University of Southern California Dissertations and Theses

Late medieval Japan's Seto Inland seascape: shipping, sailors, and seafaring

(USC Thesis Other)

Late medieval Japan's Seto Inland seascape: shipping, sailors, and seafaring

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

LATE MEDIEVAL JAPAN’S SETO INLAND SEASCAPE:
SHIPPING, SAILORS, AND SEAFARING

by
Michelle M. Damian

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(HISTORY)

May 2015

Copyright 2015 Michelle M. Damian

i

Table of Contents

Acknowledgments ........................................................................................................................................ iv
Dedication .................................................................................................................................................. viii
List of Figures .............................................................................................................................................. ix
List of Tables .............................................................................................................................................. xii
Chapter 1: Japan is an Archipelago............................................................................................................... 2
The Early Inland Sea Seascape ................................................................................................................. 4
Constructs: Ships and Maritime Infrastructure .................................................................................... 5
Sociologies: Maritime Communities and Sailors ................................................................................ 11
Governance: Power Players and Hierarchies .................................................................................... 17
Transgressions: Piracy and Rebellion ................................................................................................ 22
The “Seascape Gap” in the Historiography of Medieval Japan .............................................................. 24
Medieval Constructs: Ships and Maritime Infrastructure .................................................................. 25
Medieval Sociologies: Court, Warriors, Peasants… Sailors? ............................................................ 28
Medieval Governance: Control and the Economy .............................................................................. 32
Medieval Transgressions: Pirates and Sea Lords .............................................................................. 36
From Seascape to Maritime Cultural Landscape .................................................................................... 39
Methodologies: An Interdisciplinary Approach...................................................................................... 42
GIS Methodology ................................................................................................................................ 45
Conclusion .............................................................................................................................................. 47
Chapter 2: The Ships and Ports of the Inland Sea: Shipping, Production, and Piracy ................................ 50
Ship Technologies: Medieval Innovations .............................................................................................. 51
What Makes a Port? ................................................................................................................................ 63
Governance of Port Towns and Checkpoints .......................................................................................... 72
Checkpoints: Controlling Sea Traffic ................................................................................................. 74
Ports: Shipping and Strategic Administration .................................................................................... 78
Port Governance: Final Thoughts ...................................................................................................... 85
Culture: People of the Ports .................................................................................................................... 86
Saltmaking and Ports .......................................................................................................................... 86
Fishing and Ports ................................................................................................................................ 92
Women and Port Towns ...................................................................................................................... 98

ii

Sociologies: Conclusion .................................................................................................................... 100
Outliers: “Pirates” in the Inland Sea ..................................................................................................... 100
Conclusion ............................................................................................................................................ 108
Chapter 3: The Sea Connects: Tracing the Flow of Trade ........................................................................ 110
Dynamism in the Late Medieval Economy ........................................................................................... 111
The Register of Incoming Ships at the Hyōgo North Gate: An Introduction ........................................ 113
Salt Shipping ......................................................................................................................................... 121
Shipping Routes ................................................................................................................................ 121
Pricing............................................................................................................................................... 126
Salt: Final Thoughts ......................................................................................................................... 129
Rice Shipping ........................................................................................................................................ 130
Rice Production and Packaging ....................................................................................................... 130
Shipping Routes ................................................................................................................................ 133
Pricing............................................................................................................................................... 137
Red (Champa) Rice ........................................................................................................................... 140
Rice: Final Thoughts ......................................................................................................................... 141
Beans, Wheat, and Barley ..................................................................................................................... 142
Wood Products ...................................................................................................................................... 149
Marine Goods........................................................................................................................................ 155
Egoma (Perilla), Indigo, and Miscellaneous Specialty Goods .............................................................. 165
The Register and Archaeological Evidence .......................................................................................... 174
Jige: A Special Case .............................................................................................................................. 184
Conclusion ............................................................................................................................................ 185
Chapter 4: Rules of the Sea: Seafaring Practices and Collaboration ........................................................ 189
Laws of the Sea: The Maritime Codes (Kaisen shikimoku) .................................................................. 190
Contracting for a Voyage .................................................................................................................. 194
Provisioning and Manning the Ship .................................................................................................. 196
The Ship at Port ................................................................................................................................ 203
Cargo Concerns ................................................................................................................................ 208
Other Perils at Sea ............................................................................................................................ 217
Navigation Problems and Consequences .......................................................................................... 218
Encountering Other Ships at Sea ...................................................................................................... 220
Collaborative Voyages and Captains’ Networks: The Hyōgo Register Meets the Maritime Codes .... 222

iii

Captain Gyōbu Shirō: A Typical Range of Networks ....................................................................... 224
Hyōe Tarō: Ship’s Captain and Ship Owner? .................................................................................. 228
The Western Inland Sea Region: A Port-Centered Model ................................................................ 233
Collaborative Voyages: Final Thoughts ........................................................................................... 237
Conclusion ............................................................................................................................................ 238
Chapter 5: Conclusion: Filling in the “Seascape Gap” ............................................................................. 241
Constructs: Environment and Maritime Infrastructure ......................................................................... 241
Control: Governing the Seascape .......................................................................................................... 244
Sociologies: The Maritime Community ................................................................................................ 246
Transgressions: Piracy in the Inland Sea .............................................................................................. 248
The Medieval Maritime Cultural Landscape ........................................................................................ 249
Future Directions and Final Thoughts .................................................................................................. 251
Appendix 1: List of Early Fifteenth Century Military Governors ............................................................ 253
Appendix 2: Determining Disputed Port Sites .......................................................................................... 255
Appendix 3: Translation of Version B (KI 3071) of the Maritime Regulations ....................................... 260
Works Cited .............................................................................................................................................. 274

iv

Acknowledgments

Graduate students do not produce a dissertation alone. I owe a great debt to many individuals and
organizations for helping make this project possible.
First and foremost, I must thank the members of my committee. My advisor, Prof. Joan Piggott,
allowed me the freedom to combine disciplines as I brought my archaeological training and perspectives
to a “traditional” history department, while pushing me to look further into the documents and terrestrial
background to create a more robust study. Dr. David Bialock has helped bring a new perspective as he has
pointed out ways in which literature can further enhance our understanding of the seascape. Dr. John E.
Wills, Jr., has inspired me to widen this study even further in the future with his questions about global
connections and trends beyond the Inland Sea. I am greatly indebted to them for their support and
suggestions.
In Japan, a number of institutions opened their doors, their archives, and their boxes of artifacts to
me. Dr. Yoshida Sanae graciously arranged for me to be able to study at the Tokyo Historiographical
Institute, and Dr. Takahashi Toshiko often let me query her regarding various estate-related matters. Dr.
Sakurai Eiji provided insight into medieval economic concerns. I was very lucky to be able to attend one
of Dr. Murai Shōsuke’s last seminars before his retirement from Tokyo University, and the practice with
reading sources related to the western medieval maritime seascape as seen in the Aokata monjo provided a
fascinating counterpoint to my own research. His input regarding my interpretation of the maritime codes
was invaluable, and I was most grateful for his help in letting me see the actual Hyōgo Register.
Yokoyama Ayano, a fellow seminar student, very kindly helped me tremendously with reading and
interpreting a number of primary sources.
At Meiji University, Dr. Yoshimura Takehiko and Dr. Sasaki Ken’ichi provided me with access
to that university’s libraries and their wonderful repository of archaeological site reports from throughout
Japan. Any analysis of the artifacts found throughout the Inland Sea would have been impossible without
their assistance. Once I ventured outside of Tokyo, numerous archives and archaeological repositories let

v

me into their vaults to look at their collections. I am extremely grateful to Saiki Iwao at the Kobe City
archaeological center, Miyazaki Tetsuji at the Kagawa prefecture archaeological center, Tai Yoshiaki at
the Seto Inland Sea Folk History Museum, Tominaga Taka and Takada Kinji at the Hokudan-chō Folk
History Museum, my volunteer tour guide Tanaka-san on Naoshima, Ōta Ayumu and Saitō Tsutomi at the
National Museum of Japanese History, Suzuki Rikio at the Okayama Prefectural Museum, Masada Taka
for sharing his stories about diving on the Mizunoko-iwa shipwreck site, Yano Hitoshi and Tanaka Ken of
the Murakami Suigun Museum (particular thanks to Tanaka-san for arranging the rapids tour and visit to
the Noshima archaeological site), and Iwasaki Kumi and Hirahara Junji for turning an unplanned stop at
Fukuoka-Bizen to one of the best site visits of my entire stay in Japan.
As I am still a beginning user of GIS software, I must acknowledge Dr. Jenny Swift for her
instruction in the subject. Tatsumi Chiaki in Japan very generously provided both raw data and
suggestions. Dr. John Knoerl came to my aid when I was tearing my hair out trying to figure out why the
maps wouldn’t map!
Funding from the University of Southern California Provost Fellowship and the Fulbright
Doctoral Fellowship allowed me to conduct research in the USA and Japan that was integral to
completing this dissertation. Additional funds from Duke University and the University of Chicago let me
access library materials during the writing process in the USA.
I was fortunate enough to be able to present aspects of this research at various gatherings,
including the fifth world conference for the Society for East Asian Archaeology and USC’s “Reassessing
the Shōen Conference.” I owe particular thanks to Randy Sasaki and the members of the Asian Research
Institute of Underwater Archaeology for inviting me to present at the Fourth Symposium on Underwater
Archaeology at Obirin Daigaku in Tokyo, as well as to Dr. Mark Staniforth and Dr. Maria-Cruz Berrocal
for including me in their panels at the Society for Historical Archaeology and the Second Asia-Pacific
Conference on Underwater Cultural Heritage. I am grateful for the additional perspectives and feedback
garnered at each venue, and I apologize for the cases when I have not been able to fully incorporate all of
the suggestions into the present work.

vi

I took a circuitous academic route that has culminated in this dissertation, and would like to
acknowledge the inspiration of the professors that shaped my approach. Dr. H. Mack Horton and Dr.
Mary Elizabeth Berry at UC Berkeley were the first to encourage my interest in premodern Japan.
Though I didn’t recognize it at the time, by allowing me to incorporate sculpture, literature, and history
into my undergraduate thesis they set me on an interdisciplinary path from the outset. Dr. Brad Rodgers,
Dr. Nathan Richards, and Dr. David Stewart provided me with the tools of maritime archaeology and
armed me with the idea of the maritime cultural landscape. Dr. Hans van Tilburg reassured me that yes,
one can study Asian maritime topics and still eventually be gainfully employed.
On a personal level, the following people were sounding boards academically, personally, and
emotionally – sometimes all at once! I’m grateful to Dr. Lori Meeks, Dr. Nobuko Toyosawa, Dr. Phil
Ethington, Dr. Charlotte von Verschuer, Lori Rogers, Sandra Hopwood, Kristina Buhrman, Rieko Kamei-
Dyche, Dan Sherer, Kevin Wilson, Nadia Kanagawa, Jillian Barndt, Jesse Drian, Cassandra Deirolf,
Viktor Shmagin and Makiko Ogawa, John Mehringer, Yōko Takau-Draubin, Michael McCarty, Park Su-
Cheol, Daniel Schley, Sherry Funches, Lisa Hew, Pat Hughes and Kazumi Nakagawa, Miwa Takano-Lee,
Niinae Mio, Uehara Sōya, the Kunimatsu family, and Yuki Kent for their kindnesses, input, and
encouragement throughout my PhD studies. Dr. Jan Goodwin generously took time to read and critique a
draft of this dissertation. Dr. Aratake Ken’ichi provided suggestions and encouragement at a critical
moment. Dr. Gaynor Sekimori opened her house to me in Japan, allowing me not only to live in a much
more luxurious space than any graduate student in Tokyo has a right to expect, but also to broaden my
perspective to learn to appreciate cats. The folks at the Aquarium of the Pacific in Long Beach, CA, and
the members of Taiko Charleston, SC, reminded me that occasionally it’s OK to step away from the
computer.
I must particularly acknowledge certain individuals who have gone above and beyond anything I
ever had a right to hope for in their encouragement over the years. Dr. Brett Sheehan was a constant
source of support and encouragement, and I am tremendously appreciative of his patience and generous
mentorship. I am exceptionally in debt to Kagawa Tomomi and her family for their longtime friendship,

vii

as well as for acting as my personal chauffeur service and hotel during my explorations around Onomichi
and its environs. I am grateful to Dr. T. Kurt Knoerl for setting me on the road (or on the water) toward
the realm of maritime archaeology, for his valuable insight into the shaping and editing of this dissertation,
and above all for his friendship. Erin Brightwell and Sachiko Kawai were constant sources of both
academic and personal support from even before I began formally studying at USC and have helped me
greatly throughout the whole process. Kristen Bradanese has been my touchstone for sanity when I have
needed to step back and get perspective on this project, and I am lucky to call her friend.
Finally, I am most grateful to my family for their enduring support and encouragement
throughout my graduate career: my brother Kevin and his wife Corrie, my father’s presence in spirit, and
most of all to my mother Sandra and stepfather Jerry.

viii

Dedication

To my mom, for her encouragement of all my geographical and academic exploring. I couldn’t have done
this without you.

ix

List of Figures

Figure 1. Map of medieval Japanese provinces. Courtesy Joan R. Piggott. ................................................. 1
Figure 2. Rock from the Owada-no-tomari foundation, dating to the late Nara to mid-Heian periods and
recovered in 2003. Owada-no-tomari today is covered with landfill and has been absorbed into
the city of Kobe. Photo by author (2012). ................................................................................... 9
Figure 3. Rocks recovered from near the Kyōgashima site and said to have been buried with prayers for
successful completion of the manmade island. Pictured here on display at the temporary
exhibit, “KOBE de Kiyomori” history pavilion, on loan from Hoteiji Temple ( 布袋寺) in Kita-
ku, Kobe city. Photo by author (2012). ....................................................................................... 9
Figure 4. View of Wagaeshima at low tide from shore. Note that fishing boats still moor nearby today.
Photo by author (2011). ............................................................................................................ 10
Figure 5. Wagaeshima. The piles of rocks still hold countless pottery sherds, as well as modern anchors,
historical anchor stones, and other remnants of maritime activity. Photo by author (2012). .... 11
Figure 6. Hōnen shōnin gyōjō eden. Goodwin (2007, cover image). ......................................................... 53
Figure 7. Kōbō Daishi gyōjō e zanketsu. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 149). . 53
Figure 8. Ippen Shōnin eden mohon. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 150). ....... 54
Figure 9. Map of places named in Chapter 2. ............................................................................................. 56
Figure 10. An in situ photo of the dugout excavation site. The boat is upended, forming one side of a well.
Kobe-shi Kyōikuiinkai (2001, photograph section p 32, plate 4). ............................................ 57
Figure 11. Closeup of scene from Shido-ji engi, late Kamakura or early Muromachi era. Note how the
boats pull up directly to the shore, with no evidence of dock infrastructure. Kokuritsu gekijo
Nōgaku-dō chōsa yōseika(1991, plate 1). ................................................................................. 58
Figure 12. Shiogama scenes from Ippen Shōnin voyages. Jōshōji scroll. Miya (1979, Scroll 2 Section 3).
.................................................................................................................................................. 60
Figure 13. Shinnyōdō engi. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 154). ...................... 61
Figure 14. Tamanoiwa boulder today at Senkōji temple, viewed from the ropeway cable car. Photo by
author (2011). ............................................................................................................................ 65
Figure 15. View towards Tamanoiwa from the original medieval shoreline. The boulder is encircled in red
for easier visibility. Photo by author (2011). ............................................................................ 66
Figure 16. Location of Abuto Kannon in relation to the port sites of Onomichi, Tajima, and Tomo. ....... 67
Figure 17. Abuto Kannon today, part of Bandai-ji Temple in Hiroshima prefecture. Photo by author
(2011). ....................................................................................................................................... 68
Figure 18. From left to right: A growing egoma plant, the seeds from the egoma plant ready to be ground
into oil, and a statue on the precincts of Ōyamazaki Hachiman today honoring the egoma and
lamp-oil sellers. Photos by author (2012). ................................................................................ 78
Figure 19. View from a pier at Shido Port looking towards shore. Note Shido-ji temple’s pagoda. Photo
by author (2012). ....................................................................................................................... 82
Figure 20. Photo of reconstructed irihama-style salt flat in Utazu, Kagawa prefecture. Note the irrigation
ditch that allows seawater to flow directly from the ocean to the salt flats. Photo by author
(2012). ....................................................................................................................................... 88
Figure 21. Map of large volumes of salt (more than five hundred koku) being shipped to Hyōgo in 1445.
.................................................................................................................................................. 91
Figure 22. Map of Yuge Island, fourteenth century. Shapinsky (2014, p. 78) ............................................ 93

x

Figure 23. Various thirteenth - fifteenth century fishing weights from the Hamanochō excavation site,
Kagawa prefecture. Similar types of weights were found at Ōurahama. (Photo by author; taken
with permission from the Kagawaken Maizō Bunka Senta-. 2012). ........................................ 95
Figure 24. Discoveries of fishing weights in Ōurahama Cove. From left to right, the maps indicate rod-
shaped weights; large, medium, and small grooved weights; and single-groove tube weights.
Kagawa-ken Kyōiku Iinkai and Honshu Shikoku Renraku Hashi Kōdan (1988, 159-160). .... 95
Figure 25. Sardine shipping in the Register. ............................................................................................... 97
Figure 26. The swirling currents facing the coast of Noshima island. Photo by author (2013). ............... 102
Figure 27. Two postholes (front right and far left) on the "funadamari" terrace at Noshima island. Photo
by author (2013). ..................................................................................................................... 103
Figure 28. Identifiable ports noted in the Register. ................................................................................... 120
Figure 29. Annual salt shipments to Hyōgo in the Register. .................................................................... 122
Figure 30. Ports in the Register shipping more than five hundred koku of salt to Hyōgo. ....................... 123
Figure 31. Annual rice shipments to Hyōgo in the Register. .................................................................... 132
Figure 32. Ports in the Register shipping more than five hundred koku of rice to Hyōgo. ....................... 134
Figure 33. Annual Champa Rice shipments to Hyōgo in the Register. .................................................... 139
Figure 34. Annual bean shipments to Hyōgo in the Register. .................................................................. 144
Figure 35. Annual amounts and types of bean shipments to Hyōgo in the Register. ............................... 145
Figure 36. Annual wood shipments (types) to Hyōgo in the Register. ..................................................... 150
Figure 37. Annual wood shipments (types and amounts) to Hyōgo in the Register. ................................ 152
Figure 38. Annual sardine shipments to Hyōgo in the Register. .............................................................. 156
Figure 39. Annual sea cucumber shipments to Hyōgo in the Register. .................................................... 159
Figure 40. Annual less-frequently shipped marine goods to Hyōgo in the Register. Various units of
measurement are used for these items, but as all were shipped infrequently all of the relative
volumes are small, so they have been combined on this map. Sharkskin and jellyfish are
measured in gō (approximately 1/10 of a masu, a unit of measure commonly used for rice).
Mackerel, flathead mullet, parrotfish, nishi (possibly a spiral shellfish), nikomi (unknown) and
abalone are in koku. Crab, squid, and shrimp are in horseloads. ............................................ 161
Figure 41. Annual seaweed shipments to Hyōgo in the Register.............................................................. 162
Figure 42. Annual egoma shipments to Hyōgo in the Register. ............................................................... 167
Figure 43. Annual indigo shipments to Hyōgo in the Register. ................................................................ 168
Figure 44. Annual other infrequently shipped items to Hyōgo in the Register. ....................................... 170
Figure 45. Iron shipments to Hyōgo in the Register and archaeological iron sites. .................................. 176
Figure 46. Pottery shipped to Hyōgo in the Register and archaeological sites. ........................................ 179
Figure 47. Sample of Mizunoko-iwa artifacts. Back Row: two kame and a tsubo. Front row: small
suribachi, medium and small tsubo, stacked suribachi. Photo by author (2012). Used with
permission of the Okayama Prefectural Museum. .................................................................. 182
Figure 48. Diagram of jars packed within a larger vat, from Kugui-no-oki site. NPO hōjin suichū
kōkogaku kenkyūjo (2010, p. 74). .......................................................................................... 183
Figure 49. Rudder from an eighteenth-century shipwreck. The lower area covered by copper sheathing is
intact (left), as is the charred tip of the post (close up, right), while the interior is eaten away
by shipworms. St. Augustine Lighthouse and Maritime Museum. Photo by author (2013). .. 200
Figure 50. “Fuji with a Rocket,” by Katsushika Hokusai. Note the men charring the base of the ship.
Hokusai and Smith (1988, pp 42-43). ..................................................................................... 200

xi

Figure 51. Egara Tenjin engi. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 147). ................ 207
Figure 52. Gyōbu Shirō networks. ............................................................................................................ 225
Figure 53. Originating ports for Hyōe Tarō networks. ............................................................................. 231
Figure 54. Shirō Saemon networks. .......................................................................................................... 235
Figure 55. Disputed port sites from the Register. ..................................................................................... 255

xii

List of Tables

Table 1. List of Identifiable Ports Noted in the Register (numerical order). ............................................ 118
Table 2. List of Identifiable Ports Noted in the Register (alphabetical order). ......................................... 119
Table 3. Pricing: Salt. ................................................................................................................................ 127
Table 4. Pricing: Rice. .............................................................................................................................. 137
Table 5. Pricing: Beans and Grains. .......................................................................................................... 147
Table 6. Monthly Shipping: Rice, Beans, and Grains. .............................................................................. 148
Table 7. Monthly Shipping: Lumber and Planks. ..................................................................................... 153
Table 8. Pricing: Wood Products. ............................................................................................................. 153
Table 9. Monthly Shipping: Sardines. ...................................................................................................... 158
Table 10. Monthly Shipping: Other Infrequently Shipped Marine Goods. .............................................. 163
Table 11. Monthly Shipping: Other Marine Goods, Unique Shipments. .................................................. 163
Table 12. Versions of the Maritime Codes. .............................................................................................. 193
Table 13. Gyōbu Shirō Networks. ............................................................................................................ 226
Table 14. Awaji-based Hyōe Tarō Networks. ........................................................................................... 230
Table 15. Shirō Saemon Networks. .......................................................................................................... 234

1

Figure 1. Map of medieval Japanese provinces. Courtesy Joan R. Piggott.

2

Chapter 1: Japan is an Archipelago

Travelers in Japan today are often impressed by the ease of using the modern train systems. Rushing
Tokyo to Hiroshima and beyond at speeds of nearly two hundred miles an hour, high-speed train travel is
part of the contemporary image of Japan. The routes tend to parallel the centuries-old roads that
connected the major city centers – the Tōkaidō between Tokyo and the Kansai region, the San’yōdō from
Kansai south towards Kyushu. Transportation historians are often fascinated by the continuing
development of these routes, from the checkpoint system that developed as early as the Nara period to the
lodgings that sprung up along the roads in medieval times. They cite these routes as the major connectors
throughout Japan, and draw parallels between development of these roads and development of the
economy and communities.
This approach, however, ignores the fact that Japan is an archipelago. While the road system
connected the interior of Honshu and provided a gateway to Kyushu, there is more to Japan’s
development than overland travel. The seas surrounding not only the major islands of Japan but the
smaller island chains within the Seto Inland Sea and elsewhere were connective routes as well.
Environmental historian William Cronon argues that “‘nature’ is a fundamental category of historical
analysis,” demonstrating its influence on economic and cultural development.
1
This dissertation explores
“nature” in the form of the Seto Inland Sea, showing how that sea shaped littoral communities during the
late medieval
2
era when domestic maritime trade and transport saw tremendous growth. As the volume of
maritime shipping increased, so too did the need for infrastructure to support it, codes to regulate it, and

1
William Cronon, “The Uses of Environmental History,” Environmental History Review 17, no. 3 (October 1,
1993): 4. Cronon problematizes the term “nature” by putting it in quotes, but does not fully define what he means by
the term. Based on his argument in the article, however, he seems to conceive of “nature” as an all-encompassing
term equal to and as nuanced as the concepts of gender, class, or race. He argues that environmental history draws
on numerous fields, from ecology to anthropology to economics and more, for its analyses. I use “nature” in a
similar sense here, referring not only to the physical environment of the Inland Sea but also encompassing the
practices of the people living in coastal communities.
2
For the purposes of this dissertation, the term “medieval” refers to the Kamakura through Sengoku eras (1185 –
1573). “Early medieval” refers roughly to the Kamakura period (1185 – 1333) and “late medieval” to the
Muromachi-Sengoku era (1333 – 1573). The question of periodizing Japanese history continue to be debated, and
these dates should not be considered absolute. See Joan R. Piggott, “Defining ‘Ancient’ and ‘Classical,’” and
Andrew E. Goble, “Defining Medieval,” in Karl F. Friday, Japan Emerging: Premodern History to 1850 (Boulder,
Colo.: Westview Press, 2012), for recent treatments of the issue.

3

people to ply the waters. I argue that the geography of the maritime environment shaped the physical
development of seaside communities, in turn affecting economic growth and creating a distinct maritime
culture unique to that setting.
This dissertation will focus on development in the Seto Inland Sea area, a waterway semi-
enclosed by the islands of Honshu, Shikoku, and Kyushu. Stretching nearly three hundred miles, the
Inland Sea coastline extends from the modern-day Osaka region, with inland waterways connecting to
Kyoto, to Fukuoka in the west, serving as the gateway to continental Asia. The sea is dotted with islands,
from rocky uninhabitable outcroppings to the largest island of Awajishima (about 270 square miles,
population in 2000 approximately 159,000).
3
While its average depth is approximately one hundred feet,
several of the straits exceed four hundred feet, with its deepest point, the Hayasui Seto between Shikoku
and Kyushu, plunging to over 1500 feet.
4
Tidal change along the islands can be as much as eight to ten
feet. Currents can exceed ten knots along some of the island chains, and certain areas along the northern
Shikoku coast are known for their whirlpools, making for challenges to navigation.
Based on a geographical and historical analysis of the maritime communities along the Inland Sea,
this dissertation will argue that the natural environment shaped the attitudes and practices of the people in
littoral communities, who in turn made use of and shaped the natural environment to suit their needs. The
geography and environment of the Inland Sea was in many ways ideal for seafaring, with naturally
accessible shoreline along coasts and islands that resulted in the development of ports serving as
transshipment hubs. Other areas that may not have had such ideal conditions to accommodate great
numbers of larger ships took advantage of natural endowments to function as production centers for
various marine commodities. As those ports developed, so too did a thriving maritime trade relying on
that production as well as shipping connections to create a robust trade network. With that increased
maritime-based contact, a culture of seafaring took shape that affected practices both on land and at sea.

3
Minami-Awajishima City. Awajishima no sōjinkō suii.
http://www.city.minamiawaji.hyogo.jp/soshiki/soumu/awajishimajinkousuii.html. Accessed March 22, 2015.
4
Japan Coast Guard, Hydrographic and Oceanographic Department, “Setonaikai no umi no fukasa wa.”

4

Taken together, ports, ships, commodities, and cultural practices are the various elements of the maritime
cultural landscape that helped shape medieval development in Japan.

The Early Inland Sea Seascape

The term “landscape” is widely understood in common parlance and has spawned its own set of studies.
Different approaches to landscape studies include examining the clearly defined or intentionally created
landscape, the organically evolved landscape created through socioeconomic development associated
with the environment, or the associated cultural landscape that includes sacred sites and outstanding
geographic features.
5
The concept of a “seascape,” however, is a newer idea. Throughout the world the
land has been privileged in studies of human history, relegating the role of the sea to a secondary position.
Particularly when looking at manifestations of political and economic power, developable land space is
traditionally seen as what mattered, rather than the untamable, uncontrollable sea. Studies of preindustrial
Europe, for example, often value the sea from an economic standpoint – as a connector used by trade
ships, but not as a space to be possessed by any land-based power.
6
The sea itself has been an
afterthought, simply the means by which people transported goods – it was the goods themselves that
were important.
Recently scholars have come to consider the “seascape” in its own right, examining the idea that
the ocean itself is important in shaping culture, technology, economy, and governance. Geographer and
historian Kären Wigen introduced one such perspective on the seascape that examines it from four
vantage points: constructs, such as islands, ships, port infrastructure and geographic space; sociologies,
comprised of the mariner’s world both at sea and in port; empire, or the projection of power over water
space; and transgressions, focusing on smuggling and piracy.
7
These four approaches provide a useful

5
Wendy Ashmore and A. Bernard Knapp, Archaeologies of Landscape: Contemporary Perspectives (Malden, MA:
Wiley-Blackwell, 1999), 9.
6
Philip E. Steinberg, The Social Construction of the Ocean, Cambridge Studies in International Relations
(Cambridge; New York: Cambridge University Press, 2001), 109.
7
Kären Wigen, “Introduction,” in Seascapes: Maritime Histories, Littoral Cultures, and Transoceanic Exchanges,
ed. Jerry H Bentley, Perspectives on the Global Past (Honolulu: University of Hawai’i Press, 2007), 3. The

5

start for understanding the history of the Seto Inland Sea and its maritime development, and they are a
helpful framework for this project in its entirety. The following chapters will argue that constructs are
often a direct product of the environment and geography, both shaping and being shaped by the local
population. Those constructs affect the people associated with the area, influencing governance and
culture (“empire” and “sociologies” in Wigen’s terms) and directing the actions of transgressors. First,
however, we shall consider the background of Seto Inland Sea maritime development, exploring the
classical and early medieval seascape and the relevant historiography of the medieval period.

Constructs: Ships and Maritime Infrastructure

For classical and early medieval times, constructs of the early Japanese seascape are difficult to grasp
clearly. Relatively little is known of the physical construction of ships or the layout of early ports,
whether in the Inland Sea or elsewhere along the archipelago. Extant evidence suggests that the
geography of classical and early medieval port sites could accommodate relatively small boats with
shallow drafts, and in later years as construction technology improved and ships grew larger, additional
infrastructure was built. Archaeological evidence has yielded some information about the earliest ships,
progressing from simple dugout canoes (maruko-bune or maruki-bune). Dugouts from as early as the
Jōmon period (12,000 – 300 BCE) ranged from three to eight meters long but averaged about six meters
(approximately twenty feet). Larger vessels used additional strakes (planks) to raise the waterline of the
vessel, and sailing technology was probably known from at least the Yayoi period.
8
Longer voyages such
as those along the trade routes to the mainland were likely undertaken in Chinese-style junks, but these

Seascapes volume was the result of a conference that had as it main goal to put ocean-oriented scholars together
without prerequisites of specific geography or era, and to examine the nature of the “seascape” worldwide. The four-
pronged approach of construct, sociologies, empire, and transgressions was born at this conference. I find it a useful
approach to examine the background of Inland Sea history at a holistic level and as such, I employ it here. Note,
however, that I use “empire” in Wigen’s sense to encompass control and administration of an area. I do not imply
that there was a push to establish an Inland Sea “empire” by any of the parties involved.
8
Okita Masaaki, “Kodai no fune,” in Toshi to kōgyō to ryūtsū, ed. Tanaka Migaku and Kanaseki Hiroshi, vol. 3,
Kodai no ronten (Tokyo: Shōgakkan, 1998), 287–8. Okita notes that although graphic representations in the Yayoi
and Tumulus periods and haniwa (statues placed surrounding tombs in the Tumulus Period) of boats rarely show
evidence of masts or sails, a Yayoi-era jar excavated from the Arao-minami site in modern Gifu prefecture depicts
two boats with what appear to be sails.

6

vessels were often poorly constructed and many apparently sank.
9
There was likely at least some effort to
duplicate Chinese construction techniques, as one report indicated that a ship broke in half but yet both
halves somehow made it to shore. This implies the use of Chinese watertight bulkheads on these ships,
which would have at least ensured that even if the vessel itself was severely damaged, the separate parts
would have been able to float.
10
As there were seventh-century references to the court ordering “Paekche
ships” to be built in Aki province, is it possible that the actual shipwrights were themselves Koreans and
employing their technologies in Japan.
11

A few hints at the mechanics of seafaring come from poetry and travel diaries. Sailing practices
regarding the journey to Silla are alluded to in poems from the Man’yōshū, Japan’s earliest poetry
collection. One such poem suggests that night sailing was not uncommon, indicating the need to rely on
the tides and the moon for light.
12
Another (verse 15:3646) alludes to the fearful nature of such voyages,
stating, “We have been rowing/ through the crescent cove in our ship/but the wind blows fierce/so out in
this fair bay/must we now find shelter?”
13
Though ships from this era, particularly those attempting to
cross to the Korean mainland, would have been equipped with sails, they were likely fragile bamboo sails
and not made of sturdier fabric.
14
Strong winds would have required the sails to be reefed (taken down),
and the sailors to rely on rowing to safety.
For domestic seafaring, one of the earliest accounts is the tenth-century Tosa Nikki by the courtier
poet and official Ki no Tsurayuki.
15
He traces his journey from Tosa (modern Tokushima prefecture on

9
Wang Zhenping, Ambassadors from the Island of Immortals: China-Japan Relations in the Han-Tang Period
(Univ of Hawaii Pr, 2005), 78. Wang notes that of the forty ships that Japan sent as trade missions to China in the
eighth and ninth centuries, twelve sank, suggesting the relative unseaworthiness of these craft.
10
Shoku Nihongi, Shinsōban, vol. 2, Kokushi Taikei (Tokyo: Yoshikawa Kōbunkan, 12), 287. The description of the
broken ship is from the entry on the fourth month, seventeenth day of Tenpyō Hōji 6 (762).
11
William Wayne Farris, “Shipbuilding and Nautical Technology in Japanese Maritime History: Origins to 1600,”
Mariner’s Mirror 95, no. 3 (August 2009): 266. Farris suggests that these may have been members of the relatively
frequent Paekche embassies to Japan.
12
H. Mack Horton, Traversing the Frontier: The Man’yōshū Account of a Japanese Mission to Silla in 736-737,
Bilingual edition (Cambridge, Mass: Harvard University Asia Center, 2012), 84. Horton translates Verse 1:8 as “At
Nikitatsu/we awaited the moon/to board our ship/and we have our tide as well/let us now row forth!”
13
Ibid., 103.
14
Ibid., 146.
15
The Tosa diary is not a typical travel journal due to the fact that the author (Ki no Tsurayuki, former governor of
Tosa) thinly disguises his identity by writing as a lower-ranking female attendant recording the actions of her

7

Shikoku) to Kyoto, a voyage that took fifty-five days. While his diary format is generally accepted to
have been a framework for a poetry anthology, Tsurayuki’s use of his journey as a backdrop provides
insight into the mechanics of seafaring. Numerous references to rowing the ship and occasionally to it
being towed when near shore suggest that the boat’s main form of propulsion was manpower, not sails.
The vessel was likely not very large, as repeated notes about the lack of food on board, records of who
brought food on board at different times, and incidents of trading poems for food in poetry contests
suggest that there was not much space on board for food storage.
16
While in the final example Tsurayuki
may have been trying to emphasize the value of poetry as a medium of exchange, apparently the ship did
not moor in open water. Instead it traveled from port to port, which would have let the passengers procure
food on a daily basis. It seems likely that any storage space on board ship would have been used for
shipping personal cargo and other items, and would not have been used for consumables. As for later
times, a more detailed discussion of later Kamakura shipbuilding and its evolution into Muromachi-style
ships follows in Chapter 2, but the little evidence that remains of shipbuilding through both
archaeological and literary references suggests that throughout the Heian (794 – 1185) and early
Kamakura (1185 – 1333) periods, most of the domestic vessels that plied the Inland Sea remained little
more than modified dugout canoes with minimal space on board.
Even less has been written about port sites and infrastructure. Matsubara Hironobu has traced the
known early histories of a number of ports in the Inland Sea region, particularly in Harima (modern
Hyōgo prefecture) and the Bisan Seto region (encompassing the area from modern coastal Okayama to
coastal Kagawa prefectures). Records are sporadic, however. Matsubara uses mention of the port names
in gazetteers or poetry collections such as the Man’yōshū to note existence of the ports, but little else can

superiors. Gustav Heldt argues that by taking on this persona and hiding his true identity, it is possible that
Tsurayuki was trying to create a “diary” that was not specific to his particular house or family. Gustav Heldt,
“Writing like a Man: Poetic Literacy, Textual Property, and Gender in the ‘Tosa Diary,’” The Journal of Asian
Studies 64, no. 1 (February 1, 2005): 7–34.
16
Ibid., 24. Heldt notes that this also reinforced the importance of poetry as a type of currency in the hinterlands.

8

be gleaned about most sites.
17
Many smaller port towns likely had no permanent docks or mooring points
for ships, but would have simply allowed vessels to pull up onto a sandy shore. As long-distance trade
increased and locals saw the advantage in creating mooring points for larger vessels, however, some
exceptions to this trend arose. The most notable would be Owada-no-tomari (modern day Kobe), which
was reportedly created in the eighth century. In the twelfth century, as Taira no Kiyomori (1118-1181)
moved to consolidate his power near his future capital at Fukuhara (modern-day Kobe), he realized the
economic advantages that monopolizing the China trade would provide him and his men. To actively
court the China trade – in opposition to the existing state laws (ritsuryō 律令) that curtailed overseas trade
-- he first needed to provide an infrastructure that could accommodate larger seafaring vessels. He thus
refurbished the old port of Owada to better accommodate the deeper drafts of the open-water vessels from
China. Bringing rocks down from the mountains nearby for landfill and inscribing some of them with
Buddhist sutras to ensure the island’s safety (figs. 2 and 3), he created a man-made island known as
Kyōgashima (“sutra island”) that allowed larger vessels both from China and domestic locales to moor
safely offshore. The port was burned in the Genpei War (1180 – 1185), becoming unusable until the Nara
royal temple Tōdaiji took over its administration to ship needed timbers to rebuild the Nara Great
Buddha.
18
As will be discussed in more detail in the following chapter, in medieval times Tōdaiji would
eventually come to administer the northern checkpoint at Hyōgo, on the site of Ōwada-no-tomari, while
Kōfukuji Temple took over the southern checkpoint.

17
Matsubara Hironobu, Kodai Setouchi no chiiki shakai (Tokyo: Dōseisha, 2008). See in particular chapter 6, pp
177 – 200, for a summary of ports and water transport.
18
Kamiki Tetsuo and Sakiyama Masahiro, Kaidō rekishi no ta-minaru: Hyōgo no tsu no monogatari (Kobe, Japan:
Kobe Shinbun Sogo Shuppatsu Sentaa, 1996), 37 – 44.

9

Figure 2. Rock from the Owada-no-tomari foundation, dating to the late Nara to mid-Heian periods and recovered in
2003. Owada-no-tomari today is covered with landfill and has been absorbed into the city of Kobe. Photo by author
(2012).

Figure 3. Rocks recovered from near the Kyōgashima site and said to have been buried with prayers for successful
completion of the manmade island. Pictured here on display at the temporary exhibit, “KOBE de Kiyomori” history
pavilion, on loan from Hoteiji Temple ( 布袋寺) in Kita-ku, Kobe city.
19
Photo by author (2012).

19
The pavilion was a temporary construct in Kobe to commemorate the 2012 NHK “Taiga Dorama,” a year-long,
weekly broadcast of a historical drama depicting the life of Taira no Kiyomori.

10

In the following centuries as political power shifted to Kamakura, a similar situation arose. The
city of Kamakura, located alongshore a protected cove, was ideal for welcoming ships, but the water was
again too shallow to accommodate larger vessels. The manmade island of Wagaeshima, created in 1232,
would have served a similar function as Kyōgashima.
20
The deeper waters surrounding the island would
have allowed ships to moor there and offload their cargo onto smaller ships to bring it to shore. Even
today a pier-like isthmus made of rocks extends from the shoreline to a large offshore island (figs. 4 and
5).

Figure 4. View of Wagaeshima at low tide from shore. Note that fishing boats still moor nearby today. Photo by author
(2011).

20
A stele placed in 1924 on the shoreline end of the rock isthmus states that an itinerant priest named Ōamidabutsu
往阿弥陀仏 made a behest to create this port and received the support of Taira no Moritsuna. Though a priest by
that name does appear in Kamakura-era records as having solicited funds for temple repairs, none specify his work
in the Wagaeshima area.

11

Figure 5. Wagaeshima. The piles of rocks still hold countless pottery sherds, as well as modern anchors, historical anchor
stones, and other remnants of maritime activity. Photo by author (2012).
These port constructs of the Heian (794 – 1185) and Kamakura (1185 – 1333) eras, then, in large
part owe their development to the China trade. Domestic trade apparently required only shallow landings
that could accommodate the smaller vessels based on a dugout canoe structure. Once open-water
seafaring became more prevalent, however, ships became larger and used Chinese shipbuilding styles
with various degrees of success. Those vessels needed deeper waters to moor, resulting in the creation of
better infrastructure in the form of the ports at Owada-no-Tomari or Wagaeshima. People therefore were
shaping the environment around them to support more types of trade.

Sociologies: Maritime Communities and Sailors

If little historiography has analyzed classical and early medieval maritime constructs, even less work has
been devoted to seafarers and the residents of port towns. Coastal communities were located far from the
central court and likely had fewer literate residents than did the capital, resulting in a distinct lack of
written records. What is known can generally be gleaned from early travel diaries or an occasional

12

document referring to the residents of maritime communities, usually written by someone from the central
court who regarded these areas as underdeveloped and quaint. Literary descriptions of the sea and
seafaring can therefore be problematic, since they were colored by the preconceptions of terrestrial
elites.
21
Life in the capital cities of Nara and Kyoto was the center of existence for most of those people,
and travel to the seaside was associated with either pilgrimages or exile. Descriptions of such travel, then,
are usually tinged with melancholy or an emphasis on long-distance travel as particularly arduous. While
it is unlikely that the seaside communities portrayed in these writings were indeed so bleak, or the seas so
storm-tossed, it is still possible to glean an idea of the composition and general tenor of seafaring and
seaside communities from these sources.
Once again, Ki no Tsurayuki’s Tosa Diary demonstrates a courtier’s view of seafarers. The
narrator notably distances himself from the commoner sailors and oarsmen, even the ship’s captain. On
several occasions the captain is reported to speak in a poetic tone, but it falls short of accepted poetic
conventions. The narrator recounts shock and surprise from the other passengers that a sailor could utter
even a near-poetic phrase, indicating that sailors were indeed seen as uneducated compared to their
passengers returning to the capital. Later on in the journey as the boat encounters rough seas, the captain’s
prayers were not enough to turn away a storm, so the sea gods were appeased by the sacrifice of a mirror.
Not only was the narrator contemptuous that the captain could not formulate words eloquently enough to
placate the gods, but he also seemingly sneers at the sea god himself for being soothed more by material
goods than beautiful verses.
22
Incidents such as these reveal the low esteem in which seafarers were held
by capital elites.

21
Michael McCarty, “Divided Loyalties and Shifting Perceptions The Jōkyū Disturbance and Courtier-Warrior
Relations in Medieval Japan” (Ph.D., Columbia University, 2013), 18–32. McCarty addresses the recent shift in
premodern Japanese studies from focusing nearly exclusively on court-related documents to the incorporation and
value of literary sources in understanding early Japan. He argues the need for careful consideration of the biases
present in the authorship and circumstances of creation of such sources, but posits that useful information is
available from these largely ignored sources in historical study. I agree with his contention and as such incorporate
literary sources into my studies where applicable.
22
Heldt, “Writing like a Man,” 19–20.

13

Nor did the view of seafarers and seafaring communities change much in the literary records of
later decades. In her early eleventh-century Pillow Book, Sei Shōnagon (c. 966 – c. 1025), lady-in-waiting
to the Queen Consort Teishi (976 – 1000), devotes nearly the entire section titled “Times When One
Should Be on One’s Guard” to the dangers of traveling via boat. She notes the rapidly changing seas as
she recounts how a voyage went from smooth sailing to high winds and waves tossing the boat about. Her
entry distinguishes between types of boats, claiming that “people of quality” traveled in boats with
deckhouses – she implies that smaller boats for people of lesser quality were no sturdier than bamboo
leaves. The most telling indicator of her attitude, however, is her description of sailors. While she seems
to hold them in some esteem, calling them the “bravest people in the world” for taking flimsy vessels out
on the open waters, there is a contemptuous tone to her writing that looks at sailors as people apart.
Writing that the “common people who man the boat run up and down, never giving a thought to the
danger,” she implies that perhaps they do not see the danger inherent in such travel that she and her
contemporaries do. She ends her tale by stating, “When all is said and done, only common people should
go in boats.”
23
Of course, many of the entries throughout Sei Shōnagon’s Pillow Book display open
contempt for lesser members of society, and it is also possible that her attitudes were extreme. But
coupled with Tsurayuki’s earlier attitudes towards sailors as being uneducated and inferior, and given
other writings discussed below that seem to treat members of maritime communities as a breed apart, it
seems likely that Shōnagon’s attitudes were not unique.
In actuality, however, records indicate that sailors were relatively important members of their
local communities. For instance, documents from a sixty-year time span in the thirteenth century
concerning Yuge estate (Iyo province) name eighteen pilots (kandori 梶取, a post roughly equivalent to a
ship’s captain that implies navigational and piloting skills) who transported rents and dues to Tōji temple.
At least seven of those men were elite cultivators who were rent collectors (myōshu 名主). And in a
census record from 1189, Yuge’s population was approximately five hundred people, of which only

23
Sei Shonagon, The Pillow Book, (Penguin Classics, 2007), 246–247.

14

twenty-three were myōshu.
24
So, a pilot who had myōshu status would have occupied a relatively high
position within the local estate hierarchy. Similarly, records from Yanai estate in Suō province show
instances of pilots being awarded tax-exempt fields for their services, which would have put them in the
upper echelons of that community as well.
25
Pilots, with their specialized skill sets, were far from the
lowest classes in the local social hierarchy.
26

One of the most enduring tropes in Japanese classical literature is the lonely seaside as a place of
exile, best exemplified by Genji’s exile to Suma in Murasaki Shikibu’s early eleventh-century Tale of
Genji.
27
Although Suma was less than a day’s journey from the capital at Kyoto, Genji bemoans his fate
as he contemplates other men who had also been banished there, “dripping brine from the sea grass.”
28

Throughout the chapter, his poems refer to boiling down brine to make salt, to gathering sea grasses, and
to the rolling of incessant waves that sounded “like moans of helpless longing.”
29
While the fisherpeople
and saltmakers in the poems are not fully fleshed-out characters, their inclusion in this narrative of exile
clearly demonstrates the perceived gap between life in the capital and that along the coast. The people of
littoral communities are seen as representatives of occupations quite foreign to court elites, and they
symbolize a lifestyle far removed from that of the capital city. Although Murasaki Shikibu does not show
Genji treating these figures with the contempt that Sei Shōnagon did in her Pillow Book jottings, the sense
of maritime communities as a world apart is reinforced in the Genji’s poems and imagery of the Suma
coast.
Later travel diaries record additional impressions of seaside towns and seafaring, although these
too were written with an eye towards poetic tropes, and they were often intended to extol the prowess of

24
Shinjō, Chūsei suiunshi no kenkyū, 67–70.
25
Ibid., 75.
26
Common sailors, however, may have been a different matter. Oarsmen (kako or suishu 水手) did not necessarily
require special training, and it is possible that they were commoner estate residents impressed into service for short
times only. (Shinjō, Chūsei suiunshi no kenkyū, 72).
27
Haruo Shirane, The Bridge of Dreams: A Poetics of “The Tale of Genji” (Stanford, CA: Stanford University Press,
1988), 19. Shirane notes that the focus on Suma in Genji gave rise to its common use as an utamakura, or poetic
place-name, including the maritime images of salt burning, fisherpeople, seaweed, boats, waves, and winds.
28
Murasaki Shikibu, The Tale of Genji, trans. Edward Seidensticker, 1st ed. (Knopf, 1978), 231. The narrator notes
that Genji is near the place of exile for the poet Ariwara Narihira, quoting from the latter’s poem about Suma in the
Kokinshū.
29
Ibid., 236.

15

the voyager. One such diary, written by Koga Michichika in 1180, recorded the pilgrimage of the retired
tennō Takakura to Itsukushima Shrine in Aki province, then a locus of power for the Ise Taira clan.
Unlike the more literary style of the Genji or the Pillow Book, the relatively prosaic descriptions of sailing
and coastal towns in the Takakura account seem not overly embellished. There are hints of the mechanics
of seafaring, noting the practice of numerous boats traveling together with Takakura’s royal boat. The
royal boat was larger, propelled by about twenty oarsmen. It also was said to have a deeper draft than the
accompanying vessels. In one instance the harbor was too shallow for the royal boat, and as a result
smaller watercraft were required to relay people ashore.
30
Though rowing seems to have been the
accepted form of propulsion and sometimes even continued through the night, in good weather sails were
used as well.
31
The propensity to travel in fleets was a custom followed for many centuries, and will be
discussed further in Chapter 4.
Takakura’s entourage felt some of the same apprehension as had Sei Shōnagon earlier,
underscoring the challenges facing voyagers in an unfamiliar world. For example, upon Takakura’s
departure his scribe Michichika recorded that, “Everyone who was to be in attendance on the journey
would, to their discomfort, have to travel by boat.… [the ladies in waiting] disregarded the taboo against
speaking and lamented amongst themselves, ‘what kind of a recreational excursion is this!...”
32
The
universal reluctance to travel by sea from the capital suggests that the pilgrimage itself may indeed have
been a scheme of Taira no Kiyomori’s to keep Takakura and his entourage under his influence, rather
than simply the whim of the retired tennō. Upon arrival at Takasago harbor, Michichika notes that, “the
sound of the waves, which none of us was accustomed to hearing, all of a sudden became frightening and
the shouting of fishermen rang in our ears.”
33
Once again the imagery of the seaside, so foreign to
members of the court, is presented as fearsome and foreboding.

30
Herbert E Plutschow, ed., Four Japanese Travel Diaries of the Middle Ages, Cornell University East Asia Papers
(Ithaca, N.Y: China-Japan Program, Cornell University, 1981), 34–35.
31
Ibid., 42.
32
Ibid., 30.
33
Ibid., 34.

16

A final aspect of note in seaside towns is the portrayal of women there. Jan Goodwin, in her
comprehensive study of early sexual entertainers, Selling Songs and Smiles, cites the late tenth-century
Asobi wo miru’s account of the governor of Iyo’s travels along the Inland Sea. He stopped at Kaya
(Yamazaki port, on a river near Kyoto), where women were waiting at the anchorage. Goodwin translates
the sight described therein as, “The younger women melt men’s hearts with rouge and powder and songs
and smiles, while the older women give themselves the job of carrying the parasols and poling the
boats.”
34
Women of all ages were there to welcome travelers to the port. There is also a notable female
presence in two entries of Takakaura’s journey to Itsukushima. At Muro harbor, sexual entertainers
35

(yūjo 遊女) approached by boat near the temporary residence of the retired tennō, but when they were
ignored, they retreated. Then at Kojima harbor, the scribe Michichika notes that there were no entertainers,
and he wonders at their absence since they approached “every other traveler.”
36
The fact that the absence
of sexual entertainers is noteworthy suggests that by this time they were a fixture in port towns. The same
is suggested in Lady Nijō’s 1307 memoir of her travels along the Inland Sea, as she notes the presence of
such women in Tomo-no-ura to “ease the traveler.”
37
Taken together, these scenes certainly imply the
near-permanent presence of sexual entertainers in port communities.
From the perspective of travelogues and literary sources, then, seafaring was generally seen as a
rough, dangerous pastime, by and large undertaken only in the direst necessity. It was a communal affair,
as voyages usually proceeded in groups that hugged the shoreline. Early records romanticize the
loneliness of the coast and fishing villages, while later travelogues minimize those tropes in favor of
narrative accounts of the geography. The typical scenes popularized by Genji’s exile at Suma continue to
appear, as later travelers noted the smoke of saltmaking huts, the crashing waves and pine trees of the
shoreline, the sights of small fishing boats along the shore, and the presence of women as “one-night

34
Janet R. Goodwin, Selling Songs and Smiles: The Sex Trade in Heian and Kamakura Japan, (University of
Hawaii Press, 2007), 15.
35
Goodwin prefers this translation, as “prostitute” has connotations of legality that were not necessarily applicable
to premodern Japan. See Goodwin, Selling Songs and Smiles, esp. 2-3.
36
Plutschow, Four Japanese Travel Diaries of the Middle Ages, 36.
37
Nakanoin Masatada no Musume, Towazugatari, trans. Tamai Kōsuke (Tokyo: Iwanami Shoten, 1997), 244.

17

wives” that visited incoming ships. The courtier elites that recorded such accounts saw sailors as a breed
apart, commoners uneducated in the finer arts of courtly speech and behavior.
On the other hand, records of coastal estates like Yugeshima indicate that pilots were important
members of their local societies. Such a difference hints at an early Japanese manifestation of the
“mariner’s perspective,” a viewpoint particular to those who live and work on and near the sea.
38
As we
shall see, this mariner’s perspective would color the developments of seafaring and its governance over
time, as will be discussed more thoroughly in Chapter 4.

Governance: Power Players and Hierarchies

To explore governance of the seascape, which includes matters of administration and control, requires
examining both domestic and foreign concerns. Much historiography concerning domestic government of
the Heian provinces and peripheries is concerned with the collection of taxes, rents, and dues, whether
from landed estates (shōen 荘園) or public land unites (kokugaryō 国衙領). As will be discussed further
below in this section, the period also saw significant interaction with China, the Ryūkyū Islands, and
other foreign parts – issues of governance include concern for the Japanese role in the greater East Asian
sphere that spanned several oceans. I will show here that control of coastal sites was integral to central
authorities and their regional and local delegates for purposes of economic development, for their
livelihoods, and to demonstrate their prestige.
The late Heian period, at least from the eleventh century on, witnessed the development of the
estate (shōen) system. Estates submitted rents and dues directly to an overseeing institution (sometimes
called a “proprietor”), such as a temple or shrine, or a member of the court, royal family, or regent’s
lineage. Public lands, on the other hand, included administrative districts (gun 郡), hamlets (gō 郷),
neighborhoods (ho 保), and fields (myō 名), and taxes on those lands were collected under the authority

38
Paul Rainbird, The Archaeology of Islands, 1st ed. (Cambridge: Cambridge University Press, 2007), 49. Rainbird
describes the development of a “mariner’s perspective” by suggesting that people associated with the sea have
distinct practices and mindsets that create a maritime identity. See the section below for further details.

18

of the provincial governor.
39
As these rents and dues were often assessed in rice or other products, much
of the scholarship on the estate system focuses on agricultural production, providing fewer insights into
the activities and contributions of non-agrarian communities. One reason for this may have been the fact
that particularly in the earlier Nara (710 – 794) period, the state itself largely ignored maritime
governance. Indeed, the classical historian Matsubara Hironobu, in his article “Kakai no kōtsū,” points
out that the eighth-century ritsuryō (legal and penal 律令) codes focused on regulating overland
checkpoints for terrestrial travel. Even regulations for maritime checkpoints, particularly those on rivers,
were referred to as wataritsu ( 渡律), which places the focus on terrestrial crossings rather than waterways.
Matsubara suggests that the ritsuryō state simply did not focus on maritime travel until it began to
increase substantially in the eight century.
40
Takeda Sachiko, another classicist, carries this thought even
further, noting that the sea route was preferred by foreign embassies, but that the courts wanted envoys to
use roads. The architectonics of the land route – a straight, far-reaching road with Chinese-style
architecture – had been designed to convey a sense of the monarch’s terrestrial authority. An approach by
sea approach would fail to adequately impress foreign travelers.
41
It was no wonder, then, that the state
overlooked governance of the sea in favor of the land.
The medievalist Amino Yoshihiko (1928 – 2004) was one of the first scholars to challenge the
perception that maritime communities were less important than terrestrial ones as he explored the role of
fishing villages and salt producers, among others, within the estate system, and I will look more closely at
his contributions later in this chapter. Then in 1994, Shinjō Tsunezō produced perhaps the most complete
study of medieval seascape governance and economy in his Chūsei suiunshi no kenkyū (Studies on
Medieval Maritime Shipping). He looks closely at coastal and riverside communities and their
involvement in maritime shipping, examining the roles of seafarers (pilots and sailors) and their status

39
Amino Yoshihiko et al., Rethinking Japanese History (Ann Arbor, Mich.: Center for Japanese Studies, University
of Michigan, 2012), esp. 65.
40
Matsubara Hironobu, “Kakai no kōtsū,” in Hito to mono no idō, ed. Yoshimura Takehiko (Tokyo: Iwanami
Shoten, 2005), 47–49.
41
Sachiko Takeda, “Roads in the Tennō-Centered Polity,” in Capital and Countryside in Japan, 300-1180:
Japanese Historians Interpreted in English, trans. Joan R. Piggott (Cornell University East Asia Program, 2006),
163.

19

within communities, their interactions with central authorities, and the types of cargoes being shipped
throughout the archipelago. Shinjō shows that in reality control of coastal territory was critical to the
court authorities. For instance, he notes that the members of the provincial administrative office (kokuga
国衙) near the coast would often establish an official marina (funasho 船所) to ensure a firm grasp on
maritime transport for rents and dues.
42

Certain studies have focused on control of coastal areas by local powerful families or proprietary
institutions. For early times, Matsubara has done studies of early clans along the Inland Sea, particularly
those on Shikoku, and he examines their ties and administrative loci in the ritsuryō and early Heian
periods (the “classical period”). While documentation is scattered, he is able to demonstrate that local
Shikoku families had significant ties with the court, and he suggests that there was a significant amount of
trade carried out between Shikoku sites and central Yamato-based authorities even in the early classical
period.
43
He also has studied the types of items sent as tribute goods by coastal provinces as listed in the
tenth-century Engishiki (Procedures of the Engi Era), highlighting some of the traditional items that were
produced or procured in those areas. As the estate system developed in the later Heian period, the
wealthiest or strongest proprietors gained significant control of coastal venues and their communities,
allowing them access to the marine goods produced there as well as control over the ships that passed
through those ports. For instance, Amino has noted that many of the most prominent powers, including
Ise Shrine, Iwashimizu Hachiman Shrine, and the royal and regent families had significant coastal
holdings. He particularly remarks upon the lands of the high ranking ministerial Saionji family in
Kamakura times: in addition to several inland estates, their holdings included an entire county (Uwa) on
Shikoku, two coastal settlements (Nuta and Kayō) in Aki and Bitchū provinces, and several riverside
estates on the Uji and Yodo rivers near the capital. The presence of their properties on these waterways

42
Shinjō, Chūsei suiunshi no kenkyū, 33. At the marinas provincial officials could register and conduct inspections
of the boats, dispatch sailors, or oversee cargo loading and unloading.
43
Matsubara Hironobu, Kodai no chihō gōzoku (Tokyo: Yoshikawa Kōbunkan, 1988). Matsubara traces the
evolution of major clans on Shikoku from the Kofun Period, looking at tomb clusters and patterns, and continues to
examine documentary evidence of clans throughout Sanuki, Iyo, Tosa, and Awa provinces (all on Shikoku) through
the early Heian period. In many cases he has been able to discover several major clans whose members were
awarded court titles, indicating strong ties with the central court.

20

allowed them to exert control over riparian shipping, and afforded them direct access to the marine goods
that could be procured from the Inland Sea. Throughout Kamakura times the Saionji family further
expanded their lands, allowing them control of the Bungo Straits that separated the Inland Sea from
northern Kyushu, and which served as the gateway to the mainland.
44
Control of these coastal
communities helped make the Saionji one of the wealthiest and most powerful courtier families from the
late Heian through the early Muromachi eras.
There is also scholarship that examines early Japan’s role within the East Asian sphere. As
mentioned above, Taira no Kiyomori was determined to expand the China trade, and he created an
infrastructure to encourage larger trading vessels. In her monograph Across the Perilous Sea, Charlotte
von Verschuer notes that Kiyomori set a new precedent by dealing personally with Chinese
representatives, giving and receiving gifts and going so far as to arrange an audience for them with the
retired sovereign. Previously royals had very little direct contact with foreigners, and Kiyomori’s acts met
with official outrage. Yet after this disruption of tradition, more Japanese “castaways” were discovered on
China’s coasts – likely, individuals or groups seeking foreign trade partners.
45

In his book To The Ends of Japan, Bruce Batten explores borders and networks in premodern
Japan. He suggests that borders were both geographic (the ocean) and mental, as represented by the
presence of an opposing power (Silla on the mainland or the Emishi people in the north).
46
He notes that
during most of the late Heian and all of the Kamakura periods, until the first Mongol invasions in 1274,
Japan neither sent nor received any full-scale diplomatic voyages from the mainland, suggesting a period
of relative isolation.
47
Yet he also argues that as the control of the court over the periphery began to
fragment, local powers became more autonomous and took control over maritime trade with the mainland.

44
Amino et al., Rethinking Japanese History, 110–112. For a detailed treatment of the Saionji family maritime
holdings in the Kamakura and early Muromachi eras, see Rieko Kamei-Dyche, “Tools of Authority: The Saionji
Family and Courtier Society in Early Medieval Japan” (University of Southern California, 2013), 162 – 167.
45
Charlotte von Verschuer, Across the Perilous Sea: Japanese Trade with China and Korea from the Seventh to the
Sixteenth Centuries, Cornell East Asia Series (Ithaca, NY: East Asia Program, Cornell University, 2006), 46–47.
46
Bruce L. Batten, To the Ends of Japan: Premodern Frontiers, Boundaries, and Interactions (Honolulu: University
of Hawai’i Press, 2003), 32–33.
47
Ibid., 150.

21

The best example of this was the Sō family based in Tsushima. They were able to leverage their
geographic proximity to the Korean peninsula to eventually dominate trade affairs in that sphere.
48
While
diplomatic contacts may have diminished during the later medieval period, unofficial trade continued to
thrive. Similar to the case of the Saionji family increasing their power through control of the domestic as
well as interregional waterways, families such as the Sō could use their foreign ties to obtain additional
wealth and prestige.
49

Archaeological evidence demonstrates the frequency of this unofficial trade. Japanese
archaeological excavations regularly unearth Chinese copper coins from the twelfth and thirteenth
centuries. In his recent study, Ethan Segal has described a debate among late Heian nobles about the use
of foreign coins, noting that nearly all were in favor of a ban on coins and suggesting that their main
concern was maintaining political authority. Segal argues that although the ban was enacted, coins still
flowed into Japan, suggesting a thriving unofficial trade – and a flouting of the central authority’s rule.
50

Batten notes that 135 cached hoards of coins have been discovered, most from the thirteenth through
sixteenth centuries, and most found in eastern Japan. He suggests that the geographic distribution reflects
the relative political instability in the east, as people would have been more likely to hoard their precious
coins.
51

Evidence from the Ryūkyū Islands also suggests significant unofficial interaction between those
islands and the Japanese mainland. The great green turban shell (Turbo marmoratus), prized for its
iridescent inlay, was a major Ryūkyūan export and was in great use in Kyoto as early as the beginning of
the Heian period. Ornaments used for the statuary at Byōdoin in Uji included turban shell inlay, as did
Heian-period horse saddles, chests, and mirrors. Turban shell inlay can also be found in the decorations of
the twelfth-century Konjikidō at Chūsonji in Hiraizumi (in modern Iwate prefecture), suggesting a wide

48
Ibid., 39.
49
The Saionji were not without their own foreign ties. They held Munakata Shrine in northern Kyushu, which had
deep ties with the Chinese mainland. For two generations the wives of the head priests were actually Chinese.
Amino et al., Rethinking Japanese History, 112.
50
Ethan Isaac Segal, Coins, Trade, and the State: Economic Growth in Early Medieval Japan (Cambridge, Mass:
Harvard University Asia Center, 2011), 56–58.
51
Batten, To the Ends of Japan, 196 –197.

22

trade range.
52
Other artifacts found on Kikaigashima, in the northern part of the Ryūkyū island chain,
include eleventh- and twelfth-century pottery from Sanuki and Higo provinces on Shikoku and Kyushu,
suggesting direct interaction between those areas.
53
As no documents survive, the particulars of such trade
are unclear, but based on the lack of official trade regulations during this era, it seems likely that
individual entrepreneurs carried on such trade.
In short, domestically the court exerted its control through an administrative presence along the
waterways, gaining wealth and power from such governance. Within a greater East Asian context, the
central court also wished to control foreign interactions, but it enjoyed less success there. Regardless of
official policies, both written and archaeological evidence confirm the existence of regular overseas trade
networks. The geographic situation played a part in determining which port areas became important trade
loci, allowing those who held control over those venues to enjoy the wealth and goods brought in by trade
and enhancing their prestige.

Transgressions: Piracy and Rebellion

Piracy is a universal problem on the water, regardless of time and place. The small islands within the
Inland Sea made it a prime location for pirates, providing many small coves that skilled mariners could
use to escape pursuit. Ki no Tsurayuki, in the Tosa Diary, alludes to pirates taking revenge on provincial
officials (a thinly veiled allusion to himself – he was fearful of pirates who threatened to attack him for
his rulings against them as governor of Tosa). Although he notes reports of pirate activity, and in one
entry claims that the boat should not set sail before midnight in an attempt to evade the pirates, he does
not record any actual pirate sightings.
54

One of the most notable cases of an individual taking to the seas to assert power, and the focus of
much of the scholarship on early piracy in Japan, is the tenth-century rebellion led by Fujiwara no
Sumitomo (? – 941). This former provincial official of Iyo province turned from his government post in

52
Richard J. Pearson, Ancient Ryūkyū: An Archaeological Study of Island Communities, 2013, 140–141.
53
Ibid., 159.
54
Ki no Tsurayuki and William N. Porter, The Tosa Diary, Bilingual (Tuttle Publishing, 2005), 75–79.

23

an active attempt to challenge the rule of the court and assert his own power. He gathered crews of
seamen under his command to terrorize ports along the Inland Sea and Yodo River. Initially the court
attempted to lure Sumitomo back to a legitimate position, offering him a promotion and issuing arrest
warrants for his followers.
55
The court’s initial response was surely due to the shortage of options – it had
no centrally organized and controlled naval fleet. But after initially accepting the promotion, Sumitomo
once again amassed a fleet and began to attack ports in Iyo and Sanuki provinces. At this point the court
commissioned Ono no Yoshifuru to pursue Sumitomo, and only when Yoshifuru ordered two hundred
ships to be built was there the creation of an organized military fleet.
56
Sumitomo proceeded south to
Kyushu, finally burning the government seat of the Kyushu headquarters (Dazaifu) in the fifth month of
941. But shortly after, Yoshifuru and his fleet met Sumitomo and sank nearly eight hundred pirate
vessels.
57
Though Sumitomo escaped, a mere two weeks later he was captured and beheaded.
Organized piracy on Sumitomo’s scale would not be seen again for several centuries, although on
occasion the court exerted control over smaller-scale pirate groups.
58
Fifty-one years after Sumitomo’s
death, for example, a Minamoto no Tadayoshi was appointed to search and destroy a pirate band based in
Awa province.
59
Then in 1135, pirates were enough of a threat in the Inland Sea that the court sent Taira
no Tadamori to subdue them.
60
As there are a number of similar such records, these references suggest
that piracy was indeed still a concern of maritime travel, but that large-scale pirate attacks were relatively
rare. Sets of maritime codes said to have been written in the 1220s have very few regulations either

55
Karl F. Friday, The First Samurai: The Life and Legend of the Warrior Rebel Taira Masakado (Hoboken, N.J.:
John Wiley & Sons, 2008), 143.
56
Ibid., 144.
57
Kōen and Eiyū, Fusō ryakki. Teiō hennenki, Shinsōban (Tokyo: Yoshikawa Kōbunkan, 11), 218–219. 940
(Tengyō 3) 11.21. As records of battle often exaggerated for effect, we should hesitate before accepting numbers of
troops or ships on either side at face value.
58
Peter Shapinsky notes the sometimes nebulous nature of early “pirates.” Terminology could be ambiguous, as
sometimes those referred to as “pirates” were officially enveloped into estate hierarchies to perform duties such as
escorting ships delivering rents and dues. The line between licit and illicit maritime acts was often blurry,
particularly in later centuries. Peter D. Shapinsky, Lords of the Sea: Pirates, Violence, and Commerce in Late
Medieval Japan (Ann Arbor: Center for Japanese Studies, University of Michigan, 2014), 56–62.
59
Amino Yoshihiko, Higashi to nishi no kataru nihon no rekishi (Tokyo: Kōdansha, 1998), 111. Throughout this
dissertation, “Awa” refers to the province on Shikoku ( 阿波, modern Tokushima prefecture), not the former Awa
province ( 安房) on the modern Chiba peninsula.
60
William Wayne Farris, Heavenly Warriors: The Evolution of Japan’s Military, 500-1300, (Cambridge, Mass.:
Council on East Asian Studies, Harvard University: Harvard University Press, 1992), 253.

24

outlining punishments for pirates or compensation for victims of piracy.
61
One document from 1294
shows the manager of new royal edict lands (shin-chokushiden 新勅旨田) in Aki province
acknowledging the need to provide documentation in case of trouble on the seas due to piracy.
62
And in a
notable case in 1308, the shogunate stationed guards in fifteen provinces along the San’yō and Nankaidō
roads in defense against a pirate uprising. Six years later Kobayakawa Tomohira was dispatched by the
government, and he successfully apprehended the alleged pirate Gagaku Saemon Jirō.
63
In the early
medieval period, however, these isolated incidences provide the only evidence that piracy was a
significant concern in the Inland Sea region. Chapter 2 will discuss how in later years, likely due to
increased maritime traffic and trade, pirates (or, as some have called them, “sea lords”) were a more
visible presence in the area.

The “Seascape Gap” in the Historiography of Medieval Japan

The early Japanese Seto Inland Seascape elaborated above provides some hints as to the prominence of
the sea in daily life. The sea affected nearly everything, from the need for infrastructure to the roles –
perceived or actual – of seafarers and members of maritime communities, and to questions of governance,
taxation, and piracy. That fact notwithstanding, surprisingly few maritime-centric studies exist for early or
medieval Japan: there is a historiographical “seascape gap” that is particularly notable in English-
language scholarship. Again, the focus for many historians of Japan has been the landscape, especially the
control of rice and dry fields. When scholars mention maritime matters at all, they generally relate it to
overseas trade with China and Korea, foregoing discussion of the importance of the sea and coastal
dwellers in Japan’s historical affairs. Studies of the sea as domestic space-that-shapes, then, have only
recently begun.

61
There is some debate, however, whether the maritime laws (kaisen shikimoku) were actually dated from this time
or were written in the fifteenth centuries and backdated. This will be discussed more thoroughly in Chapter 4.
62
Shinjō, Chūsei suiunshi no kenkyū, 776.
63
Karaki Hiroshi and Hashitsume Shigeru, Chūsei no Sanuki (Takamatsu, Japan: Mikōsha, 2005), 115.

25

Medieval Constructs: Ships and Maritime Infrastructure

Part of the dearth of studies on shipbuilding or maritime infrastructure is simply due to the paucity of
sources, be they written, archaeological, or artistic. Shipwrights did not begin to create lines drawings
64

until the early modern period, so no such records remain; and few examples of boats have been
discovered in archaeological excavations. As for infrastructure, far more energy has been expended on
reconstructing agricultural communities than maritime ones. In looking at what scholarship does exist, we
see the historiographical “seascape gap” in domestic Japanese maritime constructs even more clearly.
Ishii Kenji is the foremost scholar of ship construction history in Japan, having produced several
volumes on Japanese boats (wasen 和船). He introduces the different types of vessels and technological
innovations from prehistoric dugout canoes through large trade vessels (kaisen 廻船) of the early modern
period. In his discussions of late medieval vessels, however, save for a single-page discussion of ship
sizes as seen in a set of port records,
65
his focus is squarely on the envoy ships (kenminsen 遣明船) used
for trade and exchange missions with China.
66
While these ships were certainly an important part of
medieval Japanese maritime history, his lack of interest in domestic-use ships demonstrates the dual
tendencies of highlighting the sea’s importance for overseas trade and ignoring its importance for
domestic trade.
Even less information is available in English. Charlotte von Verschuer’s Across the Perilous Seas,
while wonderfully detailed in her descriptions of the types and volumes of cargo typical of trade missions
to mainland Asia, largely neglects the mechanics of that trade (i.e., the ships and manpower needed to
make those journeys).
67
Furthermore, she focuses on the maritime connections between Japan, China, and

64
“Lines drawings,” in the plural, refers to the plans used to build a ship’s hull. These usually consist of three views:
one showing the longitudinal elevation (sheer plan), a vertical cross-section (body plan), and the horizontal planes
parallel to the waterline (half-breadth plan). See I. C. B. Dear and Peter Kemp, The Oxford Companion to Ships and
the Sea (Oxford: Oxford Univ. Press, 2005), 60, 378–80.
65
Ishii uses figures from the Register of Incoming Ships at the Hyōgo North Gate (Hyōgo Kitaseki Irifune Nōchō) to
determine ship sizes. These records will be discussed in great detail in later chapters of this dissertation.
66
Ishii, Zusetsu wasen shiwa, 44–60. Ishii even titles the chapter on Muromachi sailing as “Envoy Ships: The Large
Vessels of the Muromachi Era.”
67
Von Verschuer, Across the Perilous Sea. See particularly Chapters 3 and 5 for her descriptions of trade
commodities.

26

Korea, with little acknowledgment of the maritime infrastructure needed domestically to facilitate such
trade, and without any note of how domestic shipping may or may not have been affected. William
Wayne Farris has written an article entitled, “Shipbuilding and Nautical Technology in Japanese Maritime
History: Origins to 1600.” While serving as a helpful introduction to the topic overall, the essay largely
abstracts much of Ishii’s work on medieval ships. Farris does, however, devote a few paragraphs to
domestic maritime concerns during the Muromachi era (1337 – 1573). Most importantly, he notes how
expanded demand for products led to the development of more efficient maritime trade; it resulted not just
in construction of larger, faster ships but also in the growth of port towns and merchant facilities.
68
This is
one of the few acknowledgments of the need for maritime infrastructure that developed as a result of
increased maritime trade in later medieval times.
The historiographical “seascape gap” is further evident in what is missing from landscape and
settlement studies. Few scholars focus on the development of port cities – unless they are critical to
overseas maritime trade. For example, Yamamura Kōzo, in “The Growth of Commerce in Medieval
Japan,” writes, “we know many [port towns]… played important roles in the economic activities of the
capital trade region,” and he goes on to name several sites near Kyoto, on the Yodo River, and Lake Biwa.
But he provides no details about those ports, citing insufficient documentation, and the reader must
speculate as to what those “important roles” may have included. Later in the same chapter, however, he
devotes several pages to the discussion of trade with Song China, emphasizing the role of foreign trade at
the expense of domestic trade.
69
Yamamura does at least note that many late medieval ports were situated
in coves and lagoons, often in close proximity to shrines and temples that sometimes doubled as port
authorities.
70
In Kinda Akihiro’s relatively recent A Landscape History of Japan, only a scant chapter by
Yamamura Aki is devoted to medieval towns and cities, and of that, merely two pages discuss port towns

68
Farris, “Shipbuilding and Nautical Technology in Japanese Maritime History: Origins to 1600,” 275.
69
Yamamura Kōzo, “The Growth of Commerce in Medieval Japan,” in The Cambridge History of Japan, ed.
Yamamura Kōzo, vol. 3 (Cambridge: Cambridge University Press, 1991), 356–360.
70
Yamamura Aki, “Medieval Towns,” in A Landscape History of Japan, ed. Kinda Akihiro (Kyoto: Kyoto
University Press, 2010), 81.

27

that were perhaps best known for overseas trade: Hakata and Obama.
71
And while Soranaka Isao has
written on the history of Obama, a port on the Japan Sea (Wakasa province, modern Fukui prefecture), he
devotes little time to discussions of the physical layout of the port zone, only noting that in the fifteenth
and sixteenth centuries the volume of international trade coming through Obama rose due to its proximity
to mainland Asia.
72
In Hakata’s case, its pivotal location on the northern coast of Kyushu has led it to be
called the “Gateway to Japan,” from China, Korea, the Ryūkyūs, and Southeast Asia.
73

Many of the conclusions in Bruce Batten’s study of Hakata can be applied to other domestic ports.
For example, he demonstrates that cross-border contacts overseen by the Kyushu Headquarters at the
Dazaifu near Hakata helped the area become a natural center of regional power. The Dazaifu acted as
defense headquarters, diplomatic site, and outpost of court control. A number of these same functions
would also have been carried out on a smaller scale at local ports in the Inland Sea region. In this regard,
one recent work by Ichimura Takao, entitled Chūsei Sanuki to Setouchi sekai (Medieval Sanuki and the
World of the Seto Inland Sea), begins to close the “seascape gap” as it shifts away from the focus on
overseas trade to look at the role of Sanuki province (modern Kagawa prefecture) in the medieval Inland
Sea.
74
Though most of the contributors to that volume highlight the trade goods and fishing activities of
ports, Kitayama Ken’ichirō examines the physical layouts of several ports in medieval Sanuki in a manner
that highlights the geographical aspects of port development. Through archaeological evidence and
reconstruction of thirteen ports, he identifies six ideal aspects of any port: a place where a boat can moor
(wharf or seawall), a place to load or unload cargo, places that can be used as navigational markers (often
temples or castles), institutions that can administer the port, a road running through or crossing the port,
and access to inland areas.
75
All of these aspects point to the importance of a port site functioning as an
easy place to exchange goods. Boats could easily come in and out, anchor, offload, and transport their

71
Yamamura, “Medieval Towns,” 80-81.
72
Soranaka Isao, “Obama: The Rise and Decline of a Seaport,” Monumenta Nipponica 52, no. 1 (April 1, 1997): 90.
73
Bruce Batten applies this title to Hakata in his monograph: Batten, Gateway to Japan.
74
Ichimura, Chūsei Sanuki to Setouchi sekai.
75
Kitayama Ken'ichirō, “Chūsei minatomachi no chikei to kūkan kōzō,” in Chūsei Sanuki to Setouchi sekai, ed.
Ichimura Takao (Tokyo: Iwata Shoin, 2009), 124.

28

goods to inland destinations. As will be discussed in chapters following, local ports not only became loci
for trade, but like the ports at Dazaifu and Obama, the ports were also used as strategic stopping points in
dangerous areas, and they provided overseers – those who controlled them – with political power.
Understanding medieval constructs lays the groundwork for examining the rest of the seascape.
Without knowing the capabilities of both ships and ports, which are often determined by geographic
location and factor endowments (including natural resources, climate, and population), it is impossible to
understand the mechanics of maritime shipping and shipping routes. Examining which areas developed
their maritime infrastructure and why also provides key insights into the dynamics of maritime
development at the local level. Chapter 2 will explore more aspects of maritime constructs, from the ships
themselves to the physical layout of port facilities. But first we shall explore the historiography of the
society that developed around medieval port sites.

Medieval Sociologies: Court, Warriors, Peasants… Sailors?

Scholars of medieval Japan working in English have not focused on the people and cultures of maritime
communities, though there have been more studies produced in Japanese. Central power in the medieval
period shifted with the fall of the Kamakura shogunate (1333), attempts by the court under Go-Daigo (r.
1318 – 1339) to reassert its own power, and the rise of the Ashikaga shogunate (1336 - 1573). Many
scholars of this era focus on the prominence of warriors and subsequent jockeying for power among its
members. Even studies of influential families in coastal provinces, such as Peter Arnesen’s The Medieval
Japanese Daimyō that focuses on the Ōuchi family, regional elites of western Honshu, largely ignore
maritime aspects of the region.
76
Littoral communities receive little to no mention in English-language
medieval scholarship unless they are involved with the China trade, in which case discussion focuses on

76
Peter Judd Arnesen, The Medieval Japanese Daimyo: The Ōuchi Family’s Rule of Suo and Nagato (Yale
University Press, 1979). Arnesen looks at the role of power relations and the development of military leadership in
the Muromachi age through the lens of the Ōuchi family based in Suō and Nagato provinces. While he does focus on
the commoner classes to an extent, highlighting the results of the crumbling estate system and the need for those in
power to secure steady revenue, he does not speculate as to the advantages or disadvantages of overseeing maritime
estates, nor does he examine the people in those communities.

29

overseas interactions.
77
Several scholars of Edo-period history have tried to bridge the “seascape gap,”
most notably Arne Kalland in his study of fishing villages, but there is a sad lack of equivalent studies for
earlier eras.
78

One happy exception is Peter Shapinsky’s recent monograph, Lords of the Sea, which devotes a
short section to the “sea peoples” (ama 海民、海人), drawing from literary depictions that portray them
as a class apart from those authoring the accounts.
79
Much of his evidence is drawn from classical
literature and diaries, however, and with less discussion of late medieval people of the coasts and sea. His
discussion of Yuge estate is an exception – it delves deeply into the lives of the residents of this small
island in Iyo province. By examining shipping manifests, records of dues being submitted to the
proprietor (Tōji temple) of the estate, and other related documents, he argues for a distinct maritime
culture on the island. Indeed, records show that parts of the estate were divided into areas for cooperative
net-fishing endeavors, as well as other areas used for making salt.
80
The pilots on Yuge Island, however,
were not afforded special consideration by Tōji, though records of appropriation of dues for their own
purposes suggests that the pilots felt entitled to certain benefits.
81
Such are some of the only hints in
English about the sociology of medieval maritime communities.
Japanese scholarship, much of it by Amino Yoshihiko as mentioned earlier, offers more insight.
Amino is renowned for his studies of marginalized aspects of Japanese history: both agrarian and “non-
agrarian” (hinōgyōmin 非農業民) commoners on medieval estates, sacred spaces and forgotten places,
and most importantly for this study, the role of the sea in shaping Japanese history. Even in recent years,
however, his research has been considered as tangential to what “should” be studied to understand Japan.

77
V. Dixon Morris, “Sakai: From Shōen to Port City,” in Japan in the Muromachi Age, ed. John W. Hall and
Takeshi Toyoda (University of California Press, 1977). In a chapter in the widely read Japan in the Muromachi Age
(ed. John Hall), Morris focuses more on issues of governance and merchant activities than the makeup of the
community.
78
Arne Kalland, Fishing Villages in Tokugawa, Japan (Honolulu: University of Hawai’i Press, 1995).
79
Shapinsky, Lords of the Sea, 42–48. Shapinsky’s monograph discusses the sociologies of the pirates or sea lords
in great detail. I will discuss his contributions to that topic more fully in the “transgressions” section of the seascape,
below, which focuses on pirates and their roles.
80
Ibid., 71–82.
81
Ibid., 79–80.

30

The best example of this attitude can be seen in a 1992 incident, when Amino produced a series of
videotapes entitled Japanese History as Viewed from the Sea for a national chain of Japanese college
preparatory schools. The school never used the series, explaining that they were afraid it would be
“counterproductive” to their students’ understanding of Japanese history, as measured by their scores on
college entrance exams.
82
This popular reluctance to embrace his approach notwithstanding, Amino
produced a number of volumes exploring historical aspects of maritime communities throughout Japan,
with a particular focus on the Japan Sea coastal area.
As groundbreaking as Amino’s work was, nevertheless, much of it explored maritime
communities through the lens of relations with the court, a connection that will be addressed more
thoroughly in the next section. For example, members of coastal communities were still discussed in
terms of their responsibilities to an estate’s proprietor, such as the amount of rents and dues submitted by
maritime estates or by networks created by shrine attendants who thereby brought maritime shipping
under their own purview. Amino was the first to question specifically what the contributions from
maritime estates were, however, and he focused on items such as salt and marine goods submitted as rents
and dues instead of rice. He also suggested that shrine affiliates may have doubled as shipping agents
themselves.
83
Still, his focus was often more on governance aspects of the medieval seascape than on an
exploration of a mariner’s perspective. Fortunately Amino does focus on the mechanics of saltmaking and
fishing in particular, which will be discussed in more detail in Chapter 2.
Some of Amino’s work shows how products of mariners’ labors made an impact on other aspects
of society. One such example is his discussion of abalone. This shellfish had been an important
commodity since at least the Heian period, as indicated by multiple references in the tenth-century
Engishiki, where it is described as a ritual or ceremonial offering. The twelfth-century Heiji Monogatari
also contains a scene in which abalone was presented as a gift to the gods. Since abalone was associated
with the gods and gift-giving, Amino suggests that stories such as these indicate the origin of the practice

82
Amino et al., Rethinking Japanese History, xxxi. Translator’s introduction.
83
These are common themes in many of Amino’s books and articles. See Umi no kuni no chūsei, Chūsei
hinōgyōmin to tennō, and his contributions to Nihon gijitsu no shakaishi Vol 2: Engyō, gyogyō in particular.

31

of drying strips of abalone (known as noshi 熨斗), to be given as gifts or offerings to Ise Shrine. For
instance, in 1458 a document (mōshitsugiki 申次記) recording items presented to the Ashikaga shōgun
shows that he received one thousand strips of abalone on the first days of the second, seventh, and twelfth
months.
84
It is clear that the shellfish was valued at the highest levels of society. While few scholars look
specifically at daily lives of coastal communities, research like this demonstrates how maritime endeavors
affected those who were completely uninvolved with the sea.
Given the historiographical “seascape gap” in terms of medieval sociologies, then, what we know
is an odd hodgepodge. Relatively little scholarship is devoted to understanding the people who actually
lived in the maritime communities. Through Amino’s work we understand more of their role as procurers
of marine goods for the court and other central elites, as well as some of the mechanics of saltmaking and
fishing. The importance of these goods to the center is apparent, yet we still understand little about the
sailors who transported the goods, or the men and women who fished or burned salt, or even how
authorities perceived and interacted with those groups. We also know little about the rituals and practices
of the members of the seaside communities, or how they ordered their own daily lives. Medieval
sociologies of the seascape, then, leave much to be explored. Chapter 2 below will explore more about the
lives of these seaside community residents and their roles in medieval ports, while demonstrating how the
type of port led to the development of different sub-communities. Furthermore, through some of the
extant documents including sets of maritime codes, Chapter 4 will show how increasing maritime trade
gave rise to a specific seafaring culture, with its own values and practices. The development of ports and
trade gave rise to more than just a particular maritime culture, however. As the next section will show, it
also warrants attention for its importance in economic development and the exercise of authority.

84
Amino Yoshihiko and Inaba Nobumichi, Kaimin no shakai, vol. 10, Amino Yoshihiko chosaku shū (Tokyo:
Iwanami Shoten, 2007), 225–227. Amino does not record who donated the abalone.

32

Medieval Governance: Control and the Economy

Control of maritime communities and ports was particularly important for economic benefits, and
governance is a key question: who controls the port and can take advantage of trade often has a direct
correlation to the development of local power and influence. As Western scholars of Japan have focused
on governmental structure and its changes over the centuries, the historiographical “seascape gap” comes
closest to being bridged in studies of governance and the economy. That fact notwithstanding, detailed
studies focusing on maritime conditions are still rare. John Whitney Hall’s groundbreaking Government
and Local Power in Japan, 500 to 1700, the first English-language study of one of the outer regions in
premodern Japan, reflects this gap. Hall focuses on eastern Kibi (later, Bizen province; modern-day
Okayama), which borders the Seto Inland Sea and includes a number of islands offshore. While he
thoroughly details hierarchies of land ownership and rights to cultivation evidenced in the records, there is
virtually no mention of the importance of overseeing port facilities for shipping cultivated goods from
place to place. Even a discussion of the court noble turned pirate Fujiwara no Sumitomo is largely limited
to making the point that during his rebellion Sumitomo killed the deputy governor of Bizen. His
development of naval power goes undiscussed.
85
Hall does note the import of the port of Onomichi in his
discussion of the transport of rents and dues from Ōta Estate (Bingo province) to the capital, and he
suggests that a similar system probably existed in Bizen. But the analysis ends there.
86
Control of the sea
is not a factor in economic and political power in his study; it is land that counts.
As we have already noted, scholarship on maritime trade expands when the focus is medieval
interactions with the mainland. Bruce Batten and Von Verschuer deal with these issues in depth, but again
they focus on higher-level government interactions. Batten’s studies of Hakata demonstrate the role of
that port city in diplomatic and trade ties. While Hakata was recognized as the gateway between Japan
and the mainland, Japanese authorities consciously laid out facilities in the area to simultaneously
welcome and control foreign interaction. In classical times, a guest facility, the Kōrōkan, housed foreign

85
John W. Hall, Government and Local Power in Japan, 500 to 1700: A Study Based on Bizen Province (Michigan:
University of Michigan Press, 1999), 129.
86
Ibid., 178.

33

diplomats with sufficient grandeur for their station, but it was located far enough away from the
government headquarters at the Dazaifu to prevent spying or other incursions.
87
During the Heian period,
trade was highly regulated, and at one point ships were allowed to come only at two-year intervals. Trade
was conducted through the Kōrōkan, overseen by Dazaifu authorities.
88
Such examples of the conscious
exertion of power to regulate both trade ships and the physical location of trade and diplomacy
demonstrate how highly authorities valued control of those resources.
Von Verschuer explores the role that the provinces might have played in a power struggle among
domestic political factions within Japan and the resulting ramifications on maritime trade. For instance, in
the early fifteenth century, Korea authorized trade with several Japanese shipping families in addition to
formal trade with the Ashikaga shogunate. The Sō family of Tsushima, capitalizing on that island’s
geographic proximity to the peninsula, actually became a vassal of Korea and so gained the legitimate
authority to dispatch permits for commercial trade between Tsushima and Korea.
89
Authorization to deal
directly with the Korean officials weakened the hold the Ashikaga shogunate had on that trade route. A
similar scenario unfolded with the Ōuchi family as early as 1398, when Ashikaga Yoshimitsu requested
that the Ōuchi, headquartered in Suō province (modern Yamaguchi prefecture), forward a request to the
Choson court for a copy of the Korean Tripitaka, a set of Buddhist sutras.
90
Despite a series of ups and
downs in the meantime, the Ōuchi pressed their advantage in 1468. Officially accompanying a shogunal
trade mission, the Ōuchi family not only seized part of the shogun’s cargo on its return, but also
commandeered the tallies – official trade permits presented by the Chinese court to the shogunate – for
future trade missions. This crippled immediate shogunal attempts to continue direct trade with China,
though they soon used Korean intermediaries to explain the situation to the Chinese court.
91
Through such

87
Batten, Gateway to Japan, 67.
88
Ibid., 109–114.
89
Von Verschuer, Across the Perilous Sea, 134.
90
Charlotte von Verschuer, “Ashikaga Yoshimitsu’s Foreign Policy 1398 to 1408 A.D.: A Translation from ‘Zenrin
Kokuhōki,’ the Cambridge Manuscript,” Monumenta Nipponica 62, no. 3 (October 1, 2007): 271–2, 279-80.
91
Von Verschuer, Across the Perilous Sea, 131.

34

examples we see the effects on central governance when coastal residents along a significant trade route
exercised or expanded their own power.
Like Batten, in his study of the port at Obama Soranaka Isao notes that ships from China and
Korea trafficked the Obama area and were an important source of trade. Unlike in the power struggles in
Kyushu, however, Soranaka highlights Obama’s strategic importance during the Ōnin war (1467-1477),
when it was under the control of the powerful Takeda warlord family. Takeda allies would ship rice and
other provisions through the port to their forces in Kyoto, heightening Obama’s military importance for
provisioning. As the war continued and other regular shipping routes were disrupted, Obama, relatively
untouched by battle, became a safe port and therefore shipping volume grew.
92
It is clear then that
command of ships and shipping, seen in the examples of the Ōuchi domains, of Hakata, and of Obama,
was an important aspect of medieval governance.
Amino Yoshihiko, whose importance we have already noted, was the first Japanese scholar to
examine in detail maritime influences on power relations, land reform structures, and state formation.
Even his research, however, was originally inspired by the drive to understand more about land-based
concerns. In one case concerning the Noto peninsula, for instance, Amino recounts the apparent paradox
of large, relatively opulent houses in a “poor” agricultural area. His subsequent exploration of that
discovery revealed that although seventy percent of the Kamakura-era port districts in the Oku-Noto
region were classified as atamaburi (頭振 “poor farmer” districts), there were nevertheless a number of
affluent residents. These were usually merchants or shippers involved in maritime trade. Since they held
little or no land, and since district records noted only land holdings, district leaders classified the region as
“poor.”
93
Wealth obtained from maritime occupations was not considered. For the first time, Amino’s
examination of maritime-related occupations provided a different approach to understanding the actual
economic dynamics of hinterlands.

92
Soranaka, “Obama,” 90–92.
93
Amino Yoshihiko and Mori Kōichi, Uma/Fune/Jōmin: Tōsai kōryū no nihon rettō shi (Tokyo: Kodansha, 1999),
98.

35

Yet Amino still generally regarded these places through the lens of relations with central powers
– his explorations of fishing and saltmaking communities approach the topic from the viewpoint of their
proprietors. For example, the Kamo Shrines in Kyoto were affiliated with the Adogawa mikuriya ( 御厨 a
tribute land for a shrine) in Ōmi province as early as the tenth century. These lands were responsible for
producing marine goods for the shrine, and Amino presents a thorough history of how Kamo Shrine
grasped this and other tribute lands along Lake Biwa, the Kii Peninsula waterways, and through the Seto
Inland Sea into the Bungo strait. He discusses how shrine affiliates were free to move between these areas
to obtain fish, shellfish, and seaweed for the shrine.
94
As the mikuriya tribute lands eventually became
virtually indistinguishable from shōen (estate) lands, Amino highlights how these coastal communities
also became loci for shipping rents and dues to their proprietors.
95
His overall focus, then, is to
demonstrate the importance of maritime communities to the central powers. Certainly this is due in great
part to the fact that most extant written records refer to concerns addressed by the central court or other
estate proprietors. While his contributions cannot be overlooked, as prior to his studies few scholars
realized the impact of non-agrarian settlements on central authority control, the “seascape gap” is still
present in the records’ primary concern with top-down relationships. By examining more of governance
and economic issues at the community level, including the use of the archaeological record, it should be
possible to see more clearly the impact that these communities had on each other’s economic growth.
Shinjō Tsunezō, also mentioned earlier, has conducted perhaps the most comprehensive study of
maritime governance concerning estate shipping and trade routes. A key aspect of his work is his
exploration of the checkpoint (sekisho 関所) system.
96
It was an important aspect of travel and shipping,
as goods passing through the checkpoints were to be assessed a toll, and those fees were then given to the
overseeing institution or proprietor. For example, if an estate whose proprietor was Kōfukuji Temple sent
goods that passed through the northern Hyōgo checkpoint, those items were subject to tolls that went to

94
Amino Yoshihiko, “Kodai-chūsei-kinsei shoki no gyorō to kaisanbutsu no ryūtsū,” in Koza nihon gijutsu no
shakaishi. Dai 2 kan: Engyō gyogyō, ed. Nagahara Keiji and Yamaguchi Keiji (Tokyo: Nihon Hyōronsha, 1985),
220–227.
95
Amino, Umi no kuni no chūsei, 148–149.
96
Shinjō, Chūsei suiunshi no kenkyū, 559–752.

36

Tōdaiji Temple, administrator of the checkpoint. As such, control of the checkpoints – and exemptions for
some users – were critical concerns for those both high and low.
As we have seen, scholars working in English have examined port cities such as Hakata and
Obama and their roles in overseas trade. Many of the principles of controlling port towns and trade routes
can be applied to the smaller scale domestic trade routes, but they have yet to be examined in depth. The
Japanese scholarship is more comprehensive, but it still has lacunae. For example, Amino’s work focuses
extensively on maritime communities and their role in central governance, without looking closely at
administration on a local level or examining the economic interrelationships of those communities
themselves. Even Shinjō’s study lacks an intensive exploration of lateral ties between estates. Chapters 2
and 3 will address these ties and shed more light on late medieval maritime governance and economic
growth as seen through the control of ports and checkpoints in strategic locations.

Medieval Transgressions: Pirates and Sea Lords

Piracy is the one aspect of seascapes that has been the subject of a fair amount of scholarly interest. Like
many other aspects of medieval maritime scholarship, however, some of the earliest English-language
work focuses on piracy not necessarily within Japan, but between Japan and the mainland. Japanese
Piracy in Ming China: 1368 – 1644 by So Kwan-wai examines instances of piracy along the China coast.
Though the book’s very title spotlights the Japanese role, the focus of the book is actually on China. So
claims to place the wo-k’ou (the Chinese reading of the characters generally referring to “Japanese
pirates”) in the context of Chinese history, yet his becomes more a tale of Chinese pirates. Though the
wo-k’ou are often referred to in Chinese documents and though the term supposedly specifically refers to
“Japanese” seamen, So uncovers evidence of extensive Chinese participation in piratical activities. There
is a brief discussion of ways in which the Japanese turned pirate, including the possibility that Chinese
ringleaders living in the Japanese islands encouraged Japanese participation. This description, however, is
followed by the claim that the majority of those pirates were actually Chinese who were possibly using

37

Japanese names or who had hiredJapanese mercenaries to crew their ships.
97
The rest of the volume is
devoted mainly to factors within China that influenced piracy, including waxing and waning government
intervention, the availability of desirable trade goods, and geographical factors that facilitated both piracy
itself and later cooperative mainland efforts to work together against the pirates. The title of So’s
monograph belies his argument: the majority of “Japanese” pirates in the Ming period were actually
Chinese.
Peter Shapinsky’s book, Lords of the Sea: Pirates, Violence, and Commerce in Late Medieval
Japan, places the focus squarely back on Japanese piracy while also arguing that the pirates were not
always the lawless bands that the Western mind imagines. Shapinsky examines interactions mostly within
the Seto Inland Sea area, paying particular attention to classification systems and noting that many so-
called “pirates” were actually what should be called “sea lords”
98
who governed their territories much as
did other terrestrial lords of the time. The focus of his research is on political interaction and patronage,
including an examination of the dynamics in the sixteenth century between Hideyoshi and Inland Sea
maritime lords. Deconstructing the evolution of the term kaizoku ( 海賊), Shapinsky compares the role of
the “sea lord” with that of a landholding lord. Though the word kaizoku is usually translated as “pirate,”
in fact in late medieval Japan the term was used to describe both legal and illegal maritime activities.
Kaizoku could refer, for instance, to a seafarer attacking another ship illegally, or it could denote a
seafarer legally contracted as an escort or protector for another ship. Much of Shapinsky’s book focuses
on incorporating sea territory into land-based estates, as well as the fluctuation of illegal and legal
practices in sea-based domains. Throughout the medieval period, particularly with the decentralization of
power in the later Kamakura era (early fourteenth century), the sea lords’ authority grew and they came to

97
So Kwan-wai, Japanese Piracy in Ming China During the 16th Century (East Lansing: Michigan State University
Press, 1975), 27.
98
Shapinsky, Lords of the Sea, 13–14. Specifically, Shapinsky defines sea lords as leaders of seafaring bands who
may have been considered pirates by mainland powers, but who likely perceived themselves as rulers of their
maritime-based domains. One such leader, Noshima Murakami Takeyoshi (1533 – 1604), used the title “Yamato no
kami” (governor of Yamato province) or the clan name “Minamoto” in correspondence, calling to mind a land-based
administrative legitimacy. In practice, such sea lords controlled access to maritime space and the means of maritime-
based production.

38

administer their own territories. In the case of Yugeshima Estate, for example, Tōji temple was the
estate’s supreme proprietor but gradually relied more and more upon the Kurushima Murakami sea lords
to protect and manage the island.
99
Shapinsky also notes that in the sixteenth century, the Kōno provincial
governor family in Iyo province licensed several sea lord families as administrators of its territory.
Eventually those families received allotments of lumber, wheat, and marine goods procured locally, and
even recruited residents to participate in collecting protection fees from ships passing through the
region.
100
These activities suggest the sea lords’ growing local influence in the later medieval period.
Shapinsky further examines the sea lords’ administrative techniques, including marital and
military alliances, and instances of local residents inserting themselves into land-based administrative
infrastructures. For example, a legal contract between the sea lord Murakami Takeyoshi (ca. 1533 – 1604)
and the terrestrial daimyō, Mōri Motonari (1497 – 1571) came as the culmination of Murakami’s petition
to Mōri requesting title to maritime locations. The language of the petition and subsequent loyalty oath
that Murakami swore to Mōri was largely couched in long-established terrestrial terms, including a title
(kamon no kami 掃部頭, director of the Office of Housekeeping) that derived from the classical ritsuryō
administrative code. On the other hand, Murakami included Mishima Daimyōjin, a sea-based deity, in the
list of gods witnessing his oath, together with the terrestrial-based Hachiman Daibosatsu and
Tanmandaijizai Tenjin.
101
For Shapinsky, then, the Murakami were not actual pirates in the illicit sense,
since they worked legitimately within the terrestrial power structure. But they did insert their mariner’s
perspective into that structure. In Wigen’s “seascape” terms, then, Shapinsky’s work shows the gradual
blurring of lines between “transgression” and “governance” in the later medieval period.
Japanese scholars have also examined piracy, particularly in the Inland Sea region. Amino
incorporates piracy into his study of sea peoples, suggesting that pirate bands evolved from the shrine
purveyors mentioned above as they expanded their rights to “procure” items from locals.
102
Sanagi Nobuo

99
Ibid., 92.
100
Ibid., 95–96.
101
Ibid., 115–7.
102
Amino, Nihon chūsei no hinogyōmin to tennō, 245–6.

39

examined piracy in the Shiwaku island chain of Sanuki province. Like Shapinsky, he draws attention to
the blurry lines between transgressions and governance, noting that kaizokusen ( 海賊船) were indeed
pirate ships but that they might also have been involved in naval defenses and shipping activities.
103

The primary contemporary scholar of Inland Sea piracy is Yamauchi Yuzuru. He particularly
notes the geography of the Inland Sea islands and the establishment of sea castles – that is, pirate enclaves.
By examining both documentary and archaeological evidence, he extrapolates the types of interactions
pirate clans may have had with their land-based neighbors.
104
Though he frames his studies within the
bigger frame of pirate clans’ relationship with the central authorities, Yamauchi’s work is of particular
interest since he also attempts to understand daily life in pirate enclaves. From his interpreting the types
of artifacts found on castle sites in order to determine common trade goods and luxury items, to his
elaborating concerns of how to procure drinking water and tie off vessels, Yamauchi is one of the few
maritime scholars who has examined the seascape at the community level.
Piracy, then, has been relatively well addressed within seascape studies. Scholars generally agree
that the kaizoku were not necessarily “pirates” in the Western sense, but that they often occupied a quasi-
legal status. Although there were warnings about pirates on the seas, often “sea lords” were actually
legitimate actors in maritime shipping and travel, functioning as (sometimes forcefully) protective escorts
on the water. Yamauchi’s work even highlights sea lords as a maritime community in their own right. In
the following chapters, I will argue that pirates took advantage of geographically undesirable port sites to
establish their strongholds, and that their presence changed the course of certain trade routes and seafaring
customs.

From Seascape to Maritime Cultural Landscape

The concept of a seascape as outlined above is helpful for gaining a long-term view of the socioeconomic
development of coastal communities over a wide geographic area. I have examined the premodern

103
Sanagi Nobuo, Setonaikai ni okeru shiwaku kaizokushi (Tokyo: Kyozaikenkyusha, 1934), 133.
104
Yamauchi Yuzuru, Kaizoku to umijirō : Setouchi no sengokushi (Tokyo: Heibonsha, 1997).

40

Japanese seascape by considering the infrastructure of coastal communities through documents and
literature highlighting the role of the residents and seafarers as well as their relationships to the court and
central elites, and through evidence concerning those who challenged authority. The next line of inquiry
seeks to answer the question of why the seascape developed as it did, particularly in a specific area.
William Cronon has noted that environmental historians’ greatest challenge is to integrate three levels of
analysis: the dynamics of the natural ecosystem, the political economies of those ecosystems, and the
beliefs and perceptions about both.
105
Through using the dual concepts of the “maritime cultural
landscape” and the “mariner’s perspective,” it is possible to understand how nature, economic
development, and cultural practices were related in the medieval Japanese Seto Inland Sea region.
The idea of the “maritime cultural landscape” was introduced by maritime archaeologist Christer
Westerdahl in 1992. He describes it as “human utilization (economy) of maritime space by boat:
settlement, fishing, hunting, shipping, and its attendant subcultures.”
106
Since the environment shapes how
people approached and used the landscape, the environment itself becomes an active player in historical
development. As people adapted to maritime surroundings, they created a particular infrastructure –
constructs of the seascape – to best take advantage of the resources available in maritime locales. These
could include pier and wharf infrastructure, shipbuilding facilities, salt-making fields, and navigational
markers. Examining patterns and trends in maritime infrastructure provides clues to how attuned the local
inhabitants were to the sea, and how much they made use of both sea and shore.
Taking Westerdahl’s ideas one step further, the environment also shapes how people use both
land and sea. His ideas also demonstrate the difficulty of looking at Wigen’s four seascape categories
separately, since each aspect affects the others. The presence or absence of natural resources influences
the desirability of an area to live in, govern, or exploit. In this way, the environment shapes governance of
a place as well as lifestyles (in Wigen’s terms, the sociologies) of the people who live there.

105
William Cronon, “Modes of Prophecy and Production: Placing Nature in History,” The Journal of American
History 76, no. 4 (March 1, 1990): 1123.
106
Christer Westerdahl, “The Maritime Cultural Landscape,” The International Journal of Nautical Archaeology 21,
no. 1 (1992): 5.

41

In the Inland Sea, certain natural “chokepoints” (narrow shipping channels or key points in the
seafaring infrastructure) became vital for seafarers as safe stopping points, or for authorities as places to
control the inward and outward flow of people and goods. Historically, the availability of salt, marine
goods, lumber, and other commodities played a part in determining which maritime communities and
economies would thrive. This in turn made locations subject to control in one form or another. Such
control could assume many forms, whether it was direct control such as when Kamo Shrine attendants
procured goods directly from a tribute land or estate; or whether it was indirect control such as when fees
were levied on ships passing through checkpoints. As residents improved technology and techniques, they
were further able to shape their own environment. The creation of higher-producing salt fields or the
construction of a better infrastructure for larger boats are examples of manmade influence on the local
environment as well as manifestations of the maritime cultural landscape.
Piratical activity is another manifestation of environmental influence. The Inland Sea is dotted
with small islands, fierce currents, whirlpools, and navigation hazards. Only very skilled mariners were
able to maneuver through some of the shoals and rocky shallows, in places battling currents of ten or
more knots. These small islands with their natural barriers became the perfect locations for pirates or sea
lords to evade other authorities. They were limited by those same surroundings, particularly with regard
to procuring fresh water; and they adapted their infrastructure accordingly, building relatively small “sea
castles” and having only semi-permanent moorings to prevent others from landing on their shores.
107

Mastery of their environment was especially critical for the sea lords, and the environment helped deter
others from interfering with the “pirate” clans.
A complementary concept, that of the “mariner’s perspective” as put forth by Paul Rainbird, is
more challenging to define, but it is still an important aspect of maritime life. Rainbird reminds us that
people associated with the sea – not just seafarers, but residents of coastal communities – have particular
practices and outlooks unique to those with close, prolonged contact with the sea. In other words, such

107
Yamauchi, Kaizoku to umijiro, 24.

42

people have a distinct maritime identity.
108
The mariner’s perspective is evident in such situations as
pilots being valued as high-ranking members of their local society (while being disdained by the courtiers
they ferried), or seafarers coming together to create laws of the sea that land-based authorities did not
consider. The mariners’ perspective is even evident in language, manifested in the various terms for
harbors, ports, and coves (tsu 津, ura 浦, minato 港, and the like) that represent distinctions in the
topography important to seafarers.
109
As the mariner’s perspective asserted itself more forcefully, littoral
residents actively shaped their physical and social environments, be it through creating additional
infrastructure, becoming active in local governance and trade, or by turning to more illicit acts. Thus the
mariner’s perspective is another factor that both shapes and is shaped by the maritime cultural landscape.

Methodologies: An Interdisciplinary Approach

By examining the maritime cultural landscape and mariner’s perspective, the influence of the sea in
shaping Japanese history has become clearer. This study will treat the geography of the Inland Sea region
as an active vector in medieval socioeconomic development, since the sea serves as a barrier in some
ways but as a connector in others. A multifaceted approach provides the richest picture of how geography
affects the physical infrastructure, aspects of production and trade, and cultural developments. This
section will introduce the main methodologies used in this study, including analysis of documentary and
archaeological evidence, literary sources, and geospatial analysis, to understand the development of the
maritime cultural landscape on land and at sea.
Although the central documents for this project will be discussed in depth in the following
chapters, I should note here that several of the major resources include a variety of legal records, such as
the Register of Incoming Ships at the Hyōgo North Gate (Hyōgo Kitaseki Irifune Nōchō; below, Register

108
Rainbird, The Archaeology of Islands, 49.
109
Ichimura Takao explores the question of the difference between a harbor (tsu 津) and a port (minato/kō 港). Tsu
were often located at river mouths and often served the function of a port as well. As time progressed and the
location began to function more as a port or simpler harbor, the appropriate suffix (tsu or kō) was appropriated as
part of the location’s name. In general, ports usually had easier access to inland areas via roads or other rivers. See
Takao Ichimura, “Chūsei kōki no tsu/minato to chiiki shakai,” in Tsu, tomari, shuku, ed. Amino Yoshihiko, Shohan
(Tokyo: Shin Jinbutsu Ōraisha, 1996), 96–105.

43

兵庫北関入船納帳) and several versions of maritime codes (kaisen shikimoku). Geographic connections
can be found throughout all of these types of documents. As I have suggested, many of the estate and
legal documents have been studied up to the present – particularly in the body of Amino’s work –with an
eye to understanding more about the relationships between particular locales and central authorities. By
looking at these documents in conjunction with other evidence, it is possible to learn more about the
lifestyles of actual residents of maritime communities. In understanding more about the roles of sailors,
pilots, fisherpeople, and saltmakers in their own communities, as well as the physical layout of the port
towns themselves, we can clarify more of the sociologies of the medieval Inland Sea.
Very few scholars in Japan and almost no one in the West have seriously studied Japan’s
medieval maritime codes. Farris’s article on shipbuilding technologies and Shapinsky’s book note the
existence of such codes, but they do not explore them in depth. In contrast, Kubota Hiroshi has compared
all extant versions of the codes, from premodern to early modern copies, and compiled a detailed list of
linguistic variants and possible interpretations of the individual terminology of the codes.
110
While
Kubota’s study is an extremely helpful resource, it does not provide a deeper analysis of the content of the
codes and ramifications for seafarers. But it is my view that such regulations provide clear information
about the types of concerns held by seafarers, from right of way on the seas to provisions needed for
seafaring, to financial concerns for payment of the crew, shares of profits on the cargo, and
reimbursement in case of disaster. These issues provide great insight into the mariner’s perspective.
Several regional variants exist, suggesting deviations from the norm that responded to local customs, to
governing concerns, or even to climate and ocean conditions – the impact of geography was critically
important to the mariner’s perspective.
We will see too that literary sources can provide important information, particularly with regard
to seafaring customs and littoral residents. As noted above, they must be used with some caution, with
consideration of a writer’s motivations. Unlike in the West, most diaries in premodern Japan were written

110
Kubota Hiroshi, Kaisen taihoko: Sumida Shōichi hakushi kaisen shikimoku no kenkyū shūi. (Yao: Osaka Keizai
Hoka Daigaku Shuppanbu, 1989).

44

with the assumption that they would at some point be made public material. Oftentimes, as was the case
in Ki no Tsurayuki’s work, travel diaries were written as frames for poems that would highlight the
author’s skill. One translator of medieval diaries has gone so far as to suggest that “diaries provide little
realistic place description and refer more to the myths, legends and history, as well as to past poets and
poems connected with such ‘famous’ places."
111
While it is certainly true that many of the poems and
tales recorded in such works make use of commonly accepted literary tropes, to dismiss the descriptions
unilaterally seems shortsighted. Many of the medieval diaries were travel accounts, recounting the
author’s journey through different geographical areas. By taking into account the author’s potential biases
and background, as well as comparing the information therein to that of other diaries, we can often
determine which tales were unique articulations of the author’s own experiences. It is therefore possible
to glean information about people and places along the traveler’s routes. Most of the descriptions were
written by outsiders from the upper echelons of society rather than by members of maritime communities,
so they provide an external and sometimes exoticized view. Though these accounts must be treated as
such (for instance, Sei Shōnagon’s contempt for the lowly sailor), they still provide insight into common
perceptions of maritime communities, to understand aspects of the sociologies and infrastructure of
maritime communities.
Elites were more likely to leave written records; indeed, non-elites mostly did not have the luxury
of education, time, and paper to write down their thoughts and practices.
112
Incorporating evidence from
the archaeological record, however, can help shed further light on the lifestyles of those whose voices
have been lost over the centuries. Japanese archaeological reports meticulously detail the range of
artifacts found at any given site, and from these studies certain conclusions can be drawn about foodways,
occupations, infrastructure, technology, and the exchange of goods. One drawback is that not all

111
Plutschow, Four Japanese Travel Diaries of the Middle Ages, 3.
112
William Wayne Farris, Japan’s Medieval Population: Famine, Fertility, and Warfare in a Transformative Age,
(University of Hawaii Press, 2009), 159, 237. Farris suggests that “local notables and some wealthy peasants”
probably achieved literacy sometime after 1300, but that village children probably were not afforded the opportunity
for an education until the 1500s. Medieval ships’ captains may have had the capability to record information
pertinent to their ships or cargo, though no such documents remain. It is unlikely that common sailors or members of
seafaring communities were literate during this time.

45

archaeological sites are created (or documented) equally. Very rarely do scholars of historical eras find
that their research questions have been the foci of systematic excavation.
113
The site reports that exist,
while detailed, are frequently a result of cultural resource management projects undertaken before a local
authority builds a new road or building today. Archaeologists complete an (often hurried) excavation to
assess the cultural heritage present on the site, subsequently producing a report detailing the location,
excavation techniques, artifacts or formations found, and placing the site into a historical context. It is
thus largely a matter of luck whether a site is excavated at all. This overall lack of research-driven
archaeological excavation means that to date there have been no systematic surveys of, for example,
littoral communities along the Shiwaku island chain in the Inland Sea that might help guide a study of
saltmaking facilities. If in the past several decades the local authorities have built a new pier or coastal
road, however, then the archaeological reports produced for those projects may have uncovered evidence
of a premodern salt field. The types of information gathered rely on the scattering of sites that have
produced reports. Though the artifacts may provide valuable information about the processes of
saltmaking, fishing, or other concerns, sweeping generalizations about maritime communities are difficult
to draw from archaeology alone.

GIS Methodology

A critical methodology used to collate the various types of information from documents, literary sources,
and archaeology, while placing geography in the forefront of the analysis, is the use of Geographic
Information Systems (GIS) software. GIS is defined as “an information science that focuses on the

113
Japanese archaeology in universities and research centers today is heavily weighted towards the pre- and proto-
historic eras. Many research questions are formulated and explored for issues such as early human migration, burial
rituals, settlement patterns, and trade routes, and as such prehistoric archaeology has a robust theoretical and
practical body of research to draw from. For unclear reasons, very little historical archaeology is being
systematically undertaken under the auspices of research institutes. Although I could not verify the truth of this
statement, one archaeologist teaching at a university in Tokyo even calmly informed me that there were no historical
archaeologists employed as full-time faculty members at any higher education institution in Japan.

46

collection, modeling, management, display, and interpretation of geographic information.”
114
While GIS
has been used extensively in recent years for tasks such as city planning, environmental management, and
even analysis of crime patterns, its application in historical analysis is a relatively recent trend. Yet it can
be a powerful tool even for historians when processing different types of information simultaneously. GIS
allows for both separate and overlaid geographic displays of different categories of information that
highlight the geographical aspects of a research question.
115

As will be demonstrated in the following chapters, geography-based information such as written
records denoting the ports of origin for ships, goods, and people; archaeological evidence of trade goods
or maritime-related tools; and place-specific information from travel diaries can all be collated easily
using this software. The resulting maps help reveal a history that cannot be known from one source of
information alone. For example, by approaching the Register of Incoming Ships at the Hyōgo North Gate
holistically through a geography-based analysis, it becomes possible to understand more about the lateral
flow of trade goods between ports, thus avoiding the tendency to focus on center-periphery and top-down
relationships. The use of GIS also highlights the role natural factor endowments played in commodity
production. Furthermore, incorporating the archaeological evidence of trade – including pottery, iron
byproducts, and fishing gear – reveals additional dimensions of Inland Sea trade. For instance, based
solely on the mid-fifteenth century Register, one could conclude that Bizen pottery was produced in a
single location and shipped only to Hyōgo. Overlaying archaeological information disproves that
hypothesis, however, showing that in reality Bizen ware was transshipped through certain areas,
suggesting a thriving local market; or it was also shipped directly to other sites for immediate use.
Chapter 3 explores the different trade routes traceable through the use of GIS analysis of the Register and
archaeological evidence.

114
Karen Kemp, “Geographic Information Science for Spatial Analysis and the Humanities,” in The Spatial
Humanities: GIS and the Future of Humanities Scholarship, ed. David J. Bodenhamer, John Corrigan, and Trevor M
Harris (Bloomington: Indiana University Press, 2010), 31.
115
Ian N. Gregory and Paul S. Ell, Historical GIS: Technologies, Methodologies, and Scholarship (Cambridge; New
York: Cambridge University Press, 2007), 3.

47

By expanding the analysis beyond cargoes alone to the people named as ships’ captains and
warehouse managers, significant connections between both adjacent and far-flung ports also become
evident. Many of the ships’ captains have common names that are nearly impossible to trace through
other written records, and as a consequence most scholars have cited lack of corroborating evidence as a
reason to forego a detailed analysis of the individual captains. Through the use of GIS to highlight the
captains’ geographic affiliations, it becomes possible to pinpoint particular individuals and reveal
connections with other captains. The results of this analysis, discussed in more detail in Chapter 4,
demonstrate a high level of interactions between ships’ captains but with geographic variants in
collaborative practices. In short, GIS analysis becomes a powerful tool for assessing influences on the
production and trade of commodities and patterns of interaction throughout the medieval Inland Sea.
Maps can provide important clues to the patterns of the flow of both goods and people in
medieval Japan, but by themselves they cannot necessarily explain that history.
116
GIS is an extremely
effective tool for visualizing the geographic aspects of a question and for focusing on the geographic and
environmental influences on development. Nevertheless the resulting maps must be placed within a
greater historical context, as I will do throughout.

Conclusion

The following chapters will examine the maritime cultural landscape, including aspects of the mariner’s
perspective, in medieval Japan’s Seto Inland Sea. Coastal communities there naturally created a maritime-
based landscape that responded to the geography of the area, and they made use of its surroundings to
take best advantage of the resources available. The sea shaped the social and economic development of
those communities, and in turn those communities shaped the region in which they lived.
Chapter 2 will discuss the development of port communities that created the most efficient
transport system for the various cargoes shipped through the Inland Sea. Ship construction and port

116
Ibid., 90. The authors caution the user about the GIS trap of becoming too enthralled with the different types of
map displays the software affords and losing track of the actual research question. GIS is a tool for researchers, not
an end in itself.

48

infrastructure form an important part of this development, but so do more intangible aspects such as
governance of those areas and trade routes. Checkpoints, systems of taxation and exchange, and a social
hierarchy developed to control the region. This chapter will also explore what we know of the people of
the maritime community, including those involved in procurement of marine goods such as fishermen and
saltmakers; and people involved in shipping and exchange such as sailors and merchant houses, women,
and pirates.
The port communities provided the overall infrastructure necessary to facilitate a thriving
maritime trade. Chapter 3 will focus on the flow of trade goods and economic development within that
maritime network. By tracing the patterns of trade throughout the Inland Sea, it becomes possible to
pinpoint local transshipment hubs. Often these locations are not the same as the major ports that were
used for the China trade, and they have often been forgotten in historical studies of maritime trade. Yet
through determining how local trade worked and which regions held particular local or regional
prominence, our common preconceptions of which were important shipping ports undergo a shift.
Examining the types and volumes of cargoes shipped also provides important information about the
mechanics of maritime trade and the factors that influenced local economic development, including
geography, seasonality, the creation of “brand names,” and the role of personal power in trade.
As maritime trade expanded and thrived, so too did the need for sailors to carry out that trade.
Chapter 4 therefore concentrates on a thorough examination of shipboard life, including seafaring
practices while at sea and in port. By looking at documents written by and for seafarers, the problems
mariners grappled with and the solutions they proposed become clear. They responded directly to the
needs of their profession and the particular environment of maritime shipping. The very existence of
codes for seafarers suggests a particular “mariner’s perspective” that gave rise to the regulations.
Combining information from the codes with a study of geographical ties between ships’ captains in the
Register from Hyōgo reveals patterns of collaboration in shipping practices. This suggests a wider
maritime community far beyond a single port town, and provides evidence of how maritime geography
and experiences shaped customs and culture.

49

The concluding chapter will return to the concept of a seascape to determine the broader import
of the development of the maritime cultural landscape. The concept of the seascape is woven through the
entire project, from issues raised by discussing constructs, sociologies, governance, and transgressions. In
revisiting the medieval seascape after a thorough examination of the maritime cultural landscape, the
impact of the unique geography of the Inland Sea on the development of Japan’s medieval economic and
social development will be apparent.

50

Chapter 2: The Ships and Ports of the Inland Sea: Shipping, Production, and
Piracy

This chapter will examine different aspects of the seascape as embodied in the port town or coastal
community, and in the ships that plied medieval waters. I argue here that the geography and environment
are key aspects determining port specialization. Locals responded to the natural factor endowments of the
region, resulting in the creation of a maritime-centric culture and strategies for control of the ports. For
example, sites with deeper water and protected coves were naturally more able to accommodate larger
watercraft. Moreover, without an area conducive to landing ships, loading and unloading cargo, and
subsequently transporting that cargo overland, a town could not become a thriving center of maritime
commerce or production. Infrastructure alone did not characterize a port town, however. By the very
nature of its location next to a body of water, a maritime settlement engendered the development of a
particular maritime community that reflected the seascape sociology. Governance of a maritime region
often became a contested right, as control of a port or related facilities could become a direct source of
revenue or a strategic foothold for its overseer. Finally, piracy was certainly a concern on the Inland Sea,
and pirate enclaves had their own unique infrastructure to facilitate their activities. In examining the
aspects of the maritime community through the seascape lens, we see how seaside settlements, both licit
and illicit, had their own unique culture and concerns. That culture and those concerns reflected the
surrounding environment and in turn shaped it as well.
While a certain infrastructure must be put in place for a port to function in any way, specialization
in ports can generally be categorized as focusing on shipping, production, or illicit activities (pirate or sea
lord enclaves). Larger ports often became hubs of trade that would accommodate larger boats, while
smaller ports usually drew their importance from procurement of resources, often through saltmaking or
fishing, and did not necessarily require infrastructure for large-scale shipping vessels. The characteristics
of each of these types of ports were often quite different, and clearly reflected the influence of the
environment around them in their infrastructure, residents, and governance. This chapter will begin with

51

an introduction to medieval ship technologies and construction, as it is critical to first understand the
types of vessels that any port needed to accommodate.

Ship Technologies: Medieval Innovations

Ships are constructed in response to their environment and intended use. Shallow waters call for flat-
bottom vessels that will not scrape the seafloor, while rougher open seas require a distinct keel and stem
that aid in cutting through the waves. We begin the discussion of medieval maritime infrastructure, then,
with a discussion of the types of ships used in the Inland Sea. Without knowing the capabilities of the
ships, it is impossible to understand the distance they could sail, the number of men needed to operate
them, the amount of cargo they could carry, or their susceptibility to wrecking. All of those factors are
eventually reflected in port development as well. For example, if the sailing capability and seaworthiness
of most ships used along frequently traveled routes were limited, more ports might develop along those
routes, as ships would need to regularly moor at places where they could resupply or take shelter.
There are no extant lines drawings or construction plans for ships that predate the early modern
period; information about medieval ship construction must be gleaned from contemporary illustrations,
port records, and archaeological evidence. The issue is further complicated by the fact that much of the
existing scholarship on medieval ship construction is far more concerned with ships used for the China
trade than with domestic travel. Ishii Kenji, in his comprehensive treatment of Japanese shipbuilding,
focuses almost exclusively on the envoy ships used for the China trade (kenminsen) in his introduction of
Muromachi-era vessels, and he overlooks most domestic watercraft.
117
Innovations that enabled long-
distance travel were likely adopted for domestic use, but the pace of change is unclear.
William Wayne Farris, in his article “Shipbuilding in Japanese Maritime History,” calls the
period from 900 – 1300 a “transitional era” in shipbuilding techniques, yet notes that the ships used in the
China trade were mostly of Chinese design. Domestic shipbuilding meanwhile underwent little change in
style, simply reflecting growth in the size and number of vessels. As noted in the previous chapter, the

117
Ishii Kenji, Zusetsu wasen shiwa (Shiseido, 1983), 44–61.

52

earliest Japanese boats were simple dugout canoes, which were eventually modified to have additional
planks (strakes) along the side of the vessel and were propelled by oars. This type of vessel construction
has come to be called “planked-up dugout” (jun-kōzō-sen 準構造船), a precursor to the later, more
complex ship construction referring to a ship with keel structure and planked hull (kōzō-sen 構造船).
118

Small vessels such as the planked-up dugout could and would be used for voyages even to the continental
mainland, but without the ability to stabilize themselves against rough seas or harness wind-power
effectively, they were required to hug the coastline, mooring overnight and in storms. During the Heian
and early Kamakura periods when the China trade increased, the majority of vessels used for overseas
voyages were of Korean or Chinese design, while Japanese vessels remained largely unchanged.
119

The modified dugout canoe should not, however, be considered a primitive vessel. Kamakura-era
scroll paintings show that these could be quite complex in their own right. The lower part of the hull, stem,
and stern were each carved out of logs and joined at sharp angles at each end, suggesting a predecessor to
a stem and sternpost.
120
Perhaps the best representations of the planked-up dugouts are from scrolls
depicting journeys of various priests. The sharply angled bow visible in images of the voyages of Hōnen
and Kōbō Daishi suggests the foundation of the stem created by a large hollowed-out log (figs. 6 and 7).
Additional strakes attached to the hull raise the freeboard, creating a larger craft able to carry as many as
sixteen people, as shown in the Hōnen image. While the log canoe structure is not as clearly defined in a
ship in the Ippen Shōnin scroll since it is lying low in the water, this vessel entering the Hyōgo harbor
carries at least nine people and is laden with additional cargo (fig. 8).

118
My thanks to Jun Kimura for his input regarding translation of these terms. Personal Communication, February
13, 2015.
119
William Wayne Farris, “Shipbuilding and Nautical Technology in Japanese Maritime History: Origins to 1600,”
Mariner’s Mirror 95, no. 3 (August 2009): 267-271.
120
Hiroshima Kenritsu Rekishi Hakubutsukan, Shōnintachi no Setouchi: Aki no kikakuten. Umi no michi kara chūsei
wo miru II (Hiroshima-ken: Hiroshima Kenritsu Rekishi Hakubutsukan Tomo no Kai, 1996), 142.

53

Figure 6. Hōnen shōnin gyōjō eden. Goodwin (2007, cover image).

Figure 7. Kōbō Daishi gyōjō e zanketsu. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 149).

54

Figure 8. Ippen Shōnin eden mohon. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 150).
Archaeological excavations of a late Heian or early Kamakura site in present-day Futaba-chō,
Kobe city (fig. 9), have revealed one example of the central and side portions of a planked-up dugout boat
that had been discarded and subsequently reused in the construction of a well (fig. 10). While not a large
vessel – the hull log is only about one meter wide – the physical remains provide valuable information
about the joinery and durability of the vessel not immediately visible from artistic depictions. The
hollowed-out log formed the lower portion of the boat, and several planks were found alongside. Rust
marks around holes suggest the presence of fasteners that held the planks together, and x-rays of the
remaining wooden parts reveal a number of fasteners embedded into the boat. Taken together, over sixty
iron fasteners were likely used to affix the planks to the hull log and to each other.
121
It is not apparent
why the boat was discarded, but it is telling that even after abandonment the hulk was reused as part of
the well. It would have already had the natural curve desirable in well siding, and it was durable enough
to withstand additional years of usage in its new position. Although some of the fasteners deteriorated

121
Kobe-shi Kyōikuiinkai, Futabachōiseki hakkutsu chōsa hōkokusho. 3. (Kobe, Japan, 2001), 174–179.

55

over the centuries, nearly a quarter of them remained when the boat was excavated in 2001. The boat
section is a testament to the technology and know-how that shaped the watercraft for durability.

56

Figure 9. Map of places named in Chapter 2.

57

Figure 10. An in situ photo of the dugout excavation site. The boat is upended, forming one side of a well. Kobe-shi
Kyōikuiinkai (2001, photograph section p 32, plate 4).
Both large and small boats in all medieval scroll paintings cited above have deck structures,
suggesting that these vessels all could have been used for longer distance voyages to accommodate
passengers on board or sailors at rest. Although only the Ippen scroll image shows the mast stepped
(standing upright in a position to hoist a sail), the larger boats in the other two scenes also have the
infrastructure to support a mast, as evidenced by the supports above the deckhouses where the mast might
lay when it was unstepped (removed from its standing position). None of the boats, however, are actually
using the sails, and instead two images show oarsmen propelling the vessels. This is likely due to the
locations depicted in each image; Kōbō Daishi is at the mouth of the Yodo River, Ippen is entering the
harbor at Hyōgo, and Hōnen is at the port of Muro. All three locations were along shallower seas
requiring a finesse of movement that would have been more difficult under sail. The lack of clear open-
water sailing depictions prevents a detailed analysis of rigging capabilities.

58

All of the ships depicted have sharply angled, wide stems and sterns (the front and back ends of
the vessel, respectively). Additional planks are affixed athwartships, perpendicular to the sides, further
stabilizing the ends. Examining artistic depictions of ships approaching land, it seems that very few
locations had the docks or wharfs typical of medieval Western ports. As will be further discussed below,
the limited archaeological evidence available for port sites suggests that pier infrastructure was a rarity in
medieval ports. Without tying off to a manmade dock ships often would have been pulled up directly onto
shore and people and cargo would debark over the angled bow (fig. 11). Instead of adapting the landscape
to suit the design of the ship, the mariners conformed to the constraints of the landscape and designed the
vessels for efficient use of the extant shoreline.

Figure 11. Closeup of scene from Shido-ji engi, late Kamakura or early Muromachi era. Note how the boats pull up
directly to the shore, with no evidence of dock infrastructure. Kokuritsu gekijo Nōgaku-dō chōsa yōseika

(1991, plate 1).
It was not until the Muromachi period (1336 – 1573) that domestic ship construction changed
substantially. Not only did increased interactions with the mainland require larger ships capable of

59

transporting men and goods over the open seas, but the domestic economic shift from transport of estate
rents and dues to a burgeoning commercial trade, as we shall see below, demanded larger ships as well.
This range of watercraft is clearly understood from the data in the Register of Incoming Ships at
the Hyōgo North Gate. While the majority of ships carried less than two hundred koku
122
worth of goods
to Hyōgo, forty-five ships carried between four hundred and one thousand koku, and four ships carried
over one thousand koku of cargo, or approximately 150 tons.
123
The largest vessels would have been
unheard of before this time, even in travel to the mainland.
124
Seafarers also apparently were coming to
understand more about weather patterns and navigation even in open water. Shinjō Tsunezō suggests that
in the early medieval period, approximately one-third of the ships used on China trade missions sank,
while there is little note of shipwrecks in eighteen trade missions of about ninety ships total during the
later Muromachi era.
125
While he attributes this in part to a better understanding of typhoons,
improvements in ship construction also likely helped preserve the vessels.
Very little pictorial or archaeological evidence remains of late medieval ships used for domestic
trade and travel. Highly stylized ships depicted in a Nanbokuchō (fourteenth century) version of Ippen’s
voyage (fig. 12) hint at the presence of larger cargo vessels in Shiogama (medieval Mutsu province; today,
Miyagi prefecture), far from the Inland Sea. They appear to have cargo piled high on deck, and have both
mast supports and oars for propulsion. More construction details are revealed in the Shin’nyōdō engi
scroll from 1524, which includes a scene that ostensibly recreates the eighth-century voyage of envoys to
the Tang court. The ship is not a Chinese junk, however, nor is it representative of Heian-period ship

122
1 koku was equal to approximately 330 lbs
123
Hayashiya Tatsusaburō, ed., Hyōgo kitaseki irifune nōchō (Tokyo: Chuo Koron Bijutsu Shuppan, 1981), 245.
124
In her work Across the Perilous Sea, Charlotte von Verschuer has compiled lists of the amounts and types of
typical cargoes sent to China and Korea from Japan during the 10
th
– 12
th
centuries. Though the exact amounts are
often unrecorded, most manifests consisted of luxury items such as bolts of silk taffeta or damask, precious woods
and stones, pearls, and swords. The heaviest cargoes recorded were one ship with 700 pounds of sulfur and another
with over two tons of silver. She suggests that most Japanese ships used for overseas shipping were at most about
one hundred tons.
125
Shinjō Tsunezō, Chūsei suiunshi no kenkyū (Tokyo: Hanawa Shobo, 1994), 757–777.

60

construction. Scholars believe that the image is actually modeled on the type of boats that would have
been found in Inland Sea waters (fig. 13).
126

The double sails of Ippen’s ship are one of the first instances of complex rigging seen on a
Japanese ship. No longer is a single hollowed-out log used as a hull; instead planks are joined to form a
relatively flat base. This represents the next step in the evolution of shipbuilding that moved beyond the
planked-up dugout. The resulting ships are much wider than the earlier Kamakura vessels, with the ability
to carry far more cargo in an era of increased trade. Sail technology also improved with the importation of
plant fibers and the development of cloth production. No longer were sails made out of woven mats;
rather, they were made of cotton cloth, a more durable and flexible material allowing for larger sails to
propel bigger ships.
127
Although no Japanese ship from this era has yet been discovered, based on the
progression of construction techniques from planked-up dugouts to later Edo-period ships as well as a
comparison with wrecks found off the Korean and Chinese coasts, scholars believe that there would have
been interior beams running athwartships for additional reinforcement.
128

Figure 12. Shiogama scenes from Ippen Shōnin voyages. Jōshōji scroll. Miya (1979, Scroll 2 Section 3).

126
Hiroshima Kenritsu Rekishi Hakubutsukan, Shōnintachi no Setouchi: Aki no kikakuten. Umi no michi kara chūsei
wo miru II, 133. Also see Farris, “Shipbuilding,” 273.
127
Kagawa-ken (Japan), Kagawa-ken shi, vol. 2 Chūsei (Kagawa-ken: Shikoku Shinbunsha, 1989), 550.
128
Hiroyuki Adachi, Nihon no fune: Wasen hen (Tokyo: Nihon kaiji kagaku shinko zaidan: Fune no kagakukan,
1998), 33.

61

Figure 13. Shinnyōdō engi. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 154).
Although the ship remains have been lost to time, one shipwreck site does provide an inkling of
the size of the trade vessels within the Inland Sea. The Mizunoko-iwa wreck site, six kilometers offshore
of Shōdoshima (formerly Sanuki province; modern Kagawa prefecture), is thought to date from the
Muromachi era. The ship contained not only nearly two hundred pieces of Bizen pottery, but also
approximately ten tons of ballast weight in the form of river rocks. This is the first archaeological
example of ballast weight used in Japan.
129
The practice was likely known, however, as a reference in the
fourteenth-century war tale Gikeiki notes the process of taking on rocks before leaving port to stabilize a
ship in the face of a typhoon.
130
While the ship’s intended route has been lost to time, the Bizen pottery on

129
San’yō Shinbunsha, ed., Kaitei no kobizen: Mizunoko Iwa gakushutsu chōsa kiroku (Okayama-shi: Sanyo
Shinbunsha, 1978), 64.
130
Satō Genzō, Gikeiki, vol. 1 (Tokyo, Japan: Heibonsha, 1979), 233. The Gikeiki chronicles the saga of Minamoto
no Yoshitsune (1159 – 1189) and his followers. It was compiled sometime in the late fourteenth century by an
unknown author. In the passage in question, Yoshitsune and his follower Benkei are traveling via boat in the Inland

62

board indicates the likely starting point of Katakami, the center of Bizen pottery production. Possible
destinations may have been Kishū or Kumano on the Kii peninsula via the Awaji straits, since remnants
of Bizen pottery have been found in archaeological sites in that region.
131

The Muromachi innovations in ship design, while resulting in a wider, larger craft, also created a
shallower draft than those of earlier vessels, in turn requiring heavier weights to stabilize the ship on
rough seas. If the Mizunoko ship’s route was indeed through the Awaji straits, it would have needed to
prepare for more exposed waters as it left the Inland Sea, which would explain the need for the large
amount of ballast weight. Though scientists have examined the ballast stones in the hopes of determining
a point of origin, which would shed further light on the travel routes of the unnamed wreck, to date they
have unfortunately been unsuccessful. Tests show that the rocks were not from the likely port of origin
near the Yoshiigawa river in Bizen, but certain stones were common to several areas in Shikoku and
Kishū.
132
The sheer amount of ballast from a likely origin somewhere along the eastern Inland Sea,
combined with the Bizen cargo and possible destination of Kumano, demonstrate the ability of the larger
Muromachi ships to range further afield with a heavy cargo. Increased trade required increasingly large
ships; and as will be discussed in more detail below, port infrastructure gradually changed to
accommodate the changing vessel styles.
The development of shipbuilding throughout the medieval period reflects increased maritime
shipping and the growing importance of domestic maritime trade. Although we know little about the
people involved in building ships domestically, innovations in shipbuilding technology such as use of
ballast stones and reinforcing beams suggest the willingness to adopt open-water seafaring practices in
domestic vessels. As the volume of trade and the size of vessels increased, communities needed to adapt
the local infrastructure to facilitate larger ships and cargo loads.

Sea and have been caught in a storm. The narrator describes how the ship had taken “many stones” on board on
departing Kawajiri (Aki province) for additional stability.
131
San’yō Shinbunsha, Kaitei no kobizen: Mizunoko iwa gakushutsu chōsa kiroku , 65–66.
132
Ibid., 87–88.

63

What Makes a Port?

Although the size and construction of the ships affected the types of ports at which ships could land,
certain criteria were necessary for all ports. In Kitayama Kenichirō’s list of criteria for a successful port
town, five of six items refer to the geography and infrastructure of the port. He cites the need for a place
where a boat can moor, navigational markers visible from the sea (often temple or castle buildings), the
ability to load and unload cargo, a road crossing the port, and easy inland access.
133
A sixth factor related
to infrastructure, unremarked upon by Kitayama, would have been access to a supply of lumber. It was
not only important for shipbuilding, but it was also critical in seaside saltmaking to fuel fires that would
evaporate the water. Though Kitayama was specifically examining ports in Sanuki province, applying
these criteria to other port sites helps clarify why certain areas became such important ports in the
medieval period. Onomichi (Bingo province) is the best embodiment of most of Kitayama’s criteria,
while other ports such as Utazu (Sanuki province), as well as certain smaller ports, demonstrate the
influence of the natural environment on port development and the evolution of shipping and production
ports.
Onomichi, located on the coast of Bingo province, was an ideal site for the development of
maritime shipping. The large island of Mukaishima sits only two kilometers south of the port, creating a
narrow waterway that provided both natural shelter from inclement weather and a deep channel that could
accommodate larger vessels. This environment satisfied the requirement to have a safe mooring place as
well as easy offloading of cargo.
The first historical references to Onomichi are found in records from the late Heian period.
Documents from 1168 (Nin’an 3) mention “Five chō of paddy and dry fields of Onomichi-mura” that
were then designated as a harbor and warehouse area for use by Ōta Estate.
134
From the first, then,
Onomichi served as a transshipment hub, acting as the loading point for rents and dues from Ōta Estate to
be sent to its proprietor (initially the retired tennō Go-Shirakawa, who ceded it to Kōyasan in 1186).

133
Kitayama, “Chūsei minatomachi no chikei to kūkan kōzō,” 124.
134
Dai nihon komonjo, Kōyasan monjo Vol. 1 #37. Ōta-no-shō gesu narabini satanin tō ureijō.

64

Though it is unclear which overland route was taken between the estate and the port, Onomichi’s official
designation as the port for shipments from Ōta Estate indicates that there was indeed easy access to inland
areas to allow for cargo distribution and traffic to and from the port. In later centuries, as ship
construction technology improved and larger ships were built more regularly, ports such as Onomichi or
the similarly sheltered Setoda were able to accommodate larger ships. Of the sixty-two ships registered to
Onomichi that arrived in Hyōgo in 1445, forty-nine carried over one hundred koku of cargo (with thirty-
four of those carrying two hundred koku or more), and of the sixty-nine Setoda ships, only five carried
less than one hundred koku. Ports with better mooring and cargo loading facilities could accommodate
more and larger vessels.
Kitayama’s stipulation of the need for navigational markers is amply fulfilled in Onomichi’s
natural environment. The Onomichi area is hemmed in by hills, providing some natural shelter for the
port and a ready source of trees for shipbuilding or repairs. The hills were geographically distinctive in
their own right. Earlier records from Saikokuji temple, said to have been founded by the monk Gyōki in
729 in what would become Onomichi, describe the location as “Tamanoura.” The “tama” of that name
may refer to a specific boulder that is located within the precincts of Senkōji temple, situated near the
summit of one of the mountains surrounding Onomichi. According to legend, at night the boulder glowed,
emitting a light to guide ships safely into harbor (figs. 14 and 15).
135
Regardless of the veracity of this
claim, the boulder itself would likely have been a distinctive marker for ships coming through the strait
between Onomichi and Mukaishima. Legends sprouting up relating to the navigational effectiveness of
this boulder reinforce the idea that this was a special place, ideal for the safe passage and mooring of
ships.

135
Onomichi Shiyakusho, Onomichi shishi, vol. 1 (Onomichi, Hiroshima: Onomichi Shiyakusho, 1939), 208. Today
a round ball has been placed on top of the boulder to symbolize a glowing pearl, and it is lit up manually at night.

65

Figure 14. Tamanoiwa boulder today at Senkōji temple, viewed from the ropeway cable car. Photo by author (2011).

66

Figure 15. View towards Tamanoiwa from the original medieval shoreline. The boulder is encircled in red for easier
visibility. Photo by author (2011).
An account from a Korean envoy actually describes one such navigation marker in the area. Upon
leaving Onomichi, their ship exited Bingo province heading toward the capital, and en route travelers
viewed “an islet, with a small hut situated within the rock walls. People claim that ‘A priest always
resides there with the “Water Moon” Kannon.’ In front of the hut is an old pine tree, and [in front of it] is
the wide ocean on three sides.”
136
This was likely what is today known as the Abuto Kannon, a small
temple located on the tip of the peninsula jutting out between Onomichi and Tomo (figs. 16 and 17). The
temple sits at the mouth of the strait in between Tajima and the mainland, and it does indeed appear that
there was ocean on three sides: the strait behind it, the body of the Inland Sea before it, and the coves and
harbors of Tomo and Kasaoka to the east. The small temple visible on the “island” and its Kannon statue
would have served as an excellent navigational marker for ships entering or leaving the strait and heading

136
Song Hui-gyong, Rōshōdō Nihon kōroku: Chōsen shisetsu no mita chūsei Nihon, trans. Murai Shōsuke (Tokyo:
Iwanami Shoten, 1987), 89.

67

to or from the wider Inland Sea. The prominence of this navigational marker likely was an asset not just
for Onomichi, but also for Tomo. Chinese coins and pottery found at Tomo (also known as Kusado
Sengen) indicate that it likely enjoyed a lively trade with other regions, suggesting that Tomo, too, was a
shipping port.

Figure 16. Location of Abuto Kannon in relation to the port sites of Onomichi, Tajima, and Tomo.

68

Figure 17. Abuto Kannon today, part of Bandai-ji Temple in Hiroshima prefecture. Photo by author (2011).

Utazu (Sanuki province) had a similar developmental trajectory as did Onomichi. Situated at the
mouth of the Daisoku River, the port was initially used as a loading point for rents and dues (nengu 年貢)
shipped from Kawazu Estate to its proprietor, Kasuga Shrine in Nara.
137
In the late medieval period it
rapidly became one of the most booming ports in Sanuki, though not quite on the scale of Onomichi or
Setoda. In 1445, forty-seven ships sailed from Utazu to Hyōgo, twenty-seven of which were larger than
one hundred koku. In a description from 1389 of Ashikaga Yoshimitsu’s pilgrimage to Itsukushima
Shrine via the western provinces, Utazu is described as follows: “Rows of fisherpeoples’ houses extend to
the north along the beach…. Ancient pine and other trees line the rocky shores. The eaves of many
temples are visible.”
138
Although none of the temples seem to have attained the legendary prominence

137
Karaki and Hashitsume, Chūsei no Sanuki, 192–193.
138
Quoted in Yamauchi Yuzuru, Chūsei Setonaikai no tabibitotachi (Tokyo: Yoshikawa Kōbunkan, 2004), 32.

69

that Tamanoiwa in Onomichi had, the fact that they were clearly visible from shore attests to their
suitability as navigational markers.
Archaeological information in conjunction with the historical record reveals more about the
physical layout of mooring areas in various parts of the archipelago, providing reference for our Inland
Sea study. Unfortunately, to date no archaeological investigations have been carried out at major
medieval port sites such as Onomichi or Utazu, and the Hyōgo checkpoint port area is today largely
covered by landfill. Stone remains have been found at some smaller port sites, however, including the
mouth of a stream feeding into the sea on Naoshima and in excavations at Nohara near Takamatsu Castle
(both in Kagawa prefecture).
139
The former did not appear at all in the 1445 Hyōgo Register, but thirteen
ships registered to Nohara sailed to Hyōgo. Not only was this a far smaller number than the ships that
sailed from Utazu, Onomichi, or Setoda, but the ships carried less cargo altogether, with only nine vessels
shipping more than one hundred koku of goods. The fact that these ports did not enjoy the geographic
advantages of the larger shipping hubs required them to create their own mooring infrastructure. It
generally consists of stone laid near the waterline, likely established as a stable place to unload cargo on
the sandy beaches. In the case of Nohara, the stones are each between ten and twenty centimeters long
and are laid to follow the natural shoreline. Wooden remnants likely used as stabilizers for mooring posts
were found along the same line.
140
Naoshima’s situation is similar, as stone foundations line a sandbar
within Tsumiura Bay close to the shoreline.
141
Sites built up in the same manner have also been
discovered in Tosaminato (Aomori prefecture, fifteenth century), Hakata (Fukuoka prefecture, sixteenth
century), and Tokuzōdani (Saga prefecture, thirteenth – fifteenth centuries).
142
At both Naoshima and
Nohara, the cargo loading and offloading still would have been difficult, as it would have been awkward

139
Kagawa-ken Kyōkuiinkai, Kendō kankei maizō bunkazai hakkutsu chōsa: Muraguro iseki, Tsumuura iseki
(Kagawa-ken, March 2003), 37.
140
Satō Ryoma, Kanehara Masaaki, and Kitano Nobuhiko, Takamatsu- jō seki (Nishinomaru-chō chiku) Sanpōto
Takamatsu shōgō seibi jigyō ni tomonau maizō bunkazai hakkutsu chōsa hōkoku 2 (Kagawa Prefecture: Kagawa ken
kyōiku iinkai, 2003), 236–237.
141
Matsumoto Kazuhiko, “Chūsei minatomachi Nohara no jisshō,” in Chūsei Sanuki to Setouchi sekai, ed. Ichimura
Takao (Tokyo: Iwata Shoin, 2009), 20.
142
Kagawa-ken Kyōikuiinkai, Kendō kankei maizō bunkazai hakkutsu chōsa: Muraguro iseki, Tsumuura iseki, 37.

70

to maneuver heavy loads over sandbars and shoreline. The archaeologists for the Nohara site hypothesize
that this setup may have been due to the fact that it was a relatively small port, and that it was only in the
nineteenth century that the ports in this area truly began to see significant shipping.
143

A recent investigation of a mooring site near a seventeenth-century quarry used to transport stone
for Osaka Castle may provide further insight into the mechanics of shipping and the types of
infrastructure used. The Iwagatani site, on the east coast of Shōdoshima, consists of a number of large
rocks completely submerged offshore as well as a three-meter tall rock set in two-meter waters, known as
the “kamome ishi ( かもめ石 seagull stone),” with a stone post atop it. While analysis of this site is still
ongoing, preliminary results indicate that the submerged rocks are placed in a nonrandom pattern that
appears to help channel the water flowing in stronger currents. Archaeologists hypothesize that the
seagull stone with its post was used as a mooring site for vessels, letting them tie off safely in a stable
area. They suggest that boards were used as a temporary bridge between the seagull stone and shore,
allowing rocks from the quarries to be loaded efficiently, and then using the channeled currents to refloat
the boat at high tide to safely move it away from shore.
144
It is possible that a similar technique was
employed at Naoshima and Nohara, as the rocks placed along shore could have channeled the currents in
the same manner. Instead of building a permanent infrastructure to lade cargo, planks could have been
used there as well.
Though the above-mentioned sites provide only a few scattered examples of mooring sites that
have been discovered to date, the Naoshima, Nohara, and Shōdoshima investigations suggest important
information about shipping even in smaller ports. As Chapter 3 will demonstrate, records of trade goods
shipped throughout the Inland Sea do not highlight any of those three ports as major sources of cargo
shipments to the central capital area. According to those records, Nohara and Shōdoshima appear to have

143
Sato, Kanehara, and Kitano, Takamatsu-jō seki (Nishinomaru-chō chiku) Sanpōto Takamatsu shōgō seibi jigyō ni
tomonau maizō bunkazai hakkutsu chōsa hōkoku 2, 237.
144
Mogi Kotarō, “Examining Coastal Shipping Processes around Shōdo Island during the Tokugawa Period,” in
Proceedings of the 2nd Asia-Pacific Regional Conference on Underwater Cultural Heritage, ed. Hans Konrad Van
Tilburg et al., vol. 2 (presented at the 2nd Asia-Pacific Regional Conference on Underwater Cultural Heritage,
Honolulu, HI: 2014), 624–626.

71

shipped mainly locally produced salt to Hyōgo, with little indication of transshipment of other products,
so they were mostly production ports.
145
Yet even without documentary evidence of frequent shipping,
the port residents labored to create an infrastructure that could support maritime shipping. Building out
the stone foundations at Nohara and Naoshima was a task that likely required a significant amount of time
and effort to transport and place such a large quantity of rocks. At Shōdoshima, the submerged rocks are
much larger and would have needed substantial manpower to place the rocks, not to mention know-how
about the currents to situate the rocks most effectively for using high and low tides for shipping. The local
population put forward quite an effort to improve their port infrastructure in these ways.
This sampling of what we know about medieval ports shows that most fit Kitayama’s
prerequisites for a viable port site, but that geographic factors helped shape their role as larger trade ports
or smaller local ports. Sites such as Onomichi or Utazu had the deep waters necessary to support larger
vessels, temples or natural landmarks that could be used as navigation markers, and protected waters that
would facilitate cargo loading and offloading. Both were located near rivers that provided inland access,
by raft or other small watercraft. All of these factors contributed to their development as prominent
shipping ports. Smaller ports such as Nohara or Naoshima were not as naturally endowed with similar
facilities. Their archaeological remains suggest that they needed to construct a safe mooring area that
could accommodate vessels, and that even with those efforts it was still difficult to manipulate cargo.
Natural navigational landmarks were not as prevalent. These factors likely contributed to the inability of
those ports to sustain the volume of shipping that Onomichi or Utazu could accommodate, and thus their
main role lay in shipping locally produced items.

145
In the Register of Incoming Ships at the Hyōgo North Gate, which records nearly two thousand boats that passed
through the Hyōgo checkpoint, Nohara is mentioned only twelve times. Boats registered to “Shima” are thought to
have been from Shōdoshima, although there is no way of knowing if the port site noted in the Register was the one
found at Iwagatani. Regardless, Shima sent only twenty ships to Hyōgo – a not insignificant number, but also not a
major source of Inland Sea shipping. Naoshima does not even have a single mention in the Register.

72

Governance of Port Towns and Checkpoints

Ports could be important sites for governing bodies to control. Through overseeing a port, particularly one
focused on shipping, a proprietor had access to incoming and outgoing products and the potential to tax
that revenue. Production ports became direct sources for local products. Control of ports in the medieval
Inland Sea is sometimes unclear due to a lack of documentation. Overall trends of proprietorship in the
late medieval period, however, indicate a shift to local assertion of control over ports. The increasing
desire for control underscores the importance of port sites and maritime checkpoints.
During the late Kamakura and early Muromachi eras the relationship between estate proprietors
and estate management shifted. As power struggles occupied court and shogunal factions during the
Nanbokuchō period, the central powers (including court families, regental families, or religious
institutions such as Tōji or Tōdaiji) tended to maintain estates close to home, near the capital at Kyoto. In
outlying areas, local influential families began to take direct power, administering the estate as their own
property. In an effort to reassert control over their estates, some central powers abolished parts of the
hierarchy of intermediate overseeing posts (shiki 職) and asserted direct control. These directly
administered lands, with either central powers or local powers as their proprietors, came to be known as
ichienryō (“fully held lands” 一円領).
146
Furthermore, as the central proprietors’ hold on their lands
weakened, cultivators and other estate residents began banding together. As early as the late Kamakura

146
Nishitani Masahiro, “Shōen sei to saigoku shakai,” in Saigoku ni okeru seisan to ryūtsū, ed. Kawaoka Tsutomu
and Koga Nobuyuki (Osaka-shi: Seibundō, 2011), 27–28. Also see Shimizu Ryō, “Nanbokuchō ki – Sengoku ki no
shōen,” in Shōenshi kenkyū handobukku, ed. Shōenshi kenkyukai (Tokyo: Tokyodō, 2013), 154–155, who describes
the rise of jisha honjoryō and bukeryō, or lands that were administered by shrines and temples or military families,
respectively. In the case of the temples and shrines, these lands were formerly estates with rights granted to high-
ranking nobles or institutions, and in the military families’ cases they were usually estates and public lands to which
those individuals held local rights. By the 1340s, however, those proprietors were generally administrators in name
only and were at the top of an administrative hierarchy. In an effort to reform the system, a decree by Ashikaga
Tadayoshi effectively gave the military families the right to administer their lands as direct suppliers of military
provisions. To maintain control over their own lands, shrines and temples therefore needed to take a more direct
hand in administration. As internal strife continued, the shogunate passed further reforms that removed additional
administrative layers, resulting in the growth of ichienryō, or lands fully held by their direct proprietors.

73

period, there is evidence of cultivators collectively presenting complaints or appeals to central authorities,
representing their own interests directly rather than going through local authorities.
147

The early fifteenth century saw the rise of several families as military governors for provinces
throughout the Inland Sea region. The Hosokawa were the major presence, at one point controlling Settsu,
Bitchū, Awaji, Awa, Tosa, and Sanuki provinces, and in 1455 briefly adding Iyo, previously under Kōno
administration, to their list. The Yamana were based in Bingo and Aki, but would eventually take over the
military governorship of Bizen and Harima from the Akamatsu family. In 1467 the Akamatsu would
regain control of Harima, but Bizen remained under the Yamana. The Ōuchi family had long been
ensconced in Suō and Nagato provinces and would continue to hold power there in the fifteenth
century.
148
As the different families’ influence increased, they were able to bring other areas under their
control. As we shall see in the following chapters, in some cases those ties may have helped facilitate not
only administrative endeavors, but also led to different connections among the trade networks
encompassed by their various spheres of influence.
With the increasing civil unrest throughout the later Muromachi period, military governors still
needed funds from the estates to support their military endeavors in particular, but often relied on proxies
or contractors to physically collect the rents and dues. Since these proxies were locally powerful figures,
this also transferred more power and authority to estate residents.
149
The proxy collectors made use of
other local powerful individuals to collect rents and dues, further strengthening ties with cultivators and
residents while the absentee landlords became further removed from daily administration.
One of the most lucrative forms of port control was that of overseeing a maritime checkpoint,
providing direct access to toll monies. With the rise of individual control in the late medieval period came
an increase in the creation of maritime checkpoints, particularly along rivers, as local powers sought to
take advantage of passing ships. Shipping ports were also likely to be directly administered by central

147
Shimizu, “Nanbokuchō ki – Sengoku ki no shōen,” 158. Shimizu cites a case from 1306 of the cultivators in Tara
Estate in Wakasa province appealing to their proprietor at Tōji for a reduction in rents and dues. Kamakura ibun
(below, KI) #22717.
148
See Appendix 1 for a detailed list of provincial/military governor appointments.
149
Shimizu, “Nanbokuchō ki – Sengoku ki no shōen,” 167–169.

74

proprietors or proxies, while production ports had less of a visible administrative presence. Looking at
different types of ports such as the transshipment hubs of Onomichi and Setoda, as well as the
development of smaller ports in Sanuki, suggests ties between governance and maritime concerns, while
disputes concerning checkpoints such as those at Hyōgo or on other rivers highlight the desirability of
control over those areas.

Checkpoints: Controlling Sea Traffic

Checkpoints (sekisho) were a critical aspect of control over travel routes. The checkpoint system was
initially established during the ritsuryō period, as a wide network of roads was built. Aida Jirō suggests
that they served three major functions: economic (a means for taxing travelers), military (preventing
enemies from approaching), and monitoring (observing the flow of traffic).
150
Maritime-based
checkpoints were usually found along rivers, particularly at river mouths, to best target maritime shipping.
The decentralization of power and the rise of local proprietors were directly reflected in the confusion and
jockeying for control visible in the records concerning checkpoint administration.
For the purposes of this study, the most important checkpoint by the mid-Muromachi period was
the northern Hyōgo checkpoint, introduced in the previous chapter. It came under the proprietorship of
Tōdaiji in 1308, when the director (bettō 別当) of the temple petitioned the court to establish a checkpoint
there, with fees to be used for both rebuilding the port and to maintain the Hachiman Shrine at Tōdaiji in
Nara. The retired tennō Fushimi agreed to the request, and Tōdaiji was awarded the right to administer the
northern checkpoint.
151
Nagashima Fukutarō suggests that the checkpoint was initially used as a
storehouse for oil, and a Tōdaiji priest (jisō 寺僧) was on site to oversee monthly offerings and any
necessary repairs to the checkpoint. A magistrate (bugyōsho 奉行所) office was later established at the
checkpoint as well. Nagashima further postulates that while Tōdaiji assigned an agent-manager (zasshō)
to the site, the magistrates were actually from local prominent families working in conjunction with the

150
Aida Jirō, Chūsei no sekisho (Tokyo: Arimine Shoten, 1976), 3–7.
151
Ibid., 108. See KI # 23491, 23494, 23598 (Vol 31).

75

temple. This would seem to be consistent with the late medieval tendency for local families to amass
power in a particular region.
152
In this case, however, Tōdaiji did apparently maintain a physical presence
at the checkpoint rather than relying entirely on proxy overseers.
The Hyōgo checkpoint sat at the entrance to the river system connecting the Inland Sea with the
marketplaces of the central capital at Kyoto, and so the potential for toll revenue was considerable.
Control of the checkpoint administration demonstrated Tōdaiji’s ties with the central court and resulted in
significant revenue. The head of the Tōnan-in chapel within Tōdaiji held the rights of inspection at the
checkpoint. The monk Shōchū (1268 – 1319) was assigned the administration of the inspection office,
possibly because he was on good terms with the retired monarch Fushimi.
153
Throughout Tōdaiji’s
proprietorship of the checkpoint, revenues from tolls collected were used for the temple’s benefit. In one
case Tōdaiji was able to build a new pagoda and to repair walls within its precincts.
154
From 1341 through
1387, there were twenty instances of special requests from Tōdaiji officials for additional funds from the
checkpoint. These special cases usually were in conjunction with specific events or activities, such as
paying for envoys to be sent to the capital for court cases, for visits to the shogunate, or for repairs to the
checkpoint itself.
155
Tōdaiji continued to maintain the checkpoint throughout the Muromachi era and
enjoyed a steady stream of revenue from the increased amounts of trade goods being shipped from ports
throughout the Inland Sea, as will be discussed in more detail in the following chapter.
Less information is available about the southern Hyōgo checkpoint (exact location unknown),
administered by Tōdaiji’s rival, Kōfukuji temple, but the records that do remain suggest that there was a
power struggle between the two temples over control of shipping through Hyōgo. Though Kōfukuji had
administrative rights over checkpoints at Kanzaki (Settsu province) and Fukudomari (Harima province),
those were apparently not as lucrative as the Hyōgo checkpoint. In 1338, the Ashikaga shogunate granted

152
Quoted in Hatakeyama Akira, “Chūsei Tōdaiji ni yoru Hyōgo-Seki no keiei to sono soshiki,” Journal of Japanese
History 494 (October 2003): 48. Refers to Nagashima Fukutarō’s “Hyōgo-seki wo meguru jisha no atsureki,”
Kokushigaku 19 (1934).
153
Ibid., 50–51.
154
Hayashiya 1981, 224–226. See Dai Nihon shiryō, Vol 18 (Iewake Tōdaiji monjo) part 5, (Tokyo: Tokyo
University, 1968) 128 – 136, document #72.
155
Hatakeyama, “Chūsei Tōdaiji ni yoru Hyōgo-seki no keiei to sono soshiki,” 56.

76

Kōfukuji’s appeal for the right to collect fees at Hyōgo as well, and a southern checkpoint was
established.
156
The administrative differences between the northern and southern checkpoints are unclear,
but it is possible that the northern checkpoint levied its fees on ships entering the harbor, while Kōfukuji
assessed fees on ships leaving through the southern checkpoint. Later documents from as late as 1404
show complaints from Kōfukuji that too many tax exemptions were being granted to ships passing
through their checkpoints, and that those exemptions should be revoked.
157
The fact that Kōfukiji
continued to fight at least until 1404 for the right to assess those fees indicates the value of having control
over such a checkpoint.
Although most of this study focuses on seaside ports and coastal communities, it is worth noting
here the importance of checkpoints along rivers as well, to emphasize the importance of control of
waterways by local powers. As the medieval era progressed, more and more riparian checkpoints were
established, and fees rose dramatically. In 1397, rents and dues from Higashi-Nagao estate in Sanuki that
were shipped via Amagasaki and the Yodo River to the proprietor, Daigoji Sanbōin, had a ten percent fee
levied on them. Ten years later that fee had doubled, and by 1459, shipping fees had increased again to
thirty percent of the cargo’s value.
158
In 1453, there were said to be 180 checkpoints along the Yodo River.
The exact length of the river that encompassed those checkpoints is unclear, but the distance from the
river’s mouth to Ōyamazaki-chō, where Iwashimizu Hachiman Shrine was located, was only
approximately thirty-six kilometers long. Many of these checkpoints were “private” checkpoints, put into
place by provincial governors and their proxies, or by other aristocrats, temple affiliates, and others from
all levels of society. This suggests the overall breakdown of centralized power in the late medieval period,
and the propensity for local individuals and institutions to exert influence over a particular locale.
159

A series of documents to and from the shrine affiliates at Ōyamazaki over the course of a century
attests to the importance of control over riparian checkpoints and the commodities shipped through them.

156
Hayashiya, Hyōgo Kitaseki Irifune Nōchō, 226. Also see Kasuga Jinja monjo, 252.
157
Kasuga Jinja Jimusho, Kasuga Jinja Monjo, vol. 1, 3 vols. (Nara Meishinsha, 1928). # 244 Ōe 11 (1404), 5.25,
#245. (no year) 3.29.
158
Shinjō, Chūsei suiunshi no kenkyū, 676.
159
Ibid., 678.

77

The community at Ōyamazaki was affiliated with Iwashimizu Hachiman near Kyoto, and it was awarded
the sole right to provide lamp oil to that shrine, a commodity that was critical for many religious rites.
Shikoku was the major source of the egoma berry as well as of sesame, the former being the main
component of lamp oil and the latter possibly a secondary ingredient (fig. 18).
160
The series of documents,
dating from 1311 to 1412, outlines a dispute regarding a checkpoint in Awa province along the Yoshino
River. Egoma shipments were meant to be exempt from any fees or taxes, as they were earmarked
directly for shrine usage. The Ōyamazaki affiliates lodged a complaint regarding the unauthorized
establishment of the Yoshino river checkpoint, claiming that the administrators had seized the egoma as
payment for passage through the checkpoint. The complaint progressed through the Awa provincial
governor all the way to the Ashikaga shogunate in1373, which ruled in favor of the Ōyamazaki affiliates
and confirmed their right to have exemptions from tolls along the Yoshino river.
161
This dispute
demonstrates first the trend for smaller locales to exert their power and gain revenue by establishing
individual checkpoints along shipping routes. Second, it suggests that even though individuals may have
been attempting to increase their power through those checkpoints, certain entities such as Iwashimizu
Hachiman were still influential enough to gain an exemption from these tolls.

160
Suzanne Gay, “The Lamp-Oil Merchants of Iwashimizu Shrine: Transregional Commerce in Medieval Japan,”
Monumenta Nipponica 64, no. 1 (2009): 14–15, 17. Gay outlines the relationship between the lamp-oil purveyors
(neither true shrine affiliates nor true merchants, but something of a hybrid, she argues) and Iwashimizu
Hachimangū, and explores the extent of the network necessary for the procurement and production of lamp oil.
161
KI 24343, 29942, 30768, and Ōyamazaki chōshi shiryō-hen #12, 39, 84, 85

78

Figure 18. From left to right: A growing egoma plant, the seeds from the egoma plant ready to be ground into oil, and a
statue on the precincts of Ōyamazaki Hachiman today honoring the egoma and lamp-oil sellers. Photos by author (2012).
By examining control over checkpoints, then, we see the enduring economic benefit to
proprietors of checkpoints. Revenues garnered from the checkpoints were critical for places such as
Tōdaiji, which reaped the benefits of tolls from the thousands of vessels that passed annually through the
checkpoint at Hyōgo. Even in the late medieval period when the estate system was breaking down and
many estates were taken over by local proprietors, institutions such as Tōdaiji and Kōfukuji kept control
of checkpoint administration. In addition, the ability to pass untaxed through checkpoints was coveted and
protected at the highest levels, indicating the importance of checkpoint control and administration. In the
next section we will explore more about checkpoints, but from the point of view of those passing through
them, not their proprietors.

Ports: Shipping and Strategic Administration

Documents pertaining to medieval port administration, particularly for the smaller production ports, are
often extremely limited. Production ports such as Onomichi sometimes have more to draw from, but even
then there are gaps. What we do see, however, is a tendency for individuals to exercise their power
through control of port sites or ships, either by direct control of the port or by using their power to
command exemptions for shipping fees and tolls.

79

As noted above, Onomichi’s geographic setting, including the sheltered landing, deep water, and
highly visible navigational markers, gave it a natural advantage as a shipping port. Ōta Estate, an inland
estate under the proprietorship of Kōyasan, began to use Onomichi as its “warehouse” (Kurashiki 倉敷)
to store rents and dues, setting Onomichi on its path as a local hub for shipping. In 1194, one Kōyasan
monk, assigned as overseer of Ōta Estate, additionally recognized the strategic importance of controlling
the port. Mindful of earlier uprisings from the local Tachibana family, he wanted the port to be physically
separate from the estate in order to keep the estate dependent on the port and to better exert the temple’s
control over the area.
162
By requiring Ōta Estate to ship its goods through Onomichi, the estate was forced
to make use of a shipping area outside its own borders, which in turn heightened Onomichi’s importance
as a local shipping port. Rents and dues would have been brought by horse from the landlocked estate to
Onomichi and loaded there to be sent to Kōyasan, possibly through the port at Sakai.
163

Although Kōyasan recognized the importance of Onomichi as a useful port in its own right,
individuals were also able to strengthen their own power through management of workers affiliated with
the port. From 1278 to 1302, for instance, Enshin, assigned by Kōyasan as Ōta Estate’s custodian
(azukaridokoro 預所), manipulated the assignment of oarsmen and sailors in the area to bring maritime
shipping under his own control.
164
There are suggestions that Onomichi shippers sometimes hired out
boats from adjacent areas such as Tomo to help transport rents and dues, indicating early tendencies
towards collaboration between shippers from different port towns.
165
Other areas, including Yuge Estate
(formerly Iyo province; modern Ehime prefecture), record instances of borrowing ships from Aki and
Bingo provinces, and as Onomichi was such a major shipping center, it is highly likely that it could have

162
Matsushita Masashi, Umoreta minatomachi Kusadosengen, Tomo, Onomichi (Tokyo: Heibonsha, 1994), 188.
163
Aoki, Onomichi shishi (Shinshū), 1:477. Records of a shipment of rents show that the goods were shipped to
Sakai, where they were exchanged for coins that were subsequently shipped to Kōyasan (Kōyasan monjo #165,
Bingo Ōta-no-shō nengu hikitsuke). This undated document describes the accumulation of heretofore unsubmitted
rents from 1439 to 1447. It is possible that earlier payments may have taken a different route.
164
Shinjō, Chūsei suiunshi no kenkyū, 41. This also suggests the importance of those skilled laborers to the overall
development of the region, as control of the mariners allowed the overseer to effectively have control over the port.
165
Nagahara Keiji, Shōen (Tokyo: Yoshikawa Kobunkan, 1998), 173. From records of Onomichi’s shipments of
rents and dues from 1439 – 1447.

80

been a source for those vessels.
166
Efficient administration – even if motivated by personal gain –
linking Onomichi with neighboring ports helped it to grow as a shipping center.
Indeed, sometimes that prominence as a shipping center put Onomichi in jeopardy. Monks of the
Kongōbuji temple complex at Mount Kōya recorded an incident in 1319: “As the location of the estate’s
storehouses and harbor, Onomichi is a port for ships. As a result of the many ships that dock here,
residents have prospered. For that reason, the provincial constable Nagai Sadashige assiduously made
plans to invade and capture it when he had the opportunity….” The record goes on to describe the
destruction caused by the invasion, including the burning of residences, offices, and temples, and notes
that the marauders loaded objects from “religious centers and residences” onto ten large ships.
167
Nagai
recognized the importance and potential for treasure in such a vibrant port town, and moved to make it his
own. It is also noteworthy that his “ten large ships” could moor close enough to accomplish the
(presumably speedy) loading and retreat, as this bustling port likely was already accommodating a
number of ships. The document thus gives some sense of the overall size of Onomichi’s mooring capacity.
Onomichi also occupied an unusual position in that it was also noted for its military importance.
While there are no records of a permanent naval force stationed there, several travel records allude to
Onomichi’s ability to accommodate a large number of ships sailing for military purposes. The Taiheiki
notes that when Ashikaga Takauji was mobilizing his forces to move south in 1340, he split his sea forces
to land at Tomo and Onomichi. His own ceremonial ship (gozabune 御座船) went to Onomichi, where he
debarked to pray at Jōdoji temple.
168
Onomichi thus was not only a port physically capable of
accommodating his fleet, but was also important psychologically as the place chosen by the military
leader himself. Onomichi, then, held prominence as both a shipping and a strategic port, one that was
overseen directly by its religious proprietor, the Kōyasan monastery.

166
Kamakura ibun (KI) 28836. Petition from the cultivators of Yugeshima Estate, Iyo Province, to Tōji.
167
Kongōbuji monjo, vol. 2, #17. Gen’ō 2 (1320) 8 (no day). Kōya-san monjo, vol. 2 (Kyoto: Kōyasan monjo
Kankōkai, 1937). Translated in Shapinsky, Lords of the Sea, 59–60.
168
Hasegawa Tadashi, Taiheiki, Dai 1-han (Tokyo: Shōgakkan, 1994), 110. Book 22; “Yoshisuke ason byōshi no
koto.”

81

No Sanuki port stands out as prominently as Onomichi does for its strategic or overall shipping
importance, although several became more notable after Ashikaga Takauji’s ascent to power. There are a
few allusions to earlier port administration in Sanuki, most notably in Shido (formerly Sanuki province;
modern Kagawa prefecture). In 1261, a Tachibana Ason wrote of opening up a new entryway to the ocean
(shioiri shinkai 潮入新開), specifying the length, width, and total size of the area and exhorting the
cultivators in the area to perform their duties [to create the canal].
169
This was likely to facilitate maritime
shipping, as a wider canal could accommodate larger ships. The plan seems to have borne fruit, as later
records indicate that there was a thriving marketplace on the grounds of Shido temple. Even today, the
temple – with a very prominent pagoda – is located directly in view of the port at Shido (fig. 19),
suggesting a function both as navigational marker and, due to the location of the marketplace inside the
temple precincts, an affiliation with trade itself. The eventual scale of the marketplace is reflected in a
much later document, ordering the closure of the horse market at Shidoji in 1473 by the Hosokawa
military governor (shugo 守護) of Sanuki.
170
The fact that horses were the main commodity suggests the
size and scale of the market as a whole – a small market, and by extension a small port, could not
accommodate large live animals. Early direct action by Tachibana Ason resulted in the development of a
port town large enough to host such a market.

169
Shido Chōshi Hensan Iinkai, Shinpen Shido chōshi (Kagawa-ken Okawa-gun Shido-cho: Shido-cho, 1986), 120.
This document does not appear to be in Kamakura ibun,
170
Kagawa-ken (Japan), Kagawa-ken shi, 2 Chūsei:549. Though the reason for the closure is unclear, it is possible
that the absentee Hosokawa military governor was attempting to control the flow of wealth and goods to local
powers.

82

Figure 19. View from a pier at Shido Port looking towards shore. Note Shido-ji temple’s pagoda. Photo by author (2012).
The growing power of the Hosokawa during the fourteenth and fifteenth centuries shaped several
Sanuki ports into more than small production ports. The Hosokawa warrior family had aligned with
Takauji during the upheaval of 1333, and after the latter’s ascent to the position of shogun in 1337, he
appointed several members of the Hosokawa family, his relatives, to posts in the Shikoku region,
including as the military governor of Sanuki, Awa, and Iyo provinces, among others. In turn, the
Hosokawa deputy shōgun (kanrei 管領) sent proxy administrators (hikan 被官) to various areas in Sanuki
especially, including members of the Kagawa, Yasutomi, and Sogōshi families.
171
From 1394 – 1428 the

171
Little is known about these families. See Kokushi Daijiten Henshū Iinkai, Kokushi Daijiten, Dai l-han (Tokyo:
Yoshikawa Kōbunkan, 54) Vol 3, p 162 for the Kagawa, who in the Sengoku period would have a stronghold at
Tadatsu; Vol 8 p 636 for the Sogōshi, who in the Sengoku period would abandon their Hosokawa affiliation and join
with the Miyoshi based in Awa; and Vol 14 p 61 and Kagawa-ken (Japan), Kagawa Sōsho (Tokyo: Meicho Shuppan,
1972), Vol 1 p 685 and Vol 2 p 280 for documents referring to Yasutomi as a magistrate (bugyō) affiliated with the
Hosokawa family. The first is undated, while the second is from 1462.

83

Kagawa and Yasutomi were equally considered proxy provincial constables (shugodai 守護代), while the
Sogōshi remained as proxy administrators until the late sixteenth century.
Unfortunately there is very little documentation regarding these families’ administration of the
ports, but as certain notations in the Register of Incoming Ships at the Hyōgo North Gate demonstrate,
their influence in the region was certainly considerable. A number of entries in the Register reflect some
form of toll exemption. Thirty-four of these are marked as “kokuryō” (国料) and an additional sixty-nine
have a “kasho” ( 過書) notation. The kokuryō notation dates back to the Nanbokuchō era, when tax
exemptions were given to building materials being shipped through a checkpoint, that were to be
presented as tax or tribute items themselves.
172
Kasho denoted a pass through the checkpoint that
exempted some taxes, but not necessarily the entire cargo.
173
Twenty-five ships coming from the Sanuki
ports of Aji, Katamoto, Tadatsu, and Utazu were granted kokuryō toll exemptions under the names of the
Kagawa, Yasutomi, or Sogōshi, amounting to nearly two-thirds of all kokuryō exemptions in 1445. Of
these ports Aji and Katamoto could be considered “production ports,” since taxable cargoes from twelve
ships registered to those ports were almost universally salt produced in Katamoto.
174
In contrast Tadatsu
sent only eleven ships to Hyōgo, while Utazu sent forty-seven. But both ports shipped a wider variety of
products, ranging from various salts to rice, grains, and dried sardines, suggesting that they functioned
more as shipping ports. Utazu was also the seat of the provincial governorship for the Hosokawa,
indicating its overall importance to control of Sanuki, and therefore it would naturally have been a
destination for any number of products.
In looking at the associations between toll exemptions and various ships and cargoes, the
realization of administrative power over ports becomes clearer. Seven boats from Tadatsu and one from

172
Kobayashi Yasuō, “Irifune Nōchō ni miru kokuryō to kasho,” in Hyōgo Kitaseki Irifune Nōchō, ed. Hayashiya
Tatsusaburō (Tokyo: Chuo Koron Bijutsu Shuppan, 1981), 291.
173
Ibid., 294. Many of these exemptions are noted as for “court duties” (kuji 公事), but the named exemption
holders include Hachiman Shrine, Shōkokuji temple, Tenryūji temple, and the Ōuchi family.
174
Aji was located on the same peninsula as Katamoto, so salt produced near either port likely was designated
Katamoto salt. Of the exceptions, one Katamoto-registered ship carried ten koku of Sanuki-baled rice along with 480
koku of Katamoto salt, and another carried ten koku each of wheat and barley in addition to 460 koku of Katamoto
salt.

84

Utazu were exempted from paying tolls (affixed with the kokuryō notation) at Hyōgo under the Kagawa
name, while eight more from Utazu and one from Katamoto were exempted under the Yasutomi name.
Four additional Aji vessels and six additional Katamoto vessels received exemptions under the Sogōshi
name. The other five non-exempt Katamoto ships all arrived in Hyōgo on the same day as other toll-
exempt vessels from Aji under the Yasutomi name, or additional Sogōshi boats from Katamoto or Aji.
This suggests that the Sogōshi proxy administrators, holding less total power than the proxy provincial
constables Kagawa and Yasutomi, relied on safety in numbers by sending multiple toll-exempt ships
together or sailing their regular vessels with ships receiving exemptions (and likely protection) from the
more powerful Yasutomi family. The overall power structure in the region was even reflected in the
Inryōken, a travel record from a Chinese mission in 1493 that stated “There are thirteen districts in Sanuki.
Six are the Kagawa lands, seven are Yasutomi [as is Shōdoshima].”
175
The Sogōshi influence is omitted
from the envoy’s records. While they still held enough power to command certain toll exemptions, their
control of the ports and ships was not equal to that of the Kagawa or the Yasutomi.
A final example of local powers attempting to use control of maritime shipping as a way to assert
additional rights is that of the Ikuchi, a branch of the Kobayakawa family who administered much of Aki
province (ryōshu). The case of vessels sailing under the Ikuchi clan’s mark (from Ikuchi island,
associated with the port of Setoda in the Register) exemplifies the waxing and waning of such power and
the potential for its abuse. In 1422, the Ashikaga shogunate military commander Hatakeyama Mitsuie
issued an exemption to Ikuchi vessels for both northern and southern checkpoints at Hyōgo, but in the
following year a complaint was lodged that not only official vessels but also boats carrying commercial
goods were passing through the checkpoints without paying. The exemption was subsequently revoked.
176

Unfortunately for the Ikuchi, the shogunate then tightened its own control as it quickly moved to block
the Ikuchi’s attempts at tax evasion. Kōfukuji, administrator of the southern checkpoint, held this case up

175
Kodama Kōta and Kihara Hiroyuki, eds., Kagawa-ken no rekishi, vol. 37 (Tokyo: Yamakawa Shuppansha, 1997),
114.
176
Setoda-chō kyōikuiinkai, Setoda chōshi shiryōhen (Hiroshima-ken: Setoda-chō Kyōikuiinkai, 1997), 34–35.
Documents 13 – 15 pertain to this situation.

85

as an example of the need to abolish all such privileges, but the fact that the Register still notes a number
of partial and full exemptions reiterates the fact that certain individual powers could still command
exemptions.
177

These examples make it clear that particularly with the rise of local powerful individuals, control
of ports or shipping became a key aspect of exercising power. Either through using influence to control
sailors or to take over the port itself, as in Onomichi’s case, or by using power to command exemptions,
such as in the Sanuki provincial officials’ situation, individual associations with power were intertwined
with associations with the ports.

Port Governance: Final Thoughts

One commonality in the administration of ports and checkpoints both large and small is that control over
those places resulted in direct economic gains. Regardless of whether it was a shipping or a production
port, or a checkpoint that monitored ships passing through, the proprietor had direct access to goods
arriving or leaving those shores. Larger official temples and shrines such as Tōdaiji and Kōfukuji lobbied
directly for the ability to oversee the checkpoints and obtain revenues. As central governing powers
weakened and local authorities gained strength, they exerted their own influence on the creation (or
cessation) of port facilities, shipping, and local markets. Manipulation of manpower, such as we see in the
examples of the Onomichi custodian’s hold on sailors or the Hosokawa proxy families’ shipping
collaborations, was another way to tighten control over a port area. The ability to command and maintain
exemption from fees at the checkpoints was perhaps the strongest indicator of local power, as individuals
and institutions alike claimed that right when shipping their goods through large and small checkpoints.
The overall trend toward decentralized power in the late medieval period was reflected in increasingly
local governance over shipping and tolls.

177
Kobayashi, “Irifune Nōchō ni miru kokuryō to kasho,” 295.

86

Culture: People of the Ports

Differences between production and shipping ports were reflected in the livelihoods of their residents.
Fishing and saltmaking were key occupations of production ports, and exploring the written record in
conjunction with archaeological evidence provides insight into the roles of those groups in the community.
Shipping ports, as seen in the previous section, often experienced a stronger physical presence of
administrative powers in the form of proprietorial overseers or representatives. They also likely had a
group of resident warehouse managers to help process the intake of goods passing through the port. While
women are rarely specifically mentioned in documents referring to seaside communities, one of their
roles in shipping ports is stated bluntly by one early twentieth-century historian of Onomichi: “Port towns
have prostitutes.”
178
As we shall see, to a certain extent, the type of port shaped the lives of the residents.
Saltmaking and Ports

Salt production along the Inland Sea was a critical aspect of port communities, especially in its influence
on the development of local economies, usurping the traditional place of rice or other dry field crops as
the major form of payment for rents and dues. Yuge Island has been studied extensively and has become
most representative of medieval salt production. Peter Shapinsky’s study of Yuge salt production reveals
the manifestation of the mariner’s perspective in this coastal community, reinforcing the idea that people
associated with the sea have distinct practices that shape their identity. Standard practice at this time was
to divide land into taxable production units (myō 名). Overseers took a portion of the crops – usually rice
– cultivated there. Since Yuge’s main product was salt, although there were smaller fields that grew crops
such as wheat and mulberry, the rents and dues for those lands were assessed in salt. As time went on, salt
production far outpaced the capability to produce crops on dry fields, and rents and dues from the latter
did not reflect the actual value of goods produced on the island. So by 1313 salt was incorporated into the
production unit assessment.
179
Though Shapinsky sees this as an attempt to adapt terrestrially based

178
Onomichi Shiyakusho, Onomichi shishi, 1:542.
179
Shapinsky, Lords of the Sea, 74.

87

systems to the littoral environment, it can also be seen as the mariner’s community making their own
mark on the terrestrial practices and forcing a recognition of the marine-based contributions to the system.
Although the littoral communities did need to conform to the assessment of the value of myō
lands, incorporating salt fields into the assessment was unique to the littoral communities and helped
establish salt as a “brand name.” As will be discussed in Chapter 3 in more detail, salt shipments in the
Register are usually labeled only with the name of the area in which it was produced, a convention not
followed for any other product except for sesame.
180
Instead of a forced adaptation of terra-centric
practices, using salt production in Yuge and other areas along the Inland Sea as a measure of production
worth was an assertion and recognition of the value and importance of salt by both the local producers
and governing assessors.
Salt’s value is rooted in historical precedent. Evidence of salt trade can be traced back as far as
the Jōmon era (approximately 12,000 – 300 BCE), and it is mentioned in wooden slips (mokkan 木管)
from the Heijō-kyō area and in the tenth-century Engishiki as an item critical for ceremonies.
181
It was a
key product throughout the coastal areas of Japan, produced by commoners, shrine affiliates, and lower
class cultivators; and from early on it became an element of the rents and dues (nengu) that were shipped
across the Seto Inland Sea region to the capital.
182

As noted in Chapter 1, saltmaking scenes became a literary trope embodying the atmosphere of a
coastal community, with smoke wafting up from fires stoked to evaporate the seawater. Those fires hint at
an important element of the overall composition of saltmaking communities: the need for fuel. Amino
notes that in records from Yugeshima Estate, holders of tax-free paddy lands who paid rents and dues in
salt also had forested lands for a readily available fuel source under their purview.
183
By the fourteenth

180
Although rice shipments in the Register do sometimes have place names recorded, those names refer to the
style/size of the bales, not necessarily the production point. Further, those entries are recorded as “Rice – Buzen
Bale” or “Rice – Awaji Bale,” while salt entries do not even record the product itself but only have the production
point name (“Bingo,” “Aga,” “Shiwaku,” etc.).
181
Amino Yoshihiko, Chūsei minshū no seigyō to gijutsu (Tokyo Daigaku Shuppankai, 2001), 82–84.
182
Hiroyama Gyōdō, Shio no nihonshi (Tokyo: Yuzankaku books, 1997), 45.
183
Nagahara Keiji and Yamaguchi Keiji, Koza nihon gijutsu no shakaishi, dai 2 kan: Engyō gyogyō (Tokyo: Nihon
Hyōronsha, 1985), 56.

88

century, saltmaking shifted from agehama ( 揚浜) to irihama ( 入浜) fields, resulting in increased salt
production. The agehama fields involved physically transporting seawater to a sandy area, allowing the
water to evaporate and the salt to remain. Irihama salt fields piped in seawater (fig. 20) to saturate a sandy
area, again letting the water evaporate and the salt to remain. In both cases the sand would have been
rinsed with seawater, further increasing the salinity, and boiled to leave the salt behind. The irihama style
decreased the amount of labor needed, allowing for increased production over a wider space. Production
decreased during the rainy season, as constant saturation slowed the repetitive evaporation and
resalination process.
184
The whole process, then, required a significant area that included forestlands to
facilitate the entire endeavor.

Figure 20. Photo of reconstructed irihama-style salt flat in Utazu, Kagawa prefecture. Note the irrigation ditch that allows
seawater to flow directly from the ocean to the salt flats. Photo by author (2012).

184
Shapinsky, Lords of the Sea, 75–76. Shapinsky details the salt production and land allocation on Yugeshima
Island, suggesting that the forest lands may have been a conscious attempt to create a sustainable fuel source.

89

An important clue to the layout and workings of medieval saltmaking is found in the Shidoji engi,
a late Kamakura or early Nanbokuchō era scroll that depicts the origin legend of Shido Temple (in Sanuki
province; modern Kagawa prefecture). In the corner of one of the first scenes is a small salt flat at the
mouth of the Benten River. In the foreground is a man leading a horse and drawing a harrow, smoothing
out the sand of the salt flat to evenly distribute the seawater for the initial evaporating process. Nearby a
man carries a bucket, likely either bringing salt water to be spread on the sands or carrying salt-drenched
sand. To their right are nine small buildings, with a red glow emanating from the entrances and smoke
wafting upward. Nearby are three other men, two holding poles, and one is poking the pole into a nearby
building. These were shacks where the highly salinated water was heated in the final stages, and excess
water evaporated. Another small building with an oven-like structure stands slightly apart, also with
embers aglow. This may have been a remnant of an earlier saltmaking process in which water was
steamed down in earthenware or iron pots over a fire.
185
Though it is difficult to determine the scale of
this saltmaking operation from the image alone, it does evoke a sense of the manpower necessary for the
process. At least five people were involved, between smoothing the sand, carrying water, and stoking the
fires, plus the horse that pulled the harrow. Nine buildings were devoted entirely to the firing process,
suggesting the large scale of the production as well as the amount of firewood necessary to maintain the
evaporation. Sites that produced enough salt to be used as payments of rents and dues or to be used as a
trade good would have required significant manpower, land, and resources to ensure a smooth and
sustainable operation.
This suggests that production ports that focused on saltmaking likely had a greater percentage of
land devoted to salt production facilities than did shipping ports. Archaeological evidence provides some
support for this theory. Seventy-eight sites within Kagawa prefecture alone have yielded evidence of salt
production, although many of them date to ancient or even prehistoric times. This does not represent a
comprehensive survey, as these sites have been discovered mostly through preliminary excavations
undertaken prior to a construction project rather than through a systematic survey. Even without a

185
Karaki and Hashitsume, Chūsei no Sanuki, 184–185.

90

thorough exploration of the entire coastline in the hopes of unearthing evidence of saltmaking, however,
certain trends are apparent. The majority of saltmaking facilities from any era have been found on the
islands of the Inland Sea, with only twenty or so such sites along the Shikoku coastal area.
186
This
suggests that far more land was devoted to saltmaking on the smaller islets, rather than on the mainland.
Chapter 3 will explore salt trade patterns in greater detail, but here we should also note the relative
volume of salt shipped from various locales within the Inland Sea to Hyōgo in the late medieval period
(fig. 21). Specifically, the vast majority of salt from the modern Kagawa prefecture area was sent directly
from the islands to Hyōgo – note on figure 21 locales such as Shiwaku or Shima, each producing its own
salt. Certain ports such as Udatsu on the mainland also shipped significant quantities of salt to Hyōgo, but
a closer look at the types sent reveals that many of those salts were not produced locally in Udatsu.
Instead, Udatsu was a shipping port, handling salt from nearby locales (Takuma, Hiketa, and Shiwaku
salt) as well as salt from further areas such as Bingo or Katamoto. While not conclusive evidence, the
quantity of archaeological sites on the islands (i.e., closer to production ports) combined with the relative
lack of such sites closer to shipping ports may be an indication that medieval ports specializing in
shipping set aside less land for saltmaking.
187

Salt production was ubiquitous throughout the Inland Sea, becoming a symbol of littoral
communities in literary tropes. The physical space and resources necessary for saltmaking is apparent in
the artistic depiction of salt flats and in archaeological evidence, indicating its prominence in the daily
lives of seaside residents. In Yuge’s case it supersedes rice as the most valuable economic component of
the estate. Salt and salt production were critical components that shaped the image of medieval maritime
communities.

186
Kagawa-ken Kyōikuiinkai and Honshu Shikoku Renraku Hashi Kōdan, Ōurahama iseki honbunhen, Seto Ōhashi
kensetsu ni tomonau maizō bunkazai hakkutsu chōsa hōkoku V (Kagawa Prefecture: Kagawa-ken Kyōikuiinkai,
March 1988), 23–25.
187
Ibid., 25. The authors of the Ōurahama site report hypothesize that the lack of saltmaking sites along the Shikoku
coast may have been due to a general lack of saltmaking in those areas, or that perhaps earlier sites were razed
entirely in later centuries. If the latter case, however, it is unlikely that there would be absolutely no archaeological
signature in multiple areas.

91

Figure 21. Map of large volumes of salt (more than five hundred koku) being shipped to Hyōgo in 1445.

92

Fishing and Ports

The ability to procure marine goods was another important characteristic of maritime communities. That
fact notwithstanding, written records that help us understand the scale of fishing near ports are
unfortunately scarce. Even Amino Yoshihiko, premier voice for understudied maritime communities in
Japan, admits that there are few documents remaining from the medieval period and earlier that describe
fishing villages and customs.
188
A few travel diaries provide hints as to the makeup of fishing
communities. Imagawa Ryōshun’s Michiyukiburi (1371) chronicles his journey from the capital at Kyoto
to Kyushu, and he notes upon his approach to Onomichi that “in passing by the rocky shoreline the homes
of the fisherpeople are close to the mountains.... Along the foot of the mountain the homes stand built
close together, with barely even enough room in the garden to dry the [fishing] nets.”
189
Fifty years later,
the Korean envoy stopping at Onomichi described a similar scene: “The residences were along the
shoreline, and the homes were touching.”
190
The authors evoke an image of the crowded area around the
bustling port. Even in smaller areas, the residences seem to have been close to shore, as the Korean envoy
also noted when passing by Kamadoseki in Suō province: “The priest’s residence is along the mountains,
and the people’s residences are near the water, with fishing boats tied slightly further off.”
191
It is possible
that the “priest’s residence” is a temple building, which may be indicative of the tendency noted above
that highly visible pagodas or religious buildings served as navigational markers. The villagers
themselves lived closer to the shoreline and their ships, allowing for easier access to the sea.
A fourteenth-century map from Yuge provides an important clue to the organizational aspect of
fishing (fig. 22) in a smaller community. Several areas on the map are designated as “net sites,” which
refers to the area designated for larger-scale net-fishing operations. The area extends from the shoreline
out into the sea, implying that the right to administer the fishing areas required specific delineation.
192
The

188
Amino Yoshihiko, Umi to retto no chūsei (Tokyo: Nihon Edita Sukuru Shuppanbu, 1992), 308.
189
Imagawa Ryōshun, “Michiyukiburi,” in Chūsei nikki kikōshū : Kaidōki, Tōkan kikō, Ben no Naishi nikki, Izayoi
nikki hoka, trans. Toshinori Inada (Tokyo: Shogakkan, 1994), 398.
190
Song, Rōshōdō Nihon kōroku: Chōsen shisetsu no mita chūsei Nihon, 86.
191
Ibid., 82.
192
Shapinsky, Lords of the Sea, 78.

93

fact that there were only five fishing net sites designated along the coastline suggests that fishing was
limited to those particular areas.

Figure 22. Map of Yuge Island, fourteenth century. Shapinsky (2014, p. 78)
The tendency to limit fishing sites is also suggested from evidence in the archaeological record.
The Ōurahama site on Hisuishi Island in Kagawa prefecture comprises the centermost area of a C-shaped
cove on the island. Had fishing been conducted without any restrictions, it would have been likely that
discoveries of fishing weights (used to weigh down nets; fig. 23) would have been equally distributed
along the shoreline, indicating no difference in overall usage. Instead, almost all types of fishing weights
were excavated at either end of the site, with a wider unused space in the middle of the cove (fig. 24). The
single exception to this trend was in simple tube weights, which were found throughout, but with the
greatest concentrations discovered at either end of the site.
193
As it is difficult to date fishing weights with
great certainty, these maps represent all weights that date from after the Yayoi period, representing a wide

193
Kagawa-ken Kyōikuiinkai and Honshu Shikoku Renraku Hashi Kōdan, Ōurahama Iseki Honbunhen, 157–160.

94

time frame. In general, later eras tended to use larger fishing weights, reflecting the growing ability to
build larger boats that could haul wider nets and transport heavier cargoes. As such, the map showing
discoveries of large grooved weights (second from left in figure 23) at the ends of the site most reliably
indicates medieval usage patterns. Indeed, large grooved weights have been found in late medieval sites
such as Kusado Sengen, which provides additional evidence that those from the Ōurahama site were
indeed from the Kamakura era or later.
194
Though not pictured on the site map here, octopus jars (tako-
tsubo 蛸壺) were also unearthed almost exclusively in the lower section of the site. This may be an
indication of a particularly limited fishing site, either due to administrative restrictions or smaller numbers
of octopi elsewhere near the shore. Regardless of type of fishing, the concentrations of remains at the
ends of the site suggest that, similar to the restrictions placed on Yuge’s net sites, fishing in the Ōurahama
area was limited as well.

194
Ibid., 156.

95

Figure 23. Various thirteenth - fifteenth century fishing weights from the Hamanochō excavation site, Kagawa prefecture.
Similar types of weights were found at Ōurahama. (Photo by author; taken with permission from the Kagawaken Maizō
Bunka Senta-. 2012).

Figure 24. Discoveries of fishing weights in Ōurahama Cove. From left to right, the maps indicate rod-shaped weights;
large, medium, and small grooved weights; and single-groove tube weights. Kagawa-ken Kyōiku Iinkai and Honshu
Shikoku Renraku Hashi Kōdan (1988, 159-160).

96

The lack of both written records and a systematic archaeological survey of seaside communities
creates difficulties in determining differences between production ports and shipping ports with regards to
fishing. Information from the Hyōgo Register indicates that a significant amount of various types of fish
were shipped from ports throughout the Inland Sea, as will be discussed in more detail in the following
chapter. Shipping patterns for sardines (the most widely shipped sea creature) reveal one noteworthy
trend (fig. 25). While transshipment hubs did send some sardines to Hyōgo, the greatest quantities were
sent from smaller ports (most notably Nohara and Muro), suggesting that they focused on procuring the
fish directly rather than concentrating on trading. Although a comparable archaeological study has not yet
been done at Muro, excavations around Nohara have yielded fishing weights, corroborating the idea that
fishing was a key activity at this port.
195
In contrast, the largest amounts of salted fish (usually snapper)
were shipped to Hyōgo from transshipment hubs such as Utazu and Ushimado. Salted fish was preserved,
and could well have been brought from other regions, traded at those transshipment hubs, and later sent
on to Hyōgo.

195
Sato, Kanehara, and Kitano, Takamatsu-jō seki (Nishinomaru-chō chiku) Sanpōto Takamatsu shogo seibi jigyō ni
tomonau maizō bunkazai hakkutsu chōsa hōkoku 2 Volume 1, 85, 87, 89, 91–92.

97

Figure 25. Sardine shipping in the Register.

98

The relative lack of information about fishing practices and villages is surprising, but in
combining the information from travel accounts, the Yuge estate map, archaeological evidence, and
shipping records, the picture becomes slightly clearer. As evidenced by the authors’ impressions of their
travels along the Inland Sea, fisherpeople lived close to the water in both major shipping areas and
smaller ports. The Yuge map and locations of fishing weights suggest that large-scale fishing was
regulated, limiting it to certain areas. Production ports focused on sending what were likely freshly-caught
fish to market towns, while shipping ports not only sent items that were locally produced, but also had the
advantage of shipping or transshipping preserved fish. Additional archaeological studies will help clarify
more about fishing practices in communities of all sizes.

Women and Port Towns

It is difficult to determine if women had specialized roles in the everyday life of the port town. In one of
Imagawa Ryōshun’s poetic tributes to the sights he visited en route to the Dazaifu, he claims that the bay
of Mekari was “named for the fisherwomen harvesting seaweed,”
196
but it is unclear whether seaweed
harvesting was considered women’s work everywhere. Archaeological evidence similarly reveals little
information specifically about the roles of women, although it does hint at their daily lives. For instance,
excavations at Kusado Sengen have yielded hair ornaments and implements for teeth blackening, but the
items were not necessarily used only by women.
197
Their presence does suggest, however, that some elite
fashion trends enjoyed popularity throughout the late medieval period even far away from the central
court.
As discussed in Chapter 1, the presence of sexual entertainers (yūjo 遊女) in ports was noted in
the early medieval period, as accounts from the retired tennō Takakura’s pilgrimage to Itsukushima

196
Imagawa, “Michiyukiburi,” 401. He says that Mekari Bay was near Kujirajima, a name for two small islands in
between Onomichi and Numata in Aki province. It is unclear what Kujirajima actually refers to, but it may have
been part of Mihara-chō or Mukaishima near Onomichi.
197
Matsushita, Umoreta minatomachi Kusadosengen, Tomo, Onomichi, 100–101.

99

Shrine (1180) and Lady Nijō’s travel memoirs (1307) show. Although the twentieth-century history of
Onomichi unabashedly declares an affiliation between port towns and pleasure women, the latter were
generally found in shipping ports, not necessarily production ports. The earlier travelers’ accounts refer to
Onomichi, Tomo, and Muro as places where they encountered pleasure women. By the late medieval
period, Muro appears to have become less of a shipping hub, but Onomichi and Tomo continued as
important shipping ports. Imagawa Ryōshun, in his description of Onomichi, wrote not only of the many
ships moored in the harbor but also of the “one-night wives” who were “like small birds, coming back
and forth on the smaller boats, floating and drifting.”
198
The “one-night wives” likely did not provide only
physical services to the travelers, but also probably took care of more mundane chores such as washing
clothes.
199
The highly-trafficked shipping ports, servicing ships from far-off regions, would have required
those types of services more than did production ports, which probably saw far fewer travelers and
focused more on shipping their goods to other areas. Watanabe Keiji further suggests that the highly
mobile population of port towns was attractive to itinerant entertainers such as biwa players, imayō
singers, and blind female musicians (goze 瞽女). In the area around Shimonoseki’s Inari-machi, for
instance, their performances gained such recognition that they became an important asset for the port
town and even took part in local festivals. He suggests that this was unique to port towns, as in smaller
communities these performers were often barred from participating in religious festivals.
200
As smaller
production ports such as Yuge have little to no mention of pleasure women or performers, it seems that
contrary to the Onomichi historian’s declaration, not all ports had pleasure women – just the ones
focusing on shipping and trade over production alone.

198
Imagawa, “Michiyukiburi,” 398.
199
Aoki, Onomichi shishi (Shinshū), 1971, 1:398.
200
Haneda Masashi, Rekishigaku kenkyūkai, and Watanabe Kenji, “Minatomachi no yūkaku to yūjo - Shimonoseki
Inarimachi wo chūshin ni,” in Minatomachi ni ikiru (Tokyo: Aoki Shoten, 2006), 41.

100

Sociologies: Conclusion

Although it is difficult to fully understand the sociologies of port towns, piecing together literary,
documentary and archaeological evidence has revealed clues to the lives and practices of residents. While
sharing some common characteristics, the individuals and activities associated with port towns
demonstrate a certain amount of specialization according to the type of port. Production ports devoted a
certain amount of land and manpower to the procurement of particular goods. Salt fields required a large
area to heavily salinate the water and subsequently evaporate it, which in turn called for significant
forestland for firewood. Fishing areas were likewise regulated, restricting large-scale net fishing to certain
regions offshore. Shipping ports were crowded with people. Descriptions of Onomichi consistently
remark upon the closeness of the houses, suggesting a density not found in smaller production ports. The
constant flow of people required different services than production ports, and comments on the presence
of women reflect a more diverse population. Further systematic study of these areas will hopefully reveal
more about their residents.

Outliers: “Pirates” in the Inland Sea

“Pirate” ports were a breed apart.
201
Successful pirate families established a wide network throughout the
Inland Sea, establishing enclaves throughout the smaller islands and threatening travelers and shipping
concerns. Yet as they were involved in illicit activities and needed to remain undiscovered, they needed to
keep their ports relatively small and inaccessible to all but the most skilled mariners. Pirates in the Inland
Sea therefore needed to create their own specialized communities that could both support successful
maritime operations and yet remain relatively unassailable.

201
In his Lords of the Sea, Peter Shapinsky argues convincingly that the translation of wakō or kaizoku as “pirate” is
something of a misnomer, particularly in the late fifteenth and early sixteenth centuries. He suggests the term “sea
lord” instead, as the maritime families attempted to use land-based conventions to secure their place as rulers over
sea spaces compatible to rulers over land spaces (pp 9 – 18). Here, however, I choose to use the term “pirate” to
characterize these seafarers. The push for legitimization occurred slightly after the time period that is the major
focus of this study, and travelers and legal documents alike use the term kaizoku in a more pejorative manner that
calls to mind the illicit actions of piracy rather than sea lords. It should be noted that many of the early pirate clans
would indeed become more accepted “sea lords” during the Sengoku era and later.

101

Returning to Kitayama’s criteria for successful ports, pirate enclaves defied most of the
established standards. Most of the Inland Sea “pirate strongholds” (umishiro 海城) were located on
smaller, remote islands, to block easy access from the mainland. Navigational markers of any sort were
virtually nonexistent. Often pirate strongholds were located on islands that were surrounded by strong,
rapid currents, making it extremely difficult to navigate to the island shores.
202
(fig. 26) This of course
protected the skilled seafarers who used those strongholds, as only the initiated would be easily able to
navigate the currents, and outsiders would be deterred from approaching for fear of being dashed to
pieces by the rapids. The strait between Hakata and Ushima islands near Noshima Island (modern Ehime
prefecture; likely Iyo province in the medieval period), site of one of the Murakami pirate family
branches, even today bears the name “Funa-ori,” or “boat-breaking.”

202
Yamauchi Yuzuru, “Kudako-jō no ikō to Kudako-shū,” Iyo Shidan 300 (January 1996): 28. Yamauchi describes
the rough rapids that separate Uwama and Kudako strongholds, which were only about 1.2 km long but were
extremely difficult to navigate. Even today, tourists can take a “tidal currents experience” boat tour off Noshima
Island (site of a Murakami pirate stronghold) and see the high-speed currents that medieval pirates would have
maneuvered with oars and sails (Noshima suigun chōryū taiken:, “Chōryū taiken: Noshima Suigun ha Setouchi
Shimanami kaidō/Miyakubo Seto,” accessed March 23, 2015, http://www.noshima.jp/choryu.).

102

Figure 26. The swirling currents facing the coast of Noshima island. Photo by author (2013).
Mooring ships and loading or unloading cargo on these islands would have followed a different
process from that used in licit shipping and production ports. While a safe cove to debark and load or
unload cargo was still important, a permanent infrastructure would have defeated the purpose of making
the strongholds inaccessible to outsiders. The best evidence of the pirates’ solution to this dilemma comes
from Noshima Island. Along the rocks on Noshima’s shoreline, holes of various sizes have been hollowed
out and they are thought to have been for posts used to moor incoming vessels (fig. 27).
203
After the ships
left the island, the posts could have been removed and stored either on shore or on the vessels themselves.
The temporary nature of this type of makeshift mooring point would allow for easy removal to deter
unwanted visitors while disguising the embarking point.

203
Imabari Kyōikuiinkai, “Chūsei kaizoku no shiro: Noshima iseki,” Bunka Kyōkai Ehime, August 2008.

103

Figure 27. Two postholes (front right and far left) on the "funadamari" terrace at Noshima island. Photo by author (2013).
Governance of pirate strongholds is a murky issue. On one hand, the popular perception of a
pirate was as a figure to be feared, as seen in Imagawa Ryōshun’s musings at the beginning of his journey
to Kyushu. Near Ōkuradani, he noted that, “I heard rumors of pirates targeting the passenger ferries for
attack, and so in a panic rushed past it.”
204
Even more telling, however, are the reactions of the Korean
envoy Song Huigyong in his travel diary of 1420. He recounted several instances on his journey to Kyoto
of perceived or actual “pirate” encounters, beginning with a description of the need to moor for the night
on the Kyushu coast opposite Dannoura near Shimonoseki, “a place where pirates lurk.” Upon the
approach of a smaller vessel that turned away upon the sight of their escort ships, everyone became wary,

204
Imagawa, “Michiyukiburi,” 394.

104

and a strange noise during the night kept Song awake in fear.
205
He subsequently composed a poem
worrying that all of the coastal residents were actually pirates, and he wrote that his “thoughts were in a
daze.”
206
Later, in the waters near Aki province, Song’s ship actually encountered a small pirate vessel
and the crew “beat the drum, raised the flag, blew the horn, rang the bells, put on helmets, and stood with
arrows nocked” to defend their ship. Song’s escort vessels also came to their defense, and the pirate ship
quickly sped away.
207
Even under the protection of the escort vessels (who may have been sea lords
themselves), pirates were a force to be feared.
Piracy in the Kamakura period is the maritime parallel to the rise of terrestrial akutō (evil bands
悪党) and is characterized by the increased documentation of maritime protection.
208
As violence
escalated on land in the form of uprisings from the evil bands, the same lawlessness was feared on the
seas. This concern was reflected in the written record by previously unseen references to “protection on
the sea” (kaijō keigo 海上警固). In the early fourteenth century, documents from Harima, Aki, and Suō
refer to the need for such protection, and indicate that “protection offices” (keigo yakusho 警固役所)
were in place in those provinces.
209
By 1420, Iyo province was also implementing similar measures.
210

Protection against seaborne threats was a concern for these areas. Shapinsky notes that by the sixteenth
century, the range of activities encompassed by the term keigo (protection) had expanded to include what
he terms “violent services,” including, ironically, the suppression of piracy.
211
The Noshima Murakami
best exemplified the institutionalization of what Shapinsky terms the “sea protection business,” as they

205
Song, Rōshōdō Nihon kōroku: Chōsen shisetsu no mita chūsei Nihon, 81. The following morning Song and his
companions realized that the “strange noise” was merely the cry of a pheasant.
206
Ibid., 82.
207
Ibid., 84.
208
This term would come to take on a different meaning in later years, referring to the payment to pirates of escort
fees.
209
Amino Yoshihiko, Akutō to kaizoku: Nihon chūsei no shakai to seiji, Shohan (Tokyo: Hōsei Daigaku
Shuppankyoku, 1995), 257–259. See Tōji hyakugō monjo, kyō #126, Miura-ke monjo #9, in Dai nihon komonjo,
Iewake Vol 14. Though Amino’s book title translates to “Evil Bands and Pirates,” of the entire volume only one
brief chapter is devoted to early piracy, focusing on the final years of the Kamakura era.
210
Matsuoka Hisato, “Nachi suigun to Nanbokuchō no dōran,” in Kōno-shi to Iyo no chūsei (Ehime Prefecture:
Ehime-ken Bunka Shinkō Zaidan, 1987), 99. Matsuoka quotes a communiqué from the Rokuhara government,
located in the Sonkyōkaku bunkosho Musashi-ke shukan 尊経閣文庫所武蔵家手鑑 Gen’ō 3 (1321) 2.13.
211
Shapinsky, Lords of the Sea, 60–61.

105

established protocol for accepting fees for protection and escort services, overseeing harbor passes, and
accepting gifts as “processing fees.”
212
As many of the terrestrial powers worked in conjunction with the
Noshima Murakami, the latter’s role became more ambiguous – were they illegal pirates, or legally
sanctioned sea lords? There was no definitive answer to that question in the medieval period.
While there is less documentation available for piracy in the Nanbokuchō and Muromachi periods,
there are some hints at the difference between “evil band” types of pirate and specialists in maritime
protection that used piracy as a form of sanctioned violence. There is some evidence of the latter in
examples such as Ashikaga Takauji’s agreement with Kumano-based pirate bands in 1340 to protect trade
and tax ships within the Inland Sea.
213
In 1349, Tōji temple contracted with the Noshima pirate clan to
perform guard duty on Yuge Estate in return for rice and other gifts.
214
Nearly a century later, the
Murakami family name came to the fore, as a member of the Kurushima branch of the Murakami wrote to
Tōji regarding his control of Yuge Estate in 1456.
215
The author describes his willingness to mediate with
others who were laying claim to parts of the island, including other pirate families such as the Noshima
and Yamaji. In addition, a letter six years later from a Tōji official to his superior names the Noshima and
the Sanuki-based Yamaji clan as pirates, noting that they provided services for the Kobayakawa, who
were proprietors (ryōshu 領主) of much of Aki province in the late medieval period.
216
The implication is
clear that sea lords who were perceived as allies were “not pirates” while those who perhaps provided
identical services for another lord were “pirates.”
By the Sengoku period, the great sea lord clans that included the Murakami and Noshima families
began to command attention, as documents indicate more and more often that they were working in

212
Ibid., 122–124.
213
Ibid., 89.
214
Ibid., 88. Also see Nihon Senbai Kosha, Nihon engyō taikei, vol. Shiryō-hen Kodai, chūsei (Tokyo: Nihon Senbai
Kosha, 1980), 210 – 211, #139 Jōwa 5 (1349) 12
th
month.
215
Yamauchi Yuzuru, Chūsei Seto Naikai chiikishi no kenkyū, Shohan (Tokyo: Hōsei Daigaku Shuppankyoku,
1998), 119. Yamauchi suggests that an earlier estate custodian (azukaridokoro) of Yuge in 1420, Murakami Uemon
no jō, may have been a member of the Kurushima branch of the Murakami. Shapinsky also describes this case in
Lords of the Sea, 2014, 93.
216
Shapinsky, Lords of the Sea, 92–94. Also see Nihon engyō taikei, vol. Shiryō-hen Kodai, chūsei, 414, #291-3
Kanshō 3 (1462) 5.17.

106

conjunction with terrestrial lords. As Peter Shapinsky describes, in the later fifteenth and early sixteenth
centuries the pirates began to consciously imitate terrestrial-based systems and appropriate their
terminology in writing official documents. Shapinsky sees this as one way in which the pirate clans strove
toward legitimizing their role as lords of the sea, on equal footing with the daimyō on land. One such
example of this can be seen when Noshima “pirates” swore reciprocal loyalty oaths with terrestrial lords
such as Mōri Motonari to “perform services,” which makes it seem that some pirates were accepted into
the general domainal hierarchy of the Sengoku era.
217
We might argue, then, that the late medieval pirate
was a transitional figure moving from the fearsome image held by Song and Ryōshun, to that of an image
not that different from the terrestrial daimyō.
While we can gain insight into pirate relationships with terrestrial powers through documents
such as those noted above, pirates rarely left written records describing their lifestyles in their strongholds.
Archaeological evidence can provide more information on that subject. Excavations on Noshima have
begun to reveal some information about lifestyles in the pirate strongholds. Most of the artifacts from the
site date from the late fifteenth and early sixteenth centuries, roughly corresponding to the transitional era
when pirates were transforming into sea lords. In addition to the mooring sites, remnants of at least five
buildings have been discovered to date, including a main stronghold and something that may have been
an armory, though no evidence of weaponry has yet been unearthed. There is evidence that the buildings
were reconstructed several times, indicating long-term, if not continuous usage of the island.
218
Most of
the buildings may have been workspaces and offices rather than residences. Though there may be a well
located within the main stronghold area, there has as yet been no discovery of a source of fresh water on
the island. Across the strait on Ōshima was a water collection site (mizuba 水場) used by the Noshima

217
Shapinsky, Lords of the Sea, 2014, 117-120. For more information on late medieval and early modern piracy, see
Shapinsky’s monograph.
218
Imabari-shi Kyōikuiinkai, Kuni shitei shiseki Noshima-jō ato. Heisei 23 nendo hakkutsu chōsa genchi
setsumeikai shiryō (Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan, March 4, 2012), 3.

107

clan.
219
Several caches of pottery and loose Chinese coins near the buildings are thought to have been
ritual deposits, buried when new structures were being erected or old ones destroyed.
220
The presence of
the coins as well as fragments of Chinese and Korean pottery hint that trade in those items (or forceful
procurement) was desirable to the pirates as well as in formal trade. Perhaps most indicative of the self-
sustaining nature of the stronghold, evidence of a forge and bellows have been found near the main
structure.
221
This suggests the likelihood that the pirates were producing their own iron products locally,
possibly including nails or other implements necessary to repair and maintain their ships. Excavations
continue on the island, which will hopefully add new information about the activities and daily lives of
the pirates or sea lords there.
Pirate ports, then, were a strongly felt presence in the Inland Sea region, but they were developed
and maintained quite differently from other types of ports discussed here. Their infrastructure was much
more minimal than that of shipping or production ports, with relatively difficult access for the uninitiated
and few buildings in the settlement. There is a stark contrast between these ports and the construction of
permanent mooring points seen at Nohara and Naoshima, or the densely populated, protected waterways
of Onomichi. The legality of pirate or sea lord actions was fluid, as they sometimes acted in conjunction
with terrestrial proprietors such as the case of Tōji and Yuge Estate, but other times independently as
pirates threatened seafarers on their own initiative. They created their own checkpoints and demanded
fees from passing vessels, or contracted with other entities to act as “escort services” for their ships. As
Shapinsky points out in his discussion of the language of oath-swearing, by later years they presented
themselves as the equals of terrestrial lords instead of conforming to an assumed hierarchy of tribute to a
proprietor. Though little is known about the daily lives of the pirate clans, the evidence from excavations
on Noshima that show the pirates’ ability to self-sustain, such as forging their own iron implements,

219
Personal communication, Tanaka Ken, January 17, 2013. I am extremely indebted to Tanaka-san for his help in
arranging a rapids tour and a guided tour of the Noshima ruins themselves, as the island is currently officially closed
to casual visitors.
220
Imabari-shi Kyōikuiinkai, Kuni shitei shiseki Noshima-jō ato. Heisei 21 nendo Hakkutsu chōsa genchi
setsumeikai shiryō (Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan, January 23, 2010), 7.
221
Imabari-shi Kyōikuiinkai, Kuni shitei shiseki Noshima-jō ato. Heisei 22 nendo hakkutsu chōsa genchi
setsumeikai shiryō (Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan, February 19, 2011), 8.

108

reinforces their independent, non-subordinate image. The lack of obvious residential areas raises
questions about where and how pirates passed their spare time, whether on the mainland or on their ships.
Further studies promise to reveal more about the lives, roles, and influence of pirate clans in the Inland
Sea.

Conclusion

In many ways the medieval port community escapes easy characterization, with few universal attributes.
Even Kitayama’s six criteria for a port vary according to the size and function of the port. The
infrastructure of ships and ports affected each other, since as ship construction improved and resulted in
larger vessels, port infrastructure needed to accommodate those watercraft if the port were to stay viable.
Ports such as Onomichi, with natural endowments such as its deep waters and protected coves as well as
geographic and manmade navigational markers, had an advantage over smaller ports such as Nohara or
Naoshima. Smaller ports required the addition of manmade mooring points to service increasingly larger
ships and cargoes. Illicit ports reflect a different tendency: not only was there no permanent mooring
infrastructure, but the locations of strongholds were chosen specifically for their difficult, churning waters.
Geography, technology, and infrastructure all varied according to the type of port in question.
The administration of ports and checkpoints became an important way of demonstrating power.
Control over a checkpoint, especially, was a critical source of revenue for both institutions and
individuals. Disputes among administrators, local residents, and the central government regarding the
rights to levy tolls or be exempt from them demonstrate the significance of the checkpoints. Control of
ships and ports, as in the cases of the custodian at Onomichi or the proxy rulers under the Hosokawa in
Sanuki, shows how local administrators used that control to heighten their own influence through
manipulation of manpower and taxes. Pirates refused to bow to terrestrial-based powers and instead met
them as equals; as lords of the sea, indeed. Local administrators, either by legal appointment or
independently as pirates, exerted control over an area by controlling the port sites.

109

Finally, port communities both created and were sustained by different types of sub-communities
that characterized the port itself. Fishing and saltmaking were the most common occupations of
production ports, requiring their own particular governance to regulate use of both land and sea. In
shipping ports, women played a very specific role that enhanced the business reputation of the port and
eased the travelers’ burdens. Pirate enclaves were apparently relatively self-sufficient, but further work
needs to be done to understand more about their daily lives.
The medieval Seto Inland Sea ships and ports, then, were shaped by infrastructure, geography,
governance, and residents, but the first two factors were more critical in determining in which direction
the port would grow and the type of vessels that could be accommodated. Natural factors (deep, sheltered
seas with easy navigation, or conversely rough, treacherous seas) often dictated the initial stages of
developing a shipping port or pirate enclave. Those areas in turn became more desirable as checkpoints or
major ports for administrative control, and welcomed different sub-communities such as the entertainers,
sexual or otherwise. Smaller ports, less naturally endowed with geographic or infrastructure assets, were
left to focus on the procurement of goods that could be shipped directly to market areas. Fewer records
indicate that production ports sustained the diverse sub-communities of the shipping ports, suggesting
instead that those residents were occupied more with fishing and saltmaking. The following chapter will
explore more fully the types of goods sent from both shipping and production ports, and will revisit the
idea that smaller ports were not always relegated only to production.

110

Chapter 3: The Sea Connects: Tracing the Flow of Trade

In the literary record of the Heian and Kamakura periods, areas further away from the center at Kyoto
were often depicted as lonely, far-off lands. Lady Murasaki wrote about the fictional Prince Genji sighing
over his exile at seaside Suma, only fifty miles from the capital. The disgraced sovereign Sutoku Tennō
was banished to distant Shikoku, separated from the mainland by the waters of the Inland Sea.
The sea was seen at best as a place of isolation and at worst as a barrier, where the threats of
storm, waves, and piracy were very real. Yet the actual evidence of seafaring as seen through records of
commercial development belies the role of sea as separator. It was in truth a connector that allowed the
shipping of goods not only from remote areas to the center, but also connected remote ports with one
another. The previous chapter showed the development of different types of ports that were the Inland
Sea-wide infrastructure to facilitate maritime transport; here we will turn to the cargoes transported
through that network. I argue here that as the ports specialized as shipment or production hubs, the goods
they accommodated reflected the economic, environmental, and cultural concerns of the medieval period.
Sakurai Eiji suggests that while the majority of premodern exchanges of goods were made either
as barter, tribute, or gifts, the medieval period saw the lines between tribute and trade blur even further.
222

Much of the scholarship on the development of the medieval economy focuses first on tribute, namely the
nengu, the rents and dues paid by cultivators on public lands or estates to their proprietors. While rice was
the major medium of that payment, non-agrarian estates substituted other products such as silk, salt, paper,
iron, or many other goods. Over time certain areas became associated with the production of specific
items, largely due to their factor endowments: a particular geography or climate that facilitated the
procurement or cultivation of a commodity.
223
Records regarding rents and dues, however, focus mainly
on the flow of goods from the periphery to the center, implying a one-way system and largely ignoring
the potential for lateral trade between different venues. In tracing the flow of trade goods throughout the

222
Sakurai Eiji and Nakanishi Satoru, Ryūtsū keizaishi, vol. 12, Shin taikei Nihonshi (Tokyo: Yamakawa
Shuppansha, 2002), 199.
223
Sakurai and Nakanishi compare records throughout the tenth through fourteenth centuries to compile a list of
provinces and their “brand items.” Ryūtsū keizaishi 12, 200 – 201.

111

Inland Sea, it will become clear that connections among ports created a network reliant upon on different
areas’ factor endowments to produce and ship specialized commodities.

Dynamism in the Late Medieval Economy

Although barter, tribute, and gifts may have been the basis for much earlier flows of goods, medieval
developments in regional politics and farming techniques ushered in a change. Hitomi Tonomura argues
for economic changes from a variety of perspectives. She suggests first that the weakening of the estate
(shōen) system resulted in increased local power, which in turn strengthened local commercial dealings
and market development. Individuals who prior to the Muromachi era had needed to pay rents and dues to
a proprietor could, as those proprietors’ grasp weakened, project their own power as they tried to take
more profits for themselves.
224
In the late fourteenth and early fifteenth centuries, the ability of
individuals to make their own profits increased with the growing productivity of the land, the
development of a cash economy, and the availability of a growing surplus.
225
New developments in
agriculture included double- and triple-cropping of paddies, better fertilizing and irrigation techniques,
and a wider use of livestock, all resulting in greater growth.
226
As local economies enjoyed the resulting
surpluses, production of specialized goods also expanded. Inventories of peasant tools from the fourteenth
through fifteenth centuries show an increase in the numbers of tools used by small cultivators, indicating
a greater availability and use of these items at the individual level.
227
Records of tributes to the capital in
the fourteenth century show increased production of luxury goods in the provinces, such as swords from
Bizen or small boxes from Iyo.
228
As more and more goods were produced outside of the capital, there
was greater potential for trade among neighboring areas in luxury items, thus resulting in the increase of
markets and merchant networks.

224
Hitomi Tonomura, Community and Commerce in Late Medieval Japan: The Corporate Villages of Tokuchin-Ho
(Stanford, Calif: Stanford University Press, 1992), 8.
225
Ibid., 86.
226
Farris, Japan’s Medieval Population, 129–136.
227
Ibid., 130.
228
Sakurai and Nakanishi, Ryūtsū keizaishi, 12, 202.

112

The cash economy also helped market growth. During the early medieval period the practice of
importing coins from China became widespread and cash transactions gained popularity. Rice or other
commodities given as rents and dues were no longer the primary source of revenue for the proprietors of
the estates; instead, cash could (and often had to) be paid. Land sale deeds during the late Kamakura
period increasingly record cash transactions, and the numbers of individuals donating cash offerings to
temples or paying smaller levies in coins also grew.
229
As coin usage expanded, it became easier to
purchase goods directly outside of the capital region. Pricing became less fluid as toimaru (warehouse
managers or people who handled the intake and output of products 問丸) could calculate the cost of items
in a uniform manner, accounting for variables such as the cost of shipping from far-flung regions.
230

Records of disputes between local farmers and the agents of estate proprietors suggest that the accepted
prices were duly reported to the estate proprietors, reinforcing the need for accountability and involving
the upper levels of estate administration in economic affairs.
231
The growing ease of exchanging goods
resulted in even more marketplaces springing up outside the capital district.
Markets were often located on borders, on islands in the middle of a river, on beaches bordering
ocean and land, or in the foothills between mountains and flatland.
232
Merchants organized themselves
into associations or guilds (za 座), which often had their roots in the shrine associations called miyaza ( 宮
座). Early miyaza oversaw the provisioning of items required by the shrine, such as lamp oil or cloth
fibers.
233
They might then gain a monopoly of sales in the local area. And as local production of
specialized goods increased, so did the number and types of guilds, with their sights set on expansion of
their commercial monopolist rights. By the late fourteenth century, the most prominent guilds had

229
Ethan Segal, “Awash with Coins: The Spread of Money in Early Medieval Japan,” in Currents in Medieval
Japanese History: Essays in Honor of Jeffrey P. Mass, ed. Gordon Mark Berger (Los Angeles, CA: Figueroa Press,
2009), 348.
230
Sakurai and Nakanishi, Ryūtsū keizaishi, 12, 210.
231
Ibid., 12, 214.
232
Amino et al., Rethinking Japanese History, 153.
233
Sakurai and Nakanishi, Ryūtsū keizaishi, 12,121–2. Also see Suzanne Gay, “The Lamp-Oil Merchants of
Iwashimizu Shrine,” discussed in more detail below.

113

representatives in most major trading centers, including Kamakura, Hakata, and the Kinai.
234
Their
increased presence in different marketplaces also required reliable trade networks that connected remote
locales. One such network involved the use of ashiko (“leg children” 足子), itinerant traders who traveled
between established marketplaces along Ise Bay and Lake Biwa, and who assisted local merchants’
organizations.
235

While overland routes are easier to trace thanks to records kept by the different localities and
checkpoints along the routes, sea-based shipping and the developments it engendered are less well
documented. Through looking at port records, estate documents referencing local markets, waterway tolls,
and archaeological evidence, however, it is possible to trace some of the trade routes used in the Inland
Sea area. Although the vast majority of the records indicate one-way shipping routes from the provinces
to the capital region, a more in-depth analysis suggests the locations of important transshipment hubs and
specialization in shipping products. Local economies grew and shifted as different centers sprang up to
accommodate the increased flow of goods. In the following sections I will examine the different types of
cargoes being transferred over water, while highlighting the influence of the environment on the goods
produced and associated shipping practices.

The Register of Incoming Ships at the Hyōgo North Gate: An Introduction

The document that forms the foundation of most analyses of medieval maritime shipping is the Register
of Incoming Ships at the Hyōgo North Gate (Hyōgo Kitaseki Irifune Nōchō, below, Register 兵庫北関入
船納帳). This register is a list from 1445 – 1446 (Bun’an 2 and 3) that records the nearly two thousand
ships that passed through the northern checkpoint at Hyōgo, carrying from under one hundred koku to
over one thousand koku of cargo.
236
It was compiled under the auspices of Tōdaiji, the proprietor of the

234
Tonomura, Community and Commerce in Late Medieval Japan, 105.
235
Ibid., 109–112.
236
In the Register, the term koku is used as a measurement of quantity or weight for rice, salt, and other bulk goods,
not as a measure of value. The exact volume of a koku in the medieval period is unclear, though in the Edo period
(1603-1868) one koku is thought to have been about 150 kilograms, or about 330 pounds.

114

Hyōgo North Gate as described in the previous chapter. Each entry cites the port of registry for the ship,
the type and volume of cargo, the amount of taxes levied on the cargo and the date they were collected,
the name of the ship’s captain, and the name of the warehouse manager in charge of receiving the load.
The fees collected were remitted to Tōdaiji and earmarked for the maintenance and repair of both the
Hyōgo checkpoint and the temple itself. Some entries are marked as tax-exempt, usually indicating
property being shipped on behalf of a particular prominent local individual or institution. The types of
products range from bulk goods such as rice, salt, wheat, and beans, to specialty items including sesame,
pottery, indigo, cloth, and iron. While scholars have examined this document to learn more about the
types and volume of goods being shipped to the capital, here I will examine each type of cargo to
determine where it was produced, the shipping (and transshipping) routes taken from various points along
the Inland Sea to Hyōgo, and the impact of regional product specialization. Examining the different
factors that affect this trade, including seasonality and environment, shipping practices, and desirability
(reflected in price and “brand name” factors) reveals the complexity of medieval maritime trade on the
Inland Sea.
The Register is not without its problems. Although the entire document comprises records from
the first month of Bun’an 2 (1445) through the tenth day of the first month of Bun’an 3 (1446), there is a
gap in the 1445 register from the ninth day of the second month until the third day of the third month. The
reasons for the lacuna are unclear. The first two months’ records are a single document housed in the
Tokyo University archives, while the rest of the original register is in the Tōshin archive of the Kyoto
City Historic Documents Building. The latter portion was discovered by Hayashiya Tatsusaburō (1914 –
1988), former professor at Ritsumeikan University and Kyoto University.
237
The format and layout of
both documents is slightly different, which indicates that they were compiled separately, rather than as a
continuous volume. Faint lines in the later records separate the entries. It is possible that an official
register for each fiscal year was prepared in advance, with uniform space for each entry. But records for
the first and second months lack lines. For whatever reason, at the beginning of Bun’an 2 (1445), the

237
Hayashiya, Hyōgo Kitaseki Irifune Nōchō, 222.

115

official register had not been ruled in advance. The first two months’ records would have therefore been a
more informal affair, and when a register had been set up by the beginning of the third month, entries may
have been shifted to that volume.
238
If this were the case, then perhaps the end of the second month’s
records were simply lost, since they were not bound into the formal volume. Regardless, the lack of
several weeks’ worth of records necessarily impacts a discussion of the seasonality of shipping different
products, although overall trends can still be analyzed. Only the records from 1445 are included below,
since incorporating half of the first month of the following year skews analysis of annual trends.
The other major problem with formulating an analysis of maritime trade from the Register is that
while we know those cargoes passed through the checkpoint at Hyōgo, and as will be shown below we
can extrapolate where many of those items were produced or loaded on the vessels, there are still two
major lacunae. The lack of documentation from most ports of registry results in a corresponding lack of
knowledge regarding the people involved in shipping from those ports. Whether the commodities were
privately owned or were surplus from accumulated rents and dues, or whether the cargoes belonged to a
single owner or if multiple owners shared transportation to Hyōgo – these questions cannot be answered
from the Register, and indeed are difficult to answer at all. Additionally, the final intended destination for
the commodities is likewise unknown. As Hyōgo was a gateway to the Kinai, it is assumed that after
passing through the checkpoint the goods were brought up the Yodo River to the markets at Kyoto.
Courtier journals, as shown below, provide some hints to the consumption of certain items that may have
been purchased at those marketplaces, but there is no direct evidence. The Register, then, should not be
considered the definitive answer to all questions regarding maritime shipping in the medieval period.
With those reservations in mind, the Register still does provide a wealth of information. It lists
nearly two thousand ships hailing from 107 ports (fig. 28, Tables 1 and 2). Many of those locations have
retained their names throughout the centuries and are easily traceable even today. Takefuji Nao (1981)

238
Ibid., 230.

116

was the first to do a comprehensive mapping of the locations of the ports in the Register.
239
Three
locations of Imagō, Okihama, and Tsuchishima are still unknown; the small number of boats sent from
each port over the course of the year (one, one, and two, respectively) and lack of corroborating
information does not yield enough evidence to place them on a map. Several locations that Takefuji
suggests also seem suspect after additional review, and the analysis in the pages that follow amends his
original assessment of the locations of Mihara, Inunoshima, Hirayama, and Minamiura.
240

While the Register clarifies the ports of registry of each boat, it is not a given that the ships came
directly to Hyōgo from those ports. Their cargoes could have been loaded anywhere, even beyond the 107
locations recorded in the list.
241
A full analysis of the cargoes and men who captained the ships strongly
suggests, however, that even if the ships made additional stops en route to Hyōgo, they likely originated
in the ports in which they are registered. As we shall see, several cases in the following two chapters
outline scenarios in which the ships may have had multiple destinations before arriving at Hyōgo. In
general, the ships should be thought of as having a strong affiliation to and often originating from the
ports in which they are registered.
The following sections will detail the different types of cargoes that appear in the Register in
approximate descending order according to the total quantities shipped. This breakdown of items first and
foremost reveals the role of the transshipment hubs described in the last chapter, suggesting a specialized
role for those ports that supported local economic development. Places such as Onomichi that shipped
large quantities of trade goods annually to Hyōgo had strong commercial ties with warehouse managers
there, allowing Onomichi products to command higher prices. Other factors affecting shipping included

239
Takefuji Nao, “Chūsei no Hyōgo-Tsu to Setonaikai suiun: Irifune Nōchō no fune sekichi hitei ni kanren shite,” in
Hyōgo Kitaseki Irifune Nōchō, ed. Hayashiya Tatsusaburō (Tokyo: Chuo Koron Bijutsu Shuppan, 1981), 260–261.
240
The ports in question and Takefuji’s suggested affiliate provinces are Mihara (Bingo province), Inunoshima
(Bizen province), and Minamiura and Hirayama (Bitchū province). While there are certain entries marked Mihara-
Bingo in the Register, the majority of the Mihara entries likely refer to a location on Awaji Island. On the maps in
this chapter, the disputed locations are marked (with prioritized denotations first): Mihara (Bingo)/Mihara,
Innoshima/Inunoshima, Hirayama-S/Hirayama-B, and Minamiura-Aki/Minamiura. See Appendix 2 for a detailed
discussion indicating the process of determining a likely location.
241
Kamiki and Sakiyama, Kaidō rekishi no ta-minaru: Hyōgo no tsu no monogatari, 112. Kamiki is adamant that
the port names should not be accepted at face value. The following analyses, though, show a strong correlation
among the ports, people, and products.

117

seasonal fluctuations on prices, geographic proximity to specialized goods, and the ability to create a
“brand name” for a particular item. By examining the different types of trade goods shipped from Inland
Sea ports to Hyōgo, the roles of the transshipment and production ports are clarified and the effect of
environmental factors on trade becomes apparent.

118

1 Mihara (Bingo) 30 Shimozui 59 Tarumi 88 Kannonji
2 Sanaki 31 Uno 60 Sakai 89 Shido
3 Hirayama-B 32 Hibi 61 Fukudomari 90 Yasuda (Tosa)
4 Tajima 33 Naba 62 Hayashi 91 Sanbonmatsu
5 Hyōgo 34 Sakoshi 63 Murotsu 92 Kōsai
6 Tomo 35 Kōri 64 Uozaki 93 Katamoto
7 Setoda 36 Katakami 65 Nishinomiya 94 Nohara
8 Kamagari 37 Azu (Kojima) 66 Amagasaki 95 Tadatsu
9 Yanai 38 Banda 67 Kuise 96 Eigashima
10 Takehara 39 Shima 68 Iwaya 97 Tangashima
11 Tomita/Tonda 40 Ushimado 69 Tsushi 98 Ejima
12 Nogami 41 Hachihama 70 Sumoto 99 Tsuruhashi
13 Shimonoseki 42 Hinase 71 Takenoguchi 100 Nakasho
14 Moji 43 Inunoshima 72 Tachibana 101 Utazu
15 Kaminoseki 44 Inbe 73 Tosadomari 102 Sojiin
16 Obatake 45 Mushiake 74 Kaifu 103 Nishiate
17 Takasaki 46 Muro 75 Anaka 104 Yoi
18 Yuge 47 Shikamazu 76 Bekku 105 Mihara
19 Mitsunosho 48 Imazaike 77 Yura 106 Jige
20 Nio 49 Aga 78 Nawari 107 Innoshima
21 Hakata 50 Itsu 79 Sakihama 108 - 128 No ports
22 Iwagi 51 Matsubara 80 Hirajima 129 Hirayama-S
23 Minamiura 52 Aboshi 81 Muya 130 Minamiura-aki
24 Warae 53 Bessho 82 Mugii
25 Shiwaku 54 Matsue 83 Kannoura
26 Onomichi 55 Ihozumi 84 Shishikui
27 Kasaoka 56 Futami 85 Maenohama
28 Tejima 57 Suma 86 Hiketa
29 Tsurajima 58 Funage 87 Aji
Table 1. List of Identifiable Ports Noted in the Register (numerical order).

119

52 Aboshi 50 Itsu 33 Naba 17 Takasaki
49 Aga 22 Iwagi 100 Nakasho 10 Takehara
87 Aji 68 Iwaya 78 Nawari 71 Takenoguchi
66 Amagasaki 106 Jige 20 Nio 97 Tangashima
75 Anaka 74 Kaifu 103 Nishiate 59 Tarumi
37 Azu (Kojima) 8 Kamagari 65 Nishinomiya 28 Tejima
38 Banda 15 Kaminoseki 12 Nogami 11 Tomita/Tonda
76 Bekku 88 Kannonji 94 Nohara 6 Tomo
53 Bessho 83 Kannoura 16 Obatake 73 Tosadomari
96 Eigashima 27 Kasaoka 26 Onomichi 29 Tsurajima
98 Ejima 36 Katakami 60 Sakai 99 Tsuruhashi
61 Fukudomari 93 Katamoto 79 Sakihama 69 Tsushi
58 Funage 35 Kōri 34 Sakoshi 31 Uno
56 Futami 92 Kōsai 2 Sanaki 64 Uozaki
41 Hachihama 67 Kuise 91 Sanbonmatsu 40 Ushimado
21 Hakata 85 Maenohama 7 Setoda 101 Utazu
62 Hayashi 51 Matsubara 89 Shido 24 Warae
32 Hibi 54 Matsue 47 Shikamazu 9 Yanai
86 Hiketa 105 Mihara 39 Shima 90 Yasuda (Tosa)
42 Hinase 1 Mihara (Bingo) 13 Shimonoseki 104 Yoi
80 Hirajima 23 Minamiura 30 Shimozui 18 Yuge
3 Hirayama-B 130 Minamiura-aki 84 Shishikui 77 Yura
129 Hirayama-S 19 Mitsunosho 25 Shiwaku 17 Takasaki
5 Hyōgo 14 Moji 102 Sojiin
55 Ihozumi 82 Mugii 57 Suma
48 Imazaike 46 Muro 70 Sumoto
44 Inbe 63 Murotsu 72 Tachibana
107 Innoshima 45 Mushiake 95 Tadatsu
43 Inunoshima 81 Muya 4 Tajima
Table 2. List of Identifiable Ports Noted in the Register (alphabetical order).
120

120

Figure 28. Identifiable ports noted in the Register.
Damian – Chapter 3 draft 121

121

Salt Shipping

Coastal communities had the natural resources to both produce large quantities of salt and to easily ship
such a bulky commodity, making salt the primary item associated with seaside settlements. As discussed
in Chapter 2, its production shaped the governance of the affiliated area as it affected tax assessments. In
the Register, salt was the most common item brought to Hyōgo from throughout the Inland Sea region,
making it the foundation for much of medieval maritime trade. Tracing salt trade routes provides
information about transshipment hubs for salt produced in various regions. In Chapter 2 I noted that
saltmaking was an intense endeavor, requiring substantial land and manpower devoted to the process.
Below we shall see how those efforts were rewarded at the marketplace as certain areas began to have a
“brand name” association with their product, suggesting the importance of salt production at the local
level and beyond.

Shipping Routes

The Register shows that in 1445 salt shipments totaled about 113,347 koku over the year (fig. 29). The
propensity to record the production point of the salt (“Aga salt,” “Mihara salt,” “Bingo salt,” etc.), greatly
helps to identify the different routes that salt took to Hyōgo. Even though the map shown here reflects a
large amount of data, the overall color distribution at first glance suggests that the salt, by and large, was
going directly to Hyōgo from nearby its point of production. Suō salt came from the ports of Tomita,
Yanai, and Kaminoseki, all within Suō province. Bingo salt in general was shipped from Bingo and Aki
ports, Aga salt from the Harima ports near Aga, Kojima salt from the ports surrounding Kojima on the
Bizen peninsula, and so forth. Closer examination, however, reveals information about transshipment
trends, particularly when major shipping ports are identified. Removing from the analysis ports that
shipped less than five hundred koku of salt to Hyōgo over the year shows greater detail about the flow of
salt throughout the Inland Sea (fig. 30).

122

Figure 29. Annual salt shipments to Hyōgo in the Register.

123

Figure 30. Ports in the Register shipping more than five hundred koku of salt to Hyōgo.

124

One of the first ports of note is Takasaki. Particularly in comparison to its neighbor ports of
Onomichi and Setoda, Takasaki was not a major trade port. We see in the Register that it shipped
approximately 5200 koku of goods to Hyōgo throughout the year, paying about 25,500 coins in toll fees
levied at the checkpoint.
242
In contrast, Setoda sent nearly 18,300 koku of goods and paid fees of 122,770
coins, while Onomichi shipped approximately 12,700 koku and paid fees of 78,275 coins. Yet even
though Setoda and Onomichi were busier ports, Takasaki played an important role in Inland Sea travel.
We know, for instance, that on his pilgrimage to Itsukushima in 1389, Ashikaga Yoshimitsu spent the
night in Takasaki, as did a vessel with Korean envoys en route to the capital in 1420. In 1451, a China
trade ship (kenminsen) employed sailors based out of Takasaki harbor.
243
So despite Takasaki’s overall
lack of domestic trade destined for Hyōgo, it was clearly a key stopping point for ships traveling longer
distances.
The Register shows us another standout aspect of Takasaki’s role in medieval trade: it was an
important transshipment point for Suō salt. Although all other Suō salt in the Register was shipped from
ports in Suō province itself (Kaminoseki, Yanai, and Tomita), Takasaki ships brought six hundred koku of
Suō salt to Hyōgo, more than even Suō ports shipped. No other ports east of Takasaki shipped Suō salt,
which strongly suggests that Takasaki first became an initial destination for Suō salt and then sent it on to
Hyōgo. Some possible reasons for such a role are discussed below in conjunction with Takasaki’s
neighbor port of Kamagari.
There are other similar transshipment trends to note. Hiketa in Sanuki actually shipped more of its
neighbor Katamoto’s salt (810 koku) than its own Hiketa salt (566 koku). On the eastern edges of the
Inland Sea, Amagasaki also sent diverse types of salt to Hyōgo. With 469 koku of Mihara salt, 160 of
Bingo salt, one hundred of Shima salt, fifty-five of Aga salt, and forty-five of Higashiyama salt (referring

242
Various scholars have examined the intricacies of payment at the Hyōgo checkpoint. See Shinjō 1994, 521;
Takebayashi 1994, 2; and Yamatoya 1996, 5 - 10. It is unclear, however, whether coins actually exchanged hands or
whether a value was simply placed on the cargo and then a portion of the cargo was turned over to pay the fees. The
Register simply notes that the fees were assessed as numbers of mon ( 文 coins, usually copper), but does not specify
the method of payment.
243
Nihon rekishi chimei jiten 35, 1982, 462.

125

to salt mostly distributed from points scattered in Harima and Settsu), most of the salt being shipped on
Amagasaki ships to Hyōgo did not even come from areas on the route between the two ports. Either the
ships left Amagasaki and visited other areas as part of their regular trade routes, picking up cargoes of salt
and then proceeding to Hyōgo before returning to their home port; or other places sent their salt to
Amagasaki and Amagasaki-registered ships then brought it to Hyōgo. The latter possibility may have
been an attempt to avoid middleman fees, as we will see below. Either case indicates a high level of
interaction regardless of geographic location. Similar to the case of Takasaki and Suō salt above,
production points and shippers were not linked by a proprietor or other central authority. What connected
them was trade.
Data concerning the ports lining both sides of the strait between Bizen and Sanuki provinces,
such as Ushimado, Tsurajima, and Utazu, also show evidence of transshipment. Ships registered in each
of these ports brought a wide variety of types of salt to Hyōgo. Utazu ships brought salt from Bingo
province, Kojima, Takuma, Shiwaku, and Katamoto. These areas were located on both Honshu and
Shikoku, situated north, east, and west of Utazu, suggesting either that Utazu-registered ships visited
those locales before heading to Hyōgo, or that those regions shipped their salt to Utazu. A similar pattern
is visible for Ushimado, which dealt with salt from Bingo province (Bingo, Tajima salts), Bizen (Kojima
salts), and Sanuki (Takuma, Hiketa, Tejima, and Shima salts).
Perhaps the most unusual trend was embodied in Tsurajima. Its ships carried salts from the
Kojima area in Bizen province, on the same peninsula where it sat (935 koku), from Bingo (140 koku),
and from Shiwaku (380 koku), located across the strait in Sanuki. But by far the greatest volume of the
salts shipped from Tsurajima to Hyōgo was from Shima (2280 koku), a port on the island now known as
Shōdoshima, which is located between Tsurajima and Hyōgo. For Tsurajima ships to carry Shima salt,
they would have needed to stop in Shima en route to Hyōgo, or to have Shima salts shipped to Tsurajima,
in the opposite direction from Hyōgo. Regardless of which route the salt took, this suggests a high volume
of inter-regional trade and contact. In this case, however, we need to note that Hosokawa family

126

governance may have been involved. The Hosokawa oversaw both Shima and Tsurajima, so they may
have encouraged connections between ports under their jurisdiction.
Pricing

Looking at price differences among types of salt highlights the importance of the production point.
Certain types commanded higher prices, suggesting a higher quality product, special treatment by
warehouse managers who handled the salt upon its arrival at Hyōgo, or both. Bingo salt was the most
lucrative, averaging 547 mon per koku. Awa was the least, averaging only 179 mon per koku.
244
Samples
of Awa salt were few, however, as it was only shipped from Yura (three hundred koku total) and Jige (305
koku). A full list of the calculable types of salt and their respective worth follows below (Table 3).

244
Yamatoya, “‘Hyōgo Kitaseki Irifune Nōchō’ kara miru bukka no hendō,” 7. Yamatoya calculated pricing
practices by subtracting forty-five mon to account for portage fees from the total taxes assessed on a vessel and
multiplying the resulting sum by one hundred to equal the approximate total sale value of the cargo. Dividing that by
the total amount of cargo yielded a price-per-koku value for the goods in question. Yamatoya calculated the value of
cargo on ships carrying a single item, as the taxes were assessed only on the total cargo, not on individual products.
If a ship brought only salt or only rice, for example, the amount could be easily calculated, but for a vessel carrying
salt, rice, and sardines together it would be more difficult to determine which percentage of the total taxes came
from each type of cargo. I have used Yamatoya’s equations to determine pricing, throughout the following analysis.
Since it can account only for shipments that consisted solely of one type of salt, however, entries with multiple
cargoes were eliminated from these calculations. For example, prices for Suō and Konoshima salts could not be
determined, as they were only sent to Hyōgo as one of multiple cargoes on board ship. Salts with fewer than ten
calculable shipments, while included in Table 2, will not be considered in the overall salt shipping trends due to the
paucity of data.

127

Salt Type Avg Worth Avg koku shipped
Total Calculable
Shipments Notes
Bingo 547
183 194

Kojima 371
133 40

Aga 319
47 52

Mihara 289
54 127

Hiketa 282
50 10

Katamoto 252
206 21

Takuma 250
189 22

Shiwaku 238
183 14
*
Shima 188
216 38

Fewer than 10 shipments:
Higashiyama 386
34 6

Awa 179
90 5

Unnamed 325
55 3

Tejima 256
120 3

Tajima 444
80 1

Ako 329
35 1

Table 3. Pricing: Salt.
* One entry from Jige is eliminated in this calculation. That single shipment averaged 4833 mon per koku,
nearly twenty-four times the average of all other Shiwaku salt shipments. Since the reason for that
discrepancy is unaccounted for, to conform to a more realistic annual average it is not included in these
calculations.

With some exceptions the trend seems to have been that the closer to Hyōgo, the higher the salt
value. Salt from the mainland nearest Hyōgo, i.e., from Kojima and Aga, fetched higher prices than salt
coming from Awaji island (Mihara) or the Shikoku area and surrounding islands (Hiketa, Katamoto,
Takuma, Shiwaku). This may have been a manifestation of increased costs in shipping items from further
away that chipped away at the profit earned. Longer distances required longer journeys, which in turn
needed more provisions and faced greater dangers. This becomes another example of geography affecting
the process of maritime shipping. The cost of outfitting such voyages may have eaten into the profits of
the cargo, causing it to be devalued at the checkpoint. It is unclear why prices for that cargo would not
have been raised to offset those costs, however, unless it was a reflection of a relatively poorer quality of
salt for those areas.
Quality of salt also might account for some of the exceptions in pricing. The most obvious of
these was Bingo salt, which was some of the furthest traveled salt and yet it commanded the highest value

128

by far. Another possible discrepancy is seen in the higher-end Kojima salt, which fetched a higher price
than Aga salt although it traveled a longer distance. This may also have been due to the disparity in the
types of shipments and relationships with the intake warehouse managers. Kojima salt shipments were
less frequent but larger, and were often coming from the larger transshipment hubs of Shimozui,
Ushimado, Tsurajima and Utazu. Captains from those hubs may have had better connections in Hyōgo.
According to the Register, many of the boats registered to Bizen and Bitchū ports had dedicated
warehouse managers in Hyōgo who oversaw their incoming cargoes. Of the calculable shipments of
Kojima salt, the manager Dōyū handled fifteen from Shimozui and Tsurajima, while Emon Kurō oversaw
nineteen from Ushimado, Banda, Azu, Uno, and Hachihama. Together these two men accounted for over
three-quarters of the calculable Kojima salts and as such had a near-monopoly on the product.
Aga salts, on the other hand, were sent more often but in smaller quantities. The majority of ports
sending Aga salts do not even appear on the map of major salt shipping centers sending more than five
hundred koku of any type of salt annually. Aboshi, Matsue, Amagasaki, and Jige were the exceptions,
with only Aboshi specializing in Aga salt; the other three ports sent far more of other types. While only
five managers total were affiliated with Kojima salt handling, with the majority overseen by Dōyū and
Emon Kurō, Aga salts had at least eight. Aboshi shipments were divided between Saburō Tarō (nine
shipments) and Taifu Shirō (two). Matsue boats were handled by Kinoya (ten) and Michinaga (one).
Mago Tarō oversaw six shipments total from Imazaike and Hayashi, and one from Nakashō, while Awaya
also handled one Nakashō shipment. Many of the shipments came from Jige and Kuise, with no
identifiable managers. This scattering of overseers may account for the variance in cargo values. Kojima
managers such as Dōyū and Emon Kurō, who were in charge of all shipments from their related ports,
would be more likely to establish ongoing connections with local merchants and command a higher price
in return for trusted ongoing transactions. While Aga salts had some specialized managers in Kinoya and
Saburō Tarō, those men did not oversee all shipments from the related ports, nor were they in charge of
all incoming Aga salts. Buyers may not have been as established, resulting in a lower asking price
manifested in the valuation of the cargo at the checkpoint.

129

There may also have been a “brand name” attitude toward certain salts. Bingo salts may have
been thought to be of remarkably good quality, for example, while Shima salt was seen as less desirable,
as suggested by its unusually low price. Shima was close to the area shipping Kojima salts, and Shima
salts were shipped only from the ports of Shima and Ushimado, relatively close to Hyōgo. It was also
closer than the unusually highly priced Bingo salt area. Moreover, the managers overseeing Shima salts
were also specialized: all sixteen shipments from Shima were handled by Dōnen, and Emon Kurō again
acted on behalf of Ushimado’s twenty-two shipments. The reason for Shima salts’ unusually low prices is
unclear, unless it was overshadowed by the brand name of the neighboring Bingo and Kojima salts.
245

Salt: Final Thoughts

The analysis of the Register clearly shows the role that transshipment and production ports played in
moving salt from different locales to Hyōgo. Places such as Ushimado, Utazu, Tsurajima, and Takasaki
were important transshipment hubs for salt, and the volume that passed through each of these areas
demonstrates those ports’ status in the overall Inland Sea maritime network. Simple geography played a
major part in this. For instance, as will be discussed in more detail below, Takasaki would have been a
convenient stopping point for captains from the western provinces who did not know the eastern waters,
or who feared pirate enclaves located on the smaller islands there. Tsurajima and Ushimado were west
and east of the Kojima area, which today is a peninsula but in medieval times was still a series of islands.
Not only would Tsurajima have been a convenient transshipment point for western ships, but Tsurajima-
registered ships would have been able to pass through Shima on a more direct route to Hyōgo and pick up
salt there. Vessels from Ushimado could also proceed in nearly a straight line directly to Hyōgo, making
Ushimado a useful stopping point for western-based ships to offload their wares and return home, and
allowing Ushimado captains to complete the final leg of the journey. Utazu would have served a similar
function on the Shikoku side.

245
Bingo salts were brought in from ports from Aki to Bizen, with a myriad of warehouse managers overseeing the
incoming cargoes. There does not appear to have been specialization in intake at Hyōgo.

130

As discussed in the previous chapter, salt’s influence on the development of the littoral
communities was first evident in the shift to using salt to replace rice as a method of payment for rents
and dues. Furthermore, the “brand name” mindset of any item from the Register was most evident in salt.
Bingo and Kojima salts were the most obvious examples, with those names commanding the highest
prices by far. This may have been due not only to the inherent quality of salt produced in that location, but
to the strong connections between captains and warehouse managers. More than any other commodity,
obtaining a near-monopoly on particular salts allowed the managers to command a higher price. We can
surmise that individual managers thus worked harder to strengthen their hold on particular salt types.
While goods of all kinds were shipped throughout the Inland Sea, salt was the driving force of the
maritime-related economy.

Rice Shipping

Though rice was the second most commonly shipped cargo in the Register, overall it did not represent the
different locales in the same manner as salt. While salt was usually designated in the Register by a
geographic location, rice was not usually associated with a specific production point. Other notations
allow us to trace its trade routes, revealing additional geographic trends influencing lateral trade. Through
this analysis additional evidence regarding Takasaki and Kamagari’s roles in transshipping comes to light.
Finally, environmental factors affected the cultivation of red rice (Champa rice), resulting in its status as a
Sanuki “brand name” product.

Rice Production and Packaging

As evidenced by the Register, in the Inland Sea region specific locales were not synonymous with rice
production. Notations of rice sometimes included a regional name (Bungo, Sanuki, Muro, etc.), but this is
not thought to have been the production point for the rice. Instead it refers to how it was packaged – the
size of the bale of rice. The typical base measurement of rice was in masu (ten masu 升 = one to 斗, and

131

ten to = one koku 石), and although most of the rice in the Register was recorded as koku measurements,
there were often additional notations specifying the particular bundling style of koku as “Sanuki bale,”
“Buzen bale,” and so on. Pricing for the rice thus varied somewhat, likely reflecting the different
measurements that made up the various types of bales. Although the bundling styles generally show a
geographic bias, the notations were far less specific than the salt production points, making it more
difficult to pinpoint a production point for the rice. Hanshō ( 半双) rice, for example, most often came
from areas close to the Kinai, including ports such as Aboshi, Uozaki, Nakasho, Matsue, and Aga. The
term hanshō possibly denoted a small receptacle originally used for hand-washing, used as a measure of
volume in homes throughout Yamato, Yamashiro, and Harima provinces.
246
The Register also noted
three-, four-, and five-to rice bales, though the former two especially were only very rarely used (fig. 31).
Although there were some rice bales with place names, except for Sanuki bales most were very
limited in their usage. That very limitation helps in some cases to track the trade routes and determine the
roles of the various ports in transshipping. “Sanuki rice,” for instance, was shipped to Hyōgo from as far
away as Shimonoseki and Moji. From this geographic distribution, however, it appears that the Sanuki
style of baling rice was widely adopted throughout the western Inland Sea. Geographic biases can also be
seen in the “Awaji” rice centered in the ports on Awaji island (fig. 31), although Awaji ports did not ship
exclusively Awaji bales either. A better understanding of the rice production and shipping trends seen in
the Register, then, requires a closer look at the major Inland Sea ports.

246
Takefuji, “Chūsei no Hyōgo-tsu to Setonaikai suiun: Irifune Nōchō no fune sekichi hitei ni kanren shite,” 255.

132

Figure 31. Annual rice shipments to Hyōgo in the Register.

133

Shipping Routes

The comparatively fewer varieties of rice bundling styles, combined with the propensity to use a common
style over a widespread geographical area, makes it more difficult to pinpoint trade routes as easily as the
preceding salt examples. Nevertheless, certain trends do become apparent through the geographic analysis.
Awaji bales of rice were sent only from Awaji ports, suggesting less transshipment of Awaji rice than
Mihara (Awaji-based) salt. Furthermore, although hanshō rice was occasionally shipped from Onomichi
(thirty koku), Hibi (125 koku), Anaka (twenty-two koku), and Sumoto (thirty-five koku), the vast majority
of it came from the ports in Harima and Settsu provinces in the Kinai region. This suggests that the Kinai
nomenclature of “hanshō” for measuring rice was mostly used locally. Other bundling styles such as
Ikawa, Akashi, and the 3- and 4-to bales are too infrequently used to trace any significant trading patterns.

134

Figure 32. Ports in the Register shipping more than five hundred koku of rice to Hyōgo.

135

The greatest revelations come from focusing on the major ports (shipping over five hundred koku
of rice), which suggest additional transshipment trends (fig. 32). Focusing on major ports first highlights
the prominence of the westernmost ports of Moji and, to a lesser extent, Tomita. Shimonoseki, adjacent to
Moji, shipped lesser quantities of rice (four hundred koku) but those two Nagato province ports did not
ship any salt at all. This is the first evidence of geographic specialization in types of cargo, with the
western ports focusing on rice instead of salt. Yet Moji and Tomita shipped either entirely Sanuki bales or
Sanuki and “unspecified” bales (no geographic or other designation) – the lack of variety suggests less
transshipment. Geography also plays a part in the type of voyages undertaken from those two ports. Moji
sent six ships total to Hyōgo over the course of the year, one of which held over one thousand koku of
cargo (one of only two ships that large that passed through the northern checkpoint in 1445), and Tomita
sent merely three. Ports from further distances sent fewer, but larger, ships to Hyōgo.
Places that sent a wide variety of rice to Hyōgo, on the other hand, are more likely to have been
transshipment ports. Traveling from west to east, the first port, major or minor, with more than two styles
of rice was Kamagari. Throughout the year it sent seven hundred koku of Sanuki rice, 410 koku of Bungo
rice, twenty koku of Buzen rice, and seventy koku of unspecified rice to Hyōgo. It was the only port that
shipped either Buzen or Bungo bales of rice to Hyōgo, suggesting a tie with those two provinces located
on the southern island of Kyushu. Though the rice may not have been produced in Buzen or Bungo, the
bale denotation indicates some relationship with those areas. Even the ports in the Register that were
closer to Kyushu, such as Shimonoseki and Moji, did not send Kyushu-baled rice. Similar to Takasaki’s
role in salt shipping, Kamagari became a rice transshipment port, with shipments of rice arriving from
Bungo and Buzen and then in turn shipped on Kamagari-registered vessels to Hyōgo.
The comparison with Takasaki, a port that shipped total annual amounts of cargo to Hyōgo
similar to Kamagari’s totals, also suggests the possibility of specialization. Kamagari was west of
Takasaki and could therefore have been a convenient transshipment point for Suō salt as well. Yet there
are no records of Suō salt coming from Kamagari. Shipments bypassed it instead for the slightly more
eastern port of Takasaki and progressing from there to Hyōgo. Conversely, if Takasaki had been the

136

region’s major transshipment hub for goods from westernmost Japan, the rice from Bungo and Buzen
would likely have foregone Kamagari in favor of Takasaki. Since this was not the case, it is possible that
the different ports specialized in particular goods.
It is also worth considering the role of production ports, or ports that showed little or no
indication of interaction with other areas. Muro, for example, shipped twenty-two bales of rice but they
were all Muro bales, suggesting a distinctive bundling and measuring style. No other ports shipped Muro
rice. Compared to the other ports closest to Muro (Naba, Sakoshi, Nakasho, Ejima, Tangashima, Itsu,
Aboshi), Muro was second only to Aboshi in sending to Hyōgo the greatest amount of cargo (nearly 1700
koku) with the greatest cumulative worth (15,920 coins). While not conclusive, this evidence taken
together could indicate Muro’s relative prominence as a port in that immediate area.
The absence of transshipment evidence is also important. In salt shipping, Sakai, Kuise, Uozaki,
and Amagasaki appear to have been important transshipment hubs. Yet evidence for rice transshipment is
notably lacking in the same ports. While the majority of the rice from all of those locations was
“unspecified” and could have originated anywhere, most other types were hanshō or 4- and 5-to bales.
There is no indication that Sanuki or Awaji rice bales were shipped to Hyōgo via those ports. This
contrasts with the trends in salt shipment, which had a number of western-origin salts being shipped
through those same ports.
247
Reasons for this discrepancy could simply be differences in volume: with far
more salt than rice being shipped, different captains with different networks may have taken additional
routes. It could also be another aspect of specialization, if Uozaki and easterly ports were more willing to
transship salt than rice. If the latter were the case, that may have been due to rice’s status as a tax payment
medium. Incoming rice from other regions may simply have been appropriated at the initial intake port –
in this case, Amagasaki, Uozaki, Kuise, or Sakai – and exchanged for other trade goods at hand, most
notably salt. Locals could then have used the incoming rice to pay their own rents and dues.

247
Most notably, Amagasaki shipped 469 koku of Mihara salt, 160 of Bingo, and 100 of Shima, as well as 55 of Aga
and 45 of Higashiyama salt. Kuise shipped 45 koku of Mihara salt and 135 of Aga. Sakai shipped 20 koku of Kojima
salt and 15 of Higashiyama.

137

Pricing

Though overall there were fewer rice shipments to analyze for price fluctuations, certain trends reflect the
correlation of geographic proximity to Hyōgo and rice valuation. Unspecified rice was the most valuable
of all types, averaging 1,513 mon per koku. This is in direct contrast to the previous trends seen in salt,
where particular places commanded higher prices, and reinforces the idea of salt as being important for its
“brand name” value as a maritime product, while rice was not necessarily valued simply because of its
production point. Of rice types with five or more calculable shipments, hanshō rice was the cheapest,
averaging 985 mon per koku, or only about two-thirds the price of unspecified rice (Table 4).
Rice Type Avg Worth Avg koku shipped
Total Calculable
Shipments
Unspecified 1513
38 39
Akashi 1338
29 6
5-to 1312
28 7
Awaji 1260
35 6
Sanuki 1203
120 9
Hanshō 985
38 61
Fewer than 5 shipments:
4-to 1167
21 2
3-to 900
15 1
Ikawa 500
2 1

Muro --
-- 0
Table 4. Pricing: Rice.
Much of the “unspecified” rice came from locations close to Hyōgo, as did the 5-to, Akashi, and
Awaji rice. Further ports tended to ship mostly Sanuki bales, which fetched the lowest prices. This is
reminiscent of salt pricing: with some exceptions, salt coming from further locales was cheaper than salt
from closer areas. Occasionally even the same types of rice bales were priced differently depending on its
port of origin. Hanshō and unnamed rice were most expensive on average coming from Jige, followed by
Amagasaki, Aboshi (for hanshō rice only), and Kuise. Hanshō rice differed by sixty-eight mon between

138

Jige and Kuise, while for unnamed rice the discrepancy was fifty-four mon.
248
It is unclear why those
discrepancies existed.
Unlike salt pricing, however, there is less of an obvious correlation between warehouse managers
and rice price fluctuation. Over half of the most common hanshō and unspecified rice were shipped from
nearby locations such as Jige, Kuise, and Amagasaki – ports which rarely had a warehouse manager
affiliated with them. This is a striking trend, as these two types of rice commanded the lowest and highest
prices respectively. The presence or absence of a warehouse manager did not affect the pricing. No clear
correlation existed between the managers and the ports of origin for the other types of rice either. With
hanshō rice, for example, twenty-seven of sixty-one cases recorded one of seven manager names. Those
men did not have any affiliation with a particular port: Nakasho rice is overseen by both Dōyū and Saburō
Tarō, Uozaki by Emon Kurō and Dōyū, Aboshi by Saburō Tarō and Mago Tarō, and so forth. The single
exception was Awaji rice, as all five examples of calculable rice were overseen by Kinoya. Kinoya also
was responsible for the intake of Akashi rice from Matsue (four cases), as well as one instance each of
hanshō rice from both Matsue and Anaka. With relatively few examples to draw from, though, it is nearly
impossible to draw correlations between the warehouse managers and price trends.

248
Yamatoya, “‘Hyōgo Kitaseki Irifune Nōchō’ kara miru bukka no hendō,” 5.

139

Figure 33. Annual Champa Rice shipments to Hyōgo in the Register.

140

Red (Champa) Rice

One subset of rice that is best addressed individually is that denoted as “red rice” (akamai 赤米) in the
Register (fig. 33). This was almost certainly Champa rice, an Indian red rice called tōboshi or futomai. It
had been introduced to Japan via China and likely was first planted in the Kinai area, although references
to it are obscure before the fifteenth century.
249
In Sanuki province, the earliest written reference to red
rice is from a 1398 document from Nagao-no-shō, detailing a shipment of fifteen koku of “futomai” as
part of a back payment of rents.
250
Records from the early 1400s show that it was also adopted in Harima
prefecture, but fetched a lower price than regular rice.
251
Red rice did not seem to grow as well in colder
climates, however, and it became something of a Sanuki brand.
252

Evidence from the Register supports this claim. Red rice was brought in by boats registered to ten
different ports, seven of which were from Sanuki province.
253
Of the total 1,469 koku of all rice brought in
on Sanuki boats, 275 koku (18.7%) was red rice. The other three ports shipping red rice were Tsurajima
(relatively close to Sanuki, and as shown above in the salt analysis, had an established record of shipping
Sanuki goods), Amagasaki, and Kuise (both in Settsu province). The latter two ports shipped red rice in
hanshō bales, while all other ports except Utazu and Tsurajima had no notation as to bale type. It is
possible that there was some local red rice cultivation in the Settsu province area. But the vast amount of
other goods being transshipped through those ports (in conjunction with the lack of other Settsu, Awaji,
or Harima ports shipping red rice) suggests that red rice from Amagasaki and Kuise also had its origin in
Sanuki. The change in bundling style is noteworthy, however. Hanshō was the term used for rice mostly
originating in the Settsu-Harima region. If this red rice did indeed originate in Sanuki – possibly baled in

249
Farris, Japan’s Medieval Population, 132.
250
Daigoji monjo, Ōei 4 (1398) 4.16. Quoted in Kagawa-ken (Japan), Kagawa-ken shi, 2 Chūsei:555. It notes the
rents and dues that Nagao-no-shō had not paid for Ōei 3 (1397) and that sixty-eight koku of rice remained unpaid. Of
that, sixty-five koku were shipped to Amagasaki, which took eleven koku as payment, leaving fifty-four koku. Of
that, fifteen koku were futomai.
251
Farris, Japan’s Medieval Population, 133.
252
Kagawa-ken (Japan), Kagawa-ken shi, 2 Chūsei:555.
253
One of the ports was the disputed location of Hirayama. With the connections between red rice and Sanuki,
however, it seems likely that it was indeed located in Sanuki, not Bitchū.

141

the Sanuki style – it was apparently repackaged in Amagasaki and Kuise into hanshō bales, perhaps for
more uniform transport. It is also possible that local warehouse managers or intake overseers took
possession of the original Sanuki bales and redistributed some of the red rice locally before re-bundling it
and sending it on to Hyōgo. This too suggests a dynamic local market for goods from afar.
Pricing for red rice cannot be determined. The hanshō red rice from Amagasaki and Kuise was
always shipped with either hanshō or unspecified regular rice. Boats registered to Sanuki ports had more
varied cargoes, shipping different types of salt, Sanuki and unspecified regular rice, as well as beans and
soba along with the Sanuki baled red rice.
Rice: Final Thoughts

Although it is more difficult to trace the points of origin of the various types of rice than those of salt,
looking at the volumes and baling styles does provide some information about rice’s role in the maritime
trade network. Ports further from Hyōgo, such as Moji and Shimonoseki, shipped rice almost exclusively
in large vessels making infrequent voyages. For ship owners and cargo owners facing a longer journey to
Hyōgo, this exclusivity may have reflected the desire to send a stable commodity that would certainly
find a market upon its arrival. This may also relate to the lack of warehouse managers specializing in rice,
as it would have been a universally desired and handled commodity. There was also less value attached to
specific “brands” of rice, suggesting that salt locations were more critical to economic valuation than was
the case with different types of rice. This may relate to rice as a payment for rents and dues: there is little
mention in estate records of one type of rice being worth more than another. Salt, on the other hand, as
seen in the case of Yuge and elsewhere, was valued on a case-by-case basis when it was determined how
much salt would replace traditional rice payments.
254
As assessments changed due to location or overseer,
those brand values developed.

254
Yamauchi Yuzuru, Yugeshima-no-shō no rekishi (Ehime-ken Ochi-gun Yuge-chō: Yuge-chō, 1985), 70–76.
Yamauchi details several instances in the Kamakura period when Yuge submitted salt as rents and dues in lieu of
varying amounts of rice.

142

Even though trade routes for rice are more difficult to track than salt, rice reveals information
about transshipment. Kamagari’s handling of rice from western origins mirrors its neighbor Takasaki’s
role with salt, reinforcing the notion that that area was a stopping-off point for westerly ships before they
reached unfamiliar waters. Relative lack of transshipment of rice is notable as well in Muro’s handling of
only its own rice. Red rice was a minor exception, possibly due to its quasi-brand status as a Sanuki-based
product. While an important commodity for medieval Japan’s economy, rice did not attain as much of a
status-of-place as various types of salt did. In both economic value and nomenclature, salt became a brand
name item in large part due to its geographic production point. Rice, on the other hand, was not seen as
representative of particular locations to the extent that salt was, suggesting that salt quality gradations
were more valued in maritime-centric trade.

Beans, Wheat, and Barley

Other agricultural products such as beans, wheat, and barley were shipped to Hyōgo from many of the
locations throughout the Inland Sea. While their production points are more difficult to trace, seasonal
influence and the development of new farming technologies becomes clear through an analysis of these
items.
Notations of beans in the Register are somewhat problematic, as it is not always clear what was
being referenced. There were records of “beans” (mame マメ), soybeans (daizu 大豆), “bean rice”
(possibly “mamegome,” 豆米), red beans (azuki 小豆), and “bean wheat” (mame komugi マメ小麦). It is
unclear if the beans alone were soybeans or red beans.
255
Based on the meticulous nature of the Register
overall, as well as the fact that the same ports apparently shipped both beans and either soybeans or red
beans, it seems more likely that “beans” referred to a distinct item. Though prices of these products are
difficult to ascertain due to the low number of shipments, beans averaged about 150 mon per koku more
than soybeans (Table 5). “Bean rice” and “bean wheat” are also questionable items, although these may

255
Kamiki and Sakiyama, Kaidō rekishi no ta-minaru: Hyōgo no tsu no monogatari, 124.

143

have been a reference to, respectively, a mixture of beans and rice or beans and wheat. The latter was only
recorded once, in a ship from Shiwaku in the sixth month, while bean rice was more widespread. The
largest quantities of bean rice were sent from the transshipment hubs of Kamagari, Onomichi, Setoda, and
Ushimado (fig. 34). This suggests that the “bean rice” appellation was for something commonly used.

144

Figure 34. Annual bean shipments to Hyōgo in the Register.

145

Figure 35. Annual amounts and types of bean shipments to Hyōgo in the Register.

146

In looking at the distribution of bean types and ports of origin, Sanuki-baled beans came mostly
from Sanuki province. Similar to rice bundling styles, however, that appellation was also used in bean
shipments from Bingo, Aki, and Suō provinces. This reaffirms the idea that the Sanuki bundling style was
widespread throughout the western Inland Sea. The largest quantities of any type of beans being shipped
came from Kamagari and Onomichi, followed by Ushimado, Setoda, and Takasaki (fig. 35). The
Kamagari and Takasaki quantities are especially interesting due to the previously acknowledged
importance of these ports as transshipment hubs for goods from areas further west. The production points
of the beans cannot be determined, but based on the other shipping trends from those locations it is likely
that a significant quantity of the beans shipped from Kamagari and Takasaki to Hyōgo actually originated
in Suō or Kyushu.
Wheat, barley, and buckwheat (soba) also were shipped to Hyōgo from ports throughout the
Inland Sea. Wheat alone was sent from more ports, although not always in great quantities. Barley was
somewhat more prevalent in the eastern Inland Sea, and wheat was more common in the western
provinces, although neither area had a monopoly on either product. Wheat was a far more lucrative
commodity than barley, averaging nearly twice the price (Table 5). There does not seem to have been a
specific correlation between warehouse managers and the pricing of any of the beans or grains.

147

Item Avg Worth Avg Koku
shipped
Total
Calculable
Shipments
Notes
Beans 1571 6 2
Soybeans 1339 22.4 5
Bean Rice 1096 81.75 4
Wheat 1299 16.2 5
Barley 640 21.6 8 *
Goods with no calculable shipments:
Bean Wheat --
Red beans --
Beans -
Sanuki
--
Soba --
Table 5. Pricing: Beans and Grains.
*This total does not include one shipment of thirteen koku of barley that was taxed 245 mon, giving it a valuation of
about 1538 mon per koku – nearly three times as much as some of the other shipments. All calculable barley
shipments were from the fifth and sixth months, and with no apparent reason for this high valuation it was removed
from the calculations to prevent skewing.
The most relevant points in tracing bean and grain shipping, however, are the implications for
seasonality and its effect on shipping. Wayne Farris argues that double-cropping had its advent in Japan
as early as the 1160s.
256
Documentary evidence from the 1300s shows many examples of estates in the
Kinai and western Japan that produced rice and other grains in two planting seasons throughout the year.
Although early efforts were only intermittently successful, by the fourteenth century certain areas had as
much as thirty percent of their fields devoted to double-cropping. Looking at the rise and fall of the types
of shipments in 1445 demonstrates that there was indeed a propensity for double-cropping, evident in the
times of year when certain items were shipped (Table 6). Although the practice of double-cropping may
not have been widely documented in the rest of the Inland Sea region, the types and volumes of cargo
shipped strongly suggest that it was indeed accepted practice outside the Kinai, and a successful one as
well.

256
Farris, Japan’s Medieval Population, 130–131. Farris outlines a number of records from Wakasa, Harima, and
Yamashiro that provide evidence of taxation on double-cropping, protection of productive farmland from evil bands,
and references to “autumn wheat” and “summer wheat.”

148

Table 6. Monthly Shipping: Rice, Beans, and Grains.

Although the quantities were vastly different, rice and beans loosely followed the same shipping
patterns, rising sharply after the third month and falling in the sixth through eighth or ninth months, then
rising again dramatically toward the end of the year. Wheat and barley, clearly the opposite, showed their
greatest increase in the summer months, dropping down to literally nothing in the cold season of the year.
This is strong evidence for substantial, consistent double-cropping from throughout the Inland Sea region,
resulting in enough of a surplus to be able to ship thousands of koku at the peak of a given season.
Like rice, the production points of the different types of beans and grains are difficult to pinpoint.
Also like rice, these items were shipped from throughout the Inland Sea region. Grains were more
prevalent in the eastern Inland Sea area, being shipped in greater quantities from ports in Bitchū, Bizen,
and Sanuki. The ports that supplied the most of all types of beans were Kamagari, Onomichi, and Setoda,
followed by the more eastern port of Ushimado. As some beans were sent from westerly ports such as
Moji and Obatake, it is possible that the beans sent from Kamagari and Takasaki were again
supplemented by shipments from western provinces. What is clear, though, is that double-cropping had
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
1 2 3 4 5 6 7 8 9 10 11 12
Shipments per Month (in koku)
Rice
Beans (all types)
Wheat and Barley

149

rapidly become a critical aspect of Inland Sea commodity production. Few places relied solely on one
agricultural crop to ship to Hyōgo, instead making as much use of their arable land as possible throughout
the year. Beans and grains therefore comprised an important part of the economic growth within the
Inland Sea area, as growers took advantage of the relatively mild climate for double-cropping and
increasing surplus commodities.

Wood Products

Analyzing trends in shipments of wood products helps to highlight the specialization of products from
particular locations, drawing attention to environmental factor endowments that became salable
commodities. While the point of origin for products such as lumber (unspecified wood species), pine,
magnolia, and planks (an unspecified species of wood that was already shaped into boards) is impossible
to trace, examining the volumes shipped and associated key ports demonstrates the distribution of
particular products in specific locales. Specifically, shipments of wood products to Hyōgo came mostly
from the eastern Inland Sea area, no further west than Setoda and Onomichi (fig. 36). Pine and magnolia
were limited in their shipping sources, with the former coming from Setoda, Onomichi, Muro, and Jige,
and the latter coming only from Onomichi. The Bingo-Aki (Setoda, Onomichi) connection here in
particular is unsurprising, as these major transshipment ports were close to the Chūgoku mountain range
where wood products of all kinds would have been easy to procure.

150

Figure 36. Annual wood shipments (types) to Hyōgo in the Register.

151

Important sources of lumber are made even clearer by looking at the volumes being shipped (fig.
37). Slash-and-burn agriculture had depleted many of the forests in the Kinai, requiring imports of wood
products from further away.
257
The eastern ports on Shikoku and Awaji stand out as major players. Kaifu
in Tosa prefecture and Yura in Awaji, in particular, shipped inordinately high volumes of planks to
Hyōgo. Although the reasons for that cannot be determined from the Register alone, one possibility is that
there was a local abundance of lumber. All Awa ports south of Bekku save for Sakihama shipped only
lumber, planks, or a combination of both to Hyōgo. This certainly speaks to local production power.
258

Furthermore, shipments of planks in particular remained relatively consistent throughout the year (Table
7), with a lull in the winter months but never fully stopping. Calculable shipments (Table 8) showed no
seasonal price fluctuation. The steady pricing likely resulted in a reliably stable source of income for the
shippers. This may be an indication of an organized production system, procuring and cutting lumber
throughout the year. As many of the Awa ports faced the open Pacific Ocean, they would not have been
as sheltered as most of the Inland Sea ports. Fewer shipments in the colder months could therefore be a
response to rougher winter seas. Finally, taken in conjunction with the salt analysis, the high-lumber
shipping areas are places that shipped relatively little or no salt to Hyōgo. Saltmaking was a wood-
intensive process, as lumber was used to fuel the fires needed to evaporate the water and create salt. Areas
that produced more salt would therefore have had less of a lumber surplus.

257
Conrad D. Totman, The Green Archipelago: Forestry in Preindustrial Japan (Athens: Ohio University Press,
1998), 35.
258
Ibid., 45. Totman notes that lumber for use in the Kinai began to be cut from forests in Kii, westward to Aki, and
inland through Awa and Tosa as early as the mid-thirteenth century.

152

Figure 37. Annual wood shipments (types and amounts) to Hyōgo in the Register.

153

Table 7. Monthly Shipping: Lumber and Planks.
* Of the total unspecified lumber shipments, 557 koku were cypress (hinoki). Since cypress was not recorded separately, those
volumes are included in the general lumber numbers.

Item Avg Worth Avg koku shipped Total Calculable Shipments
Lumber 282 125 29
Pine 266 48 4
Planks 256 139 223

Magnolia 257 35 1
Table 8. Pricing: Wood Products.
Still further evidence of specialization in Awa province focuses on cypress (hinoki) shipments.
Unlike magnolia and pine, which were recorded separately, cypress was usually shipped in conjunction
with other lumber. Those entries were labeled “lumber, 1/3 of which was cypress.” Within those
shipments, approximately 557 koku of cypress passed through Hyōgo, with 237 of that coming from
Hirajima in Awa. None of the other ports close to Hirajima shipped cypress. Of the remaining cypress,
180 koku came from Ushimado in one shipment, and 140 from a single shipment from Minamiura-aki.
259

259
Though Minamiura is one of the disputed port sites, evidence from salt shipments suggests it was indeed located
in Aki. That location’s proximity to Setoda and Onomichi, other ports with specialized type of wood shipments that
were close to the Chūgoku mountains, further strengthens the argument for Aki as the correct location. See
Appendix 2 for further discussion.
0
1000
2000
3000
4000
5000
6000
1 2 3 4 5 6 7 8 9 10 11 12
Shipments per Month (in koku)
Lumber*
Planks

154

Ushimado was an important transshipment hub, and Minamiura-aki was close to the Chūgoku mountain
range with a wide variety of wood types likely close at hand. It is unclear why cypress should have been
recorded simply as a percentage of total lumber shipments, unlike pine or magnolia. It is also unclear why
Hirajima should have been the only port in the Awa region to specialize in its shipping, particularly when
there seem to have been connections between captains and cargoes of the other nearby ports.
Yura’s cargoes, though mostly planks, included salt (three hundred koku from Awa, sixty-five
from Mihara, and forty-five from Kojima) and some indigo. As will be noted below, indigo originated in
Awa province. The different types of salts and indigo going through Yura suggest a general tendency
towards transshipment that is reflected in wood products as well. The exceedingly high volume of
shipments may be a reflection of Yura’s proximity to both Shikoku and the Kii peninsula. Awaji did have
its share of woodlands that may have been a source of planks, but more importantly, Yura was directly
across the strait from modern Wakayama prefecture and quite close to the mouth of the Kinokawa River.
It would have been relatively simple for lumber to have been brought across the strait to the Yura area,
processed there, and then sent to Hyōgo. Evidence suggests that in the late medieval period, control of
woodlands lay largely with local villagers who would have influenced the disbursement of wood
resources.
260
Although no written records remain to corroborate this theory, there may have been a land-
based merchant network, similar to the ashiko (“leg children”) network that existed between Lake Biwa
and Ise Bay, that could have facilitated connections between Kii and ports in Awaji seeing to the sale of
lumber and transport to Hyōgo. Again, it would have been a stable source of income that could be
processed nearly year-round. If this were the case, it is perhaps unusual that the lumber or planks would
not have been shipped directly to Hyōgo from the source on the Kii peninsula, but as the next chapter will
show, this may have been due to established shipping networks in Awaji and Awa provinces. Ships’
captains from these areas often collaborated, and the predominance of planks sent from those ports may
have been one aspect of that collaboration.

260
Totman, The Green Archipelago, 41.

155

While lumber shipping trends are more difficult to pinpoint than rice or salt shipments, the
prominence of Awa, Awaji, and Tosa shipping of lumber – often to the exclusion of any other products –
demonstrates a different type of product “branding.” Different types of wood may not have carried the
names of the point of origin as salt did, but the volume of wood products clearly shows the importance of
the eastern Shikoku and Awaji-Kii regions, with their still-lush forests, for lumber and plank exports. It
also suggests a different type of maritime community less dependent on salt production and with the
ability to use surplus lumber for export.

Marine Goods

Perhaps some of the most obvious types of goods to come from the coastal communities, marine goods –
fish, shellfish, and seaweed – are some of the most difficult to analyze. While types were varied, specific
items were sometimes limited to single shipments from only one locale. The lack of data for many of the
items makes the identification of overall trends impossible, but seasonality is apparent in certain items,
and close examination of the cargoes suggests particular practices of both shipping and pricing. The types
of items being shipped as marketable goods also imply consumer trends that may have been different
from consumption trends practiced in the elite echelons of society.

156

Figure 38. Annual sardine shipments to Hyōgo in the Register.

157

The fish most commonly shipped to Hyōgo (fig. 38) were sardines. While scholar Morimoto
Masahiro believes that this indicates their common usage in side dishes, a more thorough exploration
suggests that the destination for sardines was not elite tables.
261
Most medieval journals written by
courtiers rarely mention sardines, leading some historians to conclude that only in more recent times were
sardines fished on a large scale.
262
The volumes recorded in the Register are not insignificant, however,
particularly in comparison to the quantities of other fish types. Instead of sardines not being widely fished,
then, in actuality it is more likely that sardines were more of a food commodity for commoners. Diaries of
the Yamashina family, a court-affiliated family that was in charge of supplying foods to the tennō’s
palace, list the types of foods presented to the family from various estates throughout Japan. A few entries
included sardines sent from Hosokawa Estate (Harima province), but they were sent together with
snapper.
263
There are very few mentions made of sardines alone in the journals.
Four different types of sardines were noted in the Register, with the most being small sardines,
followed by “red sardines,” large sardines, and dried sardines. No sardines of any type were shipped in
the first three months of the year, or in the seventh month; and the high volume of small sardines sent
during the ninth through twelfth months shows a seasonal bias towards the autumn months (Table 9).
Though red sardines were only shipped in the eleventh and twelfth months in quantities comparable to
small sardines at the same time, there was a geographical bias towards ports in Sanuki, Bitchū, and Bingo.
Unfortunately the limited number of shipments of all but the small sardines prevents an accurate
analysis of pricing. For small sardines, however, there seems to have been a correlation between
seasonality and pricing. Values of the sardine shipments ranged from 314 mon per koku to 1712 mon per
koku. Those two extremes, however, were both shipments from Ushimado, with the low pricing in the

261
Morimoto Masahiro, Zōtō to enkai no chūsei (Tokyo: Yoshikawa Kōbunkan, 2008), 135.
262
Yamaguchi Kazuo, Nihon gyogyō shi (Tokyo: Tokyo Daigaku Shuppankai, 1957), 60-61. Yamaguchi examines
Oyūdono no kami no nikki and other fifteenth century diaries for records of fish consumption. He notes that although
there was significant archaeological evidence of sardine consumption in the premodern era, lack of notation in the
elite diaries meant that they were not as widely fished as they would later be during the early modern period.
263
Lake Biwa Museum, Nihon chūsei gyokairui shōhi no kenkyū: 15 Seiki Yamashina-ke no nikki kara, Research
Report of the Lake Biwa Museum (Kusatsu-shi: Lake Biwa Museum, March 2010), 31.

158

tenth month and the high pricing in the eleventh. Since the reasons for this are unclear, and the prices
varied by more than two hundred mon from the next calculable price, these have been removed from
consideration. Of the remaining seventy-one calculable shipments, sixty-two range from 844 to 1188 mon
per koku, with no apparent seasonal bias. Of the highest seven prices for small sardines, however, the
most expensive two (1417 mon/koku) were from shipments in the fifth month, or the first sardine
shipments for the year. The remaining five expensive shipments, ranging from 1300 – 1413 mon per koku,
were within the first seven shipments of small sardines from the eighth and ninth months, or when the
product became available once again. This is the clearest evidence of market response to the availability
of an item: in times of short supply, demand increased and the price rose accordingly.

Table 9. Monthly Shipping: Sardines.

0
200
400
600
800
1000
1200
1400
1600
1 2 3 4 5 6 7 8 9 10 11 12
Shipments Per Month (in koku)
Small Sardines
Red Sardines
Large Sardines
Dry Sardines

159

Figure 39. Annual sea cucumber shipments to Hyōgo in the Register.

160

Other marine goods show evidence of seasonal pricing as well. Sea cucumbers (fig. 39) were
available only in the winter months, mostly in the eleventh (seven shipments) and twelfth (five
shipments), with just a single shipment in the first month. The average price of sea cucumbers was only
thirty-one mon per koku. The price was higher for the first-month’s sole shipment (thirty-eight mon),
however, subsequently dipping to the high twenties in the eleventh and twelfth months, and then
increasing sharply for the final twelfth month’s shipment (seventy mon). The higher prices may have been
a response to the end of the season for sea cucumbers, as consumers would have been aware of less
availability in the warmer months.
Sea cucumber shipping also provides a hint about shipping practices. With the exception of
Mushiake in the easternmost part of Bizen province, they were shipped only from ports in Harima
province nearby Hyōgo. Sea cucumbers were also the only item shipped multiple times as the sole
cargo.
264
Most other goods were shipped in conjunction with other items at least some of the time, but sea
cucumbers were the single exception. There was no captain who specialized in sea cucumber shipping,
nor did any single warehouse manager oversee their intake, so it is unlikely that the practice was specific
to one individual. Combined with the proximity of the supplying ports to Hyōgo, we can surmise that they
were shipped live to Hyōgo.
265
More space for holding containers would have been required. It would
also preclude the ability to carry other items, particularly goods such as rice or salt that would have been
ruined had the water from the sea cucumber containers spilled.

264
Items such as coins, “nishi,” parrotfish, mackerel, and sharkskin were also the sole cargo carried, but there was
only one shipment for each of those items and as such trends cannot be pinpointed.
265
This contrasts with later sea cucumber shipping practices in the north. Brett Walker, in The Conquest of Ainu
Lands, details the volume of sea cucumber shipments from Ezo (Hokkaidō) to central Japan and further on to China.
Dried sea cucumber was a staple of that trade, totaling nearly 132,000 pounds in 1788. See Brett L Walker, The
Conquest of Ainu Lands: Ecology and Culture in Japanese Expansion, 1590-1800 (Berkeley: University of
California Press, 2001), 95–96. There were no notations in the Register to suggest that these were dried sea
cucumbers as well, although items such as snapper and sardines were sometimes specified as dried.

161

Figure 40. Annual less-frequently shipped marine goods to Hyōgo in the Register. Various units of measurement are used for these items, but as all were shipped
infrequently all of the relative volumes are small, so they have been combined on this map. Sharkskin and jellyfish are measured in gō (approximately 1/10 of a masu, a
unit of measure commonly used for rice). Mackerel, flathead mullet, parrotfish, nishi (possibly a spiral shellfish), nikomi (unknown) and abalone are in koku. Crab,
squid, and shrimp are in horseloads.

162

Figure 41. Annual seaweed shipments to Hyōgo in the Register.

163

Other types of fish and seaweed that appear in the Register are more difficult to define due to a
paucity of data (figs. 40 and 41, Tables 10 and 11). Most of these items were shipped only a handful of
times to Hyōgo, with some only appearing once throughout the entire year of records. Some seasonal
trends are hinted at in items such as jellyfish (sent in the sixth through eighth months), shrimp (sent in the
third through fifth months), and preserved fish (dried snapper and salted fish, sent mostly in the fourth
and fifth months). Overall, unlike for sardines, the fishing season seems to have been spring and early
summer.

Table 10. Monthly Shipping: Other Infrequently Shipped Marine Goods.

Table 11. Monthly Shipping: Other Marine Goods, Unique Shipments.
0
50
100
150
200
250
300
1 2 3 4 5 6 7 8 9 10 11 12
Shipments per Month (in koku)
Salted Fish
Dried Snapper
Arame
Jellyfish
Shrimp
Nishi
0
50
100
150
200
250
300
350
1 2 3 4 5 6 7 8 9 10 11 12
Shipments per Month (in koku)
Parrotfish
Abalone
Blowfish
Mullet
Squid
Wakame
Crab
Mackerel

164

What is notable about the fish shipped less frequently, however, is that many of these types of sea
creatures were often found at the tables of aristocrats, or they had been earmarked as traditional tribute
goods from classical times. Squid, arame and wakame (types of seaweed), and abalone are mentioned
multiple times in the tenth-century Engishiki as required items for festivals, rituals, and offerings to
deities.
266
Wakame and arame had been noted on tenth century wooden slips (mokkan 木簡) excavated
from shrines at Fujiwara and Heijō in the Kinai and were likely food tributes presented to rulers over and
above regular taxes (nie 贄) at that time.
267
At least some of these items apparently retained their status as
prestige goods in the medieval period. Records note that as late as 1458, the shōgun was presented one
thousand strips of dried abalone as tribute on the first day of the second, seventh, and twelfth months of
the year.
268
The Yamashina family diaries show that snapper was presented to the royal palace year-round.
Because of its use in various ceremonies, it would have been purchased or received as payment of dues
and rents.
269
The fact that these items were being procured on a regular basis, yet appeared only
sporadically in the Register, suggests that the items recorded as cargo passing through Hyōgo were
destined more for a popular market, distinct from the goods submitted regularly as payments of rents and
dues.
Perhaps the most inexplicable omission from the Register is octopus. The Inland Sea is known
even today for its octopus, and archaeological evidence shows that this is not a recent phenomenon.
Octopus pots have been found in great quantities throughout the region from as far back as Yayoi times
(approximately the fourth century BCE – third century CE).
270
The pots grew larger and their shapes and

266
References from the Engishiki are based on the Japanese Historical Text Initiative version.
http://ucblibrary3.berkeley.edu/jhti/Engi%20shiki.html. Accessed March 23, 2015.
267
Amino and Inaba, Kaimin no shakai, 10:228.
268
Ibid., 10:225. As noted in Chapter 1, Amino does not record who donated the abalone.
269
Lake Biwa Museum, Nihon chūsei gyokairui shōhi no kenkyū: 15 Seiki Yamashina-ke no nikki kara, 30.
270
In particular hundreds of octopus pots have been recovered in the waters off Hokudan-chō on Awaji Island, as
well as a great number from modern Shōdoshima, and coastal Ehime and Kagawa prefectures (see Kodai no iidako
tsubo nawagyo ni kansuru ikkōsai, parts 1 and 2, from the Setonaikai reskishi minzoku shiryōkan kikō, 2003-4). In
the eastern Inland Sea, a particular style of pot for “madako” (Octopus vulgarius Cuvier) was seen along the Kii
peninsula as early as Yayoi times, and its use spread westward over the centuries, with modern examples found off
the coast of Kagawa prefecture (Tamura Takaaki, “Tōbu setouchi ni okeru madako tsubo gyo no kentō,” in Kagawa
kōko, Vol 11, 2009).

165

styles of pots changed over the centuries, but the numbers did not diminish. Octopus was mentioned
regularly in several medieval courtier journals.
271
The Yamashina journals note their presentation to the
royal residence and their use in ceremonies or as gifts for drinking parties, often with sake or other fish.
272

Although the widespread procurement and use of octopus is thus apparent, there is no hint of its presence
as a typical cargo in the Register. This may again suggest a different market destination, with the cargoes
on the Register earmarked for a popular audience and octopus considered more of an elite item.
Alternatively, since the archaeological findings of octopus pots consistently show widespread fishing it
may have been an incredibly common item that would not have been shipped far.
In the case of marine goods, their relationship to the maritime-based economy of the Inland Sea is
defined by reactions to them at the central markets. The overall lack of preserved fish, save for the rare
appearance of dried snapper or sardines or salted fish, suggests that marine animals and plants were
transported fresh to Hyōgo. This is directly reflected in the seasonality of the shipments, with goods such
as sardines and sea cucumbers fetching higher prices in times of scarcity. The division between marine
goods shipped as cargo in the Register and those named in elite contemporary journals suggests a very
different final destination for most of the items shipped to Hyōgo. Traditionally important goods for
ceremonies or tribute such as abalone, seaweeds, and squid appear only rarely in the Register, while
sardines and sea cucumbers, two goods rarely found in the journals or other elite sources, appear most
often. Examining marine goods reveals that there were likely two different types of commodity
consumption taking place simultaneously in the capital region: a market for the elite table, and one for the
common people.

Egoma (Perilla), Indigo, and Miscellaneous Specialty Goods

A number of items appear relatively infrequently in the Register or are specific to particular locations.
These can be thought of as specialty or “prestige” items. Prestige goods, while not necessarily luxury

271
Yamaguchi, Nihon gyogyō shi, 19. Yamaguchi mentions octopus as one of several saltwater fish that were
recorded in several 15
th
century diaries.
272
Lake Biwa Museum, Nihon chūsei gyokairui shōhi no kenkyū: 15 Seiki Yamashina-ke no nikki kara, 32.

166

items, were produced in more limited quantities only in certain areas, and they usually fetched a higher
value for smaller amounts.
273
Since shipments of these items were limited, it is difficult to draw sweeping
conclusions about their ramifications, but their presence reflects different types of specialization within
different Inland Sea regions that often correspond to local environmental factors.
Shipments of egoma (perilla), a small, flowering bush that produced berries used to make lamp
oil, are denoted in two ways in the Register, as either “goma” (sesame コマ) or “Yamazaki-
butsu/Yamazaki-goma” (Yamazaki goods, Yamazaki sesame 山崎物・山崎コマ) (fig. 42).
274
While the
reason behind this differentiation is unclear, the Yamazaki notation came from the monopoly that
affiliates of Ōyamazaki Rikyū Hachimangu in Yamashiro province had on lamp oil production, described
in the previous chapter.
275
Most of the shipments of egoma were indeed labeled “Yamazaki,” and they
were centered in the Sanuki and Bizen provincial ports, notably Ushimado, Shiwaku, and Banda. Prices
for calculable shipments of egoma varied wildly, from as little as seventy-seven mon per koku for
Yamazaki goods to 2500 mon per koku for egoma.
276
As the captains and warehouse managers all varied,
there was no common human factor that appears to have affected prices. Shipments from the same port
fluctuated as much as several hundred mon, so there was no apparent “brand name” status such as that for
salt. Three-quarters of all shipments, and all but four of the calculable shipments of both Yamazaki and
generic egoma, were shipped in the last three months of the year, so seasonal fluctuations were not the
culprit. Egoma, regardless of port of origin, time of year, or people affiliated with it, had odd fluctuations
in prices that are not easily explainable.

273
Batten, To the Ends of Japan, 162. I use Batten’s definition of prestige goods here, although as he notes those
categories are not fixed and can fluctuate due to supply and demand.
274
Suzanne Gay, “The Lamp-Oil Merchants of Iwashimizu Shrine,” 14. The nomenclature here is somewhat unclear,
as none of the Register entries specify egoma (perilla frutescens var. Japonica), but instead use “goma,” most often
translated as “sesame.” Gay and other scholars feel that that discrepancy notwithstanding, it was most likely egoma
being shipped to Hyōgo, as it was more commonly used in lamp oil production – partially because it was less
odiferous than sesame oil.
275
Ibid., 11–14. Gay’s article discusses in detail the business and rights involved in procuring egoma for lamp oil.
276
Fifteen of fifty-nine imports of Yamazaki goods and/or egoma could be calculated using Yamatoya’s equations.

167

Figure 42. Annual egoma shipments to Hyōgo in the Register.

168

Figure 43. Annual indigo shipments to Hyōgo in the Register.

169

Indigo was another product affiliated with Shikoku (fig. 43). It was once thought that indigo was
harvested in Awa no earlier than about 1585, but the Register disproves that allegation.
277
Imports of
indigo were relatively lucrative, averaging about 1021 mon per koku for calculable shipments.
278
Two
ports that shipped indigo are of particular interest. Sōjiin is the only port noted in the Register that was
not directly on the coastline. Instead it is well inland, situated on the Yoshino River of Awa province. No
other riverside port sent ships directly to Hyōgo, making Sōjiin a unique shipper. The quantity of indigo
shipped from Jige, far more than any from its source in Awa, is also unusual. Of the fifteen shipments
from Jige, three arrived on vessels piloted by Hyōe Tarō, and four by Nōfuku. As will be shown in the
next chapter, Hyōe Tarō had ties with a captains’ network based on Awaji Island, which would have
placed him directly at the midpoint between Awa and Jige, and allowed him easy access to shipments
originating from Awa. All of Nōfuku’s ships were registered in Jige, but all also had co-captains. Half of
his ships were co-captained by Gyōbu Shirō, who also captained ships registered in the Awa ports Kaifu
and Tachibana.
279
This strongly suggests that the Jige-based shipments of indigo had been transshipped
from their source in Awa, using the individual ties of those captains to arrange the shipments. Possible
reasons for this will be explored below.

277
Sakurai Eiji, Zōyo no rekishigaku: Girei to keizai no aida (Tokyo: Chuokoronshinsha, 2011), 112.
278
Eleven of twenty imports of indigo could be calculated using Yamatoya’s equations.
279
Gyōbu Shirō’s network will be more fully explored in the following chapter.

170

Figure 44. Annual other infrequently shipped items to Hyōgo in the Register.

171

Leather was shipped from Hirayama with no additional notation, and the somewhat curiously
notated leather “shoes” ( 皮クツ) originated in Tsurajima (fig. 44). There are records of a group in
Shiwaku (Sanuki province) that worked with leather, so this seems to be additional evidence placing
Hirayama in Sanuki, not Bitchū.
280
The affiliation with Tsurajima is less obvious. There were no common
captains or warehouse managers linking the two ports, so it is unclear whether the leather was shipped to
Tsurajima and reworked there, or if it was produced locally. Prices for leather are impossible to determine.
Increased use of paper products dates from in the early thirteenth century. Sanuki in particular
was known as the producer of the Teikin ōrai, a textbook-like work used for kanji practice. Increased use
and circulation of this document is thought to have been an indicator of economic growth, as Sanuki
produced more tanshi (檀紙, a high-quality wood pulp paper).
281
Bizen and Bitchū were also traditionally
known as producers of paper, with shipments of paper recorded as part of rents and dues paid as early as
the eleventh century and increased notations in the fourteenth century.
282
According to the Register, paper
was sent to Hyōgo twice from Shiwaku (Sanuki province), and once each from Hibi and Banda in Bizen
province, Nishiate in Bitchū, and Jige (fig. 44). No paper was shipped after the seventh month, suggesting
that its production usually took place during the warmer months. Paper was always shipped with other
items, including salt and “bean rice,” and so its values alone cannot be calculated.
The only woven or sewn goods noted in the Register were “Chinese grasses” (karamushi 苧),
cloth, and woven mats (mushiro 筵) (fig. 44). The fibers from the grasses were often separated and made
into threads or cloths, but there is little to no direct evidence in the Register of processing grasses in that
way.
283
Grasses were shipped from Ushimado, Inbe, Sakai, and Jige, but the most interesting correlation
here provides more hints into shipping practices. Of the thirty shipments of grasses, thirteen were from

280
Kagawa-ken (Japan), Kagawa-ken shi, 2 Chūsei:579. Refers to an unspecified document from Sutoku-in Goeidō
Bunkō 崇徳院御影堂文庫 that discusses leather production.
281
Sakurai, Zōyo no rekishigaku, 113.
282
Amino, Chūsei minshū no seigyō to gijutsu, 212–213.
283
Kamiki Tetsuō, “15 Seiki nakaba Setonaikai ni okeru kaisen to shōhin yusō: Bun’an 2 (1445) nen. ‘Hyōgo
Kitaseki Irifune Nōchō’ no kisō bunseki: 5,” Nara Kenritsu Daigaku kenkyū kihō 14, no. 2 (December 10, 2003): 81.

172

Jige and therefore likely were transshipped at some point. All eight shipments from Inbe and one of two
from Sakai were sent along with pottery items (to be discussed in more detail below). It is highly likely
that the grasses were used as packing materials for fragile pottery. Pottery was large and heavy and could
be used as ballast, but space on board a merchant vessel is always a premium. Grasses could have easily
served double-duty as a packaging medium that fit in empty spaces and protected the pottery, while
commanding a profit in its own right at the final destination.
Woven mats were generally shipped from ports west of those that shipped grasses, making a
connection between those grasses and weaving less likely. Both items exhibited a seasonal bias, however,
as four out of seven mat shipments arrived in the seventh month, and twenty-three of thirty grass
shipments arrived between the seventh and ninth months. Although there were few calculable shipments
for either commodity, shipments of mats were worth fifty mon per mat, while grasses ranged from 1250
to 2010 mon per bundle. It is unclear how much grass comprised a “bundle,” but it appears that the
unprocessed grasses were likely more valued in the market than the premade mat. Two notes in shipments
from Onomichi in the twelfth month, however, suggest that mats did represent some value. The quantity
delivered is unrecorded, and neither were assessed as taxable goods. Both had notes that they were to be
given to a Lord Yamana (likely the military governor of Bingo province or a member of his family), and a
Lord Shitami. Though the latter’s identity is unknown, the respectful “lord” (dono 殿) designation, and
the fact that these items were tax-exempt, strongly suggest that these were items intended as gifts or
tribute.
284

Cloth shipments were notable from Takasaki (126 bundles) and Kamagari (1463 bundles), with a
small amount sent from Sakihama (twelve bundles). Cloth-bearing ships rarely arrived alone in Hyōgo,
but instead traveled in groups of two to five vessels either solely from Kamagari or Takasaki; or, a
combination of ships from both locations might arrive in Hyōgo on the same day.
285
The original source

284
Ibid., 84.
285
The sole Sakihama vessel was piloted by Shirō Jirō, whose name also appears affiliated with ships from Utazu
(Sanuki province), Sakoshi (Harima province), Minamiura (likely Aki province), Onomichi and Mihara (Bingo
province). While it is unclear if these were all the same man, the timing of the Sanuki-Tosa voyages and the Bingo-

173

of the cloth, while difficult to pinpoint with certainty, may have been from the southern Kyushu area.
286

Taken together with the earlier discussions of rice and salt shipping routes, this reinforces the idea that the
Takasaki/Kamagari ports were key transshipping stops for ships from western Japan and Kyushu.
Kyushu-based products apparently were offloaded and then reloaded on to ships piloted by captains more
familiar with the eastern waters.
Other items shipped through Hyōgo included one cargo of chestnuts from Muro, eight whetstones
from a single Jige ship, and one thousand coins from Kōri (Bizen province). Two other unknown items
were kachi (かち) and nikomi ( 二コ見). The former was shipped three times (four koku total in two
shipments from Ushimado and three koku from Jige), and the latter once each from Moji (two hundred
koku) and Kamagari (150 koku). As most of these items were limited shipments, there is little to note in
the way of shipping trends. It is impossible to determine if nikomi (regardless of what exactly it was)
originated in Suō or Kyushu and was transshipped at Kamagari, or if it was procured directly at Kamagari.
Rice, salt, and cloth from the western provinces of Suō and Kyushu were transshipped at Takasaki and
Kamagari, so it is not impossible that nikomi also originated in the west and was transshipped at
Kamagari.
The scarcity of many of the prestige goods makes it difficult to analyze them effectively, but
nevertheless, local specialization is visible through the trade of these items. The westernmost ports of
Kamagari and Takasaki again suggest their role as a transshipment port for ships from even further west,
as they were the major sources of cloth for Hyōgo. Aki, Bingo, and westernmost Bitchū provinces were
the major shippers of woven mats. Paper was produced in the central Inland Sea area, especially in Bitchū
and Bizen. Chinese grasses formed a supplementary cargo for pottery from Inbe, and also were shipped

Aki voyages sometimes overlap, suggesting there were two different captains. Still, the proximity of Onomichi and
Minamiura to Kamagari and Takasaki, combined with the fact that the Sakihama ship arrived in Hyōgo on the same
day as two Kamagari ships, does not rule out some sort of connection between this Sakihama voyage and the cloth
from the Takasaki-Kamagari area.
286
Matsui Teruaki, “Chūsei no Setonaikai suiun ni okeru Onomichi no ichi,” in Chūsei Setouchi no ryūtsū to kōryū,
ed. Shibagaki Isao (Tokyo: Hanawashobo, 2005), 106. Matsui suggests that the cloth may have come from Korea.

174

mostly from nearby Ushimado. Muro with its chestnuts and Hirayama with its leather also sent single
produced items.
It is also worth noting that many of these items were shipped from relatively major ports, many of
which also functioned as transshipment ports. We have noted Kamagari and Takasaki’s transshipment
roles throughout this analysis. Even through the total volumes of cargo shipped annually to Hyōgo from
both ports was small compared to other areas, the diversity of products from the west shows that they
were a key point in the overall Inland Sea trade network. Onomichi, Tsurajima, and Ushimado were some
of the most heavily trafficked ports in the entire Inland Sea. Hirayama and Shiwaku both shipped
thousands of koku of salt to Hyōgo. Inbe’s total cargo was relatively small, but it was one of the sole
providers of Bizen pottery, discussed in greater detail below. Finally, Sakai’s total cargo shipped to
Hyōgo was comparatively small, but the port itself held an important place as a maritime gateway into the
Kinai area.
287
These ports demonstrate their importance for local markets in both the amount and variety
of cargoes shipping through them. Larger local markets provided a place to sell or trade relatively unusual
items, and a thriving local economy would have allowed residents the freedom to produce surpluses of
their local wares. Although the prestige goods in the Register may have been relatively few and far
between, their presence implies the prominence of local economies and the role of the ports that shipped
them in the greater maritime cultural landscape of the Inland Sea area.

The Register and Archaeological Evidence

Pottery and iron were two items in the Register with a significant archaeological signature that provides
further insight into medieval trade routes and practices. Instead of relying solely on analyzing the
periphery-to-center flow described by the Register, we can incorporate the physical evidence from
archaeological sites that suggests further connections between different locales. Pottery was fired and iron

287
Sakai had long played a role as an intake port for rents and dues submitted to Kyoto and Nara area proprietors,
particularly Kōyasan. Its Kamakura-era trade may have focused more on accommodating China trade ships, but in
the Muromachi era records show that products submitted as rent payments were regularly changed into cash money
at Sakai, and the merchant population transformed from toimaru, or warehouse managers, to ton’ya, or wholesalers.
For more on Sakai, see Morris, “Sakai: From Shōen to Port City.”

175

was forged in limited areas, further suggesting the influence of available resources (in this case, iron ore
and malleable clay) on the production of a specialized trade good.

176

Figure 45. Iron shipments to Hyōgo in the Register and archaeological iron sites.

177

The Register showed only two ports shipping iron to Hyōgo: Onomichi in Bingo province (one
hundred horseloads) and Setoda in Aki province (120 horseloads) (fig. 45). As all entries indicate
shipping with other cargoes, it is difficult to calculate the value of the iron alone. The Chūgoku mountain
area, on the Honshu mainland inland from Bingo, Bitchū, and Bizen provinces, was rich in iron ore, and
iron mined there was likely brought to Onomichi to be shipped to Hyōgo. Setoda’s iron shipping is
particularly notable as it was an island without a significant local iron mine, yet it shipped even more iron
than mainland Onomichi that had more direct access to the iron supply. The travel diary Inryōken
nichiroku (fifteenth century) contains an entry regarding the procurement of iron to forge a sword,
describing how one hundred men were required to carry twenty horseloads of iron. It would have been far
more economical and efficient to ship the iron by boat.
288
Setoda’s role as a major iron shipper, then,
likely reflects that practicality and demonstrates yet again the tendency for places producing iron to look
outside their own borders for trade and shipment of locally produced resources.
Examining the locations of iron forges further supports that idea. Kamakura- and Muromachi-era
remains of iron forges have been uncovered at several sites in Aki, Bingo, Bitchū and Bizen provinces,
nearby a local supply of iron. These sites are most often identified by the presence of fuigo, a component
of the bellows used in iron smelting. A lone site in modern Hokudan-chō, near the medieval port of
Murotsu in Awaji province, possibly suggests that Murotsu imported ore from elsewhere to be forged
there. That cannot be assumed, however, as forging iron through the use of iron sand was also a common
practice. The Hokudan-chō site may have used local iron sand instead of importing iron from elsewhere.
Also notable is the lack of any records of Bizen province ports shipping iron to Hyōgo. Areas along the
Asahikawa River were known for their iron production since the Heian period, and would ship their
products in riverboats to the Kojima peninsula, where they would be transshipped to the Kinai.
289
Yet

288
Imaya Akira, “Tetsu wo hakonda fune,” in Zoku tetsu no bunkashi: Nihon no kiseki to tōzai no kaikō, ed. Shin
Nihon seitetsu kōhōshitsu, New Edition (Toyo Keizai Shinposha, 1988), 50.
289
Matsushita, Umoreta Minatomachi Kusadosengen, Tomo, Onomichi, 218.

178

according to the Register, iron was not an export item from that area.
290
Neither the information from the
Register nor the additional information from the archaeological sites fully explains the complex
particulars of iron trade routes.
Incorporating archaeological evidence concerning pottery is more revealing. The Register notes
that small jugs (tsubo 壺, written ツボ in the Register) were shipped to Hyōgo from Sakai, Katakami, and
Inbe (fig. 46). Suribachi ( 擂鉢, written スリはち in the Register), shallow bowls used in cooking, were
shipped only from Jige, and only once throughout the entire year of our extant records. There is a notation
in one shipment from Ushimado that records six suribachi as part of the fees levied on a shipment of
egoma seeds. As that pottery was not recorded as part of the official taxable cargo, it is an unusual case
and cannot necessarily be factored into an analysis of typical cargo shipments. While the pottery type is
unspecified in the Register, the total volume shipping from the Bizen area combined with the presence of
kiln ruins from the Inbe area strongly suggests that the majority if not all of the pottery was Bizen ware.
291

The pottery that came from Sakai is more questionable, but in that case the pottery was brought in on the
same boat that contained Kojima salt, also a Bizen peninsula product. This implies that that vessel also
visited or was shipping items from Bizen, and could therefore have been transporting Bizen ware to
Hyōgo as well.

290
The possible exception here may be the port of Kōri. One ship brought 1000 coins to Hyōgo. Since it is not clear
where those coins were manufactured, though, it is impossible to say whether they were made of iron from local
forges or were perhaps even imported Chinese coins.
291
San’yō Shinbunsha, Kaitei no kobizen: Mizunoko Iwa gakushutsu chōsa kiroku, 62. One kiln in the Furōzan area
of Inbe was nearly forty meters long, which suggests that it had been producing pottery from at least the early
Muromachi era.

179

Figure 46. Pottery shipped to Hyōgo in the Register and archaeological sites.

180

While the Register provides information about the flow of Bizen pottery from its production point
to Hyōgo, that is only a small part of the entire picture. In looking at archaeological site reports from
along the Inland Sea, Bizen ware is ubiquitous. Excavations in terrestrial settlements in Bingo province
(Onomichi, Kusado Sengen), Sanuki (Nohara, Michi-shita site), Harima (Aboshi, Murotsu-Hyōgo),
292
and
even far inland in Aki (Yoshikawa Genshun site) reveal significant amounts of Bizen ware, ranging from
suribachi cooking dishes, to tsubo jars, to larger vats and other implements. Underwater recoveries, which
were not systematic excavations but serendipitous discoveries often brought up in fishing nets, also dot
the Inland Sea. These were mostly located in the central inland sea area at sites near Naoshima, Aji, and
Kugui-no-oki, and likely are the remnants of either shipwrecks or jettisoned cargo. The final type of site
is represented only by Mizunoko-iwa, which was a shipwreck site that yielded nearly two hundred pieces
of Bizen pottery.
293

Looking at the sites alone certainly suggests a wide variety of usage for Bizen ware throughout
the Inland Sea. Not only flowing from the production point east to Hyōgo and north to the Kyoto capital,
Bizen ware was also shipped out to destinations far west of the kilns. More Kamakura- and Muromachi-
era Bizen ware has been discovered on Honshu than on the western areas of Shikoku, possibly suggesting
less direct interaction with the estates in Sanuki and Iyo. Two sites are of particular interest. The first is
Aboshi in Harima, a destination between the Inbe kilns and the Hyōgo checkpoint. A number of pieces of
Bizen pottery from the fifteenth century were found here, especially larger vats (kame 甕).
294
They show
evidence of use and thus were likely imported to Aboshi as a final destination.
295
If much pottery were
found in one area with no signs of usage, this would more likely indicate a transshipment location.
Information from the Register shows Aboshi as a small-to-medium sized port, which sent sixty-two ships
to Hyōgo over the course of the year, usually carrying less than one hundred koku of goods. It does not

292
Murotsu-Hyōgo is different from the port of Murotsu in the Register. The latter is in Awaji province.
293
San’yo Shinbunsha, Kaitei no kobizen: Mizunoko Iwa gakushutsu chōsa kiroku, 80–81.
294
Nakagawa Takeshi, “Ceramics Unearthed at the Furuaboshi Site in Hyōgo Prefecture,” Trade Ceramics Studies,
2008, 119.
295
Ibid., 124.

181

appear to have been a major source of transshipping, but the evidence of Bizen ware used in Aboshi
suggests that there was direct trade with other venues along the Inland Sea. Instead of being a
transshipment port, however, Aboshi seems to have mostly been a production port that imported certain
wares.
The second significant archaeological site is the shipwreck at Mizunoko-iwa. Located
approximately six kilometers offshore of modern-day Shōdoshima, it is one of the only known
discoveries of a Muromachi-era shipwreck site. Although the ship itself has been lost to the elements, the
pottery cache recovered yielded thirty-five large jars (kame), seventy-one large jugs (tsubo), fourteen
small jugs (tsubo), and sixty-six cooking bowls (suribachi), totaling 189 objects (fig. 47). This provides a
clear picture of the minimum number and types of items that were shipped longer distances. It is unclear
where the ship’s final destination might have been, though it may have been the Kumano area on the Kii
peninsula. Perhaps the pottery was commissioned by the Kumano Shrine itself. Bizen ware has often been
discovered along the coast of Wakayama prefecture, and the wreck site places a ship leaving the kilns at
Katakami and Inbe squarely along that route.
296
The first archaeological evidence of ballast stones used
on ships also comes from the Mizunoko site. Although the origin of the ballast stones has not yet been
identified, studies have shown that they were not from the region near the kilns, which suggests that the
vessel had loaded them at yet another port, possibly in the Kishū or Shikoku area.
297
Again, this is clear
evidence of direct interactions between different locales.

296
San’yo Shinbunsha, Kaitei no kobizen: Mizunoko Iwa gakushutsu chōsa kiroku, 65.
297
Ibid., 87–88.

182

Figure 47. Sample of Mizunoko-iwa artifacts. Back Row: two kame and a tsubo. Front row: small suribachi, medium and
small tsubo, stacked suribachi. Photo by author (2012). Used with permission of the Okayama Prefectural Museum.
The pottery from the Mizunoko site combined with other archaeological evidence suggests a
possible reinterpretation of the notation styles in the Register. For pottery shipments, the entries show the
number of tsubo and the notation “large/small” ( 大小). The meaning for this is unclear: it may refer to
sixty large and small tsubo in a shipment, or sixty large and sixty small. The number of pottery items
being shipped through Hyōgo ranged from fifteen to one hundred tsubo per shipment – vastly smaller
numbers than those recovered from the Mizunoko site. At another site near Kugui-no-oki in modern
Okayama prefecture, a large Bizen-yaki vat was recovered in a fishing net. Within that large vat were two
smaller jars (fig. 48).
298
This packing method would not only have saved space onboard ship, but would

298
NPO hōjin suichū kōkogaku kenkyūjo, “2009 Nendo Kagawa-ken, Okayama-ken no shiryō chōsa ni tsuite,”
Suichū kōkogaku kenkyū, No. 3 (March 2010): 74.

183

have provided space to use packing materials around the smaller jars for safer transport, which
corresponds to the abovementioned combination of a joint cargo of grasses and pottery.
299
It is possible
that the large/small notations in the Register may refer to this packing process, with smaller jars packed
within larger ones in double the numbers recorded.
300
If this were indeed the case, then the numbers of
most of the shipments to Hyōgo would be closer to the numbers found at the Mizunoko shipwreck site.

Figure 48. Diagram of jars packed within a larger vat, from Kugui-no-oki site. NPO hōjin suichū kōkogaku kenkyūjo
(2010, p. 74).

Looking at the archaeological evidence in conjunction with the documentary evidence reveals a
much wider range of trade, particularly in the case of pottery. Even with evidence from the Register about
transshipment of salt or rice, much of that is couched in speculation and deeper analysis of the
correlations between cargoes, nomenclature, and the captains and warehouse managers who handled the
cargoes. The underwater recoveries of the Bizen pottery with jars packed inside jars even suggests
alternate ways to interpret the notations of the Register. The tangible evidence remaining of both iron and

299
Chapter 4 will also show evidence from maritime laws that suggest using rice as a packing material around
fragile items.
300
Yoshiko Makabe, “Bizen-yaki memo: ‘Hyōgo Kitaseki Irifune Nōchō’ no tsubo to suribachi” in Kōkogaku no
shiten: Seikatsu ishiki no kōkogaku (Okayama-shi: Unizuka Insatsu Kabushiki Gaisha, 2009), 232.

184

pottery provides incontrovertible direct evidence of the flow of goods between locales, rather than simply
the shipping of goods from the points of production in the periphery to the central markets.

Jige: A Special Case

Boats that are labeled as coming from “Jige” were something of a special case. This notation refers to
boats that originated from nearby Hyōgo, claiming it as their home port. The majority of ships (295 out of
1978 entries) passing through the checkpoint throughout the year were registered as “Jige,” and as a result,
the volume and value of the goods from Jige ships were the highest of all ports in the Register. It is
possible that Jige ship captains voyaged to other areas and picked up trade goods to bring back through
the checkpoint. But it is also likely that other captains brought their wares to trade directly with local Jige
captains, who would then bring those goods headed to the inland markets through the Hyōgo checkpoint.
Examining the timing of the ships manned by captain Awaya/Saemon Gorō
301
and the goods he
transported through the checkpoint as one case, it becomes apparent that he could not have personally
gathered each type of good at its production point. In one example, he passed through Hyōgo on the fifth
day of the fourth month with salt from Mihara, and then again four days later with salt from Bingo. The
distance from Hyōgo to the Bingo area is approximately two hundred kilometers; with the sailing
technology available at that time, such a round-trip journey would have been close to impossible. It is far
more likely that he brought in another person’s shipment of Bingo salt, either from a transshipment port
or from a Bingo-province captain who brought Awaya/Saemon Gorō his own product for individual trade.
The reasons for such a large amount of local trading nearby Hyōgo are not totally clear, but one
hypothesis involves the role of warehouse managers. In the Register, few of the entries for ships from
local ports (most notably those registered to Jige, Amagasaki, or Kuise) included the name of a warehouse
manager. It is possible that the local captain acted as his own agent,
302
allowing the captain to avoid going
through a middleman and thus needing to pay a third party an additional service fee. Similarly, captains

301
These records specify both names in each entry. It may have been meant to designate a Saemon Gorō affiliated
with Awaya, likely representing a particular warehouse, from another captain with the same name.
302
Hayashiya, Hyōgo Kitaseki Irifune Nōchō, 291.

185

from more distant ports might have local connections and traded their goods to the local captains at a fair
price, offloading their wares without even needing to pass through the checkpoint. As will be seen in the
next chapter, ship captains had a widespread maritime network that encouraged collaboration among
many parties, and local unofficial trade may have been another manifestation of those practices.
Jige cargoes saw the widest variety of goods from any single port. Five different types of fish and
seaweeds, all types of wood products save for magnolia, four different styles of rice bales, thirteen of the
seventeen recorded types of salt, sesame, grasses, mats, pottery, wheat, whetstones, barley, indigo, and
beans: no other port in the Register had such diverse shipments as those noted “Jige.” Most of the ships
were relatively small, with only eight vessels bringing more than two hundred koku of goods through the
checkpoint. Twenty-three brought between one and two hundred koku, sixty-three between fifty and one
hundred koku, and one hundred seventy-one ships brought less than fifty koku worth of goods through the
checkpoint. Though they may have held smaller cargoes, the high frequency of voyages combined with
the variety of items suggests the ability to procure and turn over goods quickly. A well-connected
network would have been critical to such operations.

Conclusion

The economic development in the Inland Sea both shaped and was shaped by various aspects of the
maritime cultural landscape: the geography, environment, and infrastructure as represented by the
different types of ports. Simply looking at the total volume of cargo shipped from each of the ports
immediately marks Onomichi, Setoda, Ushimado, Yura, and Jige as important shipping hubs, but that was
far from the complete picture. Geography played a part in the specialization of transshipment or
production ports, both due to its influences on the types of goods that could be produced in particular
climates as well as to simpler aspects such as location or easily navigable waters. As noted in Chapter 2,
Onomichi had long been known as an important port due to its deeper waters that allowed larger ships to
moor more easily. The total volume and various types of goods it shipped to Hyōgo demonstrate that its
role as a transshipment hub continued in the late medieval period. Less immediately obvious was the role

186

of Kamagari or Takasaki, which shipped relatively small amounts of goods to Hyōgo. This fact is brought
into focus by noting the volumes of goods that originated in Suō and Nagato provinces as well as in
Kyushu, and then were transshipped through Kamagari and Takasaki. It was likely the geographical
location of those ports, falling west of the island chains that hid the pirate enclaves, that made them safe
transshipping hubs. Western captains, less familiar with those treacherous waters, could offload their
goods at Takasaki and Kamagari and defer to the captains who were used to navigating those areas.
Tsurajima, too, shipped relatively smaller amounts, but transshipped salt and red rice as well as the luxury
goods of mats and leather. Combined with its proximity to the mouth of the Takahashi River that
connected it to inland points, it held an important role as a local hub and gateway to the inland areas.
In other cases, different products could be procured more easily than others due to the geography
and climate. Marine animals were some of the best examples of this, as certain fish were more prevalent
in particular areas, such as abalone from Setoda or parrotfish from Hiketa. Alternatively, there were items
such as sea cucumbers or sardines that were likely shipped fresh, resulting in a thriving local industry at
Muro and adjacent ports close to Hyōgo. Lumber products were also clearly influenced by geographical
factors, including the presence of forestland or saltmaking facilities. Many of these items came from areas
of Shikoku and Awaji that had not been severely deforested or did not prioritize lumber use for salt firing.
The ability to procure goods in quantity increased the volume shipped from any particular port, allowing
that port to become a major or minor area hub.
Seasonality was also a factor in local shipping and production of goods. The Register clearly
demonstrates evidence of double-cropping in the seasonal fluctuations of rice, beans, wheat, and barley,
suggesting that farmers were maximizing their ability to make a profit from their cropland. Certain types
of fish also were only shipped in particular months. The ports that shipped those fish, particularly ones
that claimed a higher price such as mullet or mackerel, may have found their local economies fluctuating
yearly with their take in any given season. Higher yields would have allowed them to ship more as a
profit, while lean years may have caused problems.

187

Personal connections and power were part of the human factor of the maritime cultural landscape.
Though difficult to understand from the Register alone, it is possible that individual connections may
have facilitated the lumber trade between the Kii peninsula and the port of Yura. Personal networks
almost certainly influenced the pricing of salt, as seen through the inflated prices of Bingo and Kojima
salt and the near-monopoly enjoyed by a handful of warehouse managers. Ships local to Hyōgo (denoted
as “Jige”) likely enjoyed enough personal connections to eschew the services of a warehouse manager,
and those captains could process their own goods. The next chapter will demonstrate in more detail how
ship captains collaborated as well, strengthening those personal networks. With the tightening and
loosening of different personal networks and control, profits would ebb and flow accordingly, further
influencing the standings of particular ports and people.
Finally, certain items began to be associated with their production points, taking on a “brand
name” status. This “brand name” trend was best reflected in salt production, with salts labeled by the
names of their production point. Salts from Bingo and Kojima commanded the highest prices and the
warehouse managers who oversaw their intake at Hyōgo had the tightest networks. Red rice became one
such item associated with ports in the Sanuki area. While indigo did not take on the names of the ports
with which it was associated, it became known as an Awa product, centered in Sōjiin, Muya, and
Tosadomari.
The legacy of some of these products is capitalized on even in the present day. Salt from what
was the Bingo-Aki area is still promoted as a high-quality item, and is often found in locally produced
confections and snacks.
303
Hakata, a port from the Register that shipped only Bingo salt to Hyōgo, today
is the home of the “Hakata no shio” salt factory that promotes its product as retaining the “saltiness of the
ocean water.”
304
Visitors to Tokushima (formerly Awa) prefecture can visit the Aizumi-chō Historical
Museum to learn about indigo production, and throughout the prefecture can visit smaller indigo dyers

303
JB Highway Service Company Limited, “Ohama Souvenirs,” accessed March 23, 2015, http://www.jb-
highway.co.jp/souvenir/ohama_up.php. The parking area souvenir shop on Innoshima offers salted jelly, salt bean
cakes, salted chocolate, salted caramel, and more.
304
“Hakata No Shio,” accessed March 23, 2015, http://www.hakatanoshio.co.jp/. Ironically, today Hakata no Shio
uses salt from Mexico and Australia as raw ingredients for its processed salt.

188

and experience the dyeing of indigo firsthand. These products, so critical to maritime shipping and the
development of the local economies in the medieval period, even today demonstrate the long-lasting
effects of that centuries-old maritime cultural landscape.

189

Chapter 4: Rules of the Sea: Seafaring Practices and Collaboration

I have so far explored the infrastructure associated with maritime shipping. As seen in the previous
chapter, the quantity of trade items shipped along the Inland Sea was considerable. This constant flow of
trade required not only the physical infrastructure to support it, but also manpower to service it.
Fisherpeople and saltmakers in the production ports were not enough to support the growing maritime
trade. Skilled sailors and ships’ captains were responsible for transporting goods from place to place, and
as trade continued to grow, so did their influence. They faced different challenges from land-bound
travelers. There are no clearly visible “roads” in maritime travel; boats are subject to wind and waves that
can drastically alter the course of sail. Authorities therefore often found it more difficult to monitor
maritime travel, resulting in the sailors and captains enjoying an unusual autonomy.
This autonomy manifested itself in what Paul Rainbird terms a “mariner’s perspective.” He
argues that seafarers and those who lived in maritime communities shared particular experiences that thus
created a distinct culture in comparison to the wider society.
305
The concerns and challenges they faced
differ from those in non-maritime communities and therefore resulted in new customs and codes that were
at best peripheral and at worst inapplicable to land-based workers and travelers. This chapter will examine
how ships’ captains and others reacted to the growing maritime trade, particularly as seen through the
development of sets of maritime codes or regulations. Geography was again a factor, and different
regional adaptations and practices are apparent in the different versions of the codes and in the records of
actual captains’ voyages. Yet the widespread adoption of the maritime regulations and the commonalities
of practices throughout the Inland Sea suggests that for the most part the mariner’s perspective knew no
geographic constraints. I argue that seafarers took advantage of the autonomy offered by their
occupational mobility to develop a complex network of collaborative practices, both among themselves
and with littoral residents, and that that autonomy transcended physical and administrative boundaries.

305
Rainbird, The Archaeology of Islands, 49.

190

Inland Sea mariners enjoyed high mobility and the freedom to regulate their own world – an unusual
combination in the medieval period.

Laws of the Sea: The Maritime Codes (Kaisen shikimoku)

Much has been made of the rules and regulations governing overland travel. Entire books have been
written about the checkpoints along the roads throughout Japan, instituted as part of the ritsuryō codes in
the seventh and eight centuries and surviving through the Edo period. The idea of a mariner’s perspective
suggests that seafarers should have had similar codes to govern water-based transportation, but serving
the particular needs of a maritime community. And indeed, examining six sets of extant kaisen shikimoku
or kaisen taihō (maritime codes or laws)
306
said to date to the Kamakura era shows that seafarers did
indeed create specific sets of rules that delineated the responsibilities and rights of various parties
involved in the transport of goods.
There are few other such contemporaneous compilations under the name taihō, suggesting that
this was an extraordinary set of codes complied to respond to specific circumstances.
307
Historically and
globally the sea has been a contested space; it is not often subject to ownership and therefore it is difficult
to determine who should administer justice in the case of a maritime dispute.
308
There are countless
examples of legal documents detailing disputes on land in medieval Japan, with appeals from estate
residents to authorities regarding property ownership, taxation, boundary violations, or theft. In those
cases, residents would submit their paperwork through proper local authority channels and await
judgment from above. At sea, however, the legalities become murkier. Questions of responsibility when a
ship wrecks are not so easy to resolve if the ship is from one place but wrecks off the shore somewhere

306
Kaisen 廻船 literally means “a ship making the rounds,” and usually refers to cargo ships that stop from port to
port to trade goods. While many of the laws do indeed refer to kaisen, there are also references to riverboats, port
protocols, miscreant personnel, and other non-ship specific items. I have therefore translated the title of these codes
as “Maritime” to imply that wider range of seafaring concerns.
307
Shinjō, Chūsei suiunshi no kenkyū, 502–503. Shinjō notes that the other documents using that term are generally
concerned with selling humans.
308
Philip E. Steinberg, The Social Construction of the Ocean, Cambridge Studies in International Relations
(Cambridge; New York: Cambridge University Press, 2001) 39-67. One of the most telling examples of the
confusion about maritime jurisdiction dates to Roman imperial maritime law. The Romans claimed the right to
command in the Mediterranean, but did not claim ownership of the seas, a subtle but distinct difference (64-66).

191

else, or if pirates abscond with a boat and abandon it in an entirely different area. If a captain facing the
threat of capsizing jettisoned another man’s cargo but the cargo owner felt it was an excessive response to
the impending storm, who was the arbiter of that dispute if it occurred miles offshore?
Mariners therefore took it upon themselves to compile codes addressing these issues and
submitted them to the authorities, hoping for widespread adoption. As we shall see, aside from occasional
references to the need to consult with local authorities, for the most part the codes call for disputes to be
resolved by the involved parties themselves, independent of consultation with governing bodies. Seafarers
thus were relatively autonomous and removed from the usual process of legal consultation, as they held
jurisdiction over their own ships and cargo on the seas. Late medieval and early modern Dutch captains
were allocated the right “in the absence of other judges” to pass judgment on foreign offenders against
their authority, but they sailed under their country’s laws and in a dispute needed to adhere to those
rules.
309
The Japanese kaisen captains, instead of sailing under central authority, proposed their own set of
codes to the government in the hopes of universal adoption, effectively creating their own governance. A
careful analysis of the items recorded in these codes offers insight into the world of maritime shipping
and its denizens, including hardships faced, dispute resolution, and sailing logistics.
The origins of the codes remain something of a mystery. The texts are dated the third month
sixteenth day of Jōō 2 (1223) and are published in Kamakura ibun (Records of the Kamakura Age), a
series of volumes published in the late twentieth century that has compiled remaining documents from the
Kamakura period. Japanese scholars agree, however, that the codes are more likely to date from the late
Muromachi era. It is unclear why the codes would have been backdated as far as the Kamakura era,
though it may have been an attempt to argue that the ideas codified within were actually long-standing
precedents. The oldest extant versions of the codes, however, are Edo period (seventeenth – nineteenth
centuries) copies of codes with dates from the fifteenth and sixteenth centuries.
310
Furthermore, Kubota

309
Lauren A. Benton, A Search for Sovereignty: Law and Geography in European Empires, 1400--1900 (Cambridge,
New York: Cambridge University Press, 2010), 134–135.
310
Shinjō, Chūsei suiunshi no kenkyū , 495. In the following discussion, KI 3078 is dated 1573, and KI 3073 is a
copy from 1775, though both are collected in the Kamakura ibun.

192

Hiroshi has pointed out that the use of the term “kaisen” itself rarely appears in any documents prior to
the Nanbokuchō and Muromachi eras (fourteenth – sixteenth centuries) and became more commonly used
in the late Sengoku and Edo periods (sixteenth – nineteenth centuries), lending further credence to the
idea that the codes were not as old as the 1200s.
311
Even if they were penned as late as the sixteenth
century, however, the practices recorded were likely in place long before formal codification. As such, I
believe that the customs outlined within them were likely practiced widely in the 1400s, or the time
period of this study.
The attributed authorship may also have been a deliberate attempt to make the codes seem
universal.
312
Most versions of the maritime codes note a trio of authors from places as diverse as Hyōgo,
Bōnotsu in Satsuma, and Urado in Tosa. There is no traceable relationship among the three individual
authors, although one scholar suggests that they may have worked together along maritime routes
culminating at Tsuruga (Echizen province, modern Fukui prefecture), but that is a tenuous connection at
best.
313
The three locations roughly correspond to three corners of the Seto Inland Sea, and having
representatives from each of those ports may indicate agreement on the need for universal laws across the
region. Other versions of the codes from Owari, Ise, and elsewhere also attribute the original authorship
to that trio, and suggest additional amendments.
314
Of the six versions discussed here, four cite the
Hyōgo-Satsuma-Tosa authors, one (KI 3070) notes the original attribution but also cites Ise, Shima,
Owari, Mikawa, Omi, and “other provinces” as adhering to the codes, and the sixth version (KI 3068,
dated 1573) notes “three ports and seven harbors” that will adhere to the codes: ports of Ise, Hakata,
Izumi; harbors at Echizen (Mikuni), Kaga (Motoyoshi), Noto (Washima), Etchū (Iwase), Echigo
(Imamachi), Dewa (Akita), and Ōshū (Tsugaru Jūsan). To facilitate the discussion of the different
versions of the codes, the following sections will refer to each of them as follows:

311
Kubota, Kaisen taihoko, 8–9. The term only appears three times in records from the Heian period, with the
earliest referring to a trade vessel from China (Heian ibun, Jōkan 5 (863) 11.13, in a record by the monk Enchin).
312
Also noted in Shapinsky, Lords of the Sea, 63.
313
Shinjō, Chūsei suiunshi no kenkyū, 496.
314
See Kamakura ibun 3070.

193

Referred name Kamakura ibun (KI)
number
Authors’ Geographic Affiliation
Seven-harbor version 3068 “Three ports and seven harbors,” as above
Version A 3069 Hyōgo-Satsuma-Tosa
Ise-Shima version 3070 Ise, Shima, Owari, Mikawa, Omi, and “other
provinces”
Version B 3071 Hyōgo-Satsuma-Tosa
Version C 3072 Hyōgo-Satsuma-Tosa
Version D 3073 Hyōgo-Satsuma-Tosa
Table 12. Versions of the Maritime Codes.
Even accounting for certain site-specific variants, the common contents suggest a widespread
following. Including later texts, versions of the maritime codes have been found associated with
communities from as far north as Hokkaido and as far south as Kyushu – literally from coast to coast.
315

While the original codes were probably submitted to shogunate to be considered for use in the Inland Sea
area and took time to disseminate throughout the land, the fact that they were so widely adopted does
imply that mariners shared common values regardless of geographic location and that they realized the
practicality of these codes.
The collections of maritime codes discussed below range from thirty to forty-three individual
articles.
316
Categorizing the types of codes reveals some of the major concerns about maritime shipping
and travel. Personnel issues are a common theme, including the hiring of sailors and the rights and
responsibilities of the ship owner, ship’s captain, cargo owner, and the individual who commissioned the
voyage. Profit was of course a major concern of any trading ship, and many of the codes delineate cargo
allocation, loading, and pricing. Steps to deal with maritime disasters such as storms and piracy are
addressed as well. There are a number of items detailing the practical matters of sailing, including tools

315
Kubota, Kaisen taihoko, 8. In this monograph, Kubota has complied twenty-six different versions of the codes
from the Sengoku and Edo periods to compare linguistic differences and meanings. Those versions are associated
with places as diverse as Esa in Hokkaidō, Tsuruga in Fukui prefecture on the Japan Sea, Bōnotsu on Kyushu, and
Hachijōjima, nearly two hundred miles south of modern Tokyo.
316
In many cases the discrepancies in the numbering result from the rewording of a particular item, separating one
longer law into two shorter ones or combining two items into a group.

194

and shipboard provisioning, as well as intangible concerns including taboos and defilements. A more
detailed examination of the codes will suggest ways in which the mariner’s perspective manifested itself
in the shared responsibilities of operating a trading ship.
Cooperation among many people is critical to a smooth sea voyage. In reading the various articles
in all six versions of the regulations, it becomes clear that collaboration among the ship owner (senshu 船
主, or in some cases “lender” or kashite 貸し手), the ship’s borrower (karite, 借手), the captain (sendō,
船頭), and the cargo owner (ninushi 荷主) was the foundation for trade voyages. These may have all been
disparate individuals, or a single person may have taken on more than one role simultaneously – the codes
are not specific on this point. The fact that these four roles are clearly noted, however, suggests that it is
possible that each could have been a different individual.

Contracting for a Voyage

The first step of a journey required that the ship be contracted for sail. Although there are no extant
contracts that delineate an agreement between a ship owner and a borrower, the codes clearly show that
such contracts existed and were a standard practice before starting a voyage.
317
The items related to
contracting show the primacy of the ship’s captain. As the items below will show, he was responsible for
the ship, and many of the items concerning contract fees ensure that he was repaid for his work.
Several of the articles refer to “promises” (yakusoku 約束) among the various parties, and they
discuss procedures when those promises go unfulfilled. All of the collections, with the exception of
Version B, emphasize the responsibility of the ship owner to uphold his end of the contract, saying that if
the lender provides a ship different from the agreed-upon vessel, the borrower has the right to exchange
the ship and receive the new one as his own, apparently indicating that he became the new ship owner.
318

317
A record from 1324 sent to Tōji from the residents of Yugeshima (Iyo province) notes the possibility of
borrowing ships from Aki and Bingo provinces to transport rents and dues (KI 28836), indicating a precedent for
borrowing and lending of boats.
318
KI 3068-30, 3069-31, 3070-30, 3072-40, 3073-40.

195

Unsurprisingly, a number of articles focus on payment issues, including the allocation of profits,
transport fees, and fees for use of the cargo space. All six versions of the codes contain an article
specifying payment of transport fees:
Item: On borrowing a boat, when collecting transport fees on its return, one-third of the
fees are to go to the ship owner to use as he pleases. However, when borrowing it, if on
its return it is carrying cargo, the ship owner can take less than one-third [of the fees].
319

The Seven-harbors, Ise-Shima, and Version A collections use the term “captain” instead of “ship
owner.”
320
This terminology is not consistently applied throughout each collection, as other articles do use
the “ship owner” characters, so these may have been regionally specific variants indicating different
people and different practices. In this item, we see the distribution of labor and charges for borrowing a
vessel. It is unclear exactly who would be collecting the transport fees, since only a one-third share would
go to the ship owner or captain. It is possible that the remaining fees would have gone to the other party
(ship owner or captain, whoever was not the original recipient) to pay for his services. Although not
explicitly stated, the fact that the recipient could choose to take less than the one-third standard fee if
cargo was on board implies that in lieu of cash, a share of the cargo itself was also acceptable. The
implications for this on local economic development are worth noting. While that share of the cargo may
have been simply for personal use, it is also very likely that the recipient could have sold it locally for
additional profit.
All six versions placed responsibility on the ship’s captain for his actions en route, even if he
were in a borrowed ship. If the captain was involved in a court dispute (kuji, 公事)
321
at the destination
and his ship was taken from him, then the captain was to pay reparations.
322
While the recipient of the
reparations is not specified, it seems likely that it was the ship owner. Although the borrower would
normally have been liable for the ship’s safe return, since the captain was the person involved in a court
case en route, the onus fell on him. An additional article says there will be unspecified “problems for the

319
KI 3068-22, 3069-23, 3070-22, 3071-21, 3072-19, 3073-19.
320
KI 3068-22, 3069-23, 3070-22.
321
While commonly translated as a form of taxes or court duties levied on individuals, during the Muromachi era
kuji was also used to denote a court case or suit (soshō 訴訟).
322
KI 3068-23, 3069-24, 3070-23, 3071-22, 3072-21, 3073-20.

196

captain” if he is found piloting a ship that had been stolen by pirates and subsequently sold, regardless of
where in the archipelago the ship was taken or resold.
323
The captain therefore had some responsibility to
ensure that the ship he piloted was legitimately owned.
The Inland Sea-based versions of the codes (versions B, C, and D) also include a clause
pertaining to additional agreements struck between the captain and cargo owner. Though there were some
variants in wording, the common theme was that even if the cargo owner deviated from agreed practices
and therefore loading cargo caused delays, the captain was still to collect transport fees as per the initial
agreement.
324
The fact that these regulations are present only in three of the four Inland Sea-centered
versions suggests that such delays were more common in these areas. It is possible that the number of
small island ports, the high rate of interaction and collaboration between ship captains (discussed in
greater detail below), and the development of ports such as Ushimado for transshipment (as discussed in
the last chapter) all combined to create an environment of occasional confusion. Interactions between
cargo owners and captains from disparate regions would have been common, and individual quirks in
cargo loading practices may have been more difficult to overcome, causing delays. This article is also the
only explicit evidence of contracts being struck directly between the captain and the cargo owner,
suggesting the captain’s autonomy. There are articles referencing contracts between the ship owner and
the borrower, and the ship owner and the captain, but for the captain to be able to strike independent deals
with the cargo owner implies that the captain had additional power over his own ship to decide the types
and methods of loading cargo. While the ship owner may have agreed to the original parameters of the
voyage, the captain seems to have had the flexibility and responsibility to arrange matters aboard ship.

Provisioning and Manning the Ship

After the voyage was contracted, the ship needed to be fitted out and manned. Many items outlined the
types of provisions needed on board, and who was responsible for repayment in the event of their loss.

323
KI 3068-7, 3069-6, 3070-7, 3071-11, 3072-33, 3073-33.
324
KI 3071-33, 3072-27, 3073-26.

197

Interestingly, the items regarding the ship’s infrastructure – line (ropes), anchors, and the like – are
largely universal to all six versions. This suggests the truest manifestation of the mariner’s perspective, as
regardless of where the ship sailed it still faced the same basic structural needs. Manpower concerns,
however, demonstrate some geographical differences. This suggests that while the inanimate objects
associated with seafaring may have generated common concerns, human nature resulted in different local
practices that needed to be addressed.
With regard to tools and physical provisions, when someone borrowed a ship, the owner was
responsible for making the vessel seaworthy. Several items discuss overtly or imply the types of tools that
should be supplied when someone borrowed the ship. All six versions of the codes note that when a ship
was borrowed, “various tools” (shodōgu 諸道具) were also to be prepared and handed over to the
borrower at that time, implying that it was the ship owner’s responsibility to supply them.
325
Half of the
collections – Versions B, C, and D – also suggest that the ship owner was expected to absorb some of the
costs of wear and tear on other implements, such as line (ropes), mats, or anchors. The borrower was
directed to return the ship with the line and any woven mats in the same condition in which they were
supplied, although if there were only one or two lines missing, that did not need to be reported back to the
owner.
326
Another item in all six versions further clarifies regulations about ropes, saying that if a line
broke during the voyage, then reparations were not needed. If it let loose and fell in the water, however, it
needed to be paid for, as would any anchor lost en route.
327
Version D also incorporates the provision that
the rudder and any other tools needed to be repaid if lost at sea.
328
An additional item in all versions of the
codes stipulates that the borrower had to make reparations if the propulsion systems were damaged. All
the collections but Version A specify the rudder post, but Version B is the only one to include the sculling
oars (ro 櫓).
329
Version A does not mention the rudder, but instead requires reparations to be made for

325
KI 3068-13, 3069-12, 3070-13, 3071-27, 3072-20, 3073-32. The wording suggests that the tools are included
when a borrower contracts with the ship owner for a vessel.
326
KI 3071-13 and 14, 3072-11 and 43, 3073-10 and 43.
327
KI 3068-12, 3069-11, 3070-12, 3071-26, 3072-31 and 32, 3073-30 and 31.
328
KI 3073-31.
329
KI 3068-1, 3070-11, 3071-25 (also includes the oars), 3072-25, 3073-24.

198

damages to the mast, unless the mast had been in poor condition when the boat was originally
borrowed.
330
These specifications demonstrate the variety of types of propulsion used for the watercraft,
including oars and sails.
These articles clearly demonstrate items of economic importance to the owner, as well as an
understanding of what can happen at sea. Line and mats in particular are items that need replacing on a
regular basis, and line can easily snap while in use. While these items show that the ship owner accepted
responsibility for those basic supplies, it also placed the responsibility for competent sailing upon the
borrower. Rough seas could cause the loss of an anchor or even the rudder, which on Japanese ships was
exposed to the elements and often snapped off the stern. But the responsibility for those losses fell on the
shoulders of the borrower, who had to replace more permanent, costly items. Anchors, especially, were
becoming more of an investment in the Muromachi era. Earlier anchors had been made of stone or wood,
but as iron forging technologies improved, anchors could be produced out of iron – a strong, heavy
material that would take up less space on board ship than a stone anchor, but one which also cost more to
produce in terms of time, manpower, and resources.
331

Shipworms or other wood-damaging insects were a concern for mariners, as two items common
to all six versions attest.

Both items discuss the fact that the ship can be eaten by bugs ( 虫, here a
reference to shipworms) and that the vessel should be treated to prevent damage.
332
The first item
stipulates that if the ship was damaged (“consumed” by the worms), then the borrower must make
reparations, unless the organisms had already been present. Even so, if it was found that the ship was
damaged through the borrower’s carelessness, then once again the borrower was responsible for
repayment.
333
This implies that the borrower was responsible for the upkeep of the ship en route, and
needed to ensure that the wood was treated properly to prevent infestation.

330
KI 3069-10.
331
Ishii, Zusetsu wasen shiwa, 322–323.
332
The word used is たでる, defined as extending the life of the ship by burning straw and brushwood underneath
the base of a ship to protect against shipworms and water seepage. Nihon kokugo daijiten.
333
KI 3068-10, 3069-9, 3070-10, 3071-24, 3072-24, 3073-23.

199

A simple treatment found worldwide for wooden ships was to smoke the hull of the ship as a
deterrent to shipworms. A graphic display of the effectiveness of this treatment can be seen in a rudder
from an eighteenth century ship recovered off the coast of Florida. The copper sheathing protected the
parts of the rudder that were regularly in the water, but the rudder post was left exposed. The sunken ship
appears to have been burned, based on the char marks present at the tip of the post. During the years the
wreck was submerged, the metal-covered areas and charred areas continued to protect the wood, while
shipworms severely damaged the exposed areas (fig. 49). Edo-period woodblock prints document that
such a charring process was known in early modern Japan (fig. 50). As another item in the maritime
regulations discusses the potential need for reparations in the case of the ship actually being damaged
through burning, it is not unlikely that medieval ship owners employed this method as well.
334
Most
versions indicate that the borrower was responsible for reparations in burning the ship while treating
against worms.
335
It further suggests that such treatment needed to be applied repeatedly. Vessels usually
followed the coastline and round-trip voyages generally did not exceed a month, so these rules implied
that the borrower may have needed to treat the wood again at some point during the journey.

334
KI 3068-27, 3069-27, 3070-27, 3071-30, 3072-41, 3073-41.
335
Version B requires the lender to pay (kashinushi, 借し主), Versions A and C specify that the borrower (karinushi,
かり主) pay, and the Seven-harbors version calls on the proxy captain (kari sendō 仮船頭) to pay. The Ise-Shima
version and Version D have no grammatical indicators and thus are unclear, but it seems likely that the
responsibility lay with the borrower. If we consider the proxy captain as an equivalent role to that of the borrower,
then it is also likely that the kashinushi of the first version may be a copy error and should actually be karinushi
instead.

200

Figure 49. Rudder from an eighteenth-century shipwreck. The lower area covered by copper sheathing is intact (left), as
is the charred tip of the post (close up, right), while the interior is eaten away by shipworms. St. Augustine Lighthouse
and Maritime Museum. Photo by author (2013).

Figure 50. “Fuji with a Rocket,” by Katsushika Hokusai. Note the men charring the base of the ship. Hokusai and Smith
(1988, pp 42-43).

Several of the items in the maritime codes give clues to the manning of ships, as well as to the
problems caused by sailors. Sailors had an individual responsibility to protect the cargo, and were
compensated for their actions with a share of the profits from the voyage. Items relevant to allocating

201

profits are discussed in greater detail below. There are several specific clauses that further address the
considerations of hiring sailors (kako or suishu, 水主) and the circumstances of administering discipline.
Three Inland Sea versions of the codes (variants B, C, and D) are the only ones to contain an
article about hiring sailors.
336
These versions may well reflect hiring practices specific to the Inland Sea
area. The first clause states that if a sailor was hired but, “if there were discrepancies [presumably from
the original contract when hiring, i.e., if the sailor did not perform his duties adequately],” then the sailor
was to be handed over to the captain. While not explicitly stated, it is likely that this would be for
discipline. The item specifies that only sailors of equal ability to the employer
337
should be entrusted to
the captain, suggesting a highly capable crew. The second clause further states that if a sailor was hired
who was of no use to the captain or caused trouble aboard ship, that he should be put ashore immediately
and provided funds to travel home by himself. This speaks to the need to maintain discipline aboard ship,
as dissent or disruption would not be tolerated, and it also suggests an inherent need for fairness. The
sailor was not marooned, but was left with sufficient funds to travel.
A second item recorded in the same three versions as the above is somewhat unclear but hints at
the shipboard hierarchy and the expendability of sailors.
Item: When in a borrowed boat, if it is damaged and ten sailors die, if the captain alone
survives, then the contracted cargo fees should be given over [to the captain]. However if
the ten sailors live and the captain dies, there should be no cargo fees [paid to the
sailors].
338

The ambiguity of this article lies in two different points. In two versions the word translated here as
“cargo” is the funadoko (船床), literally the baseboard of the ship, but also thought to refer to the vessel’s
hold. The third version reads funatoku (船徳), which is unclear, but if it is an accurate copy it may refer to
the value of the ship and/or the items on board. Kubota Hiroshi, in an analysis of later versions of

336
KI 3071-15 is a longer article, but the same information is split into two articles in KI 3072-6 and 7 and KI 3073-
5 and 6.
337
我におとらぬ水主. おとらぬ is probably taken from 劣る, “to be inferior.” The employer may have been the
ship owner or the captain himself. Presumably in this case the ship owner would have had some experience himself
with sailing, at least to be able to judge others’ capabilities. See the annotated version of KI 3071-15 in Appendix 3
for a discussion of the ambiguity regarding the subject of the item.
338
KI 3071-26, 3072-9 and 10, 3073-8 and 9.

202

maritime codes, has suggested that funadoko could also refer to fees charged for use of the cargo space.
339

This definition, if applied to these earlier versions of the maritime codes, helps clarify the second
ambiguity: who is to be repaid. Since this item notes that it was a borrowed ship, if the ship was damaged
but the captain lived, then it is possible that repayment was to be made to the ship’s owner. If that were
the case, however, then if the captain died but the sailors lived, it seems odd that the owner would not
have received any compensation for the damage. It makes more sense that the captain would have
received the agreed-upon cargo fees even if the men perished, since the captain had been hired to bring
the ship to its destination and back. If the captain died, then the sailors were not entitled to any share of
those fees, since they had not been contracted as the responsible party for the voyage itself. Especially
when taken in conjunction with the previous article, we see a hint of the hierarchy of hiring men aboard
ship. The previous item, which required that the sailors be “handed over” to the captain, suggests that the
sailors were contracted for separately from the rest of the agreements among ship owner, captain,
borrower, and cargo owner. As such, it is possible that they were viewed as more expendable and without
the responsibility that lay on the shoulders of the captain. They were therefore entitled to less or even no
profits if disaster struck.
Theft was apparently a common problem aboard ship, since all versions of the codes detail
responsibilities in case of thievery and again demonstrate the hierarchy on board:
Item: When the boat has cargo loaded and a sailor steals something and escapes, the
captain must pay. However, if the sailor is caught and handed over to the cargo owner,
even if the stolen cargo is gone [discarded], the captain is not responsible for
repayment.
340

The captain again seems to have been to be the final arbiter aboard ship, as he bore responsibility for the
actions of his crew. If the sailor in question absconded with the cargo, the captain was responsible for
reparations. If the sailor was found, however, the captain was absolved of responsibility, and decisions
regarding repayment were turned over to the cargo owner. Here we also see the autonomy on board ship.

339
Kubota, Kaisen taihoko, 114–115.
340
KI 3068-28, 3069-29, 3070-28, 3071-37, 3072-38, 3073-38.

203

Instead of turning the thief over to local authorities for discipline, the culprit was handed over to the cargo
owner instead, and the punishment is left unspecified.

The Ship at Port

Several items suggest a strong connection between the ships and the places to which they sailed. They
emphasize the responsibility of the captain to protect his ship while at port, not just while under sail. They
further suggest the connection between the ships and the people in the port communities, requiring the
latter to aid the ships and their crew in a crisis. There were some geographical variants concerning the
captain’s responsibilities, but the need for local support was largely universal.
Although someone else may have contracted to borrow the ship, after embarking on a voyage the
captain needed to take responsibility for his own actions even when in ports en route. All six variants
noted that if the captain became involved in a court case (kuji 公事) and his ship was taken from him,
then the captain was to make reparations for the ship.
341
The Inland Sea-centered versions (Versions B, C,
and D) noted the captain’s responsibility for the ship when moored at a port in a storm. If he absented
himself from the ship at various ports and harbors and went onto land, and if the vessel sustained damage
when winds and waves hit, then according to Version B, the captain was responsible for the losses.
342

Versions C and D stipulated that the borrower was responsible.
343
This discrepancy may result from the
fact that in some cases the borrower and the captain were the same individual. All three versions note,
however, that if the captain stayed with the vessel and tried to protect the ship, and if the boat still
sustained damage beyond repair, then the cargo and value of the boat were to be distributed as payment
(haitō 配当; to be discussed in greater detail below) to the captain and sailors, absolving the captain of the
need to make reparations. Here the expectations for the captain are made clear, as he is to ensure the
safety of his vessel whether at sea or in port.

341
KI 3068-23, 3069-24, 3070-23, 3071-22, 3072-21, 3073-20.
342
KI 3071-12.
343
KI 3072-26, 3073-25.

204

When the ship was moored, the entire community was responsible for protecting the vessel. All
six variants of the codes call on villagers to help secure the ships when multiple vessels were anchored in
a storm, suggesting that this was a particular danger in busy ports. This demonstrates the idea that it was
indeed a maritime community, in which all were expected to act for the greater preservation of the vessels
that connected far-flung ports. The items begin, “When there are many ships moored, in times of high
winds the village
344
should provide aid.” It then went on to describe the need to prioritize tying the ships
upwind so that they did not snap their moorings and smash into any vessels downwind.
345
In contrast to
some of the other items that discuss the need for reimbursement in case of damage to vessels, there are no
such specifications here. The implication is that in ports with multiple ships suggesting lively trade, it was
the community’s responsibility as a whole to protect the ships that were so important to the livelihood of
the community. If the vessels were loosed from their moorings and sank or damaged other ships, it would
be to the detriment of the entire village, which relied on those watercraft to provide trade goods and
supplies that could not be otherwise obtained.
Version B of the maritime codes contained three additional articles stipulating reliance on local
aid in times of trouble. If the flagship and tools were lost at sea, the remaining sailors had to make do as
best they could, likely boarding accompanying vessels to arrive at port. There, survivors were to appeal to
local village officials
346
and receive guidance on how to proceed.
347
It is implied that some evidence
needed to be presented to officials, presumably documentation indicating that the flagship had been
wrecked at sea. Furthermore, the following item noted that if the flagship was not in need of repairs, then
the owners of lodgings in the area were to provide aid.
348
This wording is unclear, and as the item only
appears in one version, there is no alternate wording for comparative purposes. It is possible that if the
vessel did need repairs, the sailors were expected to sleep on board ship, perhaps to act as guards. If the
flagship was intact, then local lodgings were to be made available. Finally, a third article reads that if a

344
This is noted variously as village (mura 村, four versions), cove (ura 浦), or place (tokoro 所).
345
KI 3068-3, 3069-3, 3070-3, 3071-4,3072-4, 3073-3.
346
村役人江及訴
347
KI 3071-17.
348
KI 3071-18.

205

ship must be rowed in to port backwards, then the people there should without question lend aid.
349

“Backwards” rowing (gyakusen 逆船) refers to a way of securing the ship at port in rough seas, involving
turning the ship to have the bow face upwind and reversing it into the mooring. The stern and rudder were
notoriously weak parts of Japanese ships, and tying the ship down in a manner that would protect that
vulnerable area from direct storm damage would have been an important maneuver. In the days before
engines and propellers, however, this would have been a much more difficult undertaking, particularly for
larger vessels. People on land would have been enlisted to help, probably using tow lines and poles. All
three of these articles imply expectations of ongoing interactions between the people manning the ship
and the local population. It is one manifestation of the mariner’s perspective, demonstrating how the
community as a whole functions to maintain smooth operations aboard ship and in port.
The Ise-Shima version is even more specific about port infrastructure and maneuvering, a
stipulation not seen in the Inland Sea or eastern sea versions of the codes. In the Inland Sea, medieval
maritime communities with active ports were usually located either directly on the coastline of one of the
main islands, or on one of the many smaller islands scattered in the sea. Using the 107 locations noted in
the Register as a representative example, only about fifteen ports were located even slightly upriver and
inland. Of these, most were in Settsu and Harima, with one each in Bizen (Hachihama), Bitchū (Nishiate),
and Awa (Sojiin).
350
This suggests that shallow water was less problematic in these regions, and did not
require stringent rules. On the other hand, as the Ise-Shima version alludes to shallower waters and river
travel, it is likely that shoals or difficult navigation was more of a challenge in that area. For instance, one
item discusses port infrastructure, as follows:
…Harbors and such at the ocean’s or inlet’s mouth, do not need to erect a depth gauge.
For river ports, if there is a shallow beach [within the inlet], to accommodate the
changing passable water routes, the various ships [may need] a set of poles
351
[to
indicate] the passable route.
352

349
KI 3071-19.
350
The Settsu and Harima ports were Sakai, Kuise, Amagasaki, Ihozumi, Bessho, Matsubara, Shikamazu, Aga,
Imazaike, Aboshi, Naba, and Nakasho.
351
The “set of poles” here is kidachi ( 木立), defined as a line of trees specifically planted in a row or wooden posts.
352
KI 3070-32.

206

The Ise-Shima version also contains an item regarding provisioning not found elsewhere. It notes
that steering poles ( みさを, misao, or probably 水棹) should be available on ships from the entryway to
the bilge (the lowest compartment on the ship) to the stern of the ship, and also hung from the gunwale
(the topmost rail) to the stempost.
353
In fact, the Egara Tenjin engi from 1319 recounts the tale of
Michizane’s voyage to Kyushu (fig. 51), and shows a pole extending from behind (aft) of the mast out
beyond the stern of the ship. Of course, the scene portrayed is nowhere near the Ise-Shima area, but the
Egara Tenjin Shrine, repository for the scroll, is located in Kamakura, relatively much closer to Ise-Shima
than Kyushu. The ship depicted is clearly more technologically advanced than typical Heian vessels, and
so it is possible that the artist simply modeled his subject on contemporary watercraft plying the local
waters.

353
KI 3070-35.

207

Figure 51. Egara Tenjin engi. Hiroshima Kenritsu Rekishi Hakubutsukan (1996, plate 147).

The concern for navigation is evident in these two items. While exposed ocean ports were
presumed to be deep enough for ships to navigate, ports along the mouths of rivers would change and
grow shallow over time due to the gradual accumulation of silt. Steering poles would be necessary to push
the vessel off from sandbars or to help navigate in waters too shallow to maneuver by sail or oars. The
fact that these two provisions are not specified in the other versions shows clearly how the geography of
local seas and the overall environment directly affected the rules that mariners created to assign
responsibility for and to ensure safe navigation.

208

Cargo Concerns

Many of the items in the maritime codes referred to cargo concerns, from accurately recording the cargo
on the manifests, to following standard procedures when jettisoning cargo in an emergency, to splitting
the profits after the sale of cargo in various situations. A key term was haitō ( 配当), referring to the
process of allocating or distributing the worth of the cargo, sometimes even including the value of the
ship itself. While a successful voyage required the cooperation of the ship owner, borrower, and captain,
the cargo owner often held the most influence and had the most to lose. While the percentages of
distribution and the specific recipients were not always clearly detailed in the maritime codes, multiple
references to splitting the allocation imply agreements among the ship and cargo owners, captains, and
sometimes the crew. Profits could have been increased or diminished according to the response to various
disasters encountered at sea. These items demonstrate the priority placed on the economic aspects of a
ship’s voyage. They are nearly all universal, suggesting that cargo concerns and their related profit
concerns were common problems at sea.
Accurate bookkeeping is always an important part of commerce, whether maritime or land-based.
Two items in all six versions of the codes reflected the need for an accurate ship’s manifest to account for
all of the cargo on board, and suggested the existence of a system of checks and balances. The first item
indicates the importance of the entire crew working in unison: “When giving a cargo manifest to the
captain, the others on board [ 乗衆; the group of people aboard ship] should also affix their seal to it.”
354

This indicates a mutual acknowledgement and recognition of the contents of the hold by both captain and
crew (and possibly even by any passengers), and it implies that all were in agreement as to the value of
the items. The article also stated that items on board that did not have seals attached – i.e., those not
officially recognized as part of the cargo – could not be included in any profit allocations.
Five of the six versions include a second item about cargo manifests; the Seven-harbor version
incorporates the two items into one. The second item reiterates the need for all items on board ship to be

354
KI 3068-21, 3069-22, 3070-21, 3071-34, 3072-18, 3073-18.

209

reported in the manifest, presumably to aid in reparations if cargo was lost en route. The item further
emphasizes this need by stating that if the cargo manifest was not brought to the captain, even if silver
and gold were jettisoned (in the event of storm or other emergency), then that jettisoned cargo should not
be entered into the calculations when distributing profits or remuneration.
355
This item, viewed in
conjunction with the previous one, indicates the need for the captain and crew to be honest in their
recording of cargo loads. The first of these two articles further stated that, if there were items that were
not included on the official manifest but were discovered on inspection of the ship’s interior, then those
items should be included in the profit allocations. While it is not clear who was carrying out these
inspections, if it was by the cargo owner, at a port checkpoint, or by someone else, the ship clearly needed
to pass through certain areas that required outside scrutiny. Combined with the second item that prevented
unrecorded goods from being included in the allocation, the mutual responsibility and demand for honesty
from the captain and crew is strongly accented. If one person secreted personal trade items on board ship
and they were discovered, then his additional profits would have been distributed equally among the
entire crew, and if undiscovered but later reported lost he was not entitled to any compensation. This
implies that trade of personal items, separate from the contracted cargo transport, was not uncommon on
these voyages. A third article that focuses on the process of repayment to the lender mentions too that the
captain or ship owner may have “internal dealings” or secret items ( 内談, naidan) that could be
incorporated into the repayment.
356
This is another indicator that personal trade was common practice on
these voyages, since even the power players of the captain and ship owner could be taking items on board
for their own profit. These items would of course go unrecorded on official manifests, and would not be
accounted for in the official documents levying cargo fees or taxes at checkpoints, so we must rely on
these kinds of hints to indicate that personal trade in all likelihood existed.
Sea voyages could be difficult, even in the relatively protected Inland Sea area, and cargo could
easily be damaged, jettisoned in the case of storm or fire, or otherwise lost. Many of the articles discussed

355
KI 3068-21, 3069-21, 3070-20, 3071-43, 3072-17, 3073-17.
356
KI 3068-29, 3069-30, 3070-29, 3071-29, 3072-39 , 3073-39.

210

reasons for cargo losses and compensation. Certain losses were expected as a matter of course in sea
voyages, while others were considered preventable and required compensation. Rats were omnipresent;
all six versions of the codes included an item declaring that goods eaten by rats should still be calculated
into the profit allocations, regardless of the size of the items.
357
This implies that rats were a known
danger to the cargo, and one that could not be completely defended against. Damages were expected, and
the crew or power players were not penalized by withholding profits in this case, suggesting that they
were not to be held responsible for rodent infestations.
On the other hand, water damage or cargo losses due to fire at sea could be blamed on negligence.
A different article that appears in all six versions stipulated that if the cargo got wet, unless that was due
to heavy winds and waves at sea, then the captain was to make reparations. If the ship was still in port and
rain seeped through to wet the cargo, then the captain was responsible.
358
While rats were inevitable and
damage was expected, water seepage and subsequent damage was preventable, except possibly in the case
of unpredictable, rough seas. One document from a Yuge shipper of rents and dues to Tōji pled his case to
the temple, saying that nearly a hundred ships in Harima Bay had encountered a strong typhoon, and he
was able to escape with only twelve bales of salt having water damage. He begged Tōji to reward his
honesty about the losses by exempting him from the need to repay their value.
359
Another article in all six
versions noted that if cargo was loaded and a fire broke out, reparations were to be made in the same way
as if the cargo had been damaged by heavy winds and rains at sea. The cause of the fire, however, was to
be considered human error.
360
All collections but Version B specify that this should be the case whether
the ship was at sail or at port.
361
But at port, where the captain should have been able to secure the ship
and the cargo safely, damage to the cargo was solely his responsibility. This idea corresponds with the

357
KI 3068-15, 3069-15, 3070-15, 3071-38, 3072- 14, 3073-13.
358
KI 3068-14, 3069-14, 3070-14, 3071-35, 3072- 13, 3073-12.
359
Yamauchi, Yugeshima-no-shō no rekishi, 85. Yamauchi does not specify the date of the incident in his citation of
a document from Tōji hyakugō monjo, “shi” volume #499. Printed versions of the same volume do not include this
specific document, but include others from between the 13
th
and 15
th
centuries, likely placing this incident in the
same time frame.
360
KI 3068-26, 3069-28, 3070-26, 3071-36, 3072- 37, 3073-37.
361
The exception is KI 3071-36.

211

earlier item that placed the responsibility on the captain for the entire ship’s safety at port, and chastised
him for absenting himself in cases of storms blowing up while the ship was moored. The captain’s duties
did not end when the ship was at anchor.
The most extreme example of responsibility for cargo losses is found in the Inland Sea-centered
versions (versions B, C, and D). They indicate that the final responsibility was on the shoulders of the
shipbuilder and presumably the ship owner.
Item: When sailing in open waters, if the ship is weak and the oakum
362
is wet, and waves
get in and tools and excessive cargo are lost, then the boat itself must be factored into the
allocations.
363

This refers to a fundamental flaw in the construction of the ship itself. Were the vessel to be watertight,
there would be no seepage and the cargo would not be damaged. The punishment for launching an
unseaworthy vessel, then, required the ship owner to incorporate the value of the ship itself into the profit
calculations and distribute that value among crew and investors as compensation for the damaged cargo.
The importance of the seaworthiness of the vessel cannot be overstated here.
The practice of incorporating the value of the ship itself into allocations and the mutual
responsibility of the captain and crew to get the ship and cargo safely to shore are noted in other articles
as well. Cargo might be thrown overboard in the case of a storm, if the ship was taking on water and the
load needed to be lightened to prevent sinking. That would also help to stabilize the ship, since in high
seas the vessel would pitch back and forth. Cargo might shift in the hold, causing the ship to list and
possibly capsize. Jettisoning cargo, then, would have been an accepted emergency measure in the case of
storms, as recorded in the case of one boat shipping rice as nengu from a Tōdaiji estate in Chikuzen in
1341. Upon encountering strong winds near Sunoe, the rice was thrown overboard. The crew of two
nearby ships from Buzen served as eyewitnesses that verified the testimony, though it is unclear who bore
the economic responsibility for the losses.
364

362
The word translated here as oakum is makihada, referring to fibers from the bark of the maki tree 槇, a type of
pine. It is often used to line fastener holes or the seam between planks to prevent seepage.
363
KI 3071-44, 3072-42, 3073-42.
364
Shinjō, Chūsei suiunshi no kenkyū, 777.

212

Several articles in the codes suggested that overzealous, panicked sailors may have had the
tendency to throw more items overboard than was strictly necessary. All collections but Version C noted
that when cargo had been jettisoned excessively, then the value of the ship was to be included in the
allocations.
365
They specify “the ship” ( 其船) that jettisons cargo excessively, not a “borrowed ship” ( 借
船), as noted in several other articles. This suggests that in the case of borrowed ships, the ship owner was
expected to be present on board, and he likely was making decisions about the amount of cargo to throw
overboard. He thus bore the responsibility for excessive jettisoning, and therefore would have needed to
sell his ship and divide the value among the rest of the crew to recoup the losses of the cargo. The crew
also appears to have borne some responsibility, as the item also stated that if the ship was rescued, then
the sailors’ personal effects were also to be included in the allocations in the case of excessive jettisoning,
but not if jettisoning was minimal. This again speaks to the collective nature of shipboard life; the men
shared responsibility for their actions in an emergency and overreaction was a collective misdeed.
One article in all six versions is specific about the type of cargo and method of jettisoning it, but
there are variations among versions. The basic item reads as follows:
Item: When unhusked rice (momigome 籾米) is loaded, or if Chinese goods (karamono
唐物)
366
are included in the cargo, if the owner of the Chinese goods throws away his
Chinese items, the [value of the] unhusked rice should not be part of the allocations
[distributed to the person who threw away the Chinese goods]. If the rice owner or the
sailors should throw away Chinese goods hastily, regardless of whether there are Chinese
goods packed inside [the rice], even if they were unaware of [the valuables in the rice], it
must be reported.
367

This item further illuminates the relationship between the crew and cargo owners. Based on this
description, it is likely that the Chinese or foreign goods aboard ship were fragile, not very large, and
valuable. Porcelains and ceramics would have been common cargo, and the rice would have made a
natural packing material to protect those items on a ship that pitched and rolled en route. If the crew were
not clear on which containers stored rice and Chinese goods, however, in the throes of a fierce storm it is

365
KI 3068-16, 3069-16, 3070-16, 3071-39, 3073-14. The exception is KI 3072.
366
This can mean either specifically Chinese goods or foreign items in general.
367
KI 3071-32.

213

easy to imagine a panicked crewmember throwing the bulky, heavy rice containers overboard, without
thinking about the more valuable items packed inside.
Version B, noting specifically the owner of the rice and the owner of the Chinese goods (a
distinction not made in all versions), suggests that multiple cargo owners could have their goods on board
the same ship. Many entries in the Register show ships bearing different types of items; it is possible that
in those cases multiple owners commissioned the ship to take their goods to Hyōgo. It is therefore also
possible that cargo owners from different places or estates traveled aboard the same ship. For example,
the Tsurajima ship arriving in Hyōgo on the sixth day of the sixth month (1445) brought salt from both
Kojima and Shima. Perhaps in that sort of situation the various salt types were commended to the
captain’s care by different individuals.
The above article also implies that the cargo owners worked together, as it would be highly
unlikely to pack the fragile items in the rice without obtaining the rice owner’s permission. If the owner
of the Chinese goods threw his own items overboard, then he was responsible for the loss and would not
receive compensation in the form of a share of the rice allocation. Versions A and B as well as the Ise-
Shima version included the captain as well as the sailors and rice owners as potential culprits in the
overzealous cargo jettisoning.
368
If they were unaware of the presence of the Chinese goods, it appears
that they would not be penalized in the allocation of the profits, although the inadvertent discarding was
to be reported. Although the recipient of the report is not specified, it is likely to have been the ship owner
who would have been the final arbiter of the allocations, or possibly it would have been the local
authorities who at times would have had a hand in determining profit margins. In the case of a ship
bearing rents and dues, the report would have been submitted to estate authorities. The co-dependent
relationship among cargo owners, captain, and ship-owner is again demonstrated here, and the item shows
the strong ties among the principals.

368
KI 3069-26, KI 3070-25, KI 3073-36.

214

Notably, the Seven-harbor version of this same item is quite different from the rest.
369
Instead of
specifying unhusked rice, the packing material was simply listed as “rice.” Chinese goods are mentioned
only at the very end of the item, with most of the discussion focusing on “heavy” and “light” objects ( 重
物 and 軽物).
370
It is unclear what the “light” objects may have been, although they were likely also
packed in the rice. In this version, if the captain threw heavy items and rice overboard, then they were
also to be included in the allocation. The cargo owners were therefore not penalized for the captain’s
decision to lighten the load. But if lighter items, including Chinese goods, were jettisoned instead of the
heavier goods, then that would need to be reported. It is unclear why this item is written so differently
from the other variants. This is the only version with signatures from the Japan Sea coastal provinces,
including areas as far north as Noto and Dewa and as far south as Hakata, long known as the gateway to
Asia, and being on the Japan Sea coast would presumably have increased direct trade with the mainland,
leading to greater amounts of Chinese goods on board. Perhaps it was the very rarity of Chinese goods
that led to them being considered of greater value and therefore deserving of special mention in the Inland
Sea versions of the maritime codes, again demonstrating a geographical bias.
Version B, which was signed by the original three authors from Hyōgo, Tosa, and Satsuma, had
an additional item regarding actual reporting of cargo jettisoning to local authorities, therefore reinforcing
the concept of administrative ties between the ship’s captain and the local population of the port. The
article required that the captain apply to local officials (yakunin 役人) for reimbursement of both travel
fees and cargo when cargo was jettisoned, and that a written receipt should be issued.
371
It is possible that
local officials might actually have reimbursed the captain and crew for any losses or travel fees, but they
may simply have certified that the ship had made landfall without cargo. Other articles suggest that, while
official reports may not have been filed, local pricing and customs influenced profit allocations. Although

369
KI 2068-25.
370
The “light” objects may simply be alternate characters for Chinese goods, as both could be pronounced karamono,
but the additional specification of heavy objects, not present in the other versions, suggests that the term does indeed
reflect the size or weight of the object.
371
KI 3071-20.

215

there are slight variants in wording, all six versions of the maritime codes state that the value of jettisoned
cargo should be calculated according to the buying price at the intended destination, which supports the
theory that local authorities provided documentation as to the local pricing.
372
Similarly, Versions B, C,
and D state that if the ship planned to travel a certain distance but only managed to make part of the
journey, then the cargo prices should be determined at the place where the ship actually landed.
Allocations, however, were to be distributed according to the original contracts.
373
While there are no
documents that detail such an event actually occurring, in 1470 a ship carrying eight hundred koku of rice
was blown off course and the majority of the rice was distributed among the locals. Whether it was
actually sold or simply appropriated is unclear.
374
Regardless, the incident demonstrates the disbursement
of cargo after landfall.
If the voyage was unsuccessful and the intended destination not reached but the ship nevertheless
returned to its home port, other articles in all versions specify that if cargo had been jettisoned the value
of the cargo was to be calculated at the rate “at the current venue” (zaisho 在所), presumably that at the
place to which the ship returned.
375
Versions A and B, the Ise-Shima version, and the Seven-harbor
version clarify this further in a separate article, stating that if the ship did not reach its destination but also
was unable to retrace its path to its home port, then if allocations were made, they should be calculated
according to a price at whatever destination was able to be reached.
376
This implies the direst scenario,
when a ship would have been so badly damaged that it could not make the return voyage. Presumably the
crew would be paid off from the allocation at that point and left to find their own way home. In each of
these scenarios, though, the captain or cargo owners would have needed to communicate with the local
population to determine pricing, which would vary in part due to the distance traveled and consideration

372
KI 3068-18, 3069-18, 3070-18, 3071-41, 3072-16, 3073-16.
373
KI 3071-28, 3072-30, 3073-29.
374
Quoted in Shinjō, Chūsei suiunshi no kenkyū, 771.
375
KI 3068-19, 3069-19, 3070-19, 3071-40, 3072-15, 3073-15.
376
KI 3068-20, 3069-20, 3070-19, 3071-42.

216

of shipping costs for the voyage as a whole.
377
This all suggests that even if an area did not have an
established local market, residents were aware of general pricing practices and values.
Only one item, found in all six versions, discusses cargo damage while the ship was at port.
378
If a
moored ship incurred losses (presumably from storms occurring while the ship was anchored), the wet
items were to be dried in the local village and returned to the captain. In order to prevent disputes, and
possibly also to ensure that fees could be paid before any damages were incurred at port, anchorage
taxes
379
(hobetsu 帆別) were due upon renting the port slip.
380
In this case there should have been no
complaints from the provincial constable or governor (kokushu 国主 or, in two versions, shugo 守護).
The relationship here between the ship and the port is telling. Similar to the earlier item that required local
cooperation in the case of a storm at port, the locals were also expected to help with the aftermath. But the
ship also needed to fulfill its responsibilities – paying for the privilege of anchoring and therefore
securing the locals’ aid. Here again we see the nature of the maritime community at work, with both sides
having responsibilities to fulfill.
The extensive nature of the provisions regarding cargo and allocations highlights the fundamental
reason behind kaisen voyages, i.e., trade and profit. All parties – captain, ship owner, cargo owners, and
crew – needed to profit from the journey, but the passage could be rough and dangerous, and cargo was
obviously lost or damaged on a regular basis. These detailed provisions reveal the extent to which
different players shared responsibility on board ship, and their need to consult with the local population in
cases of loss. Collaboration among all parties was a critical concept in these sea voyages.

377
Sakurai Eiji and Nakanishi Satoru, Ryūtsū keizaishi (Tokyo: Yamakawa Shuppansha, 2002), 210.
378
KI 3068-2, 3069-2, 3070-2, 3071-3, 3072-3, 3073-2.
379
Aida, Chūsei no sekisho, 73. While Aida equates this tax with ships passing through a checkpoint and suggests
that such examples are limited in the written record, the wording of this law indicates that anchorage taxes could be
levied at other ports as well, not necessarily only at official checkpoints.
380
One version of the regulations uses the more generic “court costs” 公事 and is not specific to port fees.

217

Other Perils at Sea

Though storms and piracy are addressed in several articles, other items discuss additional difficulties at
sea. Trouble on the water did not always result in the loss of cargo or life, and those situations needed to
be addressed as well. These items, too, are largely universal regardless of geographic location. In each
version, either the first or second article discusses procedures to follow when a ship is blown off course.
The basic item reads:
Item: Wayward ships and ships blown off course. They [items from the ship] should [be
used to] repair the local Shinto shrine or Buddhist temple. If there are sailors
381
on the
ship, [the situation] should be dealt with according to [the wishes of] the ship owner.
382

Since there is no mention of losses or jettisoning cargo in this item, even though the ship had been blown
off course, it likely arrived safely at shore, although not at the originally intended destination. Presumably
to give thanks for a safe landing, an offering to the nearest religious institution was required in the form of
repairs to the shrine or temple. All versions of the article place decision-making responsibility, regardless
of whether it is about the sailors or cargo, on the shoulders of the ship owner. It is unclear what “the
situation” (italicized above) was: disputes over cargo pricing, profit allocation, payment for an unfinished
voyage, or something else. In any event, the ship owner’s authority is made clear even if the ship did not
reach its final destination.
If the ship was at sea and sustained damage but people on board were rescued, there appears to
have been an accepted stance of “every man for himself.” All six versions state that if someone was saved
but found to be holding on to even gold and silver from the ship, then that person should not be blamed.
383

The article does not specify that the valuables should have originally belonged to the rescued person, so it
is possible that in the case of emergency, salvage was allowed and was not required to be split among

381
The term used for sailors, italicized here, varies from version to version. In the Ise-Shima and Seven-harbor
versions it clearly refers to human passengers or workers, either using the characters norite ( 乗手, KI 3070-1) or
suishu ( 水主, KI 3068-1). In Versions A, B, and D it is ambiguous, as it uses the less descriptive notarumono ( 乗た
るもの, KI 3073-1 ) or norimono ( 乗物, 乗者, KI 3071-2 and 3069-1, respectively), possibly denoting sailors or
cargo, although the second reading of norimono is more likely to refer to people than things. Version C uses the
term sekinibutsu ( 積荷物, KI 3072-2), or laden cargo specifically.
382
KI 3068-1, 3069-1, 3070-1, 3071-2, 3072-2, 3073-1.
383
KI 3068-24, 3069-25, 3070-24, 3071-31, 3072-35, 3073-35.

218

survivors or given to rescuers in gratitude. This seems to be a fairly unusual attitude, as so many of the
other articles specify in detail the types of allocations for losses. Perhaps in the case of such severe
wrecking, when rescue was required in order to escape with one’s life, then anything that could be
salvaged was seen as simply taking good advantage of the spoils of disaster.

Navigation Problems and Consequences

Several of the articles in the Inland Sea-based versions of the codes (versions B, C, and D) address very
practical matters of maneuvering and navigation at sea. These “rules of the road” determined right of way
and responsibility in the case of accidents. Through these regulations, we see the hierarchy of different
types of vessels at sea. These types of item also yield the only references to riverine travel in the codes,
which go unremarked upon in the Ise-Shima and Seven-harbors versions (as opposed to riverside ports
and shallow waters, which are noted in the Ise-Shima variant). This suggests that travel on rivers may
have been more important in the Inland Sea region, particularly as any cargo destined for the capital at
Kyoto would have been shipped at least partway by river. This situation highlights the influence of local
geography in the creation of the codes.
To begin, the first item found in the Inland Sea-centered versions (Versions B, C, and D)
demonstrates the consequences of differing propulsion technologies. Sailed ships would likely have been
faster and more responsive to shifting direction with the wind more quickly than oared vessels. As such,
they would have borne the brunt of the responsibility for getting out of the way in an impending collision.
Given this, the item notes that sailing ships were to make way for other vessels, and if a rowed ship was
overtaken and damaged by a sailing vessel, then the latter was to make reparations. The sailing ship was
responsible for refitting the damaged rowed boat, and someone on the ship also had to proffer an apology.
If an oared vessel hit a sailed ship, however, “there should be no debate;
384
” i.e., the sailed ship was

384
“ 異義有間敷事.”

219

responsible for outmaneuvering the slower oared vessel and therefore the operators of the oared vessel did
not need to make reparations.
385

As for river travel, addressed in Versions B, C, and D and thus perhaps indicating more travel
along inland waterways in the Inland Sea region, sailors on a boat coming downriver were required to
make way for upriver-bound boats.
386
Two versions add that if the upriver boat caused damage to the
downriver-bound boat, the onus to move away was on the operators of the downward-bound boat.
387

Similar to the sailing ship in the above item, the downriver boat was harnessing the power of the current
and would likely have been able to maneuver more quickly to avoid the boat struggling against the water
flow. If the boat coming upriver was struck and damaged by the downriver boat, the operators of the latter
vessel were responsible for repairing the damaged boat and apologizing.
These two items clearly show a sense of responsibility manifest in the maritime world. Damages
incurred by storm and fire were acts of nature and, save for cases when the captain was found negligent,
the damages were to be split amongst the principals, including the crew. In contrast, human error that
affected another ship made the operators of the first ship liable. The emphasis here was on competent
seamanship and respect for others on the sea. Those who could maneuver more easily, such as sailors on
the sailing ship or the downstream-bound boat, were required to be more considerate of others on the
water. If they failed to do so, or worse, if they were not skilled enough to outmaneuver the less agile
vessel even with their advantageous position, then they were required to pay damages, ensure that the ship
was repaired, and offer an apology. There was thus a strong sense of respect due to another member of the
maritime community.
Another item, occurring only in Version B, also demonstrates the need to be considerate of one’s
fellow mariners. It is unclear why the other Inland Sea-centric versions do not include this item. If a ship
bearing cargo was in the way of an unladen ship at port, the ship with an empty cargo hold was to retreat

385
KI 3071-8 and 9, 3072-29, 3073-28.
386
KI 3071-6, 3072-8, 3073-7.
387
KI 3072-28, 3073-27.

220

to the mouth of the waterway.
388
The laden ship would have been heavier and therefore slower to
maneuver, while the empty, lighter ship was freer to make room for the laden ship. Though riverboats
were not specified in this instance, this article precedes the aforementioned riverboat traffic law, and may
have been an additional admonition for riverboats in narrow waterways.
The need to respect seamanship is reflected in an item present in all six versions. All but the
Seven-harbor version agreed that if a ship was at port and the captain misjudged when to put to sea, with
the result that no progress was made, then the captain was “defiled.”
389
The Seven-harbor version simply
said that the captain “must not return.”
390
Versions A, C, and D, as well as the Ise-Shima version,
specified that defilement occurs when the captain sets sail against the will of the crew.
391
This is
particularly notable, since in matters of cargo jettisoning, overall navigation, and protection of the ship,
the captain had final responsibility in most cases. Yet, when a vessel put to sea from port, if the captain
launched the ship against the advice of the crew and then made no headway (presumably due to rough
seas or lack of a headwind) then the captain’s seamanship was publicly compromised and he was
considered defiled. The article does not specify what purification measures were to be taken to combat the
taboo, though perhaps there is a hint in the Seven-harbor version’s warning that the captain “must not
return.” Since the ship was at port, it is possible that the captain might have had to put to land and a new
captain then was hired or chosen from qualified crewmembers.

Encountering Other Ships at Sea

One of the most interesting geographical discrepancies can be found in the Ise-Shima version, and it
suggests that the hierarchical nature of society was more of a concern in this region. This version contains
three articles that were not found in any of the other versions of the codes, all dealing with interactions
with other ships and with noblemen. The first unique article noted that if requesting an item from another

388
KI 3071-5.
389
KI 3069-13, 3070-13, 3071-10, 3072-34, 3073-34. Though the characters used vary, all versions use some form
of “kegataru,” meaning to be under defilement or a taboo.
390
KI 3068-13, 船頭不可返之事.
391
KI 3069-13, 3070-13, 3072-34, 3073-34.

221

ship, the operators of that ship were to hail the other ship three times and then explain the details. The
rationale behind this ritualized greeting is unclear, save for the simple fact that a single greeting might be
lost to the noise of wind and water.
392
The next two articles were concerned with encountering envoys
from other vessels, and imply that the envoy may in certain cases board one’s own ship:
Item: When a ship is summoned by a noble’s ship, place your hand on the base of the
akamaguchi by the port side of the rudder, and attend to the message of the envoy.
Item: When the envoy is aboard ship, he should not be in the stern area; those with [him]
in the akamaguchi should [be given] unattractive mats; be aware of this.
393

The meaning of akamaguchi is unclear. “Akama” may be referring to 淦間, which is defined as an area
fore of the midship (closer to the bow) where the gunwale is shallow and bilgewater is prone to
accumulate.
394
As the first article stipulates that the location is actually near the rudder, however, it is
possible that akama refers to the upraised part of the stern in smaller boats (see the vessel depicted in the
foreground of figure 7, Chapter 2, Kōbō Daishi yūgyō emaki zanketsu, for an example) which could have
a similar shallow-draft construction. The second article, though, stipulated that the envoy should not be in
the stern area, but that his companions in the akamaguchi were to be provided with unattractive woven
mats upon which to sit. It is possible that the deckhouse, generally situated near the stern or the vessel,
may have been called the akama(guchi). This would situate the respondent to the summons in the stern of
the ship, ready to respond to the noble’s request with his hand at the base that supported the ship’s
superstructure, while the envoy and his entourage were within the deckhouse itself. The stipulation for
“unattractive mats” may have been out of concern for bilgewater overflow ruining a fancier item.
Regardless of the exact location of the akamaguchi, the content of these articles is still important.
Since the stipulations of meeting envoys or nobles are noted only in the Ise-Shima version, it seems that
this may have been a more likely occurrence in that region. The protocol to be observed is also worth
noting for both ceremonial and tactical reasons, as stationing a responder in the upraised stern of the ship
would likely have kept him visible to the noble’s ship, but the location near the rudder would also allow

392
KI 3070-36.
393
KI 3070-33 and 34.
394
Nihon kokugo daijiten.

222

for quick changes in navigation if a problem occurred. Keeping the envoy out of the stern area implies a
more honorable reception to be held elsewhere on the ship but similarly reduces the possibility of hostile
hands gaining control of the rudder. These rules, then, were likely ceremonial as well as practical
measures for receiving visitors on board ship. The specific nature of maritime-based shipping leading to
the development of these types of rules is another manifestation of a particular mariner’s perspective.

Collaborative Voyages and Captains’ Networks: The Hyōgo Register Meets the Maritime Codes

References to flagships and accompanying vessels provide important clues to understanding the ways in
which ships and captains worked together. As noted above, one item in Version B of the maritime codes
stipulated actions to take if the flagship (honsen/motobune 本船 or 元船) was lost at sea. A different item,
present in all six versions, discusses measures when a flagship and/or accompanying vessels (edabune 枝
舟) jettisoned cargo. An accompanying vessel would have been between one-third and one-fifth the size
of the flagship.
395
If the flagship had to jettison items but the cargoes for the accompanying vessels were
undamaged, the allocations were to be split only among the accompanying vessels.
396
In Version D and
the Seven-harbor version, the roles were reversed; if an accompanying vessel suffered losses, the
flagship’s cargo was exempt from the allocations.
397
The implication is that although the ships were
traveling together, they were not necessarily responsible financially for each other’s losses. Although the
earlier article exhorted accompanying vessels to lend aid in an emergency, fiscal responsibility was
viewed more separately.
The relationship between flagships and accompanying vessels is further emphasized through
analysis of the Register. By coordinating the names of the ship captains involved in bringing cargo to
Hyōgo, it becomes clear that the practice of traveling in groups was relatively common. The concept of

395
Kobayashi, “‘Hyōgo Kitaseki Irifune Nocho’ ni miru edabune,” 795.
396
KI 3069-5, 3070-5, 3071-11, 3072-22.
397
KI 3068-5 and KI 3073-21. Though the latter simply says if “a ship’s cargo is jettisoned,” not specifying an
edabune accompanying vessel, it continues by saying that the flagship’s undamaged cargo should not be entered into
the allocation, implying that the unspecified “ship” is indeed an edabune.

223

“safety in numbers” was certainly in play here, as there was always a threat of piracy, and as the articles
in the codes noted, if one ship sank others could lend aid. Shinjō also suggests that medieval ships may
have traveled together to serve as trustworthy witnesses to each other in the case of a dispute.
398
In the
incident from 1341 described above, when a Chikuzen boat needed to jettison nengu rice, the incident was
corroborated by the two ships from Buzen and verified as true testimony – valuable evidence when
arguing for the repayment of losses. The various vessels did not even need to be affiliated with the same
port of registry, as the Chikuzen and Buzen boats were not.
399
In the cases visible from the Register, often
even the overseeing institutions were varied when ships registered to ports affiliated with different places
traveled together, likely meeting up en route to Hyōgo and voyaging together to their destination.
The following brief case studies of captains’ activities based on the Register demonstrate the
different types of networks visible in the Inland Sea. The Register has nearly two hundred notations of
edabune alongside the name of the captain, although almost no specific flagship designations.
400
By
looking at the names of the captains, the ports of origin of the ships, and the dates of entry into Hyōgo, it
becomes apparent that captains from nearby ports were likely journeying together to Hyōgo. The ability
to take on dual roles of ship captain and ship owner, as implied in the maritime codes, is also suggested
here. Often the same captain’s name was listed arriving in Hyōgo on the same day, attached to two or
more ships that might have an edabune designation. It is highly likely that in this case, both ships would
be recorded under the same man’s name, as he would have had primary responsibility for the ships and
cargo as they passed through the checkpoint. Though most Japanese scholars have assumed that the
individuals are largely untraceable, since their names are commonly found and they mostly lack other
documentation, examining their geographical ties suggests otherwise. The ability to successfully navigate
a ship through the sometimes treacherous Inland Sea waters would not have been a common skill, and

398
Shinjō, Chūsei suiunshi no kenkyū, 777.
399
In this case, the Chikuzen and Buzen boats were both affiliated with estates overseen by Tōdaiji temple, which
may have provided added incentive to collaborate as they transported rents and dues to the same destination.
400
Kobayashi’s article delves into questions of which ships in the Register would have been accompanying vessels,
including certain notations which may not have explicitly stated that they were smaller ships but still served that
function. For the purposes of this analysis, I am using those entries that specifically are labeled “edabune” in the
Register.

224

while there may have been captains of the same name operating out of different ports, in many cases it
seems more likely that the same man was hired by people from different places to take their goods to
Hyōgo.

Captain Gyōbu Shirō: A Typical Range of Networks

Examining the voyages of the captain Gyōbu Shirō provides a clearer example of an individual captain
operating in multiple areas of the eastern Inland Sea (fig. 52, Table 12), as well as the complexities of
interregional collaboration.
401
Over the course of the year, he piloted six boats from the port of
Shishikui
402
to Hyōgo. Five out of those six boats arrived in Hyōgo on the same days as boats from
Kaifu.
403
One of those five Gyōbu Shirō boats also arrived on the same day as a boat from Nawari.
404
The
sixth Gyōbu Shirō boat, while not with any Kaifu boats, arrived on the same day as three boats from
Kannoura.
405
Though none of these boats were specifically labeled as accompanying ships, the proximity
of Shishikui to Kaifu and Kannoura strongly suggests that these ships left the Awa-Tosa border and
traveled to Hyōgo together.

401
Many of the captains’ names in the Register appear to have some sort of official designation, including gyōbu
(member of the law enforcement), emon (gate guards), hyōe (official military guards), and so forth. It is unlikely that
these men held dual posts as court officials and as skilled ships’ captains, and is therefore probable that these were
hereditary names. Jeffrey Mass, in Antiquity and Anachronism in Japanese History, devotes a chapter to the
intricacies of naming, including the practice of taking names from official posts (Mass 1992, “Identity, Personal
Names, and Kamakura Society,” esp. 97 – 99).
402
Awa province. Shishikui was almost certainly within the boundaries of Shishikui Estate. This estate had been a
Kōyasan holding in the thirteenth and fourteenth centuries, but more often than not rents and dues went unpaid, or
local Kōyasan appointees appropriated lands for their own use. In the fifteenth century it came under control of the
Hosokawa military governor. Amino Yoshihiko, Shikoku, Kyushu chihō no shōen: tsuketari sōsakuin (Tokyo:
Yoshikawa Kobunkan, 2005), 31–32.
403
Awa province. It is unclear exactly which estate Kaifu may have been affiliated with, though it was certainly
located within the confines of Kaifu-gun, which also contained Shishikui Estate. From the Nanbokuchō through
Sengoku periods it was overseen by the Kaifu family, working in conjunction with the Hosokawa military governor.
Heibonsha. Chihō Shiryō Sentā., Tokushima-ken no chimei, vol. 37, Nihon Rekishi Chimei Taikei; (Tokyo:
Heibonsha, 2000), 687.
404
Tosa province. Little remains about Nawari. A reference from a 1401 document refers to Kitayama Seiganji
Temple on Nawari Estate, but the administrative hierarchy is unclear . Heibonsha. Chihō Shiryō Sentā., Kōchi-ken
no chimei, vol. 40, Nihon Rekishi Chimei Taikei (Tokyo: Heibonsha, 1983), 71.
405
Tosa province. No documentary evidence remains specifically about Kannoura in the medieval period. In late
medieval times it is thought to have been affiliated with Asama Estate but there is little direct knowledge of that area
or its governance. Ibid., 40:48.

225

Figure 52. Gyōbu Shirō networks.

226

Province Port Name # of
Boats
Accompanying Captain Port Affiliation (s),
province
# of
Accompanying
Voyages
Awa Shishikui 11 Fuji Emon Kaifu (Awa) 2

Nakatsukasa Taro Kaifu (Awa) 2

Shōkan Hyōe Tarō Kaifu (Awa) 2

Mago Saemon Kaifu (Awa) 3

Saemon Saburō Kaifu (Awa) 5

Saemon Saburō Nawari (Tosa)

Saemon Saburō Kannoura (Tosa)

Saemon Saburō
Hirayama (Sanuki
or Bitchū)

Saemon Shirō Ushimado (Bizen) 1

Saburō Jirō (Yuraguchi) Yura (Awaji) 1
Settsu Jige 5 Nofuku/Nofuku Zushi Jige (Settsu) 4
Awa Tachibana 1 Nofuku/Nofuku Zushi Kaifu (Awa)

Sanuki Hiketa 3 Saburō Jirō (Yuraguchi) Yura (Awaji) 1

Saemon Shirō Ushimado (Bizen) 1
Table 13. Gyōbu Shirō Networks.
Incorporating the other entries in which Gyōbu Shirō is listed as captain of an accompanying
vessel (edabune designation) adds credence to the idea that vessels from places that were physically
separated actually coordinated their travel. In addition to the six Shishikui vessels noted above, Gyōbu
Shirō captained five edabune from Shishikui (Awa province). Three of those five ships arrived on the
same day as boats from Kannoura or Kaifu, similar to the combinations noted above. A fourth arrived
with two other accompanying ships from Hirayama
406
(Sanuki province) and Ushimado
407
(Bizen
province), as well as a ship from Yura
408
(Awaji province) with the notation of “nagabune,” meaning
“long ship,” which may have been an alternate reference to the flagship. If the Yura vessel was indeed the

406
There is some uncertainty as to whether Hirayama refers to a location in Bitchū or Sanuki, though it is more
likely to have been Sanuki (see Appendix 2). They are therefore both noted on the map, as Hirayama-B and
Hirayama-S respectively. The lack of any corresponding documentation for either specific location makes it difficult
to ascertain which estate or proprietor it may have been affiliated with.
407
Bizen province. While there are few detailed early records about Ushimado that provide information about its
direct oversight, it was referred to as Ushimado-ho in late medieval documents related to Honrenji Temple. It was
widely referred to in travel diaries as a stopping-off point for trade and pilgrimage voyages, suggesting a relatively
public nature. Amino Yoshihiko, Chugoku chihō no shōen (Tokyo: Yoshikawa Kobunkan, 1998), 16–17.
408
Awaji province. Although records are sporadic, it was almost certainly part of Yura Estate, overseen by a
shogunate-appointed administrator (zasshō) as of 1394. In 1476 a percentage of its rents and dues were given to
what is today Kibune Shrine in Kyoto, suggesting that in general Yura was still affiliated more with central court
and government powers, rather than local administrators. Heibonsha. Chihō Shiryō Sentā., Hyōgo-ken no chimei, vol.
29-1, Nihon Rekishi Chimei Taikei (Tokyo: Heibonsha, 1999), 611.

227

flagship to those three vessels, it indicates collaboration among captains not only from four different ports,
but four different provinces. In 1445, Sanuki, Awa, and Bizen were all under the provincial governorship
of members of the Hosokawa family, and those ties may have helped facilitate trade or other links. Bizen,
however, had come under the Yamana provincial governorship four years previously, and prior to that
had been under the Akamatsu family. As such there is no administrative tie linking all of the ports,
suggesting the presence of individual ties independent of administrative boundaries.
The fifth ship arrived on the same day as one Jige (as noted in a previous chapter, ships labeled
jige were registered to Hyōgo) and three Kuise ships. While this combination is unusual in comparison to
the geographic configuration of the rest of Gyōbu Shirō’s voyages, the very next day Gyōbu Shirō is
listed as the second captain on a Jige ship. As Jige ships were registered to Hyōgo, it would not have been
impossible for him to arrive at the Hyōgo checkpoint on his own ship one day and then assist an
acquaintance in piloting a local ship the next. There were four other notations of Gyōbu Shirō as second
captain of a Jige ship, all under Nōfuku or Nōfuku Zushi as primary captain. Three of those ships passed
through the Hyōgo checkpoint within twelve days of Gyōbu Shirō landing there as captain of a Shishikui
vessel, while the fourth example was nearly a month later. It is extremely likely that this was the same
Gyōbu Shirō, cooperating with the local captain Nōfuku. If that was the case, it speaks to the mobility of
captains and the potential for them to change from ship to ship relatively easily, as well as suggesting a
specific collaboration between the two captains, Gyōbu Shirō and Nōfuku.
Gyōbu Shirō was also listed as captain of an edabune-designated vessel from Tachibana. On the
same day that vessel arrived in Hyōgo, ships from Bingo, Bizen, and Sanuki provinces landed, as did five
ships from Kaifu (Awa province). Two of those Kaifu ships also were marked as accompanying vessels.
While the flagship is unclear, there is a notation by Gyōbu Shirō’s name indicating “Shishikui,”
suggesting that this was the same captain of the vessels noted above. There are also records of a Gyōbu
Shirō captaining three boats from Hiketa, all arriving on the same day as boats from Yura. Although it is
less clear that this is the same man, since Hiketa is further removed from Shishikui or Tachibana, it is not
impossible that he operated out of that port as well.

228

Expanding Gyōbu Shirō’s case to include other individual captains that accompanied him
suggests additional maritime networks. Two of the Kaifu ships that arrived in Hyōgo with Gyōbu Shirō’s
Shishikui vessels were captained by Fuji Emon. Nakatsukasa Tarō and Shōkan Hyōe Tarō, both from
Kaifu, each arrived twice with Gyōbu Shirō. Mago Saemon’s vessels from Kaifu also accompanied
Gyōbu Shirō’s boats on three different arrival days.
409
One name often found on the same day as Gyōbu
Shirō is Saemon Saburō, who was associated with ships from Kaifu, Aji, and Shiwaku in Sanuki, and
even Setoda in Aki. Although it is not certain that the same man was piloting all of those ships, the
frequency of his arrivals with Gyōbu Shirō on five different occasions and on ships from Sanuki and
Kaifu implies more than a coincidental connection. If nothing else, the frequency of Kaifu captains
arriving together certainly suggests collaboration among captains from the same port. Finally,
strengthening the argument that Gyōbu Shirō of Shishikui and Hiketa were the same man are records
showing that the captain Saburō Jirō (Yuraguchi), from Yura on Awaji, and Saemon Shirō from
Ushimado, both accompanied Gyōbu Shirō ships from Hiketa and Shishikui to Hyōgo. It seems highly
likely that there were connections between these men that transcended geographic boundaries.
Based on the number of similar examples of captains voyaging together in the Register, Gyōbu
Shirō’s case was typical of such individual networks, particularly in the eastern Inland Sea area. The
accompanying vessel notation was not always present, but tracing the associations of names and arrival
times reveals many similar networks in play.

Hyōe Tarō: Ship’s Captain and Ship Owner?

Hyōe Tarō’s case suggests that a single man could own ships or shares in ships registered in completely
different provinces, implying a surprising degree of autonomous investment potential on an individual
level. In looking at all ships captained by a Hyōe Tarō (fig. 53, Table 13), it appears that there were at
least two men by that name plying Inland Sea waters between Bingo and Harima. The main ports that

409
The number of boats here is unclear, since Mago Saemon is listed twice on the same day as captain of both an
unmarked ship and one with an edabune designation. It is possible that he was the owner of the smaller vessel.

229

Hyōe Tarō was associated with were Matsue (nine ships), Mihara (eight ships), and Yuge (five ships), as
well as a number of other ports with only one or two vessels.
410
Looking at the timing of the entries into
Hyōgo, the Yuge ships likely could not have been captained by the same man. Sometimes arriving in
Hyōgo between two and six days after another Hyōe Tarō-captained vessel, there would not have been
enough time for him to make another round-trip journey between Hyōgo to Yuge.
411
The travel times
from the Kyūshū-dō no ki from 1573 record an eight-day journey from Tadanoumi (on the mainland near
Yuge) to Hyōgo. Although this included several stops for sightseeing and poetry readings, it still would
have probably taken about four or five days between Yuge and Hyōgo without any delays. Especially
after factoring in time to load and unload cargo, the timing of the Yuge voyages would have been nearly
impossible for the same captain. The accompanying captains’ names also share few similarities with those
from the ships from the eastern ports, indicating a different maritime network at play and lowering the
chances that Hyōe Tarō owned the Yuge boats as well. Therefore we will regard the Yuge-based Hyōe
Tarō as a separate individual from the Awaji-based Hyōe Tarō.
The Awaji Hyōe Tarō captained vessels from Matsue, Mihara, and Ushimado (once as primary
captain, twice as secondary), and provides the more intriguing case. At first glance it seems impossible
that it could be the same man captaining all the vessels. The ships often arrive within one or two days of
each other, and in one case a Hyōe Tarō-captained vessel from Matsue and Mihara arrive on the same day.

410
The names of the accompanying captains for the less frequently used ports do not clearly indicate to which
network they belonged, so this analysis focuses on the traceable ports.
411
Nagasaki Ken, Chūsei nikki kikōshū: Kaidōki, Tōkan kikō, Ben no Naishi nikki, Izayoi nikki hoka (Tokyo:
Shogakkan, 1994), 563–569.

230

Province Port Name # of Boats Accompanying
Captain
Port Affiliation (s),
Province
# of
Accompanying
Voyages
Harima Matsue 9 Emon Tarō Mihara (Awaji) 1
Emon Tarō Hibi (Bizen) 1
Gorō Tarō Mihara (Awaji) 3
Mata Gorō Mihara (Awaji) 2
Kanimori Tarō Mihara (Awaji) 3
Taifu Jirō Mihara (Awaji) 1
Awaji Mihara 8 Emon Tarō Mihara (Awaji) 6
Gorō Tarō Mihara (Awaji) 3
Mata Gorō Mihara (Awaji) 2
Kanimori Tarō Mihara (Awaji) 1
Kanimori Tarō Yura (Awaji) 1
Taifu Jirō Mihara (Awaji) 1
Emon Jirō Murotsu (Awaji) 3
Hyōe Jirō Matsue (Harima) 2
Bizen Ushimado 3 Gorō Tarō Mihara (Awaji) 1
Emon Jirō Murotsu (Awaji) 1
Hyōe Jirō Matsue (Harima) 2
Table 14. Awaji-based Hyōe Tarō Networks.

231

Figure 53. Originating ports for Hyōe Tarō networks.

232

If, however, Hyōe Tarō was not the actual pilot of all of these vessels but rather the owner of the
ships, that would have allowed for the close voyages.
412
Many of the same names were recorded as
captains of the ships landing on the same days as Hyōe Tarō’s vessels: Emon Tarō (Mihara) arrived with
six Mihara ships and twice with Matsue vessels. Gorō Tarō (Mihara) appeared three times with both
Matsue and Mihara vessels and once with an Ushimado boat, and Mata Gorō from Mihara arrived with
two Matsue and two Mihara boats. As noted on the above table, additional captains also accompanied
Hyōe Tarō multiple times.
While the repetitive names certainly suggest a collaborative network in voyages, the timing of the
partnerships is of particular interest in determining Hyōe Tarō’s role. Although the Hyōe Tarō vessels
from different ports may have arrived within days of each other, none of the accompanying captains were
recorded in such quick succession. Except for Hyōe Jirō in Matsue (Harima province), all of the recurring
accompanying captains were from ports in Awaji province (Mihara, Yura, Murotsu). This implies a
conscious collaboration among the Awaji-based captains, including the Mihara Hyōe Tarō boats, but it
does not so easily explain that same network including the Matsue and Ushimado Hyōe Tarō vessels.
There were no other often-recurring accompanying captains from regions outside of Awaji. Were Hyōe
Tarō simply a common name shared among three different ship captains, there would likely be others
from diverse locations included in this network resulting in a more random scattering of names.
Furthermore, all of the cargo from Hyōe Tarō Mihara and Matsue vessels as well as their
accompanying ships consisted of Mihara salt, with Hyōe Tarō’s ships often carrying the largest cargo.
The Ushimado vessels are exceptions; where Hyōe Tarō was primary captain his vessel carried rice and in
both secondary captaincies his cargo was Shima salt. It seems more reasonable to conclude that Hyōe
Tarō likely based himself in Mihara where he was accepted into that network of other ships’ captains, and
also owned boats (or in the Ushimado cases where he is listed as secondary captain, perhaps shares in

412
The Matsue/Mihara based Hyōe Tarō was less likely to have also owned the Yuge ships, as there are far fewer
common ties with the individuals accompanying those ships. It is more probably that it was another man with the
same name.

233

vessels) registered in other ports. Even though he may not have physically piloted those ships, the fact
that he was already an established member of the extant network allowed his boats from Matsue to join
the voyages to Hyōgo, all carrying salt from Mihara. If this was indeed the case, not only was the same
man working as captain out of different ports, but he also owned property in the form of the ships
registered in entirely different provinces. Further indicating his independence from traditional hierarchical
ties, in 1445 those provinces shared no common provincial governors. Harima and Bizen were under
governance by different members of the Yamana family, but the Hosokawa family controlled Awaji.
Hyōe Tarō’s ability to captain or own numerous vessels in two provinces led by completely different
families may indicate a higher level of labor mobility and individual economic ties to diverse locations.

The Western Inland Sea Region: A Port-Centered Model

While the above records indicate individual captains’ mobility throughout the inland sea area,
individually based collaborations were not necessarily embraced wholly by all captains who made
multiple voyages. Instead of individual connections, such as Hyōe Tarō voyaging with the same people
repeatedly as noted above, networks based in the western Inland Sea area appear to have centered more
around places. Multiple captains from the same port journeyed together, but the individuals were
interchangeable from voyage to voyage. Captain Shirō Saemon provides an example (fig. 54, Table
14).
413
In examining his ships registered in the western provinces, there are far fewer instances of
consistent collaboration. Shirō Saemon’s boats hailed twice each from Setoda, Tajima, and Yuge, and
once from Yanai and Onomichi. One of his Setoda boats arrived on the same day as an accompanying
vessel from Onomichi that lists him as a secondary captain, so it is likely that he was at least part owner in
that boat.

413
There may have been two captains named Shirō Saemon, as ships under that name come from several ports in
Aki, Bingo, Iyo and Suō provinces (likely all the same man) and then another series of ships from Kaifu in Awa.
Although the timing would not necessarily have prevented Shirō Saemon from making all of the voyages, the lack of
geographical or personal ties suggests that there were two different individuals, an Awa-based captain and another
captain from the Aki/Bingo area.

234

Province Port Name # of
Boats
Accompanying Captain Port Affiliation (s),
Province
# of
Accompanying
Voyages
Aki Setoda 2 Saburō Saemon Setoda (Aki) 2
Hyōe Tarō Yuge (Iyo) 1
Emon Tarō Shimozui (Bizen) 1
Jirō Saemon Onomichi (Bingo) 1
Mago Saemon Kaminoseki (Suō) 1
Hyōe Saburō Onomichi (Bingo) 1
Setoda (Aki) 1
Bingo Tajima 2 Mago Saemon Mihara (Bingo) 1
Iyo Yuge 2 Tarō Saemon Warae (Bingo) 1
Hyōe Tarō Yuge (Iyo) 1
Emon Tarō Shimozui (Bizen) 1
Jirō Saemon Mitsunosho (Bingo) 1
Suō Yanai 1 - - -
Bingo Onomichi 1 Tarō Saemon Warae (Bingo) 1
Table 15. Shirō Saemon Networks.

235

Figure 54. Shirō Saemon networks.

236

Accompanying captains’ relationships in the western Inland Sea region are simultaneously
less frequent and more complex than those seen near the eastern Inland Sea provinces. While the
earlier examples showed numerous examples of the same multiple captains arriving together in
Hyōgo, there were far fewer recurring names from the western provinces. There were multiple ships
from the same port arriving in Hyōgo – for instance, on the days when Shirō Saemon ships from
Setoda and Onomichi arrived in Hyōgo, anywhere from four to eleven boats from those ports
arrived with him – but as seen in Table 14, accompanying individuals show far greater variation.
While there were repeated names that suggest collaboration, there are some notable
differences from the earlier examples. First, each captain’s name occurred far less frequently. This
may reflect the fact that ports further away from Hyōgo generally sent fewer (but larger) vessels
over the year, so there simply may not have been as much time to make as many voyages as one
could from ports in the eastern Inland Sea provinces. It also suggests less of a network among
specific individuals and a stronger connection to a particular port. It may not have been as important
to voyage with certain people as it would have been to simply journey with several other captains
from the same area.
Captains in the same locale also showed more mobility than in the eastern provinces. While
the Shishikui, Mihara, and Matsue-based captains noted above generally made multiple voyages
from the same port and were accompanied by the same men, in Shirō Saemon’s case the captains’
names occur more than once but with much more variation among the ports of registry. Setoda,
Mitsunosho, and Yuge are all islands adjacent to one another, and Tajima is another island not far
away. Onomichi and Warae are located on the mainland but in close proximity to those other ports.
Geographic proximity would allow the men to move easily from one place to the next, resulting in
more variation between captains and ports. Furthermore, unlike the eastern Inland Sea region,
where the Hosokawa family branches were posted as provincial governors over multiple provinces,

237

governance of the western provinces was less consolidated. While the Yamana family did hold
power in Aki and Bingo provinces, the Ōuchi were governors of Suō, the Kōno governors of Iyo,
and the Hosokawa governors of Bizen. The presence of multiple powerful families in this region,
combined with relatively fewer voyages, may have resulted in less opportunity for individual
captains to create strong networks with each other that were facilitated by common administrative
ties.

Collaborative Voyages: Final Thoughts

Examining the ties between ships’ captains, then, can reveal much about the maritime networks.
The maritime codes suggested the practice of multiple ships journeying together, and even these
few examples from the hundreds of ships that passed through Hyōgo demonstrate the prevalence of
that custom. Even without the hierarchical relations inherent in flagship/edabune designations, ships
frequently journeyed in the company of others. Ties between individuals were more common in the
eastern Inland Sea provinces, particularly evident in the Awaji-based captains. In the western areas
personal connections were not always as prevalent, but ships affiliated with the same ports often
traveled together. The timing of these voyages was likely a factor, as the shorter, more frequent
journeys between Awaji and Hyōgo may have offered more opportunity for individuals to
collaborate. The longer trips between Aki/Bingo and Hyōgo, with more potential for delays en route,
would likely have resulted in more difficulties in coordinating so many people’s movements.
The ability of the captains to pilot ships from disparate ports also suggests a high rate of
mobility amongst skilled workers. Fujimoto has noted that the position of captain developed out of
the kandori (oarsman) becoming more specialized and taking on a relatively high ranking position
within their estate.
414
This implies that originally they would have been tied to their estates, perhaps

414
Fujimoto, “Chūsei zenki no kandori to chiikikan no kōryū,” 20–21, 27.

238

mostly responsible for shipping rents and dues. While Fujimoto does say that the transition from
oarsman to captain happened during the breakdown of the estate system, individual mobility was
likely a relatively new innovation. Hitomi Tonomura, in her study of local merchants in the Lake
Biwa region, suggests that the decentralization of power in the late medieval period led to the
propensity of local powers to cultivate economic opportunity.
415
It is very possible that the ability of
men to ply their trade in different areas is another example of individuals taking advantage of those
opportunities. Captains in the Register often came not only from different ports, but also from
entirely different provinces. They were able not only to work out of different ports, but possibly
even to own boats registered in entirely different places, as shown in Hyōe Tarō’s case above.

Conclusion

Examining the extant recordings of seafaring, particularly as seen through trade records and
maritime codes, helps reveal some of the mariners’ perspective of the medieval period. That
perspective is not limited to sailors alone; it is manifested in the actions of coastal residents and in
their adaptations to accommodate and interact with seafarers. Maritime regulations reflect not only
a distinct mariners’ perspective, but also the complexity of the connections between sailors and
non-sailors. The very existence of the maritime codes, drafted and proposed by people affiliated
with ports around the Inland Sea, demonstrates the self-recognized need for regulations specific to
maritime shipping. There are no similar medieval-era codes to regulate, for example, whether
someone who borrowed another man’s horse to transport goods would need to supply his own
saddlebags or if they should be included in the contract. Overland shipping in the medieval period
has no records of requiring town dwellers help tie up merchants’ horses in the case of a storm. The
maritime codes regulated not only profits and fees for the sailors and cargo owners, but also the

415
Tonomura, Community and Commerce in Late Medieval Japan, 99.

239

expectations placed upon both seafarers and non-sailing residents for the safe completion of a
voyage. Land-based residents’ roles were in many ways just as important to the maritime
connection as that of the actual ships’ captains and sailors. While the trade ships brought their wares
to various locales, they were dependent on local communities for everything from aid in case of
emergency to determining the pricing for the cargoes. Local markets became more prevalent and
robust as maritime trade increased, which would only have served to encourage still more ships to
bring their goods to those ports. The mutual dependence between seafarers and coastal communities
was unique in that it was clearly codified in the maritime regulations, and in that those rules were
widely adopted.
Perhaps even more striking than the collaboration between sailors and land-based residents,
though, were the relationships among seafarers themselves. The maritime codes suggest that the
practice of ships traveling in groups was widespread, which is in many ways common sense. There
would have been safety in numbers, and one boat could come to another’s aid in case of an
emergency. The analysis of the Register demonstrates just how common it was for captains to travel
together, either embarking from the same port or bringing together men piloting ships registered to
several different ports. The complexity of the networks is astonishing.
Most studies of medieval Japanese commercial development focus on merchants or
agricultural production, groups that are usually connected with a specific location. Yamamura
Kōzō’s study of guild development in medieval Japan notes that regardless of the commodity
produced, from lamp oil to needles to sedge hats, the people involved were generally anchored to
one location. The y could not move from guild to guild even within the same industry.
416
The
itinerant nature of the ships’ captains, who appear to have been free to pilot boats from numerous
ports, speaks to a great mobility among individuals. At the very least, captains communicated

416
Yamamura Kōzō, “The Development of Za in Medieval Japan,” The Business History Review 47, no. 4
(December 1, 1973): 438–465.

240

regularly with other captains from other ports and provinces to organize the timing of the voyages,
suggesting strong ties that transcended any affiliation to a particular locale. The mobile nature of a
seafarer, particularly one with the skills needed to successfully captain a ship, created the ability to
cross borders easily and frequently, further changing the face of labor mobility in medieval Japan.

241

Chapter 5: Conclusion: Filling in the “Seascape Gap”

This study has used an environmental history approach to narrow the historiographical “seaside gap”
in medieval Japanese studies. Through exploring the role of the sea in shaping ports, coastal
communities, economic development, and shipping customs on land and at sea, it becomes clear
that the sea has had a great influence. I have argued that the particular geography and environment
of the Seto Inland Sea affected the development of the specialized ports, laying the framework for a
burgeoning maritime trade that produced its own seafaring culture and customs.
In this conclusion, I will return to the concepts of the maritime cultural landscape and the
mariner’s perspective of medieval Japan as has been explored through the seascape aspects of
infrastructure, control, culture and piracy. By understanding these ideas, it is possible to see the
extent to which the sea impacted each of those categories, and to understand the importance of
maritime development in the medieval period.

Constructs: Environment and Maritime Infrastructure

The Seto Inland Sea’s environment and layout were primary factors influencing the development of
ports throughout the region. Areas that were naturally endowed with deep, sheltered waters and that
had visible navigational markers grew rapidly as important port sites, particularly if they were also
easily accessible to inland areas. Ports such as Onomichi, Setoda, and Hyōgo reflect those
circumstances as evidenced by their critical role in maritime trade. Written records describe
Onomichi’s bustling markets from early days, and the records from Hyōgo reflect the great volumes
of cargo coming from those ports and through Hyōgo itself. The size of these ports also allowed
ever-larger ships to moor there, reinforcing the high volume of trade goods that came through those

242

areas. As ships grew in size so did the ports that accommodated them, and the opposite held true as
well.
Yet sites facing more challenging locations were not always limited by their environments.
Smaller ports strove to overcome natural disadvantages by creating their own infrastructure,
building mooring sites to facilitate loading cargo. This allowed them to welcome larger vessels and
increased their opportunities for participating in maritime trade. Tracing the flow of goods from
various ports along the Inland Sea, it becomes apparent that by virtue of their location even certain
smaller ports held critical positions in the greater Inland Sea trade patterns. Takasaki and Kamagari
were relatively small ports shipping cargo volumes that were almost insignificant when compared
with their neighbors of Onomichi and Setoda. Nevertheless, they played a key role in east-west
trade routes. Products from western regions, including cloth, rice, and salt from Suō and Kyushu
provinces, were shipped through Takasaki and Kamagari. Although those ports did not have the
natural advantages of Onomichi or Setoda, they were the last “safe” stops for captains from the west
before reaching the treacherous waters of the Geibi island chain, with fierce currents and the
potential to encounter pirates. Takasaki and Kamagari captains could and did take advantage of
their familiarity with those waters to bear the western products safely to Hyōgo. Similarly, a port
such as Yura on Awaji Island took advantage of the geographic proximity to the forests on the Kii
Peninsula to become a hub for processing lumber into planks to be shipped to central markets.
Shipping smaller overall quantities of goods did not necessarily mean that the port was an
insignificant one in the larger network of medieval trade.
The environment not only affected port development but also local economies in terms of
the types of products that could be cultivated and sold. Coastal communities in general became
major sources of salt, but areas additionally able to secure a sustainable lumber supply such as Yuge
and Mihara (Awaji Island) saw even greater quantities of salt production. Conversely, eastern and

243

southern ports on Shikoku that did not produce as much salt for trade instead had more wood
surplus, as reflected by the cargo records from the Hyōgo Register. Climate affected different
products, resulting in the limited cultivation of Champa rice and indigo in the milder climates of the
Sanuki and Awa locales or the ability to practice double-cropping of rice, wheat, and beans for a
nearly year-round surplus output. Seasonal changes also affected fishing, most clearly visible in the
shipping of sardines for trade. The ability to procure certain goods sometimes helped determine the
relative importance of a port, as products took on regional “brand names” or the area became
known for a particular item.
Sites with the most difficult navigation sometimes became a locus for a different type of
maritime group: pirates. Unlike seafarers engaged in licit trade, pirates needed to be able to
withdraw to hidden port sites, and they took advantage of rapid currents and isolated isles to create
their strongholds. Even the maritime codes hint at the pirates’ ability to navigate ships far and fast.
The codes remonstrate captains who pilot ships stolen by pirates regardless of location, implying
that the ship may have been resold far from where it was originally captured. Pirate presence
directly affected the traffic and customs involved with trade, as seen in the examples of Takasaki
and Kamagari, and in the tendency of ships’ captains from different ports to travel together. There
was safety in numbers as they navigated through pirate-infested waters.
Environment and geography, then, were critical factors in shaping the development of the
maritime cultural landscape. These factors determined where ports could naturally develop easily,
leading to strong local economic development, or where pirates could ensconce themselves in
places that less skilled mariners could not penetrate. Climate and seasonality affected the production
of particular items, further influencing a port’s prominence in shipping those goods. Yet areas that
did not have natural environmental advantages were in turn shaped by the people that lived there.
Humans influenced the landscape as coastal communities built the infrastructure to accommodate

244

ships, began double-cropping to maximize production year-round, or took part in fishing and
marine harvests to support their community and trade. Landscape and environment both shaped and
were shaped by the people living there.

Control: Governing the Seascape

Central and local powers alike recognized the importance of controlling geographically desirable
port and river sites. Places such as the bustling port of Onomichi or the lucrative checkpoint at
Hyōgo represent important sites that directly benefited their proprietors. Kōyasan, the proprietor of
landlocked Ōta Estate, recognized the importance of controlling a port and designated nearby
Onomichi as its “warehouse” to store and ship Ōta Estate’s rents and dues. The port, endowed with
naturally favorable conditions to accommodate maritime shipping, continued to prosper and in 1319
even became a target for takeover precisely because of the wealth of commodities coming through
the port. The checkpoint at Hyōgo was similarly situated in an extremely desirable location, sitting
at the mouth of the Yodo River that served as a gateway to the markets of the Kinai region. It was a
critical source of revenue for Tōdaiji, so much so that even when the estate system was breaking
down in the later medieval period, Tōdaiji retained control of the checkpoint and assured itself of a
steady stream of revenue. Control of a port or checkpoint site provided immediate economic
benefits for its proprietor.
Proprietorship of a port site was not required to demonstrate local power, however. As seen
through the notations of exemptions in the Register, while the Sanuki proxy officials of the
Hosokawa provincial constable did not control the Hyōgo checkpoint, they were powerful enough
to regularly be granted toll exemptions on their cargoes. In contrast, while documents show that the
Ikuchi clan in Aki had at one point also been granted such exemptions, they abused the privilege
and subsequently had the exemption revoked. While they may have at one point had enough power

245

to command exemptions, they were unable to maintain that status. The conflicts along the
Yoshiigawa River checkpoints regarding exemptions for egoma transport to Ōyamazaki for
Iwashimizu Hachiman Shrine further emphasize the importance of controlling the checkpoint. As
the shrine still commanded substantial power in the Muromachi era the dispute was resolved in their
favor. But had it been in decline it is likely that the local checkpoint administrators would have been
able to successfully demand toll payments.
At sea, governance and control were manifested in different ways. Mariners created their
own set of maritime regulations to determine ways of settling disputes. Unlike terrestrial disputes,
where landmarks were generally more fixed and ownership of land was related to a definable space,
sea-based arguments could be less clear. Consider this case: a ship owner from one port hires a ship
captain who, en route to a second port, must put ashore unexpectedly at a third port and loses profit
as a result. Who is ultimately responsible for the captain’s actions, and which location (origin,
intended destination, or actual destination) has jurisdiction over the case? The authors of the
maritime codes recognized the complexity of such situations and created the regulations in an effort
to clarify some of the ramifications of maritime trade that differed from terrestrial endeavors. The
variants among the codes further indicate that while many customs were universal, certain local
customs were still influential. Regional variants sometimes indicate particular concerns, such as the
emphasis in the Ise-Shima version of the regulations requiring specific greetings when encountering
envoys or nobles. The lack of such regulations in other areas suggests that those encounters were
more common in the Ise-Shima area and required special rules.
Control of the seascape was largely directed by the geography and environment. Desirable
locations that easily accommodated sea traffic became a source of revenue for their holders, and as
such, they were targets for control through official proprietorship or forceful takeover. Seafaring

246

itself involved very different circumstances than overland travel, and mariners created their own
rules of control that acknowledged those differences.

Sociologies: The Maritime Community

Littoral communities often depended upon the sea for their livelihood, through cultivation of marine
products, maritime trade, and providing hospitality for travelers. Seafarers also adhered to specific
cultural norms and codes, evidenced in the maritime codes that they themselves wrote and
disseminated. The sociologies of the seascape often developed as a result of the type of port and
environment, but the residents also had a hand in developing their community, bringing their unique
mariners’ perspective into their actions.
The scope of seafaring occupations is perhaps most obvious through analyzing the literature
on and archaeological remains from port towns. Travel diaries and picture scrolls recount the near-
constant presence of saltmaking facilities and the close quarters of fishing villages. Women’s roles
are highlighted most in larger ports since they performed a variety of services for travelers, giving
those transshipment hubs a different atmosphere than ports focusing more on production. The
infrastructure associated with fishing and saltmaking is visible both in the written record, where
maps delineated salt fields and fishing grounds, and in the archaeological record, as artifact scatter
defines the areas used for those activities.
The impact of fishing and saltmaking communities is most clearly seen in the flow of goods
throughout the Inland Sea region. Salt, in particular, was the most important product of medieval
maritime production. The fact that the production point lent its name to the type of salt produced
indicates not simply the point of origin but also signals a sense of local pride. Price differentials
among types of salts indicate that location was indeed a factor in value (either due to naturally
superior quality or to personal connections with warehouse managers who increased prices). The

247

“brand name” of Bingo or Kojima commanded higher prices than many other locales that produced
salt. Other specialized goods such as egoma, indigo, and pottery were generally associated with a
particular production point and not cultivated as widely as salt or rice, indicating that a smaller
community produced those items. Such near-monopolies became points of pride in different ways,
particularly in the case of egoma, which was used for the manufacture of lamp oil at Ōyamazaki.
Although these were not marine goods, traders transported them throughout the Inland Sea.
Seafarers codified their cultural practices in the maritime regulations, inscribing their
obligations and responsibilities involved in maritime trade. Fundamentals of maritime shipping and
challenges faced by seafarers were highlighted in those codes. A ship at sea, when faced with an
emergency, needed to respond to the situation with what was at hand – there would have been no
opportunity to obtain additional tools or supplies. The maritime codes responded to that atmosphere
by outlining basic provisioning, from tools to anchors, and by advising ways to protect against
disaster or obtain aid. The reciprocal relations among captains, cargo owners, and ship owners are
visible in the codes, highlighting the connections between seafarers and the likely land-based cargo
owners. One could not function without the other: cargo owners relied on the ships’ owners and
captains to transport their goods, but without the cargo suppliers the seafarers would have had no
goods to sell. Agreements among these parties were complex and binding. If one side or the other
failed to live up to expectations, penalties ensued; but the compilers accounted for the effects of
storm and other emergencies as well as the resulting potential profit losses. These codes were
unique in that they appear to have been written by seafarers and submitted to the government for
approval and wider dissemination; in effect, the seafaring community created its own regulations. In
doing so they codified seafaring practices, giving practitioners binding legal responsibilities.
Perhaps the most interesting aspect of seafaring culture is manifested in the ties and
collaborations evident both in the codes and in practice. The maritime codes mandated connections

248

both between different ships and between boats and ports, most often in times of emergency. At
port, residents were required to come to a ship’s aid when faced with an oncoming storm, or to help
sailors moor the vessels in rough seas. Ships at sea were enjoined to come to the aid of other vessels
caught in storm or other emergency, rescuing injured or distressed sailors. Tracing the voyages of
individual captains noted in the Register suggests that collaboration was common among captains,
either making use of agreements between individuals who sailed together regularly, or relying on a
more port-centric agreement among several captains based at the same port to travel together for a
single voyage. Throughout all of these examples, however, the collaborative nature of seafaring is
highlighted repeatedly. These communities of marine cultivators, sailors, and port residents did not
exist separately from one another, but relied on internal and external ties to facilitate smooth
production and trade. Medieval Inland Sea communities had a highly interconnected network
encompassing geographically diverse ports.

Transgressions: Piracy in the Inland Sea

As Peter Shapinsky and others have explored piracy more thoroughly, this study has focused less on
the intricacies of piracy in favor of the other aspects of the seascape. Their presence and effect on
the maritime cultural landscape cannot be ignored, however. Records show that pirates were a
constant concern for travelers and traders as they harassed passing vessels. Pirate enclaves sprung
up on smaller islands protected by rushing currents, and much of the infrastructure was designed to
be temporary to allow for quick dismantling. As seen through tracing the flow of trade goods,
captains unfamiliar with the waters around the pirate enclaves altered their routes to avoid those
areas. This resulted in smaller ports such as Takasaki and Kamagari taking on a more important
transshipping role than they may have done had pirates not been a problem. Had they not felt

249

threatened, sea captains from the western provinces may have sailed directly on to Hyōgo, but
instead they chose to stop at the threshold to the pirates’ domain.
Piracy also affected sailing practices. Several of the maritime codes addressed piracy and
suggested harsh repercussions for those who dealt with pirates. Travel diaries describe how pirate
ships fled when additional ships came to the defense of the initial target, while safety in numbers
was reflected in the groups that sailed together from areas throughout the Inland Sea to Hyōgo.
Although the types of networks varied, from ties between the individual captains to more
geographically based agreements to sail together, the tendency to travel by convoy was likely a
direct response to the threat of pirates. While extant records that directly address pirate attacks in
the medieval period are relatively scarce, pirates’ influence was still felt in those adaptations to
seafaring both in the maritime codes and actual sailing practices.

The Medieval Maritime Cultural Landscape

Understanding the maritime cultural landscape of the Seto Inland Sea sheds light on the role the sea
played in shaping medieval Japanese history. Each of the four aspects of the seascape – constructs,
empire, sociologies, and transgressions – interact to form the maritime cultural landscape and to
color the mariner’s perspective. Actions on one level of the seascape affected the others. Smaller
ports, for example, could accommodate only smaller ships, which therefore required more frequent
voyages to transport large amounts of cargo. More trips could, in turn, strengthen networks among
sea captains, as they coordinated sailing with their counterparts in other ports to guard against
pirates and other maritime emergencies. Smaller ports were also more likely to focus on the
production of a particular product rather than transshipment, making the saltmakers, fisherpeople,
or farmers the driving force behind local economic development. Yet to ensure that they could fully

250

participate in maritime trade, certain smaller ports created more stable infrastructure to
accommodate shipping, setting the stage for increased growth.
Larger ports, on the other hand, were more likely to become transshipment hubs as they
accommodated the shipping of goods from a wide variety of locales. Control of these ports –
particularly ones with checkpoints such as those at Hyōgo – was often a critical concern for both
local and higher authorities. Major religious institutions, such as Tōdaiji, used the revenue garnered
from checkpoints to fund their own endeavors. Riparian checkpoint control and checkpoint fee
exemptions often reflected the rise and fall of political players. Those who held greater influence
could command exemptions, but when their fortunes fell, so too did their right to pass through the
checkpoints uncharged. Larger ports also became hubs for travelers, giving rise to bustling
marketplaces and attracting entertainers, helping to further stimulate local economic development.
The late medieval period has been characterized by the breakdown of traditional
hierarchical authority, allowing individuals more freedom to act independently at the local level.
Even in this relatively more permissive atmosphere, mariners appear to have enjoyed a mobility
unlike many others. While pirates would have been the ultimate expression of an unfettered seafarer,
even those who operated within legal limits were able to command an extensive network. The
Register indicates that the same captain could command ships from different ports affiliated with
different places and possibly even simultaneously own ships registered in various ports. This
suggests that captains could cross borders freely and could operate out of numerous areas to
maximize profit. It is unlikely that land-based cultivators would have had a similar freedom of
movement.

251

Future Directions and Final Thoughts

The role of the sea in local development cannot be overstated. This study has been an introduction
to the importance of the seascape, but additional research will reveal more ramifications of
maritime developments even on inland institutions or communities, and vice-versa. There have been
hints at this throughout this study, particularly in aspects such as proprietorship over port sites or
tracing trade routes inland – recall the Bizen pottery found in the mountain regions of modern
Hiroshima prefecture, or river-based transport of the egoma stretching far inland on Shikoku. While
we have been able to determine much about factors that affected maritime shipping routes, and have
a better sense of production trends near ports, questions about the origin and final destination of
these goods remain. Who contracted to send the products to Hyōgo, for example, or in between
other ports? After passing through the Hyōgo checkpoint, were those items transported up the Yodo
River to Kyoto marketplaces, and if so who arranged for that transport and who finally purchased
the products? With whom did the end profits finally rest? More work needs to be done to further
illuminate the beginnings and ends of the trade routes.
I also believe that the trends of local autonomy highlighted in this study provide important
clues to later development that will challenge certain preconceptions about the Edo period. Recent
scholarship has suggested that Japan was not as “closed” to foreigners as has long been assumed.
417

Other studies by scholars such as Robert Hellyer demonstrate that individuals in regions including
Shikoku and Kyushu regularly violated the restrictions on foreign trade in the later Edo period.
418

Archaeological excavations in Inland Sea smaller port areas and even pirate enclaves have yielded
significant numbers of foreign objects, including pottery sherds from the Korean and Chinese

417
For details in English regarding this idea, see Ronald Toby, State and Diplomacy in Early Modern Japan:
Asia in the Development of the Tokugawa Bakufu, 1st ed. (Stanford University Press, 1991).
418
Robert I. Hellyer, “The Missing Pirate and the Pervasive Smuggler: Regional Agency in Coastal Defence,
Trade, and Foreign Relations in Nineteenth-Century Japan,” The International History Review 27, no. 1
(March 1, 2005): 1–24.

252

mainland. It is quite possible that the strong network among ships’ captains that already existed in
the Muromachi era was maintained through later centuries, facilitating trade of foreign goods even
when such official trade ostensibly was discouraged. Additional studies should continue to shed
light on these topics.
Finally, further studies should explore seafaring practices beyond the maritime codes. War
tales and travelogues could supplement this study through their discussions of naval battles or ritual
practices. The Edo period sees a proliferation of maritime-related religious practices, such as
providing a shrine with a scale model before launching a new vessel. Many of those models are still
preserved today in the collections at Kotohira Shrine in modern Kagawa prefecture.
419
The roots of
these practices are not well understood. Additional research is needed to understand the role of
seafaring rituals in the medieval period and why they suddenly rose in popularity in the Edo period.
Historians of Japan must remember that it is, indeed, an archipelago, and recognize that the
sea itself influences the development of communities, trade, and everyday practices. The seascape
affected the landscape, as seafarers and residents of maritime communities adapted to their unique
environment. In turn they shaped that environment – both physical infrastructure and social
practices – to facilitate their own activities, through practices as localized as building a better
mooring point for ships or as widespread as penning a sweeping range of maritime regulations for
all seafarers to follow. As we further narrow the historiographical “seascape gap” I am confident
that scholars will continue to move away from the concept of Japan as a “rice country” and
reconsider it as an island country, recognizing the importance of Japan’s maritime past.

419
For a detailed discussion of Kotohira Shrine (also known as Konpira-san), see Sarah Thal, Rearranging the
Landscape of the Gods: The Politics of a Pilgrimage Site in Japan, 1573-1912 (Chicago: University Of
Chicago Press, 2005).

253

Appendix 1: List of Early Fifteenth Century Military Governors
Province Date Person Notes
Settsu 1397 - 1426 Hosokawa Mitsumoto * Concurrent with Sanuki, Tosa

1426 - 1429 Hosokawa Mochimoto * Concurrent with Sanuki, Tosa

1429 - 1442 Hosokawa Mochiyuki * Concurrent with Sanuki, Tosa

1442 - 1473 Hosokawa Katsumoto * Concurrent with Sanuki, Tosa
Remained under the Hosokawa until 1558
Harima 1371 - 1427 Akamatsu Yoshinori

1427 Akamatsu Mochisada

1427 - 1441 Akamatsu Mitsusuke

1441 - 1454 Yamana Mochitoyo
Remained under the Yamana until 1467, when it reverted to
the Akamatsu until 1521
Bizen 1371 - 1427 Akamatsu Yoshinori * Concurrent with Settsu

1427 Akamatsu Mochisada * Concurrent with Settsu

1427 - 1441 Akamatsu Mitsusuke * Concurrent with Settsu

1441 - 1466 Yamana Noriyuki
Remained under the Yamana until 1467, when it reverted to
the Akamatsu until 1521
Bitchu 1393 - 1405 Hosokawa Mitsuyuki

1405 - 1430 Hosokawa Yorishige

1430 - 1460 Hosokawa Ujihisa

Remained under the Hosokawa until 1518
Bingo 1401 - 1433 Yamana Tokihiro

1433 - 1454 Yamana Mochitoyo *Precedes appointment at Harima
Remained under the Yamana until 1497
Aki 1394 - 1400 Shibukawa Mitsuyori

1404 - 1406 Yamana Mitsuuji

1406 - 1411 Yamana Hiroshige

1418 - 1419 Yamana Noritaka

1429 - 1433 Yamana Tokihiro *Precedes appointment at Bingo

1433 - 1454 Yamana Mochitoyo
* Precedes appointment at Harima,
concurrent with Bingo
Remained under the Yamana until 1475
Suo 1402 - 1433 Ōuchi Hiroshige **All three men are listed simultaneously

Ōuchi Moriharu *Semi-concurrent with Nagato

Ōuchi Mochimori *Semi-concurrent with Nagato

1431 - 1441 Ōuchi Mochiyo *Concurrent with Nagato

1441 - 1465 Ōuchi Norihiro *Concurrent with Nagato

254

Remained under the Ōuchi until 1551
Nagato 1401 - 1431 Ōuchi Moriharu *Semi-concurrent with Suō

1432 Ōuchi Mochimori *Semi-concurrent with Suō

1432 - 1441 Ōuchi Mochiyo *Concurrent with Suō

1441 - 1465 Ōuchi Norihiro *Concurrent with Suō
Remained under the Ōuchi until 1551
Awaji 1398 - ? Hosokawa Mitsutoshi

1402 - 1404 Hosokawa Mitsuhisa *Precedes Awa appointment

1407 - 1441 Hosokawa Mitsutoshi

1446 Hosokawa Mochichika

Remained under the Hosokawa until 1519
Awa 1392 - 1402 Hosokawa Yoshiyuki

1411 - 1430 Hoshikawa Mitsuhisa *Follows Awaji appointment

1430 - 1449 Hosokawa Mochitsune

1449 - 1478 Hosokawa Shigeyuki

Remained under the Hosokawa until 1553
Sanuki 1397 - 1426 Hosokawa Mitsumoto * Concurrent with Settsu , Tosa

1426 - 1429 Hosokawa Mochimoto * Concurrent with Settsu , Tosa

1429 - 1442 Hosokawa Mochiyuki * Concurrent with Settsu , Tosa

1442 - 1473 Hosokawa Katsumoto * Concurrent with Settsu , Tosa
Remained under the Hosokawa until 1508
Tosa 1400 - 1402 Hosokawa Yorinaga

1400 - 1402 Hosokawa Motoyuki

1403 - 1426 Hosokawa Mitsumoto
* Semi-concurrent with Settsu and
Sanuki

1426 - 1429 Hosokawa Mochimoto *Concurrent with Settsu, Sanuki

1429 - 1442 Hosokawa Mochiyuki *Concurrent with Settsu, Sanuki

1442 - 1473 Hosokawa Katsumoto *Concurrent with Settsu, Sanuki
Remained under the Hosokawa until 1508
Iyo 1394 - 1411 Kōno Michiyuki

1414 - 1435 Kōno Michihisa

1435 - 1444 Kōno Norimichi

1449 - ? Kōno Michiharu
Under Hosokawa Katsumoto from 1455 - 1459, Kōno
Michiharu from 1459 - , under Hosokawa Katauji from 1465,
then returned to Kōno Norimichi and remained under the
Kōno until 1519

255

Appendix 2: Determining Disputed Port Sites

Figure 55. Disputed port sites from the Register.

256

Takefuji Nao was the first scholar to create a comprehensive map of likely locations for 107
port sites noted in the Hyōgo Register.
420
Several locations, however, may have actually been at
different sites from his initial assessment. To ensure the most accurate geographic-based analysis of
the flow of trade goods, it is necessary to pinpoint each port’s location as best as possible.
Takefuji’s locations for Mihara (Bingo province), Inunoshima (Bizen province), and Minamiura
and Hirayama (Bitchū province) were likely incorrect (fig. 55). While there are certain entries
marked Mihara-Bingo in the Register, the majority of the Mihara entries probably refer to a location
on Awaji Island. Inunoshima was actually more likely to have been Innoshima in Bingo, Minamiura
at the southern tip of Ikuchi Island (medieval Aki province; labeled above as Minamiura-Aki),
421

and Hirayama in Sanuki. Below I will outline my rationale for challenging Takefuji’s original
placements of these ports.
Hirayama’s location is least clear, but in tracing salt shipments it seems to be more
associated with Sanuki than Bitchū (See fig. 29, Annual salt shipments from the ports of origin as
noted in the Register). Hirayama vessels brought 2140 koku total of Takuma salts, 150 of Kojima
salts, and forty of Tajima salts to Hyōgo. Tajima salts came only from Tajima (which only shipped
twenty-seven koku of its own salt and 4970 koku of Bingo salt), Hirayama, and the transshipment
port of Ushimado. Hirayama-B lies closer to the Tajima production point. Takuma, while not a port
listed in the Register, in all likelihood referred to an area in Sanuki province near the ports of Nio,
Tadatsu, Utazu, and the location of Hirayama-S. Finally, Kojima salt was produced across the strait
on the peninsula in Bizen, but was transshipped from there to other points in Sanuki including
Kosai and Utazu. Shipping salt from both Kojima and Takuma to Hirayama-B would have been
backtracking en route to Hyōgo, while Hirayama-S was more central to the production of Takuma

420
Takefuji, “Chūsei no Hyōgo-tsu to Setonaikai suiun: Irifune Nōchō no fune sekichi hitei ni kanren shite,”
260–261.
421
Takebayashi Eiichi, “Chūsei Setouchi no shōhin ryūtsū,” Kenkyū hōkoku, Okayama kenritsu hakubutsukan
15 (March 1994): 3.

257

salts and was closer to other transshipment points for Kojima salt. Based on salt analysis, then, it is
likely that Hirayama was located in Sanuki, not Bitchū province.
It seems quite clear, on the other hand, that except for the cases in which the salt was
clearly labeled “Mihara (Bingo),” “Mihara” refers to a port on Awaji Island. Save for the westerly
port of Nakasho, all ports shipping Mihara salts were located on Awaji or points east towards
Hyōgo. Kuise and Amagasaki, east even of Hyōgo, were transshipment ports and would have
handled several different types of salt. Had the Mihara salt come from Bingo then there would have
been no reason for it to bypass all of the Bitchū, Bizen, and Sanuki ports in favor of Awaji – it is far
more likely that the production point was Mihara on Awaji. Tracking rice shipments supports this
theory, as Mihara shipped only Awaji bales, providing more evidence that Mihara was indeed on
Awaji and not in Bingo province.
The analyses for Minamiura and Inunoshima are most complex. In determining the accurate
location of Minamiura, first note that Minamiura shipped 1240 koku of Bingo salt to Hyōgo, a
significant amount. Sites east of Minamiura in Bitchū that also shipped Bingo salt included
Inunoshima (also a disputed site) with 2340 koku, Utazu with 580 koku, and Ushimado with 2912
koku. Utazu and Ushimado were transshipment hubs, as both ports also sent a combination of
Kojima, Tejima, Takuma, Shima, Shiwaku, Hiketa, and Katamoto salts. As such, it is unsurprising
that they would also include Bingo shipments on their vessels. The Bingo-only shipments from
Minamiura and Inunoshima, though, are further east than the alternate choices of Minamiura-aki
and Innoshima, both of which are surrounded by other ports shipping Bingo salt. The quantities and
proximity to other ports handling Bingo salt suggest that the accurate locations of both these ports
are in Bingo or Aki provinces.
Factoring in the network of ships’ captains only strengthens this hypothesis. In
Inunoshima’s case, twelve boats arrived in Hyōgo over the course of the year. Ten arrived with

258

ships from Aki or Bingo provinces (most often Setoda or Onomichi, but also with ships from
Kamagari, Takasaki, Tajima, Warae, and Mitsunoshō), while only eight arrived with ships from
Bizen province (usually from Banda, Shimozui, or Ushimado, but occasionally with ships from
Hachihama and Inbe), suggesting stronger ties with the Aki and Bingo areas. Saburō Kurō, who was
a member of a network based around Setoda, piloted six of the Inunoshima ships, and his name was
not associated with any other ports besides Inunoshima. Kurō Gorō piloted four of the Inunoshima
boats, and his name was also recorded once on a boat from the adjacent Yuge and once from the
far-removed Hirajima in Awa. Mori only captained one Inunoshima boat, but he was also recorded
on five boats registered in Mitsunoshō, a port located on the same island of Innoshima. The twelfth
boat was captained by Sōa, the only time that name is seen in the Register. Except for the lone
instance of Kurō Gorō in Hirajima, the geographic proximity of all of these men combined with the
accompanying ships and Bingo salt cargo strongly suggests that Inunoshima was in fact located in
Bingo province, not in Bizen.
The nine Minamiura boats show a similar trend. Eight of the nine boats arrived with vessels
from Aki, Bingo, or Iyo provinces, usually from ports on one of the adjacent islands (Mitsunoshō,
Yuge, Iwagi, Tajima, Tomo, Takasaki, Warae, and Setoda, which is also located on Ikuchi island).
Only six arrived with boats from Bitchū province, and three of those six arrived in Hyōgo on the
same day. Shichirō Hyōe, also a member of the Setoda-based captain’s network, was only affiliated
in the Register with five Minamiura vessels. Shirō Jirō was also part of that same network and was
associated with three Minamura boats, twice with the notation “Sunoe” next to his name. Sunoe is a
place name on the southern tip of Ikuchi island, strengthening Minamiura’s connection to that
location. Shirō Jirō’s name was associated with single vessels from the nearby Onomichi and
Mihara (Bingo province) ports, as well as the further Utazu, Sakihama, and Sakoshi ports. While it
is unclear if these were all the same Shirō Jirō or if there were multiple individuals by that name,

259

the Bingo-based ships were all likely to have been captained by the same man. Kurō Hyōe piloted
the ninth ship, and his name was found in the Register on only that single boat. His ship arrived in
Hyōgo on the twelfth day of the twelfth month, accompanied by both Shichirō Hyōe and Shirō Jirō.
Again, the geographic relationships between the men and the cargoes strengthen the likelihood that
Minamiura was in Aki province.

260

Appendix 3: Translation of Version B (KI 3071) of the Maritime
Regulations

Version B is thought to have been the original version of the maritime regulations, with
other modifications coming later, and so I have chosen to translate this version as an example of the
codes. While the original does not number each item, I have done so for easier cross-referencing
with articles discussed in Chapter 4. Unless noted otherwise, term definitions are taken from the
online version of the Nihongo kokugo daijiten. I am also deeply grateful to Dr. Murai Shōsuke and
Yokoyama Ayano for their help in deciphering some of the more obscure phrasings.

KI #3071-1
Item: When performing the farewell
422
before departing from home by ship, one should drink sake
with the family of the house and even their group of retainers, and should proceed en route.
423
When
arriving at the destination, one should excuse oneself from one’s lodgings and the port merchant,
and celebrate the [safe] arrival of the ship at that place, giving an appropriate gift to the person
acting as steward
424
there to give thanks for the ship’s [safe arrival].

422
門出 (kadoide or kadode). Upon undertaking a long journey, a particular farewell was performed outside
the home.
423
It is unclear whether this refers to the ship proceeding en route, or if the sailors, family, and retainers
should all proceed along the road to where the ship is moored.
424
地頭 (jitō)

261

KI #3071-2
Item: Ship hulks
425
(yoribune 寄船) and ships blown off course [that run aground]. They [items
from the ship] should [be used to] repair the local Shinto shrine or Buddhist temple. If there is
anyone
426
on the ship, he is given the rights and responsibilities of the ship owner.

KI #3071-3
Item: When at a port a moored ship
427
has losses, the wet items should be dried at that village and
given to the captain. For this reason the anchorage tax
428
is [levied on the] sails
429
and when the port
[slip] is rented. In this case [since payment was received in the form of the anchorage tax] there
should be no complaints from the provincial lord.

KI #3071-4
Item: When there are many ships moored,
430
in times of high winds the village should provide aid.
First, prioritize help for the ships upwind, as even if the downwind ships have tied-down their
anchors, if the upwind flow catches hold [shifts the location of the upwinds ships], all of the ships’
lines will not be able to hold fast. If the upwind ships happen to cut their lines, they will be dashed
into the downwind ships, and the ships will sustain damage.

425
寄船 (yoribune): A ship damaged due to severe weather, or a lost ship that had washed up on shore. It is
assumed that there is no one left aboard such a ship, and items salvaged from the hulk rightfully belonged to
the residents near the place where the remains washed up.
426
The term here used is 乗物 (norimono). It may refer to people on board the ship or to cargo, but based on
the context of the item it appears to be referring to any sailor left alive.
427
繋き船 (tsunagibune). Any ship tethered to a post at a seaside or river’s edge, or a boat in a port.
Interestingly, 繋 is also Tokushima dialect for “keel.”
428
碇役 (ikariyaku): Upon mooring, a tax was levied on ships. These fees were usually used for the upkeep of
the port, and in return the ship was guaranteed a safe berth.
429
帆前 (homae): Kubota suggests that this refers to the Muromachi-era howake-zeni 帆別銭, a tax probably
based on the size of the ship (represented by the number of masts/sails). Kubota, Kaisen taihoko, 51.
430
懸船 (kakaribune): moored vessels

262

KI #3071-5
Item: When a ship loading [laden ship] is in the way of
431
a ship unloading [empty ship], the
unloading ship must retreat to the mouth of the waterway.
432

KI #3071-6
Item: For riverboats, when a boat going upriver meets a boat coming downstream and the stems
433

abut each other, the downriver boat should make way. If that doesn’t occur and the upriver boat
runs into it, the [operators of the] downriver boat must repair
434
the damaged parts, apologize, and
[return]
435
the vessel.

KI #3071-7
Item: When sailing
436
on the open seas, when overcome and hit by a downwind ship, if even one
person transfers over from the damaged ship to the upwind ship,
437
that defiles
438
the upwind ship.

431
From 軋ろう (kishirō): To be so crowded as to bump into one another, or to crash into an obstacle and be
unable to proceed.
432
The term here is 寄船口. While 寄船 refers to a ship’s hulk (see note 3), based on context it seems likely
that in this case it refers to the opening to area where ships come by ( 寄る). For the same reason, while it is
not entirely clear here whether this is at the point of loading/unloading, as in at a port or wharf, or if this is
referring to laden/empty ships en route, as through a riverway. Based on the fact that the following item
specifies riparian travel and that the route is classified there as upstream or downstream, the lack of such
specificity in this item suggests that it is referring to loading/unloading at a port.
433
The term used is 船頭 (sendō), which usually refers to the ship’s captain but in this scenario is more likely
to refer to the “head” of the physical ship, or the stem.
434
作事 (sakuji): While referring generally to construction or repair of a building, by the Muromachi era it
came to indicate woodworking specifically.
435
船渡すべきなり. The person making repairs likely needed to return the vessel to its original owner.
436
はせる (haseru) in the original. Likely indicates 馳せる: to run fast; to run far using a horse or cart. In this
case the ship is doing the “running.”
437
This is not totally clear, since the downward ship damaged the upwind ship by hitting it, yet seems to be
referred to here as the damaged vessel.
438
けがたる (kegataru). Presumably 穢, meaning defiled or under a taboo.

263

KI #3071-8
Item: The sailing ship must make way [for other ships]. If this rule is not followed and if a rowed
ship
439
is overtaken and rammed, fittings [for the ship] should be made, an apology proffered, and
[the ship] given over.

KI #3071-9
Item: If a rowboat damages a sailing ship, there should be no objections.
440

KI #3071-10
Item: If at a port a group of sailors launches the ship but makes no progress, the captain is defiled.
441

KI #3071-11
442

Item: When there are tenders
443
for the flagship: [if?] the main ship’s cargo has been thrown away,
and if the tenders’ cargoes
444
are undamaged, they
445
should be distributed amongst the tenders.
Inasmuch as a parent’s neglect extends to the child, the child’s neglect should not be tolerated.
However, when the foremost flagship and the tenders’ cargoes are loaded together, there should be

439
押船 (oshibune): propelled by rowing, rather than by sails.

440
異儀有間敷事 (igi arumajiki): Presumably this means that the rowboat does not owe anything to the
sailing ship.
441
穢有べき (kegare arubeki): Here, the character is used for kegare instead of the phonetic spelling only, as
above.
442
In most other variants of the maritime regulations, this item is split into two separate articles. The second
article usually begins from “Even if the ship is stolen…”
443
枝船 (edabune): lit. “branch ships.” Accompanying vessels, usually smaller than a flagship.
444
The term used here is 配当 (haitō), meaning to apportion or allocate. From the context of the maritime
regulations overall, it refers to the process of distributing the profits or proceeds from the voyage and
indicates how those proceeds are apportioned to the captain, crew, and ship and cargo owners.
445
It is unclear whether the cargoes should be distributed or the transport fees. The latter seems more likely –
it sounds as if in this case the tenders should not be shorted from their fees even though the flagship lost its
cargo/was lost at sea.

264

a promise amongst the sailors and as such a report should be given.
446
Even if the ship is stolen,
[i.e.,] taken by pirates, if a northern ship is in the western provinces or if a western ship is in the
northern provinces, [no one should] buy or sell that ship and continue to sail. If [the ship has] cargo
loaded and is sailing, as soon as the ship owner sees this it will be returned, and there will be
problems for the captain as well.

KI #3071-12
Item: [In this scenario,] A borrowed ship
447
is taken from port to port and harbor to harbor, and the
captain is not concerned with his ship and goes up on land. [When] wind and waves [come] the
captain [returns to be with] his ship, and though he strives hard he is unequal to the task [of
stabilizing the ship]. If there is damage, if that flotsam can be saved it should be allocated to the
sailors.
448

KI #3071-13
Item: When borrowing a boat, on its return if there is line (ropes) from straw or vines,
449
you should
return line that was [in the condition that they were] when you borrowed it. However, if the line is
not tattered, and if that was the original situation, then this does not apply.
450

446
沙汰 (sata): It is unclear to whom the report should be given. It may simply refer to the promise made and
distributed amongst the sailors to distribute things equally.
447
借船 (karibune): To borrow a ship, especially in the case of a ship’s captain or cargo owner borrowing a
vessel from a ship owner for the purposes of transporting trade goods. The borrowed ship itself.
448
The term here is 舟子 (shūshi). While it could refer to the captain, it can also mean the sailors or oarsmen.
Based on the need for the captain to take responsibility that is demonstrated in other items, it is unlikely that
he would be rewarded for causing damage to his ship, and is more reasonable that any usable items would be
given to the sailors.
449
かづら (kazura, probably 葛 or 蔓): Vines, bamboo, wicker.
450
Here, the authors of the laws seem to anticipate that the ship owner would have provided line in good
condition, but that it would have suffered wear and tear en route and therefore new or good condition line
would need to be returned with the boat. The second sentence indicates that if the line was undamaged on the
voyage then it was sufficient to return the original line.

265

KI #3071-14
Item: When borrowing a ship, regardless of whether its woven mats
451
are large or small, as above
the mats should be returned in the same condition that they were when borrowed.

KI #3071-15
Item: When hiring a sailor, there is the custom of hiring them directly [by the ship owner?].
452
Give
sailors that are of equal ability as [yourself] to the captain. Moreover if you hire sailors and they are
of no use to the captain, if they cause trouble on board ship no matter where you are they should be
put ashore with travel funds
453
[to go home].

KI #3071-16
Item: When in a borrowed boat, if it is damaged and ten sailors die, if only the captain lives the
determined ship’s cargo
454
should be given over [to the captain]. However if the ten sailors live and
the captain alone dies, there should be no cargo [paid].

KI #3071-17
Item: When at sea and the flagship and tools are lost,
455
preparations
456
should be made according to
the abilities of the people on the ship.
457
When the ship enters the harbor, you should appeal
458
to
the village officials and receive guidance
459
[about how to proceed].

451
苫 (toma): A woven mat of rushes.
452
The term here is 手前 (temae), indicating a first-person perspective or placing someone directly under
one’s lead. It is unclear who would be doing the hiring, but it was likely the ship owner or cargo owner, since
the ship’s captain is referred to separately.
453
旅銀 (ryogin): Travel fees.
454
船床(funadoko): This literally refers to the ship’s hold, but based on the context here likely indicates the
cargo itself.
455
It is ambiguous if the flagship and tools were lost, or if the flagship’s tools were lost.
456
It is unclear if the “preparations” are for docking, repairs, presentation of evidence, or something else.

266

KI #3071-18
Item: When the flagship is not in need of repairs, the various local lodgings owners should take care
of [the people on the boat].
460

KI #3071-19
Item: When a ship is rowed in
461
“backwards,”
462
the people there [at the port] should without
question help out.
463

KI #3071-20
Item: When the cargo is jettisoned, the local official should be appealed to for [reimbursement of]
both the travel fees and the purchased cargo. A written receipt should be issued.

KI #3071-21
Item: On borrowing a boat, when collecting transport fees on its return, 1/3 [of the fees] are to go to
the ship owner to use as he pleases. However, when borrowing it, if on its return it is carrying cargo
and [the ship owner so desires?] the ship owner can take less than 1/3 of the fees.
464

457
船中才覚 (senchū saikaku): Saikaku can refer to making use of knowledge, or to the monies needed for
construction projects.
458
This likely includes giving evidence to prove the case of a shipwreck.
459
差図 (sashizu): To indicate methods, placement, or ordering; to give guidance or orders.
460
This may indicate that if the ship did need repairs, the sailors needed to sleep on or near the boat, perhaps
to keep watch over it.
461
漕入 (kogiru/kogiiru): Rowing a ship into an interior area.
462
逆船(gyakubune): In bad weather, when a ship is turned to have the stem face upwind and is “backed in”
to the mooring.
463
This would have been a difficult maneuver, particularly for larger vessels, and aid from those on land
would have been helpful.
464
This seems to imply that repayment to the ship owner for borrowing a boat could take the form of cash (i.e.,
1/3 of the transport fees), or a share of the cargo.

267

KI #3071-22
Item: On borrowing a ship, if the captain has [is involved in] a legal dispute
465
at his destination and
the ship is taken from [the captain], the borrowed ship’s captain must pay.

KI #3071-23
Item: Even with borrowed boats, if the boat suffers losses, the borrower should not have to pay. Due
to the captain’s indiscretion the captain shoved off and launched the ship, and at that point if the
ship is damaged the borrower pays. However, it should be [done] according to the original
agreement.
466

KI #3071-24
Item: When, in using a borrowed boat, that boat is eaten by insects,
467
this is not the borrower’s
responsibility. However, if the items on board [are seen to be damaged]
468
upon tying up at port, the
[borrower]
469
should pay for his carelessness.

KI #3071-25
Item: When the sculling oar or rudder is damaged the borrower must make reparations. However,

465
公事 (kuji): Usually translated as “court fees” in the classical and early medieval times; by the Muromachi
era, it can also refer to a legal dispute ( 訴訟).
466
最前約束 (saizen yakusoku): It is unclear what shape the “original agreement” may have taken, since no
such documents have yet been found. Presumably it was a form of contract between the ship owner, borrower,
and/or the captain.
467
虫喰たる (mushi kuitaru): This almost certainly refers to teredo worms (shipworms) infesting the hull of
the ship.
468
船附之物約委に入所を懸り. The wording here is unclear as to what happens to the items after tying up
at the port. Based on other versions which suggest that the state of the cargo is verified or come to be
understood ( 種々に理処 or 常々理る処に), it is likely that some sort of assessment of the cargo is being
performed and if damage is discovered then the borrower must pay.
469
“Borrower” here is 手前 in the original item (see note 452). Other versions of the regulations clearly
indicate borrower (karite), so it seems likely that that is who is being referred to here as well.

268

on borrowing it, if there is already damage [scars] on the rudder pedestal, if you inform the captain
of it there is no need to pay.

KI #3071-26
Item: When a line (rope) breaks reparations are not needed. However, if a line is let loose and falls
then it must be paid for, and if an anchor falls it must be paid for.

KI #3071-27
Item: When receiving a ship, the various tools should be ordered and gathered at that time.
470

KI #3071-28
Item: When one takes a borrowed boat 100-ri
471
, or when one decides to return it, if there is some
kind of problem there, consider this at that time: If you ride out for 20 or 30-ri, if something is
different [from the arrangement] pay [any debts] with the cargo
472
at that 20 or 30-ri mark, and keep
to the earlier promises
473
with regards to cargo and other things.

KI #3071-29
Item: If there is a difference in arrangement on the part of the borrower
474
when a ship is borrowed,
it should be handed over in accordance with the [original] boat contract.
475
At that time repayment

470
This seems to indicate that any necessary tools should be ordered at the same time as arrangements are
made to borrow the ship.
471
1 ri was about 500 – 600 meters.
472
船床. See note 454.
473
Similar to the “agreement” described in note 466, it is unclear what “difference from the arrangement” 相
違の儀 or “promises” 約束 exactly denote here.
474
The “difference in arrangement” 相違之儀 here may refer to the borrower wanting to break the original
contract. Personal communication, Dr. Murai Shōsuke.
475
約束(yakusoku).

269

to a third party is halted.
476
However, according to internal dealings
477
of the boat owner, if there are
some personal items brought and paid out, it should be pointed out to the people affiliated with the
aforementioned boat.

KI #3071-30
Item: When a borrowed ship is treated for shipworms through burning the base and the ship itself
burns, the lender must make reparations.

KI #3071-31
Item: When the ship is damaged and someone’s life is saved, for example, if even one of those
people holds onto gold and silver [from the ship], no one should complain about it.

KI #3071-32
Item: When unhusked rice
478
is loaded, or if Chinese goods
479
are included in the cargo, if the owner
of the Chinese goods throws away his own Chinese items, the unhusked rice should not be
distributed [to the person who threw away his Chinese goods]. If the rice owner or the sailors
should quickly throw away Chinese goods, regardless of whether they knew about those Chinese
goods or even if they say they didn’t know, it should be reported.
480

476
其時右之上下仕戻し留め置なり. It is unclear to whom or what the 右之上下 refers. Dr. Murai Shōsuke
has suggested that the ship borrower may have sold items to a third party independent of any prearranged
contract, and therefore had violated the terms of the original contract. As a result, the borrower needed to buy
those items back.
477
船主内証 (senshū naishō): Secrets of the boat owner, likely indicating a private trade.
478
籾米 (momigome).
479
唐物 (karamono): Refers both to items from the mainland or Chinese-style objects.
480
The rice may have served as a cushion for smaller, fragile items, so the “Chinese goods” may have been
packed in the rice. If an emergency arose that required cargo to be jettisoned, in a rush the sailors may have
thrown heavy bags of rice overboard whether they realized or not that valuable items were packed inside.

270

KI #3071-33
Item: When a contract is made with the captain to load cargo, if there is an agreement or a
difference with the cargo owner, the place where cargo is loaded should leave space for a boat [to
load cargo] and receive [the cargo] according to the agreed transport fees. [If there is a natural
precedent for receiving goods, follow the examples on both sides].
481

KI #3071-34
Item: When giving a cargo manifest
482
to the captain, the others on board should also affix their
seals
483
to it. Ones without [their seals affixed] should not be included in the allocation even a little
bit. However, on inspecting the interior of the ship, even if the remaining items are not included in
the cargo manifest, they should be entered and included in the distribution, while discarded items
should not be included in the distribution.

KI #3071-35
Item: When cargo gets wet then the captain must make reparations. However, in the open sea if the
ship encounters high winds and the cargo is dampened through waves and rain, the captain does not
have to pay. [But] in port if rain seeps through and wets the goods then the captain is responsible for
it.

KI #3071-36
Item: With cargo loaded, or if even on the open seas a fire breaks out on deck, it should be dealt

481
自然以前之中より扱ひ持入候はゞ、双方可任其扱事. The wording here is unclear, but I have
interpreted as best as possible to incorporate the meanings from other versions of the maritime laws.
482
積日記 (tsumi-nikki): A cargo manifest. It may have been a letter sent detailing the contents of the cargo.
Kubota, Kaisen taihoko, 202.
483
加判 (kahan): In the Muromachi era, this seal was affixed by the responsible party in a leasing agreement.

271

with in the same way [with regards to reparations] as if the ship meets with high winds. However,
the cause of the fire should be regarded as [human] error.

KI #3071-37
Item: When the boat has cargo loaded and a sailor [steals something and] runs away, the captain
must pay. However, if the sailor is caught and is given over to the cargo owner, even if the stolen
cargo is gone [discarded] the captain is not responsible for repayment.
484

KI #3071-38
Item: Regardless of whether items within the ship are large or small, those items cut [nibbled on] by
rats should be included in the allocations.

KI #3071-39
Item: When cargo is jettisoned excessively, the ship should also be included in the allocations.
Further, if there is jettisoned cargo and if the ship is rescued, sailors’ personal effects should also be
allocated. If there is only a little [cargo jettisoned], though, then the sailors’ personal effects should
be kept out [of the calculations].

KI #3071-40
Item: When cargo is jettisoned, the destination is not reached, and the ship returns with an
allocation, cargo should be calculated at local prices
485
and allocated.

484
In this case, since they caught the thief presumably he would have been responsible for repaying the cargo
owner, and the captain was not penalized.
485
在所之買直: the buying price at that place.

272

KI #3071-41
Item: When cargo is loaded, if the cargo is lost, it should be allocated [the value should be
calculated as if it were bought]
486
at the destination.

KI #3071-42
Item: When cargo is jettisoned and the destination cannot be reached, nor can you retrace your route,
if the allocations are made en route it should be done according to the selling price at that
destination.
487

KI #3071-43
Item: When the ship’s cargo manifest is not brought to the captain, even if it is said that gold and
silver was thrown overboard, it should not all be entered in to the allocations.

KI #3071-44
Item: When sailing in open waters, if the ship is weak and the oakum
488
is wet, waves get in and
tools are lost and excessive cargo is lost, the boat itself must be factored into the allocations.

On Jōei 2 [1233] mizuno-to hitsuji, 3
rd
month 16
th
day, Hyogo Tsujimura Shinhyōe no jō, Satsuma
Bōnotsu Iida Bizen no kami, and Tosa Urado Shinohara Mago Uemon no jō,
489
respectfully present

486
Kubota, Kaisen taihoko, 186.
487
其所の売値: The selling price of that place.
488
のみ（筎）(nomi): Called makihada today, it refers to fibers from the bark of the maki tree 槇, a type of
pine. It is often used to line fastener holes or the seam between planks to prevent seepage. Kubota, Kaisen
Taihoko, 347.
489
These men are basically unknown other than for this document. There is no clear relationship documented
among them.

273

the above forty-three rules to the shogunate. In addition to asking about shipping codes, [we hope
that] you will affix your seal
490
to it. Even if there might be a bad law, there should not be bad
principles behind that law. Besides these 43 items, if there are reports from other ships, those should
be reported as well.

As above, when departing Osaka we presented this to all of the descendants of the honored lords
Matsudaira Tanba no kami and the Gouchi Furuya Shin-jūrō, and Kohei Minbu possesses a copy of
it. They have heard the above 43 regulations, and an additional seven given to Osaka deputy
provincial constable Soga Matasaemon and Matsudaira Hayato, and have together presented it as
above.

Hyogo Tsujimura Shinhyōe no jō
Satsuma Bōnotsu Iida Bizen no kami
Tosa Urado Shinohara Mago Uemon no jō
Jōei 2 [1233] mizuno-to hitsuji, 3
rd
month 16
th
day

Iida Bizen

490
袖判 (sodeban): refers to signing or stamping a document in the blank space at the right edge.

274

Works Cited

Adachi Hiroyuki. Nihon no fune: wasen hen. Tokyo: Nihon kaiji kagaku shinkō zaidan: Fune no
kagakukan, 1998.
Aida Jirō. Chūsei no sekisho. Tokyo: Arimine Shoten, 1976.
Amino Yoshihiko. Akutō to kaizoku: Nihon chūsei no shakai to seiji. Tokyo: Hōsei Daigaku
Shuppankyoku, 1995.
———. Chugoku chihō no shōen. Tokyo: Yoshikawa Kobunkan, 1998.
———. Chūsei minshū no seigyō to gijutsu. Tokyo Daigaku Shuppankai, 2001.
———. Higashi to nishi no kataru Nihon no rekishi. Tokyo: Kōdansha, 1998.
———. “Kodai-chūsei-kinsei shoki no gyorō to kaisanbutsu no ryūtsū.” In Koza nihon gijutsu no
shakaishi. Dai 2 kan: Engyō gyogyō, edited by Nagahara Keiji and Yamaguchi Keiji, 197–
271. Tokyo: Nihon Hyoronsha, 1985.
———. Nihon chūsei no hinogyōmin to tennō. Tokyo, Japan: Iwanami Shoten, 1984.
———. Shikoku, Kyushu chihō no shōen: tsuketari sōsakuin. Tokyo: Yoshikawa Kobunkan, 2005.
———. Umi no kuni no chūsei. Tokyo: Heibonsha, 1997.
———. Umi to retto no chūsei. Tokyo: Nihon Edita Sukuru Shuppanbu, 1992.
Amino Yoshihiko, Alan S Christy, and Hitomi Tonomura. Rethinking Japanese History. Ann Arbor,
Mich.: Center for Japanese Studies, the University of Michigan, 2012.
Amino Yoshihiko, and Nobumichi Inaba. Kaimin no shakai. Vol. 10. Amino Yoshihiko chosaku
shū. Tokyo: Iwanami Shoten, 2007.
Aoki Shigeru, ed. Onomichi shishi (shinshū). Vol. 4. Onomichi, Hiroshima: Onomichi Shiyakusho,
1971.
———. , ed. Onomichi shishi (shinshū). Vol. 1. Onomichi, Hiroshima: Onomichi Shiyakusho, 1971.
Arnesen, Peter Judd. The Medieval Japanese Daimyo: The Ōuchi Family’s Rule of Suō and Nagato.
Yale University Press, 1979.
Ashmore, Wendy, and A. Bernard Knapp. Archaeologies of Landscape: Contemporary
Perspectives. Malden, MA: Wiley-Blackwell, 1999.
Batten, Bruce L. Gateway to Japan: Hakata in War And Peace, 500-1300. Illustrated edition.
University of Hawaii Press, 2006.
———. To the Ends of Japan: Premodern Frontiers, Boundaries, and Interactions. Honolulu:
University of Hawai’i Press, 2003.
Benton, Lauren A. A Search for Sovereignty: Law and Geography in European Empires, 1400--
1900. Cambridge; New York: Cambridge University Press, 2010.
Cronon, William. “Modes of Prophecy and Production: Placing Nature in History.” The Journal of
American History 76, no. 4 (March 1, 1990): 1122–31.
———. “The Uses of Environmental History.” Environmental History Review 17, no. 3 (October 1,
1993): 1–22.
Dear, I. C. B., and Peter Kemp. The Oxford Companion to Ships and the Sea. Oxford: Oxford Univ.
Press, 2005.
Farris, William Wayne. Heavenly Warriors: The Evolution of Japan’s Military, 500-1300.
Cambridge, Mass.: Harvard University Press, 1992.
———. Japan’s Medieval Population: Famine, Fertility, and Warfare in a Transformative Age.
University of Hawaii Press, 2009.
———. “Shipbuilding and Nautical Technology in Japanese Maritime History: Origins to 1600.”
Mariner’s Mirror 95, no. 3 (August 2009): 260–83.

275

Friday, Karl F. Japan Emerging: Premodern History to 1850. Boulder, Colo.: Westview Press,
2012.
———. The First Samurai: The Life and Legend of the Warrior Rebel Taira Masakado. Hoboken,
N.J.: John Wiley & Sons, 2008.
Fujimoto Yorihito. “Chūsei zenki no kandori to chiikikan no kōryū.” Nihon rekishi, no. 11 (2004):
19–36.
Gay, Suzanne. “The Lamp-Oil Merchants of Iwashimizu Shrine: Transregional Commerce in
Medieval Japan.” Monumenta Nipponica 64, no. 1 (April 1, 2009): 1–51.
Goodwin, Janet R. Selling Songs and Smiles: The Sex Trade in Heian and Kamakura Japan.
Illustrated edition. University of Hawaii Press, 2007.
Gregory, Ian N., and Paul S. Ell. Historical GIS: Technologies, Methodologies, and Scholarship.
Cambridge; New York: Cambridge University Press, 2007.
“Hakata No Shio.” Accessed March 6, 2014. http://www.hakatanoshio.co.jp/.
Hall, John W. Government and Local Power in Japan, 500 to 1700: A Study Based on Bizen
Province. Michigan: University of Michigan Press, 1999.
Haneda Masashi, Rekishigaku kenkyūkai, and Watanabe Kenji. “Minatomachi no yūkaku to yūjo -
Shimonoseki Inarimachi wo chūshin ni.” In Minatomachi ni ikiru, 39–65. Tokyo: Aoki
Shoten, 2006.
Hasegawa Tadashi. Taiheiki. Dai 1-han. Tokyo: Shōgakkan, 1994.
Hatakeyama Akira. “Chūsei Tōdaiji ni yoru Hyōgo-seki no keiei to sono soshiki.” Journal of
Japanese History 494 (October 2003): 47–65.
Hayashiya Tatsusaburō, ed. Hyōgo Kitaseki Irifune Nōchō. Tokyo: Chuo Koron Bijutsu Shuppan,
1981.
Heibonsha. Chihō Shiryō Sentā. Hyōgo-ken no chimei. Vol. 29. Nihon Rekishi Chimei Taikei ; 29.
Tokyo: Heibonsha, 1999.
———. Kōchi- ken no chimei. Vol. 40. Nihon Rekishi Chimei Taikei. Tokyo: Heibonsha, 1983.
———. Tokushima- ken no chimei. Vol. 37. Nihon Rekishi Chimei Taikei ; 37. Tokyo: Heibonsha,
2000.
Heldt, Gustav. “Writing like a Man: Poetic Literacy, Textual Property, and Gender in the ‘Tosa
Diary.’” The Journal of Asian Studies 64, no. 1 (February 1, 2005): 7–34.
Hellyer, Robert I. “The Missing Pirate and the Pervasive Smuggler: Regional Agency in Coastal
Defence, Trade, and Foreign Relations in Nineteenth-Century Japan.” The International
History Review 27, no. 1 (March 1, 2005): 1–24.
Hiroshima Kenritsu Rekishi Hakubutsukan. Shōnintachi no Setouchi: Aki no kikakuten. Umi no
michi kara chūsei wo miru II. Hiroshima-ken: Hiroshima Kenritsu Rekishi Hakubutsukan
Tomo no Kai, 1996.
Hiroyama Gyōdō. Shio no nihonshi. Tokyo: Yuzankaku books, 1997.
Horton, H. Mack. Traversing the Frontier: The Man’yoshū Account of a Japanese Mission to Silla
in 736-737. Bilingual edition. Cambridge, Mass: Harvard University Asia Center, 2012.
Ichimura Takao. “Chūsei kōki no tsu/minato to chiiki shakai.” In Tsu, tomari, shuku, edited by
Amino Yoshihiko, 83–116. Tokyo: Shin Jinbutsu Ōraisha, 1996.
———. Chūsei Sanuki to Setouchi sekai. Tokyo: Iwata Shoin, 2009.
Imabari Kyōikuiinkai. “Chūsei kaizoku no shiro: Noshima iseki.” Bunka kyōkai Ehime, August
2008.
Imabari-shi Kyōikuiinkai. Kuni shitei shiseki Noshima-jō ato. Heisei 21 nendo hakkutsu chōsa
genchi setsumeikai shiryō. Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan,
January 23, 2010.

276

———. Kuni shitei shiseki Noshima-jō ato. Heisei 22 nendo hakkutsu chōsa genchi setsumeikai
shiryō. Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan, February 19, 2011.
———. Kuni shitei shiseki Noshima-jō ato. Heisei 23 nendo hakkutsu chōsa genchi setsumeikai
shiryō. Imabari-shi: Imabari-shi Murakami Suigun Hakubutsukan, March 4, 2012.
Imagawa Ryōshun. “Michiyukiburi.” In Chūsei nikki kikōshū: Kaidōki, Tōkan kikō, Ben no Naishi
nikki, Izayoi nikki hoka, translated by Inada Toshinori, 389–426. Tokyo: Shogakkan, 1994.
Imaya Akira. “Tetsu wo hakonda fune.” In Zoku tetsu no bunkashi: Nihon no kiseki to tōzai no
kaikō, edited by Shin Nihon seitetsu kōhōshitsu, New Edition., 45–57. Toyo Keizai
Shinposha, 1988.
Ishii Kenji. Zusetsu wasen shiwa. Shiseido, 1983.
Japan Coast Guard, Hydrographic and Oceanographic Department. “Setonaikai No Umi No Fukasa
Wa.” Setonaikai No Umi No Fukasa Wa, March 25, 2015.
http://www1.kaiho.mlit.go.jp/KAN6/5_sodan/mame/topic39.htm#%E9%80%9F%E5%90%
B8%E7%80%AC%E6%88%B8,465m.
JB Highway Service Company Limited. “Ohama Souvenirs.” Accessed March 25, 2015.
http://www.jb-highway.co.jp/souvenir/ohama_up.php.
Kagawa-ken (Japan). Kagawa-ken shi. Vol. 2 Chūsei. Kagawa-ken: Shikoku Shinbunsha, 1989.
Kagawa-ken Kyōikuiinkai. Kendō kankei maizō bunkazai hakkutsu chōsa: Muraguro iseki,
Tsumuura iseki. Kagawa-ken, March 2003.
Kagawa-ken Kyōikuiinkai, and Honshu Shikoku Renraku Hashi Kōdan. Ōurahama iseki
honbunhen. Seto Ōhashi kensetsu ni tomonau maizō bunkazai hakkutsu chōsa hōkoku V.
Kagawa Prefecture: Kagawa-ken Kyōikuiinkai, March 1988.
Kalland, Arne. Fishing Villages in Tokugawa, Japan. Honolulu: University of Hawai’i Press, 1995.
Kamei-Dyche, Rieko. “Tools of Authority: The Saionji Family and Courtier Society in Early
Medieval Japan.” University of Southern California, 2013.
http://digitallibrary.usc.edu/cdm/compoundobject/collection/p15799coll3/id/246709/rec/1.
Kamiki Tetsuo. “15 Seiki nakaba Setonaikai ni okeru kaisen to shōhin yusō: Bun’an 2 (1445) nen.
‘Hyōgo Kitaseki Irifune Nōchō’ no kisō bunseki: 5.” Nara Kenritsu Daigaku kenkyū kihō
14, no. 2 (December 10, 2003): 81–88.
Kamiki Tetsuo, and Sakiyama Masahiro. Kaidō rekishi no ta-minaru: Hyōgo no tsu no monogatari.
Kobe, Japan: Kobe Shinbun Sogo Shuppatsu Sentaa, 1996.
Karaki Hiroshi, and Hashitsume Shigeru. Chūsei no Sanuki. Takamatsu, Japan: Mikōsha, 2005.
Kasuga Jinja Jimusho. Kasuga Jinja Monjo. Vol. 1. 3 vols. Nara Meishinsha, 1928.
Kemp, Karen. “Geographic Information Science for Spatial Analysis and the Humanities.” In The
Spatial Humanities: GIS and the Future of Humanities Scholarship, edited by David J.
Bodenhamer, John Corrigan, and Trevor M Harris. Bloomington: Indiana University Press,
2010.
Ki no Tsurayuki. The Tosa Diary. Translated by William N. Porter. Bilingual. Tuttle Publishing,
2005.
Kitayama Kenichirō. “Chūsei minatomachi no chikei to kūkan kōzō.” In Chūsei Sanuki to Setouchi
sekai, edited by Ichimura Takao, 123–37. Tokyo: Iwata Shoin, 2009.
Kobayashi Yasuo. “‘Hyogo Kitaseki Irifune Nocho’ ni miru edabune.” Ritsumeikan bungaku 605
(March 2008): 795–807.
———. “Irifune Nōchō ni miru kokuryō to kasho.” In Hyōgo Kitaseki Irifune Nōchō, edited by
Hayashiya Tatsusaburō, 289–302. Tokyo: Chuo Koron Bijutsu Shuppan, 1981.
Kobe-shi Kyōikuiinkai. Futabachō iseki hakkutsu chōsa hōkokusho 3. Kobe, Japan: Kobe-shi
Kyōikuiinkai Bunkazaika, 2001.

277

Kodama Kōta, and Kihara Hiroyuki, eds. Kagawa-ken no rekishi. Vol. 37. Tokyo: Yamakawa
Shuppansha, 1997.
Kōen and Eiyū. Fusō ryakki. Teiō hennenki. Shinsōban. Tokyo: Yoshikawa Kōbunkan, 11.
Kokushi Daijiten Henshū Iinkai. Kokushi daijiten. Dai l-han. Tokyo: Yoshikawa Kōbunkan, 54.
Kubota Hiroshi. Kaisen taihoko: Sumida Shōichi hakushi kaisen shikimoku no kenkyū shūi. Yao:
Osaka Keizai Hoka Daigaku Shuppanbu, 1989.
Lake Biwa Museum. Nihon chūsei gyokairui shōhi no kenkyū: 15 seiki Yamashina-ke no nikki kara.
Research Report of the Lake Biwa Museum. Kusatsu-shi: Lake Biwa Museum, March 2010.
Makabe Yoshiko. “Bizen-yaki memo: ‘Hyōgo Kitaseki Irifune Nōchō’ no tsubo to suribachi.” In
Kōkogaku no shiten: Seikatsu ishiki no kōkogaku, 224–36. Okayama-shi: Unizuka Insatsu
Kabushiki Gaisha, 2009.
Matsubara Hironobu. “Kakai no kōtsū.” In Hito to mono no idō, edited by Yoshimura Takehiko,
47–86. Tokyo: Iwanami Shoten, 2005.
———. Kodai no chihō gōzoku. Tokyo: Yoshikawa Kōbunkan, 1988.
———. Kodai Setouchi no chiiki shakai. Tokyo: Dōseisha, 2008.
Matsui Teruaki. “Chūsei no Setonaikai suiun ni okeru Onomichi no ichi.” In Chūsei Setouchi no
ryūtsū to kōryū, edited by Shibagaki Isao, 95–118. Tokyo: Hanawashobo, 2005.
Matsumoto Kazuhiko. “Chūsei minatomachi Nohara no jisshō.” In Chūsei Sanuki to Setouchi sekai,
edited by Ichimura Takao, 11–46. Tokyo: Iwata Shoin, 2009.
Matsuoka Hisato. “Nachi suigun to nanbokuchō no dōran.” In Kōno-shi to Iyo no chūsei, 97–114.
Ehime Prefecture: Ehime-ken Bunka Shinkō Zaidan, 1987.
Matsushita Masashi. Umoreta minatomachi Kusadosengen, Tomo, Onomichi. Tokyo: Heibonsha,
1994.
McCarty, Michael. “Divided Loyalties and Shifting Perceptions The Jokyu Disturbance and
Courtier-Warrior Relations in Medieval Japan.” Ph.D., Columbia University, 2013.
Mogi Kotarō. “Examining Coastal Shipping Processes around Shōdo Island during the Tokugawa
Period.” In Proceedings of the 2nd Asia-Pacific Regional Conference on Underwater
Cultural Heritage, edited by Hans Konrad Van Tilburg, Sila Tripati, Veronica Walker,
Brian Fahy, and Jun Kimura, 2:619–29. Honolulu, HI: 2nd Asia-Pacific Regional
Conference on Underwater Cultural Heritage, 2014.
Morimoto Masahiro. Zōtō to enkai no chūsei. Tokyo: Yoshikawa Kōbunkan, 2008.
Morris, V. Dixon. “Sakai: From Shōen to Port City.” In Japan in the Muromachi Age, edited by
John W. Hall and Takeshi Toyoda. University of California Press, 1977.
Murasaki Shikibu. The Tale of Genji. Translated by Edward Seidensticker. 1st ed. Knopf, 1978.
Nagahara Keiji. Shōen. Tokyo: Yoshikawa Kobunkan, 1998.
Nagahara Keiji, and Yamaguchi Keiji. Koza nihon gijutsu no shakaishi, dai 2 kan: Engyō gyogyō.
Tokyo: Nihon Hyoronsha, 1985.
Nagasaki Ken. Chūsei nikki kikōshū: Kaidōki, Tōkan kikō, Ben no Naishi nikki, Izayoi nikki hoka.
Tokyo: Shogakkan, 1994.
Nakagawa Takeshi. “Ceramics Unearthed at the Furuaboshi Site in Hyogo Prefecture.” Trade
Ceramics Studies, 2008, 118–25.
Nakanoin Masatada no Musume. Towazugatari. Translated by Kōsuke Tamai. Tokyo: Iwanami
Shoten, 1997.
Nishitani Masahiro. “Shōen sei to saigoku shakai.” In Saigoku ni okeru seisan to ryūtsū, edited by
Tsutomu Kawaoka and Nobuyuki Koga, 9–40. Osaka-shi: Seibundō, 2011.
NPO hōjin suichū kōkogaku kenkyūjo. “2009 Nendo Kagawa-ken, Okayama-ken no shiryō chōsa ni
tsuite.” Suichū kōkogaku kenkyū, no. No. 3 (March 2010): 72–76.

278

Okita Masaaki. “Kodai no fune.” In Toshi to kōgyō to ryūtsū, edited by Tanaka Migaku and
Kanaseki Hiroshi, Vol. 3. Kodai no ronten. Tokyo: Shōgakkan, 1998.
Onomichi Shiyakusho. Onomichi shishi. Vol. 1. 3 vols. Onomichi, Hiroshima: Onomichi
Shiyakusho, 1939.
Pearson, Richard J. Ancient Ryūkyū: An Archaeological Study of Island Communities, 2013.
Plutschow, Herbert E, ed. Four Japanese Travel Diaries of the Middle Ages. Cornell University
East Asia Papers. Ithaca, N.Y: China-Japan Program, Cornell University, 1981.
Rainbird, Paul. The Archaeology of Islands. 1st ed. Cambridge: Cambridge University Press, 2007.
Sakurai Eiji. Zōyo no rekishigaku: girei to keizai no aida. Tokyo: Chuokoronshinsha, 2011.
Sakurai Eiji, and Nakanishi Satoru. Ryūtsū keizaishi. Vol. 12. Shin taikei Nihonshi. Tokyo:
Yamakawa Shuppansha, 2002.
Sanagi Nobuo. Setonaikai ni okeru shiwaku kaizokushi. Tokyo: Kyozaikenkyūsha, 1934.
San’yō Shinbunsha, ed. Kaitei no kobizen: Mizunoko Iwa gakushutsu chōsa kiroku. Okayama-shi:
San’yō Shinbunsha, 1978.
Satō Genzō. Gikeiki. Vol. 1. 2 vols. Tokyo, Japan: Heibonsha, 1979.
Satō Ryoma, Kanehara Masaaki, and Kitano Nobuhiko. Takamatsu-jō seki (Nishinomaru-chō
chiku) Sanpōto Takamatsu shogo seibi jigyō ni tomonau maizō bunkazai hakkutsu chōsa
hōkoku 2 Volume 1. Kagawa Prefecture: Kagawa-ken Kyōikuiinkai, 2003.
Segal, Ethan. “Awash with Coins: The Spread of Money in Early Medieval Japan.” In Currents in
Medieval Japanese History: Essays in Honor of Jeffrey P. Mass, edited by Gordon Mark
Berger, 330–61. Los Angeles, CA: Figueroa Press, 2009.
———. Coins, Trade, and the State: Economic Growth in Early Medieval Japan. Cambridge,
Mass: Harvard University Asia Center, 2011.
Sei Shōnagon. The Pillow Book of Sei Shōnagon. Translated by Ivan I Morris. New York: Penguin
Classics, 1977.
Setoda-chō Kyōikuiinkai. Setoda chōshi shiryōhen. Hiroshima-ken: Setoda-chō Kyōikuiinkai, 1997.
Shapinsky, Peter D. Lords of the Sea: Pirates, Violence, and Commerce in Late Medieval Japan.
Ann Arbor: Center for Japanese Studies, University of Michigan, 2014.
Shido Chōshi Hensan Iinkai. Shinpen Shido chōshi. Kagawa-ken Okawa-gun Shido-chō: Shido-chō,
1986.
Shimizu Ryō. “Nanbokuchō ki – Sengoku ki no shōen.” In Shōenshi kenkyū handobukku, edited by
Shōenshi kenkyukai, 154–81. Tokyo: Tokyodō, 2013.
Shinjō Tsunezō. Chūsei suiunshi no kenkyū. Tokyo: Hanawa Shobo, 1994.
Shirane, Haruo. The Bridge of Dreams: A Poetics of “The Tale of Genji.” Stanford, CA: Stanford
University Press, 1988.
Shoku Nihongi. Shinsōban. Vol. 2. Kokushi taikei. Tokyo: Yoshikawa Kōbunkan, 12.
Song Hui-gyong. Rōshōdō Nihon kōroku: Chōsen shisetsu no mita chūsei Nihon. Translated by
Shōsuke Murai. Tokyo: Iwanami Shoten, 1987.
Soranaka Isao. “Obama: The Rise and Decline of a Seaport.” Monumenta Nipponica 52, no. 1
(April 1, 1997): 85–102.
Steinberg, Philip E. The Social Construction of the Ocean. Cambridge Studies in International
Relations. Cambridge; New York: Cambridge University Press, 2001.
Suzanne Gay. “The Lamp-Oil Merchants of Iwashimizu Shrine: Transregional Commerce in
Medieval Japan.” Monumenta Nipponica 64, no. 1 (2009): 1–52.
Takebayashi Eiichi. “Chūsei Setouchi no shōhin ryūtsū.” Kenkyū hōkoku, Okayama Kenritsu
Hakubutsukan 15 (March 1994): 2–30.

279

Takeda Sachiko. “Roads in the Tennō-Centered Polity.” In Capital and Countryside in Japan, 300-
1180: Japanese Historians Interpreted in English, translated by Joan R. Piggott, 147–65.
Cornell University East Asia Program, 2006.
Takefuji Nao. “Chūsei no Hyōgo-tsu to Setonaikai suiun: Irifune Nōchō no fune sekichi hitei ni
kanren shite.” In Hyōgo Kitaseki Irifune Nōchō, edited by Hayashiya Tatsusaburō, 232–71.
Tokyo: Chuo Koron Bijutsu Shuppan, 1981.
Thal, Sarah. Rearranging the Landscape of the Gods: The Politics of a Pilgrimage Site in Japan,
1573-1912. 1 edition. Chicago: University Of Chicago Press, 2005.
Toby, Ronald. State and Diplomacy in Early Modern Japan: Asia in the Development of the
Tokugawa Bakufu. 1st ed. Stanford University Press, 1991.
Tonomura, Hitomi. Community and Commerce in Late Medieval Japan: The Corporate Villages of
Tokuchin-Ho. Stanford, Calif: Stanford University Press, 1992.
Totman, Conrad D. The Green Archipelago: Forestry in Preindustrial Japan. Athens: Ohio
University Press, 1998.
Verschuer, Charlotte von. Across the Perilous Sea: Japanese Trade with China and Korea from the
Seventh to the Sixteenth Centuries. Cornell East Asia Series. Ithaca, NY: East Asia Program,
Cornell University, 2006.
———. “Ashikaga Yoshimitsu’s Foreign Policy 1398 to 1408 A.D.: A Translation from ‘Zenrin
Kokuhōki,’ the Cambridge Manuscript.” Monumenta Nipponica 62, no. 3 (October 1,
2007): 261–97.
Walker, Brett L. The Conquest of Ainu Lands: Ecology and Culture in Japanese Expansion, 1590-
1800. Berkeley: University of California Press, 2001.
Wang Zhenping. Ambassadors from the Island of Immortals: China-Japan Relations in the Han-
Tang Period. Honolulu, HI: University of Hawai’i Press, 2005.
Westerdahl, Christer. “The Maritime Cultural Landscape.” The International Journal of Nautical
Archaeology 21, no. 1 (1992): 5–14.
Wigen, Kären. “Introduction.” In Seascapes: Maritime Histories, Littoral Cultures, and
Transoceanic Exchanges, edited by Jerry H. Bentley, 1–18. Perspectives on the Global Past.
Honolulu: University of Hawai’i Press, 2007.
Yamaguchi Kazuo. Nihon gyogyō shi. Tokyo: Tokyo Daigaku Shuppankai, 1957.
Yamamura Aki. “Medieval Towns.” In A Landscape History of Japan, edited by Kinda Akihiro,
65–88. Kyoto: Kyoto University Press, 2010.
Yamamura Kōzo. “The Development of Za in Medieval Japan.” The Business History Review 47,
no. 4 (December 1, 1973): 438–65.
———. “The Growth of Commerce in Medieval Japan.” In The Cambridge History of Japan,
edited by Yamamura Kōzo, 3:344–95. Cambridge: Cambridge University Press, 1991.
Yamatoya Ikumi. “‘Hyōgo Kitaseki Irifune Nōchō’ kara miru bukka no hendō.” Hokudai shigaku
46 (November 2006): 1–23.
Yamauchi Yuzuru. Chūsei Seto naikai chiikishi no kenkyū. Tokyo: Hōsei Daigaku Shuppankyoku,
1998.
———. Chūsei Setonaikai no tabibitotachi. Tokyo: Yoshikawa Kōbunkan, 2004.
———. Kaizoku to umijiro: Setouchi no sengokushi. Tokyo: Heibonsha, 1997.
———. “Kudako-jō no ikō to Kudako-shū.” Iyo shidan 300 (January 1996): 27–33.
———. Yugeshima-no-shō no rekishi. Ehime-ken Ochi-gun Yuge-chō: Yuge-chō, 1985.

Abstract (if available)

Abstract This study combines evidence from medieval written records, archaeological excavations, and geospatial analysis to explore Japan’s late medieval (14th–16th centuries) Seto Inland seascape. I argue that the geography of the maritime environment shaped the physical development of seaside communities, in turn affecting economic growth and creating a distinct maritime culture unique to that setting. By looking at ports and people, economic trade routes and ties, and shipboard practices and collaboration between seafarers, it becomes apparent that the sea was a critical factor in shaping trade patterns and transport customs. As a result, late medieval Japanese society became highly interconnected through complex maritime‐based trade networks. Littoral communities both shaped and were shaped by their environment, creating a distinct maritime cultural landscape.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Tools of authority: the Saionji family and courtier society in early medieval Japan

PDF

Sick days in the Konjaku monogatari-shū: healing and epidemics in late Heian Japan

PDF

The stars and the state: astronomy, astrology, and the politics of natural knowledge in early medieval Japan

PDF

Saving the myriad spirits: the development and diversification of the segaki ghost-feeding ritual in medieval Japan

PDF

The lady of the eighth ward: political, economic, and military power of nyoin during the twelfth century, Japan

PDF

Kokugo and the ""nation"" in Meiji-Era Japan: language standardization, ideology, and national identity

PDF

Going offshore: studies of the maritime zone in East Asia from the eleventh to the fifteenth centuries

PDF

Power of the purse: estates and the religio-political influence of Japanese royal women, 1100-1300

Asset Metadata

Creator Damian, Michelle M. (author)

Core Title Late medieval Japan's Seto Inland seascape: shipping, sailors, and seafaring

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program History

Publication Date 04/23/2015

Defense Date 03/06/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Economy,Fishing,geospatial analysis,GIS,governance,Hyogo,maritime,maritime cultural landscape,maritime laws,medieval Japan,Muromachi,OAI-PMH Harvest,Onomichi,piracy,Ports,salt,seafaring,seascape,Seto Inland Sea,ships,trade

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Piggott, Joan R. (committee chair), Bialock, David T. (committee member), Wills, John E., Jr. (committee member)

Creator Email mdamian@usc.edu,mmddives@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-558542

Unique identifier UC11300103

Identifier etd-DamianMich-3379.pdf (filename),usctheses-c3-558542 (legacy record id)

Legacy Identifier etd-DamianMich-3379.pdf

Dmrecord 558542

Document Type Dissertation

Format application/pdf (imt)

Rights Damian, Michelle M.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA