II. The Local Setting
A. Location of Enoshima
The images below will help to orient the reader as to the location of
Enoshima.
|
|
|
|
Satellite
image of Japan
|
Sagami Bay in
relation to Tokyo
|
Satellite
image of Enoshima in Sagami Bay
|
Enoshima,
Dragon's-Mouth Hill, and Koshigoe
|
Click to enlarge images
B. The Geologically Active Natural Setting
Enoshima Island is located in Sagami Bay, which is bounded by the Miura
Peninsula on the east and the Izu Peninsula on the west. Four plates
meet in this area: the Philippines plate, the Eurasian plate, the North
American plate and the Pacific plate. This is one of the most
seismically active areas on earth, and earthquakes and volcanic
eruptions are common.
As a recent example, the Izu-Oshima volcano in Sagami Bay began
erupting in November 1986. This eruption, which was visible from the
mainland, was spectacular at night. In July 2000, the Miyakejima
Volcano
began erupting further out in the bay. Although this eruption was not
visible from the mainland, many people in the area noticed a strong
sulfurous smell from the eruption.
The point is that the local inhabitants are quite familiar with
volcanoes. The average inhabitant may expect to experience several
eruptions in a lifetime. In other words, local people are familiar with
earthquakes, volcanoes, and the normal phenomena associated with such
events. They are unlikely to treat earthquake- or volcano-associated
phenomena as special or pass on the memory of such phenomena in legend
unless the events or phenomena were truly out of the ordinary.
Earthquakes and Crustal Uplift in the Area
Earthquakes in this area can have massive effects on the crust.
According to a report by the National
Information Service for Earthquake Engineering (NISEE) on the 1923 Great Kanto
earthquake:
An unusual characteristic of the Great
Kanto earthquake was the dramatic upheaval and depression of the
ground.
The earth was lifted as high as 24 feet at Misaki [the peninsula just
before the entrance to Tokyo Bay, not far from Enoshima], substantially
changing the shape of the shoreline. This uplift lasted only about 72
hours, however, before the ground began to sink, at first by as much as
two feet per day. When the settling had ceased, an offset of some 5
feet
remained.
The island of Enoshima was found to have risen by over one meter after
the 1923 event (see 藤沢市：地図に刻まれた歴史と景観,
pg. 41).
Landslides
The NISEE report goes on to mention the severe landslides that resulted
from the 1923 event.
The dramatic uplifting and depression
of the ground resulted in thousands of landslides, the worst of which
occurred in Idu [sic] province. Here the entire village of Nebukawa was
buried by a massive mudflow, killing hundreds. Landslides were also
observed on the Miura Peninsula, the southern part of Bo-so Peninsula,
and the mountainous district of southwestern Sagami.
Uplift and Subsidence
It is widely recognized that regional and local uplift and subsidence
of the earth's crust are associated with earthquakes. Below is a
representative citation from a lecture
on earthquake hazards (note: clicking on the link will download a large
.pdf file).
Regional subsidence can have
particularly severe effects near shoreline facilities such as harbours,
canals, coastal roads etc, where changes in surface elevation may occur
relative to water level. In the 1964 Alaska earthquake subsidence
caused
flooding of piers, docks, highways, railways, airstrips, and houses
over
many hundreds of square kilometres, as they were permanently lowered
relative to sea-level. Tectonic uplift over a similarly large area
caused shallowing of harbours and waterways which restricted their
later
use.
Changes in uplift and subsidence rates are often considered to be
earthquake precursors.
Changes in Sea-level
Small changes in sea-level can have dramatic effects on low-lying
coastal areas.
A 1994 study by the Japan Society of
Civil Engineers showed that a one meter rise in sea level would put 679
square kilometers (262 square miles) of Japan, or 0.18 percent of the
nation's land, underwater. (Japan
to Study Ways to Deal with Rising Sea Level, AFP, Aug. 1, 2001)
Topography
The main river of concern to this study is the Kashio River, whose
catchment basin is mainly around Totsuka, Ofuna, and parts of Kamakura.
Nowadays this river flows at Fujisawa into the Sakai River, which flows
into Sagami Bay at Enoshima.
The dominant characteristics of the natural setting of the rivers (the
Kashio and the Sakai Rivers) leading to the island of Enoshima are that
lakes and wetlands have often formed in the course of these rivers by
flooding and that the area has long been known for the suffering caused
both by flooding, especially flash floods, and drought.
The difference between the elevation of the Kashio River at its sources
and at its mouth is not large, and the river tends to flow slowly when
not in flood. Before the banks of the river were encased in concrete,
the sand washed down when the river was in flood tended to build up,
raising the river bed. Consequently, the channel of the river changed
frequently due to floods, causing great hardship to farmers.
The city of Fujisawa has an apt description of the Kashio River on its web
page. It states: "The Kashio River, whose catchment basin is in
Yokohama and Kamakura, is a violent river that overflows frequently
(柏尾川は横浜・鎌倉を流域とし、度々氾濫をしてきた暴れ川です).
See also The
Relationship between the Puyang River in China and the Kashio River
(as
of February 2006)
Below is a representative quote from a book on the topography of this
region.
The lowland occupies the southern part
of the region, from the eastern foothills of Komayama in Hiratsuka to
Enoshima in the eastern part of the Shonan sand dune zone. The southern
lowland is new land formed about 5,000-6,000 years ago. Its surface
consists of sandy, slightly raised land such as dunes, sandbanks, and
natural levees, as well as swampy land between the banks of columns of
dunes and muddy, low-lying swampy land in the flood-plain between the
Sakai and Hikiji Rivers. In the Shonan dune zone, which consists of
sandy, slightly raised land and muddy, swampy lowlands, the slightly
raised land often suffers damage from drought, while the low-lying
swampy land often suffers water damage. Like the northern plateau, the
southern lowland is not blessed with a good environment or good
conditions for agricultural productivity. Ponds and marshes are
distributed throughout the low-lying areas, and temporary wetlands are
often formed by the flooding of rivers and tidal backflows of sea water
(see
藤沢市：地図に刻まれた歴史と景観, pg. 12).
Above is a chart (click to enlarge)
showing the areas in the region around Enoshima affected by flooding
and
landslides (see
藤沢市：地図に刻まれた歴史と景観, pg.
13). As can be seen, flooding and landslides were common along the
courses of the Kashio and Sakai Rivers (the river banks were encased in
concrete in the postwar period; since then, flooding has been rare).
In other words, the dominant threats
to the people of the region and their livelihoods in the past have been
flooding, flash flooding, and drought.
C. Discussion
This portion of the Enoshima Engi
has long been regarded as myth because the main actors in its story are
a dragon that swallows children and a deity that descends upon a newly
risen island. Moreover, the details of the story are fabulous — an
enormous, swampy lake 40 li
(150 kilometers) around, people fleeing to escape a dragon, and so on.
Another problem is that the locations of many places in the story,
especially the location of the lake, have been uncertain.
So let us begin by locating the key landmarks.
Place-names
The Enoshima Engi mentions
the
village of Tsumura (津村, meaning village on a boat landing, inlet, or
estuary), which was in a valley of South Hill (南山, meaning "the hill
south of the lake"). The village does not exist any more; however, the
place-name Tsumura is located close to present-day Nishi-Kamakura
Station on the monorail from Enoshima to Ofuna. A tiny stream, the
Gotogawa (神戸川), runs down the gentle slope of the broad valley from
this
point, entering Sagami Bay just to the west of Koyurugisaki at the
Koshigoe Station of the Enoshima Electric Railway. This stream may be
regarded as the centerline of the valley. The place-name Tsu-nishi
("west of the inlet") is found to the west of the place-name Tsumura.
At
present, this locality appears to be about 30 meters or so above sea
level (estimating from my topographic maps; however, this is not
entirely certain because the locality is so built up that map contour
lines are difficult to spot). This locality will be our prime candidate
for the location of Tsumura.
If we have correctly located Tsumura, then the ancient location of the
lake should be to its north. Walking north from Tsumura for a few
minutes, one passes over the crest of a hill that is part of the Katase
Hill (Kataseyama) complex of hills. From the top, one can see the
sinuous bend of the Kashio River and its floodplain about 1.2
kilometers
below. This floodplain contains a number of locations — high schools,
police stations, and so on — that incorporate the term Fukasawa in their place-name. The Enoshima Engi tells us that the
long
lake was named Fukasawa, or "deep swamp." Therefore, this area will be
our prime candidate for the location of the long lake (1).
The Topography of the Area in the Ancient Past
About 5,000 years ago, the entire area was under the waters of Sagami
Bay, as can be seen from this map
showing shorelines in the Kanto region at that time (from History of
Fujisawa City, pg. 9). Also see this
study on the web. It was a time after the glaciers had melted but
the land along the coasts had not yet risen above the water.
A few thousand years later, during the Jomon period (8000-300 BC),
especially the later part of the period, the land had risen enough that
today's landforms were beginning to be recognizable. Below is a map of
the area during that period taken from History
of Kamakura, Archeology Volume (鎌倉市史
考古編, pg. 4).
This map is remarkable informative. Note
first of all, Tsu Estuary (Tsu Irie), which cut into Katase Hill at
precisely the point where we have located the village of Tsumura, which
had a boat landing according to the Enoshima
Engi (2). Secondly, the course of today's Kashio River is
labeled
as Ofuna Estuary (Ofuna Irie). At that time, it was an estuary of
Sagami
Bay.
Now let us compare the Jomon-era map with a map of the area during the
Kofun era (300-710 AD), most of which falls within the time frame of
the Enoshima Engi.
Firstly, by this time, Tsu Estuary had largely disappeared and was
replaced by a relatively thickly populated valley not far above the
water level. This is the prime candidate for the location of the
village
of Tsumura.
Secondly, Ofuna Estuary had been replaced by a low-lying area that
usually was a wetland except when it flooded. This could have been the
long lake known as Fukasawa,
or "deep swamp" in Japanese. Note its enormous size, just as stated in
the Enoshima Engi. The lake
could have reached 40 li
around, depending on the volume of water present and the flood
conditions (and on whether Kokei used the shorter Chinese li or the longer Japanese li as his unit of measurement). The
former Fujisawa Estuary is not such a likely candidate for the lake
because of its relatively small size (see the Jomon-era map above).
With these two locations tentatively fixed, we can go on to identify
the location of other landmarks. The text of the Chinese version of the
story states:
At this time, the five-headed dragon
first appeared at the water gate of Tsumura Village in the valley of
South Hill (the hill south of the lake) and began to devour children.
The people of the time named this place
Hatsu-kuhi-sawa ("Swamp Where the
Dragon First Began to Devour People") and called the steep hills to the
west
Eno. This swamp was the
water gate to the waters of the lake and an estuary of the Southern Sea
[Sagami Bay] (original text and translation
here).
Looking at the Kofun-era map above, the hill south of the swampy lake
was obviously part of the complex of hills now known as Katase Hill
(Kataseyama), and Tsumura was located in a valley of this hill. The
swamp Hatsu-kuhi-sawa (a
name
that was later shortened to Hatsusawa)
was probably the low-lying strip bordering the Katase Hill complex,
running west from the valley in which Tsumura lay, past present-day
Dragon's-Mouth Temple, into the wetland that at an earlier time had
been
the estuary leading to Ofuna. The steep hills to the west named Eno were the hills that overlooked
the wetland (today's Kugenuma) and the swamp bordering it.
The above detail of the previous map shows the new additions that we
have identified (labeled in blue).
Check of Locations
Now it is possible to make a preliminary check of the locations we have
tentatively identified. The Ofuna Estuary (and its successor, the
floodplain of the Kashio River) is our prime candidate for the location
of the lake. The Enoshima Engi
says that a hill (or complex of hills) lay south of the lake and that
there was a valley in the hill, a village with a boat-landing in the
valley, and the sea to the south.
Looking at the maps, we can see that the complex of hills known as
Kataseyama lies directly south of the Ofuna Estuary (and its
successor),
there is a valley in Kataseyama that hosted a village with a
boat-landing (Tsumura), and Sagami Bay (the southern sea) lies to the
south of the village. All of these locations fit the description in the
Enoshima Engi.
As an experiment, let us substitute the Fujisawa Irie (and its
successor, the Sakai River) as the candidate for the lake. If the
Fujisawa Irie (and its successor) was the location of the lake, then
there should have been a hill to its south. However, the maps show that
there was no hill, only sand dunes at best, and no places for a valley
or a boat-landing to have existed. Therefore, the Fujisawa Irie (and
its
successor) is not a suitable candidate for the location of the lake.
The fact that the above-named locations — the Ofuna Estuary (and its
successor), Kataseyama Hill, the Tsu valley, the village named Tsumura
and its boat-landing, and the southern sea (Sagami Bay) — all fit the
description of the Enoshima Engi
while other candidate locations do not, indicates that our
identification of the locations has a reasonable probability of being
correct.
Dimensions of the Lake
We have rough dimensions for the size of the lake's
predecessor (the Ofuna Estuary). The History
of
Kamakura,
Archaeology (pgs. 3-4) lists the dimensions of the Ofuna Estuary
(大船入江),
stating: "The Ofuna Irie
(estuary)
was no more than about 800 meters in width but it was 8000 meters long,
and a number of rivers flowed into it" (
この大船入江は幅800m内外に過ぎないが奥行は8000mに及び幾つもの川が流れ込んでいた).
In other words, a
rectangle 800 meters in width and 8,000 meters in length can be
drawn over a map of the Ofuna Estuary during Jomon
times, from
Kawana in present-day Fujisawa to the
vicinity of present-day Totsuka. Such
a rectangle would be 17.6 kilometers in length. Of course, the
Ofuna Estuary was not a perfect rectangle, as can be seen in the detail
of the Estuary below, which shows such a rectangle superimposed (not to
scale).
The
estuary had small inlets and peninsulas. The later lake would have had
similar inlets and peninsulas. If a person walking around the lake to
measure its perimeter followed the lay of the land, doing so would have
added somewhat to the length of the lake's perimeter.
Kokei states that the lake had a perimeter of 40 li
(有四十里之湖水).
The first question is whether Kokei using the Japanese or Chinese li
as his unit for measuring distance.
He probably was using the Chinese li. His history contains numerous parallels
between Lake Dongting (洞庭湖) (3) in
present-day Hunan (湖南) Province in China and the details
of the Enoshima Engi (refer
to
Parallels with Dongting Lake). Dongting Lake is also known as
"Eight-hundred Li Dongting" (八百里洞庭) because of its vast size when
flooded.
The next question concerns the length of the li Kokei was using.
Li of
varying
length have been used in China. In ancient times, the li was 360 paces (古代一里為三百六十步), which was 1890 feet or 576 meters, according to
Matthews Dictionary of Chinese. Similarly, Enshin
Saito (斉藤圓眞) states that in the Tang Dynasty one li =
559.8 meters (4). In
contrast, other
sources say that the length of the flight of an arrow (一箭之地) was 500 paces (五百步), or about 690 meters. Accordingly, 360 paces, or one
li, would have been about 359
meters.
So let us consider Kokei's li
to have been within the range of
359-576 meters. Thus, 40 li
could have been anywhere from 14.36 kilometers (1 li = 359 meters) to 23.04 kilometers (1 li = 576 meters), depending on the length of the li (5).
The lake left after the land rose above the water of the bay would have
varied in size, depending on how flooded it was. Nonetheless, its
maximum size would have been limited by the size of its predecessor,
the Ofuna Estuary.
Therefore, we can regard
the dimensions of the Ofuna Estuary as the maximum for the size of the
lake. The perimeter of the Ofuna Estuary was 17.6 kilometers, at the
minimum, and was perhaps somewhat longer. Kokei's 40 li
could have been anywhere from 14.36 kilometers to 23.04 kilometers, depending on the length of the li unit. It is clear that the
dimensions of the lake would thus fit within these parameters.
Before going further, we must first discuss the dragon.
Go to III. The Dragon
Report broken links
(1) Here are some maps to help orient the reader. This digital map (scale: 1/6500) locates
Tsumura near present-day Nishi-Kamakura Station on the monorail line
from Enoshima to Ofuna. A wider view
of
the same area (scale: 1/21000) shows Tsumura in relation to Enoshima,
Koshigoe, Dragon's-Mouth Hill, the Kashio River, and the numerous
locations around the Kashio River that incorporate Fukasawa (the name of the long-gone
lake where the dragon had its lair) as part of their name. These maps
can be viewed on the web here.
Note that Tsumura lies far inland nowadays, an indication of how much
the crust in the area has risen.
(2) I am aware that what I have labeled Tsu Estuary was not a proper
estuary. However, I have labeled it so to reflect the parallelism in
the
Japanese, which uses the term irie
(入江) both for the Tsu Estuary and the Ofuna Estuary (which was a true
estuary in the English sense of the term).
(3) Note
that the region around Lake Dongting has long been known as 瀟湘湖南
(sometimes abbreviated as 湘南), and the region around Enoshima in Japan
is often known as Shonan (湘南)
because of its supposed similarity to the region around Lake Dongting
in China. In other words, medieval Japanese saw parallels between the
Lake Dongting region in China and its counterpart (centered around
Enoshima) in Japan. When Kokei wrote his history, he may have had in mind
a parallel between the dragon in the lake in the Enoshima Engi and the dragon at the
bottom of Lake Dongting.
(4) Enshin Saito (斉藤圓眞) in The Biography
of Jikaku Daishi Ennin (慈覺大師傳) (Sankibo Busshorin Inc.,
山喜房佛書林, Tokyo, Japan, 1992, pg. 45) states that in the Tang Dynasty one li =
559.8 meters.
(5) By way of contrast, the Japanese
li,
which is 2.44 miles or 3.92 kilometers, is too long to have
been used. By the Japanese li,
the perimeter of the lake would have been almost 160 kilometers, much
bigger than the combined perimeters of the three estuaries (the Ofuna,
Fujisawa, and Oba estuaries) in the area.
