From SamuraiWiki
Jump to navigationJump to search
  • Japanese/Chinese/Korean: (wakou / wōkòu / waegu)

The wakô were medieval East Asian seafarers often described as raiders, pirates, or brigands, active in East Asian waters in the Kamakura to early Edo periods, with the phenomenon peaking in the 16th century (the late Muromachi or Sengoku period). The term might be literally translated as "Japanese pirates," the wa (倭) denoting Japan, but many wakô were in fact Chinese, Korean, Ryukyuan, or from mixed or ambiguous ethnic or national backgrounds.[1][2] Because of the Muromachi shogunate's weak control over Kyushu, and the lack of any centralized authority in the Ryukyu Islands (especially prior to the 16th century), these regions became major centers of wakô activity, and Japanese came to be used as a common language among the wakô. However, while many wakô groups enjoyed some kind of relationship with local authorities such as samurai houses or the Seiseifu in Kyushu, they were not in truth controlled by or otherwise "belonging to" Japan, or Ryûkyû.[3] While Chinese and Korean perspectives on these figures has colored their characterization today as "brigands" and "pirates," scholars such as Peter Shapinsky suggest the English term "sea lords"; while not denying the fact that wakô were often violent, self-serving, power-seeking, and so forth much like warlords more generally, thinking of them as "sea lords" may help permit or encourage a different (no less violent, but perhaps less criminal or marginal) perspective on their activities and political positions.[4]

Chinese primarily sources of the mid-16th century identify the wakô problem at that time in particular as stemming chiefly from the activities of merchants and others in China, who hired or otherwise encouraged Japanese to be involved. Some scholarship suggests that from the very beginning of the Ming Dynasty in China (1368-1644), the anti-maritime policies of the first Ming emperor - forcing coastal communities to move inland, and trying to monopolize all maritime trade under the throne - were a chief cause of, rather than a response to, the proliferation of smugglers, who then became brigands or pirates.

Though traditionally regarded as "brigands" or "pirates," many scholars today describe the wakô in broader, more complex terms. Though certainly operating on the fringes of the law, and in many cases resorting to violence, wakô represented a complex mix of adventurers, smugglers, traders, mercenaries, and so forth, more often engaging in violence for economic gain than out of a pure desire for violence or chaos.[3] Many of those based in the Inland Sea and in coastal areas around Kyushu lived on houseboats known as ebune, housing a mixed community of merchants, families, adventurers of various sorts, the remnants of defeated warrior bands, fishermen, and others down on their luck. When they did engage in violence, however, wakô were known to not only fight with bow and sword, but also sometimes to use battering rams or other devices to punch holes in enemy vessels; it was also not unknown for wakô to ravage far inland, traveling on horseback.[5] Wakô typically sourced their horses from Jeju Island, the Gotô Islands (near Nagasaki), Tsushima, and Iki Island.[6]

Only after the Korean Invasions of Toyotomi Hideyoshi in the 1590s, in which organized samurai invasion forces were labeled by the Chinese and Koreans as wakô, i.e. as pirates or brigands, it would seem, did the earlier history of the wakô come to be colored, in Chinese and Korean sources, by implications or assumptions that the wakô were somehow agents of a central Japanese authority. Though documents written in the 16th century generally identify Chinese as having been the source of encouragement for piratical activities, those written in the 17th century and later, especially the Ming shi ("Official History of the Ming Dynasty") generally implicate the Japanese authorities in organizing and backing the wakô, or at the very least refusing to take action to curb wakô activities. In the 20th-21st centuries, scholarship and school textbooks, closely based upon these later 17th century sources, have come to link the wakô, and the foreign relations policies of the Japanese authorities at the time, with transhistorical notions of the Japanese as militant and expansionist.

Early Wakô

The earliest extant usage of the term "wakô" (K: waegu) may be in a 1223 entry in the History of Goryeo (Goryeosa); piracy or brigand-like activity in the East China Sea of course dates to long before that, however.[7]

During the reign of the first emperor of Ming, great efforts were made to establish coastal fortifications to defend against the so-called "Japanese pirates." However, raids and attacks on the Chinese coast at this time were led primarily not by Japanese, but by the Emperor's Chinese political rivals.[8] The Hongwu Emperor sought to restore relations with the Ashikaga shogunate, and to get the shogunate to take action to curb wakô attacks, but, perhaps in part due to the chaos and disunity of the Nanboku-chô period, the Emperor's efforts were unsuccessful. The Chinese Prime Minister Hu Weiyong was executed in 1380 for allegedly conspiring with the Japanese and with Yuan loyalists to overthrow the Ming; as a result of the discovery of this conspiracy, efforts to restore relations with Japan were ended.[9]

The shogunate officially acknowledged Ming China as its suzerain in the 1390s, and entered into tributary relations which would continue through the 15th century. However, wakô raids occurred during this time as well. On occasion, the Japanese authorities would capture Japanese pirates and present them to the Ming authorities as proof that they were taking action, but the shogunate did not always demonstrate willingness to take action against the pirates. More to the point, despite the fact that the shogunate did not always have the ability to enforce order on the seas, to apprehend pirates, and to otherwise deal with the situation, the Ming Court seems to have been convinced that the shogunate had full power to control the raids.[10] This insistence on the part of the Chinese authorities would continue through the 16th century, and would prove a major obstacle to friendly relations between China and Japan.

Wakô raids were a major problem for Joseon Dynasty Korea as well at this time, and remain a prominent issue in anti-Japanese sentiment in Korea today. The period from roughly 1350 to 1420 is considered the earlier peak period (zenki) of wakô activity.[5] Attacks on Korea grew in intensity around 1350, and especially around 1375 to 1388; though the Korean court sent embassies to Kyoto in 1366 and at other times asking the Ashikaga shogunate to take action to curb the piracy, this was ultimately not within the shogunate's power, as the shogunate had little actual control over Kyushu, and a number of powerful wakô groups were both supporting and being supported by figures such as Prince Kaneyoshi of the Southern Court, head of the Seiseifu and political rival to the emperors of the Northern Court with whom the Ashikaga were close. Knowing that spoils from wakô raids on Korea were supporting Kaneyoshi's armies, figures such as Imagawa Ryôshun, Kyushu Tandai in service to the Northern Court, attempted to suppress wakô activity, but overall had little success.[11]

Unable to secure agreements from the Ashikaga shogunate to take efforts to curb the piracy, the Joseon Court took matters into their own hands. In 1419, it launched a Joseon fleet of over 200 ships in an effort to destroy pirate bases on the island of Tsushima. The raids were successful, destroying many pirate ships and villages, but within a few years, the pirate activity returned. The court then turned to a different set of methods, granting titles and seals to members of certain samurai clans, including the Ôuchi of western Japan; these titles and seals conferred official permission to engage in authorized trade, in exchange for samurai assurances that they would take real steps to combat the piracy. In 1443, the Joseon court then entered into an arrangement with the Sô clan of Tsushima, granting the Sô an annual stipend of 200 koku of rice, official permission to send fifty trading ships to Korea each year, and permission to exact maritime fees and taxes on cargoes traveling to Korea, in exchange for the Sô ensuring that all trading ships bound for Korea from Japan were properly authorized, and taking action against those which were not (i.e. the pirates).[12] This marked a special relationship between the Sô and the Koreans, which would continue down to the 19th century. In the Edo period, after the wakô problem had essentially come to an end, the Sô would continue to enjoy a unique position of power as the sole intermediaries effecting communication and interaction between the Joseon Court and Japan (now under the Tokugawa shogunate).

The Ming court, meanwhile, engaged in a different tactic. It curried favor with Prince Kaneyoshi; while many have supposed that the Ming did not understand the true power hierarchies in Japan at the time, and named Kaneyoshi "King of Japan" as a result of this misunderstanding, some scholars now suggest that perhaps the Ming court knew exactly what it was doing, in attempting to ally itself with the authorities who actually had the power to curb or control wakô attacks.[13]

Early 16th Century

Scholars identify a period from roughly 1470 to 1600 as the "latter period" (J: kôki) of wakô activity.[5] Though relations between Ming China and the Ashikaga shogunate were quite good for a time, with merchants from both countries engaging in official trade, by the early 16th century, tensions arose between the two powers. The increasingly weak shogunate did not wield strong control throughout Japan, let alone overseas, and was unable to curb or halt the attacks by Japanese pirates, acting independently, upon the Chinese coast.[14] By the 1530s, Sengoku (i.e. civil war in many parts of Japan) was in full swing, the shogunate held little power, and relations with China had fully soured.

Roughly 1,200 junks, large and small, could be found along the China coast on any given day around this time. Most were simple traders, armed to defend themselves where the Ming authorities wouldn't, and considered "smugglers" under Ming law. Others made a living as armed arbiters, helping to resolve disputes and collect debts where the Ming authorities failed to intervene.[15]

Though the term wakô would come to be applied to a wide range of people, engaging in a wide range of activities, including Chinese traders and pirates, and Japanese traders, that is not to say that there were not, in fact, genuine Japanese pirates, raiders, brigands, or whatever term may wish to apply active on the seas at this time. The Murakami clan of Iyo province, known for their piratical activities in the Inland Sea, were among these; Murakami Zusho, lord of Nôshima is recorded as having led attacks on the Chinese coast, the Philippines, and parts of Indonesia. Iida Kôichirô of Iyo and Kitaura Kanjûrô of Bingo are also known to have commanded raiding parties around this time. One contemporary source relates that "the seven bands" of wakô, though presumably there were many more groups than that, grew to number as many as 1,000 men by 1555, if not earlier, incorporating people from a wide range of walks of life, including ronin, fishermen, and others, mainly from Kyushu and Shikoku.[16] Despite Chinese conceptions or assertions that all of these pirates were "Japanese pirates," in fact the Ming authorities paid out different levels of rewards to anyone who captured or killed a pirate, depending on the identity (e.g. Chinese, Japanese, or Korean) of the pirate.[2]

Concurrent tribute missions sent by the Ôuchi and Hosokawa families clashed in 1523, and burned Ningpo, becoming labeled as wakô. The Chinese authorities responded by banning foreign trade in the area around Ningpo. This led in turn to a rise in illegal trade between the coastal Chinese on the one hand with Japanese and other foreigners. A number of Chinese officials and merchants came to owe great debts to Japanese (or other foreign) traders, and though they sought aid from the local authorities, the foreigners resorted to piracy in order to reclaim the funds owed them, and for survival in the face of Chinese authorities seeking to capture them for the crime of participating in illegal trade.[17] Denied access to a satisfactory volume of official trade[18], the Ôuchi clan remained prominent for some time in commanding, backing, or otherwise encouraging some wakô bands including those led by prominent figures from other provinces and regions. The Ôuchi are often cited as among the chief backers of the wakô, and it is through them that many draw connections between the wakô and Japanese national ambitions. However, the clan was destroyed by the Môri in 1557, while wakô activity continued.[19]

As Chinese demand for, and Japanese supply of, silver rose in the 1530s-40s, a number of Chinese merchants established themselves at bases in Kyushu, selling expensive Chinese silks for Japanese silver, in violation of the Chinese bans. These merchants, including Wang Zhi (d. 1559), Chan Hai (d. 1556), Chen Dong (d. 1556), and Ye Ming (d. 1556), along with their mixed Chinese and Japanese crews, were considered 'wakô by the Chinese authorities as well, despite not being Japanese, and not being involved in any true piratical or raiding activities.[20] One Chinese primary source indicates that the proportion of ethnic Chinese among the so-called "Japanese pirates" may have been as high as ninety percent.[21]

Over time, the raids spread to encompass much of the South China coast. Though the pirates who emerged in the aftermath of the 1523 bans were largely traders, now regarded as smugglers under the ban, their crews gradually came to include people with little or no interest in trade, and more interest in violence and thievery. Huangyan, in Zhejiang province, fell in 1552 to a party of wakô said to number as many as 10,000.[22] Raiders traveled up the Yangtze and attacked cities along its shores the same year, and attacked Nanjing and Chaozhou in 1555. By this time, many of the raiding parties made use of arquebuses (teppô);[16] Meanwhile, the Ming armies were equipped with rather inferior firearms, as there were no centrally-coordinated factories or distribution depots, and generals were left to supply their armies on their own; as a result, Western-style firearms made by individual armies based on models had high failure rates, often failing to ignite, or even exploding in the soldier's hands.[23]

These coastal raids, in which wakô came onto land - sometimes fairly deeply inland - and attacked villages and outposts on land, constituted the majority of their activity, not seaborne combat. In fact, some Chinese sources from the time suggest that the raiders were far more effective at land battle than on the seas, and advocated trying to defeat them at sea, and to prevent them from coming on land.[24] Over the course of the early 1550s, the wakô seized or simply defeated nearly every defense post along the coast, transforming many into their own bases of operations. The situation became particularly severe for the Chinese authorities when the pirate threat spread to Nanjing.[22]

After 1555, the wakô threat to the central Jiangnan region diminished, as the pirates turned their attentions to Fujian to the south, and Anhui to the north. Raids became fewer. Yet, altogether, eleven cities had been captured by the raiders, and countless coastal unwalled market towns attacked. After 1561, wakô attacks diminished even further, except in and around Fujian,[25] and in 1563, Chinese military forces expelled a number of wakô from that region as well.[14]

For six months in 1556, Zheng Shungong, an envoy sent by Yang Yi, the Chinese official in charge of dealing with the wakô, resided in Japan and collected information about the wakô. In his report, published as "A Mirror of Japan" (日本一鑑, Riben Yijian), he writes extensively about Chinese spurring Japanese to engage in piracy and raids on the Chinese coast; he makes no mention of the involvement of Japanese daimyô, and portrays the wakô as decidedly headed by Chinese.[26]

Late 16th Century

While there were certainly many Japanese who did engage in violent acts of piracy and raiding, however, one of the chief factors contributing to the growth of the phenomenon was the Chinese hai jin ban on overseas travel and trade, imposed in 1557. Formal trade with foreign countries (including Japan) was only allowed to occur within the framework of tributary relations, and only at certain designated ports. Strict restrictions were placed on Chinese contact or trade with foreigners. In theory, this was intended to prevent Chinese merchants or seamen from becoming involved with the wakô or other foreign forces, but in practice, such policies were ineffective in preventing contact and trade - a great many Chinese settled abroad and conducted trade and other interactions as "overseas Chinese" no longer subject to Ming law.[1] Furthermore, in the eyes of the Chinese Court, Japanese seamen who sought to trade with Chinese, or to make port in China, as well as many Chinese seeking to trade with Japanese, were considered in violation of the bans, and were labeled criminals, and wakô. In this way, the numbers of the wakô, and their perceived presence, grew dramatically.

The wakô are generally said to have made their bases on Formosa, in the Ryukyus, and in ports, castle towns, and more remote coastal sites on Kyushu. However, the question of the extent to which regional daimyô, particularly in Kyushu, supported and enabled wakô activity is a contentious one, and one of the chief issues involved in the subject of wakô. Arano asserts that the regional daimyô must have provided tacit consent, if not outright invitations, for these Chinese merchants to engage in such activities within their domains; the Chinese smugglers had similar relationships with local officials in China.[20] As noted above, much of what was described as "wakô" activity was simply trade - illicit or otherwise - and not true piracy, in the sense of violent raids on coastal towns or on other ships. Chinese communities in Kyushu flourished in the 16th century, many of them located in castle towns, and directly encouraged and supported by the local daimyô. Chinese communities brought Chinese trade, i.e. income, as well as skilled craftsmen and other talented workers, thus making the idea of supporting a local Chinese community quite attractive for daimyô.[27]

Most, if not all, residents of these Chinese communities in Kyushu traveled to Japan illegally (travel to Japan was, itself, after all, illegal under Ming law), though many also came against their will, either as prisoners of the wakô, or of the samurai forces of Hideyoshi, who brought back many prisoners of war to Japan when he invaded Korea in the 1590s.[27] Many captured by the wakô were sold as slaves; some 200-300 Chinese, for example, are known to have been kept as slaves in Takasu, Satsuma province, in the mid-16th century. Chinese and Japanese pirates captured Chinese and Japanese alike, selling them as slaves to willing buyers in the opposite land.[28]

Decline of the Wakô

The later years of the reign of the Ming Jiajing Emperor (1521-1567) saw a peak in wakô activity, which subsided when, in 1567, the Ming Court lifted the bans on Chinese trade and interaction in Southeast Asia[29], thus allowing many so-called "wakô" to become legitimate traders and seafarers in the eyes of the Chinese authorities. Many smugglers still engaging in activities deemed illicit, such as trade with Japan, moved their bases at this time to Taiwan or the Philippines.[29] Wakô attacks on Korean vessels and territory subsided at the same time, due in large part to these factors; agreements between the Korean court and the Sô clan of Tsushima in 1557 and 1567 allowing the Sô to send thirty ships a year to engage in legal trade may have contributed to the decline in wakô activity as well.[30]

Toyotomi Hideyoshi helped further weaken the wakô with a 1588 edict banning piracy. Hideyoshi established a definition of "Japanese waters," and declared that force could not be used to settle disputes within those boundaries; further, this edict severely weakened the ability of provincial daimyô to support, benefit from, or otherwise directly associate with pirates, i.e. the wakô.[31]

Though the actual wakô were somewhat weakened by these and other steps taken by Hideyoshi, his invasions of Korea in the 1590s were viewed by China and Korea as part and parcel of the wakô phenomenon. His samurai forces, who raided, plundered, and pillaged, destroying and stealing much financial and cultural property, and kidnapping many craftsmen (especially potters) and others, were seen as no different from the wakô pirates.[32] This is somewhat ironic, as, according to some sources, Hideyoshi's goal in invading Korea was to press China for access to the so-called "tally trade" (kangô bôeki)[33], the very same formal trade relations which were cut off by the Ming court in response to Japanese refusal or inability to curb wakô raids.

In the early years of the Tokugawa shogunate, various efforts were made to gain access to formal Chinese trade, thus curbing the wakô phenomenon by diminishing the situations under which traders between China and Japan would be considered "pirates."

A letter sent by the shogunate to the Governor-General of Fujian in 1610 was among these efforts. The letter was drafted by Hayashi Razan and Nagasaki bugyô Hasegawa Sahyôe, and passed through the hands of Honda Masazumi and the Chinese merchant Zhou Xingru, who had in fact come to Japan in order to complain about the pirates. The letter was aimed chiefly at seeking normalized relations with China, and access to the official tally trade. It offers that Nagasaki will be formally opened as a port for traders from Fujian, and offers that a formal Japanese mission will be sent to China once access to formal trade is obtained. The letter also requests that wood, water, or other supplies be provided to Japanese sailors who find their way to Chinese shores in an emergency (shipwreck, drifting off-course, etc.), but explains the red seal ships system, and grants permission to the Chinese authorities to punish as a pirate anyone not carrying a vermillion-sealed license.[34] By 1620, however, the shogunate gave up on trying to restore relations with China.[35]

The imposition of maritime restrictions in the 1630s dealt a major blow to the wakô. All but three ports[36] were closed to foreign trade, and Japanese were forbidden from leaving the country or returning. Wakô activity still continued among Japanese, and others, based overseas, who traded (or raided) in China, Korea, and Southeast Asia, as well as among, presumably, some small number of smugglers who continued to engage in illegal operations along the Kyushu coast. Following the fall of the Ming dynasty in 1644, Ming loyalists continued to fight against the Manchu conquest for forty years; these loyalists, and others associated with them, may have been at times referred to as wakô in Qing documents.

It was only with the turn of the 18th century that the wakô phenomenon really petered out and came to an end. The Tokugawa shogunate solidified its control over Japan - including, to the extent it ever would, over the Kyushu daimyô who allowed or encouraged wakô activities in earlier times. Meanwhile, greater European presence and activity in the region (though not in Japan itself) brought a degree of stability.[31]


The issue of terminology is thus quite central to discussion of the wakô. The question arises as to why Chinese and Korean authorities, and the official records they produced, employed the term when, arguably, in many cases, that being described was either not Japanese, or not a band of pirates. Yet, as with many historical terms, the issue is complicated by the fact that officials, historically, had no intention, and made no conscious effort, to employ the term in a consistent manner. Rather, the term was often used, or avoided, for explicit political purposes - such as to attack or protect certain parties or interests; its usage also varied from individual to individual, from case to case, and from time to time.[8]

Nevertheless, in aggregate, an image emerges of Chinese and Korean use of the term wakô, or "Japanese pirates," to refer not only to Japanese raiders, brigands and the like, but also to raiders and brigands of a number of other ethnicities (mainly Chinese), to smugglers and traders less involved in violent or predatory activity, and to the formal samurai invasion forces of Toyotomi Hideyoshi. The 17th century Official History of the Ming (Ming shi), among other primary sources, links the wakô closely to Japanese foreign relations policy, implying if not stating outright that the wakô were agents of the central Japanese authorities, and conflating pirate raids and the like with Hideyoshi's formal invasion forces. These sources also make little or no distinction between those who committed violent acts of piracy and coastal raids, and those who simply engaged in maritime trade in violation of Ming law. Though other prominent sources such as the Ming shi-lu (The Veritable Records of the Ming) do not misrepresent the subject in this way, the conflations and distortions seen in the Ming shi and elsewhere are reflected widely in modern scholarship, and in modern-day textbooks as well.

Some contemporary sources, however, were much more clear on the actual identities and activities of the so-called "Japanese pirates." A 1596 text by Vice Commissioner for Military Defense Cai Fengshi describes the "Wo" as not being all barbarians, i.e. Japanese, but as actually being led by spurious barbarians, i.e. Chinese, and goes on to recommend that the concern should be not with the "barbarians," but with the "spurious barbarians."[37] Similarly, Xie Jie (d. 1604), wrote that "the pirates and the traders were the same people. When trade flourished, the pirates became traders; when trade was banned, the traders became pirates."[38] He goes on to cite Chinese bans on trade and other restrictions and policies as the chief causes of wakô activity, suggesting that if trade were allowed to take place legally, illegal activity would diminish.[39]


  • Arano Yasunori. "The Formation of a Japanocentric World Order." International Journal of Asian Studies 2:2 (2005). pp185-216.
  • So Kwan-wai. Japanese Piracy in Ming China during the 16th Century. Michigan State University Press, 1975.
  1. 1.0 1.1 Arano. p186.
  2. 2.0 2.1 Watanabe Miki, “Shifting Representations of Ryukyuans between Early Modern China and Japan,” in Caroli (ed.), Imagined Okinawa: Challenge from Time and Space, Ca' Foscari University in Venice (2015), 197.
  3. 3.0 3.1 Gregory Smits, Maritime Ryukyu, University of Hawaii Press (2019), 39-40.
  4. Smits, Maritime Ryukyu, 121.; Peter Shapinsky, Lords of the Sea: Pirates, Violence, and Commerce in Late Medieval Japan, University of Michigan Press (2005).
  5. 5.0 5.1 5.2 Smits, Maritime Ryukyu, 41.
  6. Smits, Maritime Ryukyu, 70.
  7. Smits, Maritime Ryukyu, 37.
  8. 8.0 8.1 So. p209.
  9. So. p3.
  10. So. p4.
  11. Smits, Maritime Ryukyu, 46.
  12. Robert Hellyer, Defining Engagement, Harvard University Press (2009), 31.
  13. Smits, Maritime Ryukyu, 46-48.
  14. 14.0 14.1 So. p204.
  15. Ray Huang, 1587: A Year of No Significance, Yale University Press (1981), 163.
  16. 16.0 16.1 So. pp15-16.
  17. So. p5.
  18. The Ôuchi sent the last Japanese tribute mission to China in 1549, but as tribute missions only occurred once in a number of years, and only consisted of a small number of ships (and thus, brought limited revenues), the clan desired a greater volume of trade.
  19. So. pp16-17. While the Ôuchi are mentioned in some contemporary Japanese sources, So Kwan-wai notes that most contemporary Chinese sources emphasize the Chinese involvement in organizing and leading wakô bands, and do not mention the Ôuchi at all.
  20. 20.0 20.1 Arano. p188.
  21. So. p205.
  22. 22.0 22.1 So. p6.
  23. Huang, 170-171.
  24. Huang, 169.
  25. So. pp6-7.
  26. So. p22.
  27. 27.0 27.1 Arano. p194.
  28. Arano, 195.
  29. 29.0 29.1 Arano. p189.
  30. Marius Jansen, China in the Tokugawa World, Harvard University Press (1992), 6.
  31. 31.0 31.1 Arano. p190.
  32. Arano. p197.
  33. Arano. pp206-207. Arano asserts that Hideyoshi believed that by demonstrating Japan's military supremacy, Japan could thus rightfully claim "civilized" non-barbarian status within the Sinocentric world order, and a right to access to trade.
  34. Arano. p209.
  35. Arano. p210.
  36. Plus the more land-based "port" of access in Matsumae han for interactions and trade with Ainu.
  37. Ts'ai Feng-shih. An Illustrated Discourse on the Maritime Defenses of Wen-chou and Ch'u-chou (Wen-Ch'u hai-fang t'u-lüeh). Cited in translation in So. p212.
  38. So. p214.
  39. So. pp215-216.