Skip to content
Topics

Search History Omakase

Start typing to search.

    ← Back

    HO Guides

    Chihaya-Akasaka: The Army Below the Ridge

    Wound reports and a medieval siege narrative reveal the different tasks hidden inside an attempt to capture a mountain stronghold.

    Opening

    A report submitted after fighting at Chihaya in 1333 lists wounded men from the force of Kumagai Naotsune. He was serving in the army attacking the castle. Marks beside names record the checking of claims; the character meaning “deep” identifies the severity of wounds. Two officials endorsed the document after examining the injuries. One entry, reporting a blow from a stone, has no confirming mark.[1]

    This is a small, uncomfortable view of a famous siege. The men appear as people who had been hurt and whose injuries required recognition, rather than as an anonymous mass beneath the stronghold of Kusunoki Masashige. An assault had ended. Its consequences still had to be handled.

    Chihaya's resistance is usually approached from above, through the defenders and the steep ground they occupied. The reports invite a view from below. An attacking army had to get men close enough to fight, decide what to do when direct attacks failed, and remain in place while events elsewhere continued. The ridge did not confront it with one problem. It created a succession of them.

    1. Three strongholds, more than one campaign

    Masashige's resistance began before the siege of Chihaya. In 1331, he supported Emperor Go-Daigo's challenge to the Kamakura shogunate, the warrior government based far to the east. Go-Daigo occupied Kasagi, southeast of Kyoto, while Masashige raised resistance at Akasaka in the foothills of Mount Kongō. The imperial position at Kasagi fell, and the emperor was captured. The lower Akasaka stronghold also fell.[2]

    That setback belongs to the place as much as Chihaya's later endurance. The surviving historic area includes Shimo-Akasaka, or Lower Akasaka; Kami-Akasaka, Upper Akasaka; and Chihaya farther into the mountains. They are separate ridge positions, not successive courtyards inside one castle. Their differing locations allowed resistance to recur in the same region without requiring the same enclosure to hold throughout.

    At Shimo-Akasaka, the castle features are now faint. The village's account locates the first rising here, but the surviving ground does not offer a complete plan of its hurriedly prepared defences. Kami-Akasaka retains more extensive earthworks. Chihaya occupies a higher mountain position. Reading the three together gives the conflict room to develop: a position could be lost while resistance continued elsewhere nearby.[3]

    The surrounding settlement also mattered. Excavation during construction of the village's Kusunoki Hall in 1990–1992 revealed buildings within a double-moated compound. It is one candidate for a Kusunoki residence, not an authenticated birthplace. The remains nevertheless put a substantial enclosed residence into the landscape below the mountain strongholds.[4] The castle hills were part of inhabited country, rather than detached refuges whose entire history began when an army arrived.

    By late 1332, the government was issuing new summonses against Masashige and Go-Daigo's son, Prince Moriyoshi. The Kumagai family's records place its men in fighting at Akasaka and then Chihaya in 1333.[5] The renewed campaign thus reached a district where earlier fighting, settlement, and several defensible positions already overlapped. Taking one position would not necessarily finish the work of suppressing resistance in the area.

    2. Where numbers had to become an approach

    Chihaya stands on a ridge branching westward from Mount Kongō, with deep valleys beside it. The village's archaeological guide identifies a connection behind the castle towards the higher mountain and the Yamato side, in the direction of present-day Gose and Gojō. That geographical relationship is important: the ridge was part of a mountain landscape extending beyond the settlement immediately below.[6]

    The surviving fortification is principally shaped ground. Levelled terraces, known as kuruwa, provided usable surfaces on the ridge. Slopes separated those surfaces from the approaches. The archaeological plan presents a sequence of spaces rather than the footprint of a great keep. Some supposed defensive features remain uncertain, and the visible earthworks cannot all be assigned securely to the same month of 1333.

    Kami-Akasaka makes the alteration of terrain particularly legible. Its main and secondary enclosures occupy a projecting ridge, with smaller terraces and steep cut slopes around them. A defensive arrangement called Soroban-bashi combines terraces with dry ditches cut across the ridge. Despite the name, this is not a surviving wooden bridge. It is an interruption of the ground over which someone approaching the inner works would have to move.[7]

    Such an arrangement separates the size of an army from the number of people who can reach a particular obstacle together. Men elsewhere on a hillside cannot all bring their strength to bear on the same narrow approach. Each terrace or cut requires another movement before the next part of the position becomes reachable. This is a reading of the surviving layout, not a reconstruction of exactly where individual soldiers stood.

    The medieval war chronicle Taiheiki builds its Chihaya account around that difficulty. It describes early attackers pressing towards the entrances, where stones and arrows strike them on the slopes. Its enormous army and casualty figures belong to an expansive literary narrative, not a usable battlefield count. The spatial problem is more durable than the numbers: an army spread around a hill still has to concentrate at places where men can actually climb.[8]

    The physical landscape and the chronicle therefore meet at a limited but useful point. Both give the approach a determining role. Neither the earthworks nor the narrative permits every celebrated stratagem to be plotted onto an exact surviving feature.

    3. The business after an attack

    Naotsune's wounded-men report belongs to a different kind of writing. It was submitted in connection with the fighting, rather than composed as an extended story of the campaign. The published explanation identifies the endorsements as those of officials who visited his camp to examine injuries. Marks distinguished checked entries from one left unconfirmed.[1]

    The report shows a second task following the attempt to climb the ridge. Injury had to become a recognized claim. The person making the report, the injured men, and those endorsing it performed different parts of that process. Fighting alone did not write the document or verify its contents.

    This also changes the way the attacking force appears. “The army” conceals the smaller bodies whose service could be reported separately. Naotsune's men had a particular connection to him, while the inspecting officials supplied recognition beyond that immediate group. The document preserves one of the transactions through which a large force was made administratively intelligible.

    The Taiheiki includes wound inspection too. After describing the early assaults, it turns to officials registering the dead and injured. It then reports an order prohibiting further fighting without a commander's permission. The narrative moves from men competing to advance to an attempt to regulate their participation.[8]

    The contemporary report does not authenticate the chronicle's totals or every stage of that sequence. What it does establish is that checking injuries was an actual part of the campaign. The later narrative uses that work to mark a change in its siege story: the attackers have to stop treating proximity to the castle as an invitation to charge. Their commanders now face the problem of turning assembled contingents into a sustained operation.

    4. A blockade depends on what it can reach

    The chronicle's next proposal is to take away the defenders' water. A commander invokes the experience at Akasaka, where, in his account, cutting the supply brought surrender. He reasons that Chihaya must likewise depend on water obtainable outside the castle. Troops are placed beside a stream, with obstacles across the paths by which the defenders might descend.[9]

    This changes the object of the operation. The attackers no longer need to cross the castle's threshold immediately. They need to identify a necessity that its occupants cannot obtain without leaving. A water blockade would make the defenders' own routine bring them into reach.

    The Taiheiki defeats that proposal by giving the defenders a different arrangement. It describes springs known to mountain practitioners, stored water in hollowed timber containers, and gutters directing rain from roofs into reservoirs. These are elements of the narrative's explanation, not excavated installations that can now be pointed out on the ridge. Its causal sequence is nevertheless clear. Storage and collection allow the garrison to avoid the stream the attackers are guarding.

    The story's most revealing omission is an encounter that never happens. The guards wait for water carriers, but none come. Their position has been chosen on an assumption about the castle's dependence. Occupying it does not produce control over the resource that actually keeps the defenders supplied.

    The account then gives the garrison a sortie against the waiting detachment. Other attacking troops cannot immediately assist because valleys and ridges lie between their positions. The terrain that obstructed the first assault now obstructs cooperation among people stationed around the castle.[9]

    No precise water budget or night attack can be recovered from this passage alone. Its useful contribution is the distinction between surrounding a place and isolating it. The former describes the attackers' position; the latter depends on their knowledge of the defenders' needs and connections. The real ridge's continuation towards Mount Kongō makes that question geographically meaningful, without proving that the chronicle's particular supply arrangements existed exactly as described.

    A blockade is consequently a demand for information as well as men. The besiegers have to know which route, source, or activity they must interrupt. A large force at the wrong stream could remain a large force without becoming an effective water blockade.

    5. Waiting was an operation too

    After further attacks, the Taiheiki has the commanders order a blockade intended to exhaust the castle's food. The army is to remain around the position rather than continue losing men in direct assaults. The change makes time, instead of immediate movement up the slope, the principal means of pressure.[10]

    The narrative fills the ensuing interval with linked verse, board games, tea, and other diversions. It makes much of the contrast between an army gathered for combat and people finding ways to pass the days. These scenes belong to its portrayal of the siege; they are not a surviving duty roster from the camps.

    Yet the account also says that the defenders find the cessation of attacks troublesome. This is a consequential turn. If remaining outside is the attackers' method, avoiding battle need not mean abandoning their purpose. From the garrison's side, an enemy that waits can present a different difficulty from one that repeatedly approaches the defended entrances.

    The famous straw figures enter at this point. In the chronicle's version, armoured dummies, shields, and a brief appearance by real defenders entice attackers forward. The living men withdraw, leaving the figures behind. When the attackers gather around what they take to be opponents, stones strike the concentrated group.[10]

    The episode is often detached from its place in the story and offered as a self-contained example of Masashige's ingenuity. Within the sequence, its purpose is more specific. It brings an army that has been ordered to wait back towards a dangerous approach. The false figures matter because they offer an apparent opportunity to do something different from maintaining a blockade.

    The chronicle's distinction between the two methods is sharper than its dramatic casualty figures. Attacking and waiting require different behaviour from the same people. A position maintained patiently can be lost as an advantage when a detachment rushes towards an apparent opening. Conversely, refusing an opening can look like inactivity to observers expecting visible progress.

    The Taiheiki makes that pressure explicit when a messenger arrives from Kamakura with an order against letting the days pass without fighting. The commanders respond with another proposal: a large timber structure that would bridge the gap to the castle's cut slope. Carpenters are summoned from Kyoto. In the narrative, the attempted crossing ends in fire.[11]

    Here the siege acquires another distance. The men below the ridge answer to an authority elsewhere, whose demand for action reaches them by messenger. The story has moved from an obstacle in the ground to the difficulty of sustaining one method of siege under pressure to try another.

    6. The army had a rear of its own

    The defenders were not the only people who needed supplies. Near the end of its Chihaya account, the Taiheiki brings in armed groups from the surrounding regions, acting under Prince Moriyoshi's instructions. They obstruct routes used by the attacking force. The narrative reverses the direction of the supply problem: the army attempting to wear down a castle now has difficulty maintaining itself outside it.[12]

    That reversal links Chihaya to a war larger than its ridge. The chronicle's account of shortages and departures cannot establish an exact attrition rate. It does explain why a siege army's security extended beyond the perimeter it was holding. Food, reinforcements, and communication had to continue reaching people who were stationary in front of the stronghold.

    Contemporary documents provide another way to see the campaign opening outward. A summons dated to the first day of the fourth month of 1333 calls on Naotsune to act against Hōjō Takatoki, the figure at the centre of the Kamakura regime. Its issuer is Moriyoshi. The recipient is the same man whose earlier service is documented on the attacking side at Chihaya.[13]

    An invitation to change sides is not proof of immediate compliance. It is evidence that the opposing cause was addressing a warrior whom the government had mobilized. Command could no longer be understood only as an instruction descending through a stable army. A participant could be approached by another authority seeking his service.

    The decisive losses of the shogunate followed elsewhere. Ashikaga Takauji turned against it; the government's Rokuhara establishment in Kyoto fell in the fifth month. Later that month, Nitta Yoshisada's attack on Kamakura ended with the destruction of the regime there. Chihaya's resistance had continued during this expanding crisis.[14]

    The castle's endurance should not be made the single cause of those changes. Its local history shows a different, more limited relationship. The attackers needed time and continuing command to reduce the stronghold. The government directing them was simultaneously losing its position in other places. A siege might remain unfinished because its wider political conditions ended first.

    Chihaya-Akasaka is therefore more intelligible as a group of positions within a changing campaign than as the stage for one uninterrupted victory. Lower Akasaka had fallen; other positions sustained renewed resistance. At Chihaya, approaches, supplies, coordination, and instructions from afar presented different tasks. The defenders did not have to solve the attackers' problems for them. Holding the ridge left those problems accumulating below, while the war beyond the mountains changed who could still command the army.

    Next topic · HO Guides

    Wakayama Castle: Making Room Below the Keep

    A covered crossing, a buried boundary wall, and a palace plan explain how the castle accommodated a ruling household.

    Continue reading
    Sources and NotesView sourcesHide sources

    Web materials were consulted on 16 September 2026. Numbered months in medieval records retain the historical calendar, including its intercalary month; they are not modern Gregorian month names. The Taiheiki is read as a medieval narrative with its own explanatory and literary purposes. Its speeches, stratagems, water arrangements, and numbers are not treated as independently verified field observations. Contemporary documents below were accessed through published scholarly descriptions and document abstracts, not through an independent decipherment of the originals.

    1. [1]

      Minato City, 港区史 通史編 原始・古代・中世, “軍忠状による戦功認定システム,” pp. 223–226, paragraph beginning “『熊谷直経合戦手負注文』” and figure 2-1-2 caption. It describes the report submitted after Chihaya fighting, the two inspecting officials' endorsements, checking marks, the word 深, and an unmarked stone-injury claim. The original document image could not be retrieved; these observations are used through the published explanation. The article does not adopt the explanation's stronger conclusion that the unmarked claim was certainly rejected, or project its whole account of later Ashikaga procedures backward into 1333. The caption supplies an intercalary-second-month date; the body uses only the secure year and campaign identification.

      Return to the reference ↑
    2. [2]

      Hiroshima Prefectural Archives, *広島県史年表(中世1)治承4(1180)~永享12(1440)*, PDF pp. 34–35, entries for 1331–1333; The entries identify Go-Daigo's Kasagi campaign, capture, and the wider sequence. For Lower Akasaka's place in the first rising, see the village account in note 3. The chronology cites its underlying sources; it is not a claim that all those original works were independently consulted.

      Return to the reference ↑
    3. [3]

      Chihaya-Akasaka Village, “史跡赤阪城跡(下赤阪城跡)”, complete site account, and “歴史・文化”, historical account of the castle sites. These describe the lower site's association with 1331 and the indistinct surviving features. The article does not adopt the narrative's precise tower count or treat every episode of withdrawal and recapture as independently established archaeology. The relationship among the three locations is not presented as proof of one surviving original connecting road or a single pre-planned defensive system.

      Return to the reference ↑
    4. [4]

      Chihaya-Akasaka Village, “歴史・文化”, subsection on the traditional birthplace and the excavated residence, including the 1990–1992 Kusunoki Hall construction investigations. The double-moated building complex is archaeological evidence; attribution to a Kusunoki residence remains a candidate identification alongside another proposed location. No occupant's identity or birth event is inferred from the enclosure.

      Return to the reference ↑
    5. [5]

      Hiroshima Prefectural Archives, chronology cited in note 2, entries for 1332, twelfth month, ninth day, and 1333, second and intercalary second months. These identify the government summons, Kumagai fighting at Akasaka's main approach (Kumagai document 40), and Chihaya's main approach (documents 41 and 211). The summaries were read on the actual chronology pages. The essay does not merge all named Kumagai family members into one individual or infer that every summoned man complied immediately.

      Return to the reference ↑
    6. [6]

      Chihaya-Akasaka Village, *千早城跡* archaeological leaflet, both pages, including the description of the ridge, valleys and connection towards Mount Kongō, and the earthwork plan credited to Murata Shūzō. The explanation distinguishes identifiable terraces from less certain features. The rear connection is geographical evidence, not proof of a particular supply journey or an open route throughout the siege. No on-site inspection or entry to the restricted principal enclosure is claimed.

      Return to the reference ↑
    7. [7]

      Chihaya-Akasaka Village, “史跡赤阪城跡(上赤阪城跡)”, complete earthwork description: main and secondary enclosures, smaller terraces, cut slopes, and Soroban-bashi's combination of terraces and dry ditches. The explanation of interrupted movement is a bounded reading of that described arrangement, not a reconstruction of the exact attack sequence or a date for every surviving cut.

      Return to the reference ↑
    8. [8]

      Taiheiki, book 7, section “千剣破城軍事”, opening passage through the order beginning “今より後は、大将の御許なくして.” See the entire Chihaya section in its original-language public transcription. The account describes approaches, stone and arrow attacks, injury inspection, and a prohibition on unauthorized fighting. Its huge totals and its assessments of the participants' character are not adopted. Comparison with note 1 establishes the existence of injury reporting, not authentication of the chronicle's complete scene.

      Return to the reference ↑
    9. [9]

      Taiheiki 7, same section, from “爰に赤坂の大将金沢右馬助” through the account of the detachment at the stream and the captured banners. See the passages on springs, timber water containers, roof gutters, and the difficulty of reinforcement across ridges and valleys. The article analyses the text's sequence without claiming surviving reservoirs, measured flow, water-treatment efficacy, exact garrison endurance, or a verified sortie. Its discussion of encirclement versus isolation is an inference from the explicitly attributed narrative, considered alongside the site's geographical description.

      Return to the reference ↑
    10. [10]

      Taiheiki 7, same section, “唯取巻て食責にせよ” through the straw-figure episode. See the original-language account of the food blockade, diversions, defenders' response, and renewed attack. The essay interprets the episode's position in the narrative rather than presenting the dummies as archaeologically demonstrated equipment. No exact reconstructed dialogue, emotional state, casualty count, or continuous camp schedule is supplied.

      Return to the reference ↑
    11. [11]

      Taiheiki 7, same section, passage beginning “同三月四日関東より飛脚到来して,” through the burning of the timber crossing. The text describes a message, commanders' consultation, carpenters summoned from Kyoto, a bridge-like assault structure, and its destruction. This is used as the chronicle's representation of pressure on siege command. Its dimensions, workforce, rope count, technical feasibility, and losses are not independently established or reproduced.

      Return to the reference ↑
    12. [12]

      Taiheiki 7, same section, final sequence beginning “去程に吉野・戸津河・宇多・内郡の野伏共.” The text attributes obstruction of the attackers' routes to groups acting under Moriyoshi's orders. The essay reports this narrative relationship, not a verified map of road closures or an independently calculated rate of desertion and supply loss. Its sweeping claims about weapons lost throughout Japan are not adopted.

      Return to the reference ↑
    13. [13]

      Hiroshima Prefectural Archives, chronology, PDF p. 35, 1333, fourth month, first day: Moriyoshi's summons to Kumagai Naotsune against Hōjō Takatoki, citing Kumagai document 34. See also the earlier Chihaya entries. The event is the issuing of a summons, not a documented same-day change of allegiance, a statement of Naotsune's private calculation, or proof of his physical location when it arrived.

      Return to the reference ↑
    14. [14]

      Same chronology, PDF p. 35, 1333 entries for Takauji's change of side in the fourth month, the fall of Rokuhara on the seventh day of the fifth month, and the Kamakura fighting and destruction on the twenty-first and twenty-second days. Chihaya's continuing resistance during the regime's fall is also the sequence given in the village's historical account, note 3. The final synthesis does not assign the shogunate's collapse to the siege alone or claim that Masashige had planned the later actions of other leaders.

      Return to the reference ↑

    Table of Contents

    1. 00Opening
    2. 011. Three strongholds, more than one campaign
    3. 022. Where numbers had to become an approach
    4. 033. The business after an attack
    5. 044. A blockade depends on what it can reach
    6. 055. Waiting was an operation too
    7. 066. The army had a rear of its own
    Sources and notes

    Reading preferences

    Text size
    Glossary intensity

    Choose how much explanation appears as you read.

    Saved on this device. All terms remain available in the glossary.

    Browse the glossary →
    History Omakase

    Opening reading…