Japanese Fortifications and strongholds I

Fortunately for the economic survival of Japan, in subsequent decades defensive strategies, particularly in large-scale campaigns, began to center on entrenchments and fortifications, rather than on evasion and refusal of battle. Whether bushi perceived a problem and responded directly to it, or simply stumbled onto a solution for other reasons, is difficult to assess. Whatever their genesis, however, in the event, the new tactics helped prevent recurrences of devastation on the level of the Tadatsune episode.

The first significant campaign in which fortifications played a major role appears to have been Minamoto Yoriyoshi’s so-called Former Nine Years’ War against Abe Yoritoki and his sons, waged from 1055 to 1062. This contest took place in Mutsu, in the northeast, a region where warriors were heir to a three century-old tradition of establishing stockades as bases from which to control the local population. The Abe’s strategy throughout the conflict centered on ensconcing themselves and their followers behind bulwarks and palisades, in an effort to outlast Yoriyoshi’s patience and resolve. Such tactics played on the eagerness of Yoriyoshi’s troops to get back as soon as possible to their own lands and affairs. As Yoriyoshi’s lieutenant Kiyowara Takenori warned him:

Our government army is made up of mercenaries, and they are short of food. They want a decisive fight. If the rebels were to defend their strongholds and refuse to come out, these exhausted mercenaries could never maintain an offensive for long. Some would desert; others might attack us. I have always feared this.

If the Mutsuwaki, a nearly-contemporaneous literary account of the war, is to be believed, the forts the Abe manned, and the defenses they employed, could be elaborate:

On the north and east sides of the stockade there was a great swamp; the other two sides were protected by a river, the banks of which were more than three jo [about 10 meters] high and as unscalable as a wall. It was on such a site that the stockade had been built. Above the stockade the defenders stood towers, manned by fierce warriors. Between the stockade and the river, they dug a trench. At the bottom of the trench they placed upturned knives and above the ground they strew caltrops. Attackers at a distance they shot down with oyumi; at those who drew close they hurled stones. When, intermittently, an attacker reached the base of the stockade wall, they scalded him with boiling water and then brandished sharp swords and killed him. Warriors in the towers jeered the besieging army as it approached, calling for it to come forth and fight. Dozens of servant women climbed the towers to taunt the attackers with songs. . . .

Yoriyoshi’s tactics against this stockade were equally elaborate – and ruthless as well:

The attack began on at the hour of the hare [5:00-7:00 am] on the following day. The assembled oyumi shot throughout the day and night, the arrows and stones falling like rain. But the stockade was defended tenaciously and the besieging army sacrificed hundreds of men without taking it. The following day at the hour of the sheep [1:00-3:00 pm] the besieging commander ordered his troops to enter the nearby village, demolish the houses, and heap the wood in the dry moat around the stockade. He further told them to cut thatch and reeds and pile these along the river banks. Accordingly much was demolished and carried, cut and piled, until at length the stacks towered high as a mountain. . . . The commander then took up a torch himself and threw it on the pyre. . . . A fierce wind suddenly sprang up and the smoke and flames seemed to leap at the stockade. The arrows previously fired by the besieging army blanketed the outer walls and towers of the stockade like the hairs of a raincoat. Now the flames, borne by the wind, leaped to the feathers of these arrows and the towers and buildings of the stockade caught fire at once. In the fortress thousands of men and women wept and cried out as with one voice. The defenders became frantic; some hurling themselves into the blue abyss, others losing their heads to naked blades.

The besieging forces crossed the river and attacked. At this time several hundred defenders put on their armor and brandished their swords in an attempt to break through the encirclement. Since they were certain of death and had no thought of living, they inflicted many casualties upon the besieging troops, until [the deputy commander of the besieging army] ordered his men to open the cordon to let the defenders escape. When the warriors opened the encirclement, the defenders immediately broke for the outside; they did not fight, but ran. The besiegers then attacked their flanks and killed them all. . . . In the stockade dozens of beautiful women all dressed in silk and damask, minutely adorned in green and gold, wept miserably amidst the smoke. Every one of them was dragged out and given to the warriors, who raped them.

Yoriyoshi’s experiences with the Abe may have become the inspiration for increasingly widespread use of fortifications elsewhere in the country; nevertheless defensive works as elaborate or permanent as those Yoritoki and his sons occupied remained rare outside the northeast until the fourteenth century. Most Heian- and Kamakura-period fortresses were comparatively simple structures erected for a single battle or campaign.

Unlike the castle homes – protected by deep moats, wooden palisades and earthworks – of Sengoku-era warlords, early medieval bushi residences were scarcely distinguishable from those of other rural elites, and differed only in size and opulence from the dwellings of nobles in the capital.

Heian, Kamakura and Nambokucho warriors built their homes on level ground, usually on relatively high points in or very near the alluvial lowlands of rivers, and immediately adjacent to paddies and other agricultural fields. The main houses, stables and other key buildings were surrounded by water-filled ditches and hedges or fences, and accessed through wooden- or thatch-roofed gates. None of these features, however, appear to have been designed for military expediency.

Ditches were narrow and shallow – less than a meter wide and 30 cm deep  – and enclosed areas of 150 by 150 meters or more, presenting an impractically long line to defend with the small number of men normally available to early medieval landowners. They seem, therefore, to have served primarily as components of irrigation works, used to warm water and as a safeguard against droughts. Similarly, fences depicted in medieval artwork are low – a meter or so in height – and constructed of wood, thatch or natural vegetation, making them more suitable for controlling wandering animals than for keeping out marauding warriors. Careful archeological studies indicate that deeper moats and earthworks did not appear around warrior homes until the fourteenth century, and did not become widespread until the fifteenth.

The terms “shiro” or “jokaku” (usually translated as “castle” in later medieval contexts) appear frequently in diaries, chronicles, documents and literary accounts of late twelfth- and thirteenth-century warfare, but only in wartime situations, and nearly always in reference to field fortifications, erected for a particular battle. Such breastworks were intended to be temporary, and were rudimentary in comparison to the castles of the later medieval period, but they were not always small in scale. Some, like the famous Taira defense works erected in 1184 at Ichinotani, near Naniwa on the Harima border of Settsu province, could be quite impressive:

The entrance to Ichinotani was narrow; the interior was broad. To the south was the sea; to the north were mountains – high cliffs like a folding screen. There seemed not even a small space through which horses or men could pass. It was truly a monumental fortress. Red banners in unknown numbers unfurled, blowing toward heaven in the spring wind like leaping flames. . . . The enemy would surely lose its spirit when it looked upon this.

From the mountain cliffs to the shallows of the sea they had piled up large boulders, and over these stacked thick logs, on top of which they positioned two rows of shields and erected double turrets, with narrow openings through which to shoot. Warriors stood with bows strung and arrows at the ready. Below this, they covered the tops of the boulders with brush fences. Vassals and their underlings waited, grasping bearclaw rakes and long-handled sickles, ready to charge forth when given the word. Behind the walls stood countless saddled horses in twenty or thirty rows. . . . In the shallows of the sea to the south were large boats ready to be put to oars instantly and head to the deeper water, where tens of thousands of ships floated, like wild geese scattered across the sky. On the high ground they readied rocks and logs to roll down upon attackers. On the low ground they dug trenches and planted sharp stakes.

These descriptions, drawn from later literary accounts of the Gempei War, doubtless incorporate considerable exaggeration, but they nevertheless offer important clues about the nature of late twelfth-century fortifications. Two points, in particular, merit special attention. First, the preparations for battle involved provisions for escape – “countless saddled horses in twenty or thirty rows” and “large boats ready to be put to oars instantly,” to ferry troops to “tens of thousands of ships” waiting in deeper water – in addition to the defensive works. And second, as formidable as Ichinotani was, it was neither a complete enclosure nor fortified in all directions. In fact, the Taira defeat there was brought about, in part, by Minamoto Yoshitsune’s attack from the hills behind it. Similar tactics decided other key battles of the age as well.

Late Heian and early Kamakura “jokaku” were defensive lines, not castles or forts intended to provide long-term safe haven for armies ensconced within. Many were simply barricades erected across important roads or mountain passes. Others were transient wartime modifications to temples, shrines or warrior residences. Their purpose, in either case, was to concentrate campaigns and battles: to slow enemy advances, thwart raiding tactics, control selection of the battleground, restrict cavalry maneuver, and enhance the ability of foot soldiers (who could be recruited in much larger numbers) to compete with skilled horsemen. And they were expendable, as well as expedient; they were never the sites of sustained sieges or – by choice – of heroic final stands. Contingency planning normally provided for withdrawal and reestablishment of new defensive lines elsewhere.

Picture scrolls indicate that most of the defense features cataloged in the descriptions of Ichinotani were commonly deployed by the late thirteenth century, and most appear in descriptions of other Gempei War-era fortifications in Heike monogatari and its sister texts. Curiously, however, some of the simplest devices – brush barricades (sakamogi) and shield walls (kaidate) – cannot be corroborated in more reliable sources for the 1180s.

Shield walls were exactly what the name implies: rows of standing shields erected behind or on top of other defense works. Standing shields had been used as portable field fortifications since the ritsuryo era, and were also deployed as counter-fortifications by besieging armies. Kaidate were used on boats as well, to convert what were otherwise fishing vessels to warships.

Sakamogi (literally, “stacked wood”) appear to have been essentially piles or hedges of thorny branches placed in front of the principal defensive palisade. They served as an application of what is sometimes called “the principle of the curtain”: a light barrier designed to break the momentum of an enemy charge, dissipating its shock power and holding the enemy under fire before he can bring force against the main walls. Brush fences of this sort were architecturally simple, yet extremely effective for the task: Martin Brice notes that, during World War I, thorn enclosures, called boma or zareba, built by the Masai of Tanzania and Kenya proved as difficult to cross, and as resistant to high explosive bombardment, as barbed wire!

Masai thorn fences represented a wartime application of a device normally used to contain and protect livestock. Japanese sakamogi may have had similar origins. Such a military adaptation of a technology developed for animal control was entirely apropos for early medieval warriors, whose main concern was restricting the movement of enemy horsemen. Brush curtains are, however, vulnerable to fire, which, as we have seen, was a favorite weapon of early bushi.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.