Roman Bath E-Book

Peter Davenport

• Herausgeber: The History Press
• Kategorie: Geisteswissenschaft
• Sprache: Englisch

Seitenzahl: 327

Veröffentlichungsjahr: 2021

Beschreibung

For almost three hundred years, excavations have been carried out in Roman Bath. At first these were rare and sporadic and archaeological finds were made by chance. Even fewer were reported. But from the 1860s, deliberate investigations were made and increasingly professional methods employed. The Roman Baths were laid open to view, but little was published. From the 1950s, interest accelerated, professionals and amateurs collaborated, and there was never a decade in which some new discovery was not made. The first popular but authoritative presentation of this work was made in 1971 and updated several times. However, from the 1990s to the present there has been some sort of archaeological investigation almost every year. This has thrown much new and unexpected light on the town of Aquae Sulis and its citizens. In this book, Peter Davenport, having been involved in most of the archaeological work in Bath since 1980, attempts to tell the story of Roman Bath: the latest interim report on the 'Three Hundred Year Dig'.

Leseprobe

To Lisa, Sarah and Beccaand in memory ofJohn C. Clarke, amicus romanorum

First published 2021

The History Press

The Mill, Brimscombe Port

Stroud, Gloucestershire, GL5 2QG

www.thehistorypress.co.uk

© Peter Davenport, 2021

The right of Peter Davenport to be identified as the Author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without the permission in writing from the Publishers.

British Library Cataloguing in Publication Data.

A catalogue record for this book is available from the British Library.

ISBN 978 0 7509 9643 3

Typesetting and origination by Typo•glyphix, Burton-on-Trent

Printed in Great Britain by TJ International Ltd, Padstow, Cornwall

eBook converted by Geethik Technologies

CONTENTS

Acknowledgements

List of Illustrations

Foreword

Introduction

Prelude

1 The Romans Arrive

2 The Army at Walcot

3 Taming the Waters

4 Early Baths and Temple

5 The First 100 Years Around the Temple and Baths

6 The First Town of Aquae Sulis

7 The Mid-Second Century: Another Beginning

8 The Monumental Centre

9 The Town Around the Baths

10 People of Aquae Sulis

11 The Countryside of Aquae Sulis

12 Late Roman Aquae Sulis

13 The End of Aquae Sulis

Afterword: The Three-Hundred-Year Dig

Bibliography

Notes

Index

ACKNOWLEDGEMENTS

I would first like to thank Stephen Clews, whose title of Roman Baths and Pump Room Manager does not do justice to his academic knowledge and interest in the archaeology of the Roman Baths. He asked if I would write a new and up-to-date survey of the Roman town that owed its existence to the baths and spring. His colleagues at the museum have been unfailingly helpful, especially Susan Fox, collections manager and her assistant, Zofia Matyjaskiewicz, in providing access to the museum’s archaeological archives and images and providing new ones. I must also thank my colleagues at the former Bath Archaeological Trust, especially Mark Beaton, Robert Bell and Marek Lewcun, who ran or supervised nearly all the archaeological investigations in the town between 1983 and 2005. Marek has also been involved in most of the work since that latter date and has helped fill in gaps in the visual record with images he had taken.

Photographs and other images are individually credited, except as follows, and with the exception of Fig. 2, whose owner I have not been able to trace. Should he or she come forward, proper credit will be given. I should also like to thank Ian R. Cartwright, Chief Photographer of the Oxford Institute of Archaeology, for providing images from the Institute’s archive, and Oxford University School of Archaeology and the Library of the National Museums Scotland for allowing the reproduction of images without charge. The Ussher Society also allowed the use of Figs 5 and 6 free of charge. Cambridge University Collection of Aerial Photography also kindly allowed reproduction without charge but were unable to provide the image itself for reasons beyond their control. Other specific permissions are as follows: Figs 24, 71 and 107 are reproduced by kind permission of the Society of Antiquaries of London. Fig. 29 is reproduced courtesy of the British Library and Fig. 100 of the British Museum. Fig. 92 is copyright West Air Photography, now part of the Historic England Archive. As the archive has been completely closed during the Covid-19 pandemic, it has not been possible to formally acquire permission. The situation will be regularised as soon as it becomes possible to do so.

My wife, Lisa Brown, acted as proofreader and added what elegance there might be to the text. Finally, I have to acknowledge Emeritus Professor Sir Barry Cunliffe, who brought his ex-student to Bath all those years ago and generously provided the foreword to this edition.

LIST OF ILLUSTRATIONS

Chapter 1

Fig. 1 Flints from the Hot Bath spring. (Roman Baths Museum)

Fig. 2 The hot springs as they may have looked before the Romans came (actually a spring in Armenia).

Fig. 3 The causeway into the sacred spring (based on Cunliffe and Davenport, 1985, Fig. 27).

Fig. 4 Celtic coins from the spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 5 The deep geology of the Bath area and the spring catchment (redrawn from Gallois, 2006, Fig. 2).

Fig. 6 The long-buried springs break through at Bath in the late Pleistocene (redrawn from Gallois, 2006, Fig. 6).

Fig. 7 The Dobunni and their neighbours.

Fig. 8 Roman roads and the Avon Valley topography around Aquae Sulis.

Fig. 9 Bath, sitting in the Avon Valley, looking south-west from Solsbury Hill. The valley continues towards the Severn on the right.

Fig. 10 The Fosse Way road surfaces and roadside ditch exposed at Hat and Feather Yard. (Marek Lewcun)

Chapter 2

Fig. 11 The tombstone of cavalry trooper Lucius Vitellius Tancinus. (Roman Baths Museum)

Fig. 12 The octagonal building at Hat and Feather Yard. It has been cut into the hillside and then cut away itself on the left by a later building. (Marek Lewcun)

Fig. 13 Early, probably pre-AD69, pottery from Nelson Place and Hat and Feather Yard. Locally made, military-type flagons and a honey pot and a complete example of one of the imported amphora types found at Walcot (here from Verulamium, © Verulamium Museum), common on military sites, usually containing olives or olive pressings.

Fig. 14 The early road junction at Walcot and the octagonal building.

Fig. 15 Excavations at Bathwick, one of the early Roman, probably military ovens. (Marek Lewcun)

Fig. 16 Early, typically military, glass and pottery from Nelson Place. (© Bath and North East Somerset Council, Roman Baths Museum)

Chapter 3

Fig. 17 Elevation of Temple front (after Cunliffe and Davenport, 1985, Fig. 11).

Fig. 18 Coins of Nero from spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 19 The spring under construction. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 20 Oak piles in the spring and wall during the 1979 dig. (© Barry Cunliffe/Oxford University School of Archaeology)

Fig. 21 Sluices from the baths and possibly of the sacred spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 22 The baths and temple as first built, around AD 70.

Fig. 23 The south wall of the spring and the West Bath windows. (© Barry Cunliffe/Oxford University School of Archaeology)

Fig. 24 The lead lining of the sacred spring tank in 1878. (© The Society of Antiquaries of London)

Fig. 25 The sculptured pediment of the temple, coloured as it might have been.

Fig. 26 Close-up photo of the Gorgon in its roundel. (© Bath and North East Somerset Council, Roman Baths Museum)

Chapter 4

Fig. 27 Plan of the Period 1 baths.

Fig. 28 Photo of the Great Bath drained. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 29 Hoare’s depiction of the East Baths in 1754. (© The British Library Board [Add. MS 21577b])

Fig. 30 Elevation and cross-section of the Great Bath (based on Cunliffe 1969, Fig. 37 with modifications).

Fig. 31 Reconstruction of interior of the Period 1 Great Bath.

Fig. 32 The fallen vault and hypocaust in the West Baths tepidarium in 1869 (Irvine’s record, redrawn from Cunliffe 1969, Fig. 48).

Fig. 33 Reconstruction of the temple precinct in the late period. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 34 Four of the deities on the altar corners, clockwise from top left: Bacchus, Hercules, Apollo and Jupiter. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 35 Minerva’s head. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 36 The ‘Vilbia’ curse and coins from the sacred spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 37 Personal items, bracelets, earrings and brooches, and the ivory handle from a folding blade, probably a cosmetic implement (70mm long) all from the sacred spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 38 The ballista washer (83mm diameter) and gemstones from the sacred spring and the culvert. The gems are typically 10 × 12.5mm. (washer © Bath and North East Somerset Council, Roman Baths Museum and the gems from Cunliffe 2000, Pl. 23)

Fig. 39 The tin mask from the drain (330mm high) and the ivory breast ex-voto (70mm across) from the sacred spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 40 Pewter vessels from the sacred spring and culvert. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 41Paterae from the sacred spring, all inscribed ‘Deae Sulis Minervae’ or variations. The bronze patera is the probable Hadrian’s Wall souvenir. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 42 Drawings of the silver (top) and bronze (bottom) paterae (from Cunliffe, 1988, Figs 8 and 9).

Fig. 43 Fragments of probable priest’s regalia from the sacred spring (© Bath and North East Somerset Council, Roman Baths Museum) and a reconstruction of the filigree as hat decoration.

Fig. 44 A selection of items from the sacred spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 45 The penannular brooch from the sacred spring: probably late fifth century. (© Bath and North East Somerset Council, Roman Baths Museum)

Chapter 5

Fig. 46 Plan of the central part of town AD70–150 (including early ‘administration’ building).

Fig. 47 The early road pre-dating the temple precinct, smoother surface, top left and the junction with the rerouted section, more cobbly surface, bottom right.

Fig. 48 Irvine’s plan of the later road around the precinct (the dark areas show what he was able to see). (By kind permission of National Museums Scotland)

Fig. 49 The early ditch under the courtyard building with its fill of column fragments from the demolished early building.

Fig. 50 The columns reconstructed from the fragments found in the ditch under the courtyard building.

Chapter 6

Fig. 51 Plan of London Street excavations (all periods).

Fig. 52 The early strip building and footbridge at Hat and Feather Yard.

Fig. 53 The rear of the terrace for the first wooden strip building at Hat and Feather Yard. (Marek Lewcun)

Fig. 54 Plan of the second phase of building at Hat and Feather Yard.

Fig. 55 The second century building with probable shrine. (Marek Lewcun)

Fig. 56 The bust from Hat and Feather Yard. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 57 The massive footings of the latest stone building at Nelson Place under excavation (a) and the plan of the latest stone buildings at Nelson Place (b).

Fig. 58 Architectural fragments (large pier and column capitals) from the Methodist Burial Ground at Walcot.

Fig. 59 Plan of Walcot Street/Bathwick settlement.

Fig. 60 Architectural fragments (colonnette and part of a frieze[?]) from the Hat and Feather Yard excavations.

Fig. 61 Plan of the later stone buildings at Hat and Feather Yard.

Fig. 62 Photo of the later back street at Hat and Feather Yard. (Marek Lewcun)

Fig. 63 Reconstruction painting of the street frontage of Hat and Feather Yard by Jane Brayne.

Fig. 64 The plan of the large house at St Swithin’s Yard.

Fig. 65 The side wall of the St Swithin’s Yard house and the lane alongside.

Chapter 7

Fig. 66 The central area in the mid to late second century.

Fig. 67 The north-south road at the west end of the possible garden beyond the temple precinct.

Fig. 68 The mosaic under the Crystal Palace public house.

Fig. 69 Cutaway view of the spring enclosure building. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 70 The stylobate of the eastern precinct portico (a) and the datable deposits against it surviving from the Victorian excavations (b).

Fig. 71 Mann’s record of the eastern wall of the precinct and the ‘monumental structure’. (© The Society of Antiquaries of London)

Chapter 8

Fig. 72 Plan of the Period 2 baths.

Fig. 73 Plan of the Period 3 baths.

Fig. 74 The fallen window arch at the west end of the Great Bath.

Fig. 75 Reconstruction of interior view of Period 3 Great Bath.

Fig. 76 Reconstruction of exterior view of Period 3 baths and temple.

Fig. 77 A section of the box tile and concrete vault over the Great Bath, displayed inverted.

Fig. 78 A portion of the ridge rib and adjacent box tiles of the Great Bath vault of Period 3.

Fig. 79 The collapsed vault in the spring. (© Barry Cunliffe/Oxford University School of Archaeology)

Fig. 80 The stylobate wall of the Period 3 corridor. (© Cotswold Archaeology)

Fig. 81 Façade of the Four Seasons with the Luna pediment superimposed.

Fig. 82 The theatre hypothesis: the evidence.

Fig. 83 The theatre hypothesis: possible reconstructions.

Chapter 9

Fig. 84 Cross-section of the enclosing bank (after Cunliffe, 1969, Fig. 63).

Fig. 85 The courtyard building and its surroundings.

Fig. 86 Plan of the Hot Bath baths. (Based on Irvine’s 1864 drawing, by kind permission of National Museums Scotland.)

Fig. 87 Two dog figurines from the courtyard building and just north of it (drawings from Davenport, 1999, Fig. 1.78 and photo © Bath and North East Somerset Council, Roman Baths Museum).

Fig. 88 Shoes from the cobbler’s rubbish pit on Walcot Street. (© Bath and North East Somerset Council, Roman Baths Museum)

Chapter 10

Fig. 89 Three military tombstones (Gaius Murrius, Julius Vitalis and Marcus Valerius) plus the memorial to Rusonia Aventina. See also Fig. 11 (© Bath and North East Somerset Council, Roman Baths Museum and antiquarian drawings after Cunliffe 1969 Plates LXVII and LXVIII)

Fig. 90 A selection of curses from the spring. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 91 The grave of the elderly lady from Bathampton. (Marek Lewcun)

Chapter 11

Fig. 92 Aerial photograph of Bathampton Down prehistoric and Roman field system. (© West Air Photography)

Fig. 93 Plan of Bathampton Meadows Farm Roman phase.

Fig. 94 A stone pewter mould for the stand of a vessel, from Julian Road, probably early fourth century, and a pewter chalice from the sacred spring.

Fig. 95 Reconstruction painting of the Durley Hill Roman villa at Keynsham. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 96 Location map of the suburban villas.

Fig. 97 Plan of the villa at Lower Common, Bath.

Chapter 12

Fig. 98 Plan of late Roman Bath.

Fig. 99 The discovery of the Beau Street hoard. (© Cotswold Archaeology)

Fig. 100 Bags 1 and 4 of the hoard being ‘unpicked’. (© Trustees of the British Museum)

Fig. 101 The internal face of the Roman city wall at Terrace Walk.

Fig. 102 The external face of the city wall near to Terrace Walk at the Compass Hotel in 1965 (from Cunliffe, 1969, Plate LXXXIV).

Fig. 103 James Irvine’s restoration painting of the collapsed mosaic he discovered under the Gainsborough Hotel in 1864 (by kind permission of National Museums Scotland) and the possible baths mosaic from the Bluecoat School. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 104 Reconstruction of the temple precinct and baths in the mid-fourth century.

Fig. 105 The pediment of the quadrifrons buttress.

Fig. 106 The plan of the Period 4 baths (fourth century).

Fig. 107 Richard Mann’s long section of the drain and the buildings north of it. (© The Society of Antiquaries of London)

Fig. 108 The Great Drain (from a video © Bath and North East Somerset Council, Roman Baths Museum).

Fig. 109 The eastern arch to the destroyed manhole over the Great Drain. Cf Fig. 107. (© Bath and North East Somerset Council, Roman Baths Museum)

Fig. 110 Cross-section and plan of one of the Great Drain manholes (from Cunliffe, 1969, Fig. 41).

Fig. 111 Plan of the Period 4 temple precinct.

Chapter 13

Fig. 112 Photo of late cobblings in the precinct. (© Barry Cunliffe/Oxford University School of Archaeology)

Fig. 113 Reconstruction of the baths and precinct in the fifth century.

Fig. 114 The columned doorway and step of the entrance into the room built between the quadrifrons arch and the eastern buttress in the temple precinct. (© Barry Cunliffe/Oxford University School of Archaeology)

Fig. 115 Wear in the Great Baths walkways. a. Note the unworn early paving under the partly removed later paving and the sharp edge to the worn section where it was protected during erosion by the overlying slabs. b. A remnant of later paving, its edges worn by the wear through it.

Fig. 116 Extreme wear in the paving around the Circular Bath.

Fig. 117 The collapse process of the vault over the spring (compare Fig. 79).

Fig. 118 The fallen rubble of the reservoir enclosure in the temple precinct. (© Barry Cunliffe/Oxford University School of Archaeology)

FOREWORD

When workmen, digging a sewer trench along Stall Street in 1727, uncovered the gilded bronze head of Minerva it heralded the beginning of the long and exciting process leading to the discovery of the remarkable Roman healing shrine of Sulis Minerva, deep beneath modern Bath. That so much is now known of the Roman complex is a tribute to the many who, over the last three centuries, have laboured in the mud, often in dangerous conditions, to explore and to rescue the basic evidence upon which Bath’s former glories can be reconstructed. That the process of discovery continues, and new and surprising evidence may appear any moment, is what makes archaeology such a satisfying pursuit.

Along with the privilege of discovery comes the responsibility to make the new information available to the general public whose story this is. So it was, in 1864, that the Rev. H.M. Scarth published his Aquae Solis or Notices of Roman Bath, bringing together all previous finds including the stunning remains of the temple façade found when the foundations of the Pump Room were dug in 1790. In the years following the publication of Scarth’s book, new finds came thick and fast. Remains of the temple podium were being recorded during building work in the 1860s, soon to be followed by the excavation of the Roman Baths – a hugely ambitious programme driven forward by city architect, Major Charles Davis to the lasting benefit of the city. More of the Baths was exposed in the 1920s but thereafter things quietened down and little that was new was added for several decades.

It was the foresight of Sir Ian Richmond, an archaeologist who had made a study of the Baths, that inspired a renewed effort. In 1963 Richmond encouraged the setting up of the Bath Excavation Committee, of which I was appointed director. Alongside carrying out rescue excavations we decided to focus on completing Richmond’s work on the Baths and to begin a new study of the Temple. The first stage of this programme was published in a scholarly monograph, Roman Bath (1969), and a popular account, Roman Bath Discovered (first edition, 1971). A few years later, encouraged by the city’s farsighted chief executive, David Beeton, we began a more ambitious programme – to excavate the entirety of the temple precinct lying beneath the Pump Room and explore the nearby sacred spring. One of my young colleagues, who played an active part in the work, was Peter Davenport and together we published the spectacular results in our volume The Temple of Sulis Minerva at Bath (1985). To make this work and some later finds more available to the general reader I updated Roman Bath Discovered, the third edition being published in 2000.

In the forty years following the temple excavation, Peter Davenport has continued to play an active role in the archaeology of Bath and has been directly responsible for much of the excavation and publication work. During this time a great deal that is new has come to light, particularly about the town outside the boundary of the sanctuary, while work in and around the Baths has clarified the picture, leading to a fuller understanding and new interpretations. Perhaps the most important result of all this is that it is now possible to present the development of Bath and its sanctuary as a narrative of change spanning the Roman period – that is what the present book sets out to do.

And what of the future? First and foremost there is the need for past excavations to be published – it is the professional duty of archaeologists to do so. Much has been done, but there are still some significant excavations of the last twenty-five years awaiting full analysis and publication. We also need a change in attitude to excavation in the city. It is customary to argue that archaeological deposits should be preserved wherever possible and there is much good sense in this. But such an attitude can hinder research. Every new building proposal offers an archaeological opportunity. That opportunity should be taken in full if there are sound research reasons for doing so, even if it means excavating archaeological deposits that would not otherwise need to be removed. Only in this way will future generations be able to add to our understanding of the remarkable remains of Roman Bath in the way that our predecessors have done so effectively over the last three centuries.

Barry Cunliffe, 2020

INTRODUCTION

As this book is being completed, it has been twenty years since the last edition of the standard book on Roman Bath was published.1 Several important excavations and smaller investigations that have taken place since then, along with some re-evaluation of older work, mean that we are justified in producing ‘yet another book on Roman Bath’. Stephen Clews, the manager of the Roman Baths Museum complex, has christened the growth in our understanding of the Roman town of Aquae Sulis ‘The Three-Hundred-Year Dig’, underlining the point that the discovery is a continuous process, never completed. In fact it is nearly 500 years since John Leland noted Roman antiquities built into the ancient walls of Bath, and since the mythical foundation stories of Bath were first challenged. I have outlined this process in the Afterword.

The rate of discovery has varied, as has the degree of interest, but has never completely flagged since the dramatic discovery of the gilded bronze head of Minerva in 1727 (Fig. 35). It is not a linear progression and for much of this period investigation was based on chance discoveries, and even more on the presence of someone capable and willing to recognise and record them. Where this happened, the degree of competence and understanding varied enormously. The clearance of the baths in the 1880s and ’90s under the City Architect, Major Davis, was extensive but not archaeological, although great pains were taken by his assistant, Richard Mann, to record the structures uncovered. Professional excavations only started in the 1920s on a limited scale, followed by targeted investigations in the Baths in the 1950s and ’60s. There has been some kind of investigative work in the Baths and Temple in every decade since. In the rest of the town opportunistic ‘rescue excavations’ started in the 1950s and continued into the ’70s. Professional provision has been available to record and even preserve archaeological remains since the early 1980s. The excavation report on this three-hundred-year dig must be revised every so often; ideas change and old models are questioned or revived. This book is the latest interim.

PRELUDE

In 1998 geological drilling into the sediments of the Hot Bath spring was monitored by archaeologists, who recovered a remarkable collection of 494 high-quality stone tools dating from the early Mesolithic period (Fig. 1). They were as much as 9,500 years old. The tools showed virtually no signs of wear and were an unusual selection of types in carefully selected raw material. The numbers from the narrow sample (a 100mm diameter borehole) implied that there were as many as 1,700 artefacts per cubic metre in the muds in the spring. Whether or not this was the result of one event or several, we are almost certainly looking at the earliest known act of veneration or propitiation at the hot springs in which supplicants deliberately threw offerings into the hot water at what was to become the World Heritage Site of Bath.

Fig. 1 Flints from the Hot Bath spring. (Roman Baths Museum)

This was a time when hunter-gatherers, people who had not yet adopted farming, were just beginning to colonise Britain, free of ice and still not yet an island. Spreading across southern England along wooded river valleys and tree-covered downland, small groups followed game, exploiting plant foods as they came into season, reaching further and further west; and, one year, they came to a bend in the river, wooded with alder and willow and rich in wildfowl, fish and other game. And they found the springs (Fig. 2).

Fig. 2 The hot springs as they may have looked before the Romans came (this is actually a spring in Armenia).

It must have been a shocking discovery. The nearest hot springs were over 200 miles away in Belgium; it is very unlikely that our wanderers had ever seen anything like it. If it was a cool day, the wooded valley bottom would have been sliced by a band of mist drifting through the foliage and spreading down to the river. As they drew closer, the slight metallic smell would be noticed and then the weird orange and emerald green colours of the algae that grows around the hot waters would become apparent. If they dared to come even closer they would see the three pools of hot water bubbling and steaming and streaming down to the river.

We have little idea of the spiritual beliefs of the Mesolithic hunter-gatherers, but it is hard to imagine that their reaction to the springs was not one of wonder, if not awe and fear. The deposition of a group of high-quality flint tools, of a few restricted types, made of carefully selected flint from as much as 20 miles away, might not, therefore, be an unexpected reaction, a gift to whatever power was responsible for this worrying break in normality.

Recent discoveries indicate that the river bend, despite being flood prone, was visited regularly after its first discovery in this period, and down to about 5,000 or 6,000 years ago, but not by more than wandering hunting bands. When farmers settled the surrounding hills and valleys, still no one stayed for long and it seems that the springs themselves were avoided or only rarely visited, perhaps under the protection of rituals and invocations. In any case, until the very end of the Iron Age, the centuries before the Roman Conquest in AD43, there was no activity around the springs that has left any trace to modern investigation. This matches the idea that Celtic holy sites tended to be natural places or enclosures: groves, caves or springs, liminal places, haunted by the gods, to be approached only by priests or those protected by them. The Roman poet Lucan, over-dramatically perhaps, describes such a place near Marseilles in the first century BC:

There was a grove from a bygone age, never ravaged, caging within its laced branches dusky gloom and icy shadows; high above, the banished sun. Here no rustic Pan holds sway, no powers of the forest – Silvani or Nymphs – but, barbarous in its rituals, a cult of Gods: altars heaped with hideous gifts, every tree around them splattered with human gore.

The God’s images, grim and crudely fashioned, started forth, rough-hewn, from felled trunks. The very earth, the pallor of heartwood long since rotted down to powder, left men thunderstruck. Divinities consecrated in common shapes can never cause fear like this – so much does it add to human terror not to know the Gods we fear!2

After these millennia of apparent inactivity, probably in the early first century AD, a causeway of gravel and mud kept in shape by hurdles was thrown out into the main spring pool (Fig. 3). This allowed worshippers to reach the heart of the spring and make offerings. These included seventeen coins of the local British tribe (the Dobunni) found in excavation in 1979 (Fig. 4). Another was found in the Hot Bath spring pipe in 1998. Society was changing in the later Iron Age, and this change in the way the springs were used may reflect aspects of this.

Fig. 3 The causeway into the sacred spring (based on Cunliffe and Davenport, 1985, Fig. 27).

Fig. 4 Celtic coins from the spring. (© Bath and North East Somerset Council, Roman Baths Museum)

A few rare sherds of pre-Roman pottery have been found in the vicinity of the springs, but no Iron Age settlement has been found closer than 0.7km away, and these, sensibly, on the lower slopes and terraces of the Avon Valley, not its floor. It was the Romans who had the presumption to build on the springs, to tame and enclose them and to found a town based on their exploitation.

1

THE ROMANS ARRIVE

THE SOURCE: THE ORIGINS

The water that bubbles up in the centre of Bath is old, very old. It was old when the first hunters stumbled across it and when the Dobunni tribesmen and women made offerings; it was old when the Romans channelled and confined it, and it is old today. It is, however, always the same age: constantly flowing, constantly recharging, constantly puzzling. Ever since the end of the last ice age, around 12,000 years ago and now, it appears, for much, much longer, rain falling on the absorbent carboniferous limestone of the Mendips (and now we believe from similar rocks to the north and west) has soaked, percolated, flowed and dwelled in the beds that run from there under the clays and mudstones of the hillsides and valley floors, heated by the thousands of feet of rock pressing down on it from above, and then, following the rising strata under Bath and finding a break in the sealing layers, bursting upwards to emerge in that low-lying river bend.

This continuous, extraordinary journey is now known to take at least 2,500 years, but the latest studies show that the process is still somewhat mysterious. While the source and characteristics of the water are no longer quite the mystery they once were, we are still looking for the particular reason why the three springs appear together here and nowhere else.

In the first place, the geology around Bath is extraordinarily complex (Fig. 5). Deep below our feet the scene is one of stratigraphic chaos. This is the result of ancient folding and fracturing of even more ancient sediments, resulting in deep structures that provide the environment for the capture and transport of the waters beneath, but which also make modelling the flow of water extremely difficult. The latest theory is that the springs are, in fact, many millions of years old, first bubbling out of a limestone pavement or karst in a knoll in the carboniferous limestones now buried at around 50m under central Bath (Fig. 6).3 The spring was then buried by around 250m of marine mudstones and limestones of the Triassic and Jurassic periods (250 to 65 million years ago) and the waters were confined to the deep geology. Much later, the cutting down of the Avon Valley brought the frozen ground of the last ice age close enough to the buried hot waters so that they could melt their way through the permafrost to establish permanent ‘pipes’ to the surface. In our current warm phase the pipes remain open due to the water pressure (Fig. 6).

Fig. 5 The deep geology of the Bath area and the spring catchment (redrawn from Gallois, 2006, Fig. 2).

Fig. 6 The long-buried springs break through at Bath in the late Pleistocene (redrawn from Gallois, 2006, Fig. 6).

The three springs still pump 1.44 million litres (over 300,000 British gallons) of water to the surface every twenty-four hours, all within a radius of little more than 100m. They vary slightly in temperature, the Hot Bath spring being a few degrees warmer than the others, but all are at a constant 44–46°C, or around 111–115°F. It is the temperature that makes Bath unique in the British Isles: they are the only truly hot springs in these islands.

They are mineral springs as well as thermal, and contain at least thirty-eight minerals. These include calcium, sodium, potassium, magnesium, hydrogen carbonate and chlorine. The others are mostly traces, with calcium and sodium together making up 35 per cent of the total.4 Iron is present and its oxides are the source of the bright orange colouration. The waters are also mildly radioactive and when this was discovered in 1908, the Radium cure was promoted. It was discontinued when the hazards of radiation became better appreciated. Nonetheless, the levels in the water are so low that it is doubtful any harm was ever done, even in concentration.

The rise of modern medicine led to a decline of confidence in the value of taking or bathing in the waters during the twentieth century, until the National Health Service stopped funding treatment in 1976. While bathing in warm water certainly has a therapeutic effect on various physical ailments, it was not thought that the spring water was any more efficacious than any other hot water. However, in the absence of modern treatments, we should not sneer at the symptomatic relief that immersion in the spring water undoubtedly gave to earlier visitors, or the benefits that a regime of eating less and drinking spa water rather than beer, wine and spirits, could bring.

House painters had particular reason to be grateful to spa water. Palsy, or paralysis, was a common complaint in this trade. In the eighteenth century it was not known that this was caused by lead poisoning (traditional white paint contains a high level of lead), and neither was the reason understood for the relatively good cure rate for painters who took the water. For adults, at least, drinking spa water while staying away from lead sources can help flush out the lead and the paralysis will be much reduced.

But this is based on what we know about more recent times, the eighteenth- and nineteenth-century heyday of the spa. What about the Romans?

They had little idea that the water contained nearly forty minerals, was mildly radioactive, or where it came from. That three springs rose in this constricted area was clearly the result of divine intervention.

The Romans brought their gods with them, but were never exclusive in their beliefs. Wherever they went they expected to meet local divinities, other beliefs. Arriving at the hot springs, with which they were well-acquainted on the European mainland, they would naturally expect to discover who was the deity in charge. As the Romans later recognised Sulis Minerva as the presiding deity of the springs, we can infer that that divinity was the goddess Sulis. The conflation of a Roman goddess with a local one was a normal Roman practice, and presumably means that Sulis was enough like Minerva that she could be seen as equivalent, or the local version. This was the interpretatio romana and such double deities are common all over the empire. When they founded a sanctuary here in her honour and a town grew up, it was naturally named Aquae Sulis, the waters of Sulis.

A SENSE OF PLACE

We have uncritically talked about ‘the Romans’ and later (Chapters 2 and 10) we will consider who we mean when we do: but, for now, we should ask, ‘Why were the Romans, what brought them here?’

In the broader political sense, it was the desire of the Emperor Claudius (AD41–54) to cement his recently acquired and shaky hold on the imperial power by demonstrating his military prowess in the traditional way. The Roman army invaded the island in the summer of AD43 and quickly conquered the dominant tribes in the south-east of the country. With a combination of military might and diplomacy, the southern half what is now England was brought under political control within a year or two at most, although we do not know the timetable of the first part of the conquest with any precision.

With the exception of the Durotriges of modern-day Dorset and south Somerset, the western tribes in England rapidly made peace or even allied themselves with the invaders. Britannia was not a country, merely an island divided into many tribal groups, some politically sophisticated, other perhaps less so.

The springs fell into the territory of the Dobunni, which spread across modern-day Gloucestershire and Worcestershire, parts of Oxfordshire and down into north Somerset (Fig. 7). It is thought that they saw the invaders as a useful ally against the tribes east and south. The latter fought hard against the Roman army; there is no evidence that the Dobunni did. Indeed, we hear that the Dobunni were among tribes who hastened to make peace with the Romans before their neighbours had been defeated.5

By AD48 the Romans had set up a province with its de facto western and northern boundaries along the rivers Severn and Trent.6 Inside this border a limes, that is, a military road, was laid out, linking the south coast to the Humber. Most of this road still exists, and we know it as the Fosse Way. If we look at the route of this road as it passes through Bath, and the others described in the following section, we can see how quickly and thoroughly the Romans had grasped the character of the topography around the springs and, specifically, the strategic significance of the site (Fig. 8).

Fig. 7 The Dobunni and their neighbours.

All these roads, vital to the Roman army’s control of the new province, had to find a route across or through the Avon Valley and the hills either side. The valley itself is a natural route from the east, the heartland of the new province, and onwards to the west, to the Severn and the line of control of those early years. After the initial conquest there was a continuing campaign against the Silures of south Wales. Securing the route of troops to a crossing of the Severn and their communications would have been paramount.7 The first phase of this campaign was concluded in AD51