20 April 2021

Siege of Ulsan

This will be my first blog post about Imjin War that is not a direct translation of other's work. Since my knowledge and access to primary sources are quite limited as compared to the authors of my translated articles, I will still be heavily referencing others' works, including the writings of aforementioned authors, pilfering incorporating anything I find useful into this blog post. Therefore, I expect this blog post to be a lot more messy, with less references but more rewrites and edits. Nevertheless, I still strive to present a most detailed and accurate account of battle to the best of my ability.

蔚山城の戦い
Section of "Urusan Rōjō-zu Byōbu (《蔚山籠城図屏風》)", a Japanese folding screen painted in 1886 to commemorate the defence of Ulsan, depicting the highly misleading scene of Ming troops assaulting a lone Ulsan castle.

Prelude

After the peace talk that dragged on for years fall apart, an enraged Toyotomi Hideyoshi (豐臣秀吉) ordered the resumption of hostilities. The renewed Japanese invasion was fiercer than ever, quickly overrunning the defences of Sacheon, Namhae and Gwangju. The Koreans, still weakened by the devastation of the first invasion, fared even worse this time around, as even the proud Joseon navy was all but annihilated in Battle of Chilcheollyang. With Joseon navy destroyed, Japanese army turned its attention to Jeolla Province. Soon Namwon fell, followed by Hwangseogsanseong, Jeonju and Gongju. Even the valiant effort of Admiral Yi Sun-sin (이순신 or 李舜臣) during Battle of Myeongnyang did little to stem this unstoppable tide of Japanese onslaught, as what little gains he made from the tactical victory failed to make up for the devastating strategic loss—Japanese navy captured Yi's naval base and made headway into the Yellow Sea (for the first time since the beginning of Imjin War), and Jeolla Province fell into Japanese hands shortly after.

With Joseon navy virtually annihilated and Gyeongsang, Jeolla, as well as Chungcheong Province trampled underfoot, Japan had achieved all initial objectives of the second invasion. All hope seems lost for the Koreans, and even the Chinese reeled from the ferocity of the attack (after Battle of Jiksan, many Ming commanders and troops were seriously contemplating a full retreat from Korea). Yet, in a strange twist, Japanese invaders suddenly halted their advance, retreated southwards, and began to construct a series of fortresses along the entire coast, among them Ulsan fortress complex, the focus of this blog post.

It turns out Hideyoshi learnt well from his past mistakes and took steps to correct his failures (perhaps a little too well). The Japanese were overeager during the first invasion and tried to advance as fast as they could, which resulted in severe logistics issues that left them unable to mount an effective defence against Ming counterattack. They were also unaccustomed with the harsh Korean winter, and wanted to wait out the winter before resuming attack. Thus for the second invasion the Japanese decided to take it slow. The overall strategy of the second invasion was to quickly crush Korean defence, then consolidate and dig in. Once Japanese footholds in Korea were sufficiently secure, a new phase of invasion would begin, followed by yet another phase of consolidation and digging in. Nevertheless, while Hideyoshi's overall strategy was sound, his decision to have the Japanese army halt its offensive and consolidate gave precious breathing space to his enemies, allowing the Koreans to stabilise from the initial shock and the Chinese to send in reinforcement.

The Battlefield

蔚山倭城群
Important locations of the siege campaign marked on a modern satellite map. Made by myself using Google Earth and Adobe Illustrator (click to enlarge).

A serious misconception regarding the Siege of Ulsan is that the so-called "Ulsan Japanese Castle" is mistakenly believed to be but a lone castle. In reality, it was a massive fortress complex with no less than FOUR separate and independent castles, protected by a five-kilometre-long line of multi-layered ditches, earthen walls and palisades, and guarded by two additional satellite forts. Regrettably, the only visible ruin, that of Dosan Fortress, is now called "Ulsan Japanese Castle", which only helps to perpetuate the misconception.

Modern reproduction of a Japanese occupation era land survey map showing the defensive layout of Ulsan fortress complex, taken from Stephen Turnbull's 'Japanese Castles in Korea 1592–98'. I highlighted the three main fortresses inside the complex in red.

Important fortresses involved in the Siege of Ulsan were:

Ulsan Fortress (蔚山城)

蔚山城と島山城
Section of a Ming period map 'Si Lu Wo Zhai Zong Tu (《四路倭寨總圖》)' depicting Ulsan Fortress (right) and the smaller Dosan Fortress (left). Note the double-layered palisades of Ulsan Fortress.
Known to Ming Chinese as Wei Shan Qing Zheng Da Zhai (蔚山清正大寨, lit. 'Kiyomasa's Ulsan great stronghold') and to the Koreans as Seonghwangdang Togul (성황당토굴 or 城隍堂土窟, lit. 'Seonghwangdang burrow' as the fortress was built near the site of an ancient shrine of Seonangshin), this fortress was the largest fortification in the vicinity and served as a major staging ground for Japanese army to thrust into Gyeongju.

Katō Kiyomasa actually built the fortress during the first invasion, although at the time it was a relatively small and unimportant fort. After Japanese army retreated from Hanseong (한성 or 漢城, present-day Seoul) near the end of the first invasion, Katō Kiyomasa began to expand and fortify the original fort, entrusting Asano Yoshinaga to carry out the task.

The fortress was also known as Dongbudongseong (동부동성 or 東部洞城) during the period of Japanese occupation of Korea. Japanese historians generally regard this fortress to be simply a dejiro (出城, lit. 'Satellite castle') subservient to Dosan Fortress, downplaying its importance (even though it was actually larger than Dosan Fortress).

After the war, the hill which Ulsan Fortress was built on became the site of a Confucian academy known as Gugang Seowon (구강서원 or 鷗江書院), although the academy had since been relocated and the hill flattened to make way for modern urban development.

Dosan Fortress (島山城)

Ruin of Dosan Fortress
Ruin of Dosan Fortress now reclaimed by nature.
Sitting atop Dosan (도산 or 島山) hill at the west-southwest side of Ulsan Fortress, this fortress was the site where the most intense battles took place. Katō Kiyomasa built Dosan Fortress and the rest of the fortress complex during the second invasion in anticipation of the coming Ming offensive. Due to delays, harsh winter, and difficulties in gathering materials, the fortress was still incomplete when Ming army arrived.

Some Japanese sources call the fortress Urusan Shinjō (蔚山新城, lit. 'Ulsan New Fortress') to differentiate it from the original Ulsan Fortress. Nowadays, it is also known as Ulsan Japanese Castle, and the hill which the fortress was built on is also home to Hakseong Park.

Bangujeong Fort (伴鷗亭城)

A secondary fort located at the south of Ulsan Fortress, built atop the ruin of a Goryeo-era walled settlement. It was named after Bangujeong (반구정 or 伴鷗亭, lit. 'Pavilion of accompanying seagulls'), a pavilion built by righteous army leader Yi Eung-chun (이응춘 or 李應春) before the war.

Seobudongseong (서부동성 or 西部洞城)

A fortlet located between Ulsan Fortress and Dosan Fortress, overseeing the passageway between two major fortresses. Being a relatively minor fort, it was only given a name during the period of Japanese occupation of Korea.

Eastern Camp

A military camp situated somewhere outside Bangujeong Fort (its exact location is currently unknown as few traces of the palisade survive). It was garrisoned by Asano Yoshinaga, although he left the camp before Ming army attacked in the earnest.

Taehwa River Fort (太和江寨)

A river fort located at about 2 ~ 3.5 kilometres west of Dosan Fortress, overseeing Taehwa River from both sides.

Byeongyeongseong (병영성 or 兵營城, lit. 'Barrack fortress')

Surviving Korean wall section of Byeongyeongseong (click to enlarge).
A Korean hill fort that once served as the headquarters of Joseon Dynasty's Gyeongsang Left Army, although the control changed hand early in the war. Katō Kiyomasa asked Mōri Hidemoto to set up a Jinsho (陣所, lit. 'Battle camp') here, utilising existing Korean fortifications as well as erecting his own, to protect Ulsan Fortress proper. It was also known as Cheongjeong-ui Byeolyeong (청정의 별영 or 清正別營, lit. 'Kiyomasa's other camp') by the Koreans, as well as Kariei (仮営, lit. 'Temporary camp') in Japanese sources.

Note: While both castle and walled settlement are known as Shiro or (城) in Japanese, the same cannot be said for Chinese and Koreans that seldom build castle-like fortifications, and only perceived walled settlement as Cheng/Seong (城). Instead, contemporary Chinese generally used Ying (營, lit. 'Camp'), Zhai (寨, lit.'Stockade' or 'Stronghold') or Wo Zhai (倭寨, lit. 'Japanese stockade') to refer to Japanese castles, while contemporary Koreans used some variations of Togul (토굴 or 土窟, lit. 'Burrow'), Waegul (왜굴 or 倭窟, lit. 'Japanese burrow'), Jeoggul (적굴 or 賊窟, lit. 'Bandit burrow') and Jeognu (적누 or 賊壘, lit. 'Bandit fort') to refer to Japanese castles, all of which are derogatory. The modern name, Wo Cheng/Waeseong/Wajō (倭城 in Chinese, Korean and Japanese pronunciation respectively) only came into use after the war.


Belligerents

Ming-Joseon coalition

Ming Army

Regulator of Korean Military Affairs: Yang Hao (楊鎬)
Military Superintendent: Ma Gui (麻貴)

Middle Division leader: Gao Ce (高策)
Middle Division commander: Zu Cheng Xun (祖承訓), Po Gui (頗貴), Wu Wei Zhong (吳唯忠), Li Hua Long (李化龍)
Strength: 11,690

Left Division leader: Li Ru Mei (李如梅)
Left Division commander: Chen Yin (陳寅), Mao Guo Qi (茅國器), Yang Deng Shan (楊登山), Bai Sai (擺賽), Lu De Gong (盧得功), Dong Zheng Yi (董正誼)
Strength: 13,006

Right Division leader: Li Fang Chun (李芳春), Jie Sheng (解生)
Right Division commander: Lu Ji Zhong (盧繼忠), Yang Wang Jin† (楊萬金), Chen Yu Wen (陳愚聞)
Strength: 11,630

Total Strength: 36,326*
Casualties: 798 killed; 823 succumbed to injury or sickness; 2,908 wounded
Casualties (horse): 213 killed, 2,330 died to attrition (injury, sickness, cold, malnutrition, fatigue etc.)

*Note: The primary Ming source 'Liang Chao Ping Rang Lu (《兩朝平攘錄》)' erroneously included a few Ming commanders that were yet to enter Korea into the tally, so the strength of Ming army was actually lower than 36,326.

Joseon Army

Commander-in-chief: Gwon Yul (권율 or 權慄)
Notable commander: Kim Ung-seo (김응서 or 金應瑞), Yi Si-eon (이시언 or 李時言), Seong Yun-mun (성윤문 or 成允文), Gwon Eung-soo (권응수 or 權應銖), Jeong Gi-ryong (정기룡 or 鄭起龍), Go Eon-baek (고언백 or 高彥伯)
Notable official: Jang Un-ik (장운익 or 張雲翼), Yi Deok-hyeong (이덕형 or 李德馨)

Strength: 12,500*
Casualties: 298 died; 876 seriously wounded (casualty record only up to date to February 7, 1598. Casualties for the remaining days unknown)

*Note: This was only the on-paper strength. In reality, the initial strength of Joseon army was only 3,500, with more troops being sent in during the siege. As a whole, a total of 9,965 Joseon troops participated in the fighting at one point or another, although as many as 4,982 deserted. Despite rampant desertion and abysmal quality, by Korean standard these troops were considered some of the most battle-hardened soldiers Joseon military had to offer.

Joseon Navy*

Notable commander: Yi Un-ryong (이운룡 or 李雲龍)
Strength: Unknown
Casualties: N/A

*Note: While Yi Un-ryong's fleet was the closest Joseon navy in the vicinity, and could have joined the battle, he remained inactive throughout the entire siege campaign.

Japan

Japanese Army

Commander-in-chief: Katō Kiyomasa (加藤清正)
Notable commander: Asano Yoshinaga (浅野幸长), Reizei Motomitsu† (冷泉元満), Asunuma Motohide† (阿曾沼元秀), Tsuno Ieyori† (都野家頼), Shishido Mototsugu (宍戸元続), Ōta Kazuyoshi (太田一吉), Katō Yasumasa (加藤安政), Katō Yasuyuki (加藤安之), Katō Yosaemon (加藤与左衛門)
Rōnin: Aso Kazuo (阿蘇大夫)

Strength: 10,000 ~ 23,000+
Casualties: 18,360+ killed on the first day; 896+ died to starvation or cold during the siege (according to Ōkōchi Hidemoto)

Japanese Relief Force (Seosaengpo)

Commander-in-chief: Mōri Hidemoto (毛利秀元)
Notable commander: Yamaguchi Munenaga (山口宗永), Kuroda Nagamasa (黒田長政), Ankokuji Ekei (安国寺恵瓊), Takenaka Shigetoshi (竹中重利), Nabeshima Naoshige (鍋島直茂), Nabeshima Katsushige (鍋島勝茂), Hachisuka Iemasa (蜂須賀家政), Ikoma Kazumasa (生駒一正), Katō Yoshiaki (加藤嘉明), Wakisaka Yasuharu (脇坂安治), Hayakawa Nagamasa (早川長政), Kakimi Kazunao (垣見一直), Kumaga Naomori (熊谷直盛), Chosokabe Motochika (長宗我部元親), Nakagawa Hidenari (中川秀成), Ikeda Hideuji (池田秀氏), Ikeda Hideo (池田秀雄), Mōri Katsunobu (毛利勝信), Mōri Katsunaga (毛利勝永), Akizuki Tanenaga (秋月種長), Takahashi Mototane (高橋元種), Itō Suketaka (伊東祐兵), Sagara Yorifusa (相良頼房)

Japanese Relief Force (Other)

Notable commander: Shimazu Toyohisa (島津豊久), Kikkawa Hiroie (吉川広家), Mōri Takamasa (毛利高政), Tōdō Takayoshi (藤堂高吉), Tōdō Yoshikatsu (藤堂良勝), Matsuura Shigenobu (松浦鎮信), Kurushima Hikozaemon (来島彦左衛門), Kan Uemonpachi (菅右衛門八)

Overall strength of Japanese relief force: 16,780+
Casualties: 2,800+ killed (according to Ōkōchi Hidemoto)

The Battle

『此賊七千,足當倭奴十萬。』
"Seven thousand Jurchens could resist a hundred thousand Japanese."

— Li Ru Mei's impression of the capabilities of Japanese army.

While the Japanese were busy constructing their fortresses, Ming reinforcement from Xuanfu, Datong, Liaoji, Yansui, Baoding and Zhejiang, numbering roughly 42,000, began to pour into Korea and slowly gathered at Hanseong. In December 1597, Xing Jie (邢玠), Ming military commissioner of Korea and supreme commander of Ming army in Korea, crossed Yalu River and entered Korea, arriving at Hanseong on January 6, 1598. He held a war council with Regulator Yang Hao and commander-in-chief Ma Gui to discuss the matters of countering Japanese advance. After repeated discussions, Ming military leadership decided to strike at Ulsan first, due to its significance as a launchpad for incursion into Gangwon Province, Hamgyong Province, and even Hanseong itself.

After reorganising Ming army into three divisions, Xing Jie ordered Yang Hao and Ma Gui to lead Left and Right Division to Ulsan, departing from Chunju, passing through Mungyeong Saejae, Andong and Gyeongju. Xing Jie was concerned that Konishi Yukinaga (小西行長) at Suncheon might come to Ulsan's aid, so he arranged a detachment from Middle Division to be stationed at Uiryeong, both to defend against possible Japanese reinforcement from Jeolla Province, and to reinforce Ming army attacking Ulsan Fortress when needed. On top of that, Xing Jie also picked 1,500 Ming cavalry and ordered them, along with some Joseon troops, to march to Suncheon through Cheonan, Jeonju and Namwon while putting up an appearance that they were about to attack, as well as spreading false rumours that he was about to personally lead a huge army of 200,000 to Namwon to lay siege to Suncheon Fortress. These were all diversion tactics to pin down Konishi Yukinaga's force. In reality, Xing Jie himself stayed behind in Hanseong with only 1,500 Ming troops.

Upon receiving Xing Jie's order, Yang Hao and Ma Gui crossed Han River and departed for Ulsan. Ma Gui arrived at Mungyeong City on January 13, 1598 and held a secret war council attended by both Ming and Joseon military leadership. During the council, Ma Gui secretly asked Gwon Yul to dispatch Joseon Navy under Yi Un-ryong to support the siege campaign, and promised him several hundred Southern troops, along with 200 Ming and Joseon arquebusiers, to support Yi Un-ryong's fleet.

On January 24, 1598, Yang Hao arrived at Uiseong. After consulting his Jubbansa (접반사 or 接伴使, a Joseon official appointed to a specific Ming delegate to serve as his adjutant, aide-de-camp, and sometimes translator) Yi Deok-hyeong, he decided to send a number of Ming and Joseon scouts, among them a Japanese defector known only by his Korean name Yeo Yeo-moon (여여문 or 呂汝文), to reconnoitre Ulsan fortress complex ahead of the main army.

On January 26, 1598, all three divisions of Ming army arrived at Gyeongju. Yang Hao held a war council to discuss the finer details of the siege campaign, and Ming army began final preparation before the attack. Ma Gui intended to commit the full might of Ming army into attacking Ulsan fortress complex. However, he too was concerned about the potential reinforcement from Suncheon and Jinju, so he ordered Dong Zheng Yi to lead his troops to defend Namwon and Gurye (the fact that Ming military leadership made multiple arrangements as a precaution against Konishi Yukinaga's reinforcement shows that they saw him as a very dangerous foe). On top of that, Ma Gui also dispatched Yu Cheng En (于承恩), leading a small contingent of Southern troops and Joseon naval troops, to sneak to Seosaengpo and set up a dummy army (using scarecrows for troops, long poles for weapons, fabric scraps for clothes and banners, as well as making noise with horns and drums) to sow confusion.

The next day, Yeo Yeo-moon returned from his reconnaissance mission, bringing with him valuable intelligence and a detailed map of various facilities inside Ulsan fortress complex drawn by himself. With all preparation complete, Ming army departed Gyeongju and marched south towards Ulsan on January 28, 1598. Yeo Yeo-moon once again moved ahead of Ming army and blended into Japanese territory unnoticed.

慶長の役
Overall strategic map of the siege campaign (click to enlarge).
Unfortunately, the promised food and supply to support the siege campaign by the Koreans was nothing more than a puff of hot air. Despite boasts from Ryu Seong-ryong (류성룡 or 柳成龍), Chief State Councillor of Joseon Dynasty, claiming that he already requisitioned vast amount of supply in Gyeongsang Province, as well as assurance from Yun Bang (윤방 or 尹昉), quartermaster in charge of transporting military supplies, that there was enough food to feed entire Ming army for one and a half month, such supply was nowhere to be found. In fact, Korean logistics incompetence was of such a mind-boggling degree that when Yang Hao arrived at Yeongcheon, all local Joseon officials that were supposed to meet him vanished without a trace, leaving Ming army unable to resupply. Faced with severe supply problem before the siege campaign even begun (Ming army ran out of food as early as January 26, 1598), Yang Hao was forced to leave artillery train behind and rush to Ulsan, hoping that he could win the battle decisively before food shortage turn into a major issue.

Day 1: A payback for Byeokjegwan

January 29, 1598 (25th year of Wanli reign, 12th month, 23rd day)

Japanese positions on January 29, 1598, before the arrival of Ming army (click to enlarge).
The first Ming attack came just before the dawn of January 29, 1598. A small advance force of about 3,000 cavalry led by Li Ru Mei, Yang Deng Shan and Bai Sai arrived at Ulsan ahead of the main army, with the intention of neutralising Japanese garrison of Byeongyeongseong. Li Ru Mei originally planned to launch a feint attack on Byeongyeongseong with 300 cavalry to lure out its garrison into an ambush set up by Yang Deng Shan and Bai Sai. Much to his surprise however, the garrison of Byeongyeongseong posted no sentries and was caught completely off-guard by the feint attack, so he immediately took advantage of the situation and ordered his troops to attack for real. The camp of a commander known only by his surname Asakuchi (浅口) was the first to come under attack. Japanese troops inside the camp were quickly defeated and fled to Asunuma Motohide's camp. Alerted by the chaos, Asunuma Motohide and his troops tried to resist the attack, but he was quickly killed and his troops scattered. Li Ru Mei followed up by attacking Reizei Motomitsu's camp. In a hurry, Reizei Motomitsu, a seasoned old warrior (he was 56), tied two swords to his sleepwear using a sash and then rushed to meet Ming cavalry in combat with another longer sword in his hands. Ming cavalry killed him and crushed his troops all the same.

In a flash, what was originally meant to be a harassing attack quickly evolved into a serious assault, and then a wanton slaughter. Japanese garrison fell into complete disarray, and all three of its commanders (Reizei Motomitsu, Asunuma Motohide and Tsuno Iyeori) were killed in action. The chaos at Byeongyeongseong eventually alerted the main Japanese force at around 7 am, prompting Shishido Mototsugu, Ōta Kazuyoshi and Asano Yoshinaga to rush to Byeongyeongseong's rescue with a large army of more than 10,000 troops.

蔚山襲撃
Li Ru Mei crossed Dongcheon River with 300 cavalry to attack Byeongyeongseong, then lured the Japanese reinforcement into an ambush (click to enlarge).
Realising that a large Japanese army was upon him, Li Ru Mei ordered his troops to revert to their original mission and lured the (now much larger) Japanese reinforcement to the position where Bai Sai and Yang Deng Shan lay in waiting. After crossing Dongcheon River (동천 or 東川, lit. 'Eastern River'), Japanese army drew up in battle formation, then launched an attack against Ming advance force. The vanguard of the Japanese army ran into the ambush set up by Bai Sai and Yang Deng Shan, and was quickly surrounded and defeated. With the ambush sprung, Japanese army attacked Ming advance force head-on twice more, but still failed to gain the upper hand.

At around 8 am, additional Ming troops led by Ma Gui arrived at the scene. After witnessing the heated battle between Ming advance force and Japanese army, Ma Gui personally led a unit of 200 elite Mongol cavalry armed with iron whips and stormed into the Japanese army with such ferocity that Japanese arquebusiers were unable to even react to his cavalry charge. By 11 am, Ming advance force and Ma Gui's reinforcement had decisively crushed the Japanese army on the field. Ōta Kazuyoshi was wounded in battle but managed to escape thanks to a relief column sent by Katō Yasumasa, who stayed behind at Ulsan fortress complex. Meanwhile, Asano Yoshinaga split his army into two, and personally led one half of his army to cover the retreat and river crossing of the other half.

At the end of the first day of battle, Ming army successfully burned down the Japanese camp at Byeongyeongseong and collected some 460 Japanese heads as well as numerous equipment abandoned by fleeing Japanese troops. After the battle, Li Ru Mei's advance force and Ma Gui's reinforcement waited for the rest of the Ming army to arrive (other Ming troops gradually arrived since noon, and the entire army finally assembled by nightfall) while Japanese army dispatched a messenger to Seosaengpo Japanese Fortress to inform Katō Kiyomasa about the disastrous defeat. Shocked by the news, Katō Kiyomasa immediately left Seosaengpo and rushed to Ulsan, arriving at Dosan Fortress by midnight.

Despite overwhelming Ming victory on the first day, a very regrettable incident also happened: Japanese defector Yeo yeo-moon, who scouted ahead of Ming army, was still inside Japanese camp when Ming advance force attacked. Amid the chaos, he rose against his former kinsmen and killed four before escaping with their heads. Unfortunately, he ran into Bai Sai, who promptly killed him and stole the heads for himself.

Day 2, dawn: Fall of Ulsan Fortress

January 30, 1598 (25th year of Wanli reign, 12th month, 24th day)

The preparation for the next attack began at 3 am of January 30, 1598. Ming army intended to launch a simultaneous multi-pronged attack on the fortress complex: Li Ru Mei's Left Division would attack the fortress complex directly, whereas Li Fang Chun's Right Division would attack Taehwa River Fort. On the other hand, Gao Ce's Middle Division was tasked with securing Jeontan (전탄 or 箭灘, lit. 'Arrow shore'), a shore about 5 km west of Ulsan fortress complex, in order to ward off potential Japanese reinforcement from Busan coming to Ulsan through Yangsan and Eonyang or Taehwa river. Yang Hao and Ma Gui, highest-ranking commanders of Ming army, would personally supervise Left and Right Division respectively.

Japanese positions at the break of dawn of January 30, 1598 (click to enlarge).
Meanwhile, Japanese army caught wind of Ming activities at around 4 am, and began preparing for the defence of Ulsan fortress complex. After receiving an order from Katō Kiyomasa, Asano Yoshinaga left his camp and moved into Ulsan Fortress proper. Katō Yosaemon also relocated to sannomaru (三の丸, lit. 'Third circle', or tertiary ward of a castle) of Dosan Fortress.

Ming army launched a three-pronged simultaneous attack against Ulsan fortress complex (click to enlarge).
The actual attack began at 6 am with an opening salvo of artillery and rocket barrage, setting alight large swaths of buildings inside Ulsan fortress complex and even some ships at the river dock. Ming Left Division launched a direct assault on Ulsan Fortress (the Chinese mistakenly believed that Katō Kiyomasa was inside the fortress), and Asano Yoshinaga's camp quickly fell to the attack of Mao Guo Qi's contingent. Asano Yoshinaga, now at the helm of Ulsan Fortress's defence, put up a fierce resistance to Li Ru Mei's Left Division and managed to stall his attack for several hours. However, by 11 am a detachment led by Chen Yin managed to break through the multi-layered palisades at the northwest corner of the fortress complex and defeated both Ōta Kazuyoshi and Shishido Mototsugu's forces, forcing them to retreat into Dosan Fortress. Realising that Ulsan Fortress had become increasingly vulnerable, Katō Kiyomasa ordered Asano Yoshinaga to abandon his position and withdraw to Dosan Fortress as well. While Asano Yoshinaga was retreating, Ming troops poured in and overran the defenders of Ulsan Fortress, then pursued him all the way to Dosan Fortress. Mao Guo Qi also captured Bangujeong Fort and Seobudongseong along the way.

Meanwhile, Ming Right Division troops were initially hesitant to simply storm the walls of Taehwa River Fort. To inspire courage in his troops, Li Fang Chun ordered one of his retainers to climb up the wall and plant his personal yellow umbrella on the wall. This feat of courage successfully roused the Ming troops to launch the attack, and they quickly captured the fort. After the capture of Taehwa River Fort, Ming Right Division began to approach Ulsan fortress complex from the west, mopping up any Japanese troops and fortifications along the way. 

Ming army rapidly capturing one fortress after another (click to enlarge).
While the battle was raging on, Gao Ce and Wu Wei Zhong of the Middle Division sent a request to Yang Hao to ask for his permission to join the battle. Yang Hao was displeased with the request and cut off one of the ears of the messenger. Nevertheless, as Ming army gained more grounds, Yang Hao had a change of mind and ordered Ming Middle Division to move closer to Dosan Fortress by about 2.5 km.

Day 2, midday: Assault on Dosan Fortress

January 30, 1598 (25th year of Wanli reign, 12th month, 24th day)

Layout of Dosan Fortress, reconstructed to the best of my ability and superimposed on an aerial photo of the hill. The tone of the red colour indicates elevation of the fortress (lighter red indicates higher elevation). In Japanese castle terminology, this type of Japanese castle has a teikakushiki nawabari (梯郭式縄張).
By noon, the only Japanese fortification left standing was Dosan Fortress. Yang Hao and Ma Gui issued order to relocate the main siege camp to Hakseongsan (학성산 or 鶴城山), a small hill at the northwest of Dosan Fortress, so that they could personally oversee the battle from there. All three Ming divisions began a direct assault on Dosan Fortress. Ming troops utilised large grapping hooks to pull down some of the walls, then poured into the fortress through the gaps. At the same time, incendiary rockets engulfed the fortress in fire and smoke, killing thousands of Japanese troops and labourers. From the east side, Mao Guo Qi captured the gate of the obiguruwa (帯曲輪, lit. 'Sash compound', subsidiary ward of a castle) of Dosan Fortress, and began attacking the ninomaru (二の丸, lit. 'Second circle', secondary ward of a castle) of Dosan Fortress. From the west side, Chen Yin that broke through the outer palisades joined force with Ming Right Division. He tore through another layer of palisade, attacked and captured the sannomaru of Dosan Fortress, and began attacking the ninomaru as well. However, the Japanese commander guarding ninomaru was none other than Katō Kiyomasa himself. With his masterful command of arquebusiers, as well as the most elite core of Japanese army by his side, Katō Kiyomasa successfully resisted both Mao Guo Qi and Chen Yin.

Ming army attacking Dosan Fortress from three directions. Blue colour indicates captured portions of the fortress.
As the battle raged on, even Li Ru Mei himself partook in the assault. About 200 Northern troops under his command attacked the eastern gate of honmaru (本丸, lit. 'Main circle', main ward of a castle) directly. Unfortunately, while they broke through the gate and almost succeeded in capturing the most critical portion of Dosan Fortress, Southern troops (who did not see eye to eye with Northern troops) intentionally withheld support, and even Ma Gui forced Li Ru Mei to cancel the attack. Without adequate backup, the attack ultimately failed and most of the Northern troops perished in battle.

Japanese flotilla sailing from Seosaengpo Fortress to Ulsan Fortress (click to enlarge).
Ming army attempted to intercept the Japanese flotilla (click to enlarge).
At around 2 pm, a flotilla of 40 Japanese ships from Seosaengpo reached Ulsan, carrying much needed reinforcement and supply. The arrival of Japanese ships alerted the Ming army, so Yang Hao quickly dispatched 1,000 cavalry and 2,000 Zhejiang infantry to the river confluence to try to intercept the flotilla. Wu Wei Zhong also dispatched some Zhejiang infantry to cross Taehwa River (by wading through the shallowest part of the river at Jeontan) in order to intercept the flotilla from the other side (Asano Yoshinaga, who was in Dosan Fortress at the time, witnessed the river crossing from afar and mistakenly believed that Wu Wei Zhong's detachment was sent to intercept a Japanese relief force that came to Ulsan by land). Unfortunately, since Ming army did not have any warships, and Joseon navy under Yi Un-ryong was nowhere to be found, the intercept attempt was ultimately an exercise in futility. As Japanese ships inched closer to Dosan Fortress, Chen Yin, who was attacking the fortress from the west side, ordered his artillerymen to open fire at the ships, sinking 3 ships and successfully drove the rest away, but the Japanese flotilla did not give up easily and loitered outside the range of Chen Yin's guns until the end of the day.

Ming assault on Dosan Fortress would continue until around 4 to 5 pm, but it was already clear that the battle had turn into a deadlock and no new progress could be made. Realising that Ming troops were completely exhausted after entire day of fighting, Ma Gui requested Yang Hao to signal a retreat. Yang Hao agreed and ordered the Ming army to pull back. Meanwhile, the Japanese also realised that Ming troops were pulling out. Asano Yoshinaga immediately opened the western gate of honmaru, and Japanese troops poured out from the gate into ninomaru, killing dozens of Ming troops that could not evacuate the premises in time. Ōta Kazuyoshi's men reportedly killed and collected seven heads, whereas Asano Yoshinaga and Katō Kiyomasa's men collected one head each. In addition, Katō Kiyomasa ordered Katō Yasumasa to lead 500 troops and chase after the Ming army, until he was fully certain that Ming army would not return.

Known Ming positions at the end of January 30, 1598 (click to enlarge).
Position of Lu Ji Zhong at the end of January 30, 1598 (click to enlarge).
Ming army withdrew outside the range of Japanese arquebus but maintained encirclement of Dosan Fortress. Yang Hao and Ma Gui went to the main siege camp on Hakseongsan, Gao Ce (Middle Division) camped at the east side of Dosan Fortress, whereas Li Fang Chun (Right Division) camped at the west side. To prevent potential Japanese reinforcement from Busan, Wu Wei Zhong's detachment that crossed the river remained at the southern side of Jeontan (Wu Wei Zhong himself did not personally cross the river). They were supported by Zu Cheng Xun and Po Gui, both of whom were assigned to the northern bank of Jeontan. Meanwhile, Li Ru Mei and Bai Sai were assigned to guard the northern bank of Taehwa River near the fortress, while Lu Ji Zhong was sent to the river mouth of Taehwa River, about 6 km to the east of Dosan Fortress. As the day was already late, Ming army simply camped in the wild for the night. Yang Hao also ordered Ming artillerymen to continue firing for the entire night to disrupt the Japanese army.

Return trip of the Japanese flotilla (click to enlarge).
At around 7 pm, the tide of Taehwa River began to rise. Taking advantage of the fact that Ming encirclement was not yet airtight, Japanese flotilla caught the rising tide and rushed to Dosan Fortress. After unloading its troops and supply, the flotilla took in a few passengers from Dosan Fortress and then sailed eastwards to Japanese naval base at Yeompo (염포 or 鹽浦). Yang Hao caught wind of this incident and mistakenly believed that Katō Kiyomasa had escaped with the flotilla. He immediately gave chase to the flotilla along with Ma Gui, only to return empty-handed. After returning, Yang Hao called for a war council with other Ming commanders to discuss the next course of action, and it was decided that Ming army would attempt to burn down Dosan Fortress in the coming days. Ming troops would spend the rest of the day gathering firewood and making preparation for tomorrow's attack. 

Meanwhile, the Japanese did not sit idle inside their last remaining fortress either. After receiving fresh troops and supply, Katō Kiyomasa began overseeing the repair of Dosan Fortress. Drawing lessons from their nightmarish experience throughout the day, Japanese troops repaired damaged fortifications, plugged and remodelled various weak spots of the fortress, reinforced the gates, and added more gun loops to allow for higher concentration of arquebus fire.

Late into the night, a Ōta Kazuyoshi's subordinate named Tanaka Kozaemon (田中小左衛門) requested audience with Katō Kiyomasa and self-recommended to lead a night raid on Ming camps. Katō Kiyomasa approved his request. Tanaka Kozaemon silently slipped out of Dosan Fortress with 50 Japanese troops, only to discover that Ming camps were heavily defended. He returned to Dosan Fortress without actually attacking.

And thus ended the second day of the siege.

In mere two days, Ming army had decisively crushed the Japanese army in the field and captured all but one fortress of the entire Ulsan fortress complex. Yang Hao explicitly forbade the Ming army to count its gains and losses until the battle was won completely, and since Ming army failed to capture Dosan Fortress, no formal report was filed that day. Nevertheless, Yang Hao's own Ye Bu Shou (夜不收) reported to him that they managed to collect more than 800 Japanese heads, whereas other Ming sources report more than 1,200 heads taken. In his own assessment, Ma Gui wrote that Ming army only lost 33 troops as well as a few captains on the second day, but he was clearly downplaying the casualties. Actual Ming death toll is estimated to be in the range of several hundred.

On the flip side, most Japanese sources intentionally downplay the casualties, or simply leave it vague or unreported. Later period Japanese sources generally cite the Chinese assessment (i.e. 1,200 heads taken) and treat it as the total casualties suffered by Japanese army for the second day. Nevertheless, Ōkōchi Hidemoto (大河内秀元), witness and survivor of the siege, actually recorded that Japanese army suffered a mind-boggling 18,360 casualties on the first day alone. The staggering death toll likely caused the collapse of the entire Japanese defence, leading to rapid fall of Japanese fortresses on the second day.

(It should be noted that even though Ōkōchi Hidemoto was a firsthand witness, his account of the siege campaign contains embellishment, exaggerations and errors, and should be taken with a grain of salt. For example, he recorded that Ming army was 800,000-strong, and that it left behind 15,754 dead bodies during the retreat. He also believed that Kobayakawa Hideaki was the leader of the Japanese relief force and personally killed 13 Ming troops, yet in reality Kobayakawa Hideaki did not go to Ulsan and only sent Yamaguchi Munenaga in his stead. Nevertheless, Ōkōchi Hidemoto was still the only witness that did not deliberately keep the Japanese casualties ambiguous, so his witness account should not be dismissed out of hand. Japanese casualties on the fist day may not be as high as he claimed, but it's undeniable that they suffered catastrophic loss.)

Day 3: Fierce assault

January 31, 1598 (25th year of Wanli reign, 12th month, 25th day)

Like the previous day, the attack on Dosan Fortress began before dawn. Ming army launched its most ferocious attack on Dosan Fortress, commencing no less than seven assaults over the entire day. Wearing a green robe and carrying a white banner, Katō Kiyomasa personally commanded the defence of the fortress, leading the arquebusiers from the front and even participating in the shooting along with Asano Yoshinaga. 

Having received fresh reinforcement and supply, as well as repaired/upgraded the fortress, Japanese army was much better prepared than the day before. In contrast, hunger and exhaustion caused by being forced to camp in the wild began to take their toll on Ming army. As Ming troops pushed closer to the fortress, they were met with thunderous hails of arquebus fire. Such was the intensity of Japanese firepower that Katō Yasuyuki, who was armed with two arquebuses and aided by four loaders, single-handedly spent more than 280 rounds while defending the western portion of Dosan Fortress. Ming army suffered heavy casualties under fire, losing several captains, a Joseon commander, several righteous army leaders, as well as 200 dead and more than 1,000 wounded. Chen Yin, who was leading the vanguard, also suffered a serious gunshot wound in his right thigh outside the west gate of sannomaru, and had to be evacuated away from the battlefield. Realising that the attack had failed, Yang Hao had no choice but to call off the attack at around 7pm.

Japanese flotilla sailing from Yeompo naval base to Ulsan Fortress (click to enlarge).
Li Ru Mei engaging Japanese flotilla, supported by Gao Ce and Zu Cheng Xun (click to enlarge).
While the attack on Dosan Fortress was under way, another Japanese flotilla, this time sailing from Yeompo naval base, arrived at Ulsan and began approaching Dosan Fortress. Japanese arquebusiers on the ships opened fire at Ming troops guarding the river bank, which prompted Li Ru Mei to order his gunners to return fire, sinking several Japanese ships in the process. Yang Hao ordered Gao Ce and Zu Cheng Xun to reinforce the river bank, and Ming army successfully drove the flotilla away.

Later that day, Yang Hao dispatched a messenger carrying Shang Gong Qi (賞功旗, lit. 'Banner of reward') and Mian Si Tie (免死帖, lit. 'Badge of pardon') to Dosan Fortress to call for Japanese surrender, of which Katō Kiyomasa replied that he will only surrender if Koreans also agree to the terms. Yang Hao however saw through Katō Kiyomasa's ruse/stalling tactic and withdrew the offer.

Despite the failure of today's attack, Ming army managed to capture four Japanese troops. In addition, several Korean women and children that were taken captives by the Japanese successfully escaped. Through questioning them, Ming army acquired intelligence that food and especially water inside Dosan Fortress were running low.

Day 4: Respite

February 1, 1598 (25th year of Wanli reign, 12th month, 26th day)

A handful of Korean supplies from Gyeongju—enough to sustain the troops (but not the horses) for a few more days—managed to make its way to the frontline on the first day of February. Having received some supplies at last after enduring days of hunger and restless battles, Yang Hao decided to allow Ming army to rest for a day. He summoned Gwon Yul and informed him of the decision, but ordered Joseon army to continue the attack as well as bury all water wells near Dosan Fortress.

Gwon Yul duly followed the command and ordered Joseon army (along with some defected Japanese troops) to launch the attack. Joseon troops approached the fortress with their wooden shields held high, intending to pile up firewood and other flammable material at the base of the fortress and set it on fire. They were quickly beaten back by Japanese defenders and suffered heavy casualties, causing the army to fall into panic. To restore order and reassert control of the chaotic situation, Gwon Yul made a brutal example out of Jeonje (전제 or 全霽), district administrator of Youngsan, as well as several other officials. Their execution scared the rest of the Joseon army straight. Gwon Yul and Yi Deok-hyeong then went on to personally lead the assault. With renewed rigour, Joseon troops tore through the palisades of Dosan Fortress and moved closer to the base of the fortress proper. Unfortunately, hails of arquebus gunfire stopped even this determined assault dead in its track, and Joseon army was forced to retreat after suffering another wave of casualties.

Relief column from Seosaengpo Fortress and flotilla from Yeompo naval base going to Ulsan Fortress (click to enlarge).
Li Ru Mei engaging Japanese flotilla (click to enlarge).
While Joseon army was busy attacking Dosan Fortress, a Japanese relief column from Seosaengpo arrived at the outskirt of Ulsan and met with another Japanese flotilla coming from Yeompo. The two joined force and attempted to send supply and reinforcement into Dosan Fortress by water. However, their movement alerted Li Ru Mei, who promptly ordered Ming troops guarding the river banks to bombard the Japanese flotilla with cannons and rockets. After some exchange of fire, Japanese flotilla lost one ship to Ming barrage and was forced to retreat.

On the same day, Yang Hao once again dispatched a messenger carrying Shang Gong Qi and Mian Si Tie to Dosan Fortress. However, for this time the messenger went to the outskirt of Dosan Fortress and shouted for the entire Japanese garrison to hear, enticing them to surrender. One Japanese troop capitulated and ran away from the fortress to surrender to Ming army. Yang Hao was greatly delighted by the news and rewarded the defected Japanese troop handsomely. He gifted a fine steed to the Japanese troop and paraded him around Dosan Fortress to show off, enticing even more Japanese troops to defect. Powerless to stop Yang Hao's ploy, Katō Kiyomasa was forced to shut off the fortress entirely to prevent further desertion.

At around 3 pm, the weather suddenly changed. Heavy winds began blowing and rain started pouring. By evening, the rain had turn into a heavy downpour which persisted for the entire night. Due to the fact that it was winter season in Korea, many Ming troops that camped in the wild were drenched by freezing rainwater and frostbitten.

Day 5: Raging storm

February 2, 1598 (25th year of Wanli reign, 12th month, 27th day)

The downpour showed no sign of stopping, turning paddy fields surrounding Dosan Fortress into knee-deep mud fields. Early in the morning, a few Korean women and children once again escaped and were brought to Yang Hao for questioning. From their testimonies, Yang Hao learnt that Dosan Fortress had run out of food and water. Desperate Japanese troops had resorted to collecting charred remains of rice from the burnt down portion of the fortress to sate their hunger, as well sneaking out during the night to collect water at water wells around the outskirt of Dosan Fortress. Japanese morale was low, and even Katō Kiyomasa deeply regretted his decision of coming to defend Dosan Fortress. Nevertheless, heavy rain alleviated their water issue somewhat.

At around 8 am, a flotilla of 30 Japanese ships from Yeompo once again came to Ulsan under heavy rain and attempted to make a contested landing, alerting Ming army. Yang Hao and Ma Gui quickly assumed command at the main siege camp at Hakseongsan and mobilised the army. Joseon army was ordered to encircle Dosan Fortress to prevent Japanese reinforcement from breaking in, while Ming Left Division scrambled to intercept the flotilla and was soon joined by Zhejiang infantry guarding the river banks. After a protracted shoot-out with Ming troops, Japanese flotilla suffered heavy casualties and had to retreat.

Japanese flotilla once again sailing from Yeompo naval base to Ulsan Fortress (click to enlarge).
Ming Left Division and Zhejiang infantry scrambled to engage Japanese flotilla (click to enlarge).
Realising that their hope for receiving aid through Taehwa River was once again crushed, Japanese defenders inside Dosan Fortress had no choice but to contemplate a peace negotiation with Ming army. Minobu Kintaifu Kihachirō (美濃部金大夫喜八郎), Katō Kiyomasa's retainer and vice commander, wrote a letter of peace and ordered several Japanese troops to tie it to a bamboo pole with a white flag, then plant the pole at the outskirt of the fortress. The letter was retrieved by order of Yang Hao, who immediately noticed that it was merely a bluff to buy time (the letter attempted to deceive the Chinese into believing that Katō Kiyomasa was still at Seosaengpo, and even requested Ming army to dispatch a Joseon official to accompany Minobu Kintaifu Kihachirō to Seosaengpo for peace negotiation).

In turn, Yang Hao ordered Kim Ung-seo to bring Park Dae-geun (朴大根), a Korean translator, as well as two Japanese defectors, Tahara Shichizaemon (田原七左衛門) and Okamoto Echigo-no-kami (岡本越後守, believed to be the same person as famous Japanese defector Kim Chung-seon/Sayaka) to Dosan Fortress to make a counter-offer to the Japanese, claiming that if Katō Kiyomasa surrender to him, he would spare all the Japanese inside the fortress, and even reward Katō Kiyomasa handsomely. However, Japanese inside Dosan Fortress still tried to deceive the negotiators by insisting that Katō Kiyomasa was at Seosaengpo, and even demanded that Ming army lift the blockade of one side of the fortress and send them a commander as hostage before they would consider negotiation. Unsurprisingly, such outrageous demands were immediately rejected.

After the negotiation failed, Yang Hao held a war council to discuss the next course of action. He summoned Gwon Yul and Yi Deok-hyeong and, in a rare moment, actually praised the performance of Joseon troops for the past few days. He then ordered Gwon Yul to lead the elites of Joseon army to conduct a feint attack, and asked Yi Deok-hyeong to organise Joseon troops that did not participate in the fighting to collect grass for use by Ming army. 

After receiving Yang Hao's order, Gwon Yul ordered Joseon army to launch the feint attack. Joseon cavalry rode around Dosan Fortress and fired warning shots with their guns constantly, forcing the Japanese to be constantly on their toes, as well as baiting them to waste their shots. While the feint attack was ongoing, Joseon army also launched a probing attack on the gate of sannomaru of Dosan Fortress. Katō Yosaemon, who was in charge of the defence of sannomaru, deployed taketaba (たけたば or 竹束, lit. 'Bamboo bundle') and ordered his arquebusiers to repel them. While the attack was unsuccessful, Joseon troops managed to pile up firewood at the base of sannomaru before retreating, hoping that they could set fire to the fortress once heavy rain subside. 

At around 4 pm, Yang Hao once again sent Kim Ung-seo and Japanese defectors to Dosan Fortress to call for surrender, although this time they were simply ignored. As the sky turned dark, Yang Hao, concerned that Japanese troops may attempt another night raid, ordered Ming troops to stand guard over the night, and even specifically ordered Joseon troops to stand guard together with their Chinese counterparts. On top of that, Yang Hao also sent Kim Ung-seo and his Japanese defectors to patrol the water wells near Dosan Fortress and attempt to entice defection of Japanese troops that came to collect water. Despite all these defensive arrangements, Yang Hao was still paranoid about the lie about Katō Kiyomasa's whereabouts, so he repeatedly summoned Korean captives for questioning to ascertain that Katō Kiyomasa was really inside Dosan Fortress. Afterwards, Yang Hao and Ma Gui returned to the main siege camp at Hakseongsan to oversee the construction of semi-permanent thatched shelters (up until now, entire Ming army including its highest-ranking commanders simply camped in the wild).

Just as Yang Hao predicted, Katō Yosaemon and Kondō Shirō Goemon sallied out of the fortress under the cover of night to launch a night raid on Ming camps, although they quickly discovered that both Ming and Joseon troops were on high alert and had to abort the plan. However, as they were returning to Dosan Fortress, they preemptively burned away the firewood.

Day 6: A turn for the worse

February 3, 1598 (25th year of Wanli reign, 12th month, 28th day)

Ming army launched an artillery barrage against Dosan Fortress from across Hakseongsan (click to enlarge).
The heavy rain slowly subsided on the sixth day. According to some Japanese sources, on this day some Ming troops moved their cannons to Hakseongsan and launched an artillery barrage across the hill into Dosan Fortress. Several Japanese troops were pulverised by cannon fire, causing a panic to spread. Japanese troops attempted to run for cover, but they were stopped by Katō Kiyomasa, who remained unfazed even as some cannonballs landed near him. Katō Kiyomasa's calmness and non-reaction misled Ming artillerymen into believing that their shots missed the mark, so they adjusted their guns and launched the next salvo with a higher trajectory, overshooting Dosan Fortress. It was at this point that Katō Kiyomasa ordered his troops to act as if they were getting shot at to further deceive the Chinese. As a result, Ming artillerymen continued to fire their cannons using high trajectory, missing most of their shots. 

It should be noted that Chinese and Korean sources make no mention of such artillery barrage taking place, as Ming army left its artillery train behind (to its own detriment). If this artillery barrage actually happened at all, it is likely that Ming artillerymen only employed a few lightweight anti-personnel pieces such as Fo Lang Ji (佛狼機) for suppressive fire, rather than heavy artillery for wall-smashing bombardment. Incidentally, some Korean sources do mention Japanese troops shooting at Hakseongsan with "cannonballs as large as chicken egg", presumably from an ō-deppō (大鉄砲) handheld matchlock cannon and nearly hitting Yang Hao (he was equally unfazed), although no specific date was given for this barrage. Perhaps Japanese barrage was a counter-barrage in response to Ming bombardment, although the possibility that these were unrelated incidents, or even propagandistic embellishments that did not actually happen, cannot be ruled out.

Like the days before, Joseon troops launched an assault on Dosan Fortress, and once again they were repelled by the Japanese. At around 11 am, yet another Japanese flotilla arrived at Ulsan and began sailing closer to Dosan Fortress. Japanese troops inside the fortress soon noticed something unusual about today's flotilla: instead of Katō Kiyomasa's own banners, these ships were flying the uma-jirushi (馬印, lit. 'Horse insignia') of Yamaguchi Munenaga and Mōri Katsunobu.

This can only mean one thing—new Japanese reinforcement was under way.

Yamaguchi Munenaga and Mōri Katsunobu's flotilla sailing from Seosaengpo Fortress to Ulsan Fortress (click to enlarge).
Ming troops at the river bank engaging Japanese flotilla (click to enlarge).
As it turned out, before Katō Kiyomasa rush back to Dosan Fortress on January 29, 1598 (Day 1), he sent out messengers to surrounding Japanese commanders to call for aid. Japanese commanders at Yangsan and Busan, being the closest to Seosaengpo, received the call on January 30, 1598 (Day 2) and dispatched their relief forces the soonest. Yamaguchi Munenaga (from Yangsan) and Mōri Katsunobu (from Busan) arrived at Seosaengpo on the evening of February 2, 1598 (Day 5), and sent out their ships the next day. Since both of them had only just arrived, the flotilla only scouted the area and briefly communicated with Japanese troops trapped inside Dosan Fortress using flag signals before reporting back to Seosaengpo. Nevertheless, signs of incoming help raised hopes for Japanese troops inside Dosan Fortress. At around 6 pm, additional relief forces led by Kuroda Nagamasa, Ankokuji Ekei, and Takenaka Shigetoshi also arrived at Seosaengpo.

Meanwhile, Katō Kiyomasa's messenger only just arrived at Suncheon Fortress (which was farther away from Seosaengpo). Incidentally, construction of Suncheon Fortress was completed on the same day, and Shimazu Yoshihiro (島津義弘) with his son Shimazu Tadatsune (島津忠恒) were hosting a celebratory feast with other Japanese commanders inside Suncheon Fortress when they received the news about Ulsan under siege. Due to Xing Jie's diversionary attack, Japanese commanders at Suncheon Fortress were reluctant to send out their full force to relief the siege. Kakimi Kazunao even specifically ordered Shimazu Yoshihiro to stay behind to defend the fortress, and only went to Ulsan with his own troops. Perhaps not wanting to lose reputation for not sending out help, Shimazu Yoshihiro later wrote a letter to his nephew Shimazu Toyohisa, asking him to go to Ulsan in his stead. He also dispatched a few retainers and 50 arquebusiers to reinforce Shimazu Toyohisa's 500 troops.

The rain finally stopped by nightfall, only to be replaced by strong westerly winds. The freezing winds took a serious toll on Ming troops, in particular Zhejiang infantry guarding the river banks. At midnight, Katō Shigetsugu (加藤重次), Shōbayashi Hayato (庄林隼人) and Kondō Shirou Goemon (近藤四郎右衛門), leading a contingent of 100 mounted samurai and 300 arquebusiers, sallied out of the eastern gate of ninomaru of Dosan Fortress to harass Ming army. They launched a few volleys of fire arrows and arquebus shots at Ming camps at the east side of Dosan Fortress before turning back.

Day 7: Renewed attack

February 4, 1598 (25th year of Wanli reign, 12th month, 29th day)

The strong wind that began to blow since last night persisted into the seventh day of the siege. At dawn, Yang Hao ordered Ming army to gather more firewood in preparation for another attack on Dosan Fortress, as he felt the strong wind could be helpful in spreading the fire. At noon, another 26-ship flotilla from Yeompo arrived at Ulsan and began approaching Dosan Fortress, and once again Ming troops guarding the river banks bombarded the flotilla with cannons. While both sides were busy exchanging fire against each other, one samurai and a few of his followers suddenly dashed out of Dosan Fortress to the river bank and began shouting to the flotilla. Although he did not understand Japanese, Yang Hao was alerted enough that he immediately dispatched Ming Right Division to reinforce Wu Wei Zhong's Zhejiang infantry at the river banks, as well as asking them to be extra vigilant. Ming army repelled the flotilla at around 5 pm.

Japanese flotilla sailing from Yeompo naval base to Ulsan Fortress for the fourth time (click to enlarge).
At around 6 pm, Yang Hao ordered Ming army to get ready for the attack. Ming troops silently approached Dosan Fortress under the cover of twilight, carrying firewood and with their shields readied. Unfortunately, they were spotted by Japanese sentries as they came near the palisades of Dosan Fortress, and were forced to retreat after suffering heavy casualties under fusillades of arquebus fire. By nightfall, Ming and Joseon army launched a second attack with much greater ferocity than before, ignoring casualties inflicted by Japanese arquebusiers and moving closer to Dosan Fortress. Realising that they may not be able to resist the assault with arquebus fire alone, Japanese defenders were forced to bust out the ō-deppō to supplement their firepower. The battle raged on until around 9.30 pm before Ming army retreated.

After today's attack, Yang Hao appeared to have a change of mind. He probably felt that trying to take the fortress by force was no longer practical after days of consecutive failures, and wanted to adjust his strategy to that of encirclement and investment. To this end, Yang Hao ordered his troops to upgrade their tents into semi-permanent thatched shelters, and urged Gwon Yul and Yi Deok-hyeong to hasten the next delivery of supply. Ma Gui suggested Yang Hao to lift the blockade of one side of Dosan Fortress, then ambush the Japanese as they came out. However Yang Hao rejected his idea.

Late into the night, a small Japanese ship sneaked back to Dosan Fortress, presumably due to the shouting communication earlier in the day. About 30 Japanese troops came out of Dosan Fortress and attempted to board the ship. However, they were immediately attacked by Ming troops laying in ambush. After a brief skirmish, Wu Wei Zhong's Zhejiang infantry managed to behead six Japanese troops, whereas Ming Right Division troops beheaded one. The rest ran back into Dosan Fortress.

In the meantime, Mōri Hidemoto finally arrived at Seosaengpo at around 4 pm. Japanese commanders at Seosaengpo held a war council that night to decide their next course of action—whether to send a relief force to Dosan Fortress immediately, or wait for further Japanese reinforcement to arrive before sending out help. They concluded that an advance party would be dispatched the next day, but the majority of Japanese forces should stay and wait for more reinforcement.

Day 8: Living hell

February 5, 1598 (25th year of Wanli reign, 12th month, 30th day)

It was February 5, 1598, last day of the year on Lunisolar calendar, and the eighth day of the siege. Dosan Fortress already ran out of food and water, and days of exhaustion, starvation, dehydration, and freezing winter turned the fortress into a living hell (in fact the Japanese only managed to survive this long thanks to the heavy downpour of the past few days alleviating some of their water shortage problems). Japanese troops had slaughtered the last of their pack animals for food, and resorted to munching paper and dirt on the wall for subsistence, as well as drinking urine to sate their thirst. They were so desperate that some sneaked out the fortress to scavenge for leftover field ration on the frozen corpses of their fallen comrades-in-arms, while others drank water from ditches full of dead bodies and blood. Only the highest ranking Japanese commanders could have some semblance of meal, but even them had to ration their food into pinches of rice. To stave off the cold, samurai, ashigaru, and levied labourers disregarded their social standing and bunched together in groups of dozens of people each. Despite their best effort, incidents like people unexpectedly drop dead due to the cold or froze to death in their sleep were daily occurrence. In his diary, Ōkōchi Hidemoto recorded that he was malnourished to the point of becoming literal skin-and-bone, and that a friend of his, who he described as a large and burly man, refused to take off his armour to conceal his rapidly emaciating frame.

While Japanese troops were suffering inside their fortress, Chinese troops besieging them hardly fared any better. Despite some Korean sources claiming that Ming army was well-supplied, the reality of Korean logistics mismanagement cannot be made more apparent by multiple reports from Joseon officials actually present at the frontline. To sum it up, Yi Yong-sun (이용순 or 李用淳), overseer of the entire supply operation, ignored his duty and returned to Gyeongju for no reason, and many Joseon quartermasters were not doing their jobs at all. Ming army ran dangerously low on food, and the warhorses, which had not eaten anything for nine straight days, were dying by the hundreds. In fact the situation was so bad that Jang Un-ik and Yi Deok-hyeong, two of the highest ranking Joseon officials, had to take the matter into their own hands and do the jobs of their inferiors. Despite their best effort, they only managed to secure stable food supply for the highest ranking Ming commanders, Yang Hao and Ma Gui (as the Koreans were afraid to anger them). Even Ming commanders had depleted their salt and sauce provisions with no hope of resupply, and the rank and file were simply left to their own devices. The death of more than a thousand warhorses (and rapid weakening of the rest) had a disproportionate impact on combat readiness of the predominantly-mounted Ming army, not to mention the exhausted and starving Ming troops exposed to freezing temperature for days. Joseon officials were painfully aware that given the sorry condition of Ming army, the entire siege campaign would be thrown into jeopardy the moment Japanese reinforcement show up. Unfortunately, they were incapable of salvaging the situation.

Japanese advance party and flotilla from Seosaengpo went to Ulsan (click to enlarge).
Leaving aside the suffering of both armies, by 7 am in the morning, Japanese advance party from Seosaengpo had arrived at Ulsan by land and set up a camp on a hill known as Songsan (송산 or 松山), some 12 km away from Dosan Fortress. Meanwhile, another Japanese flotilla also arrived at Ulsan. Unlike previous days, it did not attempt to break through the encirclement, but waited patiently on the river.

Later in the day, Katō Kiyomasa wrote a letter to Ming army, informing the Chinese of his intention to negotiate for surrender. He requested permission to let in a Buddhist monk on the Japanese flotilla to serve as translator and draft the written agreement, on the pretense that no one in Dosan Fortress spoke Chinese. On the Ming side, Yang Hao already knew that taking Dosan Fortress by force would be difficult, and his troops were in a terrible shape, so when he received news that the Japanese were trying to sue for peace, he gladly accepted it and let the Buddhist monk enter Dosan Fortress. After some back and forth communication, both sides agreed to hold a peace talk in three days, in which Katō Kiyomasa was required to personally attend. Little did Yang Hao know, the flotilla was not Katō Kiyomasa's, but a scouting flotilla sent by Mōri Hidemoto, Kuroda Nagamasa, Yamaguchi Munenaga and Takenaka Shigetoshi. It had already made contact and coordinated with Japanese troops inside Dosan Fortress using flag signals beforehand, so when the "Buddhist monk" was allowed to enter the fortress, he brought with him the news of the arrival of Japanese relief force, which boosted the morale of Japanese troops tremendously. The flotilla returned to Seosaengpo afterwards.

(It should be noted that Korean and Japanese sources strongly disagree on which side initiated the negotiation. Many Japanese sources claim that it was the Chinese that initiated the peace talk after being repeatedly defeated, and the peace talk was initiated on Day 7 rather than Day 8. However, it makes no logical sense that the Chinese would sue for peace during the day and then immediately attack in the same evening, so Japanese version of the event simply cannot be true. This logical problem did not go unnoticed by Japanese historians either. However, instead of acknowledging the mistake, later Japanese sources such as Nihon Senshi simply alter the time of the Day 7 attack from evening to morning to forcibly harmonise the discrepancy.)

By noon, additional reinforcement led by Nabeshima Naoshige and his son Nabeshima Katsushige, as well as Hachisuka Iemasa, Ikoma Kazumasa, Katō Yoshiaki, Wakisaka Yasuharu and Hayakawa Nagamasa, also arrived at Seosaengpo.

Later in the night, Katō Kiyomasa dispatched two parties out of Dosan Fortress to conduct a so-called "night raid" mission. Despite the name, they were actually foragers desperately looking for food and water. One of the parties, of about 30 men strong, headed straight to a water well near the outskirt of Dosan Fortress to collect water. Unfortunately, they ran into Korean commander Kim Ung-seo and a contingent of defected Japanese troops under his command, who were guarding the water well since Day 4. Kim Ung-seo immediately attacked them, killing five and captured another five. Yang Hao was notified of the incident and immediately ordered the captives to be brought to him and interrogated. After witnessing the severely famished captives, and learnt of the dire situation inside Dosan Fortress (according to the captives, Dosan Fortress had completely run out of food and water, and out of about 10,000 starving souls still alive, less than one thousand were combat capable.), Yang Hao was finally able to put his mind at ease, confident that the Japanese wouldn't be able to hold out for much longer.

Day 9: An uneasy new year

February 6, 1598 (26th year of Wanli reign, 1st month, 1st day)

Korea welcomed its Lunisolar New Year in tension and uneasiness. Nevertheless, Joseon King Yi Yeon (이연 or 李昖, more commonly known as King Seonjo) personally visited Xing Jie to give him a new year greeting. Xing Jie gifted the king with two calligraphic scrolls to congratulate him in advance on the victory of Imjin War, in return King Yi Yeon wished him a prosperous new year and congratulated him in advance on the victory at Ulsan. The two had a great time together. 

In stark contrast, Ryu Seong-ryong went to Ulsan at dawn to give Yang Hao and Ma Gui his new year greeting only to be met with cold shoulder, as no one at the frontline was in any mood of celebration. Yang Hao urged him to return to Gyeongju as soon as possible to send in the next delivery of supply, going so far as to declare that the supply problem was so critical that even a single dan of rice should be delivered to the frontline at the double.

The Japanese however were in even more sombre mood compared to the Chinese and Koreans. Dosan Fortress was literally hanging by a thread, with less than 6,000 Japanese troops alive and casualties mounting at an alarming rate due to starvation and freezing temperature. Katō Kiyomasa and Asano Yoshinaga wrote a joint distress letter to seven Japaneses commanders, among them Kobayakawa Hideaki and Mōri Hidemoto, to explain the dire situation inside the fortress. They stated that if reinforcement didn't come soon, the entire garrison of Dosan Fortress was prepared to fight to the last man. If the fortress fell, they hoped that the recipients of the letter can at least bring the news of their valiant last stand back to Japan.

Zu Cheng Xun crossed Taehwa River (click to enlarge).
Later in the day, Ming mounted scouts detected Japanese presence at Songsan. Since Japanese advance party was few in numbers, Yang Hao only ordered Zu Cheng Xun (that guarded the northern bank of Jeontan) to cross Taehwa River to join force with Wu Wei Zhong's detachment to monitor Japanese activities from afar.

Meanwhile, Shimazu Toyohisa arrived at Eonyang and captured Eonyang Fortress after defeating a small garrison there. At around 2 pm, Kakimi Kazunao and Kumaga Naomori arrived at Seosaengpo. They were followed by Chosokabe Motochika, Nakagawa Hidenari, Ikeda Hideuji, Ikeda Hideo, Mōri Katsunobu, Mōri Katsunaga, Akizuki Tanenaga, Takahashi Mototane, Itō Suketaka and Sagara Yorifusa, who arrived in succession at around 4 pm. Later in the evening, two retainers of Asano Yoshinaga and one retainer of Ōta Kazuyoshi slipped out of Dosan Fortress and delivered the distress letter to Seosaengpo, arriving by nightfall. Upon receiving the letter, Japanese commanders at Seosaengpo held an emergency meeting, and decided that they couldn't wait any longer—the relief force would immediately depart the next day. Mōri Hidemoto also initiated a joint letter, signed by Japanese commanders at Seosaengpo, to report the current status of Dosan Fortress and Seosaengpo to Hideyoshi.

Day 10: Reinforcement

February 7, 1598 (26th year of Wanli reign, 1st month, 2nd day)

Japanese relief force marched to Ulsan while Japanese flotilla sailed to Yeompo Naval Base (click to enlarge).
On February 7, 1598, the Japanese army that gathered at Seosaengpo finally made its move. The massive relief force departed for Dosan Fortress by both land and water route. On land, Nabeshima Naoshige and Kuroda Nagamasa led the First Division and departed first, with Hayakawa Nagamasa, Kakimi Kazunao, Kumaga Naomori and Takenaka Shigetoshi acting as ikusa metsuke (軍目付, senior military supervisor/army superintendent). Katō Yoshiaki, Nakagawa Hidenari, Ikoma Kazumasa, Wakisaka Yasuharu, Yamaguchi Munenaga, Ikeda Hideo led the Second Division and departed after them, and Mōri Hidemoto led the Third Division and departed last. On the water, Chosokabe Motochika and Ikeda Hideuji set sail to Yeompo to rendezvous with Katō Kiyomasa's fleet, with Katō Kiyomasa's troops that remained at Seosaengpo hitching a ride on their ships.

Beside the main relief force from Seosaengpo, there were other Japanese commanders that bypassed the Seosaengpo gathering and headed to Dosan Fortress directly. Shimazu Toyohisa that arrived at Eonyang on Day 9, as well as Kikkawa Hiroie and Mōri Takamasa, who arrived on Day 11, were some of the more notable examples. Tōdō Takatora (藤堂高虎), who just completed the construction of Suncheon Fortress, dispatched his adopted son Tōdō Takayoshi and subordinate Tōdō Yoshikatsu to relief Ulsan. Matsuura Shigenobu, who also took part in the construction of Suncheon Fortress, personally led his troops to Ulsan. Kurushima Hikozaemon from the currently-leaderless Kurushima Michifusa (来島通総)'s fleet (Kurushima Michifusa was killed in Battle of Myeongnyang), as well as Kan Uemonpachi, son of naval commander Kan Michinaga (菅達長), also brought a portion of their respective fleets to support Ulsan.

Bai Sai set up a second line of defence at Jeontan while Mao Guo Qi moved to guard the river bank (click to enlarge). It should be noted that Mao Guo Qi's movement is merely an educated guess, as his exact whereabouts before today's order was unknown. Nevertheless, given that Mao Guo Qi was part of Ming Left Division and fought together with Li Ru Mei on Day 2, it's likely that he stayed with Li Ru Mei.
As Japanese relief force gradually gathered at Songsan, the once-small camp set up by the advance party now became filled with all sort of war banners. Upon receiving scout report of this sudden increase in activities, Yang Hao immediately realised that something was not right. He ordered Bai Sai to lead a contingent of cavalry to reinforce Po Gui at the northern bank of Jeontan and set up a second line of defence, as well as sending Mao Guo Qi and his southern troops to reinforce Wu Wei Zhong's Zhejiang infantry and guard the river bank around Dosan Fortress.

In stark contrast to the quick reaction to Japanese activities at Songsan, the massive increase of ships and activities at Yeompo somehow went unnoticed by Lu Ji Zhong guarding the the river mouth only several kilometres away from it. Perhaps this was due to carelessness, or perhaps daily harassment of Japanese flotilla numbed his sense of danger. In any case, Yang Hao remained oblivious to the danger and made no arrangement to defend against Japanese reinforcement from the river, and this grave mistake ultimately costed him the entire siege campaign. Yang Hao was so furious over this negligence that he sacked Lu Ji Zhong later.

Later that night, Katō Kiyomasa once again dispatched his retainer to Songsan to plea for help.

Day 11: Day of reckoning

February 8, 1598 (26th year of Wanli reign, 1st month, 3rd day)

February 8, 1598 was the day when the peace talk was due to take place. Yang Hao dispatched a messenger to the outskirt of Dosan Fortress and urged Katō Kiyomasa to come out and attend the negotiation. The Chinese had no intention to actually negotiate with the Japanese, however. The peace talk was but a ploy to lure out and capture Katō Kiyomasa. 

Nevertheless, Katō Kiyomasa refused to show up. Yang Hao's messenger was greeted by Ōta Kazuyoshi from inside the fortress, who gave an excuse that Katō Kiyomasa, along with other high-ranking commanders inside the fortress, were gravely ill, so he couldn't attend the negotiation personally, and no one was healthy enough to act as his representative. And thus the peace talk fell apart before it even started.

It should be noted that different Japanese sources give different reasons on why Katō Kiyomasa went back on his word. Some sources claim that a Japanese defector in the Ming army secretly leaked the ploy to him, while other sources claim that he was stopped by Asano Yoshinaga and his own subordinates at last minute. Regardless of the actual reason, Katō Kiyomasa took a great gamble in doing so. Dosan Fortress was literally on the brink of capitulation, so botching the peace talk at this critical moment put the entire fortress at risk of being massacred should Ming army successfully capture it later. Fortunately for him, the gamble ultimately paid off. Help would arrive on this very day.

Japanese fleet sailed to Ulsan, cutting off Lu Ji Zhong in the process (click to enlarge).
Early in the morning at around 4 am, Japanese ships that gathered at Yeompo—now a full-fledged war fleet, began to set sail to Dosan Fortress. Meanwhile, Joseon naval commander Yi Un-ryong, who stayed at Gyeonju doing absolutely nothing over the entire duration of this siege campaign, suddenly remembered he had a untouched fleet at his disposal, and went to reconnoitre the river. What laid before his eyes was nothing short of terrifying, as he witnessed hundreds of Japanese ships entering Namgang River (남강 or 藍江, lit. 'Blue River', not to be confused with another more famous Nam River) from the sea and began sailing upstream, thronging the entire river with hulls and masts. Shocked by the discovery, Yi Un-ryong wrote an emergency report to Ryu Seong-ryong (who was at Gyeongju), and then fled as far as he could. Not only Yi Un-ryong did not engage the Japanese to delay their advance, nor retreat to Ulsan to prepare for a defensive engagement under more favourable conditions (i.e. with Ming support from land), he did not even bother to send a warning to Ming army about the incoming Japanese fleet. His selfish and cowardly action, along with Lu Ji Zhong's negligence, wasted away precious time that Ming army could have use to prepare for a countermeasure. As a result, at around 4 pm Japanese fleet arrived at Ulsan in force and blockaded entire section of Taehwa River and Dongcheon River. By that time, Ming army was powerless to stop them (Yang Hao was so embittered by the incident that he later urged the Koreans to build more warships).

Japanese relief force set up a new camp and began harassing Zu Chen Xun and Wu Wei Zhong's detachment (click to enlarge).
Around the same time, the last of the Japanese relief force, Kikkawa Hiroie and Mōri Takamasa, finally arrived at Ulsan. With the relief force finally assembled, Japanese Second and Third Division marched north and set up a new camp on a hill just south of the southern bank of Jeontan, right before the camp of Zu Cheng Xun and Wu Wei Zhong's detachment. Trapped between the new camp and Japanese fleet occupying Taehwa River behind their back, both of them were cut off from the rest of the Ming army. Worse yet, they were now being harassed by multiple Japanese war parties, each numbering 50 to 60 troops, from the new camp.

First Division of the Japanese relief force attempted to cross Taehwa River only to be beaten back by Li Ru Mei and Jie Sheng (click to enlarge).
Seeing that Zu Cheng Xun and Wu Wei Zhong's detachment were pinned down, Kuroda Nagamasa, Hachisuka Iemasa, Nabeshima Naoshige of the First Division, along with Kikkawa Hiroie and Mōri Takamasa, decided to march straight to Dosan Fortress. With support from Japanese fleet on the river, they attempted to forcibly cross Taehwa River at a river bank about 2 km west of Dosan Fortress. Unfortunately, their attempt was quickly discovered by the Ming army. Li Ru Mei and Jie Sheng led a contingent of cavalry, along with a number of Joseon troops, and attacked them. After a fierce battle, First Division was beaten back to the southern side of Taehwa River.

Standoff at Taehwa River (click to enlarge).
At this point, Taehwa River was filled to the brim with Japanese ships, and the entire length of the southern river bank was bristle with Japanese troops and banners. Yang Hao realised that his was quickly running out of options. He could either call for a full retreat before Japanese relief force cross the river en masse, or gamble everything and attack Dosan Fortress one last time, hoping that the defeat of Katō Kiyomasa could seriously damage the morale of the Japanese, allowing Ming army to deal with them somehow. Yang Hao picked the second option, and ordered Ming army to prepare torches for the night attack. However, before Ming army launch its final assault, Yang Hao had to make a few adjustments to Ming deployment. He tasked Ming Right Division with maintaining the encirclement of Dosan Fortress as well as preparing for the attack, and drew up the rest of the Ming army into three defensive positions.
  1. Bai Sai and Po Gui remained at Jeotan, keeping a look-out against Japanese relief force in the new camp. Joseon commanders Yi Si-eon and Seong Yun-mun and some Joseon troops were sent to assist them.
  2. Li Ru Mei, Jie Sheng and some Joseon troops defended the river bank between Dosan Fortress and Jeontan against potential landing of Japanese fleet on Taehwa River, as well as potential river crossing of Japanese troops from the other side.
  3. Southern troops under Wu Wei Zhong and Mao Guo Qi were stationed around the river confluence, guarding against both Japanese troops at the southern bank of Taehwa River and Japanese fleet on Dongcheon River.
Final repositioning of Ming army before it launches its last attack (click to enlarge).
As for Zu Cheng Xun and Wu Wei Zhong's detachment at the southern bank of Jeontan, they were more or less abandoned and left to their own devices. Regrettably, since the only naval power on the coalition side had fled, Ming army without any ships of its own would be hard-pressed to mount a rescue operation across the river while being threatened from at least three directions (i.e. Japanese defenders inside Dosan Fortress, Japanese fleet on the river, as well as Japanese relief force troops at the southern bank of Taehwa River). Ironically, as dangerous as their position was, they were still somewhat better off than Lu Ji Zhong, who was positioned farther away from the main army. Lu Ji Zhong was trapped between Yeompo naval base, Japanese relief force at Ulsan, not to mention Japanese fleet on Dongcheon River could cut off his line of retreat at any moment. In fact, Ming army lost all contact with Lu Ji Zhong after Japanese fleet blockaded the rivers, and no one knew what happened to him for the remainder of the siege campaign.

While Ming army was busy preparing, Mōri Hidemoto dispatched two of his retainers to sneak into Dosan Fortress (the fact that Katō Kiyomasa and Mōri Hidemoto were able to sneak messengers in and out of Dosan Fortress so easily shows the rapid weakening of Ming troops due to lack of supply. Katō Kiyomasa had to ask for Chinese permission to let in the "Buddhist monk" just a few days prior, but now the encirclement had become extremely porous). The messengers updated Katō Kiyomasa with the current status of the relief force and its planned rescue operation in the coming days, encouraging the defenders to hold out just a little longer.

Later that night, Shimazu Toyohisa left Eonyang and marched towards Ulsan.

Day 12: Final Attack and Retreat

February 9, 1598 (26th year of Wanli reign, 1st month, 4th day)

Ming army finished its preparation by midnight. Then, under the leadership of Yang Hao, it launched one final attack on Dosan Fortress with a ferocity second only to Day 3 assault. Ming artillerymen relentlessly battered Dosan Fortress, setting the fortress alight, while Ming troops set up siege ladders and attempted to climb over the wall. Japanese defenders inside Dosan Fortress poured hails of arquebus and ō-deppō rounds into Ming ranks, and cut down anyone that managed to climb up. Unfortunately, days of starvation, camping in the wild under heavy rain and freezing wind, and the death of hundreds of warhorses had taken a heavy toll on the morale and combat readiness of Ming troops. In stark contrast, Japanese troops were able to put everything on the line and fought with crazed determination, knowing full well that help was within reach. As a result, Ming army suffered heavy casualties without making significant progress, and began to show signs of faltering and fear. Realising the hesitation of his men, Yang Hao executed several retreating troops for cowardice, and ordered the faltering cavalry commander Li Hua Long to be tied up and paraded in front of the army to maintain discipline. Ming army resumed attack with increased ferocity. Many Ming captains led their men from the front during the attack, often losing their lives or being seriously wounded in the process. Among the fallen was Yang Wang Jin, a brigade commander from Datong leading 1,000 cavalry, and the highest ranking Ming officer to perish in the siege campaign.

A flotilla of 90 Japanese ships on Taehwa River began to slowly sail upstream (click to enlarge).
While the battle raged on, a letter claiming that as many as sixteen Japanese commanders from Gadeok Island, Angolpo, Jugdo, Busan, and Yangsan had arrived at Ulsan with 60,000 troops was intercepted by Ming Right Division, which reported it to Yang Hao and Ma Gui. At the same time, Yang Hao also received report that a flotilla of about 90 Japanese ships split from the main fleet and was sailing upstream of Taehwa River (i.e. sailing pass Jeontan), and could easily make uncontested landing beyond the reach of Bai Sai and Po Gui and then attack the rear of the Ming army from the west. Realising that Ming army was under serious risk of being surrounded, Yang Hao had a brief discussion with Ma Gui, and finally decided to call off the attack and make a full retreat.

Shocked upon hearing the news about the retreat, Yi Deok-hyeong and a number of Joseon officials rushed to Ming siege camp and passionately tried to persuade Yang Hao against the idea. The Koreans even suggested that Ming army should send a 10,000-strong detachment to occupy the (what they considered) advantageous ground around Jeontan and the road to Eonyang, and engage the Japanese in open battle. However, the foolish suggestion failed to take into consideration that Japanese had total naval supremacy, and could make landing at any point of the entire length of Taehwa River and even Dongcheon River, so defending only the west side of Dosan Fortress was completely pointless. Yi Deok-hyeong also conveniently left out the fact that Joseon troops, who were so unreliable that nearly one-half of the army deserted at the time Ming army still had the upper hand, were now fleeing en masse in light of the worsening situation. 

Unsurprisingly, Yang Hao and Ma Gui knew the unreliability of their ally very well after fighting side-by-side with the Koreans for so long. Given that Joseon army had abandoned the Ming army, they obviously had no intention of being treated like cannon fodder by the Koreans. At around 7 am in the morning, the order to cease attack and prepare for retreat was formally issued. By 9 am, the order to retreat was also issued, and Ming army began a full retreat. Ming infantry, as well as wounded troops, were the first to leave the battlefield, crossing Dongcheon River and heading east towards Gyeongju, whereas Ming troops surrounding Dosan Fortress lifted the siege and retreated northwards into the mountains. Cavalry commanders such as Bai Sai and Po Gui at Jeontan, as well as Li Ru Mei and Jie Sheng at the river bank west of Dosan Fortress, were ordered to act as rearguard and cover the retreat for the rest of the army. Yang Hao also ordered Yang Deng Shan to lead a contingent of cavalry to support Bai Sai and Po Gui at Jeontan. By 3 pm, the majority of Ming army (with the exception of rearguard and Yang Hao himself) had left, so Yang Hao ordered the dismantling of the main siege camp on Hakseongsan and prepare for the retreat of his own troops.

Ironically, the supply that Yang Hao had been repeatedly asking for was finally delivered to the frontline earlier today, although at this point the supply was nothing but extra burden for the retreating Ming troops to carry back to where it came from. Since Ming army was unable to carry all the supplies during the retreat, Yang Hao had to order the leftover to be burned down to prevent them from falling into Japanese hands. Despite this, Yang Hao still could not put his mind at ease. While waiting for the rearguard cavalry to catch up with his troops, he personally went to the stockpile area of the supply (at the foothill of Hakseongsan) to make sure that everything was properly burnt down. Yang Hao also ordered his own servants to seek out stragglers and cavalrymen that lost their mounts, as well as scouring the battlefield to collect discarded armours, weapons, and other materials to be burned. The destruction of Ming war material was so complete, that the Japanese did not even find trash inside the dismantled siege camp when they visited the site days after, although this also caused further delays to Yang Hao's own retreat. After making sure that nothing of value was left behind, Yang Hao himself finally retreated at some time after 3 pm.

Ming army lifted the siege and began to retreat. Ming cavalry were recalled to perform rearguard action (click to enlarge).
As for the Japanese, the defenders of Dosan Fortress quickly noticed that Ming troops were retreating, and immediately dispatched messengers to notify the Japanese relief force across Taehwa River. However, after witnessing the discipline and prowess of Ming rearguard, Japanese relief force deemed them too dangerous to attack, and thus chose a wait-an-see approach, wasting away nearly half a day doing nothing. It wasn't until they saw the smoke from the burning of war material, the evacuation of the main siege camp on Hakseongsan, and even the rearguard at Jeontan had begun to leave, that they finally decided to commence the rescue operation. Even so, Kuroda Nagamasa and Hachisuka Iemasa of the First Division were fearful of Ming army and hesitant to move, further delaying the operation (unbeknownst to both of them, this act of cowardice was witnessed and recorded by ikusa metsuke of the First Division, and they would be severely punished by Hideyoshi because of it).

Ming troops stranded at the southern bank of Taehwa River attacked into Japanese camp (click to enlarge).
While the First Division was hesitating, Second and Third Division of Japanese relief force inside the new camp suddenly came under attack from Zu Cheng Xun and Wu Wei Zhong's detachment. Despite being stranded and abandoned (Yang Hao did not even bother to inform them about the retreat, and they were not aware that the rest of the Ming army had left), they nevertheless stood their ground and even launched an attack uphill. In spite of literally every odds against them, Ming troops still fought the Second and Third Division troops led by Mōri Hidemoto to a standstill. As such, the Second and Third Division of the Japanese relief force were pinned down and unable to cross the river.

Japanese First Division crossed Taehwa River (click to enlarge).
Meanwhile, Kikkawa Hiroie, who was positioned behind Kuroda Nagamasa and Hachisuka Iemasa, finally had enough of their cowardice and decided to cross the river on his own. He was reprimanded by Ankokuji Ekei for stepping out of line and disobeying order, but rebuked him by saying that a monk should not interfere with the matters of a samurai (Ankokuji Ekei was a Buddhist monk) and crossed the river anyway. Katō Kiyomasa witnessed the river crossing from Dosan Fortress, and was so impressed by his bravery that he thought Kikkawa Hiroie's original three flap uma-jirushi banner was unbefitting of a samurai of such calibre. As a sign of gratitude for being the first to save Dosan Fortress from danger, Katō Kiyomasa gifted his own personal banner, a silver-coloured nine flap uma-jirushi, to Kikkawa Hiroie after the battle was over, of which Kikkawa Hiroie later added another four flaps to the banner and changed its colour to red. On the other hand, due to the little episode between the two, Ankokuji Ekei would later deliberately withhold information of Kikkawa Hiroie's bravery and exploits in his report to Hideyoshi and Mōri Terumoto (毛利輝元).

Replica of Kikkawa Hiroie's thirteen flap uma-jirushi displayed at the entrance of Sengoku no Niwa Museum of History (戦国の庭歴史館), Hiroshima, Japan.
With Kikkawa Hiroie taking the lead, the rest of the First Division finally got their act together and crossed Taehwa River to attack Ming rearguard. At the same time, Japanese fleet on Dongcheon River began to made landing, whereas Japanese troops inside Dosan Fortress also opened its gates and poured out to support the First Division. As most of the Ming army had left, Ming rearguard realised that they could not resist the landing of Japanese relief force while being threatened from three sides, so they only briefly clashed with the Japanese before retreating (Kikkawa Hiroie, who was the first to cross the river and engaged in combat, only managed to kill six Ming troops). After repelling Ming rearguard, Japanese First Division established a beachhead at the northern bank of Taehwa River. Kikkawa Hiroie, being bold as ever, raced ahead of the rest of the First Division and recaptured Byeongyeongseong. His decisive action successfully cut off Yang Hao's line of retreat to Gyeongju.

Kikkawa Hiroie recaptured Byeongyeongseong, forcing Yang Hao to change direction. Ming rearguard engaged and repelled Japanese pursuers from Dongcheon River (click to enlarge).
With his eastern line of retreat cut off, Yang Hao was forced to retreat westwards to Eonyang. Seeing that Ming army was retreating, funateshū (船手眾, naval troops) of Japanese fleet and Katō Kiyomasa's Seosaengpo troops that were hitching a ride on their ships seized the opportunity and disembarked at Dongcheon River with 200 to 300 arquebuses to chase after Ming army. Unfortunately, they ran into Ming rearguard a mere 300 m from the river bank. Bai Sai and Yang Deng Shan immediately launched a cavalry countercharge, shooting dead several Japanese troops, beheading eight more, and drove the rest back to their ships. In addition, the First Division of the Japanese relief force was still in the process of crossing Taehwa River and had not yet amassed enough troops to begin the pursuit operation, whereas Second and Third Division were pinned down in their camp thanks to the action of Zu Cheng Xun and Wu Wei Zhong's detachment. As such, Yang Hao was able to retreat in relative safety.

With the last of the Ming army retreating, Dosan Fortress was finally spared from its doom. Japanese ships on the river began delivering food and supply into the fortress, and many starving troops immediately gorged themselves full the moment they saw food. Unfortunately, this resulted in even more death due to refeeding syndrome.

Shimazu Toyohisa blocked Yang Hao's line of retreat, forcing him to turn north (click to enlarge).
Meanwhile, Shimazu Toyohisa, who came to Ulsan from Eonyang, joined force with the Japanese troops that disembarked from the 90-ship flotilla and blockaded the road, cutting off Yang Hao's line of retreat once again and forcing him to turn north and take the long mountain route to Gyeongju. While Ming army was switching route, Shimazu Toyohisa personally rode ahead of his army and attacked alone, beheading two Ming troops but was lightly injured in his left ear. However, Shimazu Toyohisa's troops consisted of foot soldiers that could not keep up with him, so he was unable to prevent Ming army from leaving. 

Wu Wei Zhong's detachment forcibly crossed Taehwa River under heavy fire (click to enlarge).
Zu Cheng Xun stormed through Japanese camp and sneaked to Seosaengpo Fortress (click to enlarge).
At the southern bank of Jeontan, the fierce battle in the new camp had finally begun to shift in Japanese favour. It was an unwinnable battle from the start, as Ming troops were starving, exhausted, and outnumbered, not to mention they were attacking uphill against a well-defended Japanese position held by fresh troops. As the situation became untenable, Wu Wei Zhong's detachment decided to call off the attack and retreat northwards, forcibly crossing Taehwa River under heavy fire from Japanese ships on the river and pursuing Japanese troops from the camp. As a result, Wu Wei Zhong's detachment suffered heavy casualties, losing as many as 200 troops in battle and during retreat. On the other hand, Zu Cheng Xun had a different idea. Instead of retreating to the north, he gathered his retinue cavalry and charged through the camp. In the ensuing fierce battle, Zu Cheng Xun's own horse was shot out from under him, and many of his retinue cavalry also lost their mounts. Nevertheless, they still managed to break out of Japanese encirclement and escaped south. Still unsatisfied with the outcome, Zu Cheng Xun and his troops sneaked to Seosaengpo Fortress (now largely empty since most Japanese troops had left for Ulsan) later that night and stole the signboard on its bridge before slipping back to friendly territory.

Japanese Second and Third Division crossed Taehwa River, captured Hakseongsan, and rendezvoused with First Division (click to enlarge).
After the attack on Japanese camp was dealt with, Second and Third Division of the Japanese relief force were finally able to cross Taehwa River. They occupied a high ground near the northern bank of Jeontan, but did not immediately chase after Ming army. Instead, Mōri Hidemoto assigned his troops to guard the high ground, while Second Division moved towards Dosan Fortress to capture the (now vacant) Hakseongsan and rendezvous with First Division. In the mean time, First Division also completed its river crossing. 

Final encounter behind the hill of Baegamsa Temple (click to enlarge).
With Dosan Fortress completely secured and large numbers of troops congregating together, Kuroda Nagamasa finally gathered enough courage to begin the mopping up and pursuit operation in the earnest. Japanese relief force successfully killed a number of stragglers, and was able to quickly close the distance with the retreating Ming army, finally catching up with Yang Hao's troops behind the hill of Baegamsa Temple (백암사 or 白奄寺, present day Baeg-yangsa Temple), about 4.8 km away from Ulsan fortress complex. To shake off the pursuers, Yang Hao once again ordered Ming rearguard cavalry to cover the retreat. Li Ru Mei and Jie Sheng launched a cavalry charge against the Japanese, killing a number of Japanese troops and drove the rest away. However, after Ming rearguard cavalry returned to their formation, Japanese relief force resumed its pursuit and began trailing the Ming army from a safe distance for another 3 km. The tense stare-off was finally broken when two mounted samurai carrying white banners rode closer to the Ming army to probe its response. Both of them were promptly beheaded by Ma Yun (麻雲) and Wang Guo (王果), Ma Gui's retinue cavalry. Seeing that Ming army closely guarded its retreat, Japanese relief force finally gave up and returned to Ulsan.

And that left us with Lu Ji Zhong. Due to the fact that he lost all contact with the rest of the Ming army and no one knew what happened to him, most Korean sources presume that his entire unit of 2,100 troops was wiped up to the last man. However, it can be known from other sources that Lu Ji Zhong was later sacked by Yang Hao, and the command of his troops was transferred to his successor, Chen Chan (陳蠶), who later had a merger with another 1,600 troops to make a 3,000-strong combined regiment. In other words, despite suffering the heaviest casualties among all Ming units, Lu Ji Zhong still managed to escape with large portion (1,400 troops out of 2,100 total) of his unit intact.

Aftermath

Having rid of the pursuers at last, Yang Hao returned to Gyeongju safely, although he only made a brief stop at Gyeongju before heading to Andong. Meanwhile, Japanese relief force also returned from the pursuit and encamped at Ulsan Eupseong (Ulsan walled town). Later that night, commanders of the relief force went to Dosan Fortress to meet the commanders of Dosan Fortress. Katō Kiyomasa, Asano Yoshinaga, and Ōta Kazuyoshi then wrote a joint final report back to Japan, detailing the entire siege and relief of Ulsan.

A day after the siege (February 11, 1598), Ryu Seong-ryong saw that there were still many surplus supply meant for Ming army left in Gyeongju, so he ordered Seong An-ui (성안의 or 成安義) to distribute the supply among Joseon troops, under the pretense of preventing the supply from falling into Japanese hands (he later proclaimed that the Koreans did their utmost to keep the frontline well-supplied, notwithstanding the fact that Ming troops starved at the frontline while supplies continued to pile up in Gyeongju). In addition, Ryu Seong-ryong and Gwon Yul managed to rally about 800 Joseon stragglers returning from Ulsan and stationed them in Gyeongju to defend against potential Japanese attack, but allowed the rest to return home.

On the Chinese side, despite the failure of the siege campaign, Ming army still took up the defence of Korea. At the beginning of March, some Ming commanders that returned from Ulsan (as well as additional commanders that entered Korea after the siege) were reassigned to defend various places in Korea: Li Fang Chun, Lu Ji Zhong, Li Hua Long, Lu De Gong and Niu Bo Ying (牛伯英) were assigned to defend Andong; Ye Bang Ron (葉邦榮) to Yonggung; Wu Wei Zhong to Chungju; Chen Yu Wen to Suwon; Lan Fang Wei (藍芳威) to Jiksan; Li Ning (李寧) to Gongju; Dong Zheng Yi, Chai Deng Ke (柴登科) and Qin De Gui (秦德貴) to Jeonju; as well as Bai Sai to Anseong. The rest of the Ming army gradually returned to Hanseong, as were Ma Gui and Yang Hao, who returned to Hanseong on March 14 and March 22 respectively.

Analysis

From the onset, this blog post dispelled several prevailing myths about Siege of Ulsan, as well as Imjin War in general, including but not limited to:
  • "Ulsan Japanese Castle" was a single castle (it was actually a massive multi-fortress complex, as were all other Waeseong in Korea);
  • Ming army suffered heavy casualties due to undisciplined retreat, leaving its troops vulnerable to Japanese pursuit (Ming army actually retreated in good order and protected its rear remarkably well, and only suffered relatively light casualties);
  • The disorganised retreat was caused by Yang Hao fleeing before his army, causing the rest to break ranks (Yang Hao actually stayed behind and was among the last to retreat); 
  • Japanese fortification design and mastery of matchlock firearms conferred them unique advantages against contemporary Chinese siege tactics (on the contrary, Ming army quickly captured all but one fortresses of the entire complex, and nearly captured the last one). 
That said, the strategic implication of Siege of Ulsan actually goes well beyond debunking a few myths. However, in order to fully grasp the impact and importance of this battle, one must look at the big picture of the second invasion, both before and after the siege campaign.

As mentioned in the prelude, after the breaking down of peace talk in 1597, Toyotomi Hideyoshi launched the second invasion with never-before-seen ferocity. Neither Joseon military nor righteous army was able to put up any meaningful resistance against the Japanese onslaught, and Joseon navy, the crown jewel of Joseon military might, was virtually obliterated during Battle of Chilcheollyang. Even Jeolla Province that survived the first invasion fell into Japanese hands. Essentially, the war unfolded much like the first invasion, only this time Joseon Kingdom was already devastated, its military in shambles, and its naval dominance completely undermined. 

To put things into perspective, in 1592 (before the first invasion), Joseon army had 180,000 troops stationed around the vicinity of Hanseong alone, and could muster as many as 400,000 conscripts using Hosu-Boin system (호수보인 or 戶首保人, a system where one soldier, known as Hosu, was supported by tax revenue from several common folks, known as Boin) should the need arise. Even after Joseon army suffered heavy losses during the first invasion, a census conducted in early 1593 shows that the combined strength of Joseon army, Joseon navy, and various righteous army groups of entire Korea still numbered 172,400, with Jeolla Proving having 10,000 army troops and 15,000 naval troops, and Gyeongsang Province having 35,000 troops stationed at Ulsan and Andong. However, by mid-1597 (before the second invasion), Joseon military had deteriorated to the point that there were only 1,500 troops left in Jeolla Left Province, as well as 11,100 army troops and 5,000 naval troops in Gyeongsang Province. 

As if such terrible state wasn't bad enough, Japanese army once again wreaked havoc across Korea during the second invasion, further degrading the strength of Joseon military. By the time Ming army was preparing to attack Ulsan, Gyeongsang Province could barely scrape together 3,000 troops to support the effort. Moreover, Jeolla Province fared even worse than Gyeongsang Province. At the beginning of February 1598, Yi Kwang-ag (이광악 or 李光岳), Army Commander of Jeolla Province, only had 500 troops under his command, most of them rabbles. Other Jeolla commanders fared even worse, even better-off commanders barely had 70~200 men under arms, and many could not muster even a single troop. Even the once-numerous righteous army of Jeolla Province had been shattered into fractured warbands numbering only 20~70 men per band.

The crisis was so dire that Hanseong once again came under grave danger of being captured by the Japanese. Residents of Hanseong fled the city en masse, almost hollowed out the capital, and even Joseon King Yi Yeon was contemplating to send his princes and harem away to safety. Although Hanseong was ultimately spared from this terrible fate because Japanese invaders decided to halt the attack and consolidate, the Koreans understood that the crisis was far from averted: The Japanese held all the cards and could resume aggression at any moment; both Joseon army and navy were ravaged and toothless, and the prospect of evicting the invaders from Korean lands was looking increasingly bleak.

As such, Ming intervention was the last, and in fact only, hope for Korea. The fact that Ming Dynasty sent in reinforcement alone was already greatly inspiring, and the news of Katō Kiyomasa's defeat at the hands of Ming army during the early phase of Siege of Ulsan was cause for celebration. Even after the failure of the siege campaign, contemporary Koreans mostly expressed disappointment and regret that Ming army wasn't able to finish off Katō Kiyomasa once and for all, rather than seeing it as a debacle/complete failure.

As much as Siege of Ulsan motivated the Koreans, its impact on the Japanese was even more profound. On October 19, 1597, two days after Battle of Jiksan, Japanese commanders held a war council at Jincheon. During the council, Ōta Kazuyoshi (one of the ikusa metsuke of Japanese army) instructed the Japanese commanders to return to the southern coasts of Korea in order to recuperate and wait for the winter to pass. He also announced that the offensive should resume in the coming spring (April 1598), where a well-rested Japanese army would march straight to Hanseong. However, the Japanese did not expect a Ming counteroffensive to come so early, before they were able to fully settle down, much less preparing for the 1598 spring offensive. Although Ulsan Fortress ultimately prevailed against Ming attack, it suffered severe damage in the process, losing most of its facilities, stored food, war materials, garrison troops, and more importantly, its function as a forward base to support the offensive. Furthermore, the threat of another Ming attack still loomed over the heads of Japanese commanders. Fearing for the safety of their own fortresses, many Japanese commanders in Korea petitioned Hideyoshi to abandon Ulsan and Suncheon Fortress in order to narrow down the battlefront to a more manageable size, of which Hideyoshi angrily declined. Not one to give up easily, Japanese commanders petitioned Hideyoshi for the second time, this time adding Yangsan Fortress into the list of fortresses to be abandoned. Hideyoshi was understandably furious and harshly criticised the cowardice of Japanese commanders, but even he realised that the situation was untenable and eventually ordered the abandonment of Yangsan Fortress and Gupo, and later pulled out one-half of Japanese invasion force from Korea.

By this point, no one on the Japanese side was thinking about the 1598 spring offensive anymore. In fact, it was not until April 4, 1598 that Hideyoshi brought up the resumption of the offensive again. In a letter addressed to Tachibana Muneshige (立花宗茂), Hideyoshi mentioned that he wanted to sent another army to Korea in 1599 to resume the offensive, and ordered Japanese commanders in Korea to procure food, gunpowder, and other war materials to support the operation. Another letter written by Fukuhara Nagataka (福原長堯) et al. to Shimazu Yoshihiro in June 29, 1598 reaffirmed the plan, and named Fukushima Masanori (福島正則), Mashita Nagamori (増田長盛), and Ishida Mitsunari (石田三成) as the commanders that would lead the 1599 offensive. Additionally, Ulsan was selected as the landing point for the new invasion force.

Essentially, Japanese army's entire strategic plan for the second invasion was derailed. Not only the planned 1598 spring offensive was completely ruined, forcing Hideyoshi to delay the invasion for another year, Japanese army was forced into defensive by an enemy one-third its size (only 40,000 Ming troops had entered Korea at this point, while Japanese troops in Korea numbered about 140,000), losing all the initiative and momentum it built since the beginning of the second invasion.

In July 1598, due to deteriorating health, Hideyoshi ordered Katō Kiyomasa to restart the peace talk, only this time he dropped everything in his list of demands (including the cession of Korean provinces, sending a Joseon prince to Japan as hostage, yearly tribute, and submission of Joseon Kingdom to Japan) and only asked for one thing: an apology from the Koreans. Whether that apology was made by Joseon King or some unnamed nobody mattered not to Hideyoshi, as long as he received one, he would end the war. After spending seven years waging a fruitless war, throwing away tens of thousands of lives as well as untold amount of wealth in the process, Hideyoshi was now eager to end it. The demand for apology was no more than a face-saving gesture to satisfy his ego, as well as a last-ditch attempt to hold onto some kind of moral high ground.

Thus it's fair to say that Siege of Ulsan was the single most important battle of the second invasion, as well as its real turning point. Even though Ming army retreated without accomplishing its objective, the actions of Japanese commanders in Korea as well as Hideyoshi himself after the siege campaign clearly show that they were rapidly losing control of the situation. Moreover, all this while Ming army was able to build up its strength in Korea unhindered, and by September 1598 as many as 74,400 Ming troops and 24,000 horses had gathered in Korea. From then on, Ming-Joseon coalition went from being constantly on the defensive during the early phase of the second invasion to having secured defence and capable of going on the offensive. In contrast, Japanese army was forced from a dominant position in full control of the war into a vulnerable position constantly preoccupied with reacting and responding to the changing situation. All of these were directly or indirectly caused by Siege of Ulsan.

Missed opportunity

It should be noted that neither the Chinese nor the Koreans were fully aware of the instability and chaos on the Japanese side. Due to the massive debacle that was the previous peace talk, which resulted in the execution of chief negotiator Shen Wei Jing (沈惟敬), no one in the Ming army dared to even entertain the idea of re-enter negotiation with the Japanese anymore. As a result, Ming army continued to amass troops and gather supply in Korea while purposely ignoring repeated attempts from Katō Kiyomasa and Konishi Yukinaga to make peace. 

Unfortunately, just when Ming army completed the mustering of troops and Yang Hao was about to put his plan of a new offensive into motion, a memorial to the throne written by Ding Ying Tai (丁應泰) in July 1598 sparked a massive internal feud in both the Ming court and the Ming army in Korea, to the point that even Joseon court was dragged into the chaos. Ming army was paralysed by the scandal, wasting away three whole months (July to October) doing literally nothing. By the time the dust began to settle, Yang Hao was already discharged in disgrace, Ding Ying Tai ascended to power and began to lord over the rest of the Ming military leadership in Korea, and Ming army was heavily disheartened and beset with confusions and internal strives. As a result, it severely underperformed in the three sieges that followed. For example, there was little to no coordination between Ming army led by Liu Ting (劉鋌) and Ming navy led by Chen Lin (陳璘) during Siege of Suncheon (Liu Ting barely had any motivation to fight), whereas a gunpowder accident during Siege of Sacheon resulted in the most humiliating Ming defeat of Imjin War. The most egregious one, however, was none other than Second Siege of Ulsan. The second siege failed not because of any mistake on Ming army's part, but because Ding Ying Tai forcibly ordered Ma Gui to cancel the siege and return to Gyeongju "for inspection". In stark contrast, thanks to the internal strife that paralysed Ming army, Japanese army managed to weather through its worst period of instability and weakness, and began to slowly regain footing. By the time a demoralised Ming army relaunched its offensive in October 1598, Japanese army had sufficiently stabilised and fought off the three sieges with remarkable competency. In fact, the whole offensive could've ended up as a massive blunder if not for Battle of Noryang, of which the Ming-led coalition navy inflicted the single heaviest casualties to Japanese navy since the beginning of Imjin War, thus preserving the reputation of Ming military somewhat.

To summarise, Siege of Ulsan decisively tipped the balance of war in favour of Ming-Joseon coalition, forever preventing the Japanese from ever achieving their objectives. However, both Ming court and Ming army were soon caught up in a massive internal feud, not only wasting away a golden opportunity to take advantage of the favourable situation, but also dragging out the war unnecessarily. Regrettably, despite committing an ever-increasing number of troops and resources into the war, in the end Ming army still fell short of achieving a complete victory.

Attributions and Special Thanks

While originally I planned to write this blog post based on my original research, I quickly came to realise that neither my knowledge on this topic, nor access to historical documents, nor my ability to understand and interpret those documents to construct a comprehensive narrative are up to the task. As such, the completion of this article would not be possible without massive amount of inputs and guidance from Mi Zhou Zhai (米粥斋), an expert in the field, through an intermediate (who wishes to remain anonymous). I also borrowed heavily from 《万历朝鲜战争全史》written by Zhu Er Dan (朱尔旦), who also penned the three-part critique of Samuel Hawley's book that I translated. For that, I owe them my utmost gratitude.

Reference

It should be noted that the following list of reference materials are not so much the references I use, rather they are references used by those who advised and guided me. In retrospect, I am glad that I consulted them. Such a massive collection of primary and secondary sources far exceeded what I can realistically access (and this list is far from exhaustive, I only jog down what I can remember), let alone study.

Chinese sources

  • 《明史》
  • 《明史稿》by 萬斯同
  • 《明史紀事本末》by 谷應泰
  • 《明神宗實錄》
  • 《國榷》by 談遷
  • 《萬曆邸鈔》a collection of Wangli-era Dibao (邸報) gazette/newspaper.
  • 《萬曆野獲編》by 沈德符
  • 《兩朝平攘錄》by 諸葛元聲
  • 《東征記》by 徐希震
  • 《經略禦倭奏議》by 邢玠
  • 《太僕奏議》by 張輔之
  • 《東師奏報失實疏》first-hand casualty report of the siege campaign co-authored by Yang Hao and Ma Gui. Included in《太僕奏議》
  • 《全邊略記》by 方孔炤
  • 《武備志》by 茅元儀
  • 《牧齋初學集》by 錢謙益
  • 《崖山西生鎮都督洞層岩鐫刻記》by 片碣頌

Korean sources

  • 《宣祖昭敬大王實錄》
  • 《宣廟中興志》
  • 《征蠻錄》by 李擢英 
  • 《再造藩邦志》by 申炅
  • 《象村稿》by 申欽
  • 《懲毖錄》by 柳成龍
  • 《慕夏堂文集》by 金忠善
  • 《洪吉童傳》attributed to 許筠
  • 《月沙先生集》by 李廷龜
  • 《請留經理再奏》included in《月沙先生集》
  • 《看羊錄》by 姜沆
  • 《金陵集》by 南公轍
  • 《磻溪隨錄》by 柳馨遠
  • 《星湖先生僿說》by 李瀷
  • 《宋經略書》by anonymous

Japanese sources

General

  • 『豊臣秀吉譜』by 林羅山
  • 『朝鮮蔚山合戦之図』not to be confused with the famous folding screen painting.
  • 『義演准后日記』by 三宝院義演
  • 『西笑和尚文案』by 西笑
  • 『藩翰譜』by 新井白石
  • 『日向記』
  • 『岩国市史』

Asano clan

  • 『浅野家文書』
  • 『浅野幸長蔚山籠城以下万事覚書』included in『浅野家文書』
  • 『浅野幸長家臣某蔚山籠城覚書』included in『浅野家文書』
  • 『幸長手記』

Katō clan

  • 『清正公行状』
  • 『清正記』by 古橋左衛門又玄
  • 『清正高麗陣覚書』by 川兵太夫
  • 『本山豐前守安政父子戰功覺書』
  • 『下川文書』

Kikkawa clan

  • 『吉川家譜』
  • 『吉川家文書』
  • 『吉川広家朝鮮役戦功覺書』included in『吉川家文書』

Kuroda clan

  • 『黒田家譜』
  • 『黒田家文書』
  • 『黒田記略』

Shimazu clan

  • 『島津国史』
  • 『島津家文書』
  • 『義弘公御譜中』
  • 『旧記雜錄後編』
  • 『西藩烈士干城録』
  • 『本藩人物誌』
  • 『征韓錄』
  • 『又七郎豐九譜中』
  • 『朝鮮役録』by 五代秀尭 

Mōri clan

  • 『毛利家記』
  • 『毛利秀元記』by 黒川真道
  • 『陰德太平記』by 香川正矩
  • 『萩藩閥閲録』
  • 『安西軍策』by anonymous
  • 『宍戸記』
  • 『松井物語』

Nabeshima clan

  • 『鍋島家文書』
  • 『鍋島直茂譜考補』

Ōta clan

  • 『朝鮮日々記』by 慶念
  • 『大河内秀元陣中日記』by 大河内秀元

Tachibana clan

  • 『立齋舊聞記』

Tōdō clan

  • 『高山公実録』

Early modern sources

  • 『南藤蔓綿録』by 梅山無一軒, 1804~1817
  • 『征韩偉略』by 川口長孺, 1831
  • 『大日本編年史』by 小西惟沖, 1883
  • 『如蘭社話』by 邨岡良弼 et al., 1912
  • 『朝鮮征伐記』by 黑川真道, 1916
  • 『日本戦史: 朝鮮役』by 参謀本部, 1924
  • 『碧蹄戦史』by 渡邊刀水, 1938
  • 『秀吉の朝鮮経略』by 京口元吉, 1939

Modern references

  • 『近世日本国民史·豊臣氏時代: 朝鮮役』by 徳富蘇峰, 1981
  • 『関ヶ原合戦と近世の国制』by 笠谷和比古, 2001
  • 『加藤清正―朝鮮侵略の実像』by 北島万次, 2007

Further Reading

164 comments:

  1. Great write up so far. Looking at the modern pic of Dosan Fortress on the raised hill it looks like a great place to build a fortification, being on high ground and overlooking the surrounding area. You mention the Chinese and Koreans rarely build 'castles' in the Japanese (and medieval European) style, ... I wonder why? Probably because of the 'feudal' nature of the Japanese and Middle Ages Europeans with their plethora of daimyo and barons constantly fighting each other in worthless feuds, while China and Korea were more state level actors perhaps ??

    ReplyDelete
    Replies
    1. Chinese and Koreans did build similar-sized fortresses, but those are generally of the "wall with buildings inside" kind (essentially a "shrunk down" walled city), without an obvious keep/tenshukaku. Chinese called these smaller fortresses Zhai (寨), Bao (堡) etc, and that's what they called Japanese castle too.

      Delete
  2. Such a nice work im waiting for its continuation!

    I want to ask a guestion, Im from Turkey and i really love Chinese-Korean-Japanese history. I want this passion to spread therefore im going to make a roleplay game of Imjin War with even Jianzhou Jurchens playable. I want units to be accurate and not as simple as "pikemen swordsmen archer cavalry". I want to add some egzotic units and weapons such as fire lances and rocket wagons. But i really lack the information, can you tell me some Late Ming Chinese units and weapons to me so i can make Ming Chinese army enjoyable?

    Thank you!

    ReplyDelete
    Replies
    1. Good day and welcome to my blog.

      Imjin War is a highly complex topic given the multiple sides (and sources) involved, so it's difficult to accurately reconstruct the exact equipment of the armies involved.

      Jianzhou Jurchens did not participate in the war. Only Kato Kiyomasa briefly intruded into their territories and fought with them (it went badly form him).

      Delete
    2. Did Kato Kiyomasa's skirmish with the Jurchen go badly for the Japanese? I've read that his muskets made mince meat of Jurchen cavalry. Do we have any more details about this amazing encounter? What did the Japanese make of the Jurchen? did they even know who they were? what did they make of their tactics, what did the Jurchen think of the Japanese? It's been said Nurhaci offered to aid the Ming and Koreans in the Imjin War, but the Ming declined. Interesting if the Jurchen did intervene against the Japanese, how would they have faired against the might of the entire Japanese invasion force?

      Delete
    3. I will make every nation playable therefore everyone will have the freedom to declare war on whatever country they want thats what i meant :D

      Delete
    4. @Der
      Those are generally heavily exaggerated later period sources intended to glorify and justify Japanese invasion of China.

      Basically, Kato Kiyomasa crossed Tumen River into Yanbian and attacked/destroyed several Jurchen settlements, which provoked the Jurchens to attack him. According to Korean source, his force suffered heavy casualties and returned to Korea.
      According to "Kiyomasa Korai-jin Oboegaki", a primary source written by Shimokawa Heitayuu (Kato Kiyomasa's subordinate and participant of the battle), some 8,500 Japanese troops fought valiantly against "tens of millions" of Jurchens, they were surrounded, but saved by a timely "divine downpour" (rain version of Kamikaze basically) and retreated to Korea the next day.

      The Jurchens then lay siege to Kilju (then under Kato Kiyomasa's control, but he was not present there). Kato requested Nabeshima Naoshige's aid to lift the siege.

      Delete
    5. The Japanese were aware of the Jurchens and described their settlements as "foreign" but noting much beyond that. They called the Jurchens they fought "Orankai (兀良哈)" which seems to be used loosely/derogatorily to refer to generic "barbarian".

      In Chinese context, that name usually refers to a specific Mongol subgroup (Uriangqais), which the Japanese certainly did not encounter. I am not entirely certain which Jurchen tribe was encountered by Kato Kiyomasa, as I am not familiar with early Jurchen/Manchu history, although I've read discussion that suggest that it was Warka Jurchens.

      Delete
    6. @Ali Emre Azgın
      Yes, I understand the draw of "what-if" scenario and exotic gadgets of war, although they are by definition not historically accurate.

      I can only give you a broad and crude overview:
      The majority of Ming troops that went to Korea were Northern Chinese border troops, most of them can be classified as multipurpose medium cavalry (armoured rider on unarmoured horse, armed primarily with bow and arrows + sabre, many also wielded some kind of polearm or matchlock/handgonne). Rocket was used fairly extensively by Ming cavalry.

      Ming infantry were much more varied and generally came from Southern China and included everything in the usual roster (pikemen, swordsmen, archers, matchlockmen, artillerymen, also cavalry). Many of them were "double-armed" (usually one melee and one ranged primary weapon, i.e. pike + bow, polearm + rocket etc)

      Fire lance was no longer used by Ming army except in naval battle and sometimes for siege (the so-called fire lance was actually a powerful one-use flamethrower, ranther than a weak peashooter attached to a spear). I am not aware of explicit record of Chinese use of rocket wagon during Imjin War, but given the amount of rocket they brought, they probably did.

      Delete
  3. @ Ali Emre Azgin

    Most the Ming troops sent to Korea during the Imjin War were northern border cavalry, and they wouldn't be that much different from the more iconic Mongol and Manchu cavalry of the same era. Their main weapons were bow and arrows, saber, and lance or polearm. Their armor would be brigandine, basically a long cotton coat riveted with squarish metal plates on the inside. The northern troops carried some firearms too, but mostly cannons, rockets, and outdated handgonnes like the iconic three-eyed gun, with few to no matchlocks. And presumably they would have used some war wagons too, both for offensive (especially as a platform for launching rockets) and defensive purposes, as well as for transporting equipment and supplies.

    The southern Chinese troops, on the other hand, were a completely different matter. They were mostly infantry with very few cavalry. In terms of firearms, they were quick to adopt the matchlocks from the Portuguese merchants and the Japanese wokou, unlike their northern counterparts who chose to stick with the old-fashioned handgonnes. They would also have been equipped with some rockets and small anti-personnel cannons like the Folangji breech-loading swivel guns or the Hudun Pao hand mortars. Regarding to their ranged weapons, they were less proficient with bows than their northern counterparts, but they were better at using crossbows and javelins. Some of their crossbow bolts and javelins would have been poison-tipped. And in terms of their melee weapons, there was a strong anti-cavalry focus, so they would have been equipped with polearms like pikes, spears, Zhanmadao horse-chopping cleavers, and some Ming paintings (like Wokou Tujuan) even showed southern troops carrying strange-looking hook-shaped polearms called Xiezi Wei or Scorpion's Tail. And speaking of strange-looking polearms, one cannot ignore the Lang Xian wolf brush, another unique polearm coming from Southern China in the Ming period. It was basically a bamboo spear with its frontal branches still attached to block enemy's vision and to entangle / parry enemy's weapons. It proved to be very successful during the anti-Wokou campaigns of Qi Jiguang in the 1560's, and would ultimately spread to Korea. And last but not least, in terms of their defense, since metal armors would become rusty rather quickly under the humid conditions of Southern China, a lot of southern troops would wear non-metal armors made from mulberry paper, lacquered rawhide, and rattan, etc. And they were avid shield users too, with the most common shield types being the round rattan shield and the V-shaped Yanwei Pai or Swallow Tail Shield. With all that being said however, due to the great distance involved, I don't think the southern troops played any major role in the Imjin War, as only a small amount of them had been sent there.

    And just as the admin said, fire lances were mostly used for naval battles and siege operations. The Ming fire lance or Pen Tong was essentially a large bamboo tube reinforced with hemp ropes and leather bands, and could spit out various types of projectiles including flame, poison smoke, ceramic / iron / lead pellets and shrapnel, and semi-solid burning discs that could attach onto enemy clothes.

    Hope that my explication would help you make your game, cheers.

    ReplyDelete
    Replies
    1. Now i know what i should put, thank you so much!

      Delete
    2. Great explanation. But in your opinion, which side had the better armies? Chinese, Japanese or even the Koreans with the warrior monks and navy seamen.

      The Japanese had huge numbers (120k troops), the largest military force in the Imjin War and were battle hardened and led by ferocious samurai, but all their experience was gained in Japan fighting against other Japanese. They had matchlock arquebuses but no artillery. And their cavalry was nearly nonexistent. Their military leadership was primitive, with feudal lords leading their own private armies acting independently instead of a clear chain of command with a top Supreme Commander like the Allied Chinese/Koreans. And of course they had a horrible naval arm that failed to keep the Japanese adequately supplied and were soundly defeated by the Korean Navy.

      The Koreans neglected their land forces, hence the easy victories by the Japanese on land. But Korean naval technology, strategy and tactics, home ground advantage and of course Adm. Yi's leadership was decisive in the Imjin War since the Japanese supply lines were devastated and the Japanese attempt to extend naval control to the western Korean coastlines failed. I would say the Koreans did as much as the Chinese in freeing Korea from the Japanese invaders. And the Korean Buddhist Warrior Monks were formidable fighters along with the Righteous Armies forming guerilla bands.

      And the Chinese, contributing the smallest contingent (~50k troops) to the war but the most professional (as in professional military fighting for payment from the central government), with Li Ru Song in overall command of all Chinese and Korean forces coordinating with a clear chain of command, leading excellent battle hardened northern cavalry and experienced southern infantry. The Chinese were hampered by overstretched supply lines, inadequate supply due to Korean incompetence, inadequate numbers where they couldn't sustain heavy losses, and musketry inferior to the Japanese.

      Delete
    3. @Der

      I think the three were at a roughly similar level of military development, and neither side has complete advantage over the other.

      I disagree with the claim that Chinese musketry were inferior to Japanese. It's true that Northern Chinese troops were not proficient with matchlock muskets, but the Southern Chinese troops were quite good at using them. By the time of the Imjin War the Southern Chinese were already using them for at least several decades, and Ming firearm inventor Zhao Shizhen (from Zhejiang province) was able to improve upon existing matchlock designs and create new types that were more suitable for Chinese troops. The Southern Chinese had access to various matchlock designs through trade, including Japanese, Portuguese, Southeast Asian, Indian, and even Middle-Eastern / Ottoman.

      The reason why during the Imjin War the Chinese troops didn't seem to use matchlocks was because most of the troops sent to Korea were northern border cavalry and they preferred to use old-fashioned handgonnes like the three-eyed gun. Only small groups of southern infantry were sent there and they only served a complementary role.

      Delete
    4. @Der @Unknown
      Stay tuned to find out.

      Delete
    5. @Der @Unknown
      My latest update should be a good reference for the comparison between Ming and Japanese army.

      IMHO, while Japanese army was good at what it was supposed to do, it was "defective" in the sense that it didn't have much in the way of cavalry, artillery, and experience in fighting either of them (or fighting both at the same time). This put the Japanese army at a remarkably disadvantagous situation against Ming army (with cavalry and artillery up the wazoo) in both field battle and siege.

      Delete
    6. The Ming Infantry Southern China were originally staffed by Commanders who were trained and recruited during Gen Qi Jiguang days. Their experiences and training came useful in Imjin Waeran , because Northern Based Ming Army were mainly cavalary based. However by Wanli reign, corruption and influence peddlers were already rampant in Ming court. Since Gen Qi jiguang himself fell out of favour in the past due to the political intrigues, so such things usually also affected his followers and students in the military ranks. It was no surprise, the southern Ming troops were sent to Korea to deal with the Japanese, but politically their commanders were weakened by past associations with Gen Ji Qiguang. In the booknof Jingbirok , by Rye Seong Ryeong, he mentioned a Ming commander who " helped " to train Joseon soldiers , and that commander was from the Southern Ming infantry commander.

      Delete
    7. @Unknown
      IMO the negative effect of Qi Jiguang's fall from grace would be very limited, because surprisingly few Southern commanders were directly related to him. Qi himself was also "merely" sidelined and then fired instead of being pinned with some kind of grave crime, which was still a pretty shitty fate for him, but not the kind that will cause others to avoid association with him.


      Among the Southern commanders, only Wu Weizhong directly worked under/with Qi Jiguang before, and actually led Southern troops from Zhejiang. However his troops were mostly fresh recruits.

      Luo Shangzhi, another Soutern commander, was not directly related to Qi Jiguang, although he likely studied Qi's materials. He brought the only Southern troop veterans (about 600, possibly from Ji Garrison) to Korea, and I think it was him that trained the Koreans. At the time he was also a commander of Shenjiying (Firearm branch of Imperial guards) and unlikely to be affected by Qi Jiguang's fall.

      Qi Jin was Qi Jiguang's nephew, but he commanded war cart troopers from Ji Garrison (i.e. Northern troops) and usually not counted among "Southern commanders".

      Liu Ting was leading troops from Sichuan and not related to Qi Jiguang at all.

      Qian Shizhen was another commander from South China (Shanghai), but he led the cavalry from Ji Garrison, so he was a "Northern commander".

      Delete
  4. Thanks everyone, you guys really helped me

    ReplyDelete
  5. Would it be fair to say the only Imjin War battle that the Southern Chinese units participated in majority numbers compared to Northern forces, would be the Battle of Noryang Strait? Chen Lin and his marines/navymen I'm assuming is mostly from Guangdong. I know it diverges from the main topic of Ulsan, but what kind of weaponry, equipment, and armor they would use?

    ReplyDelete
    Replies
    1. Yes, AFAIK that was the only battle which the Southern Chinese troops participated in numbers. Chen Lin was indeed from Guangdong, though I'm not so sure whether his marines were all from Guangdong. They could be from other places in Southern China like Zhejiang, Jiangxi, or Fujian.

      They would have used cannons like the Folangji breech-loading swivel gun, which was probably the dominant Ming navy gun in the 16th century and was mass-produced. And they also carried many other firearms like rocket arrows, muskets, hand grenades, fire lances, and cold weapons like javelins, crossbows, pikes, swords, etc.

      In terms of armor they would most likely used lighter non-metal armors for naval operations, like paper armor or leather armor.

      Delete
  6. All in all, a total of 9,965 Joseon troops participated in the siege at some point, and among them 4,982 deserted.

    This retention rate is incredibly low, not sure if I heard of army behaving in such manner without dissipating completely, what are reasons which led to this poor morale of Korean army?

    ReplyDelete
    Replies
    1. It's better to treat this as one combined army with a particularly weak contingent (i.e. the Koreans) rather than two allied armies working together but otherwise acting independently. Otherwise, most armies wouldn't survive half its troops deserting.

      Joseon Dynasty was militarily very weak to begin with (as it was remarkably peaceful for the past centuries), and the carnage of the First Japanese invasion only made things worse. Korea had yet to recover from the invasion (and wouldn't for a long time to come), basically.

      Delete
    2. We can also consider the Joseon soldiers are on home territory. Just having that option available is a big factor in desertion. Especially if a battle is not going well, and on your side the main contingent is a foreign ally. It probably made more sense to not go all out in this battle.

      Delete
  7. What sources did you primarily use for this battle?

    ReplyDelete
    Replies
    1. Currently I am looking at Seonjo Sillok, but I mostly just referenced works from other more knowledgable people than me.

      Delete
    2. Did you use Kenneth Swope's work because some of this seems very familiar

      Delete
    3. @Unknown
      No. From the onset you can see some serious errors in Kenneth's book, and some events unfold differently. For example, Kenneth lists Yang Dengshan, Po Gui and Bai Sai as the vanguard (it was Li Rumei, Yang Dengshan and Bai Sai) and they "met with a hail of fire arrow (wut? the primary ranged weapon of the Japanese was arquebus)". The dates are also off, and Kenneth also implies that Ma Gui reached the outer defence first before sending out the vanguard.

      The way he describe the siege seems to reinforce the misconception that Ulsan was but ONE castle. This leads to many weird mistakes like Li Rumei's crack troops "attack on Ulsan proper" and draw out the defenders during the day of the siege, rather than on outlying Byeongyeongseong the night before.

      He also wrote that Ming attacked the inner sanctum and the arrival of Kato Kiyomasa...on the fourth day? The heck? What a complete mess.

      Delete
    4. @春秋戰國
      Swope also seems to mistakenly refer to Asano Yoshinaga as Asano "Nagayoshi" (he seems to have gotten the kanji mixed up as he writes his name as 淺野長吉).

      Looking over the end notes he has for his section on the battle/siege (#79-89), he seems to refer to a mixture of sources, both primary and secondary, that might have distorted his account. Listing a few here in no particular order based on those and his abbreviations

      -Li Guangtao, “Yang Hao Weishan zhi yi,”
      -Li Guangtao, Chaoxian “Renchen Wohuo” shi liao
      -Sin Kyöng, "Zaizao fanbang zhi [Chaejo pönbang chi]"
      -Kuwata and Yamaoka, "Chôsen no eki"
      -Kawaguchi, "Seikan iryaku"
      -Gu, "Ming shi jishi benmo"
      -Keinen's diary
      -Taiko-ki (Hideyoshi's biography)

      Delete
    5. Yes his book does seem very muddled at times and a bit too pro ming as well. Do you know any good english language sources on this topic as I am currently in the process of learning Chinese

      Delete
    6. @Bleb
      Isn't Asano Nagayoshi another name for Asano Nagamasa?

      Kenneth isn't entirely to blame for the mess. There were serious political struggles among the Chinese at the time (Yang Hao vs Ma Gui, the doves vs the hawks), and the doves would stop at nothing to stop the war, including exaggerate Ming casualties and rampant smearing and false accusations against frontline generals not on their side, leading to Yang Hao being disgraced after this battle.

      Those documents also passed down and became "historical sources", they eventually crept in to say, Mingshi (History of Ming) among others, causing a lot of confusions and misconceptions.


      @Unknown
      The way Kenneth describe Siege of Ulsan gives me the impression that the Chinese were a bunch of headless chickens. Seriously, Ming army in his book wasn't able to prevent Kato Kiyomasa and his relief column of FIVE HUNDRED troops from forcing his way into the castle, with 55,000 men! And then he wrote that Chinese lost 4,000 troops during retreat and the Siege of Ulsan "clearly a debacle".

      If THAT level of incompetence is still considered "too pro Ming", I wonder what people will think when they read the accurate account of the battle...




      Delete
    7. @春秋戰國
      Honestly, my level of knowledge on the Imjin War is startlingly poor so I'm not really sure. I'm inclined to think this was an honest error on Swope's part since he later refers to "Asano Nagayoshi" as a "young samurai" fighting off Ming troops on the walls at Ulsan. Since Yoshinaga was born in 1576, he should've been 22 at the time of the battle, so that at least fits the bill. I could be totally wrong though!

      With regards to the other issues--yeah, I can definitely see that being a problem. I think Swope (along with Hawley!) even ends up aligning Song Yingchang with the hawkish factions in the Ming court (page 151, 181, 182), which totally seems to be at odds with the Chinese netizen dude's critique of Hawley's book.

      Swope does dedicate a lot of time to discussing Yang Hao's tragic 'fall guy' reputation though, and I think this extends into his next book ("The Military Collapse of China's Ming Dynasty, 1618-44") which, from what I read a few years ago seems to hold up a bit better than his Imjin War book, especially since he isn't dealing with so many conflicting narratives and sources in different languages.

      IIRC, the Shogun-ki blog (which I linked in the comments of the first part of your translation of the critique of Hawley's book) were very critical of Swope's book and specifically attacked his usage of Japanese sources, relying far too heavily on "gunkimono (Japanese war tales)" compared to his Chinese and Korean sources. I suspect these might be the reason some of the Japanese perspectives in his accounts of the battles are so confusing (hell, maybe this is where the "hail of fire arrows" and descriptions like "Young samurai Asano Nagayoshi whirled his spear in an arc of death" came from).

      http://www.theshogunshouse.com/2009/12/superior-title-inferior-book-kenneth.html

      @Unknown
      If the question was for me, I've got nothing. There's not really a lot on the Imjin War other than the 'big three' as far as I can tell, though maybe Nam-lin Hur's paper critiquing Swope and Hawley which I've linked in some other comment will probably be of use to anybody interested in western historiography thus far...

      Delete
    8. @Bleb
      Even the Shogun-ki blog post is full of bias and errors and unsubstantiated claims, and the article sounds very defensive as if the author is offended (to me at least). Better to take it with a grain of salt.

      Delete
    9. @春秋戰國
      Oh yeah, the critique on that site does seem just a bit salty so I'm definitely taking a grain (or more) with me as I consider its claims. They followed up that review with the Hawley interview shortly after, and as we both know they end up getting really giddy about bashing Swope for his unprofessional behaviour. Supposedly, he posted a negative Amazon review anonymously--based on the quotes they use in the interview, I guess it's this "Top critical review" for the first edition of Hawley's book.

      https://www.amazon.com/Imjin-War-Sixteenth-Century-Invasion-Attempt/product-reviews/8995442425/ref=cm_cr_arp_d_viewpnt_rgt?ie=UTF8&reviewerType=all_reviews&filterByStar=critical&pageNumber=1

      https://i.imgur.com/ZCb0Xiw.png

      And if that really was Swope under an alias ("K-Pop Fan" is pretty hilarious 13 years down the line tbh) then that's pretty damn petty. But I really have no idea--I already feel a bit bad for digging up old internet skeletons littering their polemical battlefield.

      Anyway I don't really want to derail the comments section here any further. Keep up the great work with updating the post!

      OH, and a somewhat brief question: I noticed that one of the random facts/trivia you posted on the site mentions a punitive expedition against Japan that the Ming court had planned but never came to fruition due to a rebellion in Guangxi and the end of the Imjin War itself--when exactly was this planned and how was it situated within the wider context of how the war was going for the Ming? You write that the Ming seriously considered a full retreat from Korea after losing at Jiksan--was this counter-invasion planned in the middle of the three Ming offensives, or after Ulsan, which the Chinese critic you translated believes was the real turning point of the war?

      Delete
    10. @Bleb
      There were a few failed proposals before, but the successful one (proposed some time after Chilcheollyang) was accepted and put into motion around February 1598, nearing the end (or right after) the Siege of Ulsan.

      Delete
  8. I read the part of about Jurchen cavalry being employed at this battle with the Japanese. Fascinating. So the Jurchen did fight the Japanese outside of Kato Kiyomasa's invasion of Manchuria? Amazing. And with iron whips? how did those work? I wonder what the Japanese thought about the Jurchen cavalry? and why were they working for the Ming, aren't they enemies? I wonder what the Jurchen thought about the Japanese ??

    ReplyDelete
    Replies
    1. My bad, Ma Gui's cavalry should be Mongols, not Jurchen, I've corrected my blog post.

      Iron whip is basically a sword-shaped mace/heavy duty iron truncheon. Very good at bashing armour and caving skull.

      Ming army in the north employed quite a lot of Mongols (and other ethnics). There were a lot of mingling between people of different cultures at the border after all.

      Delete
    2. Interesting. I wonder what the Japanese and the Koreans thought about the Mongols and vice versa? Did the Mongols still have their reputation they gained from Genghis and Kublai Khan? What did the Mongols think of Japanese fighting prowess, and vice versa?

      Thanks for the interesting post.

      Delete
    3. @Der
      Not sure about the Japanese, but the Koreans were pretty disturbed. No records on what the Mongols thought about them though, as far as I am aware.


      Delete
    4. Half a century before described events Mongols still could reach outskirts of Beijing.
      Nothing close to what they did in 13th century, but nothing to laugh at either.

      Delete
    5. According to the Korean official records, the Ming Mongol Cavalry ripped the Japanese army apart on the open field...
      http://sillok.history.go.kr/popup/print.do?id=kna_13806007_002&gubun=chn
      臣曾從(道山)〔島山〕 之戰得見, 麻貴所率㺚子二百餘名, 皆持環鞭, 亂打如雨疾雷, 不及掩耳, 銃筒亦不暇放。 賊兵之走, 亦如我國人之走, 以此見之, 銃筒於馬戰, 亦末耳。

      Translation: From what I saw during the battle of Tosan, Ma Gui led 200 Mongols armed with chain whip, their speedy blow was like a storm, struck like thunder, (the Japanese) didn't have a chance to fire any shot. The Japanese soldiers were running away in the same fashion as our soldiers were running away in the past (from the Japanese army)...

      Delete
    6. @SY
      The "環鞭" should translate to hard whip/iron truncheon, not the flexible "chain whip". In any case, Ma Gui's 200 Mongol's cavalry were probably his handpicked Jia Ding and armed in the Chinese fashion (Mongols rarely use truncheon). They shouldn't be equated with average Mongol horsemen of the period.

      Delete
    7. So ring whip is not a flail but Tie bian?

      interesting

      Do you have any record additional record for 環鞭 is being refered as truncheon ?

      Delete
    8. @ssd
      Yes. The chain whip type is usually called Jiu Jie Bian (九節鞭) or "Nine-segmented whip". Huan Bian 環鞭 literally translates to "ring whip", the "ring" probably refers to the many buldges on Tie Bian. In modern Chinese martial arts, Jiu Huan Bian (九環鞭) or nine-ringed whip is a short stick-based martial arts.

      Although a later source, Muyetobotongji also attest that Ming soldiers made frequent use of Tie Bian/Tie Jian.


      Delete
    9. Personally I think it is a type of "two-section staff" (Lian Jia), the word "ring" has the literal meaning of ring as in the section that connects the staff together. There is no "ring" in Tie Bian, it would be strange if they call the bulges as "ring". Also, since you mentioned Muyetobotongji, there is reference to the use of "two-section staff" on horseback by heavy cavalry, they use a shorter version, and the book named it as "whip", and it describes the difference of "chains" between the cavalry version and the infantry version. As for the "frequent use of Tie Bian/Tie Jian" that you mentioned, I didn't find such record in Muyetobotongji. Maybe you can indicate the section number where it gives such reference?

      Delete
    10. @SY
      I remember the source also mentions Ming cavalry shooting their Tie Bian on horseback (i.e. Lei Huo Bian, the Tie Bian-handgonne hybrid), but oddly I wasn't able to relocate the text again. Maybe I confused the source or deceived by unreliable third-party sources.

      The buldges on Tie Bian do look like rings when they are worn on finger. Besides, Koreans also call sabre "Hwando (環刀)" even though Korean sabre has no visible ring.Thirdly, Ming cavalry also wasn't known for using flail-type weapon, as far as I can tell.

      Delete
    11. "Hwando (環刀)" in Korean terms referred to ring on the sheath of the sword/saber
      https://terms.naver.com/entry.nhn?docId=1155429&cid=40942&categoryId=32432
      So I don't think the same thing could be applied to Tie Bian because Tie Bian usually does not have a sheath and the Chinese style of carrying sword/saber is different from the Korean Hwando, you can look at the picture on this website for the Korean "ring" style
      https://www.pinterest.co.kr/pin/465559680207212815/?nic_v2=1a4wkXD6V
      So I think that the word "ring" in 環鞭 refers to a different thing than the the same word in 環刀.

      As for the Ming cavalry, they used various types of weapons, including the "whip staff" (Bian Gun) or "Iron Chain Staff" as recorded by the Mao Yuanyi who wrote that the Chinese soldiers were good at it. The volume 4 of Muyedobotongji described it in detail. Later in the same text, the Korean author used 鞭 ("whip") to describe this weapon.

      Delete
    12. @SY
      鞭棍/Bian Gun usually refers to a short staff in Chinese context, and a related Bian Gun Chong is a six-shot double-headed handgonne version of that.

      Mao Yuanyi's iron flail description was copied from Song Dynasty Wujing Zongyao (called, and the context was that Han foot soldiers used it better than nomadic cavalry. Chinese soldiers generally used flail behind fortification (either walls or wagon fort).

      In any case, I re-read the equipment list for the preparation of the first phase of Imjin War. One entry lists equal number of Tie Bian and flail (both were 1,500).

      Delete
    13. So is it cavalry man's weapon?

      Delete
    14. @ssd
      After reading through Muyetobotongji and find no such quote, I believe I've been deceived by unreliable articles.

      The listing is basically an inventory checklist, so we don't know who was going to use what weapon. But given that the equipment was brought by Northern Army, I wouldn't rule out cavalry use.

      Delete
    15. I know it's off the topic but is there any mention of staff or wooden club weapon in the list?

      IIRC in Silok they mention ming cavarly army use staff weapon (not sure it was gun or bang) so I wonder if this staff would be Da bang or Jia dao gun

      and another question

      did southern army ever use tie bian?

      Delete
    16. @ssd
      I have not find wooden staff in the list so far (it's an excerpt), I will need to find the complete book to give you a definite answer.

      Gun/Ban are basically the same thing (different regions of China used to call it by different names), and yes it was used by cavalry.

      Yes, Tie Bian was used universally in entire China.

      Delete
    17. Thanks


      Here is a record from silik that I referenced


      ○丙申/提督接伴使張雲翼啓曰: "卽刻自稷山戰所回來唐兵說稱: ‘天安、稷山之間, 不意倭賊先鋒, 皆着白衣, 遍野而來, 唐兵等初謂稱朝鮮人, 不爲進逼。 俄而倭先放砲, 唐兵一時跑馬廝殺, 交戰良久, 倭人中箭被棍死者, 幾至五六百, 斬級三十餘顆, 解副摠、楊叅政, 各手斬二級。 而倭賊登山擧白旗, 天安大軍, 卽刻雲集, 衆寡不敵, 各自退守。 解摠兵等四將, 去夜發稷山前來, 唐兵亦多死者云。’ 且提督卽刻發放各營, 使之盡數出陣江邊, 仍爲野營云, 且發令旗, 使擺遊擊, 抄領精兵二千五百, 迎擊於水原之路云。 敢啓。" 傳曰: "知道。"

      Delete
    18. @SSD
      “棍死” should mean "bludgeon to death", rather than referring to specific weapon.

      Delete
  9. "Seven thousand Jurchens could resist a hundred thousand Japanese."
    — Li Ru Mei's impression of the capabilities of Japanese army.

    surely this is exaggeration??

    ReplyDelete
    Replies
    1. Of course. Li Ru Mei was boasting to the (King of) Koreans.

      Delete
    2. Having said that, Li Ru Mei certainly did pretty well against the Japanese, all things considered.

      Delete
    3. Frankly it's surprising to me Li Ru Mei and the Ming did so well against the Japanese. The Ming military was overstretched and outnumbered. Whereas the Japanese had the numbers, made up of battle hardened samurai and disciplined Ashigaru led by the best generals of the Sengoku Jidai, with the best muskets in the world made in large numbers.

      Delete
    4. @Der
      For the most part cavalry still had a clear edge over infantry, so this isn't surprising.

      Delete
    5. @Der

      Japanese army in Korea severely lacked in both cavalry and artillery department, which was inexusable during this era, as both cavalry and field cannons played huge role in defeating enemy army.

      Delete
    6. @Dmitry

      There were very few good pastures to raise war horses in feudal Japan. Only in some area of Kanto region did they have adequate pasture land. In fact, the Ming also suffered the lack of war horses during the early reign of Hongwu and Yongle emperors, but the Ming had a centralized government, huge resources and border trade system to address the issue.

      Delete
    7. @Der

      They did poorly on open field battle because the matchlock was not a "quick firing" weapon, and they didn't develop efficient pike and shot tactics similar to the Spanish tercio, their tactical formations look like medieval (organized around noble knights) rather than early modern. But they performed really well in defending stronghold and castles, where the discipline and skill in matchlock were in full effect.

      Delete
  10. This is a very interesting article, hats off to your desire to shed some light on this largelly unknown conflict to the Western audience, but I can't understand one thing: how could Japanese forces constantly get surprised by Ming and Joseon army despite possessing numerical advantage and maintaining long defensive castle chain? Although they lacked cavalry and heavy artillery, it seems strange for them to essentially lose all head to head engagements against the Chinese.

    ReplyDelete
    Replies
    1. Thank you.

      Well, in the case of Ulsan the Japanese were caught off guard (my personal speculation is that they miscalculated how fast a cavalry army could march), but this wasn't always the case. A few Righteous Army contingents actually attacked Ulsan several days before Ming army arrive, and the Japanese army made short work of them.

      Both Byeokjegwan (see my other blog post) and Jiksan were Japanese victories as well. Japanese army performed poorly despite huge numerical advantage, but a victory is a victory, regardless of how poor their performance was - it's the outcome that matters.

      Delete
    2. I did read your Byokjegwan post, it was superb and even made me write a short story about - I had a blast with it so thank you once again.

      Saying "losing" is somewhat a bold statement on my part, I agree, but still it's somewhat strange for Japanese to generally pull back before Ming advance. Was it really because of severe lack of cavalry so they had troubles attacking themselves?

      Delete
    3. I have trouble replying comment atm, will get back to you later.

      Delete
    4. Yes, it is strange the Japanese didn't do better with numerical superiority, battle hardened, etc, etc.

      I can only speculate, but I think it comes down to more than just equipment and numbers. The Japanese for centuries fought only other Japanese in clan based warefare. Before the Imjin War, the last time the Japanese fought outsiders were the Mongols in the 13th century and before that was the Baekche Restoration War in the 7th century against the Tang! The Japanese being an insular island nation simply were not exposed to outside military techniques and developments. Japan is like the Galapagos Islands, isolated from the rest of the world for so long they developed their own ways of warfare. These uniquely Japanese ways of warfare were simply not up to par against the Ming, ... despite being the latter stages of decline, the Ming was still a military power than fought Mongols/Jurchens in the North, tribal peoples in Burma/Yunnan, Wako pirates on the eastern coast and even internal rebellions in China itself. The Chinese were obsessed with logistics so supply officials and bureaucrats were just as important as soldiers while the Japanese didn't seem to realize they needed Korean cooperation to maintain supply lines back to Japan. I guess the difference between the Ming and the Japanese is the difference between soldiers and warriors ... fighting as a profession and fighting because of ones social class?? In the Early Modern Period the soldier slowly displaced the warrior, technical skills outclassed bravery. We saw the same thing with the Ottoman victories against the feudal armies of Europe, like the Battle of Mohacs where professional Janissaries mowed down the brave Hungarian knights.

      Delete
    5. Indeed, for some time Ottoman military was above anything European kingdoms could field.

      Delete
  11. Note: I'm not downplaying East Asian military engagements and neither am I defending the Japanese, it's just my general observation.

    Nevertheless, Ōkōchi Hidemoto (大河内秀元), witness and survivor of the siege, recorded that Japanese army suffered a mind-boggling 18,360 casualties on the first day alone.

    I'm no specialist on the Imjin War, but this seems to be extremely and unlikely high number. It's easily the biggest number of casualties of any side suffered throughout the entirety of the war, and the one which was delivered on relatively small territory during relatively short amount of time. Few battles of this era can match this amount of casualties inflicted in a single day. Plust, that would mean the entirety of Japanese army would collapse and flee, as that number is like the entirety of garrison plus large chunk of the relief force. I have a feeling this is exaggeration. What do you personally think of this number?

    ReplyDelete
    Replies
    1. I am glad you ask that question, because there are more information that I would like to convey, but can't fit into the blog post (the blog post is already super bloated, and it's only day 2).

      Now, to make one thing clear - Japanese army had around 20,000 troops garrisoned in Ulsan fortress complex, so losing 18,360 won't wipe them out completely.

      Secondly, regardless of our feelings toward that number, Ōkōchi Hidemoto's diary is still the only primary Japanese source from a first-hand witness (that contains casualty figure). We can't simply dismiss a witness account just because we "feel" it doesn't sounds right.

      Nevertheless, Ōkōchi Hidemoto actually recorded three Japanese casualty figures: Japanese casualties on the first day (i.e. night raid/field battle), Japanese troops that starved/freezed to death for the entire duration of the siege, as well as casualties among Japanese relief force.

      Conspicuously, he did not record Japanese casualties on the second day (i.e. Ming assault on Ulsan fortress complex). Some theorise that the 18,360 figure was the combined casualties for the first and second day, rather than for the first day only. Personally I find this theory plausible.

      Delete
    2. The problem here is that even first-hand witnessess can and often do exaggerate their losses and hardships. For example, French mercenary officer on Polish service during the siege of Glughiv in 1663 Antoine de Garmon wrote down that one of the Polish assaults on the city costed them more than 4,000 killed and "innumerous number" of wounded, which isn't reflected on any other primary sources (including those on the opposing side) and doesn't match up with recorded losses of the Polish army. We also know that Ottoman chronicles describing their army in 1596 to be 150,000 strong also exaggerated their forces, even though it would make more sense to do so with the enemy.

      I know I stand on shaky ground and apologize for my probable stubborness on this matter, yet I have a feeling Okochi could exeggerate. Are there any army lists and field reports which could indicate fluctuation of personnel in Japanese armies? Surely this big loss of life would somehow reflect that.

      Although if there are no other primary sources which discuss that, I suppose I can't back my claims with anything and it simply touches on the subject of faith.

      Delete
    3. Indeed it is possible (in fact, quite likely) that Ōkōchi Hidemoto exaggerated in some ways, although we don't have other numbers to make comparison and thus the extend of his exaggeration can't be known.

      We can be fairly certain that Japanese army suffered very serious and possibly catastrophic loss during the first day (or first two days), considering that Ming army was able to rapidly capture so many fortresses with minimal loss during the second day.

      Delete
  12. When do you think you'll be done?

    ReplyDelete
    Replies
    1. No idea, the siege lasted for twelve days, and I am only on day 2.

      Delete
  13. Would you ever consider doing a post on the Tumu crisis? I can hardly believe 20,000 mongol horsemen destroyed an army of 500,000 Chinese troops no matter how badly trained they were

    ReplyDelete
    Replies
    1. I have other plan at the moment. I've been writing losing battles for a while now, so I want to write a winning one.

      I haven't done much research on Tumu Crisis, and one can't write about Tumu Crisis without discussing the subsequent Siege of Beijing (1449) anyway.

      Delete
    2. Now I'm curious, what winning battle are you working on?

      Delete
    3. A post-Imjin War battle against the Mongols.

      Delete
    4. What is the battle called?

      Delete
    5. I thought most of the mongolian conflicts stopped after the whole Altan Khan thing in the 1580s when the Ming finally agreed to trade

      Delete
    6. Battles of Huangzhong. Altan Khan already died by that point.

      Delete
    7. Do you have the characters for HuangZhong I've never heard of this battle and am curious

      Delete
    8. HuangZhong is a name of a place (Huangzhong district). The battle is relatively unheard of, and currently I am having a very hard time translating all those Mongol names.

      A minor correction, the battles actually happened concurrently with Imjin War.

      Delete
    9. Concurrently with the Imjin War? wow! the Ming did have a lot on their plate. Did the Ming leadership consider Korea a minor affair and the Japanese a minor threat compared with the Mongols and Jurchen??

      Delete
    10. How many troops were involved if this battle is unheard of?

      Delete
    11. @ Unknown
      The war involved tens of thousands of troops on multiple sides (Tibetans were involved on the side of Ming).

      @Der
      Japan wasn't a minor threat, but Mongols and Jurchens were still more concerning.

      Delete
  14. What sources are you using for this mongolian battle?

    ReplyDelete
  15. Thanks so much for posting this. I actually used to live in Ulsan a few years ago as an English teacher and was lucky enough to visit the remains of the fortress. I always got the impression that this was a tourist site the local people weren't entirely comfortable with (which is understandable considering later history). I'll be interested to read your conclusions as I remember the site information played up the siege as a strategic victory and a humiliation of the Japanese forces. On the other hand, the Japanese accounts I've seen tend to play it as a successful strategic withdrawal after the Ming forces wrecked themselves trying to take the walls.

    ReplyDelete
    Replies
    1. @anon

      Hello and welcome to my blog. Since it'll be a while before I complete this, I will spoiler you a bit about the conclusion: The battle was a Japanese victory, as Ming army wasn't able to capture Dosan Fortress. However, the Japanese talking point about "Ming forces wrecked themselves" was patently untrue, as Ming only suffered miniscule casualties for the siege.

      On the other hand, Japanes army suffered massive casualties due to the siege, and it completely derailed their plan of next phase invasion. Many commanders petitioned Hideyoshi to abandon some fortresses in order to shrink down the battlefront to a more managable size, and Hideyoshi tried to restart negotiation with drastically reduced terms. Japanese situation was actually extremely chaotic and unstable, but this golden opportunity was promptly wasted by Ming Dynasty due to serious internal political struggle (which persisted until the end of the war). By the time the next three sieges (Ulsan 2nd, Suncheon, Sacheon) were launched, Japanese had restabilised their footholds whereas Ming army had become rife with internal feuds and politicking.

      Delete
    2. -and it completely derailed their plan of next phase invasion.

      What is "their plan of next phase invasion"?

      Delete
    3. Hideyoshi was planning a third invasion in 1599.

      Delete
    4. Yes, "next phase" refers to that 1599 invasion.

      Delete
  16. Replies
    1. I am tied down by IRL stuffs, but hopefully I can write something today or tomorrow.

      Delete
  17. The siege starts to look really intriguing, any chance on update?

    ReplyDelete
  18. 站主你覺得朱爾旦的"萬曆朝鮮戰爭全史"這書如何? 我聽說評價好像很不錯,但價錢有點貴。

    ReplyDelete
    Replies
    1. 應該很不錯。我有個友人有買,正在託他想辦法幫我多買一份。
      如果能入手我大概會把這本書當作以後寫萬曆朝鮮戰爭的主要參考資料。

      Delete
  19. Maybe I'm missing something, but it seems really strange for entire army to start starving and losing horses en masse after just a week of fighting.

    ReplyDelete
    Replies
    1. IMO there are multiple reasons:

      Ming army was running out of food before reaching Ulsan, so it had starved for more than one week. On top of shortage of supply, distribution of supply was also a problem.

      It was also winter (in fact, January was the coldest month in Korea), with temperature often dropping below freezing point. This likely limited the availablilty of forage.

      Also for that entire week both sides were engaging in battle almost constantly. When they were not fighting they were busy collecting firewood for the next attack/guarding the river/fixing damaged parts of the fortress. Exhaustion alone would probably do a number on both men and horses.

      Delete
    2. I should also add that generally speaking, human can only suvive for around five to eight days without food and water, but can live much longer (up to 40+ days) with access to water. So it's not surprising that Japense troops were droping like flies inside Dosan Fortress in a siege that lasted only twelve days.

      On the other hand, Ming army actually starved for longer (food ran out before the army even reached Ulsan), but still had access to water, so it came out of the siege with comparatively miniscule casualties.

      As for the horse, reminder that horse are trickle eaters that have to constantly nibble on something to keep their digestive system in good condition, and may develop potentially life-threatening complications if not properly fed for as short as 10+ hours. Even by itself, going without food for nine days WILL kill the horses en masse.

      Delete
  20. Hey, do you have that one quote that's something like how the Japanese arquebuses can deafen you, but the Ming cannons can shake the earth? It's also from the Imjin war

    ReplyDelete
    Replies
    1. "倭銃之聲,雖四面俱發,而聲聲各聞,天兵之炮,如山崩地裂,山原震盪,不可言狀"

      Delete
    2. What is that quote from and what's the context of it? Seems like perfect one-liner for a hollywood imjin war action movie hahaha

      Delete
    3. @Fyre
      It's from Seonjo Sillok, and from a series of Q&A between Korean King and Yi Deok-hyeong & Yi Won-ik about current situation of the conflict, various aspects of China, Japan, and Korean military, logistics etc.

      Delete
  21. @C
    Sorry I mistakenly deleted your comment.

    Accurate telling of the event requires careful reading of multiple sources from all three sides as well as modern archaeological researchs, have a firm grasp on Ulsan's geography, and see through lies and exaggeration in the sources.

    To be honest it is beyond my ability to do research of this scale. I just ask around and translate the work of others.

    ReplyDelete
  22. Latter parts of the battle make me imagine it looking frighteningly similar to this picture (with Ming in the front).

    https://i.redd.it/out3msngcqb01.jpg

    ReplyDelete
    Replies
    1. LMAO. The last good part has yet to come. The retreat is one of the highlights of Siege of Ulsan for me actually (other than the first two days) as it really shows the competence of the Ming army.

      Delete
  23. It would be very interesting if you write about other battles of the imjin war

    ReplyDelete
  24. Congratulations on finishing this article, this was a massive undertaking! I know all of the English-language books seem to completely muddle this siege up big time, but it's really impressive how detailed this reconstruction is. Kudos to whoever had to go through all the hard work of sifting through all the bullshit/embellishments and figuring out the most likely progression of events...

    I guess just out of curiosity, you mentioned to me in a past comment that Japanese scholarship on the Imjin War was by far the most developed--did any of that factor into the formation of this article? Did any of the sources (if there were any secondary ones!) you referred to and the people you asked, draw upon modern Japanese historiography on the Imjin War and this battle in particular? If so, did Japanese historians reach similar conclusions? Do they share the view that Ulsan was the real turning point of the war, or do their opinions diverge in any significant ways?

    ReplyDelete
    Replies
    1. Thank you.

      The person I consulted (I will add attribution to him in this blog post on a later date) does mention that Japanese historians over the years do gradually shift from taking Japanese sources at face value (i.e. Ming army suffered tens of thousands of casualties during retreat), to referring both Korean and Japanese sources, but still lean on Japanese side (i.e. claiming that the Koreans downplayed the casualties), to siding with the more trustworthy Korean sources, using the works of Tokutomi Sohō (1863-1957), Kinzō Watanabe (1874-1965) and Manji Kitajima (1935-2018) as example.

      He does not mention on how Japanese historians view the significance of this battle though. This battle became the turning point not only because Ming army dealt a heavy blow to Kato Kiyomasa, but because other factors that happened on the Ming-Joseon side as well (Hopefully I can elaborate on it later).

      Delete
  25. I found your anecdote about "defected Japanese" troops joining the Ming/Korean forces to attack their former compatriots. I'm interested to know why they would defect? and why would they join in the fight against the Japanese, being Japanese themselves, why would they turn traitor?

    ReplyDelete
    Replies
    1. Defection could happen for any number of reasons: mistreatment by own side, escaping punishment, infighting, bribery, forced at sword point, stragglers trying to survive by any means necessary, etc.

      Delete
  26. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. I already described as such in the prelude ("achieved all initial objectives" and "consolidate"). Yes, Japanese building the fortress network to prepare for the next invasion was the original plan.

      However, the strategy fall apart because of Siege of Ulsan.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. This comment has been removed by the author.

      Delete
    4. Wait for my analysis update.

      It did though, the situation after the siege became so unstable for the Japanese that multiple prominent daimyos petitioned Hideyoshi to abandon some of the fortresses they just built to shirnk down the defensive network, and Hideyoshi himself even tried to restart the peace talk.

      The original plan, as decided by the Japanese war council held in Jincheon in 1597, was to restart the invasion after the winter, in the spring of 1598. Siege of Ulsan caused that plan to fall apart, so the invasion had to be pushed back for another year (1599).

      In other words, the 1599 invasion plan itself was the direct result of derailment caused by siege of Ulsan.

      Delete
  27. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. I like your professional articles. I don't want to comment. I want to be a quiet follower. Please delete this comment. sorry.

      Delete
    2. Jincheon war council was also recorded by Ōkōchi Hidemoto. I think there were three war councils being held by Japanese commanders around that time (before and soon after Jiksan): Jeonju, Jincheon, and Jeongeup.

      The planned 1599 Invasion would be headed by Fukushima Masanori, Mashita Nagamori, and Ishida Mitsunari.

      Delete
  28. Really enjoying this post (was going to write "enjoyed" but you're still adding more apparently!). The different and partisan nature of the Chinese, Japanese, and Korean perspectives are really interesting. I'm curious if you have any rule of thumbs for deciding on how you come to your conclusions about what actually happened. Say, if the Japanese and Chinese sources agree but the Korean source tells something very different, err on the side of the former two. Not saying I think you actually do this but just giving an example for illustration.

    ReplyDelete
    Replies
    1. It's rarely that simple, sometimes different people looked at the same event and report it in seemingly contradicting manner, even though they are both telling the truth. Sometimes different historians looked at the same source and come out with wildly different interpretations etc. So, except in the case where the record blatantly defiles logic, there really isn't a "rule of thumb".

      Very often seemingly minor details like "where is this site located" or "when does this thing happen" can drastically alter the narrative/reconstruction of the event, more so than number crunching. To be honest I wrote half the blog post before realising that, f*ck, the Japanese flotilla actually came from the east, not the west. That's the reason I keep trying to redo the maps.

      A VERY significant portion of the blog post also contains information/details/sources that are beyond my capabilities to access/analyse, so I have to rely on consulting with people more knowledgeable than me, or translating their writings. I will credit them after I am done.

      Delete
  29. Your mention of the southern troops leaving their northern buddies(?) out to dry on the second day is interesting to me--were these Zhejiang infantry or some of the Tusi contingents (Wolf troops et al?)? I'm curious because the 'popular' English narrative of the Ming military in the Imjin War (you see this in Turnbull, Swope, and Hawley!) was that the Northern troops were overconfident assholes who kinda didn't know what they were getting into, and the Southern troops were hyper-comptent but also humane and gentlemanly, with "Southern" tending to just describe the Zhejiang troops even though not that many marched all the way to Korea in the first place.

    I'm guessing this is mainly because of all the Korean accounts being very favourable to the Zhejiang infantry (not to mention our various Joseon sources really not liking Li Rusong or his buddies, and describing northern Chinese troops as wearing no armour, Qian Shizhen's diary that alludes to the rivalry between southern and northern troops, and also Swope mentioning that Li Rusong spoke highly of Qi Jiguang's tactics when facing the Japanese), and while we probably have little reason to doubt the good impression they left on the Koreans, this incident seems to be an apparently rare case where 'Southern' troops were behaving rather selfishly and compromising the siege. I get the sense from reading your post here and the others you've translated from the Chinese netizen, that one of the myths you're also trying to dispel is that the Northern Chinese contingents weren't in fact utterly clueless and hilariously overconfident in fighting the Japanese, but were fairly competent and held their own despite all of the problems they faced in both invasions.

    Just how bad was the north/south rivalry, especially at this phase of the war, and how did these troops compare in terms of discipline and fuckups (though my impression is that the worst of it seemed to be due to factionalism in the Ming court and the dove/hawk struggle)? Your post also debunks the often-mentioned messy retreat of the Ming thanks to Yang Hao, but there's also the gunpowder accident at Sacheon, though I guess any army would be in complete disarray after something as disastrous as that...

    ReplyDelete
    Replies
    1. Zhejiang infantry refers to Chinese troops from Zhejiang, which was/is a Han-majority region (and part of the core region of China or "China proper"), and famously the place where Qi Jiguang recruited the anti-wokou troops. Soldiers from Zhejiang were typically modelled after Qi Jiguang troops/tactics. They were not under the Tusi system.

      Yes, all things considered, Northern troops performed better than their Southern counterpart. They were still kinda assholish though, especially the Mongol troops. (As a side note, I recently learnt of the Japanese description of Chinese cavalry during Siege of Ulsan. Apparently Han troops wore red uniform with white armours, while Mongols wore black).

      While Northern and Southern Chinese troops generally didn't like each other, they could still tolerate each other's presence, and work together if needed be. Most of the bickering were actually about who should take credit and reward for battlefield achievement(s).

      Using Siege of Ulsan as example, the primary factionalism problem within Ming army was actually between Yang Hao and Ma Gui, two highest-ranking commanders of the army. Li Ru Mei was close to Yang Hao, while Bai Sai and Yang Deng Shan were Ma Gui's men. That's why those three were dispatched ahead of the army on Day 1 (to prevent one side from hoarding all the credit).

      Then again, Ma Gui planned the entire Day 1 night raid, so he thought he (or his subordinates) should take credit for the victory. However, Yang Hao thought that Li Ru Mei should take credit, since he was the vanguard, basically defeated the entire Japanese garrison at Byeongyeongseong (note that Byeongyeongseong was not a small garrison, it actually rivalled Ulsan Fortress in area size) with only 300 cavalry, and scored the most kill. This caused a lot of troubles between the two, and was the reason why Ma Gui forcibly ordered Li Ru Mei to retreat on Day 2.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. Dang, thanks for the detailed reply!

      I think I might have been a bit too vague with my wording regarding the Zhejiang troops! I was just wondering if our sources ever identify the southern troops (that abandoned the northern Chinese attacking the eastern gate) in question--if they were the Zhejiang infantry, or the less disciplined Tusi troops that were deployed in greater numbers during the second invasion.

      I actually recall Swope describing Ming troops assaulting Ulsan as "apparently making good targets with their red armor and white helms." (p. 257) His footnotes indicate he gleaned this information from the Taiko-ki (Hideyoshi's biography), sections of the 'Chosen Seibatsuki', and 'Chosen no Eki' for his Japanese sources, though he also refers to a bunch of Chinese sources as well. I don't he ever mentions the Mongol retinues at all in his account of the battle though.

      Delete
    4. @Bleb
      The feud between Yang Hao and Ma Gui also caused Ma Gui to kept his mouth shut and simply let Yang Hao do his thing out of anger for the duration of the siege (hence Ma Gui is largely left out of the picture after Day 2).

      The only prominent Southern commander in the Left Division at the time was Mao Guo Qi, a Zhejiangnese, so the “Southern troops" were definitely Zhejiang infantry.

      Chen Yin was also a Southern commander in the Left Division, but he was busy attacking Dosan Fortress from another direction. He was not born in Zhejiang, but his troops were Zhejiang infantry as well.

      Another Southern commander was Wu Wei Zhong, who was probably the most senior Southern commanders (a long-time former Qi Jiguang's subordinate). He was in Middle Division though.

      Swope probably got it from the same source, but somehow come out with different translation.

      Delete
    5. @Bleb

      Where did you hear that Tusi troops or Wolf Troops were deployed in greater numbers in the second invasion? Most southern troops sent to Korea were Zhejiang troops and not from other southern regions due to the greater distances involved. I'm aware that Liu Ting might have commanded some Tusi troops, but I don't think there were other Tusi troops involved. Overall their participation was very trivial and insignificant.

      Delete
    6. I don't have an exact source though I do recall GMM mentioning in a comment on the Langbing article that more of them were potentially sent to Korea than the Zhejiang troops.

      He also mentions that other Tusi like the Tu Bing had participated in the Imjin War... but I'm not really sure of the numbers! My Imjin War knowledge sucks since I don't really have access to any primary sources, and even if I did I suck too much to be able to read any of 'em!

      Delete
    7. @Bleb

      It's impossible that more Lang Bing were sent to Korea than Zhejiang troops, for two reasons. First is the greater distances involved. I think many people fail to grasp the size and diversity of Southern China. Lang Bing originated from Guangxi, which is more than 1,000 to 2,000 km further away from Korea than Zhejiang, and that's a huge distance to travel in ancient times. Secondly, Lang Bing were usually led by their tribal chieftains, or if not by a general from that region. Yet I don't recall any Tusi chiefs participated in the Imjin War, and the only generals that might have led Lang Bing or Tusi troops were Chen Lin from Guangdong and Liu Ting from Sichuan, yet they only participated at the last moment and didn't really see that much action.
      Hence, I don't think there were more Lang Bing sent to Korea than Zhejiang troops, due to the distance involved, and due to not having their chiefs with them.

      Delete
    8. @Bleb
      Note that Qian Shizhen himself was leading troops from Ji Garrison (i.e. Northern troops).

      @Unknown
      As far as I can tell, Chen Lin had at least 4,600 Wolf troops under him, that should already outnumber all Southern troops during the first invasion (3,000 under Wu Wei Zhong and 600 under Luo Shang Zhi, ).

      Delete
    9. But we're talking about the second invasion here, not the first invasion. In the second invasion there should be more Zhejiang troops. And also the only major battle that Chen Lin participated was the Battle of Noryang, hence they didn't really see that much action. The battles on land were still mostly fought by northern troops and Zhejiang troops.

      Delete
    10. @Unknown
      You'ere right. I double checked mobilisation records that I could find. There were indeed more Zhejiang troops in Korea than Guangdong troops.

      A logistics report written by Yang Hao on May 4, 1598 gives a good overview of Ming strength in Korea (at the time). From the report, we can know the status of Southern troops:

      Eastern Route
      3,900 Southern troops under Wu Wei Zhong (place of origin not specified but presumably Zhejiang)
      4,000 Southern troops under Chen Yin & Ye Si Zhong (place of origin not specified)

      Middle Route
      1,600 Zhejiang troops under Ye Bang Ron
      2,900 Zhejiang troops and 1,000 Miyun troops under Mao Guo Qi.
      4,000 Zhejiang troops (yet to arrive)

      Western Route
      12,000 Sichuan troops under Liu Ting (yet to arrive)
      3,300 Zhejiang troops under Lan Fang Qi
      1,500 Nangan troops (yet to arrive)

      Naval route
      Fujian: 2,180 (1,000 yet to arrive)
      Wusong: 2,000
      Guangdong: 8,000 under Chen Lin and Zhang Liang Xiang (3,000 yet to arrive)
      Zhejiang: 6,400 (3,100 yet to arrive)
      Langshan: 1,500 (yet to arrive)
      Jiangbei: 3,000 (yet to arrive)

      Delete
    11. Slight mistake, it was Lan Fang Wei, not Lan Fang Qi.

      Also, although Lan Fang Wei followed Liu Ting to Korea, he participated in Siege of Sacheon instead of Suncheon

      Delete
  30. I might be wrong, but I'm under the impression that only troops from Guangxi were the real Lang Bing. Granted I know that some troops from Guangdong were called Lang Bing as well, but I don't think all Guangdong troops were called Lang Bing.

    ReplyDelete
    Replies
    1. Lang Bing were generally but not exclusively from Guangxi. In 《再造藩邦志》, about 4,600~4,950 Guangdong Regiment troops under Zhang Ru Wen (張汝文), a subordinate of Chen Lin, were explicitly referred to as Lang Tu Bing.

      Wu Guang also had another 5,500 Lang Tu Bing from Guangdong.

      Delete
    2. That's still a meager number compared to Zhejiang troops, and also I don't think Guangdong troops were the same as Guangxi Lang Bing despite both can be referred to as such. The original Lang Bing were from Guangxi. And not to mention that they only participated in the last naval battle of the Imjin War which was the Battle of Noryang, and didn't really see much action elsewhere. Koreans had a much deeper impression of Zhejiang troops and of northern troops than of Lang Bing.

      Delete
    3. @Unknown
      While Guangdong troops may or may not be Lang Bing, those under Zhang Ru Wen and Wu Guang were explicitly called by that name ("Lang Tu Bing"), so they definitely count.

      (Otherwise classification of troops origin will become very weird. For example, Wu Weizhong's troops came from Ji Garrison, a Northern Garrison, but they were still treated as "Southern troops". There's also the issue of whether troops on Zhejiang fleet should count as typical "Zhejiang infantry", since they had few in common).

      Delete
    4. There are however significant confusions regarding Chen Lin and Battle of Noryang.

      While Chen Lin was the overall commander of Ming navy for the battle, he personally brought very little troops to Korea (as few as 500, possibly fewer). The majority of the Ming ships that fought in Noryang were under Ji Jin, which came from Wusong (majority), Langshan, Zhejiang and Jiangbei.

      Wu Guang didn't seem to participate in Battle of Noryang. Lang Bing were land troops after all.

      Delete
    5. If there were around 10,000 Lang Tu Bing sent to Korea as you claimed (which is not a small number considering that the total troops sent there during the Second Invasion were around 75,000), then the Koreans (and the Japanese) should have left an impression of them, since Lang Tu Bing were fairly easily distinguishable from northern troops and Zhejiang troops by their appearance, their armor, their weapons, and their fighting styles. Yet we found no such mentions from neither the Koreans nor the Japanese. That's the reason why I don't really believe your numbers regarding to Lang Tu Bing.

      And Lang Tu Bing were mostly from the southernmost part of China, which is very far from Korea. That would put severe logistic strains on the Ming if they choose to recruit troops from that area. Just for reference in 1555 the Ming indeed recruited Lang Bing from Guangxi to fight the Wokou in Zhejiang, and it took them a few months just to get to Zhejiang.

      Delete
    6. @Unknown
      《再造藩邦志》 is itself a Korean source, and both the numbers of Lang Tu Bing and the names of their commanders came from that source, so Lang Tu Bing did leave an impression on the Koreans. Doubting that number based on the lack of impression also amount to argument from silence fallacy.

      The Koreans had higher opnion on Zhejiang troops even at the expense of Northern troops, despite Northern troops outnumbering Southern troops pretty substantially and fought in nearly all land battles of Imjin War. I'd assume other troops got the same treatment by them. Also, what little Japanese impressions had on Ming infantry (that I know) seem to describe Northern infantry.

      The total troops sent to Korea was nearly 100,000.

      Eight months passed between Siege of Ulsan and Four Route Offensive, so there was plenty of time.

      Delete
    7. Now, this isn't to say you can't raise doubt about the numbers, it's just that "lack of Korean impression" and "distance of Guangdong to Korea" are extremely flimsy reasons to discredit the numbers I pulled directly from historical source. You need to do better.

      A more likely possibility is that Korean sources sometimes double-counted Ming troops or treat a detachment that split from its parent unit as additional troops, due to lack of understanding of Ming military structure/hierarchy. This applies to all kinds of Ming units and not just Lang Bing, however.

      Whether it applies to THIS case, I still need to double check.

      Delete
    8. It's not flimsy, cause we do have actual evidence about Ming employing Lang Tu Bing to fight Wokou, they departed from Guangxi and it took them a few months to arrive in Zhejiang. Korea is at least more than two times away from Guangxi than Zhejiang, so it would at least take them anywhere between half a year to a year to arrive in Korea. That's very impractical from a logistic POV. I've heard that when the Imjin War broke out even the Siam kingdom (Thailand) asked if they could send an army to Korea to fight the Japanese, however their offer was refused by the Ming. So if the Ming really had the logistics to move a substantial amount of troops from the Far South to the Far North, then why would they refuse the Siamese?

      Doubting is what improves our understanding of the issues in the world, be them historical issues or others. It's what makes science moves forward. The records were written by ordinary people like you and me, not by gods or saints, and people make mistakes. So the records aren't doubtless, especially not when concerning numbers since Confucians (most of the records regarding ancient China and Korea were written by Confucian scholars) didn't put a high value on maths, science, technology, and commerce.

      Delete
    9. And based on your description of Lang Bing about their ability to fight in small groups or squads, their ferocity, and certain of their armors and weapons (like poison crossbow and javelins for instance), if there were really that many of them who participated in the war, then Koreans and Japanese should have a much deeper impression of them, not just a very slight mention. Yet most of Korean and Japanese records were describing Zhejiang troops or Northern troops. This tells me that either they didn't really get a chance to fight due to arriving too late, or very few of them were sent there.

      Delete
    10. @Unknown
      The reasons you gave are flimsy because it is based on speculation, assumptions and hearsay.

      The Guangxi to Zhejiang case, even if true, may be caused by many other factors such as delays, time spend on recruitment/training/equipping, or simply because the troops were indisciplined and went to do other things. Those factors may not come into play during the time of Imjin War. Also, just because the troops came from Guangdong doesn't mean the supply also had to come from Guangdong. Supply came from the most convenient/available Chinese territory or contributed by the Koreans, regardless of troop origin.

      Not to mention it doesn't take anywhere that near long to move from Guangdong to Korea. Chen Lin himself was from Guangdong, and Wu Guang fought in Siege of Suncheon with his Guangdong troops. Regardless of how you feel about Korean records, we know for a fact that Guangdong troops could, and DID, make it to Korea in time.

      Siam case is even more simple: Ming Dynasty eventually refused because of political reasons, not logistics.

      "Confucians didn't put a high value on maths, science, technology, and commerce."
      Wrong and has no relavance to the topic we are discussing. You don't need calculus and abstract numbers to calculate head count. Yes, the writters were not saints but ordinary people like you and me, which means they had at least basic level of competency in math, like you and me.

      Again, not saying that you can't cast doubt on the numbers, but the reasons you give so far are too flimsy to challenge/dismiss it. (you can also read my discussion with Dmitry about Japanese casualties on Day 1 as well).

      I've repeatedly said that I am currently double-checking to see if the record holds up. One day I might even find something concrete enough to dismiss it, but it won't be the reasons in this comment chain.

      Delete
    11. “then Koreans and Japanese should have a much deeper impression of them, not just a very slight mention. ”
      I should remind you that this is a repeat of argument from silence fallacy.

      Also, a number of Zhejiang commanders and teachers stayed in Korea and helped to teach them their ways/rebuild Joseon army during the negotiation phase, which may explain deeper Korean understanding/impression of Zhejiang troops.

      Delete
    12. Speaking of Korean accounts, isn't there that comment on the Langbing article that links to a picture of uh, literal werewolves, or monkeys, embedded in the Ming army? You suspected there that it might have been referring to Liu Ting's 'weird circus legion'

      http://pds21.egloos.com/pds/201412/21/34/a0053134_5495d8c01f486.jpg


      If it's legit, it definitely seems to be the case that even some of the more 'exotic' Ming troops brought over there left a big impression on the Koreans, even if they ended up being rather fantastical. Or straight out of a furry convention.

      Delete
    13. So what you're suggesting is that when the Chinese were fighting the Wokou a mere few decades before the Imjin War, everything was terrible no organization no logistics and it took several months to travel from Guangxi to Zhejiang. And yet when the Imjin War started everything was fine the logistics was perfect. Is that what you're suggesting? Yet this contradicts with what you've written in your blog post. If Chinese logistics during the war was perfect and it can transport large quantities of troops from faraway places like Guangxi and Guangdong to Korea, then why Ming troops were starving and why they depended on Korean logistics?

      And the points that I raised are valid points, just because you cannot refute them doesn't make them flimsy.

      Delete
    14. @Bleb

      First of all, Lang Bing weren't dressed like monkeys or wolves. The word "Lang" originally was written like "俍" with a human radical on the left rather than a dog radical, and it was an ethnonym which referred to the ancestors of the Zhuang and Dong peoples. It's only later that it got confused or mixed up with "狼" the character for wolf. I don't deny that the Lang Bing might have worn some exotic forms of armor like rattan or lacquered leather, but they definitely didn't dress like animals.

      Secondly, based on the description of Lang Bing (like their ability to fight in highly coordinative small squads, their proficiency with crossbows, and their ferocity), you'd expect that they would leave a much deeper impression to Koreans and Japanese. Yet the Koreans and Japanese records were all describing Zhejiang troops and Northern troops, there were barely any record whatsoever for Lang Bing. If it was indeed the case that a large quantity of them were sent to Korea, then why there's so little mention of them in both Chinese and Korean/Japanese records?

      Delete
    15. @Unknown

      I think this is getting a bit heated, so I hope I didn't come across as offensive or something. All I was doing was pointing out that the painting/picture in question is peculiar with the artist's choice to depict auxiliaries or more exotic Ming troops as so savage that they apparently looked "beast-like". I'm only drawing attention to this because it seems to indicate that these troops DID leave quite an impression on the Koreans, regardless of how ridiculous or fantastical these impressions ended up being.

      Delete
    16. @Unknown
      Your problem is that you simply ASSUME that Guangdong-Zhejiang marsh to be the upper limit of how fast these troops could move, which is a baseless assumption. There could be many other factors that cause these troops to move slower or faster.

      And then you simply take that assumption and apply it to Korea to calculate the time required to march from Guangdong to Korea, which is what makes your point extremely flimsy (i.e. your speculation is built upon baseless assumption, which itself come from an "example" that you did not even elaborate).

      i.e. assumption, speculation and argument from silence (lack of impression part as per previous comment), as I've been pointing out. It is far too weak to use a speculation like this to challenge something clearly written in historical source.

      Also, the fact that Wu Guan and his Guangdong troops were fighting in Korea already soundly refuted your entire point. Their mere presense demonstrates that it was physically possible to march large group of people from Guangdong to Korea fast enough to participate in Imjin War.


      As for your last comment:
      No, I do not "suggest" any of that so don't put words in my mouth. Those are some examples of factor that may slow down the troops, but I do not affirm any of them as definitely true or in play (note that I used the word "may", both in this comment and my previous one). You failed to elaborate on this Guangdong-Zhejiang march, so example is all I could give.

      Delete
    17. Historical sources weren't written by gods but by ordinary people, anyone can challenge or doubt what was written. Yes, I acknowledge that the points that I raised are mostly assumptions, but they're logical assumptions nonetheless.

      The point is how many Guangdong troops were present in Korea and when did they arrive? I don't doubt that a few of them can arrive in Korea, but the point is their number were very few compared to Northern troops and Zhejiang troops. And Guangxi and Guangdong are different, I don't think you can mix the two up. Although they're both quite faraway from Korea.

      The 1555 battle that I mentioned was when commander Zhang Jing recruited a hereditary Tusi chief from Guangxi named Mrs. Wa and her Lang Bing troops to fight the Wokou in Zhejiang. You can look it up if you don't believe me. It took them a long time to get to Zhejiang.

      Delete
    18. @Unknown
      I think we are simply going in circles here.

      - Historical sources weren't written by gods and can be challenged.
      Yes, no disagreement.

      - Lang Bing were fewer than Northern troops & Zhejiang troops.
      Fewer yes (as per my comment in the previous chain), but still a significant number given the Korean record.

      - Assumptions to challenge Korean records
      There are simply too many unknowns in your assumption to make it a viable challenge. To name but a few:

      *We don't know whether Guangxi troops were marching at full speed to Zhejiang or not.

      *We don't know how applicable the above example is to Guangdong troops to Korea, given that they were decades apart, taking different route and marching to different location, as well as numerous other factors to consider.

      *The assumption that the number recorded by Koreans was too much is completely arbitrary. What makes (for example) 1,000 troops plausible, but 5,000 troops too much? What about 2,500? or 3,000? or 75,00? There is literally no standard to judge which number is too high. In other words, your doubt to that number amounts to "because I say so".



      Also, I take it the 1555 battle you refer to is the famous Battle of Wangjiangjing?

      I don't have many info on that battle. However, based on what I know, Lady Wa departed Tianzhou (Guangxi) on January 4th, 1555 (12th day of the last month of Jiajing 33th), and arrived at Jinshan (Zhejiang) on April 3rd. That's about 3 months. However, She fought in Shengdun (Pingwangzhen in modern China) beforehand, and stayed in Suzhou for about one month before going to Jinshan. Upon arrival, she waited for another month before fighting the famous Wangjiangjing battle.

      Delete
  31. @Bleb

    Everything suggests that Northern troops and Zhejiang troops formed the majority of the Chinese troops that participated in the Imjin War, whereas troops from faraway places like Guangdong or Sichuan were either too few or didn't really participate in most battles due to logistic issues. This can be supported by Japanese and Koreans sources as well. Japanese paintings (I forgot the name) showed Chinese troops to be mostly cavalry.

    ReplyDelete
    Replies
    1. I can't believe I forget to address this comment for so long.

      Actually, that's not exactly accurate. The majority of Ming soldiers that went to Korea were indeed Northern troops, but among the Southern infantry, surprisingly few came from Zhejiang.

      For the first invasion, we have two notable Southern commanders: Wu Weizhong and Luo Shangzhi, the former led 900 Ji Garrison infantry and 2,100 Zhejiang infantry, the later led 600 Ji Garrison infantry. So, overall 2,100 Zhejiang infantry, or 3,600 if you count Ji Garrison infantry as “Zhejiang”.

      For comparison, Liu Ting alone brought 5,000 Sichuanese infantry to Korea during the first invasion. He alone put Zhejiang infantry in the minority.

      For the Second Invasion, there were indeed more Zhejiang troops. From what I gather, Chen Yin had 4,000, Wu Weizhong had 3,900, Mao Guoqi had 3,900, Ye Bangrong had 4,000, (counting maximum numbers). There's another commander Lan Fangwei with ~4,800 Zhejiang troops, but his troops were cavalry (we know this from Siege of Sacheon). Another commander Zhang Bang also led 4,000~6,000 Zhejiang troops, but he supposedly arrive late and only entered Korea in 1599, after the war was over.

      Put together, there were around ~15,800 Zhejiang infantry in Korea throughout the Second Invasion, or 20,600 if you count cavalry. This does put them at a slight majority among Southern troops, although still fall short of being over 50% of all infantry (number of troops from Sichuan and Guangdong were about equal to them).

      I wouldn't call Zhejiang infantry "participated in most battles" either. For the first invasion, they fought in Siege of Pyongyang and Battle of Angang (and somehow suffered more loses than even Battle of Byeokjegwan), for the second invasion, they fought in First Siege of Ulsan, and Siege of Sacheon (worst Ming defeat during Imjin War). Not terribly good showing for the handful of battles they participated.

      It's also wrong to call troops from Sichuan and Guangdong “didn't really participate”, as they DID fought in major battle such as Siege of Suncheon. It wasn't due to logistic issues either, since that issue affected everyone, not just limited tot them.

      The reasons they didn't fought more were, ironically, largely caused by Zhejiang troops: Chen Yin kickstarted the Ding Yingtai debacle, causing entire Ming army to be beset in chaos and confusion, devastating its morale. The disastrous defeat at Sacheon caused two other routes to retreat. After those defeats, Ming army secretly enter ceasefire talks with the Japanese, allowing them to return to Japan. This was, again, initiated by Mao Guoqi, another Zhejiang commander.

      Delete
  32. Hi, sorry this comment is slightly unrelated to your original post but would it be possible if you could direct me to some books that describe these battles?; (车岘之战、万福寺之战、乡校后峰之战、南原南村之战、求礼之战、潺水驿之战)

    I saw these battles mentioned in an Imjin War article describing numerous other Ming and Japanese land battles. However, it only briefly named these and when I search up their individual names I can get no further detailed information on these battles. Thank you so much.

    https://www.sohu.com/a/462881027_100145375

    This was the original article I found it from.

    ReplyDelete
    Replies
    1. Good day ANON. I can't give you to the exact quote, but I can point you the source that records these battles: Nanjung japnok《亂中雜錄/난중잡록》. You can read about (some of) them here:
      http://www.davincimap.co.kr/davBase/Source/davSource.jsp?Job=Body&SourID=SOUR002247&Lang=%ED%95%9C%EB%AC%B8&Page=3

      Not much to be said about them, as these are skirmishes that happened between Battle of Jiksan and Siege of Ulsan, as Ming/Joseon forces pursued after Japanese forces that were in the process of retreating to southern coasts of Korea to build their fortresses.

      Delete
    2. Thank you so much! Do you think it's possible Japanese sources would record these engagements as well? Given that they are skirmishes I'm worried that Japanese sources would brush over these engagements in favor of more significant land battles.

      Delete
    3. Indeed you are unlikely to find them in Japanese sources unless someone significant took part in it. Had it not been Jo Gyeong-nam, (author of Nanjung Japnok), we might not even find them in Korean source. Jo Gyeong-nam was a Righeous Army leader that was active in Namwon, the skirmishes were mostly from that area.

      There are other skirmishes around the same time/same region recorded in other sources though (Sillok recorded a few, and one skirmish was recorded in a Japanese source).

      Delete
  33. I got the impression that Ming generals and Ming fans think very highly of Konishi Yukinaga. They took special precautions against him during Ulsan, mistook Kuroda Nagamasa to be him and blamed him for the failure to take Ulsan even though he was nowhere to be seen.

    ReplyDelete
    Replies
    1. More or less. Konishi Yukinaga was pretty competent and pragmatic, and overall can be considered some of the better-performing commanders on the Japan side. Losing Pyongyang wasn't purely his fault too, he bore the brunt of the then-unfamiliar Ming army alone while other Japanese commanders, like Otomo Yoshimune, abandoned him and fled.

      Delete

< > Home

Random Quotes & Trivia

GREAT MING MILITARY © , All Rights Reserved. BLOG DESIGN BY Sadaf F K.