Edo Castle: The Tower They Never Rebuilt

At the northern end of the Imperial Palace East Gardens, a great stone platform supports the sky.

It is the foundation of a castle keep, but not simply the surviving base of the tower destroyed in the Great Fire of Meireki in 1657. This foundation was prepared afterward, for a replacement. By 1658, the stonework was ready. The tower that would have completed it never rose.^[1]

The distinction is small enough to miss and large enough to change the story. Here is evidence of something people intended to build, began to build, and then decided they could do without.

For most of the Tokugawa period, the government of Japan worked without a keep above Edo Castle. Its absence did not turn the castle into a ruin. The busy place around it continued to receive officials, accommodate a household, and conduct affairs that reached far beyond the moats.

Today, much of that working world has disappeared too. Yet the two absences are not the same. The unbuilt tower and the vanished palace ask different questions: what made a monument dispensable, and what happened to the life that continued beside it?

Tokugawa Ieyasu entered Edo in 1590. He was not arriving at an empty shore. A castle associated with the fifteenth-century warrior Ōta Dōkan and a developing settlement already occupied the area. Ieyasu and his successors greatly enlarged both the stronghold and the political world organized around it.^[2]

The moat and masonry make that enlargement look rooted in the ground beneath them. Much of the work began elsewhere.

Along the Izu Peninsula and the neighboring coast, quarry sites preserve stones prepared for Edo Castle's early-seventeenth-century construction campaigns. Some still carry rows of wedge holes used to split the rock. The sites include working areas, places where blocks waited, hauling routes, and access to shipping. The surviving remains document an operation extending from a hillside to a waterfront and ultimately to the castle.^[3]

A finished wall brings these stages together so effectively that they disappear from view. A stone seems simply to belong where it stands. Before that could happen, someone had to judge where to split it, bring it out of difficult ground, and arrange its journey. Another person had to decide whether it would fit the work being done at the other end.

The great lords, or daimyō, participated in these shogunal building obligations. Archaeological study of the quarries therefore concerns the organization of labor and political service as well as stonecutting.^[3] The castle concentrated contributions from territories whose rulers maintained households and interests of their own.

This gives the massive defenses a second meaning. They were obstacles to an enemy, but also the accumulated result of relationships among people who were supposed to cooperate. Their solidity depended on arrangements that were much less visible than stone: orders, responsibilities, supplies, and the ability to coordinate work at a distance.

Looking across a moat, one sees the boundary of the castle. Following the making of the wall leads outward, into the country that helped sustain it.

The castle had not always lacked a tower. Successive early Tokugawa shoguns commissioned imposing keeps. The one built under Ieyasu's grandson Iemitsu was destroyed in 1657, during a disaster that spread through large parts of Edo.^[4]

Stone walls could make an army's approach difficult. They could not guarantee that fire would remain outside. The Meireki disaster involved several outbreaks driven by strong winds after a long dry spell. The castle belonged to the same vulnerable city as the streets around it.^[5]

The first response was not to announce that the age of great towers had ended. Work began on a replacement foundation under the fourth shogun, Ietsuna. Only after that investment did the undertaking stop.

The decision is traditionally associated with Hoshina Masayuki, Ietsuna's uncle and adviser, and with giving priority to rebuilding the city. The familiar explanation holds that the government no longer needed such a vast display of strength.^[6] It is an appealing story of a ruler's advisers distinguishing the useful from the spectacular.

But the empty foundation cannot disclose every calculation behind the decision. Nor does abandoning one expensive project prove that the whole regime had suddenly placed public welfare above prestige. What it establishes more securely is that the tower's loss could be accepted without abandoning the castle's principal work.

A palace had to provide usable rooms. Gates still had to control entry. Officials needed places to meet and transact business. A keep might strengthen the image of authority, but it was not the machinery by which every order became effective.

Even this judgment did not remain beyond reconsideration. A new proposal to build the tower emerged in 1712, but it too went unrealized.^[4] The long absence was not necessarily the execution of one permanent decision. Later people could reopen the question and still leave the same skyline behind.

That makes the platform more interesting than a monument to enlightened thrift. It is the remainder of an argument about priorities. Something once worth building on an extraordinary scale had become something whose replacement could repeatedly be postponed.

The shogunate still had power. It no longer needed to express all of it in the same form.

The broad lawn of the Honmaru, the castle's principal enclosure, encourages the eye to wander. A palace plan asks it to work differently.

One drawing in the National Archives records the arrangement of the Honmaru's official and domestic rooms, including their names, sizes, kinds of tatami, and the positions of wells.^[7] Those details recover a place that had to function at the scale of a doorway, a water supply, and a room prepared for a particular use.

The shogun's activities were distributed through this low, extensive complex. He met senior councillors in one room and conducted business in another. The middle part of the palace combined work and daily life; its arrangements changed with individual shoguns and with the positions of the people serving them.^[8]

A tower organizes a distant view around a single summit. Governing was less compact. A matter could pass through several people before reaching the ruler, and through several more before a decision had consequences outside the castle. The impressive height of a building tells us little about whether that passage was quick, accurate, or effective.

The people close to the shogun included attendants, physicians, and officials mediating between him and the senior council. Some offices appeared or disappeared over time. Their rooms and proximity reflected their responsibilities.^[8] This was neither an impersonal modern office nor merely a private home in which public business occasionally intruded.

The overlap carried practical consequences. A ruler needed care, meals, and a daily routine while others needed access to him. Those demands could be arranged together, but never reduced to the same thing. The person responsible for conveying a matter might become important without being the highest-ranking person who cared about its outcome.

Seen this way, the missing tower ceases to be the missing center of the government. Much of the center lay horizontally, in the relationships between rooms and the people entitled to enter them.

The lawn does not reproduce that world. It removes its partitions. What is easy to cross now was once divided into places where different kinds of conduct were expected. The freedom of the present view is itself one of the changes that has to be understood.

If the tower could remain absent, the palace demanded a different response when it burned.

After a fire destroyed the Honmaru Palace in 1844, rebuilding was completed in 1845. A surviving official's work diary records duties connected with reconstructing the inner quarters.^[7] Recovery was not simply the moment a shogun declared that his residence should return. It consisted of tasks assigned, work supervised, and decisions recorded.

The decoration has left another kind of evidence. Designs for the new palace's wall and sliding-door paintings survive in the National Archives. Among the subjects prepared by Kanō-school painters were views of Kyoto through the seasons, populated by people going about their lives or visiting familiar places.^[9]

Kyoto could therefore reappear inside a residence in Edo, even while the rooms themselves were being remade after a fire. The paintings offered a setting larger than the immediate demands of reconstruction. They brought cultivated associations, recognizable landscapes, and other kinds of human activity into the household.

Such work complicates a simple opposition between necessary repair and unnecessary beauty. The palace needed a roof, but those rebuilding it were not aiming merely to keep the rain off. They were restoring a way of receiving people and living among things that carried meaning.

That did not make every expenditure indispensable. It meant that the people commissioning the work had to decide what counted as an adequate palace, rather than an adequate shelter. Their answer depended on the institution the rooms were intended to sustain.

The designs also bring the makers into view. A finished painted room can look as though its imagery arrived naturally with its architecture. A preparatory drawing reveals planning: a subject chosen, a composition worked out, a wall considered before the finished image occupied it. The castle's recovery drew on practiced skills as well as administrative authority.

These nineteenth-century records stand at a great distance from the foundation begun in 1658. That distance is the point. The government did not make one decision about rebuilding and apply it to everything forever. A tower could remain a possibility, while rooms needed for present use were repeatedly made real.

In 1863, fire destroyed the Honmaru Palace again. This time it was not rebuilt. Government business continued in the Nishinomaru, the western part of the castle.^[6]

The contrast with 1844 is sharp. A similar kind of physical loss did not produce the same result. The shogunate would surrender the castle just five years later. The empty Honmaru belongs to that late period, not to the earlier judgment that a keep was expendable.

It would be tempting to make the unrepaired palace a perfect symbol of a government already dead. But offices still worked, people still negotiated, and the question of who would control the castle remained consequential. A damaged institution can continue to make decisions that matter enormously.

In 1868, Katsu Kaishū, acting for the Tokugawa side, negotiated with Saigō Takamori, a leading figure in the forces opposing it. The resulting surrender brought Edo Castle under the new regime without a battle for the castle itself.^[10] That achievement should not be confused with the end of the wider Boshin War, or with a claim that every dispute had been settled by two men's conversation.

The former shogunal seat became the Imperial Palace in the same year.^[11] A place developed for one ruling house now accommodated a different source of authority.

This was a substantial change, but not a fresh beginning on vacant land. The new occupants inherited an exceptionally important location, defenses, usable spaces, and a city already oriented toward the castle. The political reason for coming through its gates changed more quickly than the geography that made the gates significant.

There is an important difference between continuity of address and continuity of government. The new regime did not become Tokugawa rule under another name because it occupied Tokugawa ground. Yet it did not have to invent every practical condition of being there either.

The Honmaru's emptiness is therefore only one part of the ending. Nearby, the same larger site was acquiring another life. Loss and continued use occupied neighboring ground.

In 1871, an official named Ninagawa Noritane arranged for Edo Castle to be photographed. The photographer Yokoyama Matsusaburō and the painter Takahashi Yuichi helped produce an album of sixty-four images. Ninagawa's application explained the value of recording buildings before destruction made that impossible.^[12]

He was already thinking about people who would know the place differently from those living around it.

The album was a made object, not an unmediated window. Yokoyama took the photographs; Takahashi added color; Ninagawa supplied labels and an ordering of views. Photography, painting, and written identification worked together to make the castle intelligible.^[12] A hand-colored image is useful historical evidence, but its color is not a measurement of the light on the day the shutter was opened.

Nor did the camera arrive in time to capture the intact shogunal palace. The Honmaru had already burned. The album devotes considerable attention to its empty ground, recording the remains of a center whose principal buildings were gone.^[6]

That choice is revealing. A place did not need a complete building to be worth documenting. The relation between a wall, a gate, and an open area could preserve information that would become harder to recover after further changes.

The photographs also belong to the new imperial period. Calling them pictures of “old Edo Castle” can make them sound as though they were taken before the political transfer. They were not. They record a site already becoming something else, while some of the structures associated with its former life remained.

For a reader looking back, this is a useful discipline. A source can preserve something absent today without preserving everything that existed before it. A photograph has a date; so does the emptiness within it.

The album does not cancel the losses. It gives later viewers a way to ask more precise questions about them. That is a quieter achievement than rebuilding a tower, but one that has made a great deal of understanding possible.

The East Gardens opened to the public in 1968, following a project begun earlier in the decade. They occupy former castle compounds within the larger Imperial Palace site.^[11] Their present openness is a modern arrangement, not the survival of an old public park.

Some of what looks most timeless is also carefully made. The Ninomaru garden was reconstructed with reference to an eighteenth-century plan. Nearby woodland was planted to recall landscapes disappearing from the Tokyo region.^[13] Historical recovery and a modern concern for nature meet within the same garden, without being the same undertaking.

The gates carry similarly mixed histories. The large upper gate at Ōtemon, the castle's principal entrance, was rebuilt in 1967 after wartime destruction. A surviving roof ornament displayed nearby bears the date 1657 and belongs to an earlier structure.^[14] The visitor passes through reconstruction while encountering an old object that the reconstruction helps explain.

These differences do not make the place a collection of failed originals. They identify what kind of evidence each part can offer. A surviving stone, a rebuilt gate, a recovered garden plan, and a deliberately unbuilt tower all connect us to the past in different ways.

The East Gardens should also be distinguished from the working palace around them. Public paths do not imply that every neighboring enclosure is a museum waiting to be entered. The site continues to accommodate imperial and administrative functions as well as historical visiting.^[11]

At the keep foundation, however, the central puzzle remains available without entering any restricted room. A considerable investment in stone ends at an empty platform. The obvious next step was never taken.

The temptation is to supply the missing building in the imagination and call the scene complete. It is worth resisting for a moment. A reconstructed silhouette would help us picture a keep. The absence asks us to understand a decision.

Edo Castle endured because its occupants could change what they needed from it. They could lose a tower and continue governing, replace a palace and resume living, then transfer the whole site into another political order. None of those changes was painless or inevitable. Together, they explain why the center of Tokyo contains something more instructive than an intact monument to one ruler's ambition.

The foundation preserves the scale of what people meant to build. The sky above it preserves the fact that they did not.