which
at
settlement)
were
articulated,
and
recovered
bones were also recoveredfrom the south side of the road as were a sizable number of horse
bones
from
bones.
In
the
small
of
animal
contrast,
only
numbers
were
recovered
and pig
ditches that flank the northern side of the road and they were dominated by meat bearing
bones,
horse
bones.
Only
small
numbers
cattle
with
very
of
pig
and
and
particularly
sheep
identified
bones
Roman
bird
the
at
early
period settlement, nine of which were
were
eleven
deposited in the ditches that flank the road. The prominence of sheep bones in the ditches
leading up to entrance of the settlementtogether with containers associatedwith drinking
in
and around the structure and at the
may reflect contrasting consumption practices
boundary of the settlement.The incidencesof dress-fitting items recoveredprimarily in the
ditch that flanks the south side of the road further conjures up images of people
in
congregating this areaof the settlement.
Unlike the late Iron Age settlement,fewer artefacts and remains were recovered from the
internal featuresof the settlement.Notable exceptionsare the pits situated around the aisled
building which were quite large and held sizeable deposits. The early Roman period pits
both
kitchen
linked
Amphorae
tableware.
selection
of
to olive oil and
a
varied
and
contained
fish paste, mortaria, beakersand bowls were recovered in significant amounts. These pits
differ from the late Iron Age pits, which contained primarily jars and small amounts of
The
with
olive
oil.
contents of some of the pits may reflect the
associated
amphorae
house.
The animals bones recovered from pits
the
associated
meals
with
of
consumption
indicate different types of depositsfrom those identified in the ditches (see Figure 8.22. and
Figure 8.21.). Cattle bones are still the most common bone identified, although 59% of the
228
bones are non-meatbearing.In contrast,sheepbones are primarily meat bearing. Only small
in
identified
bones
horse
bones
horse
the
pits,
were
majority
and
pig
of
of
were
numbers
from group `C' (mostly cranial bonesfound in a single pit - see below) whereasall the pig
bones are meat bearing. Butchery marks are rare on the bones recovered from pits (a sheep
tibia and a dog tibia eachhad cut marks).
It is possible that the sheep and pig bones recovered from the pits represent the remains of
bones
that
they
the
that
or
roasted
whole
were
separated
were
at
and
meals
particular
ligaments. There is some similarity between the deposits of animal bones in late Iron Age
bones,
for
instance,
Bird
Roman
(only
bone
pits.
are
virtually
period
absent
one
and early
for
both
horse
bones
few
from
in
identified)
the
these
and
and
phases
pig
pits
are
was
held
feasible
is
It
the
towards
that
of
mores
particular
species
some
of
animals
at
contexts.
held
Roman
Age
late
Iron
the
at
early
period settlement, although
settlement were still
the
in
Roman
the
tableware
the
period pits points to a possible
early
of
presence
clearly
distinction in consumption practices.
Only a couple of early Roman (ER1) plant samples contained more than ten items, the
distribution
in
discussion
is
this
the
on
patterns
consequently
plant
remains
of
consideration
8.5.2.
)
indicate
(section
As
the
limited.
ratios
of
grains, chaff and weeds
we saw above
quite
Roman
Most
the
the
settlement.
at
early
of
product
contexts sampled
processed
more
a
indicate very small numbers of items except in the case of samples recovered from a section
547)
(this
had
ditch
(context
few
the
the
boundary
side
of
on
north
road
context
very
of
items
72%
The
it
bones).
the
of
contained
all
recovered
sample
at
settlement;
was
animal
23%
The
1%
66%
items
weeds.
ten
chaff
and
only
grains,
sample
other
with
of
or
made up
ditch
from
(context
620),
(this
had
the
the
the
taken
south
on
side
of
road
was
context
more
large number of bones, particularly sheep bones). This sample contained 16% of the total
items recovered at the settlement; it was made up of 20% grain, 31% chaff and 48% weeds.
A rotary quern was also recovered from this context. The plant samples should not be taken
it
is
interesting
from
deposits
face
that
the two sides of the
once
again
although
value,
at
from
distinguished
each other.
road are
229
'Special deposits' at the early Romanperiod settlement
At the late Iron Age settlement, it was the accumulation of particular types of species and
bones, small finds and containers in particular areas that was distinctive (rather than single
isolated deposits) - evidence it seems, of the everyday ritualized deposition of material. This
Roman
been
The
have
the
the
types of
to
at
early
settlement
case
period
also.
appears
deposits recovered from the ditches that flank either side of the road leading to the entrance
house
by
from
the
the
associated
with
appear
specialized
the
pits
what
and
settlement
of
in
they signify
comparison to other areas of the settlement.
Similarly with the late Iron Age settlementthe isolation of `special deposits' is not obvious.
A number of the pits associateddirectly with the structure contained large dumps of pottery
including table and specialty wares although no glassware.The presenceof these types of
deposits
does
in
the
themselves
special per se except when one
not make
and of
wares
deposits.
Butchery
found
the
the
animal
marks
were
rarely
on
nature of some of
considers
bones
horse
in
deposited
bones
and
particularly
were not commonly recovered
the
pits; pig
from pits; pit 2519 standsout becauseit containeda deposit of horse cranial bones,where
from
This
bone
horse
the
was
recovered
metacarpal
other
pits.
pit also
only one other
-a
large
jars
the
types
and
a
series
of
amphorae
of vessels
contained a selection of mortaria,
beaker
food
a
and a variety of cattle, sheep
preparation-as well as
that are associatedwith
deposit
This
bones.
the
could
represent
remains of a special or ritualized event that
pig
and
deposits
in
in
the
the
the other pits associated
could
as
varied
structure,
took place or around
with the structure.
One final comment on special depositsis the presenceof the antithesis of a special deposit.
None of the south Gaulish samian ware recovered from Platform 1 was recovered from
it
find
I
this
the
to
rather
unusual,
and
may
point
very special-nessof
contexts.
phased
deposition
least
to
a
of
process
unlike the other artefacts and remains
samian ware or at
recoveredat the settlement.
230
Summary of the re-contextualizedmaterial at the early Romanperiod settlement
The change in the structure of the early Roman period settlement is exemplified by the way
drinking
were deposited. The majority
the artefacts and remains associated with eating and
found
from
in
boundary
ditches.
the
were
the
settlement
recovered
and
remains
artefacts
of
Two particular stretches of the ditches - those that flank either side of a road off which lies
the main entrance to the domestic area of the settlement - stand out from the other ditches.
These ditches together contain most of the drinking paraphernalia found at the site, although
beyond this point the northern and southern stretches of ditch are distinct from each other;
dress-fittings
ditches
items
the
of
a
number
contained
and
of personal adornment as
one of
house
deposits
The
found
the
types
those
with
reveal
similar
associated
of
pits
as
at the
well.
interestingly
drinking
the
to
vessels - although
settlement,
no glassware - specialty
entrance
fish
diet,
Mediterranean
for
linked
to
the grinding of
oil,
paste
and
mortaria
a
olive
wares
important
Sheep
diet
to
the
were
also
of
sauces.
making
and
as evidenced by the
spices
bones
in
features.
The
these
quite
visible
sheep
meat-bearing
prevalence of
prominence of
in
both
bones
deposits
these
cattle
areas
may
signal
yielding
specialized
of the
non-meat
head and feet bones. At the early Roman period settlement the specialized deposits appear
divided between the house itself and entrance to the settlement.
8.8.
Discussion of the Distribution
of Artefacts and Remains at Claydon Pike
The structure and restructureof late Iron Age and early Roman period settlementat Claydon
Pike echo what occurred at Barton Court Farm. At both late Iron Age settlements, the
households
different
two
of
resident
status, or one main household with
possibility of
different expressionsof eating and drinking was presented. The reconfiguration of both
early Roman period settlementseach saw the establishmentof single rectangular structures
linking
of the entrance to the settlements to roadways. The artefacts and remains
and
drinking
from
the two sites, however, point to very
recovered
associatedwith eating and
different consumption practices and these will be discussed in more detail in the next
Pike
Claydon
At
even though the consumption practices at the late Iron Age and
chapter.
distinct,
Roman
were
quite
settlement
period
unlike at Barton Court Farm, the late Iron
early
231
Age settlement appears to have adopted a taste for preparing and flavouring their food in
in
to
the
change settlement.
particular ways prior
8.8.1. The late Iron Age settlement
A number of potential scenarios have been presented for the late Iron Age settlement at
Claydon Pike. Initial analysis of the distribution patterns of the artefacts and remains
in
habitation
the southern area of the settlement with animal
to
a single area of
pointed
living
in
However,
the
north
of
areas
metal
working
space.
paddocks, ring stacks and
in
form
features
to the
to
these
enclosures
contemporary
number
of
similar
are
a
addition
This
likely
that
puts the artefacts and remains
contained structures.
two southern enclosures
in
in
to
the
the
comparable
are
which
numerically
south,
areas,
a new
northern
recovered
light. The types of artefacts and remains that singled out the southern area, such as personal
items, early glass vessels, early mortaria, meat bearing, butchered cattle bones are in
from
These
the
to
types
the
remains
recovered
northern
and
area.
of
artefacts
contrast
do
but
the
typify
or
storage
of
goods,
waste
products
not
rather have
their own distinct repertoire of table and specialty wares and an emphasis on meat-bearing
habitation
is
bones.
It
the
that
area of the settlement shifted over time, a
quite possible
sheep
distinction
between
fabrics
in
features
the
the
are
and
a
area
earlier
northern
of
number of
had
higher
does
that
the
area
proportion of the calcareous wares.
northern
a
the pottery
show
However, as has also been established, a number of the enclosures are contemporary and the
is
in
lies
There
two
the
the
the possibility of
similar.
areas
pottery
chronological range of
two households with different expressions of consumption living together at the settlement.
Although there are suggestions that the southern area may have been of higher status or had
for
high/low
distinction
between
the
taste
the two
consumables,
status
some
a continental
because
is
tableware, mortaria and particularly amphorae, plus a greater
groups not clear cut
in
bones
found
horse
the northern area as well.
the
are
percentage of
If two householdsof differing status or social identities did live at late Iron Age Claydon
Pike, this is particularly interesting, as it has been proposed by Miles (1984) and arguedby
Rippingal (1995) that in the later early Roman period, when the early Roman (ERl)
1,
beyond
Platform
there were two separatedoccupation areas of
settlement expanded
232
differing status. Towards the end of the occupation of the late Iron Age settlement, a
boundary ditch was erectedaroundthe whole settlement,which suggeststhat the two groups
were possibly allied with eachother.
8.8.2. The early Roman period settlement
The early Roman period settlementappearsfor the most part to have been occupied by a
in
A
household
building,
rectangular
a
single
structure.
second
residing
rectangular
single
is
in
in
`barn'
described
is
the
the northeast sectionof
excavation
report,
situated
a
as
which
dateable
is
lack
there
of secure
evidence and it is not clear whether the
a
the settlement;
building datesto the ER1 or ER phaseof settlement;a sheepburial, also undated,was found
One
homes
the
two
that
the
the
speculate
structures
could
structure.
represent
second
within
lack
finds
households
the
the
although
settlement,
of
associatedwith the
residing at
of two
highly
In
further
circumstantial.
any event, structureone
speculation
secondstructure makes
house
it
been
have
the
the
to
settlement
and
at
was enlarged when the early
main
appears
Roman settlementexpandedbeyondPlatform 1.
For the purposes of this discussion, we will assume that there was one household living at
been
has
It
established that the contents of particular features in the
the settlement.
deposits
form
to
that
to
specialized
represent
appear
point
some
of
settlement
for
food
in
Mediterranean
preparing
a
as
a
predilection
as
well
way (I refer
commemoration
here to the pits associated with the structure and the two stretches of ditches that flank a
is
There
distinct
to
the
the
settlement).
a
suggestion that the entrance
main entrance
road and
distinguished
in some way. That the
domestic
the
was
or
of
settlement
marked
the
area
to
is
indicates
to
the
the
to
the
and
close
road
entrance
situated
settlement
that the
structure
domestic area, though enclosed, was in the public domain. A similar complement of
drinking vessels, table and specialty wares and specialized animal deposits were found in
house,
directly
the
suggesting that the acts of consumption
the pits
associated with
associated with the structure and the entrance to the settlement were not a contradiction of
public and private consumption practices.
233
8.9.
at Claydon Pike
The presence of sizeable amounts of imported ware, particularly
at the late Iron Age
from
Claydon
Pike
Roman
but
the
the
sets
apart
settlement
period
early
at
also
settlement
Claydon
Pike,
Miles
In
in
the
this
thesis.
accounts
of
published
and
other case studies
Palmer have indicated that the site may have had a military
connection because of the
(Miles
imported
the
the
artefacts
associated
of
with
and
recovery
military
goods
numbers of
387-8;
Miles
1984:
1983b:
199,202,208).
Palmer
Miles
91-2;
1983a:
Palmer
and
and
Miles
have
Pike
Claydon
been
have
that
the
Palmer
to
may
possibility
requisitioned
pointed
and
by the Second Legion Augusta as a legionary prata (an amphora sherd with the inscription
`LEG II A. ' was recovered at the site); in the late AD 60's and 70's the Second Legion was
have
have
been
They
Gloucester.
that
the
settlement
also
speculated
may
garrisoned at
imperial
for
by
the
the
that
an
estate
was
collection of
settlement
or
even
veterans
occupied
by
(Miles
from
Palmer
taxes
administered
soldiers
and
areas
and
surrounding
produce
1983a:92; Miles 1984:208).
The military equipment found at Claydon Pike was analysed by Francis Grew. It was unThe
in
first
includes:
late
dated
been
has
but
the
the
to
equipment
century.
question
stratified
fragment of a harnessclip, a stud used on helmets or armour, a rivet possibly used to mount
leaf
leather
alloy
vine
similar to a military pendant
strapsand a copper
the military apron of
functioned
have
back
the
Grew
the
that
as
may
a
copper
vine
reflector
at
of
an
oil
notes
lamp as there are differencesbetweenthe example found at Claydon Pike and other military
4.2).
found
finds
Other
finds
have
(unpublished
the
appendix
at
report:
small
site
pendants
include,
BB
I,
Hofheim
flagon
linked
been
type
to
and
a
a
military
presence
and a cup,
also
jars',
`honey
The
of
amphorae
and
sizeable
amounts
and
mortaria.
very
narrow-necked
indicated
however,
Green
based
to
that
ware,
samian
pre-Flavian
of
on the
small amounts
is
`very
dubious'
(unpublished
presence
military
a
pre-Flavian
pottery
ceramic evidence
intaglios
device
Two
thunderbolt
one
an
with
were
recovered,
eagle,
and
glass
globe
report).
lighted
hands
both
have
been
dated
the
a
the
and
alter,
clasped
of
signet
gems
with
other
and
be
both
AD;
first
to
thought
are
of the type that were worn by soldiers (Henig
to the
century
4.3)
find
appendix
report:
unpublishedsmall
234
Hingley (1989) and Black (1994) have offered alternative explanations of the presence of
imported
Hingley
ware
and
military
equipment.
of
numbers
cites examples of
significant
(1989:
items
181
15)
were
recovered
which
note
military
and argues that the
civilian sites at
does
not mean that there was necessarily an official presence at
objects
of
military
recovery
the settlement. Bishop (1989), for instance, has emphasized the variety of ways that military
have
in
first
the
the
archaeological
entered
record
century AD, including as
equipment may
booty. Francis Grew who conducted the analysis of the military equipment recovered at
Claydon Pike cautions that "Many of the items found can be paralleled on known military
is
in
but
he
"The
to
they
that
that
say
were
only
used
such
not
a
context";
adds
number
sites
few,
though they are obviously
associations
military
are
actually
undoubted
of objects with
finds
4.2:
4).
Hingley
for
(small
(1989),
instance,
importance"
appendix
suggests that
of
Claydon Pike is an example of a high status native settlement (1989: 160). Black (1994), on
the other hand, argues that rural sites with military equipment may have been occupied by
in
in
Roman
(1994:
108-9).
He
the
army
some
capacity
concludes
served
native people who
in
farming;
indecent
Britons
but
be
it
"...
to
to
see
engaged
anything
that
almost
seems
instead new owners, immigrants, are introduced to explain
what are regarded as
from
homes
devoted
finds
to productive agriculture"
the
of
men
unacceptable
(Black
1994: 109).
At the beginning of this chapter, I stated that most of the published discussions on the
èarly' Roman period settlementrefer to the settlementonce it had expandedfrom Platform
1 onto Platform 2 by the late secondcentury. This study is basedon the initial early Roman
from
late
Iron
imports
Age
Early
(ERl)
the
the
within
context
of
settlement.
of
settlement
Roman-style material culture were present at the late Iron Age settlement; glassware,
been
have
identified
mortaria
at the settlementprior to the reconfiguration of
amphorae,and
the early Roman period settlement.The dating of theseartefacts as pre or post the arrival of
the army in the area is not absolute,although a number of the artefacts are definitely early
first century AD (e.g. mortaria). The contextual associations of these goods were not,
however, isolated finds but appear integrated into the daily rituals of the inhabitants.
Suggestionsof continental tasteswere also evident at the re-structured early Roman period
The
its
direct
links
the
of
configuration
newly
constructed
to the
settlement
with
settlement.
house
the
consumption
practices
associated
complementary
and
and entranceto
with
road
the settlementallude to a public adoption of certain aspectsof a Roman lifestyle. Yet, there
235
that
that
the
drinking
traditions
consumption
consumption
suggest
of
meat
and
patterns
are
habits
`Romans'.
These
the
necessarily
nor
of
emulative
wholly
are
neither
practices
late
Roman
both
Iron
Age
inhabitants
the
to
the
and
early
period
of
observations point
by
(ibid).
I
Hingley
being
as
suggested
would
of
some
wealth
native people
settlements as
further suggest that the people, who resided at the early Roman period settlement, as at the
late Iron Age settlement, had both the means and desire to display particular acts of
form
Some
be
the
to
the
of
military
presence
at
settlement
cannot
public.
ruled
consumption
have
388)
Dobunnic
(1983b:
Palmer
Miles
that
tribesman of
speculated
also
a
and
out, and
Claydon
have
Pike;
in
importance
the
at
occurrence
resided
of
may
military objects
some
line
indicate
be
it
the
of
communication
with
a
military
of a collaborative,
this context could
combative, or even social nature.
8.10.
Conclusion
Claydon Pike is the fourth and final case study in this thesis. It, along with the other case
in
diversity
Upper
Thames
Valley.
Similarly
illustrates
the
the
settlements
of
native
studies,
different
households
Farm,
Court
Barton
there
of
suggestions
are
with different
to
late
different
Iron
Age
the
status,
residing
settlement.
of
at
possibly
consumption practices,
There is also evidence, as with Barton Court Farm, of the possible consolidation of the
households within one structure; although the possibility of a second structure in the
the
the
the
and
positioning
of
main structure alongside the
settlement
of
section
northern
idea
the
the
to
of public and private actions that was put
settlement goes against
entrance
forward for Barton Court Farm. Instead, at Claydon Pike we have the possibility of a
in
Age
late
Iron
that
to
that
chose
settlement
express
wealth part through the use of
wealthy
Roman-style consumablesthat refer to the preparing and flavouring of food. At the early
Roman settlementwe seethe up front and communal use of the sametypes of Roman style
large
The
decidedly
in
the
the second
context.
native
expansion
settlement
a
of
within
goods
Farm
Thornhill
(neighbouring
its
links
time)
this
was
abandoned
around
and
with
century
importance
the
to
of the settlement.
attests
other settlements
A number of recurring themeshave pervaded the analysis of consumption practices at the
four sites in this study. Theseinclude the possibility of alternate criteria for what constitutes
236
importance
the
the status of a settlement and
apparent
of deposits at the, boundaries of
following
I
In
in
discussion
these
themes
the
that
explore
chapter,
will
a
settlements.
drinking
Barton
Court
Farm,
Roughground
the
at
eating
and
social contexts of
summarizes
Farm, Old Shifford and Claydon Pike.
237
-I
--------
-i
tl
:'
<
/r
r-- --
JT
r' -
-.
%^
oOO
e.
40
r-
n
J
ER-.
Figure 8.1 The late Iron Age and early Roman period (ER 1) settlement at Claydon Pike
(after unpublished O. A. U. drawings, drawn by M. Seymour)
23
Claydon Pike late Iron Age containers and ingredients
Figure 8.2 Late Iron Age pottery forms
LIA - pottery forms

n=328
90
80
70
60
50
40
30
20
10
0
'
Figure 8.3 and Figure 8.4 Histogram of the rim diameter of late Iron Age jars and bowls
LIA -jars
LIA - bowls
7
50
Ng.
a 40
30 20
4'.
3,
E2
E 10
Z0
N
CD
OA
OD
NNWWW
0)
N
OA
co
0-N
co
NNWW
O
0)
Rim diameter
co
rim diameter
Figure 8-5 Late Iron Age N. I. S.P. (Wilson)
Fi iwe S.6 Late Iron Age N. I. S.P. (Meadows)
LIA - animal bones

n=1734
LIA - animal bones

n=1648
60
60
50
40
50
40
%30
%30
20
10
0
ED
cow
sheep
pig
horse
20
10
0
cow
sheep
pig
horse
Clavdon Pike late Iron Age ingredients cont.
Figure 8. ' Late Iron Age animal bone groups according to meat vield
LIA - animal bone groups

60
50
40
DA
0B
0c
%30
20
10
0
U,
m
CD
'U
(D
"'4()
Claydon Pike early Roman period containers and ingredients
Figure 8.8 Early Roman period pottery forms
ER1 - pottery forms

n=284
80
60
50
%40
30
20
10 -oO
0
-OaoeO
aJo
(n
3-
a)
o
Z
n
vv
41
7c
v
0o
(0
-6
cD
Figure 8.9 and Figure 8.10 Histograms of the rim diameter of early Roman period jars and bowls
ER1 - bowls
ER1 -jars
U)
0' 40
5
N4
co
30 ,
w
0 20
w3
10
d2
a
ZN
Co
AO
IV
(7)
WW
IV
Co
c0
-N
Rim diameter
Co
NNWWA
O
CY)
Co
Z.
Rim diameter
Figure 8.11 Early Roman period N. I. S.P. (Meadovvs)
Figuue & /'
Earle Roman period N. I. S.P. IVdiIsonl
ER - animal bones
n=405
ER - animal bones
n=392
60
50
40
60
50
40
%30
%30
20
20
10
10
0-Q
0-Q0
cow
sheep
pig
horse
cow
sheep
pig
horse
'll
Cladon Pike carl Roman period ingredients cunt.
l r,Kurt 8 /3 1 arl Roman pcru)d ml imit I burr grutlp accot difw, to meat icid
ER - animal bone groups

80
60
DA
m8
O (--
aG
cow
sheep
pig
horse
24
Clavdon Pike: distribution of late Iron Age containers and ingredients
Figure 8. /4 Distribution of late Iron Age pottery
LIA - distribution
of pottery
120
ioo
i
eo
O ditch
eo
gully
40
O pit
zo
0
Figure 8. /5 Distribution of late Iron Age animal bones - south north
LIA - animal bone distribution

southn=912 northn=736
60
50
40
0 south
m north
%30
20
10
0
cow
sheep
pig
horse
Figure 8.16 and Figure S. I" Distribution of late Iron Argeanimal bone groups

south
60
50
40
south. north
north
60
0A
OB
oc
%30
20
10
0
50
C3A
40
"9
%30 1
0C
20
10
0
(D
oL
(D
V)
CD
,43
Clasdon Pike: distribution of gat: Iren :_: in_redlent> and carl\ Roman period containers and ingredients
Figure 8.18 Distribution of late fron Age animal bone groups by feature

pits

gullies

ditches
60
50
(3"
lii
iiii!
40
DA
OE
C3,
4 }0
OB
OC
cow
,n eep
pig
horse
Figure 8.19 and Figure S.20 Distribution of early Roman period potter, and animal bones
ER - distribution
of pottery
ER - animal bone distribution

120
100
Oo itches
ESgullies
80
60
100
80
Opt
60
0 wall
40
0 ditch
pit
0 wall
40
20
20
Ii
0-
c"
3vm-
CDO
mea
W
5i
oQaN
CD
L0
my
Figure -4.' l and Figtu L' "
Distribution of animal bone groups ditches pits

-

pits

ditches
80
70
ea
60
OA
%40
B
ac
20
.. n
see0
O-Y
"o se
cow
sheep
244
Chapter 9
Food and Drink Consumption and the Study of Social Change
Introduction
9.1.
fabrics,
forms
Behind the permutationsof pottery
meat-yields, butchery patterns, plant
and
drinking.
The
like,
the
the
minutiae
of
eating
the
of
and
contexts
social
are
ratios and
informative
digestible,
in
interesting
detail
is
a
and
much
so
way
challenge of presenting
deal
been
has
throughout
the
to
of
review
a
good
working of this
subject
and
considerable
thesis. No doubt this will continue.
detail
integrate
that
to
This chapter will serve
and provide an opportunity to flesh
some of
drinking
four
A
the
the
practices
and
at
eating
sites.
on
number
out some of my observations
further
have
the
they
case
studies
themes
preceding
and
warrant
permeated
of recurring
include:
These
comment.
"
How set notions of statusaffect the interpretation of the settlementsin this and other
Upper
Thames
in
Valley.
the
studiesof native settlement
"
How the identification of two or more groups with differing consumption practices
dynamics
the
within a settlement.
revealssomeof
"
How boundariesboth within and around the settlementsmight be considered from

imperial
context.
within an
This chapterwill concludewith a commentaryon early Romanperiod settlementin the Upper

ThamesValley asviewed through the presentstudy of the social contextsof consumption.
245
9.2.
Food and Drink Consumption at the Late Iron Age Settlements
The background for the consideration of diet and culinary practices at early Roman period
`native' settlements in this study has been the consumption habits at the late Iron Age
In
them.
the process of considering the early Roman period from
that
preceded
settlements
late
Iron
believe
I
Age,
than
the
the
that some sense of the
rather
vice
of
versa
context
within
level
has
been
household
imperialism
I
the
realized.
at
would also suggest that this
of
effects
late
Iron
Age consumption in and of itself (see
to
our
understanding
of
study will contribute
lack
by
Willis
late
Iron Age in its own right
to
the
the
points
of
study
who
of
comments
1994: 142). What follows is a review of some of the more salient aspects of the eating and
drinking habits of the peoples who resided at the three', late Iron Age settlements in this study.
9.2.1. Late Iron Age Barton Court Farm
At Barton Court Farm, there were few examplesof pottery imported from outside the local
from
beakers,
the
of
which
were
southeastEngland. Most of the pottery
exception
area with
fabrics.
This
lack
imported
together
the
of
coarse
ware
with
out
of
made
was
ware would mean
that the settlementwould be situatedquite low on a site hierarchy - if the use of imported
by
the
vehicle
which the inhabitants expressed their status. For some
containers was
be
irrelevant
(Brown 1997:100) and this could
the
of
a
of
origin
vessel
may
consumers point
have beenthe caseat Barton Court Farm. It is possiblethat the surprisingly high proportion of
bowl-like vessels and other types of vesselsassociatedwith serving, particularly beakers,
indicatesthat the giving of feastswas one meansby which the inhabitantsgained status(and
createdobligations) either within the settlementor in the community. Particular types of food
These
include unusually high quantities of bread wheat,
the
settlement.
with
are associated
for
its
baking
its
flavour but also as being labour intensive
good
qualities
and
acknowledged
(Allen et al. 1993:176; Garnsey1999:120-122),and possibly dairy products as seenthrough
the kill-rates of the animal population and the high proportion of cheese-makingvessels.Both
the cheese-makingvesselsand the grains were recoveredfrom ritualized contexts.I have also
suggestedthat the presenceof young animals and vesselswith perforatedbasescould instead
be an indicator of heightenedritual activity and the production of alcohol (rather than cheese).
1Pleasenotethat RoughgroundFarmwill not be discussedin this
sectionbecauseit did not havelate Iron Age
settlement.
246
In either case,the settlementcould have had a reputationfor producing and serving particular
drink
food
(see
Hastorf
1998:
777
to
drink
foods
the
as
to
and
gifts
re:
community
types of
and
foods).
feasts
Hayden
1996:
137
than
and
special
re:
non-localgoods;
the community rather
In addition to questioning the parameters for status, I am also interested in how the traditions
day-to-day
life
in
inhabitants
For
the
the
within
settlement.
expressed
were
and mores of
bones
both
long
late
it
the
the
that
of
cattle
and
sheep
at
many
of
also
observed
example, was
Iron Age settlement were chopped into sizes appropriate for cooking pots. I have suggested
found
bowls
high
type
and
cooking
pots
on the site could indicate the
the
of
that
number
however,
There
bones
occasions,
were
of
stews.
when
serving
were
and
preparation
bone;
in
ligaments
the
that
the
stripped
off
was
meat
was
meat
and
prepared
at
separated
distribution
In
different
in
the
types
studying
of
vessels?
of the animal
this way served
have
I
identified
different
the
types
the
settlement,
around
of
pottery
remains and
The
the
remains of particular meals at
settlement.
specialized areas of consumption
hearthstones, decorated bowls, beakers and unbutchered articulated animal remains - for
in
house
in
These
found
one
particular
and
around
site.
pits
and other
example, were
from
be
directly
deposits
the
this
of
settlement
area
may
or
may
not
recovered
specialized
linked to the status of the group in the wider community but certainly suggest that the residents
the
settlement.
of
areas
particular
out
singled
There are a number of possibleexplanationsfor the apparentpatterning at the settlement.For

instance,we might be observingareasin the settlementthat are defined accordingto the types
deposits
Specialized
in
food
consumption.
and
one areaof the settlementin this
preparation
of
light would suggestthat this particular area was a focal point for special events (Hayden
1996:140 re: the identification of feastinglocations on archaeologicalsites). It is also possible
in
that the peoplewho engaged the ritualized activity were given statusaccordingto their age,
head
family
(see
discussions
in
by
Hingley 1989:6-8; Parker
to
the
of
a
relationship
genderor
Pearson1996). In this light, the associationbetweenchoppedup bones and large bowls and
jars with ditches and small bowls andjars and trimmed meat with house gullies points to the
possibility that the preparationand consumption of one-pot meals was a communal practice
that involved all membersof the group.
247
Farm
Old
Shifford
Iron
Age
Late
9.2.2.
identified
Farm,
Shifford
the
vesselswas quite small, no obvious specialized
Old
of
number
At
dishes,
beakers,
few
if
in
bowls
(none
identified
cups
no
and
very
and particular
vesselswere
Green)
Sarah
All
identified
the
the
typology
were
recovered.
the
of
pottery
at
use
you
handmade
locally
though
vessels
and
predominate
made
even
wheel-turned
was
settlement
is
The
low
the
settlement
as
a
result
settlement.
viewed
as particularly
vesselswere reaching
It is quite likely that therewas variability in the significanceof the pots
169).
1996:
(Hey
status
for
handmade
Some
have
been
for
the
vessels
could,
example,
of
renowned
at the settlement.
identified
for
for
(see
below,
the
practices
consumption
with
or
past
ability
and
their cooking
73) and thus were consideredby the inhabitants (if not the outside
1994:
Barley
examples
Willis
1997:
48;
1994:
(Pluciennik
is
145).
It
that
the
status
also
possible
special
of
community)
inhabitants
less
that
the
ware
signifies
serving-type
were
concerned with
lack of obvious
1989b:311) and more concernedwith ingredients and with the
(Barrett
individuals
serving
by
be
inhabitants
In
Old
Shifford
food
the
this
the
to
group.
case,
consumed
of
preparationof
have
Farm
Court
Barton
may
enhancedtheir statuswithin the local community
Farm as with
feasts.
the
of
giving
through
in
bones
the
fragmented
each phase of settlement and the prominence of
nature of
The
bone
body
than
trimming
the
rather
meat
the
off
parts
coupled
with
up
prevalenceof
chopping
However,
the placement of specific types of
on
cooking.
to
one-pot
emphasis
jars points an
bonesand speciesaroundthe settlementand the occasionalcutting of meat appearsto suggest
in
before
treated
Old
Shifford
times
they
At
special
ways
at
were
consumed.
were
that animals
Farm with the absenceof obvious serving wares,it is possiblethat we are seeinga distinction
betweenthe ways animalswere eatenratherthan the way they were served.Barley is the most
in
(albeit
identified
small
amounts),
grain
the
which
could
emphasize
useof alcohol
commonly
identify
held
though
liquid
(perhaps
the
we
cannot
the
even
that
vessels
the
settlement
at
have
been
identified
have
ovens
also
the
used);
were
at
vessels
which
could
settlement
wooden
inhabitants
The
in
the
process.
malting
beenused
of the settlementmay have consumedand/or
based
the
the
economy
settlement
was
their
wealth
probably
on
rearing of animals shared
from
the
both
the
group
and
within
outside
community.
others
with
have
identified
in and around
been
and
plant
animal
types
and
remains
pots
of
Different
also
late
Iron
Age
both
in
between
Distinctions
house
of
house
settlement.
the
phases
sites
sites
the
248
food
consumption
hand
directly
the
to
on
preparation
and
to
cooking
one
and
on
relate
appear
for
Barton
Court
Farm,
differences
These
as
was
suggested
may,
reflect on the status
the other.
food
if
its
house
the
there
other
value
was
a
placed
on
over
site
consumption
over
of one
Old
Farm
Shifford
At
ingredients
the
with
apparent
versa.
vice
emphasis
on
or
preparation
different
distribution
for
just
be
the
consumption
patterns
their
could
preparation
as
well
an
and
labour
different
the
roles of the residentsandpossibly their genderand/or age(see
indication of
Barrett 1989b:312).
Pike
Claydon
Iron
Age
Late
9.2.3.
late
by
Iron Age case studies, the settlementwas
Pike,
the
to
Claydon
other
comparison
At
imported
including
inundated
Roman-style
and
non-local
goods,
with
wares, namely
almost
but
also a wide variety of non-local coarseware pots.
mortaria, amphoraeand glassvessels,
Numerousitems of personaladornmentand dressfittings were also identified in late Iron Age
for
imported
the
An
conspicuous
consumption
of
argument
goodsas a communicator
contexts.
Claydon
Pike
it
does
for
more
comfortably
than
sits
status
at
the
any of the
settlement's
of
in
in
does
Age
itself
Iron
this
late
study,
although
such
settlements
an
observation and of
other
imported
the
the
the
of
significance
use
of
vessels.
not consider possible
The Roman-stylevesselsthat were reachingthe settlementdo not appearto centreon serving
food
flavouring
food.
but
Food
the
preparation
and
emphasize
rather
of
preparation
se
per
important to the inhabitants as mortaria were more commonly
been
have
to
quite
appears
dishes,
beakers
the
than
of
identified
numbers
and cupscombined. Similarly, equalnumbersof
identified.
Food prepared with elaboration has been
bowls and olive oil amphorae were
(Garnsey
the
127)
is
1999:
it
members
wealthy
of
more
society
with
and
possible
associated
identified
the
settlement
were
of
partly by the way they prepared and
that the residents
1997:
209;
Hastorf 1998:780; Dietler 1996:98). Serving-type
food
(Bakels
al.
flavouredtheir
et
the
bowls
evident
at
quite
settlement,
nonetheless
particularly
was
non-local
ware
and
food
drink
that
the
way
and
were served was of importance but was
tankards, suggesting
ingredients.
to
the
tastes
Similar
the
late
and
preparation
Iron
to
the
of
subservient
possibly
barley
Farm,
Old
Shifford
was the most commonly identified grain and the
Age settlementat
beakers,
tankards,
even
cups
and
glass vesselssuggeststhat partaking of alcohol
of
presence
Butchery
(or
to
the
lack
the
residents.
marks
the
on
of)
significant
various species
quite
was
249
bone
the
or stripped off the bone. The deposition
that
cooked
on
indicate
meatwas commonly
ingredients
(see
body
Rippengal
1995)
the
emphasis
on
also
parts,
and
and
of particular species
in
type
types
of
vessels
serving
select areasaround the settlement
the recovery of particular
in the preparationand consumption of specialized
inhabitants
the
that
were
engaged
suggest
62).
1995:
(Hill
meals
in
late
Age
Iron
this study, differing areas of consumption were
two
settlements
As at the other
have
Pike.
Two
been suggested, one in the northern part
Claydon
occupation
areas
identified at
in
Distinctions
between
the
the
south.
the
other
and
two
the
settlement,
areas were striking
of
in
to
lived
they
the
the
variations
may
relate
rank
of
people
who
at the
and, although
that
that
there
of
anomalies
a
number
suggest
the
are
arrangement
of
settlement was
settlement,
has
been
It
dense
there
that
established
was
an
especially
accumulation of
more complex.
in
Large
deposits
the
the
zone
of
southern
settlement.
remains
and
of pottery and
artefacts
in
bones,
cattle
particular
butchered animal remains,
pig and bird bones, items of personal
dress fittings, coins, oven fragments, mortaria, glass vessels, rivets for repairing
adornment,
from
features in and around two enclosures
loom
so
were
and
on
recovered
weights
pots,
houses.
By
have
the
to
contrast
northern area of the settlement contains ring
contained
thought
domestic
devoid
features
for
that
of
material
almost
were
probably
containing
and
stacks
in
working
area
metal
and
a
number
of
enclosures
a
possible
similar
shape to the
animals,
in
first
At
it was thought that the two areas represent
house
the
area.
southern
sites
suggested
however
the
the density of pottery and animal remains
at
settlement;
activity
different zones of
that more than one household may have lived at the
from
the
suggests
area
recovered
households with dramatically different consumption habits. The northern area
settlement items,
but a comparable proportion of vessels
few
glass
vessels
or
mortaria
personal
contained
higher
distinction
than
and
numbers
of
A
amphorae
serving
the
with
southern
area.
associated
between the fabrics of the pottery was also established, particularly in terms of their colour in
the
the
vessels
southern area was particularly vivid. The plant samples
of
the colour palette
higher
had
densities
from
the
northern area
of cereals and they were more processed
recovered
from
Cattle
the
bones
southern
area.
recovered
the
cereals
than
and sheep
were equally
distinction
(the
between
in
the
the
in
area
two
northern
the
was
species
more
overt
prominent
butchered
less
but
bird
bones
bones.
there
of
evidence
was
animal
and
area)
southern
Consumption in the southern area appears more conspicuous both in terms of personal
in
Hill
1998)
Jundi
(see
the
food
drink
and
in
and
preparation
of
and
serving
and
appearance
250
been
have
The
more
public
may
or
ceremonial,
an
area
of
area
southern
ways.
particular
house
for
Court
Farm.
Barton
feasts
the
one
of
sites
at
as was suggested
specializedmealsand
differing
indication
the
that
of
roles of the residentswith
we are seeingsome
It is also possible
(managing?
food
)
linked
to
and
processing
of
and possibly even
preparing
the northerngroup
the economy of the settlement.
9.3.
Late Iron Age Food and Drink Consumption - Final Comments
lacklustre
in
before
Iron
Age
the
the
were
rather
peoples
culinary
arts
By many accounts
food
depiction
(see
This
to
the
types
Romans.
that
they
extends
of
comments
the
ate
of
arrival
food
1995b)
to
that
1995a;
the
they
to
vessels
and
used
and
prepare and consume
by Reynolds
but
it
has
this
this
Not
challenged
stereotype,
research
also suggests that at some
drink.
only
inhabitants
ingredients,
have
the
of
may
the
emphasized
status and reputation
settlements
Late
Iron
Age
dramatically
the
repertoire
of
vessels
at
a
settlement.
affect
pottery
would
which
Roman
in
having
but
the
the
specialization
of
period
overtures
making
is characterized as not
In this regard, we have seen that the three late Iron Age
1979).
(Millett
direction
that
diverse
in
their use of containers, `specialized' or
in
quite
this
actually
were
study
settlements
lack
ideas
is
The
thereof
or
apparent
to
of
pottery
own
our
relative
specialization
otherwise.
food
drink.
In
for
have
this
up
and
serving
study,
and
seen
example,
we
preparing
about
food
for
in the way
in
and
preparing
the
cooking
of
consumption
as
methods
well
as
variability
both
been
have
drink
publicly and privately.
served,
might
food and
the
the
associations
of
contextual
all
of
of
artefactsand remains associated
The consideration
However,
is
laborious.
drinking
have
through
this
type
of methodology we
with eating and
different
aspectsof consumptionthat reflect on all facets of daily living
to
beenable realize
inhabited.
The
food
diverse
the
were
settlements
realization
the
of
acts of
way
and on
late
Iron
Age
a
single
settlement
three
the
within
at
each
sites
consumption
of
and
preparation
has
the
that
contextual reintegration of the artefacts and remains of
shown
in this study
is
the
effort.
worth
consumption
251
9.4.
Food and Drink Consumption at the Early Roman Period Settlements
is
in
the
eyesof others the overriding preoccupation
the pursuit of reputation
...
by
is
be
life,
human
the
to
which
reputation
means
although
achieved are
of
in
1979:
3
Millar
(Harre
1985:
184-5).
quoted
extraordinarilyvarious
ideas
into
discussion
incorporate
the
important
the
to
is
various
on
status
offered
above
on
It
Roman
in
Roman-style
the
this study.
at
goods
early
period
settlements
the significanceof
definition,
by
often perceived as low status because they lack
Native settlementsare,
49;
1993:
1990:
Allen
18
1).
Raven
At the root of many ideasaboutthe
(e.
et
al.
Romanization g.
buildings
is
Roman
if
the
that
the
style
goods
and
of
notion
could
natives
absence
and
presence
Roman-ness (Webster 2001:7) and thus the degree of
they
to
exude
would
afford
directly
144).
The
(Hingley
1989;
1999:
the
the
on
status
of
settlement
Romanizationreflects
Roman
in
the
to
this
of
early
at
each
period
settlements
study
serves
cattle
of
prominence
lifestyle
based
how
a
particular
assume
illustrate
on the presenceof what are
we cannot
Notwithstanding
Romanization.
in
Upper
Valley
Thames
that
the
the
of
symbols
as
considered
has
been
long-standingtraditions
to
generally
species
related
more
particular
of
prominence
Valley
(Wilson
1978;
Robinson 1992; Lambrick 1992),
different
the
of
the
elevations
linked to
bones
by
the
the
that the same
contextual
associations
have
of
considering
animal
I
shown
differently
in
is
The
this
true
the
treated
at
each
settlement
study.
same
of
was
species
including
drink
Roman
food
has
Garnsey
As
style
vessels.
and
stressed,wealth
containersof
it
(1999:
114)
is
inextricable
have
to
and
possible
statusor reputationwithin
and statusare not
In
the
to
wealth.
shifting
the
or
without
with
emphasis
contexts
of
social
on
community
one's
the
than
types
on
presence
or
absence
of
rather
particular
of vessels or
consumption,
from
the
artefacts
and
remains
consider
ingredientswe can
within an imperial context. The
in
features,
the
settlement
and
remains
artefacts
depositionof
and in particular at settlement
becomes
paramountto the understandingof consumption and the effects
thereby
boundaries,
individual
households.
imperialism
on
Roman
of
Court
Farm
Barton
Roman
Early
period
9.4.1.
òwners'
lineage
late
both
Iron Age and
that
the
the
or
same
been
occupied
theorized
It has
Court
Barton
Farm
built
in
the
two
at
the
Roman
settlements
were
settlements
period
early
252
been
have
less
location
to
(Miles
1986b:
49;
appears
more
occupation
or
and
continuous
same
has
Miles
136).
David
Romanizing
1995:
Ferrell
noted
a
slight
presenceat the early
see also
in
few
increase
Roman-stylevessels;small numbersof
the
cattle, a
Roman period settlementRoman coins and a rectangularstructurewith plasteredwalls. He suggests,however, that these
in
in
Upper
Thames
Valley
the
the
of
context
other
settlements
unremarkable
and that
are
little
the
local
the
of
settlement
was
organization
social
changed;
native
style
pottery
otherwise
family
identified
late
Iron
Age
lived
the
the
at
extended
and
settlement
under
now
was prolific
below).
50
On
hand,
1986b:
I
(Miles
Miles
the
see
that
although
one
would
agree
with
one roof
dealing
held
beliefs
that
the
with
people
their
many
possibly
of
quite
of
ancestorsseen
are
we
deposits
in
importance
the
the
of
ritualized
perpetuation
pits
and
on
with
continued
of
perhaps
drinking rites at the settlement.On the other hand, I have suggestedthat the re-structureof the
Barton
Court
Farm
fundamental
in
Roman
the
the
at
marks
settlement
a
period
shift
way
early
inhabitantslived and communicatedwith eachother and with thoseoutsidethe group.
As we have seen,the living area,which was once multi-featured and expansive,was in the
large
Specialized
deposits
Roman
within
one
consolidated
structure.
that appear
period
early
fully integratedinto the living spacesof the late Iron Age settlement seem divided between
boundaries
home
those
the
Roman
inside
the
at
the
and
of
settlement
at
early
settlement.
those
inhabitants
in
the
the
of
movement
The apparentchange
around the settlementreverberatesin
habits.
The
bases
to
their
consumption
reduction
of
vessels
perforated
with
and the
the changes
in
indicates
the
the
from
dairying
of
structure
animal
population
to the
change
a
shift
apparent
in
the
both.
Meat-yielding
horse
bones
of
a
rituals
shift
consumption,
or
meat
or
of
production
butchery
The
butchered.
less
body
there
cattle
of
also
changed;
was
chopping
up
of
now
were
in
towards
from
bone.
Ritualized
shift
consuming
meat
the
apparent
steaks
an
and
or
parts
imply
butchered
that the consumptionof the species,in this caseof cattle
animals
depositsof
in
horse,
activity
the
part
ritual
where
a
at
played
previous settlementnon-butchered
and
The
figured
in
bowls
burials
identified
the
reduction
prominently.
proportion
of
at the
animal
indicates
from
further
there
that
movement
a
away
one-pot cooking, although the
settlement
increasein large cooking-typepots could signify that consumption of one-pot meals was still
importantbut was now more of a group activity. The reduction in the presenceof bowls could
in
decline
Although
indicate
public
change
consumption
or
events.
a
no plant remains
also
it
is
from
the
settlement
quite
that
possible
cereal production or the
recovered
were
bread
(possibly
important
inhabitants;
wheat)
to
the
was
grains
still
of
the
accumulation
the
the
ground
in
above
was
raised
and
possible
cereals
were
granaries
of
positioned
storage
253
for
As
into
late
defined
Iron
the
Age
the
they
settlement.
that
access
settlement,
such a way
in
Roman
ingredients
but
the
the social contextsof
early
period,
on
there remainsan emphasis
have
changed.
consumption
in
ingredients
in
treatment
the
the
that
changes
of
is
of
some
reflect a change the
It
possible
in
(see
Miles
1986b:
46).
imposition
Certainly, we must be
taxation
comments
the
of
form of or
identified at the settlement occurred within an imperial context. The
the
that
changes
conscious
boundary
defined
the
a
within
quite
substantial
settlement
with
a
clearly
and
of
consolidation
indicates
identified
the
a
concern
with
the
approach
of
outsiders
entrance
not
at
gated
possibly
boundary
has
"...
first
Ferrell
foremost
As
the
the
stated:
and
expresses
unity
settlement.
earlier
134).
The
distinctive
(1995:
inside"
deposits
is
the
the
positioning
of
at
either
on
what
of
from
the
to
the
the
the
entrance
away
settlement
or
within
structure
settlement
of
periphery
imperial
it
context
and
appears to suggest that consumption was
itself also occurred within an
Finally, the movement to a single dwelling large enough to accommodate the
judicious.
more
doors
locked,
imperial
that
the
were
occurred
possibility
within
context
an
whole group and
between the residents or the rules that governed the segregation
distinctions
that
and suggests
The
become
to
the
Roman
had
the
security
of
whole
secondary
group.
early
period
of tasks
has more recently been referred to as a relatively high status settlement (Henig and
settlement
because
it
I
106)
the
2000:
seems
of
principal
primarily
rectangular
structure.
would
Booth
inhabitants
`wealth'
the
that was once expressed through the giving of feasts
of
the
that
suggest
house
the
(Rippengal
1993)
towards
of
a
sturdy
construction
and
secure
and a
was redirected
boundary around the settlement.
well-defined
Farm
Roughground
Roman
Early
period
9.4.2.
Roughground
Farm
in
be
Roman
at
settlement
can
period
The early
viewed a number of ways.
In one sense,the settlementappearsto epitomize a relatively low status (non-Romanized)
The
house
initially
that
probably
contained
enclosure
a
was
circular
and
settlement.
native
but
(the
became
certainly
was
not
ovoid
rectangular
more
chronology of re-cuttingof
gradually
1993:
(Allen
179)
is
et
al.
so we should not necessarilyview the
the enclosure not secure
in
heading
direction,
from
i.
(sub-)
the
a
as
particular
to
enclosure
e.
rounded
of
changingshape
Romanizedcontainerswere identified and those that were recoveredtend to
Few
rectangular).
drink
food
flavouring
it.
than
`Fine
and
rather
preparing
serving
and
wares'
be associatedwith
254
four
the
its
assemblage,
percent
of
pottery
all
to
of
which
only
contributes
constitute
designation as native and of little wealth (Allen et al. 1993:181; Booth in press).On the other
including
local
imitation
have
types
that
I
tableware,could
hand,
other
of
containers,
suggested
is
it
Roughground
Farm, ingredientswere
been
that
very
apparent
although
at
have
prestigious
The
deposits
in
treatment.
ritualized
of
querns
and
related
special
paraphernalia
pits
afforded
importance
but
imply
to
the
the
they
of
agriculture
economy
that the
might also
could signify
67)
Hill
1995:
into
food
(after
large
cereals
of
was
significant;
a
quantity of
transformation
deposited
in
bones
The
ingredients,
these
was
pits.
animal
emphasis
on
as was
meat-bearing
late Iron Age Old Shifford Farm, suggeststhat celebrations might have been
for
suggested
histograms
if
large
we
remember
of
serving-type
vessels
were
comparably
more egalitarianindicator
(see
Chapter
6
I
the
of
communal
as
an
serving
at
considered
settlement
which
`villa' in the early second century within the same
The
).
the
6.3.2.
of
construction
section
in
boundaries
of
a
and
respect
number
that
area
of
existing
settlement
could
signify
general
towards
(The
their
the
inhabitants
wealth
construction
redirected
a
of
villa.
rapid
the
its
Romanized
building
by
Allen
the
and
structure
were
considered
of
as evidence
construction
had
have
he
if
though
new
owners
that
there were two groupsat
may
adds
the
settlement
that
Romanized
than
have
the
been
that
other,
more
ownership
one
may
continuous
the settlement,
)
196).
1993:
(Allen et al.
in
distinctions
distribution
the
of
number
of the artefacts and remains at the
There were a
determination
that
the
based on consumption habits is
of
to
settlement
suggest
status
settlement
house
deposits
Around
the
that
site
(many
are
to
appear
emphasize
cooking
complex.
more
burnt
from
this
bowls
fragments
(local
were
stones
recovered
and
area)
and
serving
and
oven
containers)
items
be
that
Belgic
well
as
as
can
associated with individuals such as
imitation
Unlike
hunting
Roman
paraphernalia.
at
early
brooches and
period Barton Court Farm2, the
does not appear to have been contained within a single boundary and a variety of
settlement
been
identified
have
beyond
the
practices
enclosures that surrounded the main
consumption
West
the
the
deposits
occupation
of
main
area
are
above-mentioned
area.
ritualized
occupation
burial.
infant
In
in
beneath
stock
enclosures
an
the
as
and
well
as
pits
second century
querns
of
Roman-style
have
been
dump
(a
in
of
pottery
amounts
recovered
of
samian
cups
sizeable
villa,
bones
has
in
It
been
as
as
ware
well
samian
animal
and
cranial
pits).
enclosure
argued
a stock
household
Romanized
have
lived
may
at the settlement (Allen et al.
that a second more
habitation
does
to
2Thisobservation
and
specifically
not meanthat surroundingfields for examplewere
refers
nota partof thesettlement.
255
isolated
deposits
Roman-style
have
I
the
that
However,
of
181).
vessels as well
suggested
1993:
burial
deposits
(the
deposits
the
and
querns
elaborate
of
native
see
style
ritualized
as other
The
linked
that
to
be
the
to
the
are
about
changes
occur
at
settlement.
structural
below) could
deposits found around the settlement might also reflect different types of deposits
specialized
food
preparation and consumption associated perhaps
passage
for different seasonsor rites of
death
birth
drinking
In
the
harvest,
or
of
case,
with
people
or
even
animals.
either
and
the
with
limited
life
based
to
tends
the
criteria
at
a
on
gloss
over
of
determination
variety
status
of
the
have
brought
from
to
that
the
may
people
community
the
wider
events
various
and
settlement
the settlement.
from
Roughground
be
Farm
Roman
settlement
at
period
viewed
must
Certainly, the early
it
However,
that
imperial
appears
at
socio-political
circumstances
context.
an
within
different
has
been
Farm.
for
Court
Barton
to
Farm
what
suggested
rather
were
Roughground
been
Roman
have
does
the
to
period
settlement
early
appear
not
As was mentionedabove,
Features
boundary.
the
of
are
spread
around
a
settlement
and
number
single
a
within
contained
least
bring
to
celebrate
or
at
attention to a particular type of
appear
featuresat the settlement
deposits,
the
identity
proliferation
of
specialized
structures,
communal
pit
rounded
native
burial
burials
been
has
linked
in
the
to
an
elaborate
style
of
which
particular
and
consumption
have
been
features
the
to
the
Several
these
related
chronologically
of
reorganization
Gaul.
of
in
Thus
`villa'.
is,
least
there
the
towards
the
of
a
the
at
end
of
construction
and
settlement
both
identities
Roman-like
the
a
strong
presence
of
the
native
at
and
settlement,
of
occupation
in
fit
The
Roman.
do
traditional
notions
of
native
with
or
of
that
persistence
not
settlement
be
lifestyle
the
to
the
as
evidence
of
resistance
viewed
with
a
associated
can
symbols
native
the
within
context of someof the changesthat are about to
considered
when
Romans,although
it
is
Roman
that
the
co-presence
possible
of
culture
native
and
material
the
settlement
occur at
for
instance,
has
Webster,
Jane
"Romanized
that
recently
argued
material
is more ambiguous.
in
identities
be
ways,
ambiguous
simultaneously
and
used
new
creating
could
culture
belief
(Webster 2001:9). I concludedthe
key
and
pre-Roman
of
practice"
aspects
maintaining
Farm
Roughground
the
transformation
the
with
a
query
about
whether
of
case study of
done
trepidation or in celebration. It could
in
century
was
with
the
second
early
settlement
be
identity
these;
that
seeing
might
a
negotiated
we
social
been
was necessary
of
have
neither
in
Roman
lifestyle
Britain.
live
what
was
now
a
controlled
a particular
in orderto
256
Farm
Old
Shifford
Roman
Early
period
9.4.3.
Shifford
Farm
Old
Roman
was constructed on the same
The early
period settlement at
late
Iron
features
boundaries
Age
Iron
Age
late
and
the
settlements
and
were
as
alignment
The
into
the
the
settlement.
of
re-use and expansion of existing
re-structuring
incorporated
in
between
inhabitants
descendants
indicates
the
relationship
and
continuity
not
boundaries
below)
(see
Claydon
Pike
in
Farm
Court
Barton
though
or
evident
possibly
either
at
apparent
Old
Shifford
Farm,
is
Farm.
At
little
direct
Roughground
form
there
evidence
of
different
at
a
by
This
indicate
taken
the
the
the
to
that
was
settlement.
excavators
at
Roman
presence
a
(as for the late Iron Age settlements) was of particularly low status. I suggested
settlement
imply
less
lack
that
there
serving
ware
could
the
obvious
was
on
of
emphasis
that
above
ingredients
feasts
late
Iron
Age
the
the
and
giving
on
of
at
emphasis
more
individuals and
The same could be true at the early Roman period settlement, where a few
settlements.
Green's
Sarah
identified.
Furthermore,
typology)
to
(only
were
no
according
bowls
one
from
As
the
were
recovered
no
amphorae
settlement.
was
and
mortaria
no
ware,
samian
is
7.4.1.
)
7
because
(Chapter
this
in
section
quite
at
other
unusual
the
study
case
stated
in
Upper
Thames
few
Valley
least
the
these
`low
settlements
at
of
a
status'
similarly
fact,
In
identified
been
have
the
the
only
non-local
recovered.
pottery
at
pots
Roman-style
jars
have
in
handmade
been
Their
Malverian
that
to
appear
used
cooking.
settlement was
handmade
the
to
that
however,
possibility
credence
cooking
containers
may
adds
presence,
inhabitants,
for
ingredients
the
important
the
to
either
been
cooking
of particular
or
have
in
held
themselves
were
particular esteem.
because the vessels
Roman
in
For
the
time.
that
period
settlement
was
early
is
somehowsuspended
This not to say
items
the
two
adornment
were
recovered
personal
at
only
first
of
settlement,
albeit
time
the
horse
bones
identified
high
Roman
the
The
of
proportion
at
early
period
brooches.
relatively
horses,
have
in
in
been
held
high
indicate
that
to
the
appear
which
esteem
could
settlement
In
Valley
the
were
this regard, the
Thames
elsewhere,
reared
and
at
settlement.
Upper
for
horse
have
the
rearing
a
reputation
with
community,
maybe
gained
outside
may
settlement
Thornhill
(Interestingly,
Farm,
have
Roman
to
army.
appears
the
raised
which
with
even
Roman-style
but
brooches
few
pottery,
had
of
sizeable
numbers of
and
examples
horses,
is
have
been
)
There
below.
the
that
also
possibility
items
cereals
may
produced
at
see
toiletry
increase
in
densities
in
the
the
there
slight
a
was
plant samples.
cereals
of
as
the settlement,
Furthermore,
the two
the
quite
was
possibly
that
self-sufficient.
settlement
This suggests
257
distinctions
in
house
the
the
show
settlement
at
sites
acts of specializedpreparation
suggested
late
than
Iron
Age
consumption
preparation
or
as
was
the
suggested
rather
at
consumption,
and
deposits
The
Malverian
the
contained
enclosure
square-shaped
only
of
cooking
settlements.
identified
beaker
deposits
the
human
at
settlement,
the
cereals
with
chaff
only
no
and
of
pots,
bones in the outside boundary of the enclosure.The second suggestedhouse site featured
butchered
lots
horse
bones
burnt
a
quern
chaff,
stone
and
of
meat-bearing
with
and
cereals
horse bones - the first evidence that horses were possibly consumed at the settlement.A
brooch was recoveredfrom each house site and the status of the inhabitants appearsalmost
foods
by
important
the
that
and
consumption
preparation
of
specific
to the
were
equalized
individual at the settlement,ingredientsandthe social
the
Although
there
are
vestiges
of
group.
fore.
the
their
consumptionremain at
contextsof
Evidence of some form of direct or indirect imperial presenceat Old Shifford Farm lies
in
boundary
This
the
erection
of
substantial
a
with
a
controlled
possibly
entrance.
quite
directed
have
is
have
to
thought
the
passage
of
which
animals,
would
also
served
entrance,
The
is
different,
the
of
outsiders.
the
approach
organization
of
settlement
to control
however, from that of Barton Court Farm, which similarly erected a substantial boundary,
house
the
sites
are
quite
public
two
and
straddle
suggested
the
entranceto the settlement.
as
houses
hand,
the
the
placement
of
and erection of a substantial boundary could
On the one
have servedto protect what was a valuable commodity at the settlement, namely horses.On
hand,
we must also acknowledgethat the reaffirmation of the boundariesof their
the other
descendants,together with the apparent absence of Roman cultural influence, occurred
Some
beliefs
imperial
the
of
context.
earlier
and traditions on preparing and
within an
been
have
for
food,
because
familiar
feature
in
could
example,
retained
they
were
a
consuming
Harbottle
(see
1997:
176
familiar
world
increasingly
the
unfamiliar
on
tastes).
persistence
of
an
lack
Romanized
the
that
Timby
of
Jane
suggested
pottery at Old Shifford Farm could
lack of desire for items clearly related to Roman eating and
"a
there
that
was
signify
drinking habits" (1996:129). If this was indeed the caseand the social contexts of eating and
drinking, which accentuateingredients and the giving of feasts, appearsto suggestthis, the
bold
a
statementto all those who enteredthe settlement.This could
inhabitantswere making
(and
have
their
local
reputation
status)
enhanced
the
with
members
some
well
of
very
community.
258
Claydon
Pike
Roman
Early
period
9.4.4.
The proliferation
Claydon
Roman-style
at
goods
of
Pike
is in
contrast to many
in
Upper
Thames
Valley
the
and as a consequence outside
contemporary native settlements
including the presence of the Roman army have been invoked to explain the
explanations
late
Iron
Age
At
been
have
there
the
to
settlement
these
appears
a
goods.
similar
of
presence
items
including
Roman
habits
for
associated
with
and with
eating
goods,
non-local
thirst
imported
It
that
there
consumption
suggested
was
using
goods
was
appearance.
personal
by
inhabitants
the
their
which
means
expressed
one
and
was
maintained
and
conspicuous
in
ingredients
through
the
preparation
and
consumption
was
way
of
status; another
particular ways.
imported
Roman
the
to the settlement
proportion
of
containers
settlement,
At the early
Age
have
late
Iron
been
there
to
As
the
settlement
appears
an emphasison the
increased. with
foods
in
fragments
the
Roman
of
mortaria
preparation
common
vessels
are
quite
style
of
use
fish
pasteamphorae;once again amphoraeand
(and show signs of repair) as are olive oil and
is
This
in
direct
be
Roman-style
to
the
types
to
represented.
contrast
equally
of
bowls appear
Farm,
There
however,
identified
which
emphasized
serving.
an
was,
at-Roughground
vessels
Roman
the
in
wares
of
serving-type
at
the
early
period settlement,
proportion
increase
drinking:
beakers,
flagons,
bottles
tankards,
with
associated
glass
and
vessels
of
particularly
if
Gaulish
jars,
the
and
unstratified
south
the
samian
cups
used
were
at
narrow-necked
cups,
drinking
involving individualized drinking.
is
the
a
there
of
new
possibility
rite
then
settlement
identified
indicate
longer
the
commonly
most
grain,
which
could
an alternate
Barley was no
have
been
(Green
139-40).
1981:
types
other
of
grains
could
although
used
sourceof alcohol,
have
been
indigenous,
both
to
primarily
appear
type
were
and
non-local and
Serving
vessels
forms
indigenous
have
been
deliberate,
it
is
the
may
of
use
quite
although
locally produced;
because
these
types
the
of
of
significance
vessels
to
difficult access
of the presenceof sizeable
Gaulish
There
been
have
does
samian
to
south
ware.
appear
an
amounts of un-stratified
flagons,
beakers,
brown
type
tankards
typically
colour:
and
and
were
vessel
or
associationwith
dishes
black
factor
If
in
bowls
typically
were
or
and
grey/black.
colour
a
was
red/orangeand
`red'
have
in
feasts
for
been
then
samian
ware
may
events
and
other
used
consumption
public
in
disposed
it
but
(almost
different
of
a
to
was
completely
way
casually)
occasions,
ceremonial
serving.
with
associated
vessels
types
of
other
259
differently
Roman
the
treated
early
at
period settlement. Horse-bonesare less
Animals were
late
Iron
Age
do
butchery
the
they
than
settlement
and
at
not
exhibit
marks;
represented
however, their contextual associationssuggestthat they were at times a ritualized deposit.
Dogs and cats were both butchered (the butchery of cat is quite unusual in the Roman
in the early Roman period. Many of the deposits of thesespecies
64))
1978:
(Maltby
period
deposited
boundary
the
types
the
vessels
were
of
at
of
settlement,particularly
particular
and
the
to
the
settlement.
entrance
around
do
Pike
Claydon
boundaries
not convey the same senseof settlement consolidation
at
The
identified
Barton
Court
Farm.
Edmonds
has
that
(in
at
was
commented
security
and
ditched enclosures serve as both boundaries and
Neolithic
that
to
enclosures)
reference
by
Ardener
111;
1993:
(1993:
13;
Hingley
1990b), and this
see
comments
thresholds
been
have
Claydon
Pike.
Entrance
the
to
to
case
at
the
appears
settlement was
certainly
directly off a Roman road and the principal residential structure was situated alongsidethe
depositsassociatedwith specialized consumption were depositedin
Many
the
of
entrance.
flank
the
the
ditches
that
and
road
entranceto the settlement and these may reflect what
the
ìmportant
to
episodes of consumption' (1993: 112; see also Hingley
as
Edmonds refers
drinking,
The
the
including glass vessels,
100).
associated
of
vessels
majority
with
1990b:
interestingly
in
identified
the
deposited
this
of
settlement
part
no
glass
vessels
were
were
Items
the
structure.
of personaladornment were also found in this
in the pits associatedwith
deposits
dog,
bird,
horse
the
bones,
burnt
bones.
of
cat,
of
most
pig
and
were
and
as
area,
in
to
the
the public domain; it was an area where the
settlement
was
the
Clearly,
entrance
(if
display
conspicuous
acts
of)
specific
not
consumption
of
and
personal
were
remains
inhabitants
the
that
This
the
had
relationships
suggests
the
of
settlement
with the
situated.
including
those
intimate
perhaps
with
official
connections
community
and
outside
- were
through
food
personal
one's
appearance
through
and
marked
particular
acts
of
were
drinking.
especially
and
consumption
in
has
that
Webster
our preoccupation with establishing `Roman' or
argued
recently
Jane
ignored
have
the possibility of, what she terms, the `creolization' of
life
`native'
ways we
"...
She
explains:
creole
material culture represents not the gradual
culture.
material
by
but
life
blending
in
both,
the
another,
of
way
one
of
a clearly nonegalitarian
replacementof
is
highly
"The
in
that
it
result
a
ambiguous
and
the
that
material
culture,
sense
context",
social
in
different
different
meanings
contexts" (2001:10). Roman-style material
is imbued with
260
flavouring
highlights
food
that
of
preparing
and
ways
particular
and serving alcohol
culture
Claydon
Pike.
These
into
incorporated
the
at
event
consumption
acts of consumptionalso
was
included drinking from beakers,the use of indigenous serving vessels and the ritualized
killing
(sheep,
dogs,
horses).
The
animals
of
particular
cats,
and
and/or
contexts
consumption
èpisodes
identities
these
tied
to
of
consumption'
appear
that were
of
social
of some
The emphasisof the consumption events at the early
in
the
settlement.
earlier
established
however,
in
the public domain and appearsto be
was,
very
much
settlement
Roman period
inhabitants
the
the
of the settlementperceivedthemselvesand were perceivedby
linked to
way
98).
1996:
If,
has
(Dietler
been
inhabitants
Claydon
the
as
suggested,
community
the outside
of
Pike had a public, possibly controlling relationship with the local community, they expressed
in
their
own choosing.
a
way
of
wealth
and
their power
Thornhill
Farm, Roman style vesselswere few, but a
kilometre
at
contemporary
One
away
large
items
adornment
toiletry
of
personal
group
articles
and
were recovered
comparatively
inhabitants
Here
the
obviously had different concerns with the use of
from the settlement.
linked
that
to specific individuals (Hill 1997)rather
one
was
possibly
Roman-materialculture,
large.
Jundi
Hill
(1998:
126)
have
the
at
community
outside
and
with
than
suggestedthat
is
fact
times
characteristic
appearance
of
that
the
personal
of
anxiety
and
with
stress;
concern
Thornhill
Farm
Pike,
Claydon
to
was abandonedin the early secondcentury could
in contrast
instance.
in
this
be quite significant
9,5.
Early Roman Period Settlement in the Upper Thames Valley
history
Upper
that
the
Thames
intend
the
Valley
in
to
of
do
suggest
I
the early Roman
not
based
interpretation
be
four
I
do
however,
believe,
on
my
re-written,
of
sites.
should
period
has
that
to
shown
we
need
re-think our criteria for defining status at native
this
study
that
in
the
boundaries
from
as
well
as
consider
valley,
settlement
structure
and
settlements
Certain
imperial
four
in
have
the
characteristics
context.
this
of
an
sites
study
played
within
interpretations
formulation
in
the
current
the
of
of
political and social changesthat
a part
I
like
in
the
to add my own observationson the nature of
been
valley;
would
have
observed
settlement.
4.3.3.
4
(section
in
Chapter
),
discussed
a number of commentators have noted two
As was
in
in
Upper
Thames
the
The
Valley.
first
in
settlement
patterns
changes
occurred
significant
261
Age
Iron
late
with the abandonment of many settlements and the establishment of new
the
defensive
including
the
two
settlements;
second
period of settlement change
settlements,
in
to
the
century,
where
mid-second
many settlements were similarly
early
occurred
83-4;
1992:
Fulford
(Lambrick
1992:
27-9;
Allen
1993:
196).
expanded
or
et
al.
abandoned
in
in
the four sites in this study. Barton Court
are
settlement
represented
These changes
late
in
Iron Age, reorganized in the early Roman period, and
the
Farm was established
Old
Shifford
Farm
in
in
late
the
Iron
century.
the
second
was
established
early
abandoned
boundaries
in
the early Roman period and abandoned
existing
Age, reorganized using some
Roughground
Farm
differs
century.
in the early second
slightly in that there was settlement
in
hiatus
habitation
Age
Iron
and a
until the early Roman period; in the early to
in the early
the
settlement
was
the
reorganized,
expanded
century
and
site was occupied
mid-second
The
Claydon
Pike
Roman
Iron
the
site
of
throughout
the
period.
was
occupied
throughout
in
different
was
established
settlement
a
Age although a new
part of the site in the late Iron
in
Roman
the
again
early
period; in the early second century the
Age and was reorganized
Roman
It
throughout
the
and
occupied
expanded
was
greatly
was
period.
would
settlement
in
for
in
this
third
the
study
that
sites
a
the
episode
of
settlement
change
occurred
appear
first
AD,
half
be
Roman
the
which
century
the
may
or
to
of
may
not
related
second
Henig and Booth 2000: 106). Examples of other settlements that were
(contra
conquest
first
late
in
include:
AD
Watkins Farm, Abingdon
the
century
established
or
reorganized
Vineyards, and Appleford.
in
that
were
abandoned
early the secondcentury were the ones that
Many of the settlements
late
Iron
in
Age
(Allen
2000;
Henig
Booth
2000:
This
is
the
106-7).
and
established
were
27)
(1992:
Fulford's
in
that
late
Iron
Age
suggestion
settlements
to
the
established
counter
Roman
the
in
throughout
that
period
and
settlements
the
established
earlier
occupied
were
in
discussion
(see
the
in Chapter 4 section 4.3.4.)
Age
second
century
Iron
were abandoned
fit
this model, for example Gravelly Guy and Ashville
settlements
(although certainly some
Court
Farm and Old Shifford Farm were establishedin the late
Barton
Both
Estate).
Trading
in
Age
Iron
and abandoned the early second century. It would also appear that for the
in
determined
there
this
was
no
single
study
that
characteristic
whether
settlements
Roman
(see
the
Henig
throughout
Booth
2000:
106period
continued
also
and
occupation
74). Fulford also suggestedthat the settlementsestablished in the late Iron Age are also the
Romanized
have
structural
evidence,
while
sites
more
established
earlier
often
sites with
but
no structural evidence(Fulford 1992:27-9). Barton Court Farm and
Romanizedmaterial
262
Old Shifford Farm, which were both abandonedin the early second century, were very
in
terms of their use (or non use) of Roman style
different settlements structurally and
both
is
It
in
that
settlements
the third
noteworthy
were
re-established
culture.
material
lines
in
their
Old
Shifford
the
Farm
third
as
predecessors:
similar
century
along
centuries
Roman
material culture and in the third century at Barton Court
had only slight evidence of
Farm a villa was established.Claydon Pike and Roughground Farm, which were also very
in
terms of the use of Roman style material culture,
and
different settlementsstructurally
Roman
be
Interestingly,
throughout
the
both
to
occupied
period.
settlements
continued
have
been
have
that
in
areas
to
associated
residential
the second
with
workers
appear
Miles
191;
1984:
199).
1993:
If
(Allen
al.
continual occupation were related to the
et
century
i.
based
Roman-style
e.
structures,
degree of
on emulation (Fulford 1992:27), one would
Farm
Court
Roughground
Barton
Farm to have been occupied
and
not
have expected
continuously.
Possible causesof the dislocation of settlement in the early second century are varied. I
Lambrick
has
in
that
the
as
suggested,
agree,
would
changes
think most
settlement pattern
factors
linked
increases
to
in
than
socio-political
rather
the population or
are probably
land
84).
Fulford (1992) has suggested that people
(1992:
the
internal pressures on
have
been
`drawn
land'
in responseto the growth of
the
may
people)
off
(particularly poorer
there
that
was a movement of people to work on the more `successful'
possibly
towns or
32).
have
Others
(1992:
linked
in
the
estates
changes
settlement to the
agricultural
Iron Age structures in the face of "a much more complex social,
break-up
of
continuing
infrastructure"
(Lambrick
1992:
There
105).
have
been
administrative
and
also
economic
dislocation of settlementwas tied to the appearanceof imperial estates
the
that
suggestions
208-9).
1984:
(Miles
Valley
in the
Roman
the
difficult
to
early
period settlements in this study fit in with the
is
where
see
It
because
the
different.
One
does
settlements
were
thing
so
that
explanations
not
various
factor
is
been
have
the
Roman-style
contributing
a
to
presence
absence
or
of
appear
the
The
that
different
settlements
of
economies
were
abandoned
were
also
quite
structures.
farming).
Barton
Court
Farm
mixed
(pastoralism and
and Old Shifford Farm may have
Roman
Both
the
to
presence.
aversion
settlements erected substantial
shared a strong
boundariesaround the main occupation area and appearto have particular attitudes towards
At
Claydon
Pike
Roughground
Roman-style
Farm, the
culture.
material
and
the use of
263
in
different
in
for.
(although
have
themselves
to
ways)
a
place
negotiated
inhabitants seem
had
in
is
It
these
that
settlements
what
Roman
common were
quite
possible
world.
the
in
imperial)
dislocation
(local
that
the
the
resulted
either
of
and/or
varying alliances
its
between
the
various
regional
zones
or
that
enhancement.
existed
once
cooperation
base
to
an analysis of an apparently widespread
This is not a new model on which
Valley; this is simply an observation on four settlementsand perhaps a
in
the
phenomenon
its
to
the
native
contexts
of
consumption
and
social
relationship
to
examine
challenge
Upper Thames Valley over time. The shared characteristics of Barton
in
the
settlement
hand,
Farm
Roughground
Claydon
Shifford
Farm
Old
the
Farm
on
one
and
and
Court
and
Pike on
have
influenced
inhabitants
the
whether
moved or stayed;
the other, may or may not
drinking
the
the
and
of
eating
some
of
in
contexts
practices
we
place
social
but considering
in
hands
inhabitants.
the
behind
the
continuation
of
settlement
of
relocation or
decisions
9.6.
Conclusion
consolidate
integrate
ideas
in
has
this
to
the
and
some
of
presented
served
This chapter
formulating
By
the
importance
social
contexts
a
of
studying
consumption.
of
the
on
thesis
food
I
have
the
contexts
social
of
preparation
to
consumption,
and
explore
methodology
daily
life
hints
late
Iron
Age
the
the
multiplicity
of
of
and early
to
at
provide
been able
has
been
food
It
that
suggested
preparation, consumption and
Roman period settlements.
Various
the
specific
with
areas
of
settlement.
associated
possibilities
drinking were often
division
late
Roman
Iron
Age
the
the
to
of
space
at
explain
been
and early
offered
have
include
habits;
These
the
observance
of
private
and
consumption
public
settlements.
period
differing
the
the
residing
status
within
of
same
settlement;
possibility
groups
of
the presence
the
to
the
within
settlement
according
gender
of
responsibilities
age
or
different
and
roles
of
different
based on
drinking
types
the
of
of
eating
inhabitants;
and
customs
presence
the
harvest,
to
that
the
the
pertain
might
rites
of
passage,
maintenance
the
settlement
at
events
The
importance
itself.
the
within
changes
settlement
major
or
of
relations
of power
incorporated
into
the
both
settlement
around
was
and
emphasized
and
within
boundaries
late
Iron
Age
in
the
consumption
specific
practices
of
and early
the
ideason
significance
Romanperiods.
264
discussions
in
both
these
are
questions
on
throughout
what
constituted.
status
Resonating
have
have
through
that
I
their
may
settlements
gained
some
a
reputation
suggested
periods.
in
foods
foods
The
those
the
the
sharing
of
and
with
community.
associationwith particular
different
have
in
is
It
the
varied
among
also
groups
community.
status of a settlementmay
have
been
by
low
the
simultaneously
may
that
perceived
as
status
settlement
a
possible
local
include
Roman
but
have
been
the
the
elites
and
may
military,
which
wider community
local
feasts
by
the
through
the
in
community
possibly
of
or
giving
held particular esteem
The
imported
Roman
culture.
material
negotiated
consumption of
through acts of resisting
in
by
Upper
Thames
Valley
the
their
to
people
which
chose
means
express
one
was
goods
but
has
is
this
thesis
that
the
what
shown
community
status and reputation
identity within
different
in
different
types
been
through
have
of
consumption
practices and
sought
could
social contexts.
265
Chapter 10
Concluding Remarks
in
Roman
`native'
Britain
have
been
knowledge
that
settlements
characterized
Acting on the
diverse
inhabitants
lives
has
the
the
and
new
perspectives
on
provided
this
of
study
a priori,
the
through
the
type
consideration
of
social
contexts
of
eating
settlements
of non-Roman
has
been
to explore elementsof social changethat
this
The
thesis
drinking.
of
aim
main
and
`Romanization'
its
`native
the
theories
of
or
alternative,
within
are not contained
it is difficult when considering material culture from within the
However,
continuity'.
in
be
Roman/native
have
dichotomy
I
the
imperialism,
to
caught
up
not
of
which
of
context
`Roman'
`native'
In
to
the
outside
explore
constructs
of
and
attempts
our
been critical.
in
the
use
of
material
culture,
we
must
not remove politics
ambiguity
considering
through
did
`Roman'
`native'
denying
by
that
some
people
conceive
of
a
and/or
from the equation
knowledge
beliefs
did
to
that
to
communicate
use
certain
attitudes
and
and
cultural package
in
Perhaps,
the
the
main
one
of
problems
with
study
of
change
social
the outside world.
Roman
been
but
has
the
concept
Britain
so
much
of
not
and
native,
rather, the view
Roman
lot
Britain
in
Roman
their
the
two-dimensional
saw
political
same
way
of
the
people
that
hope
I
have
I
In
depth
described.
this
that
the
to
is
regard,
added
it
some
study of
often
that
different
`native'
have
interpreted
by
that
ways
their
people
considering
might
social change
`Roman' world.
have
drinking
I
the
minutiae
through
of
of
eating
some
and
practices,
seen
Having worked
in
be
dichotomy
knew
Romanization
I
the
theory,
that
the
to
case
of
and
in practice what
describe
did
do
did
the
types
that
adequately
of
not
changes
and
not
occur
continuity,
native
in
Upper
Thames
Valley.
detailed
the
As
this
settlements
the
as
analysis
native
of
some
at
drinking
is
of
eating
the
and
remains
and
artefacts
nonetheless
rudimentary,
of
use
my
was,
data
do
from
four
in
things
the
the
with
to
this
more
is
recovered
many
sites
it
possible
and
begun
barely
have
imperialism
I
to
the
the
study
effects
In
without
of
this
regard,
study.
266
but
I
in
Chapter
2,
Romanization;
theories
as
stated
the
we
must
of
of
encumbrances
it.
beyond
Knowing
if
I
know
I
in
to
what
were to
now,
deconstructthe concept order move
liberated
to
the
the
I
take
approach
analysis
of
artefacts
and
a
more
would
anew,
start
for
instance,
for
I
drinking.
the
consider
weed
many
uses
would,
other
and
eating
of
remains
in
their
the
to
than
besides
their
suggest
place
rather
ritual
process,
edibility,
species,
fully
I
the
the
animal
remains. would consider more
composition of
focusing primarily on
this
the
the
aesthetic
and
what
containers
might
say
about
decoration
of
colour
and
the
have
in
how
influenced
inhabitants
this
turn
the
types
the
might
of
pots
and
of
perceptions
decoratedand plain samian ware would
(1997)
to
the
Willis'
study
of
approach
they used.
indigenous
for
type
of
wares,
of
study
example.
similar
a
with
well
combine
four
been
has
in
to
the
their own right, rather
this
consider
sites
study
One of the tenets of
here
have
I
just
begun
too
`native
to
and
of
only
some
settlements'
process
simply
than as
For
I
beyond
that
dynamics
example,
would
settlement.
now
go
suggesting
of
the social
factors
influential
in
detected
have
I
the
that
and
patterning
were
roles
gender
or
status
implications
fully
the
the
of
segregation
and
spatial
acts
of
of
preparing
more
consider
households
lives
inhabitants.
drink
food
the
the
of
on
and
possibly
of
and
consuming
blur
define
however,
to
the
to
has,
that
many
of
served
categories
continue
This study
in
done
This
through
the
the
that
was
part
challenging
use of
notion
studies.
Romano-British
direct
the wealth and status of those who
`Roman'
a
was
measure
of
signifiers
particular
instead
have
lives.
I
have
into
their
that
suggested
them
consumed
people may
incorporated
feasting
it
through
thereby
the
and
away
within
status
established
given
and
their wealth
for doing so. I have also shown that people might use a number of different
community
depending on the occasion, to negotiate their place in the community. We have
signifiers,
in
Roman-style
the
that
to
use
of
there
variability
that
was
appear
emphasize
goods
seen
also
drinking,
hand,
indigenous
the
on
of
the
one
ways
and
use
of
eating
particular
and
serving
but
Roman-style
flavouring
food,
the
methods
of
preparing
drinking
on
and
containers
and
drinking
in
the
the
and
eating
of
The
customs,
use
particularly
of
contexts
social
other.
been
have
to
similarly variable, used only on particular
Roman-style containers, appears
into
integrated
daily
in
instance
This
in
consumption
and
practices
another.
occasions one
know
inside
that
to
further
cannot
assume
we
was
has
on
shown
going
what
study
houses.
their
the
based
of
shape
on
settlements
267
issues
imperialism.
have
the
to
some
of
on
wider
reflect
me
of
enabled
These observations
have
been
identified
drink
food
in
that
consumption
occurred at
and
The various changes
in
Roman
the
extended
or
established
period.
that
early
reorganized,
either
were
settlements
between
food
identified
has
the
structure
connection
of
settlements
and
a
and
This study
directly
inhabitant's
imperialism.
At
the
on
experiences
reflects
of
which
drink consumption
it
Farm,
Court
was suggestedthat the consolidation of the activities of one or more
Barton
households into a single rectangular structure within an enclosed settlement, together with
in
different
direct
identification
public
and
consumption
private
practices, was
of very
the
in
Roman
the Valley. I suggestedthat the use of `Roman' signifiers
the
presence
to
response
identity
The
`native'
to
the
in
reinforce
strategy
settlement.
of
a
sense
of
was, that case,part
Farm was particularly strong around the time that the settlement was
Roughground
at
isolated
deposits
The
Roman-style
drinking
into
of
vessels,
estate.
villa-like
a
restructured
ingredients,
determined
to
that
deposits
that
emphasize
suggested
social-politics
appear
and
inhabitants
in
for
the
that
themselves
and
negotiated
a place
the type of consumptionevent
Old
Shifford
Farm,
At
link
between
the
Roman
world.
rea
was
made
controlled
their now
feasting,
boundaries,
the
the
maintenance
of
communal
consumption
and
and
use of existing
Roman-style
to
containers,
suggest
a
strong
ancestral
association
with
any
of
absence
defiant
Romans.
Finally,
Claydon
Pike,
to
the
attitude
we saw
at
traditions and an almost
in
lived,
house
the
the
the
which
people
and
social contexts of eating and
the
settlement,
that
in
Roman
It
through
the
their
sphere
public
association
a
with
road. was
drinking, were each
inhabitants
in
form
the
that
were
engaged
a
of
conspicuous
consumption
apparent
quite
also
but
`native'
`Roman'
in
their
a
nor
reflection
of
standing
and
power
own
that was neither
in
late
Iron
Age.
the
that
present
was
the community
duration,
has
identified
this
this
work
of
a number of areasthat might
As with most studies
These
Upper
Valley,
further
Thames
to
the
from
refer
specifically
the
study.
study
of
benefit
households
in
Roman
Britain.
the
Age
to
Iron
of
study
late
and
the
Valley
Thames
studies
Upper
We might consider how the apparent similarities between certain settlementsfor
Farm
Thornhill
Shifford
Farm,
Old
both
horses
did
and
raised
which
and
example,
Roman-style
inhabitants
to
the
pots
the
use
reflect
on
of
attitudes
not appear
is
This
keeping
in
Romans.
Barton
Court
that
the
particularly
relevant,
mind
towards
268
Farm and Claydon Pike appear quite similar structurally but had dramatically
different reactionsto the Romans.
It could be interesting to examine the abandonment,expansion or reorganisationof
from
in
Valley
the
the
within
context of possible alliances with
settlements
particular
in
different
the
regional zones.
neighbouringsettlements
Late Iron Age studies

There is much room to explore the relationship between feasting and social power at
both
in
Age
Iron
late
terms of gender and labour relations within
settlements,
the
in
the
the
creation
of
obligations
and
and
maintenanceof power relations
settlements
with the outsidecommunity.
We might consider further the notion of households (or particular people) with
different consumptionpractices, statusesor roles existing at the same settlement,as
hasbeensuggestedfor someof the late Iron Age settlementsin this study.
Britain
in
Roman
Householdstudies
Without the dichotomy of Romanization and native continuity, we are free to go
beyond Roman and native and study people. This thesis has concentrated on
households within settlements; however, through the study of food and drink
the
females
identities
consider
can
also
we
negotiation
of
and
male
consumption
lines
Counihan
(along
households
the
Kaplan
1998).
of
and
within
has
been
level,
thesis
this
different
a.methodological exploration of how one would go
On a
imperialism
`non-Roman'
level,
I
And
the
at
of
effects
this
settlements.
as
on
about studying
just
begun
have
I
to
integrating
by
the
only
consider
possibilities
above,
provided
mentioned
drinking.
What
is
and
eating
of
still very apparent to me, and this
the artefactsand remains
is
Britain,
Roman
that we need to continue to deconstruct the
for
is true
any study of
define
in
to
`Roman'.
`nonsocial
practices,
that
use
terms
not
only
we
of
and
categories
label
interpret
in
the
but
we
and
way
our artefacts. We have seen, for
even
Roman',
`fine
descriptions
fabric
`coarse ware' are used to determine
the
ware'
and
that
example,
hand,
have
implied
function
the
Both
and
one
an
the
on
on
other.
status
wealth and
269
from
be
to
within the context of other types of remains and
considered
interpretations need
before
their
such assumptionsare made.
from within
social context
Romanization
be
full
theories
the
brings
why
of
can
used to
not
circle, emphasizing
This
us
in
Upper
Valley
Thames
(and
that
the
at
settlements
occurred
the
changes
describe
social
before
the
to
to
social
contexts
of
consumption
explore
we
presume
need
we
elsewhere):
know the significance of what was consumed.
270
Bibliography
finds.
Iron
Glass,
H.
(1996)
Iron Age and RomanSettlementat Old Shifford
L.
small
Allen,
and
Farm, Standlake,Hey, G. (ed.) pp. 142- 145. Oxonensia60: 93 - 175.
by
Farm,
Standlake,
Oxford
Unit.
Shifford
Assessment
Archaeological
(1988)
Old
G.
T.
Allen,
Unpublished OAU Evaluation Report.
Settlement
Iron
Age
Romano-British
Enclosed
Watkins
Farm,
(1990)
An
G.
T.
and
at
Allen,
Northmoor, Oxon. ThamesValley Landscapes:the Windrush Valley 1. Oxford: The Oxford
University Committee for Archaeology.
Oxfordshire,
background.
In
Roman
Henig, M. and Booth, P. pp.
Iron
Age
The
G.
(2000)
T.
Allen,
1-33. Oxford: Oxbow Books.
(1993)
Excavations
Farm,
S.
Green
M.
Jones
Roughground
T.
Darvill,
G.,
T.
and
at
Allen,
Lechlade, Gloucestershire:A Prehistoric and RomanLandscape.Oxford: The Oxford
University Committee for Archaeology.
line
in
(1997)
G.
Miles,
D.
A
Hey,
G.,
T.
time
the upper
to
approaches
of
archaeology
and
Allen,
World
Valley,
England.
Archaeology
29
114
129.
(1):
Thames
and middle
in
Upper
Region.
In
S.
Iron
Buildings
Thames
Palmer,
(1984)
D.
Age
Miles,
G.,
T.
the
and
Allen,
Aspectsof the Iron Age in Central SouthernBritain, Cunliffe, B. W. and Miles, D. (eds.) pp.
89 - 101. Oxford: University of Oxford Committee Monograph 2.
M.
Landscape
A.
(1993)
The
Pre-historic
Robinson,
G.
T.
and Iron Age Enclosed
and
Allen ,
Settlementat Mingies Ditch. Oxford: Oxford Archaeological Unit. Thames Valley
Landscapes:The Windrush Valley Volume 2, Oxford University Committee for
Archaeology.
do
Why
(1997)
M.
P.
excavationreports have finds' catalogues?In Not So Much a Pot
Allison,
More a Way of Life: RecentApproachesto Artefact Studies, Cumberpatch,C. and
Blinkhorn, P. (eds.) pp. 77 - 84. Oxford: Oxbow Books
Allison, P. M. (1999) Introduction. In TheArchaeology of HouseholdActivities, Allison, P. M. (ed.)
Routledge.
London:
18.
1pp.
Appadurai, A. (1986) Introduction: commodities and the politics of value. In The Social Life of
Things, Appadurai, A. (ed.) pp. 3- 63. Cambridge: Cambridge University Press.
for
Ground
(1993)
S.
Ardener,
rules and social maps
women: an introduction. In Nomenand
Space: Ground Rulesand Social Maps, Ardener, S. (ed.) pp. 1-30. Oxford: Berg.
271
Armour-Chelu, M. (1991) The faunal remains.In Maiden Castle: Excavations and Field Survey
1985-6, Sharples,N. (ed.), pp.135-151.London: English Heritage Archaeological Reports
19.
Bakels, C., Wesselingh, D. and Van Amen, I. (1997) Acquiring a taste: the menu of Iron Age and
Roman-period farmers at Oss-Ussen,the Netherlands.Analecta Prehistorica Leidensia 29:
193-211.
Fired
Clay.
Iron
G.
(1996)
Age and Roman Settlementat Old
H.,
Hey,
Glass,
A.,
Barclay,
and
Shifford Farm, Standlake,Hey, G. (ed.) Oxonensia60: 93 - 175.
from
Clay
London:
Feats
Africa.
Pots:
British Museum Press.
Smashing
(1994)
N.
of
Barley,
Death.
Grave:
Encounters
London: John Murray.
Dancing
N.
(1995)
the
with
Barley,
on
lake
Glastonbury
(1987)
The
C.
J.
Barrett,
village: models and sourcecriticism. Archaeological
Journal 144: 409 - 423.
Caesar.
In
Barbarians
West
Romans
North
in
C.
(1989a)
Afterward:
J.
Barrett,
and
renderunto
Europe, Barrett, Fitzpatrick J. C. and Maclnnes, L. (eds.) pp. 235 - 241. Oxford: British
Archaeological Reports, International SeriesS471.
In
The
Bronze
Food,
C.
(1989b)
J.
questions
of
social
Barrett,
reproduction.
genderand metal:
Iron Transition, Sorensen,M. L. and Thomas,R. (eds.) pp. 304 - 20. Oxford: British
Archaeological Reports, International SeriesS483: 304 - 320.
Sacred
In
Towards
(1991)
C.
J.
Barrett,
an archaeologyof ritual.
and Profane, Garwood, P.,
Jennings,D., Skeates,R. and Toms, J. (eds.) pp. 1-9. Oxford University Committee for
Archaeology Monograph No 32. Oxford: Oxbow Books.
Recent
Roman
Period:
Britain
Trends, Jones,R. H. J. (ed.).
Review
in
(1993)
C.
J.
the
Barrett,
of
Britannia 24: 343 - 344.
Barrett, J. C. (1997a) Romanization: a critical comment.In Dialogues in Roman Imperialism:
Power, Discourse, and Discrepant Experiencein the Roman Empire, Mattingly D.J. (ed.),
Journal
Rhode
Island:
Roman
64.
Portsmouth,
Supplementary
Archaeology
51
of
pp. SeriesNo 23.
Barrett, J. C. (1997b) Theorising Romanarchaeology.In Proceedingsof the Sixth Annual
Theoretical Roman Archaeology Conference,K. Meadows, C. Lemke and J. Heron (eds.),
Oxbow
Books.
Oxford:
8.
1pp.
Barthes,R. (1979) Towards a psychologyof contemporaryfood consumption. In Food and Drink in
History: Selections From the Annales, Economies,Societes,Civilisations, Forster, R. and
Ranum, 0. (eds.) pp. 166-173.Baltimore: John Hopkins University Press.
Bedoyerede la, G. (1989) The Finds of Roman Britain. London: B. T. Batsford Ltd.
272
River
Valley.
Survey
Thames
An
Upper
Archaeological
The
D.
(1974)
the
Miles,
D.
of
Benson,
and
Gravels. Oxford: Oxford Archaeological Unit Survey No. 2.
Roman
look
fortuna:
0
(1989)
C.
the
M.
archaeological
military
record
and
at
a
sideways
Bishop,
C.
Sources
Driel-Murray,
Evidence,
Equipment:
Military
In
Roman
the
of
van
equipment.
(ed.) pp. 1-11. Oxford: British Archaeological Reports, International Series476.
25:
99-110.
Britannia
Romano-British
Villa-owners:
(1994)
W.
E.
gentlemenand officers,
Black,
houses
in
Roman
Gaul
Britain.
In
The
Early
Roman
First-century
C.
(1990)
F.
T.
and
Blagg,
Empire in the West,Blagg, T. F. C. and Millett, M. (eds.) pp. 194-209.Oxford: Oxbow
Books.
London:
New
York
Study.
Comparative
A
Households:
Houses
(1994)
E.
and
and
Blanton, R.
Plenum Press.
from
British
The
two
(1986)
C.
Evans,
transformation
prehistory.
of space:
examples
Boast, R. and
Archaeological Reviewfrom Cambridge 5 (2): 193- 205.
implications,
Pottery
Linear
their
Ceramic
(1984)
I.
the
P.
economic
culture and
sievesof
Bogucki,
Oxford Journal ofArchaeology 3 (1): 15-30.
Warwickshire:
in
Roman
between
Inter-site
pottery assemblages
comparisons
Booth, P. (1991)
4:
1-10.
Pottery
Studies
Roman
Journal
indicators
of
of site status,
ceramic
Excavations
In
from
1990
the
Introduction
(1993)
the
at
to
excavation.
pottery
Booth, P.
Roughground Farm, Lechlade, Gloucestershire:A Prehistoric and Roman Landscape,
Oxford
Oxford:
The
)
143.
135
Jones,
M.
(eds.
S.
Green,
T.,
T.
Darvill,
pp.
Allen,
and
University Committee for Archaeology
from
Thames
Valley.
Journal
data
Upper
Quantifying
the
(in
status,some pottery
Booth, P.
press)
Studies.
Pottery
Roman
of
Latin
Britannia
Writing-Tablets.
London:
The
Vindolanda:
D.
(1983)
Thomas,
K.
A.
Bowman,
and
Monograph SeriesNo. 4.
Context,
family
dinner.
Nielsen,
I.
In
Meals
Social
in
Roman
The
(1998)
K.
and
Bradley,
a
at
Nielsen, H. S. (eds.) pp. 36 - 55. Aarhus, Denmark: Aarhus University Press.
Longman
Prehistoric
Britain.
Foundations
London:
Social
The
(1984)
R.
of
Bradley,
Bradley, R. (1990) Thirty yearsof Roman Britain, Britannia 21: 394 - 396.
Miles,
D.
Countryside,
farm
Roman
In
The
Romano-British
Celtic
to
K.
(1981)
villa.
Branigan,
(ed.) pp. 81 - 96. Oxford: British Archaeological Reports British Series 103.
273
Branigan, K. (1985) The Catuvellauni. Gloucester:Alan Sutton.
in
Specialisations
(1988)
K.
villa economies.In Villa Economies: Economic Aspectsof
Branigan,
Romano-British Villas, Branigan,K. and Miles, D. (eds.) pp.42 - 49. Sheffield: Sheffield
University Press.
Empire?
An
Images
K.
(1991)
to
the
of
archaeological
mirages
or
approach
problem.
Branigan,
- In Images of Empire, Alexander, L. (ed.) pp. 91-105. Sheffield: Sheffield Academic Press.
from
Britain
Transactions
Roman
(1994)
The
K.
the
view
west
country,
of the
new
Branigan,
-a
Bristol and GloucestershireArchaeological Society 110 (11): 9-16.
imported
Not
In
The
(1997)
H.
D.
of
medieval
significance
pottery.
social
so much a pot,
Brown,
C.
Blinkhorn,
R.
W.
(eds.
)
95
112.
Oxford:
life,
Cumberpatch,
and
pp.
more a way of
Oxbow Monograph 83.
Huexotla,
American
Anthropology
89:
Aztec
Consumption
M.
(1987)
E.
at
and
politics
Brumfiel,
676 - 686.
ideological
Aztec
Figurines
E.
M.
(1996)
testing
the
of
state:
and effectiveness
and
Brumfiel,
domination. In Genderand Archaeology, Wright, R. P. (ed.) pp. 143 - 166. Philadelphia:
University of PennsylvaniaPress.
Invasion
Response:
The
In
Pre-Roman
Romano-British
(1979)
C.
B.
urbanism.
and
and
Burnham,
Case of Roman Britain, Burnham,B. C. and Johnson,H.B. (eds.) pp.255 - 272. Oxford:
British Archaeological Reports,British Series73.
The
Case
Roman
Invasion
Response:
H.
B.
(eds.
)
(1979)
Johnson,
C.
B.
and
of
Burnham,
and
Britain. Oxford: British Archaeological Reports,British Series73.
J.
Towns'
Roman
Britain.
London:
B.
Wacher,
(1990)
The
'Small
T.
C.
B.
of
Burnham,
and
Batsford Ltd.
Scottish
Archaeological
Review
9 and 10: 15 - 21.
Post-processual
(1995)
S.
palynology.
Butler,
Carsten,J. and Hugh-Jones,S. (1995) Introduction. In About the House: Levi Strauss and Beyond,
Carsten,J. and Hugh-Jones,S (eds.), pp. 1-46. Cambridge: Cambridge University Press.
Case,H. J. and Whittle, A. W.R. (eds.) (1982) SettlementPatterns in the Oxford Region:
Excavations at the Abingdon CausewayedEnclosure and Other Sites. London: Council for
British Archaeology ResearchReport no. 44.
Chapman,M. (1992) The Celts: The Construction of a Myth. New York: St Martin's Press.
Clarke, S. (1996) Acculturation and continuity: re-assessingthe significance of Romanization in the
hinterlands of Gloucesterand Cirencester.In RomanImperialism: Post-Colonial
Perspectives.,Webster,J. and Cooper,N. (eds.), pp. 71 - 69, Leicester: Leicester
Archaeology MonographsNo 3.
274
Clarke, S. (1998) Social changeand architecturaldiversity in Roman period Britain. In Proceedings
Theoretical
Roman
Archaeology
Seventh
Annual
Conference,
Forcey, C., Hawthorne,
the
of
J. and Witcher, R. (eds.) pp. 28 - 41. Oxford: Oxbow Books.
Collis, J. (1979) City and state in pre-Roman Britain. In Invasion and Response: The Case of
Roman Britain, Burnham, B. C. and Johnson, H. B. (eds.) pp. 231-240. Oxford: British
Archaeological Reports, British Series 73.
before
Romans.
In
Europe
The
Roman
World
Celtic
(1987)
(Vol. 1), Wacher, J. (ed.)
J.
the
Collis,
London: Routledge and Kegan Paul.
Conkey M. W. and Gero, J. M. (1991) Tension,pluralities, and engenderingarchaeology:an
introduction to women and pre-history. In EngenderingArchaeology: Womenand
Prehistory, Gero, J. M. and Conkey, M. W. (eds.) pp. 3- 30. Oxford: Basil Blackwell.
blank
for
Searching
J.
(1996)
N.
the
generation:consumerchoice in Roman and postCooper,
Roman Britain. In RomanImperialism: Post-Colonial Perspectives,Webster, J. and Cooper,
N. (eds.) pp. 85 - 98. Leicester:LeicesterArchaeology MonographsNo 3.
distinction
legitimation
in
Status
(1989)
T.
Earle,
L.
C.
and
Costin,
of power as reflected
and
in
late
Peru.
American
54:
691
Antiquity
of
consumption
prehispanic
patterns
changing
714.
Gender:
Food
S.
L.
(eds.
)
Kaplan,
(1998)
Identity
Power.
C.
M.
and
and
Counihan,
and
Amsterdam: Harwood Academic Publishers.
Early
Suffolk:
Husbandry.
East
Stow,
Anglo-Saxon
West
Animal
Anglian
(1989)
J.
P.
Crabtree,
Archaeology Report no. 47, Suffolk County Planning Department.
690
Antiquity
(4):
American
55
717.
diet
Monticello,
Slave
(1990)
C.
D.
Crader,
at
In
(1997)
Towards
C.
to
the
Cumberpatch,
study of medieval pottery.
a phenomenologicalapproach
Not so much a pot, more a way of life, Cumberpatch,C. and Blinkhorn, R. W. (eds.) pp. 125
Oxbow Monograph 83.
Oxford:
152.
Cunliffe, B. W. (1976) The origins of urbanisationin Britain. In Oppida in Barbarian Europe,
Cunliffe, B. and Rowley, T. (eds.) pp. 135- 162. London: British Archaeological Reports
11.
Cunliffe, B. W. (1979) Someconcluding thoughts.In Invasion and Response:The Caseof Roman
Britain, Burnham, B.C. and Johnson,H.B. (eds.) pp.359 - 365. Oxford: British
Archaeological Reports,British Series73.
Cunliffe, B. W. (1984) Iron Age Wessex:continuity and change.In Aspectsof the Iron Age in
Central Southern Britain, Cunliffe, B. W. and Miles, D (eds.) pp. 12 - 45. Oxford:
University of Oxford Committee for Archaeology Monograph 2.
275
in
drinking
Britain:
Eating
(1979)
B.
G.
the evidenceof amphoraand
pre-conquest
Dannell,
and
invasion
Claudius.
In
Invasion
The
Response:
the
trading,
the
of
of
and
effect
and
saurian
Caseof Roman Britain, Oxford: British Archaeological Reports, British Series73.
form.
Journal
fabric,
Nice
(1989)
J.
M.
the
of Roman Pottery Studies,2: 98 pity about
Darling,
101.
diet.
Britannia
11:
122
142.
The
Roman
W.
(1971)
R.
military
Davies,
for
Journal
Botanical
W.
(1974)
R.
crop
processingactivities.
of
evidence prehistoric
Dennell,
Archaeological Science,1: 275 - 284.
horse
Durotriges.
Proceedings
Hippophagy
(1969)
L.
S.
the
H.
sacrifice
among
or
of the
Dewar,
Dorset Natural History and Archaeological Society91: 194-195.
in
drinking
by
drink:
(1990)
Driven
the political economy and the caseof
M.
the
role of
Dietler,
Archaeology
9:
406.
352
Journal
France.
Age
Iron
ofAnthropological
early
food,
in
(1996)
Feasts
the
M.
political
power and status
economy:
Dietler,
and commensalpolitics
in prehistoric Europe. In Food and the Status Quest:An Interdisciplinary Perspective,
Wiessner,P. and Schiefenhovel,W. (eds.) pp. 87 - 125. Providence,USA: Berghahn
Books.
food.
Order:
Food
Food
Social
Studies
In
in
Standard
(1984)
the
M.
of
and
of
social uses
Douglas,
Festivities in ThreeAmerican Communities,Douglas, M. (ed.) pp. 1- 39. New York:
RusselSageFoundation.
World
Towards
Goods:
Anthropology
The
B.
(1978)
Isherwood,
M.
of
of
an
Douglas,
and
Consumption.London: Allen Lane PenguinsBooks Ltd.
Interpretation
Reconstruction.
Structural
In
Approaches
Introduction.
to
(1982)
the
J.
P.
of
Drury,
Oxford:
J.
(ed.
)
British
Drury,
P.
5.
Buildings,
1
Remains
Excavated
pp. of
the
Archaeological Reports,British Series 110.
Dungworth, D. (1998) Mystifying Roman nails: clavus analis, defixiones and minkisi. In
Proceedingsof the SeventhAnnual Theoretical RomanArchaeology Conference,Forcey, C.,
Hawthorne, J. and Witcher, R. (eds.) pp. 149 - 159. Oxford: Oxbow Books.
Dyer, C. (1982) The social and economic changesof the later middle agesand the pottery of the
6:
42.
Ceramics
33
Medieval
period.
Earwood,C. (1993) Domestic WoodenArtefacts in Britain and Ireland From Neolithic to Viking
Times.Exeter: University of Exeter Press.
Edmonds,M. (1993) Interpreting causewayedenclosuresin the past and present.In Interpretative
Archaeology, Tilley, C. (ed.) pp. 99 - 142. Providence,USA and Oxford: BerghahnBooks.
276
Churches
West
Natives,
Christians
(1980).
Romans
Hill,
Uley.
In
Temples,
Ellison, A.
on
and
and
Religion in RomanBritain, Rodwell, W. (ed.) pp. 305 - 328. Oxford: British Archaeological
Reports, British Series77.
Es, W. A. van (1983) Introduction. In Roman and Native in the Low Countries: Spheres of
Interaction, Brandt, R. and Slofstra, J. (eds.) pp. 1- 11. Oxford: British Archaeological
Reports, International Series 184.
Evans, A. K. B. (1981) Pottery and history. In Roman Pottery Research in Britain and North -West
Europe, Part II, Anderson, A. C. and Anderson, A. S. (eds.) pp. 517 - 535. Oxford: British
Archaeological Reports, International Service 123 (2).
further
interpretations
be
Whether
(1982)
C.
vats:
some
Evans,
pits
of subterraneanfeatures.
Institute ofArchaeology Bulletin 19: 171-176.
Evans, C. (1989) Perishablesand worldly goods- artefactdecorationand classification in the light
8
(2):
Journal
179
201.
Oxford
ofArchaeology
research.
of wetland
literacy
in
Graffiti
(1987)
J.
the
Evans,
and pottery use Roman Britain.
and
evidenceof
Archaeological Journal 144: 191 - 204.
Evans, J. (1991) Not more pot. Journal of RomanPottery Studies4: 69-75.
Evans, J (1993) Pottery function and fine wares in the Roman north. Journal of Roman Pottery
Studies6: 95 - 118.
Evans,J. (1995) Late Iron Age and `native' pottery in the north-east.In Moorland Monuments:
Studiesin the Archaeology of North-East Yorkshirein Honour of RaymondHayes and Dan
Spratt, Vyner, B. (ed.) pp. 46 - 68. London: Council for British Archaeology Research
Report No. 101.
Ferrell, G. (1995) Spaceand society: new perspectiveson the Iron Age of north-eastEngland. In
Different Iron Ages. Studieson the Iron Age in temperateEurope, Hill, J. D. and
Cumberpatch,C. G., (eds.) pp. 129 - 147. Oxford: British Archaeological Reports,
International Series602.
Fisher, A. R. (1985) Winklebury Hillfort: A study of artefactdistribution from subsoil features.
Proceedingsof the Prehistoric Society, 51: 167 - 180.
Fletcher,R. (1992) Time perspectivism,annalesand the potential of archaeology.In Archaeology,
Annales and Ethnohistory, Knapp, A. B. (ed). Cambridge:Cambridge University Press.
Freeman,P. W. M. (1991) The study of the Roman period in Britain: a comment on Hingley.
ScottishArchaeological Review 8: 102 -104.
Freeman,P. W. M. (1993) `Romanisation' and Roman Material Culture. Journal of Roman
Archaeology 6: 438 - 45.
277
Roman
Empire.
imperialism
In
Imperialism:
Roman
(1996)
British
W.
M.
P.
the
and
Freeman,
Post-Colonial Perspectives,Webster, J. and Cooper, N. (eds.) pp. 19 - 34. Leicester:
Leicester Archaeology MonographsNo 3.
in
Haverfield:
Romanization
(1997)
Mommsen
the
W.
M.
to
P.
through
studies
origins of
Freeman,
late 19th-c. Britain. In Dialogues in Roman Imperialism: Power, Discourse, and Discrepant
Experience in the Roman Empire, Mattingly, D. J. (ed.) pp. 27 - 50. Rhode Island: Journal
No
Series
23.
Supplementary
Archaeology,
Roman
of
London:
Cardinal.
Britain.
Roman
History
A
Britannia:
(1974)
S.
of
Frere,
dependence
in
Ist
2"d
In
Britannia's
Demonstrating
(1984)
the
M.
economic
and
centuries.
Fulford,
Military and Civilian in Roman Britain. Cultural Relationships in a Frontier Province,
Blagg, T. F. C. and King, A. C. (eds.) pp. 129 - 142. Oxford: British Archaeological Reports,
British Series 136.
In
Developing
Roman:
the
Iron
Age
(1992)
M.
to
change
gravels.
on
a period of radical
Fulford,
Landscapesof Low-land Britain. TheArchaeology of the British Gravels: A Review,
Fulford, M. and Nichols, E. (eds.) pp. 23 - 38. London: The Society of Antiquaries of
London, Occasionalpapersvol. 14.
State
Studies:
A
Current
Romano-British
Pottery
The
(1991)
K.
Huddleston,
G.
M.
of
Fulford,
and
Reviewfor English Heritage. London: English Heritage.
in
history.
In
Historical
Introduction:
(1999)
M.
Hall,
S.
Jones,
P.
A.,
archaeology
P.
and
Funari,
M.
120.
London:
Routledge.
Hall,
(eds.
S.
)
Jones,
P.
A.,
P.
Funari,
and
pp.
Archaeology,
Survey
Prehistoric
Farm
Project.
An
Integrated
Maddle
The
(1989)
M.
Tingle,
V.
of
Gaffney, and
Oxford:
Archaeological
Reports,
Downs.
British
Berkshire
Landscapes
the
Roman
on
and
British Series200.
faunal
In
from
Sampling
in
information
Optimising
(1978)
C.
studies
of
remains.
Gamble,
Contemporary British Archaeology, Cherry, J. F., Gamble, C. and Shennan,S. (eds.) pp.
321-353. Oxford: British Archaeological Reports,British Series50.
impact
faunal
The
G.
(1994)
Bailey,
the
C.
specialistas excavator:
of recovery
Gamble, and
interpretation
Klithi.
Whither
In
Environmental
Archaeology?
Luff,
faunal
at
techniqueson
R. and Rowley-Conwy, P. (eds.) pp.81 - 89. Oxford: Oxbow Books.
Garnsey,P. (1999) Food and Society in Classical Antiquity. Cambridge: Cambridge University
Press.
Economy,
Society and Culture. London: Duckworth.
Roman
The
(1987)
R.
Saller,
P.
Garnsey, and
Gero,J. M. (1996) Archaeological practice and genderedencounterswith field data. In Genderand
Archaeology, Wright, R. P. (ed.) pp. 251 - 283. Philadelphia: University of Pennsylvania
Press.
278
Gilbert, A. S. and Singer, B. H. (1982) Reassessingzooarchaeologicalquantification. World
Archaeology, 14 (1): 21 - 40.
Godwin, H. (1975) The History of British Fauna: A Factual Basisfor Phytogeography.2"d edition.
Cambridge:CambridgeUniversity Press.
Gomez de Soto, J. (1993) Cooking for the elite: feasting equipment in the late Bronze Age. In
Trade and Exchangein Prehistoric Europe, Scarre,S. and Healy, F. (eds.) pp. 191 - 197.
Oxford: Oxbow Books.
Goody, J. (1982) Cooking, Cuisine and Class: A Study in Comparative Sociology. Cambridge:
CambridgeUniversity Press.
Gosden, C. (1989) Debt, production and prehistory.Journal ofAnthropological Archaeology 8:
355 - 387.
Gosden, C. (1999) Food: where biology meetsculture. In ThePrehistory of Food, Gosden,C. and
Hather, J. (eds.) pp. 1-9. London and New York: Routledge.
Gowers, E. (1993) The Loaded Table: Representationof Food in Roman Literature. Oxford:
ClarendonPress.
Grahame,M (1998) Redefining Romanization:material culture and the question of social
in
Britain.
In
Seventh
Roman
Proceedings
Theoretical
Annual
Roman
the
of
continuity
Archaeology Conference,Forcey, C., Hawthorne, J. and Witcher, R. (eds.) pp. 1- 10.
Grant, A. (1975) Animal Bones. In Excavationsat Porchester Castle Vol. P Roman, Cunliffe, B.
(ed.) pp. 378 - 408. London: The Society of Antiquities London.
Grant, A. (1981) The significance of deerremains at occupationsites of the Iron Age to the AngloSaxonperiod. In The Environment of Men: The Iron Age to the Anglo-Saxon Period, Jones,
M. and Dimbleby, G. (eds.) pp. 205 - 213. Oxford: British Archaeological Reports, British
Series87.
Grant, A. (1984) Survival or sacrifice: a critical appraisalof animal burials in Britain in the Iron
Age. In Animals and Archaeology. 4. Husbandry in Europe, Clutton Brock, J. and Grigson,
C. (eds.) pp. 221 - 227.Oxford: British Archaeological Reports, International SeriesS227.
Grant, A. (1989) Animals in Roman Britain. In Researchon Roman Britain -1960 - 1989, Todd,
M. (ed.) pp. 135-146.London: Society for the Promotion of Roman Studies(Britannia
Monograph 11).
Grant, A. (1991) Economic or symbolic? Animal and ritual behaviour. In Sacredand Profane,
Garwood, P., Jennings,D., Skeates,R. and Toms, J. (eds.) pp. 109 - 114. Oxford University
Committee for Archaeology Monograph No 32. Oxford: Oxbow Books.
279
Grayson, D. K. (1984) Quantitative Zooarchaeology: Topics in the Analysis of archaeological
Faunas. Orlando: Academic Press.
Green, C. (1980) Handmadepottery and society in late Iron Age and Roman Sussex.Sussex
Archaeological Collections 118: 69 - 86.
Green, F. J. (1981) Iron Age, Roman and Saxon Crops: The archaeologicalevidencefrom Wessex.
In TheEnvironment of Man: The Iron Age to Anglo-SaxonPeriod, Jones,M. and Dimbleby,
G. (eds.) pp. 129 - 153. Oxford: British Archaeological Reports, British Series87.
Green, M. (1992) Animals in Celtic Life and Myth. London and New York: Routledge.
Green, S. (1993) Introduction to the pottery recoveredbetween1957 and 1982. In Excavations at
RoughgroundFarm, Lechlade, Gloucestershire:A Prehistoric and RomanLandscape,
Allen, T. Darvill, T., Green, S. and Jones,M. (eds.) pp. 113- 135. Oxford: The Oxford
University Committee for Archaeology
Gregory, C.A. and Altman, J.C. (1989) Observing the Economy.London: Routledge.
Gregson,M. (1988) The villa as private property. In Villa Economies: Economic Aspectsof
Romano-British Villas, Branigan,K. and Miles, D. (eds.) pp.21 - 33. Sheffield: Sheffield
University Press.
Griffiths, K. E. (1989) Marketing of Romanpottery in second-centuryNorthamptonshire and the
Milton Keynes area.Journal of RomanPottery Studies2: 67 - 76.
Groenman-vnWaateringe(1980) Urbanization and the north-west frontier of the Roman Empire, In
RomanFrontier Studies 1979: Paperspresentedto the 12'hInternational Congressof
RomanFrontier Studies,Hanson,W. S. and Keppie, L. J.F. (eds.), pp. 1037 - 1044. Oxford:
British Archaeological Reports, International SeriesS71, part iii.
Gwilt, A. and Heslop, D. (1995) Iron Age and Roman quernsfrom the TeesValley. In Moorland
Monuments:Studies in the Archaeology of North-East Yorkshire in Honour of Raymond
Hayes and Dan Spratt, Vyner, B. (ed.) pp. 38 - 45. London: Council for British Archaeology
ResearchReport No. 101.
Halstead,P. (1985) A study of mandibular teeth from Romano-British contexts at Maxey. In
Archaeology and Environment in the Lower Welland Valley, Volume 1, Pryor, F. and
French, C. (eds.) pp. 219 - 224. Cambridge: CambridgeArchaeological Committee.
Halstead,P., Hodder, I. And Jones,G. (1978) Behavioural archaeologyand refusepatterns: a case
Review
Archaeological
11: 118- 131.
Norwegian
study.
Hamilakis, Y, (2000) The anthropology of food and drink consumption and the Aegean
Aegean,
Vaughan,
S.
in
E.
In
Paleodiet
J.
W.
(eds.
)
Coulson,
D.
the
and
archaeology.
280
from
behavioural
(1991)
Beyond
M.
J.
reconstruction
subsistence:
paleoethnobotany.
Hansen,
Archaeological ReviewFrom Cambridge 10 (1): 53 - 59.
in
In
The
Roman
S.
(1988)
Administration,
W.
the
west.
urbanization
and
acculturation
Hanson,
Administration of the Roman Empire 241 BC - AD 193, Braund, D. (ed.) pp. Exeter: Exeter
Exeter
Press.
University
18,
in
history
of
no.
studies
Britain.
Building
In
in
The
Romanization
Barbarians:
Dealing
W.
S.
(1994)
the
of
with
Hanson,
Past: papers celebrating 100years of the Royal Archaeological Institute, Vyner, B. (ed.)
Institute.
Archaeological
Royal
London:
163.
149
pp.
in
Britain.
In
Food
food
Iranians
Taste
(1997)
the
L.
of
preferences
and embodiment:
Harbottle,
Preferences and Taste: Continuity and Change, Macbeth, H. (ed.) pp. 175 - 185.
Providence, USA: Berghahn Books.
historic
Journal
Britain.
in
dog
The
(1974)
R.
A.
of archaeological
prehistoric and early
Harcourt,
Science, 1: 151-175.
Report.
Summary
Excavations
1969-70:
A
House,
Piercebridge:
Holme
W.
(1984).
D.
Harding,
Edinburgh: University of EdinburghDepartmentof Archaeology Project PaperNo. 2.
27
30.
3:
Review,
Scottish
Archaeological
Hingley,
Comment
C.
(1984a)
on
Haselgrove,
British
Gaulish
before
`Romanization'
C.
(1984b)
the
and
precedents
conquest:
Haselgrove,
Relationships
in
Britian.
Cultural
Roman
Civilian
in
Military
In
a
and
consequences.
Oxford:
63.
British
King,
C.
(eds.
)
5
F.
C.
T.
A.
Blagg,
Province,
Frontier
pp.
and
Archaeological Reports,British Series 136.
Ronan
In
Research
beyond.
in
in
Britain
later
Iron
The
Age
(1989)
C.
and
southern
Haselgrove,
Britain 1960-1989,Todd, M. (ed). pp. 1- 18. London: Society for the Promotion of Roman
Studies(Britannia Monograph 11).
In
The
Belgic
Gaul:
Romanization
The
(1990)
C.
some archaeologicalperspectives.
of
Haselgrove,
Early RomanEmpire in the West,Blagg, T. and Millett, M. (eds.) pp. 45 - 71. Oxford:
Oxbow Books.
Hastorf, C. A. (1990) The effect of the Inka state on Sausaagricultural production and crop
262
Antiquity
55(2):
290.
American
production.
food
in
Gender,
(1991)
C.
A.
Hastorf,
space,and
prehistory. In Engendering Archaeology: Women
M.
Conkey,
M.
Basil
J.
159.
Oxford:
Gero,
W.
(eds.
)
132
Prehistory,
and
pp.
and
Blackwell.
domestic
life
782.
72:
773
The
Antiquity
(1998)
C.
A.
of early
Hastorf,
cultural
plant use.
Seventh
ideologies
Romanization.
Proceedings
Motivations
In
the
(1997)
R.
of
of
Haussler,
and
Annual Theoretical RomanArchaeology Conference,Forcey, C., Hawthorne, J. and
Witcher, R. (eds.) pp. 11 - 19. Oxford: Oxbow Books.
281
Haverfield, F. (1912) The Romanizationof Roman Britain (2nd edition). Oxford: Clarendon Press.
Hawkes, G. (2001) An archaeologyof food: a casestudy from Roman Britain. In TRAC 2000:
TenthAnnual TheoreticalArchaeology ConferenceProceedings,Davis, G., Gardener,A.
Oxford:
Oxbow
Books
K.
(eds.
).
Lockyear,
and
in
W.
J.
(1998)
Pottery
J.
Hawthorne,
and paradigms the early Western Empire. In Proceedings of
Conference,
Roman
Forcey, C., Hawthorne, J.
Archaeology
Theoretical
Seventh
Annual
the
Oxford:
Oxbow
Books.
160
172.
(eds.
)
Witcher,
R.
pp.
and
food
The
Planters:
Pluckers,
Nimrods,
Piscators,
(1990)
B.
and
emergenceof
production.
Hayden,
Journal ofAnthropological Archaeology, 9: 31 - 69.
Hayden, B. (1996) Feastingin prehistorical and traditional societies.In Food and the Status Quest:
An Interdisciplinary Perspective,Wiessner,P. and Schiefenhovel,W. (eds.) pp. 127 - 146.
Providence,USA: BerghahnBooks.
in
highlands.
disposal
Maya
Where
(1983)
Cannon,
A.
the
B.
the
refuse
Hayden, and
garbagegoes:
Journal ofAnthropological Archaeology 2: 117-163.
Helbaek, H. (1952) Early crops in southernEngland. ThePrehistoric Society 12: 194 - 233.
Heller, A. (1984) EverydayLife. London: Routledge and Kegan Paul.
form
function:
(1983)
Ceramic
M.
M.
A.
F.
McDonald,
E.
an ethnographic
Hendrickson,
and
and
85:
630
645.
American
Anthropologist
searchand archaeologicalapplication.
Henig, M. and Booth, P. (2000) RomanOxfordshire. Oxford: Oxbow Books.
Hey, G. (1996) Iron Age and Roman Settlementat Old Shifford Farm, Standlake.Oxonensia60: 93
175.
Hill, J. D. (1989) Re-thinking the Iron Age. Scottish Archaeological Review, 6: 16-24.
Hill, J. D. (1994) Why we should not take the data from Iron Age settlementsfor granted: recent
In
The
in
Recent
Intra-settlement
Iron
Age
Work,
Wessex:
patterning.
studiesof
Figtzpatrick, A. P. and Morris, E. (eds.), pp. 4-8. Salisbury: Association FrancaiseD'Etude
de L'Age du Fer/Trust for WessexArchaeology.
Hill, J. D. (1995) Ritual and Rubbish in the Iron Age of Wessex:A Study on the Formation of a
Specific Archaeological Record. Oxford: British Archaeological Report, British Series242.
Hill, J. D. (1997) `The end of one body and the beginning of another kind of body'? Toilet
instrumentsand `Romanization' in southern England during the first century AD. In
ReconstructingIron Age Societies,Gwilt, A. and Haslegrove,C. (eds.) pp. 96 - 107.
Oxford: Oxbow Monograph 71.
Hinchliffe, J. and Thomas, R. (1980) Archaeological investigationsat Appleford. Oxonensia 14: 911
282
Hingley, R. (1984) Toward social analysisin archaeology:Celtic society in the Iron Age of the
Upper ThamesValley (400-0BC). In Aspectsof the Iron Age in Central Southern Britain,
Cunliffe, B. W. and Miles, D. (eds.) pp. 72 - 88. Oxford: University of Oxford Committee
for Archaeology Monograph no. 2.
Hingley, R. (1988) The influence of Rome on indigenoussocial groups in the Upper Thames
Valley. In First Millennium Papers: WesternEurope in the First Millennium AD, Jones,
R.F.J. et al (eds.) pp. 73-98. Oxford: British Archaeological Reports, International Series
401.
Hingley, R. (1989) Rural Settlementin RomanBritain. London: Seaby.
Hingley, R. (1990a) Domestic organisationand genderrelations, in Iron Age and Romano-British
Households.In TheSocial Archaeology of Houses,Samson,R. (ed.) pp. 125 - 147.
Edinburgh: Edinburgh University Press.
Hingley, R. (1990b) BoundariessurroundingIron Age and Romano-British settlements.Scottish
Archaeological Review, 7: 96 - 103.
Hingley, R. (1991) Past,presentand future - the study of the Roman period in Britain. Scottish
Archaeological Review 8: 90 - 101.
Hingley, R. (1993) Attitudes to Roman imperialism. In Theoretical RomanArchaeology: First
ConferenceProceedings, Scott, E. (ed.) pp. 23 - 27. Aldershot: Avebury Press.
Hingley, R. (1995) The Iron Age in Atlantic Scotland: searchingfor the meaning in the substantial
house.In Different Iron Ages: Studieson the Iron Age in TemperateEurope, Hill, J. D. and
Cumberpatch,C. (eds.) pp. 185 - 193. Oxford: British Archaeological Reports, International
SeriesNo. 602
Hingley, R. (1996) The `legacy' of Rome: the rise, decline, and fall of the theory of Romanization.
In RomanImperialism: Post-Colonial Perspectives,Webster,J. and Cooper, N. (eds.) pp.
35 - 48. Leicester: Leicester Archaeology MonographsNo 3.
Hingley, R. (1997) Resistanceand domination: social changein Roman Britain. In Dialogues in
RomanImperialism: Power, Discourse, and Discrepant Experience in the Roman Empire,
D. J. Mattingly (ed.) pp. 81 - 100. Portsmouth,Rhode Island: Journal of Roman
Archaeology SupplementarySeriesNo 23.
Hingley, R. (1999) The imperial context of Romano-British studies and proposals for a new
In
Historical
Archaeology, Funari, P. P. A., Jones,S. and
change.
understandingof social
Hall, M. (eds.) pp. 137 - 149. London: Routledge.
Hingley, R. (2000) Roman Officers and English Gentlemen:The Imperial Origins of Roman
Archaeology. London: Routledge.
283
Hingley, R. and Miles, D. (1984) Aspectsof Iron Age Settlementin the Upper ThamesValley. In
Aspects of the Iron Age in Central SouthernBritain, B. Cunliffe and D. Miles (eds.) pp. 52
University of Oxford, Committee for Archaeology Monograph No. 2.
Oxford:
-71.
Hirth, K. G. (1993) The householdas an analytical unit: problems in method and theory. In
Prehistoric Domestic Units in WesternMesoamerica,Santley, R. S. and Hirth, K. G. (eds.)
CRC
Press.
Boca
Raton
36.
21
:
pp.
Hodder, I. (1989) Writing archaeology:site reports in context. Antiquity, 63: 268 - 74.
discussion.
In Production and
Pottery,
(1981)
I.
theoretical
Hodder,
production and use: a
Distribution: A Ceramic Viewpoint,Howard, H. and Morris, E. L. (eds.) pp. 215 - 220.
Oxford: British Archaeological Reports,International Series S120.
Howard, H. (1981) In the wake of distribution: towards an integrated approachto ceramic studiesin
Ceramic
Distribution:
A
Viewpoint,
Howard,
H.
In
Production
Britain.
and
and
prehistoric
E. L. Morris (eds.) pp. 1 -330.Oxford: British Archaeological Reports, International Series
S120.
Huelsbeck, D. R. (1991) Faunal remainsand consumerbehaviour: what is being measured?
Historical Archaeology 25: 62 - 76.
Ingold, T. (1984) Time, social relationshipsand the exploitation of animals: anthropological
3.
Early
In
Archaeology:
Herders
Flocks,
Animals
in
their
and
reflections on prehistory.
Clutton - Brock, J. and Grigson, C. (eds.) pp. 3 -12. London: British Archaeological
Reports, International Series202.
Ingold, T. (1995) Building, dwelling, living: how animals and people make themselvesat home in
Contexts:
Transformation in Anthropological Knowledge, Strathern,
Shifting
In
the world.
M. (ed.) pp. 57 - 80. London: Routledge.
London:Routledge.
Ireland,S. (1986)RomanBritain:. Sourcebook.
Johnson,M. H. (1989) Conceptionsof agencyin archaeologicalinterpretation. Journal of
Anthropological Archaeology 8: 189 - 211.
Johnson,M. H. (1993)HousingCulture: TraditionalArchitecturein an EnglishLandscape.

London:UniversityCollegeLondonPress.
Jones,B. and D. Mattingly (1990) An Atlas of RomanBritain. Oxford: Blackwell.
Jones,G. and Levitan, B. (1993) Animal bones.In Excavationsat Roughground Farm, Lechlade,
Gloucestershire:A Prehistoric and RomanLandscape,Allen, T. G., T. Darvill, S. Green
Oxford:
171
175.
(eds.
)
Oxford
for
The
Jones
Committee
M.
University
pp.
and
Archaeology.
Jones,G. E. M. (1991) Numerical analysisin archaeobotany.In Progress in Old World
Palaeoethnobotany,Van Zeist, W., Wasylikowa, K. and Behre, K. E. (eds.), pp. 63 - 80.
284
in
Oxfordshire.
Sampling
in
In
in
Sampling
(1978)
M.
study
a
case
context:
a rescue
Jones,
Contemporary British Archaeology,Cherry, J. F., Gamble, C. and Shennan,S. (eds.) pp.
321 - 353. Oxford: British Archaeological Reports,British Series 50.
In
Beyond
beyond
Domestication
in
Archaeology
(1985)
M.
subsistence
reconstruction.
Jones,
Prehistoric Europe, Barker, G. Gamble, C. S. (eds.) pp. 107 - 128. New York: Academic
Press.
In
Archaeology
Barton
Court
Farm,
Abingdon,
The
(1986a)
M
at
remains.
Jones,
carbonizedplant
Oxfordshire, D. Miles (ed.), chapter7 (microfiche 9:A1-9: B5) Oxford: Oxford
Archaeological Unit Report 3, Council British Archaeology ResearchReport No 50.
B.
T.
Batsford
Ltd.
London:
Domesday.
Before
England
(1986b)
M.
Jones,
dynamics
In
Research
Roman
in
Britain:
Roman
Agriculture
(1989)
the
M.
of
change.
on
Jones,
Britain 1960 -1986, Todd, M. (ed.) pp. 127.134. London: Britannia Monograph SeriesNo.
11.
Trends,
In
Britain:
Recent
Roman
Food
(1991)
M.
Jones,
production and consumption- plants.
Jones,R. F. J. (ed.) pp. 21- 27. Sheffield: J. R. Collis Publications.
Jones,M. and Miles, D. (1979) Celt and Roman in the ThamesValley: approachesto cultural
B.
C.
Roman
Burnham,
The
Case
Britain,
Response:
Invasion
In.
and
of
and
change.
Johnson,H.B. (eds.) pp.315-325. Oxford: British Archaeological Reports, British Series73.
(ed.
Millett,
M.
)
1-5.
Archaeologist,
In
Pottery
Why
(1979)
J.
F.
the
R.
pp.
and
Jones,
pottery?
London: Institute of Archaeology OccasionalPublication no. 4.
in
Constructing
Ethnicity:
Identities
The
Archaeology
the Past and the Present.
(1997)
S.
of
Jones,
London: Routledge.
identities
in
First
Britain:
Brooches
Century
AD
(1998)
D.
J.
Hill,
S.
and
more than
Jundi, and
Seventh
Roman
Theoretical
Proceedings
Annual
Archaeology
In
the
the
of
eye.
meets
Conference,Forcey, C., Hawthorne,J. and Witcher, R. (eds.) pp. 125 - 137. Oxford: Oxbow
Books.
Keevill, G. and Booth, P. (1997) Settlement,sequenceand structure: Romano-British stone-built
Farm,
Stanwich
(Northants),
Redlands
(Oxon).
In
From
Alchester
at
and
roundhouses
Round House to Villa, Friendship-Taylor, R. M. and Friendship-Taylor, D. E. (eds.), pp. 1946. The Upper Nene Archaeological Society.
King, A. C. (1984) Animal bonesand the dietary identity of military and civilian groupsin Roman
Britain, Germanyand Gaul. In Military and Civilian in RomanBritain, Blagg, T. F. C. and
King, A. C. (eds.) pp. 187 - 217. Oxford: British Archaeological Reports, British Series 136.
285
Economies:
In
Villa
bones.
Economic Aspectsof RomanoVillas
C.
(1988)
A.
King,
and animal
British Villas, Branigan, K. and Miles, D. (eds.) pp. 51 - 60. Sheffield: Sheffield University
Press.
King, A. C. (1991) Food production and consumption- meat. In Roman Britain: Recent Trends,
Jones,R. F. J. (ed.) pp. 15 - 20. Sheffield: J. R. Collis Publications.
King, A. C. (1999) Diet in a Roman world: a regional inter-site comparisonof the mammal bones.
Journal of RomanArchaeology 12: 168- 202.
frontier:
Britons
Romans
(1995)
the
B.
a theoretical evaluation of the
northern
Kurchin,
on
and
Archaeology
Second
Conference
Roman
In
Theoretical
archaeologyof resistance.
Proceedings,Rush, P. (ed.) pp. 124- 132.Aldershot: Avebury Press.
in
Upper
in
Iron
Age
Pitfalls
the
G.
(1984)
pottery
studies
Lambrick,
and possibilities
- experiences
ThamesValley. In Aspectsof the Iron Age in Central SouthernBritain, Cunliffe, B. W. and
Miles, D. (eds.) pp. 162 - 177. Oxford: University of Oxford Committee Monograph 2.
627.
Valley.
Current
121:
in
Thames
Archaeology
burial
Ritual
G.
(1990)
the
Lambrick,
and
farming
Thames
Roman
late
development
The
the
G.
(1992)
on
Lambrick,
prehistoric and
of
E.
Nichols,
Fulford,
M.
Gravels:
Review,
British
A
The
Archaeology
In
the
and
of
gravels.
(eds.) pp. 78 - 105. London: The Society of Antiquaries of London, Occasionalpapersvol.
14.
Farmoor,
Settlements
Riverside
Roman
Age
Iron
(1979)
Robinson
M.
G.
at
and
Lambrick,
and
Oxfordshire. Oxford: Oxford ArchaeologicalUnit Report 2 and Council for British
Archaeology ResearchReport 32.
in
development
floodplain
Upper
The
(1988)
M.
Robinson
the
G.
grassland
of
Lambrick,
and
Thames Valley, In Archaeology and the Flora of the British Isles: Human influence on the
Oxford
University
Oxford:
M.
55
Jones
)
75.
(ed.
pp.
evolution ofplant communities,
Committee for Archaeology Monograph Number 14.
Lange, P. (1996) The Animal Bones, in G. Hey (ed.) Iron Age and Roman Settlementat Old
Shifford Farm, Standlake.Oxonensia60: 148-158.
Lawrence,D. L. and Low, S. M. (1990) The built environment and spatial form. Annual Reviewof
Anthropology 1990 19: 453 - 505.
Leech, R. (1982) Excavations at Catsgore 1970-1973:A Romano-British Village. Bristol: Western
Archaeological Trust Excavation MonographNo. 2.
Limbrey, S. and Robinson, S. (1988) Dry land to wet land: soil resourcesin the Upper Thames
Valley. In The Exploitation of Wetlands,Murphy, P. and French, C. (eds.) pp. 129- 144.
Oxford: British Archaeology Reports,British Series 186.
286
Lyman, R. E. (1987) Archaeofaunasand butchery studies: a taphonomic perspective.,Advancesin
ArchaeologicalMethodand Theory10:249- 337.

Lyons, D. (1996) The politics of houseshape:round vs. rectangulardomestic structuresin Dela
Compounds,Northern Cameroon,Antiquity 70: 351 - 67.
Maltby, M. (1979) TheAnimal Bonesfrom Exeter 1971-1975.Sheffield: The University of
Sheffield Departmentof Prehistoryand Archaeology.
Maltby, M. (1981) Iron Age, Romano-British and Anglo-Saxon animal husbandry -a review of the
faunal evidence. In The Environment of Man: The Iron Age to Anglo-Saxon period, Jones,
M. and Dimbleby, G. (eds.) pp. 155 - 203. Oxford: British Archaeological Reports, British
Series 87.
Maltby, J. M. (1985a) Patternsin faunal assemblagevariability. In Beyond Domestication in

Prehistoric Europe, Barker, G. and Gamble, C. S. (eds.) pp. 33 - 74. New York: Academic
Press.
Maltby, J. M. (1985b) Assessingvariation in Iron Age and Roman butchery practices:the need for
Investigations:
Research
Design,
Data
In
Palaeobiological
Methods
and
quantification.
Analysis, Fieller, N. R. J. et al (eds.) pp. 19 - 30. Oxford: British Archaeological Reports,
International Series266.
Maltby, J. M. (1989) Urban-rural variations in the butchering of cattle in Romano-British
Hampshire. In Diet and Crafts in Towns. The Evidenceof Animal Remainsfrom the Roman
to the Medieval Periods, Serjeantson,D. and Waldron, T. (eds.) pp. 75 - 106. Oxford:
British Archaeological Reports,British Series 199.
Manning, W. H. (1983) The cauldron chains of Iron Age and Roman Britain. In Romeand Her
Northern Provinces, Hartley, B. and Wacher, J. (eds.) pp. 132- 154. Gloucester:Alan
Sutton.
McCormick, F. (1992) Early faunal evidencefor dairying. Oxford Journal ofArchaeology 11 (2):
201 - 209.
McWhirr, A. (1984) The cities and large rural settlementsof Roman Gloucestershire,In
Archaeology in Gloucestershire:From the Earliest Hunters to the Industrial Age, Saville,
A. (ed.) pp. 212-222.CheltenhamArt Gallery and Museums and the Bristol and
GloucestershireArchaeological Society, Cheltenham.
Mattingly, D. J. (1996) From one colonialism to another: imperialism and the Magreb. In Roman
Imperialism: Post-Colonial Perspectives,Webster,J. and Cooper, N. (eds.) pp. 49 - 69.
Leicester: Leicester Archaeological MonographsNo. 3.
Mattingly, D. J. (ed) (1997a) Dialogues in RomanImperialism: Power, discourse, and discrepant
Empire.
Rhode
Roman
Island:
Journal
Archaeology,
in
Roman
the
of
experience
SupplementarySeries23.
287
Mattingly, D. J. (1997b) Introduction: dialoguesof power and experiencein the Roman Empire. In
Dialogues in RomanImperialism: Power, discourse,and discrepant experiencein the
Roman Empire, Mattingly, D. J. (ed.) pp. 6- 24. Rhode Island: Journal of Roman
Archaeology, Supplementary Series 23.
diet,
identity
In
Proceedings
Romanisation.
You
(1994)
K.
I.
and
of the
Meadows,
are what you eat:
Fourth Annual Theoretical RomanArchaeology Conference,Cottam, S., Dungworth, D.,
Scott, S. and Taylor, J. (eds.) pp. 133 - 140. Oxford: Oxbow Books.
drinking
in
Nothing:
The
About
Much
Ado
I.
(1997)
K.
social
context
of
eating
and
Meadows,
Way
Life:
Recent
Approaches
Not
Much
Pot
More
So
In
Britain.
Roman
to
a
of
a
early
Artefact Studies,Cumberpatch,C. and Blinkhorn, P. (eds.) pp.21 - 35. Oxford: Oxbow
Books.
Meadows, K. I. (1999) The appetitesof householdsin early Roman Britain. In TheArchaeology of
HouseholdActivities, Allison, P. M. (ed.) pp. 101 - 120. London: Routledge
from
Middle
France
England
Eating
Taste
in
Food:
Manners
All
the
S.
(1985)
and
Mennell,
and
of
Ages to the Present. Basil Blackwell, Oxford.
Messer,E. (1984) Anthropological perspectiveson diet. Annual ReviewofAnthropology 13: 205 49.
In
from
Upper
Thames
in
Confusion
the
(1982)
the
D.
region.
Miles,
countryside: somecomments
TheRomano-British Countryside: Studiesin Rural Settlementand Economy, Miles, D. (ed.)
103(i).
British
Series
Reports,
British
Archaeological
Oxford.
79.
53
pp. Claydon
landscapes:
Pike
integrated
An
the
(1983)
to
the
D.
study of ancient
Miles,
approach
G.
Maxwell,
S.
Reconnaissance
Archaeology,
(ed.
)
Aerial
Impact
The
In
on
pp.
of
project.
74 - 84. London: Council for British Archaeology ResearchReport No. 49.
Miles, D. (1984) Romano-British settlementin the GloucestershireThamesValley. In Archaeology
in Gloucestershire.From the Earliest Hunters to the Industrial Age, Saville, A. (ed.) pp.
191- 211. Cheltenham:CheltenhamArt Gallery and Museumsand the British and
GloucestershireArchaeological Society.
Miles, D. (1986a) The Iron Age. In TheArchaeology of the Oxford Region, Briggs, G. et al. (eds.)
Oxford
University
Department
for
Oxford:
Studies.
External
57.
49
pp. Miles, D. (1986b) Archaeology at Barton Court Farm, Abingdon, Oxfordshire. Oxford
Miles, D. (1988) Villas and variety: aspectsof economyand society, in the Upper Thames
Landscape.In Villa Economies,Branigan, K and Miles, D. (eds.) pp. 60 - 71. Sheffield:
University of Sheffield Press.
288
Miles, D. (1989) The Romano-British Countryside,In Researchin Roman Britain 1960-1989,
Todd, M. (ed.) pp. 115 - 126.London: Society for the Promotion of Roman Studies
(Britannia Monograph 11).
Miles, D., Hofdahl, D. and Moore, J. (1986) The pottery. In Archaeology at Barton Court Farm,
Abingdon, Oxfordshire, Miles, D. (ed.), pp. microfiche: 7:A1-G5. Oxford Archaeological
Unit Report 3, Council British Archaeology ResearchReport No 50.
Miles, D. and S. Palmer (1983a) Claydon Pike. Current Archaeology 86: 88 - 92.
Miles, D. and S. Palmer (1983b) Claydon Pike, Fairford/Lechlade,Gloucester.Proceedingsof the
Prehistoric Society 49: 385 - 388.
Miller, D. (1985) Artefacts as Categories:A Study of Ceramic Variability in Central India.
Cambridge: Cambridge University Press.
Miller, D. (1995) Consumption as the vanguardof history: a polemic by way of an introduction. In
Acknowledging Consumption:A Reviewof New Studies,Miller, D. (ed.), pp. 1- 57.
London: Routledge.
Millett, M. (1979) An approachto the functional interpretation of pottery. In Pottery and the
Archaeologist, Millett, M. (ed) pp. 35 - 48. London: Institute of Archaeology, Occasional
Publication no. 4.
Millett, M (1984) Forts and the origins of towns. In Military and Civilian in Roman Britain.
Cultural Relationships in a Frontier Province, Blagg, T. F. C. and King, A. C. (eds.) pp. 65
Reports,British Series 136.
Archaeological
British
Oxford:
74
Millett, M. (1990a) The Romanizationof Britain. Cambridge:CambridgeUniversity Press.
Millett, M. (1990b) Romanization: historical issuesand archaeologicalinterpretation. In The Early
Roman Empire in the West,Blagg, T. and Millett, M. (eds.) pp.35-41. Oxford: Oxbow
Books.
Millett, M. and Graham, D. (1986) Excavationson the Romano-British small town at Neatham,
Hampshire, 1969 - 1979. Gloucester:HampshireField Club and Archaeological Society:
Monograph 3.
Mintz, S. W. (1996) Tasting Food, Tasting Freedom: Excursions into Eating, Culture and the Past.
Boston: Beacon Press.
Monaghan,J. (1995) Roman pottery researchfor the 1990's. In Theoretical RomanArchaeology SecondConferenceProceedings,Rush, P. (ed.) pp. 148 - 157. Aldershot: Avebury Press.
Moore, H. (1981) Bone refuse - possibilities for the future. In EconomicArchaeology: Towards an
Integration of Ecological and Social Approaches,Sheridan,A. and Bailey, G. (eds.) pp. 87 94. Oxford: British Archaeological Reports,International Series96.
289
Moore, H. (1987) Problems in the analysisof social change:an example from the Marakwet. In
Archaeology as Long-Term History, Hodder, I. (ed.) pp. 85 - 104. Cambridge: Cambridge
University Press.
Moore, H. (1988) Feminism and Anthropology. Cambridge:Polity Press.
Moore-Colyer, R. J. (1994) The horse in British pre-history: some speculations.Archaeological
Journal 151: 1- 15.
Moreland, J. F. (1992) Restoring the dialectic: settlementpatternsand documentsin medieval
Ethnohistory,
Knapp,
B.
(ed)
112
Italy.
Archaeology,
Annales
A.
129.
In
and
pp.
central
CambridgeUniversity Press,Cambridge.
Morris, E. (1996) Artefact product and exchangein the British Iron Age. In The Iron Age in Britain
Collis,
T.
J.
R.
(eds.
)
65.
Sheffield:
Champion,
C.
Ireland:
Trends,
41
Recent
and
pp.
and
J. R. Collis Publications.
Mytum, H (1989) The recognition and interpretation of intra-site patterning on sites with small
finds:
Walesland
Scottish
6:
Rath.
Review
65
Archaeological
the
of
of
example
numbers
74.
Needham,S. P. and Sorensen,M. L. S. (1988) Runnymederefuse tip: a considerationof midden
depositsand their formation. In TheArchaeology of Context in the Neolithic and Bronze
Age: RecentTrends, Barrett, J.C. and Kinnes, I. (eds.) pp. 113 - 126. Sheffield: J.R. Collis
Publications.
Oetgen,J. M. (1984) The absorption of foodstuffs by ceramics.Bulletin of the Experimental Firing
Group 2: 41 - 55.
Okun, M. L. (1989) The Early RomanFrontier in the Upper Rhine Area. Oxford: British
Archaeological Reports International Series 547.
Orton, C. (1978) Is pottery a sample?In Sampling in ContemporaryBritish Archaeology, Cherry, J.
F., Gamble, C. and Shennan,S. (eds.) pp. 321 - 353. Oxford: British Archaeological
Reports,British Series50.
Orton, C. and Tyers, P. (1991) Slicing the pie -a framework for comparing ceramic assemblages.
Journal of Roman Pottery Studies4: 79 - 80.
Orton, C., Tyers, P. and Vince, A. (1993) Pottery in archaeology. Cambridge: Cambridge
University Press.
Parker,A. J. (1988) The birds of Roman Britain. Oxford Journal ofArchaeology 7(2): 197- 226.
Parker Pearson,M. (1996) Food, fertility and front doors in the first millennium BC. In The Iron
Age - RecentTrends. Collis, J. (ed.) pp. 117-132.Sheffield: Sheffield University Press.
Parker Pearson,M. and Richards, C. (eds.) (1994) Architecture and Order. London: Routledge.
290
Parker Pearson,M., Sharples,N. and Mulville, J. (1996) Brochs and Iron Age society: a reappraisal.
Antiquity 70: 57 - 67.
Partridge, C. R. (1981) Skeletal Green: a late Iron Age and Romano-British Site. London: Britannia
Monograph SeriesNo. 2.
Payne, S. (1972) On the interpretation of bone samplesfrom archaeologicalsites. In Papers in
Economic Prehistory, E. S. Higgs (ed.) pp. 65 - 81. Cambridge: Cambridge University
Press.
Pearce,J. and Luff, R. (1994) The taphonomy of cookedbone. In Whither Environmental
Archaeology? Luff, R. and Rowley-Conwy, P. (eds.) pp. 51 - 56. Oxford: Oxbow Books.
Perring, D. (1989) Cellars and Cults in Roman Britain. Archaeological Journal 146: 279 - 301.
Pluciennik, M. Z, (1997) Historical, geographicaland anthropological imaginations: early ceramics
in southernItaly. In Not so much a pot, more a way of life, Cumberpatch,C. and Blinkhorn,
R. W. (eds.) pp. 37 - 56. Oxford: Oxbow Monograph 83.
Price, C. R. (1985) Patternsof cultural behaviour and intra-site distribution of faunal remainsat the
Widow Harris site. Historical Archaeology 19 (2): 40 - 56.
Pryor, F. (1983) Gone, But Still Respected:Some Evidencefor Iron Age House Platforms in
Lowland England. Oxford Journal of Archaeology 2 (2): 189 - 198.
Rackham,J. (1983) Faunal samplesto subsistenceeconomy:some problems in reconstruction. In
Integrating the SubsistenceEconomy,Jones.M. (ed.) pp. 251 - 277. Oxford: British
Archaeological Reports, International SeriesS181.
Rackham,J (1987) Practicality and realism in archaeologicalanalysis and interpretation, in
Gaffney, C. F. and Gaffney, V. L. (eds.) pp. 47 - 69. Pragmatic Archaeology: Theory in
Crisis? pp. 83 - 88. Oxford: British Archaeological Reports, British Series 167.
Rackham,J. (1995) Envirorunent archaeologyas material culture. ScottishArchaeology Review 9
25.
22
10:
and
Rapoport,A. (1990) Systemsof activities and systemsof settings. In DomesticArchitecture and the
Use of Space,Kent, S. (ed.) pp. 9- 20. Cambridge:Cambridge University Press.
Raven,S. (1990) The Romano-British pottery. In An Iron Age and Romano-British Enclosed
Settlementat WatkinsFarm, Northmoor, Oxon.,Allen, T. G. (ed.) pp. 46 - 51. Thames
Valley Landscapes:the Windrush Valley 1, Oxford: The Oxford University Committee for
Archaeology.
Reece,R. (1979) Romano-British interaction. In Space,Hierarchy and Society, Burnham, B.C. and
Kingbury, J. (eds.) pp. 229 - 240. Oxford: British Archaeological Reports, International
SeriesS59.
291
Reece,R. (1988) My Roman Britain. Cirencester:Cotswold Studies at the Apple Loft.
Reece,R. (1990a) Review of Researchon RomanBritain, Todd, M. (ed.) TheArchaeological
Journal 147: 445 - 447
Reece, R. (1990b) Romanization: a point of view. In The Early Roman Empire in the West, Blagg,
T. and Millett, M. (eds.) pp. 30 - 34. Oxford: Oxbow Books.
Reynolds, P. J. (1982) Substructureto superstructure.In Structural Reconstruction: Approachesto

the Interpretation of the ExcavatedRemainsof Buildings, Drury, P. J. (ed.) pp. 173 - 198.
Oxford: British Archaeological Reports,British Series 110.
Reynolds, P. J. (1995a) Rural life and farming. In The Celtic World, Green, M. J. (ed.) pp. 176206. London and New York: Routledge.
Reynolds, P. J. (1995b) The food of the prehistoric Celts, In Food in Antiquity, J. Wilkins, D.
Harvey and M. Dobson (eds.) pp. 303 - 315. Exeter: Exeter University Press.
Revel, J. F. (1982) Culture and Cuisine, a Journey Through the History of Food, Lane, H. R.
(trans). New York: De CapoPress.
Rice, P. (1987) Pottery Analysis: A SourceBook. Chicago:University of Chicago Press.
Rippengal, R. (1991) Review of The Romanizationof Britain by Millet, M. Archaeological Review
from Cambridge 10: 2,220 - 222.
Rippengal, R. (1993) Villas as a key to social structure:somecomments on recent approachesto the
Romano-British villa and somesuggestionstowards an alternative. In Theoretical Roman
Archaeology: First ConferenceProceedings,Scott, E. (ed.) pp. 79-101. Aldershot: Avebury
Press.
Rippengal, R. (1995) Romanization, society and material culture: an archaeologicalstudy of the
in
Romanization
Unpublished
Roman
Britain.
Ph.
D.
the
thesis,
context
of
of
significance
University of Cambridge.
Robinson, M. (1981) The Iron Age to early Saxon environment of the Upper Thames Valley. In The
Environment of Man: the Iron Age to Anglo-SaxonPeriod, Jones,M. and Dimbleby, G.
(eds.) pp. 251 - 286. Oxford: British Archaeological Reports, British Series 87.
Robinson, M. (1983) Arable / pastoralratios from insects.In Integrating the SubsistenceEconomy,
Jones,M. (ed.) pp. 19 - 55. Oxford. British Archaeological Reports, International Series
S181.
Robinson, M. (1992) Environmental archaeologyof the river gravels: past achievementsand future
directions. In Developing Landscapesof Lowland Britain. TheArchaeology of the British
Gravels: A Review, Fulford, M. and Nichols, E. (eds.) pp. 47 - 62. The Society of
Antiquaries of London, OccasionalpapersVol. 14, London.
292
Robinson, M. (1994) Excavation and environmentalarchaeologyat Mingies Ditch. In Whither
EnvironmentalArchaeology?, Luff, R. and Rowley-Conway, P. (eds.) pp. 105 - 112.
Robinson, M. (1996) The plant and invertebrateremains.In Iron Age and RomanSettlementat Old
Shffford Farm, StandlakeHey, G. (ed.) Iron Age and Roman Settlementat Old Shifford
Farm, Standlake.Oxonensia60: 158 - 166.
Roe, F. and Glass,H. (1996) StoneObjects. In Iron Age and RomanSettlementat Old Shifford
Farm, Standlake,Hey, G. (ed.) Oxonensia60: 145 - 148.
Roymans,N. The north Belgian tribes in the is` century BC. In Romanand Native in the Low
Countries, Brandt, R. and Slofstra, J. (eds.) pp.71 - 103. Oxford: British Archaeological
Reports,International Series 184.
Rush, P. (1997) Symbols, Pottery and Trade, in Proceedingsof the Sixth Annual Theoretical Ronan
Archaeology Conference,Meadows, K. I., Lemke C. and Heron, J. (eds.) pp. 55 - 64.
Samson,R. (ed.) (1990) TheSocial Archaeology of Houses.Edinburgh: Edinburgh University
Press.
Sanders,J (1979) Finds: The Romanpottery. In Iron Age and RomanRiverside Settlementsat
Farmoor, Oxfordshire, Lambrick, G. and M. Robinson (eds.) pp. 46 - 54. Oxford: Oxford
Archaeological Unit Report 2 and Council for British Archaeology ResearchReport 32.
SchusterKeswani, P. (1994) The social context of animal husbandryin early agricultural societies:
from
Journal
insights
Cyprus.
of
and
an
archaeological
example
ethnographic
Anthropological Archaeologyl 3: 255 - 277.
Scott, E. (1990) Romano-British villas and the social construction of space.In The Social
Archaeology of Houses,in Samson,R. (ed.) pp. 149 - 172. Edinburgh: Edinburgh University
Press.
Scott, E. (1991) Animal and infant burials in Romano-British villas: a revitalization movement. In
Sacredand Profane, Garwood, P., Jennings,D., Skeates,R. and Toms, J. (eds.) pp. 115 121. Oxford: Oxford University Committee for Archaeology Monograph No 32 and Oxbow
Books.
Scott, E. (1993). Writing the Roman Empire. In Theoretical RomanArchaeology: First Conference
Proceedings,Scott, E. (ed.) pp. 5- 22. Aldershot: Avebury.
Scott, E. and V. L. Gaffney (1987) Romano-British villas: practical lessonsfor tactical fieldwork. In
Pragmatic Archaeology: Theory in Crisis? Gaffney, C. F. and Gaffney, V. L. (eds.) pp. 83 88. Oxford: British Archaeological Reports, British Series 167.
293
Scott, S. (1994) Patternsof movement:architectural designand visual planning in the RomanoBritish villa. In Meaningful Architecture: Social Interpretation of Buildings, Locock, M.
(ed.) pp. 86 - 98. Aldershot: Avebury Press.
Sellwood, L. (1984) Tribal boundaries viewed from the perspective of numismatic evidence. In
Aspects of the Iron Age in Central Southern Britain, Cunliffe, B. W. and Miles, D. (eds.) pp.
191 - 204. Oxford: University of Oxford Committee Monograph 2.
Shanks,M. and Tilley, C. (1987) Re-constructingArchaeology. Cambridge: Cambridge University

Press.
Sharples,N. M. (1990) Late Iron Age society and continentaltrade in Dorset. RevueArcheologie
QuestSupplement3: 299 - 304.
Sherratt, A. G. (1987) Cups that cheered.In Bell Beakersof the WesternMediterranean, Waldren,
W. H. and Kennard, R. C. (eds.) pp. 81 - 114. Oxford: British Archaeological Reports,
International Series 331(i).
Sherrat,A. (1995) Alcohol and its alternatives.Symbol and substancein pre-industrial cultures. In
ConsumingHabits, Goodman,J., Lovejoy, P. E. and Sherrat, A. (eds.) pp. 11 - 46. London:
Routledge.
Sillar, B. (1997) Reputablepots and disreputablepotters: individual and community choice in
present-daypottery production and exchangein the Andes. In Not so much a pot, more a
life,
C.
Blinkhorn,
Cumberpatch,
R. W. (eds.) pp. 1- 20. Oxford: Oxbow
and
way of
Monograph 83.
Skibo, J. M. (1992) Pottery Function: A Use-AlterationPerspective.London and New York:
Plenium Press.
Slofstra, J. (1983) An anthropological approachto the study of Romanization processes.In Roman
Countries,
Low
Brandt, R. and Slofstra, J. (eds.) pp. 71 - 103. Oxford:
Native
in
the
and
British Archaeological Reports,International Series184.
Smith, J. T. (1978) Villas as a key to social structure.In Studiesin the Romano-British Villa, Todd,
M. (ed) pp. 149 - 173. Leicester:LeicesterUniversity Press.
Smith, M. E. (1987) Household possessionsand wealth in agrarian states:implications for
Journal
ofAnthropological Archaeology 6: 297 - 335.
archaeology.
Smith, M. E. (1992) Rhythms of changein postclassiccentral Mexico: archaeology,ethnohistory,
In
Archaeology,
Braudelian
Annales, and Ethnohistory, Knapp, A. B. (ed.)
the
model.
and
Cambridge
Cambridge:
University
74.
51
Press.
pp. Spain, R. J. (1986) Millstones from Barton Court Farm. In Archaeology at Barton Court Farm,
Abingdon, Oxfordshire, Miles, D. (ed.) pp. microfiche 5:A13-B6. Oxford: Oxford
294
Stead, I. M. and Rigby, V. (1986) Baldock: The Excavation of a Roman and pre-Roman Settlement,
1968 - 1972. London: Britannia Monograph SeriesNo. 7.
Tambiah, S. (1969) Animals are goodto think with and good to prohibit. Ethnology 8: 423 - 459.
Thawley, C. (1982) The animal bones.In Early Roman Occupation at Cirencester, Wacher, J. and
McWhirr, A (eds.) pp. 211-227. Cirencester:CirencesterExcavation Committee,
Timbey, J. R. (1996) The pottery. In Iron Age and RomanSettlementat Old Shifford Farm,
Standlake,Hey, G. (ed.) pp. 125- 143. Oxonensia60: 93 - 175.
Tierney, J. J. (1959-60) The Celtic ethnographyof Posidonius,Proceedingsof the Royal Irish
Academy60: 189 - 275.
Todd, M (1985) Oppida and the Romanarmy: a review of recent evidence.Oxford Journal of
Archaeology 4 (2): 187-199.
Tringham, R. E. (1991) Householdswith faces:the challengeof genderin prehistoric architectural
remains.In EngenderingArchaeology: Womenand Prehistory, Gero, J. M. and Conkey, M.
W. (eds.) pp. 93 - 131.Oxford: Basil Blackwell.
Trow, S. D. (1990) By the northern shoresof Ocean:someobservationson acculturation processat
the edgeof the Romanworld. In The Early RomanEmpire in the West,Blagg, T. and
Millett, M. (eds.) pp. 103 - 118. Oxford: Oxbow Books.
Turmo, I. G. (1997) The pathwaysof taste:the West Andalucian case.In Food Preferencesand
Taste: Continuity and Change,Macbeth, H. (ed.) pp. 115 - 126. Providence,USA:
BerghahnBooks.
Turner, D.R. (2001) Ruminations on Romanisationin the east: or, the metanarrativein History.
htmi.
Assemblage4: www. shef.ac.uk/assem/t/turner.
Tyers, P. (1996) RomanPottery in Britain. London: B. T. Batsford Ltd.
Uerpmann, H. P. (1973) Animal bone finds and economicarchaeology:a critical study of osteoWorld
Archaeology 4: 307 - 322.
method.
archaeological
(1989)
Charred
Veen,
M.
from
der
Roman-period
grain
assembles
corn driers in Britain.
van
Archaeological Journal 146: 302 - 319.
Consumption
Veen,
M.
(1991)
der
or production? Agriculture in the CambridgeshireFens?In
van
New Light on Early Farming. RecentDevelopmentsin Paleoethnobotany,Renfrew, J. (ed.)
Edinburgh:
Edinburgh
361.
University
Press.
349
pp.
Van Es, W.A. (1983) Introduction. In Romanand Native in the Low Countries: Spheresof
Interaction, Brandt, R. and Slofstra, J. (eds.) pp. 1- 11. Oxford: British Archaeological
Reports,International Series 184.
295
Vencl, S. (1994) The archaeologyof Thirst. Journal of EuropeanArchaeology 2 (2): 299 - 326.
Veyne, P. (1987) The Roman Empire. In A History of Private Life, Veyne, P. (ed.) pp. 5- 207.
Cambridge,Mass. and London: Harvard University Press.
Wacher, J. (1974) The Townsof RomanBritain. London: B. T. Batsford.
Wacher, J. and McWhirr, A. (1982) CirencesterExcavationsI: Early Roman Occupation at
Cirencester.Cirencester:CirencesterExcavation Committee.
Wait, G. A. 1985.Ritual and Religion in Iron Age Britain. British Archaeological Reports, British
SeriesNo. 149.
Webster, J. (1996) Roman imperialism and the `post imperial age'. In Roman Imperialism: PostColonial Perspectives, Webster, J. and Cooper, N. (eds.) pp. 1- 17. Leicester: Leicester
Archaeological Monographs No. 3.
Webster, J. (1997) A negotiatedsyncretism:readingson the developmentof Romano-Celtic

Religion. In Dialogues in RomanImperialism: Power, discourse,and discrepant experience
in the Roman Empire, Mattingly, D. J. (ed.) pp. 165 - 184. Rhode Island: Journal of Roman
Archaeology, SupplementarySeries23.
Webster, J. (2001) Creolizing the Roman provinces.American Journal ofArchaeology 105: 1- 17.
Webster, J. and Cooper, N. J. (eds.) (1996) RomanImperialism: Post-Colonial Perspectives.
Leicester:Leicester Archaeology MonographsNo 3.
Wilk, R. R. (1990) The Built Environment and ConsumerDecision, In DomesticArchitecture and
the Use of Space,Kent, S. (ed.) pp. 34 - 42. Cambridge:Cambridge University Press.
Williams, D. F. and Peacock,D. P. S. (1983) Importation of olive-oil into Iron Age and Roman
Britain. In Produccion y Comercio del Aceite en la Antiquedad, Blazquez Martinez, J. M.
and Rodsiguez,J. R. (eds.) pp. 26 3- 280. Madrid: Universidad Complutense.
Willis, S. (1994) Roman imports into late Iron Age British societies:towards a critique of existing
models. In Proceedingsof the Fourth Annual Theoretical RomanArchaeology Conference,
Cottam, S., Dungworth, D., Scott, S. and Taylor, J. (eds.) pp. 141 150. Oxford: Oxbow
Books.
Willis, S. (1996) The Romanization of pottery assemblagesin the eastand north-eastof England
during the 1S`century AD: a comparative analysis.Britannia 27: 179 221.
Willis, S. (1997) Saurian:beyond dating, In Proceedingsof the Sixth Annual Theoretical Roman
Archaeology Conference,K. Meadows, C. Lemke and J. Heron (eds.) pp. 38 - 54. Oxford:
Oxbow Books.
Wilson, B. (1978) The Animal Bones.In The excavationof an Iron Age Settlement,BronzeAge
Ring-Ditches and RomanFeatures at Ashville trading Estate,Abingdon, Oxfordshire,
296
Parrington, M. (ed.) pp. 110 - 139.Oxford: Oxfordshire Archaeological Unit and the
Council for British Archaeology.
Wilson, B. (1979) The vertebrates.In Iron Age and Romanriverside settlementsat Farmoor,
Oxfordshire, Lambrick, G. and Robinson,M. (eds.) pp. 128 - 133. Oxford: Oxford
Archaeological Unit Report 2 and Council for British Archaeology ResearchReport 32.
Wilson, B. (1985) Degradedbones,featuretype and spatial patterning on an Iron Age occupation
Methods
Investigations.
Designs,
Research
England.
In
Paleobiological
in
Oxfordshire,
site
81
N.
G.
(eds.
)
Gilbertson,
D.
D.
Ralph,
N.
J.,
A.
R.
Fieller,
Data
Analysis,
and
pp.
and
94. Oxford: British ArchaeologicalReports,International Series266.
Court
Farm,
Barton
in
D.
Miles
(ed.
)
Archaeology
Bones,
The
Animal
(1986)
B
Wilson,
at
Abingdon, Oxfordshire. Oxford Archaeological Unit Report 3, Council British Archaeology
Research Report No 50.
Wilson, B. (1989) Fresh and old table refuse.The recognition and location of domestic activity at
3:
237-260.
England.
Valley,
Archaeozoologia,
in
Upper
Thames
the
archaeologicalsites
Wilson, B. and Allison, E. (1990) The animal and fish bones.In An Iron Age and Romano-British
EnclosedSettlementat WatkinsFarm, Northmoor, Oxon, Allen, T. G. (ed.) ThamesValley
Landscapes:the Windrush Valley 1. Oxford: The Oxford University Committee for
Archaeology.
Wilson, B. (1992) Considerationsfor the identification of ritual depositsof animal bones in Iron
Age pits. International Journal of Osteoarchaeology,2: 341 - 349.
Wilson, B. (1994) Projects modelling the spatial patterningof bones: limited successin publication.
In Whither Environmental Archaeology? Luff, R. and Rowley-Conwy, P. (eds.) pp.57 - 66.
Wilson, B. (1995) On the curious distortions behind the chargeof scientism againstenvironmental
67.
Archaeological
Review
9
10:
66
Scottish
and
archaeology.
Wilson, B. (1996a) Aspects of the literature on the theory and identification of ritual. In Ritual
Treatmentof Human and Animal Remains,Anderson, S. and Boyle, K. (eds.) pp. 11 - 15.
Oxford: Oxbow Books for the OsteoarchaeologicalResearchGroup.
Wilson, B. (1996b) Spatial Patterning amongAnimal Bonesin SettlementArchaeology: An English
Regional Exploration. Oxford: British Archaeological Reports, British Series251.
Wilson, D.C. (1994) Identification and assessmentof secondaryrefuse aggregates.Journal of
Archaeological Method and Theory 1(1): 41 - 68.
Woods, A. J. (1984) The old pot-boiler. Bulletin of the Experimental Firing Group 2: 25 - 41.
Woods, A. J. (1986) Form, Fabric, and Function: Someobservationson the cooking pot in
Antiquity. In Technologyand Style, Ceramics and Civilization Vol. 2, Kingery, W.D. (ed.)
Society.
Ceramics
The
72.
American
157
pp.
-
297
Woodward, A. and Blinkhorn, P. (1997) Size is important: Iron Age vessel capacitiesin central and
southern England. In Not so much a pot, more a way of life, Cumberpatch,C. and
Blinkhorn, R. W. (eds.) pp. 153-162.Oxford: Oxbow Monograph 83.
Woolf, G. (1988) MediterraneanAmphorae in TemperateEurope. Paperpresentedat the
Cambridge Iron Age Symposium.
Woolf, G. (1993) The social significanceof trade in late Iron Age Europe. In Trade and Exchange
in Prehistoric Europe, Scarre,S. and Healy, F. (eds.) pp. 211 - 218. Oxford: Oxbow Books.
Woolf, G. (1995) Beyond Romansand natives. l Vorld Archaeology 28 (3): 339 - 350.
Woolf, G. (1998) BecomingRoman: The Origins of Provincial Civilization in Gaul. Cambridge:
Cambridge University Press.
Yanagisako, S. J. (1979) Family and household:the analysisof domestic groups.Annual Review of
Anthropology 8: 161 - 205.
Young, C. (1986) The Upper ThamesValley in the Roman period. In The Archaeology of the
Oxford Region, Briggs, G. et al (eds.) pp. 58 - 63. Oxford: Oxford University Department
for External Studies.

Eating-Drinking Roman Britain

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Eating-Drinking Roman Britain

Uploaded by

Copyright:

Available Formats

THE SOCIAL CONTEXT OF EATING AND DRINKING

IN EARLY ROMAN BRITAIN

A thesis submitted for the degree of Doctor of Philosophy

An Outline of the Thesis

Roman style goods and constructs at `non-Roman' settlements

A caveatto the study of the social contextsof eating and drinking

Natives and Romans? in the Upper Thames Valley

Early Romanperiod householdsin the Upper ThamesValley

Barton Court Farm

Site reports, microfiche and archives

The containersat the late Iron Age settlement

Food and Drink Consumption at the Roman Period Settlement

Other implementsat the early Romanperiod settlement

Summary of the Artefacts and Remains of Consumption for each Settlement...

The early Romanperiod settlement

The Archaeology of Food and Drink Consumption at Roughground

The containersat the early Roman period settlement

The ingredientsat the early Roman period settlement

Summary of the Artefacts and Remains of Consumption at Roughground Farm

Pre-villa and environs

The containersat the late Iron Age settlement

Other implements at the early Roman period settlement

The early Romanperiod settlement

The early Romanperiod settlement

The Social Contexts of Imperialism

at Old Shifford Farm

Figures 7.1 - 7.13

The Archaeology of Food and Drink Consumption at Claydon Pike

The containers at the late Iron Age settlement

The other implementsat the late Iron Age settlement

Food and Drink Consumption at the Early Roman Period Settlement

Other implementsat the early Romanperiod settlement

Summary of the Artefacts and Remains of Consumption at each Settlement

The early Romanperiod settlement

The Distribution of the Remains of Eating and Drinking at Claydon Pike

Figures 8.1 - 8.22

Late Iron Age Old Shifford Farm

Early Romanperiod RoughgroundFarm

List of Figures and Tables

Figure 5.8. N. I. S.P. (Meadows)

Figure 5.11. Early Romanperiod pottery fabrics

Figure 5.25. Histogramsof early Roman periodjars and bowls by feature

Figure 7.5.LateIron Age- bothphases

Figure 8.3. Histogram of the rim diameterof late Iron Agejars

Figure 8.7. Late Iron Age animal bonegroupsaccordingto meatyield

Figure 8.13. Early Romanperiod animal bonegroupsaccordingto meatyield

Table 6.1. Early Romanperiodpotteryfabrics

Table62. Thedistributionof earlyRomanperiodpotteryfabrics

Table7.1. LateIronAge phaseoneandtwo- andearlyRomanperiodpotteryfabrics

The Social Context of Eating and

Karen Ingrid Meadows

are tied to the

I would like to thank the University of Sheffield for granting me a PostgraduateResearch

I am especially grateful for the encouragementand cooperationof various peopleat the

JaneTimby and Pricilla Langewho were very accommodatingand willingly answered

Andrew Sherratt at the AshmoleanMuseum generouslyallowed me free reign of the

For a variety of reasons,this thesis took longer than anticipatedto complete,and I am

native-type settlement (Chapter3 and 4)

The "group who usedand residedwithin a singlehouseor group of closely related

Late first centuryBC to the mid-first centuryAD

Mid-first centuryAD to the mid-secondcentury AD

Roman Britain is too important to be left to the Romanists!(Barry Cunliffe 1979)

To the memoryof Barbara Meadows