Nagaoka Palace: Government Before Completion

A narrow wooden tablet from Nagaoka-kyō requests food for scribes. The amount is four shō, a measure of volume. The writer gives a day in the tenth month and the name Yakumo. There is no account of imperial policy, no description of a palace, and no warning that the capital will soon be abandoned.^[1]

The request survives from an office serving the Council of State. It belongs to the ordinary business of keeping people at work in the capital between 784 and 794, when the court occupied Nagaoka before moving to Heian-kyō. That working establishment had ceremonial halls, stores, residences, and supporting offices. It also kept changing while officials were using it.^[2]

The excavated palace core, now within the city of Mukō, makes that overlap unusually clear. Government did not wait for a final plan to be completed. It occupied usable buildings, shifted activities into new ones, and eventually made arrangements to continue while the palace was being dismantled. The ten-year capital is a history of successive accommodations, rather than one construction project followed by one departure.

The move began by changing where resources went. In 784, shortly after appointing officials to build Nagaoka, the government ordered that the year's chō and yō taxes, together with supplies for construction workers, should be sent to the new palace. These were obligations delivered in goods or in substitution for labor, not simply money deposited in a treasury.^[3]

The destination mattered before all the buildings existed. A new palace required materials to build with and food for people working there. Changing the delivery address began drawing those resources toward Nagaoka while the court was still moving out of its previous arrangements. The order made the project a destination in the country's administrative life, as well as a place on which construction had begun.

The palace was only one part of the intended capital. Its ceremonial and governmental compounds, together with the emperor's residence, formed a core enclosed by substantial earthen walls. The broader city included other offices, households, and commercial activity. The palace core lies within modern Mukō; it should not be confused with the boundaries of today's neighboring city of Nagaokakyō.^[4]

Reused buildings helped the new establishment become usable. The local museum describes structures dismantled at Naniwa, in the Osaka area, and brought along the Yodo River for reassembly at Nagaoka.^[4] The choice of a connected site and the availability of existing architecture worked together. Transport could move construction material, while dismantling preserved some of the investment already made in it.

This did not make the entire undertaking instantaneous. Transferred buildings still had to take their places within newly prepared compounds, and the government's different activities required more than a single finished hall. What it offered was a way to establish important parts of a working center without waiting for every component to be produced anew.

The emperor moved to Nagaoka in 784. A New Year audience followed at the Daigokuden, the principal ceremonial hall, in 785.^[5] The sequence puts the capital into use early. It does not provide a completion certificate for everything behind its walls.

The ceremonial core joined the Daigokuden to the Chōdōin, a compound of halls and courts used for official assembly. Its arrangement is known through excavations spread across the modern town. In 1955, historical geographer and teacher Nakayama Shūichi, working with his former pupil Sodeoka Masakiyo, found the compound's south gate after following a report of cut stone discovered during railway-station construction.^[6]

Later investigations established more of the surrounding halls. The discovery of the western fourth hall in 1982 was particularly important: it showed that this compound's arrangement and scale followed the later Naniwa palace, rather than the ceremonial complexes at Nara's Heijō palace or at Heian.^[6] Nagaoka's place between those two capitals in a chronology had not dictated its architectural model.

The material transfer described in the museum's history consequently has a spatial counterpart. Buildings brought from elsewhere belonged to a composition with precedents elsewhere. At the new site, inherited forms helped provide a recognizable setting in which the court could gather. Reuse involved an arrangement for activities, not merely a supply of timber.

The foreground of the principal hall adds a smaller piece of evidence about that use. Excavation in 1997 found substantial postholes interpreted as supports for ceremonial banners. Their regular spacing and subsidiary post traces supported the identification. The national heritage account relates them to New Year audiences at Nagaoka, where no imperial accession ceremony took place.^[7]

A ceremony therefore required preparation in the ground as well as a hall above it. Posts needed supports; a line of banners required an ordered position in relation to the building. The preserved holes recover part of that practical setting without giving us the faces of the people assembled before it.

Nor is the broad platform an intact eighth-century building. The Daigokuden's foundation stones were lost; traces of paving and tuff helped establish the dimensions of its platform.^[7] The remains locate the ceremonial center through what was left below and around it. From there, the less monumental office tablet leads into another part of the same working capital.

The food request came from the eastern city, outside the palace's main enclosure. Excavated wooden documents identify the establishment there as the kitchen office attached to the Council of State. Its work extended to receiving, storing, and accounting for revenues that supported the council.^[8]

Mukō's historical account describes rice and salt arriving as rents from land. Clerks checked the goods against wooden labels and recorded them. Requests for scribes' food formed part of the same documentary activity. The surviving tablets show an office concerned both with what came in and with what should be issued for people doing its work.^[8]

The little request signed Yakumo belongs to this second movement. It identifies a purpose—food for scribes—and an amount. The date and name make it something more usable than an anonymous statement that supplies are needed. A recipient could recognize the request as one made on a particular occasion. Whether this individual request was granted is not recorded on the tablet.

A second tablet, dated in 789, asks for four shō for scribes and again carries the name Yakumo. Its transcription also preserves repeated numerals, units, and fragments of formal expressions; the database classifies the piece as both a document and writing practice.^[9] Alongside a request for supplies, the wood retains work on the written forms through which official business was conducted. The record does not establish who practiced on it or whether the practice preceded the request.

This also gives the palace a reach beyond its enclosing wall. The council's work depended on an establishment in the surrounding city that could receive and account for supplies. A diagram containing only the grand halls would omit an operating part of government. The connection is demonstrable through the office's name and documents, rather than an assumption that every nearby building served the palace.

The work left an unexpectedly intimate record. Official writing that once concerned a modest quantity of food now reveals a continuing establishment from which the buildings themselves have disappeared. Its value lies less in making a general point that officials needed to eat than in showing how that need entered an identifiable office's paperwork, on repeated occasions, at a particular city address.

Several years after the move, the palace was still unfinished. An edict of 788, reproduced in Mukō's chronology, acknowledged both continuing construction and the burdens of requisition imposed on the population. Tax relief accompanied that acknowledgment.^[10] The government was operating from a place whose further making continued to require resources from outside it.

The following year brought a substantial change within the palace. Emperor Kammu moved from the western residence to a new eastern one. A wooden document identifies an office concerned with constructing the Eastern Great Palace, and the chronology connects this work with the new residence occupied in 789.^[11]

Archaeology supplies the other side of that record. Investigations beginning in 1966 found corners of the residence's enclosing wall and covered corridor. From these discoveries, excavators located the central hall, a large building supported on posts set into the ground. Its configuration and roof-tile evidence led them to identify it with the eastern residence recorded for 789.^[11]

Finding the enclosure before the principal building mattered to that identification. A substantial hall by itself could have served several kinds of establishment. The surrounding boundaries placed it within a controlled residential compound. Its position, form, and associated objects could then be considered together with the written record of the emperor's move.

The result changes how a reconstructed plan should be read. The eastern residence was not necessarily the place in which the emperor had lived throughout Nagaoka's decade. The capital had acquired a different domestic center while retaining its name. A single finished-looking plan can obscure this sequence by presenting facilities from different stages as though every relationship had been settled in 784.

Nor did the new residence remove the need for other premises. Ceremonial assembly, council business, and the emperor's household were related activities, but they were not interchangeable. A usable audience hall did not make every household requirement complete. The second residence shows the palace being developed around activities that had already begun, with government continuing during that development.

East of the emperor's residence, another group of discoveries identifies the arrangements for the crown prince. During investigations in 1996–1998, archaeologists recovered discarded material carrying the name Tōgūbō, the office supporting the heir's official and everyday life. The excavators associate this establishment with Prince Ate, Kammu's son and the later Emperor Heizei.^[12]

A crown prince on a genealogical chart appears as one person waiting in a line of succession. The finds give that position a working establishment. Inscribed material names offices concerned with food and craft production. Amber, tortoiseshell, and ivory offcuts indicate the manufacture of decorated objects. The assemblage places skilled work within the provision of the prince's household, rather than leaving it as an invisible service supplied from nowhere.^[12]

Written material served several purposes there. One small wooden label reads tokudo-mon, documents concerning Buddhist ordination, and carries a date in 792.^[13] Its brevity is the point. The label could identify a category of papers without reproducing all their contents. It preserves part of an arrangement for keeping documents usable, not the full business recorded in the missing papers.

The distinction between a label and a request becomes revealing when this piece is set beside Yakumo's food tablet. One named a body of documentation; the other asked for an issue of supplies. Writing supported different operations within the palace's associated establishments. The surviving words are short because they once worked inside contexts in which their readers already knew much more.

The crown prince's office must also be kept separate from the emperor's Eastern Palace. Their Japanese names can look confusingly similar in discussions of the site, but they refer here to different establishments. Archaeology places the heir's support office east of the emperor's residence, rather than identifying the same building twice.

Together, the second imperial residence and the prince's office show why “the court moved” is a compressed description. The imperial household contained more than one person with institutional requirements. Accommodating it involved premises, supplies, objects, and documents that had to function in relation to one another.

The decade's best-known political rupture came early. Fujiwara no Tanetsugu, an official directing construction, was assassinated in 785. In the ensuing crisis, Crown Prince Sawara, Kammu's brother, was deposed and confined. He refused food and died while being sent away. Ate subsequently became crown prince.^[14]

These events affected succession at a capital still under construction. They also acquired a religious aftermath. The city's chronology records that divination in 792 attributed the illness of the new crown prince to Sawara's angry spirit. The same year brought flooding and damage to government buildings.^[14] The attribution belongs to the court's recorded explanation of illness; it is not a supernatural finding adopted here.

The sequence deserves attention because several years of institutional development lie between the assassination and the eventual move. The council's office had been put into use, the eastern residence had been occupied, and the heir's establishment was handling its own affairs. A narrative that moves directly from Tanetsugu's death to abandonment removes much of what the archaeology is recovering.

The same evidence prevents the ongoing work from becoming proof that all was well. The acknowledgment of requisition burdens in 788 and the later records of illness and flooding belong to the capital's active years. Building and governing could continue under pressures that the continuation itself did not resolve.

Why the court ultimately chose another site remains a question with more than one proposed explanation. The local museum presents political crisis, beliefs concerning Sawara, and flooding among the relevant accounts, rather than identifying one demonstrated cause.^[4] Their presence need not be erased to recover Nagaoka's working life. They need to be placed within it.

For the palace, the consequential next step was practical as well as political. Leaving required new accommodation while existing structures were taken down. The court could not simply cease to be a household or a government for the duration of another building project.

In 793, Kammu moved to an establishment called the Tōin so that work could proceed on dismantling the palace. Construction at the next capital was under way that year. The markets moved in 794, before the emperor's transfer to Heian-kyō.^[15] Departure, like arrival, consisted of stages affecting different parts of the city.

The temporary residence is particularly instructive. The court could give up its principal buildings while continuing to occupy accommodation in the capital it was leaving. Dismantling and residence overlapped. The date of the final imperial move therefore cannot stand for the day on which every building ceased to function or every person changed address.

Identifying those establishments on the ground has required further distinctions. A large detached-palace complex excavated east of the main residence in 1991 was initially associated with the Tōin. Its scale and position made that interpretation plausible. Later, pottery inscribed “Tōin” was found at another detached-palace site. The archaeological center now regards the first identification as unlikely; the excavated complex remains a palace establishment whose exact historical name is unresolved.^[16]

This revision is productive because it leaves substantial knowledge intact. A large complex was found, its arrangement investigated, and its character recognized. What changed was the name attached to it. The city's short occupation did not mean that it contained only one establishment large enough to be mistaken for the recorded temporary residence.

The scattered remains consequently preserve a more active history than one abandoned plan. Ceremonial halls established an early place of assembly. A council office handled incoming revenues and repeated requests for food. A new imperial residence changed the household's position; the heir had a supporting establishment of his own. Finally, another residence allowed the court to remain while dismantling began.

The little food tablet belongs within that sequence. It was written because there was work to continue on a particular day, not because the capital had reached a final architectural state. Nagaoka's government occupied arrangements that were usable while others were still being made—and arranged to leave before all its existing premises had stopped being useful. Its decade was short, but it contained more than one version of a capital at work.