Scotland After the Ice Age: Environment, Archaeology and History 8000 BC - AD 1000 0748617361

Table of contents :
Contents
List of Figures
List of Plates
List of Tables
List of Authors
Preface
1 Environment and People in Prehistoric and Early Historical Times: Preliminary Considerations
2 Climate Change
3 Geomorphology and Landscape Change
4 Soils and Their Evolution
5 Vegetation Change
6 Fanual Change
7 The Mesolithic
8 The Neolithic
9 The Bronze Age
10 The Iron Age
11 The Roman Presence: Brief Interludes
12 The Early Historic Period: An Archaelogical Perspective
13 The Early Norse Period
14 Environment and Archaeology in Scotland: Some Observations
References
A Guide to the Literature since 1996
Index

Citation preview

3X

l

and History, Environment,Archaeology Environment,Archaeology and > ‘ , ._ 1000,__,“;,. H A D 1000 8000 8000 BC -—AD

Scotland After the Ice Age

After the Ice Age Scotland After Scotland History, Environment, Archaeology and History, Environment, 1000 8000 8000 BC — AD 1000

Edited by Kevin J. Edwards Edited and Ian B. M. Ralston and Ian

University Press Edinburgh University Edinburgh

organisation Kevin J. Edwards © © editorial matter and organisation and and Ian B. M. Ralston, 1997, 2003.

Archaeology, 8000 BC — AD 1000 and Archaeology, Scotland: Environment and

published in 1997 by John Wiley & Sons. first published was first Edinburgh University Press Ltd Edinburgh University 22 22 George Square, Edinburgh Transferred to digital print 2008 Eastbourne Rowe, Eastboume Printed and bound by CPI Antony Rowe,

Library available from the British Library book is available A CIP record for this book ISBN 0 7486 1736 1 (paperback)

Contents

List of Figures

vii

List of Plates

ix

List of Tables

xi

List of Authors List

xiii

Preface

xvii

Times: 1 Environment and People in Prehistoric and Early Historical Times: Preliminary Considerations Kevin J. Edwards and Ian B. M. Ralston Climate Change Climate

11 ll

Graeme Whittington and Kevin J. Edwards

Geomorphology and Landscape Change and Alastair Alastair G. Dawson Colin K. Ballantyne Ballantyne and Colin

23

Soils and Their Evolution Carter Donald A. Davidson and and Stephen P. Carter Donald

45

Vegetation Change Kevin J. Edwards and Graeme Whittington

63

Faunal Change Vertebrate Fauna: Finbar McCormick and Paul C. Buckland The Vertebrate

Land Snails: Stephen P. Carter Insects: Paul C. Buckland and Jon P. Sadler Insects:

83 104 105

The Mesolithic Bill Finlayson and Kevin J. Edwards

109

The Neolithic Neolithic The Barclay J . Barclay Gordon J.

127

The Bronze Age Trevor G. Cowie and Ian A. G. Shepherd

151

CONTENTS CONTENTS

vi

Iron Age 10 The Iron

169

Ralston Ian Armit and Ian B. M. Ralston The Roman Presence: Brief Interludes 11 The

195

William S. Hanson

Perspective Archaeological Perspective An Archaeological Period: An Historic Period: Early Historic The Early 12 The Ian Armit and Ian Ralston and M. Ralston B. M. Ian B. [an

217

The Early Norse Period 13 The John R. Hunter

241

14

Observations Environment and Archaeology in Scotland: Some Observations M. Ralston Ian B. M. and Ian Edwards and Kevin J. Edwards Kevin

255

References

267

A Guide to the Literature since 1996

311

Index

321

List of Figures

1.1 1.2 2.1 3.1 3.2 3.3 3.4 3.5 3.6 3.7 4.1 4.2 4.3 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 6.1 6.2 7.1 7.2 7.3 8.1 8.2

chronological guide showing archaeological and environmental showing archaeological A chronological subdivisions subdivisions government reorganization in The counties of Scotland prior to local government 1975 Locations through time Locations of the Polar Front through Scotland, showing the distribution of The structural provinces of Scotland, major rock types Limits of Late Devensian glaciation deposits in Scotland drift deposits distribution of drift Generalized distribution Scotland Generalized Quadratic trend surface maps for (a) the Main Lateglacial Shoreline Quadratic and (b) the Main Postglacial shoreline in Scotland relative sea-level curves Time—altitude graph depicting depicting relative Time—altitude Scotland during Main the Main during the east-central Scotland inundation of east-central Extent of inundation Extent Postglacial Transgression Braeroddach Loch deposition rate at Braeroddach Sediment deposition Sediment distribution of soils for a typical area in the Southern Schematic distribution Uplands Distribution of land capability classes for agriculture Map of archaeological sites with buried soils Woodland in Scotland c. 5000 BP (3780 cal BC) Isochrone maps of Corylus and Ulmus pollen for Scotland Location of sites mentioned in Chapter 5 diagram for Black Loch II, Fife Pollen diagram Mesolithic age pollen spectra from Loch an t-Sil, South Uist Fife Pollen diagram for Black Loch I, Fife BP (5270 period c. 6330 BP the period diagram for the present the present (5270 cal BC) to the Pollen diagram Barra from Lochan na Cartach, Barra Vord, Unst, Shetland Pollen diagram for Saxa Vord,

Lateglacial to mid Holocene sites referred to in Chapter 6 Lateglacial Late Holocene sites referred to in Chapter 6 Sites and key areas mentioned in Chapter 7 artefacts Mesolithic artefacts Mesolithic Radiocarbon dates for the the Mesolithic Radiocarbon mentioned in Chapter 8 Map showing sites mentioned Distribution Distribution map of certain and possible henges and small hengiform enclosures enclosures (less than 20 m in diameter); cursus monuments; Clava and recumbent cairns recumbent stone circles caims and

8.3 8.4

Pit-defined enclosures and cursus monuments monuments in Scotland Orkney buildings

10 12 24 28 31

34 38 40 43 50 51 55 65 66 68 69 71 77 78 80 84 85 111 114 118 130

136 137 145

viii 8.5 9.1 9.2 9.3 10.1 10.2 10.3 10.4

10.5 10.6 10.7 11.1 11.2 11.3 12.1 12.2 12.3 12.4 12.5 13.1 14.1

14.2

14.3

LIST OF FIGURES

Western Isles Structures Structures on the mainland and the Western Map showing sites mentioned in Chapter 9 archaeological landscape landscape Perthshire archaeological Balnabroich, Strathardle, Perthshire 9.6 Reconstruction Reconstruction of the roundhouse shown in Plate 9.6 Age Scottish Iron Iron Age The subdivision of the Scottish The conventional scheme for the subdivision Simplified plans of Scottish Iron Age sites Sites mentioned in Chapter 10 principal phases of enclosure Schematic representation of the principal East Broxmouth, East and Broxmouth, Roxburghshire and Hownam Rings, Roxburghshire represented at Hownam Lothian Distribution of Scottish hillforts and duns Plans, and access maps indicating differential complexity, of complex Atlantic Atlantic roundhouses and wheelhouses Landscape subdivisions in the cropmark record around Castlesteads Midlothian Newton in Midlothian and Newton Antonine period (c. The Roman occupation occupation of Scotland in the early Antonine AD 142—158) extent in the The Roman occupation of Scotland at its furthest extent Flavian period (c. AD 84—87) Hadrian’s Wall, third-century Roman forts and non-Roman sites mentioned in Chapter Chapter 11 Distribution Distribution of ‘pit’ names, Pictish symbol stones and Pictish silver chains Sites mentioned in Chapter 12 architectural styles styles Pictish buildings buildings in a variety of architectural Lothian Anglian buildings in East Lothian British and and Anglian British stones on symbol stones depicted on as depicted Pictland as The The fauna of Pictland Scotland General area of Norse influence in northern and western Scotland archaeological periods, periods, The number of excavations assigned to archaeological The number excavations featuring showing the proportions of Scottish excavations

146 152 160 162 171 172 173

environmental and non-environmental investigations environmental Scottish obtained from Scottish The environmental data obtained The major types of environmental environmental all environmental excavations expressed percentages of all expressed as percentages excavations investigations investigations

258

reports featuring The percentages of Scottish excavation reports period environmental investigations investigations over the review period environmental

177 181 186 190

196 199 200

220 223 227 228 235 242

260 261

List of Plates

3.1

Clatteringshaws Loch, Pine stumps appearing from beneath peat at Clatteringshaws Kircudbright Stewartry of Kircudbright Stewartry Quartzite mountains rising above a platform of glacially scoured Quartzite gneiss bedrock, Arkle, Sutherland gneiss Glaciated landscape of the Loch Quoich area, north-western

25

3.2

Highlands mountains of the outwash deposits amid the mountains Fertile deltaic and outwash Highlands north-western Highlands deposited at the end of moraines deposited Chains of hummocky recessional moraines Skye of Skye Isle of the Loch Lomond Stadial, Luib, Isle the Relict cliffline in western Jura uplifted as a result of glacio-isostatic processes Fossil clifflines with Lateglacial unvegetated raised beach ridges in

29

2.1

3.3 3.4 3.5 3.6 4.1 5.1 5.2 5.3 5.4 7.1 7.2 7.3 7.4 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.1 9.2 9.3

9.4

Jura western Jura Braeroddach Loch, Aberdeenshire

Coppiced hazel on the Isle of Mull Black Fife Black Loch, Fife Barra Lochan na Cartach, Barra

Peat cutting cutting in blanket peat on slopes of Reineval, South Uist Mesolithic shell middens on Oronsay The machair at Bagh Siar, Vatersay background Skye in the background Cuillins of Skye Kinloch, with the Cuillins Rhum, with Isle of Rhum, Kinloch, Isle Kincardineshire Banchory, Kincardineshire Dee, near Banchory, River Dee, Terraces of the River Terraces Dumfriesshire Cairnholy, Dumfriesshire The chambered cairn at Cairnholy, Hill, Perthshire known as Herald Hill, The long barrow known

Excavations on the north-east cairn at Balnuaran of Clava monument at North Mains, Strathallan, The excavated henge monument Perthshire Perthshire Callendar Auchenlaich near Callendar The long cairn at Auchenlaich long cairn The

Douglasmuir, Angus An aerial view of the Neolithic enclosure at Douglasmuir, Daviot, Aberdeenshire stone circle at Loanhead of Daviot, Recumbent stone contents residues of its contents sherd of Grooved A sherd with residues Grooved Ware with The building at Balbridie, Balbridie, Kincardineshire, under excavation Angus Pitkennedy, Angus from Pitkennedy, Jet necklace from Jet

Headland, Uig, Lewis Bronze Age burial from Cnip Headland, adult male Reconstruction of the facial features of the Bronze Age adult Cnip from Cnip Drummond The remains of the wooden disc wheel found at Blair Drummond Moss

16

29 32 35 36 60 70 74 79 81 110 113 116 121 131 132 133 134 135 138 138 144 147 154 155

156 157

PLATES LIST OF PLATES

9.5

9.6 9.7 9.8 9.9 10.1 10.2

Balnabroich, Strathardle, Perthshire. Hut-circles, field systems and small air the air from the showing from caims showing small cairns Large Lairg, Sutherland excavation, Lairg, course of excavation, roundhouse in the course Large roundhouse Benbecula Ard marks at Rosinish, Benbecula Loch Nell, Ox Argyll Nell, Argyll bog at Loch yoke found in a bog Ox yoke

Drumturn Burn, Alyth, Traces of a relict prehistoric landscape at Drumturn Perthshire hut platforms with hut Hownam Law, Roxburghshire, with Hillfort, Hownam Hillfort, Cropmarks of an unenclosed settlement in the Esk valley, the North Esk Angus

10.3 10.4 10.5 10.6 11.1

Remnants of an experimental timber-laced wall experimental timber-laced Traprain Law, Lothian East Lothian Law, East Broch at Dun Carloway, Lewis Hut external Knowe, Roxburghshire: an enclosed settlement with an external Hut Knowe, agriculture trackway and bounded plots of cord rig agriculture The line of the Antonine Wall ditch and upcast mound across Croy Hill the air from the Hill from

159 161 163 164 167 174 175 178 180 184 191 197

11.2

pits, and an adjacent The line of the road, picked out by its quarry pits, Ridge at Westerton timber Westerton Gask Ridge the Gask timber watchtower on the

201

11.3

adjacent temporary parts of two adjacent The Antonine auxiliary fort and parts camps at Glenlochar Glenlochar

202

11.4

Angus, showing as acre (25 ha) Severan camp at Kirkbuddo, Angus, The 63 acre cropmarks cropmarks

210

11.5

Ditched field systems outside the vicus at Inveresk showing as

11.6 12.1 12.2 12.3 12.4 12.5 12.6 13.1 13.2 13.3 13.4

13.5

14.1 14.2

cropmarks hillfort at Edin’s Hall earlier hillfort settlement and earlier The broch, unenclosed settlement The hillfort at Dundurn, Perthshire oak) at Green Castle, reused oak) The timberlaced rampart (built of reused Portknockie Portknockie and the site of Buckquoy islet of the The tidal islet Birsay and Brough of Birsay the Brough The

A battle scene on the Class II slab at Aberlemno Kirkyard, Angus The Dupplin Cross, Perthshire View of cropmarks near Boysack in the Lunan Valley, Angus Aerial view Orkney during excavation eroded boat burial at Scar, Sanday, Orkney The eroded bay and sheltered bay land and fertile pocket of land Coll, featuring the fertile Sorisdale, Sorisdale, Coll, settlers early Norse settlers characteristics favoured by early — characteristics The outline of House 1 on site 2 at Skaill, Deerness, Orkney The

illustrated by Increasing use of cattle for traction at Pool, Orkney illustrated metapodial showing extension of articular end and infection, probably resulting from arthritis Scapula from a red deer at Pool showing new bone formation resulting from wound to tissue which seems to have been caused by a projectile Deflation of machair sands by wind causing damage to a Bronze Age settlement at Cladh Hallan, South Uist These rat holes have pierced the sandy soils associated with a possible Iron Age settlement on Sandray

213 215 224 225 232 236 237 238 246 247 250

252

253

262 263

List of Tables

4.1 4.2 4.3 4.4 4.5 6.1

6.2 6.3 11.1 11.2 11.3

Classification of Scottish soils Classification Scotland materials in Scotland and parent materials extent and their extent associations, their soil associations, Main soil subgroups in groups or subgroups Occurrence in rank order of major soil groups Scotland Scotland agriculture in Scotland capability classes for agriculture Extent of land capability Extent information on buried soils in Sites for which there is published information Scotland Carding and Carding Oronsay, Morton from mesolithic sites at Oronsay, bone from Morton and Animal bone Mill Bay Scottish sites from a selection of Scottish bird fragments from of bird Distribution of Distribution

Fish from a selection of sites of varying date Scotland garrisons in Scotland Estimated Roman garrisons Estimated

Dated pollen diagrams and the onset of major forest clearance Tree species from Roman forts represented by macrofossil evidence

47 47 48 48 49 52 54 89 92—6 97—8 204 209 211 211

List of Authors

Ian Armit MA, PhD (Edinburgh) is Senior Lecturer in Archaeology at The Queen’s with Monuments with Inspector of Ancient Monuments was formerly an Inspector University of Belfast and University and was Historic Scotland. He articles on Scottish and and books and articles numerous books has written numerous He has Historic Scotland. co-directing a programme of currently co-directing and is currently European archaeology and north-west European north-west excavation on the major Scottish Scottish hillfort of Traprain Law. His most recent books are Scotland's Scotland’s Hidden History, Celtic Scotland and The Archaeology of Skye and the Western Western Isles.

Colin K. Ballantyne MA (Glasgow), MSc (McMaster), (Edinburgh), DSc (St (McMaster), PhD (Edinburgh), Andrews. His Physical Geography Professor of Physical Andrews) Andrews) is Professor Geography at the University University of St Andrews. paraglacial and periglacial, paraglacial principal research interests are in the fields of glacial, periglacial, principal geomorphology and hillslope geomorphology hillslope and Late Quaternary Quaternary landscape landscape evolution, evolution, particularly particularly in Great Britain. The Periglaciation Britain. of Great Periglaciation of of The He is co-author of Scotland. He Scotland. the Proceedings (Edinburgh) is editor of the Barclay MA, Gordon Gordon J. Barclay MA, PhD (Edinburgh) Proceedings of the Society Monuments for Inspector of Ancient Principal Inspector and Principal Scotland and Antiquaries of Scotland of Antiquaries of Ancient Monuments prehistory of Historic Scotland. He has excavated and written extensively on the prehistory east-central historiography of interested in the historiography particularly interested and he is particularly cast-central Scotland and excavations Maxwell) on the excavations G. S. Maxwell) (with G. monograph (with has published a monograph He has prehistory. He prehistory. Neolithic Cleavcn Dyke, and Farmers, Temples and Tombs, a study of Neolithic of the Cleaven Scotland. Scotland.

Paul Department of the Department Professor in the (Birmingham) is Professor BSc, PhD (Birmingham) Paul C. Buckland BSc, with posts with formerly held posts Sheffield. He formerly University of Sheffield. Prehistory, University and Prehistory, Archaeology and Archaeology Lecturer in becoming Lecturer before becoming Museum before Doncaster Museum and Doncaster Trust and Archaeological Trust York Archaeological the York the

Geography Geography at the University of Birmingham. His principal interests lie in the insect biogeography of Atlantic islands and the conservation of wetlands. ArchaeHeadland ArchaeDirector of Headland (Bristol), PhD (London) is a Director BSc (Bristol), BSc ology Ltd in Edinburgh. He has worked as a consultant soil scientist and holds an honorary position in the Department of Environmental Science, University of from archaeological contexts in sediments from interested in sediments particularly interested Stirling. He is particularly Stirling. Scotland, including the use of soil micromorphology.

Carter Stephen P. Carter

the collections in the Age collections Bronze Age the Bronze MA (Edinburgh) is Curator of the Trevor G. Cowie MA Trevor field Scotland and was a field Archaeology, National Museums of Scotland Department of Archaeology, Development Scottish Development Unit, Scottish Excavation Unit, Central Excavation archaeologist with the former Central periods Age periods Bronze Age and Bronze Neolithic and the Neolithic include the interests include His research interests Department. His Department.

and he is a co-author of Symbols of Power at the Time of Stonehenge.

xiv

LIST OF AUTHORS

DepartProfessor in the (Sheffield) is Professor PhD (Sheffield) (Aberdeen), PhD BSc (Aberdeen), the DepartDavidson BSc Donald A. Davidson having held posts in the ment of Environmental Science, University of Stirling, having Universities of Sheffield Readership in Geography at the University Sheffield and Wales, and a Readership Strathclyde. He has worked on geoarchaeological aspects aspects of projects in Greece and of Strathclyde. Scotland. His current research interests include include applications of soil micromorpholmicromorpholresearch interests His current Scotland.

ogy and the spatial spatial variability of soil properties. He is the author or co-editor co—editor of

numerous books including Principles and Applications of Soil Geography and GeoPast. and the Past. archaeology: Earth Science and archaeology:

Alastair G. Dawson MA (Aberdeen), MS (Louisiana), PhD (Edinburgh) is Professor in Quaternary Science and and the Environment Environment at the School of Science Quaternary Science in the fessor palaeoclima— Coventry Coventry University. His principal research interests lie in Quaternary palaeoclimaHe Atlantic region. He Holocene climate changes in the North Atlantic tology, in particular Late Holocene has on aspects of Quaternary sea-level and tsunami sea-level changes and widely on published widely has also published Age Earth. Ice Age He is author of Ice research. research. He Kevin J. Edwards Edwards MA PhD (Aberdeen) Physical Professor of Physical (Aberdeen) is Professor Andrews), PhD MA (St Andrews), Geography in the Department of Geography and Environment, University of Environment, University of City University of The City Graduate School, The and Adjunct Professor the Graduate Professor in the Aberdeen and

Departments of Geography New York. He was on the staff of the Departments Geography at the Universities Universities Birmingham before becoming Professor and Head of the Department of Belfast and Birmingham Department University of Sheffield. Prehistory, University and Prehistory, Archaeology and of Archaeology Sheffield. His research research interests interests focus and include Scotland and include applications applications of palynology, palynology, the North Atlantic region, and and the on Scotland co-editor of was co-editor years he was archaeology. For sedimentology and For ten years and tephra studies in archaeology. sedimentology Editor of Associate Editor currently Associate the Journal of Archaeological Science and he is currently Environmental Archaeology. Environmental Archaeology.

Centre for Field Manager of the Centre was Manager MA, PhD (Edinburgh) was MA, PhD Archaeology at the University of Edinburgh Director of Edinburgh until his appointment as Director been has been He has Amman. He based in Amman. the Council Levant, based the Levant, Research in the British Research for British Council for responsible and as far afield projects in Scotland and archaeological projects variety of archaeological wide variety for a wide responsible for

Bill Bill Finlayson

focus primarily within the Mesolithic, as Jordan. Mesolithic, and focus interests lie primarily Jordan. His research interests

especially on the use of lithic materials. He is the author of Wild Harvesters. Harvesters. William S. Hanson BA, PhD (Manchester) is Professor of Roman Archaeology Archaeology in the interests His research interests Glasgow. His University of Glasgow. Archaeology, University Department of Archaeology, the Department concentrate on the northern provinces of the Roman Empire, particularly the interaction between Rome and the indigenous population. His publications include Agricola and the Conquest of the North and Rome Rome’s ’s Northwest Frontier: The Antonine Wall. He co-edited Scottish Archaeology: Archaeology: New Perceptions.

Ancient Professor of Ancient and Lund) is Professor (Durham and (Durham) PhD (Durham BA (Durham) John R. Hunter BA History Reader was formerly Reader and was University of Birmingham and the University and Archaeology at the History and survey and experienced in multi-period survey Archaeology at Bradford University. He is experienced in Archaeology excavation books His books Isles. His Western Isles. and Western Northern and the Northern Scotland, particularly in the excavation in Scotland, include Archaeology of include Rescue Excavations aatt the Brough of Birsay and Fair Isle: the Archaeology an Island Introduction to Studies in Crime: an Introduction of Studies co-editor of He is also a co-editor Community. He Island Community.

LIST LIST OF AUTHORS Forensic Archaeology, The Archaeology Management Management in the U K .

xv Resource Archaeological Resource o f Britain and Archaeological

Finbar Archaeology Lecturer in Scientific Archaeology Finbar McCormick M A (Cork), PhD (Belfast) is Lecturer UniQueen’s UniThe Queen’s and Palaeoecology in The Archaeology and School ooff Archaeology the School and Head of the and versity of Belfast. He was formerly an archaeologist with AOC (Scotland) Ltd. His research interests are primarily on the evolution of the relationship between people and animals in Ireland and Scotland. Ian B. M . Ralston M A European PreLater European (Edinburgh) is Professor of Later PhD (Edinburgh) A,, PhD Department of Archaeology, University of Edinburgh, having been a history in the Department lecturer in the Department of o f Geography, University of Aberdeen. H e maintains research research interests in the later prehistory of France and in aspects of Scottish archaeology. H e authored Les Enceintes fortifiées du Limousin, and (with O . archaeology. Hee Age Bourges. H and a study of Iron Age Bibracte and remparts de Bibracte Les remparts Buchsenschutz) Les Buchsenschutz)

co-edited Archaeological Resource Management in the UK and The Archaeology of Britain. Britain.

Lecturer in PhD (Sheffield) is Senior Lecturer MSc (Birmingham), PhD S c (Birmingham), M Biogeography in the School of Geography, Earth and Environmental Environmental Sciences, University of Birmingham and was formerly a Research Fellow in the Department Hee has carried out the University Prehistory at the and Prehistory Archaeology and of Archaeology University of Sheffield. H archaeological projects in extensive palaeoentomological associated with archaeological research associated palaeoentomological research extensive Sadler Jonathon P. Sadler Jonathon

Greenland. Scotland, Scotland, Iceland and Greenland.

Aberdeenshire Archaeologist for Aberdeenshire (Edinburgh) is Principal Archaeologist Shepherd M A (Edinburgh) A. G. G . Shepherd Ian A. Council. His Regional Council. Grampian Regional Council with the former Grampian Council and held a similar post with Europe. H western Europe. Hee is a former Age of western Bronze Age the earlier Bronze centre on the interests centre research interests research author of and the author o f Scotland Scotland and Antiquaries of Society of Proceedings ooff the the Proceedings editor o f Antiquaries the Society editor of the Powerful Pots; Beakers in North-East North-East Prehistory and Aberdeen Aberdeen and North—East North-East Scotland.

the Graeme Whittington Geography in the PhD (Reading) is Emeritus Professor of Geography BA, PhD Whittington BA, School of Geography and Geosciences, University of St Andrews. His research environmental change, interests lie in Scotland and are concerned with landscape and environmental particularly palynological and sedimentological approaches. His publications include An Historical Geography of Scotland and Fragile Environments.

Preface

This paperback represents an updated edition of Scotland: Environment and ArchaeAD 1000, ology, 8000 provide was to provide then was aim then Our aim 1997. Our Wiley in 1997. 1000, published by Wiley BC—AD 8000 BC— an overview of Scotland’s natural environment and the human communities which

inhabited it from the end of the last Ice Age until about a thousand years ago. That book represented a successful collaborative venture involving numerous colleagues across a range of disciplines. We are indebted to our original publisher, Iain working across Stevenson, and his colleagues at and to John Davey and guidance and initial guidance Stevenson, for his initial Edinburgh University support. subsequent support. their subsequent Press for their University Press The new feature selective guide to the literature that has volume is a selective present volume feature of the present published between 1996 and quantity of research published over the last few years. The quantity emerged over assistance of the contributing impressive. We trust that, with the assistance early 2002 has been impressive. flavour of authors, we have identified many of the key items and have conveyed a flavour significance of omitted and the full significance Much has had to be omitted developments. Much newer developments. some newer Final future. Final apparent in the future. undoubtedly only become apparent some such contributions will undoubtedly

responsibility for this selection rests with us. accessible format, more accessible format, Scotland After the Ice Age should continue to serve In this more

those interested in the intertwined natural and cultural dimensions of the country’s advances that have been past. We hope that readers will be able to appreciate the advances

made over recent decades and that these essays will act as a spur to increasing among the various integration among various research fields. integration wives, Rachel and Sandra, and owes much to our wives, volume owes this volume predecessor, this Like its predecessor, put up with who again have put and Tom, Tom, who Fraser, Natalie children, Calum, Fraser, to our children, Natalie and to

playing second fiddle to computer screens and keyboards. Kevin Edwards and Ian Ralston, June 2002 Maryculter and Kinross, June

1

Environment and People in Prehistoric and Early Historical Times: Preliminary Considerations Considerations KEVIN J. EDWARDS AND IAN B. M. RALSTON

INTRODUCTION INTRODUCTION During and since the final stages of deglaciation Scottish the Scottish years ago, deglaciation some 10 000 During and 000 years ago, the and human forces. forces. Climate natural and dramatically fashioned by natural has been dramatically landscape has landscape extent, territory and, to an extent, physical make-up of the territory have influenced the physical geology have and geology have dictated the biogeographical records of flora revealed by records patterns revealed biogeographical patterns have dictated flora and and fauna. fauna. The Isles, initially forming part linked part of a more substantial British Isles, The British substantial landmass landmass and and linked

Europe at the outset of the period, became fully by a landbridge to continental Europe 7000 BP. BP. detached detached by c. 7000 Taking location on the Atlantic edge of north-western north-western Atlantic edge Scotland’s location view, Scotland’s long View, the long Taking the Europe, and have not inhospitable terrain, have often inhospitable and its often not noticeably inhibited human human settlement. That said, when when the record instances of land closely, instances examined closely, record is examined

formerly occupied and now abandoned, or now used much less intensively than in previous centuries, are numerous. In the relatively recent processes such as the recent past, past, processes profoundly modified the Movement have Highland Clearances and the have profoundly the Improving Movement and the

appearance of substantial tracts of the countryside, as well as having had a serious socio—economic life of the human population. Whilst impact on the distribution and socio-economic these large—scale large-scale changes within the last few centuries have generated a substantial literature, have the countryside and that affected the earlier changes that the earlier literature, the and its inhabitants have tended in the main to be considered in specialist publications.

AIM

The principal aim of this account is to make information on the environmental and archaeological records for Scotland for the period from the first clear human B. M . Ralston. Ian B. and Ian Kevin JJ.. Edwards and by Kevin Edited by 1000. Edited AD 1000. BC — AD and Archaeology, 8000 BC Environment and Scotland: Environment Scotland: contributors. Published in 1997 by John Wiley & Sons Ltd. editors and contributors. © 1997 The editors

2

K. J. EDWARDS AND I. B. M. RALSTON

accessible. medieval state, more accessible. the medieval emergence of the until approximately presence until approximately the emergence presence forward put forward that have been put and ideas that information and together information intention is to bring together The intention The 1970s, in a since the 19705, more particularly since academic literature, more range of academic wide range in a wide indulge in historiographical discussions not the This is not overview. This readable overview. the place to indulge readable however, fields. It would be inappropriate, however, research in these development of research the development these fields. of the pioneering work, and contributors indebtedness to pioneering acknowledge present-day indebtedness not to acknowledge included reference to key early studies within their fields. The have in some cases included primary focus of this volume also differs from previous general studies of Scottish Ritchie and Piggott 1962, (Childe 1935, prehistory (Childe 1991). 1981, 1991). and Ritchie 1981, 1982; Ritchie 1962, 1982; 1946; Piggott 1935, 1946; prehistory instance, are less fully examined, and funerary practices, ornaments and funerary practices, for instance, Tools, ornaments

juxtaposed more readily with other categories of archaeological data which can be juxtaposed such as settlement environmental information, such are settlement sites and evidence for agriculture, are prominence. proportionately greater prominence. given proportionately given

preservation of data for past Scotland offers many circumstances for the excellent preservation archaeological sites provide similar environmental conditions, and to some extent its archaeological similar environmental environmental dimensions. In this treatment, these opportunities for considering environmental where this is appropriate, been highlighted perspectives have been environmental perspectives appropriate, but highlighted where environmental approaches nor, in cases where other approaches exclusion of other total exclusion neither to the total available where available neither follows What follows data. What significance of the data. overplaying the significance weak, by overplaying very weak, information is very information although it past, although Scotland’s past, archaeological account not a conventional account of Scotland’s conventional archaeological is not system of (e.g. the standard system the use of the standard the familiar retains enough deliberately familiar (e.g. enough of the deliberately retains Neither navigate a course. reader to navigate Stone, Bronze and course. Neither Iron Ages) to enable the reader and Iron Stone, is this a book about environmental archaeology as such. That would require evaluate information and evaluate extract and information approaches to extract multidisciplinary approaches practice of multidisciplinary the practice the contexts or their wider settings. Many of on the environment from archaeological contexts the approaches that underpin the present study are indeed identical, but their attention interpretation. Less attention deployment has been to furnish perspectives for wider interpretation. various suites of environmental data is paid to the strengths and weaknesses of the various than would be the case in a work where taphonomic and methodological considerations were a major emphasis. interactions human—environment interactions nature and here is upon the focus here The focus extent of human—environment and extent the nature since the final melting of the Lateglacial ice sheets. The early Holocene (Postglacial) was a time when natural environmental changes were already transforming the appearance and resources of the land: it is therefore appropriate that the imnumber of chapters in order to examined in a number mediately antecedent conditions are examined set the scene. It is shown that indications of human exploitation of, and by extenback as human the environments of Scotland are traceable as far back sion, impact on the communities are demonstrably present. At times when naturally occurring changes communities dynamic, a human presence is sometimes difficult to detect; in particularly dynamic, were particularly attributable to overwhelmed by those attributable changes seem to be overwhelmed natural changes periods, natural later periods, human action. human Environment, as considered human considered here, is not simply a mute context for human Environment, endeavour part of life in the past. Until very recently endeavour — rather, it formed an integral part — Scottish communities were certainly well within the present millennium ~ — and certainly and their natural environments formed, consciously exclusively rural and consciously or unconsciously, part of the fabric of their lives. To say this is not to imply a deterministic stance. Environmental Environmental factors may restrict, but do not eliminate, human choices. stance.

PRELIMINARY CONSIDERATIONS

3

also of importOther factors — social, economic and religious, for instance — were also whilst follow, whilst which follow, the chapters which and the and stability, and contributing to change and ance in contributing outline some of these influences. also outline influences. mind, also circumstances in mind, keeping environmental circumstances keeping

The opposition between environmental and cultural approaches to the study of the archaeological record found in some recent publications (e.g. J. Thomas 1990), is archaeological (cf. rebuttals by O’Connor [1991] and Wilson [1995]). Our position inappropriate '(cf. inappropriate remains that the environment should not be relegated to a benign neutrality and disregarded as ‘noise’ in the creation of a social archaeology, any more than it disregarded the front of which the designs, in front varied designs, should be seen as a series of backcloths with varied should actors perform with insouciance.

environment and of earlier human communities are both necessStudies of the environment multidisciplinary, the range of approaches now available exceeding the arily multidisciplinary, capacity of any individual. Whilst the contributors have collectively addressed many capacity research fields, a work of this kind makes no pretence at being fully comprehensive. Amongst resources, plant macrofossils and exploitable minerals and ores are some Amongst of the topics which might be considered to merit more extended treatment. The impact of external trade and contact and a consideration of human population the major less, the the less, prominence. None the also have enjoyed greater prominence. dynamics might also dynamics active research areas are all included. The levels of integration found in published and the cultural periods the variable and periods is variable topics or cultural particular topics with particular associated with research associated research contributions reflect this. Thus, the discussion of climate is unidirectional following contributions agricultural vegetational and landscape, soil, vegetational may influence that changing in that and agricultural influence landscape, climate may changing climate variables are not shown to have a demonstrable impact development, but these variables interrelationship between environupon climate. During the Mesolithic, a strong interrelationship constraints and opportunities and human life is assumed, not least because mental constraints because mental

of the characteristics of the archaeological record. Inferences focusing upon social the sparseness behaviour at this time have to allow within exploitable data within sparseness of exploitable allow for the behaviour Scotland. approached, Paradoxically, as more recent prehistoric and historical times are approached, less no less likely to be no and knowledge knowledge of environmental change might be thought likely and the influence of environment upon the earlier periods, consideration than in earlier the consideration of the than the literature. less prominently literature. developed in the prominently developed human communities seems to be less assertion, which, if overstated, is this assertion, There are a number of reasons underpinning this

indicators, potencertain of the environmental indicators, not grossly so. On the one hand, certain useful as measures less useful early periods, become less the early sensitivity in the great sensitivity tially of great measures of tially Other indicators remain case in point. reduced, is a case (once reduced, point. Other change. Woodland, «once

potentially useful (e.g. insects or stable isotopes), but are little-studied because of the difficulties involved in obtaining suitable deposits, while the environmental signals derived from lake sediments could result from a complex mixture of hand, this the other hand, On the and anthropogenic processes. On natural and emphasis this reduced emphasis natural may be a function of less responsive environmental indicators and hence a relative lack of interest on the part of environmental researchers. It may be that the and evidence and archaeological evidence physical archaeological weight of physical the weight overwhelmed by the record is overwhelmed than when natural information that is available, or indeed, other than historical information ., historical affairs of influence on the affairs environmental influence less environmental disasters occurred, perhaps there was less

people. At the very least, the environment had to be permissive in terms of the agricultural practices. requirements of biota and agricultural ecological requirements

4

K. J. EDWARDS AND I. B. M. RALSTON RALSTON

THE SURVIVAL AND DETECTION OF THE ARCHAEOLOGICAL ARCHAEOLOGICAL RECORD

Authors detection Authors were not asked to address in detail the matter of the survival and detection landscapes of Scotland. attributable to earlier human communities in the landscapes of sites attributable Scotland. question, well rehearsed from different perspectives by Stevenson (1975) and This question, Barclay (1992), is however intimately related to environmental concerns, concerns, as the locations of the sites of prehistoric and early historic activities have a greater greater or locations lesser chance chance of being identified substantially as a result of the materials of which they were made, their positions within the varied topography of the country, and subsequent land-use histories. Current evidence suggests that few major Current evidence major sites sites occur above above the 500 500 m contour, although although there are exceptions, such as Ben Griam Beg in Sutherland (Chapter 9). In the altitudinal altitudinal band below this, and above the intensively cultivated sectors of the countryside, and over numbers and survive in substantial numbers monuments survive upstanding monuments countryside, upstanding over considerable areas, afforested heathland, and in some afforested grassland, heathland, areas, in permanent grassland, considerable structures, other clearly visible settings. Alongside immediately other sites, less immediately Visible stone structures, Alongside clearly settings. apparent, can be noted in quantity. The banks of stones include banks latter include The latter be noted apparent, stones delineating delineating circumstances, the practices the traces of early cultivation practices field systems and, in some circumstances,

within them. Leaving aside the more massive drystone monuments, perhaps too drystone monuments, remote to have served as convenient quarries for subsequent subsequent reuse, and some land boundaries may have demarcated periods, most land for extended parcels of land demarcated parcels boundaries which may extended periods, most of these sites have survived on the surface because subsequent subsequent episodes of land use have been insufficiently intensive to prompt their demise. This zone of likely survival large by large clipped by moorlands, clipped the short grass like the — particularly in areas like survival grass moorlands, archaeological the Scottish resources of the the great resources of the one of represents one flocks of sheep flocks sheep — represents Scottish archaeological record. For here, soils, have interacted to enable the vegetation and land use have soils, vegetation

survival of extensive surface traces been appreciated that some of long _

periods. It has attributable to some of the early periods. attributable landscapes include amongst archaeological landscapes these archaeological amongst

their also the former visible remains not only the traces of stone-built structures, but also their visible

techniques, built on fieldworking techniques, stances of those built of wood. Improved Improved fieldworking particular by oblique aerial supported in particular familiarity with these landscapes and supported occasion detailed traces, and finely extensive and photography, continue to reveal extensive photography, finely detailed traces, on occasion now substance now how that substance indicative of how thus indicative and thus disappearing below peatland and disappearing communities. by human communities. previously used by areas previously obscures areas obscures In contrast, the lowlands, whether improved substantially for pasture as in the South-West, or given over to arable crops as is more frequently the case in eastern Scotland, have a much reduced tally of monuments periods. monuments surviving from earlier periods. result of cultivation where, as a result destruction where, entered the zone of likely destruction have entered Here, we have since Medieval times, and perhaps most importantly importantly during the periods of intense agricultural improvement of recent centuries, surface indications of many archaeological sites have been wholly eliminated. Artefact collection through fieldarchaeological walking, used than further less used Scotland, is less successfully employed in Scotland, be successfully can be whilst it can walking, whilst south in Britain. Such collection (sometimes supplemented by test-pitting) is of considerable period, the Mesolithic period, notably the earlier prehistory, notably for earlier significance for considerable significance identify. In difficult to identify. are difficult sites are accumulations, sites leaving aside midden accumulations, where, leaving where, favoured lowland zones, the former extent of human activity for all subsequent

CONSIDERATIONS PRELIMINARY CONSIDERATIONS PRELIMINARY

5

cropmarks. photography of cropmarks. periods is most extensively displayed through aerial photography pioneering Although scale and the 1920s, the in Scotland over made were flights Although dramatically increased has reconnaissance archaeological for overflying regularity of regularity areas where free19708, i.e. relatively recent in British terms. In areas only since the mid 1970s, soils, draining coincide with gravels, and sands glaciofluvial over located especially draining These recorded. can cropmarks multi-period rainfall, reduced These recorded. be can cropmarks of suites multi-period extensive, reduced comparable with those in the cropmarks include categories of sites not readily comparable corpus of upstanding monuments. Archaeologists are only now coming to grips information. Cropmark records also furnish numerous with this wealth of new information. indications that lower-altitude areas, more favourable for settlement on grounds, were extensively and repeatedly employed in the past. environmental grounds, environmental destruction is necessarily a This brief delineation of zones of likely survival and destruction climatic, interplay of soil, climatic, from result patterns complex reality, simplification. interplay the from result patterns complex reality, In simplification. including land-use and other factors. The insertion of modern infrastructure, including The archaeological evidence wholesale. The removed archaeological has removed sprawl, has and urban sprawl, motorways and motorways dovetailing of zones of survival on hard rock areas and long coastline sees the dovetailing sectors where where natural and human sometimes within sand-dune systems, with other sectors by peat masked Areas peat growth have work. long have processes destructive work. at been long have processes destructive

been mentioned for their potential blanketing of archaeological features, thereby numerous bodies of water (and their often ensuring good preservation. The numerous Neolithic, the Neolithic, from the immediate least intermittently from at settlement for used littorals), immediate recent more and Age Bronze late of and Age late the of the in and crannogs the of case the famously most and dates, also provide excellent milieux for the survival of organic artefactual and environmental data. The diver with drysuit and aqualung, and the airborne phototwentieth century search grapher, are perhaps the two most obvious symbols of the twentieth Scotland. They here serve to emphasize the incomfor archaeological remains in Scotland. between relationships between archaeological record, the Scottish archaeological and the direct relationships pleteness of the record, and archaeological remains, and of archaeological detection of and detection survival and land-use environmental and land-use and environmental factors.

PARAMETER POPULATION: THE UNQUANTIFIED UNQUANTIFIED PARAMETER almost all cases, assessing the scale of the impact that is detectable at a particular In almost point is fraught with difficulties, compounded by the fact that archaeology has few, if any, methods at its disposal to estimate previous human population sizes. readiness researchers of a generation ago, there is a readiness of researchers the perspectives of Compared to the Compared remain numbers absolute although populations, earlier for higher to accept numbers absolute although populations, earlier figures accept elusive. Thus the suggestion of about 70 as the Mesolithic population of Scotland Mesolithic population serious underestimate (Chapter 7), 1962, 7) is considered a serious (Atkinson 1962, only in not only 7), not the environment, regard to with regard carrying capacity of the but also with environment, but terms of the carrying population graph for Britain proposed a population Burgess (1989) has viability. Burgess reproductive viability. Britain has proposed reproductive from 5000 BC until modern statistics become available. Necessarily speculative in oscillations likelihood of oscillations diagram makes explicit numbers, his absolute numbers, terms of absolute the likelihood explicit the his diagram though in the absence of in human numbers in early times, as well as suggesting, though volcanic events, as a factors, more particularly environmental factors, particularly volcanic secure evidence, environmental assumption that the development of control (cf. Chapter 9). It is a frequent assumption

6

K. J. EDWARDS AND I. B. M. RALSTON

increasing populations, as milk-and-mush permitted increasing agricultural regimes may have permitted nutrition of more children a feasibility, diets adequate and weaning earlier made diets advantage in underand arguably, made the availability of more hands a distinct advantage The literature contains production. The agricultural production. with agricultural tasks associated with taking the tasks see though see 1968; though Atkinson 1968; (e.g. Atkinson demographic expansion (e.g. frequent references to demographic unquantified, other than at Bradley 1978), but within Scotland this is usually wholly unquantified, been made, as in the have been speculative forays have scale, where speculative individual monument scale, the individual

case of Isbister and other chambered cairns in Orkney (Hedges 1982; Fraser 1983). individual archaeological sites, population estimates have either been derived For individual assumptions about labour requirements for their construction, or, as in from sets of assumptions characteristics of the human skeletal material they contain. the Isbister case, the characteristics part of the consideration of whether all, or part This latter approach is predicated on a consideration assemblage. Even less secure within the bone assemblage. population of the locality is comprised within within population sizes based on the likely number of inhabitants within are guesses about population population buildings. At a wider scale, estimates have been made about likely population near-contemporaneity of extensive capacities where near-contemporaneity figures in relation to carrying capacities impact of a assumed (Fojut 1982). A special case of the impact sets of sites can be assumed population is offered by the Roman army (as discussed in Chapter 11), particular population and even here, although this perspective is underpinned by knowledge from written considerable the time, there remains a considerable units at the military units sources of the size of military sources measure of uncertainty. Rackham (1980) has estimated that a minimini— From an environmental perspective, Rackham population of about half a million adults would have been needed in Britain mum population to reduce ehn elm pollen productivity sufficiently for it to be registered as the wide‘national statistic’, it is, as the spread elm decline. Whilst this provides us with a ‘national author makes plain, one fraught with uncertainty to the point of improbability. All such approaches to former populations are useful heuristically as an aid to groups and the potential scale envisaging the possible structures of previous social groups of impact of communities on their environments. The foregoing examples indicate about how unsure remain about unsure we remain and thus how how tentative such attempts attempts necessarily are, and how populations. the size of previous human populations.

CONSIDERATIONS PRACTICAL CONSIDERATIONS chronology The cultural framework and its chronology

The contributors have adopted a relatively traditional perspective on the classification systems employed here. An initial note on the use of these is provided by fication way of explanation. explanation. way Radiocarbon dating understanding both of the chronologies of environmental Current environmental change and of the Current understanding by the radiocarbon timeScotland in settlement human of underpinned is duration Scotland human duration with calendar comparable straightforwardly this that scale. Early confidence straightforwardly was confidence introduced by the fact years has long proved to be misfounded. A complication is introduced

PRELIMINARY CONSIDERATIONS

7

not systematic errors, not that the radiocarbon chronology is subject to a number of systematic

long—term fluctuations in the 14C content of the earth’s atmosphere. least because of long-term accommodate Fortunately, it is now possible to calibrate the radiocarbon dates to accommodate dates radiocarbon of (dendrochronology) radiocarbon (dendrochronology) Tree-ring these of some calibration Tree-ring effects. these some demonstrated the variable discrepancies between historical years and those derived decline of c. elm decline radiocarbon date for the elm the radiocarbon (e.g. the radiocarbon method (e.g. the radiocarbon from the from 5100 radiocarbon years ago becomes, upon correction, c. 5780 calendar years ago). It has been the convention in the environmental sciences to work with radiowhere ‘present’ equals AD (BP; where present (BP; years before present carbon carbon (14C) dates expressed as years would date the decline, be presented as would be the elm the for Thus, 1950). elm the of example the for 1950). Thus, 5100 BP. This is perfectly satisfactory where the uncorrected radiocarbon time-scale is used solely, but occasions particular difficulties when other sources of dating are employed, or when the true duration of an event or process is of significance. These circumstances apply in many archaeological contexts. Moreover, since the developcircumstances expressed uncaliradiocarbon technique, archaeologists have generally ‘expressed ment of the radiocarbon reference to the BC/AD convention of historians. l4C dates by reference brated 14C As confidence in calibration procedures has increased (most notably after the Quaternary publication publication of a special issue of the journal Radiocarbon in 1986), Quaternary been has been time-scales. This trend has calibrated time-scales. adopting calibrated increasingly adopting are increasingly scientists are programs (e.g. Stuiver and calibration programs availability of computer calibration accelerated by the availability conventions Reimer 1993). To accommodate these developments, the following conventions throughout this work: have been adopted throughout

(i) (ii)

(iii) (iv)

Uncalibrated radiocarbon dates will be cited in years BP. Calibrated radiocarbon dates will be cited as years cal BC or cal AD, using the dates more only for dates (1993), but only Reimer (1993), of Stuiver and Reimer correction software of correction BP dates, and cal cal BC or cal and labelled cal BP. Figures 000 BP. recent than 10 000 Figures following BP recent ranges chronological ranges calibrated chronological the corresponding corresponding calibrated identify the appropriate, identify AD as appropriate, standard deviation. The term ‘cal BC’ will be applied to longer timeat one standard millennia), where the period concerned has been primarily defined spans (e.g. millennia), radiocarbon dates. use of calibrated radiocarbon by use The use of BC or AD in relation to a specific date will mean that this has not been derived from radiocarbon, but normally from historical sources. radiocarbon determiThe calibrated radiocarbon date derived from a raw radiocarbon i1 standard deviation associated with it) will be nation (i.e. including the :l:1 decade, as a one nearest decade, rounded to the nearest expressed, rounded standard deviation corone standard expressed, (e.g. rected range or the extremes of corrected ranges where these are produced (e.g. 3880i60 BP becomes, upon conversion and rounding 2460—2200 cal BC).

Impact of calibration on the cultural framework general terms, the availability of radiocarbon dating has had the effect of In general

from those prehistoric periods from earlier prehistoric the earlier chronological spans of the the chronological lengthening lengthening the with the reliant largely were who by envisaged with cross-dating on reliant largely were who scholars, mid-century envisaged comparable to using material comparable Basin, using Mediterranean Basin, historical civilizations of the Mediterranean historical that found in Britain from parts of continental Europe. For example, the likely start date, and the duration, of the New Stone Age (Neolithic) of Scotland was

8

RALSTON K. J. EDWARDS AND I. B. M. RALSTON

recognized, Piggott’s before Piggott’s radiocarbon dates became available, to lie centuries recognized, as radiocarbon centuries before has been calibration has estimate of c. 2000 BC; the (1954) estimate (1954) the impact impact of calibration been to extend the

timespan, and to push back the initial appearance appearance of Neolithic traits yet further. It Scotland extends human settlement the lengthening chronology that the follows follows that chronology for human settlement of Scotland extends the time-scale over which human—environment interactions can have taken place. traditional archaeological periods have also subdivisions of cultural The traditional archaeological subdivisions cultural periods also been retained (Figure embedded in the literature are still firmly embedded (Figure 1.1). These are retained literature on which which this book is based, and despite the shortcomings of divisions essentially founded on technological developments, developments, they serve as useful labels to enable correlations to be technological achieved. The Mesolithic period, associated with hunter—gatherer—fisher activity, achieved. occupies the span up to around the beginning of the fourth millennium cal BC, with the Neolithic continuing continuing for about one and a half millennia thereafter. The first appearance and copper-alloy Age from Bronze Age the Bronze heralds the copper-alloy equipment heralds copper and appearance of copper appearance as the preferred material for perhaps a little before c. 2500 cal BC. Iron’s appearance perhaps edge the around the occurred around have occurred be estimated to have may be Age, may the Iron Age, marking the tools, marking edge tools, environmental conditions conditions within but environmental seventh century BC, but within Scotland mean that ferrous material does not survive well. Use of these labels should not prejudice material prejudice the possibetween some considerable overlaps bility of considerable bility overlaps between some of the technological technological and and subsistence subsistence understood by those conventionally strategies conventionally understood notably those these terms, most notably strategies implied by these 1992). Neolithic (Armit and Neolithic Mesolithic and Finlayson 1992). and Finlayson (Arrnit and Mesolithic Geographical units

A further complication is produced by the fact that twice in the last quarter of a has been substantially Scotland has areas in Scotland authority areas of local authority map of the map century, the century, substantially reorganization effected in Scotland in 1996 disposed redrawn. The local government reorganization redrawn. two—tier system of Regions and Districts that had been in place since 1975. of the two-tier was in book was this book whilst this set in place whilst system set authority areas, a system new local authority The new The unfamiliar, and include areas both too large (e.g. Highland) and preparation, are unfamiliar, preparation, (e.g. names (e.g. service; equally, other be of great service; Renfrewshire) to be (e.g. East Renfrewshire) small (cg. too small other names Aberdeenshire) are now attached to different areas than those traditionally associAberdeenshire) two-tier system of Regions and Districts that preceded them, in ated with them. The two-tier use for only two decades, has too rapidly been consigned to history for it to have been adopted here. The most useful course of action seemed to us to be to revert to the pre-l974 counties as the best-established convenient geographical units, and any reference to counties in the following chapters thus refers to these (Figure 1.2). On a broader scale, the restriction of the scope of the present study to the modern political boundaries of Scotland, whilst conforming to recent traditions of archaeological heritage management, is culturally and environmentally meaningless for earlier times. It should thus be regarded largely as a convenient means of providing contributors recognizable geographical limits to the present study, beyond which contributors recognizable warranted. material warranted. have been permitted to stray as the nature of their subject material

9

ATIONS RATIONS CONSIDER ARY CONSIDE NARY PRELIMIN PRELIMI

TIMESOALE TIMESCALE

14

Calondar Calendar Years

r—_

1000 _— ‘°°° 1000 1000 -—"

1000

Norse Norse

-— — _____

Historic Early Histofic

/

_ —_

2000

-—

300° 3°00

_

Roman

2000

Iron Age Iton

_____ I-U l-U

4000 4000

—

---------——————————

zz

Neolithic

o

-

I-u w 5000 — 5000

r—

— 4000

0)

— -

— 3000 -

aa:0 -‘

Btonze Bronze Age —

I

I

I Quaternary

Archaeolo Archaool09! 9!

“ c BP Years

AD BC -

SUBDIVISIONS

—l

,.

—

—l

,—--J ,'—-’

I “

'9

o

2

0

E

A

—— _/ -’ -_ / ’’ _ _

._ _

500° 5000

6000 __ 600°

6000

-

7000 __-e 7000 _—

— —

_'— _..—'

._ _

8000 8000

— 9000

a: :1:

3°00 °°°°

__ _

9000 —9000 —r-

— -

— 7000 -— l-

_

O

Mesol't Mesol'thi

''mc

z a5 Lu LIJ

— 10000

— H

— 11000 -—

—

—

Loch Lomond ._. Lomond Sladlal {-3 1L: Stadial

§g, 2

3 3 12000 12000

(Quaternary environmental (Quaternary Figure 1.1 A chronological guide showing archaeological and environmental period) subdivisions subdivisions

10

Figure 1.2

K. J. EDWARDS AND I. B. M. RALSTON

reorganization in 1975. counties of Scotland prior to local government government reorganization The counties

Ross and Cromarty; 6: Inverness; Orkney; 3: Caithness; 4: Sutherland; Sutherland; 5: Ross Shetland; 2: Orkney; 1: Shetland; Moray; 9: Banff; 10: Aberdeen; 11: Kincardine; 12: Angus; 13: Perth; 14: Argyll; Naim; 8: Moray; 7: Nairn; l9: Fife; 20: Renfrew; Renfrew; 21: Lanark; Stirling; 17: Clackmannan; 18: Kinross; 19: 15: Dunbarton; 16: Stirling; Peebles; Berwick; 26: Ayr; Ayr; 27: Peebles; 22: West Lothian; 23: Midlothian; 24: East Lothian; 25: Berwick;

Selkirk; 32: Roxburgh Kirkcudbright; 30: Dumfries; 31: Selkirk; Wigtown; 29: Stewartry of Kirkcudbright; 28: Wigtown;

2

Change Climate Change Climate GRAEME WHITTINGTON GRAEME

AND KEVIN J. EDWARDS

INTRODUCTION weather they the weather relation to the populations in relation of past populations activities of the activities In considering the In experienced, it is necessary to prevent current perceptions of Scotland’s climate from intruding. Current perceptions are derived from those conditions which operate at the present, and yet it can be demonstrated that weather conditions, at the global level at least, have been far from stable over time. There have been large-scale and long-term variations in climate during the Quaternary period. Such changes occurred during the Ice Age and its aftermath, and thus it might be expected that relationship with the accumulation and wasting of intimate relationship Scotland, owing to its intimate That disturbance. That climatic disturbance. experienced climatic has also experienced ice sheets (Price 1983), the vast ice 1983), has the period due interglacial period the present interglacial well have continued into the due to the time needed could well for recovery from the severe conditions of the Ice Age, especially as Scotland is located in a particularly sensitive climatic zone. This chapter will explore the possibility of climate change during the nine millennia following the removal of the last permanent ice from Scotland. A first consideration will be the interplay between the circulation. country’s geographical location and the effect of patterns of atmospheric circulation. This knowledge must then be mediated by acknowledging the difficulties involved in extrapolating back many thousands of years. Such extrapolation requires the use of extrapolating proxy evidence which itself may be incomplete and difficult to exploit. Thus a second focus of enquiry will be the sources of evidence which are available in potential and to their inherent problems. Climate change may also relation to their potential operate over different time-scales and so it will be considered from long-term and operate short-term viewpoints.

THE GEOGRAPHICAL LOCATION OF SCOTLAND AND ITS CLIMATIC IMPLICATIONS Scotland’s location at the western extremity of the Eurasian land mass and its proximity to the northern Atlantic Ocean, has profound significance for both its emphasize the Furthermore, it is important and long-term climate. shortthe important to emphasize climate. Furthermore, short- and country’s northerly position which gives it a latitudinal equivalence with Labrador. Edwards and Ian B. M. Ralston. Scotland: Environment and Archaeology, 8000 BC — AD 1000. Edited by Kevin J. Edwards contributors. Published in 1997 by John © 1997 The editors and contributors. John Wiley & Sons Ltd.

WHI'ITINGTON AND K. J. EDWARDS G. WHITTINGTON

12

WW

30"W 30°W

0°

30°E 30"E

60“E 60°E

North Pole represents represents a flow line of the strongest surrounding the North Figure 2.1 The solid line surrounding which that deformation which the deformation winds in the example of the dashed line is an example The dashed vortex. The circumpolar vortex. the circumpolar winds

flow can undergo. numbered lines represent positions of the Polar Front at different undergo. The numbered flow times. 1: refers to the position position from 11 000—10 000 BP; 2: 13 000—11 000 BP; 3: 10 000—9000 BP; 4: 9000—6000 onwards 6000 BP onwards 9000—6000 BP; 5 from 6000

Taken together, features mean that there is not only exposure to both oceanic together, these features with associated with elements associated and continentally factors, but that the elements generated climatic factors, continentally generated are circumpolar vortex, are the circumpolar mechanisms, the one the earth’s most important climatic mechanisms, one of the heminorthern herniflow, in the northern of immediate system, air flow, part of this system, importance. As part immediate importance. sphere, sphere, occurs in a series of waves which migrate around the pole, moving from low-pressure west to east in an irregular The waves generate low-pressure (Figure 2.1). The pattern (Figure irregular pattern areas, where polar air extrudes southwards towards the equator, and ridges of high pressure which allow tropical air to intrude towards the pole. that The contrast contrast between the temperature and humidity properties of air that that means that originates in the polar regions and air that develops in the subtropics, subtropics, means Front. their contact generates a further important major climatic feature — the Polar Polar Front. but The winter than in summer, but The location of the Front varies, being further south in winter

it has also altered its position many times in relation changes in long-term changes relation to long-term atmospheric Figure 2.1). The importMcIntyre 1981; Figure atmospheric circulation (Ruddiman and McIntyre creates ance of the Polar Front to Scotland’s Scotland’s climate lies in the fact that it creates heat mid-latitude travelling depressions; these act as a mechanism transferring heat mechanism for transferring from low latitudes and cold air from the opposite structure of these opposite direction. The structure depressions ensures an with considerable air, with and colder air, warmer and of warmer an alternation of Thus track. Thus levels that lie in their track. areas that winds, for the areas strong winds, precipitation and strong levels of precipitation

CLIMATE CHANGE

13

large determines to a large long-term perspective, determines any long-tenn Front, in any Polar Front, the Polar of the the location of the climate experienced by the Scottish land mass. extent the climate depends upon the another depends climatic regime prevails over another Whether one type of climatic decades or even constant. Over decades circulation remains constant. length of time that atmospheric circulation dominant centuries, when the movement of depressions along the Polar Front is the dominant winters and cool, damp mild, wet winters climatic process, Scotland will experience mild, climatic

precipitation types summers. With anticyclones prevailing, winters will be cold, with precipitation summers. depending upon the location of the anticyclone’s centre, and summers and amounts depending will be warm or even hot.

SCOTLAND’S HOLOCENE CLIMATE

disappearance of permanent ice from Scotland was inevitably followed by an The disappearance general conamelioration of climate. Much has been written on this topic and a general nature of the Holocene sensus had been reached which allowed a synthesis of the nature view has been chal(Lamb 1977). This View north-west Europe as a whole (Lamb climate in north-west Postglacial re—evaluation of Scotland’s climate in the Postglacial lenged in recent years and a re-evaluation period is necessary. The traditional view

botanists, A. Blytt and The work in the early twentieth century of two Scandinavian botanists, Semander, led to the widespread acceptance of four distinct climatic R. Sernander, climatic periods for anticyclonic north-west Europe since 10 000 BP. During a Boreal phase, in which anticyclonic became warmer and summers those warmer than those summers became dominated, temperatures rose rose and conditions dominated, following The following frosts. The winters, although milder, while winters, today, while milder, still witnessed frosts. of today, provide the considered to provide period (7500—5000 BP; 6270—3780 cal BC) was considered Atlantic period Postglacial Climatic Optimum during which the summer temperatures, already at °C greater than those of today (cf. their maxima, were some 2—3 °C al. (cf. Atkinson et al. and rainfall was plentiful. Between about 1987), while the the ‘cold’ seasons were mild and 1987), sub-Boreal), conditions akin to those of the 5000—2500 BP (3780—630 cal BC) (the sub-Boreal),

climate was supposedly less dry due to Boreal were thought to prevail, but the climate circulation. Finally atmospheric circulation. latitudinal and meridional atmospheric oscillating patterns of latitudinal came the sub-Atlantic period (c. 2500 BP to the present) during which summer 0C, rainfall much as 2 °C, temperatures fell by as much winters rainfall totals increased and mild winters

predominant factor in Front as a predominant establishment of the Polar Front prevailed, following the establishment currency, Scotland’s climate. This climatic division has tended to remain in currency, Scotland’s and Van Van Peer Vermeersch and 1982; Vermeersch the archaeological literature (Magny 1982; especially in the (Serebryanny and Orlov 1993), 1990), but also in the writings of other disciplines (Serebryanny although like the cultural divisions of Neolithic, Bronze and Iron, more through convention convention than proven scientific accuracy. A revised view

appeared as early as 1949, when Firbas when Firbas position appeared traditional position A modification modification of the traditional Vorwarmezeit proposed a simpler tripartite division of the postglacial period into Vorwarmezeit

14

G. WHI'ITINGTON WHITTINGTON AND K. J. EDWARDS

Nachwarmezeit (the Wirmezeit (the warmest times) and Nachwéirmezeit (the time of warming), Wéirmezeit time of cooling). By 1964, McVean was commenting on how little change took place in the Scottish forests during Postglacial times. Recently, a more radical reappraisal has occurred. This has been stimulated by the recognition that the inferences about climate change have been based overrecord, and stratigraphical record, whehningly whelmingly on assumptions derived from the pollen stratigraphical Furthermore, these may be inherently flawed (Birks 1982; Davis and Botkin 1985). Furthermore, climatic modelling (Kutzbach and Guetter 1986; COHMAP 1988), the use of climatic the use 1983) and the and the Birks 1983) and Birks (e.g. Huntley and databases (e.g. development of biological biological databases

(e.g. Guiot et al. 1989; Pons et al. 1992) have led statistical treatment of pollen data (e.g. behaviour. Current thinking confirms earlier ideas to sharper insights into climatic behaviour. both on the fast rate of temperature and precipitation increase between 10 000 and 7000 BP (5840 cal BC), and also that by the latter date Scotland would have been drier and warmer than at present (Birks 1990, Figure 1). Since that time, however, it is argued that changes in climatic regime have been only minor.

CHANGES LONG-TERM CHANGES

If the pattern of climate change for north-west Europe also applies, as seems probable, to Scotland, an examination of the Scottish proxy evidence might be corroborative evidence. Three such sources will be examined in expected to provide corroborative important part played by vegetational this section and, because of the extremely important history in the exploration and defining of past climate change, evidence drawn from history that source will be considered first.

The evidence from vegetational change The nature of the evidence

The temperature and moisture requirements of plants make them potential indicators of climate change. While some plants have adapted to a wide range of cators conditions, others are extremely sensitive to climatic fluctuations, especially if they are near the limits of their climatic range. Most of Scotland was glaciated until ameliorating about 18 000 years ago (Chapter 3) and since that time a generally ameliorating colonization by plants from the east and south, especially up to climate has allowed colonization c. 7000 BP (5840 cal BC) when Britain was finally severed from the European occurred in Postglacial mainland (Funnel 1995). If major climatic oscillations have occurred times, it might be expected that changes in Scotland’s vegetation patterns would reveal them. There are two main routes to the investigation of vegetational change, one at a microscopic and the other at a macroscopic scale. biostratigraphical evidence provided by The microscopic scale lies in the study of biostratigraphical pollen analysis. Plants release pollen and spores which are incorporated into chronostratigraphical record accumulating accumulating peat and lake deposits, thus providing a chronostratigraphical of the vegetation of an area. The speed at which and the length of time over which the pollen-preserving matrix accumulates determines the detail to which the conditions, such as where optimum conditions, taken. In optimum reconstruction can be taken. vegetation reconstruction vegetation

CLIMATE CLIMATE CHANGE

15

annually varved sediments are available, such reconstruction can be very sensitive indeed (Peglar 1993). This line of enquiry into climate change does, however, present problems. Depositional hiatuses may occur, leading to breaks in sedimentary sequences. More conditions may fundamental perhaps is the lag effect in plant colonization. Climatic conditions well be suitable for particular plant species to thrive but this will depend upon many conditions to factors. It is necessary for seeds to arrive in an area and also for soil conditions be suitable for their their success. success. Furthermore, trees, example, not only need to trees, for example, migrate into the flowering flowering stage; stage; this will time before they reach the require time but require area but into an area migrate be accomplished only tardily in areas where the climate is ameliorating slowly, but more speedily where conditions change rapidly. Different trees and shrubs, even in their optimum environment, may also achieve reproductive maturity at varying rates: Quercus (oak) achieves this state more slowly than do either Betula (birch) or Corylus avellana (hazel) and thus, although present in an area, it may be temporarily palynologically silent. This problem of the sensitivity of the biostratigraphical record is illustrated very clearly by the Postglacial record of Alnus glutinosa (alder). Birks (1989) mapped the recolonization of Scotland by alder and showed that it had arrived in much of the country by about 6500 BP (5440 cal BC), but subsequent work (e.g. Whittington et al. 1991a) indicated that in one area of eastern Scotland it was present about a thousand years earlier. It is now realized that this species, previously accepted as an indicator of increasing wetness, has an extremely complicated history which may 1989; Bennett and Elliott 1989; not be directly related to climate climate change (Chambers and Elliott 1992). Birks 1990; Edwards 1990; pace Tallantire 1992).

At the macroscopic level, the remains of trees, preserved in peat (Plate 2.1) or lacustrine sediments, also serve regimes. The previous climatic lacustrine sediments, serve as evidence of previous climatic regimes. The

current treeless nature of the Outer Hebrides contrasts with the pine stumps still to be found buried there in peat (Wilkins 1984) and the recording, by Beveridge (1911), of birch and hazel wood in intertidal deposits (see also Fossitt 1996). Similarly, the higher areas of Scotland are today treeless although, well above the peat are root remains in blanket peat growth, there are pine growth, limit of pine altitudinal limit current altitudinal current (Pears that this evidence to show that use this been made to use 1970). Attempts have been (Pears 1968, 1970). greater warmth existed in the earlier Holocene, but the dates of these macrofossils are not well established and there is a danger of assuming that they are all contemporary rather than representing different phases of adjustment to climatic amelioration 1993). (Lowe 1993). amelioration (Lowe The exploitation of the evidence

With this awareness that past vegetation vegetation patterns can be difficult to use as proxy evidence for climate change, it is now possible to attempt to exploit it by addressing some of the main features of the until-recently accepted climatic variability that would have affected Scotland. The first point which might be examined is that of a rapid climatic amelioration by about 10 000 BP, since when conditions have remained more or less constant at the demand arboreal taxa, which demand broad-leaved arboreal the broad-leaved the case, the this is the macro-scale. If this the macro-scale.

thermophilous conditions, should not only have spread (sensu Birks 1989) into

16

G. WHITTINGTON AND AND K. J. EDWARDS EDWARDS

flax“

stumps appearing Plate 2.1 Pine stumps appearing from beneath peat peat at Clatteringshaws Clatteringshaws Loch, Stewartry Stewartry of Kircudbright. Copyright: Copyright: K. J. Edwards Kircudbright. Edwards

Scotland by the time of the supposed supposed Climatic Climatic Optimum, Scotland Optimum, but maintained maintained their presence and distribution distribution thereafter. Betula was established established in the central and eastern areas by 10 000 BP and in the eastern areas BP and the north and west west by 9500 BP (8560 (8560 cal BC). Corylus avellana was present in western Scotland Scotland at 9500 BP and by 9000 BP (8030 cal BC) had spread to the south and east. UImus Ulmus (elm) had advanced advanced to cover cover whole of the mainland mainland by 8500 8500 BP (7530 virtually the whole (7530 cal BC) and Quercus Quercus was north Forth—Clyde lowlands lowlands by the same date. north of the Forth—Clyde date. The The later colonizing colonizing TiIia Tilia (lime) and and Fraxinus Fraxinus excelsior (ash) had had a restricted restricted spread spread into Scotland Scotland (Birks (Birks may well serve to emphasize emphasize that it is not climatic conditions 1990). This This may conditions alone alone which control the establishment of vegetation, but that conditions of soil, soil, seed

spread and competition are as important in many cases. This is also an example of where the proxy evidence may always be incomplete and therefore misleading. It is noticeable that Birks’s maps of spread spread for Tilia and Fraxinus reveal few sites from the eastern eastern lowlands. investigated after the lowlands. Two Two sites sites in Scotland, Scotland, investigated after Birks’s Birks’s data data collection, have T Tilia collection, ilia present by 7000 BP and 6200 BP (5840 and 5140 cal BC) and Fraxinus by 8400 BP and 7000 BP (7460 and 5840 cal BC) (Whittington (Whittington et al. 1991a,b). It therefore had achieved the threshold 1991a,b). therefore appears that that temperatures temperatures had threshold demanded by the warmth-loving trees for them to become established at an early date in the Holocene. This lends support to the contention that the rise in temperature had indeed been rapid. Furthermore, an examination of pollen diagrams from across Scotland reveals stability in the nature of the woodland woodland until about 5000 BP (3780 cal BC). At that time, the effects of human human activity in the undermine the worth of vegetation patterns as proxy climate landscape appear appear to undermine climate pointed out, that even after data. It must be pointed out, however, however, that after that date there is no

CLIMATE CHANGE

’ '

17

fundamental change 1n in the distribution distribution and composition fundamental composition of the arboreal arboreal and shrub shrub r ' taxa which survived. which survived foregoing evidence suggests that the main features of Scotland’s Postglacial The foregoing climate were not only established established at an early date climate date but that they have shown very little alteration alteration since then. That such a statement is largely derived from pollen and associated data makes it vital to remember remember the problems associated problems created by vegetation vegetation inertia 1985), and the debate about vegetation vegetation spread inertia (Smith 1985), spread as against against vegetation vegetation and climate equilibrium (Davis and Botkin 1985; Webb 1986), and atmospheric atmospheric carbon variability (Isdo 1989), 1989), as well as those carbon those inherent inherent in the method method of pollen pollen analysis. problem Pertinent to this topic of Holocene climate Pertinent climate change, however, remains the problem of such a long-accepted long-accepted division as the Atlantic period period and its wetness. Can it be framework that is emerging? Two main dismissed in line with the new climatic framework main sources of proxy data are useful in this matter; one will be discussed here and the sources second in the following section. The spread and rapid expansion of Alnus, because of its affinity with wet sites, has been used as an indicator of a major change in climate brought brought about about by a marked increase however, the initiation already stated, however, increase in rainfall. rainfall. As already initiation and spread of Alnus were as synchronous as might be expected not as certainly not were certainly expected if climate climate change change over

Scotland as a whole, associated with a long-term pluvial regime, was the main cause. The proxy data is again evident. the proxy The need to consider carefully the nature of the evident. Pollen survival requires the existence of anaerobic conditions. These exist in bogs flourishes. Thus the alder and lacustrine sediments, the very areas areas in which Alnus flourishes.

appearance and rise, although real, may well be no more than a response to appearance hydrological changes of a local nature and not related to a fundamental fundamental shift in climate. The danger of modern climatic climatic perceptions clouding the interpretation of past climatic periods was emphasized emphasized above. If pollen-analytical data are used circum— circumspectly, important correction factor. Contrary to current function as an important can function they can spectly, they current perceptions season and length of the growing concerning the length perceptions concerning growing season and exposure exposure to high winds there (Lowe Hebrides and Shetland the Outer Hebrides shows that the record shows the pollen record 1993), the (Lowe 1993), Shetland were 1996a; Fossitt 1996), 1992; Edwards 1996a; 1990, 1992; (Bennett et al. 1990, were wooded (Bennett 1996), even down down to the western shoreline 8000 BP from 8000 unpublished), from Edwards, unpublished), (Whittington and Edwards, shoreline (Whittington

(6840 cal BC), if not earlier, and remained so until at least 5000 BP (3780 cal BC) — the time at which human interference landscape became much more intense. interference in the landscape intense. The evidence of peat development

The nature of the evidence A distinctive feature of Scotland’s landscape is the frequent occurrence of soligenous soligenous wetlands (valley mires) and bogs. Of these, bogs should be exploitable as a source of information infrequent, information on climatic variation. They occur in two forms: the infrequent, raised (ombrogenous) (ombrogenous) bogs are often large, as in the wetter west, e.g. Claish Moss (Argyllshire) (Wigtownshire), but smaller and less frequent in (Argyllshjre) and Moss of Cree (Wigtownshire), blanket bogs occurring blanket frequently occurring the frequently Bankhead Moss (Fife); the e.g. Bankhead east, e.g. the drier east, the are widespread (e.g. in the Outer Isles and the Cairngorm and Monadhliath Monadhliath

18

WHITTINGTON AND K. J. EDWARDS G. WHITTINGTON

Mountains), but can also appear as deep unbroken stretches, as on Rannoch Moor. Mountains), particularly high conditions, particularly atmospheric conditions, bogs rely on atmospheric Unlike the mires, bogs valley mires, the valley Unlike and continued existence. An initiation and rates, for their initiation rainfall and and low evaporation rates, the ombrogenous ombrogenous particularly of the bogs, particularly stratigraphy of bogs, examination of the the vertical stratigraphy examination vegetational comconstituent vegetational type, can often reveal considerable differences in its constituent humification and compaction which have ponents. It also exhibits differences in the humification taken place. Blanket bogs also cover large areas which peat digging (for domestic photography reveal as formerly and agricultural agricultural purposes), natural erosion and air photography agriculture. As the bogs are a response to having been exploited for settlement and agriculture. moisture availability, it might be expected that all of the above features of bog and humidity humidity amounts and precipitation amounts variability in precipitation stratigraphy would be related to variability would be initiation should therefore, the recovery of dates of initiation levels (cf. Aaby 1978). In theory, therefore, Furthermore, significantly wetter. Furthermore, indicate the time at which the climate became significantly development, cessation in peat development, drier periods should cause a slowing down or even cessation should be apparent in the marked by a change in the vegetation succession, which should period, an peat stratigraphy and should therefore be datable. After such a dry period, increase in wetness should bring about regrowth at the peat surface, especially with known as a recurrence recurrence what is known leading to what thus leading regard of Sphagnum, thus resurgence of regard to a resurgence 1993). surface surface (Gore 1993). evidence of the evidence exploitation of The exploitation The

Exploitation of these features for the deriving of proxy climatic data can be difficult. The credibility of the Atlantic and sub-Atlantic phases as contributors to climate change in Scotland has depended to a large extent upon the growth of bogs, climate equation because this equation problems in this blanket. There are major problems and blanket. ombrogenous and both ombrogenous peat growth can occur as a result of the interplay of factors other than those of a direct climatic origin. Peat may be a response to a sudden increase in rainfall, but might also represent the climax stage of soil development as part of the long-term operation operation of pedogenic processes in which leaching led to acidic and anaerobic conditions becoming dominant (Ball 1975). If that is the case, there is no need to deterioration in climate for the onset of peat. It could develop at any posit a sudden deterioration time when the appropriate critical conditions had been reached. That would also help to explain a further factor which tends to complicate the recourse to peat as a particular date. The initiation of blanket bog at one sign of increasing increasing wetness at a particular whole over an area as a whole representative for that event over location may not be representative location

(Whittington and Ritchie 1988; cf. Solem 1986). Thus, before peat initiation can (Whittington widespread undertake a widespread vital to undertake wetness, it is vital sign of increasing wetness, offered as a sign even be offered even areal survey.

climatic features are crucial in peat formation, they also interact with Although climatic determining the start and rate of peat conditions in determining topographic and soil porosity conditions topographic late Holocene phenomenon mainly a late peat inception is mainly that peat growth. It is notable that exploit(Watts 1988) and thus commonly occurs during the period of human land exploitsouth-east Perthshire Dalnaglar in south-east Perthshire Prehistoric dwellings, for example at Dalnaglar ation. Prehistoric Sithean on Islay An Sithean 1984), at An (Stevenson 1984), (Stewart 1962), at C111 a’Bhaile on Jura (Stevenson (Stewart 1962), al. 1986), have Shetland (Whittle et a]. (Barber and Brown 1984) and Scord of Brouster, Shetland agriculture, all of which lie arable agriculture, or land that was was used for arable systems or field systems associated field associated

CHANGE CLIMATE CLIMATE CHANGE

19 19

buried under blanket blanket bog. It is conceivable that over-exploitation of the soil in such perhaps also soil compaction which hampered hampered drainage, may may well well compaction which engendering soil also engendering areas, perhaps have induced peat initiation and accumulation. accumulation. Thus peat inception may be due, not change, but to alteration (Moore 1993). 1993). local hydrology (Moore alteration to the local to climate change, The existence of more than a single recurrence recurrence surface in a stratigraphic stratigraphic sequence could be interpreted cyclic changes changes in precipitation precipitation amounts. amounts. Again, Again, interpreted as evidence of cyclic could great care has to be taken before such a conclusion is drawn. drawn. In areas where wood identified in employed as fuel. Hiatuses identified has been scarce, peat has commonly been employed accumulation of peat which are followed by regrowth, may be no more than the the accumulation result of human interference with the natural processes of the peat formation. The potential of bog structures and characteristics characteristics has been reviewed by K. E. Barber (1982) and together with later work (e.g. Barber 1985, 1994; Haslam 1987; Stoneman et al. a]. 1993; use regard to the use Barber et al. 1994b), 1993; K. E. Barber 1994b), especially especially with regard of recurrence recurrence surfaces, has shown that the results obtained may be best when used at a local scale and for determining short-term short-term changes. New methods are being used change G31ackford proxy evidence bogs as providers of proxy explore bogs used to explore evidence for climate change (Blackford 1993). Until they are the existence of major wise not be wise would be realized, it would are realized, 1993). not to rest the major shifts climate, especially pluviality of Scotland’s climate, shifts in the pluviality especially such such as has been envisaged envisaged for the supposed Atlantic period, upon evidence drawn from sources derived from the peat record. event that be shown to have unlikely event the unlikely Even in the record. Even the peat that peat peat growth growth could be been initiated over Scotland synchronously, especially at the start of an Atlantic Atlantic period, that still might not be evidence for greater rainfall but merely of lowered rates 000 and and 6000 and 4870 cal BP (12 040 6000 BP Between 12 000 rates of evaporation. Between 040 and cal BC), the tilt of the earth’s earth’s axis present and than at present was greater than axis was and perihelion took took place in the northern have led would have and Guetter 1986). (Kutzbach and northern summer (Kutzbach 1986). This would led to intense intense anticyclonic the colder, the been colder, have been would have winters would and, although the winters anticyclonic conditions and, summers summers would have been much warmer than today with a comcomitant increase in evaporation evaporation rates. A climatic change took place but not necessarily one which rainfall. involved an excessive increase in rainfall. The evidence of faunal fauna] change The evidence

Many fauna operate operate within distinct climatic limits and so can be regarded as surrogates for climatic data. Certain mammals fall into this category, but probably probably only have been made for Suggestions have value for Scotland. Suggestions be of value reindeer would be the reindeer only the their late Postglacial (Lawson the late into the survival into their survival (Lawson 1984; 1984; Whitaker 1986). 1986). This would would create the early show the proxy climatic data which show potential conflict with other proxy create a potential early Holocene have been a period which would warmth, a condition which period of considerable warmth, tohave Holocene to have had major implications for the reindeer’s food source. Proof that reindeer have been present at all since the last Ice Age is difficult to acquire as the survival survival of bone or antler in Scotland’s acidic soils is unlikely. Cave deposits provide a better preservation preservation matrix and thus focus interest on the cave system of Creag Creag nan Uamh in Assynt. Reindeer remains exist there and examination has shown that the latest of them dates to 8300i90 830021:90 BP (7480—7100 cal BC) (Lawson (Lawson 1981; Murray et al. 1993). A review of all other potential remains from Scotland Scotland has failed to provide a later occurrence occurrence other than from those which are intrusive (Clutton(CluttonBrock and Thus, this faunal evidence corroborates 1988). Thus, MacGregor 1988). and MacGregor corroborates that from

20

G. WHITTINGTON WHI'ITINGTON AND K. J. EDWARDS

greater than those of today in the other proxy sources which indicate temperatures greater other early Holocene. can be and can preserve well, and mammals, insect remains preserve those of mammals, In contrast to those abundant, abundant, in all organically rich deposits that are waterlogged (Coope 1975). Thus, there is a certain similarity between fossil insects and pollen as climatic data surrogates but the former possess a distinct advantage in this category. The greater surrogates maturation, certainly compared with trees, and the capability speed of reproductive maturation, for individual movement, which is not dependent upon an indirect dispersal effect observed in the case of pollen is that the problem of the lag effect mean that mechanism, mean mechanism, dependent upon particular less significant. There are also insects which are not dependent plants for their food supply so that they can respond to shifts in climatic limits very plants quickly. Where pollen proves to be a superior source of evidence is in its ready quickly. Chironomidae (midges) are availability in small quantities of sediment. While the Chironomidae similar in this respect, the Coleoptera (beetles) can only be obtained in an adequate availability of the availability demands the material, which demands of material, kilograms of from several kilograms number from number sediment exposures. relatively rare and suitable sediment As a result, climatic evidence drawn from fossil insect study currently available for Scotland is very limited. An investigation carried out in the South-West has, been a rapid climatic had been confirmation that there had however, provided further confirmation provided further amelioration by about 10 000 BP. The presence of the Carabid Odacantha melanura Kirkcudbright (Bishop and Coope 1977) at‘ Brighouse Bay in the Stewartry of Kirkcudbright at had been attained that temperatures as great as those of today had allows the inference that by 9640 BP (8950 cal BC). This method clearly possesses great potential for insights palaeoclimates, but unfortunately palaeoentomologists are even into Scotland’s palaeoclimates, rarer than the exposed deposits upon which they depend.

SHORT-TERM CHANGES

The proposal that that Scotland’s climate may well have witnessed long-term macrostability during the Holocene does not preclude intermittent or even periodic stability changes of a lesser nature, on national or regional scales. Many such events are likely to remain undetected as the proxy evidence has too low a resolution status. There are, however, indications from surrogate climatic data that episodic flucstable isotopes records of the instance, records occurred. For isotopes of the ratios of stable tuations have occurred. For instance, tuations oscillations in probable significant oscillations Tower, Fife, reveal probable oxygen (6180) from Lundin Tower, from Lundin temperature in the early Holocene (Whittington et al. 1996). temperature Vegetational evidence Vegetational

constituents of Scotland’s the constituents have cover have suggested above Scotland’s woodland cover was suggested above that the It was distributional change since early in the Holocene. Attention has been shown no distributional recently (Gear and Huntley 1991) to the behaviour behaviour of pine forests in the drawn recently and within a cal BC), and BP (2490 cal the period period around 4000 BP north During the north of Scotland. During

period of 400 years, the tree line expanded over a distance of 70—80 km and then root precipitation totals is at the root retreated. It is proposed that marked variation in precipitation application of of this phenomenon. Perhaps confirmation of this comes from the application

CLIMATE CHANGE CHANGE CLIMATE

21

isotope ratios of deuterium in Ferguson (1985), using isotope isotope analysis. Dubois and Ferguson pine stumps from the Cairngorms, Cairngorms, have concluded that a wet episode did occur other They have 2330 cal BC). They between 4250 isolated other have also isolated (2880 and 2330 3870 BP (2880 and 3870 4250 and 6250—5800 BP (5220—4690 cal BC) and 1330 BP (cal periods with higher rainfall, at 6250~5800 establishing of these wet episodes episodes does not necessarily 680). The establishing AD 680). does not necessarily mean periods of greater activity on the Polar Front. Possible alternative explanations for the changes greater changes in rainfall, rainfall, and and also in pine pine distribution, distribution, on either either side of 4000 BP BP have have been been advanced (Blackford et al. 1992). These involved the eruption advanced eruption of Icelandic volcanoes and the injection of tephra into the atmosphere. atmosphere. The effect of this would would not only have been to increase the availability of hygroscopic nuclei but also to have ensured enhanced acidity acidity in any any ensuing ensuing rainfall, ensured rainfall, either either or both both of which may have have had a damaging growing close to their their range range limits. limits. If this pine trees growing on pine effect on damaging effect had connection is justified, any extrapolation of this event to support climate climate change over Scotland as a whole is dangerous. The crucial need is to establish the spatial spatial extent of the tephra effect, and even then it may only produce evidence from changes in arboreal cover in areas where the trees were near a critical threshold threshold (Edwards a]. 1996). 1996). (Edwards et al. evidence of peat development The peat development The evidence Large areas of Scotland, for example example the uplands to the north to the north of Strathmore Strathmore (Harris 1984) Brora, Naver, Halladale and region confined by the Rivers Brora, 1984) and the region

Helmsdale in the extreme north, still bear the impression of past cultivation practices where altitude or soil conditions now make such activity unrealistic. The peat cover such areas might involving climate change be regarded as involving might still be cover encountered in such despite the generation, especially peat generation, complications noted in peat the complications especially those related to agricultural agricultural exploitation. That having been said, the need to invoke long-term climate unnecessary. The tephra from the Icelandic eruptions change may still be unnecessary. climate change

may not only have increased rainfall, making agriculture in the short-term short-term very unrewarding, unrewarding, but its deposition might have also increased the acidity of the soil (cf. Grattan and became inevitable Gilbertson 1994) to a point where and Gilbertson where peat initiation became inevitable phenomena in such an incident would have had longand irreversible. The climatic phenomena term consequences equally longinterpreted as arising from an equally consequences but should not be interpreted term climatic shift. sand movement The evidence of sand

Evidence for periods increased activity and therefore of increased storminess, and increased storminess, periods of increased along the the location of the the location shift in the temporary shift probable temporary or even a probable Polar Front, or the Polar Polar Front, the most obvious of the One of sources. One proxy sources. from proxy identify from to identify difficult to Front, is difficult features of such events are wind-thrown trees but it is rare for conditions conditions to exist which would preserve the evidence (Lamb 1966). Most important important as a source might be shorelines, unconsolidated masses along shorelines, located in unconsolidated and located highly mobile and sand; it is highly be sand; which are also vulnerable to exposure exposure to high winds. Some 20% of the Scottish Scottish shoreline displays sand as either dunes or links, as at Forvie (Aberdeenshire), along along the Atlantic shores of the Outer Hebrides Hebrides and Luce Bay (Wigtownshire). Only under extremely favourable circumstances circumstances can sand movement be used as a climatic

22

G. WHITTINGTON AND K. J. EDWARDS

conditions emanating from one direction can be surrogate. surrogate. Response to stormy conditions negated by winds from a contrary direction, a feature which can occur very rapidly passage of deep depressions across Scotland. It is also necessary to have during the passage Fortuitous events can be measured. Fortuitous datum against which the onset of a sand blow can a datum has allowed a study may make this possible. Recent erosion, caused by deflation, has may the coastal dune area of Tentsmuir in Fife. 1996) in the and McManus 1996) (Whittington and (Whittington This with millennium AD, the dune belt with middle of the first millennium that until the middle shown that This has shown

pasture. After that date, its Calluna (heather) cover was being exploited as managed pasture. a massive sand blow engulfed the area. This change occurred in the same period as Chambers and Chambers and Wales (Blackford and Ireland and deposits in England, Ireland bog deposits the study of bog the storminess and wetness. 1991) indicated as having been a period of increased storminess The coastal sands of the northern islands appear to offer clear evidence to deposits of storminess, in that deep deposits increased storminess, support statements as to periods of increased climatic change surfaces. Thus, a climatic sand frequently overwhelmed human occupation surfaces. Dickson 1979) and that line of (Keatinge and Dickson is argued argued for at 5000 BP (3780 cal BC) (Keatinge example, at Northton on Harris sites; for example, can be extended to other sites; reasoning can

(Crawford and Switsur 1977), Rosinish on North Uist (Crawford (Simpson 1976), Udal on North Benbecula (Whittington (Whittington and Ritchie 1988) and Knap of Howar on Papa Westray relationship between increased storminess and (Ritchie 1983). Unfortunately, the relationship (Ritchie major sand movements is not one of direct cause and effect. Also involved in this human effects of human erosional effects the erosional the movements, the sand supply to provide the process are sand activity in creating conditions for sand deflation, and changes in sea level during the location and extent mid Holocene marine transgression which altered not only the location Ritchie 1988). Whittington and Ritchie of exposed sand, but also its mobility (Ritchie 1979; Whittington Periods of major sand movement may have been no stormier than those in which the sand was stable. Even a conjunction of dates for major sand blow across a wide spatial range of sites could well be related to coastal woodland clearance which molluscan evidence has shown as occurring during the Neolithic (Evans 1979).

CONCLUSION Holocene climate of explanations for changes in the Holocene The description of and explanations interim stage but also in a state Scotland must be regarded not only as being in an interim available is still at a crude level of interpretof flux. Much of the proxy evidence available that of source, that principal source, The principal misinterpretation. The from misinterpretation. suffers from ation and perhaps perhaps also suffers ation pollen data, has to be viewed much more critically and it has to be recognized that its use, in establishing long-term climate change, has in the past involved a large although currently degree of circular argument. The infusion of climatic modelling, although only at the macro-scale, is not only bringing new insights but is also encouraging encouraging a The extension sacrosanct conventions. The and rather sacrosanct of long-held and radical examination of however, it must be Scotland, however, of isotopic studies also contains great potential. For Scotland, palaeoclimatology is still in its infancy. confessed that palaeoclimatology

3

Geomorphology and Landscape A' Change COLIN K. BALLANTYNE AND ALASTAIR G. DAWSON

INTRODUCTION Throughout the period of known human occupance, the magnificent variety of the Throughout inhabitants. constraints for its inhabitants. and constraints opportunities and has posed both opportunities landscape has Scottish landscape Scottish During this time, moreover, the physical characteristics of the landscape have

undergone significant changes, some of which have strongly influenced the pattern of human activity whilst others reflect the consequences of such activity. This chapter explores the geomorphological background to the early human occupance threefold: the first part of the chapter outlines the geoof Scotland. Its focus is threefold: landscape; the second is devoted to the ‘' characteristics of the Scottish landscape; morphological characteristics disappearance of the last nature of landform changes during the period since the disappearance the form has affected the ways in which human activity has considers ways and the third considers glaciers; and interglacial of the Scottish landscape during the first nine millennia of the present interglacial period.

LANDSCAPE CHARACTERISTICS OF THE SCOTTISH LANDSCAPE PHYSICAL CHARACTERISTICS

The physical landscape of Scotland is the product of a long, complex and someunderstanding of its present An understanding (Harris 1991). violent geological history (Harris times violent form present form 1991). An times characteristics of the underlying rocks, and involves three main elements: first, the characteristics denudation of the resistance to erosion; second, the denudation their variable resistance particular their in particular Scottish land mass during the last 70 million years; and finally the effects of glaciation by ice sheets during the Pleistocene Epoch. relief Geology and relief

Traditionally, Scotland has been seen as comprising three major relief units (i.e. the Traditionally, Highlands, the Midland Valley and the Southern Uplands), but geologically a Highlands, units is oldest of these units The oldest landscape is more appropriate. The the landscape division of the fivefold division fivefold the Hebridean Craton in the extreme north-west (Figure 3.1). A fragment of a Ralston. M.. Ralston. Ian B. M and Ian by Kevin 1. Edited by 1000. Edited —- AD 1000. and Archaeology, 8000 BC —— Environment and Scotland' Environment J . Edwards and Scotland: © 1997 The editors and contributors. Published in 1997 by John Wiley & Sons Ltd.

/

”

24 p

”

BALLANTYNE AND A. G. DAWSON C. K. BALLANTYNE DAWSON E

Tertiary Volcanic Province: g lavas and intrusive rocks '

Southern Uplands: shales and of Silurian or Ordovician age greywackes oi Caledonian Fold Belt: metamorphic rocks of Moinian or Dalradlan Dalradian age

“ .

Sedimentary rocks of garboniferous or Carboniferous evonian age Devonian

I—

l

oi Carboniferous or Lavas of Devonian age

Hebridean Craton: mainly Lewisian Gneiss and Torridon Sandstone

N '

Principal faults

l

\

at H

o

8° _ -_ “' a: B

— — '

7

m

”

,

1

t

‘

“I

\

\\‘

6% 37a

g

\ I I

.

ea“

5%? gas

%

Granitic plutons

‘‘

a

A

A

/ ’’ \ Main area of theTertiary K.” l Volcanic Province K. —- 1

50 km ll

'

0 II

— —

l

i

distribution of major rock Figure 3.1 The structural provinces of Scotland, showing the distribution types. Geological boundaries based on Craig (1991)

GEOMORPHOLOGY GEOMORPHOLOGY

LANDSCAPE CHANGE CHANGE AND LANDSCAPE

25

rising above Plate 3.1 3.1 Quartzite Quartzite mountains mountains rising above a platform platform of glacially scoured scoured gneiss bedrock, Ballantyne Copyright: C. K. Ballantyne Arkle, Sutherland. Sutherland. Copyright:

primeval continent, this comprises comprises undulating rocky primeval rocky lowlands lowlands underlain underlain by resistant resistant crystalline rocks (Lewisian Gneiss) which are locally surmounted by bold sandstone and quartzite quartzite mountains mountains 700—1100 In m in altitude and altitude (Plate 3.1). A thrust thrust fault zone zone known as the Moine separates the Hebridean Craton from the second of the Moine Thrust separates

five structural structural units, the Caledonian Caledonian Fold Belt. This five This makes up most of the Scottish Scottish Highlands and represents the eroded remnants of an ancient mountain range of alpine stature. It is divided into two major subunits, the North-West Highlands Highlands to the Great Glen Fault, and and the Grampian Highlands to the south and east north of the Great (Figure 3.1). Both are dominated by folded and faulted metamorphic rocks, primarily primarily schists. Interspersed amongst these are broad outcrops of granitic rocks that sometimes sometimes form high plateaux such as the Caimgorms, Cairngorms, and less often underlie broad broad basins such as Rannoch Moor. Most of the Caledonian Fold Belt is dominated by m in altitude. In the West these form ridges and intervening mountains 800—1200 In valleys that trend W—E or SWaNE, SW—NE, but the eastern Grampians are dominated by broad plateaux that constitute a formidable barrier to communications. In Buchan, the metamorphic rocks of the fold the Moray Moray Firth area area and Caithness, Caithness, however, however, the fold belt form undulating lowlands and are partially overlain by sandstones of Devonian age. To the south of the Caledonian Fold Belt, and separated from it by the Highland Boundary Fault, lies the Midland Valley. This represents a broad trough of land that foundered foundered in Carboniferous Carboniferous times as a result of crustal stretching. It is not a valley in the conventional sense, but an area of undulating lowlands underlain by sedimentary rocks. These lowlands are interrupted interrupted by upland plateaux composed of

26

C. K. BALLANTYNE BALLANTYNE AND A. G. DAWSON DAWSON

resistant lavas (e.g. the Ochil Hills and Campsie Fells) or capped by equally Fault and the resistant sill rocks. Farther south, between the Southern Uplands Fault English border, lies the fourth major structural unit, the Southern Uplands. This area comprises interrupted by the broad broad comprises mainly rolling hills under 800 m in altitude, interrupted valleys of major rivers such as the Clyde, Nith and Tweed. Most of the Southern Uplands area is underlain by shales and greywackes of Ordovician or Silurian age, though several areas areas of granite occur in south-west

Scotland

area of and an area

Carboniferous sedimentary rocks in the south-east underlies the fertile lowlands of the lower Tweed Basin. The final (and geologically most recent) structural structural unit is the Tertiary Tertiary Volcanic Province, the main component of which comprises the Hebridean islands that lie between Skye and Mull, together with parts of Arran and the Ardnamurchan Peninsula (Figure 3.1). Most of these areas are dominated by stepped plateaux built of resistant basaltic lavas of Eocene age, together with smaller areas of intrusive rocks that form n altitude, such as form mountains over 700 m iin

the Cuillin Hills on Skye and the granite mountains of northern Arran. Tertiary evolution landscape evolution Tertiary landscape

Although the main morphotectonic units of the Highlands, Highlands, Northern Isles and established by the end of the Palaeozoic era (Hall probably established Southern Uplands were probably 1991), much of the present configuration of the Scottish landscape reflects differTertiary period, which began about 70 million years ago. ential erosion during the Tertiary In early Tertiary times, renewed sea-floor spreading in the North Atlantic was accompanied not only by eruption of lavas in the Tertiary Volcanic Province, but accompanied also by widespread uplift and erosion across much of the Scottish mainland and Northern Isles. The North-West North-West Highlands and Grampians appear to have been uplifted en masse, with dislocation and downwarping of peripheral areas. areas. Later Tertiary uplift was more modest periods of uplift were accommodest and episodic. Such periods panied by vigorous denudation, often concentrated along lines of geological weakweathering that ness. Intervening phases of tectonic stability favoured deep bedrock weathering enhanced pre-existing relief, together with widening of valleys and basins and the surfaces (Hall (Hall 1991). erosion surfaces development or extension of erosion development glacial legacy The glacial The

Whilst the broad outlines of the present Scottish landscape had been established by the end of the Tertiary, its detailed configuration owes much to the events of the Quaternary period. This is subdivided into two geological epochs: the Pleistocene (~2.6 million to 10 000 years BP) and the Holocene (10 000 years BP to the present). The Pleistocene, popularly known as ‘the Ice Age’, comprised several periods of extreme cold (glacial stages) separated by briefer temperate prolonged periods interglacial stages. The Holocene epoch is coincident with the present interglacial interglacial stage, also known known in Britain as the Flandrian. mountains of the developed in the mountains During successive glacial stages, glaciers developed During Highlands and Southern Uplands and spread over the surrounding surrounding low ground, ground, ultimately coalescing to form ice sheets that buried much or all of the present present land surface. Our knowledge of the timing and dimensions of most of these glacial events

GEOMORPHOLOGY GEOMORPHOLOGY AND LANDSCAPE CHANGE

27

obliterate the tended to remove remove or obliterate glaciation tended is very incomplete, as each successive glaciation very incomplete, evidence left by its predecessors. In consequence, consequence, only the final events of the most reconstructed in any detail. These relate to the period recent glacial stage have been reconstructed subdivided known known in Britain as the Late Devensian (c. 26 000—10 000 BP), which is subdivided

into three chronozones: the Dimlington Stadial (cold) of c. 26 000—13 000 BP, the Loch (temperate) of c. 13 000—11 000 RP, and the Loch Windermere Interstadial Interstadial (temperate) BP. 11 000—10 000 BP. Stadial (cold) of c. 11 Lomond Stadial During southwards into sheet extended southwards ice sheet last ice the last Dimlington Stadial, the the Dimlington During the Britain. present land area of Britain. two-thirds of the present occupying two-thirds Wales, occupying and Wales, England and England

mountain Almost all of Scotland was covered by this ice sheet, though some mountain

Boulton et above its surface e t al. have remained above may have (Ballantyne 1990; Boulton surface (Ballantyne summits may summits Caithness may 1991). Some commentators have also suggested that Buchan and Caithness 1984; Bowen (Sutherland 1984; Bowen et al. Devensian (Sutherland glaciation during the Late Devensian escaped glaciation have escaped

argued that the ice-sheet limit lay offshore from these areas areas 1986), but others have argued (Hall and Bent 1990; Figure 3.2). Deglaciation of Scotland appears appears to have been last ice sheet the last remnants of the 000 BP the remnants and by 13 000 BP, and 500 BP, well advanced by c. 14 500 were confined to the mountains of the western Highlands (Sutherland 1991). under very Although most of the retreat of the last ice sheet apparently took place under 000—12 700 BP witnessed rapid warming of the cold, arid conditions, the period 13 000~12 al. 1987), and by about 12 500 BP July temperatures in British climate (Atkinson et a]. southern Scotland were similar to those of the present (Bishop and Coope 1977). transition to the Loch marked the transition From 11 400 BP on, however, renewed cooling marked

Lomond Stadial, the last period of extreme cold in Scotland. During this relatively short-lived cold phase, a large icefield developed across the western Highlands, together with several smaller icecaps and icefields and numerous small valley and maximum their maximum have reached their 3.2). These appear to have glaciers (Figure 3.2). (cirque) glaciers corrie (cirque) al. 1989), before (Rose et al. 1988; Peacock et a]. extent between 10 500 and 10 000 BP (Rose

experiencing a period of intermittent retreat that was terminated by rapid warming after c. 10 100 BP (Atkinson et al. 1987; Benn et al. 1992). BP is 000 BP events prior to c. 26 000 information concerning Although information concerning Pleistocene events Although throughout nevertheless manifest throughout glaciations are nevertheless the effects limited, the successive glaciations effects of successive limited, North—West Highlands, Scotland. Glacial erosion has sculpted the mountains of the North-West Scotland. corries, arétes and western Grampians and the Hebrides into a landscape of cliffs, corries, western

eastern Grampians has been less U—shaped troughs (Plate 3.2). Erosion of the eastern U-shaped erosion plateaux that represent erosion intersecting broad plateaux deep troughs intersecting pronounced, with deep pronounced, with lowland 1959). Across lowland (Linton 1959). landscape (Linton from the pre-Quaternary pre-Quaternary landscape surfaces inherited from surfaces such as the Lewisian Gneiss or Tertiary lavas, rocks, such areas underlain by resistant rocks, topography, ‘knock-and—lochan’ topography, glacial erosion resulted in the formation of ‘knock—and-lochan’ small lochs. In the Midland numerous small with numerous interspersed with hills interspersed consisting consisting of low rocky hills

enhancement: areas of Valley, the prime effect of glacial erosion was relief enhancement: igneous sedimentary rock were readily eroded, leaving those underlain by resistant igneous sedimentary rocks standing proud. unconsolidated sedideposition of unconsolidated has been erosion has such erosion Complementing such been the deposition Complementing components: ments, collectively referred to as drift deposits. These have two main components: glaciofluvial or outwash tills, which are sediments deposited directly by glaciers; and glaciofluvial (Plate retreating ice masses (Plate margins of retreating deposits deposited by meltwater rivers at the margins composition, but most 3.3). Scottish tills are highly variable in terms of their composition,

C. K. BALLANTYNE BALLANTYNE AND A. G. DAWSON

28

L J ‘ J

‘\ 4\ I ‘

50 km

If

Maximum extent of Loch Lomond Readvance Limits of Late Devensian ice sheet according to Bowen a|.(1986) et al.(1986)


S

0

:8 a

0 a

0 8a 328» figfim m

: E=5

a

0 § n?

m.O

: 83z:

e

5

8 £ m

»

m

. w

8E r3 2 mm

a3

x

b:— m m

.

w

b 8

85g2 g

z

m8

a38

mfloouflowdo mSODHNOHmo

.

8m

:98

8

8

e 5

g

:8 3E

8E fi

?

:3o

8“

m

~

n

80m

E E

53.

:3 B o

cs

3m

2w c

Egon

380m 3~3o 5 8om 3 2 8 :8 3 880m 335m 688

8 E n

80m

8

a:8 85— 388 322» 338385 EZEBofiO ~o>8w :3 : 8 2 Em

8

m82 u

3— A9 E

n

o a

“meow :835oo Egon Envom “mp—ow 880m 380m

0

u

3383030 H38» «Baum 33.62020 “gem 3382020

é

os

8o

o

:oo 808 o

0k

n

5

655 mm

388% mm

80 2 e a “ c8n

8m5

8 1S N mm

.

235 fl mm

s

m

w

S m8 m3m 8 9 m$

>88 m

hm8oom

S 3

H

.82no

?

3

o

s

8

a

0

o6

u—

/ 8 m2

w $8»

m

8

8

o 3

fi

53 m — m3u 358

3 o8 80 o H 8 — 8mo n

S2%: .8

09 o 5>03p8

5

t

mm o8 88«

.

m

m

:

3

9

u8

3 H m8332 2n c5 o o

— m x5a

»

2

o

o

a

8

9

2

3

w

s

8

8w

%a

6E n3

3 “ 05

a

o

%

B

% %

3

E

m

o%8

5

8

2

w

38

% v

o

.. n

5

w 0: £ 8 o 3 em

5 5

l

3:

m

a 3 o

fi

3

E 3

8

8“

8

& 6m

m

0 m

s

6t

g

g

v

8

E8 u uva

3

88

08 m2a 0m8

mmu3m 0

83%;

82388— on 838058 3E S E 8:539 3383 mm 31253 mama—RD 2 5 5 % 30:3 mm mofloma 83E253% 5 85 SEQ 85 32 8 6 saw.0 mm 815 mm 02m n< >888 £ 2 88086.EE 0330 0@ m

3

8

8. 58t 8

8E

mm 2

a. 8omm

3 no a8

55

SOILS SOILS AND THEIR EVOLUTION

Brouster Brouster

Kllphedir o Shetland Shetland Islands VH

gurgheaclfi. Boghead Boghead

66

9

°

Castle Hill, Strachan o

5u > 5 ” o . c c

(0

. 5

cI./ 0,9

6. 0/ ¢

m

3 3 3 8

o

5

o wz cs 6«aw

9

9

& 4 0) myv o W9 o O0 ~ c

sx 1 00

/

1 o{s/

o_

v9 " % an m o

_

‘,

0l

_6

v.

o0 D

1 %O J

0g

0091

3E

0

W m 0s0/0 .e s

1%

#0 19

v

4 a

2

:3£

.

.

:

ov

.

on 9v o

o.

u

/o/o s. 9 1o

oo%

4 o/ a

o c 0m

J 1

.

s» «m »

5

5

ma

am§ E fi3 g 53M

0

a/

3E8

one

0?

m x

emcm

u

v

vb 0 ‘ as 0ar d

Q 1w

.

o0

5e

mm

3ea ON

o0o

, Nv

52

ay z v N o0

0

s U

{0

0

s0i

r, m

.

x

5

m

u

on.

89n84 ms 9 o m95

4

0

0

y o

,0

n

f

ma 3 5 a

Dom oow

005 .00h

.

3

.3..o

5

c— 53a

Ban 6 2 3 2 03

63 325::

68

00—

on

12:

6

o

n 1 3m m 3 oa? . 3 “. 8 . $ 3 9 8 3228 .

.om.m~u~ eon .omfimmmm .oom

com 3 : 3 3

.08. 5 0 : 2 3 £

. 23 31 25 . 5“

. .00h. . . .

p

ra

10a

0

v

.

a

fl2

n

0

u

.

5

0

w

.

.

O

9

I

0

m

I

u

z 8g 33 a8:

m

b/

e

3

9

m

N

o

3

3

h

;

u

«

«

v

o0 (x % : ® xE « 1 o oo ? o9/ l o o o 9 w x Oo s0 ¢ M w fl v§ /W n ( w0 O6/ /9 eae % o x . v%@o &2 o/ o / «0&cn/ oz Q oaz «os o0we a m fimo «4 / os / ea // y as a ”/ ao . o n & o 6 0 { 0 a / 0

o0

vJ/o o 0 1 40a/ oA a 0v o /

0yo a4 0/

. . . . . . . . . . . . . . . . . . .. 23

0w44

.

,o006v b 09o . b ~/ wo 0/a

u _

«9

o . / nss / 0

£

s

0

3

9

$ . m

78

a8

0

0

0

/

Aa28 8mm2 c5U 3m:.o58o.ommo 2n 8. 0 ENS a3m 3 afi1a338m :8u9a 5m END dm0 S m8Gr8m n H:“6 5 m

o

00 o

V0 v V O 0

90o0fla 84B

Iat

o

00

1o0 g0

V h;

/ o9 1v o

0/

e

a fi hmao2 S330 fl 8 8 ohaV m. : v m m ou Q o45 o. 25 B 8€r a w

0 afi xm x 0 ox f9 m z

0

a m 85

9 c 5

5

a3

o= h3 E8tm 8 n w 3a9 9amo. % c 5 5 m

$€ 626 Ga:

each

0006

60mm

60mm

000?

600m

oomn

600?

.ooon

600m

50mm

.aom—

3 3n 3

o t 3 ?%

u

. mm

5

VEGETATION CHANGE

79

BP, the Barra. Prior to c. 6500 was site was vicinity of this site the vicinity 6500 BP, 5.3 Lochan na Cartach, Barra. Plate 5.3 occurred, and since that time the pollen Woodland reduction occurred, dominated by birch and hazel. Woodland indicative of peat and acid grassland taxa, similar to and spores received at the site have been indicative those within the present-day Edwards present-day vegetation. Copyright: K. J. Edwards

The Shetland site of Saxa Vord is located on an exposed hillside at the far northern tip of Unst. The base of this blanket peat profile has been dated to 3760i85 spectra are dominated by grasses and 376021:85 BP (2290—2030 cal BC). The spectra heathers and peat surface), and growing on the peat vegetation growing reflecting the vegetation (largely reflecting heathers (largely nowhere tree or shrub pollen exceed 12.6% 5.8) does tree (Figure 5.8) nowhere in the pollen diagram (Figure of the land pollen sum. There is no convincing convincing evidence that cultivation took place or that trees grew in the immediate immediate vicinity of the site at any time. This is entirely consistent with a locality where a gust of 325 km h‘1 was recorded in 1979 before the persistent open Johnston 1980); but such persistent and Johnston away (Berry and blew away anemometer blew the anemometer vegetation is common in a number of upland and island areas subject to exposure, salt-spray provides some shelter topography provides salt—spray or high winds, even where the local topography (e.g. Birks and Madsen 1979; Mills et al. 1994). Thus, throughout this period at Saxa Vord, the only potential land use would appear appear to have been rough grazing.

THE SPREAD OF PEAT In this chapter the vegetation history history of Scotland up to 1000 AD has been surveyed sections preceding the In chronologically. sections there have been occasional mentions of chronologically. peat and peatlands. This should not be taken to suggest that peatlands peatlands and their attendant heathland vegetation were of minor importance. A problem with these

\

.F

0

6©

aA

l

.

J

0

5

T

f

5

.

s

0o0s0

a 6

'

3

A '

9JA

f 0

.[I.

_

A

/

V

I

r

0

l

o

0 0 s0 o

«

H

g 5 o 3 8EB U

n/0 /v 0

s «o

0

/

o2

i

1a /

fl

/

\

/

f

T

8

0O 0 « sO 9

3 D .o u

fl O

l

m

T

a

W w3

f

o

o

0 n 19 0// IV 0

A

.p_

fl

w y

09

0/ 0/

a

.E.

V

l

u fo 0

0v

0

/r‘"‘1~\ .

[L_ :

J /

80

mB3gSfi

.

oom

0

0

I

;

f

r

._

. .

K

l

a4

oow

ao

a

T

B m2 H

4o

\

\/

‘ 0

x

w

an

va o

o o 9

6 >

_ w L

.

\\

3

|

.

\

‘

«9

0/

o

!

/0

_ [I

0 0

I

9

o

I:

a/

,_

w

O / 0

\

v

l

w

=.m :a % . E a 8

,

04 O

mo

_

/

I

.

8w 5

oo ._

.4\

I /

\ \

m| .#

2 Q

1

F'T

66 90

«0

ovy 0

9

9

.

n

.

“0o

O

/0 0.

.

.

0

.

i ] '

I

.

.

3

N

T

.

.

I

.

0 O 0

. /

oI

,

.u-.

1

w.

o o0

0 9a /v

4

.

4

.0

o 90

[I [n [w

\

cw . c o . " ?m

/ »0 0c

00

M

.

0

.

m

{

.

[

.

w .

s 9s

.

1\ 9

V

o /

\

a% 5 35 = 3 ;

o.

|

f f 0d 0«

o0

/

1

°o

533%

.

x0

4« /

b

0 O

.

s0

1/ 4

I

.” m — h—

w

0 4 4

/

. |

.

‘

. w

I

.

.

.

/ m w 0 o ./ v «

.

V

or O

/

_

L

o

.

0o

/

.Fh

.I .

u

/O

0

. .

5. 3 3 5

l. .

/

0 01

f

w

z

E H

.

o

|

.

o

.

y

os

n

I

,

,

.\

0

81

VEGETATION CHANGE

//

Plate 5.4 5.4 Edwards Edwards

//;./,1(

Reineval, South blanket peat on slopes Peat cutting Copyright: K. J. South Uist. Copyright: slopes of Reineval, cutting in blanket

wetland features lies in the lack of knowledge inception. The knowledge as to their date of inception. traced back to soon after deglaciation in possible beginnings of this process can be traced the Outer Hebrides (Bennett et al. 1990; Edwards et al. 1995) and in upland heathland seems to have Perthshire (Tipping 1995b) where Calluna-dominated Calluna-dominated heathland become arguments for the well-rehearsed arguments The well-rehearsed landscape. The component of the landscape. become a component inception and extension of blanket peat, whether the result of a rather impenetrable impenetrable 1975, anthropogenic processes (Moore 1975, mix of hydrological or anthropogenic climatic, pedogenic, hydrological of climatic, mix certainty — that one certainty with one leave us with 1993; Kaland 1986; Stevenson and Birks 1995), leave

blanket peat cloaked the land surface over vast tracts of the country (Plate 5.4). The available dates for peat spread are not easy to evaluate because the general lack of sub-peat surveys make it impossible to assess the topographic origins of the peat (Edwards and Hirons 1982; but see Charman 1992 and Tipping 1995b). Dates Dates of initiation from Carn cal BC] at Cam BP [9040—8320 cal widely (e. g. 9800—9200 BP Scotland vary widely from Scotland Dubh, Perthshire [Tipping 1995b] and two statistically indistinguishable dates of 2415i25 BP [520—400 cal BC] and 2395:l:25 from the obtained from 2395i25 BP [420—400 cal BC] obtained base of peat sections 20 m apart at Starr, Loch Doon, Ayrshire [Edwards 1996b]); variations can and 7270:]:100 [6180— small areas (e.g. > 7270:|:100 very small within very found within even be found can even and variations 5980 cal BC] and 4810i60 BP [3650—3520 cal BC] at Callanish for two adjacent adjacent profiles on a small peninsula [Bohncke 1988]). As was the case for Ireland (Lynch 1981; Edwards 1985), a continuous process of peat inception and spread would probably have been in operation, and although intuitively it might seem reasonable reasonable that times of climatic climatic deterioration would accelerate peat spread, insufficient information is available to substantiate this.

82

WHI'ITINGTON EDWARDS AND G. WHITTINGTON K. J. EDWARDS

CONCLUSION CONCLUSION

exploraanalysis for the exploraSince Erdtman began the systematic exploitation of pollen analysis That the tion of vegetation history in Scotland, great strides have been made. That country took on an almost universal tree cover from about 9000 years ago is now vegetation fully established. The demands made by prehistoric communities on that vegetation cover, whether it be directly on the plants themselves or on the soils that they eventually the virtual demise of the woodland. occupied, led to an opening up and eventually introduced plants, combination of introduced that a combination The was changed so that pattern was The floral pattern especially the cereals, and native species such as grasses and many weeds of parts of the north and west, cultivation cultivation became dominant. In the uplands and in parts ubiquitous. peatland vegetation became ubiquitous. considered as definitively should not be considered vegetation history of Scotland should The vegetation capable of raising new published are capable diagrams published Virtually all new pollen diagrams established. Virtually remain. problems still remain. Major problems improve. Major questions, questions, especially as sampling strategies improve. against or in conjunction parts played by climatic and pedogenic change, as against The parts particularly with still need further investigation, particularly humanly-induced factors, still with humanly-induced knowledge of the status of the woodland inception of peat. Current knowledge regard to the inception cover, especially for the Mesolithic period, remains clouded with doubt, while the identification and characteristics of woodland management practices remain adoption of conjectural. The role of fire, the precise timing and nature of the adoption arouse floras continue continue to arouse value for land land use of weed-type floras cereals, and indicative value the indicative and the debate. Statistical techniques in the analysis of vegetational change are still underWhittington and Bennett and Humphrey 1995; Whittington utilized (cf. Birks and Line 1992; Bennett Edwards 1995). accumulated information already accumulated corpus of information substantial corpus basic level, level, the substantial At a basic provides an ecological context for human activities in prehistoric and historical times. Quite clearly, this relationship relationship should not be viewed as a passive one; the fossil record is also capable of contributing to a picture of the nature and development of human communities. Beyond this, palynology can be used proactively to Edwards predict the existence of human impacts through time (Whittington and Edwards 1994), both when the archaeological record is unknown or absent and where the landscape record is obscure.

6

Faunal Change THE VERTEBRATE FAUNA FINBAR MCCORMICK AND PAUL C. BUCKLAND

INTRODUCTION INTRODUCTION

The rapid warming that opened the present interglacial at c. 10 000 BP, changed the Scottish landscape from one of residual icecap, corrie glaciers, snowfields and tundra, with low carrying capacity in terms of vertebrates, to one in which a largely continental steppe rapidly gave way to a succession of birch—pine forest, hazel, and continental mixed oak forest; only perhaps on a few offshore islands did some form of wooded Through the mid to late Holocene, landscape fail to develop fully (Chapter 5). Through human impact led progressively to the modification and destruction of the forest cover and its associated fauna. The Holocene history of Scotland’s vertebrate fauna kept in introductions of aliens, often kept and introductions the native fauna and extinctions in the is one of extinctions inordinate numbers to lead to further losses and retractions of range in sufficiently inordinate the indigenous biota.

EVIDENCE THE NATURE OF THE EVIDENCE vertebrate fauna of Scotland is, The prehistoric and protohistoric record of the vertebrate when compared with England, singularly incomplete. The development of raised mires and acid soils soils over the fossil much of the of the removed much has removed country has the country over much of mites and coastal restricted to coastal largely restricted record is largely evidence, and, outside the Midland Valley, the record locations, usually on and in shell sands, and the limited outcrops of calcareous Dumess Limestone in the North-West (Figures 6.1 rocks, such as the caves of the Durness and 6.2). Most assemblages also derive from archaeological archaeological contexts, where the dogs further compound and dogs selective activities of human and problems. Antler, the problems. compound the tooth ivory and even bone were materials widely utilized for the manufacture of artefacts artefacts (cf. Foxon 1991; Hallén 1994; Weber 1994) and are likely to have been dispersed far from the locality where the animal was killed. The indirect impact of human interference is evident in the disturbed nature of many assemblages, particularly from machair and similar coastal sand deposits. Barratt (1995), in a particularly review of Norse and later animal animal bone groups from Caithness and Orkney, found review

cunz‘culus), a medieval introduction (Oryctolagus cum‘culus), introduction to England which only that rabbit (0ryctolagus became common and widespread in Scotland with the unrelenting suppression of its Scotland: Environment and Archaeology, 8000 BC ~ Ian B. M. Ralston. Edwards and 12111 — AD 1000. Edited by Kevin J. Edwards © contributors. Published in 1997 by John Wiley & Sons Ltd. © 1997 The editors and contributors.

.

84

F. MCCORMICK AND P. C. BUCKLAND Pimwan KnapofHowar

LinksofNommfi Links ofNoltlandfi ##5

68

G Q

81:33a Shin Brae

Q a

s

I

Q

Q, CW“: 3W5”

Scord Scordofof

"

Do

Brouster '

o

,

0

I Crag Nan Uamh ICleagNanUamh

a

Northton Northton‘

‘~ \‘

41?}

‘~

g K

RosinishQ

o O

0 0

v , L

Whoa 5’ i0 “we, °O

dfl

‘\

\

yp

‘\ 7,

rr

3 )

1 I

/,,

'.

Omnsay? 0mm?

/ x,

,

\‘x‘

m m

U Q

Morton Mouton

., / /

" Oban

'‘

Meiklewood

' " Aim",

‘

\\

r“

s‘

mom «mm (‘/

/ /

,1 ,1

/ //

o

E 5

Figure 6.1

Tm] Tah 5

150 :

Lateglacial to mid Holocene sites referred to in the text

century, was frequent predators, from Scotsmen to wild cats, during the nineteenth century, mammals from archaeoin the majority of bone assemblages. Records of small mammals logical sites, therefore, have often to be treated with some circumspection. The net known from the demise is well known and its demise the history of woodland and whilst the that, whilst result result is that, are only known palynological from a few locations, known from associated mammals are record, its associated palynological record, archaeological record provides only glimpses of the wild and domestic and the archaeological animals, which not only shaped its destiny but also formed the basis of most of the animals, activity. human activity.

85

VERTEBRATE FAUNA CHANGE: THE VERTEBRATE FAUNAL CHANGE: FAUNAL | » k 3 Mme.)

3

p

a

3% 1 %? 1 6‘:

B'

‘3‘“I

How

r Crosskirk CrosskiflrQ l?

I

seafloway, Scalloway, °

Fteswick Freswiek
5 3 8 : 0 EnumsoBu—Em amen“«82850 .“ 65

m

n9A

3

A §§$§ .33 3 3v :5» 3323—055 :5» EEBJEm A§V§fi§g

g

m

u

g

fi

s

Q

3

5

5

S

—

5

o

23

n

3

3

c

3

9

g

E

n o

g

:

Em Nw 3 zoawxonm moawxosm o S3

Noe—4

3

b :R3u 3 a

E3J

q

a

T

heaven—m zosvozm w

S hb 53 52 e9

?

w 950m 0303 5 h

fl

8%

h 32E 89E N. 2503 9502 0 2 c

mo— as

N I tm ax w “ MO v —— Iv— c fi— u —I ‘ N

s fi g

m n

%

a

3

2 vw

S2

E
”a83i

E san :

R R

v n m

1—1 v—1

m2 Q2

d fl

+

? n.

\D \O H H

N ! \ DI l\\ 4— v -H \D H

ma

5

m

n n

3

N

cmHa cm:

m NU NO

NN O «m m h $ 3 S :~

ASS Hn36S ‘

u3o

M

v aa m a wNN mg

M (“1 M

% A

E

N

+ +

3

33A

m:

a

H5a

2.mm 2 0 c mEESM . % wnfioom Mom—Em Aaaufiufiem §Euu=on~v“axiom GSEQSBR“33.9335 was EBVSum: A235 3:9: S553

xfiob 2:3..v @ 5 . 2 3 MEOH 63 2 @5 8 n 2Angus: “33.9335 058m 3 Son~33230q2 we Eon Qutufinu $538..»t Emfiefifivmo CUTE»

:"932

fi

39 o o

aޣE5 M W

o m

o5d

€m§§m§ § § = E am

E

vv

:

m

o

9 wS

o 99

o9 d58 3

a 3

l

m

=

a

e

°

l

+

030

nE 0 a

ua

an m

u
8 M03505.>G8

€25“ 9:33 3 9 FQ EEuofifiw n Anxmsfig ummsbvmats:m

eateafim g i g am 5 E 2.

E€ 8 § § e m nfimaatrmvmo wee 80m S A§M§~SE méu§tm§$ $533 §NM§~SV5 was: A§M§~§E A§=w®~umu §EEEM§S~V5 x 8 6 3 :a wQ2335 335VcooA S

flN NJ

_ N _ w m N

Sa

2a m a s

:

m 0$3=v2h :.o oE a

c

d

|

—

5. 8n —

w

Hxxfi uua

E

m

3

3

3

z

3

3

m o

E

a

0u

3 E

9

fi

m M: 3

N 2_0 N w V _w NU m 2 NO m 3S 5A w h m «2 S

0 ?

2

m

v

o

m

q

5

3

Hi m _ m

h S 2 m

N

fi

3

o

mc ~35

E

S «m _ 08S c0oQ n S m w

g

o

q

fl8

_ NJ N

o

n0Q 3 w

H

m

1m8

[\ l\

l\ v—t

h — t«h m trv— —I vt — O rr— \O (Ni—4 H N

m

0

u

8

o

ev

m3a h9n

m

N m v

.

EQ 8 :m

fi

onv

%

E

5

5

0y

S

n

S

flA

+++ : N M m fl

c 9a

3

"093

E

Q

“

§

E

3

u

$3

§

.

§

E:

k

-

E

Sm

9 m a EE

£

u

5

h.#T3

i8 m 6w

6

m B

3

fi

z

"nA >£ A

E 8 mom J £a8 Ezfiughv93m

ES m

S

9.5::t S

n

i

l

2 mmo 32:15 S ~

3

§8

©

5w

2 8 2.3m

s0 —

§

B

m 53

S E

“

§

29

y

o

g€ § § § a

a

Q

S

n

S

~

a

5 §

fi

~

-

8

afi

83x

6

fi

3

S

3

FAUNAL FAUNAL CHANGE: THE VERTEBRATE FAUNA

99

find of 15 completely articulated skeletons. The implication implication is one of active management of a herd (Plate 4.10 in Bramwell 1983a), but why the carcasses should have been left unused is unknown. Although assemblages from tombs run the risk of bias by selection processes of a South chambered tomb at Isbister animal frequencies ritual Isbister on South the chambered from the frequencies from the animal ritual nature, the Ronaldsay are similar to those from most contemporary occupation sites on Orkney, with sheep and cattle in equal proportions and red deer and pig in small numbers (Barker 1983). Although some of the bird and fish remains may reflect the activities of predators (Colley 1983), both fox (Vulpes vulpes), no longer present on Orkney, and otter were found in the tomb. Wheeler (1979) preferred to regard the bulk of bone as part of the human component in the tomb at Quantemess on Mainland. scorpion (Taurolus bubalis) or father lasher sea scorpion fish, either sea Mainland. Cottid fish, (Myoxocephalus scorpio), are most abundant at Isbister, whilst corkwing wrasse (Crenilabrus melops), presently a rather southern species, dominates at Quantemess. Quantemess. Amongst albicilla), which (Haliaeetus albicilla), white-tailed eagle (Haliaeetus the white-tailed Amongst the birds, bones of the which was exterminated in Scotland during the late nineteenth century AD and has recently been reintroduced, tomb, Isbister tomb, foundation deposit of the sealed in the foundation well sealed are well reintroduced, are the Isbister chamber (Bramwell and there are the remains of at least six other birds from the chamber Another raptor, not currently resident in Orkney, the goshawk, Accipiter 1983b). Another gentilis, is present in both the Isbister and Quantemess tombs (Bramwell 1979). settlement contemporary with Neolithic settlement Evidence for the vertebrate vertebrate fauna, contemporary from assemblages from useful assemblages sparse. There are Orkneys, is surprisingly sparse. outside the are no useful the Orkneys, outside Brouster show the from Scord of identifiable from Shetland, of Brouster fragments identifiable the few fragments although the Shetland, although BP mid fifth (N oddle 1986) presence of cattle, fifth millennium BP 1986) by the mid red deer (Noddle and red cattle, sheep and has yet Northton on Harris has from Northton bone from the bone Isles, the the Outer Isles, On the 1986). On (Whittle et al. 1986). 1981) Simmons et al. 1981) detail (Simpson 1976), but Grigson (in Simmons published in detail to be published

records a dominance of sheep bones, with hare (Lepus sp.), and seal, both grey and common, amongst the wild component; cattle, red deer and pig are also present (Finlay 1984). The presence of part of the skull of a badger (Meles meles), on this site may relate to imported pelts since the animal is not recorded from the Outer Simmons et al. Isles (Serjeantson (Grigson, in Simmons correct (Grigson, record of hare is correct (Serjeantson 1990); if the record 1981: 192), its origin is likely to have been similar. On the Scottish mainland, on a group of sites in the Firths of Forth Forth and Moray, a group of shell midden sites belong to appear to belong ovicaprid, which appear and ovicaprid, contains cattle and domestic cattle contains a few bones of domestic the fifth millennium BP, and it is possible to consider these sites in the context of

Erteballe sites of the Baltic littoral (Sloan 1989). the well-known contemporary Ertebolle The early Bronze Age evidence for the vertebrate fauna is even more restricted. Northton on Harris, red deer bones are as numerous as cattle and sheep (Finlay At Northton 1984), perhaps suggesting a regular pattern of culling. The Neolithic and later faunas from Northton also include grey and common seal, and the fish remains remains faunas

include ling (Molva molva). The earliest evidence for domestic horse in Scotland also comes from the Outer Hebrides, from the small assemblage assemblage on the Beaker site at Rosinish on Benbecula. The same site includes field mouse (Apodemus sylvaticus), field vole (Microtus agrestis) and house mouse (Mus musculus), although the sandy nature of the later, from later, burrowed in from may have burrowed they may concern that they some concern site causes some the site M musculus, at least, is Norse occupation deposits on the site (Serjeantson 1990); 1990); M.

usually regarded as a late introduction. From Ardnave on Islay, Harman (1983) has

100

F. MCCORMICK AND P. C. BUCKLAND

described a small bone assemblage dominated by sheep, cattle and pig, associated with red on present on only present latter is now only The latter fox skull. The and a fox vole and red deer, seal, field vole Skye, and formerly Mull, in the Inner Hebrides, and may represent a trophy or pelt from the mainland. brought from brought

THE LATE HOLOCENE

Whilst the pollen data show the early part of the late Holocene, the Bronze Age and early Iron Age, as the time during which the Scottish landscape rapidly approached its modern form, with the widespread reduction in woodland and expansion of the evidence of the the evidence Chapter 5), the 1994a; Chapter bog (Tipping 1994a; heather moorland blanket bog moorland and blanket associated fauna is singularly lacking. At the late Bronze Age/early Iron Age sites of associated Baleshare Point in the Uists (Figure 6.2), sheep and cattle dominate, Hornish Point Baleshare and Homish

and deer, deer, pig, seal and otter are less significant (Halstead forthcoming). An extinct large crane (Grus primigenia) primigem'a) has been described from late Bronze Age or early Iron deposits at King’s Cave, on the shore of Loch Tarbert, Jura (Harrison and Age deposits remainder of the assemblage has yet to be published. Crane is Cowles 1977), but the remainder

also present at the Iron Age wheelhouse site of Bac Mhic Connain on North Uist (Hallén 1994) and in the broch at Howe on Orkney (Bramwell 1994); in both cases, visitor very rare visitor now a very (Grus grus), now the extant crane (Grus ascribed to the the bones have been ascribed the Isles. British Isles. the British to the Northern Isles, from Howe The largest Iron Age assemblages are again from the Northern on Mainland, Orkney; the bulk of the material belongs to the latter part of the period, contemporary with Roman occupation to the south. This site shows a and, whilst it 1994) and, al. 1994) Smith et a]. Age (C. Smith Iron Age through the late Iron deer through red deer decline in red survived on Lewis and Harris, this species was extirpated from Orkney before the medieval period. The late phases of this site, associated with the broch, show sheep tending to dominate over cattle among the domestic animals, and there are also significant numbers of pig. A small horse, similar to the Shetland pony in size, and significant occurs. (Platt 1956), occurs. Shetland (Platt on Shetland contemporary deposits at Jarlshof evident in the contemporary Jarlshof on evident than a wolf rather than may have been a wolf evident at Howe, although one Dog is evident skull may one skull Dog animal, presumably imported with its pelt from the Scottish mainland (C. domestic animal, Shetland from Jarlshof, Shetland recorded from 146). A possible wolf is also recorded Smith et al. 1994, 146). was of an site was the site from the dog from skeleton of a dog 1956); in contrast, (Platt 1956); contrast, the articulated skeleton a’latt wild, arlshof cats were were thought to be wild, the JJarlshof of a terrier. Whilst the size of animal about the size ones, with the bones bones also showing furs, the Howe ones, source of furs, probably an imported source signs of skinning, appear to have been domestic. Other furs at the site were provided by both otter and fox. Compared with the Neolithic animals from Knap Iron Age cattle are small, comparable with material from contemof Howar, the Iron

porary sites elsewhere in Britain. The type, usually termed the Celtic shorthorn, is also present on the north coast of Caithness in the material from the broch at from Sollas on bone from cattle bone compared the cattle (1991) compared (Macartney 1984). 1984). Finlay (1991) Crosskirk (Macartney North Uist with the lightly built West Highland Highland or ‘black’ breed. The sheep from appears more frequent this site more resemble the Shetland type than the Soay. Pig appears on the mainland than in the island assemblages and goat makes its first definite appearance in the north at Crosskirk (Macartney 1984) and at Dun Mor Vaul on

FAUNAL CHANGE: THE VERTEBRATE FAUNA

101

Tiree in the 1974). Despite the existence Despite the (Noddle 1974). Hebrides (Noddle Inner Hebrides the Inner existence today today of feral feral flocks in Galloway and on several offshore islands, goats do not appear in numbers numbers in any of the and they appear never to have formed a bone groups and the Scottish bone appear never significant component of the domestic fauna. significant The wild fauna from Crosskirk also includes roe deer and possibly wild cat. The deer remains remains are between the large perhaps intermediate in size between are of animals perhaps red deer animals and the the smaller ones of the present forest animals present day, day, an adaptation adaptation to the poor poor range provided provided by moorland moorland (Macartney 1984). range 1984). Despite Despite its coastal location, location, there there are few fish, although wrasse and pollack are again present. Although the explanacould be taphonomic, tion could taphonomic, in the way way fish fish bone bone was disposed disposed of, Howe shows shows a the earlier similar pattern pattern in the earlier phase of its broch broch (Locker (Locker 1994). 1994). The The fish fish list from from the the later phases is extensive (Table 48 in Locker 1994), dominated by saithe probably caught inshore, inshore, small small wrasse wrasse taken from the shoreline, and smaller numbers of caught gadoid fishes, taken in fishes, including cod and ling, both of which would have been taken deeper Jarlshof on Shetland present at Jarlshof also present are also waters. Both species are deeper waters. Shetland (Platt 1956). 1956). The marine marine fauna at both sites sites also also includes includes cetacean cetacean bone, bone, usually usually used used for for artefact artefact manufacture, and seals, assem— Crosskirk. The Jarlshof assemidentified at Crosskirk. seals, with grey seal identified

blage includes both seals, and bones of walrus (Odobenus rosmarus) (Platt 1956). Whilst its ivory and skins were widely traded out of the Arctic in the early historical period (McGovern (McGovern 1985), the few records of bones from Scottish Scottish prehistoric prehistoric sites are likely to reflect the chance killing of the occasional stray animal. animal. Both Crosskirk Crosskirk and Howe have provided extensive bird lists (Macartney (Macartney 1984; Bramwell 1994). Gannet, great auk and shag (Phalocrocorax (Phalocrocorax aristotelis), aristotelis), cormorant corrnorant (P. carbo), indicate the razorbill indicate (Pufifinus pujfinus) carbo), manx (P. manx shearwater shearwater (Pu/firms pujfinus) and and razorbill the exploitaexploitation of coastal resources, but black tetrix), characteristic black grouse (Lyrus (Lyrus tetrix), characteristic of open

moorland, red grouse present at the former site, whilst red with a few trees, is present usually with moorland, usually (Lagopus lagopus), also a heath and bog bird, appears appears on the Orkney site. Chicken (Gallus gallus) occurs in the latest deposits on the site, which are probably postRoman; goose (Anser domestic goose where domestic from Howe, where material from contemporary material Roman; there is contemporary (Anser anser) and duck (Anas platyrrhynchos) represented. The great auk is evident platyrrhynchos) are also represented. in virtually but is not Isles, but Northern Isles, the Northern assemblages in the Age assemblages Iron Age contemporary Iron virtually all contemporary included in Platt’s (1956) list from the later, Norse deposits from Jarlshof Jarlshof on Shetland. easy but, flightless vary, but, may vary, extinction may island dates of extinction Individual island Shetland. Individual flightless and and an easy prey to hunters, it shows an inexorable decline; it was last seen in Orkney in 1813 and finally became extinct in 1844 (Buckley and Harvey-Brown 1891; Cramp 1983). In southern Scotland, in the region intermittently controlled by Rome in the first few centuries AD, the record record is sparse sparse apart from the excavated Roman forts forts and

vici, where the bone either survives or has been studied. Bametson (1982) has reviewed near Broxmouth, near hillfort at Broxmouth, the hillfort from the husbandry from for animal husbandry reviewed the evidence for

Dunbar Dunbar in East Lothian. The main occupation of the site ranges in date from late in the third millenium BP into the Roman period. The scatter of bones from the early phases cattle, sheep, cattle, Scotland), sheep, certain record in Scotland), earliest certain included goat (the earliest phases of the site included

pig, horse and dog, the wild fauna being restricted restricted to roe deer and a bird bone, perhaps perhaps of a heron (Ardea cinerea). Later assemblages from the site contain a similar mix, with cattle slightly predominating over sheep, with a few pigs and goats. The from comes from site comes from a Roman site published from extensive bone assemblage published most extensive The most James Curle’s excavations at Newstead, near Melrose in the Borders (Ewart 1911).

102

F. MCCORMICK AND P. C. BUCKLAND

The the assemthe remainder of the and the has already been mentioned, and The presence of elk has blage consists boat presence of boar The presence animals. The usual mix of domestic animals. largely of the usual consists largely

both on this site and at Bar Hill (Macdonald (Macdonald and Park 1906, 530) implies the continued continued existence of at least some form of forest cover in some areas. The animal is also figured on a number of later Pictish and Early Christian stones (Allen and Anderson 1993, I, 73), although by this period, like the deer also depicted, it may have become largely the hunting preserve of the nobility. Boar is also present in the contemporary native assemblage from Edinburgh Castle (F. McCormick, unpublished). In the West, both red fort at from the Roman fort recorded from deer are recorded roe deer and roe red and bone quantified bone but there is no quantified 107), but 1975, 107), (Robertson 1975, Dumfriesshire (Robertson Birrens in Dumfriesshire report from the site, a problem which also accompanies most early excavations reports to go on, it is along the line of the Antonine Wall. With the dearth of bone reports characterize the Iron Age surprising that Piggott (1958, 25) saw fit to characterize not surprising inhabitants inhabitants of the north of Britain as ‘footloose Celtic cowboys’. At Inveresk, on Forth, mid second century bone groups from the vicus attached the south side of the Forth, to the fort show a dominance of older sheep and cattle (Barnetson 1988), a contrast with most contemporary rural assemblages and perhaps a reflection of stock being the availability of fodder in the greater availability than a greater rather than garrison, rather feed the garrison, driven in to feed driven Lowlands. A similar pattern is apparent in the Antonine Wall fort at Mumrills (Macdonald and Curle 1929). Although probably a Roman introduction, domestic cats cats appear not to be recorded from any fort or vicus site, but dog, of about the size

Whippet, is evident as a paw impression from Flavian Inchtuthil on the edge of of a whippet, the Highlands (Pitts and St Joseph 1985, 340), and two sizes of dog are apparent in similar impressions from Newstead (Elliot 1991). although the assemblages, although The post-Roman period shows little change in the bone assemblages, progressive expansion of the dietary constraints of Christianity could be expected to impact. The eating of horse flesh was explicitly prohibited, and fasting have some impact. reduced the available meat to fish, seal, whale and possibly deer (McCormick 1987) Colurnban on certain days and during Lent. The high incidence of deer on the Columban monastic monastic site of Iona may reflect this dietary requirement, and the community from rookeries neighbouring seal rookeries control over neighbouring onwards exercised control the the sixth century AD onwards Although the animal bone evidence is inconclusive, finds of bog (McCormick 1981). Although butter (Earwood 1991) suggest that dairying had become widespread in the West debatable period; earlier evidence is debatable and the Highlands by the early medieval period; (McCormick 1992). Domestic fowls appear to become more common during the Pictish period, but Serjeantson (1988, 212) notes that their place in the islands was frequently taken by Serjeantson exploitation of seabirds; gannet at Buckquoy, Orkney (Bramwell 1977) and the Udal North Uist, and puffin (Fratercula arctica) at Scalloway, Shetland (T. O’Sullivan, on North emphasis on poultry in some comm, 1995). There appears to be an increased emphasis pers. 00mm, later Norse deposits. Rackham (1989) suggests that the small size of red deer antler, indicates animals from Birsay on Orkney, indicates artefacts at Birsay but the Caithness, but from Caithness, used in artefacts widespread trade in raw materials, including antler (Weber 1994) and walrus and whale ivory, indicated by such items as the Lewis chessmen, requires care to be vertebrate assemblages. exercised in the interpretation of unusual elements in vertebrate differences clear differences Although, as in much of the artefactual record, there are no clear evident assemblages, it is evident vertebrate assemblages, Pictish and early between late Pictish early Norse terrestrial vertebrate

VERTEBRATE FAUNA FAUNAL CHANGE: THE VERTEBRATE FAUNAL CHANGE:

103

that eleventh that fishing becomes increasingly important. In the late ninth to early eleventh century at Birsay on Orkney, inshore inshore fishing for small saithe dominates the marine component, although a range of other gadoids shows deep-water activity activity (Colley 1989, 258). At Freswick Links in Caithness, the massive quantities of fish bone indicate a community moving towards the commercialization of activities (Morris et nascent the nascent world of supply to the the larger world into the Scotland into locking of Scotland and the locking a1. 1992), and al. Europe is evident in such coastal Late Norse and medieval sites as urban centres of Europe (Barrett 1995) on the same coast. Robertshaven (Barrett Robertshaven

1 04 04

LAND SNAILS LAND SNAILS STEPHEN P. P. CARTER

The modern land snail fauna of Scotland totals 77 species (Kemey 1976). The distribution distribution of these is only imperfectly known, but they may be divided into two majority The majority requirements. The reflect habitiat requirements. that reflect distributions that with distributions categories with broad categories broad (61 species) occur over a wide area and are tolerant of acidic, calcium-deficient This sites. This restricted to coastal sites. species) is more or less restricted habitats. smaller group (16 species) habitats. A smaller calcareous presence of calcareous coastal distribution probably reflects their dependence on the presence coastal distribution sediments in such sediments absence of such the almost total absence and the shell sand, and case shell this case sediments, in this sediments, inland areas. inland The history of the development of this fauna is also limited by the extreme rarity of calcareous calcareous sediments in Scotland, which are required to preserve the calcium carbonate carbonate shells. Added to this restriction is the fact that none of the slugs (19 species) can be identified from their internal shells, and therefore their history is Although all fossil assemblages of land snail shells have been entirely unknown. unknown. Although north coasts west and accumulations, largely coasts of calcareous sand and north largely on the west recorded in calcareous sand accumulations, recorded since fauna] change since picture of faunal provide a picture these assemblages provide (Evans 1979), these Scotland (Evans habitat. Scottish terms, exceptional habitat. distinct, and in Scottish 5000 BP (3780 cal BC) in one distinct, 5000

detected: those attributDespite these restrictions, two causes of change can be detected: introduction. and human human introduction. migration and to species migration due to and those due habitat change and to habitat able to Nothing is known about the land snail fauna of the early Holocene; it presumably rather and the rather to rapid climate change and underwent frequent changes in response to underwent frequent Scottish fauna, modern Scottish One species of the modern vegetation. One development of vegetation. slower slower development suggesting that it may distribution, suggesting Deroceras agreste, has a restricted montane distribution, boreal adapted to boreal have been more widespread in the early Holocene. Other species, adapted vegetation, may have become extinct. The species which made up the early prehistoric fauna are probably all still present today, although some (e.g. Vertigo habitat change, change, reflects habitat probably reflects recorded. This probably rarely recorded. are rarely lilljeborgz') are and V. lilljeborgi) pusilla and wetlands. The restriction restriction of and the the draining of wetlands. of woodland and particularly loss of the loss particularly the spread of species able to has been balanced by the spread prehistoric fauna has early prehistoric the early accumulations. New species have progressively thrive in the coastal shell sand accumulations. colonized this habitat: Helicella itala followed by Cochlicella acuta, both during the Iron Age, and Candidula intersecta and Cernuella virgata probably in the medieval period (Evans 1979). These species occur earlier in south-west England (Bell 1990: perhaps 247) and there appears to have been a gradual spread northwards, perhaps humans. Other species have benefitted directly from human passively assisted by humans. southern draparnaudi and Helix aspersa are widespread in southern presence: Oxychilus drapamaudi entirely Both are entirely north. Both become increasingly synanthropic further north. but become Britain but Britain dependent on the microclimate of human settlement for survival in the north of Scotland. As a result of the changes outlined, the present-day snail fauna of Scotland probably includes more species than at any other time in the Holocene.

105

INSECTS INSECTS PAUL C. C . BUCKLAND AND JON P. PAUL P . SADLER

The insect fauna of Scotland Scotland has not been subjected to the same detailed detailed study that palynological record provides provides for the flora. Not only is the insect record record patchy the palynological temporally, it is also, with the exception of recent unpublished unpublished work on archaeorestricted to the Coleoptera (beetles). Although the fauna Virtually restricted logical sites, virtually includes no true endemics, it does does contain a number number of species not not recorded recorded elsewhere British Isles. may reflect from the survival from Several may Isles. Several the British in the elsewhere in reflect survival the cold of the the Loch Lomond mountain top locations, whilst whilst others remain in habitats subarctic mountain stadial in subarctic Lomond stadial habitats

eradicated from lowland Britain by human that have been virtually eradicated human activity — particularly old established pine (Hunter (Hunter 1977) and birchwoods. established pine particularly birchwoods. The The extent of of late late Devensian glaciation (Sutherland (Sutherland 1984) and the intensity of the cold, would appear appear to preclude the survival of any animals from previous interglacials, despite earlier arguments 1953; concerning offshore islands contrary concerning arguments to the contrary islands (Balfour-Browne 1953; Jackson 1956), must have with aa tabula rasa. 500 BP the Lateglacial must and the 1956), and have begun begun c. 14 500 BP with

More assemblages of the Lateglacial in Scotland have been examined than for most of the but the the sample remains the Holocene, but most remains inadequate. Bishop Bishop and and Coope (1977) examined four in Dumfriesshire (Roberthill, Redkirk Point, Bigholm Burn and Bennie had James Bennie The geologist James and Sanquhar). The had recovered some beetle remains from and last century, and late last and Saughton near Edinburgh late at Corstorphine and sites at from sites these have have been (1968). In similar small assemblage In addition, a similar been published by Coope (1968). these from near Airdrie, 1962) is probably also Lanarkshire (Coope 1962) Airdrie, Lanarkshire Burnhead, near from Bumhead, also of Lateglacial origin. More recently, Merritt et al. al. (1990) from small fauna from (1990) published a small beneath Loch Lomond till in the Teith Walsh (1992) Perthshire, and Walsh Teith Valley, Perthshire, beneath

examined material from a kettlehole at Logie assemblages from the Logic in Fife. No assemblages Highlands Islands have yet been studied and none of the sites provides a Highlands or Islands complete through the the into the the Windermere Interstadial into warming through initial warming from initial complete record from early 1977), however, and Coope 1977), Bishop and (in Bishop early Holocene. Coope (in however, provides a July July temperature curve, constrained radiocarbon dates and derived derived from the fossil constrained by radiocarbon Windermere c. 13 000 BP thermal maximum of the Windermere beetle evidence, from the thermal

through incorporated in and several of the Scottish sites are incorporated through to the early Holocene, and the revised revised diagram probable that some of the more Atkinson et al. (1987). It is probable diagram of Atkinson eurythermal temperatures), eurythermal elements in the insect fauna (tolerant of a wide range of temperatures), survived from immigration after deglaciation, but much of the extensive from initial immigration interstadial fauna, described described from such sites as Roberthill (Bishop and Coope 1977), would have been cut out by the return to conditions equivalent to high arctic during Lomond Stadial. the Loch Lomond The rapidity of early Holocene climatic amelioration apparent amelioration has long been apparent from the fossil insect assemblages 1980). At Brighouse 1972; Osborne Ashworth 1972; (cf. Ashworth assemblages (of. Osborne 1980). Brighouse

Bay (Stewartry of Kirkcudbright), 9640:}:180 Kirkcudbright), the fauna is thoroughly temperate by 9640i180 BP (9050—8470 cal BC), but no Scottish site has so far far provided provided insect remains which cover the actual transition. The latest evidence from the Greenland Greenland ice cap suggests that much as 7 °C occurred than a decade warming by as much that warming occurred in less than decade at

106

P. C. BUCKLAND AND J. P. SADLER

1993). Such a rapid transition ago (Alley et al. 1993). calendar years ago 500:1:200 calendar transition would l l 5001200 11 mountain top already close to mountain have led to the extinction of any arctic elements not already refugia, dissolution of the Loch Lomond icecap would consequent rapid dissolution refugia, and the consequent dispersal of biota to offshore islands have provided ice rafts and freshwater for the dispersal species of include several species Bay faunas include Brighouse Bay (cf. The Brighouse 1988). The Buckland 1988). (cf. Buckland Coleoptera which have yet to be found in Scotland, and even allowing for our inadequate inadequate knowledge of the present fauna, their presence in the early Holocene Yorkshire merits some discussion. At least ten species occur no further north than Yorkshire indicate summer and Lancashire, and cumulatively these might be seen to indicate pinehole borer broad-leaved pinehole temperatures slightly warmer than at present. The broad-leaved temperatures (Xyleborus [=Anz'sandrus [=Am'sandrus of Bishop and Coope 1977] dispar) is largely restricted to Scandinavia, deciduous tree. In Scandinavia, south-east England, boring in various species of deciduous distribution with the 16 °C July correlation of its distribution Lekander et al. (1977) note a correlation Brighouse today. The The Brighouse warmer than today. °C warmer up to 2 °C isotherm, implying temperatures up isotherm, implying species of four species The four interpretation. The caution in interpretation. assemblage, however, however, requires some caution Odacantha gilvz'pes, B. fumigatum, Agonum thoreyi and Odacantha ground beetle, Bembidion gilvipes, wetland melanura, with the sedge smut beetle (Phalacrus caricis), are associated with wetland polluted habitats, all of which have both declined catastrophically and been heavily polluted habitats, during the woodland habitats, with woodland associated with remainder are associated The remainder past century. The the past which have similarly been destroyed, and the dates of extinction of these species from Scotland Scotland is open to some doubt. Dinnin (1993) records the Colydiid Cerylon and histeroides from Hebrides, and Holocene deposits on South Uist in the Outer Hebrides, from mid Holocene period at the Eucnemid Melasis buprestoides survived at least until the post-Roman period otherwise restricted to south of the Whilst otherwise Ayrshire. Whilst Kilmarnock, Ayrshire. Buiston, Buiston, north of Kilmarnock, beetle, Calosoma inquisitor, is recorded from arboreal ground beetle, from old large arboreal the large Border, the Border, 1977). and Coope 1977). (Bishop and Crowson (Bishop Bridge in Invernesshire by Crowson Spean Bridge woodland woodland at Spean and chrysalises caterpillars and the caterpillars moths Torticid moths prey are the and Torticid Geometrid and Its usual prey chrysalises of Geometrid on oak, but it appears long before the immigration of that tree into Scotland at Osborne amelioration noted by Osborne response to climatic amelioration Bay, a pattern Brighouse pattern of response Brighouse Bay, tempting to dismiss the apparent (1974) for several other forest species. Whilst it is tempting the fauna in favour of the implications of climatic implications favour of anthropogenic factors, Crowson climatic (1981, 630—632) has pointed out that the present Scottish fauna includes a number surviving in favourable either surviving relicts, either could be regarded as thermal relicts, of be regarded species which could of species may, flightless as a result of declining temperatures. These may, now flightless microclimates or now microclimates Age’ (Grove Ice Age’ ‘Little Ice through the ‘Little of recent changes through impact of the impact reflect the however, reflect however, 1988), involving thresholds and limited corridors for redispersal during the current current absence of In the absence climate. In nature of initial the nature than the warming, initial Holocene climate. rather than warming, rather complete sequences from the Scottish mainland, the nature of the early several more complete conditions at warming to conditions Postglacial Postglacial climate remains uncertain beyond the rapid warming least equivalent to those at present. compounded Holocene fauna are further compounded The problems of the development of the Holocene (Hunter 1977). Whilst considered (Hunter when the faunas of the Scottish pine forests are considered less secure the macrofossil detailed (Chapter fairly detailed secure macrofossil is less palynological record is fairly (Chapter 5), the the palynological the be falsified distribution patterns modern distribution conclusions based upon modern and conclusions falsified by the may be patterns may and xylophagous species A number of xylophagous assemblages. A nourish(obtaining nourishfossil assemblages. species (obtaining study of fossil study been able have been century, have Scotland in the last century, restricted to Scotland largely restricted from wood), largely ment from ment and expand Forestry Commission the Forestry activities of the the activities of the expand their advantage of Commission and take advantage to take

FAUNAL CHANGE: INSECTS

107

Cossonine the Cossonine and the rusticus) and (Arhopalus rusticus) longhorn (Arhapalus like the the longhorn range southwards. Some, like species with a range range of species along with are, along chloropus]), are, ater [= Rhyncholus chloropusD, (Eremotes ater weevil (Eremotes weevil England at lowland England present in lowland 1993), present Dinnin 1993), and Dinnin (Buckland and Britain (Buckland extinct in Britain now extinct now demise of lowland lowland the demise that it is the and it is apparent that Age, and late Bronze Age, the late until the least until least

pine forests which have led to their decline. Others, like the birch bark beetle (Scolytus ratzeburgi), remain restricted to limited areas of the Highlands (Crowson despite an abundance of apparently suitable habitat in the lowlands. For this 1971), despite

particular species, the fossil record extends as far south at the Somerset Levels particular (Girling 1977); at Thome in South Yorkshire, its characteristic exit holes appear in (Girling virtually all the birches being pulled from the bog during peat extraction and it is difficult to see any reason for its present reduced distribution other than the continuous old forest habitat. Caution must be exercised when destruction of continuous wasastjernae, discussing potential extinctions from Britain. The dytiscid, Agabus wasastjemae, previously only recorded as a fossil from early Holocene deposits at Church Shropshire (Osborne 1972), has since been found living in waterfilled Stretton, Shropshire Stretton, 1992). rotholes in trees in Speyside (Owen et al. 1992). suitably studied sites on the Scottish mainland is only partly The absence of suitably (1993) and Faunas islands. Faunas the islands. others on the and others work by Dinnin (1993) for by recent work compensated for compensated from beneath machair on Barra and the Uists in the Outer Hebrides show a are forest elements are day, but but forest present day, the present not dissimilar wetland assemblage not from the dissimilar from wetland South Uist Uist and Harris, where the ant mid—Holocene faunas from both South evident in mid-Holocene

lemam‘i) dominates the assemblages during the initial phase of paludifi(Formica lemanii) forest element Uist, the forest cation of the forest floor. At South Lochboisdale on South Uist, 03innin and litter leaf and species as well as leaf xylophagous species of xylophagous number of species (Dinnin litter species to a number extends extends to work on a small wooded island in Loch Druidibeg, and work 1993) and Druidibeg, South Uist, shows 1993) surviving in a fragment still surviving several impact of The impact habitat. The fragment of this habitat. several of these species still

human activity is evident in all faunas from the mid Holocene Holocene onwards. Changes in from the Mainland, Shetland, resulting from the trophic status of Loch of Brunnatwatt, Mainland, the the attested by the fluctuation in the activity of humans in the local catchment, are attested from the basis midges). On the chironomids (non-biting evidence from beetle evidence basis of beetle (non-biting midges). abundance of chironomids and Unst in Shetland, Hoy in Orkney and Outer Hebrides, Hoy the Outer the picture is of Shetland, the the expansion of heathland and raised mire habitats, alongside a rising frequency of dung feeders and other probable anthropochores (cf. Girling and Greig 1977). The biogeographically important fauna from Skara Brae illustrate this point well, and ‘culture-steppe’ landscape implies that the anthropochorous fauna so typical of the ‘culture-steppe’ the the assemblage as early recognizable assemblage day coalesced into present day the present of the early as the into a recognizable of associated with Neolithic period. At Skara Brae, the predominance of the faunas associated stored hay, and decomposing plant and animal waste, including dung, are in accord with other palaeoecological evidence which indicates a farming economy based upon pastoral activity. The occurrence of the human flea (Pulex irritans) provides the earliest European record of a species which has probable New World origins 1989). Sadler 1989). and Sadler (Buckland and (Buckland transported been transported have been insects have anthropochorous insects many anthropochorous initial habitats, many their initial From their around the globe and several are now cosmopolitan in distribution. This by humans around is very evident in pest species which are associated with stores of food and other and weevil (Sitophilus granarius) and pests, the grain weevil grain pests, example, the grain For example, resources. For the saw-toothed grain beetle (Oryzaephilus surinamensis), first attain widespread

108

P. C. BUCKLAND BUCKLAND AND J. P. SADLER

distribution distribution along with the Romans (Buckland 1981). Members of both genera, along with the large ground beetle, Laemostenus sp., are recorded from the Roman fort at Bearsden, about AD and dating to about Glasgow, and Antonine Wall in Glasgow, the Antonine lying on the Bearsden, lying 142—158 (Dickson movement indicate further movement periods indicate from later periods Data from 1979). Data (Dickson et al. 1979). of species (Sadler 1991) household picture of squalid household consistent picture and offer a consistent 1991) and conditions. conditions. The latter is particularly evident in the Dipterous evidence from sites such as the Dark Age crannog at Buiston, where records of the housefly (Musca constitute the earliest positive records of the species in Scotland, although domestica) constitute it is recorded from Roman and later deposits in England as far north as Carlisle (Kenward et al. 1991). In some levels at Buiston the species reached almost plague populations, and under such conditions, incidences of myiasis and fly—bome fly-borne diseases may have been common in the human population (P. Skidmore, pers. com., 1994). Both Coleoptera Coleoptera and Diptera from the Viking age ‘pit’ at Earl Thorfinn’s farm at Tuquoy on the island of Westray, Orkney, indicate dumped material from a byre. Tuquoy The ethnographic ethnographic record from Scotland highlights the importance of peat as a domestic fuel, and the suite of acidophile insects from Tuquoy suggest that peat may also have been utilized as litter in the byre, a practice known from Medieval deposits at organic deposits the organic from the Humle 1991). and Humle (Fredskild and Norway (Fredskild 1991). Several species from Norway Tuquoy are no longer recorded from Orkney and the small clambid, Calyptomerus dubius, is only known as far north as the Scottish lowlands at the present day. This changing farming practices during the last may reflect the loss of habitats due to changing necessary before any firm conclusions can further modern collecting is necessary century, but further

be drawn. A fauna from the Biggins, Papa Stour, Shetland, provides the only other been Shetland has been modern fauna of Shetland The modern assemblage in Scotland. The dated Norse assemblage Bacchus (1980) and it is deficient in anthropochores. Not surprisingly, collated by Bacchus four species recorded from the Norse deposits are not presently known from the island. Of these, these, the blind colydiid, Aglenus brunneus, has been subjected to detailed island. discussion discussion by Kenward (1975, 1976). The species is found in accumulations of first occurs in archaeological plant material and sour grain residues. It first rotting plant rotting

deposits where it is a deposits in the Roman period and is more common in Norse deposits fossil from Dublin Dublin (Coope 1981) and York (Hall et al. 1983). The single frequent fossil Norse represents a Norse almost certainly represents Biggins almost from the Biggins recorded from specimen recorded specimen

introduction. Smith’s (1996) study of a black house on South Uist not only provides a useful analogue for resolving the archaeological assemblages from rural Scottish sites, but affords new records of insects not previously recorded from the Hebrides. The picture, however, remains far from complete. Outer Hebrides. complete.

ACKNOWLEDGEMENTS ACKNOWLEDGEMENTS

This chapter is an attempt to summarize the work of a number of specialists working in Scotland and it is recognized that it is impossible to do justice to all the detailed studies that have appeared. Particularly thanks are due to Barrie Andrian, Barber, Lynn Ian Armit, John Barber, Lynn Barnetson, James Barrett, Stephen Driscoll, Paul Barbara Noddle, Tanya O’Sullivan, Derek Simpson and Dale Hill, Barbara Halstead, Peter Hill, Paul Edwards, Paul comments of Kevin Edwards, The cements unpublished data. Sarjeantson data. The Sarjeantson for access to unpublished

Halstead and Ingrid Mainland are also acknowledged. Halstead

7

The Mesolithic FINLAYSON AND KEVIN J. EDWARDS BILL FINLAYSON

INTRODUCTION The term Mesolithic is generally used to refer to the period from the onset of Holocene warming, c. 10 000 BP, to the local introduction of agriculture associated associated with the Neolithic. In some areas, Mesolithic hunter—gatherers and Neolithic Neolithic agriculturalists co—existed co-existed within the same region, exploiting different parts of the agriculturalists environment. This may have been the case in Scotland, as indicated by an overlap environment. overlap of radiocarbon dates (Morrison 1980; 1980; Bonsall 1990) and by radiocarbon dates Bonsall and and Smith Smith 1990) reconstructions of environmental evidence (Edwards environmental evidence reconstructions (Edwards and Ralston 1984; Edwards Edwards

1988; Chapter Chapter 5). Although the onset of the Neolithic has been considered considered to be primarily an economic change, it has also been assumed that the social systems of hunter—gatherers and and farmers were were different different and hunter—gatherers and largely largely incompatible, and and that there was also a social transformation (Thomas 1991). Research Research on the European Mesolithic has shown it to be part of a phenomenon which extends to the Middle East, marked by an increasing use of microliths (small tools made on deliberately modified similar materials, often shaped modified flakes of flint or similar into geometric was a period onwards. It was Upper Palaeolithic onwards. late Upper the late from the geometric forms) from period when many specifically producing a great many and producing rapidly and changing rapidly was changing society was when society specifically local adaptations. Some of these accommodations accommodations in easterly areas led directly to adaptations. farming, while on the western fringes of Europe exploitation Europe there was an intensive exploitation of the environment. In southern particularly good evidence Scandinavia, where particularly southern Scandinavia, survives, intensification and specialization permitted a degree of Mesolithic Mesolithic sedentism, and arguably the introduction of a number number of material material and possibly social traits that are also conventionally associated associated with farmers, such as pottery and social hierarchies (Albrethsen and Brinch Petersen 1976; 1976; Larsson 1984; 1984; RowleyRowley-

Conwy 1985). The difference between hunter—gatherers and farmers appears appears to be especially diminished amongst communities who depended heavily on marine sources. importance given to characterized by the importance Modern Mesolithic research has been characterized environmental archaeology. Clark (1980, 38) argued argued that after the 19303, ‘It soon became apparent apparent that the most promising way of gaining a picture of the achievements of the inhabitants of Europe between the end of the Ice Age and the adoption adoption

Scotland: Ralston. M.. Ralston. Edwards and Ian B. M Scotland Environment and Archaeology, 8000 BC — AD 1000. Edited by Kevin J. Edwards © © 1997 The editors and contributors. Published in 1997 by John Wiley & Sons Ltd.

B. FINLAYSON AND K. J. EDWARDS

110

////////A/////

Plate 7.1 7.1

Ritchie Copyright: J. N. G. Ritchie Mesolithic Mesolithic shell middens on Oronsay. Copyright:

of a Neolithic way of life was to adopt an ecological approach and deploy the full approach can best be seen in the armoury Research.’ In Scotland this approach Quaternary Research.’ armoury of Quaternary Plate 7.1), on 1987; Plate 1978, 1987; 1978, (Mellars middens shell Oronsay the on work conducted middens Oronsay conducted the reinterpretation of Morton, Fife (Deith 1983, 1986), on recent work at Ulva

vegetational Cave, close to Mull Mull (Russell et al. 1995) and in studies relating past vegetational Ralston history to early 1989a, 1990, 1996a; Edwards and Ralston settlement (Edwards 1989a, early settlement 1990). Newell 1990; Edwards and Hirons 1984; 1990; Edwards 1984;

A separate strand of research research has focused on Mesolithic chipped-stone technology. This work has had considerable considerable significance given that many Scottish composed principally are sites Mesolithic scatters of flint or equivalent material. composed of scatters principally Mesolithic dominated been recently until have studies These Lacaille, most work of Lacaille, dominated by the work recently These studies have Scotland impetus has new A 1954. in Scotland in Age Stone The of publication his notably publication notably Farm, Kinloch from results of publication the by inquiry of line this to given been Farm, Kinloch from publication inquiry been given modern a of publication comprehensive first the 1990), Rhum (Wickham—Jones comprehensive (Wickham-Jones Mesolithic Previously, lithic studies tended to concentrate Scotland. Previously, excavation in Scotland. Mesolithic excavation on technology and the availability availability of raw material. The database has only recently to permit more enough become substantial tentative elaborate, although as yet tentative more elaborate, substantial 1990a). (Finlayson reconstructions reconstructions (Finlayson 1990a). Modern research has for human settlement the earliest evidence for on the has also focused on in Scotland and its extent, and regional contexts rather than single sites are emphasized. This particularly applies Hebrides (Mithen et applies to work in the southern Hebrides and Mithen 1996; al. et Finlayson 1995; Mithen and Edwards 1992; a]. 1996) and Lake 1996) al. and al. and around Oban (Bonsall and Sutherland 1992; Bonsall 1996). Sites mentioned in 1996). Sites

the text and key areas are shown Figure 7.1. shown in Figure

111 1 11

MESOLITHIC THE MESOLITHIC

I

4

3

5

'

o i

‘

9. aside D aside

.

13

11 44 1 5 \K

Valley weed Valley Tweed

‘

"

2 25

23'

O |—..._|_I_L_ha_| 100kui 100km

Galloway Hills

95‘ _

WIS"

50

Figure 7.1 Sites and key areas mentioned in the text. 1: Millfield Farm; Farm; 2: Bay of Sannick; Sannick; 3: Inchnadamph; 6: An Freswick; 4: Bettyhill; 5: Inchnadamph; An Corran; 7: Inverness; 8: Kinloch; 9: Braeroddach Loch; 10: Ballevullin; 11: l l : Ulva Cave; 12: Braeroddach l2: Carding Mill Bay and Druimvargie Druimvargie Rockshelter; 13: North North Mains; Rockshelter; Mains; 14: Friarton; 15: Morton; Morton; 16: Caisteal nan nan Gillean I; 17: Lussa Wood; 19: Bridgend; Glenbatrick; 18: Bridgend; 21: Gleann Mor; 22: Bolsay l8: Lussa Bay and Lussa Bolsay Rhoin Farm; 24: Moorlands; 25: Starr Starr Farm; 23: Rhoin

112

B. FINLAYSON AND K. J. EDWARDS

THE MESOLITHIC IN SCOTLAND In a recent review, Woodman research Woodman (1989) painted a bleak picture of the state of research and the lack of evidence. He maintained that there was no workable chronology, no workable chronology, ‘feeling for of human occuthe extent of on the no certainty on of economy’, no the type(s) of for the pation, types of artefact used, or indication Mesolithic was a single that the Mesolithic indication that 1989, 1). While there are economic problems, this undeniable problems, are undeniable (Woodman 1989, economic entity (Woodman picture is perhaps argument, of argument, the sake of 5), for the (1989, 5), Woodman (1989, too pessimistic. Woodman perhaps too has indicated indicated that Scotland would have one of the lowest densities of Mesolithic sites in Europe, perhaps one per 800 kmz, if a strictly defined diagnostic trait such alternative assessment as microliths was chosen as a basis for estimation. An alternative fauna-bearing Mesolithic sites would, however, place founded on the density of fauna-bearing Scotland as the fourth richest of 17 west European countries (Andersen et al. 1990). The Mesolithic environment Scotland environment of Scotland natural environment afforestation, the natural rapid afforestation, and rapid deglaciation and After deglaciation sea with fresh and sea consisted of a varied woodland landscape, which, combined with

water, provided an abundance of resources for hunter—gatherer groups. The hunting, gathering harvesting wild resources by hunting, resultant economy was based on harvesting resultant and fishing. Much environmental research is aimed at reconstructing the resource Mesolithic peoples. base and at assessing the levels of environmental impact by Mesolithic deliberate modification there evidence for deliberate minor and accidental, or is there Was impact minor Was

(Smith 1970; Mellars 1976a; to the natural argued (Smith natural environment? It has been argued Simmons et al. 1981) that fire could have been used deliberately, either directly as environment, both to improve grazing strategies, or to modify the environment, part of hunting strategies, for deer and to encourage hazel growth for food and the supply of twigs (Morrison charcoal in 1980). Edwards (1990) examined the evidence provided by microscopic charcoal fire-climax represent a fire-climax Scotland and one conclusion reached was that hazel did not represent markedly palynological sites there are markedly vegetation type (Chapter 5). On many palynological vegetation quantities of charcoal during the Mesolithic. These may diachronous increases in the quantities diachronous be indicative of local burning, perhaps perhaps from anthropogenic fires, either domestic or even for land management (Edwards et al. 1995), but natural conflagrations made possible by climatic dryness have also been mooted (cf. Tipping 1996). occupation in areas where Mesolithic occupation Interestingly, the fire record may indicate Mesolithic (Edwards et no artefactual artefactual evidence has yet been recovered, e.g. in the Western Isles (Edwards 1992). In such cases, it is al. 1992). and Shetland Edwards 1996a) and Shetland (Bennett et al. 1995; Edwards al. 1995; suggested that bone has disappeared in acid soils, while lithics are likely to be which have risen since the coastal sand and sea waters which hidden the early beneath peat, coastal hidden beneath Holocene (Edwards 1996a; Plate 7.2). The pollen record is full of other instances, not necessarily fire-related, where

Mesolithic occurred during the Mesolithic both major and minor reductions in woodland occurred attractive to (Chapter 5). One effect of these would be the provision of open areas attractive vertebrate fauna record does not animals. Unfortunately, the inadequate vertebrate grazing animals. provide clear indications of the extent and survival of all species through the provide (although their presence in Shetland and Mesolithic. Mesolithic. Red deer are well represented (although evidence, which, if the Outer Hebrides is only an inference from palynological evidence,

THE MESOLITHIC

1l 13

/

///

/ /

The machair at Bagh Plate 7.2 The Bagh Siar, Siar, Vatersay. Vatersay. Sea-level Sea-level rise in mid mid Holocene Holocene times and the movement movement of calcareous ‘machair’ sands inland inland may may have concealed concealed evidence evidence for Mesolithic Mesolithic occupation Copyright: K. J. Edwards Western Isles. Copyright: the Western occupation in the

correct, would imply human transport of deer: Bennett et al. 1992; Edwards 1996a). 1996a). Roe Roe deer, elk, aurochs, brown bear and boar were also present (Ritchie (Ritchie 1920; 1920;

Simmons et al. a1. 1981; Chapter 6). Coastal Coastal locations locations (Lacaille (Lacaille 1954; 1954; Coles Coles 1971; Mellars Mellars 1978) 1978) have produced produced evidence for otter and possibly domestic dog, as well as, for example, grey seal, common common porpoise porpoise or dolphin, field vole, common shrew and red squirrel. squirrel. Amongst Amongst birds, razorbill razorbill and guillemot the birds, guillemot were were probably probably used for food, food, and and fish fish remains remains and eel. include whiting and pollack, whiting haddock, pollack, cod, saithe, haddock, include cod, Mesolithic early Mesolithic Colonization and the early Colonization

seems reasonable Although it seems reasonable to suppose suppose pre-glacial and Lateglacial human occuScotland, much of the evidence pation of Scotland, evidence will have been destroyed destroyed by such such events events

as glaciation and rises in sea level and there is no definitive evidence evidence in Scotland for any pre-Mesolithic occupation. occupation. A review of lower Palaeolithic Palaeolithic tools found in Scotland reports that all are most likely recent losses from collections collections made elsewhere (Saville 1993). Late Palaeolithic occupation occupation had been proposed at Inchnadamph, Inchnadamph, Assynt, Assynt, where where there there is possible possible evidence for human human exploitation exploitation of reindeer (Lawson and Bonsall 1986). Radiocarbon dates now indicate indicate that that the material material in question may rather be part of a sequence of natural deposits (Murray (Murray et et

al. 1993), albeit extending back into Lateglacial times. A flint scraper stratified within a marine core taken between the Shetland Shetland Islands Islands and Norway, if not secondarily derived, suggests either human occupation occupation of land land

114

Figure Figure 7.2

FINLAYSON AND K. J. EDWARDS B. FINLAYSON

Mesolithic artefacts. 1, 2, 3 and 7: narrow blade microliths; 4 and 5: broad blade

microliths; 6: tanged point; 8: perforated shell; 9 and 10: bevel ended tools; 11: barbed bone microliths; (1994) Wickham-Jones (1994) not to scale). After Wickham-Jones 11 not point (number 11

RP, which had been exposed by low sea levels before the Holocene, at about 18 000 BP, a]. 1986), or a stone tool lost certainly submerged again by 10 000 BP (Long et al. and certainly Postglacial overboard during a fishing expedition. There is evidence for rapid Postglacial north of the Arctic colonization colonization in Norway (Bang-Andersen 1989) and occupation north high-latitude (Engelstad 1989). This evidence of high-latitude Circle before 9000 BP (8030 cal BC) (Engelstad settlement settlement may suggest early colonization of Scotland from the east. This hypothesis is possibly supported by the tanged points of proposed Lateglacial and early similarities to examples initially identified Postglacial Ahrensburgian affinity, with similarities Barton 1989). These are recorded as stray finds finds northern Germany (Taute 1968; Barton in northern

(Orkney), Farm and Brodgar (Orkney), from Ballevullin (Tiree) (6 in Figure 7.2), Millfield Farm Bridgend (Islay; and Bridgend Bay (Jura), and Lussa Bay findspot) and Lussa only precise findspot) (Islay; the only Lussa Wood and Lussa (Livens 1956; Mercer 1980; Edwards and Mithen 1995). Doubt has been cast on the (Morrison and rolled condition condition (Morrison heavily rolled and heavily broken and examples because of their broken Jura examples

THE MESOLITHIC

1l 15

context, and radiocarbon-dated context, Bonsall 1989). None of the finds comes from a radiocarbon-dated circumstantial evidence for an pollen evidence from Islay allows, at best, only circumstantial Berridge immediately Postglacial presence elsewhere on that island (Edwards and Bern'dge indication as a possible indication The group retains interest as Mithen 1995). The and Mithen 1994; Edwards and occupation. of relatively precocious occupation. non-geometric characterized by non-geometric Mesolithic is characterized south in Britain, the early Mesolithic Further south Further artefacts 1974, 1976b). Similar artefacts Mellars 1974, 1973; Mellars artefacts (Jacobi 1973; broad-blade flint artefacts Glenbatrick convincingly from Glenbatn'ck have been recovered from a few Scottish sites, most convincingly reasonable to on Jura (Mercer 1974) (4 and 5 in Figure 7.2). At present it seems reasonable narrow-blade assume that the appearance of such broad-blade assemblages precedes narrow-blade the presence of from the has argued from (1988) has 1988). Bonsall (1988) also (Myers 1988). ones in Scotland also mixed assemblages, at both Morton (Coles 1971, material, albeit in mixed broad-blade material, BP sites predate 9000 BP these sites of these that elements of (Mercer 1980), and ~Lussa 1983) and 1980), that _Lussa Wood (Mercer 1983)

(8030 cal BC). This dating is founded mainly on a consideration of local sea-level considered the location of the sites is considered 1984), and the Sutherland 1984), 1979; Sutherland (Dawson 1979; change (Dawson change reasonable that resources. It might thus seem reasonable marine resources. to result from exploitation of marine The cal BC). The BP (8030 cal 000 to 9000 BP from 11 000 assemblages date from broad-blade assemblages the broad-blade the west to Fife in the east, from Jura in the west sites with such material, from distribution of sites and the from Orkney and southern the southern the evidence of tanged points from combined with the

Hebrides, may suggest that most of Scotland was colonized before 9000 BP (8030 Hebrides, cal BC). More secure evidence for this earliest occupation of Scotland remains to be found. Later Mesolithic (from c. 9000 BP)

Mesolithic occupation of Scotland is at present insubstantially insubstantially While an early Mesolithic there is, by the later Mesolithic, clear evidence for established established settlement. attested, there

occupation site are those from excavations at The earliest radiometric dates for an occupation extending back to 8590i95 BP (7700—7500 cal BC) Kinloch, Rhum (Plate 7.3), extending on most Scottish sites, (Wickham-Jones 1990). There, as on the assemblage is characsites, the (Wickham-Jones component (1, 2, 3 and 7 in Figure terized by a geometric narrow-blade microlithic component 7.2). Scrapers form the other main artefact type, but it is notable that unlike in England (Mellars (Mellars 1976b), microliths dominate most assemblages regardless of site Invemess (Wordsworth Narrow-blade assemblages have been found at Inverness location. Narrow-blade location. Morton (Coles 1971), Deeside (Kenworthy 1981), the Solway coast 1985), Morton (Cormack and Coles 1964; Cormack 1970), the Galloway Hills (Affleck 1986), the many west coast islands: from many 1970) and from Valley (Mulholland 1970) Arran (Allan islands: Arran Tweed Valley and 1971, 1972; Mercer and (Mercer 1970, 1971, Jura (Mercer 1988), Jura Affleck et al. 1988), Edwards 1987; Affleck and Edwards

Searight 1987), Islay (McCullagh 1989; Mithen 1990) and Colonsay Colonsay (Mithen 1989; Mithen and Finlayson 1991), as well as Rhum. The most northerly northerly mainland Caithness sites of Freswick (Lacaille 1954) and possibly the material comes from the Caithness Bay of Sannick (Pollard and Humphreys 1993), and from Bettyhill, Sutherland (Wickham-Jones (Wickham-Jones and Firth 1990). Material has also been reported from Orkney (Wickham-Jones 1990; and see Saville 1996). Within the limits of admittedly very (Wickham—Jones can now be suggested that later Mesolithic occupation extended research, it can patchy research, patchy probably fairly dense. At some sites, Scotland and in some areas was probably throughout Scotland such as Kinloch (Wickham-Jones 1990) and Bolsay Farm, Islay (Mithen et al.

‘116 '116

EDWARDS B. FINLAYSON AND K. J. EDWARDS

Kinloch, Isle of Rhum with the Cuillins of Skye in the background. The setting of Plate 7.3 Kinloch, excavation site lay Mesolithic excavation Scotland. The Farm Fields Mesolithic occupation site in Scotland. the earliest known occupation to the centre bottom of the picture, just above Loch Scresort, and contained charred hazelnut

indicated a Mesolithic (bottom left) indicated shells dated to 8590i95 BP. The Kinloch pollen site (bottom experienced regular prominent, and which experienced locally prominent, were locally alder were hazel and alder landscape landscape in which hazel have communities have vegetational communities phases of woodland reduction. sedge-dominated vegetational and sedge-dominated Grass and reduction. Grass time existed since appearance of cereal pollen and soil erosion at this time 3950 BP, and the appearance since 3950 suggests that Neolithic Copyright: K. J. changes. Copyright: responsible for the landscape changes. activities were responsible Neolithic activities Edwards

1992), hundreds excavated artefacts have been recovered from the excavated hundreds of thousands of artefacts recurrent result of recurrent the result accumulated as the areas alone, have accumulated assemblages have these assemblages alone, although these or continuous occupation. The Mesolithic settlement of Scotland is clearly more wanderings of a single family’ substantial than occasional foraging trips or ‘the wanderings (Atkinson 1962, 6). The later Mesolithic comprises not only the narrow-blade flint scatters, scatters, but also the ‘Obanian’ sites (Turner 1872, 1895; Grieve 1882; Anderson 1895, 1898; Bishop Bishop 1914; Lacaille Lacaille 1954; Pollard 1990; Bonsall 1996). These consist of midden deposits,

preserving a wide range of fauna] faunal material including shells and the bones of fish, dietary important dietary birds and mammals, with fish probably representing the most important element. In addition, a range has survived. found elsewhere has not found artefacts not range of organic artefacts These include barbed points and antler and bone and antler mattocks, and points (harpoons), antler bevel-ended tools, often referred to as ‘limpet scoops’ (8, 9, 10 and 11 in Figure Figure Chipped-stone artefacts are also represented, but rarely microliths. This 7.2). Chipped-stone industry platformbipolar knapping, rather than the platformmostly been produced by bipolar industry has mostly core technique associated associated with the narrow-blade assemblages. While it would no longer be argued that the ‘Obanian’ (a term first used by Movius [1940b, [194%, 76]),

THE MESOLITHIC

1l 17

represents group of sites cultural entity, there is clearly discrete cultural represents a discrete clearly a group sites with a distinctive distinctive set of artefacts. Their relationship with other Mesolithic (and early Neolithic) sites is problematic. problematic. The discovery of inter-stratified ‘Obanian’ and microlithic material in rock shelter Corran, Skye (Saville and Miket 1994) promises to be of shelter deposits at An Corran, major importance major importance in furnishing furnishing a key sequence pertinent pertinent to this problem. problem.

THE NATURE OF THE EVIDENCE archaeological perceptions and archaeological Site Site types and

available evidence from Scotland is undoubtedly of restricted value for Much of the available interpretation. The lithic scatters are rarely stratified and apparently have few interpretation. fieldwalking associated structural remains (although most scatters are known from fieldwalking

excavation: cf. Mulholland 1970; Edwards et al. 1983). For many sites, rather than excavation: it is difficult to judge how many occupations are represented or their duration, and basic parameters parameters such as group size and site function remain elusive. Lithic scatters scatters are generally open air, while midden deposits occur both in the aVIellars 1987) and in caves. Open-air midden sites perhaps appear open (Mellars appear simple to disposal heaps but although assess, but assess, although these disposal heaps produce produce an abundance abundance of economic economic data, Oronsay, apparent On Oronsay, interpret. On to interpret. these can be hard to these apparent seasonality seasonality on four four middens middens a1. 1985; Mellars 1987); it is impossible, elegantly demonstrated (Andrews et al. was elegantly however, to quantify the scale of occupation or the frequency of visits. Some cave disturbed since the deposits disturbed contain deposits but contain records, but stratigraphic records, better stratigraphic have better sites sites have the Mesolithic Mesolithic by subsequent human use. Study of the period has not been assisted by the division in primary interests between researchers deposits) environmental evidence (fullest in midden deposits) researchers interested in environmental and those concerned with information on lithics (best served by flint scatters). The latter sites generally provide much less direct evidence for subsistence, while the lithic industry category is not accorded a high priority for industry from sites in the former category publication (e. (e.g. publication g. Oronsay). This distinction is almost certainly over-simplification. certainly an over-simplification. The narrow-blade while the narrow-blade artefact content, and artefact form and sites vary in form midden sites shell midden The shell content, while Farm, Islay, assemblages vary from large sites such as Kinloch, Rhum and Bolsay Farm, to small, possibly single occupation sites as at Gleann Mor, Islay (Mithen et al. 1992).

Chronological framework Chronological

Mesolithic has generally been regarded as chronological framework for the Mesolithic The chronological inadequate because of poor site stratification and the frequent reoccupation of sites. inadequate On such palimpsest sites, only radiometric dating of numerous features can date the coarse to clearly too coarse temporal resolution is clearly the temporal and the within them; and elements within elements distinguish annual or seasonal visits. The difficulties involved are illustrated by the distinguish Morton in Fife chronological interpretations of the site at Morton debate surrounding the chronological debate 1988; Bonsall (Coles 1971, 1983; Myers 1988;

Wickham-Jones 1988).

1988; Clarke and 1988; Woodman 1988;

EDWARDS FINLAYSON AND K. J. EDWARDS B. FINLAYSON

118

Guam (EU-2147 60-2145 GU-2145

H ;

"’ U

——-— —— + — —