Battle of Hansan Island

The Battle of Hansan Island was a naval battle of the Imjin War, the Japanese invasion of Korea that began in 1592. It was fought on the 8th day of the 7th lunar month of 1592 by the Joseon calendar. Converted with the table of the Korea Astronomy and Space Science Institute (KASI), that day is 14 August 1592; many reference works, including Wikipedia, give 15 August, and Japanese accounts, which followed Japan’s own calendar, date it to the 7th day of the 7th month. The fighting took place in the open sea off Hansan Island, a small island between Geoje Island and the coast of what is now Tongyeong in South Gyeongsang Province.

On one side was a combined Joseon fleet of about 55 warships from three naval commands: the Jeolla Left Navy under Yi Sun-sin, the Jeolla Right Navy under Yi Eok-gi and the remnant of the Gyeongsang Right Navy under Won Gyun. On the other was a Japanese squadron led by Wakisaka Yasuharu, which Yi’s report put at 73 ships. Yi lured the Japanese out of the narrow Gyeonnaeryang strait into open water, surrounded them in a curved formation called the “crane wing” and destroyed or captured 59 ships, according to Korean records.

Two days later, on the 10th day of the month (16 August), the Joseon fleet attacked a second Japanese squadron under Kuki Yoshitaka and Katō Yoshiaki at the harbour of Angolpo and, according to Yi’s report, wrecked most of it. Samuel Hawley puts Japanese losses in the two battles together at roughly a hundred ships. Soon afterwards Toyotomi Hideyoshi, the ruler of Japan who had launched the invasion, ordered his commanders to stop seeking battle at sea.

In Korea the battle is counted, with the defence of Jinju in 1592 and the battle of Haengju in 1593, among the “three great victories” of the war. It is widely seen as the moment when Joseon secured control of the southern coastal waters west of Geoje. How much this mattered for the Japanese war effort as a whole is a question historians still discuss; the later sections of this page set out the evidence and the different views.

The invasion began on the 13th day of the 4th lunar month of 1592 (23 May), when the first Japanese division landed at Busan. According to the Northeast Asian History Foundation’s account, Hideyoshi’s land army was organised in nine divisions with a nominal strength of about 158,700 men, plus a separately organised naval force of about 5,000; the same account notes that actual numbers were lower because many daimyo did not fully meet their quotas. The Japanese took the capital, Hanseong, about three weeks after landing (2nd–3rd day of the 5th month, 11–12 June) and Pyongyang in the middle of the 6th month (around 21 July; sources differ by a few days). King Seonjo fled to Uiju on the Chinese border.

Taking cities, however, was not the same as controlling a country. The Japanese held the main roads and walled towns, but much of what a force of that size needed, from gunpowder to fresh troops, had to be brought up a long corridor from Busan through hostile country, while food was also seized locally. Local “righteous armies” (uibyeong) soon began to attack these lines.

At sea the Japanese met their first real defeats. Yi Sun-sin, commander of the Jeolla Left Navy based at Yeosu, sailed east and on the 7th day of the 5th month (16 June) destroyed Japanese ships anchored at Okpo on Geoje Island. In a second sortie he won at Sacheon on the 29th day of the 5th month (8 July), where the turtle ship went into action for the first time and Yi was wounded by a bullet, and then at Dangpo and Danghangpo early in the 6th month (10 and 13 July). Several of these raids caught Japanese ships at anchor along the coast while many of their crews were ashore.

Hideyoshi responded by ordering three daimyo, Wakisaka Yasuharu, Kuki Yoshitaka and Katō Yoshiaki, to destroy the Joseon fleet. They gathered at Busan, but Wakisaka sailed out ahead of the other two; Japanese accounts say that he wanted the credit for himself. Meanwhile, on land, a Japanese army was pushing toward Jeonju, the capital of Jeolla province. On the same day as the battle of Hansan, the Joseon commander Gwon Yul held the Ichi pass against it. Japanese troops did occupy parts of Jeolla, such as Geumsan, but resistance on land, including Ichi, helped keep Jeonju and most of the province, Joseon’s richest rice-growing region and the home of Yi’s fleet, out of Japanese hands.

Yi Sun-sin and Yi Eok-gi assembled their fleets at Yeosu, drilled them together on the 5th day of the 7th month (11 August) and sailed the next day. At Noryang they were joined by Won Gyun with seven ships, giving a combined fleet of 55 warships by the Encyclopedia of Korean Culture’s count (56 in Samuel Hawley’s history). On the evening of the 7th (13 August), at Dangpo, a local cowherd named Kim Cheon-son told them that more than seventy Japanese ships had anchored in the strait of Gyeonnaeryang.

Gyeonnaeryang is a narrow channel between Geoje Island and the mainland, with hidden rocks. In his report Yi explained that his large warships could not manoeuvre there without colliding, and that the Japanese, if cornered, would simply escape onto land. The sea in front of Hansan Island, by contrast, was open and far from the mainland, so that men who swam ashore would be stranded. The Revised Annals of King Seonjo, compiled decades later, add that Won Gyun wanted to attack at once and that Yi overruled him; this story should be read with care, because the rivalry between the two commanders coloured many later accounts.

On the morning of the 8th, five or six Joseon warships advanced on the Japanese anchorage and then turned back as if fleeing. The Japanese ships pursued them out of the strait into open water. There Yi ordered his fleet to form the crane-wing formation, a wide curve that closed around the head of the Japanese column, and to open fire. After the first two or three Japanese ships were shattered, the rest lost their order, and the fighting went on for the rest of the day.

According to the figures in Yi’s report, as summarised by the Encyclopedia of Korean Culture, 47 Japanese ships were burned or broken and 12 captured. About 400 Japanese abandoned their ships and fled onto Hansan Island, while 14 ships at the rear escaped. Wakisaka himself got away. No Joseon warship was lost, although men were killed and wounded; for Hansan and Angolpo together, Hawley gives 19 dead and 114 wounded.

The backbone of the Joseon fleet was the panokseon, a large warship with a raised upper deck. Its high, heavy hull made it hard to board, and it could carry many cannon. Most Japanese ships, by contrast, were lighter, had been gathered largely to carry troops and supplies, and at the start of the war carried few or no cannon. Japanese crews relied on arquebuses and on boarding enemy ships to fight hand to hand, the method that had served them well in Japan’s own wars.

Joseon warships carried several sizes of gun, called by the characters of a traditional sequence, cheonja (“heaven”), jija (“earth”) and hyeonja (“black”), which could fire large arrows as well as shot. This meant that Yi could fight at a distance, outside the effective range of Japanese muskets, and destroy ships before their crews could board. The Northeast Asian History Foundation’s account says that Yi adopted the crane-wing formation at Hansan to raise the hit rate of his guns; curving the line of ships around the enemy also meant that more guns could fire on the head of an advancing column at the same time.

Modern reconstructions of the crane wing, such as the one in English Wikipedia’s article, place the heavier ships, including the turtle ships, in the centre and lighter ships on the two wings; Yi’s report names the formation but does not describe where each ship stood. How many turtle ships took part is also uncertain: some accounts say two, others three. The turtle ship, a covered warship whose exact design is still debated, is the most famous symbol of the battle, but it was a small part of a fleet that won mainly through the gunfire of ordinary panokseon.

Command mattered as well. The three Joseon commanders acted together under Yi’s leadership, while the Japanese side had no overall naval commander: Wakisaka, Kuki and Katō were separate daimyo, each with his own forces and interests, and Wakisaka’s decision to sail ahead alone meant that he met the united Joseon fleet with only part of the available Japanese strength.

On the 9th, scouts reported about forty Japanese ships at Angolpo, a harbour near today’s Jinhae. These were the ships of Kuki Yoshitaka and Katō Yoshiaki, who had followed Wakisaka. On the 10th (16 August) Yi attacked. This time the Japanese, who knew what had happened at Hansan, kept their ships close inshore in shallow water and refused to come out. Yi’s ships took turns sailing in and firing, and by evening, according to Yi, almost all the Japanese ships had been wrecked; the survivors escaped by land and, during the night, by sea. Yi wrote that he deliberately left a few boats intact so that the Japanese on shore would leave by sea rather than turn on the Korean refugees hiding in the hills nearby.

The court rewarded the victories. Yi was raised to the rank of jeongheon daebu, the senior second rank, and Yi Eok-gi and Won Gyun were also promoted.

The Japanese reaction was to change tactics. Hideyoshi ordered his naval commanders to stop seeking battle at sea and instead to hold their anchorages around Busan and Geoje from fortified positions on shore. The Japanese historian Ogawa Takaaki (2022) cites a letter under Hideyoshi’s vermilion seal to Wakisaka, dated the 14th day of the 7th month by the Japanese calendar, which ordered him to build and garrison forts on Geoje Island with Kuki and Katō and forbade separate battles with the Joseon fleet. When Yi attacked the main Japanese base at Busan on the 1st day of the 9th month (5 October 1592), he destroyed many ships at anchor but could not take the port. Ogawa notes that Yi did not attack Busan again, and that five attacks on the fortified Japanese positions at Ungcheon early in 1593 failed to destroy the Japanese there.

In 1593 Yi moved his headquarters to Hansan Island, from which he could watch the Gyeonnaeryang route toward Jeolla, and became the first Samdo sugun tongjesa, commander of the navies of three provinces. When he was later arrested, his patron Yu Seong-ryong is said to have warned that if Hansan were lost, Jeolla could not be held. In 1597, after Won Gyun’s fleet was destroyed at Chilcheollyang, the base was abandoned.

Why could an army that took Hanseong in three weeks and Pyongyang in two months not finish the conquest? Part of the answer is supply. A force of well over a hundred thousand men needs food every day. In the first months the Japanese lived partly on what they captured, including government grain stores such as those at Pyongyang, but the further they marched, the longer and more exposed the line back to Busan became. The Northeast Asian History Foundation’s account notes that this overland supply route from Busan to Pyongyang was harassed by righteous armies in the Japanese rear.

Korean writers have long linked Hansan to this problem. Yu Seong-ryong, chief minister during the war, records in his memoir Jingbirok that after taking Pyongyang the Japanese commander Konishi Yukinaga sent a letter to King Seonjo announcing that “more than a hundred thousand of Japan’s sea forces are also coming by the western sea” and asking where the king would go from there. Yu concluded that the Japanese had meant to advance west by land and sea together, and that this one battle “cut off one of their arms”: Konishi, though he held Pyongyang, was isolated and did not dare to advance further. Korean reference works, such as the Encyclopedia of Korean Culture and the Northeast Asian History Foundation’s history, call this intended strategy suryuk byeongjin, a “combined land-and-sea advance”, and hold that Hansan defeated it.

Just as important was what the defenders kept. Yu wrote that because of the battle Joseon could hold Jeolla and Chungcheong and the coasts of Hwanghae and Pyongan, move grain and pass on orders. Jingbirok gives an example from the same weeks: a shipment of 1,200 seok of tax rice from Asan in Chungcheong reached Jeongju in the far north-west by sea, and Yu used it to help feed the Ming troops arriving from China. The shipment came at about the same time as the battle, so it illustrates the value of the western sea route rather than an effect of the victory; the sea lanes served whichever side controlled them.

The limits are also clear. Hansan did not cut the Japanese army’s main crossing between Tsushima and Busan, which remained in Japanese use; Yi himself heard after Angolpo that dozens of Japanese boats a day were still coming and going near Gimhae. The Japanese forces in the north went on being supplied, poorly, through vulnerable overland communications and by what they could take locally. What the battle did was deny the Japanese secure use of the route around the south-western coast, a potentially valuable way to the north, and keep most of Jeolla and the western sea in Joseon hands. In that sense it widened the gap between how far the Japanese could march and how much of the country they could actually hold.

Even the size of the Joseon fleet varies with the source. The Encyclopedia of Korean Culture counts 48 ships from the two Jeolla commands plus Won Gyun’s 7, making 55; Hawley gives 56; Yu Seong-ryong’s Jingbirok speaks of Yi leading forty panokseon. Such differences probably reflect which commands and which kinds of vessel are counted.

For the Japanese squadron, Yi’s report gives 73 ships: 36 large, 24 medium and 13 small. The Revised Annals say “more than seventy”. An English translation of Yi’s report prints 82 ships (36 large, 34 medium, 12 small), a figure repeated in some English-language accounts, but 73 is the total that fits the report’s own tally of 59 ships destroyed or captured plus 14 that escaped. A Japanese account from the Wakisaka side, cited in Japanese Wikipedia, speaks of Wakisaka taking sixty or seventy ships and losing 39 of them burned. For Angolpo the figures diverge again: Yi wrote that almost all of the 42 ships he found were smashed, apart from a few he deliberately left; the Revised Annals say 40 were burned; English Wikipedia’s article on Angolpo says that 22 vessels, including Kuki’s flagship Nihon Maru, got away. These accounts imply different losses; reconciling them would require knowing how each treated damaged, abandoned or later recovered vessels.

Human losses are even less certain. For Hansan and Angolpo together, Yi reported that the men of the three commanders took nearly 250 heads and that many more Japanese drowned. The Revised Annals repeat a rumour from the Japanese camps that 9,000 Japanese died at Hansan, which cannot be checked; a Japanese muster list assigned Wakisaka 1,500 men for the campaign, though rowers and sailors may not be fully reflected in such figures. On the Joseon side, Hawley gives 19 killed and 114 wounded for the two battles together. Yi’s report lists the dead and the wounded by name and duty, including sailors, gunners and divers, some of them recorded as private or temple slaves, a reminder of who rowed and fought in the fleet.

Dates need the same care. Joseon used a lunisolar calendar, and converting the 8th day of the 7th month of 1592 with the KASI table gives 14 August 1592, the date used on this site. Many books and Wikipedia say 15 August; the reason for that one-day difference has not been established in the sources used here. Japanese accounts, following Japan’s calendar of the time, give the 7th day of the 7th month.

Ming China, Joseon’s ally, had its own crisis in 1592. In the spring a garrison revolt broke out at Ningxia on the north-western frontier, and the rebels held the city for months. The Ming suppressed it by October 1592, which allowed the court to turn its forces toward Korea; In the same year the Jurchen leader Nurhaci offered to send troops to help Korea against Japan, but the Joseon government refused, and on 28 November 1592 Nurhaci’s son Hong Taiji was born, the man who would later found the Qing dynasty. Nothing connects his birth to the war; it simply happened in the same year.

In Nanjing, 1592 also saw the printing of the earliest surviving edition of Journey to the West, the great novel of the monk Xuanzang’s pilgrimage and the monkey king Sun Wukong. The book had nothing to do with the war; it is a reminder that Ming publishing and urban culture carried on while the empire fought on two frontiers.

Far to the west, in 1592 Galileo Galilei moved to the University of Padua, where he would teach for the next eighteen years; the French essayist Michel de Montaigne died on 13 September 1592; and in London a pamphlet published under the name of Robert Greene, registered on 20 September 1592, mocked an “upstart crow”, usually read as the first surviving reference to William Shakespeare as a playwright. None of these events was connected with Korea. Setting them side by side shows only that the year of Hansan was also a year of new science, new literature and new theatre elsewhere in the world.

In Korea, Hansan is one of the best-known battles in the country’s history and a central part of the memory of Yi Sun-sin as a national hero. The site of his headquarters on Hansan Island, Jeseungdang, is marked today by a memorial stele and shrine buildings, and the battle is remembered as the moment the war at sea turned in Joseon’s favour.

Foreign admirers helped spread the battle’s fame, and some of their mistakes. In The Passing of Korea (1906) the American missionary and historian Homer Hulbert called Yi’s victory “the turning-point in the war”, but described it as an encounter with a Japanese fleet “bringing sixty thousand reinforcements”, which Yi’s turtle boat rammed; the Encyclopedia of Korean Culture also attributes to Hulbert a comparison with the Greek victory at Salamis. The British admiral George Ballard, in a 1921 book on the sea and Japanese history, wrote that not even Nelson could have done more, and likewise described a convoy of many hundreds of transports sailing by the western passage, attacked by Korean ships that rammed their way through it. Yi’s own report describes a squadron of 73 ships drawn out of a strait and destroyed mainly by gunfire. Part of the popular image of Hansan as the destruction of a great reinforcement convoy goes back to such early English-language accounts.

In Japan the battle has been remembered less, but not ignored. The history of the journalist and historian Tokutomi Sohō, cited by the Northeast Asian History Foundation from a 1935 edition, named Joseon’s naval superiority as one of three main reasons the invasion failed, along with the righteous armies and the arrival of the Ming army. A tradition in the Wakisaka family holds that Wakisaka, after the defeat, hid on an island and lived on seaweed for days; this is a family story, not a documented fact.

The battle has also become a subject of popular culture. It was the subject of the 2022 South Korean film Hansan: Rising Dragon, directed by Kim Han-min, and has been dramatised many times on Korean television.

The strongest interpretation goes back to Yu Seong-ryong. In Jingbirok he argued that Hansan broke a Japanese plan to advance west by land and sea together, left Konishi stranded at Pyongyang, saved Jeolla, Chungcheong and the western coasts, and even spared the Chinese coast around Liaodong and Tianjin from alarm, so that the Ming army could come to Korea by land. Modern Korean reference works follow this line: the Encyclopedia of Korean Culture says the battle frustrated the Japanese combined land-and-sea plan and turned the war in Joseon’s favour, and the Northeast Asian History Foundation, citing the national Korean history series, calls it the decisive turning point of the war. In English, Samuel Hawley (2005) interprets Japan’s strategy as requiring a sea supply route around Korea’s south-western coast to Pyongyang and treats the naval campaign of 1592 as having halted Japanese naval operations along the southern coast. Some Japanese historians take a similar view: Kasaya Kazuhiko and Kuroda Keiichi (2000), as quoted by Ogawa, write that Yi’s navy struck a decisive blow at the Japanese army by cutting its supply lines.

Other historians raise different doubts. The naval historian Yi Min’ung questions the coherence of Japan’s initial naval planning, arguing that the Japanese navy at first lacked an overall plan to defeat the Korean navy and secure command of the sea; this concerns a different aspect of planning from Hawley’s supply-route interpretation and does not by itself show whether a western sea route was intended. Ogawa Takaaki’s research note (2022) questions the evidence for a western-coast supply plan itself: he reports that the combined land-and-sea strategy does not appear in the major studies of Kitajima Manji and Nakano Hitoshi or in Yi Hyeong-seok’s history of the war, and that no order from Hideyoshi to send the fleet up the west coast appears in Nakano’s studies of Hideyoshi’s military orders; he reads Konishi’s letter as a bluff, comparable to the Ming envoy Shen Weijing’s threat of 400,000 Ming troops, and suggests that Yu, who had recommended Yi for his command, may have overrated his victories. Ogawa also cites Tokutomi Sohō’s view that Hideyoshi, at least early in the war, expected his navy to do little more than escort transports between Nagoya and Busan. The evidence itself has limits that fit these doubts: Wakisaka’s squadron at Hansan was a fighting force sent to destroy Yi’s fleet, not a convoy carrying supplies to Pyongyang, and the Japanese had raised only a small dedicated naval force.

A second question is how to weigh the navy against other causes. Konishi in fact halted at Pyongyang in the middle of the 6th month, about three weeks before Hansan, so the battle cannot explain his first pause; Yu’s argument concerns why he did not move on afterwards. Here other pressures also counted: attacks on the corridor behind him; the first Ming attack on Pyongyang in the 7th lunar month (on the 17th by the Northeast Asian History Foundation’s account, the 19th in Jingbirok), which that account counts among the reasons the Japanese gave up further advance north; the approach of winter; and the truce talks that the Ming envoy Shen Weijing opened with him in the autumn. Kenneth Swope’s study of the Ming side (2009) argues that it was the Ming intervention on land, above all the recapture of Pyongyang in early 1593, that turned the war, and that accounts centred on Yi Sun-sin and the righteous armies underrate it; he nonetheless treats Hansan as an important victory that denied Japan control of the sea. Disentangling the effect of a single battle from all of these is difficult.

Historians therefore differ both over the navy’s relative importance and over how firmly the evidence establishes a Japanese western-coast supply plan. What the sources cited here agree on is narrower. Hansan gave Joseon control of the sea west of Geoje for the rest of the first invasion and helped protect most of Jeolla and the western coast, while the crossing between Japan and Busan stayed open and the Japanese armies in the north depended on vulnerable overland communications and local procurement. Combined with the righteous armies, the Ming intervention and the winter, Joseon’s control of these waters was one of the conditions under which the Japanese gave up the north in 1593.

The most detailed source is Yi Sun-sin’s own report to the court on the battle, written within days and later collected with his other reports as Imjin jangcho. It gives routes, dates, the names of captains who distinguished themselves and the names of the dead. It is close to the events but is still a commander’s account of his own victory, and its figures for enemy ships and heads served to support claims for reward.

The Korean court records need care of a different kind. The Revised Annals of King Seonjo were compiled decades after the war and include second-hand material, such as the rumour of 9,000 Japanese dead. Yu Seong-ryong’s Jingbirok, written after the war by a minister who had recommended Yi for his command, is an invaluable interpretation but sometimes merges events: in its account of the battle near Gyeonnaeryang it describes the turtle ship and Yi being hit in the left shoulder by a bullet, a wound that other records place at Sacheon in the 5th month.

Japanese sources include Hideyoshi’s orders and records of the Wakisaka house, such as the account cited in Japanese Wikipedia as Kōraisen senki, which speaks of Wakisaka’s sixty or seventy ships and 39 burned. Its date and transmission are not set out in the sources used here, so its figures can be set beside Yi’s report but not easily weighed against it. The two sides also counted different things, such as ships burned, ships captured and ships that escaped, so even a careful comparison yields estimates rather than firm totals.

Even the exact place is uncertain. Published maps and reconstructions differ: English Wikipedia places the battle off the eastern coast of Hansan Island, and the marker on this site’s map is approximate. A 2019 Weekly Kyunghyang report describes several earlier proposals in the waters between Tongyeong and Mireukdo, and the argument of Park Jong-pyeong, a translator of Yi’s war diary, who used Won Jung-geo’s eighteenth-century account of Japan, Hwagukji, together with the Wakisaka records to place the main fighting in the waters north of the island, about twenty li from Gyeonnaeryang. That argument depends on how old distance measurements are converted and has not settled the question.