Asuka and Fujiwara: The Work of Making a Capital

Before Fujiwara Palace could have its great ceremonial buildings, a waterway running through their future ground had to disappear. Archaeologists traced a north–south canal beneath the areas later occupied by the Great Audience Hall and imperial residence. It belonged to the work of making the palace: a means of moving construction materials that was filled before the structures above it were completed.^[1]

The canal gives the finished plan a missing dimension—sequence. A route useful to builders could be incompatible with the building they were preparing. The palace could not be brought into existence simply by enlarging every useful feature already on the site. Some arrangements had to be created for a task and then removed.

That problem belongs to the larger history of Asuka and Fujiwara. Across the late sixth and seventh centuries, palaces, temples, engineered water facilities, and elite tombs accumulated in this part of the Nara basin. Fujiwara, occupied by the court from 694 to 710, brought government into a more extensive planned urban setting. The change was substantial, but it did not begin on the day the ruler moved. Its earlier stages survive across a landscape that no single palace outline can contain.^[2][8]

The Asuka Palace Site represents successive court compounds, rather than one building continuously occupied under different names. Archaeological interpretation distinguishes places for residence, ceremony, and government business. Offices were distributed inside and outside the palace. Political activity already had specialized settings, but those settings were not yet gathered into the kind of encompassing arrangement associated with Fujiwara.^[2]

It is worth resisting the picture of an unorganized countryside suddenly corrected by a city plan. A dispersed institution can still have order. People can know which residence, hall, or precinct they must approach without all of them standing inside the same boundary. The question is how that knowledge is maintained, and how easily it can accommodate a larger concentration of activities.

The surviving palace traces show that the choice was not simply between a household and a state. Residence and government shared a setting, while some official functions extended beyond it. Moving the ruler therefore raised a practical question: which of the surrounding relationships would move too, and which could continue in existing places?

This makes the wider Asuka landscape indispensable. A temple could retain a consecrated center; a water installation depended on a particular engineered arrangement; a burial remained attached to its ground. They were not interchangeable pieces waiting to be lifted into a new rectangle. Each had a different reason to be where it was.

Fujiwara's significance becomes clearer against that background. The later city imposed a more comprehensive spatial arrangement on court life, but it drew upon practices and skills already developed in multiple settings. To understand what its builders could bring together, one must begin beyond the new palace enclosure.

Asuka-dera was one such setting. Its construction began in 588, and its principal halls and pagoda were completed in 596, according to the historical account used by its excavators. The temple was supported by Soga no Umako, a leading figure at court. Building it brought Buddhist worship into an ambitious architectural establishment rather than leaving it dependent on small, isolated settings.^[3][7]

Excavation established an unusual principal arrangement: a pagoda with main halls to its north, east, and west. The report connects features of its platforms, roof tiles, and kiln with traditions on the Korean peninsula. This material evidence complements the written account of craftsmen arriving from Baekje, one of the peninsula's kingdoms, to guide construction. The plan belongs to a largely vanished complex, not a surviving sixth-century ensemble.^[3]

A building of this kind embodied several forms of knowledge at once. A patron could support the undertaking, but an intention to build did not supply the methods for producing and assembling its parts. Tile manufacture required a material and firing process; a platform required a construction method; the arrangement of sacred buildings followed a conception of what the precinct was for. Religious establishment and technical exchange were joined in the same project.

That conjunction does not make the temple merely a rehearsal for secular government. Its buildings had Buddhist purposes, and the pagoda's central place belonged to the organization of a sacred precinct. The political relevance lies in how a court-connected patron could sustain such an institution and draw upon expertise reaching beyond the archipelago. Imported skills were put to work at a particular site, rather than remaining knowledge possessed elsewhere.

The contrast with Fujiwara is instructive without requiring an unbroken line of named craftsmen between them. By the time a much larger palace demanded tiled monumental buildings, tile-making and the construction of complex precincts already had a history in the region. Fujiwara's builders faced a problem of scale and organization, not the first discovery that monumental architecture required specialist work.

Another part of the court's developing order stood near Asuka-dera. At the Mizuochi Site, excavators found a strongly constructed platform, water conduits, copper piping, and remnants of lacquered wooden boxes. They identified the complex as the water clock associated in the Nihon shoki, an eighth-century court chronicle, with Prince Naka no Ōe and the year 660. The dating of the excavated pottery is compatible with that seventh-century context.^[4]

The identification does not rest on a surviving clock face. It brings an unusual hydraulic installation together with a written account. In the reconstruction, water supplied a receiving vessel whose changing level could indicate elapsed time. The excavators also examined the proposed mechanism using hydraulic analysis. Such work tests whether a reconstruction is plausible; it does not supply a recorded operating performance for the ancient apparatus.^[4]

A shared time signal offered a different kind of coordination from a road or a gate. A road helped people reach a place; a timekeeping system could help them act at a common moment. That is the practical implication of the installation, not evidence that every activity in Asuka followed a uniform daily timetable. The court was developing means to organize both where things happened and when they should happen.

Water also occupied less easily explained settings. At Sakafuneishi, a carved stone has channels and basins whose exact purpose remains disputed. The wider site includes other stone and water features, and it has been interpreted in relation to royal ritual. The village's cultural-property account does not regard any proposed use of the famous carved stone as conclusively established.^[5]

These sites should not be collapsed into one story of hydraulic control. Measuring time and conducting a ritual would use water for different ends; in the second case, even the precise activity is uncertain. Together they show how much of Asuka's institutional landscape lay beyond the ruler's residence. Government was acquiring equipment and settings as well as buildings for assemblies.

Empress Jitō transferred the court to Fujiwara in 694, carrying forward a project associated with Emperor Tenmu, her husband and predecessor.^[13] The palace occupied an enclosure roughly a kilometre square within a much larger rectilinear capital. The city's broad extent, approximately five kilometres in each direction, should not be mistaken for the dimensions of the palace itself.^[2][12]

The distinction between those two scales is the heart of the change. A palace plan arranged the ruler's residence, ceremonial spaces, and government offices. A city plan extended the ordering effort beyond that enclosure to streets and places where the capital's inhabitants lived. The court was being situated within an urban framework, rather than treated only as a residence with activities gathered around it.

The two named capitals were not simply separate patches on the ground. A reconstruction based on excavated streets places the old central Asuka district within the southern part of Fujiwara's broader planned extent. Moving the court could therefore change the organizing center while incorporating ground already associated with earlier palaces.^[12] The new layout gave roads, institutions, and residences common spatial reference points. It does not follow that every part of the planned area was equally built up, but the scale of the intended arrangement extended well beyond the new palace walls.

Nor did drawing a rectangle settle the sequence of construction. The canal beneath the future central buildings makes that plain. The geometry of the eventual capital and the routes that made its construction possible belonged to different stages. A plan could organize the outcome while builders used arrangements that the outcome would erase.

That difference prevents the surviving open ground from becoming a falsely instantaneous picture. The palace was made through successive operations. A monumental center required access for materials before it could provide a setting for government, and the work of supplying those materials reached considerably farther than the capital's boundaries.

Roof tiles provide unusually concrete evidence for that wider effort. In her account of Fujiwara's construction, archaeologist Ishida Yukiko describes production drawing on existing kilns outside the Nara basin as well as new facilities within it. The capital's builders could make use of established connections while developing production closer to the project. The two need not have belonged to neatly successive stages.^[1]

A closer study of tiles from the construction canal tested precisely that sequence. At excavation 20, Ishida distinguished material in the canal's lower deposits from material in the layers that filled and levelled it. The lower group belonged to the period when the waterway was still functioning. Associated wooden inscriptions placed its use in the later part of Tenmu's reign. Among the tiles were products linked to different production groups, including examples made by forming clay coils around a mould.^[6]

The manufacturing detail matters because techniques had been used to arrange the tiles into a chronology. If a method thought to belong to a later production phase occurs in an early canal deposit, the sequence needs reconsideration. Ishida's analysis supports the presence of the newer basin production alongside established supplies relatively early in the undertaking. The canal context supplied a check independent of simply lining up tile styles.^[6]

For the history of the capital, the consequence is more useful than a revised list of tile types. New production capacity was not necessarily an afterthought added when the old supply proved insufficient. At least part of it was already participating while the central complex was being made. The building project drew together sources of material that were geographically separate and technically varied.

None of this yields a contractor's ledger, a total labour force, or a daily tally of loads. It does establish a relationship between production and construction sequence. An order to roof a palace became a coordination problem extending from kilns to a temporary waterway and finally to buildings whose completion required that waterway to be closed. The broken tiles recover a portion of that coordination which the finished architecture would have concealed.

Fujiwara also preserves evidence that the organization occupying the new buildings was still changing. Wooden tablets recovered around its northern palace boundary name offices whose titles differ from those established under the Taihō administrative code of 701. The excavators identify examples concerned with gardens and with check stations. These are traces of administrative arrangements that cannot simply be read backwards from the later code.^[8]

The same publication describes tablets using an earlier written designation for local districts. Here the significance of the documents differs from the daily transactions preserved at Heijō. They allow changes in administrative vocabulary to be studied close to the time when those terms were used, rather than relying entirely on retrospective court histories.^[8]

The chronology is revealing. The court had moved in 694; the Taihō code belongs to 701. The capital did not wait for every rule and office title of that later arrangement to exist before beginning to function. Conversely, the move did not freeze the bureaucracy into its final form. Construction, occupation, and legal reorganization overlapped without being one event.

This is why the label ritsuryō needs a little unpacking. It refers to government organized through penal and administrative codes, not to a particular roofline or street pattern. Law and planned space could support each other, but evidence for one does not automatically prove the complete operation of the other. A regular city could contain institutions undergoing revision; an older office name could survive on a document discarded in a newer urban setting.

Fujiwara's importance is consequently not diminished by the signs of change within it. Those signs are part of what makes the place historically informative. The palace was a working setting in which the relation between the sovereign, offices, and the wider realm continued to be organized. Its sixteen years of occupation were long enough to contain more than one administrative moment.

Beyond the government buildings, the burial landscape presents another relationship between place and order. Kitora, in the tomb area south of Fujiwara, is generally dated to the late seventh or early eighth century. Its small stone chamber contained a lacquered coffin, grave goods, and elaborate paintings. The surviving objects indicate a high-status burial; they do not provide a securely established personal name.^[9]

On its ceiling, gold-foil stars joined by red lines formed a Chinese-style star chart. The painted circles included representations of the celestial equator and the ecliptic—the sun's apparent annual path across the sky. The research institute notes an error in the ecliptic's position. The image is therefore evidence for the use of a sophisticated astronomical scheme, not a perfectly accurate observational chart produced at the tomb.^[10]

Its setting changes the meaning of that knowledge. In a burial chamber, the heavens belonged to the treatment of the dead. The chart cannot be treated as a surviving instrument from the court's timekeeping office. Yet it shows that ordered conceptions of the cosmos reached into elite funerary practice, just as other imported forms had entered temple construction and the measurement of time.

Nearby Takamatsuzuka offers a related but distinct arrangement. Its chamber paintings include groups of men and women as well as directional guardian figures and stars. The occupant is not securely identified. The human figures belong to a designed funerary image, not a documentary portrait of named court attendants. The paintings were removed with the chamber stones for conservation; their historical meaning must not depend on imagining an intact, accessible ancient room.^[11]

These tombs cannot tell us the private beliefs of every person involved in building the capital. They establish something more limited and tangible: elite commemoration drew on ordered images that joined human position to a larger scheme of space and the heavens. Fujiwara's administrative enclosure was only one setting in a landscape where such forms of order were being made material.

In 710 the court moved to Heijō, at Nara. That date closes Fujiwara's life as the seat of government, but it is a poor measure of how much happened there. A capital's duration and the effort required to create it are different quantities. So are the longevity of a building and the usefulness of the arrangements developed around it.^[8]

Across Asuka and Fujiwara, the record is unusually varied. A temple established a sacred center and brought specialist construction into a court-connected institution. The water-clock complex gave a developing timekeeping practice a fixed installation. Fujiwara joined palace and city planning to a larger material-supply effort, while the offices working there continued to change. The burial sites placed the dead within still other schemes of order.

These connections do not make every site a component of a single master plan. They explain why the landscape must be read in several places and at several dates. The court's capacity to create Fujiwara grew amid institutions with their own purposes, and moving the government did not make those purposes identical.

The clearest evidence of what changed is therefore a set of relationships: between craftsmen and a religious foundation, between water and a common time, between separate kilns and a concentrated construction project, between a city's buildings and its changing offices. Fujiwara brought more of those activities into a coordinated urban undertaking. It did not abolish their different histories, and its abandonment did not undo the fact that a new urban seat of government had been brought into use.

The court's next move belonged to that history of revision. Fujiwara had provided a setting large enough to bring government, residence, and an urban plan together while the institutions inside it were still changing. Its importance does not require every building to have been finished or every intended block inhabited. The surviving sequence is sufficient: an extensive urban government center was constructed, put to use, and reorganized within the short period that a list of capitals reduces to 694–710.