Tagajo and Sendai

8 mins read

How forts grew into castles

At first glance Japanese castles appeared to have weathered the centuries unscathed, but looks can be deceptive. Here Stephen Turnbull contrasts Sendai Castle’s picture-book fragility with the rather tougher existence in the earlier fort of Tagajo.

The massive stone walls of Kumomoto Castle support and protect the delicate structure perched on top of them. [Image: Tohoku History Museum]

Japanese castle keeps are fantastic creations of white wood and blue tiles ornamented with gold, and look more like the fairytale illustrations of a child’s storybook than a fortress or stronghold. They sit perched high on formidable plain stone walls – the only means of comparison with, to use Shakespeare’s words, the ‘worm-eaten hold of ragged stone’ that characterises the typical European castle.

Yet appearances can be misleading, and the first common mistake made by a visitor to a Japanese castle is to assume that nothing is missing. Anyone used to visiting ruined European castles, where cold winds howl through open towers, may be forgiven for thinking that the striking appearance of a Japanese castle means that the edifice has survived in an almost intact state. Yet this is usually not the case. The famous multi-storeyed keeps of castles such as Himeji or Kumomoto represent the survival of just one small part of what was once an enormous defensive complex.

Taken as a whole, Japanese castles are often as ruined as European ones, because, being built predominantly of wood, they are no strangers to repeated destruction, demolition, and rebuilding. Fire, earthquakes, and tsunami have all taken their toll through the centuries; and so has war, although the demands of peace have tended to exert a stronger influence on the likely survival and appearance of most of today’s castles. When Japan entered the modern world in 1868, no place was found for the ancient castles, and demolishing them sent a local message of loyalty to the new regime.

Sendai – a late castle

Sendai Castle was once one of these glorious creations. Its massive stone bases survive, but little is left of the ornate superstructure that marked the administrative centre. It was built in 1600 by Date Masamune (1566-1636), who became the first lord, or daimyo, of the Sendai Domain at the beginning of the Edo period, when Japan was reunified under the first Tokugawa Shogun. It was Tokugawa Ieyasu who moved the capital of Japan to Edo, modern-day Tokyo (meaning ‘Eastern Capital’), although the Emperor continued to reside in Kyoto until the restoration of Imperial power in the 1860s.

The fortress

Sendai should be compared with the earlier fort of Tagajo, which has recently been extensively excavated. This belongs to a period nearly a millennium earlier when Japan was being united for the first time, and when forts were being constructed to protect it from the external enemies. Tagajo began life in AD 724 as a rough timber outpost, but evolved into what was effectively a fortified palace that attempted to control all of Mutsu province, half of northeastern Japan. It was originally known as the ‘pacification headquarters’ (Chinjufu), a revealing term that indicates why Tagajo was built.

Tagajo derived its existence from the Yamato rulers, the earliest Japanese emperors, who used castles and other fortifications as a response to threats to their rule. In the south of Japan, the fear of invasion from continental east Asia led to the building of stone castles designed by cooperative Korean refugees; but the Yamato rulers also faced a military threat in the north-east of Japan, where the strongest resistance to their assumption of control was to be found in the distant territories that the Yamato called the provinces of Dewa and Mutsu. This was the land of the barbarians, whose savage ways precluded them from accepting the benevolent rule of the Yamato court. These truculent enemies were referred to as emishi, a term that does little to identify their racial origin, a matter that is still controversial to this day. They may have been identical to the Ainu, who still live on in Hokkaido, but whatever their racial origins, there is a clear implication that the emishi were regarded as being ‘beyond the pale’ of Yamato civilisation. Instead, they were despised as blood-drinking, hole-dwelling savages by the Yamato rulers. In the written records of the Yamato, a great deal of contempt is shown for them in every field except their military skills. In AD 724, for example, a revolt took eight months to put down.

Yamato efforts to bring the emishi under their control began in the latter half of the 7th century AD, and accelerated following the creation of Nara as Japan’s first permanent capital city in AD 710. In a statement of optimism, the designated provinces of Mutsu and Dewa were divided into administrative districts, and a long programme of conquest began. Its success was hindered by the policy of the Imperial court of appointing submissive emishi chiefs as local rulers, a good idea that backfired when the process served to increase the chiefs’ prestige and, as a direct consequence, their sense of independence and resistance. Crucial to this pacification process was the establishment of timber fortifications such as those at Tagajo.

Tagajo was originally a wooden saku, which was often read as -ki and is usually translated as stockade or fort. To refer to it in English as ‘Fort Taga’ conjures up the appropriate image of a frontier fort built from earth and wood, and the pictorial character itself resembles a wooden palisade.

Tagajo was never simply a military base. Instead it was a centre of Imperial state admin-istration that acted as the outpost of centralised rule in enemy territory. As both fort and office, Tagajo did not merely administer a territory, but actively facilitated its creation. Settler families were transplanted to Dewa and Mutsu to work the lands around places like Tagajo, which provided them with security in case of emishi attack. A network of smaller stockades linked the major bases.

Grand pacification

As a major seat of regional government, Tagajo’s outer walls sheltered a complex of administrative buildings. Unlike most of the northern forts, it was built on a hill, but otherwise it followed the usual practice of a long outer perimeter wall, set some considerable distance from the civilian buildings within. Was the large space between the walls and the administrative buildings intended as a refuge for the surrounding farmers and their flocks in case of attack?

The main means of construction of the walls was rammed earth above a shallow foundation of undressed stone. The soil, dug from a depth sufficient to guarantee that no seeds would be present, was carried to the site in a bag slung from a bamboo pole resting on the shoulders of two men. Here it would be mixed with water and carried up ladders to the latest layer. The overall shape of the wall, which was of trape-zoidal cross section to give stability, was marked out by a series of scaffolding poles. When the mix was tipped in, wooden shuttering held it in place, just like modern concrete; but instead of simply waiting for it to set, each layer was rammed down by a team of men. The presence of sand, and perhaps shells, in the mix produced a very strong wall when it dried out.

When one length of wall was complete, the next along was begun, and the joins can still be discerned on excavated sections. The final stage was to give it a coating of plaster, and add a tiled roof to weatherproof it. The tiles were laid across wooden trusses set at intervals along the wall. When done with precision, the resulting construction was both neat and attractive, with clean rectangular lines.

Some other buildings within the compound were made in a similar way, while others made more use of wood. For the latter, vertical posts resting on foundation stones gave the building its overall shape; lath and plaster walls were added between the verticals, with a more extensive tiled roof. So, for example, the innermost part of Akita castle – the seicho (government office) – consisted of five buildings within a precise rectangular courtyard, arranged like a ‘five’ in dominoes. The building in the middle was called the seiden. The seicho lay in the middle of a wide area enclosing 550m² and encircled by a defensive wall of beaten earth 2.1m high. Here no attempt was made to keep to a strict rectangular shape. There were two right-angled corners, but otherwise the wall was curved and followed the contours of the landscape.

Tagajo housed perhaps the most impressive seiden of all. As befitted the major fortress in northern Japan, the headquarters building of Tagajo was a splendid rectangular courtyard with gates at the four points of the compass and a fine two-storeyed building at its centre. Another office building lay just behind it. Several sections of Tagajo’s very long perimeter wall have survived, and show it to be of rammed-earth construction, although use was also made at places along its great length of planking walls. Taga also sported fighting towers that straddled the walls. These simple openwork lookout towers consisted of a plat-form built over the length of the tiled wall, held in place by large vertical posts on either side of the wall. Access to the fighting platform was by means of a trapdoor and a ladder from the inside of the wall. Examples of similar towers have been reconstructed at the site of Shiwa castle in Morioka. They are arranged at intervals of about 50m along the wall on either side of the main gate.

Tagajo finally became redundant when the emishi problem was solved, and wars with them were replaced by wars between rival families of settlers who had grown rich and powerful through Imperial service in distant lands. These were the first Samurai clans, whose challenge to Imperial supremacy was to change the political landscape of Japan, and – through the creation of massive stone and wood fortresses like Sendai – change its physical appearance as well.

Digging Tagajo

About 50 separate excavations have been carried out at Tagajo Fort, uncovering large quantities of artefacts – including the distinctive grey stoneware roof tiles that are found in abundance on most important sites of the period. The special historic significance of the site was recognised in 1984, with the level of designation being raised. The site is now the location of the Tohoku Historical Museum, where many of the artefacts are displayed alongside excellent reconstructions of ‘Fort Taga’. Of great importance are the extensive ceramic assemblages, which provide evidence for a detailed chronological sequence, and remains of documents recording many everyday details about life at Tagajo. These include lacquered paper documents bearing inscriptions, neat brushstrokes in black ink, and over 370 wooden tags or mokkan, which contain a wealth of information about the supplies sent to the fort from all around the Japanese empire, as well as spells and curses, indicating the rich spiritual life of the soldiers and administrators based there.

Past tribulations

Earthquakes are nothing new in the area around Fort Tagajo. Historical records describe a massive quake in the mid 9th century, possibly one of the largest to strike Japan in the millennium prior to this year’s disaster. At Sendai Castle too, earthquakes were a persistent problem. J F Morris, a scholar based at Miyagi Gakuin University, lists five major earthquakes between 1616 and 1710 that caused structural damage to the castle of the Date clan; and a further six prior to the restoration of the Meiji Emperor in 1868 and the end of the daimyo system – ending with it the power of the Date family.

Sendai Castle also suffered man-made horror: the buildings of the Second Enceinte burned down in 1883, and Allied bombing in the closing stages of the Second World War destroyed the few remaining gates. So it is that the present-day visitor to Sendai Castle is confronted only by some reconstructed stone walls, some of which are capped by plastered and tiled reconstructed upperworks, and some of which are badly damaged following the earthquake of March this year.

This article appeared in issue 49 of Current World ArchaeologyClick here for more information about subscribing to the magazine.

Leave a Reply

Your email address will not be published.