The archaeology of medieval towns : case studies from Japan and Europe 9781789694260, 9781789694277, 1789694272

Citation preview

COMPARATIVE AND GLOBAL PERSPECTIVES ON JAPANESE ARCHAEOLOGY

The Archaeology of Medieval Towns: Case Studies from Japan and Europe

Edited by Simon Kaner, Brian Ayers, Richard Pearson and Oscar Wrenn

COMPARATIVE AND GLOBAL PERSPECTIVES ON JAPANESE ARCHAEOLOGY

The Archaeology of Medieval Towns: Case Studies from Japan and Europe Edited by Simon Kaner, Brian Ayers, Richard Pearson and Oscar Wrenn

Archaeopress Archaeology

Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com

ISBN 978-1-78969-426-0 ISBN 978-1-78969-427-7 (e-Pdf)

© Archaeopress 2020 © Simon Kaner, Brian Ayers, Richard Pearson and Oscar Wrenn, and the individual authors 2020 Published with support from the Sainsbury Institute for the Study of Japanese Arts and Cultures Layout designed by Colin Edwards Cover: Excavated medieval street frontage in Kyoto, adapted by Colin Edwards. Layout designed by Colin Edwards.

All rights reserved. No part of this book may be reproduced, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners. Printed in England by Severn, Gloucester This book is available direct from Archaeopress or from our website www.archaeopress.com

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 6

Published with assistance from

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 7

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 8 0

Contents Foreword and acknowledgements

1

Chapter 2 Japanese medieval trading towns: Sakai and Tosaminato Richard Pearson

13

Chapter 4 The establishment and transformation of Japan’s medieval capital, Kamakura Oka Yōichirō

42

Chapter 6 The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe Senda Yoshihiro

61

Chapter 8 Medieval urbanism and culture in the cities of the Baltic – with a comparison between Lübeck, Germany and Sakai, Japan Manfred Gläser

84

Chapter 10 Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective Joanita Vroom

115

Glossary

129

Index

135

Chapter 1 The archaeology of medieval towns in Japan and Europe: an introduction Brian Ayers and Simon Kaner

3

Chapter 3 Ichijōdani: the archaeology of a Japanese medieval castle town Ono Masatoshi

27

Chapter 5 The development of Hakata as a medieval port town Ōba Kōji

55

Chapter 7 Five medieval European towns: Bruges, Göttingen, Norwich, Ribe and Rouen - a pictorial introduction

72

Chapter 9 Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100 Henri Galinié

107

Chapter 11 Afterword Richard Pearson

125

List of contributors

133

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 9

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 10

i

An early European map of Japan from the Libro di Benedetto Bordone, a book describing all the islands of the world, published in Venice in 1528.

Image from the Cortazzi Collection, Lisa Sainsbury Library, Sainsbury Institute for the Study of Japanese Arts and Cultures.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 11 0

Periodisation of Japanese history c. 70,000 – 16,000 BC:

Palaeolithic

c. 500 BC – AD 300:

Yayoi

c. 16,000 – c. 500 BC: AD 300 – 710: 710 – 784:

794 – 1185:

1185 – 1600: 1603 – 1868: 1868 – present:

Jōmon Kofun

Early Historic (capital at Heijō, modern Nara) Heian (capital at Heian, modern Kyōto) Medieval

Early Modern (centre of government moved to Edo, modern Tōkyō, though imperial family and court remain in Kyōto) Modern (capital moved to Tōkyō)

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 12 0

iii

G

NORTH SEA ENG N H CHAN

EL

AT L A N T I C OCEAN

G

Quentowic G Rouen G Paris

G

RUSSIAN PRINCIPALITIES

G

LAND

ENGLIS

G

Ribe

Norwich G Great Tewkesbury G G Yarmouth Winchester G G Dorestad G London G Bruges

Novgorod

Lübeck G

G

Göttingen

POLAND

Berlin G

Tilleda

Cologne G

Rothenburg ob der Tauber

HOLY ROMAN EMPIRE HUNGARY BLACK SEA

KINGDOM OF FRANCE NAVARRE

BYZ

ANT

INE

ARAGON

EMP

G

IRE

&O

Constantinople

T TO

MAN

TUR

KS

MEDITERRANEAN SEA

G

Beijing Tosaminato

G

KO R EA

G

G

Chang’an

Hiralzumi

Y E L LO W SEA

G

Kusado Sengen-chō

G

G

Ichijōdani

G Kyōto Ōsaka G G Nara G Sakai

G Hakata Dejima

ūk ū Ry

TA IWA N

Isla nd s

EAST CHINA SEA

Maps of medieval Europe and Japan indicating key locations mentioned in the text

PA C I F I C OCEAN

G

Kamakura

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 13

iv

The archaeology of medieval towns: case studies from Japan and Europe

A comparative timeline including key episodes mentioned in the texts NNMM

NMMM

1008 Murasaki Shikibu completes Tale of Genji

1054 Schism in Christianity between Roman Catholic and Eastern Orthodox

1115 Rebellion of the Saxons (modern Germany)

1066 Battle of Hastings: William of Normandy becomes King of England

1138 Old Lübeck destroyed

1073 Tsurugaoka Hachimangū Shrine in Kamakura is founded 1088 Earliest European university, at Bologna

NOMM

1206 Ghenghis Khan establishes Mongol Empire

1143 First foundation of German Lübeck by the Earl of Schauenburg

Heian 794-1185

960-1279 Song dynasty

Chinese dynasties

1381 Translation of Bible into English

1392 Establishment of Joseon dynasty in Korea

1223 Establishment of Kamakura banyaku 1226 Lübeck becomes a free city

1172 Rudolf I of the Hapsburgs recognises Rothenburg as a free city

1185 Battle of Dan’noura: Minamoto defeat Taira

1192 Kamakura becomes official capital and Minamoto no Yoshitomo appointed Sei-I Taishōgun

1199 Death of Minamoto no Yoritomo

Periods in Japanese history

1346 Battle of Crécy

1215 Magna Carta signed by King John of England

1158 Foundation of Hanseatic League

1099 First Crusade and establishment of Kingdom of Jerusalem

NPMM

1399 Ōei Disturbance in Sakai in which 10,000 houses are destroyed

1274 and 1281 Attempted Mongol invasions of Japan 1293 Kamakura earthquake

c. 1298 Marco Polo publishes his Travels and mentions Zipangu (Japan), isles of gold 1299 Establishment of Ottoman Empire

1323 Sin’an shipwreck, near Mokpo, Korea

1333 Siege of Kamakura and restoration of imperial power

Kamakura 1185-1333

Muromachi 1333-1568 1279-1368 Yuan dynasty

1368-1644 Ming dynasty

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 14 0

Timeline

NQMM

1404-1549 Trade opened up between Japan and Ming Dynasty, China, including 19 tribute missions 1405 Chinese Admiral Zheng He begins exploration of Indian Ocean 1408 Destruction of Endsee Castle, Germany

NRMM

NSMM

1590 Reunification of Japan

1517 Reformation begins as Martin Luther issues his Ninety-Five Theses

1523 Ningbo Incident: Japanese trade with China ceases

1592-1598 Japanese invasions of Korea

1550s Direct trade established between Japan and Europe

1416-1477 Ōnin War

1597 Death of 26 Christian martyrs in Nagasaki 1598 Toyotomi Hideyoshi dies

1569 Sakai is taken by Oda Nobunaga

1453 Constantinople captured by Ottoman Turks

1573 Defeat of Asakura clan by Oda Nobunaga

1478 Taking of old Kiyosu Castle, Owari plain, by the Shiba clan

1483 Asakura clan become rulers of the Echizen region of Japan

1485 Battle of Bosworth: Henry Tudor becomes King Henry VII of England

1492 Christopher Columbus reaches the New World

NTMM

1581 Construction of Himeji Castle

1600 Battle of Sekigahara won by Tokugawa Ieyasu

1603 Tokugawa Ieyasu becomes Shogun and Edo de-facto capital

1609 Establishment of trading post by Dutch on Deshima Island, Nagasaki Bay

1582 Assassination of Oda Nobunaga

1588 Spanish Armada attempts to invade England

1613 The Clove, captained by John Saris, becomes the first English ship to land in Japan

1614-15 Siege of Ōsaka Castle and destruction of Toyotomi family

AzuchiEdo Momoyama 1600-1868 1568-1600 1644-1912 Qing dynasty

v

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 15

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 16 1

1

Foreword Simon Kaner This book arises from a long-held fascination with medieval archaeology, and in particular with the particularities of medieval urban forms in different cultural contexts.

The idea for a comparative investigation into medieval towns in Japan and Europe has its origins in the mid 1990s, when two specialists in Japanese medieval archaeology, Maekawa Kaname and Uno Takao, brought groups of students from Toyama University to undertake field research at Swavesey in Cambridgeshire. In the 1200s Swavesey had been a planned medieval town, an ‘inland’ harbour on the River Ouse, with important trading links out into the Wash via the great port of Kings Lynn, out into the North Sea and so to continental Europe. With the establishment of the Sainsbury Institute for the Study of Japanese Arts and Cultures in 1999, with its headquarters in the Cathedral Close in the historic city of Norwich, one of the most important medieval centres in England, new opportunities arose to explore medieval urban archaeology in Japan and England. The urban archaeology of Norwich was being investigated by Brian Ayers, first directing the Norfolk Archaeology Unit and then as County Archaeologist for Norfolk. The excavations of medieval Norwich in the 1970s and 1980s were the largest archaeological investigations of a medieval city in Europe.

Other initiatives were underway at the same time which encouraged us to give further consideration to the benefits of bringing differing traditions of medieval urban archaeology into contact with each other. Professor Richard Hodges had established his Institute for World Archaeology at the University of East Anglia in Norwich in the mid-1990s, with a particular focus on the great classical and medieval centre of Butrint, a World Heritage Site in southern Albania. At the same time, Maekawa Kaname was establishing his ABC Medieval Archaeology Project, designed to bring an internationally-informed multi-disciplinary approach to medieval archaeology in Japan. Excavations of medieval sites in Japan outnumber those of other periods, although for many Japanese archaeologists and historians, the medieval period was firmly regarded as the preserve of the historian. Maekawa and his group set out to change that perception.

Acknowledgements

The 2004 conference was attended by a number of specialists who were not able to contribute to this volume: William

Bowden, now Professor of Archaeology at the University of Nottingham; Jane Grenville of the University of York; Matthew Johnson, now Professor of Archaeology at Northwestern University; Maekawa Kaname, at the time Professor of Archaeology at Chuo University in Tōkyō; Martin Morris of Chiba University. We are grateful to the Japan Foundation who supported not only the conference, but also the subsequent research visit to Japan by Brian Ayers. Thanks are also due to Cathy Potter who worked so hard to ensure that the conference was a success, Nakamura Oki, at the time Handa Japanese Archaeology Fellow, and the staff of the Sainsbury Institute for the Study of Japanese Arts and Cultures. We are also grateful to all those who helped with the research visit to medieval sites in Japan in 2007. Subsequent to the conference, Richard Pearson selected the papers by Ono Masatoshi, Oka Yōichirō, and Ōba Kōji, contacted the authors, and drafted translations, as well as editing a rough translation of the paper by Senda Yoshihiro.

The book would not have been completed without the enthusiastic assistance of Morishita Masaaki who as Handa Fellow at the Sainsbury Institute assisted with the translation of Chapter 6 along with Kikuchi Atsuko. Lucy North provided invaluable advice on editing the papers. The book was designed by the ever-patient Colin Edwards. Thanks also to Mary Redfern for comments on some of the papers. The editors would particularly like to thank the Trustees and Management Board of the Sainsbury Institute for their support.

Chronologies: The medieval period in Japan and Europe

Archaeologists and historians divide up the past in a number of ways. The English terms ‘Medieval’ and ‘Middle Ages’ is comparable to the Japanese term chūsei. But of course in Europe there remains debate about exactly what periods are covered by medieval. In some parts of Europe the medieval begins with the end of the Roman Empire. In China it is perhaps most applicable to the period between the end of the Han dynasty (206 AD) and the start of the Song (960 AD). In Japan the term chūsei is usually used from the move of the effective capital to Kamakura from Heian, modern Kyōto, in 1185, until the reunification of Japan at the end of the sixteenth century and the establishment of the Tokugawa

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 17 1

2

The archaeology of medieval towns: case studies from Japan and Europe

shogunate with its new capital at Edo, modern Tōkyō. Emperors, even if for long period without any real power, continued to sit on the throne throughout this often turbulent period in Japanese history, and Emperor reign dates are often used as the basis for chronology in Japan.

A note of Japanese names and terms

Japanese names are given throughout in Japanese name order (family name followed by given name, e.g. Senda Yoshihiro).

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 18 2

The archaeology of medieval towns in Japan and Europe: an introduction

The archaeology of medieval towns in Japan and Europe: an introduction Brian Ayers and Simon Kaner

Encountering urban forms in Japan and Europe

This book arises from a belief that the trajectories to urbanisation at each end of the Eurasian landmass and the details of the resulting urban forms merit comparison, and that archaeology can provide an important perspective on the developments of towns in both Japan and Europe, complemented by history, art history and other aspects of urban studies.

The first encounters between Japan and Europe were between urbanised societies. In the sixteenth century, Jesuit missionaries, who were among the first Europeans to come into contact with Japan, described some of the towns and cities they encountered, including the great ‘free city’ of Sakai. Following the Portuguese occupation of Malacca in 1511, direct trade with Japan was established by the 1550s. The first English account of Japanese urban centres is given by John Saris, captain of the Clove, the first English East India Company ship to come to Japan in 1613, a decade after most of Japan was unified in the year that King James I of England and VI of Scotland brought those two countries together under the Act of Union. This first Tokugawa shogun, Ieyasu, created a new capital for Japan at Edo, modern day Tōkyō, bringing an end to the medieval centuries of political and military strife, and establishing a new social order which was to bring peace to Japan for two and a half centuries. Not many Englishmen were to follow in Saris’ footsteps, and contact with Europe was subsequently restricted to knowledge and commodities brought by annual visits by the Dutch East India Company through their factory on Dejima, an artificial island in Nagasaki Bay on the southern tip of the main southern island of the Japanese archipelago, Kyūshū. Saris’ diary records how: ‘We were rowed through, amongst divers Ilands … well inhabited and divers proper towns built upon them’. He said of Hakata: ‘The Towne seemed to be as great as London is within the walls, very wel built, and even, so as you may see from the one end of the streete to the other’. The early English visitors provoked great interest and excitement, the people ‘making such a noise around us, that we could scarcely heare on another speake, sometimes throwing stones at us (but that not in many Townes). … The best advice

that I can give those who hereafter shall arrive there, is that they passe on without regarding those idle rablements, and in so doing, they shall find their eares onely troubled with the noise’ (Cooper 1995 cited in Batten 2006, 134). Saris’ account recalls the vivid evocations of urban life portrayed by travellers to European towns from the somewhat sniffy assessment of 10th century Viking Hedeby by the Arab Al-Tartushi (Jones 1973: 175-7) or the more approving assessment of late 17th century Norwich by Celia Fiennes in 1698 (Morris 1984: 136-8).

Japanese medieval urban forms in context

These accounts confirm that Japanese society was already well urbanised. Indeed the earliest urban forms in the Japanese archipelago date to prehistoric times. Large early settlements, occupied over centuries and on occasion for millennia, include Sannai Maruyama on the northern tip of the main island of Honshū, inhabited by up to a thousand people between four and six thousand years ago (Habu 2008). Some of the first written accounts of Japan, provided by Chinese chroniclers, describe the urban forms at the centres of polities ruled by the first regional leaders in the archipelago. Archaeology has provided direct evidence for the existence of such centres from around two thousand years ago, most notably at Yoshinogari in Kyūshū (Hudson and Barnes 1991). The definition of such settlements as urban, however, is as contentious as calling the central places of Iron Age Europe towns (Cunliffe 2005). In the eighth century, the first Chinese-style planned city was built in Japan. Laid out based on the model of the Tang dynasty capital at Chang-an, this gridded city, including an imperial palace, Buddhist temples and extensive administrative and residential facilities, was constructed at Heijō, modern-day Nara (Tsuboi and Tanaka 1991). In 784 the capital was moved to Heian (present-day Kyōto). These capitals, home to the Imperial family and the new urban elite, housed the flowering of literate civilisation in the archipelago (Hempel 1983). Archaeological investigations in Nara, Kyōto and Ōsaka have uncovered material which bring these early capitals vividly to life, complemented by early historical records (Wheatley 1978). Intriguingly, however, the urban form did not spread beyond these capital areas, and it was not until the decline of centralised absolute power that urban life was taken up extensively elsewhere in the archipelago.

3

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 19 2

4

The archaeology of medieval towns: case studies from Japan and Europe

The first planned cities in Japan, the capitals of Heijō (modernday Nara), Nagaokakyō, and Heian (modern-day Kyōto) were laid out on a grid, employed the Chinese principles of city planning, in particular feng shui, and were modelled after the Tang dynasty capital of Chang’an. Chang’an itself has been the focus of extensive study (Xiong 2000; see also Heng 1999, which also deals with Suzhou, and Xu 2000 for further studies about Suzhou). However the approach adopted to most urban historic sites in East Asia has been historical, architectural and art historical. Xiong writes: ‘Because of its disciplinary limitations, however, archaeology is prone to overemphasising the stylistic, material and technical aspects of a civilisation. It is ill-equipped to deal with the social, intellectual and institutional transformations of an urban centre as complex as that of Sui-T’ang Chang’an’ (Xiong 2000, 2-3). As archaeologists we would take issue with this statement, and instead assert that the archaeological record not only provides invaluable evidence for what actually happened in cities, but also complements use of available historical and art historical sources. Indeed archaeology can provide an important counterpoint to idealised representations of medieval cities, such as the remarkable screen paintings of the city of Kyōto, the Rakuchū rakugai-zu, sumptuous painted screens apparently showing everyday life in the metropolis, but just as affected by the biases of the producer and consumer as any other historical document (see McKelway 2006). One of the earliest of these pairs of screens, created by Tosa Mitsunobu, the chief of the Imperial Painting Bureau, was commissioned by Asakura Sadakage (1473-1512), the daimyō of Echizen province, who ruled from his regional capital at Ichijōdani, the subject of Chapter IV in this volume.

Early archaeological investigation of Chinese dynastic capitals at Chang’an and Luoyang was undertaken by Japanese scholars in the first years of the 20th century (cf Xiong 2000, 2-4). Zhu Jianfei (2004) provides a useful summary of the history of early dynastic capitals in China, leading up the establishment of the capital at Beijing in the first part of the 15th century, whose urban structure was finalised in 1553. Although a large part of the city was dominated by Imperial and other administrative structures, the southern areas, beyond the main city wall, comprised a series of bustling commercial zones. On the Korean peninsula, with which Japan engaged through diplomatic and trade missions, piracy and military intervention during the period under discussion in this volume, urban centres had existed since the Chinese Han Empire established commanderies in the 1st century BC, notably at Lelang near present day Pyonynang (Pyeongyang), the capital of the Democratic People’s Republic of Korea (North Korea). The capital of the Joseon dynasty was established by King Taejo at Hanyang (Hanseong, modern-day Seoul, on the banks

of the Han River in west-central Korea) between 1394 and 1405, replacing the capital of the earlier Goryeo dynasty (9181392) at Kaesong, or Gaegyeong, whose remains now lie a few kilometres north of the demilitarised zone that separates North and South Korea, which was the focus of recent investigations by the Museum of London (Pratt 2006, 86). This work suggested that some 1800 of the original 200,000 traditional buildings in this city survived to the recent past. As well as palaces and administrative buildings, there were extended areas given over to markets, including for horses, pork, and shops for oil, cosmetics, paper and hemp (Kim 2004, 68, 100-101). The great majority of people in Korea from the 10th to the 17th centuries lived in village communities which included the larger tile-roofed residences of the yangban class of civil and military officials, and the humbler thatched roof dwellings of the farmers (see Kim 2004, 108-109), although there were also many fortresses, commercial and industrial centres, the latter flourishing especially as government control of handicrafts gave way to private initiatives from the early 16th century. This volume is concerned with medium and small-scale medieval urban forms in Japan, rather than with the major urban centre of Kyōto. Part of the reason for this is that Japanese urban archaeology has flourished in the smaller centres, and indeed abandoned and deserted medieval towns provide opportunities for wide area investigation rather than the more constrained keyhole procedures required in still thriving cities. It has proved easier to understand the layout and development of sites such as Ichijōdani or Kusado Sengen-chō, rather than Kamakura or Kyōto. This is in contrast to much European urban archaeology, where many of the major excavation campaigns since the 1970s have been in the greater cities and towns such as London, York, Norwich, Tours and Lübeck, but as John Schofield emphasises at the end of his recent survey of the archaeology of medieval London, in the case of many large cities, it is often still too early to draw firm conclusions (Schofield 2011: 262).

Within Japan, the archaeology of medieval Nara and Kyōto, which boast the highest density of surviving medieval structures in the country, clearly deserve closer attention than the present volume can offer. Kyōto at least has already generated a western-language literature of its own, starting with the pioneering gazetteers of Ponsonby-Fane, first published in 1908 (Ponsonby-Fane 1956) and culminating in the monumental historical atlas overseen by Nicolas Fiévé (2008, Stavros 2016). The cityscapes of Kyōto are equally represented in the burgeoning literature on ancient cities as cultural heritage (Stavros 2016 and see papers in Fiévé and Waley 2003).

Medieval towns and cities in both England and Japan attract great attention as tourist destinations and as favoured locations to inhabit. In England many medieval towns and cities lost

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 20 2

The archaeology of medieval towns in Japan and Europe: an introduction

much of their historic cores through unsympathetic redevelopment in the postwar years. In Japan, the majority of historic urban centres were destroyed in the systematic programme of bombing at the end of the Second World War. And yet in both countries, there is now a common interest in enhancing rather than obliterating what has become known as the historic urban environment. In England, it is recognised that attention to the historic environment can play a very important role in fostering a sense of place and local identity. And similar sentiments have given rise to ‘town making’ (machizukuri) in contemporary Japan.

The papers in this volume are offered in response to a perceived need to bring Japanese medieval archaeology to a wider audience, to help realise the potential for the Japanese archaeological record to contribute to broader discussions in medieval archaeology. The approach adopted is comparative in nature. In this it complements the approach taken in McClain et al (1994) for the post-medieval cities of Paris and Edo, and an exhibition of post-medieval urban archaeology (Edo-Tokyo Museum 1996).

Comparisons of scale

William Wayne Farris defines medieval Japanese towns as centres with five thousand or more inhabitants (Farris 2006). During the 13th century, in the decades following the abandonment of Kyōto in 1185 which marks the start of the medieval era in Japan, and the establishment of Kamakura as the political capital of Japan, it is estimated that between 200,000 and 300,000 people, or some five percent of the total population of the archipelago, were living in urban environments which included cities such as Kyōto and Kamakura, ports like Tosaminato and Kusado Sengen, and former provincial capitals (Farris 2006: 86). By the middle of the 15th century there were over fifty identified urban centres, the largest still being Kyōto, with an estimated population of 200,000. There was a small number of towns with populations between 15,000 and 40,000, including Hakata, Ōtsu on Lake Biwa, and Kashiwazaki on the Japan Sea coast, and a series of smaller centres including Zuisenji in Etchū province and Ishidera in Ōmi. By the late 16th century the number of towns had increased sixfold. The proportion of the population living in towns remained the same as in the 13th century, at around five percent, but this number now represented between half a million to 700,000 people. Urban centres now included local political centres, towns focused on religious establishments, post stations, ports and market towns (Farris 2006, 152-3). From around 1520, improvements in food supplies, supported by the expansion of irrigation works, enhanced commercial activities leading to the establishment of new ports and markets, and shifting political contexts which saw local daimyō founding new

administrative centres around their castles, were factors that led to the intensification of urbanisation in the earlier 16th century (Farris 2006: 246). Hans Andersson, in a recent review of the development of medieval towns in Europe, notes that ‘there are important qualitative and quantitative changes beginning in the later part of the 12th century and peaking in the 13th century, which make this period especially important in the history of urbanisation’ (Andersson 2011). Towns got bigger in area and their populations increased, as many new towns were established across Europe. Intensification in trade, commercial activity and developments in the legal, administrative, and religious underpinnings of the organisation of society are all cited as reasons. The emergence of urban hierarchies with, in the case of England, London at the top and a series of provincial centres including Norwich and Winchester, is perhaps comparable to the situation in Japan. However, the scale of urbanisation does differ. For example, there are estimated to have been some 4000 towns by 1320 in Germany alone, although only fifty are thought to have had populations exceeding the 5000 required by Farris to define a town in medieval Japan. The largest, Cologne, had perhaps 50,000 inhabitants, a quarter of the population of Kyōto at the same time (Andersson 2011).

Comparisons of process

The trend towards urbanisation was not, however, a steady progression. Plague and famine took their toll, and the 14th and 15th centuries in Europe saw shifts in the urban network as many fewer, if any, new towns were founded. Major centres such as London and Copenhagen became increasingly important and distinguished from their provincial counterparts. The remains of failed towns can be found across Europe and Japan. The pattern of urbanisation established during the medieval period, however, was to shape the landscape in both Japan and Europe until the advent of industrialisation.

The process whereby Japan became an urbanised society is of interest in its own right but also bears comparison with the history and archaeology of urbanisation elsewhere. Japan has one of the richest archaeological records of anywhere in the world. Each year many thousands of archaeological sites across the archipelago are investigated and the findings meticulously recorded, albeit in Japanese. Over half of these investigations produce material dated to Japan’s medieval era, comprising a dataset unrivalled in most other countries. And yet this material remains largely unknown by medieval scholars elsewhere. Within Japanese scholarship, archaeology is only recently being incorporated into accounts of the medieval period, largely regarded as the preserve of the historian and art historian

5

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 21 2

6

The archaeology of medieval towns: case studies from Japan and Europe

(Mass 2002, Hall and Mass 1974). This book arises from an attempt to redress this situation, and it is hoped that the case studies presented will stimulate further research.

The study of urban development in Japan is beginning to change. The foreword to a new translation of key works by the pre-eminent social historian of medieval Japan, Amino Yoshiyuki, makes explicit reference to the archaeology of medieval towns (Amino 2012, xxxvii-xl), referring to the formal burial grounds and market places which he considers to be urban in character. In an excellent introduction to Amino’s book, translator Alan Christy sets Amino’s concern with such ‘urban spaces’ in terms of the prevailing emphasis by Japanese historians on the overwhelmingly agricultural and rural nature of premodern Japan, and how effectively Amino counteracts this with his view of peasants ‘also engaging in maritime commerce, proto-industrial production, or financial activities’ (2012, xvii). Amino deconstructs the myth of the self-sufficient rural community by demonstrating how important ‘networks, routes, circulations – and exchange - of goods, peoples and cultures’ have been since the prehistoric period’ (2012, xviii). Christy continues: ‘For Amino, the terms rural and urban have little to do with scale and everything to do with the character of everyday life. An urban settlement is inescapably part of a circulation network. It is a place where exchange is a fundamental activity, where production is premised on consumption elsewhere, and where equivalences between things are determined. Amino characterises as urban town after town that most Japanese would think of as hopelessly isolated and miniscule. He highlights the vast networks and constant mobility that he believes animated the Japanese past. With the proliferation of urban nodes covering the islands, even the images of rural communities where agriculture was dominant are unstable, for the ‘city” is no longer far away’ (ibid: xviii-xix). This new concordance between history, archaeology and art history is already beginning to manifest itself in new generations of interdisciplinary research projects and publications (Keirstead 1992, Maekawa 1991, 2004, Pitelka in preparation, Stavros 2016).

The archaeology of medieval towns: case studies from Japan and Europe

This volume forms part of a project whose origins lie in a conference held in Norwich in 2004. The conference had two prime purposes: to promote wider awareness and understanding of the dramatic strides that have been made in Japanese urban archaeology in recent decades; and bring Japanese and European specialists together to explore similarities and dissimilarities in urbanism as a force for change in medieval world cultures. The conference was followed by a

research visit to Japan, taking in many of the medieval urban centres discussed in the present book. The scale and scope of medieval urban archaeology in Japan and Europe places a comprehensive comparative survey beyond the bounds of this volume. Instead it seeks to leave a permanent marker of the unprecedented interaction between Japanese and European medieval urban archaeologists represented by the Norwich conference, intended to provide an inspiration to future collaboration and a starting point to insert an awareness of the Japanese record into the increasingly globalised debate about medieval towns.

We therefore selected four previously published papers which we felt were good exemplars of the Japanese approach, translated and adapted them, and set them alongside three papers by European specialists. The case studies in this volume were chosen to illustrate not only parallels and significant differences in methodological approach, but also parallels in results and the challenges that currently face urban archaeology as a discipline both in Japan and elsewhere.

Where possible at the 2004 conference, papers were paired and contributors were encouraged to seek comparanda, a difficult task as European speakers were often completely ignorant of work in Japan, and Japanese scholars similarly impeded with regard to European activities and concerns. Manfred Gläser, in seeking to compare the towns of Lübeck and Sakai, provided perhaps the bravest such attempt, identifying some eleven areas of comparison ranging from observations on each town’s role as a major trading centre to the different markets experienced by local craftsmen. His approach, however, together with a number of striking observations that occurred during the conference itself (such as the development of distinctive urban housing styles—see Grenville and Morris, 1997 and 2006 respectively), drives home the point that, while urban products will clearly vary in form and appearance from eastern Asia to western Europe, the social, economic and political considerations that drove urban formation, development and occasional decline are universal phenomena. Such phenomena, perhaps curiously (or perhaps not so curiously; there may have been hidden mutual influences and contacts), flourished within the same millennium on opposite sides of the globe.

It is the process of urbanisation as a socio-economic development, common to both Japan and Europe, which is clearly fruitful to explore. In doing so, however, it is helpful to reflect upon the nature of the archaeological resource, the methodologies by which that resource is investigated, the products of such investigation and the theoretical and analytical frameworks within which discoveries are assessed, synthesised and presented.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 22 2

The archaeology of medieval towns in Japan and Europe: an introduction

Definitions, methodologies, resources

The nature of the archaeological resource is not one that is necessarily easy to determine. The concept of what constituted a ‘town’ in medieval Europe, not only in the early medieval period but also during the ‘High Middle Ages’, is one that has taxed scholars for generations. Approaches to the issue have developed from a straight historical, legalistic definition (often revolving around such issues as whether a place had a right to a market), through adoption of a ‘criteria’-based approach (the Kriterienbündel propounded by Ennen in 1953, whereby a place can be considered urban if it contains two or more attributes such as defences, a planned street system, a mint or a diversified economic base) to a recent assessment that views urban settlements as expressions in part of the ‘cultural zone’ within which they were located (Hermann 1991). This last divides Europe into an ‘Inner Zone’, essentially countries bordering the Mediterranean but including France as far north as Normandy, a ‘Middle Zone’ that encompassed central Europe from the Black Sea to the North Sea (and included England), and an ‘Outer Zone’ of northern Europe including all of Scandinavia and European Russia. Hermann’s definitions specifically derive from the relationship of each area to the Roman Empire and the legacy of that empire. Thus the Middle Zone includes countries such as England that were within the empire but where Romanisation was effectively eradicated after the beginning of the 5th century, while most of France remained within the Inner Zone because of the enduring impact of Roman civilisation, in particular that of towns. In this context, it is particularly useful to have Henri Galinié’s paper in this volume as he explores the origins of urbanism within France north of the Loire (thus encompassing the border territory of Hermann’s Inner and Middle Zones) in an assessment of the legacy of different forms of Roman town and their impact upon the subsequent medieval urban framework of France.

The problem of definition is one that has also been taken up in Japan. Maekawa has explored the issue of urban formation, postulating that villages coalesced and became multifunctional regional centres in a process that he identifies as sonshuka, or nucleation (Maekawa 1998). Interestingly, his work extends beyond the study of urban centres alone and shows that the nucleation he observes does not necessarily lead to urban development; it can just as easily lead to the creation of residential forts (jōkan) or to large villages, which survive in a rather fossilised form to the present day. While such an observation has obvious resonances within a European context (how ‘urban’ were some of the Alfredian burhs of late 9thcentury Wessex, for example? Could some of them have been essentially Late Saxon jōkan?), Maekawa’s work also illustrates that any consideration of medieval urban centres must expect

variety and, occasionally, ‘urban’ centres may be found that do not necessarily appear to be urban.

The diversity of towns is explored in Galinié’s paper, where the approach is questioning and challenging, and he demonstrates that just as medieval towns often had different functions, so too did Roman ones, these differences also having an impact upon subsequent development. It is an approach grounded in archaeological observation, not necessarily that of excavation, but one that seeks to explore the relationship of a place to its hinterland, and to its socio-economic and political links, and to examine the importance of that place accordingly. One strength of the Norwich conference was that it made clear that such analysis is just as appropriate for other contexts. Richard Pearson, exploring the trading towns of Sakai and Tosaminato in Japan, describes two successful settlements, both with trade networks but operating within completely different spheres and therefore developing as different urban entities (this volume, Chapter Two).

Pearson’s assessment is based upon analysis of the material culture of each town, utilising material derived from excavations, notably ceramics. Such examination of recovered material is, of course, a traditional archaeological methodology; but Pearson adopts a broad approach, seeking to understand the significance of the ceramic assemblages within both their social and economic contexts. As with Galinié, these contexts are also political—the material can be seen as assisting understanding of the role of each town within its contemporary society and the needs and aspirations of both the ruling classes and those of the more affluent merchants. The influence of China, in particular, permeates Pearson’s paper—clearly, the impact of such a major advanced economy as a near neighbour to Japan was considerable.

The importance of the larger political context is clear within the northwest European case and brings a further parallel in urban studies. Not actually physically part of the European continent, England was a slow starter in post-Roman urban development: throughout the medieval period, its urban growth was clearly limited in comparison to that experienced elsewhere. European influence, however, was constant, usually through the medium of trade and, just as exploration of Japanese urban growth can be examined by archaeological study of tradable goods, so too have methodological studies in Europe grown over recent decades to encompass not only increased understanding of urban linkages but also the dynamics of urban economies. This extends beyond demonstrating trade contacts; recent work by scholars such as Astill has illustrated how the use of archaeological material can enable both the posing and the addressing of questions of longterm structural change within societies, questions beyond the reach of traditional historical sources (Astill 2000).

7

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 23 2

8

The archaeology of medieval towns: case studies from Japan and Europe

One further development in urban methodologies has been the recognition that towns never exist in isolation from their hinterland. Indeed, the concern with the relationship of a town to its hinterland - and the definition of that hinterland, which in the case of a Lübeck or a Sakai was certainly considerably greater than its immediate vicinity - has led to a growth in studies that seek to explore the urban/rural relationship. It is a theme alluded to in Pearson’s paper with his reference to Maekawa’s work, which explores the processes by which villages coalesced and, in certain cases, developed as urban centres. In Europe work is currently developing along a different trajectory; rather than examining the transition from rural to urban, it is exploring the interdependence of rural and urban economies and societies. Again it is the adoption of archaeological methodologies and the study of archaeological material that is assisting such study, as evidenced in the important British publication on town and country in the Middle Ages with the revealing subtitle Contrasts, Contacts and Interconnections, 1100–1500 (Giles & Dyer 2005).

Thematic connections: landholdings, commerce and guilds, religion and defence

One of the attractions for participants at the Norwich conference was the opportunity to observe medieval urban culture as uncovered at a range of sites in both Japan and Europe. This culture was illustrated most obviously in artefacts, such as the extraordinarily rich assemblages of finds from Lübeck, the absurdly large coin hoards from sites such as that of the early 14th-century Sin’an shipwreck located near Mokpo, Korea, and the 450,000 Chinese coins dated to 1368 found at Shinori near Hakodate in Hokkaidō, or the wealth of ceramics uncovered from a range of sites in Sakai. However, while it is possible for artefacts to move from rural to urban centres and back again, urban living dictates that certain physical attributes develop in towns that are particular to densely populated societies. Perhaps the most striking manner in which this is demonstrated, and which came up for discussion at the seminar, can be found in the Japanese land divisions known as tansaku.

Maekawa considers tansaku to be key indicators of urban life. Tansaku refer to strips, or belts, of land, usually quite long and narrow, such as those located at site SKT 200 in Sakai, often occupied by buildings along the street frontage and with patches of cultivable land and subsidiary structures to the rear. The parallel with the standard European urban morphology of long tenement plots set at right-angles to street frontages is striking and must reflect a common solution to a shared urban problem, that of maximising space as equitably as possible within the constraints of a market economy. The

morphological parallels go further: as valuable economic units, towns attracted the attention of local lords, who often built urban castles or similar structures in them; specialised zones for industrial manufacture or trade have been located in medieval Japanese towns and are also common in Europe; and religious establishments of varying size and importance were located in both Japanese and European towns.

Archaeological work on the impact of this last urban category—the urban religious establishment—has revealed other interesting comparanda. It has long been recognised in Europe that ecclesiastical institutions were not only major stimulants to the urban economy, fostering building trades, carpenters, sculptors and painters, and specialist suppliers such as those of painted glass, but also frequently engaged in mercantile trade, either at a local level through rents and management of rural estates or more widely through the export of raw materials, utilising the mercantile functions of towns. As Pearson observes, exactly the same sort of interaction between religious establishments and towns existed in medieval Japan; by the 12th century, while elite merchants were conducting trade through Japanese ports, as their counterparts were beginning to do in Europe, trade was also being undertaken between China and Japan with Song ships dealing with representatives of temples and shrines and their associated estates.

Urban life provided the framework for the development of mechanisms and institutions that enabled such trade to foster. Merchants’ guilds in England and continental Europe found expression in buildings which often still stand (such as the Merchant Adventurers’ Hall in York) or whose archaeological traces remain to be discovered (as seems to have been the case at Alfstraße 38 in Lübeck; see Gläser, this volume. Co-operative guilds, known as za, also existed in medieval Japan where medieval warehousing has been uncovered by excavation. The development of this and other urban building types is attested to not only by Morris and Grenville but also in the reports of work at sites such as Ichijōdani, where comprehensive excavation has revealed a range of urban structures representative of different classes within town society; see Ono, this volume as well as industrial buildings such as dyers’ houses and a blacksmith’s house. Trade clearly had a physical impact upon the form of towns, beyond that of urban buildings. The mercantile town of Tosaminato on the tip of northern Honshū was constructed on a long sand bar next to a lagoon with a narrow opening to the Sea of Japan. The site was hardly propitious for settlement, other than that it was a prime location to exploit trade to and from Hokkaidō and across the Sea of Japan. However the physical circumstances are paralleled to a remarkable degree by the contemporary town of Great

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 24 2

The archaeology of medieval towns in Japan and Europe: an introduction

Yarmouth in eastern England, also constructed on a sand spit with Breydon Water to its rear and the sea to the east, and therefore excellent regional and international links. The topography ensured development of long, linear settlements in both cases, with merchant quarters and, interestingly, major religious institutions at one end of the town - a shrine at Tosaminato and a large church at Great Yarmouth. Such comparative analogies should not be pushed too far but it is clear that topographic opportunities were exploited where possible for commercial advantage (as can be seen in the trading towns of Kusado Sengen and Bruges).

Urban castles were perhaps more common in Europe but they existed in Japanese towns as well (see Senda, this volume, and Turnbull 2009). During the research visit in 2008, we encountered an excellent example at Fukuyama, western Honshū, where investigations have not only uncovered the outworks of the castle there but also a dock protected from the Inland Sea. There is considerable scope for comparative work on the role of castles in towns. Of the three European castles illustrated here (Ribe, Rouen and Norwich), the recent extensive excavations at Norwich indicate the potential for such study both for increasing understanding of the role of the castle and of the relationship of the castle to the town (Popescu 2009).

Urban stone or brick defences, as seen in images from Bruges, Göttingen, Norwich and Ribe are not a common feature of Japanese towns but even there towns could be protected by earthworks, as at Ichijōdani or Tosaminato, and as seems to have been the case at Kamakura. Kamakura, like Kyōto, is an example of a town that was both an urban entity in its own right and a setting for aristocratic and royal display, as well as a focus of religious activity. The exceptional urban macroplanning evident in both Kyōto and Kamakura was practically unknown in contemporary Europe although works on a more minor scale can be seen at places such as late 9th-century Winchester (Biddle 1975) or the 13th-century bastide towns of southern France and north Wales (Coste & de Roux 2007). It could be argued that the 8th-century medieval Islamic city of Córdoba was an urban community planned with the symbolic care evidenced at Kyōto, indicating a further direction for comparative study (Murillo et al. 1997: 47ff). As hinted at the outset, this volume was not conceived as a comprehensive comparative study of medieval urbanism in Japan and Europe, but rather as a sampler, hopefully stimulating further investigations. It is worth mentioning a few areas of high potential. Further exploration of medieval perceptions of the role of religion as an important element of the urban experience in Japan and Europe would be rewarding. Religious sites acted both as a focus in urban layouts, as with the Tsurugaoka Hachimangū shrine at Kamukura, which is accessed by a

processional way (Wakamiya Ōji), and also provided ‘sacred landscapes’ with meaning for clergy and citizens alike, as recently argued regarding the cathedral precinct at Norwich (Gilchrist 2005).

Urban planning is another important area: consideration of the degree of topographical manipulation that was sometimes involved in successful town-building can be instructive. At Kamakura, excavation has demonstrated that the Nameri-gawa River east of the city was deliberately diverted in order to enable effective urban planning, while the valley site at Ichijōdani was sealed to the north and south by defensive ramparts. Ōba Kōji’s paper shows how at Hakata two separate settlements on adjacent sand spits were joined together by infilling, to make a continuous surface and Gläser records the deliberate infilling of marshes at Lübeck, work which, in the 13th century, increased the urban area by almost fifty percent.

Although it is often difficult to compare the products of two urban civilisations so separated by distance and lack of contact, attempts can still be made. Morris (in a ‘deliberately provocative’ manner) suggested that the celebrated late 15thcentury row houses of Tewkesbury in England have ‘significant points of congruence’ with the façades and layout of traditional houses in Kyōto (kyō-machiya) (Morris 2006, 21). Notwithstanding such potentially ‘more mobile and transferable’ urban design issues, it was also possible to observe remarkable similarities in both the processes of urbanisation and in the methodological approaches to the study of medieval towns. There is clearly great potential in further exploring the theoretical framework for such studies, but consideration of the analyses presented here demonstrates that medieval urban studies are in robust and challenging form in both Japan and Europe. A further topic of great significance for understanding what was happening in Japan is the economic, social and political impact of China, particularly with regard to Chinese urbanisation, which must have had an influence upon developments in the Japanese archipelago (cf. Pearson et al. 2001). The European papers here are solidly north European. Future studies could usefully reference the Mediterranean world, building on the enthusiasm of Vroom, which usefully highlights methodological ways forward for the study of Byzantine ceramics, drawing upon the influence of Japanese studies, and eastern Europe. Given known medieval trade contacts between Europe and central (and possibly east) Asia, there is scope for exploring mutual influences, including consideration of medieval Islamic influence, the towns of the Caliphate being one of the greatest expressions of Islamic civilisation. There is clearly also considerable scope for the sharing of

9

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 25 2

10

The archaeology of medieval towns: case studies from Japan and Europe

information with regard to methodological approaches and concepts of urban mentalities. An example is the recent development of spatial analysis in European archaeological theory in an attempt to understand the meaning behind certain actions and events as illustrated by physical structures. Johnson has observed that relationships within society can be explored through detailed analysis of buildings and their environmental context (Johnson 1993), an approach that Giles has adopted in an urban context to argue that the archaeological record can be seen as one which encodes ‘past cultural and ideological meanings’ (and can therefore be ‘decoded’ by archaeologists), and to examine the meaning that the structures and spaces of medieval urban guildhalls had for those who built and used them (Giles 2000). Ono Masatoshi is one Japanese scholar who has started to explore such issues. His paper in this volume contains interesting observations on the links between the ritual use of space in and around buildings. The Japanese urban environment was one certainly constructed with as much ‘meaning’ in the landscape as any European town, and the development of such methodological approaches bodes well for future explorations on how towns were used and viewed by their inhabitants.

It is perhaps also appropriate to reflect on the importance of archaeological work in medieval towns. As Schofield and Vince have observed of towns, ‘most of us live in them’ (Schofield and Vince 2003, 266) and we therefore observe the pressures that are inflicted daily on urban environments. These environments are the products of change in the past and it would be wrong to resist change in the future. However, sustainable management of that change relies upon an understanding of the environment that can only be obtained from detailed study, assessment, analysis and synthesis of information. Urban archaeologists therefore have a dual responsibility: to past societies, in order that their contribution to the present is represented faithfully and accurately; and to future societies, in order that urban landscapes and structures develop in harmony with the past, retaining a sense and spirit of place whilst adapting to modern needs and pressures. Observing the work of colleagues across the world assists both responsibilities. Towns, and the urban dynamic, were famously characterised by Braudel as being like electric transformers that ‘increase tension, accelerate the rhythm of exchange and constantly recharge human life’ (Braudel 1985, 479). The study of towns through urban archaeology is one that is itself subject to constant change, as new concepts, methodologies and understandings develop. That process can only be enhanced by ensuring that such development is undertaken within a

wide view of the world, recognising that common human needs, for food, shelter, warmth and spiritual sustenance, and common human aspirations, for social and economic enhancement, will lead to common solutions, often manifest in urban structures and systems. We hope that the contributions to this volume demonstrate that study of the medieval towns of Japan is not irrelevant to a greater understanding of the European town; nor is the consideration of European urban phenomena alien to Japanese academic concerns. They may also illustrate something greater: the universality of the human experience.

As this volume was in preparation, we have witnessed a ‘global turn’ in medieval history and archaeology (Belich et al. 2016, Franklin 2018, Holmes and Standen 2018, Jervis 2017, Sindbaek and Nixon 2012-13). A brief review of the available literature, however, reveals that despite the proliferation of courses of study with ‘global medieval’ in the title, for example at the Universities of Edinburgh and York1, the formation of large-scale research networks2, and the birth of a new journal about ‘medieval worlds’ (Frankopan 2019), we are still a long way from a widespread awareness of the potential of Japanese medieval archaeology (although see Souyri 2016) for this global turn. With enhanced appreciation of the significance of the medieval in the modern world, we hope that this volume will inspire further attempts to integrate the Japanese record into a ‘critical and global engaged medieval archaeology’ (Franklin 2018).

See for example, the MSc in Art in the Global Middle Ages at the University of Edinburgh https://www.ed.ac.uk/historyclassics-archaeology/centre-medieval-renaissance/msc-artglobal-middle-ages (accessed 2 June 2020) and The Global Middle Ages: the Medieval period in a transcultural perspective MA module at the University of York https://www.york.ac.uk/medievalstudies/postgraduate-study/modules/global-middle-ages/ (accessed 2 June 2020).

1

For example the UK Arts and Humanities Research Council sponsored research network on Defining the Global Middle Ages, based at the University of Oxford https://globalmiddleages.history.ox.ac.uk (accessed 2 June 2020) and the Global Middle Ages Project which started at the University of Texas http://globalmiddleages.org/about (accessed 2 June 2020). 2

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 26 2

The archaeology of medieval towns in Japan and Europe: an introduction

References

Amino, Y. 2012. Rethinking Japanese history. Ann Arbor, Center for Japanese Studies, University of Michigan. Andersson, H. 2011. The development of medieval towns. In: Carver, M. and Klapste, J. (eds.) The archaeology of medieval Europe. Volume II. 12th to 16th centuries. Aarhus, University Press: 370-378. Astill, G. 2000. Archaeology and the late-medieval urban decline. In: Slater, T. R. (ed.) Towns in decline AD 100 – 1600. Aldershot, Ashgate: 214 – 234. Batten, B. L. 2006. Gateway to Japan: Hakata in war and peace 500-1300. Honolulu, University of Hawaii Press.

Belich, J., Darwin, J., Frenz, M. and Wickham, C. (eds.) 2016. The prospect of global history. Oxford, University Press.

Biddle, M. 1975. Planned towns before 1066. In: Barley, M. W. (ed.) The plans and topography of medieval towns in England & Wales. Council for British Archaeology, Research Report 14. Braudel, F. 1985. The structures of everyday life: civilisation and capitalism 15th – 18th century. Fontana Press, London.

Cooper, M. (ed.) 1995. They came to Japan: an anthology of European reports on Japan 1543-1640. Michigan Classics in Japanese Studies No 15. Ann Arbor, Center for Japanese Studies, University of Michigan.

Coste, M. and de Roux, A. 2007. Bastides: Villes neuves médiévales (Walled towns: medieval new towns). Villefranche-deRouergue, Rempart.

Cunliffe, B. 2005. Iron Age communities in Britain. London, Routledge.

Giles, K. 2000. An archaeology of social identity: guildhalls in York, c. 1350 – 1630. Oxford, British Archaeological Reports British Series 315.

Giles, K. and Dyer, C. (eds.) 2005. Town and country in the Middle Ages: contrasts, contacts and interconnections, 1100 – 1500. Society for Medieval Archaeology Monograph 22. Grenville, J. 1997. Medieval housing. Leicester, Leicester University Press.

Hall, J. W. and Mass, J. P. (eds.) 1974. Medieval Japan: essays in institutional history. New Haven and London, Yale University Press.

Habu, J. 2008. Growth and decline in complex hunter-gatherer societies: a case study from the Jomon period Sannai Maruyama site, Japan. Antiquity 317: 517-584.

Hempel, R. 1983. The Heian civilization of Japan. London, Phaidon.

Heng, C. K. 1999. Cities of aristocrats and bureaucrats: the development of medieval Chinese cityscapes. Singapore, Singapore University Press.

Hermann, J. 1991. Siedlungsgeschichliche Grundlagen und geschichtliche Voraussetzungen für die Entwicklung Berlins (Basics about the history of settlement and the historical preliminaries for the development of Berlin). In: Brachmann, H. and Hermann, J. (eds.) Frühgeschichte der europäischen Stadt. Berlin, Akademie-Verlag: 7-18.

Holmes, S. and Standen, N. (eds.) The global middle ages. Past & Present 238. Supplement 13.

Edo-Tokyo Museum 1996. Unearthed cities: Edo, Nagasaki, Amsterdam, London, New York. Tokyo, Tokyo Metropolitan Edo-Tokyo Museum.

Hudson, M. and Barnes, G. L. 1991. Yoshinogari: a Yayoi settlement in northern Kyushu. Monumental Nipponica 46(2): 211-235.

Farris, W. W. 2006. Japan’s medieval population: famine, fertility and warfare in a transformative age. Honolulu, University of Hawaii Press.

Johnson, M. 1993. Housing culture: traditional architecture in an English landscape. Washington DC, Smithsonian Institution.

Ennen, E. 1953. Frühgeschichte der europäischen Stadt (The early history of the European town). Bonn, Ludwig Röhrscheid.

Fiévé, N. (ed.) 2008. Atlas historique de Kyoto. Paris, UNESCO and Les Editions de l’Amateur.

Fiévé, N. and Waley, P. (eds.) 2003. Japanese capitals in historical perspective. London, Routledge.

Franklin, K. 2018. Living with medieval things: why we need a global medieval archaeology. Global Journal of Archaeology and Anthropology 3.5. DOI: 10.19080/GJAA.2018.03.555625 (Accessed 2 June 2020). Frankopan, P. 2019. Why we need to think about the global middle ages. Journal of Medieval Worlds 1.1: 5-10.

Gerrard, C. 2003. Medieval archaeology: understanding traditions and contemporary approaches. London, Routledge. Gilchrist, R. 2005. Norwich Cathedral Close: The evolution of the English cathedral landscape. Woodbridge, Boydell and Brewer.

Jervis, B. 2017. Assembling the archaeology of the global middle ages. World Archaeology 49: 666-680.

Jones, G. 1973. A history of the Vikings. Oxford, Oxford University Press.

Keirstead, T. 1992. The geography of power in medieval Japan. Princeton, Princeton University Press.

Kim, S. (ed.) 2004. Atlas of Korean history. Singapore, Stallion Press and Department of History Education and Korea National University of Education.

Maekawa, K. 1991. Toshi kōkogaku no kenkyū: chūsei kara kinsei e no hatten (Studies of urban archaeology: development from Medieval age to Modern age). Tokyo, Kashiwa Shobo.

Maekawa, K. 1998. Chūsei sonraku to toshi - chiiki shakai ni okeru shisonka to toshi no seiritsu (Medieval villages and cities: regional village nucleation and the formation of cities). Kokogaku Kenkyu 45(2): 41-60.

Maekawa, K. 2004. Chūsei sōgō shiryōgaku no kanōsei (The possibility of an integrated study of resources on the medieval period). Tokyo, Shin Jimbutsu Oraisha.

11

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 27 2

12

The archaeology of medieval towns: case studies from Japan and Europe

Mass, J. P. 2002. The origins of Japan’s medieval world. Stanford, Stanford University Press. McClain, J. and Wakita, O. (eds.) 1999. Osaka: the merchants capital of early modern Japan. Ithaca, Cornell University Press.

McClain, J., Merriman, M. M., and Ugawa, K. (eds.) 1994. Edo and Paris: urban life and the state in the early modern era. Ithaca, Cornell University Press.

Tsuboi, K. and Tanaka M. 1991.The historic city of Nara: an archaeological approach. Tokyo, Centre for East Asian Cultural Studies, and Paris, UNESCO. Turnbull, S. 2009. Strongholds of the Samurai: Japanese castles 250-1877. Oxford, Osprey Publishing Ltd.

McKelway, M. P. 2006. Capitalscapes: folding screens and political imagination in late medieval Kyoto. Honolulu, Hawaii University Press.

Wakita, H. with Hanley, S. B. 1981. Dimensions of development: cities in fifteenth- and sixteenth-century Japan. In: Hall, J. W., Nagahara K. and Yamamura K. (eds.) Japan before Tokugawa: political consolidation and economic growth 1500-1650. Princeton, Princeton University Press: 295-326.

Morris, M. 2006. Kyo-machiya: tracing the development of the traditional town houses of Kyoto through the medieval centuries. Vernacular Architecture 37: 1-23.

Xiong, V. C. 2000. Sui-Tang Chang’an: a study in the urban history of late Medieval China. Ann Arbor, University of Michigan.

Morris, C. 1984. The illustrated journeys of Celia Fiennes c.1682-c.1712. London, Macdonald.

Murillo, J. F., Hidalgo, R., Carrillo, J. R., Vallejo, A. and Ventura, A. 1997. Córdoba: 300 – 1236 d.c., un milenio de transformaciones urbanas (Cordoba: 300-1236: a millennium of urban transformations). In: de Boe, G. and Verhaeghe, F. (eds.) Urbanism in medieval Europe (Papers of the ‘Medieval Europe Brugge 1997’ conference volume 1). Zellik, Instituut voor het Archeologisch Patrimonium: 47-60.

Ono, M. (ed.) 2006. Chūsei toshi no chōsa bunseki hōhō ni kan suru kenkyū (Methods of research and analysis of medieval towns in Japan). Kokuritsu Rekishi Minzoku Hakubutsukan Kenkyū Hōkoku (Bulletin of the National Museum of Japanese History) 127. Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan. Pearson, R., Li, M. and Li, G. 2001. Port, city and hinterland: archaeological perspectives on Quanzhou and its overseas trade.In: Scottenhammer, A. (ed.) The Emporium of the world: maritime Quanzhou, 1000-1400. Leiden, Brill: 177-236. Pitelka, M. In preparation. Daily life and destruction in medieval Japanese castle towns.

Ponsonby-Fane, R. A. B. 1956. Kyoto, the old capital of Japan (794-1869). Kyoto, The Ponsonby-Fane Memorial Society.

Popescu, E. 2009. Norwich Castle: Excavations and historical survey 1987-98. East Anglian Archaeology 132.

Pratt, K. 2006. Everlasting flower: a history of Korea. London, Reaktion Books. Schofield, J. 2011. Medieval London. Sheffield, Equinox.

Schofield, J. and Vince. A. 2003. Medieval towns. London, Continuum Press.

Sindbaek, S. and Nixon, S. 2012-13. Beyond the frontiers of medieval Europe. The European Archaeologist 38: 76-77.

Skinner, G.W. (ed.) 1977. The city in late imperial China. Stanford, Stanford University Press. Souyri, J.P.(ed.) 2013. Le Japon Médiéval. Au temps des samouraïs. Archéothema 30.

Stavros, M.G. 2016. Kyoto: an urban history of Japan’s premodern capital. Honolulu, University of Hawaii Press.

Wheatley, P. 1978. From court to capital: a tentative interpretation of the origins of the Japanese urban tradition. Chicago, Chicago University Press. Xu, Y. 2000. The Chinese city in space and time: the development of urban forms in Suzhou. Honolulu, University of Hawaii Press.

Zhu, J. 2004. Chinese spatial strategies: imperial Beijing 1420-1911. London, Routledge Curzon.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 28 3

Japanese medieval trading towns: Sakai and Tosaminato

Japanese medieval trading towns: Sakai and Tosaminato Richard Pearson Summary

Trade was essential to the development of urban forms in medieval Japan. In this paper, Richard Pearson introduces discoveries at two important medieval trading centres: Sakai, in the Kinai region, comprising the modern prefectures of Ōsaka, Nara and Kyōto, the longest established urban areas in Japan; and Tosaminato, headquarters of the Andō clan at the northern tip of the main Japanese island of Honshū, centre for the northern trade with Hokkaidō.

Pearson’s paper is divided into two main sections. The first part is a review of the historical and social background to the rise of medieval urban centres in Japan in general. Pearson identifies a series of major social and economic trends which underlay the development of towns: village nucleation; trade, with a particular focus on ceramics and coinage; changes in land tenure and local government; the development of new economic institutions including the appearance of shipping agents who acted as middle men between the rural estates which produced commodities and the markets, and cooperative guilds; changes in the composition of social classes, most notably the rise of the provincial warrior class and an urban merchant class; and changes in religion important for understanding the development of urban areas around temples or shrines.

The second section of the paper presents detailed summaries of the archaeological discoveries at Sakai and Tosaminato, which are related to the social and economic trends discussed in the first section. Combined with the results of the broad area excavations at Kusado Sengen in on the Inland Sea coast, these summaries introduce the reader to the diversity in medieval Japanese trading towns and to the factors which led to their development and demise.

The medieval history of Sakai and the Ōsaka region where it is located have been previously discussed in English by Wakita Haruko (Wakita 1999) and V. Dixon Morris (1977) and Pearson’s paper demonstrates how much the archaeological record of these port cities adds to the historical sources. This is the first time the results of the investigations at Tosaminato have been presented in detail in English.

This chapter was originally published in the Japanese Journal of Archaeology Vol 3

Introduction

In this paper I discuss some of the major historical and archaeological trends in the rise of towns in the Japanese islands, relating them to the moated trading centre of Sakai, near Kyōto and Ōsaka, and Tosaminato, the trading town of the Andō Clan and centre for northern trade.

Although there had been a long standing interest in medieval artifacts such as coins and ceramics, major excavations of medieval sites in Japan began with Kusado Sengen-chō, Fukushima City in 1961 (Matsushita 2000) (Figure 1). By the 1980s wide area excavation of many different kinds of sites all over Japan resulted in a proliferation of important publications such as the series, Yomigaeru Chūsei (The Rise of the Middle Ages) (Amino et al 1994). In 1991 Maekawa challenged earlier criticisms that the field had not yet reached a stage where summarisation and generalisation could be achieved (Maekawa 1991) and in the following decade Japanese medieval archaeology matured into a well developed area of study with a substantial literature and recent reviews (Sakazume 2000 and Ono 2001). In many prefectures medieval sites now constitute the largest category of sites excavated in government-funded salvage projects. From the 1990s there has been a trend to produce an archaeologically derived view of social history, with the co-operation of historians. This can be seen in the annual publication, Chūsei Toshi Kenkyū (The Study of Urban Centers).

Historical and social background

The medieval period (chūsei) in Japanese history constitutes the transition from the ancient period (kodai, c. 600 to the 11th century) and the early modern period (kinsei, 1600 - 1867). The precise dates of the beginning and end of the medieval period are debated; however it is generally thought to extend from the Late Heian Period to about 1600, a span of over 500 years. It extends through the Kamakura period (1185 - 1333) to the Muromachi (1333 - 1568) and Azuchi Momoyama (1568 1600) periods. In the Heian period (794 - 1185) the capital was Heiankyō, the site of modern Kyōto. During the Kamakura period (1185-1333) the capital shifted from Heiankyō to Kamakura, and back to Kyōto in the Muromachi period.

13

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 29 3

14

The archaeology of medieval towns: case studies from Japan and Europe

The medieval period can be seen as an interlude between two periods of unification of the 7th to 9th centuries and the 17th century to the present. During this time there was a decline in the central authority of the Japanese (Yamato) state and the rise of local powers. The period is also marked by the development of separate realms of military and sacral power and the rise of warrior and merchant classes. From 1470 to 1570 Japan was wracked by internal strife and shifting power among local rulers loyal to the central power and those seeking local autonomy. This is the time period of the building of fortified castles in Japan, termed the Warring States period (sengoku jidai). The medieval period came to an end with unification under military rulers, whose power was consolidated in the Tokugawa period (1600 – 1867). Despite the political upheavals which occurred throughout the medieval period, there were three centuries of economic growth between 1250 and 1550 (Totman 2000: 160-175). State control was very loose in the northern parts of Honshū; Hokkaidō and Okinawa were economically linked to the centre but were politically independent. Regional diversity and varying rates of change are very important aspects of the study of medieval Japan. The cultural and political conditions under which towns arose were extremely diverse, a wide variety of what are termed urban forms (toshi) being produced by multiple processes (Ishii 2000).

Major social and economic trends in medieval Japan

Village nucleation and the rise of towns and cities

Maekawa (1998) has identified a process by which villages coalesced and increased their administrative, service, handicraft, and religious roles, becoming multifunctional regional centres. He has termed this process sonshuka, which he has translated as nucleation. It is a process of coalescence or agglomeration which creates regional nuclei. A trend toward nucleated rural villages began in the 13th century and progressed rapidly during the 14th and early 15th centuries. It was driven by the expansion of arable land and irrigation systems by farmers who had previously lived in isolated farmsteads (Maekawa 1998, Yamakawa 1998). Maekawa identified three trajectories of nucleation of late medieval villages (1998:45). One of these led to the creation of residential forts (jōkan), a second to the development of villages which persist into the present and a third led to the development of urban centres. The development of the residential fort can be seen in the structure of the 16th century site of Ichijōdani, Fukui Prefecture (see Ono, this volume). In this site the internal land divisions (tansaku) are relatively small. They are the units given to local samurai groups, and vary according to the social status of the samurai. Maekawa

(1998) considered the appearance of the long narrow tansaku pattern of land tenure as a key indicator of urban life and the presence of artisans and merchants who were dependent on farmers for their food. The third type, the urban commercial centre can be seen in the Kusado Sengen-chō site in the 13th to early 14th centuries. In the ancient period it was already a local capital and in the medieval period it formed from a number of villages who shared a market. These nucleated settlements took different forms in different parts of Japan (Maekawa 1997).

By at least the 13th century, town-like or urban features can be found archaeologically in relatively remote areas such as Yanagigosho, the administrative capital of the Northern Fujiwara polity in Hiraizumi, Iwate Prefecture (Maekawa 2000: 93). The site lies in northeastern Japan, the ‘Eastern Country’ known is Japanese as Tōgoku. It was relatively remote from the administrative centre of Kyōto. In the Tōgoku area at the end of the ancient and in the early medieval periods there were rectangular garrisons which secured the power of the central authority. They housed administrators, craftsmen, and religious practitioners. These sites were surrounded by moats and earth ramparts and contained dwellings, a flat assembly area, and in some cases a Buddhist temple.

In the late 13th century there was a proliferation of port towns along the Inland Sea, such as Kusado Sengen, and Sakai. In northern Kyūshū, Hakata was the main port for the China trade (see Ōba, this volume). In the Kinai region, comprising the modern prefectures of Ōsaka, Kyōto and Nara, early centres of the 14th century show evidence of iron working. It appears that artisans lived around elite residences (yashiki) in a dependent relationship since workshops have been found around the edges of elite residences. Communities were also built around temples (monzenmachi).

The development of a commercial sector can be seen in the site of Kusado Sengen beginning in the 14th century. It has characteristics of both a port town and a temple town since it was the site of a substantial temple, the Teifukuji, built around 1321. In the 13th century it consisted of a rectangular group of residences and a cemetery of cairn burials. The chronology of the site can be divided into four periods. From the mid 13th to the beginning of the 14th century there was a main road running north to south. From the 14th century an area of long narrow lots was found as well as evidence of brewing, iron working, circulation of coins, and lacquer production. The number of houses declines in the early to mid 15th century and there was rapid decline in the 16th century (Ono 2001:48).

One of the most significant new urban developments of the Muromachi period was the building of castle towns (jōkamachi) by local leaders involved in territorial struggles (sengoku daimyō). By the late Muromachi period, castles were no longer built on defensible peaks but were situated on plains where daimyō could

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 30 3

Japanese medieval trading towns: Sakai and Tosaminato

government. From the end of the 10th century to the beginning of the 12th century the most important centre for trade was the Dazaifu Kōrokan, set up especially for interaction with foreign traders. There was an increasing inflow and use of coins before the end of the 12th century, although many rural people still exchanged goods without the use of coins. The volume of Chinese ceramics flowing to Japan increased dramatically in the 12th century, reaching a peak in the 13th and 14th centuries. It declined in the 15th century but was resumed in the 16th century. Trade in Chinese goods included works of art and antiques as well as coins and ceramics. The renewal of continental contacts, which had begun in the late Heian period, was fuelled by a demand for Chinese objects by political and religious elites, particularly under the rule of the Ashikaga family of military rulers (Totman 2000: 175 ).

Coinage

Figure1: Map of Japan showing major sites mentioned in the text.

combine defence with control. These communities included warriors, merchants, and craftsmen. Earlier fortified castles were demolished. By the late 16th century large numbers of villages in the Kinai were surrounded by ditches and moats. The latter were used not only for defence but also as important parts of irrigation systems. Up to this point we have been discussing regional centres. During the 14th and 15th centuries the population of the primal city of Kyōto was over 200,000, making it one of the world’s largest cities of its time. The population of London at this time is said to have been 50,000 (Yamamura 1990: 377).

Trade Medieval Japanese towns and cities served as redistribution centres. In this section I outline several of the major turning points of trade between China and Japan and summarise two of the commodities, ceramics and coinage. The first stage of the development of the medieval economy and society of Japan is linked to the rise of the Song economy in China. From the 10th century the pace of commercial development quickened in China and in the 11th and 12th centuries private traders from China came to the Japanese port of Hakata. By the late Heian Period, trade, mostly private, was conducted by elites through ports in northern Kyūshū and Japan became part of the East Asian trading sphere. Song ships came to Dazaifu and traded with representatives of various temples and shrines and their attached estates. Little trade was carried out by the central

Throughout the medieval period Japan used Chinese copper coins as the medium of exchange; therefore it was a crucial import (Totman 2000: 153). Huge amounts of Song coins were imported, despite Chinese official prohibition in the late 12th century (Yamamura 1990: 358). In the 15th century the Ming court sent Japanese leaders diplomatic gifts of millions of coins (Atwell 2002:86, Colcutt 1983: 174). From 1404 Ashikaga Yoshimitsu opened up trade with Ming dynasty China, accepting from the Chinese the title of ‘King of Japan’. There were only nineteen missions between 1404 and 1549, when the Ming terminated these missions because of internal confusion in China and piracy along the coast. In the transition from the Medieval period to the Modern period in the early 17th century, the Tokugawa government prohibited the use of Chinese coinage and minted its own currency in gold, silver, copper and later, iron denominations.

The Sin’an shipwreck of the early 14th century, found near Mokpo, Korea, contained 8.1 tons of Chinese coinage, some eight million coins (Information Section 1984, Kokuritsu 1998a: 60). It is said to have belonged to the Tōfukuji Temple, Kyōto. Dramatic hoards of Chinese coins have been found in Hokkaidō and the main Japanese Islands, but are virtually absent in the Ryūkyū Islands. In addition to chance finds such as the discovery in 1968 of around 450,000 Chinese coins dating from Han times to 1368 at Shinori near Hakodate, Hokkaidō (Nagai 1994: 16) there are now several cases of large quantities of coins from controlled excavations (Kurihara 2000). A few examples will convey the scale of these finds. Some 77 percent of the coins in the caches are from the Northern Song (960 – 1127) (Suzuki 1999: 59). In Kusado Sengen-chō two finds were made. One, weighing nineteen kilogrammes, was comprised of five thousand coins while the other consisted of 12,595 coins. All of the coins were dated before 1368, the beginning of the Ming Dynasty. From Niigata,

15

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 31 3

16

The archaeology of medieval towns: case studies from Japan and Europe

at Ishiusu, near Kamitate Castle and the Sempukuji Temple, 271,784 coins were found in association with architectural features and ceramics. There were 82 types of coins on strings of between 92 and 99 coins each, the latest being the Korean Chōsen Tsuho type, dated to 1423. From Namioka Castle in Aomori, privately minted Chinese coins were also recovered (Suzuki 2000).

Suzuki (1999) gathered data from 275 cases of hoards totaling 3,530,000 coins. Four prefectures, Akita, Shimane, Kōchi, and Okinawa are not included in his survey since their data were not yet compiled. He concluded that the coins were used in commerce and for paying soldiers. They were buried primarily for security, while a few instances were ritual or votive deposits. He concluded that most of the coins were transported to Japan in the 13th century, and that the hoarding peaked in the first half of the 16th century, declining in the latter half of the 16th century. He cites historical records of transactions in feudal manors (shōen) which show that the use of currency increased in the 13th and 14th centuries, in comparison to rice or textiles. In the 16th century coins of Yongle Tongbao type were specially favoured for commercial transactions. Both Chinese coins and trade ceramics provide important methods of dating medieval sites, as well as important information on changes in patterns of consumption. As Suzuki points out (1999: 60) dating from coins is not simple, since the coins remained in circulation for a long time. The relationship between the time of import (which can only be postulated) and the time of burial in the cache is not always well known. The Japanese also produced counterfeit Chinese coins at sites such as Sakai.

Ceramics

Chinese ceramics constituted the bulk of the imports in medieval Japanese foreign trade but there were also relatively small quantities of ceramics from Korea, Vietnam, and Thailand. The history of the ceramics is relatively well known through the intensive work of pioneer scholars. Excavation of the Kōrokan yielded ceramics from the 7th to 11th centuries, including Korean wares from the Korean states of Silla and Koryo, Yuan ceramics from Zhejiang, Changsha Tongguan ceramics, northern white wares and bluish white (qingbai) wares from Zhejiang and Guangdong. The most numerous are Chinese yue ware, a kind of proto celadon. While officials and upper class people at Dazaifu used yue bowls and plates from the 9th century, by the 11th century their use had spread to lower class people. The Kōrokan served as the distribution centre for ceramics until the mid 11th century, after which the merchants of Hakata assumed control of ceramics. Sites thought to be private warehouses of Song merchants have been found in Hakata (Kamei 1995: 132). Celadon ceramics from Longquan, Zhejiang, replaced yue ceramics in the mid 11th

century through mid 12th centuries (Kamei 2000: 136). Findings from the Mottaimatsu Site in Kagoshima Prefecture, southern Kyūshū, which yielded ceramics from the latter half of the 11th century to the 13th century, indicate that in addition to Hakata there were other distribution centres (Yamamoto 2003). At that time, bowls with lotus petal motifs incised on their exteriors are found, along with bowls and plates with a peculiar dark green glaze, from Tong’an, Fujian. By the 12th and 13th centuries, the rate of flow of Chinese ceramics into Japan increased. Some of the sites under the modern city of Hakata yield counts of Chinese ceramics as high as 24 percent. Ceramics can be found from Okinawa in the south up to the Tōhoku region, though the elite of northern Japan preferred to use lacquered wooden bowls. From the middle of the 13th century, the amount of ceramics decreases suddenly in the Japanese main islands.

In the first half of the 15th century, Chinese ceramics are very rare in the main islands of Japan, while they are abundant in the Ryūkyūs. During the ban on trade under the Ming Trade (13681522) the Ryūkyū Kingdom of Okinawa maintained close diplomatic ties with China and engaged in trans-shipment of Chinese goods. In the 16th century Chinese ceramics became more common in the central areas of Japan, in sites such as Himeji, and Ichijōdani, where the Asakura family had their regional capital from 1470 to 1573 (see Ono, this volume). In the former site about half of the ceramics are local earthenware, while the other half are Chinese celadon, white, and blue and white wares and Japanese wares such as Bizen, Tamba, Shigaraki, Tokoname and Hizen. In bowls and plates Chinese wares were seven times more common than Seto or Mino (Kamei 2000: 58). In the case of Ichijōdani, ceramics from the houses of the samurai and merchants consisted of about fifty percent earthenware and thirty percent Echizen ceramics. Chinese ceramics were about ten percent. However in the residence of the Asakura family members the proportion of trade ceramics was about thirty percent. They imported ceramics for daily use, including a rare antique white ware four-eared jar.

In the 16th century trade ceramics were found in higher frequencies in the northern areas of Tōhoku and Hokkaidō and in the southern regions of Kyūshū and Okinawa, whereas in the central area local Mino and Seto ceramics were also available (Suzuki 2000). Nevertheless even in these latter areas Chinese bowls and plates remain three to five times more abundant than the local ceramics (Kamei 2000: 60). Most of the Chinese blue and white is from the Jiajing Period (1522 to 1566) when trade was restored after the ban on trade under the Ming. At this time silver was mined in Japan and exported to China in huge quantities. Kamei believes that ceramics were brought in Chinese ships to Hakata and Sakai and were then distributed

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 32 3

Japanese medieval trading towns: Sakai and Tosaminato

through domestic networks. In domestic ceramic production, in the early part of the medieval period, production was not for daily use. Elite and ritual wares such as four-eared tsubo, ewers and bottles, were produced in Kamakura for the Tōgoku region. Three types of ceramics developed, all based on prototypes from the ancient period. These were ash glazed, sue, and kawara type wares. The ash glazed types developed into high quality hard fired ceramics. Grey stoneware was produced in round and flat bottom variants. Soft grey earthenware was also used for ritual, storage and heating.

Changes in land tenure and local government

In the 7th century under the Taihō Promulgations, the ritsuryō system of centralised state land tenure was established, despite opposition of independent uji chieftains. Ownership of land resided with the emperor. In the 10th century the system was changed from levying tax and corvee on individuals to land under cultivation. Also in Late Heian times, huge private estates (shōen) were created, removing land from the public domain and redirecting the revenues made from them to large temples and aristocratic families (Totman 2000: 100-105). They were originally parcels of land brought into cultivation by temples, but in later times they were estates of cultivated fields designated to support particular temples or other institutions, by imperial decree. In the late 12th century a second class of officials, land stewards (jitō) were assigned to both public and private lands to supervise the collection of taxes and the regulation of production. As the military rulers (shōgun) gained power in the 13th century they appointed military governors (shugo) throughout the country to maintain law and order. In the Muromachi period local lords (daimyō) exercised aristocratic control over their castle towns; however in cases of the trading ports of Sakai and Hakata, citizens’ councils governed, and also managed defences. In Kyōto groups of townsmen regulated affairs in their district but in the Azuchi Momoyama period these groups were brought under close regulation.

Development of new economic institutions

Two economic institutions that had appeared earlier and continued to develop during the Kamakura Period were the shipping agents who took rice and other products of the shōen estates on consignment for distribution to markets (toimaru) and co-operative guilds which provided favorable reciprocal trade advantages and reduced competition (za) (Yamamura 1990: 394). Many Heian artisans functioned as house provisioners, providing goods and services to elites. During the 12th century many of the elite institutions licensed specialised za to provide many kinds of supplies (Totman 2000: 109). The za developed mostly in the Muromachi period. They enjoyed the sponsorship of nobles of religious institutions to whom they paid dues in return for protection of their monopoly privileges and exemption from barrier charges and market fees.

In the Muromachi period there were advances in agriculture, commerce, transportation, village organisation, and urban development. There was a strong demand for goods and services from the shogunal court and the resident shugo daimyō. The policies of the sengoku daimyō aimed at fostering local commerce by new groups of merchants and at reducing barriers to trade within their domains. Two powerful new social forces emerged: a self-conscious mercantile group and an increasingly restive and market oriented economy. The toimaru, originally agents who supervised rent shipments for estate proprietors, began to handle the affairs of more than one patron and went into business for themselves, becoming powerful traders, as in Sakai (Morris 1981: 33).

Double cropping of rice and barley, which had begun in western Japan in the Kamakura Period, spread to eastern Japan by the Muromachi period. Improvements occurred in fertilizing, irrigation, and mining of gold and silver. The building of massive castles and castle towns by daimyō stimulated haulage and forestry. The need for peasants to convert at least part of their crops into cash to pay taxes and levies also contributed to the development of markets and commercial activity.

Changes in social classes

Each class developed its own residential and community patterns and patterns of consumption, which can be seen in archaeological sites. The aristocratic centralised life of the Heian imperial court in Kyōto was maintained by the civil aristocracy (kuge) whose courtly background set the standard of taste in Heian times. In the Muromachi period actual power was transferred from the kuge to the provincial warrior class (bushi or samurai). A great number of private landed estates (shōen) came under the control of the warrior class. In the 13th century a legal code of the warrior class set out the rights of the warrior class and the duties of the Kamakura-appointed officials such as military governors and stewards. The rise of the merchant class in the 15th and 16th centuries is particularly important, as we shall see in the case of Sakai, outlined below.

Religious changes

Space does not permit me to discuss in detail the effects of religious changes on life in medieval Japan. Two new forms of Buddhism introduced in the 9th century, Shingon and Tendai, had centres outside Kyōto, at Mount Kōya and Mount Hie. Their warrior monks exerted considerable power over Kyōto, although Buddhism did not spread widely among the masses in Heian times. Pure Land Buddhism became popular in the latter half of the Heian Period, and became a separate sect in the Kamakura Period. New Japanese sects such as Nichiren, arose in the Kamakura period through charismatic leaders. Zen Buddhism was introduced from China in the late 12th century, first to Hakata, and later to Kamakura, where a close

17

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 33 3

18

The archaeology of medieval towns: case studies from Japan and Europe

relationship between Zen and the military leadership was established. Both Rinzai and Sōtō sects of Zen flourished, along with early forms of Japanese Buddhism in large centres such as the Tōdaiji and Kōfukuji in Nara. Song traders supported Buddhist temples in the Hakata region (Kamei 1995: 134-135) . In the Muromachi period several large Rinzai temples of Kyōto became deeply involved in international trade and diplomacy (Collcutt 1981), while other Rinzai temples eschewed such activities. Christianity was introduced by Iberian missionaries from 1549. While it flourished in some areas for several decades it was snuffed out in the 17th century. In addition to the formal religions mentioned above I would argue that a new commercial mentality, rooted in the rise of industrialisation and a cash economy in South China in the Song Dynasty (960 to 1279) greatly affected the values of Japanese elites and townspeople. These changes were linked to the developments in Buddhism mentioned above, and the great Zen temples of Kyōto which were engaged in international trade (Collcutt 1981).

Sakai: historical background

As one of the major medieval centres of the Japanese islands, Sakai displays the expanding, multifunctional roles which typify towns of this period. Its rise in the 15th and 16th centuries is seen by some as the beginning of modern Japan, but its decline and loss of independence in the 17th century is also seen as a regression into Tokugawa feudalism (Morris 1981: 23-25). Both a local and international entrepot, and an important trade port for the land locked Nara Basin, Sakai, originally a fishing village, is situated on old communication routes running from Ōsaka Bay through low mountains into the Nara Basin, and also along the coast of Wakayama to the religious centre of Kumano (Figure 2). Sakai had long been a strategic transportation node. From the 5th century it was the site of monumental tombs of the rulers of the Kinai (Kyōto Nara Ōsaka region). The location of such impressive tombs outside the Nara Basin, in the Kawachi Plain, must have been partially motivated by a desire to impress visitors who voyaged into the Inland Sea, the maritime terminus of the Silk Road. It was an important stop on the pilgrimage route to Kumano, and a port for local produce. It was a military port during the Namboku-chō period (1336 - 1392) and an important port of trade within the Japanese islands. In the 15th century it became an international port and in the late 16th and 17th centuries it became a centre for manufacturing. Through its link with Okinawa it also had access to goods from Southeast Asia. The city was dominated by a number of powerful Shintō shrines situated along the main thoroughfare which ran parallel to the shoreline. In the area of Sakai port, a well thought to be used for agriculture has been dated to the late 12th and early 13th centuries, suggesting that the area was a small settlement

Figure 2: Location of Sakai

with garden plots at that time (Tsuzuki 1994:85). Early farmers may have used drainage water from nearby 5th century tumuli for agriculture and ditches from tumuli may have used part of Sakai’s moat (1994:85). Tsuzuki (1994:96) states that a substantial area within the moat was inhabited by people who had farm plots.

At the beginning of the 13th century, the northern part of Sakai lay in the shōen of Settsu, while the southern part was located in the Waizumi shōen. This location on such a boundary, at the junction of old roads, gave it a strategic advantage for internal trade in the Kinai as well as for international trade (Figure 4). Later, Sakai became the shugo territory of the Hosokawa family, who were kanrei (shogunal deputies). At this time it was already an important trading centre for the small polities of Kyūshū and the Chūgoku (southwestern Honshū). In the 15th century, the Shōkokuji Temple of Kyōto received the rents from the properties in Izumi Sakai, while Settsu Sakai was under the proprietorship of the Sumiyoshi Taisha Shrine (Morris 1981: 25)

During the Ōnin Wars (1467-1477), aristocrats, priests, and townspeople from Kyōto and surrounding areas sought refuge in Sakai and the city developed rapidly. During the 15th century it was the base for 19 tribute missions from Japan to the Ming

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 34 3

Japanese medieval trading towns: Sakai and Tosaminato

(Semboku Kōku Shiryōkan 1996). The Ming tribute missions were complex ventures with multiple sponsors for which the Sakai merchants acted as ambassadors for the Hosokawa family (Fukuoka Style 1992: 38). In the tribute mission of 1451, 1200 persons traveled in the ships and 350 travelled to the capital, Beijing (Itō 1998). Interests in the tribute missions were held by the Five Great temples of Kyōto, the Sakai merchants, the Hosokawa family of high officials in Kyōto, and the merchants of the Ōuchi clan in Hakata (Sakai Shi Hakubutsukan 1993: 8). In the latter half of the 15th century a self governing council was established. At the height of its prosperity in the 16th century it was governed by a council of 36 prosperous merchants, three members carrying out the administration in monthly rotation (Kiyonaga 1990: 151). At that time, many communities had a kind of autonomous governing council, responsible for the payment of taxes. Morris states that Sakai’s governing council (egōshū) originated as a shrine guild. It was self governing, as was the port of Hakata in Kyūshū. Sakai also traded with Okinawa, which was given preferential trade status with the Ming and had extensive connections with Southeast Asia. In addition to foreign trade, it was an important centre for casting weapons and guns (late 16th century), salt production, and the making of ink stones. It was also a centre for making gunpowder, which was made from saltpeter brought from Thailand (Semboku Kōku Shiryōkan 1996: 4), and was also a centre for the development of the tea ceremony. The military role of Sakai is important, as well as its commercial and cultural attributes. Sakai merchants were ultimately controlled by warriors since they were the only group who could afford the services and luxuries of the city (Morris 1981: 34).

The city underwent two episodes of destruction, the first, at the time of the Ōei Disturbance (1399), ten thousand houses were destroyed in the central part of the town, leaving a burnt earth layer. To cope with the threat of widespread disturbance and a takeover by the warrior Oda Nobunaga in 1569, the inhabitants began to build a moat around the main town, an area of one by three kilometers. The construction of the moat follows the practice of other smaller medieval centres. Sakai did capitulate to Nobunaga in 1569 but was not harmed, since it was valuable only when it was economically fully functional. In 1615, at the time of consolidation of power of Tokugawa and the fall of Ōsaka castle, the entire city of some twenty thousand houses, was destroyed to be rebuilt after the institution of a new survey and different land units ( Fukuoka Style 1992: 36). Trade goods moving to China from Sakai included sulphur, copper, sword blades, gold screens and art objects. Goods from China included ceramics, craft work, medicine, books, paintings, and other Chinese objects. Sakai merchants were also involved with the incense trade from southern China and

Southeast Asia (Kiyonaga 1990: 149). Sakai was a centre for metal casting and export; in the 14th century it was the shipping port for special temple bells cast in the Tannan region of Kawachi, the plain to the west of Sakai, and the city itself was a centre for casting coins and weapons. It had access to silver and iron from Tajima in the Hyōgo region. Buddhist texts, wood block printed in Sakai, were also shipped from there (Fukuoka Style 1992:36, Kiyonaga 1990: 153). Weaving of expensive types of cloth was also important and shuttles and spindles have been recovered from archaeological sites of commercial goods. Warehousing was an important activity of Sakai merchants (Watsky 2004: 20).

Sakai merchants who also traded in weapons became patrons of the arts, sponsoring the tea ceremony (Sen 1998). The merchant princes (gōshō) of Sakai commanded the technology to circulate goods from Sakai to the Japanese hinterland and vice versa; they financed the local (shugo) daimyō and shōen officials at high interest rates, and they were engaged in foreign trade. Moreover they had a bold entrepreneurial and independent mentality (Kiyonaga 1990: 151).

Sakai sites

Situated on the beach sands of Ōsaka Bay, Sakai contains archaeological deposits about four metres deep. From excavation, the history of the site before the Namboku-chō period is not clear, although there are some scattered remains from the Kamakura, Heian, and Nara periods. In 1962 archaeologists located the burned soil layer left by the destruction of the site in 1615, 1.5 metres below the present ground surface and in 1975 they began excavations which continue to the present (Sakai 1997). Up to the present hundreds of localities have been excavated. A few of the excavations are summarised below.

Commoners’ residential area

This excavation contained deposits from the fire of 1615 and yielded a wooden slip (mokkan) dating to Tensho 13 (1588). It appears to be part of a commoners’ residential area, containing a stone wall and a stratified deposit. Ceramics were: haji ware (small plates, suribachi, and hibachi) 47 percent; Japanese wares (Bizen, Seto, Mino, Karatsu) 17.8 percent; Chinese and Korean ceramics, 8 percent; Minato (local) wares 16.5 percent; local tile wares (hibachi) 10.8 percent (SKT 19: Sakai Shi Kyōiku Iinkai 1984).

Small buildings

To the north and west of the central part of Sakai, in an area which shows no sign of the burning of 1399, there are the remains of small buildings with stone pillar bases, along with wells and various kinds of pits. The artifacts include soft grey (kawara) ware large mouth storage jars, and plates, and soft red (haji) and grey plates, These were used for storage and serving

19

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 35 3

20

The archaeology of medieval towns: case studies from Japan and Europe

food. Japanese hard fired ceramics included Bizen ware, wide mouth storage jars and shredding and mixing bowls (suribachi), Tamba bowls and Seto bowls, plates, and jars. Chinese wares include white ware bowls and plates from Fujian, and celadon bowls, plates, and platters (chargers) from Longquan, Zhejiang (SKT 21, 61: Sakai Shi Hakubutsukan 1993: 89)

Warehouse site

This is situated at a slightly elevated location, 4.4 metres above sea level, about fifty metres from the shore at the time of occupation. The excavation in 1989 of about fifty square metres in area exposed fifteen layers, from the 18th century to the end of 14th century with a total depth about five metres. The finding of a long narrow line of grey tile bricks for supporting buildings indicates a group of warehouses near the shore. This type of grey brick storehouse architecture has been found from the end of the 15th century. In Layer 4, dating to the latter half of the 16th century, parts of a residential compound (yashiki) with a building facing a garden and a garden entrance (tsuri en) were recovered. In Layer 3 dating to 1615, there are warehouses and fragments of tea wares (Tsuchiyama 1996). (SKT 39 Sakai Shi Hakubutsukan 1993: 60).

Buildings destroyed in the fire of 1615

Writing hall

A large scale shoin type (writing hall) building was discovered in 2000. The finds reflect changes in the tea ceremony, in which Sakai merchants began to use Japanese and Southeast Asian wares, but retained their interest in Chinese wares. The finds included Japanese Karatsu, Bizen, Mino, Shigaraki, Iga, Tamba, and Raku Wares; Chinese blue and white from Zhangzhou and Jingdezhen, white wares, celadons, overglaze enamelled, Huanan Three Color, and brown wares; and Thai, Vietnamese and Burmese wares (Kanazawa 2004). This site exemplifies the trend toward the ‘grass hut style’ of tea ceremony, a less elaborate form removed from warrior elite control, and featuring Japanese and Southeast Asian as well as Chinese ceramics (Sleusser 2004).

Tea master’s tea house and warehouse

Architectural remains include a tea house and storage area. There were about 500 ceramic vessels, all burned, comprising imported ceramics (61%), Japanese ceramics (25%), and local earthenware (14%). Many of the food serving vessels are Chinese blue and white while the storage vessels are Japanese. (SB 301: Sakai Shi Hakubutsukan 1993: 63-64).

Nucleation and urbanisation

This area yielded a two storey building containing many tea ceramics along with the foundations of 8 other buildings all burned in 1615. One of the buildings, 6 m x 7 m, had a sunken floor paved with long rectangular grey tiles. On the east side of the building was a two storey building with door and windows. It yielded burned and fallen plaster, a tea grinding mortar, and red earth (kabe) for plastering the walls of a tea house as well as the remains of a garden with a built up hill (chikuyama). (SKT 47).

East-west transects of Sakai prepared by Tsuzuki (1994:66-67) show four stages of development from the mid to late 15th centuries to 1615. Warehouses begin to replace small houses along the front road by Ōsaka Bay in the late 15th century and boundaries for jōri land divisions can be seen in the inland areas. Tea houses were built near the warehouses. An agricultural zone can be seen in the western side, inside the moat, in the 16th century but by the early 17th century part of this zone has been transformed to craft industries such as casting.

Coin mold deposit was found in excavations. (Sakai Shi Kyōiku Iinkai 1997). The 1991 excavation contained several living surfaces dating from 1615 (layer 1) to the fire of 1399 (level 7). Of the total of 3, 841 mold fragments recovered, there were two types: molds for thin plain coins (85 %), and molds for inscribed coins (15%). The inscribed coins comprised 19 types, from Tang to Northern Song. These types were produced in Sakai in the mid to late 16th century; (Shimatani 1998: SKT 78).

Sakai was one of the premier trading ports in medieval Japan. Linked to the central government and to the heartland of Yamato, it had a particular commercial and financial role. The ceramics found in Sakai illustrate its importance as a local and international centre. From SKT 39 and other sites, Chinese ceramics were extremely popular up to the time of the conflagration of 1615, comprising sixty percent of the assemblage (Figure 3). Other ceramics were Japanese Bizen and Karatsu. After 1615, Imari comprised about thirty percent of the ceramics and Chinese ceramics declined to about thirty percent. In other localities Chinese ceramics were about thirty percent, while Bizen and Karatsu comprised seventy percent.

Coin molds

Residential area of townspeople

There are seven living surfaces, dating from the early to middle 15th century to 1615. Structures were small buildings set on supporting stone bases, along a road, with wells encased in soft grey ware circular tiles. Water was stored in large Tokoname jars. From Layer 5 there was evidence of long narrow tansaku lots. From Layer 3, Storage buildings were found with each house. (SKT 200:Tsuzuki 1994)

Evidence of trade

Shifts in the ceramics used in the tea ceremony in Sakai reflect general changes in taste and the availability of Chinese objects (Senboku Kōko Shiryōkan 1996). Up to the end of the 14th century, Chinese white wares, celadon, and temmoku were the primary tea utensils. In the late 15th century, Murata Shukō,

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 36

Japanese medieval trading towns: Sakai and Tosaminato

the pre eminent tea master, developed new fashions in tea consumption, favouring Japanese ceramics such as Bizen and Shigaraki as well as Fujian celadon (Sen 1998: 159-160). Despite his influence Sakai merchants continued to use Ming ceramics until the latter half of the16th century. From the first half of the 15th century to the first half of the 16th century, celadon comprised fifty to sixty percent of the ceramics from Sakai, and white wares, twenty to thirty percent. From the latter half of the 16th century, blue and white, which had been scarce up to that time, increased, comprising sixty percent of the trade ceramics at the end of the 16th century and the early 17th century. It was used not only for the tea ceremony but also for daily eating utensils.

The development of new economic institutions

The mercantile and financial role of the Sakai merchants can be seen in the finding of baked clay coin molds in SKT 78. The total number of Chinese and other Asian coins from the Sakai excavations is over 2,279,000. Plain coins of the type made in Sakai have been found in sites from Hokkaidō to Okinawa. They have a higher copper content than some Chinese or

Period

Korean coins. The inscribed coins are thinner than their Chinese counterparts and the casting is rougher. Coin molds have been found in Kyōto, Kamakura, Hakata, and Sakai, but only Sakai has produced the plain type (Shimatani 1998). Yamamura (1990: 385) has addressed the question of why the Kyōto government did not mint coins in the 14th and 15th centuries. He states that Japan did not constitute a full fledged nation state in the international community: it remained a part of the political economic sphere dominated by China. After the 9th century the government did not command the political power to have coins accepted and used nationally.

Changes in social classes

Differences in the scale of residential sites within Sakai indicate the emergence of the Sakai merchants, often referred to as merchant princes (gōshō). These sites have higher proportions of foreign ceramics. The appearance of large warehouses in the 16th century along the coastal road also indicates the ability of Sakai merchants to control large inventories and to arrange for their shipment. Tsuzuki (1994: 91) states that similar warehouses developed in Kyōto but at a later date.

Period 1 First half of 12th – first half of 13th century

Period 2 Latter half of 14th – first half of 15th century

Period 3 Latter half of 15th – first half of 16th century

Period 4 Latter half of 16th – Early 17th century

Major types of Trade ceramics

White ware bowls Xiamen type, Type 4

Longquan type celadon

Zhangzhou type blue and white. SE Asian ceramics

Traders

Hakata traders

Hakata traders

Longquan type celadon (Zhejiang, Fujian coastal kilns. Jingdezhen blue and white Fuzhou. Ningbo - Ryūkyū trade centre - Bōnotsu Tosa - Sakai

Zhangzhou Yuegang Taiwan - Ryūkyū trade centre - Bōnotsu - Funai Sakai

Primary factor in change of trade route

Southern Song trade

Opening of Ming Tributary trade (1404)

Aspect of Trade

Trade port and route

Domestic circulation system Ceramic model and copy

Chinese political change

Japanese political change

Xiamen. Quanzhou. Fuzhou - Taiwan - Ryūkyū - Bōnotsu - Hakata Hakata - Hyōgotsu Omono - Sakai

Local circulation

Ningbo - Ryūkyū trade centre - Bōnotsu - Hakata Sakai

Local circulation

Chinese white 4 ear tsubo to Seto 4 ear tsubo. Xiamen wanyao type, Jinjiang wares to Seto.

Jian temmoku bowl to Seto temmoku bowl. Longquan bowl to Seto flat bowl

Heian to Kamakura

Namoboku chō to Muromachi

Sakai traders

Sakai traders

Base for Ming tribute (1469)

Tally trade

Local circulation

Circulation over whole country

Zhangzhou kiln white Plate - Shino plate - blue and white plate Ming-Qing (1644)

Southern Song to Yuan

Figure 3: Trade ceramics from Sakai (from Morimura 1998).

Ōnin Wars, Warring States

Azuchi Momoyama - Edo

21

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 37

22

The archaeology of medieval towns: case studies from Japan and Europe

Changes in land tenure and local government

There is little archaeological evidence for the rise of the political independence of Sakai and the development of the ruling council. However, following the destruction of the city in 1615 new land divisions were instituted by the Tokugawa government and these can be seen in some structures which post date 1615.

The northern centre of Tosaminato and the territory of the Andō clan

The extreme northern part of Honshū constituted the edge of Japanese power for many centuries. Japan’s northern frontier, Hokkaidō, lay beyond the boundaries of the ritsuryō state. The consolidation of political control by the central government took many centuries. Although there were Hokkaidō style kofun tumuli in the 7th and 8th centuries, which have yielded iron agricultural implements and iron weapons, there is no evidence of political control from the Kinai centre. Satō (1994) mentions the importance of maritime communication routes in the Ancient period (7th to 11th centuries). In the Tōhoku region, groups of villages (mura) supported fortified centres (jō) and tribute and trade with the Yamato state flourished. Yamato attempted to prohibit private trade from time to time (Satō 1994:76). Northern products included fine horses, hawks, sea mammal skins, bear skins, and sea weed. Fukushima-jō, a large site earlier than Tosaminato, is located on the northern shore of Lake Jūsan (Senda 1994). Thought to date to the 10th – 11th centuries, it is triangular in area. About one kilometre on each side and 625,000 square metres in area. It has inner and outer enclosures. The eastern side of the outer enclosure has an earth rampart twelve to thirteen metres wide and three to four metres high, extending for one kilometre, while the inner enclosure is rectangular, two hundred metres on each side, with a moat and rampart. While there are no historical sources concerning the site it is thought to have been built by the East Branch of the Andō clan (Satō 1994:78). Satō states that its defensive works were built on the same scale as those of Dazaifu, the centre for trade with Korea and China, located in northern Kyūshū. There are several other medieval sites on the south shore of Lake Jūsan such as the Karakawa site (Aomori Ken 1996: 6) which has a rampart and ditch, two groups of semi subterranean pit dwellings, pillar bases, stone steps, and the remains of a five storey pagoda. The emergence of the Andō clan as key players in the northern trade indicates a shift in social classes in the region and the development of new economic institutions. The Andō clan was originally of indigenous Emishi (Ainu) origin (Sakakibara 1996,2001:27).

At the beginning of the 11th century, internal lug clay cauldrons and external lug iron cauldrons appear in Tōhoku while in

Figure 4: Map of the Tosaminato Site. (adapted from Saitō 1994)

Hokkaidō, Satsumon pottery declines and is replaced by internal lug pottery which spread from Tōhoku. Other changes reflecting influence from Tōhoku include the adoption of iron tools for making wooden eating utensils and a shift from semi subterranean dwellings to houses on the ground surface, as well as a shift from a baked clay fireplace at the edge of the house (kamado) to a sunken hearth. Suzu ceramics, produced in the Noto Peninsula area of the Japan Sea Coast, have been recovered from this and later time periods. In the 14th and 15th centuries, the period of the control of northern coastal trade by the Andō clan based at Tosaminato, the Japanese became established along the southern Hokkaidō coast. Other trading forts were built at that time. As mentioned in the introductory section, very large caches of coins have been found in Hokkaidō, indicating vigorous trade and the accumulation of monetary wealth.

There was a struggle in the mid 15th century for control of the Tsugaru Peninsula by the Andō and Nabu clans. The traditional region of the Nabu lay to the south of the Andō. The Ainu offered strong resistance to Japanese incursions in the 16th century, the time of the Koshiamuin Uprising. As the Matsumae clan attempted to establish control in the 16th and 17th centuries, the Ainu built fortified sites termed chasi. Over 25 of these sites are known.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 38

Japanese medieval trading towns: Sakai and Tosaminato

The Tosaminato site

The Tosaminato site is situated at the northern tip of Honshū, on the northwest shore of the Tsugaru Penuinsula, where the Iwaki River empties into Lake Jūsan, a lagoon with a narrow opening to the Sea of Japan (Figure 4). The site is located on a long sand bar which running north to south. Encompassing 55,00 sq m, it has been the scene of more than a decade of excavations, and the topic of symposia (Kokuritsu 1994) and exhibitions (Kokuritsu 1998b). Excavations by the National Museum of History and Folklore were undertaken in 1991 to 1993 and further excavations were undertaken by Shiura Township and the Aomori Prefectural Board of Education in 1995 and 1996 (Aomori 1996, 1998a,b) as as well as Toyama University (Shiura 1998).

Regional nucleation

Extending east to west across the peninsula is a rampart with a large ditch on its southern side, which divided the site into two parts. To the north of the rampart was the Andō clan residence, protected by another ditch one hundred metres to the north of the main ditch. Although east and west ditches have not been found, it is very probable that the Andō residence was surrounded by ditches which formed a rectangle of about one hundred metres. This has been determined by remote sensing (Senda 1994:56-59). Excavations in 1993 in this area yielded evidence of pit dwellings believed to have belonged to artisans. The main ditch fill contained Suzu ceramics dating to the 14th century. Uno (1994: 113) mentions that very small quantities of 9th and 10th century ceramics have been found in Tosaminato, suggesting that the site was not entirely unpopulated at this time. In the 11th and 12th centuries the trading sphere of eastern Japan, including Hokkaidō, expanded. Uno (1994:132) states that there were important links between Taga-jō (castle), Hiraizumi, and Fukushima-jō. Fuksushima-jō functioned more as an armed garrison outpost than as a trading post (Miura 1994). In the 12th century the trade goods from eastern Japan were circulating throughout all of Japan and Tosaminato played a role in this trend.

The western side of the sand bar faces a narrow stretch of water separated from the Japan Sea by a beach and sand dunes. In medieval times ships entered this channel, which was deeper than at present, or waited near its access to the Sea of Japan, to the south of the site. On the side of the sand bar lies the large lagoon, Lake Jūsan, also called the Back Lagoon (Ushirokata). Excavation areas 5,7,8,9,18, and 96 were devoted to the area of the Andō clan’s residence and headquarters. They uncovered pits, posts, planking, the remains of a post built building, a well, pit houses, some of which appear to have been used for blacksmithing, indicating that craftsmen were working in this area. Artifacts from this area date from the 13th to the first half of the 15th centuries, the most active period being the latter half

of the 14th and the early 15th centuries. To the north of the area of the Andō residence is the area of the retainers’ quarters (Localities 10,11,15,16, and 74; Saitō 1994). To the northeast of the retainers’ area there may have been long narrow fields in the tansaku pattern (Maekawa 2001:5). A major road running north to south, 7.2 metres wide, was located , as well as an east to west road 2.7 metres wide. In addition to the roads remains of buildings suggest a dense concentration of irregular structures which are thought to be merchants’ quarters. Dating of this area is similar to the Andō headquarters.

The third area, (Localities 6,17, and 71) is comprised of individual houses (yashiki) built with posts set in the ground and bounded by fences and ditches, which faced a central north to south road. Artifacts were dated from the end of the 14th century to the first half of the 15th century. Roads from the south of the site led to the Myōjin Shrine and the Tanrinji Temple. Amino (1994:24) notes that Jishū Buddhist and Hiyoshi Jinja Shintō networks were established along the Japan Sea coast. The Machiya Area, Area 3, yielded evidence of long narrow land divisions of the tansaku type. As Area 3 developed in the 15th century, Tosaminato assumed an urban appearance (Aomori Ken 1996:8).

Evidence for increased trade

The site yielded a density of hard fired ceramics which was higher than the norm for Tōhoku and Hokkaidō. These ceramics included Seto, Suzu, Tokoname, Echizen, earthenware (chūsei haji), Koryo ceramics and Chinese white, celadon, and brown wares (Suzuki 1998a, 1998b). From the Andō residence, luxury ceramics included Seto and celadon incense burners and temmoku bowls. Glass beads found in the Andō headquarters are found in many northern trade sites. Suzuki notes that abundant abalone shells and hearths that may have been used for boiling them are found in the Shimokita region of Aomori Prefecture. Abalone shell may have been traded from Tosaminato to the central parts of Japan (1998a:24).

Port Area was tested by an excavation in 2000 (Locality 4: Sakakibara 2001). This area is located on the northwest side of the peninsula on the Fore Lagoon side. The shoreline was artificially faced with gravel for a distance of about two hundred metres and there are posts from a wharf, as well as round logs, perhaps for hauling up small boats. In the gravel layer, Chinese celadon, Suzu, and Seto wares dating mostly from the first half of the 15th century were recovered.

The internal plan of Tosaminato comes from a proto type established in Kyōto and repeated in Kamakura and smaller communities such as Kusado Sengen-chō – a shrine in the extreme north, elite residences in the northern areas, commercial and factory areas further to the south and major

23

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 39

24

The archaeology of medieval towns: case studies from Japan and Europe

east-west and north-south roads (Uno 1994:122).

Tosaminato was superceded by the Katsuyama Kan, built by the Muto Clan on the Watarishima Peninsula of Hokkaidō. It flourished from the second half of the 15th century to the end of the 16th century. Consisting of buildings on three flat areas on a slope, divided by a ditch and surrounded by a stockade, it yielded large quantities of Chinese ceramics, as well as many tea containers and tea bowls (Miura 1994). In the late 16th century Kakizaki Yoshihiro, the local lord who controlled the southern tip of the Oshima Peninsula, Hokkaidō, accepted the expanding authority of the military ruler Hideyoshi. In 1599 he aligned with the Tokugawa, and assumed the family name of Matsumae. The Matsumae clan was granted control of the trade in marine products and other resources of the northern territory, which was called Ezo (Totman 2000: 219).

Discussion

Two medieval trading communities with different political economies provide glimpse of life at the centre and periphery of town life in medieval Japan. Each is embedded in a local context built up through long spans of history. In addition to having a common Japanese cultural background they are linked by the circulation of Chinese goods and diverse relations to a central power in Kyōto. Trade routes extending from them to the north and south and to mainland Asia show dramatically that the Japanese Islands consisted of a series of interlocking open systems.

Sakai, the major port for the central power in Kyōto, had several sectors. Domestic and foreign trade were important. The tribute missions to Ming were important financial projects for Sakai merchants, who were financiers. High quality casting and printing are evidence of a skilled, intensive, manufacturing centre. A distinctive elite culture can also be seen. In Tosaminato the Andō clan, an indigenous elite group. lived in a rectangular compound with retainers and attached craftsmen. Resources from the northern maritime regions were the major commodities and the site has a clearly demarcated mercantile component.

Acknowledgements

I offer hearty thanks to Dr Simon Kaner and the Sainsbury Institute for a stimulating conference and the enthusiastic exchange of information and ideas. Mr. Morimura Ken’ichi of the Sakai City Board of Education, and Prof. Ono Masatoshi of the Japanese National Museum of History have been particularly helpful in making information available to me. I also thank the members of the Shiura Board of Education, Aomori, for making available many sources concerning Tosaminato.

References

Amino, Y. 1994. Chūsei no Nihon kai kōtsū (Interaction in the Japan Sea in the Middle Ages). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed.) Chūsei toshi Tosaminato to Andō shi (The medieval city of Tosaminato and the Andō clan). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 16-36. Amino, Y., Ishii, S. and Fukuda, T. (eds.) 1994. Yomigaeru chūsei (The rise of the medieval period). Tokyo, Heibonsha.

Aomori Ken Shiura Mura Kyōiku Iinkai 1996. Tosaminato iseki: Shiura mura dai ichiji, dai niji, hakkutsu chōsa hōkokusho (The Tosaminato site: report of first and second excavations by Shiura village). Shiura, Aomori Ken, Shiura Mura Kyōiku Iinkai.

Aomori Ken Kyōiku Iinkai 1998a. Tosaminato iseki II: Aomori ken dai niji hakkutsu chōsa gaihō (The Tosaminato site II: preliminary report of the second excavation by Aomori prefecture). Aomori, Aomori Ken Kyōiku Iinkai.

Aomori Ken Kyōiku Iinkai 1998b. Tosaminato iseki III: dai 74ji, 75ji hakkutsu chōsa gaihō (The Tosaminato site III: preliminary report of the 74th and 75th excavations). Aomori, Aomori Ken Kyōiku Iinkai.

Atwell, W. 2002. Time, money, and the weather: Ming China and the “Great Depression” of the mid fifteenth century. Journal of Asian Studies 61(1): 83-113.

Collcutt, M. 1981. Five mountains: the Rinzai Zen monastic institution in medieval Japan. Cambridge, Harvard University Press. Collcutt, M. 1983. Muromachi history. Kodansha encyclopedia of Japan. Volume XX: 165-177. Tokyo, Kodansha.

Fukuoka Style 1998. Tokushu: Hakata to Sakai (Special issue: Hakata and Sakai). Fukuoka Style 7.

Information Section, Bureau of Cultural Properties (Korea) 1984. Artifacts from the Sinan sea bottom (in Korean). Seoul, Bureau of Cultural Properties, Republic of Korea.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 40

Japanese medieval trading towns: Sakai and Tosaminato

Ishii, S. 2000. Chūsei toshi ron no kadai (Issues of medieval urban centres). Chūsei Toshi Kenkyū 7: 9-31.

Itō, K. 1998. Ōuchi shi no Nichi Min bōeki to Sakai (The Japan Ming trade of the Ōuchi clan and Sakai). Istoria 16: 47-69.

Kamei, M. 1995. Nisō no bōeki kankei no tenkai. (The evolution of trade relations between Japan and Song China). In: Iwanami Kōza Nihon Tsushi Henshūkai (ed.) Nihon Tsūshi 6 (An Overview of Japanese History 6). Tokyo, Iwanami Shoten: 109-140.

Kamei, M. 2000. Bōeki tōjiki (Trade ceramics). In: Sakazume, H. (ed.) Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology). Tokyo, Yuzankaku: 57-60.

Kanazawa, Y. 2004. Porutugaru tsūji no mita keisei ki no cha kaiseki (Kaiseki cuisine in tea ceremony in its formative period as observed by a Portuguese translator). Idemitsu Bijutsukan Kanpō 126: 54-63. Kiyonaga, K. (ed.) 1990. Zusetsu: Ōsaka-fu no rekishi (Illustrated history of Ōsaka). Tokyo, Kawade Shobo Shinsha.

Kokuritsu Rekishi Minzoku Hakubutsukan 1994. Chūsei toshi Tosaminato to Andō shi (The medieval town of Tosaminato and the Andō clan). Tokyo, Shin Jimbutsu Oraisha. Kokuritsu Rekishi Minzoku Hakubutsukan 1998a. Tōjiki no bunka shi (The cultural history of ceramics). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan.

Kokuritsu Rekishi Minzoku Hakubutsukan 1998b. Maboroshi no chūsei toshi Tosaminato: Umi kara mita kita no chūsei (The phantom medieval urban centre of Tosaminato: the medieval period of the north seen from the sea). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan.

Kurihara, B. 2000. Umareta sen (Buried coins). In: Sakazume H. (ed.) Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology). Tokyo, Yuzankaku: 139-141.

Maekawa, K. 1991. Toshi kōkogaku no kenkyū: chūsei kara kinsei e no hatten (The study of urban archaeology: development from the middle ages to the modern period). Tokyo, Kashiwa Shobo. Maekawa, K. 1997. Chūsei kangō shūraku to sōgamae: kōkogaku kara mita chūsei goki shūraku no ruikei to hensen (Medieval moated villages and castle outer enclosures: late medieval village types and change seen through archaeology). Nihon Shi Kenkyū 420: 54-78. Maekawa, K. 1998. Chūsei sonraku to toshi: chiiki shakai ni okeru shisonka to toshi no seiritsu (Medieval villages and cities: regional village nucleation and the formation of cities). Kōkogaku Kenkyū 45(2): 41-60. Maekawa, K. 2000. Chūsei toshi iseki chōsa no shiten (Viewpoint on the investigation of medieval urban sites). In: Sakazume, H. (ed.) Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology). Tokyo, Yuzankaku: 90-93.

Maekawa, K. 2001. Chūsei shūraku iseki ni okeru toshi sei ni tsuite (Concerning the urban character of medieval sites). Kōkogaku Jānaru 478: 2-6.

Matsushita, M. 2000. Chūsei shi kenkyū to kōkogaku (The study of medieval history and archaeology). I n: Sakazume, H. (ed.) Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology). Tokyo, Yuzankaku: 83-86

Miura, K. 1994. Andō shi taitō izen no Tsugaru, Hokkaidō (Tsugaru and Hokkaidō before the rise of the Andō clan). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed.) Chūsei toshi Tosaminato to Andō shi (The medieval city of Tosaminato and the Andō clan). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 137-145.

Morimura, K. 1998. Sakai ni okeru bōeki tōji (Trade ceramics from Sakai). In: Nihon Kōkogaku Kyōkai (ed.) Nihon kōkogaku kyōkai 1998 nendo taikai, kenkyū happyō yōshi (Summary of papers, Japan Archaeological Association 1998 meetings). Tokyo, Nihon Kōkogaku Kyōkai.

Morris, V. D. 1977. Sakai from shōen to port city. In: Hall, J. W. and Toyoda, T. (eds.) Japan in the Muromachi age. Berkeley and London, University of California Press: 145-158.

Morris, V. D. 1981. The city of Sakai and urban autonomy. In Elison, G. and Smith, B. (eds.) Warlords, artists and commoners: Japan in the sixteenth century. Honolulu. University of Hawaii Press: 7-22.

Nagai, K. 1994. Chūsei no shutsudo sen: shutsudo sen no chōsa to bunrui (Excavated medieval coins: investigation and classification). Amagasaki, Hyōgo Ken Maizōsen Chōsakai.

Ono, M. (ed.) 2001. Zukai: Nihon no chūsei iseki (Japanese medieval sites illustrated). Tokyo, Tokyo Daigaku Shuppankai.

Saitō, T. 1994. Toshi kenkyū kara mita Tosaminato no toshi kōzō (The urban structure of Tosaminato seen from the study of urban centres). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed.) Chūsei toshi Tosaminato to Andō shi (The medieval town of Tosaminato and the Andō clan). Tokyo, Shin Jimbutsu Oraisha: 158-169. Sakai Shi Hakubutsukan 1993. Hakata to Sakai (Hakata and Sakai). Sakai, Sakai Shi Hakubutsukan.

Sakai Shi Kyōiku Iinkai 1984. Sakai kangō toshi iseki hakkutsu chōsa hōkoku: Ichi no Machi Higashi 4 chō, SKT 19 chiten (Report of archaeological excavations of the Sakai moated city: Ichi no Machi Higashi 4 chō, SKT 19 Locality). Sakai, Sakai Shi Kyōiku Iinkai.

Sakai Shi Kyōiku Iinkai 1997. Sakai shi bunkazai chōsa gaiyō hōkoku (Report of investigation of cultural properties of Sakai city) No. 161. Sakai, Sakai Shi Kyōiku Iinkai.

Sakakibara, S. 1996. Tosaminato iseki no hakkutsu chōsa (Excavation of the Tosaminato site). Chūsei Toshi Kenkyū 3: 1 53-169.

Sakakibara, S. 2001. Aomori ken Tosaminato iseki no chōsa (Investigation of the Tosaminato site, Aomori Prefecture). Kōkogaku Jānaru 472: 27-29.

Sakazume, H. (ed.) 2000 Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology. Tokyo, Yuzankaku.

25

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 41 3

26

The archaeology of medieval towns: case studies from Japan and Europe

Satō, M. 1994. Kodai kokka to Nihon kai, kita Nihon (The Ancient State and the Japan Sea and northern Japan). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed) Chūsei toshi Tosaminato to Andō shi (The medieval city of Tosaminato and the Andō clan). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 64-95. Senboku Kōko Shiryōkan 1996. Jiyū toshi: Sakai (Free city: Sakai). Sakai, Semboku Kōko Shiryōkan.

Sen, S. 1998. The Japanese way of tea: from its origins in China to Sen Rikyū (translated by Dixon Morris). Honolulu, University of Hawaii Press.

Senda, Y. 1994. Tosaminato, Fukushima-jō no chōsa (Research at Tosaminato and Fukushima Castle). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed) Chūsei toshi Tosaminato to Andoō shi (The medieval city of Tosaminato and the Andō clan). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 37-61. Shimatani, K. 1998. Chūsei no mumonsen to sono seibun sosei (Coins without Iinscriptions (plain coins) and their composition). Kikan Kōkogaku 62: 71-74.

Shiura Mura Kyōiku Iinkai and Toyama Daigaku Jinbungakubu Kōkogaku Kenkyūshitsu 1998. Tosaminato iseki: dai 77ji hakkutsu chōsa hōkokusho (The Tosaminato site: report of excavation no. 77). Shiura, Shiura Mura Kyōiku Iinkai, Fukuyama Daigaku Jinbungakubu Kōkogaku Kenkyūshitsu.

Sleusser, D. 2004. The transformation of tea practice in sixteenth century Japan. In: Pitelka, M. (ed.) Japanese tea culture: art, history, and practice. London, Routledge Curzon: 39-59.

Suzuki, K. 1998a. Tosaminato shutsudo no doki, tōjiki (Pottery and porcelain from the Tosaminato site). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed) Maboroshi no chūsei toshi Tosaminato: umi kara mita kita no chūsei (The phantom medieval city of Tosaminato: the medieval period of the north as seen from the sea). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 36-37. Suzuki, K. 1998b. Chūsei Tosaminato iseki to hoppō bunka (The medieval site of Tosaminato and northern culture). Kōkogaku Jānaru 436: 21-24.

Suzuki, K. 1999. Shutsudo senka no kenkyū (The dynamics of money circulation in 14th to 18th century Japan). Tokyo, Tokyo Daigaku Shuppan Kai. Suzuki, K. 2002. Zeni no kōkogaku (The archaeology of money). Tokyo, Yoshikawa Kobunkan.

Suzuki, S. 2000. Kōkogaku kara mita Nichi-Min bōeki (Japan-Ming Trade as seen from archaeology). In: Sakazume, H. (ed.) Chūsei kōkogaku e no shōtai (Invitation to medieval archaeology). Tokyo: Yuzankaku: 145-148. Totman, C. 2000. A History of Japan. Malden MA, Blackwell Publishers.

Tsuchiyama, T. 1996. Sakai shutsudo tōjiki sosei ni tsuite – SKT 39 chiten no chōsa o chūshin ni (The ceramic assemblage from Sakai, with particular reference to those

coming from SKT 39). In: Toki Shi Mino Tōjiki Rekishi Kan (ed.) Tokubetsu ten: Sakai shū no yakimono (The ceramics of the populace of Sakai). Toki, Mino Tōjiki Rekishi Kan: 40-43.

Tsuzuki, S. 1994. Chūsei toshi: Sakai (Sakai, medieval urban centre). Chūsei Toshi Kenkyū 1: 79-118.

Uno, T. 1994. Nihon kai ni miru chūsei no seisan to ryūtsū (Production and exchange in the Sea of Japan Region). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed.) Chūsei toshi Tosaminato to Andō shi (The medieval city of Tosaminato and the Andō clan). Sakura, Kokuritsu Rekishi Minzoku Hakubutsukan: 97-136.

Wakita, H. 1999. Ports, markets and medieval urbanism in the Osaka region. In: McClain, J. L. and Wakita, O. (eds.) Osaka: the merchants’ capital of early modern Japan. Ithaca, Cornell University Press: 22-43.

Watsky, A. 2004. Commerce, politics, and tea: the career of Imai Sōkyū (1520-1593). In: Pitelka, M. (ed.) Japanese tea culture: art, history, and practice. London, Routledge Curzon: 18-38.

Yamakawa, H. 1998. Chūsei shūraku no ronri: kōchi kaihatsu to shūraku keikan (The formation of villages in rural Japan: rural settlements and the expansion of arable lands). Kōkogaku Kenkyū 45(2): 104-123. Yamamoto, N. 2003. Juniseiki zengo tōjiki kara mita Mottaimatsu iseki no hyōka (An assessment of the Mottaimatsu site seen from ceramics dating to around the 12th century). Kodaigaku 55(3): 39-56.

Yamamura, K. 1990. The growth of commerce in medieval Japan. In Yamamura, K. (ed.) The Cambridge history of Japan. Cambridge, Cambridge University Press: 344-395.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 42 4

Ichijōdani: the archaeology of a Japanese medieval castle town

Ichijōdani: the archaeology of a Japanese medieval castle town Ono Masatoshi Summary

Excavations at Ichijōdani on the Japan Sea side of central Honshū since the 1970s were instrumental in establishing medieval archaeology in Japan. The remains of the town which developed around the headquarters of the Asakura clan now form the focus of a historical park nestling in the hills south of the Asuwa River to the north of modern-day Fukui city. The Asakura controlled the region for a century prior to their defeat in 1573 by Oda Nobunaga, one of the unifiers of Japan at the end of the medieval period. The excavations at Ichijōdani provided direct evidence for the first time for the layout and development of a complete medieval Japanese castle town.

This paper is an abridged translation of a book-length survey of the archaeology of Ichijōdani by Ono Masatoshi, who was involved in the excavations since the 1970s. It comprises four parts. The first section introduces the site of Ichijōdani and its planned urban structure. Ono suggests that Ichijōdani is representative of one particular style of castle town. A model is presented that encapsulates the relative positioning of the main residences and headquarters of the Asakura clan who built Ichijōdani, the temples and cemeteries, and the residential areas of the ordinary people and commercial areas. The second part provides an account of medieval Japanese concepts of spatial organisation within towns, and how this spatial organisation expressed the social relationships of the inhabitants. Following on from a summary of the elements of urban structure, Ono suggests that the concepts of hare (auspicious, ceremonial, special) and ke (inauspicious, ordinary and everyday), familiar from studies of traditional Japanese culture to the present day, provide a useful structuring principle behind the spatial organisation of the site. Ono draws on historical records to show how the formal visit by the shogun to the Asakura lords expressed the relationship between ruler (shogun) and vassal (lord). Ono then shows how the archaeological remains of the site show how this relationship was expressed in spatial and material ways. The main indicator

This chapter is an adapted summary of Sengoku jōkamachi no kōgaku: Ichijōdani kara no messēji by Ono Masatoshi, published by Kodansha in 1997, drafted by Richard Pearson and edited by Oscar Wrenn.

of status present in the artefact record is pottery. Imported wares from China were very highly valued, and the significance placed on antique items is seen in the way especially treasured vessels were repaired through time. The third part of the paper considers the evidence for the structuring of daily life for the ordinary townsfolk of Ichijōdani. Town houses were grouped into neighbourhoods. Excavations have revealed the plans of the houses, how they sit in their plots and how the plots were aligned with the planned grid of streets. Craft production within the town included metalworking, beadmaking and dyeing, and there were facilities for merchandising. The detailed archaeological work has revealed much about the occupational history of the site, with variation in burial over time, as well as intriguing evidence for the effects of house rental and redevelopment. The fourth section of the paper presents a detailed analysis of the ceramic assemblages from Ichijōdani. Ono discusses how the remains of pottery vessels allow archaeologists to reconstruct patterns of consumption of this important commodity within the urban context. This discussion ranges from the organisation of pottery production, variation in pottery usage between towns and villages, the significance of Chinese ceramics, and changes in the relative value of imported and domestic wares. This section ends with a brief consideration of the major regions of ceramic production in medieval Japan.

Ono’s paper provides a detailed overview of the archaeology of Ichijōdani, one of Japan’s most important deserted medieval urban centres, which flourished in the later part of the medieval period, vanishing as a consequence of the moves towards the reunification of Japan under a series of paramount military warlords that was eventually to see the Tokugawa shoguns win control of the whole country at the start of the seventeenth century. Ono shows how combining the interpretation of the spatial organisation of urban forms, detailed artefact analysis and historical evidence allows us to understand the occupational histories of medieval urban forms.

27

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 43 4

28

The archaeology of medieval towns: case studies from Japan and Europe

Figure 1: Ichijōdani and surrounding area

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 44 4

Ichijōdani: the archaeology of a Japanese medieval castle town

Introduction: the Ichijōdani site In the 1970s, excavations of Ichijōdani, in present-day Fukui Prefecture, began to reveal the life of a castle town that had lain ruined and abandoned under rice paddies and forest for four hundred years. The entire community was destroyed and burned in an attack by Oda Nobunaga on 14 August, 1573, and the way of life brought to the region by the Asakura clan one hundred years before abruptly ended.

The site then lay undisturbed until the 1970s, when archaeologists were able to examine Ichijōdani’s fortifications and gates, the residences of its rulers and samurai, its temples and its merchants’ quarters, its graves, and large assemblages of ceramics. Thirty years of excavation yielded a complete picture of a medieval town, one of only very few castle towns to have been fully investigated. Originally from a region in western Honshū, the Asakura clan was one of a group of rulers now termed sengoku daimyō, warring states lords. It was at the start of the Ōnin Wars, in 1467, that the clan established a power base here, in the region known then as Echizen (Matsubara 1978). The second generation became the region’s rulers in 1483, and they developed Ichijōdani as their capital. Despite local warfare, the next two generations consolidated their control, and Ichijōdani attained its peak of affluence and power in the fifth generation only to fall to Oda Nobunaga’s expansionary forces.

The town of Ichijōdani Topography

Ichijōdani lies in a narrow valley, about three kilometres long, connected to the Asuwa valley, about a kilometre inland from where it flows out onto the broad Echizen Plain (Figure 1). Without excavations, it would be hard to believe that it was the site of Echizen’s capital. Situated south of the River Asuwa, the valley is protected on three sides by ridges about four hundred metres high; and the valley was also fortified at points by ditches ten metres wide, earth ramparts five metres high, and had gates at the two ends of the valley, 1800 metres apart. At the entrance to the valley, on the north bank of the River Asuwa, the Jōganji mountain fortress was built. Within this defended area, various residential zones flanked a central road. In the central section of the valley, at its widest point, was the Asakura family residence and headquarters, the yakata, in addition to samurai dwellings; and to the south were more samurai dwellings and a merchant’s quarter. The names

of houses and streets are known from Meiji and Tokugawa documents.

Both spatially and functionally, the yakata formed the nucleus of the town. Defended by a ditch, the site comprised the main buildings, a garden, and the Kannon Hill; and outside this central area were two ‘palaces’ for close relatives in the Asakura clan, and two, level, open areas that served as riding grounds and assembly places (baba). At the highest point in this area was the grave plot of Takakage, the first lord of Ichijōdani.

Various groups of people were evidently allowed to live between the lower and upper gates (Figure 2). Through this area ran the central main road, intersected by roads that crossed the Ichijōdani river, named after their bridges, as well as by other, smaller roads. The roads divided the area into districts, which included Akabuchi, Okumano and Yoshinomoto, comprising townspeople’s houses, or machiya, samurai dwellings, and temples. Groups of buildings were of consistent style and scale. It is thought that the main street comprised small-scale shops which were adjoined to each other, with display shelves called misedana and entrances on the side that ran parallel to the roof. On the bank opposite the yakata, in the Kawaidon, Hirai, and Saitō districts, was a row of large samurai houses that were protected from view by a rampart and wall. Less excavation was undertaken on the east side of the Ichijōdani River, so details are fewer, but it is thought that there was an area of samurai housing there too.

Beyond the gates, routes connected the town to the rest of the world both by road and water, allowing an influx of goods and information. The River Asuwa led to the Sea of Japan, whereas the Mino Road led to the northbound Hokkoku Road, meaning that Ichijōdani was connected to the major trading routes of the Hokkoku Road and the Japan Sea Road. Goods were marshalled at the commercial districts of Abaga, Tōgō and Bishamon, and historical sources mention that, in 1498, Chinese traders brought their goods by these routes, with Abaga being noted for Chinese goods. Lying at the margins of urban civilisation, Abaga and other areas outside the lower gate were commercial centres and also probably comprised burial grounds; excavations have recovered the remains of cremations and deposits of bone. These outlying areas were outside of the Asakura clan’s power, and powerful individuals from outside Ichijōdani built residences there. Based on Ichijōdani, Figure 3 shows what is termed a Model 1 castle town. In basic terms, a town of this kind was laid out in concentric circles: the yakata was surrounded by open spaces (baba), and then an area inside the castle gates under the control of the daimyō, which was enclosed by a moat, rampart, or mountains. Outside the gates and moat was a ring of both

29

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 45

30

The archaeology of medieval towns: case studies from Japan and Europe

Figure 2: The structure of Ichijōdani

commercial zones and housing. The areas outside the moat became progressively further removed from the daimyō’s authority. The site of Nanjōchō Sōmayama Castle is another classic example of Model 1.

A different kind of castle town, which we can call Model 2, has been noted by Senda (1990; see Figure 4). This has a cluster of several walled residential areas surrounded by an outer commercial zone of markets, with the yakata and other elite residences protected in a central walled area.

Well-known sites in Aomori-ken prefecture, such as Namioka and Nejō, are laid out in this pattern, though Senda notes that this model can also be found in sites in Kyūshū. The names of a number of the inner enclosures are known for both Nejō castle, and these names correspond to important military commanders. That each enclosure was named after these commanders suggests that they also would have had control over their enclosure, and the residential system within.

Although requiring further study, we can theorise that these two models of castle towns imply different systems of governance and power. Model 1 implies comparatively unified and integrated power focused in the single yakata, whereas in Model 2 each walled residential area acts as a separate microcosm, with a general alliance between the power-holding military commanders.

The urban layout of Ichijōdani

Ichijōdani shows evidence of planned machi or chō (urban zones) divided by streets. The topography of the valley and the River

Ichijōdani affected the layout of Ichijōdani’s quarters, meaning that, unlike Kyōto, towns could not be planned in clear rectangular patterns (see Figures 5 and 6), and so many small districts were created. The clearest divisions are in Kawaidon, Hirai and Saitō, where lay the big samurai houses, facing the yakata. Within the main gates, on the west side of the river, lanes leading from the main valley road defined these small districts, at intervals of about 120 metres. The plan seems to have been to lay down the main road running northsouth along the valley and then the east-west roads from the sides of the valley to the river.

The remains of some fifty roads and lanes have been excavated. They can be divided into four 'grades': Grade I, the widest, is 7·8 to eight metres wide; Grade II five to six; Grade III is three metres wide; and Grade IV less than two1. The roads of Grades I to III have one or two rows of paving stones. Some were covered with gravel from side to side but some were only covered in the middle, another difference in grade. The main north-south road and the east-west roads linking the two banks of the river were Grade I while the branch roads were of lower grade and narrower. In excavations and by studying place names, archaeologists found that there were four groups of commercial districts, known as machiba. The first was the area within the upper and

The units of measurement used in Ichijōdani were shaku (30.3cm), and ken. The measurement for ken changed depending on location and time, but in Ichijōdani the Kyōto ken was used (1.97m) 1

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 46 4

Ichijōdani: the archaeology of a Japanese medieval castle town

Figure 3: Model 1 – Concentric structure of a castle town

lower gates. Another, Abaga, was immediately outside the lower gate, with another, Higashi Shinmachi and Nishi Shinmachi together, outside the upper gate. The last lay deeper in the valley, at Jōkyōji. Abaga in particular was a key centre for the movement of goods and people in and out of Ichijōdani, having permanent shops and a market. It existed at the boundary between the town and the outside world, with routes out along the River Asuwa and the Mino Road. In a diary from a visitor to Ichijōdani, Abaga is referred to as the ‘entrance to Ichijōdani’, and a ‘home to Chinese settlers’, as well as having a temple called Ganzōji. The presence of this temple designated Abaga as a kugaisho, a place separated from the social strata and secular power of the outside world. Some excavations searching for details about Abaga were carried out at the foot of the mountains, at the Nishiyama Kōshōji Temple (Ichijōdani 1992). In other excavations along the current prefectural road archaeologists found stone paved latrines, a road 2.4 metres wide, and ceramics similar to the wares found within the gates but at a fairly low density. A well was also discovered nearby. Given the relative paucity of excavations in the area, documentary evidence is useful in showing that Abaga was an important market centre: at the Shinju-an, a subsidiary shrine of the Daitokuji temple in Kyōto, a document of 1543 regarding contributions from Ichijōdani to temples in Kyōto describes the economic activity of Abaga, demonstrating its importance as a distribution and trading centre (Fukui Ken 1986). In particular, rice and cloth would have been the main goods in circulation. Jōkyōji was a centre for making whetstones for swords. The date of this activity is not clear from existing documents: such whetstones are found in the Sea of Japan region dating to the 15th century but there has been no archaeological investigation carried out at Jōkyōji, so the chronology is unclear. In the area of Higashi Shinmachi and Nishi Shinmachi, there are several place names, including one called nabeya, 'house of cauldrons', but at present there is no proof that metal cauldrons were made there.

Figure 4: Model 2 – Concentric clustered structure of a castle town

Establishing a town

Although Ichijōdani has thus far been described as a large castle town, it underwent a number of changes in its history. Excavations have shown that there a number of layers of buildings built on top of each other, and these can be split into periods, I to III (Figure. 7). Period I encompasses the 14th and first half of the 15th century, before Ichijōdani became a town and before it was the seat of Asakura power. At this time there were several large residences (yashiki) in the middle of the site, and though details of the construction of these residences are unclear, early genealogies of the Asakura clan show sons named Abaga, Tōgō, and Nishijima, names of districts of the town. This shows that residences of the first period were probably connected closely to the main line of the Asakura.

Stone mortuary monuments and Buddha figurines give us an idea of habitation and population increase in Ichijōdani from the inscriptions on mortuary figures such as the posthumous name (hōmyō) of the deceased and details of their age, sex and religious sect (Asakura 1975). From around 1460 there is the appearance of these monuments, though only a small number of children’s graves. This appears to be the beginning of more general habitation of Ichijōdani, and is the establishment of Ichijōdani as the capital of Echizen (period II); though this initial town was subsequently destroyed in a fire.

From the first decade of the 16th century there is a marked increase in mortuary monuments, showing population increase. This is when Ichijōdani was rebuilt after the fire, and began to take on the appearance and functions of a town, and is the beginning of period III. Within this period the layout of the machi changed a number of times, but its initial construction is

31

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 47

32

The archaeology of medieval towns: case studies from Japan and Europe

Figure 5: Town organisation and view of the Kawaiden, Hirai and Saitō districts

shown as IIIa, its layout before its destruction as IIIc, and the intervening period as IIIb.

Concepts of spatial organisation The Asakura yakata

Each of the yakata’s sides is about ninety metres from the earthen ramparts and about 120 metres from the ditch’s outer corners. The enclosure is about 6500 square metres. Figure 8 shows the site with a thirty-metre grid imposed. The 120 metre dimensions of the yakata, match the size of the mansions of the shogun and the Hosokawa kanrei, deputy to the shogun, in Kyōto, and we can surmise that the yakata would have been built with an awareness of these other grand residences.

Within the yakata there are fifteen buildings and four wells. Most of the buildings are clustered in the north-west four sections (Figure 8), with the three sections to the west and south being large open spaces. In the far corner of the site, by Buildings 8 and 9 and facing the mountain, is a group of standing rocks from a garden. Building 11 is the bathhouse, and in Building 12 are large hearth areas, which indicate that it was a kitchen. From the particular layout of Buildings 4 and 16, we can assume that they were stables. Buildings 14 and 6 have large, raised, flat stones spread out, indicating that they were storehouses. These were measured in the traditional shaku and sun measurements: group A was 6 shaku 2 sun, group B was 6 shaku 2.5 sun

2

The buildings can be divided into two groups, A or B, according to the distance between the placement of the pillars set into the foundation stones2, meaning that these groups of buildings appear to be from different periods (see dividing line, Figure 8). It is thought that the houses of Group B were built for the visit of the shogun, Ashikaga Yoshiaki. The yakata was aligned between the mountain ridge to the east and the main or ceremonial gate at the west, with the north gate functioning as the inner gate. Originally, the yakata was a rectangle of ninety metres by just sixty, but when the houses of Group B were built it was extended by digging into the mountainside at the rear.

Hare and ke

Hare, auspicious, and ke, inauspicious, are words often used in Japanese folklore studies. These terms appeared in the Late Heian and Kamakura periods in the context of ceremonies, and documents from these times often referred to clear, bright weather as hare, and rainy or dull conditions as ke. Hare and ke, respectively, connoted 'extraordinary' and 'special' types of activity, and 'ordinary’ or 'everyday' life. For spatial or architectural contexts, hare referred to spaces where ceremonial functions were held, including formal meetings with guests; while ke was used in reference to places associated with daily living.

If we consider the yakata’s orientation, and draw a straight line eastward through the main gate and Building 7, we find a smallish garden and a group of relatively small buildings on the north. This is the part of the site with ke functions. Here, for instance, are the bath, kitchen, and stable, Buildings 11, 12,

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 48 4

Ichijōdani: the archaeology of a Japanese medieval castle town

Figure 6: Town organisation of the Akabuchi, Okumano and Yoshimoto areas of Ichijōdani

and 16 respectively. Clearly, these buildings pertained to the yakata’s domestic or 'inner' function. To the south was a garden, an open space or hiroba and some larger buildings, places for receiving visitors and performing related, more ceremonial, functions. Here we can see a basic division between ke, north, and hare, south.

One way of investigating the use of space in the yakata is through historical sources which record the ceremonies and meetings that took place there, in particular the shogun’s official visits to the daimyō, known as onari. These formal visits, demonstrating the shogun's prestige, were well established custom by the mid-14th century and they continued into the Tokugawa period (1600-1868). The visits could entail recurrent or rarely performed ceremonies, including gift-giving, using different parts of the compound, including buildings and temples throughout the area. They lend themselves to structural analysis of the different uses of space and the mentalities behind those uses. The visits to Ichijōdani included ceremonies in the main building and assembly hall, of which the layout is shown in Figure 9. By looking at the use of space, we can get a better sense of the structure of the ritual programme at the ruler's mansion. The first part of the visit took place in the main building, shuden, and in the open space before it, comprising a formal banquet, with toasting, to confirm the relations of lord and vassal. The style of the space was Japanese. The second part of the ceremony took place in the assembly hall, kaisho, and garden, in a more enclosed space and out of public view, and involved a feast with entertainment. The hall was decorated with Chinese arts.

By looking at the use of these spaces we can see how the Asakura residence was laid out to display relations of power. The first part of the ritual programme, in the Japanese space, was used to confirm the relationship between peoples, in the case of the onari between shogun and daimyō, whereas the latter part of the ritual programme in the inner, enclosed space decorated with Chinese arts, was used to level social differences. These structural relations, of Japanese-Chinese, public-private, traditional-modern, on show and out of view, show the further subdivision of hare space, and its importance in the confirmation and subversion of power relations.

Ceramics and status: connections to Kamakura

Just as wealth and power were symbolised through the lay-out of the Asakura residence, power relations were also displayed through much smaller objects, such as pots. The same goes for many other medieval sites of similar rank in Japan. It appears that the Ashikaga period elite placed particularly high value on Chinese objects. In sites such as the Asakura yakata, and the headquarters of the Uesugi clan in Niigata, a great variety of Chinese ceramics of high value have been identified. Most of the ceramics that were found came from the 13th and 14th centuries, the period of the Song and Yuan dynasties in China. Most of these wares were also found in the Sin’an ship, which sank in 1323 on its way to Japan. The oldest of these pots to be found in the Asakura yakata are 12th century Ding white ware and Koryo celadon ware, which were also carried in the Sin’an ship as antiques (Kankoku Bunka 1977).

The fact that these objects mostly came from the 13th and 14th centuries shows the value of objects associated with the

33

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 49

34

The archaeology of medieval towns: case studies from Japan and Europe

Figure 7: Deaths and inferred population trends determined from dated mortuary stone pagodas and Buddha figures

Kamakura period. In the 16th century, such antiques were much prized as items particularly associated with this time, and is due to the development of a value system for ceramics by the Ashikaga shogunate in the 15th century. They were sought after by the Ashikaga shoguns and referred to in literature and paintings.

With the shift of power from Kamakura back to Kyōto during the Ashikaga period, the shoguns and samurai continued to look back to the Kamakura period as their ideal, to cultivate their identity through its values. The model for the medieval capital of Kamakura had shifted away from the traditional and somewhat effete culture of Kyōto to the more militaristic rough-and-ready values associated with Toba, Shirakawa, and Hiraizumi. Even in the Edo period, samurai families were keen to show that their lineages went back to Kamakura times.

This difference between Kamakura and Kyōto in this period was also shown through ceramics: if we compare the 13th and 14th century ceramics of Kamakura, Kyōto, and Hakata, we find an unusual abundance of large meiping vases in Kamakura. It is clear that the samurai of Kamakura and its region (Tōgoku, the eastern country) had a strong liking for these vases and bottles.

Repaired ceramics

One interesting aspect of pottery found in Ichijōdani is the presence of items repaired with metal clamps. Only four such pieces were recovered in Ichijōdani, in the Asakura yakata, the Saigōji temple, and in a doctor’s residence, and they were all items of high value. These were repaired by drilling small holes in a number of places either side of the crack, and inserting

Figure 8: Spatial features of the Asakura headquarters

small copper or iron staples. For most broken ceramics, the item would have been discarded, or repaired using lacquer as a replacement for glue. A tea bowl, named Bakōhan, in the Tokyo National Museum, also shows this kind of repair. Made in Yuwangshan, it is Southern Song. It was given to Taira no Shigemori by the Chinese Zen priest, Fo Zhao, in return for a gift of alluvial gold sand. Later, it belonged to Ashikaga Yoshimasa. Since it was cracked, Ashikaga requested a replacement from the Ming but, since no replacement could be found, they repaired it with a small clamp. When we look at the artefacts repaired with clamps in Ichijōdani, the story of the Bakōhan gives us a sense of the sort of value that the repaired items in Ichijōdani would have had. The fact that there were artefacts of the 12th century at the time when Ichijōdani burned, antiques already 400 or 500 years old, has interesting parallels with the value we place on antiques from the period of Ichijōdani, which are of the same age.

In 14th century towns there has been no appearance, excluding a few examples in the port of Hakata, of Yuan blue and white ware. In total there are less than 20 sites where they have been unearthed, and most of these are 15th and 16th century castle towns or temples of the daimyō in sites like Ichijōdani, or Shitokuji in Echigo. Of the ninety-eight lands mentioned by the Yuan traveler, Wang Dayuan, twenty-six are noted as trading in blue & white wares. They included countries in Africa, western Asia and south-east Asia, and were valued particularly in Muslim countries. The blue & white wares were also very popular in the region of Ryūkyū, in castle (gusuku) sites such as Shuri and Nakijin, where resided powerful rulers. This ware was not popular at the time in Japan, where yingqing

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 50

Ichijōdani: the archaeology of a Japanese medieval castle town

Figure 9: Reconstructed floor plan of assembly hall and main residence, Asakura headquarters, Ichijōdani

and celadon were preferred; but that changed after the 15th century, when large pieces were used to decorate audience halls (kaisho), matching also with the fashion at the time for using blue ink on white backgrounds. This ware seems to have connoted affluence and prosperity, and a piece from the Fusshoji Temple in the port site of Ishikawa, a blue & white meiping plum vase of the early 16th century shows signs of metal clamps, indicating that it had been repaired and was, thus, highly valued (Ishikawa Ken 1984). It was, indeed, a very rare piece by then.

Kawarake

Figure 10 gives percentages of a range of wares from various sites in Ichijōdani. The data allow comparison of the different social strata using places such as the residences of daimyō or samurai and temples. Surprisingly, the assemblages vary little, with people of differing social status using similar sets of crockery in daily life4. However, the distribution of low-fired earthenware, kawarake, does vary. This ware was used not for everyday activities but for special events. One notable difference is the abundance of fine art ceramics in temple sites.

4

The proportion of kawarake at the yakata is very high. Where it is fifty to sixty percent across the site as a whole, it is more than ninety percent at the yakata, the same proportion as from sites in Kyōto. A lot of kawarake was used for hare occasions: several tens of thousands of food vessels were used at big events such as the Shogun Ashikaga Yoshiaki’s official visit, when 180 people attended the ceremonies. The remains of the crockery found in a large pile near the rear gate included 250 kilogrammes of kawarake sherds.

Kawarake was first made in antiquity (c. 794-1185), but it was in medieval times that it was associated with hare rites, to be used just once, for food and wine, then discarded. The kawarake was used with unlacquered plain wooden trays, oshiki, which were also discarded.

There are texts referring to the use of these dishes, for example description of a Muromachi Bakufu occasion in which 884 wooden trays and 3375 kawarake were ordered, costing 57300 coins (mon) (Satō 1988). For the honzen regular dinner at the ceremony of rebuilding Nara’s Tonomine Tanzan Shrine, in 1520, the participants used 1496 wooden trays bought for 5400 coins and 435 pieces of Kyōto ware which cost 900, not counting 400 coins to carry the pottery from Kyōto. From the Ōshū Hiraizumi Yanagi Gosho, archaeologists excavated ten tons of kawarake.

35

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 51

36

The archaeology of medieval towns: case studies from Japan and Europe

long, narrow plots which faced onto the street, with a frontage of only six to nine metres wide. In some cases, several D houses shared a well (Figure 6). Group D houses were occupied not only by merchants and craftsmen, but also clerks and officials.

The machiya were positioned as series of houses forming neighbourhoods, with the houses facing the road in rows. The plots were long and narrow, partitioned by ditches, with a basic pattern of dwelling, toilet and well on each. Varying somewhat in size, the houses were built with pillars on stone bases, and the roofing was often wooden planks weighted down with stones. Houses also seemed to have similar external dimensions to others in the same district, with the location of the well and the length of the plot following the same layout.

Figure 10: Composition of ceramic assemblages from various areas and excavations of Ichijōdani

The discussion of valuable Chinese ceramics, repaired pottery, and kawarake gives us a sense of these items as markers of wealth and status. The value ascribed to items associated with the Kamakura period, and the stories of items such as the Bakōhan, give us a sense of how important these items would have been, but they also feed into wider power relations in medieval Japan. The ascription of value to Kamakura period objects, the careful repairing of antiques, and the large use of kawarake on special occasions mirrored that of the shogunate in Kyōto. This adoption reflected the acceptance of the legitimacy of the power of the shogunate, as the various daimyō used these objects as status symbols which had value because of their connection to the shogunate.

Daily life in Ichijōdani The business machiya

Lords and samurai were not the only inhabitants in castle towns, and others would have resided there, such as those in the commercial machiya (town houses). Four kinds of houses were found in Ichijōdani. Group A dwellings were found in the yakata. Set in extensive grounds enclosed by a moat and rampart, they were entered through a gate, and had ancillary buildings for different functions. Group B houses, including samurai houses and temples, were also enclosed by earth ramparts and had formal gates. A third group of houses, C, were smaller and more modestly appointed, and though they too were surrounded by ramparts the enclosures were not very large, though the houses themselves were nevertheless spacious. Houses of the fourth Group, D, were only separated by a fence or ditch, and were less than sixty square metres. They occupied

The houses varied in their particular characteristics and room dimensions, but each was entered from the front. In most cases, cooking was done on a brazier (hibachi) made of local stone5, but the kitchens varied considerably, including some which contained hearths set into the floor. Many aspects of the plots’ interior arrangement remain unclear, but usually the houses had an earth floor area and two to three rooms, as well as a separate work area. The smallest house frontage in Ichijōdani was 4·3 metres, and the largest 16·4 metres. The most common width of frontage (about 30%) was about 6 to 6.5 metres, with the next most common lengths around 5, and 9.5 metres. There are three ranks of house by size, around 45 square metres, 75 to 85 square metres and 115 to 125.

Craft production

By looking at some examples of excavated houses in Ichijōdani we can get a sense of the daily life and commerce of the residents of machiya. No. 8 is an example of a bead maker’s house, and was second house to the north of the intersection with the road that led to Saigōji temple. The beads were for pious citizens’ Buddhist rosaries (juzu), commerce that could only have occurred in a district with a number of temples. The bead maker’s plot was thirty-seven square metres, and was taken up mostly by a big square house with 5.6 metre sides, though this was smaller than others in the neighbourhood. The house had a well, and inside a pounded earth floor6. An adjacent latrine also had an earth floor. The work took place on a larger area of earth floor facing the street, evidenced by fragments of rock crystal at various stages of bead production that were lying there. There was also evidence of grinding with stone abraders and perforators. The house was amalgamated with an adjacent plot to make a 5

This would have been stones from Shakudani, Shakudani ishi

These floor areas were measured in tsubo, approximately 3.3 square metres. 6

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 52 4

Ichijōdani: the archaeology of a Japanese medieval castle town

standard-sized plot at a later date, but the household’s craft from this time is unclear.

House No. 12 belonged to a metal caster. The plot was about twice the size of the bead maker’s, with a long house occupying most of the area, and an open space at the rear. The main building was divided left and right and had a small Kyōto-style earth floor. Facing the road was a work space with a fireplace, and here were found several small clay crucibles and clay tuyères. Behind the house was an earth floor with a well and remains of a kitchen nearby, and to the left was a space of with two or three rooms.

House No. 5, south of the road leading to Saigōji temple, also belonged to a metal caster, with crucibles and copper slag. Its plot was particularly large, around 128 square metres, and contained a square building with 8·5 metre sides. In the wide, open space at the back were a well and toilet. The work area appears to have been towards the rear of the house, and to the left and right were two stone paved hearths. This household seems to have been more affluent than that belonging to the other caster.

House No. 62 is the remains of an indigo dyer’s establishment. Dating from not long before the destruction of 1573, it is in a section originally intended for samurai (Figure 5). It was one of five similar houses in a row on the west side of the main road, with a plot was similar to that of House No. 12, and containing a rectangular building. Along with three other houses the location of the building was separated from the road by about 1.2 metres. The rear area was fenced around, with a garden and a toilet set up against the fencing, and a garden and path stretched along the left of the property. In the building at the back there were large jars, four on the left and eight on the right, buried into the ground. Near the entrance was a well, and there also seemed to be a drying area. House No. 54, another dyer’s establishment (Figure 11), also had large jars buried in the ground. It contained two buildings, rare for a plot of this rectangular shape, with the front building facing the road leading to the town wall’s lower fortified gate, and the back of the lot facing the mountain behind. The plot was irregular, but was approximately 8·3 metres wide and 17 long. By the road was a well with paved area. The front house was reasonably large, with overhanging eaves at the rear, and had two or three rooms on the left side, reconstructed as four, six and eight tatami in size. The middle room had a hearth of river stones and the back room too had a hearth area. At the back of the property was the second, smaller building, containing 15 jars buried in the ground. Along the left side of the house was a narrow passage. No latrine was found within the plot, though it possibly could have been located in the rear space.

Depictions from the period show townspeople selling goods from the front windows of houses, rather than from a separate shop, which raises the question of whether merchants and craftsmen had platforms for displaying their wares, known as misedana. It is likely that they did, and we can imagine that there were petty vendors who lived facing the main street, though there is still no archaeological proof of this. Within the town gates were individuals supplying the daimyō but there were no general vendors in large shops.

Rented housing

The Rakuchū rakugai painted folding screens from late 16th century Kyōto give us a sense of machi from this period. It shows house frontages and plots of fairly consistent width and length. They have entrances that run parallel to the ridge of the roof (hirairi), and most are of the katadoma type, with a ground level earth passageway running along the inside of one side of the building. Some of the houses are of the nakadoma type, with this passageway running through the middle. In the Rakuchū rakugai painting, houses are illustrated with communal wells and toilets, which became an abiding image of houses from this period, but according to historical documents houses would not have had separate areas with these facilities, and artists took shortcuts when illustrating the houses. Ichijōdani seems to conform to these general rules in the layout of residential plots, though some examples of plots with communal facilities have also been discovered.

An example of this can be found in Figure 6. Separated from the main road by a small lane is a group of houses: 37-40, 101-103 and 106 (highlighted in grey). In the area between these houses are the remains of two wells and a latrine, but no such facilities exist within the houses. From the remains of an earthen rampart and deeper excavation in the area, it seems likely the area from houses 25-27, to 102, 103, 106 was once a single samurai residence. This was then redeveloped, with three machiya (25-27) created with separate facilities, and the space behind turned into a shared-facility space for eight machiya. The ditches of these yielded worked deer antler, indicating that some of that house’s occupants were craftsmen, as antler was used for knife handles, weapons and also gaming pieces. The sharing of such important facilities would imply that these houses were rented. It is not difficult to dig additional wells, and this area in particular has a high water table, so communal wells must indicate a particular system of property management, and social status for the tenants. Either the original owner of the land redeveloped the area into rented housing, or the land was passed on to separate owners. There are many interesting aspects of rental housing. Given that the residential area within the gates was controlled by the

37

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 53

38

The archaeology of medieval towns: case studies from Japan and Europe

and was clearly occupied by a craftsman. From excavations, part of a wheel for wrapping the cord of a chalk line was found. This, along with a carpenter's square, was a key tool of the trade, and given that in the medieval times ordinary people would not have owned carpentry tools, we can assume that this was a carpenter’s residence. It was a large residence, and although it did not face onto the main street, we can assume this was a carpenter of high regard.

We can see from these houses that there were three strata of non-samurai classes within the castle gates. Doctors would have certainly been had a social status close to that of this highly regarded carpenter, but as some B group residences show, they were equal to samurai.

Organisation of occupations in residential zones

How were various crafts organised in the town and how was that reflected in its layout? How much can archaeological research clarify the interesting issues about the distribution of people? The relation between production and commerce seems impossible to specify, but did specialists in particular crafts live in the same part of town, forming cohesive social units?

Figure 11: Postulated dyeing facility showing buried large jars, Merchant House No. 54

Asakura clan, who granted property rights, it is interesting that properties such as the one described above were redeveloped and let out. There are other examples of changing border boundaries that suggest that they were not scrupulous with their control in this area. It appears, then, that large properties were split up for different functions and large neighbourhoods divided into smaller units.

Group C houses

So far, we have discussed the aspects of the smallest group of houses, group D, but those dwellings that were larger than these, but not samurai plots, fit into the category of group C houses, and would have also contained craftsmen. House C77 gives a clear sense of the plot of a C group dwelling (Fig. 5). It was surrounded by an earthen wall with a gate, and a single house on the plot surrounded by ample space. There was a well adjacent and a latrine at the back, and in front of the house was a small rectangular garden (tsubo niwa) surrounded by a bamboo fence, a feature that is not seen with a Group D house. The house was slightly removed from a large road with a big temple and samurai houses behind it. House C45 is another good example of this type of dwelling,

The picture from Ichijōdani’s archaeology of the relationship between artefacts and sites is quite detailed: fourteen dyers’ houses can be recognised from large buried clay jars; six casters’ and blacksmiths’ houses can be identified by tuyères; and of the total of 67 craftsmens’ houses in Area D, 20 percent were dyers’ and 9 percent belonged to casters or blacksmiths, which seem to be a high proportion.

Few historical documents describe the distribution of occupations in medieval commercial zones, but important comparative evidence in cases dated 1572, 1587-90, 1668 and 1694 gives the ten most frequent occupations and corroborates the findings at Ichijōdani. For the modern period (1568-1867), it is known that various craft occupations were nucleated in urban quarters such as the blacksmiths’ (kajiya chō) and dyers’ (konya chō). Compared to this, occupations do not seem to have been located in the same areas in Ichijōdani. Though occupations did not seem to be gathered in specific areas, some, such as metal casting, appear to have been strongly associated with the samurai.

Production and distribution of ceramics Ceramics comprise some 90 percent of the materials excavated from medieval sites, the rest being metal objects, various wooden artefacts such as tubs and discarded lacquer utensils, and stone articles such as querns. Ceramics are more durable than wood or paper and cannot be recycled. Analysing their fabrics, we can

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 54 4

Ichijōdani: the archaeology of a Japanese medieval castle town

study the distribution of pottery and pursue the implications for both social life and exchange. In Ichijōdani the different scales of exchange give a particularly interesting picture, with Chinese ceramics traded the world over, Echizen wares circulated in eastern Japan, and those of Tokoname and Bizen throughout western Japan.

Ichijōdani’s assemblage

A great proportion of Ichijōdani’s ceramic assemblage was made locally, either earthenware from Fukube-machi within the town, or Echizen-yaki from within Echizen at Nyū-gun, and comprised up to some 80 percent of the total. From outside the domain came small amounts of other ceramics such as Seto-Mino ware; and from overseas came pottery from China and Korea and very rare examples of Thai earthenware, which comprised the remainder of the assemblage at Ichijōdani.

By analysing the functional type of pots found in Ichijōdani, we can see that most of the utility types were local: the earthenware and Echizen wares were for cooking, storage, lamps and keeping cosmetics. These local wares also were also used by craftsmen, such as Echizen jars for dyeing indigo. Other non-ceramic utility objects were also made locally: heating was provided through braziers and stone footwarmers and were made from the local Shakudani stone, and metal containers for cooking were also made nearby.

Imported wares covered all remaining functions, and were used for tea ceremonies, flowers, incense and as containers for other luxury items, as well as for food and drink. In particular Seto and Mino provided iron-glazed wares such as temmoku tea bowls, as well as ash glazed plates. Seto and Mino kilns also made suribachi (ridged vegetable grating bowls) in the same period, but these were not imported into Ichijōdani, and only local Echizen suribachi were used. The various centres of production for these ceramics had a number of competitive and complementary relationships, depending on their demand and use. One example of this is a set of tea service ceramics comprising a large water jar, caddy and bowl of Chinese ware, and a very similar but cheaper set in Seto and Mino wares modelled on Chinese types. Chinese pottery was greatly respected, scarce and expensive, so the second set was produced as a cheaper copy of the original set.

The complementary aspect of these areas of production was defined by their usage. As noted above, except for small oil lamps, kawarake plates made in local kilns were used as luxuries and for religious occasions, while Chinese and Seto plates and, in some cases, lacquer bowls were used for everyday purposes. Thus, each production centre produced wares that were used for different contexts.

The exact logic behind the separate uses of utensils at this time is unclear, but a 16th century document, Sannai kuketsu, describes the different kinds of utensils, wooden, earthenware, lacquer, celadon, and white ware bowls and their uses. It explains that the wooden ware was preferred for sake and other luxury uses and that imported celadons and white ware were used by the ruling class, who did not use lacquer bowls for ordinary meals. While the Sannai kuketsu claims that different ceramics reflected social differences, in Ichijōdani this was not the case: as mentioned previously, there is no difference in the proportion of ceramics used by different social strata. To judge from their distribution across the site, the machiya households too used ceramics for ordinary purposes.

Quantifying consumption

We may be able to compare the use of ceramics at different social levels and in different spaces by studying the distribution of used and broken items. In theory, we could also examine rates of consumption of materials such as wood, charcoal and vegetables but these do not usually leave archaeological residues, meaning that many aspects of consumption are difficult to measure. There is also the problem that virtually no medieval site has been completely excavated because the sites are very large. Ichijōdani covers 30 hectares within the gates and there were also extensive areas outside them. By extrapolating from the excavations thus far, where 12 potsherds have been discovered per square metre, there would be 3,600,000 potsherds in the entire site. By taking this extrapolation further and using percentages of ceramics from the whole site, there would be 210,000 sherds of Chinese ceramics, as well as 960,000 sherds of Echizen ceramics (27 percent), and, of this, there would be 13,000 suribachi (grating bowl) vessels. In the Sin’an ship, 28 metres long and 7 metres wide, with a seven chambered hold, there were 20,000 ceramic vessels, 28 tons of copper coins, and incense. Ichijōdani had about ten times the amount of Chinese ceramics in the ship. If one site yielded this quantity, one can imagine the scale of the trade, carried not only by ships from Fujian but also by Japan’s nineteen tribute missions from 1401 to 1547.

The Echizen wares were produced near Ichijōdani, also the capital of the region, so it would have been one of the main sites of consumption. Some details of Echizen kilns provide information on Japanese potting. Kilns at Matsuo in Taira Village, Fukui, supplied Ichijōdani with Echizen wares. The Dakenodan Kiln Group Kiln No. 1 serves as an example of its dimensions and capacity: it was 25 metres long and six metres wide and was of tunnel type; with the front half for medium size jars with wide necks (kame) and narrow necks

39

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 55

40

The archaeology of medieval towns: case studies from Japan and Europe

(tsubo), about 60 of each; while the rear held 1200 suribachi. As this was the capacity for one firing, the total number of suribachi recovered from Ichijōdani could have been produced in about ten firings.

Echizen ware was also distributed far up the west coast to Nejō at Hachinohe, and to the south it was distributed from Shimane Prefecture all the way to Kyūshū. Echizen kilns seem to have supplied ceramics to about a quarter the area of the Japanese archipelago in this period, and its wares can be found in small villages as well as those of average size.

By way of comparison, we can examine the consumption of pottery at Kamakura, of a period earlier than Ichijōdani. There are fewer detailed tabulations of material from Kamakura, but one recent excavation of storehouses near Kamakura Station provides some comparison. Overall the excavations yielded ten potsherds per square metre, slightly less than in Ichijōdani. The area and population of Kamakura was also much larger than Ichijōdani: in total it covered 226 hectares, comprising 131 hectares of samurai dwellings and 95 hectares of machiya; and the population was somewhere between 70,000 and 110,000. Over such a large area we can calculate that there would be 22,600,000 objects, of which roughly 10 percent are trade ceramics. That is about a hundred times the amount of the Sin’an cargo.

One interesting piece of information regarding Kamakura ceramics has been revealed through the work of Ishii Susumu, who has found a document of 1252 stating that, as part of a prohibition on the sale of sake decreed by the bakufu, a survey was done and 30,724 liquor storage jars were discovered (Ono 1990). Each house was subsequently only permitted a single liquor storage jar, and was made to destroy the rest. We can assume that, as these were ceramics for liquor, they were Atsumi or Tokoname ware from the Tōkai region.

Conclusion

The archaeological excavations at Ichijōdani provide us with an incredibly broad look at the development, structure and daily life of a medieval Japanese town. The preserved nature of the site has meant that analysis is not limited to single aspects of the town, but can be drawn together to provide a more holistic understanding as to how it would have functioned in the 15th and 16th centuries. By looking at a number of districts within and without the city gates, we can see the various ways in which power was spatially determined, from the highly structured layout of the Asakura yakata and its distinctions and subdivisions of hare and ke, to the kugaisho of Abaga, where the power of the Asakura clan was null.

Analysis of ceramics from the Ichijōdani site also shows the interesting ways in which value was ascribed to objects, especially antiques from the Kamakura period, and how this was part of a broader legitimisation of the Ashikaga shogunate in Kyōto. In contrast to these uses of older ceramics, town-wide analysis of the proportions of simple earthenware and imported ceramics show that, with the exception of the use of kawarake, the proportion of ceramics used did not change according to social status or location within Ichijōdani.

Looking at the function of different ceramics also gives us an insight into the use of wares from different production centres: locally made items provide the mainstay of daily life, with imported wares from outside the region and across East Asia serving ceremonial and luxury functions. Even within this rigid prescription of production centre and use, however, is space for competition, as the imitation Chinese tea sets show.

Much has been learnt from the excavations at Ichijōdani that could be of use to medieval sites and medieval history more generally, not least in depicting the lives of those who lived in the sites we excavate. It is not enough to merely list the facts of an archaeological site; archaeological materials are not silent objects, and in many ways tell stories about people from different times. It is the archaeologist's role to listen, and use these objects to paint vivid images of lives in medieval towns.

One part of the process of uncovering Ichijōdani that has been particularly interesting has the been the use of an interdisciplinary approach. From the beginning of excavations in the 1970s researchers in architectural history, literary history, archaeology and other areas gathered and worked to excavate and understand an entire town, showing that in order to study a large urban space such as Ichijōdani a multifaceted point of view is vital.

Ichijōdani has also shown the importance of the relationship between society and history. Ichijōdani has now become a historical park, allowing the creation of museum exhibitions and the maintenance of ruins which, in archaeology, are the most useful tools in connecting history and society. The process of creating this moved the focus of excavation away from ‘preserving the past’, to ‘making the past talk through preservation’.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 56

Ichijōdani: the archaeology of a Japanese medieval castle town

References

Asakura Shi Iseki Chōsa Kenkyūsho 1975. Ichijōdani sekizō ibutsu chōsa hōkokusho I (Report of Ichijōdani stone relics I). Fukui, Fukui Ken Kyōiku Iinkai.

Fukui Ken 1986. Fukui Ken shi, shiryō hen 2, chūsei (History of Fukui prefecture volume 2, medieval). Fukui, Fukui Ken. Ichijōdani Asakura Shi Iseki Shiryōkan 1992. Ichijōdani Asakura shi iseki kendœ Shino’o Katsuyama sen kairyœ kœji ni tomonau jizen chœsa hōkoku (Report of investigations prior to improvement construction on the prefectural Shino’o Katsuyama road). Fukui, Ichijōdani Asakura Shi Iseki Shiryōkan.

Ishikawa Ken Maizō Bunkazai Sentā 1984. Fusshōji iseki (Fusshōji archaeological site). Ishikawa, Ishikawa Kyōiku Iinkai.

Kankoku Bunka Kōhōbu 1977. Shinnan kaitei iseki (Sin’an seafloor site). Seoul, Kankoku Bunkazai Kanri Kyoku. Matsubara, N. 1978. Ichijōshiro Tashiro izen no Asakura shi ni tsuite (Concerning the Asakura Clan before their move to Ichijō Castle). Fukui, Fukui Ken Chiikishi Kenkyūkai. Ono, M. 1990. Byōdōdake no tanigama atogun no chōsa gaiyō (Overview of surveys into Byōdōdake Tanigama archaeological sites).

Satō, K. 1988. Bunken shiryō ni mieru chūsei no inshokki no shiyō to shoyū ni tsuite (Looking at the medieval use and ownership of tableware through historical documents). Fukui, Asakura Shi Iseki Shiryōkan.

Senda, Y. 1990. Sengokuki jōkaku, jōkamachi no kōzō to chiikisei (Structure and regional characters of castle enclosures and castle towns of the Sengoku period). Istoria 129.

41

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 57

42

The archaeology of medieval towns: case studies from Japan and Europe

The establishment and transformation of Japan’s medieval capital, Kamakura Oka Yōichirō Summary

In addition to their commercial importance, a significant aspect of medieval towns and cities is their political and administrative role. This paper by Oka Yōichirō treats the early medieval eastern political centre of eastern Japan, Kamakura, in the most significant period of its history, from 1180 to 1225. It deals with some of the structural and architectural changes which occurred as it came to rival the power of Kyōto, the ancient capital from the 8th century. Oka compares texts from the Azuma Kagami, the diary chronicle of the shogunate, the feudal military government, which was located in Kamakura, written after 1266, with the archaeological record.

Kamakura was a centre of some importance before the 12th century. In 1073 the clan shrine of the Minamoto clan, the Tsurugaoka Hachimangū, was founded. In the 1170s Minamoto Yoritomo, descendant of the Minamoto clan, exiled from Kyōto, married into the powerful local Hōjō family, regents for the shoguns, who were often replaced and sometimes were very young. Yoritomo arrived in Kamakura in 1180 and set about to establish a regional civil government from heterogeneous groups of local powers as a kind of independent kingdom in the east, known in Japan as Tōgoku (Shinoda 1960:62-63, Mass 1982:125). Gradually Kamakura changed from the centre of a local polity to an integral part of the Japanese polity centred in Kyōto. By 1185, the shogunate was in a transitional form, neither purely feudal nor yet a national institution (Shinoda 1960:137, Hurst 1982:5). Three regents (shikken) ruled on behalf of the shogun, from 1199 to 1242. Hōjō Tokimasa became the regent for Yoritomo’s young heir in 1199. The second regent was Hōjō Yoshitoki (r. 1205 to 1224), He was followed by Hōjō Yasutoki who ruled from 1224 to 1242. By the time of the third regent, a kind of consultative board of councilors (hyōjōshū) was part of the Kamakura administration (Goble 1982: 168). In terms of medieval archaeology, this paper shows the strong connection between the results of excavations and historical texts. Oka has been successful in finding texts which link historical figures and urban topography and architecture,

This chapter is an adapted summary of Kamakura no henyō by Oka Yōichirō, published by Kochi Shoia in 2006, drafted by Richard Pearson and edited by Oscar Wrenn.

showing how the urban plan and architectural layout changed within a few decades. Oka is particularly concerned with accounting for architectural patterns in Kamakura, which adapted the classical style of buildings (shindenzukuri) of the Heian period (794-1185), particularly in Kyōto, in which elite residences, set in gardens near large ponds, were composed of pavilions linked by raised, roofed, corridors. The eastern Japanese divergence from this archetype, is considered an adaptation to different socio-political requisites and well as cultural differences.

Some of the texts from the Azuma Kagami have been cited from Shinoda (1960). Those that we could not find in Shinoda’s translations were translated by Kazue Pearson from the original found at http://www.asahi-net.or.jp/~hd1t-situ/azuma.html

Introduction

Archaeology has always played an important role in exploring the medieval capital of Japan, Kamakura. It has been particularly useful not only in examining the structure of the city and reconstructing the landscape of the capital but also in determining the use of each site and its transformation through different periods. This has enabled us to trace the structural development of the inner city in specific, shorter phases of its history. The History of Kamakura City (Kamakura Shishi Hensan Iinkai, 1959) may have been almost totally dependent on literary resources, but archaeology has changed the study of the city. However, while archaeology has produced such substantial academic progress, it has also left a crucial problem unresolved: different scholars assess the dating differently (Mabuchi 1997). This has often been a serious hindrance to collating the results of various excavations in the city for analysing the characteristics of a particular site and its transformation. Of course, this situation will have to be resolved by, for example, cross-dating the results from Kamakura with those of other parts of Japan.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 58

The establishment and transformation of Japan’s medieval capital, Kamakura

Moreover, despite the valuable information produced by archaeology, various other sources and studies are equally important. Scholars need to use them more positively because they may supplement archaeological findings both by increasing the precision of study and by broadening the scope of research. Fortunately, compared to other medieval cities in Japan, Kamakura has ample documentary sources, including the Azuma Kagami which have already been studied closely. In the present paper, I use both archaeological and literary sources in order to demonstrate the efficacy of combining different kinds of sources, and to examine the transformation of the city and its background. The history of Kamakura had several important turning points. For the purpose of the present discussion, I focus on two: Minamoto no Yoritomo’s entry to Kamakura in 1180 and the large-scale redevelopment of the city in 1225.

Minamoto no Yoritomo’s entry

Twelfth month, twelfth day, 1180. Sky clear, winds gentle. At 9 pm, Yoritomo observed the ceremony of removal to a new house. The work on the building, which is situated in Ōkura Township, had begun in the Tenth Month under the supervision of Ōba Kageyoshi. At the scheduled hour his Lordship left the residence of the provisional vice-governor of Kazusa Hirotsune and entered the new residence. His Lordship was dressed in a ceremonial robe. His mount was chestnut coloured and mottled. Attending his Lordship at his front was Wada Yoshimori … After his Lordship entered his bed chamber, the men gathered in the samurai-dokoro, which had eighteen bays (ken) of 1.8 metres each (a huge spacious building), and with Wada Yoshimori seated in the centre, took their places in two rows facing one another. There were 311 men in attendance. His Lordship’s immediate vassals have also been constructing new residences there. These houses will increase the grandeur of the entire east and will make Kamakura an important centre. This had been a secluded place with but a few fixed abodes, frequented only by fishermen and rustics. But lanes and streets are being laid, and names are being assigned to hamlets and villages. Moreover, the tiled roofs of houses stand in a line and the doors of gates creak under the eaves. (Shinoda 1960: 199-200).

That is one of the most famous depictions of Yoritomo’s entrance to Kamakura. Compiling the Azuma Kagami in the late 13th and early 14th century, it was evidently intended to treat the event as the very beginning of the capital city’s development.

The second period also coincided with the many changes happening to the Kamakura shogunate. Previous studies consider the year of 1225, when a series of policies were enacted, as a turning point that affected the character of the shogunate. For example, Gomi (1990) considers that the new regime was established in the tenth month of 1225 as a benevolent government, crystalised through the transmission

of power from Hōjō Masako who died in that year to her nephew, Yasutoki. He maintains that, after this event, the shogunal initiative was transferred from the Kamakura shogun himself to the regency. In the 12th month was established the Shogun’s Guard, which played an important role in the shogunal system (Matsuo 1992). In the same year, the Shogun’s Palace was relocated from Ōkura to Wakamiya Ōji1. Matsuo (1992) argues that it was after that move that the shogunate adopted measures from Kyōto such as tax systems, and administrative policies including the system of ho. Kamakura now changed form from a mere military base to ‘the capital of the samurai class’, matching the political and commercial centre, Heiankyō (Matsuo 1992).

Amino Yoshihiko (1993), by focusing on the term, chi, used to refer to urban areas in medieval Japan, concludes that Kamakura was established as the medieval capital in this period. Considering that Kamakura was already referred to as chi in the Karoku period (1225-1227), he holds that its urban structure was consolidated by the introduction of various systems from Kyōto in the reign of Yasutoki (1224-1242).

Regarding Yoritomo’s entry, it has been commonly held that Kamakura was a remote place in the early 12th century, as described in The History of Kamakura City (Kamakura Shishi Hensan Iinkai (1959). However, Ishii Susumu (1988) and Noguchi Minoru (1993) reject this view. As for the description in the text, Noguchi argues that the conditions before Yoritomo entered were understated in order to emphasise the significance of the event. According to Matsuo (1992), the Kamakura banyaku (regional administrative office) had already been established in 1223, when, two years later, the lower-class retainers of fifteen local fiefs (kuni) submitted to Kamakura authority in the first year of Karoku.

Regarding the new policies, Matsuo and Amino agree that Kamakura was established as a capital city during Yasutoki’s reign by introducing Kyōto’s urban system. However, they argue that Kamakura had already been the centre of the local administration before then, virtually governing the south Kantō region. Samurai serving Yoritomo and other people concerned were gathered there, so the city must already have been somewhat ‘urbanised’. This period of Kamakura’s is underestimated merely because the city had neither yet reached the peak of its prosperity nor developed the urban structure of Kyōto, regarded as the standard for a capital. During the Middle Ages, many urban spaces of various sizes developed all over the Japanese archipelago. Few, however, have documentary evidence either to show whether they were structurally urbanised or to describe any aspect of them in detail. For want of such documentation and thanks to the 3

Large avenue

43

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 59

44

The archaeology of medieval towns: case studies from Japan and Europe

Map of Kamakura showing key locations and roads from the early Kamakura period

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 60

The establishment and transformation of Japan’s medieval capital, Kamakura

tendency to regard Kyōto’s structure as the only standard for judging, even a large city like Hiraizumi could be assessed as undeveloped.

We will now explore how Kamakura was transformed into a capital city from the time before Yoritomo’s arrival in 1180 to 1225, the first year of Karoku (Oka 1998). What we know of the city between these dates allows us to assess its development as a medieval capital and to understand the basis for it.

Before Yoritomo

Kamakura had been a regional centre from ancient times, as witness the remains of the local government office. Many details of Kamakura’s landscape in the periods before the Middle Ages are unknown, but Ishii (1988) and Noguchi (1993) have pieced together a conspectus of the city and its surroundings just prior to Yoritomo’s entry.

According to their reconstruction, several main roads had met at Kamakura (Figure 1). A coastal road ran toward the Miura Peninsula, whilst another led vertically to Musashi and another led horizontally to Mutsu’ura. These were connected vertically by the Ima Kōji and Musashi Ōji roads. Each of these roads still exist in Kamakura. A recent study discovered that there was another major route which diverged from the Mutsu’ura road to cross the mountains down to Kotsubo (Oka 2004a). At this stage, then, all of today’s main roads in the city were completed except Wakamiya Ōji, which was later, and Komachi Ōji, of unknown date. The local government office faced Ima Kōji. Many temples and shrines founded before the Kamakura era stood near the old roads mentioned above, including the Goryō Shrine, the Amanawa-shinmeisha Shrine, the Iwayadō Shrine, the Egara Shrine, the Sugimotodera Temple, and the Yu’i Wakamiya Shrine. Fukushima (1999) states that the Shōgenji Temple stood on the east-west road running along the foot of a mountain and the east of the Iwayadō Shrine. He argues that it was of the Tendai School, whose mother institution was in Kyōto; and his view that it was the region’s Tendai centre that was critical in establishing Kamakura’s religious importance. All of those temples and shrines were already established by the time that Yoritomo arrived. It is generally known that Kamakura had been the base of the Minamoto family in the east since Minamoto no Yoriyoshi gained power in 1051 through his close relation to Taira no Naokata. By analogy with Kyōto and Hiraizumi, Irumada (1991) and Noguchi (1993) assume that Yoshitomo lived amidst Minamoto’s family and vassals in the area around the Jufukuji Temple. Not all the samurai lived here, however. For instance, the remains of samurai residences have been found to the south

of the Sugimotodera Temple, a site which predates the Kamakura shogunate (Sugimoto 2002). Dating to the second period, from the last quarter of the 12th century and the first of the 13th, it includes Kamakura’s thirty largest buildings, measuring 190 square metres each. They demonstrate the prosperity of the temple and its vicinity. These remains are attributed to the Wada family, part of the Miura clan (Oka 1995). The Miuras were long related to this area: the first son of Miura Yoshiaki, Sugimoto Yoshimune, was also associated with a place called Sugimoto (Ishii 1986). Moreover, the Mukai Egara site in Egara, the north part of the Sugimotodera site (Mukai Egara 1985), is also associated with the Wadas on account of the style and dating of its structural remains, which correspond to the rest of the Sugimotodera site (Oka 1995).

The Wada residence in Egara

Third month, twenty-fifth day, 1213. The residence of Wada Heita Tanenaga, exiled to Kamakura, was located in front of the Egara Shrine, on the eastern side of his Lordship’s palace. Some close members of the samurai entourage desired the property. Today, Lady Gojō no Tsubone, wife of the Grand Chief Yoshimori, mentioned in distress that the property had been endowed by the late shogun and should not be given to anyone else, The property looked appropriate for short term religious use by the Egara Shrine. Would it be possible for it to be given (to her family or others)? This wish was realised and brought incredible pleasure.

Fourth month, second day, 1213. Tanenaga’s property was to be divided into two, to Yukichika and Tadaie. During that time, Yoshimori, who had previously been given the land, removed Kunoya Yajirō, the acting provincial constable. Neither of these people lived on the site. Yoshimori was distressed but he could not plead to the authorities about the issue. To contest this issue would be like a mouse fighting a tiger.

As these texts indicate, Wada Tanenaga had a residence in Egara. The way in which the site was redeveloped after it was forfeited corresponds to the land use indicated by the archaeology of the contemporary Sugimotodera and Mukai Egara sites. We can assume, then, that the Miura clan had lived in Sugimoto and its neighbourhood since the period before the establishment of the Kamakura shogunate.

Fire in a samurai house

Ninth month, fifth day, 1224. Sleet... At midnight, the house of Miura Yoshimura, the former Suruga District manager, at the west gate of the Palace, was destroyed by fire. The fire did not spread to other buildings. During that time, there was some disturbance in the lower town area.

This text seems to confirm that the Miura clan lived in Nishi Mikado on the west of Sugimoto in the Kamakura period. Nor do the Miura seem to be the only samurai family living in Kamakura. Noguchi Minoru (1993) argues that the Kazusa clan too were long settled in this area.

45

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 61

46

The archaeology of medieval towns: case studies from Japan and Europe

Tsunetane encourages Minamoto Yoritomo to move to Kamakura

Ninth month, ninth day, 1180. Adachi Morinaga, returning from his mission to Chiba, reported to his Lordship thus: ‘Shortly after I announced myself at the gate of Chiba Tsunetane’s residence I was invited in. Tsunetane, flanked by his sons Tanemasa and Taneyori, received me in his guest room. Although he listened patiently to what I had to say, he remained silent and appeared to be asleep. But his sons spoke up, almost in unison, saying ‘Yoritomo seeks to arouse the brave to action so that the cruel and avaricious might be overthrown. He has summoned us at the very outset of his venture. Is there any reason to defer our acceptance? We shall forward our letter of acceptance immediately’. Tsunetane contemplated no action except to concur. Tears fill his eyes and he is speechless, for he is grateful to restore the house of Minamoto which has been held in abeyance. Later, as I was being entertained with drink, Tsunetane suggested, ‘The location of his Lordship’s present headquarters is neither strategically sound nor historically important. His Lordship should move his headquarters forthwith to Kamakura in Izu province.’ Tsunetane also declared that he will join Your Lordship with his band of followers (Shinoda 1960: 175-176).

Chiba Tsunetane, who recommended Yoritomo to enter Kamakura in this text, appears in another document as his loyal retainer. If Tsunetane had been well-informed of Kamakura as Yoshitomo’s servant, he and his family may already have been living there (Oka 1995). Like the lifestyle of the samurai residences in Kyōto and Hiraizumi (Noguchi 1993, Irumada 1991), the whole Miura family probably lived in the same area, around the mansion of their patriarch. Thus, Kamakura must have had a number of temples and shrines in this period, with samurai family groups surrounding the Minamoto clan’s Kamakura date.

A home already urban in character

Tenth month, eleventh day, 1180. At 5 a.m., her Ladyship Hōjō Masako arrived in Kamakura, where she was met by Yoritomo’s emissary, Kageyoshi. Her Ladyship had reached Kamakura last night from Aki, in Izu County. However, the date was inauspicious, so she took lodging (probably with samurai or religious people), near the River Inase.

The commoners’ houses along the river may have belonged not to farmers or fishermen but to people who lived by serving or trading with samurai households and religious institutions. Clusters of such houses can be identified in Rokuhara in Kyōto (Takahashi 1998, Noguchi & Yamada 2003, 2004), Hiraizumi, and the remains of the Enjō-ji Temple in Nirayama (Yaegashi & Iimura 1995). Thus, Kamakura shared these characteristics with other contemporary towns. It has often been said that Yoritomo established his seat in Kamakura on account of its strategic position, as described in the text about Tsunetane above. Yet, as a place so important for transport, Kamakura

was well suited as the base and centre of the samurai government (Oka 2004b) and, if the scenes described did take place, it was quite natural that Yoritomo decided to take advantage of what the location offered. Moreover, to enter Kamakura was to publicise widely the legitimacy of Yoritomo’s succession (Matsuo 1993).

From Yoritomo’s to Sanetomo.

Yoritomo eventually entered Kamakura in 1192. The urban topography was quite different from today. First, the main axis of the capital was not Wakamiya Ōji, which ran north and south, but the street running east and west from Hase to Mutsu’ura (Ishii 1988). On this boulevard stood the Palace, around which the residences of the Minamoto clan and vassals may have been located (Irumada 1991). The city had not yet been fully urbanised by that time (Gomi 1992a). A boulevard did already connect the Tsurugaoka Hachimangū shrine and the beach, and the streets around the Palace had already been refurbished; but only the restricted areas around the beach and the River Inase were inhabited.

However, some researchers maintain that, from the early in its development, Kamakura was modelled on the Heian capital, modern-day Kyōto. For example, Ōmiwa (1989a, b) argues that the streets laid were out in a grid pattern, and Mabuchi (1994) proposes early application of urban planning to set Wakamiya Ōji as the main axis in reference to Heiankyō’s Sujaku Ōji. Yamamura (1997) criticises these theories, maintaining that, up to when the palace was relocated to Utsunomiya Zushi, in 1225, the capital’s main axis was the east-west street mentioned above, not Wakamiya Ōji. She also argues that the streets in the middle of Kamakura were derived from streets spontaneously developed out of daily needs . (Yamamura 1997). Despite conflicting points of view, these two assessments of its plan, both offer general pictures of Kamakura’s urban structure.

Three-piece set

Now I examine Kamakura through a reconstruction of its landscape. The sites on which I focus here are what might be called early Kamakura’s ‘three-piece set’, archaeological features that define the period of Yoritomo. It had quite disappeared by the Karoku period (1225), only leaving for the archaeologist lines of large postholes for raised buildings and the V-shaped ditches of moats. The first piece of the set is the rows of large postholes that were identified on the east side of the Ōkura Palace and a series of moats along Nikaido Ōji. One of these rows

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 62

The establishment and transformation of Japan’s medieval capital, Kamakura

supported 25 evenly spaced posts. Each hole was over 1 metre in diameter but the average distance between them is no more than 1.5 metres. The diameter of the wooden posts themselves and the construction above the ground are unknown because no remains of them have been discovered. Yet the excavation report assumes a large, solid construction because of the size of the postholes. The report also states that there may have been wooden plank walls supported by huge posts or large fences (Kamakura Shi 1993). There were no remains of the rammed earth which usually accompanied tile-roofed mud walls; and the distance between the columns was relatively short. These remains are considered to date from the very beginning of the Kamakura period, around 1185. In the archaeological reports identifying the site as the Shogun’s East Palace, the palace is considered to have been surrounded by a fence or palisade. Even if this was not the site of the East Palace, both its location, right next to the Ōkura Palace, and the remains of the large raised post buildings certainly show that the site was occupied by an influential man and his family. Similar remains have been discovered in Minami Mikado, the south of the Ōkura Palace (Kamakura Shi 1998a). This site, also dating back to the first half of the 13th century, is smaller than the east site. Its postholes were sixty to seventy centimetres in diameter and 122 centimetres apart on average. According to Azuma Kagami, Minami Mikado was inhabited by influential vassals from the east, including Hatta Tomoie in 1187 and Hatakeyama Shigetada in 1199, and these dates match that of the remains. Considering the power of the leaders, Hatta and Hatayamake, at the time, we can conclude that this site may have belonged to one of them. These two examples allow us to reconstruct a cityscape around the Ōkura Palace including a public institution (possibly the East Palace) surrounded by wooden walls with large wooden columns or fences and the streets lined with the grand residences of influential vassals. As for the premises of the Ōkura Palace itself, no large-scale research has been conducted yet but we can assume that there were institutions similar to those in the surrounding areas. As yet, however, no remains of that kind have been discovered on the site along Wakamiya Ōji, where, presumably, the Palace was located.

The second piece of early Kamakura’s ‘three-piece set’ is the large raised buildings. They have been found near the Ōkura Palace but I focus here on the most typical case, which was excavated among the remains around the Sugimotodera Temple. These buildings contrast with the major buildings in the Ima Kōji West site in its heyday of the late 13th to early 14th centuries (Ima Kōji 1990). Rather than longer posts set in postholes, the latter have stone pillar bases which supported

short posts. I agree with Kawano (1995), who attributes these remains to the shogunate and their influential vassals. He regards them as typical of that particular period, considering their scale and the fact that most of that evidence has been found around the Ōkura Palace.

The third piece of the set is the V-ditched moat. It is not restricted to a particular period. Few V-ditch moats of the late 13th to early 14th centuries can be matched with features on the former ground surface above them. From Surface III around the Sugimotodera Temple, a V ditch moat two to three metres wide and ninety centimetres deep was dated to about 1200. It is thought to have been buried in the mid 1200s. Another V ditch moat was found from Period I of the Mukai Egara site. It was 3.3 metres wide (its original width estimated as four metres) and 1.5 metres deep. Period I starts at the beginning of the 13th century, and this moat too was filled in in the first half of the 13th century by material from Period II.

It is along Wakamiya Ōji that the V-ditched moats are most clearly discernible. The first period is represented by Surface IV of the site of a residence on Hōjō Komachi belonging to Yasutoki, shogun from 1224 to 1242 and Tokiyori, shogun from 1246 to 1256 (Kamakura Shi 1996). It contains two rows of the moats. The first of these was 2.2 metres wide and 1.1 metres deep but, as fully reconstructed, 3 metres wide and 2 metres deep. The finds date it back to the first quarter of the 13th century or earlier. The second moat could not be completely excavated because its west side runs right under Wakamiya Ōji but it is estimated at more than two metres wide and 1.7 metres deep and it is thought originally to have been as wide as three metres. Yielding early Kamakura artefacts, this moat is the oldest identified on the site so far. Its linear ditches are clearly distinguished from a series of square profiled and timbered moats laid out over the other moat. Those linear ditches were for holding water, while the moats themselves were mainly intended to divide and defend properties (Oka 1995). Unearthed along Wakamiya Ōji was a line of V-ditched moats cut at different times to mark a row of properties off from the boulevard. The same feature can be found along Komachi Ōji, where the first moats had V ditches and then timbered moats were dug (Hara1998, Hara 1999). Thus, big V-ditched moats probably ran continuously along Kamakura’s main streets.

Among the remains around the Ōkura Palace, moats were laid out along the palace’s west wall, parallel to the River Higashi Mikado’s original bank. The opposite bank has been excavated too (Ōkura Bakufu 1990), and we therefore know that the river was nine metres wide. Its name implies that it ran along the east side of the Shogun’s Palace. We can deduce

47

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 63

48

The archaeology of medieval towns: case studies from Japan and Europe

that the moats were probably designed in relation to the river’s original course both from exposures of that course beneath the medieval surfaces and from their horizontal sections. Considering too that, once relocated, the Shogun’s Palace was not provided with moats of the same scale and that their ditches were not V-shaped, we may regard those features as typical of Kamakura at the time.

Fifth month, second day, 1213. Cloudy … Saburō Yoshihide of the Ashikaga encountered Yoshiuji near the bridge in front of the Central Government Office. Yoshihide chased Yoshiuji and caught Yoshiuji’s armoured sleeve. Desperately, Yoshiuji quickly mounted an excellent horse and rode westward from the dry moat. Although his sleeve was torn off, he escaped without falling. Yoshihide wanted to keep fighting but felt so tired and muddy that, after a few minutes, he decided to stay on the east side of the moat. In strength and courage, the warriors were equal. The onlookers got very excited, clapping and clicking their tongues. Yoshihide tried to go around over the bridge in pursuit but the Falcon Handler came to separate them. Another official was injured. Yoshiuji quickly escaped.

The text shows that the early Kamakura mandokoro (executive office of the Kamakura government [bakufu]) also had a moat too. Trenching has revealed only a part of it (Mandokoro 1991). Neither its form nor its development are well understood, and further investigation is expected.

So far, my discussion has been restricted to the sites associated with persons related to the shogunate and the samurai class, but the V-ditched moats have also been found at religious sites. For example, there was another moat at the site of the Tsurugaoka Hachimangū Shrine’s Training Area, Surface IV, around the area of medieval grave fill. This moat was some 3.3 to 5.2 metres wide and two metres deep (Kenshū 1983). It appears to have occupied the eastern edge of the shrine in the late 12th to early 13th centuries. Unfortunately, however, neither the boundaries of the shrine nor the moat’s development are known because the premises of the shrine expanded to the east after Surface IV formed.

Our examples enable us to reconstruct part of Kamakura. The evidence consists of V-shaped ditches and large moats centring on the Shogun’s Palace, large fences or wooden walls along the streets, and large raised post buildings. These remains include shogunal institutions, vassals’ residences and buildings in the premises of the Tsurugaoka Hachimangū Shrine. Thus, the early Kamakura ‘three-piece set’ is associated with the ruling classes. That the elements were found in various sites shows that they were widespread. Moreover, if we consider seriously the continuity of cityscape discussed above, we may conclude that such institutions had existed even before Yoritomo’s entry. That urban scene described above disappeared during the first half of the 13th century. The change coincided with land redistribution and initiation of large building projects for the capital. The shift

can be recognised in comparing Surface IV with Surface III above it.

According to Mabuchi (1992), in the natural ground surface of dark brown clay, the distance between the postholes of buildings constructed in the later 12th century and earlier 13th was 2.1 to 2.3 metres. Later buildings stood on thick layers of domestic and building debris and the intervals between their posts were reduced to two metres. This new layer and its features developed in various locations at once. It required extensive labour and a great deal of soil, and therefore Mabuchi surmises that this phase saw the administrative authority take over direction and management of large-scale city planning from local residents. He also associates the change with the introduction of Kyōto’s system of measurement in the first year of Karoku (1225). Other data disprove Mabuchi’s theory, however. For example, the average distance between the posts of a large building within the remains of the Tōshōji Temple precinct was 2.1 metres (Kamakura Shi 1998b). Presumably, the building was destroyed at the end of the Kamakura shogunate but, to Mabuchi, these measurements would indicate that it was substantially earlier. Nevertheless, his theory for the change that took place almost simultaneously across the city does sound convincing. Further research is needed.

Capital of the ‘East’

Is it appropriate to discuss Kamakura’s features and the changes to them in isolation? We know that the V-ditched moat was fairly common throughout the Japanese islands at the time, as, for example, the V and square-profile ditches of the double moat discovered at the Kusu-Arata site associated with the capital at Fukuharakyō (near Kōbe, in western Japan; Hyōgo Ken 2003; Rekishi Shiryō 2005). Considering that this city too was built during the civil war, we should be able to trace a similar development among Kamakura’s comparable remains.

Now let us return to the Azuma Kagami and examine its references to construction of large buildings and to their builders. Work at this scale was often conducted for a public institution. Examples would be the Shogun’s Palace and the Tsurugaoka Hachimangū Shrine. I believe that, by investigating records of the people in charge of constructing large buildings, we can trace the origin of various techniques, for even if they themselves were not necessarily the architects or builders, their subordinates may have done so. Ōba Kageyoshi often appears in literature about the construction of the Shogun’s Palace and the Tsurugaoka Hachimangū Shrine. He was one of the most influential

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 64

The establishment and transformation of Japan’s medieval capital, Kamakura

vassals in Tōgoku (Eastern Japan). That he came from a family of lords in the Kamakura area was considered more important than his ability to accomplish his tasks. Nuki (1981) points out that it is not for his talent as an architect but for his ability to provide labour and materials that Ōba is mentioned so frequently. Accordingly, I shall now examine how the Wakamiya Ōji was built.

Building projects

Fifth month, thirteenth day, 1181. The matter of timber for the building of the sub-shrine of Tsurugaoka has been attended to. The commissioners for this project are Doi Saneharu and Ōba Kageyoshi, This shrine, built last year, was intended to be a temporary structure. Thus, it is extremely simple, consisting of pine pillars and a miscanthus roof. The shrine will be decorated to enhance the glory of the gods. (Shinoda 1960: 217)

Seventh month, twenty-first day, 1181. Tarō Yoshimori of the Wada clan and Kajiwara Heiza Kagetoki and others received an order from above to accompany Sachuta who had been arrested yesterday and they headed to the River Katase. However, an order followed them through an Endo warrior, Mushiya-dokoro, who was sent to the River Katase area, stating that Kagetoki was commissioned to take charge of building a new sub-shrine. Therefore, Kagetoki returned quickly to his work. As his replacement, Hirauchi Mitsumori accompanied Yoshimori. Details of their roles are unknown, but these records show that Easterners, of Tōgoku, took part throughout this building project. The first of the two texts record that Ōba Kageyoshi and Doi Sanehiru were responsible for timber, and the second states that Kajiwara Kagetoki, who was related to the Ōba family, was appointed magistrate in charge of the construction. It is also evident from the next text that Easterners were generally skilful enough to complete large buildings on their own, without assistance from Kyōto, the capital. Seventh month, third day, 1181. Shōkan is commissioned to bring carpenter Kyōshi of Asakusa, Musashi Province, to Kamakura. (Shinoda 1960: 220)

Fifth month, twenty-fourth day, 1181. The land for the building of a residence for His Lordship’s daughter and a storehouse has been cleared. Kageyoshi, Kagetoki, and Shōkan are in charge. The immediate vassals have contributed workmen. (Shinoda 1960: 218) The latter entry refers to building the Palace. The administrators responsible for building the Minor Palace and the shogun’s stable included Ōba Kageyoshi and Kajiwara Kagetoki, and they were also charged with providing local labourers. The Azuma Kagami’s entry for 17th June 1191 refers to Miura-no-suke as the official in charge of building the large stable, and the entry for the next day to Doi Jirō and Okazaki Shirō as responsible for erecting the large posts and setting up

the framework of the inner stables. We can assume a similar process for the structure that was presumably the East Palace. Most vassals at that time came from Tōgoku, so the labourers recruited for the purpose may have been Easterners.

Fifth month, twenty-eighth day, 1181. The carpenters requisitioned from Awa arrived last night. Thus, the raising of the pillars and beams for the buildings will take place today. (Shinoda 1960: 218)

As this text shows, many Easterners, including carpenters summoned from Awa, worked on building the Shogun’s Palace. That corresponds to Gomi’s view that the shogun’s palace in Kamakura has affinities with the manor houses of the Eastern samurai (Gomi 1992a). Moreover, since the Minami Mikado site, discussed above, seems associated with the Hatta and the Hatayama families, a similar process of local eastern influence may perhaps be observed in the development of this site.

Shogun’s Palace

Ōmiwa (1989a,b) states that, to help establishing Kamakura as a town, a series of construction projects were modelled on Kyōto, where Yoritomo was born and brought up. There is no doubt about Yoritomo’s strong link to Kyōto. It was also familiar to many of his samurai, including those sent there under the reign of the Taira family (Oka 1995). According to Ōmiwa, Yoritomo’s taste for Kyōto is also evident in that his residence was built in the shindenzukuri architectural style typical of a nobleman’s residence in the Heian period. However, the Ōkura Palace’s retainers’ quarters and stables were significantly larger than those at other noblemens residences. Ōta (1987) points out that the shogunal palace was not necessarily faithful to Kyōto’s shindenzukuri but, rather, a mixture of shindenzukuri and elements characteristic of samurai residences, such as retainers’ quarters and stables. The Kamakura Palace’s structural hybridity was partly due to the distinct functions and life styles of samurai and noble houses in Kamakura and Kyōto (Ono 2004). It was not just another nobleman’s residence. In addition to the differences mentioned above, there were also large fences and moats. The urban aspects of Kamakura early in its development should be considered as rooted in the sociology of Tōgoku on account of the eastern origin its builders. That Kamakura had always been closely connected to its hinterland can be seen in the builders’ units of measurement. For instance, excavations of the Enjōji Temple area showed that, up to the first half of the 13th century, the buildings had columns 2.1 metres apart on average, just like those of contemporary Kamakura (Nirayama 1995).

49

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 65

50

The archaeology of medieval towns: case studies from Japan and Europe

Ono Masatoshi points out that, during the Kamakura period in Tōgoku, elite samurai could live in shindenzukuri residences. Examples include Koizumi-yakata in Tsukubō and the Ima Kōji West site in Kamakura but they were extremely rare. Different styles prevailed in the East. For instance, at the Ōkuboyama site in Honjō City, Saitama Prefecture, an oblong main house and other buildings around it formed an L or a U shape while the Miyakubo and the Kamihamada sites, in Kanagawa Prefecture, lacked retainers’ oblong buildings and seem not to have followed any particular model. The eastern samurai must have had plenty of opportunities to see shindenzukuri buildings in Kyōto but the style did not become normal in Tōgoku. In accounting for this divergence, Ono (2004) makes the interesting observation that it arose not simply because shindenzukuri did not match the life and rituals of the eastern samurai but also because the latter preferred a distinctive eastern style (an L- or U-shaped main house plus retainers’ quarters plus stables). In Hiraizumi, the Eastern samurai adopted both the Kyōto and the Eastern style for different occasions: facilities associated with the afterlife (e.g., temples and gardens) were given authority by employing the Kyōto style and those associated with temporal life were treated in native eastern style.

Although brought up in Kyōto, Yoritomo came to power through the support of eastern samurai, so he was obliged to conform to eastern standards in some degree. Even had he managed to recreate shindenzukuri precisely, it might have been impractical and would have highlighted differences between his taste and that of his eastern vassals. To promote himself as the east’s leader, it was essential for Yoritomo to share the eastern samurais’ values and to display his power by building in eastern style on a large scale. It was in this context that the large fences or palisades were constructed.

It was not unusual for aristocrats engaged with the military affairs of the central government in Kyōto to move to local regions. They used their high status and samurai ideology to unify provincial samurai. The retainers’ quarters and stables in the Ōkura Palace show that the aristocracy often adopted local values. The following texts are particularly significant for considering how Kamakura changed as a capital city.

Palaces

Seventh month, twenty-third day, 1213. An order came concerning the construction of the new residence (palace) for the Shogun. The directions were given in his presence. There would be some changes, including the erection of the Middle Gate. Military Chief Euemon and Civil Affairs Chief To were appointed.

When it was rebuilt in 1213, before Karoku (1225) and the transformation of the city, the palace’s architecture was in the

midst of its metamorphosis. The first construction of the middle gate (chūmon) was under way and so was the shindenzukuri corridor extending from it (chūmon-rō). Fifteen years later, a palace without a middle gate and corridor waiting room for samurai was criticised as very dilapidated.

Tenth month, eighteenth day, 1228. Yesterday at noon, the platform of the Hakone Shrine burned down. News of the fire was quickly delivered. Since the founding of the shrine, no such disaster as a fire has ever occurred. There was discussion about whether the repair of the palace would be delayed. There was a gathering of people such as the former government official of Suruga (Suruga-zenji), the Buddhist priest of Oki (Oki-nyudo) and the military chief, Gozaemon (Goto-saemon-no-jo). They concluded that there should be no hesitation about the repair of the palace, which had been damaged by strong winds. The palace, which lacks an anteroom for samurai, or the corridor attached to the middle gate, looks incredibly dilapidated. It would be better to hasten the repairs.

In those fifteen years, the concept of the Palace changed dramatically. As I have mentioned, urban development in the Karoku period (1225-1227) was meant to introduce Kyōto’s architectural system during the Kenpo (1213-1219) and Antei (1222-1224) periods, the Shogun’s Palace was occupied by Minamoto no Sanetomo and Fujiwara no Yoritsune. Among those that served the former occupant were people who moved from Kyōto, including a brother of Minamoto no Naka’akira’s, Nakakane, later assassinated with Sanetomo. Sanetomo married a daughter of Bōmon Nobukiyo in 1204 and, as a consequence, many people moved from Kyōto to the Palace in Kamakura, including Nakakane. His connection to Kyōto was even more apparent, and his family held the post of imperial regent for generations. Minamoto no Naka’akira was a man of many talents: he served not only as attendant to Sanetomo but also as mediator to the imperial court in Kyōto; and he owned a substantial library and had extensive knowledge of classical Chinese literature and history (Gomi 1990). That such a group became members of the Shogun’s palace staff reveals the earliest signs of the emulation of Kyōto. The lapse of time between emulation at the Palace and other places in Kamakura must have corresponded to the respective degrees of affinity with Kyōto. To justify this postulate, we must compare the lifestyles of those in shindenzukuri houses with those in other kinds of residence. It is also important to examine how the eastern samurai felt about the connection with Kyōto and the adoption of Kyōto manners.

Adopting Kyōto style

Ninth month, twenty-sixth day, 1213. In the evening, Naganuma Munemasa arrived from the Shimotsuke area, announcing that he brought the severed head of Jūkei Nakakane, a court noble at the shogun

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 66

The establishment and transformation of Japan’s medieval capital, Kamakura

household. He delivered the message from the Shogun (Sanetomo), saying that Shigetaka had been condemned without committing any offence. His youngest son, a priest, was involved in intrigue, but his role was not significant. Had he been captured alive and brought here, the judgment would have been pronounced here. However, he had already been killed cruelly, through a lamentable, careless and hurried decision. To this unpleasant reaction of the Shogun, Munemasa retorted angrily that the priest in question was undoubtedly a rebel and that if he had brought him to Kamakura, the court ladies and nuns would have appealed for leniency, so he dared to kill him. He warned people to be vigilant or, in future, even loyal people would hesitate to give devoted service and the Shogun would become lax and would reward people lavishly. Nor were strict orders obeyed either by religious or political groups. Munemasa wished for a military position, stating that he would discipline unruly people in the fifteen counties in the Tōkaidō region. He wanted stronger weapons. He would be grateful for such a position. He stated that the present reign appeared to ignore warriorship, with people singing and kicking kemari [football] balls. Women were given high status and there were no warriors. Properties confiscated by the authorities were given to women instead of awarding them to warriors for distinguished service. The property of Shirō Shigetomo of Hangaya was given to the Lady Gojō no Tsubone, and the former property of Shirō Shigemasa of Nakayama was given to the Court Lady Shimousa no Tsubone. There were many similar examples. The court noble Nakakane left the scene without a word and Munemasa also left.

In this text, Naganuma Munemasa criticises Minamoto no Sanetomo severely. It is particularly interesting that both kemari and the lady who interfered with political affairs were associated with the Palace. If the residents of the Shogun’s Palace in Kamakura were assimilating to the Imperial Palace in Kyōto, which was a Mecca of kemari, that would explain those aspects of the palace about which Munemasa protested. This attitude should not be attributed solely to Munemasa, of course. For example, the remains of a residence, recovered from an area of Ōkuboyama site in Honjō City, Saitama Prefecture, show varying distances between the large posts throughout the mid-12th and early 14th centuries (Waseda 1998). Unlike those of Kamakura’s, those large posts do not show a change at the beginning of the 13th century. We can conclude, then, that the eastern architectural styles which seem to have disappeared in Kamakura may have persisted outside the capital. Moreover, Ono Masatoshi points out that shindenzukuri buildings were rare in the east throughout the Kamakura period. So investigation as to why shindenzukuri building did not diffuse through the east entails study of various factors, including the distinct feelings of each samurai. Even though the central gate and corridor were added, the Shogun’s Palace did not become a complete example of shindenzukuri building because, in Kamakura, long, large retainers’ quarters were indispensable for shogunate rituals and these are not found in Kyōto. That may be the result of choice and adjustments made by the shogunate; they did not introduce styles and techniques from Kyōto uncritically. The disappearance of early Kamakura’s ‘three-piece set’

must be understood in this light. The large raised buildings declined, for reasons unknown; and the purpose of the V-ditched moats changed and they were replaced by the timbered ones as society settled. However, I suspect that, at least in the Karoku period (1225-1227), just after the fall of Hiraizumi, there was no specifically urban model on which to base a capital in eastern Japan, so an urban template had to be brought over from Kyōto in order to manage Kamakura’s growth efficiently.

Certainly, Kamakura related to Kyōto in many different ways, and many literary men from Kyōto were engaged in managing the shogunate (Gomi 1992b: 55). Moreover, as The History of Kamakura City points out, various people moved from Kyōto to Kamakura (Kamakura Shishi Hensen Iinkai 1959). It is undeniable, then, that Kamakura tried to develop its urban structure and culture by bringing ideas in from Kyōto. Nevertheless, I emphasise that, as we have seen in the archaeology, as a receiver, Kamakura did have its own traditions for accommodating Kyōto culture. Without that, most of the culture of Kamakura and other regions of the east would have been nothing but an imitation of Kyōto. We must examine critically the concept of one-way transmission of culture and civilisation from Kyōto, seeing the development of other cities as more complex and part of a wider network of influence.

Conclusion

In this paper, I have reviewed the development of Kamakura up to the 13th century. In conclusion, I discuss a problem that I have encountered in writing my review. In order to argue for continuity of urban planning from the pre-Kamakura period, I tried to identify evidence dated to before 1175, though I must confess that the attempt did not go as well as I had expected it to: the archaeology of Kamakura in this period is so limited that they could only show a ‘Dark Age’ (Mabuchi 2004). It is not necessarily that remains and traces of the period do not exist. Rather, at least in part, excavations might have overlooked them. For the period between the establishment of local authority in Kamakura and Yoritomo’s entry, research with the limited number of documentary sources is regarded more highly than archaeology. In future, we will have to re-examine the remains and traces excavated from the lowest living surfaces of the Middle Ages, and bridge the gulf between documentary findings and archaeology. We need to broaden the focus from my discussion in this paper to include the residences of the Minamoto family before Yoshitomo’s time, those of the Taira family before Taira no Naokata, and beyond. We need to excavate such remains still buried beneath the modern city of Kamakura

51

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 67

52

The archaeology of medieval towns: case studies from Japan and Europe

and examine them for the next stage of our research.

Finally, I suggest that, in order to investigate the case of Kamakura fully, we should seek a broader perspective on Tōgoku, the Eastern Country. As the region’s centre, Kamakura was closely connected to its hinterlands, and we will have to re-examine as many excavated sites as possible in Tōgoku. Together, those investigations would enable us to consider the case of Kamakura with a new perspective.

References

Amino, Y. 1993. Kyō to Kamakura: chūsei toshi toshite no, Asahi hyakka nihon no rekishi bessatsu: rekishi wo yominaosu 6 (Kyoto and Kamakura: medieval cities, Asahi Japanese history encyclopedia: re-reading history 6). Tokyo, Asahi Shimbun.

Fukushima, K. 1999. Tsurugaoka Hachimangū no seiritsu to Kamakura Shōgenji, Enoshima (The Development of the Tsurugaoka Hachiman Shrine and the Shogenji Temple, Kamakura, and Enoshima). In: Chihōshi Kenkyūkai Kyōgikai (ed.) Toshi, kinkō no shinkō to yūsan: kankō, kōryū, to henyō (Religious faith and excursions in cities and suburbs: tourism, interaction, and change). Tokyo, Yuzankaku. Goble, A. 1982. The Hojo and consultative government. In: Mass, J. (ed.) Court and bakufu in Japan: essays in Kamakura history. New Haven, Yale University Press: 168-190.

Gomi, F. 1990. Azuma Kagami no hōhō: jijitsu to shinwa ni miru chūsei (The method of the Azuma Kagami: fact and legend in the Middle Ages). Tokyo, Yoshikawa Kobunkan. Gomi, F. 1992a. Sanetomo no bunka kūkan (The cultural space of Sanetomo). Miura Kobunka 51: 1-12.

Gomi, F. 1992b. Bushi to bunshi no chūsei shi (Medieval history of warriors and scholars). Tokyo, Tokyo Daigaku Shuppankai. Hara, H. 1998. Kamakura Shi Utsunomiya Tsujiko Bakufuseki, Dai 22 kai Kanagawa-ken iseki chōsa (Kamakura City Utsunomiya crossing Bakufu Site, 22nd Kanagawa prefecture excavation report). Kanagawa, Kanagawa Ken Kōkogakukai.

Hara, H. 1999. Kanagawa Utsunomiya Tsujiko Bakufuseki (Kanagawa Utsunomiya crossing bakufu site). Mokkan Kenkyū 21.

Hurst, C. 1982. The Kobu polity: court bakufu relations in Kamakura Japan. In: Mass. J. (ed.) Court and bakufu in Japan: essays in Kamakura history. New Haven, Yale University Press: 3-28.

Hyōgo Ken Kyōiku Iinkai 2003. Kōbe daigaku igakubu fuzoku byōin rittai chūshajō setsubi jigyō ni tomonau Kusunoki Aratamachi iseki (The Kusunoki Arata Machi site excavation during improvements to parking facilities at the Kōbe University Hospital). Kōbe, Hyōgo Ken Kyōiku Iinkai.

Ima Kōji Nishi Iseki Hakkutsu Chōsa Dan 1990. Ima Kōji nishi iseki Onari shōgakkōnai hakkutsu chōsa hōkokusho (Report of excavations of the Ima Kōji Nishi iseki inside the Onari Elementary School). Kamakura, Kamakura Shi Kyōiku Iinkai. Irumada, N. 1991. Hiraizumi kan wa beisukampu datta (The Hiraizumi headquarters was a base camp). Rekishi Techō 19(7): 12-15.

Ishii, S. 1986. Chūsei Roku’ura no rekishi (The history of medieval Roku’ura). Miura Kobunka 40: 1-18.

Ishii, S. 1988. Chūsei toshi, Kamakura: rekishi no gen fūkei o motomete (Searching for the original historical landscape of the medieval city of Kamakura). In: Asahi Shimbunsha (ed.) Shūkan Asahi hyakka nihon no rekishi bessatsu: rekishi no yomikata 2,

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 68

The establishment and transformation of Japan’s medieval capital, Kamakura

toshi no keikan no yomikata (Asahi weekly, special issue: Japanese history, ways of reading Japanese history 2, reading cities and their views). Tokyo, Asahi Shimbunsha.

Kamakura Shi Kyōiku Iinkai 1993. Ōkura Bakufu shūhen isekigun Nikaidō Aza Egara 38 ban 1 No.49, Kamakura shi maizō bunkazai kinkyū chōsa hōkokusho 9 heisei 4 nendo hakkutsu chōsa hōkoku dai 2 bunsatsu (The Ōkura Bakufu Site Group No.38 Nikaido Aza Egara Rept. 49, Kamakura Buried Cultural Assets Emergency Investigation Report 9, 1992 Excavation Report, Vol. 2). Kamakura, Kamakura Shi Kyōiku Iinkai.

Kamakura Shi Kyōiku Iinkai 1996. Kitaoji Komachi yashikiseki [Yasutoki to Tokiyori yashiki, no. 282, Yukinoshita 1 chōme 377 ban chiten], Kamakura shi maizō bunkazai kenkyū chōsa hōkokusho 12, heisei 7 nendo hakkutsu chōsa hōkokusho dai 2 bunsatsu (The Kitaoji Komachi Residential Site [Yasutoki and Tokiyori Residence, No. 282, Yukinoshita 1 chōme No. 377 Locality 7], Kamakura buried cultural assets investigation report 12, 1995 excavation report, vol. 2). Kamakura, Kamakura Shi Kyōiku Iinkai.

Kamakura Shi Kyōiku Iinkai 1998a. Ōkura bakufu shūhen isekigun, No. 49, Yukinoshita 4 chōme, 620 banchi 5 chiten, Kamakura shi maizō bunkazai kinkyū chōsa hōkokusho 14, heisei 9 nendo hakkutsu chōsa hōkoku dai 2 bunsatsu (Remains around Okura bakufu, Locality 5, Yukinoshita 4 chōme, no. 620, Kamakura buried cultural assets emergency investigation report 14, 1997 excavation report, vol. 2). Kamakura, Kamakura Shi Kyōiku Iinkai.

Kamakura Shi Kyōiku Iinkai 1998b. Kanagawa ken Kamakura shi Tōshōjiseki dai 3, 4 ji ikō kakunin chōsa hōkokusho (Report of third and fourth investigations to confirm the site of the Tōshōji temple, Kamakura city, Kanagawa prefecture). Kamakura, Kamakura Shi Kyōiku Iinkai.

Kamakura Shishi Hensan Iinkai 1959. Kamakura shi shi (History of Kamakura). Kamakura, Kamakura Shishi Hensan Iinkai. Kanagawa Ken Kōkogakukai 1998. Kamakura shi Utsunomiya Tsujiko bakufuseki, dai nijū ni kai Kanagawa ken iseki chōsa kenkyū happyōkai happyō yōshi (The site of the bakufu at the Utsunomiya crossing, 22nd Kanagawa prefecture site excavation and study lecture meeting, summaries). Kanagawa, Kanagawa Ken Kōkogakukai. Kawano, S. 1995. Chūsei toshi Kamakura: iseki ga kataru bushi no miyako (The medieval city of Kamakura: the archaeological site speaks about the samurai city). Tokyo, Kodansha.

Kenshū Dōjō Yōchi Hakkutsu Chōsadan 1983. Kenshū dōjō yōchi hakkutsu chōsa hōkokusho (Report of Excavations at the Site for Religious Training). Kamakura, Kamakura Tsurugaoka Hachimangū.

Mabuchi, K. 1992. Chūsei Kamakura ni okeru Yato kaihatsu ni aru sokumen (One aspect of the development of Yato in medieval Kamakura). Kamakura 69.

Mabuchi, K. 1994. Bushi no miyako Kamakura, sono seiritsu to kōsō o megutte (Samurai capital, Kamakura, its development and plan). In: Amino, Y. and Ishii, S. (eds.) Chūsei no fūkei o yomu 2: toshi Kamakura no Bandō no umi ni kurasu (Reading medieval landscapes vol. 2: the city of Kamakura and living on the seacoast of Bandō). Tokyo, Shin Jimbutsu Oraisha.

Mabuchi, K. 1997. Shokki kara mita chūsei Kamakura no toshi kūkan (Urban space of medieval Kamakura considered through eating utensils). Kokuritsu Rekishi Minzoku Hakubutsukan Kenkyū Hōkoku 71: 431-472.

Mabuchi, K. 2004. Chūsei toshi Kamakura seiritsu zenshi (History of Kamakura before its medieval development). In: Gomi, F. and Mabuchi, K. (eds.) Chūsei Kamakura no jitsuzō to kyōkai (The true image and environment of medieval Kamakura). Tokyo, Koshi Shoin. Mandokoro Iseki Hakkutsu Chōsadan 1991. Kanagawa ken Kamakura shi Mandokoro iseki (The Mandokoro site, Kamakura city, Kanagawa prefecture). Kamakura, Mandokoro Iseki Hakkutsu Chōsa Dan.

Mass, J. P. 1982. The early bakufu and feudalism. In: Mass, J. (ed.) Court and bakufu in Japan: essays in Kamakura history. New Haven, Yale University Press: 123-142.

Matsuo, K. 1992. Buke no shuto Kamakura no seiritsu: Shōgun Gosho to Tsurugaoka Hachimangū to o chūshin ni (‘City’ of samurai: the development of Kamakura, centering on the Shogun’s Palace and the Tsurugaoka Hachimangū). In: Ishii, S. (ed.) Toshi to hina no chūsei shi (Medieval history of cities and countryside). Tokyo, Yoshikawa Kobunkan.

Matsuo, K. 1993. Chūsei toshi Kamakura no kōkei (The landscape of the medieval city of Kamakura). Tokyo, Yoshikawa Kobunkan.

Mukai Egara Iseki Hakkutsu Chōsa Dan 1985. Kamakura shi Nikaidō Mukai Egara hakkutsu chōsadan hōkokusho (Report of the excavation group of the Mukai Egara Site, Kamakura City, Nikaido). Kamakura, Kamakura Shi Kyōiku Iinkai.

Nirayama Chō Kyōiku Iinkai 1995. Izu Nirayama Enjōji iseki, Goshonai iseki dai sanji chōsa (The Izu Nirayama Enjōji Site, the third investigation of the Goshonai site). Nirayama, Nirayama Chō Kyōiku Iinkai.

Noguchi, M. 1993. Yoritomo izen no Kamakura (Kamakura before Yoritomo). Kodai Bunka 45(9).

Noguchi, M. and Yamada, K. 2003. Hōjūjidono no jōkaku kinō to jōnai no ryōbo ni tsuite (Concerning the castle wall function of the main building of Hōjūjidono and the graves inside the castle walls). Kyōto Joshi Daigaku Shūkyō Bunka Kenkyūsho Kiyō 16: 127-152.

Noguchi, M., and Yamada, K. 2004. Rokuhara no gunjiteki hyōka to Hōjūjidono o fukumeta kūkan fukugen (The military value of Rokuhara and a spatial reconstruction including the Hojujidono). Kyōto Joshi Daigaku Shūkyō Bunka Kenkyūsho Kiyō 17: 109-134. Nuki, T. 1981. Chūsei no Kamakura (Kamakura of the medieval period). Kusado Sengen 9(8).

Oka, Y. 1995. Chūsei kyokan saikō sono seikaku o megutte (Rethinking medieval fortified residence sites and their characteristics). In: Gomi, F. (ed.) Chūsei no kūkan o yomu (Reading medieval space). Tokyo, Yoshikawa Kobunkan.

Oka, Y. 1998. Yasutoki izen no Kamakura: toshi no tenkei (Kamakura Before Yasutoki: the selected landscapes of Kamakura). Kamakura 88.

53

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 69

54

The archaeology of medieval towns: case studies from Japan and Europe

Oka, Y. 2004a. Toshi shūhen o hashiru michi: Inukakezaka e no michi kara (Peripheral roads encircling the city as seen from the road to Inukakezaka). In: Fujiwara, Y. (ed.) Chūsei no michi o saguru (Excavating medieval roads). Tokyo, Koshi Shoin.

Oka, Y. 2004b. Gensō no Kamakura jō (The phantom Kamakura city). In: Gomi, F. and Mabuchi, K. (eds.) Chūsei Kamakura no jitsuzō to kyōkai (The true image and environment of medieval Kamakura). Tokyo, Koshi Shoin.

Ōkura Bakufu Shūhen Isekigun Hakkutsu Chōsadan 1990. Yukinoshita Aza Ōkura Kōchi 569 Ban 1 chiten hakkutsu chōsa hōkokusho (Report of excavations at locality 369, Aza Ōkura Kōchi Yukinoshita). Kamakura, Ōkura Bakufu Shūhen Iseki Gun Hakkutsu Chōsadan.

Ōmiwa, T. 1989a. Toshi Kamakura no michi to chiiki (Roads and regions of the city of Kamakura). In: Yasuda, Motohisa Sensei no Taikan Kinen Ronshūkankōinkai (ed.) Chūsei Nihon no shosō (Various aspects of medieval Japan) vol. 2. Tokyo, Yoshikawa Kobunkan. Ōmiwa, T. 1989b. Kamakura no toshi keikaku: seiji toshi toshite, gunji toshi toshite (The city plan of Kamakura as a political or military city). In: Ishii, S. and Ōmiwa, T. (eds.) Yomigaeru chūsei 3, bushi no toshi Kamakura (Reconstructing the medieval period 3, the samurai city of Kamakura). Tokyo, Heibonsha.

Ono, M. 2004. Chūsei bushi no yakata sono tatemono, keifu to keikan (Medieval samurai fortified residences, their buildings, genealogy, and view). In: Ono, M., Gomi, F. and Hagiwara, M. (eds.) Chūsei no keifu, higashi to nishi, kita to minami no sekai, Kokogaku to chūsei no kenkyū 1 (The genealogy of the medieval age, the world of east-west, north-south, study of archaeology and medieval history 1). Tokyo, Koshi Shoin. Ōta, S. 1987. Shindenzukuri no kenkyū (Study of Heian residential architecture). Tokyo, Yoshikawa Kobunkan.

Rekishi Shiryō Netuoku (ed.) 2005. Heike to Fukuhara kyō no jidai, Iwata shoin bukureto rekishi kōkogakukei H-01 (The Heike and the Pperiod of the Fukuhara capital, Iwata Bookstore history archaeology booklet H-01). Tokyo, Iwata Shoin. Shinoda, M. 1960. The founding of the Kamakura shogunate 1180-1185, with select translations from the Azuma Kagami. New York, Columbia University Press.

Sugimotodera Shūhen Iseki Hakkutsu Chōsadan 2002. Kanagawa ken Kamakura shi Sugimotodera shūhen iseki Nikaido aza Sugimoto 912 ban 1 hoka chiten hakkutsu chōsa hōkoku (Report of excavations and investigations at sites around the Sugimotodera and the Nikaido Aza 912-1, and other sites, Kamakura city, Kanagawa prefecture). Kamakura, Kamakura Shi Kyōiku Iinkai. Takahashi, M. 1998. Heishi no yakata ni tsuite: Rokuhara, Nishi Hachi-jō, Kujōmatsu (The fortified residence of the Heike clan: Rokuhara, Nishi Hachi-jō, at the end of Kujō). Kōbe Daigaku Shigaku Nempō 13.

Waseda Daigaku 1998. Ōkuboyama VI, Waseda Daigaku Honjō kōchi bunkazai chōsa hōkoku 6 (Okuboyama VI: Report of investigations of cultural properties on the Waseda University Honjō campus). Tokyo, Waseda Daigaku.

Yaegashi, T., and Iimura, H. 1995. Tōgoku no naka no Goshonouchi iseki gun, Enseiji Enjōji iseki: ibutsu to ikō kara (The Imperial Palace site group in the centre of Tōgoku, and the Enjōji site: artifacts and features). Rekishi Techō 23(9): 27-36.

Yamamura, A. 1997. Chūsei Kamakura no toshi kūkan kōzō (The city structure of urban space in medieval Kamakura). Shirin 80(2): 208-248.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 70

The development of Hakata as a medieval port town

The development of Hakata as a medieval port town Ōba Kōji Summary

Located at the heart of Hakata Bay, and regarded as one of the finest natural harbours in Japan, the modern city of Fukuoka in northern Kyūshū regards itself as the gateway to Japan. This is a description which can be applied to the area throughout much of Japanese history to the present day: the National Museum of Kyūshū at Dazaifu in the hinterland of Fukuoka is the latest expression of this role. As the part of the Japanese archipelago closest to the Korean peninsula and thus the East Asian continental mainland, north Kyūshū has been the point of introduction of many continental influences, from rice agriculture in the prehistoric Yayoi period, to the Chinese traders who lived in Hakata, the largest trading port in Japan during the medieval period. Hakata was probably also the first Japanese urban centre encountered by an Englishman, John Saris, the captain of the British East India Company’s ship the Clove, who visited in 1613.

Ōba Kōji provides a summary of some of the key archaeological findings from the medieval port town of Hakata, a place name now preserved in the downtown area around the main railway station of the modern city of Fukuoka. Ōba undertook many of these excavations himself. The paper discusses the development of the urban areas in relation to the natural topography of coastal sandbars, and the sequence of infilling that created the medieval townscape. He traces the archaeology of the medieval Chinese trading community at Hakata, a new manifestation of Japanese interaction with the continent, preceded by governmentcontrolled engagement with foreigners through the Kōrokan, the official guesthouse for foreign envoys wanting to engage with Japan from the 7th to 11th centuries. At this time north Kyūshū was ruled through the regional capital of Dazaifu. Ōba’s paper is complemented by other accounts of medieval Hakata (Batten 2006). Figure 1: Hypothetical map of Hakata, showing locations mentioned in text (from Ōba 2003)

This chapter is an adapted summary of Kōwan toshi Hakata no seintsu to hatten by Ōba Kōji, published in 2004, drafted by Richard Pearson and edited by Oscar Wrenn.

55

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 71

56

The archaeology of medieval towns: case studies from Japan and Europe

The excavations at Hakata, which began in the late 1970s with rescue excavations in advance of the construction of the first underground mass transport system, were of great significance for the development of Medieval urban archaeology in Japan. A richly illustrated summary was published in 2008 drawing together many of the findings of the past thirty years (Ōba, Saeki, Suganami and Tagami 2008). The location of Hakata, in the estuary of the Hie River, is very comparable to Kusado Sengen in Hiroshima and many other early trading towns. The rich assemblages of artefacts, notably ceramics imported from China, food remains including animal bones and plant matter, and the many structural features, including wells, graves, and the foundations of temples, houses and many other buildings, all create the impression of a vibrant urban trading centre. Extant historical materials allow some of the sites to be associated with named individuals and clans.

As does Oka Yōichirō’s paper on Kamakura, Ōba demonstrates how it is possible to weave together archaeological evidence and historical records to bring the development of medieval urban centres in Japan to life. Following an introduction to the archaeology of Hakata, Oka discusses the archaeological and historical evidence for the medieval Chinese trading post, and then goes into detail about the developments of two urban areas based around two beaches or sandbars, Okinohama and Hakatahama.

Introduction

The medieval city of Hakata lay where the River Hie flows into Hakata Bay, in northern Kyūshū. It developed from two towns that that had sprung up on sandbars, Hakatahama (Hakata Beach) and Okinohama (Okino Beach) (Figure 1). These two towns were not merely separated geographically at this time, but were also under different administrations, with Hakatahama being held by the Shōni and then Ōuchi clans, and Okinohama ruled by the Ōtomo. This changed in the mid 16th century

Figure 2: Stratigraphy of Locality 42 (scale 1:200)

when the Ōuchi clan fell, and the Ōtomo became the sole rulers, though it was two clans at the beginning of the Modern period, Kobayakawa and Kuroda, who set about reclaiming the estuary in between the sandbars. In the following pages, I will discuss the development of Hakata and Okinohama, and the ways in which its unusual topography has determined its urbanisation.

Hakata

I will first look at the development of Hakatahama, lying on the inland side of the two sandbars. Urbanisation at Hakatahama began when it was a post for Chinese traders who resided there from the late 11th century, trading goods from the nearby Song dynasty under a system known as jūban bōeki (Ōba 2001, 2003). Judging from excavations, however, the earliest settlement was in the Middle Yayoi period (400 BC–AD 50), though remains have also been found from the Ancient period. Features of this period indicate that there were administrative facilities on Hakatahama; in the southern half, ditch-shaped features run north-south and east-west and there are remains of large posts dug into the ground, which clearly supported buildings. The northern half has remains of pit dwellings and wells, though artefacts such as stone belt plaques, pottery inscribed with black ink and kōchō sen coins point to the existence of Nara and Heian period government officials. It is impossible to know the exact conditions of this governance, however, as there are no surviving historical records that mention the names associated with this at Hakatahama. Features and artefacts from the Hakata sites show that a trading centre was established by the mid 11th century. Records from Dazaifu, the regional government in Kyūshū from the 8th to 12th centuries, show casualties of a fire in 1047 in the 'Chinese traders' residence' (daisōshokyaku). That may have been the Kōrokan, the diplomatic guesthouse for Kyūshū at the time and a staging post for envoys and delegates from overseas. However,

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 72

The development of Hakata as a medieval port town

excavations of the Kōrokan have yielded no finds from the mid 11th century, so we can postulate that by this time the trade centre had, in fact, shifted to Hakatahama.

In the Kannon gengisoki, a document in the Saikyōji temple in Ōtsu, in present-day Shiga Prefecture, there is an entry dating to 1116 headed 'Chikuzen no kuni Hakata tsu tōbō’, or 'Chikuzen Hakata Chinese Residence'. That seems to confirm the presence of Chinese at the Hakata port, and is the oldest reference to them living in Hakata. Finds from the Hakata sites dating to the later 11th and early 12th centuries indicate a Chinese site on the west side, though in the late 12th century it expanded and was eventually assimilated into neighbouring Japanese sites. The jūban bōeki residential trader system, where Chinese Song traders who had taken up residence in Hakata could go back and forth between Hakata and their home land and carry and sell goods (Kamei 1986), was a transformation from that which had been established in the Kōrokan. In its initial phase Dazaifu officials continued to manage trade as had been done at the Kōrokan (Yamauchi 1989), and it is very likely that the reason Hakata was chosen as the location of this trade centre was because of an inclination on the side of the Dazaifu to continue this control of foreign trade, in addition to demands by Song traders for a location that provided them with a greater deal of autonomy. The Chinese residential area brought about a rapid urbanisation of the area, with archaeological features and artefacts of the early 12th century distributed all over Hakatahama, and showing a crowded urban environment.

However, the post was cut off from China in the 13th century, when the Yuan (Mongols) defeated the Southern Song. With the emergence of the Kamakura bakufu, Hakatahama became the site of the chinzei tandai, an agency governing Kyūshū and Tsushima for the Kamakura shogunate, and making Hakatahama the military and political capital of Kyūshū. At the beginning of the 14th century, Hakata’s roads and zones were laid down under the chinzei tandai’s direction, and the city that had developed out of the Chinese trading centre on Hakatahama was completed, and subsequently maintained through the medieval period.

Emergence of Okinohama

Having been nothing but sand in the Heian period (794–1185) and then a settlement of a few scattered houses in the Kamakura period (1185–1333), Okinohama became Hakata’s thriving nucleus during the Muromachi period (1392–1573). It could not have developed into a commercial town on the strength of its natural assets alone; human intervention, especially expansion of the habitable area, contributed greatly.

In an archaeological investigation of Okinohama’s inland slope,

Figure 3: Shorelines of 1646, in Shōhō period; 1699, in Genroku period; and 1891, in Meiji period (from Iso et al 1998)

at Locality 42 (see Figure 4 for locations of all excavations mentioned), a trench was dug along a slope aligned northwest to southeast (Figure 2; Fukuoka 1991), revealing stratigraphy and features. Perpendicular to this trench another, shorter, one was dug to ascertain the level to which the original land surface would have sloped. The perpendicular trench showed even accumulation and a level surface right down to the original sandbar, but the longer trench running north-west showed that it dropped dramatically, even though the current surface is barely sloped at all. Next I will look at the accumulations between the sandbar and the brown and dark brown medieval layers. In the northern half of the excavation there was an accumulation of a greytinged layer on top of the pale-yellow sand, then the medieval layer sitting on top of this. In the southern half pale yellow sand was not detected, but there was a thick accumulation of predominately grey sand (Layer III). Directly above Layer III was a darkish brown layer of mud mixed with sand, then alternating layers of grey and dark grey sand sitting above this. Through the accumulation of these layers we can see the slope becoming shallower (Layer II). Above this was Layer I, comprised of greyish-brown material 20-80cm thick, which was homogenous to the extent that initially it was mistaken for the upper layer of the sandbar. Above Layer I was the medieval layer.

According to Shimoyama Shōichi of Kyūshū University (personal communication), layers underneath the top of the sandbar were covered by water at high tide, and represent a wetland layer in which trace fossils can be seen, showing that Layer III and below are the result of natural accumulation.

57

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 73

58

The archaeology of medieval towns: case studies from Japan and Europe

In contrast to this, Layer II has, according to Shimoyama, a high probability of being made through human interaction (infilling). Layer II is connected to Surface VI at about the centre of the excavated area. Layer I was formed through wind-blown sand, with no evidence of occupation while it was accumulating.

There are a number of different surfaces found within these layers, including Surface IV, which was found in the shallow accumulation of grey sand in north part of the excavation and the upper part of Layer I in the south part of the excavation; Surface V, which was found on the upper surface of the natural sandbar in the northern part, and on the upper surface of Layer I on the south part; and Surface VI, which was found in the middle section of the excavation, on the surface of a layer comprised of dark grey mud and sand.

Together, these strata and their features allow us to piece a plausible picture together of the development of Okinohama. It is not known when Okinohama first emerged from the sea, but we can say that Layer III was part of sandbar that were covered by water at high tide, creating a wetland. Layer II is surmised to have been created by human infilling, with artefacts dating back to the late 10th through early 11th centuries, though these were only found in small quantities. From this fact we can say that sometime before the late 10th century Okinohama emerged from the sea, but would have been covered by waves at high tide. Post holes have been found in Surface VI, though it is unclear what they represent: their size and irregular arrangement suggest that the buildings were not large, and were probably small huts. Remains from the early 11th century were confined to the centre of the excavated area, showing that at this stage Hakata’s urban area had not reached this point. It is also worth paying attention to the fact that remains found in Surface VI were found on Okinohama’s inner slope, towards the mainland, which was probably chosen to avoid the wind and water damage that would have been sustained to structures on the seaward side of the island.

Layer I, sitting on top of Surface V, was formed by wind and appears to have accumulated undisturbed. In-between Surface V and VI no remains were found, showing that after the brief period of habitation and small huts in the early 11th century, there was no human activity there until the late 12th century, and Okinohama was abandoned. Those remains discovered in Surface V included post holes, wells, and wooden caskets, indicating that, as these were the same remains as those found in the dunes that formed the urban area of Hakata, by the late 12th century the urban area of Hakata had already expanded to include Okinohama.

We have looked at the remains of the inner slope of Okinohama, but how did the seaward side of the island develop? Locality 111 yielded remains that have been read as a

stone wall for defence against the Mongols (Fukuoka 2002b), with foundations placed at four metres above sea level in a position which even in its current topography lies at the highest point of Okinohama. During these excavations it was found that on the seaward side of the wall there were no features discovered which dated to before the second half of the 15th century. Nor at Locality 68, where the possibility of the wall existing was high, were there any features earlier than the 15th century (Fukuoka 1992), or at any seaward point of the line of the assumed wall. From this we can say that the expansion of Okinohama as an urban district stopped at the Mongol wall, and did not expand further towards the sea Although it does not appear there was habitation on the seaward side of the wall, the natural development of the beach continued. From the line of the shore shown in Edo period maps, as well as others made closer to the medieval period (see Figure 3), the rate of expansion has been calculated to have been about one metre per year. If we assume that the land on its seaward side of the wall, built close to the shore line, was not used for 200 years, there would have formed 200 metres of sandy beach. This area was equivalent to the urban space of Okinohama that had expanded through land reclamation on the inner slope, and the Okinohama of the Muromachi period in the 14th century would have been a contained town with the backdrop of an expansive sandy beach.

Development of Okinohama

Excavation at Locality 42 indicates urbanisation in the latter half of the 12th century. In the excavations in advance of the Fukuoka subway at the Gofukumachi intersection, one can see incontrovertible evidence of land filling that connected Hakatahama and Okinohama at the beginning of the 12th century (Fukuoka 1988). This development clearly spurred subsequent urbanisation at Okinohama (Figure 4).

Important excavation results have been reported for the low land between Hakatahama and Okinohama, on the east side of the causeway (Fukuoka 1998, 2002a, 2001a, 2001b). All the sites were excavated to their lowest point, but whereas in Localities 119 and 121 the sandbar formed the foundation, in Localities 98 and 100 the foundation was a layer of sand accumulated near the mouth of the river, showing the existence of another sandbar in-between Hakatahama and Okinohama. The height above sea level of Locality 119 was 1.8 metres, and that of Locality 121, 2.6 metres, meaning that this sandbar was higher than the same layer excavated on Okinohama in Locality 42. The remains found in these sites date back to the late 12th century for Localities 98 and 100, and the early to mid 12th century for Locality 121. From these points we can say that Okinohama was made up of at least two sandbars, and the

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 74

The development of Hakata as a medieval port town

about the other craftsmen, traders, shrines and temples that would have existed, we can imagine a thriving and crowded city.

Conclusion

Figure 4: Map showing hypothetical topography and infilling between Hakatahama and Okinohama, and locations of excavations

causeway created through land reclamation was put in a location to connect these with Hakatahama. The remains from the early to mid 12th century were spread out on the top of the middle sandbar, with remains reaching the low point between the middle sandbar and Okinohama in the late 12th century. This gives us a sense of the development of Okinohama and the causeway between it and Hakatahama during this period.

It is also interesting to note that in Locality 119, considerable quantities of artefacts from the second half of the 8th century were discovered, suggesting the possibility of habitation in that period. The fact that these objects included bronze mirrors and officials’ bronze belt plaques means it is unlikely that these are simply intrusive; and it would be reasonable to assume, as those that reported these finds did, that they are from burials.

This would suggest that the middle sandbar existed in the Nara period. While this discussion is based on a report rather than first-hand observation, it appears that this sandbar was empty space between the 8th and 12th centuries, and urbanisation occurred in this area after land filling and the construction of a causeway, as is suggested by the distribution of the remains.

As mentioned above, urban development did not take place on the seaward side of the Mongol Wall until the 15th century, with all expansion occurring through infilling wetlands on the inland side. This area would have been a bustling trade locus: the Kaitō Shokoku-ki (Records of countries in the Eastern Sea), published in 1471 in Korea, records the names of six merchants in Okinohama who traded with Korea, including the Kamiya family, who managed the tribute ships sent to Ming China, and other prosperous traders who lived in the area. There were also a number of temples, such as Sekijōzan Myōrakuji and the Doi Training Hall of Shōmyōji. These temples and merchants of this period remain only as names in documents and represent only a piece of the full picture of medieval Hakata; when we think

Medieval Hakata developed out of two distinct places, Hakatahama and Okinohama, during the Muromachi period (1336-1753). Hakatahama was originally a commercial area flourishing at the end of antiquity as a residential area for Song Chinese merchants who had moved there from the Kōrokan in the latter half of the 11th century. They created a residential area, Hakata tsu tōbō, and they flourished here until the late 12th century.

Okinohama, on the other hand, was barely emerging from the sea when the Song traders first settled. To anyone looking from Hakatahama, it would have appeared as just a sandbar, though it would have served a function as a barrier from the wind and the waves. There was also a small sandbar between Hakatahama and Okinohama. In the early 1100s, the area between Hakatahama and these two sandbars was filled in, and people began to inhabit the middle dune. Land reclamation continued in the wetland area between the middle dune and Okinohama, so that by the late 12th century Okinohama encompassed both areas, with remains of this period throughout. On Okinohama, as a response to Mongol military invasions, a defensive wall was constructed along the seaward shore, and over the coming centuries the beach on the far side expanded.

In the Muromachi period, the new area, Okinohama, took over from the commercial zone that had developed on Hakatahama (Saeki 1987). By the end of the Kamakura period, when Hakata became the seat of the chinzei tandai, the two areas were no longer distinguishable as distinct sections of land. There are not many historical references to Okinohama from this period, but according to one document, the Hakata Diary (Hakata nikki), Okinohama was the seat of the Kikuchi clan of Higo in 1333. The diary records that a fire broke out in the middle of the night of the 13th day of the third month that year, and that two Kamakura officials, Musashi Jirō and Takeda Hachirō, suspected that it had been started as part of a Kikuchi rebellion. The diary also records that these officials passed Susaki, on the western side of Okinohama (the site of present-day Suzakimachi), on their way to Kushida Hamaguchi to suppress the rebellion. Kushida Hamaguchi is on the western side of Hakatahama, near the Kushida shrine, near the west side of Okinohama. This route would have taken them across the widest point of the wetlands between Hakatahama and Okinohama, a point that would have been impossible to cross by horse at this point. In the Edo period, local accounts show that, at this place, there was a bridge, 120 ken long (216 metres),

59

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 75

60

The archaeology of medieval towns: case studies from Japan and Europe

known as Minatobashi (Port Bridge), which may have even existed by the end of the Kamakura period. It is difficult to imagine that such a long bridge would be constructed for a place that had only a few scattered houses, so it is possible that by the end of the Kamakura period urbanisation had already progressed to the western tip of Okinohama.

In 1333, Emperor Go Daigo assigned Okinohama to Ōtomo Sadamune, governor of Bungo. Hakata had been the seat of the Shōni clan up to this point; they had been located in Kyūshū since well before the Kamakura period. From this time onwards, Okinohama and Hakatahama began to be treated as quite separate administrative entities, even though physically, they were linked more closely than before.

References

Batten, B. L. 2006. Gateway to Japan: Hakata in war and peace 500-1300. Honolulu, University of Hawaii Press.

Fukuoka Shi Kyōiku Iinkai 1988. Kōsoku tetsudō kankei maizō bunkazai chōsa hōkoku 7, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 193 shū (Report 7 of buried cultural properties associated with the high speed rail line, Fukuoka City report of buried cultural properties no. 193). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Fukuoka Shi Kyōiku Iinkai 1991. ‘Hakata 17’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 245 shū (‘Hakata 17’, Fukuoka city report of buried cultural properties no. 245). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Fukuoka Shi Kyōiku Iinkai 1992. ‘Hakata 32’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 287 shū (‘Hakata 32’, Fukuoka city report of buried cultural properties no. 287). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Fukuoka Shi Kyōiku Iinkai 2002b. ‘Hakata 85’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 711 shū (‘Hakata 85’, Fukuoka city report of buried cultural properties no. 711). Fukuoka, Fukuoka Shi Kyōiku Iinkai. Iso, N., Shimoyama, S., Ōba, K., Ikezaki, J., Kobayashi, S., and K. Saeki 1998. Hakata isekigun o meguru kankyō henka: Yayoi jidai kara kindai made, Hakata wa dō kawatta ka (Environmental change affecting archaeological sites: From Yayoi to modern times, how did Hakata change?). In: Kobayashi, S., Iso, N., Saeki, K., and H. Takakura (eds.) Fukuoka heiya no kokankyō to iseki ricchi—kankyō toshite no iseki to no kyōzon no tame ni (The Fukuoka plain and site location: how did sites relate to their environment?). Fukuoka, Kyūshū Daigaku Shuppan Kai: 69-112. Kamei, M. 1986. Nihon bōeki tōji shi no kenkyū (Study of Japanese trade ceramics). Kyoto, Dohosha.

Fukuoka Shi Kyōiku Iinkai 1998. ‘Hakata 64’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 559 shū (‘Hakata 64’, Fukuoka city report of buried cultural properties no. 559). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Ōba, K. 2003. Hakata. Kikan Kōkogaku 85: 33-36.

Fukuoka Shi Kyōiku Iinkai 2001b. ‘Hakata 78’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 669 shū (‘Hakata 78’, Fukuoka city report of buried cultural properties no. 669). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Saeki, K. 1987. Chūsei toshi Hakata no hatten to Okinohama (The development of medieval Hakata and Okinohama). In: Kawazoe, S. (ed.), Nippon chūsei shi ronkō (Discussion of Japanese medieval history). Matsudo, Bunken Shuppan: 425-445.

Fukuoka Shi Kyōiku Iinkai 2001a. ‘Hakata 77’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 668 shū (‘Hakata 77’, Fukuoka city report of buried cultural properties no. 668). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Fukuoka Shi Kyōiku Iinkai 2002a. ‘Hakata 81’, Fukuoka shi maizō bunkazai chōsa hōkokusho dai 707 shū (‘Hakata 81’, Fukuoka city report of buried cultural properties no. 707). Fukuoka, Fukuoka Shi Kyōiku Iinkai.

Ōba, K. 2001. Hakata kōshu no jidai (Hakata in the age of the Song traders). Rekishigaku Kenkyū 756: 2-11.

Ōba, K., Saeki, K., Suganami, S. and Y. Tagami 2008. Chūsei toshi: Hakata o horu (Medieval cities: excavating Hakata). Fukuoka, Kaichosha.

Yamauchi, S. 1989. Nissō no shōennai mitsubōeki ni kansuru gimon –jūisseiki o chūshin toshite (Questions relating to secret trade between Japan and Song China within manors, particularly in the 11th century). Rekishi Kagaku 117: 11-24.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 76

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe Senda Yoshihiro Introduction

Castles are among the most iconic monuments of the medieval period in both Japan and Europe. But whereas in Europe many of the castle walls and other structures that we see today are medieval in date, in Japan the great majority of castles were extensively remodelled under the Tokugawa shogunate. And again, many of these great structures were demolished during the Meiji period which saw the re-establishment of imperial power. Many other castles fell victim to fire and earthquake. The result is that the castles that today form so striking a part of the historic landscape of Japan, from Osaka to Edo, are mainly reconstructions, and those that retain extensive original features, including the UNESCO World Heritage Site of Himeji and others such as Matsumoto, date in the main from after 1600. During medieval times, however, a great diversity of castles and fortified buildings were constructed. A recent survey by Stephen Turnbull (2009) provides a comprehensive summary of their construction, history and use. Senda Yoshihiro here provides something quite different: a meticulously researched comparison of the processes, observable in the archaeological record, underlying the development of various types of encastlement in both Japan and Germany, where he has undertaken extensive research.

Castles (jōkaku) and fortified residences (kan) occupy a central position in Japanese medieval archaeology. Although both types of sites have many functions, they served an important role as centres in the development of medieval society. Castle towns developed from these sites. All over Japan, the basic structure of many villages and cities derives from the 16th century and many prefectural administrative offices were originally parts of castle towns of the 16th to 17th centuries. The connections between castle towns at the end of the medieval period and the beginning of the Edo period are very important. Historical documents and maps are significant for studying the transition. Site plans are particularly useful. However, many points cannot be determined without archaeological research on the actual sites. Excavations and surface mapping fill in aspects of daily

This chapter is an adapted summary of A Formation of a Medieval Castle Town by Senda Yoshihiro, published in the Bulletin of the National Museum of Japanese History in 1999, drafted by Richard Pearson and edited by Oscar Wrenn.

life including ceramic production and use of technology of all kinds. The material culture of these sites is very useful for studying political and social change.

Excavation and survey plans show the extent of enclosure of different kinds of buildings and the amount of land within the castle boundaries. Mapping the surroundings of castle enclosures is now an integral part of castle studies. At present the use of detailed maps along with historical documents and archaeological materials requires a team approach that brings out the features of each region. Excavation of castle sites needs a definite strategy since the sites are architecturally complex and the archaeological features are difficult to read, especially from small excavations. Despite these problems, excavation is now routine, and is carried out by government agencies such as local Boards of Education and Centres for the Study of Buried Cultural Properties. Recent articles using these approaches have been written by Senda (2002) and Saitō (2002).

The study of encastlement

This paper explores the formation of fortified castles and citadels in Japan and Europe, and in particular the process of settlement centralisation around walled castle sites. Various approaches have been taken to the study of medieval castles in Japan and Europe. In Japan, the work has tended to focus on their distribution and the particular features of their construction. The study of walled castle sites began as early as the 17th and 18th centuries, promoted both by the military, who wanted to learn the history of fortification, and by local amateurs. Study began early in Europe too, with surveys in the 19th century. Today, study of castles focuses on the distribution of sites, their regional characteristics, and distinctive structures (Antonow 1993, Friedrich 1994, Murata 1987, Ōrui and Toba 1936, Piper 1912).

Another approach, beginning in Europe after 1945 and introduced to Japan by Murata Shūzō in 1979 (Murata 1979),

61

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 77

62

The archaeology of medieval towns: case studies from Japan and Europe

Map of Japan showing key locations mentioned in the text

treats the castle as a historical document that can be read and interpreted (Murata 1979). This approach has contributed greatly to the study of towns, villages and walled castles in Japan (Maekawa 1991, Senda 1991a). Meanwhile, thanks to the development of medieval archaeology, the history of rural settlements has been uncovered through excavations. In particular, Haraguchi Shōzō (1977) and Hirose Kazuo (1985), focusing on nucleation from dispersed settlement patterns, explore the transformation of rural settlement from ancient times to the Edo period. In the field of historical

geography, Kaneda Akihiro (1985, 1993) has made comprehensive studies of the formation of agricultural fields in the ancient and medieval periods.

While these innovative studies have greatly helped to deepen our understanding, we need further investigation of the significant role played by walled castles in the centralisation of settlements and the development of medieval urban forms. Most previous studies have treated urban and rural issues separately, and there has been little investigation of the interaction between them.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 78

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

Toubert’s seminal study of encastlements in central Italy, ‘Les structures du Latium mediaeval’ (1973), highlighted an important direction to follow for considering how castles related to settlement formation (Kido 1990; 1995). Pointing out the limitations of urban-centred studies, Toubert placed his emphasis on small towns, categorised as ‘pre-urban’ or ‘semi-urban’ and of which there were a large number in medieval Europe, questioning how these should be evaluated. Instead of treating cities and villages as separate for research purposes, he argued that the relationships among settlements in a given area should be understood as a single organic system, and that analysis should focus on each settlement’s function within that system. He defined a town as any settlement that functioned as a centre of, for example, politics, the economy or religion and art in a certain area. He explored a number of small castle towns that emerged simultaneously in the 11th and 12th centuries in southern Italy, and it became clear that regional urban centres underwent a large transformation at this time. Toubert called this phenomenon incastellamento, or encastlement, and argued that this was a characteristic of town formation in Europe unique to the medieval period. Toubert’s analysis calls for a re-evaluation of the role of regional powers in urbanisation, with a focus on the role of castles in the centralisation of settlement functions. The view that a community of citizens free from the feudal system played a central role in urbanisation has been re-examined. And in addition to the questions of development within cities, examining urban functions using a broader perspective has meant that the questions of how these are maintained within regions, and how cities, through movement and changes in planning, have been created up to the present day, have emerged as important topics.

A typical example of encastlement in Europe is provided by the castle town of Roca San Silvestro, in Italy. Established in the 11th century, this castle town was formed through merchants and manufacturers, including makers of earthenware, bakers, blacksmiths and priests, congregating around a central castle, and subsequently creating a regional political and economic centre (Francovich et al. 1995).

Encastlement is now a firmly established academic term in Europe. Toubert’s concept enables us to acknowledge and analyse the pivotal role of castles as settlement patterns form in the course of regional, social and economic activities, viewing village and town as one organic unit, and acknowledging the pivotal importance of the role of castles in settlement formation. As mentioned above, in Japanese castle studies the relationship of castles to centralisation and the formation of settlements in the medieval period has been left unexplored. To fill this gap the paper examines Japanese

medieval castle towns and villages together in the manner of Toubert, incorporating archaeological analysis of castles and villages.

The next section will show that encastlement is found both in the nucleation of rural settlements, and in the creation of towns around castles. I will refer to the former as ‘rural encastlement’, and the latter as ‘urban encastlement’. To properly examine in lateral perspective the relationships between castle, town, and village and the process of encastlement, it is important not to treat the formation of towns and villages as separate entities. In this way it is vital we pay attention to the various stages of the formation of urban centre, and in particular the early stages of castle town formation, as well as the connection between centralisation of rural settlements and newly formed towns. Studies of Europe north of the Alps have largely neglected or dismissed encastlement, especially in regard to rural settlement formation. One reason for that is insufficient scholarly attention to the last phase of earthwork castles. The transitional role played by castle towns prior to the evolution of true towns in 12th century Europe should not be ignored. I hope to compensate for the lacuna to some limited extent in the course of this paper.

Rural encastlement in Japan

Can we find encastlement equivalent to southern Europe’s in medieval Japan? The Owari plain, in central Japan, provides a good example. Today, its central city is Nagoya, with a population of several million, though it only dates from 1610. Prior to that, the centre was Kiyosu castle town, five kilometres northwest of the centre of Nagoya.

According to Satō Kimiyasu (1989), seven medieval villages have been excavated in the Kiyosu area. The earliest formed around the middle of the 11th century, and during the 14th and 15th century five of these villages were active. The early 14th century was a transitional period during which the rural landholding system of manors or rural estates (shōen kokugaryō taisei) decayed, and the development of these five villages can be associated with a newly emergent ‘village system’.

This new system can be seen in the systematisation of arable land in the Kiyosu area, most notably rice paddies. Although government-allotted field plots laid out in the jōri system located in low-lying ground and square paddies had been organised in the Nara period, these did not survive through to the 14th century. It has been discovered through research on medieval villages that a medieval jōri, which was oriented differently from the previous system, was instituted at that time. Thus, formation of a new village system in the 14th

63

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 79

64

The archaeology of medieval towns: case studies from Japan and Europe

century can be associated with wholesale reorganisation of arable land and the cultivation of new fields. Concerning the development of farming technology, Ishio Kazuhito (1993) points out that at this time it became common for the general populace to use cattle and horses to plough fields, which, up to the mid Kamakura period, had only been used by the privileged orders. Ishio argues that this prompted a change from paddies to dry fields, which subsequently stimulated the coalescence of villages. On the basis of this analysis, it is possible to argue that reorganisation of settlement occurred with the development of more intensive farming.

Furthermore, in the 14th century, together with the priesthood, members of the upper orders, such as landowners, came to hold all decision-making rights in villages (Uemura 1979), showing significant transformation can also be seen in the functioning of village communities. These changes, of increased productivity and the reinforcement of communal groups (kyōdōtai), parallel agricultural and sociological developments in Europe from the late 10th to mid 12th centuries. The subsequent development of medieval towns depended on these conditions. Villages in the Kiyosu area were economically dependent on wet rice cultivation, and were located on slightly raised land surrounded by rivers for natural defences. Allocation of houses was normally based on the jōri system, but when dry fields were available on higher ground, settlement formation followed the pattern established by the dry fields.

No defences such as moats have been found in these villages. Some scholars regard the large ditch that was found in the south part of the Mori Minami settlement as a moat (Katō 1990), but an aerial photograph and cadastral map suggest that it is unlikely that the village was surrounded by this ditch, and therefore it is more likely to be a channel diversion.

In the late 15th century, these villages were suddenly abandoned. It is thought that this was caused by the construction of Kiyosu Castle and its town, after the existing castle was taken by the Shiba clan in 1478, subsequently becoming the residence of the local warlord (shugo daimyō). Excavations indicate that Kiyosu started to flourish at the same time that the surrounding villages disappeared. We can therefore surmise that Kiyosu castle would have been seen as a protective location, and villages from within two kilometres would have coalesced around the castle into a single town. This reorganisation of local villages did not contribute to the commerce and industry of Kiyosu. No artefacts or features associated with industry have been recovered from excavations of the villages, and it is clear that the people of the villages continued to live by agriculture.

In the case of Kiyosu, merchants and manufacturers, but also farmers and entire agricultural villages were relocated inside the castle town. The relocation of previously dispersed villages into the castle town permitted more intensive agriculture and also made it easier for farmers to sell their produce to city dwellers. An added advantage for the villagers was surely that they were safer living in a castle town than if they had been living in unfortified villages that were vulnerable to attacks and raids. In this sense the relocation of various villages was not necessarily solely at the behest of the landowners. The centralisation of settlements was achieved through the mutual interests of landlords, merchants, and farmers.

There were a number of cases of castle-based centralisation in the 14th century Owari plain. Ōmori Ozekigashiro, located in present-day Moriyama-ku, Nagoya, was founded by the Ozeki clan at about that time, remaining active until the mid 16th century, through a number of changes in the lords of the castle. Interestingly, a 1962 map shows a dry field called Motogō, the site of the village of the medieval village of Ōmori. The Ozeki lords worked closely on the development of the Ōmori area along the River Yada and they used their own castle to promote encastlement.

From Motogō across the Yada river to the south were various branch villages such as Inokoishi, and the temple located on the Motogō site was supported by the people of these villages. There was also the Hachiken Shrine located there, dedicated to clan deities, or ujigami (Kobayashi 1980). These facilities show that Ōmori served the role of a religious centre for the people of the surrounding villages, sharing common features with the construction of central religious institutions in European encastlement.

Ōmori’s feudal lord had craftsmen under direct control, but the town did not function strongly as an urban centre. In fact, Ōmori would have had a market only irregularly, functioning as the centre of any commercial activity in Motogō, with higher-level urban functions for the district performed by the castle town of Kobata, two kilometres to the west, and the area around the castle of Moriyama, four kilometres west of Motogō.

In a way similar to Kiyosu and Ōmori, a number of other castles appeared, in succession, elsewhere on the Owari plain between 1300 and 1500, with reorganization of villages around them. In total, 384 medieval walled castles have been identified in Owari, mostly encastlements from this period. Encastlement occurred not only in obviously suitable places like Kiyosu, which later became Owari’s central castle site, but also in local agricultural villages, more generally. Encastlement is seen elsewhere in Japan too at this time,

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 80

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

although the degree of development varied. We can distinguish this process of building castles and the reorganisation of villages around them as the first stage of encastlement; rural encastlement. In order to recognise variations of rural encastlement, the following is an overview of the process in other areas.

The area of today’s Chiba Prefecture witnessed the evolution not only of encastlement of the type shown at Owari, but also of fortified residences raised on low hills, in which the residence and village were enclosed in one fortification and surrounded by trenches or earth walls. Villagers’ houses were also placed on hill slopes on a series of terraces. Examples of this type of fortified residence, with both residence and village enclosed by fortifications, are Sasamoto castle in Hikari-chō, and Sasago castle in Kisarazu (Tōsō Bunkazai Senta 1995).

Although, at first, these look like ordinary castle sites, they were, in fact, rural encastlements. The limited difference in the relations between local lords and villages is revealed through this form of encastlement (Senda 1996), evidenced by the homogeneity of the artefacts discovered at each site (Ono 1995). It is also worth noting that these types of castle villages disappeared in between the late 15th and early 16th centuries and converged into a type with a central hill or platform, and the village coalescing on the low ground surrounding it. This form of encastlement was basically the same as that confirmed on the Owari plain and is thought to be the result of the reorganisation of encastlement after an increase in the power of local lords.

When we look at the fortified residences found in other parts of the country in the late 16th and early 17th centuries, for example Namioka castle in Aomori, and Chiran castle in Kagoshima, we can evaluate these as a continuation of the precedent set in Chiba, of an enclosed, fortified area containing castle and village. In these areas the power of feudal lords was relatively weak, and clans such as Shimazu in southern Kyūshū brought many of the local farming and commercial population into the samurai class, with the form of encastlement comprising both village and castle reflecting the social composition of the community.

On the Saga plain of north Kyūshū, encastlement occurred around wealthy farmers’ houses in the 13th and 14th centuries as a function of reclaiming low-lying coastal land for paddy by wealthy farmers and minor feudal lords. In the 16th century, this type of rural centralisation had taken two forms (Figure 1). In villages where those from the upper social classes had become samurai through feudal lordship, fortified residences were built through piling up earth and creating large ditches, with the houses of those living in the village grouped by status around the castle; Anekawa castle is a

typical example of this. The second form of village found on the Saga plain did not develop to the same extent as those with high social stratification, and was just a group of houses surrounded by a moat; the best example of this is Kami Rokuchō. In both of these kinds of settlement, the castle towns seem to have functioned as religious centres, since there is evidence for a certain concentration of temples with limited development of commerce (Miyatake 1995, Kawano 1996).

In the middle of the Kansai region, the most urbanised in the medieval period, there were various kinds of rural encastlement, depending on relations between the feudal lord and village communities. Moated castles appeared in the late 12th century in sites such as Nagahara in Ōsaka, and somewhat later in other areas. Centralisation of rural settlements started in the late 13th century. In the late 14th century encastlement could be seen elsewhere in Ōsaka, with various crafting areas within the fortifications.

In the area of modern-day Shiga Prefecture, reorganisation of villages and moated settlements began after the 13th century (Kido 1992), but the buildings surrounded by a moat were not just houses of high status people, but also often temples. In the late medieval period, in areas where a lord imposed strong control, a town would have one central castle, but where villages maintained relative independence from the feudal lord, several fortified residences would be built in one village (Senda 1991b). Both examples, however, show forms of rural encastlement.

Urban encastlement in Japan

The encastlement of urban areas developed during the Ōnin Wars as local rulers sought to defend their territories, and castle towns replaced rural encastlements as Japan’s main form of fortified town. In Yamaguchi, in the fortified town governed by the Ōuchi shugo daimyō, a new castle was constructed in the early 15th century and the adjacent castle town formed rapidly in the surrounding areas in the early 16th century (Koga 1994). This development may have been the effect of the activity of craft workers in the lords’ employ.

In Mino, the stronghold of the shugo daimyō was Kawate castle. The distribution of haji earthenware found in archaeological sites suggests that a large castle town had formed by the 16th century (Takada and Uchibori 1994). From the late 16th century the stronghold was moved, successively, from Kawate to Fukumitsu, Edahiro, Ōkuwa and finally the Inabayama castle town. With each shift, the samurai, travelling merchants and artisans were brought into increasingly enclosed areas. In Inabayama the presence of district names from previous towns shows how craftsmen moved from place to place

65

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 81

66

The archaeology of medieval towns: case studies from Japan and Europe

(Kojima and Senda 1994).

There are a number of examples of urban encastlement in the 15th and 16th centuries, and several phases have been identified in this process of centralisation. The first, in the 15th century, is the formation of the castle town ruled by the resident daimyō (shugo jōkamachi). At this stage, the castle town functioned as the political centre but economic activity was greater at towns which developed around temples and shrines (monzen machi) and in port towns (minato machi). The second phase is the expansion of towns around the castles of provincial lords, the shugo jōkamachi which had first appeared in the 15th century, which occurred in the early 16th century. The expansion resulted from the influence of rural style encastlement on the formation of castle-centred towns as in the examples seen above. From around the mid 16th century, the Warring States period produced more castle towns (sengoku jōkamachi). The daimyō strongholds now became mountain castles or castles defended at multiple levels (Senda 1994), and they served both as residence for the daimyō and as political centre. At the same time, the internal structure of the castle towns changed as residences became arranged according to class divisions.

The last phase took place in the late 16th and early 17th centuries, as castle towns achieved primacy over those around temples and shrines and other urban spaces. Spatial distribution came to be almost entirely determined by socio-economic divisions: farmers and fishermen inhabited distinct agricultural and fishing villages while samurai and traders lived in towns. The division between urban and rural established during this period prefigured the basic structure of villages and towns still seen today. These modern castle towns were structured in such a way that the strict social stratification of daimyō, chief vassals, samurai, and normal townsfolk was reflected in the layout of the town, with the daimyō’s residence at the highest point in the centre. The castle precinct would sit at this highest point, and this stratification would radiate out to the castle town below (Senda 1996). As summarised above, while the formation of towns in 15th to 17th century Japan can be divided into several phases, it was castles that constantly functioned as the core of the centralisation of settlements and the formation of urban space. This period represents Japan’s second stage of encastlement: urban encastlement. Through these processes, political, economic, and commercial networks were systematised throughout the Japanese archipelago. Two clear stages are thus recognised in Japanese encastlement, suggesting a somewhat different process to what is seen in Germany, where we will now turn our attention.

Map of Germany showing key locations mentioned in the text

Encastlement in Germany

Using the model of second-stage encastlement, urban encastlement, which we have seen in Japan, I would like to re-examine the claim that the formation of settlements around castles has not been found in medieval Germany.

A good example is Rothenburg ob der Tauber and surrounding parts of Bavaria. Rothenburg has reconstructed its medieval cityscape and is now a world-famous tourist attraction. Although the centre did not survive completely intact, what remains of the old town does enable us to experience something of the quality of German medieval urban space. The development of Rothenburg has been attributed to its recognition as a special city by Emperor Barbarosa. In 1172, Rudolf I of the Hapsburgs gave it special status and designated it as a free city. Focusing on urban space in the surrounding area and the relation between the disappearance of castles and the establishment of Rothenburg, the formation of the town will be examined from an archaeological point of view.

Rothenburg is said to have begun with a castle built atop of a bluff over the River Tauber in the 11th century, on the site of a little 10th century church. In the 12th century the castle expanded, and although it has only been partly confirmed by the archaeology, it is surmised that in front of the castle a large urban space formed. There is no doubt that designation as a

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 82

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

chartered city promoted the development of a large town in front of the castle wall.

In the surrounding areas of Rothenburg, sixteen medieval castles have been found (Dannheimer and Hermann 1968). Thirteen of these were small fortified residences with no sign of the development of the rural form encastlement. Given that there were a large number of villages in this area, we can say that most of the villages had neither castles nor any equivalent buildings.

Suizu Ichirō’s precise study on the establishment and development of medieval settlements in Germany, based on field and archival research from the viewpoint of historical archaeology, does not touch on any example of castle-centred encastlement (Suizu 1976). Similarly, no attention is paid to the relationship between castles and settlements by Fehring in his overview of medieval archaeology in Germany (Fehring 1992). As has been pointed out, it is unlikely that rural encastlement occurred widely in Germany. If we take the region around Rothenburg as typical of medieval Germany, we can say that rural encastlement was not a widespread phenomenon.

Three medieval castles around Rothenburg, however, are worth noting: Endsee Altes Schlos, Nordenberg Schlosbuck, and Insingen. They were unusually big and had many features in common with their plans. Endsee and Nordenberg in particular look strikingly similar: in each, the main castle (Kernburg) was interconnected to the outer enclosure (Vorburg). Endsee is relatively well documented, as it was used by the Herrn von Entse in the 11th century, but in the early 12th century the lord and his family built a new residence and subsequently moved to Rothenburg. In 1408, Endsee was completely destroyed, along with other castles.

On the basis of this historical background, the visible remains, and repeated excavations, it seems most likely that this form of castle dates from the late 11th to 12th centuries. It was defended by an earthen ramp and ditch, and the presence of the remains of a part-stone tower in the centre seem to fit the appearance of that time. The structural similarity of Nordenberg and Insingen also suggests that they were built around the same time. It is also possible to speculate that other castles of the time around Rothenburg would have been structurally similar to Endsee Castle. How, then, did the inner castle and outer enclosure carry out their respective functions as urban centres? Tilleda, one of the German Emperor’s own castles located in in the Zanhauser area of Saxony Anhalt, provides some clues. Paul Grimm started surveys of Tilleda in 1958 and excavation was carried out until 1979 (Grimm 1968, 1990), with the site subsequently opened as a public park in 1993, featuring reconstructed

buildings, and moats and stone mounds marked out on the ground.

Tilleda Castle was partially constructed of stone and had moats and earthen ramparts, and comprised a main fortified castle with the moat within which was contained the outer enclosure. At the height of occupation, the foundations of this outer enclosure were also made of stone and there were circuit walls and a watch tower. Within the inner enclosure, in the castle section at the top of the hill, was built a palace and church. In the outer enclosure there were workshops and dwellings: houses of artisans working ivory and bone, smiths of copper, bronze and iron, potters and weavers have been excavated. The castle was one of most important strongholds for management of the imperial domains of Harz and Saxony for Emperors Henry IV and V in the late 11th century and the first half of the 12th. The excavations show that the town can be considered as the incipient centre of commerce: products from the surrounding territories were stored, processed and worked by the inhabitants of the outer enclosure, and they were sold in Tilleda itself.

Tilleda shows that the appearance of a large-scale outer enclosure would have become necessary to accommodate the movement of craftsmen to the castle. Similar outer enclosures would have worked in in the same way at Rothenburg, Endsee, Nordenberg and Insingen in the same period, and such aggregation in Rothenburg led it to become the centre of its area. The expansion of Rothenburg in the 12th century, and its conferred royal privileges, would have been prompted by the movement of craftspeople from smaller towns of more elementary organisation, a result of it out-competing these towns with regards to urbanisation. At Tilleda as well, the outer bailey’s manufacturers became the driving force toward centralisation of urban functions in the district.

Not all sites could adapt in this way, however. In response to the Saxon feudal lords’ rebellion in 1115, the Emperor Henry V built Rithauser Castle. Since it sat at more than four hundred metres up, it was impossible to build a town around the castle. Tilleda also had ramparts with defensive posts, and functioned as a military base for a time, though declined in this function in the early 12th century.

From this time the palace became deserted and only the church remained in use, with the courtyard turned into a cemetery. The artisans left the outer bailey for the foot of the hill in the late 12th century, where they built the Tilleda which lasts until today. The establishment of the city of Tilleda represents the process of urban encastlement as seen in the Rothenburg area.

67

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 83

68

The archaeology of medieval towns: case studies from Japan and Europe

In regard to the management of territory, there was no great difference between Emperor and feudal lords. The scale of the main castle and outer bailey is almost the same for each of the area’s castle towns. Grimm (1968) points out that Grimschleben, Werla, Rothenburg, Kauschberg and Freckleben were each provided with a larger outer space just like Tilleda, and he argues that castles equipped with both main enclosures and outer enclosures were widespread from the 10th and 12th centuries (Grimm 1968: 88-90). There are a number of other examples of this type of castle town elsewhere in Saxony, such as Oschatz and Ostro. Thus, Tilleda and the Rothenburg area illustrate the period’s standard style of walled castles.

It is important to note the church that was erected in Tilleda’s main enclosure. This suggests the role the town played in religious life, also shown in other locations such as the Dohna castle in Saxony, where a church of St Petri in the outer bailey served as parish church for both the people in the castle and the surrounding villagers (Spehr 1994). At this time one of the functions which a castle town performed was as a local religious centre.

These examples have shown the extent to which German castle towns were provided outer enclosures after the 10th century. In the 11th century, the towns’ urban functions of castle towns increased on account of centralisation of manufacturing and worship and, in the 12th century, small walled castles performing urban functions were superseded by larger central castle towns. Such processes of urban centralisation are paralleled by Japan's second stage of encastlement.

It is clear that urban encastlement was widespread in Germany, occurring in various regions. Fehring (1992) quotes Herman’s study of the formation of medieval towns in eastern Germany, which proposes that 9th to 11th century incipient walled castles grew into urban centres in the 12th and 13th centuries. Spandau Castle in Berlin is a typical example of incipient encastlement in northern and eastern Germany in the 11th century, showing great influence from the Slavs. Excavations exposed a number of manufacturing areas, offices, shops and dwellings in the outer bailey (Fehring 1996). The backdrop to this establishment of urban centres was the competition and eventual consolidation of a number of smaller castle towns. The functions that each of these smaller towns performed, whether religious or commercial, were brought together and absorbed into single large urban centres, representing a key aspect of urban encastlement.

Although this form of encastlement has been little recognised by previous studies in Germany, we can re-evaluate this based

on the widespread clustering of manufacturers around the castle, just as has been identified in Japan with the second stage of encastlement. While encastlement occurred in both Japan and Europe, Germany was distinctive in that it underwent only urban encastlement.

The historical contexts for the formation of castle towns

This paper has examined and compared the formation of walled castles which provided the central point for surrounding settlements in medieval Japan and Europe. It has shown that while these castles existed in both Japan and Europe, both areas had certain distinctive characteristics.

When analysing the centralisation of settlements as forms of encastlement it becomes clear that, unlike Germany, Japan shows both rural and urban encastlement. This divergence highlights the basic structure of encastlement in Japan. Eventually, in the early modern period, from the 17th to 19th centuries, it gave rise to the most developed form of encastlement, the castle town (jōkamachi).

For Europe it has been remarked that encastlement is not often seen north of the Alps. That perception has arisen largely because studies of the last phase of earth and timber castles, the focus of this paper, fall into the gap between the academic disciplines of history and archaeology, and has not received sufficient scholarly attention. Archaeological analyses used for Japanese castles are extremely useful when looking at this phase of castles. As medieval castles have begun to attract attention in England and Germany, there will be a re-assessment, in due course, of encastlement north of the Alps (Böhme 1991, RCAHMW 1991)

In both Japan and Europe, as larger towns appeared, castle towns fell into disrepair and became derelict, or were absorbed into larger conglomerations. In Europe, many castle towns established in the 11th and 12th centuries were blueprints for 13th century towns. New towns could function as urban centres more strongly by absorbing other castle towns. The disappearance of castle towns and the displacement of political, religious and cultural centres can be explained by economic rivalry and other practical considerations.

Excavations of castle towns, however, suggest that other, noneconomic, factors played a part in these changes, including the impact of inhabitants’ intentions and decisions. After the 13th century, some medieval cities in Europe attained their independence from emperors, feudal lords and bishops through a citizen movement toward a commune, centred on the merchants and craft guilds of the city. In the 14th century attempts in various places by manufacturer’s guilds to

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 84

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

participate in city councils resulted in the guild placing its members on the city council. In some cases, the guild occupied every seat.

Thus, after the 12th century, towns were managed by independent communes and democracy was promoted. Not, of course, that communes were inherently anti-feudal; but they sought independence within the framework of the feudal system. Independence of the communes is no longer regarded as the essential condition for the evolution of medieval towns. Rather, the inhabitants’ aspiration to independence and freedom was one of the major factors that characterised medieval towns. In Lübeck, excavations of shops, studios and the city walls have demonstrated intentions and power of the inhabitants (see Gläser, this volume). In contrast, the excavation of Tilleda Castle showed that the merchants and manufacturers who occupied the outer enclosure were clearly subordinate to the lord. By the 12th century the castle town clearly became incompatible with the aspirations of the urban citizenry. From the broader historical perspective this explains the reason why the castle town played only a transitional role.

Interestingly, the interim nature of the European medieval castle town is similar to the development of large moated settlements in Japan in the Yayoi period. Referring to a number of examples from around the world, Tsude Hiroshi considers them as defensive settlements and characterises them by their transitory use, since this type of settlement was constructed exclusively for defence and was only used only during times of political strife (Tsude 1997).

Although it is difficult to compare the Yayoi period with the Middle Ages in Europe, there were political aspects in common: central government had lost its power, the state system was in the process of reformation, and small factions were in conflict. Yet the transformation of castle settlements in medieval Europe was directed toward forming more appropriate urban centres. Some of these transformations were forced through by authorities but some were affected through the intentions or agency of the townspeople, which is not the case in the Yayoi period. The large moated communities of the Yayoi period were succeeded in the Kofun and Ancient periods, by towns subdivided into four districts by fences strengthening the attachment of craftsmen more closely than before to rulers and towns. In the capital, Heiankyō, chō divisions were eliminated and the tanzaku system of long narrow lots facing the main streets appeared in the 12th century. These transitions underpinned rural encastlement in the 13th century and urban encastlement in the 15th. In the 16th century, wealthy communities appeared in cities such as Sakai

and Kyōto. Yet, as the shokuhō political structure succeeded in imposing stability from above, and the feudal lords were located in the imperial court, it was these urban communities that were used as the basis for the management of castle towns (Niki 1997).

The new system of residential land division was systematically unified with the rectangular system of merchants’ allotments under the lords’ control. Of course, the castle towns of the early modern period (kinsei jōkamachi) would not have developed without popular consensus; but the centripetal structure, with castles at the centre, show that local authority played a leading role in urban planning. The roots of major Japanese cities today, to be found in 13th century encastlement, indicate the significance of this process.

Acknowledgements

I wish to thank Dr Sahara Makoto for providing me with the opportunity to study in Germany. I would also like to give thanks to Dr. Eckhart Schubert of the German Archaeological Institute, Roman-Germanic Commission, who helped me when I was in Germany in 1995 and 1997, and Dr. Anna Helena Schubert of the Westfalen Cultural Properties Commission.

69

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 85

70

The archaeology of medieval towns: case studies from Japan and Europe

References

Antonow, A. 1993. Planung und Bau von Burgen. Frankfurt am Main, Alexander Antonow.

Böhme, W. (ed.) 1991. Burgen der Salierzeit. Teil 1/2 RomischGermanisches Zentralmuseum. Sigmaringen, Jan Thorbecke. Dannheimer, H. and Hermann, F-R. 1968. Rothenburg o. T. Katalog zur Vor und Fruhgeschicte in Stadt und Landkreis. München, Lassleben.

Fehring, G. P. 1992 (2nd edition). Einführung in die archäologie des Mittelalters. Darsmtadt, Wissenschaftliche Buchgesellschaft. Fehring, G. P. 1996. Stadtärchaologie in Deutchsland. Archäologie in Deutschland, Stuttgart, Sonderheft, Konrad Theiss: 25-26.

Francovich, R., Guideri, S. et al. 1995. Rocca San Silvestro. Siena, Universita Degli Studi Siena.

Friederich, K. 1994. Burgen des Deutschen Mittelalters Grundriss Lexikon. Wurzberg, Weldlich. Grimm, P. 1968. Tilleda, eine Königspfalz am Kyffhäuser: I, die Hauptberg. Berlin, Akademie.

Grimm, P. 1990. Tilleda, eine Königspfalz am Kyffhäuser: II, die Voorburg und Zusammenfassung. Berlin, Akademie.

Haraguchi S. 1977. Kodai chūsei no shūraku (Villages of the Ancient and Medieval Periods). Kokogaku Kenkyu, 23(4): 16-23.

Hirose, K. 1985. Chūsei e no taidō (Indications of the beginning of the Medieval Period). Iwanami Koza Nihon Kōkogaku 6: 296-356.

Ishio, K. 1993. Chūsei teichi shūraku no keisei to tenkai (The formation and development of lowland settlements). Historia 138.

Kaneda, A. 1985. Jōri to sonraku no rekishi chirigakuteki kenkyū (Historical geographical study of the Jōri field system and villages). Tokyo, Daimeido.

Kaneda, A. 1993. Bichikei to chūsei sonraku (Detailed topography and medieval villages). Tokyo, Yoshikawa Kobunkan. Katō, Y. 1990. Mori Minami iseki (The Mori Minami site). Jinnki, Jinji Chō Kyōiku Iinkai.

Kawano, S. 1996. Anekawa jōseki (The Anekawa Castle site). Kanzaki, Kanzaki Chō Kyōiku Iinkai.

Kido, M. 1992. Mizube no shūraku no genfūkei (The original setting of shoreline villages). In: Watanabe, M. (ed.) Mizumi no kuni no rekishi o yomu (Reading the history of the lake country). Tokyo, Shinjimbutsu Oraisha: 140-174.

Kido, T. 1990. 10-13 seiki nan’ō shakai no shiro to teijū (Society and stability of castles in southern Europe in the 10th to 13th Centuries). Shigaku Zasshi 99(6): 65-82. Kido, T. 1995. Incasuteramento, sonrakuka, toshi (Encastlement [incastellamento], formation of villages, and urban space). In: Egawa, A. and Hattori, Y. (eds.) Seiō chūseishi chū (European medieval cities, middle volume). Tokyo, Minerva Shobo: 129-150.

Kobayashi, H. 1980. Yadagawa monogatari (Account of Yadagawa). Aichi, Aichi Ken Kyōdō Shiryō Kankōkai. Koga, N. 1994. Shugo daimyō Ōuchi shi no kyokanseki to jōka Yamaguchi (Fortified residential sites of the shugo daimyō Ōuchi clan and castle towns of Yamaguchi). In: Kaneko, T. and Maekawa, K. (eds.) Shugosho kara sengoku jōka e (From the sites of the shugo daimyō to the castle towns of the Warring States period). Tokyo, Meicho Shuppan: 171-198.

Kojima, M. and Senda, Y. 1994. Shiro to toshi (Castles and urban spaces). Iwanami Kōza Nihon Tsūshin 10: 195-232.

Maekawa, K. 1991. Toshi kōkogaku no kenkyū (Study of the archaeology of urban spaces). Tokyo, Kashiwa Shobo.

Miyatake, M. 1995. Saga heiya no mura to yakata (Villages and elite residential sites of the Saga Plain). In: Amino, Y. and Ishii, S. (eds.) Chūsei no fūkei o yomu (Reading the medieval landscape) 7. Tokyo, Shin Jimbutsu Oraisha: 54-103.

Murata, S. 1979. Jōseki chōsa to sengoku shi kenkyū (Castle site investigations and the study of the Warring States). Nihon Shi Kenkyūkai Taikai Hōkoku, 1980 Nihon Shi Kenkyu 211: 82-106. Murata, S. (ed.) 1987. Zusetsu chūsei jōkaku jiten (Illustrated sictionary of medieval walled castle sites) Volumes 1-3. Tokyo, Shinjimbutsu Oraisha.

Niki, H. 1997. Kūkan, ōyake, kyōdōtai (Space, public place, and communal groups). Tokyo, Aoki Shoten.

Ono, M. 1995. Sasamoto-jō no yakimono ga kataru koto (Information from Sasamoto Castle excavated ceramics). In: Tōsō Bunkazai Sentā (ed.) Oomigaeru Sasamoto jōseki (Bringing Sasamoto Castle to life). Tōsō, Tōsō Bunkazai Sentā: 41-50.

Ōrui, S. and Toba, M. 1936. Nihon jōkaku shi (The history of Japanese walled castles). Tokyo, Yuzankaku. Piper, O. 1912. Burgenkunde. München, Piper and Co.

RCAHMW (Royal Commission on the Ancient and Historical Monuments of Wales) 1991. Glamorgan: early castles. London, HMSO. Saitō, S. 2002. Chūsei tōgoku no ryōiki to jōkan (The territory and defensive sites in medieval Eastern Country). Tokyo, Yoshikawa Kobunkan.

Satō, K. 1989. Kiyosu shūhen no chūsei sonraku (Medieval villages in the vicinity of Kiyosu). In: Tōkai Maizō Bunkazai Kenkyūkai (ed.) Kiyosu kenkyū hōkoku hen (Kiyosu, edited research reports). Tōkai, Tōkai Maizō Bunkazai Kenkyūkai: 3-11. Senda, Y. 1991a. Jōkaku kenkyū no kōsō (Plan for Study of Walled Castles). In Ishii, S. and Hagihara, M. (eds.) Chūsei no shiro to kōkogaku (Medieval castles and archaeology). Tokyo, Shinjinbutsu Oraisha: 21-52.

Senda, Y. 1991b. Mura no jōkaku o meguru itsutsu no moderu (Five models for the study of village castles). Chūsei shi kenkyū 16: 103-126.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 86

The formation of medieval castle towns: a comparative archaeology of encastlement in Japan and Europe

Senda, Y. 1994. Shugosho kara sengokuki kyoten jōkaku e (From the places of the shugo to the castles as central places). In: Nara Daigaku Bunkazaigaku Ronshū Kankōkai (ed.) Bunkazaigaku Ronshu (Essays on cultural heritage studies). Nara, Nara Daigaku Bunkazaigaku Ronshū Kankōkai: 457-468.

Senda, Y. 1996. Kinsei daimyō to ryōkoku shihai (Modern period daimyō and the power of the national domain). In: Machida, A. (ed.) Kōkogaku ni yoru Nihon rekishi 5: seiji (Japanese history through archaeology 5: Politics). Tokyo, Yuzankaku: 98-104.

Senda, Y. 2002. Jōkaku kenkyū no yukue (The direction of castle studies). In: Kokuritsu Rekishi Minzoku Hakubutsukan (ed.) Jinrui no totte tatakai to wa 4: kōgeki to bōei no kiseki (What does war mean to human beings 4: The locus of attack and defence). Tokyo, Toyo Shorin: 122-144. Spehr, R. 1994. Christanisierung und früheste Kirchenorganisation in der Mark Meisen. In: Oexle, J. (ed.) Fruhe kirchen in Sachsen. Stuttgart, Konrad Theiss Verlag: 9-63.

Suizu, I. 1976. Yōroppa sonraku kenkyū (Study of European villages). Tokyo, Chijinshobo.

Takada, T., and Uchibori, N. 1994. Mino ni okeru 15, 16 seikidai no shugosho no hensen (Changes in the places of the shugo in Mino in the 15th and 16th centuries. In: Kaneko, T. and Maekawa, K. (eds.) Shugosho kara sengoku jōka e (From the sites of the shugo daimyō to the castle towns of the Warring States period). Tokyo, Meicho Shuppan: 109-140. Tōsō Bunkazai Sentā 1995. Yomigaeru Sasamoto jōseki (Bringing Sasamoto Castle to life). Tōsō, Tōsō Bunkazai Sentā.

Toubert, P. 1973. Les Structures du Latium mediaeval, Le Latium meridional et la Sabine du IX Siecle a la fin du XII Siecle (The structures of medieval Latium: southern Latium and the Sabine from the 9th to 12th centuries). Rome, Ecole Francaise de Rome.

Tsude, H. 1997. Toshi no keisei to sensō (Formation of urban spaces and warfare). Kōkogaku Kenkyū 44(2): 41-57. Turnbull, S. 2009. Strongholds of the samurai: Japanese castles 250-1877. Oxford, Osprey Publishing Ltd.

Uemura, K. 1979. Chūsei Owari no mura ni tsuite (Concerning villages in the Owari area in medieval times). Nagoya Shi Hakubutsukan Kenkyū Kiyō 2: 1-14.

71

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 87 8

72

The archaeology of medieval towns: case studies from Japan and Europe

Five medieval European towns: Bruges, Göttingen, Norwich, Ribe and Rouen – a pictorial introduction Simon Kaner One of the objectives of the 2004 conference was to provide a forum to bring specialists in Japanese medieval urban archaeology together with some of their counterparts in Europe, to foster greater understanding of medieval Japanese archaeology (see also Souyri 2013). The five short case studies in this chapter represent five offerings by archaeologists responsible for some of the most significant excavations of medieval towns in northwestern Europe. The selection was made by Brian Ayers and reflect aspects of his network in European medieval urban archaeology (Ayers 2016). They are intended as a response to the more detailed discussions of the Japanese cases in the preceding chapters and lead into the second part of this volume, which include a detailed case study (Lübeck) and two essays: the first, by Galinié on economic networks of which towns were such an important part; and the second, by Vroom, a reflection on the importance of ceramic production in the medieval Aegean, which the editors feel speaks to some of the observations on medieval ceramics in Japan, discussed in particular to Ono’s contribution in Chapter 3.

Each author was requested to provide a small selection of images, a timeline summarising some of the key events in the history of their town, and a brief introduction. These five picture essays briefly introduce the archaeological work that has taken place in a small selection of northwest European medieval towns. Our original intention was to match these picture essays with five further examples of Japanese towns, but we eventually decided that the coverage of Japanese examples in the other chapters in this volume were sufficient (Hakata, Ichijōdani, Kamakura, Sakai, and Tosaminato). These essays do, however, raise two specific issues which we feel are deserving of further elucidation: trading networks and maritime technologies.

Trading networks

Trade was a critical factor in the development of each, as detailed in Brian Ayers’ (2016) study of the development of medieval urban forms around the North Sea. The development of medieval guilds and the Hanseatic League was symptomatic of the networks and frameworks of

medieval commerce, which often took place against a backdrop of military engagement, while at the same time facilitating the rise of a new middle class of merchants who spent at least some of their fortunes on cultivating their taste in art and culture. Indeed it is said that the northern Renaissance would not have taken place had it not been for the Hanseatic League. In Japan, investigations from the 1960s to the 1990s at Kusado Sengen on the Seto Inland Sea coast of western Honshū, in modern Hiroshima prefecture, revealed the well-preserved remains of landings, workshops, storehouses and other buildings associated with a port that grew up on sand bars in the mouth of the Ashida River (Matsui 1994, Matsui and Inoue 2012). Chinese coins and wooden tags for other goods from China, along with the half a million other artefacts indicated that this was a key harbour on the trade routes between the Ōsaka/Kyōto region, Kyūshū, and the wider East Asian world. The exchanges that were so important at this time were highlighted by a special exhibition at the National Museum of Japanese History in 2005, which also presented the northern trade routes through the Japan Sea to Hokkaidō and what is now the Russian Far East. Complementing the finds at Tosaminato (see Chapter 2), port facilities including boat landings large enough to accommodate large ocean-going vessels, over 170 buildings, workshops, wells and burials were found at the sites of Okite and Nakazu, again located in river estuaries but here opening onto a coastal lagoon in modern-day Masuda city, Shimane prefecture on the Sea of Japan coast of western Honshū (Japanese 2010), dating from the 11th to the 16th centuries, at which point changes in coastal topography favoured the development of the slightly further inland at Imaichi.

Maritime technologies and piracy

Underwater archaeology in recent decades has transformed our understanding of the medieval maritime technology that was so central to the development of trading networks both in Japan and Europe. Discoveries such as the Mongol ship at Takashima on the coast of Kyūshū, and the Sin’an wreck off the coast of Korea, bring to life, respectively, the Mongol invasions of the 13th century and the scale of East Asian

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 88 8

Five medieval towns - a pictorial introduction: Bruges

trade, which was so important for the development of urban forms in Japan (Kimura 2016). In Europe, the Ijssel cog, a twenty-metre long, fifty-tonne flat-bottomed wooden ship from the early fifteenth century, discovered in 2011 in the Ijssel River in the Dutch city of Kampen provided a rare actual survival of one of the ships that were the workhorses of the seas and rivers connecting the Hanseatic League (Ghose 2016). The rich pickings of the trade networks of course gave rise to a further issue: piracy, a topic also illuminated by recent studies in Europe (e.g. Heebøll-Holm 2013) and Japan (Shapinsky 2014). The very public fates of spectacular warships such as the Vasa and Mary Rose are indicative of the flamboyant folly that resulted from royal patronage of ship-building, reminiscent of the failure of the Mongol invasions. It was those very ship-building skills that brought the first Englishman to Japan, William Adams (Milton 2003), who went on to become advisor on shipbuilding to Tokugawa Ieyasu, the first shogun of the Edo period, whose victory at Sekigahara in 1600 brought the Japanese Medieval period to a close.

References

Ayers, B. 2016. The German Ocean: medieval Europe around the North Sea. London, Equinox Publishing Ltd.

Ghose, T. 2016. Medieval shipwreck hauled from the deep. Live Science. https://www.livescience.com/53734-dutch-shipwreckhauled-from-water.html (accessed 26/05/2020)

Heebøll-Holm, T.K. 2013. Ports, piracy and maritime war: piracy in the English Channel and the Atlantic, c. 1280-1330. Leiden, Brill

Japanese Archaeological Association 2010. Okite and Nakazu: at a port and settlement, where ships of foreign seas came and went, facilities of a medieval harbour on the Japan Sea. http://archaeology.jp/sites/2010/okite.htm. (accessed 26/05/2020). Kimura, J. 2016. The archaeology of East Asian shipbuilding. University of Florida Press.

Matsui, A. 1994. Kusado-Sengen: a medieval town on the Inland Sea. In Burenhult, G. (ed.) New World and Pacific civilizations: culture of Americas, Asia and the Pacific. San Francisco, HarperCollins. Matsui, A. and Inoue, T. 2012. Wetland sites in Japan. In Menotti, F. and O’Sullivan, A. (eds.) Oxford handbook of wetland archaeology. Oxford, Oxford University Press: 175-194. Milton, G. 2003. Samurai Williams: the adventurer who opened Japan. London, John Murray.

National Museum of Japanese History 2005. Interaction in the Medieval East Asian Sea. Sakura, National Museum of Japanese History.

Shapinsky, P.D. 2014. Lords of the sea: pirates, violence and commerce in late medieval Japan. Ann Arbor, Center for Japanese Studies, University of Michigan. Souyri, J.P.(ed.) 2013. Le Japon Médiéval. Au temps des samouraïs. Archéothema 30.

73

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 89

74

The archaeology of medieval towns: case studies from Japan and Europe

Brugge (Bruges)

Hubert de Witte and Katelijne Vertongen Bruges, called Brugge in its native Dutch (which presumably was used to signify landing stage), is the capital of the province of West Flanders in present-day Flanders, Belgium. The city developed from a small settlement in the 8th-9th century into one of the most important late-medieval cities in Western Europe with more than 40.000 inhabitants. Its wealth was mainly due to the important role it played in the cloth trade (12th-13th centuries) and its key position in international trade and banking. Bruges was pre-eminently the place where merchants from the south and the north could meet and do business. The settling of the Dukes of Flanders in Bruges (from the 9th century on) and later (in the 15th century) of the Dukes of Burgundy, strengthened this position and made Bruges into a major artistic centre.

Summary table of events

c. 800 onwards A fortification stood at the crossing of the River Reie and a major road

9th century 950-965 1127 1127 1297 1376 1430

1556-1565 1562

1559

1614 1860’s

Presence of a fortification with minting of coins (coins inscribed ‘BRYGGIA’)

Construction of Carolingian church of St Donatian and founding of Chapter

Murder of Count Charles the Good in the St Donatian Church. Fortification of the city perimeter First city charter and establishment of first bench of aldermen

Construction of second city ramparts and city gates (four of them preserved today) Construction of city hall

Founding of Order of the Golden Fleece in Bruges

Digging of new canals between Bruges and Damme/Sluis to improve access from the sea Drawing of city map by Marcus Gerards

Founding of diocese of Bruges. Church of St Donation becomes cathedral

Transformation of medieval city walls to fortifications with bastions. Demolition of 3 city gates

1887

Construction of Neo-Gothic Provincial Court

1907

Inauguration of new port: Zeebrugge

1899 2002

Tower of the Church of Our Lady, one of the oldest parish churches in the medieval town. The actual building dates mainly from the 13th, 14th and 15th centuries. Its tower, at 122 metres in height, remains the tallest structure in the city.

Demolition of Cathedral of St Donatian. St Saviour’s becomes cathedral Bruges Cultural Capital of Europe

The Gent-gate, one of the four remaining medieval city-gates, originally finished in 1407. The building is now one of the seven sites of the city-museum of Bruges (the ‘Bruggemuseum’).

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 90

Five medieval towns - a pictorial introduction: Bruges

From left to right: 18th century mansion of the ‘Free of Bruges’, built in 1726, the ‘Civil Registry’ (Renaissance-style building, constructed between 1534 and 1537), gothic Town Hall (build from 1376 onwards) and the Chapel of the Holy Blood, with a 15th century gothic upper chapel and a mid 12th century lower chapel in Romanesque style.

The guild house of the Tanners, originally built in the 15th century, received its actual form in the 17th and 18th centuries.

The Gruuthuse mansion: this house, built during the 15th century, belonged to a family who had the right to sell ‘gruut’, a mixture of dried plants used to flavour beer .

18th-century warehouse next to the 17th century harbour-site,

75

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 91

76

The archaeology of medieval towns: case studies from Japan and Europe

Göttingen Betty Arndt

A proto-urban settlement, gutingi, is known from the 7th century by archaeological evidence. Although a rural village at this is period, it had gained some wealth by the time of the 11th century. The town of Göttingen was founded near gutingi in the late 12th century with burgesses and consuls being mentioned in 1269. The founder is presumed to have been Henry the Lion. The area of the old village was integrated into the city as late as the 16th century. Göttingen rapidly proved itself to have enormous political and economic dynamism, growing quickly and with a new fortification of wall and rampart being built in the 14th century. The wealth resulted from trade in the Hansa, based on a favourable position on an old and important trade route, the Hellweg. Particularly significant was the trade of local cloth. Cloth seals of Göttingen have been found as far afield as Novgorod and Helsinki. About 1400 the city had some 6000 inhabitants, five churches and two mendicant (Bettelmönche) friaries (those of the Dominicans and the Franciscans). The citizens and their town council were self-confident, independent and strong and, in 1319, they acquired the Neustadt (New Town) which had been founded before 1300 by Duke Albrecht the Fat of Brunswick-Lüneburg while, in 1387, they destroyed the castle of Duke Otto. In the 17th century, however, during the Thirty Years War, the town suffered and did not thereafter recover its former importance, only 3000 inhabitants remaining in 1690, although the city benefited from the founding of the university through Georg August of Hanover in 1733. The old town hall and many fine medieval timber-framed buildings still survive.

Glasses like these Stangengläser with the horizontal layered bulges (right hand side) have to date only been found in Göttingen. (Foto: M. Vladi, Stadtarchäologie Göttingen).

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 92

Five medieval towns - a pictorial introduction: Göttingen

Aerial view of Göttingen, situated in the Leine-Valley. On the left: church of St. Paul’s (Paulinerkirche), in the middle: St. Jacob’s (Jacobikirche), on the right: St. John’s (Johanniskirche) (Foto: B. Arndt, Stadtarchäologie Göttingen)

The ‘Big Mill’ was constructed 1492 and is one of five mills situated at the Leinekanal. It is the successor of a mill which lay outside the town walls from 1305 (Foto: T. Krüger)

Summary table of events 7th – 12th century

953

1024 1162

c. 1200

c. 1230

1270

1268

1289

1294

Nikolausberg Cloister (Augustinerinnen = female Augustinians) is documented

Construction of the first city wall

Gaining of town rights (Stadtrechte), the name Göttingen switching to the new built town (the village is called ‘the old village’)

Town hall is built

Founding of the Franciscan friary (‘Barfüßerkloster’)

First Jewish burgess

Founding of the Dominican friary (‘Paulinerkloster’)

1368

Large tourney of knights (Ritterturnier ) in Göttingen

1370/71

Extension of the town hall

1334 – 1364 From 1368 1387

1402

Several waves of plague (Black Death)

Building of the town rampart and ditch

The citizens destroy the town’s castle

High altar of St. Jacob’s is finished (today in the Landesmuseum Hanover)

1475

Göttingen cloth economy hires workers from England and Flanders

1525

Göttingen is place of refuge in the Peasants’ War

1529

1542

1562

1572

1593

1597

1632

1690

Rebellion against the merchants’ guild Beginning of the Reformation

Founding of the Pädagogium (School)

Burning of witches (Hexenverbrennung)

Withdrawal from the Hansa

Establishment of a girl’s school

Plague in Göttingen

Göttingen conquered during Thirty Years War

End of autonomy of town

Göttingen becomes a fortress

1733

The German Kaiser (Georg August of Hanover) allows a university

1831

‘Göttinger Revolution’

From 1762

The marked positions are extant monuments (Repro: Stadtarchiv Göttingen)

Kaiser Heinrich II dies on the Palatinate (Pfalz = a temporary royal residence) Grone

Town buys the “Neustadt” (New Town)

1715

Map of Göttingen from Matthäus Seutter 1750

‘Gutingi’ is mentioned for the first time in written sources

1319

1514

Grätzelhaus: the house of the wealthy 18thcentury cloth manufacturer Johann Heinrich Grätzel, who had 500 workers on 100 looms (Foto: T. Krüger)

Proto-urban settlement of gutingi (known from archaeological evidence)

1854

1866

1889

1918 1932

1942

1945

Demolition of the rampart

Opening of the Railway Station Göttingen becomes Prussian

First display of medieval urban archaeological finds in the museum November Revolution

Hitler makes speech in Göttingen

Deportation of the Göttingen Jews Göttingen remains undestroyed

77

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 93

78

The archaeology of medieval towns: case studies from Japan and Europe

Norwich Brian Ayers

Norwich was the largest city by area in medieval England, larger than London and Southwark combined although with a smaller population. Even so, its likely population of 30,000 people in the early 14th century probably ensured that it was the second most populous city in the country. It contained a royal castle, a cathedral and attendant Benedictine monastery, four great friaries, numerous hospitals, over 60 parish churches (of which 30 remain), several market places, a river port and warehousing, great houses of city merchants and rural gentry, and over 120 different crafts and trades. Norwich dominated the rich region of East Anglia and maintained its economic preeminence, one largely based upon the textile trade, until the mid-18th century.

Bishop Bridge, completed c. 1340 at the cost of the merchant Richard Spynk, the only surviving example of the five medieval river bridges of Norwich (photo: Brian Ayers)

St Stephen, an 11th- or early 12th-century foundation rebuilt in the late Middle Ages (photo: Brian Ayers)

(Right) Bacon House, the 16th-century façade on Colegate (photo: Brian Ayers)

Norwich Castle Keep, constructed c.1095 – 1110 as the centre of royal administration for the counties of Norfolk and Suffolk (photo: Brian Ayers)

Map of the medieval town showing location of principal surviving medieval buildings, churches (marked with a church symbol), the cathedral, castle and city walls. The site of the medieval quarries is also still visible while the street pattern is almost entirely medieval (the broad road running east of the castle being one of the few exceptions) (David Dawson: © NAU Archaeology)

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 94

Five medieval towns - a pictorial introduction: Norwich

Summary table of events c. 720

c. 850 c. 870-917

917

924-939 c. 980 1004 1066

1068-1075 1075 1086 1094 c. 1120 c. 1140 1174 1216 1226 1266 c.1280-1340 1349 1398/9 1404 1410 1430 15th century 1507 1536-1539 1549 1565 onward 1578 1648 c. 1660-c.1730 1790-1810 19th century 1930s 1942 1945 1948 1950s/1960s 1970s

1991 - 1994 1994 2001

Establishment of small villages on both banks of river Wensum Probable dominance of one settlement – Northwic Danish occupation including construction of defensive earthwork Conquest of East Anglia by Edward the Elder and probable use of Norwich as administrative centre Coins of Aethelstan minted in Norwich First documentary reference to Norwich (in Liber Eliensis) Norwich sacked by the Danes Entry in Domesday Book (1086) suggests Norwich had at least 25 churches as well as 1,320 burgesses Establishment of Castle and ‘French Borough’ Siege of Norwich Castle Domesday Book entry; number of burgesses fallen to 650 Establishment of the Cathedral Expansion south along King Street Expansion north-east of the river Sack of Norwich by Flemings Castle falls to Louis, Dauphin of France Establishment of first Friary (Franciscans) The ‘Disinherited’ raid Norwich Construction of the city wall Black Death Cow Tower built Charter creates offices of mayor and aldermen Guildhall under construction City water mills built Considerable rebuilding of churches Disastrous fires Dissolution of the Monasteries City besieged during Kett’s Rebellion Settling of ‘Strangers’ in Norwich Visit of Elizabeth I Civil War riot ‘Second City’ of England in terms of wealth Demolition of the city gates Development of printing, leather, food industries Slum clearance ‘Baedeker’ air raids City of Norwich Plan First post-war archaeological excavation Urban redevelopment including Central Library site and Inner Ring Road Further redevelopment including Anglia Square and completion of Inner Ring Road Construction of Castle Mall Destruction of Central Library by fire Opening of The Forum building

The Cow Tower, constructed 1398-99, one of the earliest freestanding artillery fortifications in England (photo: Brian Ayers)

A panel of 15th-century painted glass, manufactured in Norwich and now in the Guildhall (photo: Brian Ayers)

Archaeological excavation and surveying west of the church of St Peter Mancroft (© NAU Archaeology)

Archaeological excavation at the rear of Dragon Hall, a building constructed in 1427 for the textile merchant Robert Toppes (© NAU Archaeology)

79

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 95 8

80

The archaeology of medieval towns: case studies from Japan and Europe

Ribe

Jakob Kieffer-Olsen Ribe is the oldest town in Denmark and Scandinavia. Situated on the western coast of Jutland, it was an important link in the trading system around the North Sea. However, even though Ribe was one of the most important towns in Viking Age and medieval Denmark, it probably never contained more than some 5000 people although the town did have a royal castle, a cathedral, four monasteries, two hospitals and six parish

The church of St. Catharina and the surviving part of the Dominican friary seen from the tower of the Cathedral on a foggy day.

churches. Only the cathedral, together with the church and other buildings belonging to the Dominican Friary, still exist. In the 17th century Ribe was hit by several setbacks: flooding, plague, and occupation and destruction by foreign troops. At the same time power was concentrated in the capital (Copenhagen) and the harbour condition in the narrow Ribe River proved insufficient. In consequence, Ribe lost its position and played merely a regional role. Today Ribe is a little town, where the central part is dominated by the cathedral and houses from the 15th to 19th centuries.

The oldest prospect of Ribe (Braun and Hogenberg 1598). Information for this drawing was given by Anders Sørensen Vedel, who was writing a history of Denmark. He has chosen to show Ribe before the Reformation of 1536, after which many churches were demolished

The Old Town Hall was converted in 1709 from a domestic building originally erected in the 15th century

The royal castle, Riberhus, was demolished in the later part of the 17th century. Today only banks, moats and a few ruins can be seen. This photograph was taken on a day when the wind from the west made it impossible to let water from the river through the dike into the sea. The situation – Ribe as an island – was quite common before the sea dike was built in 1914

Flooding of the area round Ribe happened occasionally until the sea dike was built in 1914. The biggest floods are marked on this column

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 96

Five medieval towns - a pictorial introduction: Ribe

Summary table of events c. 710

Establishment of a market place on the northern bank of the river

c. 855

First mention of Ribe when the missionary Ansgar was permitted to build a church in the town

c. 720 948

c. 960 1077

The first Scandinavian coins were struck in Ribe

The first bishop in Ribe, Liufdag is mentioned. According to the legend, he was killed by the pagans – escaping the crowd he was hit by a spear in the middle of the river On the big runic stone in Jelling, the King, Harald Blåtand, claims to have Christianised the Danish people

First archaeological dating of activity on the southern bank of the Ribe River

1145

First mention of the chapter and the school

c. 1250

Building of the dam nowadays used as the main street

c. 1175 1228 1232 1259

1269 1283 1233 1350 1362

Building of the standing Cathedral begins

Map of surviving part of the medieval town showing location of the medieval churches, the cathedral and the castle. The street pattern is almost entirely medieval.

Establishment of Dominican Friary Establishment of Franciscan Friary

The King, Christoffer 1, dies in Ribe (according to the legend he was poisoned at Communion) The King provides Ribe with its town charter The tower of the Cathedral falls down

The new north-western tower of the Cathedral is standing Black Death

The first known big flood called ‘De grote Mandranck’

1398

A union between the three Nordic countries are established in the Swedish town Kalmar

1536

Danish Reformation causing demolition of churches and monasteries

Medieval brick with graffito depicting a monk.

1486-89 Building of the oldest timber framed house still standing

1580

Disastrous fire hits 213 houses

1634

Flood killed thousands of people along the Wadden Sea

1627-29 Ribe occupied by German troops due to the Thirty Years’ War 1645

Ribe occupied by Swedish troops

1659

Last major outbreak of plague killing 900 inhabitants and some thousands of peasants

1657-60 Ribe occupied twice by Swedish troops 1660 1849

1855 1868

Danish absolute monarchy established Danish Constitution passed

First provincial museum established in Ribe

Town and harbour of Esbjerg, some 30 km north of Ribe, is established by law

1899

Establishment of the Tourist Board

1943

First planning of the city development, including preservation

1914

1955

2007

Building of the sea-dike completed

First major town-archaeological excavation in Denmark takes place in Ribe Ribe became part of Esbjerg Municipality

The Cathedral seen from the main street. In 1283 the north-western tower fell down and killed many people during Christmas services.

81

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 97

82

The archaeology of medieval towns: case studies from Japan and Europe

Rouen

Dominique Pitte

Rouen, the principal city of Upper Normandy, was a Roman foundation (Rotomagus). It was the capital of the Véliocasses and, by the 2nd century, had an extent of some 80 hectares. A bishopric was established in the 3rd century with a wall being constructed around the town at the end of that century, protecting some 20 hectares. Ecclesiastical importance was maintained until the city was sacked by the Vikings in 841. With the donation of Normandy to the Viking Rollo in the 10th century, Rouen became the capital of the Norman dukes. It remained in Norman hands until taken by Philip Augustus of France in 1204 who pulled down the Dukes’ Tower and built a new castle dominating the town from the north. A new city wall was started in the 14th century, being completed at the beginning of the 15th century. The town was an important medieval industrial centre, particularly for the production of cloth and retains numerous timber-framed medieval buildings.

Summary table of events 0

1st – 2nd centuries Around 260

End of the 3rd century

Shortly before 400

5th – 8th centuries 841

911

By the end of the 10th century 1066

11th – 12th centuries

Third quarter of the 12th century 1204

1346

1418 – 1419

1419 – 1449 1449

Rouen cathedral (photo D. Pitte)

The Romans found a new town on the right bank of the River Seine Town flourishes

Mellon, first bishop of Rouen.

Rouen becomes the capital of a new province: the Seconde Lyonnaise

A basilica is built by bishop Victrice, in the centre of the protected area. Rouen is an important ecclesiastical metropolis. Town is devastated by the Vikings.

King Charles the Simple gives Normandy to Rollo: Rouen becomes the capital of the first dukes.

A stone castle is built in the south-east corner of the wall.

William the Conqueror Duke of Normandy and King of England.

The donjon of Philip Augustus’ castle. Built in the early 13th century, it symbolises the domination of the King of France over the Norman capital (photo D. Pitte)

Rouen is a very prosperous city, in the centre of a most important economic and political entity.

Henri II gives the Rouennais a charter of liberties. The Etablissements de Rouen, organising the Commune, serve as a model for numerous cities.

City falls to the army of Philip Augustus; Normandy is annexed to the kingdom of France After the first English raids in Normandy, King Philip VI fortifies the city town. Rouen is besieged Henry V, King of England.

During their presence, the English raise a huge fortress in the south-west corner of the city wall. Rouen is back in the kingdom of France.

At the foot of the cathedral and churches, timber framed houses, together with stone buildings, composed the frontage of the streets of Rouen (photo D. Pitte)

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 98

Five medieval towns - a pictorial introduction: Rouen

A view of Rouen today (photo D. Pitte)

In 1724 the city wall, built during the second half of the 14th century, still imposed its shape on Rouen (Rouen, ville capitale de Normandie, by Nicolas de Fer, 1724)

Rebuilt at the end of the 14th century, the belfry stands in a most symbolic place: on the course of the Roman wall, close to the town-hall and an old city gate (photo D. Pitte)

83

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 99

84

The archaeology of medieval towns: case studies from Japan and Europe

Medieval urbanism and culture in the cities of the Baltic – with a comparison between Lübeck, Germany and Sakai, Japan Manfred Gläser Summary

This paper summarises the current state of knowledge for the important Hanseatic town of Lübeck. It draws upon archaeological data drawn from some thirty years of excavation and examines in particular the urban culture of Lübeck and its influence on towns around the Baltic Sea. The paper concludes with a short comparison of the evidence from Lübeck with that from Sakai in Japan, noting the common themes of urban structures, functions and even institutions.

Introduction

settlement of Eastern Europe. One of the greatest Slavic settlements was Lübeck (Alt Lübeck or Old Lübeck). Figure 2 shows a reconstruction with the fortifications, the harbour and a special settlement for the merchants. Since 1852 great excavations have taken place and many golden finds have been discovered. In recent years a planked path, dating to the year 760, was uncovered (Figure 3) and a part of a reliquary was found (Figure 4). Old Lübeck was destroyed in 1138. Five years later, German Lübeck was founded. Lübeck is by far the oldest German settlement on the Baltic coast. The foundations of Rostock, Wismar, and Stralsund followed more than 50 years later. The city is considered to

The eastern border of the German Empire in the 10th century ran roughly along the Elbe and the Saale (Figure 1). With Bohemia in the south-east and Poland in the east, independent Christian Slavic kingdoms emerged. As far as northern Europe was concerned, the Norman age was coming to an end. Here, Christian kingdoms of the Danes, Swedes and Norwegians were coming into being. In contrast to this, the area between Elbe and Oder was still settled by pagan Slavic tribes, who bitterly defended themselves against any kind of foreign rule and Christianisation. These Slavic tribes never succeeded in uniting under a single leadership. It was not until the middle of the 12th century that the German rulers were able to conquer the region east of the Elbe. These military operations are associated with the names of Duke Heinrich the Lion or the Danish king Waldemar the Great. The Slavic princes submitted to the German feudal lords and a period of intensive Christianisation and Germanisation followed. In the following two centuries thousands, perhaps hundreds of thousands, of German immigrants entered the country: knights, peasants, craftsmen, merchants, but also monks and nuns. The immigrants founded towns, villages and monasteries and convents. This process is called German settlement in the east or colonisation of the east. The native Slavic population was forced back and frequently acquired an inferior legal status, retreated into reservations and was absorbed into the German population.

The Slavs, of course, had markets, trade and the beginnings of towns but the town in a legal sense, with a wall and the idea of citizenship, only came into being as part of the German

Figure 1: The German Empire in the 10th century.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 100

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Figure 2: Reconstruction of the Slavic settlement Old Lübeck in the 11th century.

Figure 3: A wooden path dating in the year of 760.

be the prototype of the modern planned town of western origin (v. Brandt 1974. Graßmann 1988) and to have served as an example for the whole Baltic region. It represents a ‘new type of port’ (Ellmers 1972), and it was the birthplace of the Hanseatic League. Today the whole city is on the world heritage list of Unesco with the buried archaeology being included in the preservation order.

added. Originally, only the flood-free area was suitable for settlement, that is the ridge which can still be seen stretching from north-to-south across the whole peninsula. This part became wider in the middle of the peninsula, so that spurs of land offering favourable settlement conditions extended down into the low-lying marshes of the Trave (in the west) and the Wakenitz (in the east).The top of this ridge, as well as several smaller elevations, did not rise as high above sea level as they do today, but the relief was probably much more irregular. Five elevations or hillocks are still visible in the townscape, each of them corresponding - certainly not by accident - with the location of a church or a monastery (from south-to-north: the cathedral precincts, St Peter’s hillock, the market area, the so-called Koberg, and finally the area around the castle, which later was turned into a friary).

Topography

The hill on which Lübeck is situated was once a peninsula but nowadays forms an oval extending some two thousand metres from north-to-south and some one thousand metres from west-to-east, covering an area of almost 120 hectares (Figure 5). Since 1900 the historic town centre has been entirely surrounded by water, thus becoming an island which can be reached by five bridges. When the first German settlers arrived, the hill presented itself in a different light, its present shape, morphology and island character being the result of intensive settlement during the past 850 years. Even though evidence of a pre-German occupation has been found, this did not lead to any major changes in the topography. The chronicler Helmold (I 57) was the first to give a description of the town hill. Reporting on the first foundation of Lübeck by the Earl of Schauenburg in 1143, he explains in an unusually detailed manner for a medieval chronicler that the town is situated on a peninsula enclosed by the rivers Wakenitz and Trave and their marshy banks. According to Helmold, there was only one point where the banks of the Wakenitz could be approached on dry ground.

Due to underground engineering and a number of drillings it is possible to reconstruct the original topography to a large extent. These findings have been largely confirmed by excavations, but some aspects also had to be revised or

Figure 4: Part of a reliqary, bone of a walrus, 11th century.

The marshes along the Wakenitz and especially the Trave were of a completely different character. Extensive areas in the south-west and the north-west, altogether approximately one third of the area of the town hill, must have been covered by marshland, with thick peat layers overlying the clayey subsoil. In the 12th century, this would have been a zone highly liable to periodical or even permanent flooding, interspersed with tributaries of the meandering Trave and numerous water holes - a landscape of water meadows (van Haaster 1991) totally unsuitable for occupation.

Early Settlement

The prominent town hill did not only attract settlers in the last millennium (Jestrzemski et al. 1988). The oldest, though residual, finds date from the Neolithic period. They were retrieved from the Burgkloster (’castle friary’) site in such large numbers that an intensive occupation can be postulated in this area (Gläser 1992a, 71f.). They comprise

85

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 101

86

The archaeology of medieval towns: case studies from Japan and Europe

Figure :5 Lübeck in the Slavic period (9-11th century).

Figure 6: Lübeck in the second half of 12th century.

thousands of items: flakes in particular, but also scrapers, chisels, axeheads, and pottery. Also residual and thus associated with younger finds are sherds from the Bronze Age and the Early Iron Age. The oldest undisturbed features on the town hill belong to the Later Bronze Age, being the remains of hearths uncovered at Dr. Julius-Leber-Straße 11 (Schmidt 1993). Finally, the first fortification of the town hill dates back to the Roman Iron Age (Fehring 1982; Gläser 1992a; Lagler 1982). It consisted of a ditch, almost eight metres wide and three metres deep, which was discovered at the Burgkloster site. The ditch ran straight from east-to-west, barring access to the peninsula. Presumed accompanying settlement traces further south have yet to be traced.

country. This theory was, however, refuted by the Burgkloster excavations, conducted by Günter P. Fehring, which produced clearly Slavonic features (an occupation layer and a defensive ditch) (Fehring 1982; Gläser 1992a) while, recently, even more evidence has come to light (Radis 2002).

Regarding Slavonic evidence on the town hill the situation is rather different: The chronicler Helmold (I 57) does mention a deserted stronghold of the pagan lord Cruto but, until recently, there was absolutely no sign of a Slavonic settlement on the town hill. Consequently, it has been considered by many historians to have been unoccupied when it was chosen by Earl Adolf of Schauenburg to be the site of his new town. They have suggested that Lübeck was ‘planted’ in open

Castle

Helmold makes no allusion to the erection of a German castle in the foundation year of 1143 but for 1147 (I 63) he reports the existence of such a building, which was situated at some distance from the civic settlement and strong enough to withstand a siege of several days. In spite of its importance, the castle is not mentioned in the following decades. Only for 1217 is its existence confirmed by the chronicler Detmar (59). Nevertheless, it is commonly agreed that it must have been present in the period not covered by documentary evidence as well. It is assumed to have been the residence of the bailiff of the town’s respective sovereigns, until it was eventually transformed into a Dominican friary after Lübeck’s victory in the Battle of Bornhöved.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 102

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

The strategic situation of the hillock at the only access to the peninsula was the obvious place for such a castle (Fig. 6). Here, there was enough high and almost level building ground for the erection of permanent buildings. At the same time, boats could land at the bottom of a steep slope, ensuring supply by water. From this location both the settlement emerging further south and the traffic could be controlled. This included shipping on the Wakenitz and Trave as well as a long-distance trading route which, coming from Bardowick/Lüneburg, led across the hill to the Baltic coast. The Slavonic defences were probably rebuilt as early as 1143 or shortly after. A U-shaped system of trenches was laid out which opened towards the steep slope in the west, consisting of ditches up to 22 metres wide and five metres deep and enclosing a square with a side length of about seventy metres. The northern flank of this structure was directly continued by an interrupted ditch which probably extended down into the marshes of the Wakenitz. There is evidence for a gate or passageway in the north-east of the central structure, while another one across the interrupted ditch is to be presumed.

Thus, it was possible to control the supraregional trading route running immediately along the eastern flank of the central structure without being disturbed in the castle itself. Another route, leading east to Holstein, probably branched off in the area of what was to become the Koberg in later times and reached the artificially raised and enlarged bank of the Trave at the foot of the castle. Here, the river could be crossed by ferry.

During the first years of its existence the castle only contained wooden buildings. They consisted primarily of single-aisled buildings with earthfast posts, the plans of which vary in size and intersect each other. At least in one case there is evidence that the walls rested on interrupted sills which had been joined to the posts by mortise-and-tenon joints. To provide the inhabitants with drinking water, a quadrangular well (Figure 7) of massive planks was built during the winter months of 1155/56, measuring about1.2 metres across and reaching some eleven metres down into the ground. The first brick buildings were not erected much before 1200. Beneath a massive post-medieval masonry wall in the west of the castle site, the unexpected remains of a two-metre thick brick wall were unearthed. This wall, running along the banks of the Trave at the foot of the castle, was underpinned by posts. The second brick building dating back to this time is still largely extant, so that it is not necessary to rely solely on archaeological sources. It is the so-called Long Hall, the northern wing of the monastery. This is forty metres long and about eleven metres wide, originally two-storeyed building which probably served as the summer refectory and was long regarded as the nucleus of the Dominican friary. According to

Figure 7: Hansestadt Lübeck, the well of Burgkloster-castle, dating in the winter 1155/56.

the excavation results, the building was once divided into three parts and features an arched foundation unusual in Lübeck. Architectural details, correlations with contemporary buildings in Denmark, the location within a former defensive circuit, two elaborate fireplaces on the western wall, Romanesque window openings in the southern wall next to the cloister as well as parallels with the ’Grey Friary’ in Schleswig have all led to its being interpreted as the former hall of the castle, held in the early 13th century by the Danish king. This assumption, however, still has to be verified by an exact dating. When Lübeck had become a free city in 1226 and the Danish king had been defeated in 1227, the castle in the north of the town not only lost its function but also become a threat to the young community striving for independence. To ensure that the settlement should never again be controlled from the castle by the representative of a sovereign, it was decided to raze it to the ground. According to a much cited legend, the terrain was passed to the Dominicans for the erection of a friary. Their presence in Lübeck is indeed documented from 1228 onwards, and they have been shown to be the owners of

87

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 103

88

The archaeology of medieval towns: case studies from Japan and Europe

Figure 8: Hansestadt Lübeck, Excavation Alfstraße/Fischstraße, the development of a plot in the 12th century.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 104

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

the property. Naturally, the presumed former hall was not pulled down but incorporated in the monastic complex as its north wing.

Urban settlement

Although features dating from the first decades of Lübeck have been recorded at several excavations, clear evidence of the settlement of 1143 is still lacking. The ‘castle well’ could be dated to the winter months of 1155/56 (Fehring 1982; Gläser 1992a, 77) but the castle itself was not really part of the settlement. Consequently, the different theses put forward by historical research as regards the location of the early settlement have not been verified by archaeological investigations. The only possible exceptions are structures uncovered on St Peter’s hillock and on a plot in Fischstraße (for the following passages cf. Gläser 1992b, 48f. & 59). These comprise a posthole building dated ‘around or after 1145’ and a well dated precisely to 1152.

Thus the exact location of the settlement founded by the Earl of Schauenburg is still unknown, but there can be no doubt that the centre of the second foundation by Duke Henry the Lion in 1158/59 is to be sought in the area between the market with St Mary’s church and the harbour, that is in a part of the town which has always been considered to be the ‘founding district’ in literature (Fig. 6). This zone between Mengstraße and Holstenstraße stands out on the map due to its five linear streets leading down to the waterfront. Unlike the areas to the north and south, it is obviously the product of deliberate planning by a political authority. Between 1985 and 1990 this area, also called the ‘merchants’ quarter’, was the scene of the so far largest excavation campaign in Lübeck, taking place in the block immediately west of St Mary’s, between Alfstraße and Fischstraße. The great expectations connected with this investigation (Dumitrache et al. 1987), which covered no less than nine former properties, have been completely fulfilled: for the 12th century only, a sequence of four different settlement periods could be recorded (Figures 8.1-8.4, Legant-Karau 1993). No buildings belonging to the initial phase could be identified (for the following passages cf. Legant in preparation), but there were traces left by hoes or spades, indicating that the terrain had been used as a garden. Pole and post holes belonging to fences illustrate the demarcation of property rights. This assemblage was complemented by the above-mentioned well. The associated pottery has to be assigned to the German settlement, although it contains a relatively high proportion of residual Slavonic material. Considering the entire pottery assemblage, the features of phase I can be dated to the mid12th century but whether they go back to the Schauenburg

period remains an open question. The oldest timber buildings appear in Periods II and III. They are single- to four-aisled posthole constructions which were erected at the street frontage. Some of the plots now also contained outbuildings, again posthole or rather moderate log constructions, as well as rubbish pits and wells. The oldest and at the same time the largest building dates back to the time ‘around 1159’, which, according to documentary evidence, is the year of the Welfian re-foundation.

In Period IV, there was a great leap in the quality of house building, the leap from traditional posthole and log constructions to fully-framed buildings. While the posthole buildings on the street frontage were still being used, basemented fully-framed buildings, probably multi-storeyed, were erected on five properties. These were extremely massive constructions with a quadrangular plan of much smaller dimensions (Figure 9). In at least one case the remains of a tiled stove were found. Referring to heated Kemenaten (chamber blocks) and tower-like Steinwerke, Günter P. Fehring (1989) interpreted these buildings as ‘timber-built Kemenaten’ incorporating both a high-standard domestic dwelling and a storage basement. Four of these five houses could be dated by dendrochronology, providing two exact (‘1180’and ‘1188’) and two approximate dates (‘around 1186’ and ‘around 1187’). The associated wells, however, have surprisingly been found to be of a somewhat earlier date (‘around 1167’ and ‘1175’).

Figure 9: Hansestadt Lübeck, Excavation lfstraße/Fischstraße, reconstruction of the wooden houses with cellars, the so-called ‘Holzkemenaten’, dating in the last decades of 12th century.

89

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 105

90

The archaeology of medieval towns: case studies from Japan and Europe

Figure 10: Hansestadt Lübeck, Excavation Alfstraße/Fischstraße, Reconstruction of the so-called ‘Saalgeschoßhaus’, early 13th century.

In Period V, a brick building was raised at the corner of Alfstraße/Schüsselbuden, a so-called Saalgeschoßbau, which was 18.4 metres long and 9.8 metres wide and had a vaulted cellar (Figure 10). This new building also marks the end of the timber-building period. With the exception of outbuildings, brick now became the prime building material. In Period VI, the ‘timber-built Kemenaten’ were rapidly replaced with towerlike Steinwerke adjacent to the street frontage (Radis in preparation). These activities took place in the first decades of the 13th century. In the latter half of the century, building density increased again. The Steinwerke, which had covered only a small area, were succeeded by large Dielenhäuser, set at right angles to the street and occupying half of the plot. In late medieval and early post-medieval times these buildings were usually enlarged by a narrow side wing and, if there was enough space, a rear building parallel to the street.

Market

A market in Lübeck is mentioned by Helmold as early as 1147 (I 63), around 1152 (I 76) and for 1156 (I 84). In 1159 at the latest there is evidence for a ‘food market’, that is a market where regional produce was sold. For reasons of convenience

Figure 11: Hansestadt Lübeck, Excavation Markt, the so-called ’Marktschicht’, a layer of humus, dating second half of 12th century.

this will be called central market here, and it is identical with the present-day marketplace. Thus, the results of an excavation carried out on the marketplace in 1976 came as a real surprise. They suggested that markets were first held in this area in the early 13th century, in contradiction to all historical evidence (Stephan 1978). This comparatively late date was to be re-examined by a second excavation in 1986 (for the following cf. Mührenberg 1993). Again, an occupation

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 106

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

layer representing the oldest phase of intensive use was found (Figure 11). It was rich in humus and up to thirty centimetres thick and probably had been created over several decades by continual walking, driving and riding as well as the dumping of organic matter. Due to the large-scale documentation of this layer and its reinterpretation as a gradually accumulated deposit, its dating could be corrected. Even though it was not possible to subdivide it into phases, globular pots were found to dominate in the lowermost parts, while the youngest pottery, on which - absolutely correctly - the dating of the whole feature had been based before, was confined to the upper parts. In connection with a restructuring of the place and the building of a new town hall, the market zone was covered up with planks and furnished with public rubbish dumps around 1220/30.

Waterfront

As early as 1143, the founding year, the chronicler Helmold (I 57) pointed out the excellent harbour at the foot of the hill.

For 1147 he mentions ‘ships loaded with goods’, which were lost because they had been set on fire. The essential role that the harbour played for the young settlement is illustrated by the events of 1158/59, when Henry the Lion’s plan to plant a new town on the Wakenitz failed, primarily because the site could only be reached by small vessels and thus proved totally unsuitable for a settlement relying on long-distance trade. When the citizens had returned to the peninsula of Lübeck, they were granted ‘considerable freedoms’ by Henry the Lion who had become their new lord. He also ensured ideal trading conditions by offering Scandinavian and Russian merchants access to free trade (Helmold I 86). In addition, exchange with merchants from Gotland was regulated in a charter.

But where was that oldest harbour of the town? Topographical conditions for the establishment of a harbour on the Trave were very varied (cf. Figure 5). There were only three sites where solid building ground extended down to the river: at the foot of the cathedral in the extreme south; at the bottom of the castle in the north; and finally on a spur of land

Figure 12: Hansestadt Lübeck, Excavation Alfstraße 36/38, Period I, second half of 12th century.

91

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 107

92

The archaeology of medieval towns: case studies from Japan and Europe

Figure 13: Hansestadt Lübeck, Excavation Alfstraße36/38, Period I, second half of 12th century, reconstruction.

stretching down to the Trave valley between Mengstraße and Braunstraße. The last area was certainly the most suitable spot, as it did not only lie high and dry but, compared to the other sites, also offered enough space and rose very gently to the town hill.

Excavations beneath the houses of Alfstraße 36/38 and in the area to the west, between the houses and the quay, did indeed produce the first unmistakable evidence for the existence of a harbour in the 12th and 13th centuries. Basically, two occupation levels could be distinguished (Gläser 1985;

Schalies 1992). In the first phase, a row of densely set piles was driven into the ground parallel to the riverbank and about 2m away from it in the water (Figures 12 and 13). Then the zone between the revetment and the former bank was infilled and consolidated, so that a then comparatively modern quay was created which allowed ships with a draught up to 1m to come alongside. These efforts were necessary in order to make the harbour accessible to a new type of boat, the cog. Unlike the Slavonic and Scandinavian boats, these could not be hauled up on to the beach for unloading because of their steep bow. They needed a construction where they could berth while still afloat (also Ellmers 1985; 1992). The revetment has been dated by dendrochronology to the years ‘around 1157’.

About forty metres east of the quay, the remains of a brick wall also running north-to-south were unearthed, probably the remnants of the urban defences from the 1180s. The space between the quay and the city wall was used in many different ways. Up to now three timber buildings from the last two decades of the 13th century have been recorded in spite of the small scale of the excavations. These quadrangular houses were of a modest size, with side lengths of four to five metres. They were fully-framed buildings slightly sunk into the rising slope. ‘Around or after 1187’, the riverbank and the town were connected by a planked path which probably led to one of the city gates and can be considered the predecessor of present Alfstraße. Figure 14: Hansestadt Lübeck, Excavation Alfstraße 36/38, Period II, early 13th century.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 108

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Figure 15: Hansestadt Lübeck, Excavation Alfstraße 36/38, Period II, early 13th century, reconstruction.

In the second phase (Figures 14 and 15), the overall structure of the waterfront was radically altered. A new revetment was built still further out into the river, consisting of horizontally laid timbers which were held in place by piles. The new quay was suitable for ships with a draught up to two metres. The old city wall also had to make way for the expanding town. It was pulled down, and, probably in 1217, a new wall was built even further to the west, only 5m away from the new quay. One of the merchant houses which was under construction at the waterfront at that time was investigated thoroughly from an architectural and an archaeological point of view (Gläser 1985; Holst 1985). It is the corner building at Alfstraße 38, a house of impressive dimensions hardly surpassed even by later secular buildings. It measured 23 by twelve metres, its gable walls being 1.15 metres thick. Its undercroft was divided by a central row of elaborate pillars and was furnished with two large doorways leading to the harbour. Above, there were two storeys and an enormous attic with a large storage capacity. The joists of the undercroft could be dated to the years ‘around 1220’ by dendrochronology, while thermoluminescence provides a date of ‘1216’. Presumably the erection of the building began immediately after the city wall had been moved to the west. Because of its exposed location and certain architectural details this more or less extant building has been supposed to be the first guildhall of the merchants engaged in longdistance trade (Holst 1985, 134f.).

Urban defences

The defensive walls were of vital importance to medieval cities. They have, quite rightly, been recognised by historians as a crucial stage in the development of medieval towns. Apart from the notion of ‘burgher’ and town charters, they can be regarded as an important feature of the new type of German towns which evolved in the former Slavonic territory (cf. Ennen 1975). The defences formed a clear demarcation line between urban and rural society and protected newly

created legal, administrative and economic zones from incursions by feudal lords. Thus, the fortification of the city was the most important task of the civic community, probably even more important than church building. The defences had to be erected, kept in good state and, if necessary, to be defended. The fate of the individual and the entire community depended on their effectiveness. However, town walls also signified boundaries, so that their recording and dating is important for the investigation of settlement patterns as well (for the following cf. Gläser 1990).

Apart from the castle, Lübeck’s defences are only mentioned on five occasions in the 12th and early 13th centuries. Although the Schauenburg settlement of 1143 is likely to have been fortified in some way, be it by fences, palisades or ditches, Helmold does not refer to them, neither is there any documentary evidence from the following decades. In connection with the re-foundation, however, Helmold (I 86) does report the rebuilding of ‘churches and walls’, but this is an expression which leaves much room for interpretation. Archaeological evidence of the late 12th and early 13th century town walls is indicated on Figures 6 and 16.

For several decades at least two defensive structures existed side-by-side on the town hill: the defences of the civic settlement in the centre, with a total length of three kilometres, and the castle of the town lord in the north. Whether the cathedral precincts in the southern part of the peninsula were protected by their own defences has yet to be resolved. Such defences are not improbable, as Helmold (I 87) informs his readers that the Duke strengthened the bridgehead at the Wakenitz shortly after 1160. By the early 13th century the town had expanded so much that its defences were no longer sufficient. In addition, as described below, the open area around the Koberg had been built up, and extensive areas in the south-west and north-west were to be reclaimed. Detmar’s reference to King Waldemar as the first to enclose the city and the castle with a wall in 1217 was long regarded by historical research as reference to the first town wall, an opinion now disproved by the early dates of the above-mentioned structures. Obviously what Detmar meant was not the first wall but the first common town wall shared by both parts of the settlement, the city and the castle. This was indeed an important undertaking, which throws some light on the topographical and legal situation in the north of the hill.

‘Artisans’ Quarter’

The largest excavation in the so-called ‘artisans’ quarter’ east of Königstraße or east of the central market zone with

93

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 109

94

The archaeology of medieval towns: case studies from Japan and Europe

St Mary’s church took place between the upper part of Fleischhauerstraße and Dr. Julius-Leber-Straße from 1990 to 1993. The excavation encompassed sixteen former plots with approximately 4,500 square metres, the cultural layers being up to five metres thick. The excavation data are still being analysed, and so far only a short preliminary report has been published (Schalies 2002). The dendrochronological dates, however, already indicate that the oldest features have to be assigned to the late 12th and the early 13th centuries. These are a number of buildings with earthfast posts, the oldest of which was erected ‘around 1190’, timber-built wells dating from the last decades of the 12th century, and a fully-framed building from ‘around 1232’. Several fences are also noteworthy as well as posthole building from 1211/12 which in 1230 was enlarged by a brick-built annexe - the earliest evidence of a masonry building in this area. Apart from this construction, none of the characteristic stone buildings of that time - neither Saalgeschoßhäuser nor Steinwerke - could be traced in the large excavation area. Only at the end of the 13th century did large brick houses become a common sight in this district and were represented, as in other parts of the town, by Dielenhäuser built at right angles to the street.

The excavation results in the so-called ‘artisans’ quarter’ have largely confirmed the documentary evidence. The plots east of Königstraße were first occupied during the last decades of the 12th century, some time after the foundation of the town. In the more remote parts of the quarter, the earliest structures cannot be dated before the beginning of the 13th century, e.g. in Hundestrafle (Mührenberg 1989) or in Hüxstraße (Müller 1992). The development of this area, between St Egidiusí and Glockengießerstraße, was probably incited by the foundation of St John’s priory in the extreme east of the ‘new town’ (Gläser 1989) and the allocation of monastic property to settlers. Unfortunately, very little is known about the settlement process in the south of the peninsula. Here, no excavations of any relevance have taken place, so that the assumption that not only the cathedral precincts but also both sides of Mühlenstraße were occupied by the end of the 12th century has yet to be confirmed. In the north of the peninsula the situation is different. In this district, between Glockengießerstraße and the castle/friary with the Koberg at its centre, a number of excavations have been carried out. Besides the castle area, the plot of the later Holy Ghost Hospital has been investigated with especial care. Günter P. Fehring (1997) was able to demonstrate that the initial occupation phase dates back to the beginning of the 13th century, that is several decades before the erection of the hospital. The earliest dendrochronological dates, both ‘around 1236’, have been obtained from the wooden

Figure 16: Lübeck in the middle of 13th century.

reinforcement of two pits and the sill beams of a large twoaisled framed building.

Expansion of settlement area

When the ‘artisans’ quarter’ and the Koberg area had been incorporated in the expanding town, all the high and dry ground had been used up. The only potential building land that was left were the far-stretching marshes in the north-west and south-west of the town, a rather inhospitable environment. The reclamation of this area was going to be extremely labour-intensive and material-consuming, but the only alternative would have been an expansion of the settlement to the other sides of the Wakenitz and the Trave, which would have caused many problems with regard to supply and defence (for the following passages cf. Erdmann 1982; Gläser 1992b). It is not known which group - it was probably not a single person - took the final decision to realise the reclamation project in the marshes, a project which is not recorded in any documentary sources. The town council, first mentioned in

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 110

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Figure 17: Hansestadt Lübeck, Excavation Große Petersgrube.

1 First half of 13th century.

2 Second half of 13th century.

1201, is a possible candidate, but it may also have been a co-operative of venturesome investors. The capital needed was immense, but on the other hand the town area would be enlarged by about fifty percent and the harbour could be extended a long way to the south (Figures 6 and 16). Therefore, it seems likely that the town lord himself, King Waldemar, or his representative initiated the work. The river meadows at the bottom of St Peter’s hillock were originally about eighty metres wide. To the north, in the direction of Holstenstraße, this zone became increasingly narrow, but to the south it widened to almost two hundred metres. The area in total covered some thirteen hectares. Excavations (Figure 17) and numerous drillings have shown the topography to have been very varied. The peat layer would have stretched down to depths of a few decimetres to three metres below water level, with the exception of a few flat islands jutting out above the water. In hot summers even larger areas might have fallen dry. Thus, between three and five metres had to be infilled in order to create a flood-free building ground, a fact also confirmed by excavations. However, the first step would have been to separate the reclamation zone by a dam from the waters of the Trave, so that the marshy ground could gradually dry up. This presumed dam has not been traced by any excavations

although it probably runs below the present street ‘An der Obertrave’. The reclamation process was presumably furthered by drainage ditches. It is noteworthy that most of the street names both north and south of the spur end with grube, for example Beckergrube, Fischergrube and Engelsgrube in the north, and in the south all streets between Kleine Petersgrube and Effengrube. These denominations are likely to derive from drainage ditches.

Presumably, the erection of heavy stone houses was intended from the beginning. This meant that subsidence and shifting of the ground had to be reduced to a minimum, a result which could only be achieved by large-scale timber foundations which ensured that the weight was spread as evenly as possible. The best solution, also traced in excavations, was a grid (Figure 18) consisting of horizontal timbers laid at right angles and secured against lateral pressure by upright pegs and piles. The timbers were not connected by any joints. The gaps in between were infilled with soil and debris of all kind.

In the following phase, the foundations for the walls had to be constructed. Narrow substructures of four to five layers were laid out, basically another kind of grid of horizontal timbers. These substructures obviously corresponded to the plot boundaries. Apparently, a continuous street frontage of right-

95

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 111

96

The archaeology of medieval towns: case studies from Japan and Europe

Figure 18: Hansestadt Lübeck, Excavation Große Petersgrube, The grid consisting of horizontal timbers, first part of 13th century.

angled houses had been planned from the outset. At the same time, the interior of the plots was divided into rectangles by planked walls forming provisional ‘boxes’ which then could be infilled. Finally, on top of the substructures, the actual foundation walls for the brick houses were built of boulders.

The timber used for the foundation works did not only come from oaks but also comprised beech, alder and maple. Nevertheless, more than a hundred dendrochronological dates could be obtained, in some cases giving the exact felling year, although the dates could not be ranked chronologically according to their spatial context, neither downhill from eastto-west nor from bottom-to-top within the individual sectors, because a large number of timbers were used for a second time. Accordingly, quite often comparatively old timbers dating to the 12th century were found on top of younger ones from the 13th century.

Figure 19: Hansestadt Lübeck, the street ‘Große Petersgrube’ today.

Since the stratigraphical sequence and the absolute dates do not correspond to each other, the whole complex - as with pottery assemblages and treasure finds - is dated by the youngest finds’ material. First of all, there was a remarkably large number of timbers from the 1220s, 1230s and 1240s, the youngest having been felled ‘around or after 1244’. Thus, the reclamation, at least in the area of Große Petersgrube, must have been carried out within a surprisingly short period of time, namely between 1210 and 1250, and in one single effort, not sector by sector down the hill. This means that it was started in the so-called Danish period, i.e. before 1225.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 112

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Figure 20: The Hanseatic town of Stade.

In the later 13th century, Dielenhäuser were erected on the new plots with their gables to the street (Fig. 19), at the same time as in the districts which had been settled earlier. According to documentary evidence, all plots were occupied by brick-built houses by 1313 at the latest.

Urban Culture

The term ‘culture’ derives from the Latin word ‘cultura’ which means ‘cultivation of the field’, videlicet specific human groups examining their environment as a reaction to the prevailing demands in special regions in special periods of time. Thus it concerns the totality of all results of human action and behaviour, but especially those aspects such as language, religion, ethics, institutions of family and state, rite, technique, art, music, literature, philosophy and science, as well as (mentality) fashion, sport, recreational activities. architecture, nourishment and much more. It is, of course, possible to enlarge this list of criteria as necessary in order to standardise and identify an urban culture, as felt by its members and observers. It is sufficient for this paper to refer to those criteria which were of enormous importance for medieval Hanseatic towns. Apart from affiliation to the Christian community, the Middle Low German language must be cited as a key cultural factor, a language which was understood in all important trading centres around the Baltic Sea. From the 13th century it also increasingly superseded Latin as the language of administration and Middle High German as the language of

poetry. The intensive contacts of German craftsmen and merchants with the local residents in Scandinavia, the Baltic provinces and the Slavic settlement area also resulted in a manifold linguistic approach, so much so that the share of Middle Low German borrowings of the total thesaurus of medieval Swedish is estimated to be fifty percent. At the same time German merchants also took on originally Slavic terms such as ‘Dornse’ or ‘Prahm’.

A connecting language as an instrument of communication is not only a requirement for the development of an independent culture, but is also required for the development of independent institutions. The culture of the Hanse developed in the institutional ‘town’ dominated by merchants amidst an otherwise agrarian-structured feudal society. Actually these towns with a municipal law, a conception of the term citizen and a town wall had different grades of liberty from the ruler, varying from wide liberty to an empire sovereignty. However, they were all characterised by common economic and social factors: trade and craft were the crucial branches of the economy while agriculture and even fishing were not very important. A level of prosperous merchants, of which the Council was recruited, dominated in policy and administration.

So it is not astonishing that the ground plans of all Hanseatic towns (for example, Figure 20) are surprisingly similar even if they have different dimensions: they always have a circular or oval ground plan and mostly they are situated at a river surrounded by a town wall with many towers. The national long distance roads go through two, three or even four gates

97

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 113

98

The archaeology of medieval towns: case studies from Japan and Europe

The harbour was obviously located at the waterside with its specific facilities such as quay walls and bridges, as well as berths for the ships, storehouses for the goods, cranes and scales. If the town was situated at a river, the shipyards, wrecking places and places for rope-making were located on the opposite bank. In addition, chapels, hospitals for incurables, places of execution and pest cemeteries also belonged to the municipal area of a town.

Figure 21: Hansestadt Lübeck, Excavation Hundestraße 917, the surface of a cess pit.

Figure 22: Hansestadt Lübeck, an imported vessel from Venice or Syria.

into the town. At the crossroads of these streets the central market with the pillory and the town hall with the court arcade were situated as were the prison, the house of the executioner, the scales, the stalls of the money-changer and the mint, all mostly found near the market. In the bigger towns special markets, such as those for trade in meat, cattle or wood, emerged.

Usually the biggest church in town was also in the immediate vicinity of the market. Depending on the size of the parish, other churches existed or, in case of a diocesan town, the cathedral of the bishop. Monasteries were at the edge of the town, often directly at the town wall, and, in some cases, even outside the fortification. If the monasteries were centrally situated, mostly this can be put down to later enlargements of the town. Furthermore, in many towns there was a Holy Ghost Hospital as well as numerous houses of clerical brotherhoods or beguines.

The burgess house can be cited as an obvious instance for a common culture of the Hanseatic towns. The streets of houses of all great Hanseatic towns, whether it be Lüneburg or Wismar, Danzig or Tallin, were characterised by houses with street-oriented gables and a high hallway, a low first floor and a steep roof, so-called hallway houses. The facades of surviving buildings have often changed since the Middle Ages but the common eaves walls mostly retain a medieval building fabric.

In single cases, even for a building researcher, it is not possible to assign an uncaptioned photograph of a particular individual row of houses to a specific town. The differences exist only in tendencies, such as the building material utilised. Whereas brick nearly exclusively dominates in the towns from the southern Baltic Sea coast as far as the Baltic states, further north in Visby or Tallin its place is taken by limestone. Other differences result from different building periods. Apart from some few exceptions, pure Gothic or even pure Romanesque facades can only be found in Lübeck, Stralsund, Tallin and Visby. The renaissance dominates in Danzig and Lüneburg, and Wismar is characterised by facades totally changed by baroque additions and alterations. Sometimes there are also certain stylistic devices, which are very often in one town, such as the windows and doors framed by cable moulding in Lüneburg, crossbars made of limestone in Tallin or the terracottas in Lübeck and Lüneburg.

The similarities are continued inside the buildings: the large front building was rarely used for dwelling, but for trade and industry. The living space was in the wing buildings behind the front buildings. Storehouses, stables and above all the indispensable cess pits, were located in the rear of the narrow, rectangular plots.

While on the one hand these mainly regional parallels naturally result from practical considerations (the same problems lead to the same solutions) they also reflect a standardised legal position. Magdeburger Law applied in many towns of the east German hinterland, but Lübecker Law was given to most of the coastal towns. Lübecker Law, which was recorded in both Latin and the Middle Low German language about 1224, stipulated to builders in a very detailed way as to how they had to build their houses (such as

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 114

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Figure 25: Hansestadt Lübeck, a wool-knitted cap.

Figure 23: Hansestadt Lübeck, Stoneware.

Figure 27: Hansestadt Lübeck, some examples of jewellery.

Figure 26: Hansestadt Lübeck, one of the so-called ‘Trippen’. Figure 24: Hansestadt Lübeck, a belt of silk with applications of silver, 250 cm long, second half of 12th century.

material, orientation, drawing off the rainwater, location of the cess pits and stables, and common eaves walls).

Material culture can also be examined in everyday artefacts. The following almost exclusively discusses finds from Lübeck but bibliographical reference is also made to comparable finds from other towns of the Baltic Sea. In fact, the similarity of finds located within these towns is striking, an observation which has only become clear in recent years. While there were already manifold contacts between colleagues in Lübeck and the German Democratic Republic, Poland and the Soviet Union before the border opening of 1989, the possibilities for intensive comparison studies were strongly limited. In addition, urban archaeology in the east, especially in towns of the former German Democratic Republic, was strongly neglected, apart from intensive research in towns such as Elbing or Riga. Since 1989, however, this inconsistent state of research no longer obtains with many area excavations in

Figure 28: Hansestadt Lübeck, a tripartite travelling set.

locations such as Rostock, Stralsund and Greifswald. It is now possible to make intensive comparative studies in spite of the fact that there is still a shortage of synthesised data. In addition, the situation has been much improved by access to the finds’ assemblages from Lithuania, Latvia and Estonia.

Knowledge of medieval household objects has been exceptionally well enhanced through numerous finds from Lübeck, mainly from cess pits (Figure 21). It is not possible here to list them all but some should be mentioned such as spoons, forks, baskets, boxes, barrels, tubs and mirrors. Most utensils for the kitchen or the table were made of wood or ceramic. Turned plates or shallow bowls as well as beakers and bowls made of single staves have been found in hundreds during the larger excavations. Glass vessels from regional production as well as high-quality imported vessels from Venice and even Syria (Figure 22) can be added. Equally, glass vessels also exist from Greifswald, Stralsund, Elbing and Tartu.

99

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 115

100 The archaeology of medieval towns: case studies from Japan and Europe

plastic reliefs such as rosettes, blossoms, prunts or similar things and with figurative reliefs of faces or figures of nuns or knights. The finds from pottery waste show that such elaborately decorated vessels were also produced in Lübeck.

Figure 29: Hansestadt Lübeck, Medieval games. 29.1-29.4.

Clay was, after wood, the favoured material of the Middle Ages. Clay was plentifully available, easy to quarry, arbitrarily shapeable and, after firing, water impermeable, hard and persistent. The palette of objects made of clay is numerous. Apart from dishes, which will be examined below, the range includes architectural ceramics (bricks, tiles, terracottas and floor-tiles), as well as small pieces of furniture, firecovers and spit-supports. There are also small objects such as marbles, spindle whorls or sculptures.

Household dishes remain the most important find category for medieval archaeology. During every large excavation, tens of thousands (and, in some cases even hundreds of thousands) of sherds are recovered. Pottery enables chronological classification of associated features such as walls, wells, cess pits and similar features. It has been possible to date the high medieval pottery assemblages from Lübeck to relatively small periods of time (essentially quarter-centuries). Globular pots clearly dominated in the medieval households. This was the classical cooking pot without any standing device or standing base, sometimes with rod or normal handles. The spectrum of functional types was not enlarged until the end of the 12th and the early 13th centuries, when the so-called tripod cooking pots appeared, globular pots with three feet and rod handle, which required a plain hearth surface. In addition, there were jugs, pitchers, beakers, frying-pans and bowls. Generally, the locally-produced pottery, the hard greyware, was a short-living bulk good without any decoration. The drinking and pouring dishes were an exception, being made either of redware or of proto-stoneware or stoneware. Mostly the redware is glazed, often polychrome, and decorated with

Whereas the near water impermeability of the redware was achieved by glazing, the proto-stoneware and later the stoneware had to be fired much harder than the earthenware. For this process, special clay was required, which was not available in Lübeck, as well as more complex kilns to achieve higher temperatures. The stoneware was therefore imported from the Rhineland and southern Lower Saxony. The size of this pottery trade must not be underestimated; the pottery assemblages in Lübeck often consist of ten or even twenty percent stoneware sherds (Figure 23). These elaborate import goods were by no means only bought by members of the upper social levels. The equivalent sherds were also found in the waste of craftsmen-households, but only in lower proportions.

As well as buildings and their contents, clothes belong to the elementary requirements of everyday life, even if they were more subject to casual temporary fashions than house building. Apart from illustrations and the late medieval dress codes, manifold information has come to light through archaeological investigations, especially through cess pit finds. On occasion thousands of foursquare clothing fragments have been located in waste chutes, examples of what was obviously medieval toilet paper. Completely preserved clothes are much rarer. The fragmentary discoveries, however, are informative concerning production techniques such as fabric weaves or knitted fabrics, if less frequently useful with regard to costume historical questions. However, questions about the distribution of special materials such as silk (Figure 24) or about the first appearance of wool-knitted fabrics (Figure 25) or felt, can be answered. As a rule leather, unlike clothing pieces, was not cut up before it was dumped into the cess pits. As a consequence, often complete or nearly complete preserved shoes, boots or the so-called ‘Trippen’ (Figure 26) have been recovered, enabling the development of reliable chronologies of footwear.

All artefacts belonging to clothing or which were worn on the body have been combined in the collective term ‘costume accessories’. This means that jewellery (Figure 27) such as finger rings, jewellery needles or necklaces (which were often made of precious metal or were studded with gemstones, but which were mostly of bronze or copper, and, in single cases only of wood, iron or polished bone) form part of such assemblages. Finds of precious metal are extremely rare. Such objects were carefully guarded and were rarely lost; jewellery artefacts, which became unusable, were melted down again, as were buttons, girdles and laces as well as bronze buckles

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 116

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

and belt-attachments belonged to the costume accessories.

Artefacts used for hygiene or for personal hygiene are found surprisingly often, such as elaborately decorated combs of bone or horn, as well as tweezers of iron, bronze or even gilded. In one cess pit from Lübeck, a tripartite travelling set was found, consisting of a tooth brush, a pick and an ear scoop, which were fastened together by a stud (Figure 28). Knives also belong to costume accessories and, located in single cases, daggers. The handles of knives were often made of wood or bone. In the broadest sense it is also necessary to include pilgrim badges and pilgrim shells within this group of costume accessories.

Nutrition is a third elementary requirement. It must be emphasised that written sources or illustrations contain surprisingly many references about gala dinners for special events, menu sequences and even recipes, but they are not informative about everyday nutrition of the citizens. Urban archaeology is able to close this gap to a certain extent. Thousands pieces of food waste are recovered during every large excavation. Depending on the thoroughness of the recovery, these are small finds such as fruit stones, grains or fishbones and much larger animal bones or shells, which are hard to overlook. Botanical and zoological analysis of these finds shows an impressive variety of foodstuffs, although it is not possible to draw any conclusions about the amount of consumed food per meal. Apart from different cereals, onion, leek, beans, peas and fruit species such as plum, cherry and apple have all been identified.

The bones of domestic animals clearly dominate against game animals, although deer, rabbit, wild boar and seal have been found as, in single cases, has the brown bear which was still native in the early 13th century but which must have come rarely on the table. More than ninety percent of the bone assemblages consist of the bones of pigs, cattle or goats/sheep; bones of goose or chicken can be added. However, dogs, cats, horses and rats are relatively rare. Apart from the proportions of animal species it is possible to draw conclusions about the height of the animals, the age at slaughter and the proportion of sexes. It is more difficult to find fishbones but numerous fish for everyday consumption such as perch, pike, plaice and herring have been identified through excavation material, as well as oyster and mussel by their shells. Surprisingly no striking differences of species diversity have been determined either between different excavations or between different medieval households. Nearly the same range of animals was located in the assemblage of the excavation on the area of the former Benedictine monastery as were found in the assemblage of the mercantile plot at Alfstraße 38 and in the assemblages from the craftsmen quarter. Even the proportions were comparable.

The more pleasant sides of medieval live, namely games and free time (Figure 29) are also observable from numerous archaeological finds and, together with illustrations and written sources, disprove the assertion that medieval society was anti-children. However, in individual cases, it is sometimes not possible to distinguish toys for children from toys for adults. Thus, small clay-balls with a diameter of 1.3 to 2.2 centimetres which are found quite often, were almost certainly used as marbles. The marble game was extremely popular in the Middle Ages as numerous regulations of the court attest, in which the game was prohibited. These injunctions were obviously without much success. In a siltedup well, 44 partly-glazed marbles were found during an excavation in the Cistercian monastery St Johannes. It is not possible to clarify whether the nuns or the novices indulged in the game.

Dice have also been recovered. This game was popular as is verified by regulations and prohibitions as well as limitation on special person groups. According to current understanding, dicing would belong to the world of the adults as would playing counters and playing boards. The same applies to wooden bowling bowls or the balls with which games like tennis could be played. Many such finds were recovered during investigations at the former waterside market at the Trave between the wharfage and the town wall. They consisted of bowling bowls and dice but also toys of made of wood and bone. An iron jew’s-harp and a bone flute show that music was also made. It can be assumed that not only were ships unloaded and goods offered at the waterside market but also that this site was a place of amusement.

Examples of those toys which can be imagined in the hands of children include small rattles made of clay with a hollow body filled with small stones and midget vessels such as small jugs, pitchers, pots or tripod cooking pots. These vessels were formed exactly as the larger archetypes, including the tiny handle, spouts, glazes or tripod feet. They must have functioned as doll dishes.

Small clay figures are very impressive. Finds from pottery waste show that these dolls and also animals such as horse, deer, dog or bird were produced in astonishing numbers. An ensemble of two small horses and a bird were found in a single well, again in Lübeck’s Johannis monastery. The horse with mounted rider and modelled shield is particularly interesting; modern associations would assign tournament toys to boys but it was located in a nunnery.

Metal toys, including ones made of precious metal, were rare: only a single small horse and a bird, both made of tin, have been found in Lübeck. Wooden toys were relatively cheap to produce and must have been widespread, but are only preserved under humic burial conditions. However, several

101

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 117

102 The archaeology of medieval towns: case studies from Japan and Europe

-

Fig 30 Sakai, Cross section of Sakai, 15th-16th centuries.

Figure 31: Sakai, Cross section of Sakai, 16th century.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 118

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

carved wooden dolls with formed arms and faces exist as well as numerous spinning tops, which were similar to those used until few years ago. In addition, the hull of a small sailing ship and midget weapons (a crossbow, an axe and a bow) have been found. It must be assumed that naturally, as it is still today, weapons and other utensils of adults were copied in wood. In addition, it must also be remembered that in the Middle Ages, play was not without any intention but served as preparation for the future life as an adult.

Lübeck’s position in 1300

As set out above, Lübeck developed to be the only German medieval metropolis at the Baltic Sea by the end of the 13th century. The entire peninsular between the Wakenitz and the Trave with an area of about 120 hectares was settled, the buildings mostly consisting of hallway houses with streetoriented gables on to the main street and street-oriented eaves in the side roads. Around 1300 the town had about 20,000 inhabitants, Lübeck being at the peak of its political and economical power and, for contemporaries, considered to be the head of the Hanseatic League. Its cultural influence must have been extensive accordingly, covering areas such as architecture, urban infrastructure and case law but also aspects of everyday culture such as fashion and recreational activities.

Of course, it is not possible to identify the bearers of the cultural contacts by archaeological methods, but it is known from written sources that professional as well as familial connections of the merchants were extremely close between often distant Hanseatic towns. Many of Lübeck’s merchant families sent their sons for business or for education to Riga or Dorpat, but also to London, Bergen or even Novgorod. Many of them settled in these locations permanently. The manifold connections become apparent through wills for example, in which legacies were granted to single persons or institutions of other towns. Vice versa, many merchants from the Baltic provinces and Flemish merchants came to Lübeck. Insofar, it is not astonishing that there must have been a permanent cultural influence, although with tendencies to marginal temporal delays from west-to-east and, alternatively, in single cases occasionally the other way around.

Lübeck and Sakai: a comparison

The Norwich conference provided an opportunity to compare urban development in medieval Europe and Japan in general terms. Comparisons of individual towns must, however, remain a doubtful process. Any comparison is affected by different research intensity and different states of publication. Nevertheless an attempt can be made, mainly by examining

things in common and things in contrast. Eleven areas of comparison can be observed between the towns of Lübeck and Sakai as follows (see Figures 30 and 31):

Both towns were trading centres, which had their beginning in the 12th century and reached their peak in the period from the 14th to the 16th centuries. The long-distance trade was in both towns the basis of the wealth.

The craftsmen in Lübeck only produced for the local requirements, whereas an export oriented industry is verified for Sakai. From the 13th century the settled area in Lübeck covered an area of about 120 hectares. In Sakai the area was twice this large and covered, in the 16th century, about 250 hectares. The town of Lübeck had roughly 30,000 inhabitants in the year 1500. Figures are not available for Sakai but, after the great fire in the year 1615, more than 20,000 houses seem to have been destroyed.

In both towns the settlement was very intensive. In Lübeck as well as in Sakai the thickness of archaeological deposits approaches four metres. Occasionally, deposits reach a depth of five metres. The structure of the settlement was probably different. Lübeck shows a complete densely built-up area with stonebuilt houses with street-oriented gables, whereas it seems that there were large gaps between the houses in Sakai. From the 12th century there are no indications for agriculture in Lübeck, whereas the land was still farmed in the eastern districts of Sakai in the 16th century.

In Lübeck residential buildings and store-houses in the 12th century were still built of wood, but from the early 13th century they were mostly built of brick. It seems that in Japan there were still half-timbered houses in the 16th century. Toilets and wells belonged to the infrastructure of the plots in both towns. Harbours, markets, streets, waterfronts and public buildings such as town halls belong to the infrastructure of both towns.

Lübeck was a fortified settlement from the very beginning. A city wall made of bricks was built in the early 13th century which encircled the entire town. Initially, however, Sakai remained unfortified. The town was protected by a broad ditch and earthen ramparts but not until the late 16th century.

In both towns the wealthy merchants formed the upper strata of the society. The members of the council were drawn from the patriciate (or the merchant princes). Both towns were relatively independent from the feudal princes of the surrounding countryside.

103

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 119

104 The archaeology of medieval towns: case studies from Japan and Europe

In summary, it is quite possible to establish structures, functions and even institutions in common. However, this is not really astonishing, because similar problems result in similar solutions everywhere. As has been illustrated above, much of the urban environment of Lübeck was replicated, at least in part, by smaller but similar settlements around the Baltic Sea. These towns shared similar functions and thus developed in a similar manner. The cultural differences between the medieval Baltic and Japan were profound but the social and commercial requirements were clearly similar, leading to the developments of urban functions which were also similar. However, such a conclusion must stay in these general terms, until more detailed comparative data is available. Translated by Ulrike Oltmanns and Jonathan Scheschkewitz

References

von Brandt, A. 1974. Die gesellschaftliche Struktur des spätmittelalterlichen Lübeck (The social structure of Late Medieval Lübeck). Untersuchungen zur gesellschaftlichen Struktur der mittelalterlichen Städte in Europa, Reichenau-Vorträge 1963-1964 (Vorträge und Forschungen 11) (Investigations into the social structure of European towns in the Middle Ages. Lectures and Research 19631964). Sigmaringen: 215-239. Koppman, K. (ed.) 1884. Detmar-Chronik von 1101 bis 1395 (The Chronicles of Detmar, 1101-1395). Die Chroniken der deutschen Städte, Band 19. Die Chroniken der niedersächsischen Städte, Lübeck, Bd. I. (The Chronicles of the German Town Volume 19. The Chronicles of the Towns of Lower Saxony, Lübeck, Volume 1. Leipzig.

Dumitrache, M., Kliemann, K., Legant-Karau, G. and Remann, M. 1987. 800 Jahre Besiedlung im “Kaufleuteviertel”. Bisher größte Ausgrabung der Nachkriegszeit in Lübeck. (800 years of settlement in the merchants’ quarters. Major excavations in postwar Lübeck to the present day). Lübeckische Blätter 147, Heft 6: 85-89.

Ellmers, D. 1972. Frühmittelalterliche Handelsschiffahrt in Mittelund Nordeuropa (Maritime trade in the early Middle Ages in central and northern Europe). Offa-Bücher Band 28. Neumünster, Wachholtz.

Ellmers, D. 1985. Bodenfunde und andere Zeugnisse zur frühen Schiffahrt der Hansestadt Lübeck. Teil 1: Bauteile von Koggen (Discoveries and other evidence for early sea trade from the Hanseatic town of Lübeck. Part 1: Elements of the ‘koggen’ ships). Lübecker Schriften zur Archäologie und Kulturgeschichte 11: 155-162.

Ellmers, D. 1992. Bodenfunde und andere Zeugnisse zur frühen Schiffahrt der Hansestadt Lübeck. Teil 2: Bauteile und Ausrüstungsgegenstände von Wasserfahrzeugen aus den Grabungen Alfstraße 38 und An der Untertrave/Kaimauer. Koggen (Discoveries and other evidence for early sea trade from the Hanseatic town of Lubeck. Part 2: shipping equipment from the excavations at Alfstraße 38 and An der Untertrave/Kaimauer. ‘Koggen’ ships). Lübecker Schriften zur Archäologie und Kulturgeschichte 18: 7-21.

Ennen, E. 1975. Die europäische Stadt im Mittelalter (The European town in the Middle Ages). Göttingen, Vendenhoeck and Ruprecht.

Erdmann, W. 1982. Hochmittelalterliche Baulandgewinnung in Lübeck und das Problem beider Gründungssiedlungen: Erste Befunde aus den Gebieten Große Petersgrube und An der Untertrave (Landfill of the High Middle Ages in Lübeck and the problem of the foundation settlement: preliminary observations from the area of the Große Petersgrube and An der Untertrave). Lübecker Schriften zur Archäologie und Kulturgeschichte 6: 7-31.

Fehring, G. P. 1982. Grabungsbefunde zum slawischen Burgwall Bucu und zur landesherrlichen Burg mit zugehörigem Brunnen im Burgkloster zu Lübeck - ein Zwischenbericht (Excavations of the Slavonic castle of Bucu, the Lord’s Castle, the associated pond and the castle cloister at

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 120

Medieval urbanism and culture in the cities of the Baltic –with a comparison between Lübeck, Germany and Sakai, Japan

Lubeck: a preliminary report). Lübecker Schriften zur Archäologie und Kulturgeschichte 6: 77-98. Fehring, G. P. 1989. “Domus lignea cum caminata”. Hölzerne, turmartige Kemenaten des späten 12. Jahrhunderts in Lübeck und ihre Stellung in der Architekturgeschichte (Wooden kemenaten with towers of the late 12th century in Lübeck and their meaning in terms of archirectural history). Hammaburg N.F. 9: 271-283.

Fehring, G. P. 1997. Besiedlung und Bebauung vor Errichtung des Heiligen-Geist-Hospitals (Settlement and construction prior to the building of the Holy Ghost Hospital). Lübecker Schriften zur Archäologie und Kulturgeschichte 25: 31-38. Gläser, M. 1985. Befunde zur Hafenrandbebauung Lübecks als Niederschlag der Stadtentwicklung im 12. und 13. Jahrhundert. Vorbericht zu den Grabungen Alfstraße 36/38 und An der Untertrave 111/112 (Finds of the buildings on the harbour edge in Lübeck as an example of urban development in the 12th and 13th centuries. Preliminary report on the excavations at Alfstraße 36/38 and An der Untertrave 111/112). Lübecker Schriften zur Archäologie und Kulturgeschichte 11: 117-129.

Gläser, M. 1989. Archäologische und baugeschichtliche Untersuchungen im St. Johanniskloster zu Lübeck. Auswertung der Befunde und Funde (Investigations into the archaeology and history of construction at the Church of St Johannis in Lübeck: an evaluation of the findings). Lübecker Schriften zur Archäologie und Kulturgeschichte 16: 9-120.

Gläser, M. 1990. Die Lübecker Burg- und Stadtbefestigungen des 12. und 13. Jahrhunderts (The urban defences and castle of Lübeck in the 12th and 13th centuries). Archäologisches Korrespondenzblatt 20(2): 227-234.

Gläser, M. 1992a. Die Ausgrabungen in der Großen Petersgrube zu Lübeck. Befunde und Funde. Mit zwei Beiträgen von W. Erdmann (Finds and observations from the excavations in Großen Petersgrube in Lübeck. With two contributions by W. Erdmann). Lübecker Schriften zur Archäologie und Kulturgeschichte 18: 41-185.

Gläser, M. 1992b. Archäologische Untersuchungen auf dem Gelände des ehemaligen Burgklosters zu Lübeck. Ein Beitrag zur Burgenarchäologie (Archaeological investigations at the location of the former castle cloisters in Lubeck. A contribution to the archaeology of castles)’. Lübecker Schriften zur Archäologie und Kulturgeschichte 22: 65-121. Graßmann, A. (ed.) 1988. Lübeckische Geschichte (History of Lübeck). Lübeck, Schmidt-Römhild.

van Haaster, H. 1991. Umwelt und Nahrungswirtschaft in der Hansestadt Lübeck vom 12. Jahrhundert bis in die Neuzeit. Ein Überblick über bisherige paläoethnobotanische Untersuchungen (Environment and subsistence in Hanseatic Lübeck from the 12th century to modern times. An overview of palaeoethnobotanical studies)’. Lübecker Schriften zur Archäologie und Kulturgeschichte 21. 203-222.

Holst, J. C. 1985. Zur Baugeschichte der Häuser Alfstraße 36 und 38 in Lübeck. Ein Zwischenbericht (The construction history of houses in Alfstraße 36 and 38 in Lübeck). Lübecker Schriften zur Archäologie und Kulturgeschichte 11: 131-143. Jestrzemski, D., Kühne-Kaiser, S. and Mührenberf, D. 1988. Katalog vorgeschichtlicher Funde in der Hansestadt Lübeck: die neue Bearbeitung der archäologischen Karte. Rahden, Verlag Marie Leidorf.

Lagler, K. 1982. Einige vorgeschichtliche Keramikscherben aus der Lübecker Innenstadt (Prehistoric ceramic sherds from central Lübeck). Lübecker Schriften zur Archäologie und Kulturgeschichte 6: 335-337.

Legant-Karau, G. 1993. Vom Großgrundstück zur Kleinparzelle. Ein Beitrag der Archäologie zur Grundstücksund Bauentwicklung Lübecks um 1200 (Dividing large properties into small parcels. A contribution from archaeology of the construction and property history of Lübeck around 1200). In: Gläser, M. (ed.) Archäologie des Mittelalters und Bauforschung im Hanseraum. Rostock, Konrad Reich: 207-215.

LKSH 1997-2004. Lübecker Kolloquium zur Stadtarchäologie im Hanseraum I-IV (Lübeck Colloquia on Urban Archaeology in the Hanseatic League), edited by Manfred Gläser. Lübeck, SchmidtRömhild.

Mührenberg, D. 1989. Archäologische und baugeschichtliche Untersuchungen im Handwerkerviertel zu Lübeck. Befunde Hundestraße 9-17. Mit einem botanischen Beitrag zu den spätmittelalterlichen und frühneuzeitlichen Pflanzenresten von Henk van Haaster, Amsterdam (Archaeological Investigations and the history of construction of the artisans’ quarters in Lübeck. Findings at Hundestraße 9-17. With a botanical contribution to the plant remains from the late Middle Ages and Early Modern period by Henk von Haaster) Lübecker Schriften zur Archäologie und Kulturgeschichte 16: 233-290. Mührenberg, D. 1993. Der Markt zu Lübeck. Ergebnisse archäologischer Untersuchungen (The market at Lübeck: results of archaeological investigations). Lübecker Schriften zur Archäologie und Kulturgeschichte 23: 83-154.

Müller, U. 1992. Archäologische und baugeschichtliche Untersuchungen der Budenzeile Hüxstr. 78-82 in Lübeck (Archaeology and the sequence of construction of the market stalls at Hüxstraße 78-82 in Lübeck). Lübecker Schriften zur Archäologie und Kulturgeschichte 22: 167-200.

Neugebauer, W. 1964/65. Der Burgwall Alt Lübeck. Geschichte, Stand und Aufgaben der Forschung (The castle wall of Old Lübeck: history and the state of knowledge)’. Offa 21/22: 127-257.

Radis, U. In press. Die Steinbauperioden der Ausgrabungen im Lübecker Kaufleuteviertel (The phase of stone buildings in the merchants’ quarters in Lübeck). Lübecker Schriften zur Archäologie und Kulturgeschichte 27.

Radis, U. 2002. Neue Erkenntnisse zur frühen Besiedlung im Norden des Lübecker Stadthügels (New findings about the settlement to the north of the town hill in Lübeck). Archäologische Gesellschaft der Hansestadt Lübeck. Jahresschrift 4: 44-47.

105

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 121

106 The archaeology of medieval towns: case studies from Japan and Europe

Schalies, I. 1992. ‘Archäologische Untersuchungen 1992 zum Hafen Lübecks. Befunde und Funde der Grabung An der Untertrave/Kaimauer (Archaeological investigations in 1992 at the harbour of Lübeck. Finds and observations from the excavations at Untertrave/Kaimauer)’. Lübecker Schriften zur Archäologie und Kulturgeschichte 18: 305-344.

Schalies, I. 2002. Die Großgrabungen in der Lübecker Altstadt (Large excavations in the Old Town of Lübeck). Jahresschrift 4 der Archäologischen Gesellschaft der Hansestadt Lübeck. Lübeck: 13-19. Schmeidler, B. (ed.) 1973. Helmoldi chronica Slavorum. Hannover.

Schmidt, J.-P. 1993. Jungbronzezeitliche Siedlungsreste aus der Lübecker Innenstadt (Remains of settlement from the early Bronze Age in central Lübeck). Lübecker Schriften zur Archäologie und Kulturgeschichte 23: 327-332.

Stephan, H.-G. 1978. Archäologische Untersuchungen auf dem Markt in Lübeck; Diskussionsbeiträge zur frühen Besiedlung des Stadthügels (Archaeological investigations in the market in Lübeck: a contribution to discussions of the early settlement of the town hill). Lübecker Schriften zur Archäologie und Kulturgeschichte 1: 81-91.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 122

Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100

Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100 Henri Galinié Summary

This paper explores the development of the post-Roman urban network in France north of the River Loire. It examines the roles of urban centres, establishes a typology and argues that, in contrast to development in neighbouring regions such as England or the Rhine Valley, the inherited topography of Roman towns, notably civitas capitals, provided a permanent framework, or armature, which was supplemented by economic networks of new settlements but not replaced by them. Rather, the armature of Roman urbanism proved particularly durable, persisting for two millennia despite varying degrees of importance influenced by changing political and economic conditions.

Introduction

In France, the question of continuity of towns from the Roman to Medieval periods does not need to be addressed as far as the civitas capitals of the Roman province are concerned; a large majority of these Roman towns remained in use as medieval towns. Rather, the question is to determine whether or not it is possible to define a specific urban character for the intervening centuries. A context for this examination is provided by the observation, even if it remains a matter of debate, that the antique town can be considered as a political centre and the medieval one as a productive centre according to the still useful Idealtypen of Max Weber (1921).

It is now accepted among scholarly specialists of antiquity that, from the late third-century up to the seventh century, Gaul should be looked upon as late antique (l'antiquité tardive). Thereafter, they see the onset of the Early Middle Ages (haut Moyen Age) although, for their own part, medievalists would view this period starting in the sixth century. In consequence, as far as towns are concerned, it means that, according to the standing point, at least two centuries can be looked upon either as being the end of something or the start of something else. It must also be admitted that the Early Medieval town lasts up to the eleventh-century.

A large time span is thus opened to discussion, from c. 300 to c. 1050. When discussing Roman towns, AD300 coincides with two phenomena: the real start of Christianisation of the urban space of almost all civitas capitals as well as the construction of a rampart defending a small part of the previous town in most cases. For archaeologists it also fits with the start of formation of

dark earth over large areas intra-and-extra muros. These deposits are most often interpreted as a proof of abandonment or ruralisation of previous urban areas. Consequently very little fresh archaeological information is available (Macphail et al 2003).

If, however, study concentrates on the centuries between c.700 and c. 1100, the urban situation of France can be examined from three different points of view, namely what happened to Roman towns?; what new happened concerning towns?; and what led to the 1100 urban situation?

Of course these three questions are interrelated but the first one is more interested in changes to the Roman urban heritage, the second with the definition of a specific urban phenomenon within the period, and the last one with the process which led to the medieval urban armature or framework. The first two questions deal with towns as such in the period, while the last one deals with settlements which appear to be medieval towns whatever they were before. All three have implications with regard to the political and economical functions of towns.

In the course of a short paper these three aspects, which would each deserve separate development before being synthesised, can only be discussed quickly.

1. The three main roles of the urban settlements

Much effort needs to be made in order to make sense of the very sparse and heterogeneous information available from archaeological and written sources. Accordingly, in order to facilitate a comprehensive approach to the urbanisation of France, 7th to 11th century urban settlements need first be considered according to the main roles that they fulfilled:

- administrative control over inherited territories distinct from kingdoms; - economical control over production and/or circulation of goods in networks at various scales;

- everyday service to a relatively dense population.

These roles do not directly define types of settlements as certain towns played more than one role. Some places can be considered in the historical process which led from classical to modern towns while others can be considered as assemblages in networks because they functioned as part of a system whose time span was relatively short. For instance, if the civitates lasted as chief towns for more than a millennium and as towns for

107

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 123

108 The archaeology of medieval towns: case studies from Japan and Europe

more than two millennia, wics only lasted for a few centuries, mainly in relationship with civitates.

What does the written evidence say?

Between 700 and 1100, five main terms were in use to designate complex settlements with some urban character: civitas (or its synonym urbs), castrum, vicus, portus and burgus.

All of these terms with an urban connotation had several (as well as changing) meanings in the course of the period under consideration, except for burgus. It is possible to summarise the overall change in their use as follows:

At the start c. 700

Civitas 1 is dedicated to regional political centres, both civilian and religious, including bishoprics. It always concerns Roman walled cities, most of them being walled towns of Late Roman date.

Castrum 1 is dedicated to small walled settlements of Late Roman origin, often forts on main roads or rivers like Chinon or on the limes and also to fortified Roman vici such as Maastricht or Namur. They often are also designated as Vici 2.

Vicus 1 is dedicated to villages with a church or to suburban settlements with no defensive aspects like those around the monasteries of Saint-Denis near Paris or of Saint-Martin in Tours.

Vicus 2 is dedicated to small towns also known as Castrum 1.

Vicus 3 starts to be used (sometimes as a suffix) to designate some sort of emporia or toll stations such as Quentovic or Dorestad. Its use in Gaul in this sense is very limited. It often refers to settlements abroad and is thought to be associated with long distance trade or exchange.

Portus 1 as a technical term describing a riverine part of a town or a simple wharf (or landing-stage?), and not a type of settlement as a whole, comes to an end.

Portus 2 designates emporia like Vicus 3. It is used for Quentowic, Dorestad and Sandwic(h).

Burgus is absent of the terminology in use around 700.

c. 950 and onwards

Civitas 2 keeps its original use but is extended to new settlements which are recent episcopal sees with no relationship either to a Roman origin or even to a rampart. Examples include Liège and Antwerp. Castrum 2 designates newly-built strongholds around suburban monasteries at a little distance from a walled civitas. It also starts to be used for fortified monasteries (such as Corbie and SaintOmer), seigneurial or royal manors and for castles like Blois, Compiègne or Nimègue and later Douai or Brugge.

Vicus 2 and Vicus 3 reach the end of their use.

Portus 3 occurs more frequently. It designates fluvial settlements related to local or regional traffic such as like Valenciennes, Deventer or Tiel.

Burgus is used to designate a human concentration around (extra-muros) a civitas or a castrum. It does not define a legal status for a given area at the time. It replaces Vicus 1.

What appears clearly is that it is impossible to rely on the use of these terms to describe and fully understand the urban phenomena at work.

The next part of this paper is dedicated to an explanation of the necessary temporary typology in order to examine how these centres or places formed the urban armature of France and the way in which they were linked in various networks. It is necessary to distinguish, on the one hand, between armature which corresponds to the structural frame of towns in the longue durée (which, in France, is more than 20 centuries) and, on the other hand, successive as well as contemporary networks which correspond to short-term organisations, that is lasting a few centuries, linked to economical conjuncture. The discussion concentrates on that part of Gaul which lay north of the Loire.

2. A typology of urban settlements

This typology seeks to stress the uses and roles of complex settlements in the period 700-1100. The three proposed functions are intended to underline the particular distribution and changing character of roles which can be noted in Northern Gaul during these centuries. They should be found at both the start and at the end of the considered time span under different conditions. The three main functions are administrative control, economical control and everyday service to urban population.

Administrative control of territories

This kind of control was exerted over those permanent territories which can be described as historical, traditional and inherited from Roman administration. Their geography was independent from that of kingdoms.

The places where that control was exerted are to be found at two scales, those of chief towns and local centres: civitates and castra.

Chief towns were part of the heritage of the administrative organisation of Roman Gaul whether of early or late dates, that is from the 1st century AD to the late 4th century AD. From 700 onward they appear to have been the traditional places of location for any sort of authority, whether civilian or ecclesiastical. They were the centres of a stable territories.

Local Centres were either ancient small towns or new places which were situated in the territory of chief towns. They reinforced the urban armature and were, or became, the head of

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 124

Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100

administrative subdivisions located in Castra 1 and Vici 2.

Residences were new places of post-Roman date related to political power and visited at intervals by kings and their courts for the administration of their kingdoms. Originally they could be rural villas, castles, monasteries or part of them. Residences could also be in the hands of kings' representatives for the administration of a special district in a realm. This type of settlement can be ascertained by two main criteria: originally they presented no urban character and they were not a permanent location for the central authority. They were visited at irregular intervals.

Economical control over production and/or circulation of goods in networks

Chief Towns

The Roman urban armature can easily be defined if one deals with administrative centres: the civitas capitals. The long-term project La Topographie chrétienne des Cités de la Gaule (TCCG 1986-), which will soon be completed, gives the best controlled available data of the Roman urban heritage as far as civitas capitals are concerned, up to the mid 8th century. By 400 AD, about a hundred civitas capitals covered the part of Gaul now in France. Most of these administrative centres were still present almost 1400 years later when the new chefs-lieux de départements were chosen in 1791. This permanence can be interpreted as a sign of the stability given to a place by its status of chief town in France. However, in no way does it ascertain that every chief town had been a town (as defined supra) permanently throughout the centuries.

Control places were permanent settlements established by some sort of authority at strategic locations for military or economical purposes, on borders, rivers, and at coastal lines. They did not have their own territory. They were linked to long distance exchange, whether barter or trade (Vicus 3, Portus 2).

These chief towns are better documented than any other urban centres because of their administrative status. They appear in all sorts of sources at many occasions, whether written as they relayed the central authority, or architecturally by the presence and maintenance of public buildings. Nevertheless they remain documented only on administrative and political aspects; their economic role appears underestimated.

It must be accepted, even if there is no direct evidence for measuring the quantities of goods which were exchanged, that chief towns – civitates - controlled the economical activity in their territory to ensure their regular victualling all year round.

Control places

Local centres, as seen before, were also inserted in local exchange networks within territories of civitates or between them (Vicus 1, Castrum 1, Castrum 2).

Everyday service to a relatively dense population

Service places appeared as concentrations of population and were to be found in the four types of settlements: chief towns, control places, local centres and residences. However, they may also have been characterised by a new type of complex settlement originally (c. 700), non-urban monasteries. The question is to determine if such locations were related to trade or not (in 700: Vicus 1; after 850: Burgus).

As such towns do not represent a category here. They are defined by their supposed multifunctional capacities. They can integrate all the preceding types of features but they have to fulfil the criteria of a diversified economic base and a socially stratified population, both of which are difficult to assess and are subject to controversy (Samson 1994, 1999). This outline typology is not of general interest in the study of European medieval towns. It simply aims to help in the study of change in the urban organisation between Antiquity and the Middle Ages in the particular historical situation of Northern Gaul. It lays more stress on the use made of places than on their status. The places themselves must now be considered separately.

On the other hand, it is impossible to enumerate the small towns of Roman Gaul. Some have totally disappeared but field archaeology in the past two decades has revealed that a great number of supposed medieval creations were in fact superimposed on some sort of Roman settlement, the urban character of which remains to be proved. Such study may ultimately affect the pessimistic interpretations of the impact of economic or demographic crisis during the so called Dark Ages. For instance, Caen in Normandy, Vervins ChâteauThierry in Picardy, Blois in the Loire Valley and Issoudun in Berry (among many others) have revealed traces of Roman and/or Early Medieval occupation. The same is true for many towns in Belgium (Verhaeghe 1990). Here, in this paper, it is necessary not to deal with the phases of town creations as such but merely to note the phenomenon, one which cannot be quantified but only observed on the basis of examples.

Of course, the chief towns were also the control places for their own traditional territories inherited from Roman times, but other places acquired a similar position during the high Middle Ages. Ancient frontier military forts, on rivers or on sea coasts as well as small local centres, were promoted as control places by kings or princes. Maastricht on the River Meuse (Dijkman 1994), or Boulogne (Delmaire 2004) on the Channel are good examples. These, plus new creations, were used to control flows of people or goods at very different scales. Some were of purely local importance while others were toll stations in long distance

109

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 125

110 The archaeology of medieval towns: case studies from Japan and Europe

exchange such as Dorestad in the Rhine delta or Quentowic on the Channel (Lebecq 1991).

Local centres

Among local centres were the small towns of Roman origin about which little is known, as stated above, as well as new developments of small agglomerations around monasteries, residences or castles from the 10th century onward. They were established in the territories of traditional chief towns. Most of them are linked to some sort of river traffic such as Valenciennes or Douai (Verhulst 1999). Monasteries, either of late Roman date like Saint-Denis near Paris or new ones like Corbie, give good examples of these new typical local centres of the Early Middle Ages. Their role was first related to their own internal functioning as centres of their domaines and as centres of public financial administration.

Residences

Residences must be understood here as architectural complexes adapted for the reception of the court at irregular intervals. Consequently they were not necessarily permanent structures of unifunctional character. Some appear to have been new palaces in chief towns, other were independent rural villas or domaines, others parts of urban or rural monasteries. Their diversity stresses the itinerant character of political control in the period. Residences underline the places which were considered as important as well as reliable by kings or princes. Clichy near Paris, Aachen in Germany, Attigny or Compiègne are good examples. Only a few of the rural residences had an urban development.

Service Places

In order to fulfil the need for specialised products not necessarily linked to trade but rather linked to the expression of social roles (at least at the start of the period), craftsmen and their families could be grouped outside formal settlements whether in a permanent way or at irregular intervals, seasonal for instance. In Gaul, many large monasteries and chief towns generated such service places. Indeed a great number of them were to be found in the immediate surroundings of chief towns, as in Tours, Soissons, Arras or Reims. The earliest use of the Latin term burgus in the 9th century referred to such settlements. They are to be seen as a correlate of the consumption role of the above types in the Early Middle Ages.

Towns

The word town is used here as a technical term and not as a generic one. The term denotes where urban life takes place. It is not necessarily related to the administrative status or to the temporary function of a capital. Any of the above places can

also be described as a town. It is a matter of numbers of individuals. In northern Gaul, the places which show the most developed urban character at the end of the period were certainly Rouen and Metz.

In return this meant that the above categories did not necessarily have to be, or produce, town-like settlements in order to play their roles in the period 700-1100.

3. Urban armature and urban networks

The considered time span c. 700 - c. 1100 can be described from several points of view even if one only focuses on the urban question. At the scale of Northern Gaul, on political and administrative grounds, these centuries see the last decades of the Merovingian kingdoms in the 8th century, the few decades of the Carolingian empire in the early 9th century and its subsequent subdivision into three kingdoms after 843, then the emergence of territorial princes in the 10th and 11th centuries as well as that of the Capetian kingdom in the 11th century. On economic grounds, these centuries see the importance of the Mediterranean surpassed by that of the new interior sea formed by the Channel, the North Sea and even the Baltic, from the 8th century onwards. Indeed, Northern Gaul from the late 7th century is part of a wider economic system. In addition, the impact of the Viking raids (c. 850 - c. 920) is not to be neglected as, even if these were voluntarily overestimated by the ecclesiastical written sources, they accelerated changes in the control of land and populations by local lords, whether religious or laymen. This phenomenon is well documented by the generalisation of Castrum 2 in the 10th and 11th centuries.

Urban armature

The urban armature of Northern Gaul can be seen from at least the Late Roman period, as the Christian Church adopted the Roman hierarchy of civitas capitals and provinces in order to install its episcopal and archiepiscopal sees. This armature proved to be very stable throughout the centuries between c. 400 and c. 750 where the previous Roman armature had been strong. However, on the outskirts of Roman Gaul, it appears to have been more fragile as in the province of Köln (see Fig. 1).

Ecclesiastical organisation is the only means of evaluating the duration of urban armature as it was the only centralised system producing some sort of coherent archive recorded through bishops' assemblies or councils. Nevertheless, the duration of the armature in no way means its permanent effectiveness. For instance, in the Tours province, it is obvious that the Breton bishoprics were not under the authority of the Tours archbishop in the Early Middle Ages.

The civil organisation can be viewed at two levels, that of the Merovingian kingdom's capitals and that of the local representatives of the kings, such as counts. Most of the

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 126

Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100

kingdom’s capitals were located in ancient Roman civitas capitals. As the kingdoms were shared between heirs and thus often reorganised, their geography seems to have been very mobile, so that the capitals of kingdoms could appear as temporary royal residences for a few years or decades, rather than acting as stable centres of kingdoms. Nevertheless, some places were often chosen as capitals such as Orléans, Paris, Soissons and Senlis.

At the same time, such places together with others which were never among the temporary capitals but had permanent royal representatives such as counts at local level, acted as chief towns whose tenants were often changed according to political events. They remained the stable seats of some sort of local administration and their geography was closely linked with the ecclesiastical one.

The same place could thus be, successively or at the same time, a royal residence as well as an ecclesiastical and civilian chief town in the 8th century. The History of Gregory of Tours at the end of the 6th century gives a vivid picture of this situation. The importance of adopting a traditional location in order to legitimise power has long been stressed and has been well illustrated in the case of early Medieval Gaul (Brühl 1975). Being a chief town was therefore more decisive than being the capital of a kingdom at that time in Gaul.

By 750, some 38 chief towns were the seat of a bishopric in a wide area of northern Gaul (restricted to the limits of France). All of them were of Roman origin and had had the status of civitas capital since at least the 4th century, except for the castra of Dijon and Châteaudun which in Table 1 do not appear as bishoprics but as residences respectively for the bishops of Langres and Chartres. On the other hand very few known Late Roman civitas capitals are absent from this 8th-century list, only Jublains, Saint-Quentin and Boulogne for different reasons (Delmaire 2004, Beaujard and Prévot 2004). If one now looks downwards, it must be noticed that all these chief towns were present in the late 18th century when the location of the then new prefectures was discussed, except for Thérouanne which had been sacked by Charles V of Spain.

In considering control towns, it is difficult to ascertain the real role of small Roman towns before the 9th century when texts show a hierarchy in which castra and vici of Roman origin often appear as intermediates of civitas capitals in the organisation of counties or pagi with no major structural change. A specific character of Northern Gaul as far as its inscription in a wider economic system is concerned is the importance of the administrative tradition. As Stéphane Lebecq showed (1991), Quentowic and Dorestad are first to be considered as toll stations. They controlled the flow of goods from and to England in the case of Quentowic; and from and to the Far North in the case of Dorestad. Both places lay at the border of the Frankish kingdom. Rouen and Amiens, although civitas capitals, seem to have played a similar role.

If castra and vici show a real longevity, as well as civitates, in the Gallic-French urban armature, on the other hand Quentowic and Dorestad were obviously so tightly linked to the 7th - 9th century economic system that they disappeared with it, being replaced at a short distance by places more capable of addressing the new organisation of the 10th century, one no longer based on the control of long distance trade but on that of territories in which new forces were exerted by emerging landlords and castles.

Networks

The discussion of networks is far more difficult to develop, mainly as it could well be biased by a lack of information considering the existing situation/ background around 700. Determination of networks remains quite elusive. The centuries 700-1100 obviously appear to be subdivided into two main phases which of course are part of a continuum. To model the situation, however, almost all ancient Roman civitates capitals at this period can be considered as chief towns but ones with no real urban life. They cannot be qualified as real towns by comparison to Roman and Medieval ones but must be looked upon as consumption and ceremonial centres for a less stratified population. Nevertheless, as such, they are the real towns of the time while emporia were no more than specialised settlements as Samson has pointed out (1999). In Gaul, the main actors within or assembled with the chief towns were monasteries as well as royal fiscal authorities and domaines whose dispersed properties generated a permanent flow of goods for their own use between rural estates and chief towns (or, in some cases, the seat of the monastery or palace).

In parallel, but in a totally separated system of exchange even if it used the same technical means, prestige goods circulated at a long distance scale. Emporia, which Hodges (2000) describes as type A, could have been part of this system but could also have functioned as rural palaces. Their existence is difficult to ascertain as the socially complex societies which produced them chose a double form of investment: one was expressed in ephemeral realisations, of which prestige artefacts (which were never on sale) as well as burials, present some rare vestiges; the other, more traditional, expression was in permanent realisations such as religious or civilian architecture, whether on the ground or in texts and memories.

Between 750-850, the internal functioning of monasteries, royal administration and domaines changed little. However, it seems that the economic system entered a new phase, moving from exchange to trade with a pre-eminent place given to coinage and monetary-based activities. Obviously the greatest monasteries, royal authorities and domaines also entered into that new sphere. The main urban characteristic of the first period is that most activity did not take place in traditional ancient urban centres or

111

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 127

112 The archaeology of medieval towns: case studies from Japan and Europe

territories. Rather, emporia were the new settlements as well as monasteries or fiscal centres which worked in networks at a supra-regional scale.

The second period was characterised by a concentration of activity in ancient locations whether towns, monasteries or fiscal centres, but also by newly developed monasteries or castles. Castra and portus of the 9th-10th centuries are the best examples of the new settlements which were then transformed into towns in the 11th-12th centuries. These became the centres of new districts and territories of various sizes in the feudal control of space.

In conclusion, when considering urban armature and networks, a specificity appears when comparing the situation of Northern France or Gaul with that of surrounding zones like England or the Rhine Valley. It is the ascertained permanence of administrative control which ensured a centrality of importance to Roman towns while residential and economic aspects appear to have been less crucial. A new form of medieval urbanisation also emerged, however, one based on the capacity of particular centres, first of all monasteries and palaces and later of castles, to organise control and exchange in the limits of smaller territories.

Bibliography

Anderton, M. (ed.) 1999. Anglo-Saxon trading centres: beyond the emporia. Glasgow, Cruithne Press.

Atsma, H. (ed.) 1989. La Neustrie, les pays au nord de la Loire de 650 à 850 (La Neustrie: the country to the north of the Loire from 650 to 850). Actes du colloque international (Rouen 1985), 2 volumes.

Bayard, D. 1999. Amiens. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 199-210.

Beaujard, B. and Prévot, F. 2004. Introduction à l’étude des capitales ‘éphémères’ de la Gaule (Ier s.-début VIIe s.) (Introduction to the study of ephemeral capitals in Gaul (1st-7th centuries). In: Ferdière, A. 2004 (ed.) Capitales Éphéméres, des capitales de cités perdent leur statut dans l’antiquité tardive (Ephemeral capitals: the loss of status of capital cities in late antiquity). Actes du colloque de Tours (mars 2003), 25e Supplément à La Revue Archéologique du Centrede la France, Tours: 17-37.

Benredjeb, T. 1999. Abbeville. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 187-198.

Blary, F. 1999. Château-Thierry. In Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue

Archéologique de Picardie, Special No 16: 55-59

Brogiolo, G. P. and Ward-Perkins, B. (eds.) 1999. The idea and ideal of the town between late Antiquity and the early Middle Ages. Leiden, Brill. Brühl, C. R. 1975. Palatium und Civitas. Cologne-Vienne, Böhlau.

Callebaut, D. 1994. Résidences fortifiées et centres administratifs dans la vallée de l’Escaut (Ixe-Xie s.) (Fortified residences and administrative centres in the valley of the Escaut (1st-10th centuries). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 93-112. Cantino Wataghin, G. 1999. The ideology of urban burials. In: Brogiolo, G. P. and Ward-Perkins, B. (eds.) The idea and ideal of the town between late Antiquity and the early Middle Ages. Leiden, Brill: 147-180.

Collart, J.-L. 1999. Saint-Quentin. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 67-128.

De Witte, H. 1994. La fortification de Bruges (Belgique), les fouilles de 1987-1989 au ‘Burg’ de Bruges (The fortifications of Bruges (Belgium), excavations from 1987-1989 at the ‘Burg’ of Bruges). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 83-91.

Delmaire, R. 2004. Permanence et changements des chefs-lieux de cités au Bas-Empire: l’exemple du nord-ouest de la Gaule Belgique (Permanence and change from towns to cities in the late Empire. In: Ferdière, A. (ed.) Capitales Éphéméres, des capitales de cités perdent leur statut dans l’antiquité tardive (Ephemeral capitals: the loss of status of capital cities in late antiquity). Actes du colloque de Tours (mars 2003), 25e Supplément à La Revue Archéologique du Centre de la France, Tours: 39-50.

Demolon, P. 1994. Naissance díune cité médiévale flamande, l’exemple de Douai (The birth of a Flemish medieval city: the example of Douai). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 47-58.

Demolon, P., Galinié, H. and Veraeghe, F. (eds.) 1994. Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale.

Desachy, B. 1999. Noyon. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 171-178.

Desachy, B. and Guilhot, J.-O. (eds.) 1999. Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 128

Permanent urban frameworks (‘armature’) and economic networks in Northern France c.700 – c.1100

Dijkman, W. 1994. Maas-Tricht, lieu de défense et centre religieux (Maas-Tricht, defences and religious centre). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in northwest Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 35-40.

Drinkwater, J. and Elton, H. 1992 (eds.). Fifth-century Gaul: a crisis of identity?. Cambridge, Cambridge University Press.

Durand, M. 1999. Senlis. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 179-186.

Duvosquel, J.-M. and Dierkens, A. (eds.) 1991. Villes et campagnes au Moyen-Age (Towns and countryside in the Modern Age). Liège.

Favreau, 1995. La formation d’un nouveau réseau urbain dans le Centre-Ouest du Xe au XIIIe siècle (The formation of a new urban network in the Centre West from the 10th to the 13th centuries)’. In Guignet, P. and Hiernard, J. (eds.) Les réseaux urbains dans le Centre-ouest atlantique de l’antiquité à nos jours, Actes du colloque de Poitiers (mai 1993) (Research into urban networks in central-west Atlantic from antiquity to the present day). Proceedings of the Poitiers Colloquium, May 1993). Mémoires de la Société des Antiquaires de l’Ouest, 5e sèrie, T. III: 91-108.

Fémolant, J.-M. 1999. Beauvais. In: Desachy, B. and Guilhot, J.O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 145-152. Ferdière, A. (ed.) 2004. Capitales Éphéméres, des capitales de cités perdent leur statut dans l’antiquité tardive (Ephemeral capitals: the loss of status of capital cities in late antiquity). Actes du colloque de Tours (mars 2003), 25e Supplément à La Revue Archéologique du Centrede la France, Tours.

Galinié, H. 1994. Emergence ou ré-émergence des villes dans le Nord-Ouest de l’Europe (VIIe-Xe s.) (The emergence or reemergence of towns in north-west Europe 7th – 10th centuries). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in northwest Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 9-16.

Gauthier, N. 1989. Rouen pendant le haut Moyen-Age (650850) (Rouen during the high Middle Ages (650-850). In: Atsma, H. (ed.) 1989. La Neustrie, les pays au nord de la Loire de 650 à 850 (La Neustrie: the country to the north of the Loire from 650 to 850). Actes du colloque international (Rouen 1985), Vol. 2: 1-20.

Gauthier, N. 1999. La Topographie chrétienne entre idéologie et pragmatisme (The Christian topography between ideology and pragmatism). In: Brogiolo, G. P. and Ward-Perkins, B. (eds.) The idea and ideal of the town between late Antiquity and the early Middle Ages. Leiden, Brill: 195-210.

Germand, J. and Morel, M. 1999. Chaumont-en-Vexin. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 153-156.

Guignet, P. and Hiernard, J. (eds.) 1995. Les réseaux urbains dans le Centre-ouest atlantique de l’antiquité à nos jours, Actes du colloque de

Poitiers (mai 1993) (Research into urban centres in central-west Atlantic from antiquity to the present day). Proceedings of the Poitiers Colloquium, May 1993). Mémoires de la Société des Antiquaires de l’Ouest, 5e sèrie, T. III

Guyon, J. 2004. Introduction au colloque Capitales Éphéméres (Introduction to the colloquium on ephemeral capitals). In: Ferdière, A. (ed.) Capitales Éphéméres, des capitales de cités perdent leur statut dans l’antiquité tardive (Ephemeral capitals: the loss of status of capital cities in late antiquity). Actes du colloque de Tours (mars 2003), 25e Supplément à La Revue Archéologique du Centrede la France, Tours: 11-15. Hall, R. A. 2000. The decline of the Wic?. In: Slater, T. R. 2000 (ed.) Towns in decline AD 100-1600. Aldershot, Ashgate Publishing: 120-136 Hodges, R. 2000. Towns and Trade in the Age of Charlemagne. London, Bloomsbury.

Jorrand, J.-P. 1999. Laon. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 61-65.

Lebecq, S. 1991. Pour une histoire parallèle de Quentovic et Dorestad (Towards a parallel history of Quentovic and Dorestad). In: Duvosquel, J.-M.and Dierkens, A. (eds.) Villes et campagnes au Moyen-Age (Towns and countryside in the Modern Age). Liège: 415-428.

Leroux, P., Margerie, P. and Marin, J.-Y. 1999. Les Èglises de Caen du VIIe au XIIe s. (The churches of Caen from the 7th – 12th centuries). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 125-132.

Loseby, S. T. 1992. Bishops and cathedrals: order and diversity in the fifth-century urban landscape of southern Gaul. In: Drinkwater, J. and Elton, H. (eds.) Fifth-century Gaul: a crisis of identity?. Cambridge, Cambridge University Press: 144-155. Loseby, S. T. 2000. Urban failures in late-antique Gaul. In: Slater, T. R. (ed.) Towns in decline AD 100-1600. Aldershot, Ashgate Publishing: 72-95.

Macphail, R. I., Galinié, H. and Verhaeghe, F. 2003. A future for Dark earth? Antiquity, 77(296): 349-358.

Orselli, A. M. 1999. L’Idée chrétienne de la ville: quelques suggestions pour l’Antiquité Tardive et le Haut Moyen Age (The Christian idea of the town: some suggestions about Late Antiquity and the High Modern Age). In: Brogiolo, G. P. and Ward-Perkins, B. (eds.) The idea and ideal of the town between late Antiquity and the early Middle Ages. Leiden, Brill: 181-194.

Petitjean, M. 1999. Compiègne. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 157-166.

Picard, J.-C. 1994. L’espace religieux dans le ville médiévale (Religious space in the Medieval town). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe).

113

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 129

114 The archaeology of medieval towns: case studies from Japan and Europe

Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 115-124.

Riès, G. 1999. Crépy-en-Valois. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 167-170.

Rouche, M. 1989. Remarques sur la géographie historique de la Neustrie (650-850) (Observations on the historical geography of Neustrie (650-850). In: Atsma, H. (ed.) 1989. La Neustrie, les pays au nord de la Loire de 650 à 850 (La Neustrie: the country to the north of the Loire from 650 to 850). Actes du colloque international (Rouen 1985), Vol 1:1-2.

Roussel, D.1999. Soissons. In: Desachy, B. and Guilhot, J.-O. (eds.) Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 129-130.

Saint Jean Vitus, B. 1999. Vervins. In: Desachy, B. and Guilhot, J.-O. (eds.). Archéologie des villes, démarches et exemples en Picardie (The archaeology of towns: approaches and examples from Picardie). Revue Archéologique de Picardie, Special No 16: 139-144.

Samson, R. 1994. Populous Dark-Age towns: the Finleyesque approach. Journal of European Archaeology 2(1): 97-129

Samson, R. 1999. Illusory emporia and mad economic theories. In Anderton, M. (ed.) Anglo-Saxon trading centres: beyond the emporia. Glasgow, Cruithne Press: 75-90. Slater, T. R. 2000 (ed.) Towns in decline AD 100-1600. Aldershot, Ashgate Publishing.

TCCC 1986-, Topographie chrétienne des cités de la Gaule des origines au milieu du VIIIe siècle. Paris, 13 volumes (out of 15) published since 1986.

Verhaeghe, F. 1990. Ceramiekverspreiding en handel in de middeleeuwen: interpretatie en problemen. In: Carmiggelt, A. (ed.) Rotterdam Papers, A contribution to medieval archaeology, Teksten van lezingen gehouden tijdens het symposium ‘Handel, handelsplaatsen en handelswaar vanaf de vroege middeleeuwen in de vroege middeleeuwen in de lage Landen’ te Rotterdam van. Rotterdam, Bureau Oudheidkundig Onderzoek van Gemeentewerken Rotterdam: 2-3.

Verhaeghe, F. 1992. Continuity and change: links between medieval towns and the Roman substratum in Belgium. In: De Smet, R., Saerens, C. and Melaerts, F. (eds.) Studia varia Bruxellensia. Leuven, Peeters: 229-253. Verhulst, A. 1986. La vie urbaine dans les anciens Pays-Bas avant l’an mil (Urban life in the ancient Pays-Bas 1000 years ago). Le Moyen Age 92: 185-210.

Verhulst, A. 1987. Les origines urbaines dans le nord-ouest de l’Europe: essai de synthèse’ (Urban origins in North-West Europe: a synthetic essay). Francia 14: 57-81.

Verhulst, A. 1994. La ville et son émergence en Flandre (The town and its emergence in Flanders). In: Demolon, P., Galinié, H. and Veraeghe, F. (eds.) Archéologie des villes dans le Nord-Ouest de l’Europe (The archaeology of towns in north-west Europe). Actes du Ive congrès d’archéologie médiévale (Douai 1991), Société d’Archéologie Médiévale: 41-46. Verhulst, A. 1999. The rise of cities in north-west Europe. Cambridge, Cambridge University Press.

Weber, M. 1921. Wirtschaft and Gesellschaft. Tübingen, J.C.B Mohr.

Wood, I. 1994. The Merovingian kingdoms 450-751. London, Routledge.

Métropole de la Province ecclésiastique

Evêchés

Province de Mayence Germania Prima

Mayence, Worms, Spire, Strasbourg

Province de Cologne Germania Secunda

Cologne, Liège (après Tongres et Maastricht), Utrecht

Province de Trèves Belgica Prima

Trèves, Verdun, Metz*, Toul

Province de Reims Belgica Secunda

Reims*, Laon, Amiens*, Châlons, Beauvais*, Senlis*, Soissons*, Tournai*, Arras?*), Cambrai, Thérouanne* Lyon*, Mâcon, Chalon-sur-Saône, Lugdunensis Prima Autun* Langres*-Dijon

Province de Lyon

Province de Sens Lugdunensi Senonia

Sens*, Nevers, Auxerre, Troyes*, Orlèans, Chartres*-Châteaudun, Paris*, Meaux*, Province de Rouen Rouen*, Lisieux*, Evreux*, Sées*, Lugdunensis Secunda, Avranches*, Coutances*, Bayeux*

Province de Tours

Tours*, Angers*, Nantes*, Le Mans*, Lugdunensis Tertia Rennes*, Vannes*, Dol ?, Alet? Saint-Brieuc ?,Tréguier ?, Saint-Pol ?, Quimper ?

Table 1: The urban armature of Northern Gaul: episcopal sees c. 750

Mayence: towns outside the limits of France; Metz*: Roman civitas capital of early date; Strasbourg: promoted civitas capital in the 4th century

Sources: Topographie chrétienne des cités de la Gaule, vol. I à XII, 1986-2002, Sinclair 1985, Atlas de géographie historique de la France et de la Gaule, Loseby 1999, Urban failures in late-antique Gaul.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 130

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective Joanita Vroom Introduction: East and West It was an Englishman, Rudyard Kipling, who once wrote, as the opening lines of his most famous poem: ‘Oh, East is East, and West is West, and never the twain shall meet’, a phrase that is often quoted, inappropriately in my view, to imply a complete separation of cultures. In fact, Kipling was writing about exactly the opposite condition, as is made clear in the assertion that follows the opening lines: Till Earth and Sky stand presently at God’s great Judgment Seat; But there is neither East nor West, Border, nor Breed, nor Birth,

When two strong men stand face to face, tho’ they come from the ends of the earth! [Italics added]

An encounter between ‘two strong men’ is a good metaphor to describe the meeting that took place between the Dutch and the Japanese when the Dutch set up a trading post on Deshima Island in Nagasaki Bay in 1609, nearly 300 years before Rudyard Kipling wrote his poem The Ballad of East and West in 1889. As it happened, to pursue the metaphor further, the Dutch had to fight off quite a few other ‘strong men’ from the west who were interested in meeting, influencing and trading with, the east. This meeting between east and west has left some truly fascinating traces from an archaeological point of view, and perhaps none more so than in the field of ceramics.

In examining the production of Japanese porcelain and its distribution to the west, one can detect many mutual influences as well as crossovers in techniques, production and distribution, a process that contact between the Dutch and the Japanese only accelerated. From the mid-17th century onwards the Vereenigde Oost-Indische Compagnie (the Dutch East India Company), or ‘VOC’, began to export Japanese porcelain to the Indochina Peninsula and thence to Europe (Volker 1954: 117–17; Ōba 2004: 33–37). This was during a period when Europe witnessed a growing taste for things ‘Oriental’ (among them, the new beverage tea), and it coincided with a significant drop in the production of Chinese porcelain from the Jingdezhen kiln with the collapse of the Ming dynasty. The Dutch merchants placed This chapter was originally submitted in December 2004, with minor changes since.

large orders at Japanese kilns for vessels of specific forms and designs that it was thought would appeal to European customers (Rousmaniere 2004: 50–51), and in addition, on Deshima Island, used glazed porcelain tobacco pipes (the so-called oribe kiseru) made in Japan (Suzuki 2004).

The ceramic trade had an impact upon Japan, as well as Europe. A demand for so-called ‘kraak-style’ porcelain from the West spurred the already nascent Japanese porcelain industry in the Hizen region from the 1660s to 1740s, Japanese porcelain was imported in increasing quantities to the Netherlands, as is shown by archaeological sites in several Dutch cities (Baart 2000, 2001; Kaneda 2003). Drinking vessels such as bottles, small cups and saucers, or pots for tea and coffee were produced in Japan, after European models. Further, 18th-century dishes of Japanese porcelain sported a ‘VOC’ mark (the official company emblem for the Dutch East India Company), and even sometimes depicted Dutch merchants and their tobacco pipes (Figure 1; see also Ōba 2004: 36; and Shibata 2002: catalogue entries 276 and 466).

These imports of high-quality porcelain from the east in their turn provided an incentive to Dutch and other European manufacturers. The potters at Delft in Holland, for instance, copied Chinese blue and white designs of the Ming period and even patterns from Japanese porcelain on their ‘Delftware’ while, in Saxony, King Frederick Augustus the Strong tried to have porcelain in the Japanese Kakiemon-style dishes fired in his kilns at Meissen. This meeting of east and west can also be

Figure 1: Japanese porcelain bowl (1790-1810) with depiction of Dutch merchants (after Shibata, 276).

115

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 131

116 The archaeology of medieval towns: case studies from Japan and Europe

observed in some of the Mediterranean countries, as can be demonstrated by studies of pottery in the Aegean.

Eastern ceramics and the Aegean

Pottery finds in the Aegean that show an interaction between east and west include fragments of Chinese celadon ware and porcelain that have been recovered at several excavations on the western coast of Turkey (for example, at Ephesus, Pergamon, Miletus, Iznik, Istanbul—see Vroom 2005). It is known that Chinese celadon ware and porcelain had been highly prized for a long time in the countries of the Near East, firstly by Islamic rulers and later by the Ottoman sultans.

Ottoman appreciation of wares from the east is evidenced by the fact that the Topkapi palace in Istanbul now has one of the largest and most valuable collections of Chinese and Japanese porcelains in the world (see Krahl et al. 1986). It contains a total of 10,700 Chinese pieces, ranging from the late Song (13th-century) to the Ching period (1644–1912) as well as up to 730 Japanese porcelains (mainly Imari ware produced in and around Arita in southern Japan) dating from the 17th–19th centuries (a few Japanese porcelain vessels can also be found in the Sadberk Hanim Museum in Istanbul; cf. Carswell 1995: nos. 101, 116, 134–138, 145–146 and 163).

Further, porcelain dishes from China and from Japan have been found in the famous monasteries on Mount Athos in northern Greece, built into the walls of churches as decoration. The exonarthex of the church at Iviron, for instance, includes a dish of Japanese Imari ware, probably of the Genroku era (1688– 1703) (Carswell 1966: 84, Figure E, no. 22).

These eastern finds in the Aegean area are fascinating, and it is also instructive to compare aspects of the medieval pottery industries in the Aegean and in Japan, especially because so many technological innovations in both areas ultimately came from the east, in particular from China. In the Aegean, for instance, the use in Byzantine decorated ceramics of a whitecoloured slip (or engobe) as a coating over a darker fabric owed its origins to white-bodied Chinese porcelains. Further, the technology of using green and brown colours, together with Table 1: Chronological division of medieval and postmedieval ceramics in the Aegean

Figure 2: A so-called ‘fruit stand’ of the Middle Byzantine period (after Morgan 1942, pl. V).

linear patterns against a white background in lead-glazed sancai ware from Tang China (618–908 AD), was later also imitated by Islamic and Byzantine potters. In medieval Japan, on the other hand, brown- and black-glazed stoneware tea bowls from China (known as temmoku in Japanese) were imported from the 12th century and copied from the 14th century onwards (Rousmaniere 1996).

In this context, this paper seeks to present a short introduction to some technological and socio-economic developments with regard to ceramic production in the Aegean from roughly the 10th to the 15th centuries with a view to a general theoretical comparison to ceramic changes in medieval Japan. Before doing this, however, it will be useful to set out the chronological

Early Byzantine period

c. 7th–9th centuries

Late Byzantine/Frankish period

c. 13th–mid 15th centuries

Middle Byzantine period Turkish/Venetian period Early Modern period Modern period

c.10th–late 12th/early 13th centuries c. late 15th–18th centuries

c. 19th–mid 20th centuries c. mid 20th century

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 132

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective

Figure 3: Shapes of chafing dishes of the Middle Byzantine period (after Bakirtzis 1989, figs. 12-13).

framework for periods of ceramic change in the Aegean (Table 1).

This pottery chronology is divided into an Early Byzantine (or Early Medieval) period, a Middle Byzantine period (or High Middle Ages), a Late Byzantine/Frankish period (the so-called ‘Crusader period’ in the Aegean), a Turkish/Venetian period (the period of Ottoman and Venetian domination) and an Early Modern period (the period after the War of Independence of 1821). As in any such table, there is a variety of methodological problems involved in the terminology and chronology used but also great advantages over more traditional divisions, as discussed extensively elsewhere (see Vroom 2003: 25–29).

This article is restricted to a discussion of changes in production of the decorated tablewares in the Aegean (thus the pottery used for serving food and drink), and will not discuss the (no less interesting, but perhaps more complex) problems related to the changing shapes of the coarse domestic wares that played a role in transport, storage and cooking. While pottery finds specifically from the Aegean have been chosen as the subject of this paper, many changes in pottery technology discussed here can also be observed in many other regions in the eastern Mediterranean.

Tableware of the Early Byzantine to Middle Byzantine periods (c. 7th–10th century)

The introduction in the Aegean of lead-glazed pottery marks the transition from Late Antiquity to Early Medieval times. The relatively easy technique of lead-oxide glazing was already known in some parts of the Roman Empire but, in the 7th century, new types of lead-glazed cramics began to be produced in the area around Constantinople (modern Istanbul). During the 8th century, lead glazes, which produce translucency after firing at low temperatures (700º–800º C), became more common on tablewares, as is shown by dishes set on a pedestalled ring foot, also known as ‘fruit stands’ (Morgan 1942: 45, Plate V b; see also Figure 2). From this time onwards, lead

Figure 4: A chafing dish of the Middle Byzantine period (after Byzantine Hours, fig. 120).

glazes remained the principal method of glazing for many centuries in the Aegean: first for functional reasons as a sealant, later (from the 11th century onwards) as a decorative technique. This first phase of glazed pottery in the Aegean was characterised by chafing dishes, or vessels with a glazed bowl set on a hollow, ventilated stand (Morgan 1942: 36–42, Figures 24–28, Plates I–III; Bakirtzis 1989: Figures 12–13, see also Figure 3). The red fabric is coarse, sometimes with large lime inclusions, and relatively porous. These characteristics would make the vessel suitable for cooking. The colourless lead glaze was applied directly as a sealant to the coarse fabric, resulting in a dark brownish (or olive-green) tone. The glaze varies from some sparse spots (due to the application of the lead compound in powder form) to a thick glassy coating on the inside of the bowl.

It has been suggested that these chafing dishes were placed on the dining table during banquets, as a sort of portable brazier and cooking utensil (Figure 4). Their function could be either to heat food in the kitchen or to keep it hot at the table (Morgan 1942: 37; Bakirtzis 1989). The theory that food in the upper glazed bowl was kept warm by charcoal or by a small candle placed in the lower stand is confirmed by burnt parts in the fabric of complete vessels found at excavations. According to some scholars, chafing dishes are mentioned in Byzantine texts as saltsaria, saltsera, gararia or garera, because their main function was to prepare and serve warm sauces and, in particular, warm fish-sauces (the so-called garoi) on the table (Koukoules 1947– 55: V 162; Bakirtzis 1989: 55–65).

The evidence, however, is inconclusive, as many Byzantine pictures with dining scenes from this period show no chafing dish at all on the table (see Vroom 2003: 313–321). Being a lowvalue item made of earthenware (not silver), the chafing dish is evidently not depicted on the tables of scenes that show dining habits of the well-to-do classes in the Byzantine world (see Vroom 2007). The only exceptions are four miniatures in a Greek manuscript of the 11th century, now in the National Library in Paris, that show a chafing dish- or brazier-type vessel

117

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 133

118 The archaeology of medieval towns: case studies from Japan and Europe

above a fire next to the dining table—and always in combination with a metal samovar for mixing wine with hot water (cf. Omont 1908: fols. 42, 82, 117, 166).

Tableware of the Middle Byzantine period (c. 11th–early 13th century)

Ceramic finds in the Aegean confirm that from the end of the 11th century onwards, the practice of using glazed pottery for table or display purposes became much more widespread (Figure 5). In addition, the potters took much more trouble to ensure that vessels for the table, such as bowls and dishes, were pleasing to look at. They achieved this by covering the inner surfaces with a coating of white slip (in Greece known as bandana or astari) and a lead glaze, and further enhancing the surfaces with a colourful variety of incised and painted designs, coloured by copper (green) and iron oxides (brown). These technological innovations appear to have been inspired by the arrival of leadglazed polychrome wares from Tang China (618–908 AD), firstly in the Islamic world and subsequently in the west.

Perhaps this new zest for decoration in the west was related to the rise of a new social elite in the Aegean whose enjoyment of food and drink was enhanced by appreciation of artistic endeavour at the table. But perhaps it was even more related to the spread of glazing technology and the greater availability of raw materials. The more readily available metal ores, and the increase in figural expression in architecture and painting on buildings (evident from the 12th century), formed part of the aesthetic and technological background of the potters. The production of tablewares was part of a wider ‘socio-aesthetic’ development.

With this in mind, it is little surprise that the bulk of late 11th– 12th century tablewares found on Aegean sites consists of Fine Sgraffito ware (Figure 6). This ware includes mostly bowls and dishes decorated with fine line incisions through the white slip (or engobe) with a sharp tool. Further, on most sites, it is possible to observe Slip-painted ware, on which white slip was used as a decoration technique to paint the vessel surface (Figure 7), as well as Green- and-Brown Painted ware, on which the designs were painted with copper and iron oxides on a white slip (Figure 8; cf. in general, Morgan 1942 and Papanikola-Bakirtzis 1999). Decoration techniques from the 11th century onwards were innovative, the designs colourful and non-religious. Often the potters who engraved these wares drew upon themes from popular life for their inspiration, or upon the decorative vocabulary of the Near East. They decorated their ceramics with animals, musicians, dancers, hunters and (mythical) warriors. Sometimes they used apotropaic signs (such as the so-called ‘knot of Solomon’) or rapacious creatures (such as

Figure 5: Pottery fragments of the Middle Byzantine period (photo author).

Figure 6: Fine Sgraffito Ware of the Middle Byzantine period (photo author).

Figure 7: Slip-painted Ware of the Middle Byzantine period (photo author).

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 134

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective

Figure 8: Brown and Green Painted Ware of the Middle Byzantine period (after Morgan 1942, pl. XXII).

Figure 9: Fragment of Fine Sgraffito Ware with depiction of a dove (photo author).

Figure 10: ‘Pseudo-Kufic script’ in a wall of the Monastery of Hosios Loukas (photo author).

Figure 11: ‘Pseudo-Kufic script’ on Middle Byzantine pottery (after Morgan 1942, fig. 21).

lions, leopards, or hawks with their prey). Single doves represented birds of the chase or their preferred prey, and are thus related to the imagery of falconry (Von Wartburg 2001: 124–125; see Figure 9). Of course, these images referred to the life of knighthood, banqueting and hunting but it has also been suggested that these Byzantine designs were actually talismans with a protective significance to avert evil from the meal or perhaps from the owner or user of these vessels (DautermanMaguire and Maguire 1992: 9–10).

Inspiration for the decoration found on these incised and painted vessels of the Middle Byzantine period could also derive from the repertoire of medieval ornament employed by craftsmen working in other media and in other materials (cf. Jope 1972; Le Patourel 1968). One can observe, for instance, exactly the same design of the so-called ‘Pseudo-Kufic script’ (the angular script used in the Muslim world for inscriptions) in Fine Sgraffito ware as in the abstract wall-masonry created by masons in the walls of many 11th and 12th century monasteries and churches in the Aegean (see Megaw 1931–32: 104–106; Ettinghausen 1976). One such is the Monastery of Hosios Loukas in Boeotia, where a frieze of Pseudo-Kufic script is shown by the arrangement of brick fragments on the walls of the Theotokos church (Figures 10 and 11). Further, the spiral painted motifs on the Green-and-Brown Painted sherds found on many Central Greek sites are comparable to the ornamental, figural frescoes produced by painters in the crypt of the same monastery (Figures 12 and 13; see also Vroom 2003: 151–152).

However stylish and delicate this Byzantine pottery from the Aegean appears to be from the outside, though, this was a façade designed to hide the soft, rather coarse and porous fabric. The clay was poorly levigated and contains many calcium carbonate inclusions. These inclusions must have caused major problems for the medieval potters because they cause cracking of the clay during the firing process. What resulted, in fact, were rather unsophisticated and porous vessels, which the potters tried to cover up with a white slip and a lead glaze as sealant.

In addition, the shapes of these decorated 11th–12th-century wares are generally very simple. Thick-walled dishes and shallow bowls with a low ring foot are found but jugs are unusual (Figure 14). The open dishes and bowls come in similar shapes, but in a range of sizes. The larger open vessels were probably used for serving main dishes at the table, while those of smaller dimensions were used perhaps for side dishes or sauces. The wide, flat dishes and bowls of the Middle Byzantine period have, in general, large rim diameters, ranging from 24 to 30 centimetres (Figure 15). The function of these serving and eating vessels was probably intended for communal dining rather than for individual use at the table. However, because of their porosity, they could not have been very suitable for watery or greasy dishes.

119

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 135

120 The archaeology of medieval towns: case studies from Japan and Europe

Examples of similar shapes from the same period can be found in medieval Spain, where the Muslim community used to eat communally from large centrally placed dishes from Malaga (e.g. the ataifor) with their hands. Individual plates and small bowls were very rare here before the 14th century (Gutierrez 1997, 74). Comparisons can also be made with the so-called mokhfia (large polychrome painted dishes) from Fez in Morocco, which were used for the serving and communal consumption of couscous (Hakenjos 1988: 55; Form Table 2, numbers 4–8).

Tableware of the Late Byzantine/Frankish period (c. mid 13th–mid 15th century)

Figure 12: Frescoes in crypt of the Monastery of Hosios Loukas (photo author).

The open-shaped, thick-walled tablewares of the Middle Byzantine period seem to have gone out of use in the Aegean during the 13th century. They were gradually overtaken in the Late Byzantine/Frankish period by Sgraffito wares with more deep and narrow shapes, and sometimes one or two colours in the glaze (Figure 16).

The glaze colour of these Sgraffito wares is either made of copper (green), iron (ochre-yellow) or a combination of copper, iron and manganese (brown-purple). A new range of designs was also introduced. The emphasis of the engraved decoration was now rather on monograms, geometric and stylised floral patterns, although animals and human figures can still be found (Figure 17). In addition to the vast improvement in the quality of the lead glaze (which became thicker, with a more glassy appearance), a finer, thinly potted and more hard-fired ware replaced the previous thick, soft and coarse tablewares of the Middle Byzantine period.

An important innovation during the 13th century was the introduction of the tripod stand in the potter’s kiln, that allowed better distribution of heat and flow of air around vessels compared to simple stacking (Figure 18). This was required when very controlled conditions were necessary for potters to produce highly decorated pottery and complex vessel forms. The introduction of the tripod stand resulted in a tighter packing in the kiln and consequently a substantially increased output for the Aegean workshops, which had to compete with western and eastern imports during this period.

Especially interesting, though, is the change that occurred in shape: from the 13th century onwards, a clear increase of small, deep bowls with a high ring foot instead of shallow dishes with a low ring base can be seen (Figure 14). The rim width of the bowls is generally smaller than their Middle Byzantine predecessors, ranging from 17 to 20 centimetres. One of the most remarkable aspects of the Late Byzantine/Frankish period in the Aegean is the enormous increase of bowls and the dearth of cups, compared to the use

Figure 13: Pottery fragments of Brown and Green Painted Ware (photo author).

Figure 14: Average diameters of vessel shapes from the Middle Byzantine period, from the Late Byzantine/Frankish period and from the Turkish period (after Vroom 2003, table 7.3).

Figure 15: Shapes of dishes of the Middle Byzantine period (after Morgan 1942).

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 136

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective

of other types of pottery (mainly jugs, jars and dishes) in all periods (Figure 19). These bowls could have been used for liquid mixtures, or were perhaps even used as drinking vessels (like the so-called hanaps in medieval Europe). As vessels with fairly high sides and thicker glazes, bowls clearly imply that the contents were watery.

Figure 16: Elaborate Incised Ware of the Late Byzantine/ Frankish period (after Talbot Rice 1930, pl. I).

The introduction of new shapes in the Late Byzantine/Frankish period may also reflect, apart from new dining habits and new diets, a growing demand for ceramic versions of more expensive metal ware utensils, or perhaps the well-recorded increase in the wine trade (and therefore winedrinking among various classes) in late medieval society (McCarthy and Brooks 1988: 110). Cups, chalices and goblets are, for instance, also typical shapes of the later Middle Ages, and they clearly demonstrate the influence of metal forms on pottery (Figure 20). In addition, it is known from written sources that during the 14th century earthenware tablewares were generally considered rather low-value items in the hierarchy of materials (Kazhdan 1991: 2146). Prestigious metal drinking jugs and beakers (especially those made of silver, bronze and pewter) were much more valued, and some pottery of this period clearly imitates metal shapes. As always, the ‘missing artefacts’ (such as silver, pewter, bronze and treen) must be borne in mind when considering the full range of behavioural patterning evident in any one ceramic or glass assemblage found at excavations (Smart Martin 1989: 1–27).

Conclusion

Figure 17: Sgraffito bowls of the Late Byzantine/Frankish period (after Thessaloniki, 79).

Figure 18: Vessels stacked with tripod stilts (after PapanikolaBakirtzis 1992, fig. 17).

The study of the relationship between vessel shape and function as a source of historical and socio-cultural information is still a new and underdeveloped approach among archaeologists working in the eastern Mediterranean (see Vroom 2003: 237–238 for other examples). However, it is not hard to see that a farmer’s frying pan must have met needs quite different to those met by a nobleman’s wine jug, and these functions or requirements must have had a socio-economic and cultural context. Likewise, it seems quite obvious that the changes in the function of pottery can be related to changes in the fabric, the thickness of the walls of the vessel, the shape, the decoration and the applied slip or glaze.

The medieval pottery finds from the Aegean show, at least in case of the tableware, a number of very clear changes over time in the general shapes and techniques of the pottery. Noteworthy is the change from shallow, open vessels in the Middle Byzantine period to smaller, deeper bowls in the Late Byzantine/Frankish period. At the same time, innovations in decoration techniques in tablewares were introduced in the Middle Byzantine period that were practised in many areas of Greece without much change until recent times.

121

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 137

122 The archaeology of medieval towns: case studies from Japan and Europe

Figure 19: Function analysis of pottery finds in central Greece (after Vroom 2003, table 7.5).

In addition to the search for models of ceramic production and distribution in the medieval Aegean, it would seem fruitful to explore the links between fabric and shape of pottery on the one hand, and changes in cultural context and developments in social habits on the other. In the search for archaeological explanations, with all the technicalities of fabrics, slips and glazes, it is easy to forget that the main function of pottery has always been that of a common, everyday utensil for use in the kitchen, on the table, in the cupboard, and in transportation. Whether made by rural potters for the local consumer, or by organised workshops as a luxury item for long-distance trade, the main criterion for judging the shapes and proportions of medieval pots is how well they performed the function for which they were designed.

The questions to be addressed therefore have to do with the extent the technological and functional changes in pottery production from Byzantine times onwards in the Aegean were the result of technical innovations in the potter’s craft, the result of socio-economic factors (such as an increased demand for imports, from the east and other places), the result of changes in the consumption of food, or the result of changes in dining habits of the local population. Assuming that this last factor played a substantial part in changing pottery styles, which seems probable, the issue is how to get an adequate picture of the changing socio-economic circumstances and dining habits. In other words, we have to search for the people behind the broken pieces of pottery. One obvious way to proceed is to explore written sources and pictorial evidence for dining habits and the cultural concept of

Figure 20: Shapes of drinking vessels of the Late Byzantine/Frankish period (after Papanikola-Bakirtzis 1996).

food. Needless to say, however, the relation between written sources and iconographical representations on the one hand and shape and function of pottery on the other is not a simple one (see Vroom 2000; id. 2003, 303–334). It is beyond the scope of this short paper, whose aim is merely to provide some methodological perspectives on ceramic production in the Aegean, to explore this matter further. But the arduous process of dating pottery fragments and the use of challenging models to understand long-term developments in both medieval ceramic production and the wider medieval economy make one thing clear: the potential contribution of medieval pottery to the understanding of the history and material culture of Greek lands during this period.

Rousmaniere has pointed out that as archaeology in the historical periods in Japan receives more support, a clearer picture of the use of ceramics and of exchange mechanisms there is beginning to develop (Rousmaniere 1996: 51). Changes in pottery production and distribution in medieval Japan were connected to the introduction of pottery technology from China and Korea, the emergence of the tea ceremony and the increased trade in ceramics with China and southeast Asia. While direct links between East Asia and the Aegean may prove elusive—or indeed non-existent—in the medieval period, examination of the influences behind Japanese ceramic development may provide a fruitful parallel for similar work on ceramics of the medieval Aegean.

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 138

Medieval ceramic production in the Aegean, 1100-1600 AD: some considerations in an east-west perspective

Bibliography

Baart, J. M. 2000. Japanese porcelain finds in Amsterdam. In: Saga Kenritsu Kyūshū Tōji Bunkakan (ed.) The voyage of OldImari porcelains. Saga, Saga Kenritsu Kyūshū Tōji Bunkakan: 216-220. Baart, J. M. 2001. Nieuwe inzichten in oude huishoudens (New insights into old households). Jaarboek van het genootschap Amstelodamum 93: 195-209.

Bakirtzis, C. 1989. Byzantine ëTsoukalolaginaí (in Modern Greek). Athens. Byzantine Hours. 2001. Byzantine Hours. Works and Days in Byzantium, Athens.

Carswell, J. 1966. Pottery and tiles on Mount Athos. Ars Orientalis 6: 77-90.

Carswell, J. 1995. Chinese ceramics in the Sadberk Hanim Museum. Istanbul, Vehbi Hoc Foundation.

Dauterman-Maguire, E. and Maguire, H. 1992. Byzantine pottery in the history of art. In: Papanikola-Bakirtzis, D., Dauterman Maguire, E. and Maguire H. (eds.) Ceramic art from Byzantine Serres. Illinois, University of Illinois Press: 1-20.

Ettinghausen, R. 1976. Kufesque in Byzantine Greece, the Latin West and the Muslim world. In: A colloquium in memory of George Carpenter Miles (1904-1975), New York, American Numismatic Society: 28-47. Gutérriez, A. 1997. Cheap and Spanish. Meaning and design on imported Spanish pottery. Medieval Ceramics 21: 1-8.

Hakenjos, B. (ed.) 1988. Marokkanische Keramik (Moroccan ceramics). Düsseldorf, Hetjens Museum.

Jope, E. M. 1972. Models in medieval studies. In: Clarke, D. L. (ed.) Models in archaeology. London, Routledge: 963-990.

Kaneda, A. 2003. Japanese porcelain in a Dutch archaeological context. An archaeological study of Hizen porcelain from the cesspits of Amsterdam, Alkmaar, Delft and ës-Hertogenbosch. Unpublished MA thesis, University of Amsterdam. Kazhdan, A. P. (ed.) 1991. Oxford dictionary of Byzantium. Oxford, Oxford University Press.

Koukoules, F. 1947-55. Byzantine life and civilisation I-V (in Modern Greek). Athens.

Krahl, R., Ayers, J., Erbahar, N., Yücel, Ü. and Raby, J. 1986. Chinese ceramics in the Topkapi Saray Museum, Istanbul. London, Philip Wilson Publishers.

Le Patourel, J. 1968. Documentary evidence and the medieval pottery industry. Medieval Archaeology 12: 101-126.

Megaw, A. H. S. 1931-32. The chronology of some Middle-Byzantine churches. Annual of the British School at Athens 32: 90-130. McCarthy, M. R. and Brooks, C. M. 1988. Medieval pottery in Britain AD 900-1600. Leicester, Leicester University Press.

Morgan, C. H. 1942. Corinth XI: The Byzantine pottery. Cambridge, American School of Classical Studies at Athens. Ōba, K. 2004. Japanese porcelain: Imari for Japanese shoguns and European kings. In: Rousmaniere, N. C. (ed.) Jiki. Porcellana giapponnese tra Oriente e Occidente 1610-1760. Milan, Electa Mondadori: 26-37.

Omont, H. 1908. Évangiles avec peintures Byzantines de 11e siècle: reproduction des 361 miniatures du manuscrit grec 74 de la Bibliotheque Nationale (Byzantine Gospels with paintings from the 11th century: a reproduction of 361 miniatures from Greek manuscript 74 in the Bibliotheque Nationale). Paris, Impr. Berthaud Frères.

Papanikola-Bakirtzis, D. 1992. Serres: A glazed pottery production center during the Late Byzantine period. In: Papanikola-Bakirtzis, D., Dauterman Maguire E. and Maguire H. (eds.) Ceramic art from Byzantine Serres. Illinois, University of Illinois: 21-35.

Papanikola-Bakirtzis, D. 1996. Medieval glazed pottery from Cyprus: The workshops of Paphos and Lapithos (in Modern Greek). Thessaloniki.

Papanikola-Bakirtzis, D. (ed.) 1999. Byzantine glazed ceramics: the art of Sgraffito, Athens, Museum of Byzantine Culture.

Rousmaniere, N. C. 1996. Jian Ware tea bowls imported into Japan. In: Mowry, M. D. (ed.) Hare’s fur, tortoiseshell, and partridge feathers: Chinese brown- and black-glazed ceramics, 400-1400. Harvard, Harvard University Press: 43-58.

Rousmaniere, N. C. 2004. White gold: the porcelain for export manufactured in Japan and the diffusion of new beverages in Europe. In: Rousmaniere, N. C. (ed.) Jiki. Porcellana giapponnese tra Oriente e Occidente 1610-1760. Milan, Electa Mondadori 46-53.

Shibata 2002. Complete catalogue of Shibata collection, part 8. Arita, Kyushu Ceramic Museum.

Smart Martin, A. 1989. The role of pewter as a missing artefact: Consumer attitudes toward tablewares in late 18th-century Virginia. Historical Archaeology 23: 1-27.

Suzuki, B. T. 2004. Het verschaffen van Nederlandse emigranten in Japan met kleipijpen in de 17de en 18de eeuw (The provision of clay pipes for Dutch expatriates in Japan in the 17th and 18th centuries). KnasterKOPF 17: 7-8. Talbot Rice, D. 1930. Byzantine glazed pottery. Oxford, Clarendon Press. Thessaloniki 1986. Thessaloniki. Istoria kai techni (in Modern Greek), Athens.

Volker, T. 1954. Porcelain and the Dutch East Company (as recorded in the Dagh-registers of Batavia Castle, those of Hirado and Deshima and other contemporary papers 1602-1682). Leiden, Brill Archive.

Von Wartburg, M-L. 2001. Bowls and birds: some Middle Byzantine glazed bowls from Swiss private collections. In: Herrin, J., Mullet. M. and Otten-Froux, C. (eds.) Mosaic. Festschrift for A.H.S. Megaw. London, British School at Athens: 115-129.

Vroom, J. 2000. Byzantine garlic and Turkish delight. Dining habits and cultural change in Central Greece from Byzantine to Ottoman times. Archaeological Dialogues 7: 199-216.

Vroom, J. 2003. After Antiquity: ceramics and society in the Aegean from the 7th to the 20th centuries A.C. A case study from Boeotia, Central Greece. Leiden, Leiden University.

Vroom, J. 2005. Medieval pottery from the Artemision in Ephesus: imports and locally produced wares. In: Krinzinger, F. (ed.) Spätantike und Mittelalterliche Keramik aus Ephesos. Vienna, Austrian Academy of Sciences.

Vroom, J. 2007. The changing dining habits at Christ’s table. In: Brubraker, L. (ed.) Eat, drink and be merry (Luke 12:19). The production, consumption and celebration of food and wine in Byzantium. Aldershot, Ashgate Publishing.

123

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 139

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 140 1

Afterword

Afterword Richard Pearson In this volume we have introduced case studies of medieval towns and trade in the hope of encouraging dialogue among medieval archaeologists of western Europe and Japan. We also hope to provide western readers with an impression of the richness of the field of Japanese medieval archaeology in general. But what is there to be learned from such a sampler beyond superficial comparison and introductory anecdote? Hopefully archaeologists in co-operation with other specialists will ponder the historical processes which led to the rise of towns in the Middle Ages and similarities and differences in economic structures and mentalities in the two regions. In both Japan and Europe, we can see the emergence of towns with commercial quarters and storage facilities, craft production, defined public space, planned layout, religious edifices, elite residences, and communal facilities such as wells and latrines. The utensils of daily life include substantial numbers of traded objects as well as domestic products. Roads are well developed and in some cases there are port facilities. In both regions economic and political expansion occurs in the late first millennium AD. It is of interest that the European medieval world emerged with the waning of the Roman Empire and the decline of contacts with the Mediterranean, while the Japanese medieval period began at a time of declining diplomatic ties with China, increasing power of a new political force of warrior clans, and the decline of the Heian political system based on Chinese models.

The stimulation of commercial activities at the onset of the medieval period has captured the attention of archaeologists and historians (Hodges 1982, 1988). Hodges (1988: 4, 5) states that the medieval town is the focus of a new production system. In the beginnings of English urbanism, there is a shift from “Type A emporia (periodic market places), points where native gifts are treated as commodities, where the inalienable become alienable” to Type B emporia, tribute-based production centres, markets which involved the mobilisation of social labor to build enclosures, ditches, roads, etc. Hodges states that “urban archaeology points to the brilliant use of commodity production for seducing the English. His comment that coinage was used to lubricate this transition (1988: 6) resonates with the Japanese case in which Chinese coinage was circulated in abundance.

There are other directions in which comparisons of European and Japanese medieval sites might take us. We could discuss medieval mentalities. Keikō Hongō (2004) investigated the economic mentality of medieval people in Japan through paintings and documents. She investigated the nature of commodities, the mutual relations of gender and currency, and the use of money in activities as diverse as buying political positions and gambling. In a discussion similar to that of Ono Masatoshi in this volume, she compared the relative values of medieval and contemporary commodities through their prices in relation to a unit of rice. Archaeology has much to contribute to the future study of medieval mentality and practice. It is also important to acknowledge the religious functions of early towns, in addition to the focus on the market. All of the medieval towns discussed in this book have substantial religious spaces. They were also cultural and intellectual centres. The chapters by Galinié and Gläser discuss cases in which medieval towns grew up on the sites of older Roman centres, as well as the rise of new trading centres in northern Europe. This same phenomenon can be seen in Japan. As outlined in the paper by Ōba, the commercial port of Hakata was located immediately adjacent to the earlier administrative centre of Dazaifu, the locus of Nara and Heian official government trade and diplomatic interaction with China. At Hakata, trade shifted from government monopoly to commerce undertaken by private traders, many of whom were Chinese. On new frontiers, new centres taking advantage of opportunities to the north can be seen in both Japan at the site of Tosaminato and Lübeck in Germany. In the case of Sakai, the importance of private commercial interests in financing and supporting the cultural and political capital of Kyōto can be documented.

The perspective of production is useful in envisioning medieval towns as trading centres. They were, however, religious and political centres as well. In the case of Kamakura, Oka places significance on the designation of Kamakura as the capital of the shogun, and the building of appropriate infrastructure. In the case of Ichijōdani, Ono mentions the establishment and rapid development of a political and commercial centre established by the granting of a region to the Asakura Clan. Presumably the control of trade

125

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 141 1

126 The archaeology of medieval towns: case studies from Japan and Europe

from Kyōto through across Lake Biwa, to the Japan Sea coast and to the north was an important consideration.

How does Senda’s discussion of encastlement relate to the growth of emergence of market economies mentioned by Hodges? Encastlement involves agglomeration and multifunctionality. The castle walls enclose merchants and manufacturers and are the spatial expression of commodification. In the German example of Tilleda, the outer enclosure forms a zone for artisans. Senda notes that castles played a transitional role in northern Europe, before the evolution of true medieval towns in the 12th century, the villages coalesced around castles. The castle plays an active role in transforming the landscape, both in Europe and Japan. The two stages of encastlement, rural and urban, are significant. In the Japanese cases, the presence of the samurai class, administrators, but not cultivators or craftsmen, can be seen in the samurai quarters of castle towns. The case of Ichijōdani is particularly detailed.

In the chapters by Ōba, the importance of foreign traders and their connection to overseas markets is clear. Marginal lands become the centre of access to transportation and commercial wealth. The same theme can be found in the paper by Gläser, with the mention of the important role of foreign traders. In the case of Hakata, Tosaminato, and Sakai, the town is circumscribed and defended by walls, sea, ditch, or rampart, but there is no actual castle.

Oka’s contribution appears different with its emphasis on texts and historical figures, yet it contributes substantially to the general arguments of the other chapters. It deals with the specifics of establishing the administrative site and its infrastructure within a few decades at the beginning of the 13th century, and the architecture of the rulers: moats, roads, and elevated buildings. Medieval towns such as Kamakura involved central planning, and the recruitment of paid specialists. What is the significance of the eastern style of palace architecture? It appears that towns functioned in systems with structural similarities yet cultural distinctiveness. Kamakura functioned within a particular political system of military generals and regents yet it shares common features with centres such as Lübeck. Gläser’s paper stresses the watery environment of Lübeck and its access to water transportation. The Lübeck excavations provide early dates for the actual presence of the defined market area, in the 12th century. A defensive system protected the castle and merchant quarters. Foreign traders played an important role in Lübeck’s trade, in a manner similar to that of the Chinese in Japan. The merchant houses of Lübeck contain storage space, while those of Sakai have adjacent warehouses. The guild hall is of particular note.

The Japanese examples do not mention specific space for craft or merchant organisations, although they are known to have existed in medieval times. As the centre of the Hanseatic League around 1300, Lübeck had specialised markets, money changers, and a mint. How far can one proceed in comparing it with Sakai? At Ichijōdani, Ono’s discussion contributes a great deal to our knowledge of Japanese towns. The temple quarter and the samurai quarter provide some sense of the cultural specifics of spatial organisation. Aspects of town life such as rental of housing and renovation of land owned by temples are particularly informative. Ono’s specifics of the wage economy of towns gives us some idea of the extent of the market economy. The investigations into the production and distribution, and social roles of ceramics of all kinds illuminates cultural differences in social roles of ceramics, as does the paper by Vroom.

By the 10th to 11th centuries there is archaeological evidence of a trading centre in the vicinity of Tosaminato, constructed by the Andō clan. It reached its peak in the14th and 15th centuries, when it contained a blacksmithing area and merchant’s quarters. It had ramparts and ditches for defence. The plan of Tosaminato is said to replicate a prototype which goes back to Kyōto and Kamakura: shrine in the extreme north, elite residences in the northern section, and commercial and factory areas further south. Gläser speaks of culturally specific characteristics of Hanseatic towns as well. How do these townscapes reflect culturally distinctive variations on the theme of towns?

Sakai arose at the junction of two manors. It became the centre for tribute missions to China, and an important regional port. Its warehouses were particularly substantial. As a centre of high value manufacturing, such as casting and printing, it served as an international entrepöt. The importance of the culture of the tea ceremony for Sakai underlines its intellectual and cultural importance. It appears to have been nucleated first, then moated in the 16th century. Can it be accommodated with the framework of encastlement? It also had a different political structure, the merchant’s council. It shows the strong links between commerce and culture in the life of early towns. That cultural creativity, within the common structures of towns, is something that we hope to have shown with these case studies.

Professors Senda and Maekawa and other Japanese medieval archaeologists have visited and described European medieval sites, and have conducted field work at Swavesey, Cambridgeshire. The methods of excavation and analysis seem to be similar in the two regions. Nevertheless Japanese

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 142 1

Afterword

medieval archaeology remains unknown in the west, and few attempts have been made to explore the social and historical processes underlying the rise of towns in the two areas.

Historical archaeology has been considered the archaeology of the expansion of western capitalism; however, it may be considered in a much broader context as the archaeology of societies which produced written historical documents. We hope that this volume will stimulate archaeologists to ponder the meaning and implications of medieval towns and the notion of medieval, and the exploration of medieval life in all of its aspects. Are there differences which define the development of European urban experiences in distinction to those of the Japanese? In the archaeological study of daily life can we extract the cultural essence of each case? Can we draw out their particular ways of ordering space and organising exchange? Hopefully these will be topics for future dialogues in archaeologists from various historical archaeologies.

Bibliography

Hodges, R. 1982. Dark Age economics: the origins of towns and trade AD 600-1000. New York, St. Martin’s Press.

Hodges, R. 1988. The rebirth of towns in the early middle ages. In: Hodges, R. and B. Hobbley (eds.) The rebirth of towns in the West AD 700 to 1050. CBA Research Report No. 68: 1-7. Hongō, K. 2004. Chūseijin no keizai kankaku (The economic mentality of medieval people). Tokyo, NHK Books.

127

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 143

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 144 1

Glossary

Glossary Ashikaga shogunate (see Muromachi period) Asuka period (538 – 710) – marked by the establishment of the capital in Asuka Azuchi-Momoyama period (1568 – 1600) – a period following the collapse of the Ashikaga shogunate, in which political unification was brought about through the actions of Oda Nobunaga and Toyotomi Hideyoshi. Ends with the defeat of Toyotomi by Tokugawa Ieyasu, and the subsequent establishment of the Tokugawa shogunate. bakufu (see shogun) banyaku – a system by which responsibilities are shared, particularly prominent in the Kamakura period when lower vassals would perform roles as public duty to the shogun, predominately in military or guard duties. daimyō – a general term referring to the largest landholding military lords in premodern Japan. With the establishment of the Kamakura shogunate, provincial houses were given titles such as shugo (military governor). Under the subsequent Muromachi shogunate, shugo were given jurisdiction over one or more provinces, eventually becoming provincial magnates, and shugo daimyō. In the Muromachi period, shugo daimyō were supplanted by military governors (sengoku daimyō) during the Ōnin wars. These new daimyō built their domains by accretion though military action so that all land within them was held either directly or in fief by pledged vassals. daimyō (sengoku) (see daimyō) daimyō (shugo) (see daimyō) date (see yakata) Dazaifu – government headquarters in northern Kyūshū from the late Kofun period (ca 300 – 710), through the Heian period (794 – 1185). Echizen – an old province, located in modern-day Fukui prefecture. Edo period (1600 – 1868) – marks the rule of the Tokugawa shogunate and named after their seat of power in Edo

(modern-day Tōkyō). Began with the victory of Ieyasu Tokugawa, the first Tokugawa shogun, in the battle of Sekigahara, and ended with the Meiji Restoration. Heian period (794 – 1185) – ran from the establishment by Emperor Kammu of the capital of Heiankyō (modern-day Kyōto), from which the period is named, to the beginning of the Kamakura shogunate. ho (administrative system) – a system introduced at the end of the Heian period for grouping houses into units responsible for public order and taxation. Japanese units of measurement – the traditional Japanese system of measurement was based around the shaku (approx. 30.30 cm), which was comprised of 10 sun, with 6 shaku making 1 ken. The exact measurements for these units varied with use and location. jōkamachi – a city developed around a central castle, which originally came into being in the Kamakura period with the development of power of the provincial shugo, who began to perform governmental functions from their provincial bases (shugo jōkamachi). These developed further in the Age of Warring States (sengoku jōkamachi). jōkamachi (sengoku) (see jōkamachi) jōkamachi (shugo) (see jōkamachi) jōri system – a system of land division in the 7th and 8th centuries by which the government could allocate land to individuals, and encouraged cultivation of new land by rationalising field boundaries. Kamakura period (1185 – 1333) – marks the rule of the Kamakura shogunate, and named after the seat of the government at that time. The shogunate was established when Minamoto no Yoritomo defeated the Taira family, and created a military government through the appointment of shugo. Succeeded by the Muromachi period. kanrei – shogunal deputy, a high official post in the Muromachi shogunate, often helping mediate between the shogun and powerful shugo.

129

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 145 1

130 The archaeology of medieval towns: case studies from Japan and Europe

Kantō region – an area of central Japan on the island of Honshū, containing the city of Tōkyō. Karoku period – name of the era running from 1225 to 1227. ken (see Japanese units of measurement) Kodai period – translated as the Ancient period, this encompasses the two first historical periods of Japanese history, Nara and Heian. Kofun period (ca AD 250 – 538) – named after the distinctive burial mounds (kofun) which proliferate during this time. It is during the Kofun period that the Yamato province expands to become a kingdom that encompasses the Japanese archipelago. Kyūshū region – the most southern of Japan’s three main islands. machiya – wooden townhouses that typify streets from the late Heian period, and were often owned and used by craftsmen. mandokoro – originally private administrative offices established by great noble houses of the Heian period, they were also mandokoro founded by Minamoto no Yoritomo in the Kamakura period, and by the Muromachi shogunate. Meiji period (1868 – 1912) – covers the reign of Emperor Meiji, and marks the beginning of Japan’s modern period. It began after the collapse of the Tokugawa shogunate and a number of sweeping reforms that transformed Japan from a feudal polity into a modern industrial state. Ended with the death of the emperor in 1912. Ming dynasty – the ruling dynasty in China from 1386 – 1644. Mino ware – a general term for ceramics produced in Mino province (modern-day Gifu prefecture). Mongol wall (see Mongol invasion) Mongol invasion – a series of two invasions, in 1274 and 1281, during the rule of the Kamakura shogunate, caused by the shogunate’s refusal to pay tribute to Kublai Khan, China’s first Mongol emperor. The invasions created a state of emergency in Japan that lasted for 30 years, and led to the construction of defensive walls at points on Japan’s coast.

Muromachi period (1333 – 1568) – marks the governance of the Ashikaga shogunate. It was established when forces led by Ashikaga Takauji destroyed the Kamakura shogunate, and is defined by cultural achievement and social disorder, including the Ōnin War. Succeeded by the Azuchi-Momoyama period. Nara period (710-794) – named after the seat of the government during this time, Heijō-kyō (modern-day Nara). onari – visits by high ranking individuals to other places. In the Muromachi period this came to mean visits by the Ashikaga shogun to the residences of daimyō. Ōnin War – a war fought mainly in Kyōto from 1467–1477 over the succession to the head of the Ashikaga shogunal house, and precipitated by the growing autonomy of the shugo daimyō and the declining power of the Ashikaga shogunate. The war marked the beginning of the Age of Warring states, or sengoku jidai. Ryūkyū – a kingdom based on the Ryūkyū islands until their establishment as Okinawa prefecture in 1879. Seto ware – important glazed ceramic ware which was produced in Japan from the 12th century up to the modern day in Seto, Aichi prefecture. shaku (see Japanese units of measurement) shindenzukuri – one of the two major styles of traditional Japanese domestic architecture, and marked by its central feature, the shinden (main hall), it was found predominately in the mansions of the aristocracy, and residences of high-ranking warriors and Buddhist priests. It was perfected in the Heian period, and used up until the 15th century. shogun – refers to a military office dating to the 9th century, originally meaning ‘commander-in-chief ’. Over time, samurai families used the office of shogun as a way of legitimising themselves and the rule of their country, in the name of the emperor, becoming military dictators. Their regimes were known as bakufu, or ‘tent governments’, often translated as shogunate. The three main periods of shogun rule run from the Kamakura shogunate, through the Ashikaga shogunate, to the final Tokugawa shogunate. shogunate (see shogun)

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 146 1

Glossary

shokuhō political structure – created through the regime of national unification founded by Oda Nobunaga and developed by his successor Toyotomi Hideyoshi in the AzuchiMomoyama period (1568 – 1600), marking the transition from the medieval to the early-modern age. Song dynasty – the ruling dynasty in China from 960 – 1279, split into two periods; Northern Song and Southern Song. sun (see Japanese units of measurement) tatami – traditional rice straw mats that are often used as measurements for room size. Tendai school of Buddhism – Buddhist sect founded in Japan in 806 by Saichō, and dominant in the Heian period. Tōgoku region – a region of Japan from before the modern period. It translates as Eastern Country, and broadly included those provinces around the Tōkai or Tōsan routes, though varied in definition depending on use and era. Tōkai region – a region of central Japan in Honshū that runs along the Pacific Ocean, named after the route in Edo period Japan from Edo (modern-day Tōkyō) to Kyōto. Age of Warring States (sengoku jidai) – a period of Japanese history marked by political instability and conflict between powerful local warrior lords. Began with the Ōnin War and ended with moves toward the unification of Japan at the end of the 16th century. yakata – originally referring to mansions or large residences, this term came to signify a fortified residence from the medieval period, and from the Age of Warring states were improved by sengoku daimyō to the extent that they became castles. yashiki – a term used in the medieval period to refer to a plot of land containing a residence of large size, including a garden and separated from other plots by a wall or other structure. Yayoi period (ca 300 BC – ca AD 250) – the first period of intensive agriculture, and bronze and iron use in Japanese prehistory, with trends towards social stratification and polity formation.

131

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 147

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 148 1

Contributors

List of contributors BRIAN AYERS is Honorary Senior Lecturer at the University of East Anglia. Formerly County Archaeologist for Norfolk, he also served as Chief Executive of the Butrint Foundation. He specialises in the archaeology of towns and his publications include Norwich: Archaeology of a Fine City (Stroud, 2009) and The German Ocean: Medieval Europe around the North Sea (Equinox Books, 2017). He is a Fellow of the Society of Antiquaries of London where he served as Secretary, and is a Member of the Institute for Archaeologists.

HENRI GALINIÉ is the former Director of the Centre National d’Archéologie Urbaine and worked with the Centre National Recherches Scientifique (CNRS), based at the University of Tours. After excavating at Winchester with Martin Biddle, in 1973 he established the Laboratoire d’Archéologie Urbaine de Tours, creating what was in effect the first urban archaeology unit in France. Tours and Galinié are synonymous with the birth of modern urban archaeology in France. In 2007 he edited Tours antique et médiéval. Lieux de vie, temps de la ville: 40 ans d’archéologie urbaine [Roman and Medieval Tours: land uses and urban fabric. 40 years of urban archaeology] (Revue Archéologique du Centre de la France Supplement 30, numéro spécial de la collection Recherches sur Tours). This book presented in a comprehensive fashion the archaeological discoveries in Tours over the preceding four decades.

MANFRED GLÄSER is Director of Bereich Archaölogie Lübeck, responsible for the Department for Preservation of Historical and Archaeological Monuments and also for the direction of excavations in Lübeck. He is the editor of the major urban archaeology publication series Lübecker Kolloquium zur Stadtarchäologie im Hanseraum VII: Die Befestigungen, the eighth volume of which was published in 2012.

SIMON KANER is Executive Director of the Sainsbury Institute for the Study of Japanese Arts and Cultures where he is also Head of the Centre for

Archaeology and Heritage, and Director of the Centre for Japanese Studies at the University of East Anglia, Norwich. His publications include The Power of Dogu: ceramic figures from ancient Japan (London, British Museum Press, 2009) and An Illustrated Companion to Japanese Archaeology (Oxford, Archaeopress, 2016). He is Co-Editor of the Japanese Journal of Archaeology.

ŌBA KŌJI is an archaeologist working for the Fukuoka City Board of Education. As well as being involved in the excavations at Hakata, he has published extensively on the archaeological discoveries from the medieval port city. His publications include Chūsei toshi: Hakata o horu [Medieval city: excavating Hakata], edited by Ōba Kōji, Saeki Nobuyuki, Suganami Masato and Tagami Yūichirō (2008, Kaichōsha).

OKA YŌICHIRŌ is principal archaeologist for the Hiraizumi World Heritage Site in Iwate prefecture, Japan, and Lecturer in Archaeology at Aoyama Gakuin University, Tōkyō. He is also a specialist on the Honedera Mura Shōen archaeological site, an example of a Kamakura period medieval village, at the Ichinoseki Museum in Iwate. His books include Daidō: Kamakura jidai no kansen dōro [Daidō: Kamakura period main roads] (Yoshikawa Kobunkan 2019).

ONO MASATOSHI is formerly Deputy Director of the National Museum of Japanese History and Professor at the National Institute for the Humanities. The leading expert on medieval Japanese archaeology, his books include Zusetsu: Nihon no chūsei iseki [The archaeology of medieval Japan: An illustrated handbook] (Tokyo University Press, 2001) and Higashi ajia chūsei kaidō [The interaction in the medieval East Asian sea] (National Museum of Japanese History, 2005).

RICHARD PEARSON was Professor of Anthropology at the University of British Columbia and is Senior Research Adviser at the Sainsbury Institute for the Study of Japanese Arts and Cultures. He has researched and published widely on the archaeology of East Asia. His books include Ancient Japan (Braziller, New York 1992), Ancient Ryukyu The Archaeology of Island Communities (Hawaii University Press, 2013), Ryukyu: the archaeology of island communities (Oxford, British Archaeological Reports, 2013), and Osaka Archaeology (Oxford, Archaeopress 2016).

JOANITA VROOM is Professor in Archaeology of Medieval and Early Modern Eurasia at the Faculty of Archaeology, Leiden University (NL). Her books include After Antiquity. Ceramics and Society in the Aegean from the 7th to the 20th century A.C. (Leiden, 2003), Byzantine to Modern Pottery in the Aegean. An Introduction and Field Guide (Utrecht, 2005; Turnhout, 2014), Medieval and Post-Medieval Ceramics in the Eastern Mediterranean: Fact and Fiction (MPMAS 1) (Turnhout, 2015) and Medieval MasterChef. Archaeological and Historical Perspectives on Eastern Cuisine and Western Foodways (MPMAS 2) (Turnhout, 2017). She is a specialist in ceramics and the archaeology of the Eastern Mediterranean.

SENDA YOSHIHIRO is Professor of Archaeology at Nara University, Japan, where he previously served as President. Prior to that he was based at the National Museum of Japanese History. Professor Senda has also spent extended periods researching castle archaeology in Germany, based at the German Archaeological Institute in Berlin and at Tübingen University. The additional contributors to Chapter Seven are:

Betty Arndt MA FSA, Stadt Göttingen Stadt archäologie - Rote Straße 34, 37073 Göttingen, Germany.

Jakob Kieffer-Olsen PhD, Senior Researcher; formerly Director of the museum in Ribe, Nationalmuseet, Frederiksholms Kanal 12, 1220 København K, Denmark.

Dominique Pitte, Ingénieur d’études, Direction régionale des Affaires culturelles de Haute-Normandie, Service régional de l’Archéologie, 12 rue Ursin Scheid, 76140 Petit-Quevilly, France.

Hubert de Witte, Deputy Director; formerly Town Archaeologist Brugge, Musea Brugge, Dijver 12, B-8000 Brugge, Belgium.

Katelijne Vertongen, Adjunctconservator Bruggemuseum/ Conservateur adjoint Bruggemuseum, Musea Brugge, Dijver 12, B-8000 Brugge, Belgium.

133

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 155

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 150 1

Index

Index A

Abaga, Ichijōdani 阿波賀, 29, 30-31 Aegean: and eastern ceramics, 116-117; tableware, 117-122 agriculture, 18, 63-64

Ainu, 22

Amino Yoshiyuki, 6

Ancient period, Japan古代, 69 Andō clan 安東氏, 13, 22-23 animal bones, 101

archaeology: comparative approach to, 3, 4-5, 6, 9, 99, 125126; historical texts and, 4, 43, 51, 127; of Japan, 5, 13; urban, 4, 6, 99. See also under Kamakura; Kyōto architectural styles, 49-51

Asakura clan 朝倉氏, 16, 29, 31, 38

Ashikaga shogunate 足利幕府, 32, 34, 35

Asuwa River 足羽川, 29, 31 Athos, Mount, Greece, 116

Azuma Kagami 吾妻鏡, 43-51

B

Bakōhan tea bowl 馬蝗絆, 34 Baltic region, 84-85, 98 Bavaria, 66 Beijing, 4

Bronze Age, 86

Brugge (Bruges), 9, 74-75 Buddha figurines, 31 Buddhism, 17-18

burial goods, 16, 59 burial grounds, 29

bushi 武士. See samurai

C

castles: castle towns, 14-15, 17, 29-32, 61-69; Lübeck castle, 86-89, 93; study of, 61-63; urban, 9. See also encastlement; yakata centralisation of settlements, 61-69. See also nucleation

ceramics: Aegean, 116-122; appreciation of, 33-34, 116; blue and white ware, 21, 34-35; Byzantine, 117-122; celadon ware, 16, 21, 39, 116; chafing dish, 117-118; decoration and imagery, 100, 118-119; Delftware, 115; development of, 116-122; Echizen ware 越前焼, 39; exchange between east and west, 119; glazing of, 119, 120; Green-and-Brown Painted ware, 120, 119; hard greyware, 100; Imari ware 伊万里焼, 20, 116; imitation of metal ware, 121; Islamic, 116, 118; Japanese

porcelain production, 115; Kakiemon-style 柿右衛門様式, 115; kawarake かわらけ, 35-36, 39; kraak-style porcelain, 115; Mino ware 美濃焼, 16, 39; production and distribution, 38-40; redware, 100; repaired, 34-35; Seto ware 瀬戸焼, 16, 39; Sgraffito ware, 118, 119, 120; shape and function, 119-122; slip coating, 114, 116, 117; and status, 33-36; stoneware, 98; tableware, 117-121; technological exchange, 115, 116; technological innovation, 116, 120; temmoku 天目, 20, 116; trade of, 100; yue ware, 16. See also under China, Japan, Korea, Lübeck, Sakai ceremonial spaces, 32-33 Chang’an, 3-4

Chiba prefecture 千葉県, 65

Chiba Tsunetane 千葉常胤, 46 China: art, 15, 33; ceramics, 15, 16, 20, 36, 38-39, 115, 116; coinage, 15-16; dynastic capitals of, 4; Japan’s relationship with, 7, 125; Jiajing period, 16; Ming dynasty, 15, 16, 19, 59, 115; Song dynasty, 15, 18, 56, 57; Tang dynasty, 3-4, 116, 118; trade with Japan, 15, 19, 29, 56-57, 59; urbanisation in, 9. See also under traders chinzei tandai governing agency 鎮西探題, 57, 59

Chiran castle 知覧城, 65 Christianity, 18, 84, 107, 110. See also towns: as religious centres clothing, 100

coinage, 15-16, 20, 21 Cologne, 5

communes, 68-69

Constantinople, 117

Córdoba, 9

costume accessories, 100-101

craft production, 36-38

D

daimyō 大名: authority within towns, 17, 29-30; ceramic use, 35-36; sengoku daimyō 戦国大名, 14, 17, 29; shogun, relation to, 33; shugo daimyō 守護大名, 17, 19, 64, 65, 66; supply of goods to, 37 Dazaifu 太宰府, 15, 22, 56, 57

Dejima (Deshima) Island, Nagasaki 出島, 3, 115. See also traders: Dutch Denmark, 80

Detmar (chronicler), 86, 93 Dohna castle, Saxony, 68 Dorestad, 108, 111

Dutch East India Company, 3, 115

135

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 151

136 The archaeology of medieval towns: case studies from Japan and Europe

E

East Asia, 4

Heijō 平城, 3-4 Hellweg trade route, 76

Echizen 越前, 29. See also under ceramics economic institutions, 17

Helmold (chronicler), 85, 86, 90, 91, 93

Edo period 江戸時代, 34

Hiraizumi 平泉, 45, 50, 51 historic urban environments, 4-5, 10

Edo (city) 江戸, 3

egōshū governing council 会合衆, 19 emporia, 108, 111-112 encastlement, 63-68 Endsee castle, 67

England, 4-5

Europe, 3, 5-9, 125-126 F

Fehring, Günter, 67, 68 Fez, Morocco, 120

fortified residences, 65

France, 9: medieval urban situation, 107; roles of urban settlements, 107-108; typology of urban settlements, 108-110; urban armature, 110-112 Frederick Augustus the Strong, King, 115

Fujiwara clan 藤原氏, 14

Fukushima-jō castle 福島城, 22, 23 Fukuyama castle 福山城, 9 G

Gaul. See France

German settlement of Eastern Europe, 84

Germany, 66-68, 84

gōshō merchant princes 豪商, 19, 21 Göttingen, 9, 76-77

Great Yarmouth, 8-9 guilds, 8, 17, 68-69 H

Hakata 博多, 3, 5, 9, 15, 56-60 Hakata Diary 博多日記, 59

Hakatahama 博多浜, 56-57, 58-60 Hanseatic League, 72, 85, 97-98, 103. See also under towns harbours, 91-93, 98

hare and ke ハレとケ, 32-33

Hatakeyama Shigetada 畠山重忠, 47 Hatta Tomoie 八田知家, 47 Heian period 平安時代, 17 Heian political system, 125 Heiankyō 平安京, 3-4, 69

Henry the Lion, Duke, 84, 89, 91 Herrman, 7

Hizen region 肥前国, 115 Hodges, Richard, 125, 126

Hōjō Masako 北条政子, 43

Hokkaidō 北海道, 22, 23, 24 Holy Ghost Hospital, Lübeck, 94, 98 Hongō, Keikō, 125

Honshū 本州, 14 Hosios Loukas, Monastery of, 119 Hosokawa family 細川氏, 18-19 household objects, 99-100

houses: burgess, 98; Dielenhäuser, 90; layout, 36-38, 98; merchant, 93; regulation of building of, 37-38, 98-99; rented, 37-38; samurai 侍, 29, 30, 36, 37, 38, 45; Steinwerke, 90; structure and material, 89-90, 94, 95-96, 98; timber-built Kemenaten, 89-90. See also machiya I

Ichijōdani 一乗谷, 4, 8, 14, 16: ceramics, 33-36, 38-40; craft production, 36-37, 38; establishment of, 31-32; excavations, 29; machiya 町屋, 36; topography, 29; urban layout, 30-31

Ima Kōji, Kamakura 今小路, 45

Inabayama castle town 稲葉山城下, 65 incastallamento. See encastlement

Insingen castle, 67

Iron Age, 86

J

Japan: ceramics, 16-17; coinage, 15-16; early Japanese urban forms, 3-4, 14; Europe, encounters with, 3; Inland Sea 瀬戸内海, 14, 18; land tenure, 17; medieval history, 13-14; religion, 17-18; social classes, 17; trade with China, 15, 19, 29, 56-57, 59; trade with Europe, 3; trade with Korea, 22, 59; urban archaeology, 4; urban development, 14-15, 43-44 jewellery, 100-101

jōkamachi 城下町, 14, 66, 68, 69 jōkan 城館, 7, 14

Jōkyōji, Ichijōdani 浄教寺, 31 jōri system 条里制, 63, 64

jūban bōeki system 住蕃貿易, 56, 57. See also traders: Chinese

Jūsan, Lake 十三湖, 22, 23

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 152 1

Index

K

leisure activities, 101-103

Kamakura (city) 鎌倉市, 5, 9, 40: administrative policies, 43; archaeological features, 46-48; archaeology, 42-43, 51; architectural style, 49-51; banyaku 番役, 43; early urban development, 43-46; Kyōto, model for, 43, 46, 49-51; local government office, 45; religious importance, 45; strategic position, 46; urban layout, 45-46

Lübeck, 69, 84-85: Artisan’s Quarter, 93-94; castle, 86-89; ceramics, 99-100; comparison with Sakai, 103-104; early settlement, 85-86; market, 90-91; topography, 85; urban culture, 97-103; urban defences, 93; urban settlement, 89-90, 94-97; waterfront, 91-93

kaisho 会所, 33

Kamakura period 鎌倉時代: development of Kamakura, 4252; economic institutions, 17; value of objects from, 33-34, 36 Kamakura shogunate 鎌倉幕府, 43, 51, 57

Kamiya family 神屋一族, 59 kanrei 管領, 18, 32

Kansai region 関西, 65

Karakawa site (Lake Jūsan) 唐川遺跡, 22

Kawachi plain 河内平野, 18, 19 Kawate castle 革手城, 65 Kazusa clan 上総氏, 45 kemari 蹴鞠, 51

Kikuchi clan 菊池氏, 59

Kinai 畿内, 14-15, 18, 22 Kiyosu 清須, 63-64

Kobata (castle town) 小幡城下, 64

Kobayakawa clan 小早川氏, 56 Koburg, Lübeck, 85, 87, 93, 94 kofun 古墳, 22

Kofun period 古墳時代, 69 Korea: ceramics, 16, 39; coinage, 16; trade with Japan, 22, 59; urban centres, 4 Kōrokan 鴻臚館, 15, 16, 56-57, 59 kugaisho 公界所, 31, 40

kuge civil aristocracy 公家, 17 Kumano 熊野, 18

Kuroda clan 黒田氏, 56

Kusado Sengen-chō 草戸千軒町, 4, 5, 13, 14, 15

London, 4, 5

M

machi 町, 30, 31

machiba 町場, 30-31

machiya 町屋, 29, 36-38, 39 Maekawa Kaname, 7, 13, 14

mandokoro 政所, 48 maritime technologies, 72-73 markets, 90-91, 98

material culture, 7, 99-103.

medieval mentalities, 125

Mediterranean, 9. See also Aegean merchants. See traders

Merovingian kingdom, 110 metal vessels, 121

Minami Mikado, Kamakura 南御門, 47

Minamoto clan 源氏, 45, 46, 51: Minamoto no Naka’akira 源 仲章, 50; Minamoto no Nakakane 源仲兼, 50; Minamoto no Sanetomo 源実朝, 50, 51; Minamoto no Yoritomo 源頼朝, 43-51; Minamoto no Yoriyoshi 源良守, 45; Minamoto no Yoshitomo 源義朝, 45, 46, 51 minato machi 港町, 66. See also towns: port towns Ming tribute missions, 19, 59 Mino 美濃, 65

misedana 見世棚, 29, 37

Miura clan 三浦氏, 45, 46 moats, 19, 29-30, 47-48, 51, 64, 65, 69

monasteries, 85, 98, 108, 110, 111

Mongol invasion of Japan, 58, 59 monzen machi 門前町, 14, 66.

Kyōto 京都: archaeology, 4; central power of, 24; temples in, 31; Kamakura, model for, 43, 46, 49-51; mansions, 32; medieval culture, 34; medieval population, 5, 15; planning of, 9; religious change, 17-18; social classes, 17

Moriyama castle 守山城, 64 mortuary monuments, 31

L

Murata Shukō 村田珠光, 20-21

Kyūshū 九州, 15, 16, 18, 30, 56, 57, 60 land reclamation, 57-59, 94-95, 96

land tenure, 17 language, 97

Motogō site, Ōmori 元郷, 64

Mottaimatsu site (Kagoshima prefecture) 持躰松遺跡, 16 Muromachi bakufu 室町幕府. See Ashikaga shogunate

Muromachi period 室町時代, 14-15, 17-18, 57

137

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 153

138 The archaeology of medieval towns: case studies from Japan and Europe

N

residential fort, 7, 14

Naganuma Munemasa 長沼宗政, 51

Rithauser castle, 67

Nabu clan, 22

Nagaokakyō 長岡京, 4

Namioka castle 浪岡城, 65 Nara (city) 奈良, 3

Nara period 奈良時代, 56, 59

Nejō castle 根城, 30, 40 Neolithic period, 85-86 Nordenberg castle, 67

Norman dukes, 82

Norwich, 5, 9, 78-79

nucleation, 7, 14, 20. See also centralisation of settlements

nutrition, 101

O

Ōba Kageyoshi 大庭景義, 48-49

Oda Nobunaga 織田信長, 19, 29

Ōei Disturbance 応永の乱, 19

Okinawa 沖縄, 14, 16, 19. See also Ryūkyū Kingdom

Okinohama 息濱, 56, 57-60

Ōkuboyama site, Saitama 大久保山遺跡, 50, 51

Ōkura Palace, Kamakura 大倉, 46-47, 49-50 Ōmori Ozekigashiro 大森尾関ヶ城, 64

onari 御成, 33

Ōnin Wars 応仁の乱, 18, 29, 65

Ōsaka 大阪, 3, 65

Ōsaka Bay 大阪湾, 18, 19, 20 oshiki 折敷, 35

Ōtomi clan 大友氏, 56, 60 Ottoman empire, 116, 117

Ōuchi clan 大内氏, 19, 56, 65

Owari plain 尾張平野, 63, 64, 65

Ribe, 9, 80-81

ritsuryō system 律令制, 17, 22 roads, 30 Roca San Silvestro, 63

Roman Empire, 7, 125: towns, 82, 107-112

Rothenburg ob der Tauber, 66-67, 68 Rouen, 82-83, 110, 111 royal authority, 111

royal residence, 110, 111 ruralisation, 107

Ryūkyū Kingdom 琉球, 16, 34. See also Okinawa S

Saga plain 佐賀平野, 65

Sakai 堺, 8, 24, 69, 103-104: ceramics, 20-21; excavations of, 1920; historical background, 18-19; land tenure, 22; merchants, 19, 21; military role of, 19; social classes, 21; trade, 19, 20-21; urbanisation, 20

samurai 侍: dwellings, 29, 30, 36, 37, 38, 45; power, 17; rituals, 50; social status, 38; tansaku land divisions 短冊, 14; values, 34, 50 Sannai kuketsu 三内口決, 39 Saris, John, 3

Schauenburg, Earl of, 85, 86, 89 Shiba clan 斯波氏, 64

Shiga prefecture 滋賀県, 65

Shimazu clan 島津氏, 65

shindenzukuri architectural style 寝殿造, 49-51 shōen 荘園, 16, 17, 63

Shōgenji Temple, Kamakura 生源寺, 45

shogun 将軍, 17, 33-34. See also onari Shogun’s Palace, Kamakura, 47-51 shoin writing hall 書院, 20

shokuhō political system 織豊政権, 69

P

Shōni clan 少弐氏, 56, 60

prestige goods, 111

shugo 守護, 17 Silk Road, 18

piracy, 72-73 Q

Quentowic, 108, 111

R

Rakuchū rakugai folding screens 洛中洛外図, 4, 37 religion, 8, 9, 17-18, 45. See also under towns rented housing, 36-38

shuden 主殿, 33

Sin’an ship, 8, 15, 33-34, 39

Slavic kingdoms, 84. See also under towns

sonshuka, 7, 14. See also centralisation of settlements Spain, 120

Spandau Castle, Berlin, 68

Sugimotodera Temple 杉本寺, Kamakura, 45

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 154 1

Index

T

urban layout, 29-31, 36-38; 97-98. See also towns: planning of

tansaku (tanzaku) land division 短冊, 8, 14, 69 tea ceremony, 20-21

urban networks, 111-112

Taira family 平氏, 49, 51

urban mentalities, 10

urban population increase, 5, 31

Tendai School 天台宗, 45. See also Buddhism Tewkesbury, 9

urban settlements. See towns

Tilleda, 67, 68, 69

urban/rural relationship, 6, 8

urbanisation, 3, 5, 6, 9, 20, 56-60, 67: periods of, 89-90

Thirty Years War, 76

urbanism, 6

Tōgoku region 東国, 14, 49-50, 52

V

Tōhoku region 東北, 22

toimaru shipping agents 問丸, 17

Tokugawa government 徳川幕府, 15, 22 tools: carpentry, 38; hygiene artefacts, 101; knives, 101 Topkapi Palace, Istanbul, 116

Tosaminato 十三湊, 8-9, 22, 24: excavations of, 23; regional nucleation, 23; trade, 23-24 Toubert, Pierre, 63

towns: as administrative centres, 108, 109; commercial activities within, 14, 19, 64, 125-126; control of economic activity by, 109, 111-112; cultural characteristics, 126; cultural influence, 103; definition, 7-8, 63, 110; everyday service to population, 109; function, 14, 63, 64, 67, 107-110; Hanseatic, 97-98, 103 (see also Lübeck); as industrial centres, 80; Japanese early, 3-4, 14; laws of, 98-99; planning of, 4, 9, 30, 48, 51, 89; port towns, 14, 18; as religious centres, 8, 14, 64, 65, 68, 110, 111, 125; Roman 82, 107-112; scale of, 5; Slavic, 86, 89, 97; typology, 14, 108-110 toys, 101-103

trade: control of, 57, 87; networks, 6, 72; North Sea, 78; route, 24, 29, 74, 87. See also under China; Korea; Japan; Sakai; Tosaminato

traders, 93, 97, 126: Chinese, 15, 29, 56-57, 59; Dutch, 115; German, 97; Gotland, 91; Russian, 91; Sakai, 19, 21; Scandinavian, 91

Trave (river), 85, 87, 91, 94, 95, 103

Tsurugaoka Hachimangū shrine, Kamakura 鶴岡八幡宮, 9, 46, 48 Turkey, 116 U

Uesugi clan上杉氏, 33 urban armature: of Medieval Gaul, 107, 110-111, 112; Roman, 109, 110

urban centres, 4, 5

urban culture, 8, 51, 97-103

urban defences, 9, 93

urban economies, 7-8. See also towns: commercial activities within urban expansion, 94-97

Viking age, 80, 82, 110

W

Wada family 和田氏, 45

Wakamiya Ōji, Kamakura 若宮大路, 45, 47 Wakenitz (river), 85, 87, 91, 93, 94, 103 Waldemar the Great, Danish king, 84, 93, 95

Wales, 9

Winchester, 5, 9 X Y

yakata 館, 29-30, 32-33, 35, 36 Yamato state 大和, 20, 22

Yanagigosho 柳之御所, 14 yashiki 屋敷, 14, 23, 31

Yayoi period 弥生時代, 56, 69

Z

za co-operative guilds 座, 8, 17

139

Medieval Towns text 080620.qxp 08/06/2020 13:18 Page 155