The Journal of Korean Art and Archaeology

A tanged arrowhead with two wings excavated in Sindang-ri, Gyeongju (Umehara material 3401) is of the same type as a bronze arrowhead excavated from a site in Shangcunling, Shanxian from the early Spring and Autumn Period (Chunqiu Period).20 It is presumed that the bronze arrowheads with two-stepped tang (Umehara material 1569-1/1872/2248/3401) discovered at this site were all from the Spring and Autumn Period of China and are thought to have been introduced to Korea during the Liaoning-type bronze dagger culture period. Relatively few bronze tools have been found. There are two types of bronze axes: fan-shaped bronze axes and rectangular-shaped bronze axes. Fan-shaped axes were found at sites including Misong-ri, Uiju. Molds for bronze axes have been found in Yeongheung-eup and Songguk-ri, Buyeo. A bronze axe has been found at the Toseong-ri site, Pyeonganbuk-do Province, but it is uncertain whether it is fan-shaped. Fan-shaped axes are small—most are about five centimeters long. However, one of those found in Yeongheung-eup, Geumya is longer than 10 centimeters and bears an incised triangular pattern on the head. In the Liaoning Province, more were found in the Liaotung region than the Liaoxi region. Fan-shaped bronze axes have been discovered in stone cists of early Korean-type bronze dagger culture. Only one rectangular bronze axe was found in Gosan-ri, Jaeryeong, together with a Liaoning-type bronze dagger that has a T-shaped dagger handle. Two bronze chisels were found on the Korean Peninsula. One was made by processing a tang fragment of a Liaoning-type bronze dagger and was found in a stone cist in Songguk-ri, Buyeo. The other was found at dwelling site no. 8, Geumtan-ri, Pyeongyang. The one from Geumtan-ri21 is thin and flat without a socket. Bronze chisels with sockets like those found in the Liaoning region have not yet been discovered in Korea. One bronze hand-knife was discovered in Korea. It was excavated from the Yongheung-ri site in Gaecheon. It has three protuberances at the handle. In the Liaoning region, hand-knives of this type have been discovered mostly at sites related to the Upper Layer Culture of the Xiajiadian site. It was found together with a Liaoning-type bronze dagger, a jade pendant, and a stone axe. The body of the bronze dagger narrows towards the cutting edge, is short in length, and has an edge on the spine. Personal ornaments made of comma-shaped jades, tubular jades and round jades have been found. Comma-shaped jades or gogok have a hole at one end and most have been found in burials such as stone cists and dolmens. Most of the comma-shaped jades discovered from Liaoning-type bronze dagger culture sites are made of amazonite and can be divided into two types: those in the shape of a semi-circle with a large head and a small, straight tail slightly trimmed down toward the edges and those of smaller sizes and varied shapes. Jade pieces of the first type were found in a stone cist in Songguk-ri and in dolmens in Usan-ri, Seungju. Jades of the second type were discovered in dolmens in Usan-ri, Dolmen no. 17 at Hwangseok-ri, Jewon (formerly Jecheon); Stone cist at district IV-1, 3 in Sinchon-ri, Changwon; and a shell mound in Gonam-ri, Anmyeon-do Island. Those discovered in Gonam-ri, Anmyeon-do Island are reported to be jadeite. Comma-shaped jades excavated from Yongheung-ri, Gaecheon are semicircular. In Liaoning Province, jades of this shape were discovered at the Chengchiawatzu burial, Shenyang22 and were used for necklaces together with tubular jade beads. We cannot know exactly what types of personal ornaments such comma-shaped jades were used for. However, it is thought that they were used for earrings because larger ones have generally been discovered in pairs around the area where the ears of the buried must have been at the time of burial. Such examples were found in stone cist no. 4 at Daepyeong-ri, Bukchang and in another stone cist in Chopo-ri, Hampyeong (a Korean-type bronze dagger culture site).23 It is also possible that comma-shaped jades were used for long necklaces threaded with tubular jade beads. ( Plate 2 ) Liaoning Bronze Dagger Culture 1 the Upper Layer Culture of the Xiajiadian site 2 Liaotung Liaoning-type Bronze Dagger Culture 3 Xituanshan Culture 4 Korean Liaoning-type Bronze Dagger Culture A tubular jade bead or gwanok is a long, pipe-shaped jade bead commonly strung on a necklace. In many cases, tubular jade beads came in a set with comma-shaped beads. Most tubular jade beads are made of Egyptian jasper, but some are made of tuff or clay. They can be classified as either large or small. The large ones are longer than three centimeters and wider than one centimeter. Such jade beads were found in a stone cist in Songguk-ri, Honghyeon-ri, Baekcheon, Daepyeong-ri, Bukchang. Small ones were discovered in a dwelling pit, a jar coffin in Songguk-ri, and a burial (presumably dolmen) in Mugye-ri in Gimhae, a dolmen from Usan-ri in Seungju including other dolmens from Bonggye-dong in Yeocheon, Cheokryang-dong in Yeocheon, and Pyeongyeo-dong in Yeocheon. Another type of jade bead is smaller than one centimeter in diameter and is called so-ok. Most beads of this type are round and they were found from the sites in not only Usan-ri in Seungju but also Bonggye-dong, Cheokryang-dong, and Pyeongyeo-dong in Yeochen including Orim-dong in Yeosu. Most of these tiny jade beads are found in Southern-type dolmens in Jeollanam-do Province. Tubular jade beads and these tiny round jade beads are presumed to have been used for long necklaces. No pottery has been discovered together with Liaoning-type bronze daggers from burial sites in Korea, unlike burial sites in Liaoning Province. This may be due to differences in burial customs between the two regions. Misong-ri type pottery was discovered along with a fan-shaped bronze axe24 at the Misong-ri cave site in Uiju, Pyeonganbuk-do Province. Misong-ri type pottery was also discovered at a Liaoning-type bronze dagger culture site in Jilin Province, China, suggesting that Liaoning-type bronze dagger culture was related to Misong-ri type pottery. Top-shaped pottery, small jars with handles and perforated rims, and Songguk-ri type pottery are thought to have been closely related, given the form of stone arrowheads excavated together with pottery from stone cists in Pyeongannam-do and Hwanghae-do provinces, top-shaped pottery and small jars with handles and perforated rims from the site in Yeongheung-eup, Hamgyeongnam-do Province, and Songguk-ri type pottery from a dwelling sites in Songguk-ri south of the central part of the Korean Peninsula. In summary, early bronze culture in Korea basically originated from Liaoning-type bronze dagger culture in today’s Liaoning Province, China. It can be said that Korean bronze culture comprises a part of Liaoning-type bronze dagger culture. Liaoning-type bronze dagger culture is divided into four sub-cultures: Liaoxi (the Upper Layer Culture of the Xiajiadian site), Liaotung (Liaotung Liaoning-type bronze dagger culture), Jicheng (Xituanshan Culture), and Korean Peninsula (Korean Liaoning-type bronze dagger culture). Thus, Korea is one of the four largest regions of Liaoning-type bronze dagger culture. The discoveries of more than 78 Liaoning-type bronze daggers, T-shaped dagger handles, and Liaoning-type bronze axes that are indicative artifacts of Liaoning-type bronze dagger culture at some 63 sites in Korea prove that the Korean Peninsula was more important than the Liaotung region, not to mention Jicheng. Of the four regions, Korea was the second most important after the Upper Layer Culture of the Xiajiadian site. Until recently, the Korean Peninsula had generally been considered a peripheral region of Liaoning-type bronze dagger culture but this view should be revised, separate from the discussion of the origin of the culture. Liaoning-type bronze dagger culture in Korea had the same features as Liaoning-type bronze dagger culture in other regions. This may simply be due to the geographical location. As it is a peninsula, new cultural influences were continually being transmitted into Korea from the north. The type of spearhead of the bronze mold discovered at the site in Yeongheung-eup was also discovered in Jicheng, China. The hand-knives with serrate handles like the one found at the Yongheung-ri site were also found in Liaoxi and are from the Upper Layer Culture of the Xiajiadian site. The fan-shaped axe is a characteristic of Liaotung Province. Liaoning-type bronze dagger culture represented by the Chengchiawatzu site, Shenyang is similar to the early Korean-type Liaoning bronze dagger culture. The most common types of tombs of each region, that is, box-shaped stone coffin, stone-lined tombs, stone cists, and pit tombs are all found in Korea. However, dolmens are a characteristic tomb type that is only found in Korean Liaoning-type bronze dagger culture. Also, sites of stone mounds on hills have been discovered only in Korean Liaoning-type bronze dagger culture thus far. Another difference is that no indication of horse-riding or chariot fixtures has yet been found in Korea. Broken and whole bronze daggers, processed bronze chisels, and bronze axes were buried as votive objects in Korea. These patterns represent one of the features of Korean Liaoning-type bronze dagger culture, which cannot be found in neighboring regions. Bronze daggers with grooves on the tangs have not been found in the cultural sphere of other Liaoning-type bronze culture. In Korea, they are distributed only in the central western part and southern part of the country, and the distribution is similar to that of Songguk-ri type dwelling sites. Furthermore, most of them have been found in Southern-type dolmens, which are similar to Songguk-ri type culture in southern region. A shard of a tanged bronze dagger with grooves was also found in North Kyushu in Japan suggesting that this type of bronze dagger might have been introduced to Japan at the height of Songguk-ri type culture that is represented by Songguk-ri type dwelling sites, Southern-type dolmens, and rice farming. There has been a great deal of debate as to who was the major player of Liaoning-type bronze dagger culture. The Dongho, Sanyung, Dongi, Yemaek, Gojoseon, Yemaek Toungus, and other tribes have been mentioned as a possible candidate. Because the features of archaeological sites and artifacts (especially pottery) in each region show great differences, it is difficult from an archaeological point of view to determine the origin of Liaoning-type bronze dagger culture. Many hypotheses have been proposed about the beginning date of Liaoning-type bronze dagger culture in Korea. A Liaoning-type bronze dagger excavated at the site of Yejeon-dong, Cheongdo is 34.8 centimeters long and has a heavy wide blade and very short tip. Compared with other daggers, the protrusion is oriented toward the tip and the lower part of the blade displays a rather circular shape. These features are exactly the same as those of pieces from the early stage of Liaoning-type bronze dagger culture in Liaoning Province (bronze dagger with curved edge and tang Type B) and this Korean dagger is of an ancient style close to the bronze daggers of the Shiherhtaiyingtzu Assemblage and Shuangfang Assemblage. If there is any difference at all, the Korean dagger has no edge on the spine. However, given that the edge on the spine moved to the lower part of the cutting edge in the latter period bronze daggers, there is no reason at all to believe that Korean daggers are of the later period. As mentioned earlier, given the tanged arrowheads with two wings from the end of West Zhou period to Spring and Autumn Period excavated at Gyeongju nearby Cheongdo, I think that Korean Liaoning-type bronze dagger culture could have begun as early as the first half of the Spring and Autumn Period, which is the 8th century B.C. Stone arrowheads were found together with bronze daggers of atypical shapes that are similar to the bronze daggers discovered in Daea-ri and Seonam-ri. These stone arrowheads are of similar appearance as those of Okseok-ri and Gyoha-ri in Paju.25 The fact that these arrowheads date from 9th to the 7th century B.C. lends further credence to the argument. It is also difficult to determine exactly when the latest period of Liaoning-type bronze dagger culture ended in Korea. Nevertheless, the discovery of small polished stone daggers with grooves on the tangs and stone daggers with single-stepped handles at dolmens in Jeollanam-do Province, where Liaoning-type bronze daggers have been discovered in sites dating to the latter half of the late Jomon period to the early Yayoi period in Japan,26 suggests that Liaoning-type bronze dagger culture in Korea continued until around the 4th century B.C. A Liaoning-type bronze dagger excavated at the Gosan-ri site can be seen as a type of the latest period and, because Chinese-type bronze daggers excavated along with it can be seen as imitated ones, Liaoning-type bronze dagger culture may have actually continued into the 3rd century B.C. in Korea. III KOREAN-TYPE BRONZE DAGGER CULTURE If Liaoning-type bronze dagger culture shaped the early bronze culture of Korea, Korean-type bronze dagger culture represents the late bronze culture of Korea. It exhibits unique typology and styles especially in bronze daggers, spearheads, halberds, multi-knobbed mirrors with geometric designs (a bronze mirror with knobs on the back), and ritual objects. It is a derivation from the Liaoning-type bronze dagger culture, which originated in Liaoning Province in China. But Korean-type bronze dagger culture developed its distinctive styles as to be considered uniquely Korean by incorporating the bronze cultures of Siberia and Scythia in addition to those of northeastern China. It is believed that Korean-type dagger culture crossed to Japan and shaped the bronze culture of Japan. It ultimately declined with the introduction of the iron culture from China. 01 KOREAN-TYPE BRONZE DAGGER CULTURE AND LIAONING-TYPE BRONZE DAGGER CULTURE ( Plate 3 ) Korean style Bronze Dagger 1 Tip 2 Edge on Spine 3 Shallow arc 4 2nd protrusion 5 2nd indentation 6 1st indentation 7 1st protrusion 8 Spine 9 Base 10 Tang As many scholars have indicated, the origin of Korean-type dagger culture can be clearly traced back to Liaoning-type bronze dagger culture by comparing artifacts from the two cultures.27 Korean-type bronze daggers and typical Liaoning-type bronze daggers share some features in common. First, both are about the same size. The body and handle were separately made from different pieces and joined together later. Second, the lower section of cutting edges on both daggers has grooves that trace a shallow arc. Third, the dagger handles are T-shaped and the handles have pommel fittings. Bronze daggers with such features are only found in Korean-type bronze daggers and Liaoning-type bronze daggers. Bronze axes, chisels, multi-knobbed mirrors, and stone molds similar to corresponding objects discovered at the excavated sites related to Liaoning-type bronze dagger culture have been recovered from the archaeological sites related to Korean-type bronze dagger culture. The trumpet-shaped bronze implement and the shoulder-armor-shaped bronze implement and other features have been commonly discovered in sites from both cultures as well. We cannot, nonetheless, find evidence at any archaeological site that suggests that Korean-type bronze dagger culture indeed developed from Liaoning-type bronze dagger culture. What is more, the structures of stone cists of the two cultures show some differences. In the southern regions, a number of Liaoning-type bronze daggers but very few Korean-type bronze daggers have been discovered in Southern type dolmens. Not a single Liaoning-type bronze dagger has been discovered along with bronze mirrors with coarse lines on the back. There have also been no reports of later Liaoning-type bronze daggers being found along with Korean-type bronze daggers. If Liaoning-type bronze dagger culture and Korean-type bronze dagger culture had been correlated in their successive development, both the number and quality of bronze artifacts should have increased in the later period. It seems that Korean-type bronze dagger culture was more closely related to Liaoning-type bronze dagger culture in Liaoning Province in China, rather than Liaoning-type bronze dagger culture in the Korean Peninsula. It is believed that in the latter period of Liaoning-type bronze dagger culture in Korea, the culture in Liaozhong Province around Shenyang deviated from Liaoning-type bronze dagger culture and was then introduced to Korea. 02 CLASSIFICATION OF KOREAN-TYPE BRONZE DAGGER CULTURE Korean-type bronze dagger culture can be divided into three phases: Phase I, Phase II, and Phase III.28 Phase I is the period when Korean-type dagger culture came into being. Due to the spread of Liaoning–type bronze dagger culture, Korean-type bronze daggers, bronze mirrors with coarse linear design, shield-shaped bronze implements, split-bamboo-shaped bronze implements, and trumpet-shaped implements began to appear in Korea. The influence of Liaoning-type bronze dagger culture was dominant during the formative period of Korean-type bronze dagger culture, so much so that the shapes of bronze objects from Liaoning–type bronze dagger culture and those in Korea were almost the same. However, elements of the northern culture were also introduced to Korea and incorporated into Korean culture. Only burial sites have been discovered from the first phase. The stone cists represent this phase and they are characterized by the use of trimmed stones covered with stone layers on top. Given the structure of the tombs, it is unlikely that they are directly related to the burial system of Liaoning-type bronze dagger culture. The one type of bronze artifact that best represents the first phase is the Korean-type bronze dagger. The Korean-type bronze dagger has a central edge on the spine that runs only from the tip to the first indentation and a blade that widens toward its lower section, and it traces a gentle curve from the base of the hilt to the lower blade. Although the archetype of the Korean-type bronze dagger is the Liaoning-type bronze dagger, it can be considered unique to Korea because the shape originated in Korea and its distribution was not related to the northeast region of China. The shape is basically the same as the Liaoning-type bronze dagger. The greatest difference between the two, however, is that the body of the Korean-type dagger is straighter and sharper. Both the shallow arcs of blades and the indentations on spine are more clearly defined on the Korean-type bronze dagger. Dagger handles have not been discovered because they were made of wood, which obviously does not survive under most conditions. Most pommel fittings are made of stone but there are also many made of iron. The pommel fittings have various shapes, such as a cocoon, cross, and cross with protrusions. Of these, the type in the shape of a cross with protrusions is unique to Korea.29 It is believed that the archetype of the cross with protrusions was a cocoon-shaped pommel fitting of Liaoning-type bronze dagger culture. Except for bronze daggers, no other bronze weapons have been discovered from Phase I. Bronze objects, except for weaponry from this period, include ritual implements such as shield-shaped and split-bamboo-shaped bronze implements, tools such as fan-shaped axes, and chisels. The pottery from this phase was pottery with clay stripes and long-necked black-burnished jars. Arrowheads without tangs and ornamental jade and tiny jade beads from this period have been discovered as well. Very few bronze mirrors with coarse linear designs have been found together with Korean-type bronze daggers. They were discovered from Goejeong-dong site in Daejeon, Dongseo-ri site in Yesan, and Namseong-ri in Asan. In addition, there is another example of a ritual bronze implement attributed to Jeollabuk-do Province. Early bronze mirrors with coarse linear designs produced in Liaoning Province have double-lined Z-shaped designs. In Korea, there are bronze mirrors with coarse linear designs and with double-lined designs, which have been attributed to Pyeongyang, Seongcheon, and Chungcheongnam-do Province.30 It is uncertain whether they were produced during the same period as Korean bronze daggers. Based on what has been found in Liaoning Province, they are thought to have been produced together with Liaoning-type bronze daggers. However, the fact that not a single mirror with coarse lines on the back had been discovered along with Liaoning-type bronze daggers suggests that those with double-lined designs were produced along with Korean-type bronze daggers. Besides mirrors of this type, bronze mirrors excavated from Yeonhwa-ri in Buyeo, Goejyeong-dong in Daejeon, and Dongseo-ri in Yesan generally belong to Phase I. Of heterotypic bronze vessels, trumpet-shaped and shoulder-armor-shaped objects had been excavated at the Chengchiawatzu Site, Shenyang.31 It is thought that these bronze objects were used to decorate horse equipment, purses for hand-knives, and gimlets in Liaoning Province. Similar objects found in Korea, however, are conjectured to be ritual objects because the designs of a deer, hand, and falcon engraved on these heterotypic bronze implements seem closely related with Siberian shamanism. It is also possible that one object can be used for different purposes from region to region in different contexts32 and the fact that Bronze Age horse equipment has never been discovered in Korea at the time further suggests the ritualistic purpose of these objects. That these bronze implements appeared as ritual objects in Korean-type bronze dagger culture is thought to be related to the social conditions in Korea of those who accommodated the culture. Fan-shaped bronze axes were discovered from Namseong-ri site in Asan, Yeoui-dong site in Jeonju, and Jeongbong-ri site in Singye. They were small and may be considered adzes rather than axes. They can also be considered artifacts typical of Phase I. Of pottery from this period, long-necked black-burnished jars have been discovered at Liaoning-type bronze dagger culture sites in Liaoning, so they are naturally thought to have been related. Pottery with clay stripes, however, has been found south of the central region on the Korean Peninsula, which is too closely related with Korean-type bronze dagger culture to be separate. Stone arrowheads without tangs and with a hexagonal cross section began to appear in this period. Some arrowheads of this type have been found along with bronze mirrors with coarse lines on the back, archaic Korean-type bronze daggers, and archaic ritual implements. These stone arrowheads are as closely related to Korean-type bronze dagger culture along with pottery with clay stripes. On the other hand, except for examples such as multi-knobbed mirrors with coarse linear designs attributed to Seongcheon and Pyeongyang, a multi-knobbed mirror from Soa-ri in Yeonan, and a mold for mirrors from Maengsan, no site has yielded a multi-knobbed mirror in the northern province during Phase I. As discussed above, the sites such as Seongcheon and Pyeongyang can be related to Liaoning-type bronze dagger culture. Korean-type bronze dagger culture developed during Phase II. Breaking from the influence of Liaoning-type bronze dagger culture and with new northern cultural elements introduced to Korea, bronze ritual implements that came in sets of bronze bells began to appear. Due to rapid changes including advancement in casting technology, the design of bronze mirrors became more refined. The coarse linear design was replaced with fine linear designs and bronze bells began to be used as ritual implements. The change in the second phase occurred rather suddenly and this rapid change cannot be attributed merely to technological development or changes in designs. Such changes could be possible only through contact with a new bronze culture. The only conclusion we can make is that a new northern culture was introduced to Korea. The impact of this new bronze culture can be found to some degree in split-bamboo-shaped bronze implements and the designs of shield-shaped bronze implements. Some designs with northern cultural elements continued to appear on ritual objects from Phase II. In particular, the sunlight design that symbolizes the sun (that is believed to have been a northern cultural element) was used on round bronze implements with designs, bronze pole-tops, and bronze mirrors with fine linear designs since Phase II.33 Given the designs on bronze ritual objects such as deer, hand, falcons, sun cross, and hunting scene, these objects are presumed to have been shamanistic tools used for rituals related to agriculture, hunting, and shamanism. The owners and users of these kinds of ritual objects are believed to have been in charge of ritual services and political affairs (i.e. they had political power and supervised religious rituals). At this time, following the spread of northern culture, the Culture of Central China began to influence Korea and early type bronze spearheads, bronze halberds, and bronze engravers began to appear. All of these bronze implements were so unique that they could be considered Korean type, suggesting that the influence of the Culture of Central China was not very strong. In North Korea, cast iron axes, which are believed to have been introduced from China, appear in some places but were not common. The burial practices of Phase II were almost the same as the one in Phase I. However, as seen from the example in Daegok-ri, Hwasun, it is believed that wood coffin tombs with stone layers were built in the southern region.34 Although artifacts related to Korean-type bronze daggers have been found at some dolmen burial sites, we do not yet have enough examples. During Phase II, breaking away from the elements of the Liaoning-type bronze dagger culture, a new bronze dagger culture unique to Korea began to develop. The body and shoulder of the Korean-type bronze dagger became straighter than that of the Liaoning-type bronze dagger. Bronze daggers identified as Type II were produced along with Type I (Dr. Yun Mu-byeong’s classification system). Bronze spearheads began to appear in this period. Most of them are short, only around 20 centimeters long. Some spearheads with no ears have holes and raised bands around the bottom of the sockets. Bronze spearheads excavated in Jeongbong-ri in Singye, Dunpo-ri in Asan, Tanbang-dong in Daejeon, and Gubong-ri in Buyeo are representative of this period (Type I spearheads by Dr. Yun Mu-byeong’s classification). Although they are believed to have been designed after Chinese spearheads, certain details such as the shape of the tip, spine, and the notch are different. Given that Liaoning-type bronze spearheads had been used in the preceding phase (Phase I), there is a need for further investigation and discussion on the development of shapes. During the period from the end of Phase II through Phase III, the spearheads were elongated. Some spearheads discovered at archaeological sites in Chopo-ri, Hampyeong and Ehwa-dong, Hamheung are longer than 20 centimeters. Bronze halberds also began to appear in Phase II and became a representative artifact of this period. It can also be said that the Korean halberd was based on the Liaoxi bronze halberd but, as is the case with bronze spearheads, each has its own very unique features. The bronze halberd is so unique that it can rightly be called the Korean-type bronze halberd. It shows one aspect of how Korean-type bronze dagger culture developed. There are two types of Korean halberds: one with an angle on the back ridge and one without an angle on the vein. It is, however, unclear which one was developed earlier.35 Bronze artifacts that have been excavated along with weaponry include bronze mirrors with fine linear geometric designs, bronze bells, bronze axes with shoulder, bronze engravers, and bronze chisels. Bronze mirrors with fine linear geometric designs began to appear during Phase II. The designs are exquisite and the geometric compositions are outstanding. The surfaces of many mirrors are divided into three decorated sections and most of the mirrors have two knobs. Compared with bronze mirrors with coarse lines on the back from the first phase, mirrors from Phase II are larger and the knobs are standardized. The engraved designs are so minute that it is difficult to determine how the designs were made. The outer section is bordered with sun-rayed design, which suggests that the mirror was used as a ritual object symbolizing the sun.36 Entering the latter half of Phase II, the design composition became simplified and smaller mirrors began to appear.37 Bronze bells by this time came into existence as can be seen in the cases of pole-top, two-jingle, eight-jingle, and hybrids of two-jingle types. Bronze bells excavated in Nonsan and Hwasun are exquisitely made and mirrors excavated with them were among the most exquisite of any ever found indicating that the production technique of bronze objects reached its apex in Phase II. Bronze axes produced in this period are based on the fan-shaped axes of the first phase. The area from the socket to the body is a curved shoulder. The body is an elongated rectangle. In fact, this kind of piece is more like an adze than an axe. Large bronze axes were rectangular or shaped like a bronze axe with the blades on both sides, and they were all discovered at Phase II archaeological sites. The artifacts found in Sunan suggest that straight rectangular axes were being produced continually into Phase III.38 A bronze engraver is a carving tool. In China, bronze engravers have been discovered in the former territory of Chu from the Warring States Period (403-221 B.C.). Bronze engravers excavated in Korea are similar to those of China in shape but the relationship between the two is not clearly understood. However, in China, engravers with raised lines on the back ridge dating as far back as the Spring and Autumn Period have been discovered but most such engravers that have been discovered are from the Warring States Period. Therefore, Korean bronze engravers are thought to be closely related to engravers from the Warring States Period. Many engravers from that period have are triangular or shaped like convex-lenses in section.39 Chisels dating to Phase II have been discovered in a greater quantity than those of Phase I. There are two types of chisels. One has a thick and long socket and a raised surface between the socket and body. The other has a socket linked to the body without a raised part in between. Many chisels are found as part of a set that includes a bronze axe and an engraver. Gimlets that are triangular in section also seem to be tools. Gimlets were discovered at the Izwestov Site and in a mold excavated in Yeongam, Korea.40 Molds for gimlets had been also found in Liaoning Province, confirming that they were influenced by Liaoning-type bronze dagger culture. Accordingly, gimlets are believed to have been produced in Phase I. Around the end of Phase II (end of the 3rd century B.C.~early 2nd century B.C.), the iron culture of the Warring States, especially that of the Yen, was introduced to Korea. Cast iron axes and iron chisels began to be produced, marking Korea’s entry into the Early Iron Age. Iron axes similar in form to the cast iron axes of the Warring States Period have been found in Yongyeon-dong, Sejuk-ri in the northern region and Songsan-ri, Seoksan-ri, and Ihwa-dong where Korean-type bronze daggers had been excavated.41 Such iron artifacts were also found from Soso-ri, Dangjin, and Hapsong-ri, Buyeo in the southern part of the Korean-type bronze dagger culture. In Phase III, Korean-type bronze dagger culture declined. Bronze weaponry was made less and less functional and instead became increasingly ceremonial. The tang became shorter, the tip part became longer, or decoration was incised on the groove. This was entirely due to the spread of iron culture that was introduced at the end of Phase II. The introduction of more utilitarian iron implements changed the concept of bronze tools. Afterwards, due to influence from the Chinese Han Dynasty, Chinese Han-style mirrors replaced the bronze mirrors with fine linear designs. The shift from bronze to iron accelerated and, when production of iron objects began in full-scale, bronze production barely survived. It is believed that Korean-type bronze dagger culture finally came to an end by the Late Han period. The burial practices in Phase III are assumed to have been dominated by wood coffin tombs, although we cannot confirm this.42 Although bronze daggers identified as Type I and II according to Dr. Yun Mu-byeong’s classification system were both produced in the third phase, more of the latter were made. Towards the end of the period, daggers with veins split up to the tang and multiple grooves were produced. Most pommel fittings were made of bronze. Spearheads became longer compared to those from the second phase and semi-circular ears were attached to the sides of the socket. Besides, spearheads with more than two grooves incised on the cutting edges or decorated with designs on the surface of the socket began to appear in this phase. Although there were short spearheads, they all had ring-shaped ears or incision of multiple grooves even if they had rivet holes. As for bronze halberds, those from the second phase and third phase show little difference. However, those with grooves decorated with coarse lines were excavated from Sincheon-dong in Daegu, Gujeong-dong and Ipsil-ri in Gyeongju and those with narrower and shorter butts and longer tips can be regarded as characteristic of the third phase. Besides horse equipment, Chinese Han-type artifacts (mirrors and small coins), small bells, and bronze rings have been excavated along with weaponry. Furthermore, axes, spearheads, and daggers made of iron from this phase have been discovered. Different types of horse equipment had been produced in the northwestern and southern parts of the Korean Peninsula. In the North Korean region, various types including chariot axle ornaments, 乙-shaped implements, upright cylindrical ornaments, carriage ornaments, axle fittings, and parasol caps had been excavated while in South Korea, only a few carriage ornaments and parasol caps have been discovered.43 This can be explained by the fact that the western part of North Korea is geographically close to China, allowing the northern part of the Korean Peninsula to accommodate Chinese culture more easily. Along with establishment of Nangnang (Lolang) by Chinese Han Dynasty, the iron culture of Han China began to spread rapidly southward, replacing Korean-type bronze dagger culture. As can be seen from the examples of sites at Joyang-dong in Gyeongju and Yangdong-ri in Gimhae, Korean-type bronze dagger culture remained in part until the early Proto-Three Kingdoms Period. However, the fact that the Proto-Three Kingdoms culture succeeded Korean-type bronze dagger culture has only symbolic meaning because of such archaeological patterns. IV PRODUCTION TECHNOLOGY OF KOREAN BRONZE OBJECTS Numerous pieces of bronze objects including daggers and axes had been produced throughout the Bronze Age in Korea. Determining how bronze pieces were made is very difficult because the processes involved were complex. It involved mining, smelting, making of molds, casting, and repairing and each of these processes was done by specialists. The most difficult and important tasks of all were the smelting of ores to be used to make metal alloys and casting. The technologies of mining raw ore and smelting are not yet known. Only the composition of bronze implements and some of the production and repair technologies are understood. This chapter addresses the production technology of Korean bronze objects based on the findings of composition analysis and casting of bronze as well as the traces of repairs on bronze artifacts. 01 COMPOSITION OF KOREAN BRONZE OBJECTS Mining of ores is the first step of production of bronze objects. Copper and tin are the essential alloys of bronze. Lead and zinc are also added to improve the quality of the finished bronze products. It is, therefore, believed that metal workers looked for smithsonite, cerusite, and galena in addition to malachite and cassiterite.44 The Augmented Survey of the Geography of Korea (Dongguk yeoji seungnam), compiled in 1481, states that of five metals (gold, silver, copper, iron, and lead), copper was produced in the largest volume45 indicating that it was mined at many locations throughout the country. Several ancient lead, tin, and zinc mines have been found suggesting that these metals were not in short supply at the time. However, no mines dating to the prehistoric era have yet been found leaving us know nothing of the mining technology of the Korean Bronze Age. In order to make an alloy, it is believed that each of the raw ores was smelted and the metals were then mixed at an appropriate ratio. This is supported by the results of composition analysis of weaponry, such as Korean bronze daggers, which show a consistent ratio of each component. The properties of bronze ware vary according to the ratio of alloys. It tends to be most durable when the tin content is 19%. If the tin content is over 19%, it becomes brittle. The metal composition ratios of bronze artifacts discovered in Korea are shown in the table below.46 ( Table ) Chemical Composition of Bronze Implements Bronze ware Element (%) Total Cu Sn Pb Zn Fe Sb Nb As Bi Co Ni 01 Korean style dagger 78.2 17.12 4.32 0.05 99.69 02 Korean style dagger from Suncheon 73.14 19.77 6.39 99.3 03 Korean style dagger from Suncheon 70.3 14.84 14.22 99.36 04 Korean style dagger from Pyeongyang 78.09 14.3 8.39 100.78 05 Korean style dagger from Pyeongyang 75.94 15.08 9.45 100.47 06 Chisel from Songsan-ri, Bongsan 40.55 18.3 7.5 24.5 1.05 91.9 07 Small bells from Ipsil-ri, Gyeongju 59.18 29.99 7.72 1.09 0.57 0.9 0.55 100.0 08 bamboo-shaped implement from Goehyeong-dong, Daejeon 56.2 19.9 7.8 0.006 0.2 84.106 09 Personal ornament from Chodo, Najin 53.93 22.3 5.11 13.7 1.29 100.0 10 Conical ornament from Eoeun-dong, Yeongcheon 77.56 15.71 6.08 0.11 0.09 0.25 0.38 100.18 11 Mirror from Namseong-ri, Asan 39.5 27.3 11.4 0.05 0.7 78.95 12 Mirror with fine linear design from Songsan-ri, Bongsan 42.19 26.7 5.56 7.36 1.05 82.86 13 Dagger from Yongje-ri, Iksan 75.3 17.1 6.8 0.001 0.001 0.01 99.212 14 Personal ornament from Chodo, Najin 83.4 7.2 8.0 0.05 0.12 0.85 0.3 0.08 100.0 15 Ingot from Chodo, Najin 67.23 25.0 7.5 0.05 0.14 0.24 0.05 0.002 100.212 16 Disk-shaped vessel from Toseong-ri, Bukcheong 57.7 25.0 7.0 1.0 2.0 2.0 5.0 0.3 100.0 17 Dagger from Seokdang-ri, Sincheon 83.33 10.0 6.4 0.09 0.01 0.05 0.05 0.065 99.995 18 Dagger from Gangan-ri, Onseong 96.96 0.25 2.0 0.009 0.55 0.09 0.02 0.05 0.07 99.999 19 Dagger from Sincheon 83.6 12.0 4.0 0.13 0.15 0.08 0.04 100.0 20 Dagger from Cheonju-ri, Hwangju 84.7 8.0 5.2 0.1 0.35 1.4 0.08 0.08 0.09 100.0 21 Dagger from Ilgok-ri, Baekcheon 88.88 11.0 0.025 0.04 0.045 0.01 100.0 22 Dagger from Hwangju 78.87 8.5 11.0 0.01 0.19 1.2 0.07 0.025 0.035 99.9 23 Dagger from Yeontan 92.64 4.0 3.1 0.11 0.01 0.02 0.06 0.06 100.0 24 Dagger from Sariwon-s 76.47 12.0 7.0 3.0 1.2 0.025 0.1 0.1 99.895 25 Dagger from Hamju 73.05 20.0 5.0 0.06 0.9 0.2 0.04 0.05 0.1 99.4 26 Dagger from Joyang-ri, Hamju 67.28 25.0 70 0.09 0.15 0.04 0.02 0.02 99.6 27 Dagger from Hasedong-ri, Bukcheong 67.02 25.0 7.0 0.04 0.45 0.3 0.05 0.05 0.09 100.0 28 Dagger from Nagyang no. 7 77.64 8.0 11.0 0.15 1.0 0.04 0.07 0.1 98.0 29 Dagger from Sincheon-gun 82.69 13.5 3.5 0.01 0.1 0.17 0.009 0.02 99.999 Composition analysis of bronze objects shows that the Korean-type bronze daggers are composed of 79.2% copper, 13.4% tin, and 6.8% lead on average. Compared with daggers, the content of copper in ritual objects, bells, and personal ornaments is much lower, only 59.65% while the content of tin is much higher at 22.12%. The lead content is about the same at 7.35%. The content of zinc varies from very little to as much as 24%. The ratio of the different metals was adjusted to increase the hardness, improve resistance to corrosion, and allow for easier casting as needed depending on the type of products being made. For example, bronze daggers have higher copper content and ritual objects like mirrors have higher tin content to reflect more light. Clearly, the Korean metal workers understood the principles of alloying metals. Compared with Chinese bronze objects, Korean bronze objects have lower copper content and higher tin and lead content. Noteworthy is Korean bronze with zinc. Zinc makes molten bronze more fluid giving the finished bronze product the same texture, hardness, and corrosion resistance as those with less tin.47 Korea had more abundant deposits of zinc than tin or lead, so bronze with zinc was naturally produced in Korea at an earlier stage. A bronze axe discovered in Songsan-ri has a very high zinc content of 24.5%. Personal ornaments discovered in Chodo and a bronze mirror with a fine linear design found in Songsan-ri also have high content of zinc. This suggests that the lineage of the Korean bronze objects is different from Chinese bronze objects that have almost no zinc. Without artificial manipulation, it is impossible to have a content of additive elements in bronze of more than 1.0%. It indicates that the method of alloying to improve the quality of bronze had already been developed by the prehistoric era.48 02 CRUCIBLES Extraction of metal from ore and meltdown of metal requires a crucible. In Korea, a crucible from the prehistoric era has not yet been found but a number of crucibles from the Three Kingdoms and later periods have been found. In other countries, earthen crucibles49 were in from the Bronze Age and continued to be used with little change until the early Iron Age. It is thought that the crucibles used in the Bronze Age on the Korean Peninsula were very similar to those of the Three Kingdoms and later periods. If we assume that the same type of crucible was used in the Bronze Age as in the Iron Age, the crucible was made of clay and had a capacity of about 50-360cc.50 Clay was mixed with sand and the wall and bottom of the crucible were much thicker than those of an ordinary pot. There were five types of crucibles in Korea.51 Type I is shaped like a bowl with a round bottom. It has a spout. Its mouth is relatively wide in comparison with its height. Type II has a bottom resembling a round cup. It also has a spout. Type III is conical with a pointed bottom and does not have a spout. Type IV is similar to Type III, except that the pointed bottom has a protrusion and it has a spout. Type V is like a cup with a flat bottom. These crucibles are small in size and similar in shape to those of ancient Western civilizations and their periphery such as Great Britain and Sweden. Compared with Korean crucibles, those from ancient China are shaped like a bucket or a so-called upside down helmet with a narrow flat bottom and the lower part of the body widening towards the upper part. Many Chinese crucibles are large compared to the Korean ones.52 03 MOLDS ( Plate 4 ) Two dagger molds fitted together from Chobu-ri, Yongin ( Plate 5 ) Two molds fitted together from Yeongam Bronze implements were cast in a mold. Numerous molds have been found all across the Korean Peninsula proving that bronze ware was locally made. Most molds are made of soapstone. Only one made of schist was reported to have been found at a Korean Bronze Age dwelling site at Songguk-ri.53 It is believed that molds made of clay were widely used but none have been found in Korea. A mold was produced by flattening the surface of the material until it was planar and then carving the desired shape into it. Most molds are believed to have been made of two pieces in the same shape that were fitted together before the molten liquid was poured into them. Most molds were made of soapstone because it was soft enough to carve the desired shapes into it; could be used repeatedly because it did not crack during casting; and gave the cast object a smooth surface. Almost no soapstone molds have been found in China, except in regions related to Liaoning-type bronze dagger culture such as Chifeng, Liaoyang, Luta, Tangshan in northeast China. Most Chinese molds were made of clay. In Japan, most molds were made of sandstone. A mold made of clay should have a mouth for pouring in the molten liquid and a hole that allowed gas to escape and it could be used only once. A mold made of sandstone was not durable enough to be used more than several times and often needed to be repaired for the next casting. A soapstone mold does not need a gas hole and is extremely resistant to heat. Therefore, the soapstone mold can be used over and over again.54 Nonetheless, no two bronze products that were made with exactly the same mold have been discovered in Korea. Therefore, we have no way of knowing for certain if a multiple number of bronze pieces were actually cast from the same mold. We can often see a line on the side of a two-piece mold, which was intended to make sure that the two pieces joined together precisely front and back and left and right. Several molds which clearly have a sprue were found including a mold for a bronze mirror with a coarse linear design found in Maengsan and a mold for a bronze dagger found in Jangcheon-ri. As can be seen from examples attributed to Yeongam, given the traces of heating in fire, molds for bronze daggers, axes, halberds, and spearheads have oval clay sprue towards the tang or socket, which were larger than the tang and socket so that molten liquid copper could be poured in easily.55 Bronze implements such as axes, spearheads, and bells that have sockets or a hollow part inside need an inner mold (core) made of clay or sand. The inner mold should not touch the outer mold in order not to cause any deformity during casting. Thus, when making a socket, the inner mold was made wider towards the top (toward the socket) and the lower part (the cutting edge) was made in a narrow ladder shape. The sprue was made narrower towards the bottom to prevent the inner mold from falling down as we can see in molds attributed to Yeongam. Also, on the one side of the sprue, two small grooves were made to help fix the inner cast in place. In order to fix the inner mold in place, a rivet was commonly used. This rivet linked and locked the inner and outer molds. Since the molten liquid could not be poured into this portion, a hole was made. Square holes on the body of a bell with clapper or mid-section of a two-jingle type bell are traces of such locking rivets. A hole on a bronze bell was intended to give the bell a certain sound but it was also meant to hold a locking rivet. The inner mold was made of clay or sand so that it could be scraped out completely after casting. However, as can be seen from a bamboo-hat-shaped bronze artifact excavated from Angye-ri in Gyeongju, only the socket part of the inner cast was scraped out and the hat part was left without being scraped out.56 It is likely that this was meant to maintain the proper weight while reducing the amount of raw material. The inner mold made of sand was found, as can be seen in the inside of the mid-section of a two-jingle type bell excavated in Hwasun. That the mid-section was unnecessarily made hollow suggests that the makers wanted to save raw material. If we look at the carved surfaces of molds, we can see that for some, both the front and back sides were used and for others, a multiple number of forms were carved on one side so that several products could be made simultaneously. As seen in the examples of fish hooks and buttons, two forms were carved on top and bottom pieces of molds, which were, then, aligned, so that two objects could be made at the same time. In addition to stone molds, clay molds are thought to have been widely used. Given that no stone molds for bronze mirrors with fine linear designs have been discovered and that the designs on mirrors were repaired as can be seen in bronze mirrors excavated in Yangyang,57 it is highly likely that the molds for the mirrors were made of clay. Among bronze objects, bronze bells such as a pole-top, two-jingle, or eight-jingle types all come in pairs but judging from the size and design, not a single pair was made from the same mold. That a pair of bronze bells, which were difficult to make, was produced separately indicates that they were made from a clay mold rather than a stone mold. Some bronze implements are thought to have been produced with a lost-wax (cire perdue) cast although a clay mold was used. Many ritual ornaments are believed to have been made with lost-wax cast because the surfaces of split-bamboo-shaped implements and shield–shaped implements used as ritual objects are curved and decorated with designs that are hard to engrave in a stone mold and there are no seams on semi-circular knobs with suspending rings. Designs were engraved directly on the mold or the original model. In some cases, it is believed that designs were stamped. As can be seen in the pair of pole-top style bells excavated in Jukdong-ri, Gyeongju, the decorative patterns are exactly the same. Even the deformed patterns show no variation.58 Carved and incised designs on ritual implements are presumed to have been stamped.59 Traces of stamp use for decorative designs can also be found in T-shaped dagger handles from Liaoning-type bronze dagger culture.60 04 REPAIR ( Plate 6 ) Trace of solder on a dagger handle from Sincheon ( Plate 7 ) Trace of solder on a bamboo-hat-shaped bronze artifact from Heukgyori If a mold was not properly made due to poor molding technique during the casting process or some kind of defect, the end product would also be defective. Among other things, the patterns might not show clearly: the product might have a hole in it because the molten bronze liquid failed to reach some part of the mold cavity or some part of the product might be deformed. The product would also be more likely to break when used. In such cases, the products were repaired. Engraving of patterns or soldering were two means of repair. Only one piece for which the decorative design was redone has been found: a Liaoning-type bronze dagger handle attributed to Sincheon, Hwanghae-do Province.61 Although the soldered part was polished and engraved in order to match surrounding designs and obscure the boundaries, the repair work done on this piece is still noticeable. A bronze mirror with a fine linear design found in Yangyang, Gangwon-do Province had been made with a mold that had itself been repaired before casting, rather than re-engraving the pattern on the mirror. The most common means of repair was soldering. It was done to fill in holes or to repair breakage of almost every type of bronze product such as daggers, mirrors, axle ornaments, and rings. Soldering was done throughout the Bronze Age and continued into the Goryeo and the early Joseon periods in Korea. In order to apply solder, the back side was plastered with clay and the area around the spot that required soldering was bordered with clay. Then molten bronze was poured in to fill in the crack or hole before the surface was polished.62 There were four basic soldering techniques. The first method was to solder only holes or broken spots. On the less noticeable side, the spot to be repaired was soldered a little bit wider so that the new bronze would adhere more strongly to the original piece. When the spot to be repaired was on the less noticeable back side the area surrounding the spot was also soldered for further reinforcement. The second method involved cutting the edge of one side of the part to be soldered in a serrate pattern and then applying solder as seen in bronze rings and bronze halberds. This method was not used for holes but to solder a missing part when one side of a bronze vessel was broken off. The serrated edge reinforced the soldered part by preventing it from sliding off. The third method was employed in repairing bronze rings. Soldering was done in the small semicircular ear form, which was also intended to prevent the reattached piece from sliding off. The fourth method involved carving out tiny holes on either side of the broken spot and linking the two holes before soldering to reconnect the broken part. Of the four methods, the third method was commonly used for Japanese bronze ware. We can also see that the severed parts of a split-bamboo-shaped bronze implement excavated in Namseong-ri and the trumpet-shaped bronze object excavated in Dongseo-ri were repaired in this way. Although not a repair technique, a rivet was used to join the separated parts.

Of all the masters in Korean painting, Kim Hong-do is the best known one to Koreans. Nevertheless, defining the artist and his art is not an easy task. What kind of person was he? What and how did he paint? What circumstances was he under? More specifically, what kind of life did he lead and what do his paintings mean to us today? To complicate the matter more, there are serious misunderstandings in the already well-known facts about him. This paper is a summary of what has been discussed in my book, Danwon, Kim Hong-do: Joseon, All Too Joseon which was published in 1998 to shed some light on the artist. II A PAINTER EXCLUSIVE TO THE KING Kim Hong-do was born in 1745, the 21st year of King Yeongjo’s reign, and died in 1806, the sixth year of King Sunjo’s. His lifetime of 62 years was a time of serenity and economic abundance. Unaffected by any war, the period is often described as the renaissance of the latter half of Joseon. Such peace and affluence were attributable to two kings who made genuine efforts to care for their people: Yeongjo, the 21st king of the dynasty who reigned for 52 years from 1724 to 1776 and Jeongjo, the next king who ruled for 24 years from 1776 to 1800. Not surprisingly, his paintings are full of optimism and pride for Joseon culture. One should not overlook this historical background to truly understand Kim and his works. Kim Hong-do was a court painter. The position was equivalent to a low-ranking government official of today. Neo-Confucianism was the governing principle of Joseon at the time and art was not of primary importance (albeit often considered as necessary) for high-class yangban (male Confucian scholars). The court painter position was a specialty occupation mostly filled by jungin (“middle people,” a hereditary class of technical specialists) and the highest position a court painter could aspire to was just grade level six. However, Kim Hong-do was too brilliant to remain ordinary. Under King Jeongjo, there were about thirty court painters. The ten who excelled the most were dispatched to Gyujanggak (a research library and institute of the court) and specially treated for their talents. Kim Hong-do never belonged to this elite group of court painters although he produced Gyujanggakdo (a painting of Gyujanggak that had just been established) for King Jeongjo (Plate 1). How was he able to create so many paintings for the court then? This author surmises that Kim Hong-do was a ‘designated painter’ invited to respond exclusively to the king’s order. This would explain why he was frequently exempted from the typical duties of other court painters such as creating documentaries for Uigwe (Royal Protocols of the Joseon Dynasty) but produced special works on the king’s direct behest while staying close to him. For the same reason, Kim’s name was seldom found on the list of court painters whose duty was to illustrate court events and affairs except during the early years of his artistic career. ( Plate 1 ) Gyujanggak Archives, 1776; Ink and color on silk; 144.4x115.6cm; National Museum of Korea The first floor of Juhapnu Pavilion, the large, two-story building with a hipped-and-gabled roof at the center, is Gyujanggak. Immediately after his accession to the throne, King Jeongjo ordered the establishment of a research library and institute within the palace called Gyujanggak, a center of academic research and politics. The peak rising on the left is a mountain to the rear of Changdeokgung Palace, which reminds us of the landscape painting style of Jeong Seon, pen named Gyeomjae. The refinement of the original was almost entirely lost because of the major repair work done in later years. In fact, Kim Hong-do appeared in the album of King Jeongjo despite the fact that he was just a low-ranking government employee. In 1800, the last year of his reign, King Jeongjo said in retrospect, “I know the painter Kim Hong-do well and have let him do all the major paintings of the court for thirty years.” The major works the king meant include: three portraits of the king; a drawing of Haesang Gunseondo (The Immortals on the Sea)—a huge painting on the wall of Changdeokgung Palace; a Buddhist painting of the main hall of Yongjusa Temple made in memory of Prince Sado, the father of King Jeongjo (Plate 2); the illustrations in such significant books as Wonhaeng Eulmyo Jeongni Uigwe (King Jeongjo’s Royal Progress to Suwon in 1795) and Oryun Haengsildo (a book elaborating on the moral rules to govern the Five Human Relations of master and servant, father and son, husband and wife, siblings, and friends); and Jubujasiuido, a painting on Daehak (Great Learning) that the king had considered essential for a ruler to learn and had studied all his life (Plate 3). ( Plate 2 ) The Three Tathagatas (The Three Buddhas of Past, Present, and Future), 1790; Ink and color on silk; 440x350cm; Yongjusa Temple With Sakyamuni at the center, Amitabha to the left, and Bhaisajyaguru to the right, the Three Tathagatas and the Eight Great Bodhisattvas are depicted in ovals. At each of four corners is one of the Four Divas. The canvas is densely filled with Tathagatas, Bodhisattvas, and other Buddhist images. It depicts the world of Buddha and suggests that this is a perfect space of law and wisdom. ( Plate 3 ) The Full Moon and the Calm Water, fourth panel of Jubujasiuido painted on a folding screen, 1800; Ink and color on silk; 125x40.5cm; Leeum, Samsung Museum of Art This is the fourth panel of a folding screen. It depicts the ‘Honest Mind.’ The full moon rising in the dead of night shed lights evenly on the mountain, where nobody is. On the bottom is water flowing from the fall. It fills the valley with crystal clear water. This scene implies the condition of a man who has reached a state of proper will and complete honesty. Meanwhile, King Jeongjo once requested Kim Hong-do to journey to Mt. Geumgangsan as well as the eight beautiful scenes of Danyang on behalf (Plate 4 and 5) and considerately gave special orders to the villagers to serve Kim well during his sketching trip. The king also asked the artist to copy the portrait of General Im Gyeong-eop (1594~1646), whom he had respected greatly. With the king as his benefactor, Kim Hong-do served as a chalbang (a government position equivalent to that of a stationmaster today) and as a hyungam (a magistrate) of Yeonpung in Chungcheong-do Province. Both positions were the highest honor that a court painter could attain. ( Plate 4 ) Guryongyeon Waterfall, a section from Haesancheop, an album of Mt. Geumgangsan paintings, 1778; Ink and color on silk; 30.4x43.7cm; Private collection Haesancheop had long been known only through photographs until 1995, when the album was actually opened to the public. At the time of its release, the authenticity of the album was in dispute because of a fake seal stamped on each of its leaves and unidentifiable calligraphy on each page. Ironically, the seemingly fake seal and calligraphy make this album all the more authentic because works for the king did not bear seals of the painter, and when recording the name of the place, the name was written on a separate leaf outside of the painting. The seals and calligraphy were likely added when the genuine album was disassembled to be sold in parts. The writings of the original owner of the album are not seen. ( Plate 5 ) Oksunbong Peak, a section from Danwon Jeolsebocheop, an album of Mt. Geumgangsan paintings, 1796; Ink and color on silk; 26.7x31.6cm; Leeum, Samsung Museum of Art The peaks rising along the river look like bamboo shoots after rain. Thus, these peaks are named Oksunbong, meaning peaks of bamboo. The highest peak is at the center, and the other peaks are leaning down on the left. On the right is a distant view of a mountain. The domineering appearance of the huge cliff and the sense of vast space are truly amazing. On the river at the bottom is a small boat in shade in which two gentlemen are enjoying leisure time. III BIRCH TREE GARDEN Records clearly show that Kim Hong-do belonged to the jungin, “middle people” as he was a descendant of a low-ranking military officer. Considering the fact that court painter positions were hereditary, Kim must have displayed exceptional talent in painting for he was the first in his family to become a court painter. When he was only about seven years old he was brought to Gang Se-hwang (1713~1791), a noted art critic and learned painter. Kim Hong-do presumably spent his childhood not far from the house of Gang’s wife because Gang Se-hwang then was staying with his in-laws in Ansan. Although not proven, it was Gang who recommended Kim for an admission into Dohwaseo (Royal Bureau of Painting). As Gang Se-hwang was assigned to a government post in his sixties and Kim Hong-do had already made his fame in his twenties in the art world, the recommendation by Gang was possibly made indirectly via some other authority in the court. Another safe guess is that Kim Hong-do lived in Mapo, one of the ports of the Hangang River during his stay in Seoul because the artist used ‘Seoho (West Lake)’—the name used to describe the beauty of Mapo—as his pen name. Kim Hong-do allegedly spent his later years in a secluded house built within the estate of Kim Han-tae, an aficionado of calligraphy and painting. Kim Hong-do used a number of pen names and Danwon is the most well known one. Danwon literally means ‘birch tree garden’ and was originally used by Li Liufang, a literati painter of Ming Dynasty whom Kim tremendously respected for his noble grace and profound knowledge. At that time, Li was best known for his teaching materials—the main source material for Jieziyuan Huazhuan or Manual of the Mustard Seed Garden (a manual of Chinese Southern School painting complied during the early-Qing Dynasty). Kim Hong-do also had several courtesy names and Saneung is the most frequently used one. The name signifies that ‘only a man of virtue can behave properly all the time without being swayed by material things’ and implies what kind of man Kim himself aspired to be (Plate 6). The pen name Danwon was first used in the summer of 1781 in an inscription by Seo Yu-gu on Kim Hong-do’s painting ‘Segeomjeong Ajipdo (Social Gathering at Segeomjeong).’ ( Plate 6 ) Pouipungryudo (A Gentleman Chanting Poetry); Ink and color on silk; 27.9x37cm; Private collection “I will chant poems until the day I die in this humble house with paper windows and clay walls,” is what the inscription on this painting means. This inscription is a kind of autobiography, and the gentleman depicted seems to be Kim Hong-do himself. In other words, this piece is a self-portrait. Books, scrolls, a reed instrument, a flute, a gourd filled with rice wine, and a plantain leaf tell us what kind of person Kim Hong-do was. The gentleman is wearing a hat but is barefoot, evincing his disinterested attitude. As suggested by the fact that he adopted the same pen name as Li Liufang, Kim Hong-do strived to grow into a cultivated scholar of high class with comparable knowledge that could not be easily achieved by a middle class painter. In fact, Kim seemed to endeavor to keep his social life rarified, being acquainted with then leading artists, scholars, and bureaucrats such as Yi Yong-hyu, Gang Se-hwang, Seong Dae-jung, and Yi Byeong-mo. Considering the fact that two of the most representative intellectuals of the time such as Sin Wi and Hong Hyeop-ju called the artist ‘Danwon-ja’ (a suffix that meant the same as dear in English to show some respect), the hearsay that the artist once drew obscene paintings is absurd and not worth an explanation. Numerous records confirm that the artist was a tall, handsome man with a sturdy frame like a Taoist hermit. Therefore, it is not a farfetched assumption that his painting styles—subjects such as men, deer, and donkey are depicted slender and trees are illustrated straight with lower branches trimmed off—are closely related with his physical appearance and spiritual fortitude. Kim Hong-do was also famous for his wine drinking and love for humor, and the behavior can be found in his biography written by Jo Hui-ryong. One of his numerous pen names ‘Chihwasa’ is also indicative of this tendency of Kim as it means an artist who can enjoy wine and the pleasures it brings. In truth, Kim left several paintings created under the influence of alcohol. His excessive drinking, however, seemed to affect his health and made him suffer a minor but chronic disease in his later life. IV SPLENDOR OF YOUTH, EVANESCENCE OF OLD There has been an assertion that Kim Hong-do pompously regarded himself as elite once he became a hyungam in 1971. Soon, the artist was disfavored and disentitled by King Jeongjo and started leading a hard life as a consequence. However, the assertions are serious misunderstandings based on mere guesses. As for the latter assumption, the author was recently able to confirm that Kim Hong-do was granted a royal pardon just ten days after his arrest for a trial at the State Tribunal. Meanwhile, some academics conclude that the artist suffered from financial straights in his later life because Kim was financially dependent on Kim Han-tae, a lowly merchant selling salt. However, considering the fact that the salter made a fortune as an interpreter and was influential enough to control even the ministers of Joseon as one of the richest men in Hanyang (the capital of Joseon), the conclusion is shortsighted. On the contrary, the life of Kim Hong-do was affluent as he was the most famous painter in the country during the reign of King Jeongjo (Plate 6). A record that even Nam Gong-cheol, an aristocrat from a noble family, could not easily acquire a painting by Kim Hong-do clearly indicates that he had little time for any work other than for those ordered by the king. The paintings, which only the king could order and attain, must have been highly prized by the people, and the rarity must have made Kim’s artistic style influential not only on his fellow artists of the 1800s but also on those of the first half of the 1900s. Despite all these facts, however, a couple of absurd assertions about Kim Hong-do have recently surfaced: he went to Japan and engaged in espionage working as an unidentified Japanese painter and he had to return to Japan to relieve himself from the financial misery in his later years. It is truly deplorable that such ill-informed historical assertions have been irresponsibly distributed through newspapers, radio, and television. Kim at that time was working hard as the magistrate of Yeonpung, Chungcheong-do Province to relieve the starvation resulting from a three-year-long drought of the region and he was ordered to come to the capital for a trial to review his performance in regard to this duty to manage the drought in his jurisdiction. It is true, however, that Kim Hong-do suffered poor health and lived in such poverty in his last years that he could not even afford tuition for his son, Yang-gi. He really seemed to be deeply depressed for some reason at that time. The author reasons that the suffering was due to the death of King Jeongjo in 1800 and the political as well as situational changes that occurred to the people whom the dead king had favored including Kim. The fact that Yang-gi, Kim’s only son, was born around 1792 makes this author’s reasoning probable because the aforementioned situation of Kim Hong-do, which Yang-gi might have delivered, happened after King Jeongjo’s death. In fact, Kim Hong-do was no longer exceptional and assigned to Gyujanggak in 1804 during the reign of King Sunjo. He had to undergo tests like any other court painters who were young enough to be his nephews. The old artist usually earned high scores but the very fact that he was put to a test was shameful enough for him. In fact, he sometimes disgracefully received the lowest scores of all. To make it worse, even the post of court painter did not last long as he had to quit due to his illness in 1805. At the end of that year, Kim Hong-do created Chuseongbudo (the Sound of Autumn), which illustrates the emptiness of his life that he must have felt at that time (Plate 7). The painting is the last work of Kim Hong-do. ( Plate 7 ) Chuseongbudo (The Sound of Autumn), 1805; Ink and color on silk; 56x214cm; Leeum, Samsung Museum of Art Old Ouyang Xiu, a Chinese poet, historian, and statesman, in the last days of his life, heard a strange sound while he was reading a book, and sent his servant to the courtyard to investigate it. The servant pointed to the sky and said, “Only stars and the moon are shining brightly and there are no human traces at all. The sound comes from between the trees.” The trees shaking in the cold moonlight look like the sorrow of an old man. Seen from the extant letters by Kim Hong-do, the artist was a man of good personality. He seemed not only very affectionate, affable, and sociable but also of playful humor and mischievous zest as he expressed in his genre paintings: the humorous facial expressions and interesting eye contacts in mix with complex composition of the people coming and going. Interestingly, Gang Se-hwang, the teacher of Kim Hong-do, was also a man with great sense of humor as portrayed in his self-portrait. Kim Hong-do seemed to have been a true believer in Buddhism (Plate 12). The artist made a huge offering to Sangamsa Temple near Yeonpung where he served as a magistrate. According to an epitaph, he had his only son after praying in this temple. Especially in his later years, he created a number of masterpieces that embody his sentiments about Buddhism like Gwanseeumbosal (Buddhist Goddess of Mercy). These works show the highest level of perfection that religious paintings could ever attain. More compelling evidence to show that Kim Hong-do was a true Buddhist can be found in Danwon Yumukcheop (Manuscripts left by Kim Hong-do) compiled by his son Yang-gi. The book is a collection of literary works by Kim Hong-do and its cover is decorated with densely filled tiny patterns of ‘卍’—the symbol of a temple. V JOSEON, ALL TOO JOSEON Kim Hong-do was a master painter in all types of paintings: landscapes, flowers, birds and animals, genre paintings, figures from ancient history, portraiture, and even Buddhist paintings and illustrations. He was undoubtedly, so to speak, the ‘national painter, the best in Joseon.’ Kim Hong-do was also so talented in calligraphy and literature as to compose Chinese poetry impromptu. What is more, he was famous as a musician excellent at playing the daegeum (a large transverse bamboo flute) and geomungo (a traditional Korean musical instrument with six strings). Two Chinese poems by Kim that the author has recently verified the authorships of are the representative examples of his excellent connoisseurship in literature. Kim Hong-do was not only a versatile artist whose expertise covered almost every field of fine arts ranging from poetry, calligraphy, and music to painting but also a man of fine appearance and good personality. It deserves a thought what these recently found facts about his talents signify. King Jeongjo who favored Kim Hong-do was a great king comparable to King Sejong the Great. During his reign, King Jeongjo was a real practitioner who acted upon his devotion to the social good and his knowledge was profound enough to instruct his subjects in person. He was also a great calligrapher and artist who could even do seal carving as a hobby. He was no doubt a man of excellent artistic gifts. Examining historical documents in detail, King Jeongjo paid close attention to the small details of paintings that are otherwise likely to be overlooked as trivial matters and often provided creative solutions to the court painters. The king also had a great interest in maintaining the court painters’ qualifications by promoting the outstanding and expelling the inattentive. It was only natural that even the fairly good painters could hardly meet the standards set by the artistically talented king. Kim Hong-do was a painter with profound learning in every field of art (poetry, calligraphy, painting, and music) and a fine figure with an all-around, integrated personality. Kim’s art must have met the exacting standards of King Jeongjo and the words of Jo Hui-ryeong (1789-1866)—a literati critic and painter—prove so: “Every painting submitted by Kim Hong-do immediately satisfied King Jeongjo.” Kim Hong-do was much more than merely a talented painter. He represented the Joseon Dynasty. The recent rumors—he produced pornographic paintings or that he went to Japan and worked in secret—are groundless fabrications in complete disregard of the historical facts. Kim Hong-do is often considered to be a truly Joseon painter. As proven by the fact that he did three portraits of the king, he was recognized as the best painter of the time in realistic rendering. The hyper-realistically depicted tiger attests to his proficiency in realistic expression (Plate 8). The more important thing about him is, however, not the fact that he painted extremely well but that he embodied the sentiment of the Korean people in his paintings, regardless of the subject. Every painting ranging from real figures of ancient history to imaginary landscapes (the form that was originated in China) convey the very sentiments of Joseon that we can see in his genre paintings and landscapes of Korea and its people. For example, Samgongbulhwando (Pleasure of Rural Life) is based on a story of ancient China but is full of Korean sentiment (Plate 9). With the same title and theme of Chinese classical poetry, Kim was flawlessly able to illustrate the life of Joseon people. ( Plate 8 ) Songhamaenghodo (A Tiger Under the Pine Tree), tiger by Kim Hong-do, pine tree by Yi In-mun; Ink and color on silk; 90.4x43.8cm; Leeum, Samsung Museum of Art The moment a tiger suddenly turned its head as if sensing something while walking slowly was captured in this piece. Thousands of thin brushstrokes were used to depict the hair and brindled body of the tiger, which look extremely natural. Although depicted hyper-realistically, the massive weight of the body is apparent, and the quick action of the tiger is true to nature. ( Plate 9 ) Samgongbulhwando (The Nature Better than the Official Ranks), 1801; Ink and color on silk; 133.7x418.4cm; Leeum, Samsung Museum of Art Samgongbulhwan is quoted from poetry by Dai Fugu of Southern Song China. It means that the pleasure of living countryside will not be foregone for any of the highest government posts, even one of the three ministerial posts. A huge house on the slope of a mountain above a big river and its surrounding fields are depicted on a folding screen. The bottom part of the first panel of the folding screen was damaged by fire. On his famous Seondongchwijeokdo (Hermit Boy Playing a Bamboo Flute), the wimpling of the robe is depicted in thin and thick lines as if they are moving to the music. It seems that the tunes and melodies that the Korean boy is playing are visualized into the undulating lines of the boy’s clothing (Plate 10). This was only possible with his combined aptitudes in both music and painting. Kim’s paintings also have his unique poetic sentiment that only his literary sensitivity can invoke. Last but not least, Kim Hong-do is well versed in using blank space, generously applied but not without artistic sensibility. Seonsanggwanhaedo (Boating on the Waters) is a stellar example of his wide void space in which his literary sensitivity is delicately expressed (Plate 11). This author sees that such characteristics result from the generous and relaxed personality of Kim Hong-do himself. ( Plate 10 ) Seondongchwijeokdo (Hermit Boy Playing a Flute), a panel from Sinseondo (Hermits) painted on a folding screen, 1779; Ink and color on silk; 130.7x57.6cm; National Museum of Korea In the inscription on this painting, Gang Se-hwang wrote, “How wonderful is the technique to make nine holes into green jade to make this flute. People call the boy playing a flute by the name Okjagyo, but I don’t think it is the right name.” We don’t know what the hermit boy’s real name is. The boy has a flat face and broad forehead, and smart-looking eyes, and he is undoubtedly a Joseon boy. The contours of the folds of the boy’ s clothing rhythmically represent the tune of old Korean music the boy is playing. That is, the strong brushstrokes resemble the beating and rhythm of Korean music, and the lines flowing thick and thin are melodies of Korean music as they are. ( Plate 11 ) Seonsanggwanhaedo (Appreciating the Plum Blossoms); Ink and color on silk; 164x76cm; Private collection Dimly through the haze in the distance is a steep cliff with a few blossoming trees. Thecenter of the cliff and trees are well focused, but the scene gradually blurs to the left and right. The brushstrokes become less and less pronounced towards the end as if they are lingering around the boundary between existence and non-existence and finally disappear into a blank space. This is a subjective landscape seen from the eyes of an old man boating on the bottom, rather than objective landscape in the real world. That is, the viewers of this painting are seeing the hill and trees the old man is looking up at as they are. On the other hand, the viewers are also simultaneously looking at the old man. The empathy of Kim Hong-do towards the old man is so palpable that the boat is depicted as if it is looking up from the surface of the water. ( Plate 12 ) Yeombulseoseungdo (Old Buddhist Monk Praying to Buddha); Ink and color on silk; 20.8x28.7cm; Gansong Art Museum The back of the old monk’s head is so clean and transparent that it is shining to charm the viewers. We can almost feel the thin shoulder bone of the monk under his robe. The monk is flying on something into the east sky; perhaps a lotus flower, clouds, or lotus leaves. The bright light fills the fine ramie robe of the monk. We don’t know whether the light is of a halo or the full moon. Art and religion are one. For most people, art is art and religion is religion, nothing more or nothing less, because they consider neither art nor religion to be truly genuine. However, when both religion and art become something truly lofty, art is religion and religion is art. When this state is attained, art and religion become life to you. VI A PAINTER WHO LOVED HIS TIME AND PEOPLE Kim Hong-do was indeed the best painter in the latter half of Joseon to define Korean beauty. Mun Il-pyeong (1888~1939), a historian and critic, referred to the artist as a ‘Painter Immortal.’ Mun coined this name based on a quote of Jo Hui-ryeong who had described the appearance, character, and disinterested attitude of Kim Hong-do to be similar to those of an immortal, and the title implies the loftiness and the depth that he felt from Kim’s art. The word immortal was also attached to the artist to indicate that he was a government employee who closely attended the king because, in Joseon, even the lowest-ranking officials were called immortals to respect the honor to serve the dynasty. The paintings of Kim Hong-do reveal the artist’s love for the time he lived. In his paintings exist a tall, good-looking, and big-hearted man with an intrepid spirit and a delicate nature; an erudite with deep knowledge in every field from poetry to calligraphy, painting, and music; and the ancestors who lived their time in the land of beauty and of peace (Plate 13). His paintings are imbued with the generous and humorous nature of the artist himself and his teacher, Gang Se-hwang and the keen connoisseurship of King Jeongjo, the absolute supporter of the artist and the powerful but caring ruler to his people. ( Plate 13 ) Giroseyeongyedo (Social Gathering of the Elderly), 1804; Ink and color on silk; 137x53.3cm; Private collection This clean canvas imparts the feeling of cleanness we have immediately after washing our faces. This piece depicts the scene of a social gathering of elderly gentlemen in Gaeseong in September 1804, following the example of Goryeo people who had gathered some two hundred years previously at Manwoldae at the foot of Mt. Songaksan, the site of a palace of the Goryeo Dynasty (918-1392). Under the shade, each of the sixty-four old men is served at his own small dining table. At the center is one big table with a porcelain vase on it. The inscription on top states, “Gaeseong is an old city with a great reputation. The mountains and rivers are spectacular and splendid, and the people of Gaeseong are well-dressed. This is because the city still holds the elegance as an old capital of Goryeo.” The writings on the bottom are names and indications of the family origins of the 64 gentlemen who participated in this meeting. Kim Hong-do illustrated in his art the attitude, the spirit, and the cultural pride of the times that he lived. There is no incompatible disharmony, no self-confliction, or no loss of self-identity, although his later paintings touch upon the evanescent nature of life that he felt as an old man and the depth of the religious spirit that the artist experienced as a devoted hermit (Plate 12). All these pictorial features are the epiphany of the life that Kim Hong-do lived as an artist and of the lives of people who lived in Joseon together with the artist.

The Baekje Kingdom (18 B.C.~660) and Japan were closely related in ancient times. Historical records as well as stone and metal artifacts show that many Koreans from Baekje introduced advanced culture and technology to Japan. In truth, the influence of Baekje can be found not only in Buddhist art but also in other aspects of material culture in Japan now. In addition, archaeological research on the corridor-style stone chamber tomb and roof tiles has been seeking for evidence of the influence of Baekje culture on Japan. However, the relative scarcity of artifacts from Baekje compared to those from the Silla or Gaya kingdom has been a major hindrance to the study and the authentication of the vestiges of Baekje in Japan. Fortunately, increased excavation of Baekje Kingdom sites has led to discoveries of new, important artifacts that have helped archaeologists better understand how Baekje influenced Japan. Special exhibitions like this one that present the cultural brilliance of Baekje will serve as good opportunities for Japanese researchers to realize the cultural importance of Baekje in Japan and to reexamine the relationship between the two ancient countries. This paper is based on these recent findings and archaeological achievements and will discuss a few issues in tracing Baekje’s influence on Japan from an archaeological viewpoint. II BAEKJE AND JAPAN BEFORE THE 4TH CENTURY Historical records suggest that the official diplomatic relationship between Baekje of Korea and the Yamato Kingdom of Japan was established during the reign of King Geunchogo (346-375) of Baekje. During this period, Baekje was a rapidly rising power. Given that Baekje and Goguryeo were rivals, it was diplomatically advantageous for Baekje to establish a friendly relationship with the Yamato Kingdom of Japan. To the Japanese, building a relationship with Baekje was also important as it allowed the importation of various resources including iron and a variety of advanced technologies. This does not mean that there was no interaction before the 4th century between the Baekje region and the islands of Japan. The recent recovery of artifacts especially of earthenware proved so. From Tonokubi stone coffin no. 3 in Nagasaki Prefecture on Tsushima Island, pieces of earthenware with a straight mouth and two ears were found. (Plate 1-1) (Oda et al, 1974). Discovered together with Yayoi earthenware, archaeologists dated these to between the latter half of the first century and the first half of the second century (Oda, 1978). Meanwhile, earthenware vessels with their shoulders decorated with serrate patterns were discovered at SD159 during the 17th excavation of Hakata historic site in Fukuoka Prefecture (Plate 1-3) (Fukuoka Municipal Educational Commission, 1985). Previously, most examples of this type of earthenware with ears and serrate patterns were those found in the 3rd and 4th century jar-coffin in Jeolla-do Province. However, the same type of earthenware has recently been discovered at the historic sites of Oseok-ri, Seocheon (Plate 1-2) (Yi Nam-seok 1996) and the Bungang/Jeoseok-ri tumulus, indicating that such earthenware was also used in the west coast of Chungcheongnam-do Province (Yi Nam-seok 1997). Another noticeable discovery was made from a shell mound in Daognae shell mound, Busan. It contained the earthenware jar with a straight mouth, saw-toothed patterns around the shoulder, and a raised band around the neck (Bokcheon Museum, Busan 1997). This type of straight mouth jar was presumably produced in Baekje at around the same time as the jars with two ears were (Plate 1-5). Albeit limited, the aforementioned material proofs are enough to imagine the connecting route that connects from the southwest of the Korean Peninsula to the south coast and then continues to the northern Kyushu of Japan via the Tsushima Islands. They are indicative of some interaction between Baekje and Kyushu of Japan. ( Plate 1 ) Types of earthenware jar with a straight mouth and two ears excavated in Japan 1 Tonokubi Stone coffin no.2 2 Pit tomb at Oseok-ri no. 95-3 3 SD159 of the 17th excavation at Hakata site 4. Dongnae Shell mound 5. Collection of Chungnam National University As for the region further east of the northern Kyushu, the kiln sites from the 4th century discovered at Deai site in Hyogo Prefecture are noteworthy (Kameda 1989). The kiln site was two meters in length and one and a half meters in width and was built on a slope of 10 degrees. Only the floor remains today but the kiln characteristically has a partition between the combustion chamber built vertically underground and a kiln chamber. This type of kiln was once discovered at the sites of Samnyong-ri and Sansu-ri kilns in Jincheon, Chungcheongbuk-do Province. More studies on theses sites should be done to evaluate the relationship between the kilns, but it is conceivable that the Baekje region could have been the home of those who made the Deai kiln. Other telltale artifacts about the relationship between Baekje and Japan include a horse-shaped bronze belt hook presumably excavated from Sakakiyama mounded tomb in Okayama Prefecture (Wada 1919). Because the circumstances surrounding the discovery of this belt hook are dubious and the other artifacts excavated together in the ancient tombs date back to the 5th century, the authenticity of this hook could not be verified (Azuma 1992). If genuine, it should be studied in relation to the numerous other horse-shaped belt hooks that have been excavated from the Cheongdang-dong site in Cheonan. As pointed out earlier, relatively few archaeological materials are available, but there is enough to show that Baekje culture in the 3rd and 4th centuries was introduced to the northern Kyushu and Kinki area via the route from the south coast of the Korean Peninsula. Furthermore, newer discoveries that could shed light on the relationship between Baekje and Japan before the 4th century are expected as excavation of the 3rd and 4th century historic sites of the Baekje Kingdom has been increasing. III BAEKJE AND JAPAN IN THE 5TH AND 6TH CENTURIES During the 5th and 6th centuries, new techniques in various fields such as iron and metalwork production, pottery, and civil engineering were introduced from the Korean Peninsula to Japan, causing great socio-cultural change in Japanese society. The change in Japan wrought by the introduction of Korean culture was comparable in magnitude to that experienced by Japan during the Meiji Restoration in the 19th century when Japan embraced western civilization. Koreans who came to Japan with new techniques and civilization were the driving force behind this change. Historical records indicate that there were various groups of people who came to Japan from Baekje including a new group of technicians and academics. From an archaeological perspective, the influence of Baekje culture on Japan is apparent during this period. This paper will discuss these influences with special focus on a number of issues. 01 SUEKI POTTERY AND BAEKJE EARTHENWARE Except for the earlier mentioned case of the Deai kiln, Japanese began to produce pottery using kilns long after Koreans used tunnel-type wood-firing kilns for their earthenware or stoneware. It was not until the 5th century when many people from Korea crossed to Japan that sueki earthenware began to be produced in Japan. It was confirmed that the first sueki pottery was produced at the kilns scattered in such regions as Fukuoka, Kagawa, and Okayama but the major producers were the kilns of Suemura site in Osaka Prefecture. Many kilns and pottery excavated in the course of land development projects revealed that Sumemura kilns site was the center of sueki pottery production in the 5th and 6th century Japan. The most daunting task in the study of sueki pottery was to find the oldest sueki pottery and its origin. In this regard, very important discoveries were made at the Obadera kiln site. Two kiln operations (No. TG231 and TG232) were found along with a settlement for potters who are believed to have produced sueki earthenware (Osaka Educational Commission et al. 1995). The pottery found includes a pedestaled dish, jar stand, and jars that are similar to earthenware discovered in areas on the south coast of the Korean Peninsula including Busan, Gimhae, Masan, and Haman. Meanwhile, there are some examples of pedestaled dish that were made using the production method of hajiki earthenware (traditional Japanese ware). It seems that the production of pottery started as Korean potters from the south coast of Gyeongsangnam-do Province made the kilns in Obadera and began to teach the Japanese potters who made hajiki wares. Interestingly, however, the dish with fitted cover (Plate 2-1) and hasue, sue ware wine server (Plate 2-2)—the two most typical forms of sueki ware—have hardly been discovered at the Obadera kiln site in Japan or in Gyeongsangnam-do Province in Korea but from the sites of the Baekje Kingdom, especially in Jeollanam-do. Also, rice-bale shaped bottles of the early sueki period were discovered only in Jeollanam-do Province as was the case of Mansu-ri Tomb no. 1 in Yeongam. These new styles of sueki were produced at kiln no. TK73 or TK216, which was established later than the kilns at Obadera. These facts suggest that Baekje potters joined in the production of sueki pottery a little later than the potters from the south coast of Gyeongsangnam-do Province. ( Plate 2 ) Sueki pottery similar to Baekje pottery 1 Dish with fitted cover (Kiln TK216) 2 Sue ware wine server (Kiln TK85) 3 Rice-bale shaped bottle (Kiln TK73) From a Japanese viewpoint, what part of Baekje these potters came from is meaningful. It is a difficult question but not without a clue: it could be solved if we closely study how the exterior bottom of the dish with fitted cover was made. After examining in detail the pots produced in Baekje, I found that various methods had been used to shape the bottom of the dish with fitted cover: cutting on the wheel; cutting while holding with hands; and wiping after cutting. It is not yet clear whether those various methods were regional or specific to any time. However, because most sueki ware dishes with fitted cover in Japan were shaped by cutting on the wheel, the region where the method was the most commonly used will be likely the starting point of Baekje potters in Japan. 02 IMMIGRATION OF BAEKJE PEOPLE TO JAPAN Around the beginning of the 5th century, there was a major transformation in cooking techniques that were introduced to Japan from Korea. Previously, the Japanese cooked rice in a reddish brown soft clay pot over fire. By the 5th century, however, a huge number of Koreans had immigrated to Japan and introduced a fireplace combined with earthenware steaming vessel and other pots. As this cooking method became common in Japan, a new type of earthenware was produced by Japanese pottery-making techniques. Soft-paste earthenware produced using beating technique had been introduced from Korea to Japan and named by Japanese academics Korean-type pottery. However, because this soft-paste pottery was used everyday unlike hard-paste pottery, it is likely that much of this pottery production used the techniques that had been around in Japan despite the fact that the Korean-type pottery was introduced from Korea to Japan. For this reason, this Korean-style pottery has been attracting a great deal of interest from Japanese archaeologists as an important material that could provide some detailed information about the lifestyles of Korean immigrants to Japan. Of all Korean-style pottery, that beaten with bird feet patterns was introduced by Baekje. This type of pottery with beaten bird feet patterns has been discovered in the northern Kyushu and Kinki area, mainly from the excavations at the Nagahara historic site in Osaka. Particularly noteworthy is the excavation NG95-36 where six pieces of soft-paste pottery including a steaming vessel, a bowl with flat foot, and a jar were discovered together (Plate 3-1, 3, 4, 6, 7, 8) (Sakurai 1998). Given that the color of and the clay used for these six pieces are the same, they all seem to be produced near the Nagahara site. The same site also yielded soft-paste pottery with surfaces beaten with parallel lines and cross strips (Plate 3-2, 5) and cups with a handle that are very similar to the ones discovered at the Jukmak-dong historic site (Plate 3-9), testifying that a large number of Koreans from Baekje indeed lived in the area. ( Plate 3 ) Pottery beaten with bird feet patterns excavated in the Nagahara site no. NG95 during the 36th survey (revision of Sakurai 1998) Korean-style pottery is important to understand the everyday lifestyles of Korean people who lived in Japan. As the study on the characteristics of Baekje-style soft-paste pottery progresses the residential areas of Baekje people in the other parts of Japan will be located. 03 HISTORIC SITE OF MT. JEONGJISAN AND PILLAR-WALL STRUCTURES The pillar-wall structure found in 1996 at the historic site of Mt. Jeongjisan in Gongju has been attracting the public’s attention (National Museum of Gongju 1999). It was possibly a very important ritual facility of Baekje that was used for keeping the wife of King Muryeong (?~526) prior to burial. The discovery was tremendously important for Japanese academics studying Baekje Koreans in Japan because it could be one of the houses of Korean immigrants to Japan. The construction of pillar-wall structures began with digging a square or rectangular foundation. Erecting several pillars along the inner edges of the foundation follows before filling the space between the pillars with walls of clay. In Japan, a pillar-wall structure built in the 7th century was discovered in 1982 at the Ano site north of Otsu City in Siga Prefecture. It was the first of its kind ever found in Japan despite the historical records of numerous Koreans having settled in the area. Numerous stone chambers with annexed coffin room built on square foundation were discovered along with clay miniatures of cooking fireplaces used for ancestral rituals in this area. Accordingly, there is both written and physical proof that Koreans lived in this area after the 6th century. Since the first excavation in 1982, these pillar-wall structures have been discovered in the vicinity of this site. Archaeologists strongly believe that they are Korean. Also, pillar-wall structures thought to have been constructed during the 5th to 7th century were discovered in Shiga, Nara, Osaka, and Hyogo. Korean-style pottery has been unearthed from the Nango site in Nara Prefecture and Oagata site in Osaka. In their vicinity were found massive iron working sites, which are more solid evidence that the pillar-wall structures and Baekje were related. In Korea, pillar-wall structures were found at Gongsanseong Fortress and at the historic site on Mt. Jeongjisan. Although no formal study was conducted, it has been said that similar structures were found at Pungnap-dong Fortress in Seoul and Wolseonghaeja Waterway in Gyeongju. Given the relatively few excavations of settlements, pillar-wall structures could be found in areas outside the former Baekje Kingdom. Available evidence needs to be investigated further to determine if there is any Baekje element in historic sites in Japan where pillar-wall structures were discovered. As pillar-wall structures have been found at historic sites for ancestral rituals and fortresses, rather than in ordinary residential settlements, it is likely that the structure itself likely served a special function. As structures with stone foundations were also discovered at the Nango site in Japan, a comparative study of the function of the pillar-wall structures in both Korea and Japan is needed. 04 INTRODUCTION OF METALWORK TO JAPAN Along with techniques for the production of ironware and pottery, the Kingdom of Baekje also introduced metalwork and goldsmith techniques to Japan in the 5th and 6th centuries, bringing great changes to Japanese culture. Hoping to discover the origin of gold metalwork excavated from ancient tombs in Japan, researchers have examined artifacts found throughout the Korean Peninsula. The findings of their researches, especially their comparative analyses of artifacts from Japan and the ancient tombs of Songsan-ri (including the artifacts discovered from the royal tomb of King Muryeong (501-523), the 25th monarch of Baekje) show that some artifacts—such as a belt with demon design, gold earrings, gilt bronze shoes, sword with a decorated ring pommel, and bronze bowls—that had been discovered in Japan were related to the Baekje. However, recent excavations at an ancient Gaya site are producing some types of artifacts that were previously believed to be only available in Baekje. In a broad sense, gold metalwork commonly found at sites of Baekje, Gaya, and Japan can be argued to be the artifacts of Baekje. However, there is a need to determine their provenance (whether those excavated in Japan were brought directly from Baekje or made in Japan, brought to Japan via Gaya, or made by Baekje artisans on the islands of Japan). This is not an easy task as it requires continued research. The next part will discuss this issue as exemplified by the studies of the gilt bronze shoes (Yoshii 1996). There are three types of gilt bronze shoes that were produced on the Korean Peninsula: the Goguryeo Kingdom type, of which only the sole was made of gilt bronze; the Silla Kingdom type, which had the top and heel parts made of plates of gilt bronze jointed on the sides; and the Baekje Kingdom type, which had two side plates fixed on the top and the heel. Except for one case, all the gilt bronze shoes discovered in Japan were made by fixing two side plates to the top and the heel. Accordingly, they are considered Baekje-style shoes. After probing the production techniques of the Baekje shoes, I could classify them further into two sub-groups based on the shape and attachment techniques used for the side plates. Most of the first sub-group of Baekje shoes (Plate 4-2) were found on the Korean Peninsula but none have been discovered since the excavation of the tomb of King Muryeong (501-523), suggesting that no such shoes were made after his time. In contrast, all of the second sub-group were found in Japan (Plate 4-3) after King Muryeong’s reign and continued to be made until the early 7th century. The shoes that were produced in Japan after this period are assumed to have been made by artisans from Baekje. Meanwhile, although it is only natural to assume that the types of objects found in Baekje originated there, some cases are not that simple. For example, the shoes excavated from the Etafunayama mounded tomb in Kumamoto Prefecture (Plate 4-2) may be considered to have originated from Baekje based on their structure and the production techniques employed. However, the gilt bonze crown made by the same techniques (Plate 4-1) is in the style of a two-mountain broad-banded crown, which was popular in Japan in the 6th century. Technically, the shoes may be considered Baekje-style, but the gold crown is of a uniquely Japanese shape and was made along with the shoes. This raises two questions: Where were they made and how were they distributed? The answers are elusive. In seeking the answers, however, the interrelationship between Baekje and Japan becomes apparent. ( Plate 4 ) Baekje-style gilt bronze shoes and crown 1, 2 Etafunayama mounded tomb, Kumamoto Prefecture 3 Fujinoki mounded tomb 05 INTRODUCTION OF THE CORRIDOR-STYLE STONE CHAMBER TOMB TO JAPAN In the 4th century, the main burial facility of the keyhole-shaped mounded tomb in Japan was the pit-style stone chamber. This type of chamber was completely closed with ceiling stones, after the coffin of the deceased was buried. Some coffins of tombs were made of stone instead of wood, and the some chambers were closed by clay instead of stone. However, pit-style chambers were made for only one person. Compared to the pit-style stone chamber, the corridor-style stone chamber was made to accept more than two coffins as it is a crypt with a stone passageway that can be reopened and closed. The emergence of the corridor-style stone chamber tomb in Japan was due to Baekje’s influence. Furthermore, recent studies showed that corridor-style stone chamber tombs were introduced to the northern Kyushu region at different times and in different ways than in the Kinki region. The earliest corridor-style stone chamber tomb in northern Kyushu is the Taniguchi mounded tomb in Saga Prefecture dating from the latter half of the 4th century (Plate 5). There are two stone chambers placed side by side in which stone coffins are laid. The artifacts discovered indicate that this tomb was a typical pit-style stone chamber tomb. However, when a second excavation was conducted for its preservation, a corridor leading to the outside through an entrance on top of the setback was found revealing that this is a pit-corridor-style stone chamber tomb (Hamatama-machi Educational Commission of Saga Prefecture 1991). After the Taniguchi mounded tomb was constructed, genuine corridor-style stone chamber tombs and pit-corridor-style stone chamber tombs were constructed in the Sukisaki mounded tomb and Roji mounded tomb in Fukuoka Prefecture. The chambers have rectangular foundations with the front area wider than the back area, and the passage into each tomb is at the center. Among the ancient tombs concentrated around Kumamoto Prefecture, a stone chamber with a rectangular foundation and vault ceiling was discovered. During the fifth century, numerous corridor-style stone chamber tombs were built in northern Kyushu. In other regions of Japan, pit-style stone chamber tombs were still being constructed during that time. ( Plate 5 ) East stone chamber in Taniguchi mounded tomb Archaeologists disagree as to the origin of corridor-style stone chamber tombs in northern Kyushu. Most believe that corridor-style stone chamber tombs were introduced to Japan by Baekje when Hanseong (today’s Seoul) was the kingdom’s capital, given that the structure of stone chambers of Sukisaki and Roji mounded tombs resemble corridor-style stone chamber tombs in Garak-dong and Bangi-dong, Seoul. However, it was confirmed that pottery excavated from the corridor-style stone chamber tombs in Garak-dong and Bangi-dong were made of Silla-style pottery from the latter half of the 6th century, giving rise to the theory among Korean archaeologists that these ancient tombs were the tombs of Silla period or later. As a result, Korean and Japanese archaeologists disagree on the origin of corridor-style stone chamber tombs in Japan. On the other hand, it is also noteworthy that corridor-style stone chamber tombs that date back to the Hanseong period were newly discovered on former Baekje territory, as in the cases of tomb no. 12, 13, and 14 in Bungang/Seojeok-ri (Yi Nam-seok 1997). Interestingly, the entrance to the stone chamber of Bungang/Seojeok-ri tomb is one level higher, a common feature of the early pit corridor-style stone chamber tombs in northern Kyushu. If the changes that had occurred in the tombs of this structure on former Baekje territory are identified, the influence Baekje exerted on the birth and development of corridor-style stone chamber stone in northern Kyushu will be revealed. While corridor-style stone chamber tomb in northern Kyushu originated from Baekje, the composition of burial goods and funeral ritual were of traditional style which greatly differ from those discovered in Kinki. In the Kinki region, a different type of corridor-style stone chamber tomb came into being in the latter half of the 5th century. The Takaida-yama mounded tomb in Osaka (Plate 6) is exemplary of early corridor-style stone chambers from this area (Kashiwara Municipal Educational Commission 1996). At the center of the round tomb (22 meters in diameter), there is a corridor-style stone chamber with a rectangular foundation (2.34 meters in width and 3.73 meters in length) and an entrance at the right. The ceiling is missing and is assumed to have been a vault ceiling or similar structure. Although the tombs were robbed, numerous artifacts were still there at the time of excavation. A bronze iron was among them, and it is similar to the one that had been found in the royal tomb of King Muryeong. ( Plate 6 ) Stone chamber in Takaida-yama mounded tomb Stone chamber of Takaida-yama mounded tomb is similar to the corridor-style stone chamber tombs in Songsan-ri in that they were both constructed with thin slabstones, have an entrance on the right and the entrance was closed up with a stack of stones. The only difference is found in the chamber foundation. The Songsan-ri type stone chamber is square, and Takaida-yama’s is rectangular. Some academics believe this difference supports the speculation that the rectangular stone chamber of Takaida-yama mounded tomb appeared after the royal tomb of King Muryeong had been constructed. This means that the construction of the rectangular stone chamber of Takaida-yama was done later than previously known. However, some stone chambers with rectangular foundations, like Tomb no. 1981-6 in Pyojeong-ri, date back to the 5th century. More excavations of corridor-style stone chamber tombs should be conducted before specific details on the dates and the styles can be known. The Takaida-yama mounded tomb is also note-worthy in that the influence of Baekje can be seen not only in the structure and burial goods but also in the funeral ritual. The findings show that two people were buried in the tomb. They were laid parallel to each other along the main axis of the chamber and it is speculated that the heads probably faced the rear wall. The coffins were probably made of wood because coffin pegs were found around the bodies. Ancient Japanese tombs traditionally have two kinds of wooden coffins: one was a carved out log and the other was an assembly of thick wooden planks. Typically, nails were not used. However, nails and an iron clamp have been found in some pit-style stone chamber tombs and it is possible that these pit tombs were influenced by the pit tombs of the Gaya Kingdom. Compared with Gaya-style pit-style stone chamber tombs, wooden coffins built for the Takaida-yama mounded tombs are presumed to have been a box-type for which nails were used to affix the panels (short and long) as well as the base and the side. Although this type of wood coffin shows some differences in nail sizes, assembly methods, and the like, it is similar to the corridor-style stone chamber tombs of Baekje. Placing two corpses in a stone chamber was also something new. Based on the burial accessories and the size of the wood coffin excavated, it is presumed that the man was laid on the left and the woman was laid on the right, as seen from the entrance. The practice of burying a man and a woman together was common in corridor-style stone chamber tombs of Baekje. The couple in the royal tomb of King Muryeong is confirmed to have been a husband and wife. Many corridor-style stone chamber tombs discovered in Japan have two chambers or more in parallel. According to physical anthropologists, most couples found in Japanese tombs were presumably brother and sister until the 5th century. Tombs in which husband and wife couples were buried together appeared only in the 6th century in Japan. In the case of the Takaida-yama mounded tomb, there is no evidence showing the relationship of the two corpses, but if the couple turns out to be a husband and a wife, it will be the first such known case in Japan. Another interesting practice found in the Takaida-yama mounded tomb is the burial of pottery at the entrance of the chamber. In most ancient tombs excavated thus far, pottery was scattered on top of the entrance to the tomb or on its slope. The custom of burying pottery inside the stone chamber was done only after corridor-style stone chamber tombs were introduced to the Kinki region. Such a change suggests that the funeral ritual performed with the use of pottery became of greater importance. The Takaida-yama mounded tomb can be regarded as an early example of the custom of burying pottery as burial gifts. The Takaida-yama mounded tomb may have been influenced by elements introduced from Baekje in terms of not only the structure of the stone chamber and burial accessories but also the funeral ritual procedure. Given the characteristics of the tomb and the location of the Oagata-nami site to the south (a huge blacksmith area where Korean artisans likely lived), the archaeologist who participated in the excavation of the tomb presumed that the occupant of the tomb was a member of the royal family of the Baekje Kingdom and was the leader of the artisans in the area. Since Baekje tombs were introduced to the Takaida-yama, the features of the Baekje corridor-style stone chamber tomb were adopted by the Japanese and became widespread throughout the country, resulting in drastic changes in the funeral system in Japan. The influence of Baekje is palpable in the Takaida-yama tombs and in the general burial practices. IV BAEKJE AND JAPAN IN THE 7TH CENTURY During the Sabi years, Baekje developed a unique Buddhist culture, which exerted great influence on the Kingdom of Silla and Japan. Baekje tried to expand its power in alliance with the Goguryeo and Japan, but was defeated and eventually destroyed by the allied forces of Silla and Tang in 660. After the fall of Baekje, many of its people went to Japan. The next section discusses the changes that occurred in Japan as a result of the massive influx of Baekje culture. 01 INFLUENCE OF ROOF TILE POTTERS Buddhism was officially introduced to Japan when King Seong (523-554) of Baekje sent a statue of Buddha and a sutra to King Kinmei of Japan. At that time, there were confrontations within Japan over the acceptance of Buddhism. Soganoumata won over Mononobenomoriya who was originally against Buddhism and, in 588, the construction of Japan's first full-scale Buddhist temple known as Asukadera Temple began. In 596, the construction of a pagoda and a part of a Buddhist monastery was completed and in 609, the Sakyamuni main Buddha statue, which was fashioned by Kuratsukurinotori, an artisan with Korean roots, was completed. The building of Asukadera Temple marked the beginning of construction of many temples in the rest of Japan and the proliferation of Buddhist culture. In order to construct Asukadera Temple, the know-how to build a structure with a tiled roof that had not been unavailable in Japan then was required. Historical documents state that a number of technicians and artisans that included four potters who specialized in making roof tiles were dispatched to Japan from Baekje. This record is easily confirmed by the round eave tiles of Asukadera Temple, which resemble those with lotus flowers, the most representative design of Baekje. There are two types of lotus designs used for the round eave tiles of Asukadera Temple in expressing warped petals. One type uses a small papilla ornament to decorate the ends of the petals (Plate 7: 3~8) while the other type make the ends of the petals spitted (Plate 7: 1, 2). Japanese academics distinguish them by calling the former group the “star group” and the latter the “flower group.” Recent studies revealed that these two groups share common features with those of Baekje round eave tiles with respect to not only decorative designs but also production technique (Hanatani 1993). The so-called “star group” round eave tile was made by a circular antefix made on the wheel and round tile that has a lip on the butt end to receiving a pipe joint. The “flower group” round eave tile was made by circular antefix, the back of which was flat because it was not made on the wheel, and round tile, which was made by combining the flat back side of the tile. This production technique was always used, though the designs of roof tiles were changed. The production technique of tiles and the technique to circular antefixes with round tiles were similar to those used to make round eave tiles in Baekje. Therefore, the “star group” and “flower group” tiles were created by two different groups of Baekje artisans skilled in different production techniques. They began producing roof tiles on the islands of Japan. ( Plate 7 ) Baekje-style round eave tiles excavated in Japan (Hanatani 1993) 1, 2, 3, 5 Asukadera Temple 4, 6 Toyuradera Temple 7 Sada site, Gose 8 Ishigami site One important task for academics engaged in the study of early roof tiles in Japan is to find out how these two groups of tile makers from Baekje exerted influence on the construction of early Buddhist temples in Japan. For example, it turned out that the “star group” tile makers from Baekje, who were involved in the construction of Asukadera Temple, also produced the roof tiles for the Toyuradera Temple as well as the Wakakusa-garan area of the Horyuji Temple where the wooden statue of the Buddhist Goddess of Mercy of Baekje is housed. More research on roof tile production techniques in ancient Korea and Japan should soon lead to a greater understanding of the roles that Baekje tile makers played. 02 BAEKJE ARISTOCRATS LIVING IN JAPAN IN EXILE In 660, the kingdom of Baekje fell to the allied armies of Silla and Tang. There was soon a movement to restore Baekje under the leadership of Prince Pung-jang, son of King Uija (641-660), who had been staying in Japan. Prince Pung-jang returned from Japan after the fall of Baekje, but the restoration movement failed. Many Baekje people then came to Japan as exiles. Prince Seong-wang (son of King Uija) and his family were at the center of these refugees, and their last name was Kudara-ou (meaning Baekje Wang family). Their settlement in Nanba, Osaka was established as Kudara (Baekje) County before 715. Kudara County was smaller than the other counties divided into several areas named “East side,” “West side,” and “South side,” which were related to area names of the five-division system of the Baekje Kingdom. Most of the people who lived in these places were Koreans from Baekje. It is also said that there used to be a temple called “Baekjesa.” All these facts suggest that the Kudara County served as a base for Baekje people who came to Japan as refugees. In 1996 and 1997, the Saikudani site in old Kudara County was excavated and valuable artifacts that may provide clues to their situation were discovered together (Osaka Cultural Heritage Association 1999). Numerous artifacts that are mostly from the 7th and 8th centuries were unearthed at the site that is in a small valley. Some were pieces of pottery with inscriptions in Chinese characters of “白濟尼” (Plate 8-2), “白尼寺” and “白尼” (Plate 8-3). As pointed out earlier, it is known that a temple called “Baekjesa” (Baekje Temple) used to be in this region. This theory was given support by a pot discovered at this site with the inscription of “僧寺 (Buddhist Temple)” (Plate 8-1). These calligraphic inscriptions strongly suggest that there were the Baekje Temple and another temple for Buddhist nuns called “白濟尼寺” in Kudara County. Remains of two temples have not yet been discovered. When such relics are found, we will be able to glimpse into the religious life of Baekje people who lived in exile in Japan. ( Plate 8 ) The pottery with inscriptions in Chinese characters excavated in Saikusyo site 1 “僧寺” 2 “百濟尼” 3 “百尼” Another important accomplishment of the excavation of the Saikudani site was the discovery of copper coins first minted in 708 called wadokaijin. The excavation team reported that it was highly likely that wadokaijin were minted in the vicinity of the excavation site. However, the problem is that no historical documents make reference to coinage facilities in the area. On the other hand, the excavation report suggested two possibilities: coins were minted in private or coinage was permitted by the Japanese government due to the close relationship between the family of Kudara-ou and the Japanese Emperor. Currently, it is difficult to judge which theory is correct. Whatever the case may be, there were technicians capable of minting coins independently under the Baekje Wang family. By the 8th century, descendants of the family of Kudara-ou led active lives in the northern part of Japan in particular. The most famous member of the Kudara-ou family in Japan was Gyeongbok, a great-grandson of Prince Seon-gwang. He lived in the northeast region for a long time. In 749, he made the first-ever discovery of gold in Japan. He donated the gold he found to Todaiji Temple to be used for the gilt on the Vairocana Buddha statue then being fashioned. By this time, the Kudara-ou family had gained influence in Katano, Osaka and they built a Baekje Temple there too. In addition to assuming active roles in Japanese society, the descendants of Baekje kept their culture alive. This paper approached the relationship of Japan and Baekje from archaeological viewpoints, focusing on only a few issues. In addition to questions raised in this paper, many issues remain to be solved to better understand the influence of Baekje culture on Japanese society. Some of the compelling evidence of Baekje influence are the strong resemblance between the square-shaped, moat-surrounded tombs that have been discovered on the west coast of Chungcheongnam-do Province in Korea and those from the Yayoi period in Japan; the relationship between the stone chambers in Neungsan-ri, Korea and the mounded tombs built at the end of the Kofun Period in Japan; and the Baekje-style ancient mountain fortresses scattered in western Japan. The academics in Korea and Japan should exchange findings of new excavations and conduct joint researches to prove definitively whether these and other remains of the past are attributable to Baekje influence.

Situated in Unjeong-dong, Gangneung-si, Gangwon-do Province, Seongyojang is an exceptionally unusual Korean traditional house. It is an excellent example of vernacular architecture. On its premises stand nine separate residential buildings that have a combined total perimeter of 102 kans (a kan is the traditional unit of area surrounded by four columns). A typical traditional Korean house has only one drawing room, called sarangchae. Seongyojang has five sarangchaes with different names and functions: Yeolhwadang, Dongbyeoldang, Seobyeoldang, Jageunsarang, and Hwallaejeong. Seongyojang does not follow the unique characteristics of the region as seen in the functional layout of space and the composition of space2 (Plate 1). In order to fully appreciate the architectural originality of Seongyojang, we must look at the original architectural purposes of its previous owners rather than Seongyojang as it is today. However, there are no known records on the concepts and the intentions of the original owners, requiring us to delve into the history of the family and the construction of Seongyojang. ( Plate 1 ) Ground plan of the main house of Seongyojang 1 Palgakjeong (pavillion) 2 Huwon (rear garden) 3 Yeolhwadang 4 Seobyeoldang 5 Duian (backyard) 6 Banchim (storage room) 7 Golbang (walk-in closet) 8 Geonneonbang 9 Araetbang 10 Anchae 11 Daecheong 12 Dongbyeoldang 13 Shrine (Ojaedang) 14 room 15 Jageunsarangchae 16 Yeonjidang 17 Bedroom 18 Chanbang (pantry) 19 Inner gate 20 Servants’ quarters 21 Maebang (mill) 22 Kitchen 23 Storeroom 24 Medical doctor 25 Teacher 26 Guest room 27 Carpenter 28 Butler 29 Paperhanger 30 Plasterer 31 Main Gate This research seeks to discover the architectural purposes of Seongyojang, to analyze the concept of planning and design elements to be employed for such purposes, and to evaluate the meaning and the value of the architectural purposes. In other words, this research is intended to trace every process of architectural planning from the constructional objectives of Seongyojang to the selections of design elements for the complex building. By doing so, the author will derive the architectural theory of Seongyojang. The foundation of Seongyojang was laid in 1756. It underwent many alterations over the next 200 years, gradually expanded and altered until it was configured as we see it today. The last alteration was made in the 1960s to anchae or the main house. Therefore, analyzing the architectural composition at any specific time in isolation is clearly inadequate in ascertaining the whole planning process of Seongyojang. This research focus on collective relationship not only to address the architectural composition at any one point but also to identify all the changes made through a long period of time. With this methodology, all the planning concepts that were applied to Seongyojang will surface.3 II YI FAMILY TREE, THE OWNER, AND THE HISTORY OF SEONGYOJANG 01 THE GENEALOGY OF YI CLAN (SEE TABLE 1) ( Table 1 ) The Genealogy of Yi Clan The first owner of the Seongyojang manor was Yi Nae-beon (1693~1781), the 11th descendant of Prince Hyonyeong who was the second eldest brother of King Sejong (1418~1450). Yi Ju-hwa (1635~1708), Yi Nae-beon’s father, was from a well-established aristocratic family in Chungju.4 Yi Ju-ha had three wives. The first wife came from the Nam family of Uiryeong and the second wife, the Jeong family of Gyeongju. The tombs of the first and the second wives are in Chungju and Jincheon, respectively. The birth mother of Yi Nae-beon was the third wife from the Gwon family of Andong. It is said that Yi Nae-beon’s mother moved to Gangneung with her son because her parents had some connections there.5 Yi Nae-beon and his mother settled down in Jeo-dong near Gyeongpoho Lake. In the 1760s, they moved to Baedari-gol where Seongyojang is located.6 Yi Nae-beon had no son, so he adopted Si-chun (1736~1785), a son of his half-brother Jung-beon, as his own son to succeed the family. Yi Si-chun died only four years after he became the head of Seongyojang. He had three sons, Hu, Seung-jo, and Hang-jo. The eldest son, Yi Hu, became the third generation head of Seongyojang at the age of only 13.7 Yi Hu took the state examination four times.8 He never managed to pass and finally gave up trying.9 Yi Hu had two sons and three daughters, and he took care of the families of his two younger brothers who died before him. Unlike Yi Hu, who spent his entire life as literati in seclusion without any government post, his two sons, Yong-gu (1798~1837) and Bong-gu (1802~1885) passed the state examination and were appointed to government posts. Yi Bong-gu, the younger one, became the magistrate of Cheongan in 1850 and then was promoted to county head of Tongcheon in 1853. He was highly respected for his good administration and it was he who made the Yi family of Seongyojang widely known and represented it extremely well.10 Yi Yong-gu, the fourth generation head of Seongyojang, died at the young age of 39. He was survived by two sons, Hoe-suk, 14 years old and Hoe-won, 7 years old. Like their father, the fifth generation Hoe-suk (1823~1876) and Hoe-won (1830~1909) passed the state examination and were appointed to government posts. Yi Hoe-suk became the fifth generation head of Seongyojang and he served as a magistrate. His younger brother, Hoe-won, was adopted by Yi Bong-gu (younger brother of his birth father) and grew up in Seoul and served at a government office for a long time.11 The social life of the Yi family became very active after Hoe-suk and Hoe-won obtained their government posts while the head house of Seongyojang had difficulty to succeed family lineage because few children were born to the fourth and later generations. As in the first generation, the house of the eldest brother and the younger brother exchanged sons by adoption to preserve the family lineage.12 Yi Hoe-suk had no son, so Yi Hoe-won’s eldest son, Yi Geun-u (1877~1938) who came down to Gangneung with his father and lived in Seongyojang, became the sixth generation head of Seongyojang. Yi Geun-u was very sociable. He invited many literati well known for poetry, calligraphy, and painting to Seongyojang13 and actively corresponded with literati of every town. Since Yi Geun-u, the sixth generation head, the Yi family of Seongyojang thrived and had many children. It is said that there were many houses for the offspring of the Yi family in the vicinity of Seongyojang.14 02 REMODELING OF SEONGYOJANG If Yi Nae-beon is the original builder of Seongyojang, Yi Gang-baek, the present head of Seongyojang, is the ninth generation. The heads of Seongyojang have steadily expanded and remodeled Seongyojang. Major architectural changes were made three times. The first change was made by Yi Nae-beon, the first generation; the second change was made after the third generation; and the third change was made during the sixth generation by Yi Geun-u. A great deal of research was done to ascertain the history of construction of Seongyojang and to trace the changes made to the house. Historical documentation at Seongyojang including the records on land transactions, papers on the survey of households, national census records, reports of state examinations taken by Yi family members, doctrines given to descendants, and posthumous manuscripts were studied and the genealogy was examined. Yi family members were interviewed as well. According to the findings, the construction and remodeling of Seongyojang occurred as shown in Table 2. ( Table 2 ) Chronology of Construction of Seongyojang Building Year Documentation Construction of a new house Survey on household Census? 1756 • Records on land transactions Construction of small housing units Survey on household 1795~1798 • Records on land transactions Construction of Yeolhwadang 1815 • Posthumous manuscripts by Yi Hu Creation of Lotus Pond and Hwallaejeong 1816 • Documents on Hwallaejeong Extension of anchae 1853 • Documents on the remodeling of anchae Re-construction of Banghaejeong 1859 • Documents on the re-construction of Banghaejeong• Records on land transactions Extension of Hwallaejeong 1906 • Documents on the remodeling of Hwallaejeong Re-construction of Banghaejeong 1924 • Documents on the remodeling of Banghaejeong Construction of Dongbyeoldang Changes made to anchae circa 1920 Extension of the kitchen in anchae 1960s 03 ARCHITECTURAL PURPOSES OF THE HEADS OF SEONGYOJANG Over a period of some two hundred years, Seongyojang was successively expanded and remodeled. Each change was made for different reasons and under different conditions. The author was able to ascertain the architectural purposes through the years by determining the living environment and achievements of the heads of the house from the first generation (Yi Nae-beon), after the third generation, and the sixth generation (Yi Geun-u). 1) Establishment of the House by Yi Nae-beon Yi Nae-beon and his mother settled down in Jeo-dong near Gyeongpodae Beach. They accumulated wealth by purchasing land step by step. The Yi family needed a new residence as they had prospered and become established in Jeo-dong. They decided to build a new home in today’s Baedari. The Baedari area was reportedly the home of the Jo family clan of Changnyeong. Before then, it was where the Bak family clan of Gang-neung had lived. The Yi family of Yi Nae-beon bought land in Baedari, built a house, and set roots there.15 If not for geomantic reasons, Yi Nae-beon chose Baedari-gol because it was a strategic location for economic purposes and would, therefore, bring him wealth. At that time, the waters of Gyeongpoho Lake extended to the front of Hwallaejeong, one of the drawing rooms of Seongyojang, and people crossed a bridge from a ferry there to get to the mainland. The location commanded a sweeping view of the wide fields and rice paddies all over the valleys in the area, making it easy to manage them, and it was an important junction of transportation. Seongyojang was originally constructed as a typical upper-class house, with anchae as its main house. The original house built by Yi Nae-beon was reportedly in the shape of □, the usual configuration of upper class houses in Gangneung. 2) Residence of Yi Hu for Extended Family Yi Hu, pen named Oeun, made the basic foundational preparations for expansion of Seongyojang as we see it today. He transformed what was an average upper-class residence into a huge manor from both economic and architectural viewpoints. Yi Hu’s purposes in expansion of Seongyojang can be gleaned from the history of his family. Yi Hu’s father passed away when he was only 13. Even though he became the head of Seongyojang at such a young age, Yi Hu raised his two younger brothers, Seung-jo and Hang-jo. Both of them would eventually die young, leaving surviving sons, and Yi Hu raised the young nephews. Gangneung still was a strange place for the Yi Hu family and his young nephews had no other relatives to depend on in the area. Yi Hu himself said that he worked hard and lived frugally his entire life to accumulate wealth and survive in Gangneung in the face of pressures from the local gentry and to financially support the families of his younger brothers. Yi Hu had his nephews live in the same house together with him and his own two sons. He extended the annex of the main house for the family of Seung-jo and built the Yeolhwadang house for the family of Hang-jo.16 The name Yeolhwadang means “listening to stories of relatives tenderly and affectionately,”17 suggesting that Yi Hu longed for happiness and love in his entire extended family. Although the construction of Seobyeoldang (meaning a separate house to the west) was completed after Yi Hu died, it seems that Yi Hu himself had planned the construction. Seobyeongdang was a kind of library and educational institute for children of the house. No other wealthy family built such a huge private educational facility within the premises like Seongyojang and it bespeaks a great deal of the importance Yi Hu attached to the extended family at the time. All members of the extended family lived together under the same roof. 3) Construction of Huge Manor by Yi Geun-U The fifth generation Yi Hoe-suk and the sixth generation Yi Geun-u began to make marked changes to Seongyojang. The Yi family of Seongyojang was able to hold its own in the face of the local gentry of Gangneung by accumulating wealth through effective management of agricultural business and emerged as the wealthiest family in Gangwon-do Province. Since the fourth generation of Yi Bong-gu, men from Seongyojang succeeded in advancing into central government posts and the Yi family could extend its social activities nationwide. As more and more children were born in Seongyojang, it became impossible for the entire extended family to live together under one roof. Houses were built outside Seongyojang and the family was dispersed. Yeolhwadang was made as lodgings for visitors from all over the country and Seongyojang became a playground for famous people from all over the country rather than merely a residence for a big family. The architectural changes made at this time are noteworthy also. Hwallaejeong was remodeled as it is today. Some part of the main house was removed and today’s Dongbyeoldang was built in its place. In front of the servants’ quarters, houses were built for second and third wives. In addition to expanding Seongyojang, these three buildings changed the entire character of the village of Baedari-gol. With the construction of houses for second and third wives and more farmers’ huts within Baedari-gol, the premises of Seongyojang effectively included the entire village of Baedari-gol. Previously, they had been enclosed by the servants’ quarters. Hwallaejeong was a building on the edge of the new, expanded premises. The Yi family also expanded Banghaejeong, a villa used as country house. A huge pine-tree forest was developed in the vicinity of Banghaejeong and the forest was named Yigawon, meaning ‘Garden of the Yi family.’ The border of Yigawon was Okjamam of Gyeongpoho Lake. Now, the conceptual boundary of Seongyojang expanded to the north of Gyeongpoho Lake from Baedari-gol. III CHANGES IN ARCHITECTURAL COMPOSITION The entire two-hundred-year history of construction of Seongyojang over nine generations can be divided into three important periods as seen in Table 3 and Plate 2. ( Table 3 ) Phases of Construction of Seongyojang Construction period Head of Seongyojang Buildings Phase 1 1st generation Yi Nae-beon ~ 2nd generation Yi Si-chun • □-shaped building newly constructed Phase 2 3rd generation Yi Hu ~ 4th generation Yi Yong-gu • Small houses newly constructed• Yeolhwadang newly constructed• Lotus Pond and Hwallaejeong newly constructed• Seobyeoldang newly constructed• Banghaejeong newly constructed Phase 3 5th generation Yi Hoe-suk ~ 6th generation Yi Geun-u • Anchae remodeled• Hwallaejeong expanded• Dongbyeoldang newly constructed• Houses for concubines newly constructed• Banghaejeong remodeled ( Plate 2 ) Changes in Architectural Composition 1 Phase 1 (Yi Nae-beon) 2 Phase 2 (Yi Hu) 3 Phase 3 (Yi Geun-u) 4 Present 01 FIRST PHASE (1756~1785) During the first phase, the foundation of Seongyojang was laid. The main house as we see it today is known to have been a part of the original building.18 Records state that when Dongbyeoldang was constructed in 1920, part of a □-shaped building was removed, evidencing that Seongyojang was indeed originally of such a shape. In the mid eighteenth century, the Bak family clan from Gangneung had long been established in Baedari-gol and the premises of Seongyojang did not likely extend beyond the boundary of the □-shaped house. The anchae at present has several wings, including the master’s room, a main floored hall called daecheong, lady’s room called anbang, and kitchen. A room at the rear of the anbang and master’s room was used as a closet or storage room. The floor plan before the construction of Dongbyeoldang shows that storage rooms were connected down to the kitchen.19 As seen in Plate 3, the layout of the main house was typical of an upper-class house in Gangneung at the time. This shows that the Yi family, who had been living there for only twenty years by then, followed the norms of upper class residences in the area. ( Plate 3 ) Upper Class Residences in Gangneung A House of Choe Dae-seok 1 Stable 2 Anbang 3 Kitchen 4 Sangbang (room) 5 Tteularaetbang (room) 6 Gotgan (storeroom) 7 Haengnangbang (servants' room) 8 Dwiju (rice chest) B Haeunjeong and House of Sim Sang-jin 1 Kitchen 2 Room 3 Storeroom 02 SECOND PHASE (1786~1843) Yi Hu, the third generation head of Seongyojang, was the most active of all in construction of the buildings on Seongyojang. He constructed Yeolhwadang (which represents Seongyojang at present), the pond, and Hwallaejeong. In the second phase, new buildings were constructed and the boundary of Seongyojang was gradually extended. Expansion continued into the fourth generation of Yi Yong-gu and Bong-gu. Seobyeoldang was built on the main premises of Seongyojang and Banghaejeong, a villa, was built near Gyeongpodae Beach.20 During the third generation, three brothers lived together in Seongyojang and during the fourth generation, two brothers lived together there. Compared with preceding generations, the family rapidly increased in size in the second phase. In order to accommodate the new family members, more buildings were constructed on the premises of Seongyojang including Yeolhwadang and other small houses. Yeolhwadang was later used as a study and Yi Hu used it for the education of the children of his two deceased younger brothers. The inscription on the tombstone of Yi Hu states that brothers, nephews, and servants lived together, suggesting that small houses for the grown-up nephews of Yi Yong-gu were built near the main house of Yi Hu.21 Oebyeoldang, a small separate house outside the main house, is presumed to have been used by the nephews of Yi Hu. The sixth generation of Yi Geun-u is said to have lived here until it was separated from the head family of Seongyojang.22 Seobyeoldang and Yeonjidang located between Yeolhwadang and the main house are presumed to have been constructed in the second phase. Given the function of the space, Seobyeoldang belonged to the area of Yeolhwadang while Yeonjidang was used as servants’ quarters of the main house. The madang, a courtyard surrounded by buildings, enclosed by — shaped Seobyeongdang and L-shaped Yeonjidang formed a space between Yeolhwadang and the main house, separating the two areas from each other. The long servants’ quarters were arranged in a line from Yeolhwadang and the main house was reportedly built in this phase.23 With the lotus pond and Hwallaejeong having been created at the entrance of Baedari-gol, the premises of Seongyojang, an inner village previously, came out to the front of the area. 03 THE THIRD PHASE (1844~1938) (SEE PLATE 4) ( Plate 4 ) Plot plan of Seongyojang in the 1930s (reproduced based on historical evidence) 1 Dongjin Academy 2 Jageunsarang 3 Yeolhwadang 4 Seobyeoldang 5 Yeonjidang 6 Anchae 7 Dongbyeoldang 8 Shrine 9 Haengnang 10 Hwallaejeong 11 Houses for concubines 12 Gaeksarang Yi Hoe-suk, the fifth generation head of Seongyojang, served in a govern-ment post for a long period of time and died without a son. Yi Hoe-suk’s death marked the beginning of a hiatus in construction at Seongyojang. Construction resumed when Yi Geun-u, the sixth generation, became the head of Seongyojang. Unlike the second phase, the function and the features of buildings were changed by remodeling or redecorating in the third phase. During this period, buildings of the Yi family of Seongyojang were constructed at several locations in Baedari-gol. The premises of Seongyojang then covered the entire valley of Baedari-gol. Yi Hoe-won returned to Seongyojang in 1894 after a long time in a government post in Seoul. Seongyojang was given public space for visitors. Whereas Seongyojang had exclusively been the residence of a family, it now had both a private area and a guest area.24 It is reported that when Yi Geun-u was alive, many guests used to stay at Seongyojang for as short as one week to as long as several months.25 Depending on the social status of the guests, they were put up in Yeolhwadang, a small drawing room, or in a drawing room at the entrance. The guests entered through a tall gate in the servants’ quarters to the main house. The guest area was clearly separated from the residential area for the family. The area between Seobyeoldang and Yeonjidang was partitioned by a brick wall. The main house and servants’ quarters were also separated by a wall with a small door. Entering through the tall gate, guests encountered a smaller gate to Yeonjidang that leads to the madang of the main house. The long servants’ quarters had more than 20 kans and were used for three purposes. The servants’ quarters in front of the main house were used as rooms for servants and for storage of household goods for the main house. The area between the main house and servants’ quarters was partitioned by walls and there was a small door in the wall through which servants ran errands for guests and their master. The servants’ quarters in front of Yeonjidang and Yeolhwadang were residences for artisans working for Seongyojang. It is said that there were carpenters, plasterers, tutors for children, pharmacists, Oriental medical doctors, paperhangers, a blacksmith, and gold and silversmiths, all of whom worked for Seongyojang.26 These servants stayed alone at Seongyojang, separated from their families. IV PURPOSE OF ARCHITECTURE AND CONCEPT OF PLANNING 01 PURPOSE OF ARCHITECTURE: ESTABLISHING A GRAND MANOR The architectural purposes of the head of Seongyojang varied from generation to generation. At the initial stage, the goal was to construct a house for a family to live in and lay down roots in a strange, new place. Then, the goals changed. Seongyojang was expanded into a manor where an extended family including families of brothers lived all together. Finally, the goal became to make Seongyojang a center of social activity. With the flourishing of the Yi family, a great manor came into being. A manor can be said to be a socio-political, economic, and architectural compound. On huge tracts of land, which yielded more than tens of thousand sacks of rice, were hundreds of tenant farmers working for Seongyojang. The grain storage house served as the financial foundation of Seongyojang.27 With such a strong economic foundation, Seongyojang became associated with those in political power and master artisans as well as foreign diplomats. Seongyojang was a socio-political center of nationwide importance.28 To serve public social activities, Seongyojang was divided into two separate areas. The inner space of the main house was for the private life of family members of the direct line while the Yeolhwadang area was used to put up guests. At the same time, the residential area, which previously included only the inside of the main house, was extended to cover the entire area of Baedari-gol village. Houses for concubines and tenant farmers’ huts were built outside the main house. Seongyojang at last became a public residence as the regional power center. Yi Geun-u classified his guests and friends into several groups and prepared separate guest areas appropriate for each group. Yeolhwadang was used as an official place to receive the guests of Seongyojang and at the same time a place of large social gatherings. It was an open and public space to serve guests. In contrast, Hwallaejeong was a relatively closed space. Only a few special guests who were very close to Master Yi of Seongyojang were invited to stay in Hwallaejeong. Dongbyeoldang within the main house area was a private space. It served as an ordinary master’s room. Here, members of the family got together to discuss family matters. Besides these three spaces, there was Banghaejeong by Gyeongpoho Lake. It was a villa and was used to put up guests who stayed for long periods of time. 02 SPACES FOR FAMILY AND GUESTS (SEE PLATE 5) ( Plate 5 ) Extension of Seongyojang over the phases 1 Guest area (public area) 2 Semi-public area 3 Family area (private) 4 1st phase areas (groups of buildings) 5 2nd phase area (aggregation of building groups) 6 3rd phase area (construction extended to entire valley of Baedarigol) 7 Guest area (public) When it became a manor, Seongyojang was no longer only a residence. It became an aggregate consisting of a residence for ‘a big family’ and buildings for guests. The main house and Dongbyeoldang in the east can be understood to be a private area for members of the family while the Yeolhwadang area in the west is for guests. Dongbyeoldang is the drawing room for the Yi family and Yeolhwadang is the official residence for the head of the manor. Dongbyeoldang and the main house comprise a perfect residence and Yeolhwadang is effectively an independent house. Between the two areas is the Seobyeoldang complex. It includes the Seobyeoldang, the main building, and an auxiliary building, Yeonjidang that encloses Seobyeoldang. Seobyeoldang was used as a library and study room where the children of the family were educated. It is between Yeolhwadang for the adult men and the anchae. These three buildings were reportedly at one time connected by a secret wood floored corridor. The respective locations of these three buildings and their connections show the composition and relationships of the members of the Yi family. Seobyeoldang was the inner part of Seong-yojang and was the place where members of the family connected with each other. Yeonjidang separated the family area from the guest area. Most of the residents of Yeonjidang were maids who watched the guests to protect the children of Seongyojang. Yeolhwadang and the main house are clearly demarcated by the L-shaped layout of Yeonjidang. In the front, Yeonjidang separates the two areas of Yeolhwadang and the main house, and inside, Seobyeoldang connects the two areas. This is an exquisite way to manage space and make it practical. The private and public areas are connected by a long line of servants’ quarters. The servants’ quarters are one long facility consisting of many one-unit rooms and they serve two functions. The part belonging to the main house is used as storage for household goods and the other part belonging to Yeolhwadang is used as drawing rooms for non-family members including tutors, medical doctors, technicians, and guests. There are two gates attached to the servants’ quarters. An ordinary gate in the east to the main house is used for everyday passage by members of the family and is closed to outsiders while the tall gate in the west to Yeolhwadang is open and used by guests.29 If we set aside the guest area in the west, which starts with Seobyeol-dang, we can see the composition of the private area surrounding the main house much more clearly. Between Yeonjidang and the main house are two inner gates and the madang between these two gates serves as a secret entrance for the family. Dongbyeoldang, a drawing room, and the ㄷ-shaped main house are connected diagonally to form living quarters. In front of Dongbyeoldang is a madang and a hill is at the back. The family shrine is on the hill close to the drawing room. This composition alone satisfies all needs for ordinary upper-class housing. 03 ENCLOSED EXPANSION OF RESIDENTIAL AREA (SEE PLATE 5) The two areas of Seongyojang in the east and west are not situated parallel to each other. They are not of equal status. The private area is enclosed by guest area. We can understand the characteristics of aggregated composition of Korean architecture through Seongyojang. It is an aggregate of buildings and, at the same time, it is a cumulative record of the construction of Seongyojang. The area consisting of the main house and Dongbyeoldang serves the most fundamental function, that is, a residence for family. We can think of this area as an expanded version of the □-shaped house built by Yi Nae-beon when he settled down in Gangneung. With the new extension of the Yeolhwadang and Seobyeoldang compound, the character of Seongyojang changed. As discussed before, the newly extended area was for the happy living of the families of three brothers. Because the building clusters were connected by a line of servants’ quarters in the front, the physical area of the house was very large. The area of existing living quarters became a part within the whole that was encircled by the servants’ quarters. The center of social gatherings, which used to take place in the drawing room, was moved to newly constructed Yeolhwadang and the nature of the area changed. Another change was made in the sixth generation Yi Geun-u. Within the valley of Baedari-gol, dozens of huts were built for Seongyojang’s tenant farmers. This meant an expansion of the premises of Seongyojang to the entire valley of Seongyojang. The ridges of Baedari-gol valley are the natural boundary to the extended area. Hwallaejeong and Bangji (pond) at the entrance to the village is the artificial boundary connecting the ridges in the front and rear. The existing Seongyojang area enclosed by the lined servants’ quarters becomes a part of the whole, that is, Baedari-gol, and the function of Yeolhwadang as guest area extends to Hwallaejeong. Creation of Banghaejeong and Yigawon in the vicinity of Gyeongpodae Beach meant that the territory of Seongyojang extended beyond the valley of Baedari-gol. Then, the function of Hwallaejeong as a guest space was extended to Banghaejeong. Since it started as an ordinary upper-class house, the premises and lands of Seongyojang were expanded over three phases. It was expanded into a complex residence for a large family, then into a manor covering the whole valley, and further to a vast area including Gyeongpoho Lake. In every expansion phase, a new public building was constructed, thus giving Seongyojang a new function and new place each time. The three new buildings were Yeolhwadang, Hwallaejeong, and Banghaejeong. The expansion of territory and construction of new public buildings tells us about the management or architectural history of Seongyojang. V INTERPRETATION: DIMENSION OF SET AND PLANNING ELEMENTS A large complex systematically constructed inevitably has a structure of sets and order. Because various buildings were constructed and expanded over generations, Seongyojang has a very complex structure but each element is connected with each other in a very systematic and efficient way. The buildings and spaces of Seongyojang form four different dimensions, each of which consists of a set, and the sets are overlapped in a certain order. Unit rooms form one building and individual buildings form a set of buildings. On the premises of Seongyojang, there are more than four building sets. Being interrelated, these groups comprise a larger building group. This big group, in turn, has a relationship with the natural environment and it becomes a complex at a geographical level. If we divide these interrelated complexes in four levels such as individual building, building group, aggregation, and geographical level, the smallest unit of architectural elements in each level can also be named. 01 COLLECTIVE ELEMENT AT INDIVIDUAL BUILDING LEVEL (SEE PLATE 6) ( Plate 6 ) Dimension of Set – Level 1: Composition of Yeolhwadang Rooms 1 Daecheong (main floored room) 2 Numaru (elevated floored room) 3 Rooms There are ten buildings at Seongyojang. Although the shape and floor plan of each building is unique, each building is a group of several rooms. That is, the collective element at the level of a building is a room. Thus, the unit for a building is a room. There are three types of rooms that serve as units: ondol, an under floor heated room, maru, a wooden floored room, and a room with a clay floor. Of course this classification is based on the materials used for the floor. Even so, each category has a unique feature as a space. What is important is the collective relationship of the three types of rooms, rather than the specific feature of the ondol, maru, or clay floored room. For example, the set of <clay floored room+clay floored room> makes a building that is used as a storage space or workplace. The set of <maru+maru> becomes a pavilion. The common set for a traditional Korean home is <ondol+maru>. Many buildings of Seongyojang, especially Yeolhwadang, clearly show the collective feature of an element, <ondol+maru>. Yeolhwadang is a kind of guest building. It consists of four marus, three ondols, and two elevated marus. Traditional upper-class Korean houses have this kind of veranda elevated above the ground. Noteworthy are the three ondols and two elevated marus. The ondols in typical residential houses in Gangwon-do Province are laid out in a ⊞ configuration of four ondols. In Yeolhwadang, one unit of the four-unit ⊞ is an elevated maru instead of an ondol. Thus, the remaining three ondols compose a L-shaped layout, and two elevated marus create a — shape. Next to the L-shaped ondol is a big wooden floored hall and in the front of the ondol is an elevated maru. Accordingly, all three ondols adjoin the maru. If the ⊞-shaped floor plan was used, one room out of the four could not adjoin the maru and it should be a set of <ondol+ondol>. Underneath the elevated floor is a wood-firing hole, thus helping the collective element of <ondol+maru> function more effectively. An ondol and maru cannot function independently in a building. Each has only a neutral function. The function of a space is defined by the collective relationship of the ondol and maru. At the level of a building, the <ondol+maru> is the most basic element of a collective relationship. 02 BUILDING GROUPS AND OUTSIDE SPACE (SEE PLATE 7) ( Plate 7 ) Dimension of Set – Level 2: Composition of Main House Group 1 Dongbyeoldang 2 Kitchen 3 Shrine 4 Anchae 5 Inner servants’ quarters There are four groups of buildings on Seongyo-jang: the main house group in the east, Yeolhwadang group in the west, Seobyeoldang group in between the main house and Yeolhwadang groups, and the Hwallaejeong group consisting of Hwallaejeong and Lotus Pond. There used to be a number of building groups such as the Oebyeoldang group, group of buildings for concubines, Dongjin Academy group, and farmers’ hut group outside the long servants’ quarters of Seongyo-jang. These no longer existing building groups are not addressed in the discussion here. The main house group is the oldest of Seongyojang. This area we see today is the result of changes and expansions made to the □-shaped house common in the Gangneung area. The main structure in the north and rooms in the west are remains of the original □-shaped house. L-shaped Dongbyeoldang in the east was added in the 1930s and the kitchen in the south was added in the 1970s. Later, the long servants’ quarters were attached to the front and the family shrine was added to the side to form the main house area. When individual buildings were clustered to form a group of buildings, one element emerges as the most important medium indispensable to the group. It is the madang. The four buildings: main house, rooms in the west, kitchen, and Dongbyeoldang form a perfect group of buildings as they encircle the madang. When Dongbyeoldang was extended, the outside space was extended into two areas. The square madang was maintained. The long outside space was configured lengthwise on one level of lowered ground, in consideration of L-shaped Dongbyeoldang. This newly created space was placed in parallel with the above-mentioned inner madang enclosed by the servants’ quarters. Thus, you pass through the madang in the rear of the servant’s quarters and then the madang of Dongbyeoldang to reach the corner of the inner madang of the main house. Three outside spaces, madang in the rear of servants’ quarters, madang of Dongbyeoldang, and madang of main house are continuous with one space leading to another. This special continuity is a powerful collective element, which aggregates the five buildings of the main house group. 03 LEVEL OF AGGREGATION AND DATUM ELEMENT (SEE PLATE 8) ( Plate 8 ) Dimension of Set – Level 3: Composition of Main Quarters 1 Yeolhwadang group 2 Seobyeoldang group 3 Anchae group Of the four building groups of Seongyojang, the main house group, Seobyeoldang group, and Yeolhwadang group form an aggregation and are thus separated from the Hwallaejeong group. This aggregation of three groups of buildings can be regarded as the ‘main quarters’ and are commonly called ‘Seongyojang.’ Each of the three building groups of the main quarters has its own unique character. The main quarters are the residence of the direct descendants of the family. As such, it must be a closed area. In contrast, the Yeolhwadang group is an area open to numerous dependents of Seongyojang and its guests. The Seobyeoldang group, on the other hand, is a half public and half private space where a live-in seamstress and nannies live, taking care of children and doing other work for the main house group and Yeolhwadang group. The degree of openness is in the decreasing order of Yeolhwadang group, Seobyeoldang group, and main house group. The largest madang is in the Yeolhwadang group as it was intended to accommodate large numbers of people. The smallest is in the main house group. The style of building is the most magnificent and unique at Yeolhwadang as it is a space to serve guests. The line of 23 one-room servants’ quarters connects the three different building groups with one ‘main quarters.’ Because the servants’ quarters are in a line and serve as a wall, the madang of the Yeolhwadang group could be separated from the outside and the space of main quarters was more closed. Since the gate for guests is in front of the Seobyeoldang group, the route to the entrance to the madang of Yeolhwadang is indirect. The servants’ quarters lean slightly on the Yeolhwadang side, making the perspective view of Yeolhwadang look more spectacular. Spaces each of which is of a different size and function are divided but at the same time integrated into the main quarters by the line of servants’ quarters. Building groups of different sizes and shapes are unified into one collective form by the line of servants’ quarters. With the use of the datum30 called the ‘lined servants’ quarters’ in Seongyojang, Seongyojang could become a systematic aggregate. 04 GEOGRAPHIC LEVEL AND HWALLAEJEONG (SEE PLATE 9) ( Plate 9 ) Dimension of Set – Level 4: Main quarters group Two groups of Seongyojang—the main quarters and Hwallaejeong group—have very different functions and sizes in comparison. The main quarters consisting of three building groups with ten buildings are mostly for everyday household tasks and serving guests. In contrast, the Hwallaejeong group consists of only one building and a pond. The building is small and is used by only a few people. Nevertheless, the two groups are on equal footing because they are related with the topography of the whole valley of Baedarigol. However big it is and however important its function may be, the main quarters are but a part of the whole of Baedari-gol valley. The Hwallaejeong group, on the other hand, serves as an important structure that connects the hills in the front and at the rear. Due to the location and existence of Hwallaejeong, Baedari-gol valley constitutes one single architectural property. In this regard, Hwallaejeong is simultaneously one building, one building group, and aggregated group. That is, Hwallaejeong is a collective element with a complex character, which exists in all the levels: building-building group-aggregated groups. Hwallaejeong is in a complex form where one building is at right angles to another building. It is as if the spirit of the mountain in the rear were transferred to the mountain in front. Even the form is designed to be part of the topography. If the lined servants’ quarters are a powerful datum at the aggregated group level, Hwallaejeong is an indispensable datum at the geographic level. VI CONCLUSION Throughout the two-hundred-year history of its construction, the architectural goal of Seongyojang, which started as an ordinary upper-class residence, was to become a ‘great manor’ beyond geographical limitation. In order to achieve that goal, the architects of Seongyojang adopted the planning concept of ‘repeated expansion of residential area.’ The outside area as a center of activities and ‘collective elements as datum’ such as lined servants’ quarters or Hwallaejeong were selected as tools to be used for materialization of that concept. The earliest part of Seongyojang was built for family members in the direct line and was expanded to accommodate three groups of buildings as it came to serve dual functions: residence for family members and lodgings for guests. The three building groups became a set by the powerful datum element of the lined servants’ quarters and formed a single complex of main quarters. The territory of Seongyojang does not end here. Through the datum element of Hwallaejeong, Seongyojang’s territory extends to the entire valley of Baedari-gol below the mountain ridges. The extension of Seongyojang’s territory over the three phases then moves further to cover the Gyeongpodae Beach area through villas, Banghaejeong, and Yigawon. The processes of extension of territory and selection of design elements can be understood more clearly through collective interpretation because design elements are closely related at the aggregate level. At the individual building level, the combination of individual rooms, that is, relationship of room units <ondol+maru>, is the collective element. At the level of the building group, an outside space element of the <madang> is adopted as a collective element. At the level of aggregation, the unit building called the lined servants’ quarters is the collective element, and at geographical level, the Hwallaejeong group of buildings serves as the collective element. If we can say that the processes of setting architectural goals and concepts of planning, selecting design elements appropriate for those goals and concepts, and gathering together the selected design elements, and the collective relationship principle applied to each process constitute an architectural theory, the architectural theory of Seongyojang can be summarized as a ‘collective composition for repeated extension of territory.’

The shards of Goryeo celadon excavated from the Hakata and Kamakura sites together with Chinese pottery have been of great interest since World War II.1 The shards, albeit shattered pieces, could easily be discerned because of the unique embellishment of inlaid decorative designs. In 1953, a thesis on these Goryeo ceramics by Koyama Fujio was published.2 However, this thesis was based on extremely limited resource due to the lack of available information on porcelain kilns in Japan at that time. For example, white porcelain that had been produced in China was often mistaken for Goryeo’s and the distribution of Goryeo ceramics and their development over each period of time could not easily be mapped as a result. For the past twenty years, however, archaeological excavations and studies on the Middle Ages and modern times have so advanced that the public interest in foreign ceramics excavated from historic sites has increased considerably. Most of the ceramics that have been excavated are, of course, Chinese and the excavation case of Korean ceramics is comparatively far fewer than those of Chinese. However, the number of reports on newly excavated Korean ceramics has been steadily increasing. In 1983, Nishitani Tadashi published a thesis on chronological recording of the distribution of Korean ceramics and its users based on the comprehensive study on the Korean ceramics excavated in Kyushu and Okinawa.3 In 1984, the Fifth Conference on Trade Ceramics was held under the theme of ‘Goryeo and Joseon Ceramics Excavated in Japan,’ giving us a better understanding of this area of study. Since then, a number of theses on this subject have been published.4 The purposes of this paper are to overview these preceding studies on Goryeo ceramics excavated in Japan (or preserved in Japan), to summarize several issues involved in the study results, and to present my views on the issues. II Needless to say, Goryeo celadon represents all the ceramics of the Goryeo period. Celadon is a type of porcelain unique to Asia first developed in now the northern part of Zhejiang Province during the late Han Dynasty. China has produced fine quality celadon unique to each period. Of all the celadon produced in China, those produced at the Yue Kiln during the Tang and Five Dynasties periods and at the kilns in Yaozhou, Ru, and Longquan during the Song period are the most renowned. Goryeo also introduced a unique celadon that is tantamount to the Chinese celadon in color and decoration, so deep and calm and yet rich and warm. It is widely recognized that the Yue kiln in China deeply influenced the creation of Goryeo celadon on the Korean Peninsula. The Yue kiln collectively refers to a group of massive celadon kilns concentrated around Shanglin Lake (today’s Cixi, Zhejiang Province that encompasses the cities of Yuyao and Ningbo as well as Shangyuxian Province). By the Late Han, celadon production in this area had gradually matured and a unique celadon called Ancient Yueware such as Shentinghu and Tianjihu was produced during the periods of Three Kingdoms China, Western Jin, and Eastern Jin. By the late Tang, the Yue kiln had already produced high-quality celadon in a massive volume and part of this fine-quality celadon was exported overseas in concurrence with the celadon trade boom then. Beautiful celadon thinly applied with clear bluish green celadon glaze was called celadon of mysterious color and widely praised by the world. Bowls of a simple shape with a broad circular foot that broadens straight upward to make a wide mouth were typical of Yueware style, and Goryeo produced a similar type of bowl as Yueware. Studies on Goryeo celadon have advanced considerably thanks to important discoveries on the early phase of Korean celadon in recent years. Although there is no concensus on when celadon came into being on the Korean Peninsula, it has been pointed out that the first celadon may date back to before the ninth century, during the late Unified Silla period. Looking at the Goryeo celadon that has been excavated in Japan thus far, no case has been confirmed to be a product from before the mid eleventh century and almost all of them have been discovered along with Chinese ceramics. This implies that the imports of Goryeo celadon into Japan were supplementary to the imports and distribution of Chinese ceramics.5 At historic sites in Japan, imported ceramics dating to the late eighth century have continually been discovered. Most of the ceramics imported into Japan at this time were celadon from the Yue kiln, white porcelain, and yellow glazed pottery produced at the Changsha Kiln. These three types of Chinese ceramics are collectively called ‘Early Trade Ceramics’ as they were the first major ceramics to open an era of earnest ceramic trade. During this early trade ceramics period, there seems no indication that Goryeo celadon was imported into Japan, however. From the late eleventh century through the twelfth century, these early trade ceramics were replaced by white porcelain produced in southern China and began to pour into Japan. Albeit in a small volume, there are some excavation cases of Goryeo celadon that were exported to Japan around this time. Yamamoto Nobuo6 and Morita Tsutomu7 conducted some researches on these Goryeo cases. The celadon pieces dubbed ‘Early Goryeo Celadon’ by Yamamoto Nobuo can be classified into three groups. The first group is refined celadon produced with fine quality clay and glaze and it has two types: one with a haemurigup foot (a flat type of foot-ring with its centre hollowed out) and the other with a yunhyeonggup (wheel-shaped foot-ring). The second group is of bowls that widen towards the mouth and that are decorated with lotus patterns. This group has two types in terms of quality: fine and poor. The third group is mass-produced, poor-quality celadon of which the clay, glaze, and shapes are all poor. Fired in stacks (piled one upon another), this poor-quality celadon has marks of supports on both the interior and the exterior surfaces. Celadon of this third group was recovered in a massive volume from a sunken vessel discovered in 1983 on a seabed off of Wando Island, Jeollanam-do Province.8 It has been confirmed that this type of celadon was produced at kilns in Jinsan-ri, Jeollanam-do Province.9 Yamamoto Nobuo estimated the date of the ‘Early Goryeo Celadon’ that had been excavated from Dazaifu based on the form of Hajiki (earthenware produced in the Tumulus period through the Ancient period) and concluded that both fine and poor quality celadon began to be produced in the late eleventh century and increased in volume well into the twelfth century. The discovery of ‘Early Goryeo Celadon’ is concentrated in the northern part of Kyushu since many pieces were discovered in Dazaifu and Hakata and some in Heiankyo (today’s Kyoto),10 Chikugokokufu (today’s southern Fukuoka-ken Prefecture),11 and Buzenkokufu (area covering eastern Fukuoka Prefecture and the northern part of Oita-ken Prefecture at present)12 as well as Tsushima.13 Nevertheless, the volume is tiny compared with that of Chinese ceramics leading to a speculation that the early Goryeo ceramics were not produced for export. In particular, the excavation of the poor-quality celadon fired in stacks was strictly limited to Dazaifu, Hakata, Chikugokogufu, and Tsushima. Meanwhile, looking at the foreign ceramics imported into Japan from the latter half of the eleventh century to the first half of the twelfth century (when early Goryeo celadon appeared in Japan), white porcelain accounted for the bulk of Chinese ceramics while the percentage of celadon in Chinese imports decreased sharply. Based on this fact, some scholars speculate that increased demand in the Japanese market for celadon led to importation of Goryeo celadon.14 If so, Goryeo celadon should have been intentionally imported into Japan. The characteristics and shipment route of early Goryeo ceramics imported into Japan need to be more closely examined. It is possible, for example, that part of Goryeo celadon that had been imported into Tsushima as a commodity was shipped to Kyushu. There is some difference in approaches between researches in Japan, the consumer and Korea, the producer on the production date of Goryeo celadon excavated from historic sites in Japan. Of the early Goryeo celadon, the Japanese and Korean academics almost agree on the age of the third group of mass-produced, poor-quality celadon.15 However, they disagree on those of the first group and the second group. In particular, the Korean academics believe that the production date of the first group is slightly earlier than that of the celadon pieces excavated in Japan.16 It is thought that the reason for the discrepancy between the dates of production in Korea and of importation into Japan is that the Goryeo celadon was not originally produced for export. If the early Goryeo celadon excavated from historic sites in Japan was not for trade, we need to be very careful in associating the date of artifacts excavated with the actual production date. However, given that all of the early Goryeo celadon excavated in Japan has been discovered together with Chinese porcelain from the latter half of the eleventh century to the twelfth century, it would be reasonable to surmise that the age of artifacts excavated reflects the date of importation and production. Pottery with haemurigup foot-rings is commonly found among early trade porcelain and celadon bowls produced at the Yue kilns in China. Haemurigup means ‘haloed sun’ and is a flat type of foot-ring with its centre hollowed out. Celadon and porcelain with a haemurigup foot-ring was also produced in Goryeo. According to the thesis by Kamei Meitoku on the appearance and disappearance of haemurigup in China, pieces with haemurigup began to be produced in the latter half of the eighth century, flourished in the first half of the ninth century, and disappeared in the latter half of the ninth century.17 Kamei also wrote that Goryeo celadon with haemurigup continued to be produced until the third quarter of the ninth century at the latest given that the original and the imitation were in general produced during the same period and that celadon produced during the Silla period should be the true predecessor of Goryeo celadon with haemurigup.18 The haemurigup on celadon excavated in Japan is not similar to the ‘Chinese-style haemurigup’ of the celadon produced at the Yue kilns. All the Goryeo celadon pieces are Korean-style haemurigup having flat bottom inside instead. This style represents the later haemurigup phase, and Choe Gun confirmed that this type of haemurigup continued to be produced in Korea until the end of the tenth century.19 Some academics have noticed the similarity in shapes between early Goryeo celadon bowls excavated in Japan and porcelain bowls produced in Northern Song China.20 It means that, although there is no doubt that Goryeo bowls with haemurigup were influenced by Chinese celadon produced at the Yue kilns, Chinese ceramics produced during the Northern Song period exerted much greater influence on the shape of bowls with Korean-style haemurigup, which was produced later than the Chinese-style haemurigup. Considering all, I support the opinions of Yamamoto Nobuo and Morita Tsutomu. They date the Goryeo celadon excavated in Japan to the eleventh and twelfth centuries. Many questions remain unanswered regarding the history of Goryeo celadon because of a lack of documentation on the dates.21 The date of birth of Goryeo celadon and the time when it began to be produced in larger volume for practical everyday use also need to be ascertained based on the available evidence. Because relatively little Goryeo celadon has been excavated from historic sites in Japan, it does not offer enough evidence as to when Korean celadon was created and developed. Nonetheless, since there are few excavations outside of kiln sites on the Korean Peninsula, those excavated in Japan deserve attention in the study of Goryeo celadon. III Gradually breaking away from the influence of Chinese celadon since it was first introduced to Goryeo, its potters ultimately developed a unique type of celadon. The glazes on Goryeo celadon became more transparent and beautiful jade-green, and many different decorative techniques, such as incision, carving, openwork, molding, and painting were developed. Good quality Goryeo celadon produced in its heyday was of exquisite beauty and had a dynamic feeling. This high-quality celadon was made for the aristocrats of Goryeo and was produced at kilns in Sadang-ri, Gangjin-gun, Jeollanam-do Province and Yucheon-ri, Buan-gun, Jeollabuk-do Province during the twelfth century. There is one extremely important document that describes the Goryeo celadon in the first half of the twelfth century. It is the Travelogue to Goryeo of a Chinese Envoy, an account of a visit to Goryeo written by Xu Jing in 1123. About his visit to Gaeseong, he wrote, “Goryeo people call the green color used in ceramics ‘kingfisher blue.’ Current techniques have become more sophisticated and the glaze, more beautiful than before. The color resembles the mysterious green of the vessels produced at the Yue and Ru kilns. The lion-shaped incense burners decorated with lotus patterns are most intricately made.” Although the illustration originally included in the Xu Jing’s book is lost, the book clearly proves that Goryeo celadon reached a very high level of refinement and sophistication. The celadon excavated from the Jangneung Royal Tomb of King Injong at Jangdo-myeon, Jangdan-gun, Gyeonggi-do Province (now in the collection of the National Museum of Korea) proves that Goryeo celadon was produced at least as far back as 1146. The transparent tint of green glaze is filled with the tranquility and grace that makes Goryeo celadon unique and we can understand every reason why the color is specifically differentiated as ‘jade-green,’ that is, the color of kingfisher blue. Like the incense burner in the shape of lion as described in Travelogue to Goryeo of a Chinese Envoy, other celadon pieces in the shapes of various animals, plants, and human figures in full or in part were called “sanghyeong-cheongja” or celadon modeled after figures (e.g. animal or plant). There are many pieces of this type of celadon in such shapes and they are considered the most representative exemplars of all Goryeo celadon. These animal or plant-shaped celadon pieces were imported into Japan. The fragments of a lid of an incense burner shaped as a mandarin duck were discovered at the historic site of Dazaifu.22 The feathers are incised exquisitely on the body that is covered with transparent jade-colored glaze and they attest to the technique described in Xu Jing’s Travelogue to Goryeo of a Chinese Envoy. It is shaped to emit incense from the mouth of the mandarin duck and the Museum of Oriental Ceramics Osaka holds the perfect example of this type of celadon. A lion-shaped Goryeo celadon water dropper was excavated from the island of Tsushima, between Kyushu and the Korean Peninsula. It is in the collection of Kamiagata-machi Central Public Institute at present. It is hollow inside, has a big open mouth, and is decorated with an incised mane. The entire body is covered with jade-colored glaze and the big eyes of the lion are painted with iron pigment. This celadon was reportedly excavated from Koshitaka historic site in Kamiagata-machi Prefecture but no other information on the excavation remains, regrettably. A piece of celadon of exactly the same shape and size as this piece was discovered in a sunken vessel off of Dodeok-do Island, Jungdo-myeon, Sinan-gun, Jeollanam-do Province.23 The artifacts from the Sinan seabed include seven pieces of Goryeo celadon including the lion-shaped water dropper mentioned above,24 five pieces decorated with inlaid designs, and one maebyeong bottle decorated with a carved peony scroll. There is a gap between the estimated time of production and the 1320s, when the ship foundered. Chung Yang-mo speculates that Goryeo celadon that had been taken to China earlier was loaded and transported later on this ship.25 In fact, some Goryeo celadon has been excavated in China.26 About as many as 20,000 ceramic pieces were excavated from the Sinan seabed. Of these, only seven were Goryeo celadon (the remaining were Chinese celadon and porcelain). The three Goryeo celadon pieces recovered from the third salvage operation had been packaged in a container deep inside the hold of the ship,27 showing clearly that this Goryeo celadon was handled differently from the other ceramic pieces shipped in massive volume. It is very possible that these Goryeo celadon pieces were ‘antiques’ that had been handed down from one generation to another for some time. Of the pieces found on the Sinan seabed, many resemble those excavated from historic sites in Japan. This suggests that Goryeo celadon, which was widely sought after, was distributed along with Chinese ceramics. Animal-shaped celadon is dated by many to the first half of the twelfth century, as the Travelogue to Goryeo of a Chinese Envoy records. However, given that the lion-shaped celadon water droppers discovered from the Koshitaka historic site in Kamiagata-machi and the Sinan seabed are of somewhat poorer quality and have exaggerated shapes, we can assume that these two pieces were not produced during the same period when the best animal-shaped Goryeo celadon was made. Logically, it appears that the piece excavated from Koshitaka, Kamiagata-machi, Tsushima and the others recovered from the Sinan seabed were produced later, when animal-shaped celadon was in decline. Shards of the legs of animal-shaped Goryeo celadon were also discovered in Kamakura.28 The range of the sites where Goryeo celadon decorated with incised or stamped patterns has been discovered is broader than that of early Goryeo celadon. Celadon dishes decorated with incised designs were excavated in Hakata and Dazaifu in northern Kyushu and Miyaji, Jonan-machi, Kumamoto-ken Prefecture.29 A celadon bowl decorated with an incised lotus design was discovered on Gaja Island of the Tokara Islands (Tokara-retto), Toshima-mura, Kagoshima-ken Prefecture.30 Shards of a celadon bowl with a stamped lotus scroll design were discovered in Heiankyo. Shards that seem to be of a mouth of a celadon incense burner were discovered at the Ichijodani Asakura Family Site, Fukui-shi, Fukui-ken Prefecture.32 Decorated with an incised design of clouds and coated with transparent jade-colored glaze, these shards seem to have been produced when Goryeo celadon reached its apex. As there is a considerable time gap between the period when the piece was produced and the time the relic was kept on the site of excavation, it is thought that the piece had been highly prized as an antique and handed down through the generations to the sixteenth century. Goryeo celadon that was produced after its uniqueness in style had been developed to perfection is more widely distributed. Although there are still relatively few examples, some pieces were held by families for generations. Goryeo celadon must have been highly valued and appreciated by the Japanese for the green color, and was called jade color celadon. IV By the latter half of the Goryeo period, the inlay decorative technique called sanggam had been developed and became the major means of decoration for Goryeo celadon. To make a piece of sanggam inlaid celadon, some designs were incised on the surface of a vessel, the designs were filled with white clay or iron-rich red clay and then the piece was bisque fired. After bisque firing, the surface was coated with celadon glaze. When fired, the white clay remained white while the red turned black on the jade-colored background. Although seldom used, the reverse inlay technique entailed cutting the background away and filling the carved out spaces with white clay, leaving the design in relief. In this case, the design in jade color is shown on a white background. We can find inlaid decoration on Chinese ceramics. However, only on Goryeo celadon, inlay technique developed as a major decoration making the sanggam inlay technique unique to Goryeo. Academics do not agree on the exact origin of sanggam inlaid celadon or when it was widely distributed. There is only one such piece of Goryeo celadon that is clearly documented. It was excavated in Jireung from the tomb of King Myeongjong (1131~1202).33 However, it seems clear that sanggam inlaid celadon became widely sought after from the mid to late twelfth century. Perhaps, the inlay decorative technique that originated from gold and silver filament inlay used for metal craftwork or lacquerware appealed to Goryeo people. Designs created by clear contrasting lines in black and white on the background of a calm shade of jade unique to Goryeo celadon appear as if they were painted, imparting a feeling of richness on the calm and quiet “canvas” of jade-green. It is clear that Goryeo sanggam inlay celadon was imported into Japan by the thirteenth century. Goryeo celadon from this period has been excavated in larger volume and at more locations over wider areas. Goryeo sanggam inlay celadon has continually been excavated in such areas of northern Kyushu as Tsushima, Iki, Hakata, and Dazaifu and Kusadosengen-cho, Fukuyama-shi, Hiroshima-ken Prefecture in the Setouchi region.34 Quite a large volume of Goryeo celadon has been discovered in the Kanto region as well. Goryeo celadon discovered in Kamakura, which had flourished as the Japanese capital and important center of politics and culture from the end of the twelfth century to the first half of the fourteenth century, was outstanding both in quantity and quality.35 Fifty three Goryeo celadon pieces were discovered at 14 historic sites in 1985 alone and the discoveries continued beyond that year. Most of the Goryeo celadon excavated in Kamakura is decorated with inlaid designs. Designs of clouds and cranes, peonies, and grapes were used for decoration. Some bottles are decorated with ruyi pearl designs on the shoulder and designs of maze-like, jagged, broken lines or lotus on the bottom. A few pieces are decorated with incised reverse inlay designs and one piece is decorated with an iron-painted design. The glazes are transparent bluish-green, bluish-gray, or olive. Interestingly, the discovered pieces were mainly bottles. This is especially true in Kamakura, where the percentage of bottles is far greater compared to the pieces excavated in Kyushu. This is also true of Chinese pottery excavated in this region. Bottles with small flaring lips, broad swelling bodies, and narrow waists are called maebyeong or plum blossom bottles. A greater number of Chinese porcelain maebyeong bottles have been excavated in Kamakura than in other region. This shows that maebyeong bottles were highly prized by the ruling class of Kamakura. The Goryeo celadon maebyeong bottle is one of the most beautiful pieces that have the unique beauty of Goryeo porcelain. It has a small flaring mouth and elegantly S-lined body and is decorated with inlaid, incised, or iron-painted designs. The fragments of celadon pieces with inlaid designs that were discovered at 1-210 Yukinoshita around the Wakamiya area are thought to be part of a gourd-shaped ewer.37 The most famous complete piece of this type of Goryeo celadon is a ewer decorated with inlaid designs of playful boys dangling on grapevines. It is now in the collection of the Osaka Municipal Museum of Art and it is a true masterpiece that shows the characteristics unique to Goryeo. Despite the fact that relatively few Goryeo celadon pieces have been excavated in Kamakura compared with Chinese ceramics, maebyeong bottles account for a certain percentage of pieces excavated in Kamakura. Many maebyeong bottles have been found at historic sites of the residences of the ruling class and of temples. This indicates that maebyeong bottles were widely sought after in Japan as they satisfied the tastes of the Japanese people. In the Kanto area, a maebyeong bottle decorated with a ruyi pearl design in white inlay on the shoulder was discovered in a historic site of Kamihamada, Ebina-shi, Kanagawa-ken Prefecture.38 A beautifully glazed maebyeong bottle decorated with an incised lotus scroll design was discovered in a historic site of Utsukidai, Hachioji-shi, Tokyo.39 A maebyeong decorated with iron-brown design was discovered at the Minamihiromaji historic site, Hino-shi, Tokyo.40 All these historic sites are in the vicinity of Kamakura, which is thought to have been the residential center of the ruling class. Clearly, a huge percentage of all Goryeo celadon maebyeong bottles excavated have been found in the Kamakura region. Artifacts that are used as important references in studying the imports and distribution of Goryeo celadon in Japan remain on Tsushima Island. Located on route between Kyushu and the Korean Peninsula, Tsushima served as an important gateway through which products shipped from Korea were distributed to Japan in large volumes.41 Bottles and ewers produced in the Goryeo and Joseon periods remain in shrines all over the island. Among them, those at the Kaijin-jinja Shrine in Minemachikisaka, the central southern part of Tsushima Island are most interesting. The Kaijin-jinja Shrine, the most magnificent shrine on Tsushima Island, is famous for the standing Buddha statue from the Unified Silla (designated as Important Cultural Property in Japan). Eleven pieces of Goryeo and Joseon ceramics remain in this shrine.42 Six of them are Goryeo celadon. Four are maebyeong bottles, of which two are decorated with incised designs and the other two with inlaid designs. The remaining two celadon are ewers. One of the ewers is decorated with a scroll design in reverse inlay. The upper part of the body, spout, and handle of this ewer are missing but we can see that the shape of the missing handle was a twisted strip. It is assumed that this piece was a large gourd-shaped ewer like the celadon ewer decorated with an inlaid grape scroll design in the collection of the Seikado Bunko Art Museum. The other interesting piece is a ewer in the shape of a bird with a human figure on its back. The human head and the tail of the bird are missing but it has a horn on its head and its feathers are exquisitely incised. There are still traces of the broad and flat tail attached to the body suggesting that the tail of the bird was connected to the back of the human figure as a handle as is the case of the bird-shaped ewer in the collection of the Art Institute of Chicago. The bird is decorated with a dot design painted with iron-brown pigment and white clay. In addition to these six, there is one unglazed maebyeong bottle from Goryeo. Regrettably, there is no record that tells us about the origin of the Goryeo celadon in Japan. Mikami Tsugio, who participated in the excavation of the Tsushima historic site as a member of the Society of East Asian Archaeological Studies in the summer of 1948, wrote, “Goryeo celadon maebyeong bottles were so carelessly placed on the altar... I keenly felt that Goryeo and Japan were so close.”43 Besides the Kaijin-jinja Shrine, pottery from Goryeo and Joseon has also been preserved at other shrines all over Tsushima Island. It may be that these pieces were brought to Tsushima during the medieval period, offered to shrines, and handed down through the years as valuable objects up to the present day. Some academics think they were presented in prayer for safe voyages. The Goryeo celadon preserved at the Kaijin-jinja Shrine and that excavated from historic sites in Japan including Kamakura share many features in common with respect to shape and decoration. Many of the pieces of Goryeo celadon excavated from historic sites including Kamakura are only debris, but they provide clues to the style and the stages of development of the Goryeo celadon imported into Japan. Studies on the period when Goryeo sanggam inlaid celadon was introduced to Japan have been conducted based on the excavations in Kamakura.44 Fragments of celadon pieces painted in iron-brown pigment are the oldest artifacts to be found there, presumed to be produced before the mid thirteenth century. The amount of Goryeo celadon excavated in Kamakura increased in the mid thirteenth century and continued to increase well into the first half of the fourteenth century. Although academics differ in their opinions about the stylistic development of sanggam inlay, the shards of Kamakura celadon clearly defy the common belief that the production of Goryeo celadon reached its peak in the mid to late twelfth century and began to decline in the early thirteenth century. Given that Goryeo celadon cannot be defined as a trade commodity for wide distribution in general and maebyeong bottles and ewers are items that could be easily handed down to later generations, the increasing amount of Goryeo celadon in Japan does not necessarily mean an increase of its production in Korea at the time. Some academics argue that these Goryeo celadon pieces were discarded after having been handed down in Japan while others think that they had been kept in Korea for a certain period of time before being taken to Japan. As is the case with early Goryeo celadon, if we look at Goryeo celadon and Chinese ceramics excavated together as a group, we can find certain relations. Goryeo celadon produced during the golden age of sanggam inlay celadon is distributed along with the so-called ‘pale grayish celadon,’ the third group of Longquan celadon. Therefore, the production date of the excavated artifacts cannot be totally different from the date of importation of the artifacts to Japan. In fact, China excavated an example of Goryeo sanggam inlay celadon maebyeong bottles with clear purchase date. At the site presumed to have been tomb of Shi Tianze (buried in 1275) in the Village of Hou Tai Bao, City of Shijiazhuan, Hebei Province, a Goryeo celadon maebyeong bottle decorated with a flower and clouds design along with the lid of a celadon jar produced at the Longquan kiln were found45 and its burial was dated to the latter half of the thirteenth century. That the period when Goryeo celadon decorated with inlaid designs was produced and the dating of Goryeo celadon artifacts excavated from historic sites in Japan did not match could be explained if we see that the decline in the quality of Goryeo celadon occurred later than we have previously believed. Up to the present time, the prevailing theory was that Goryeo sanggam inlay celadon lost its refined quality after the Mongol invasion of 1231. It was believed that the quality of designs began to deteriorate sharply and the color of glaze turned to grayish and brownish. However, as Ito Ikutaro shows,46 I think that high-quality Goryeo celadon continued to be produced until the latter half of the thirteenth century. Because the Goryeo celadon excavated from historic sites has only been found in small shards, it is difficult to determine the characteristics of the shapes or decorative designs of the original pieces. Compared with the Goryeo sanggam inlay celadon produced during its golden age in Goryeo, the Goryeo celadon discovered at the Kaijin-jinja Shrine is of poor quality: the decorative designs were increasingly stylized and the decorations around the mouth and bottom of body were less refined. All told, it is reasonable to believe that there was quite a time gap between the two. V Goryeo celadon was not the only type of ceramics imported into Japan. Charcoal-gray unglazed stoneware, which originated from the stoneware of the Three Kingdoms and Unified Silla periods, continued to be produced in Korea even after celadon came into being and was also brought into Japan. Little study on this unglazed stoneware has so far been done relative to Goryeo celadon. It has drawn little academic interest because it was used as miscellaneous vessels. Also, due to a scarcity of materials on it and the lack of study on the kiln sites where it was made, this unglazed stoneware hardly receives any mention at all in the history of ceramics in Japan. Saying, “At present, we have not reached the stage where the shapes or development periods of this stoneware excavated in Japan can be discussed. For now, I will present materials on artifacts excavated and point out the shapes and production techniques that can be seen from these artifacts such that they can be used as references for future study,” Akashi Yoshihiko elaborated on the characteristics of excavated unglazed stoneware imported from Korea and findings in excavated sites around Kourokan, the house for foreign affairs and international trade during the Heian period; Hakata, Kanzeon-ji Temple and its vicinity; Honman-zan Mountain; and Inayoshimotoyatugi in Ogori-shi City.47 According to Akashi, the only types of unglazed stoneware excavated in Japan were the jars used for storage and transportation. No stoneware bowls have ever been found. Most of the pieces are jars with wide large mouths. Some have handles or ears attached to their bodies. No unglazed stoneware in the shape of a maebyeong bottle has been excavated in Japan but an unglazed stoneware maebyeong bottle is preserved at the Kaijin-jinja shrine. Little unglazed stoneware produced during the Goryeo period has been discovered at historic sites in Japan, except for Tsushima, Nagasaki-ken Prefecture. More recently, historic sites of medieval times, including the site of Maitreya Hall, Kaijin-jinja Shrine, Kisaka, Mine-machi,48 site of Oishibaru, Kamiagata-machi, 49 and site of Mizusaki, Mitsushima-machi50 have been excavated one after another, allowing archaeologists to do quantitative analysis of artifacts from this period. In particular, reports on excavations at Oishibaru and Mizusaki include studies based on the composition of artifacts found. A huge volume of Goryeo celadon, Chinese pottery, and other artifacts were discovered from the site of Maitreya Hall, Kaijin-jinja Shrine. The highest percentage of all artifacts excavated was of ceramics and there was more Korean pottery than Chinese pottery. And the bulk of the pottery from Korea was large, unglazed stoneware jars. These are believed to have been produced in the twelfth and thirteenth centuries, the Goryeo period. They were likely used for everyday purposes, such as holding water and the like. A number of jars and bottles thinly coated with brownish and green brownish glazes were also found here. This kind of glazed stoneware rarely draws much attention from academia but it is believed to have been produced in the Goryeo period. The historic site of Oishibaru is in Kamiagata-machi in the north on the west coast of Tsushima, close to the Korean Peninsula. The pillar foundations of buildings were found on Tsushima for the first time and a huge volume of trade ceramics has been excavated therein. Most of these pieces are believed to date back to the twelfth and thirteenth centuries. An in-depth report on the excavation complete with quantitative analysis on the pottery discovered was done, and it pointed to some interesting facts. First of all, more than 80% of the ceramic pieces excavated were imported and less than 20% of the pieces were Japanese-made earthenware. Although roughly the same number of Chinese ceramics and Korean ceramics were found, they were of very different types. Most of the pieces of Chinese ceramics were white porcelain and celadon bowls to be used for offerings. Most of the Korean porcelain pieces were also bowls, but they only amount about half the number of Chinese bowls found. More than half of the Korean ceramics were stoneware and 70% of these stoneware pieces are unglazed. Most are jars that were used for storage. The vast bulk of the ceramics used for offerings are Chinese and very little of the ceramics found are Japanese. In contrast, more than 90% of vessels used for storage are Korean stoneware. Most of the vessels for cooking are Japanese-made sueki ware or stone pots. The historic site of Mizusaki in Osaki, Mitsushima-machi, Simoagata District is believed to date from the latter half of the fourteenth century to the first half of the fifteenth century, which is a little bit later than the historic site of Maitreya Hall, Kaijin-jinja Shrine and the Oishibaru site. A large volume of trade ceramics was excavated at the Mizusaki site. Noticeably, many ceramic pieces from Southeast Asia were discovered here and considering the volume of ceramics found, the Hayata family clan was even more powerful on Tsushima Island than the So- Family from the first half the fourteenth century to the first half of the fifteenth century. A huge volume of export ceramics excavated from Mizusaki was due to active international trade by the Hayata clan. According to quantitative analysis, export ceramics accounted for 97% of all ceramics excavated and Korean-made ceramics accounted for 70%. Up to 5% of the ceramics found were from Southeast Asia, especially Vietnam. Of Korean-made ceramics, about 50% was unglazed stoneware and the remaining was buncheong ware. In the determination of these percentages, those ceramics coated with only thin layers of glaze were classified as unglazed stoneware. A large volume of Korean-made stoneware was found and most of this was of jars for storage. Little study has been conducted so far. It may be necessary to study the characteristics of the historic site more closely but it seems that Korean-made stoneware was widely used on Tsushima Island for storage from the twelfth century through the fifteenth century. It is obvious, based on the sheer number of pieces excavated and the percentage of vessels used for storage, that Korean-made stoneware was integral to the everyday lives of the Japanese people on Tsushima Island. It was not used as packaging or containers for other objects or products. On Tsushima Island during the medieval times, Chinese-made bowls and plates were most widely used, along with some celadon and buncheong ware produced in Korea. The large jars used by the Japanese for storage were unglazed Korean stoneware. It is clear that unglazed stoneware continued to be produced in the Joseon period and was imported into Japan. Full-scale studies on this type of unglazed stoneware are about to start. Due to the lack of materials, changes in the production period have not yet been found for the most part. There are presently few reports on Korean-made unglazed stoneware excavated in Japan. However, given the fact that this type of unglazed stoneware only recently began to draw serious attention, there are many cases that need to be thoroughly investigated. For example, some of the stoneware that was considered earthenware produced before the Unified Silla period may actually be the unglazed stoneware produced after the Goryeo period. It is hardly likely that unglazed stoneware produced on the Korean Peninsula was widely distributed in Kyushu or Honshu but the unglazed stoneware brought into Tsushima was most likely obtained through international trade. In studying trade between Goryeo and Japan, unglazed stoneware should not be overlooked but to be studied in depth. VI Entering the thirteenth century, Goryeo fell into a national crisis. The Mongols invaded Goryeo Korea in 1231, forcing the Goryeo king to flee to Ganghwa-do Island. More serious tribulations were to come. In the latter half of the fourteenth century, Goryeo was wracked by constant raids and invasions by the Japanese marauders. Celadon continued to be produced until the end of Goryeo but it became less and less refined. The sanggam inlay designs lost their rich expressive beauty; works had a less mature appearance; and pressed patterns were more commonly used. Towards the end of Goryeo, the clay used was very coarse and the glaze became grayish and opaque. Goryeo celadon was declining noticeably in every respect including the shape and the firing. Some Korean-made ceramics from the latter half of the fourteenth century have inscriptions of ‘Jeongneung.’ Jeongneung is the name of the royal tomb of King Gongmin’s wife, Mongolian Princess Noguk, who died in 1365. King Gongmin passed away in 1374. Accordingly, celadon with inscriptions of Jeongneung was likely the best quality celadon of that time. This celadon was probably used as ritual vessels starting in 1365 the earliest, the year when the princess died, and continued to be used until 1374, when the king died. Shards of bowls with the inlaid inscription of Jeongneung in black were excavated from the historic site of Dazaifu.51 Although it not known how it was brought into Japan, it is valuable material as it clearly indicates the year of production. The quality of celadon produced in the closing years of the Goryeo period was certainly declining, but the quantity of Goryeo celadon brought to Japan actually increased. There was a greater volume of imported celadon and it was distributed over a wider area of Japan. Some Goryeo celadon at this time were even sent as far south as Okinawa and examples of artifacts were excavated from Shuri-jo Castle, Izena-jo Castle, and Nakijin-jo Castle.52 In the northeastern region, Goryeo celadon excavated from the historic site of Tosaminato, Shiura-mura, Aomori-ken Prefecture was thought to be produced at the end of the Goryeo period.53 Shards with inlaid designs excavated from Ne-jo Castle,54 Hachinohe-shi and Namioka-jo Castle, Namioka-machi,55 Aomori-ken Prefecture are also believed to have been produced at the end of Goryeo. In the latter half of the fourteenth century, which marked the ‘period of decline’ for Goryeo celadon and the turbulent change between dynasties on the Korean Peninsula, the volume of Korean-made ceramics brought into Japan increased and was distributed more widely. This trend continued into the Joseon Dynasty. Buncheong ware produced in the early Joseon period succeeded sanggam inlay celadon from the Goryeo period but changed in terms of shape and decorative technique as a new form of expression came into being. Development from sanggam inlay celadon at the end of Goryeo to buncheong ware at the beginning of Joseon can be understood as a continuation of trends. Also, from the viewpoint of distribution and accommodation, it is difficult to declare that there was severance between sanggam inlay celadon and buncheong ware. When discussing the distribution of Korean-made ceramics in Japan, we need not divide the period between ‘Goryeo’ and ‘Joseon.’ On Tsushima Island, the gateway through which Korean ceramics were brought to Japan, ceramics that were seemingly produced from the end of Goryeo to the early Joseon period have been excavated in large volume. In addition to the Mizusaki historic site mentioned earlier, shards were found at the site of a residence of Yoritsugu, the second son of So Morikuni, at Nii, Toyotama-machi. The historic site of the capital castle at Iki used to be a residential area that flourished in the fourteenth and fifteenth centuries. Korean-made ceramics excavated from this area account for 26.4% of total imported ceramics.56 It was pointed out that the volume of Chinese-made ceramics excavated on the mainland of Nagasaki-ken Prefecture shows the volume from the first half of the fourteenth to the first half of the fifteenth centuries decreased. Miyazaki Takao speculates, “The volume of trade ceramics from the fourteenth and the first half of the fifteenth centuries excavated in the Iki and Tsushima region shows little decrease which may be attributable to trade with Korea.”57 Kamei Meitoku years ago pointed out that so many issues remained unsolved as there were no historic sites that preserved imported ceramics for one hundred years from the mid fourteenth to mid fifteenth centuries.58 The situation has hardly changed since then. Even so, Korean-made ceramics brought into Japan during that period are relatively distinguishable. Then, did Korean-made ceramics fill the gap created by the decrease in imports of Chinese ceramics into Japan? Although excavation revealed that the volume of Korean-made ceramics increased by this time, it was far below that of Chinese ceramics. Korean-made ceramics were not imported in large enough volume to make up the decrease in Chinese ceramics imports. On Tsushima Island, on the other hand, porcelain, celadon bowls, and dishes made in China were used together with bowls and dishes made in Korea. It may be reasonable to think that when the supply of Chinese ceramics decreased, imports of Korean-made ceramics increased to make up the shortfall. We can believe that some of these ceramics imported into Tsushima were transported to Iki, Kyushu, and finally all over Japan through trade. Interestingly, Korean-made ceramics were imported during the period when trading by the Ryukyuoukoku Kingdom was at its zenith. Furthermore, as frequently cited, a large volume of ceramics from Southeast Asia was also imported into Iki and Tsushima. The latter half of the fourteenth century was a huge transition period in East Asia. The fact that ceramics produced in Korea and Southeast Asia were imported into Japan during this period can be seen as a reflection of the rapid changes in ceramic production and trade following the new international order that was being created at that time. The specific factors and reasons for this development such as the traders need to be studied in depth through philological studies but there is no doubt that ceramics from Korea and Southeast Asia are important proof of the roles that Tsushima and Okinawa played in international trade.

It is difficult to ascertain when in Korean history the first portraits were created. Based on the available historical records and a few extant examples of figure paintings, the earliest portraits date from the Three Kingdoms period. Apparently, the figures found on the murals in Sasinchong Tomb and Ssangyongchong Tomb in Maesan-ri and a tomb in Deokheung-ri from the Goguryeo Kingdom (37 B.C.~668) were drawn to indicate the buried in the tombs (Plate 1). However, it is unlikely that these figures were faithfully portrayed because they look too typical to represent specific people. Meanwhile, a book of the Goryeo period (918~1392) describes about people paying their respects before the portraits of Buddhist monks of the Three Kingdoms period. It is very possible that these portraits were created during the Three Kingdoms period but this cannot be confirmed. ( Plate 1 ) Husband and wife of Ssangyongchong Tomb, Murals from Goguryeo Kingdom, Yonggang-gun, Pyeongannam-do Province, 5th-6th century Nevertheless, various records on the epitaphs and documents show that, by the Unified Silla period, not only portraits of kings were drawn on the walls of temples as murals but also numerous images of monks were produced. As for portraits of aristocrats, a portrait of Choe Chi-won (857~?), a great poet, scholar, and statesman of the 9th century Unified Silla Kingdom, exists as documented today. Although it is a double-copied version, the Portrait of Choe Chi-won featured at this exhibition has drawn great deal of interest by many viewers and scholars alike. In the Goryeo Dynasty, portraits of the king and queen were frequently produced and enshrined. Various types of portraits of loyal subjects were often made under titles such as ‘Byeoksang Gongsin’ or ‘Dohyeong Gongsin,’ which were given in reward for and to praise their various contributions to the dynasty. In addition, a number of private records account for the inscriptions and the eulogies in regard to portraits of the literati. A considerable number of portraits of women (queens’ in particular) and of Buddhist monks were also created in this period. Regrettably, however, only a few pieces of these portraits remain today (copied versions included) and they are not sufficient enough to map the stylistic changes and characteristics of the portrait paintings of this era. In contrast, the Joseon Dynasty left a tremendous amount of portraiture. Because the dynasty promoted filial piety and loyalty to the state as the two most important virtues, numerous shrines variously called sadang, yeongdang, and seowon including the halls for kings were built to practice the virtues. Portraits of the illustrious, the wise, and the deceased were accordingly in great demand to be enshrined for tribute there. Even the kings occasionally ordered portraits of meritorious subjects to promote and to honor them for their remarkable services to the nation. A great number of portraits to respect elderly officials and Buddhist monks were also produced. To most portrait painters of Joseon, ‘rendering a portrait’ especially of a high figure was a good opportunity to become famous as a professional artist and to rise to a higher position. The painters naturally made every effort to depict the subject as realistically as possible faithfully adhering to such old principles as: “The slightest mistake, like one strand of hair not exactly resembling that of the subject, would fail the portrait.”1 According to various inscriptions and accompanying titles of portraits, the stringent evaluation criteria were to be applied to viewers and critics as well. They must have the eyes to see not only ‘the physical veracity’ but also ‘the spiritual veracity’ to determine whether it is a fine piece or not. Seven-tenths perfection was considered satisfactory even in the production of portraits of the king (a state affair so important that only the greatest painters of the times were employed) suggesting that the expectations for the artistic achievement in the genre of portrait painting were very high in Joseon. It may be due to this lofty desire to ‘learn about the spirit through figure’2 Koreans today have many excellent Joseon portrait paintings in the museums as well as in the various shrines in care of the descendants or the Confucians throughout the country. The portraits of Joseon mark a brilliant aspect of Korean painting. This paper classifies traditional portrait paintings into six categories based on the social status of the subjects and will trace the historical development of each category: portraits of kings, portraits of meritorious subjects, portraits of elderly officials, portraits of literati, portraits of women, and portraits of Buddhist monks. The portraits presented at this exhibition were selected to illustrate the unique characteristics of each category and to demonstrate how traditional Korean painting differs from those of its counterparts under the same Chinese cultural influence, such as China and Japan. The ultimate goal of this paper is to delineate the unique characteristics of Korean portrait painting through this exhibition that collectively presents the portrait painting of Korea, China, and Japan. II TYPES OF KOREAN PORTRAIT PAINTINGS 01 PORTRAITS OF KINGS, EOJIN Eojin or portraits of kings have been called by various names: eoyong, suyong, jinyong, seongyong, and wangyeong.3 It is not clear when the first eojin was created. However, there is a record in the Treatise on Ritual and Music: Goryeo Music section of History of the Later Tang Dynasty (compiled in 1060 during the Northern Song period) about a portrait of a king that was painted on the ivory decoration of a musical instrument to perform Goryeo music to Emperor Gaozu (the first emperor of the Tang Dynasty who established nine musical divisions to perform diverse court music). Although this particular eojin was made for decoration of the instrument, we cannot exclude the possibility that eojin as portraits were already being produced in earnest around this time.4 The History of the Three Kingdoms (written by Goryeo scholar Kim Bu-sik in 1145) describes the royal portraits on the wall of Buseoksa Temple verifying that portraits of kings were certainly being produced by the Unified Silla period. Changamjip, an anthology by the eighteenth century poet Kim Sang-chae (pen named Changam), also tells about a portrait of King Gyeongsun (?~978) of the Unified Silla drawn on a wall of a temple in Wonju. During the Goryeo period (918~1392), portraits of kings were being produced for various types of shrines that had been built in accordance with the development of custom to respect the deceased. Gyeongryeonjeon Shrine was founded in the capital to house the portrait of King Taejo (the founder of the dynasty) and those of numerous deceased kings that were alternately hung in a group of five in the shrine. Buddhist temples were built all over the country (usually near Gaeseong, the capital) and individually assigned for the portraits of deceased royal couple. Sadly, only two royal portraits from the Goryeo period remain today. A sketchy portrait of King Taejo Wanggeon (918~943) remains at Sunguijeon Shrine historic site but it is hardly recognizable. A portrait of King Gongmin (1351~1374), the 31st monarch of Goryeo, and his Mongolian wife Princess Noguk remains in the National Palace Museum of Korea (Plate 2). In this color portrait of single canvas, King Gongmin and Princess Noguk appear side by side. Their costumes are noticeably of the style of the closing years of Goryeo through the early years of Joseon, but neither the king nor the princess seems drawn true to life. ( Plate 2 ) Portrait of King Gongmin and his wife Princess Noguk; Ink and color on silk; National Palace Museum of Korea Throughout the Joseon period (1392~1910), a huge number of portraits were produced for the kings ranging from Taejo, the founder of the Joseon Dynasty to the last monarch, King Sunjong. From the beginning the dynasty built as many as six shrines at six locations throughout the nation to enshrine the portrait of King Taejo: Munsojeon in Seoul, Junwonjeon in Yeongheung, Yeongseungjeon in Pyeongyang, Mokcheongjeon in Gaeseong, Jipgyeongjeon in Gyeongju, and Gyeonggijeon in Jeonju. Portraits of preceding kings and queens were enshrined in Seonwonjeon Shrine on the premises of Gyeongbokgung, the main palace. The dynasty continued to build shrines systematically for royal portraits until the Japanese invasion of 1592. During the war, nearly all of the shrines were burnt down and almost all of the royal portraits inside them were lost. Although the shrines were never rebuilt after the war, they were still considered very important. Before the end of the Joseon Dynasty, Junwonjeon and Gyeonggijeon were additionally constructed to enshrine the portraits of King Taejo; Yeonghuijeon and Seonwonjeon were built inside Changdeokgung for the portraits of other kings; and various small places and pavilions were assigned within the palaces for other royal portraits.5 According to a historical document, as many as 26 scrolls of the portraits of King Taejo were produced. One of them allegedly described the king riding on horseback and many others portrayed the kings in various costumes such as a military outfit, crown and royal robe, hood, or traditional Korean hat. Regrettably, only a few eojin remain today. An eojin of King Taejo is kept at Gyeonggijeon Shrine, Jeonju and one at the Royal Museum of Deoksugung Palace is King Yeongjo’s. Only parts of portraits of Prince Yeoning (later King Yeongjo), King Cheoljong, and King Ikjong remain today after a fire broke out at the storage house where they were kept during the Korean War. In addition, a few portraits of King Gojong and the draft of a portrait painting of King Sunjong still exist. Although the full-length portrait of King Taejo (Plate 3) is a copy produced in 1872, it faithfully reproduced the original painting in terms of both formality of facial expression and painting technique (Plate 4). This portrait shows King Taejo seated facing front and wearing a king’s hat and royal robe and closely resembles the portrait of Taizu of Ming at the National Palace Museum in Taipei. The portrait of King Yeongjo is half-length down to the abdomen. It clearly shows the portraiture style of the latter half of Joseon even though it is a copy. The swollen and sunken parts of the face are expressed by dying the silk canvas in light brown. Many of the portraits of King Gojong (Plate 5) in existence are frontal, full-length portraits with the king portrayed seated on a chair, wearing a king’s hat and royal robe, with a folding screen with the sun and the moon pictured in the background. This style is typical of Chae Yong-sin (1850~1941), the most renowned portrait painter at the time.6 ( Plate 3 ) Portrait of King Taejo, Jo Jung-muk and others, 1872; Ink and color on silk; 218x150cm; Gyeonggijeon Shrine, Jeonju ( Plate 4 ) Various sides of a face ( Plate 5 ) Portrait of Emperor Gojong, Attributed to Chae Yong-shin, Early 20th century; Ink and color on silk; 130x70cm; Wonkwang University Museum Seungjeongwon Ilgi, the diary of a government agency called ‘Seungjeongwon,’ the Royal Secretariat, elaborates on the production process of a portrait of a king. The account enables us to understand the significance of a royal portrait in the era of Joseon. The making of a king’s portrait was a very complex process: First, a temporary office was established to supervise the work of the painters selected from all the court painters at Dohwaseo, Royal Bureau of Painting. The selected painters were usually those who had been recognized as the most outstanding in portrait painting at the Bureau, but when there was no painter qualified enough for the job, outside painters were brought in through recommendations by the ministers. In some cases, an open competition was held to test painters, where each candidate was required to draw meritorious subjects of the court. Once selected, the painters were divided into three teams. The first (jipilwhasa) was the lead team responsible for rendering the king’s face. The second team (dongchamwhasa) was the team of associate(s) in charge of other parts of the king’s body, and the third team (sujongwhasa) was to assist the first and the second teams with coloring. Six painters were usually involved but there could be as many as 13. In addition, there were many other workers such as artisans, mounting technicians, and sewers assigned for mounting the painting on a scroll. As for the process of rendering a royal portrait, rough drafts were made on paper first (chobonwansung) and then the actual portrait was painted in ink on a silk canvas (sangchomukwha) followed by coloring (seolsaek), mounting (whubae), entitlement (pyoje),7 designating place (jinjeon), placing (bongan), and finally evaluation of the painting by the artists and ministers in charge (nonsang). Furthermore, divination was practiced to determine auspicious days for each of these processes and the king and his ministers often oversaw the paintings in progress. The complexity of these processes indicates that the production of a royal portrait was tantamount to any major state event. While the production of a king’s portrait was for royal descendants to cherish his memory (as was the case with ordinary families) the enshrinement of the portraits was performed as a symbol of the royal wish to perpetuate the dynasty. 02 PORTRAITS OF MERITORIOUS SUBJECTS, GONGSINSANG Production of gongsinsang or portraits of meritorious subjects was ordered by the king in appreciation of those who rendered distinguished services for the nation. These portraits were developed to enhance the power and the prestige of the monarchy and to warn citizens against disobeying their monarch. While eojin were enshrined as symbols for tribute because the royal portraits were for the royal ancestors, gongsinsang were done for more practical purposes. Bestowing the title “gongsin (meritorious subjects)” on a person for his distinguished services for the nation and building a pavilion for enshrinement of the portrait of the meritorious subject were regarded as a great honor not only for the gongsin but also for the future generations of his family. It also set a strong example for other subjects and Joseon promoted gongsinsang as an essential form of portrait painting in this regard. Portraits of meritorious subjects date back to ancient times. According to historical records in China, a portrait of a subject was produced in the third year of Emperor Gaozong (5 B.C.) of the Han Dynasty and enshrined in Qilinge Pavilion. A portrait of Zhao Chongquo under Emperor Chengzong was enshrined in Ganguangong Temple and another example was enshrined in a tower in Yunnan Province during the reign of Emperor Mingzong. There are records about monetary rewards to be given to officials for distinguished services during the Unified Silla. However, it is not clear if a status like gongsin existed at the time.8 The earliest record of this practice dates to 940, the 23rd year of King Taejo’s reign. According to Goryeosa (the History of Goryeo), King Taejo ordered the construction of a shrine called Gongsindang (Hall of Meritorious Subjects) at Sinheung Temple to honor the subjects who contributed to the founding of the new dynasty. A couple of gongsinsang were painted on the east and west walls of the shrine. From that year onwards, gonsinsang were continually produced to praise various achievements in defense of the nation, expansion of national boundaries, suppression of internal revolts, and etc. In Joseon, numerous gongsinsang had been created from the beginning of the dynasty. No fewer than 28 different titles were used to commend the socalled gongsin throughout the 500-year history of the dynasty. Nearly every time gongsin status was conferred, the king ordered the construction of a shrine to house the portrait of the gongsin and the establishment of a stone monument to praise the subject for his distinguished service.9 A government agency called Chunghunbu (division of loyalty and merits) was responsible for the administrative affairs on this gongsin practice and a pavilion called Gigonggak was established to house gongsinsang, the portraits. Two copies were made for each gongsin: one was enshrined in Gigonggak,10 also called Gigak and Ingak, and the other was to reside in the home of the eldest grandson of the gongsin family. The gongsinsang of Joseon are important in two respects. First, the quality of the portraits is excellent because they were produced by the top portraitists at the king’s direct call. Second, they are valuable in Joseon studies because they bear the exact production dates. Although the clothes and hats depicted on gongsinsang vary with the passage of time, in almost all cases, the subject was seated in a chair with his hands folded on his lap and dressed in his official robe (including a black silk hat) that bear his official emblem on the chest. These chest patches are important for historical research as they reveal the subject’s rank in the government at the time the painting was created. The Portrait of Yi Jung-no (1577~1624) is typical of gonsinsang in the mid Joseon period (Plate 6). It is a full-length portrait depicting the gongsin seated in a chair with his hands folded in front and dressed in official robes and a black silk hat with his face slightly turned to the right. The floor is covered by a colorful, patterned carpet, which is typical of gonsinsang in the mid Joseon. The official emblem on his chest with haetae (a mythical and imaginary animal) and golden belt with crane tells us that Yi Jung-no served as a class 2 military official when he was bestowed the gongsin title. ( Plate 6 ) Portrait of Yi Jung-no, Anonymous, Early 17th century, Ink and color on silk, 171.5x94cm Gyeonggi Provincial Museum With albums of paintings becoming widely popular in the latter half of the Joseon period, gonsinsang were produced not only on large canvases but also in albums. By this time, in particular, Chunghunbu produced gongsinsang in albums for cost-cutting and convenience. Portraits in the form of an album were also issued in two copies: one was kept in Chunghunbu and the other was given to the family of the subject. As seen in the example of the album of Yi Man-yu, a portrait album consists of two leaves (Plate 7). One leaf bears the name, official government post, and brief biography of the subject and on the other is the portrait of the subject. In the latter half of the Joseon period, larger gonsinsang were often produced separately by the family of the subject in commemoration of being granted gongsin title (Plate 8). ( Plate 7 ) Portrait of Yi Man-yu, Anonymous, Early 18th century, Ink and color on silk, 42x30cm, Private collection ( Plate 8 ) Portrait of Yi Man-yu, Anonymous, Early 18th century, Ink and color on silk, 70x104.5cm, Private collection Production of gonsinsang ended with the bestowal of gongsin titles to military officials who helped suppress the revolt of Yi In-jwa in 1728, the fourth year of King Yeongjo’s reign. However, this style of gonsinsang: seated in a chair with his hands folded in front in a dignified manner became popular among the Joseon elite and many aristocrats had their portraits painted in similar poses. 03 PORTRAITS OF ELDERLY OFFICIALS, GIRODOSANG While fewer in number, girodosang or portraits of elderly officials are considered as important as those of meritorious subjects because they were a means to memorialize the subjects. The word giro refers to elderly men with gi meaning sixty and ro seventy. However, for one to be qualified for one of these portraits, one should be much more than merely old. One had to hold a respectable social position and be known for one’s virtuous character and other excellent personalities. The Tang and Song of China first introduced an association of lofty and virtuous elderly men. Bai Juyi, a Chinese poet of the Tang Dynasty and Wen Yanbo, who formed the Poetry Society of the Venerated Seniors, organized a society of elderly men and had famous painters produce their portraits. This practice was introduced to Korea and first realized during the Goryeo period under the name, Haedong Girohoe. Choe Dang (1125~1211) and seven other elderly gentlemen founded the organization after Choe had resigned from his government post.11 It was a fraternal society to foster friendship of the member12 and subsequently triggered the formation of such societies of aristocrats and men of letters. Records show that Yu Ja-ryang (1150~1229) and Yi Geo-yi (1348~1412) formed private societies of elderly officials in Goryeo era.13 In Joseon, a court agency known as Giroso was established in 1394, the third year of King Taejo. It replaced private gatherings of Goryeo and even the king himself joined the agency when he turned 60. Civil officials aged over seventy of minor second rank (jong-I pum) or above were selectively given membership in the agency. The practice existed until the end of Joseon but never became more than symbolic authority having little actual power. No girodosang produced before King Sukjong’s reign (1674~1720) is available today and only the documents about such portraits remain. According to Girohoedoseo (Preface to Social Gathering of the Elderly) by Kim Sang-heon (1570~1652), the members of the so-called Girohoe or Society of Elderly Officials gathered together in Cheongpung-dong in August,1635, the 13th year of King Injo’s reign, and had painters produce a painting of their social gathering. It is, however, not a portrait in a strict sense because the human figures in the painting are treated as part of the scene. Most of the girodosang extant today are album leaves produced in the latter half of the Joseon period. A few examples on folding screens and hanging scrolls exist including a folding screen depicting a gathering scene of the Girohoe Society to celebrate the introduction of Gwon Dae-un (1612~1699) as a new member. Gwon Dae-un is shown on this screen with seven other new members of the Giroso in 1689, the 15th year of King Sukjong’s reign (Plate 9). The officials are illustrated in front of a Chinese-style mansion and garden—the same style of Xiyuanyajitu (Social Gathering at West Garden) from Song.14 The officials are dressed in official robes and black silk hats but without official emblem patches on their chests. Each of the eight officials is depicted full-length, seated in a chair, and dressed in official robes and black hat. Some are talking to others and some others are sitting in dignified poses. ( Plate 9 ) Folding Screen: Social Gathering of the Elderly; Ink and color on silk; Private collection, Deposited at Seoul National University Museum These portrait albums are much like modern school albums and Korea has two albums of the 18th century. One is Gisagyecheop of 1719, the 45th year of King Sukjong’s reign and the other is Gisagyeonghoecheop of 1744, the 20th year of King Yeongjo’s reign. Gisagyecheop is presented at this exhibition (Plate 10) and was painted by Kim Jin-yeo, Bak Dong-bo, Jang Deuk-man, and Heo Suk, all of whom were leading portrait painters of the time (some were even engaged in the production of royal portraits). The album clearly shows the characteristics of the painting style during the latter half of Joseon. Nonetheless, the small, half-length portraits of figures in stereotypical postures and stylized facial expressions do not show any specific features of the individuals. There are also some examples of girodosang that were privately produced for personal celebration. ( Plate 10 ) Album of the Gathering of Aged Statesmen, Kim Jin-yeo and others, 1720; Ink and color on silk; 43.7x32.5cm each; Private Collection, Treasure No 639 As mentioned earlier, although only a few girodosang remain today, the practice of producing this type of portrait continued until the end of Joseon. According to Jeungbomunheonbigo, an encyclopedia published during the late Joseon, production of portraits was standardized in 1809, the 8th year of King Sunjo’s reign. When a chair and a walking stick were bestowed by the king as a symbol of royal appreciation to elderly subjects15, portraits were painted to commemorate the honor together, like the cases of the portraits of Ha Yeon (1376-1453) and of Gwon Dae-un (Plate 11). These portraits were similar to those of elderly officials in terms of purpose: commemoration. Also, there is an album known as Myeongsin Hwacheop (Album of Portraits of Distinguished Subjects) that collectively contains portraits of distinguished subjects of the dynasty. Meanwhile, Myeongsin Chosang Chobonjip compiled drafts of the portraits of Myeongsin Hwacheop enabling us to understand various tastes of the portrait-loving Joseon people. ( Plate 11 ) Portrait of Gwon Dae-un, Anonymous, 1689; Ink and color on silk; 178x98 cm; Seoul National University Museum 04 PORTRAITS OF THE LITERATI, SADEABUSANG The fourth type is portraits of literati or sadaebusang. Historical records and a few extant examples of literati portraits indicate that the Portrait of Choe Chi-won is the first literati portrait in Korea (Plate 12). A record states: Choe Chi-won in his last years wandered around the country and secluded himself on Mt. Gayasan where Haeinsa Temple was located. Feeling despondent about the turbulent time, he suddenly disappeared one day. Yearning for his return to the temple, a monk prayed in front of his portrait enshrined in Dokseodang Hall of the temple. Interestingly, a portrait of Choe Chi-won was housed in a shrine called Daicisi Temple in China.16 A number of the poet’ s portraits remain today and two of them are presented at this exhibition: one from Museong Seowon—a private academy that also functions as a Confucian shrine—and the other from Cheongseongsa Temple. Both of the portraits depict him in Tang-style costume, but the way in which Choe is seated and fingers are folded is unusual following the Buddhist style. Also noticeable is that, although different coloring was used for these two portraits, both are of the same form and pattern, implying that they are imitations of the same copy. ( Plate 12 ) Portrait of Choe Chi-won, Anonymous, copied, 20th century; Ink and color on silk; 113x79cm; Cheongseongsa Shrine (Choe Jong-gyu Collection) Various documents such as Dongmunseon (Anthology of Korean Literature) compiled in the fifteenth century; inscriptions recorded in many other anthologies by private individuals and various records at shrines; and numerous tombstone inscriptions verify that many portraits of the literati were done in Goryeo.17 These portraits were used for educational, ceremonial, and commemorative purposes and sometimes exchanged to enhance companionships among literati. For example, portraits of Su Dong Po, one of China's great poets and calligraphers, and Bai Juyi, a Chinese poet of the Tang, were cherished by the literati scholars.18 Despite many existing documents about sadaebusang in the Goryeo period, the quintessential Goryeo style cannot be defined because few portraits from the time remain. Among these are portraits of An Hyang enshrined in Hongju hyanggyo (County School); a portrait of Yi Je-hyeon painted by Zhen Jian Ru of Yuan; portraits of Goryeo literati which were copied during the Joseon period such as portraits of Yi Saek, Gil Jae, and Jeong Mong-ju; and portraits of Yi Jang-gyeong, Yi Jo-nyeon, Yi Po, and Yi Seung-in that are enshrined in Seongsansa Shrine in Goheung, Jeollanam-do Province. Yi Gyu-bo’s Donggukisanggukjip (Collected Works of Minister Yi of Goryeo) indicated that the portrait of Bak In-seok depicted the subject wearing a white robe detailed with black trim. In addition, Jeong Hyeong-jin, a well-known painter of the time, left a portrait of Yi Gyu-bo on his seventieth birthday. These stories tell us that portraits were quite popular among the literati in Goryeo. This can be confirmed all the more clearly from various records. In fact, King Gongmin, the 31st monarch of Goryeo, personally painted portraits of his subjects: Yi Po and Yu Won-jeong according to Goryeosa; Yun Hae according to Mogeunjip (Collected Works of Mogeun); Yeom Je-sin, Yun Taek, and Yi Gang according to tombstones; and Son Hong-ryang according to Dongguk yeoji seungnam (Augmented Survey of the Geography of Korea). In Joseon, samyo ancestral shrines were developed under the Confucian teaching, “Be devoted to your parents when they are alive; be devoted to funeral when your parents pass away; and be devoted to ancestral rituals onwards. This is filial piety.” Samyo ancestral shrines are where ancestral tablets and portraits of the deceased are enshrined and rituals are performed several times a year to comfort the souls of the deceased. These samyo ancestral shrines were developed based on gamyo (family ancestral shrines) commonly called sadang, which originated from the Confucian doctrine: “As the root of individuals and a family is ancestors, it is the obligation of descendants to pay respect for what they owe to their ancestors.” While ancestral tablets are placed in gamyo, samyo (also called yeongdang) houses portraits of the deceased and holds ancestral rituals according to the customs of the time.19 Along with the development of gamyo family ancestral shrines, Joseon witnessed yeongdang shrines increasingly established and many of them exist today (Plate 13). ( Plate 13 ) Tajindang Shrine in Chungwon-geun where Portrait of Ha Yeon’s Wife is housed Apart from the aforementioned gamyo and yeongdang shrines, sawoo or common shrine was also built to worship great men and ancient sages regardless of blood relationship. According to Sinjeung dongguk yeoji seungnam (Augmented Survey of the Geography of Korea), compiled in 1481, the shrine to worship Kim Yu-shin, a general in 7th-century Silla, had been built during the Silla period for ordinary people to worship him. During the Goryeo, a number of sawoo were built for worship including the ones for Three Gongsin of Goryeo in Andong, Gyeongsang-do Province; General Yun Gwan in Jeongbuksa Shrine; and General Gang Min-cheom in Eunyeolsa Shrine. Great men and ancient sages were widely worshiped in Joseon and shrines to worship royal subjects and ancient sages rapidly proliferated. In many cases, those who came from the same village or who had special relationships with royal subjects or ancient sages took leadership in building such shrines. In addition, a sawoo was built within a seowon. Not all of these sawoo housed portraits to be worshipped but, because there are so many sawoo, the demands for portraits were great in the dynasty. At first, there were some differences between the figures to be worshipped at sawoo and those at seowon shrines. In general, sawoo were for those who had been distinguished in their righteousness, loyalty, and filial piety while seowon were for more qualified persons. To be enshrined in a seowon, moral philosophy, scholarship, and achievement in services to the state must be satisfied and the figures must have been born in the village (where the seowon is located), lived there (even only for a short time), been assigned there as a government official, or lived in exile there.20 Most figures enshrined at seowon were literati-bureaucrats and scholars. At sawoo, most were military officials who must have been stationed in the area or died in defense of their loyalty. Toward late Joseon, however, the distinction became blurred and selection of figures to be enshrned was often made based on blood relations or other connections rather than set standards and as a result the qualifications and the quality of figures to be enshrined degraded. Joseon had as many as 417 seowon and as many as 492 sawoo. With an increasing number of seowon being established throughout the country, seowon came to exert greater influence accompanied by abuse of power. During the reign of King Sukjong (1674~1720), every province had eight to ten seowon. Worried over the increasing malfeasance by the seowon, Kim Man-jung and others presented a memorandum to bring the abuses of the seowon to attention of the king. The abuses by sawoo, collectively called hyanghyeonsa, were extremely serious in the reign of King Yeongjo (1724~1776). Although some 300 such shrines were abolished by the order of the king, there were still as many as 650 nationwide and the number continued to increase until the end of King Cheoljong’s reign (1849~63). According to Jeungbo munheon bigo (Reference to the Old Books, Enlarged with Supplements), there were as many as 670 such shrines in Joseon then. In 1864, when King Gojong succeeded Cheol-jong at age of twelve, the king’s father, Daewongun (1820~1898), who held power at the time, had most seowon closed. By the order of Daewongun, only one seowon for one figure with distinguished scholarship and fidelity was preserved. With the hundreds of seowon having been abolished, only 47 shrines (including seowon and sawoo) remained. In the wake of Daewongun’s abolishment of the seowon followed by the Japanese occupation and the Korean War, most portraits housed in seowon were scattered and lost. As has been discussed thus far, shrines for portraits developed greatly in Joseon. Those shrines not based on blood relations such as sawoo and seowon, or shrines called saengsa21 housed portraits rather than spirit tablets. The reason for this seems to be that portraits were more accessible means for people to remember the deceased than hard spirit tablets. The most commonly enshrined figure at sawoo and shrines annexed to seowon is Song Si-yeol (1607~1689). As many as 16 shrines including the ones in Maegok, Suwon, and Hyoam, Jeongeup were built to worship the scholar-official. On all these portraits, Song Si-yeol is depicted in a plain robe and headgear called bokgeon or sabangmo. The Portrait of Song Si-yeol in the collection of the Seoul Museum of History is a half-length portrait of him dressed in a white robe detailed with black trim (Plate 14). The furrows of a wrinkled face well express Song Si-yeol’s austere features and his belief in righteousness. ( Plate 14 ) Portrait of Song Si-yeol, Anonymous, copied, late Joseon; Ink and color on silk; 81.7x57.6cm; Seoul Museum of History In general, yeongdang and sawoo housed a portrait of one person. However, some shrines housed a number of portraits of different people, such as the portraits of Heo Mok and Che Jae-gong enshrined together at Dogang yeongdang. In most cases, portraits of the literati depict a man in a Confucian scholar’s robe and hat. However, towards the end of Joseon, an increasing number of saedaebusang came to resemble gonsinsang, dressed in formal official costumes and hats. Many such portraits of figures in formal outfits were housed in shrines. Of all such portraits, the Portrait of Choe Ik-hyeon (1833~1906) housed at Chaesansa Shrine is the most outstanding example (Plate 15). The portrait is one of the patriot portraits created by Chae Yong-sin, and the patriot who fought against the Japanese is portrayed full-length in the portrait wearing a deep blue official robe and a black hat for his spirit of independence. An official emblem with a crane design decorates him seated facing front in a chair placed on a tiger skin on the floor. The costume is so realistically depicted that it deceivingly invites a touch. The portrait looks as if Choe is right before us in person. ( Plate 15 ) Portrait of Choe Ik-hyeon, Chae Yong-shin, copied, early 20th century; Ink and color on silk; 97x53cm; Chaesansa Shrine (Choe Jong-gyu Collection) Most literati portraits were produced to be enshrined or worshipped but there was also a genre of literati portraits produced by a literati-painter himself as a piece of art. Self-portraits and various albums with full-length portraits belong to this genre. Records on self-portraits appeared as early as the Han period of China. The earliest record of a Korean self-portrait is that of King Gongmin in the 14th century Goryeo. It was allegedly painted based on a reflection in a mirror. Regrettably, the self-portrait does not exist today. There were two self-portraits of Kim Si-seup (1435~1493) in Joseon according to Maewoldangjip (The Collected Works of Maewoldang). No further records exist on self-portraits until the eighteenth century when self-portraits of Yun Du-seo (1668~1715), Yi Gwang-jwa, and Gang Se-hwang appeared. It seems that self-portraits were rarely painted in Joseon because the painter must first think himself valuable enough to be portrayed and have strong social recognition to support the self-respect before making a portrait in addition to an excellent skill set for realistic representation while professional court painters did not have adequate self-esteem as their social status was very low and most literati painters were not technically capable of drawing themselves. Of all these rare self-portraits, the Self-portrait of Yun Du-seo (Plate 16) is a true masterpiece. It is comparable to any other masterpiece self-portrait in the world. Only the face in ink fills the paper canvas and the artist looks so bold and firm as if he is confronting himself for some reason. This self-portrait is particularly impressive in its shrewdness in expressing the artist’s penetrating gaze. It truly shows the attitude of Yun Du-seo as the most talented realistic painter. The painting style is typical of the time in a way that numerous brushstrokes were made to express shadowed parts in his face. The pupils of his eyes are shown to be as clear and as distinctive as those of an immortal. When his friend Sim Deuk-gyeong died, Yun Du-seo painted his portrait in memory of him. The portrait the artist made was so true to life that the surviving family of Sim Deuk-gyeong burst into tears. This episode tells us how talented Yun Du-seo was in realistic expression as a literati painter. ( Plate 16 ) Self-portrait, Yun Du-seo, 18th century Joseon; Ink and color on paper; 38.5x20.5cm; Private Collection, National Treasure No. 240 Joseon also has excellent examples of portraits filled with the spirits of modern day artist. The most representative example is Im Hui-su jeonsin hwacheop, a book of portrait paintings by Im Hui-su. Im Hui-su was a prodigy artist who died at the age of only seventeen. He created a series of sketches of guests to his house in light ink or charcoal (Plate 17). Although these sketches are not completely finished, they caught the essence of subjects so spontaneously as well as objectively being free from any restraints or pressures that the professional portraitists of the time might have been subject to. Being lightly sketched, his portraits allow us to see the process of his creation as well. ( Plate 17 ) Im Su-ryun from the album of paintings by Im Hui-su, Im Hui-su, 1749; 23.8x11.5cm; National Museum of Korea There are also a few sadaebusang portrayed in very unique composition and form. Jeongsiksang is one of them and it was produced based on an anecdote: When a subject was accompanying King Sejo (1417~1468) on his way to Onyang, a fire broke out and the subject rescued the king by carrying him out of the danger on his back. Another is the Portrait of Ju Do-bok. It depicts his when King Yeongjo passed away in 1776 (Plate 18). The other is Three Brothers of the Jo Family, a group portrait of three brothers depicted side by side on one canvas (Plate 19). ( Plate 18 ) Portrait of Ju Do-bok, Anonymous, Joseon circa 1776; Ink and color on silk; 113x57cm; Private Collection ( Plate 19 ) Three Brothers of the Jo Family, end of the 18th century; Ink and color on silk; 42x66.5cm; National Folk Museum of Korea 05 PORTRAITS OF WOMEN Portraits of women are the least conspicuous category of Korean portrait painting. The earliest evidence of such portraits was found in the murals of Goguryeo tombs, including Tomb No. 3 in Anak, Sasinchong Tomb (Tomb of the Four Deities) in Maesan-ri, and Ssangyongchong (Tomb of the Twin Pillars). The faces of the women are all accompanied by men who appear to be their husbands. As was pointed out earlier in the introduction, these portraits of women on murals cannot be paintings of real people and they are rather stylized figures that were repeatedly drawn in a certain pattern. There are no records of women’s portraits from the Unified Silla period. During the Goryeo period, however, a number of queens’ portraits are said to have been done. In particular, the portrait of Princess Noguk, wife of King Gongmin, is said to have been painted by King Gongmin himself. Also, in 990, the sixth year of King Seongjong, a man named Son Sun-heung is said to have painted a portrait of his deceased mother and performed an ancestral ritual in front of it. He was commended by the king for the conduct but the quality was not so great. During the early Joseon period, portraits of queens were painted together with those of the kings and housed in Seonwonjeon Shrine within a palace. However, there is no record of the production of portraits of queens after the Japanese invasions of 1592~1598. The Annals of King Sukjong mentions an incident relating to a portrait of the queen. In August 1964, King Sukjong ordered Kim Jin-gyu, a literati painter, to paint a portrait of his wife, Queen Min, but the ministers and the artist himself all strongly objected.22 The Confucian ethical code forbade the company of males and females over the age of seven and it was unthinkable for a queen to sit for a male painter, a mere subject of her husband, the king. King Sukjong eventually withdrew his instruction and no portraits of queens were produced thereafter. In early Joseon, it was popular among the literati to commission portraits of husbands and wives together as was the case with portraits of the king and queen. This was a custom handed down from Goryeo. Among the extant examples of this genre are paintings of King Gongmin and his wife, Princess Noguk; Jo Ban and his wife; Bak Yeon, a music scholar in the court of King Sejong, and his wife; and of Ha Yeon and his wife (Plate 20). Records on the portrait of Jeong Do-jeon, a literati-bureaucrat of early Joseon, and his wife are found in Gwon Geun's Yangchonjip (Anthology of Yangchon) and Jeong Do-jeon’s Sambongjip (Anthology of Sambong). However, as these portraits are all copies and the colors are seriously faded, we can hardly appreciate the originals. Only the historical point of view on costumes and outline of the painting style can be vaguely discerned through these portraits. ( Plate 20 ) Portrait of Wife of Ha Yeon, Anonymous; Ink and color on silk; 146x93cm; Private collection It is noteworthy that a few portraits of women by Chae Yong-sin who was very active from the end of the Joseon period through the Japanese occupation period extant today. The Portrait of Un Nangja (also known as Portrait of Choe Yeon-hong) (Plate 21) is a portrait to commemorate Choe Yeon-hong who acted very bravely during the revolt of Hong Gyeong-nae in 1811. The woman here is not a real model but an ideal image of a woman seen by the artist. ( Plate 21 ) Portrait of Choe Yeon-hong, Chae Yong-sin, 1914; Ink and color on silk; 120.5x61.7cm; National Museum of Korea 06 PORTRAITS OF BUDDHIST MONKS The last group of portraits is portraits of Buddhist monks. One might think portraits of Buddhist monks preceded the development of any group discussed above. However, we cannot find any historical references to the production of portraits of Buddhist monks during the Three Kingdoms period. It is highly likely that monks who had been to China brought back portraits of revered masters, which would certainly have motivated Silla people to produce portraits of their own monks. Documents from the Goryeo period refer to portraits of monks highly revered during the Three Kingdoms period, supporting the assumption mentioned above.23 Various documents indicate that portraits of Buddhist monks were actively produced in the Unified Silla. The inscription on the tombstone of Jingam Seonsa at Ssanggyesa Temple refers to Yukjo Yeongdang (portraits of six masters) while Dongmunseon (Anthology of Korean Literature), the portrait of Haejo Guksa; Jodangjip, the portrait of Beomil Guksa (810~889); the Tombstone of Sinhaeng Guksa at Dansoksa Temple, the portrait of Sinhaeng Seonsa; and the Tombstone of Jingong Daesa (855~937) at Biroam Temple, the portrait of Doui. During the Goryeo period, portraits of Buddhist monks were a common artistic form thanks to the rising influence of Zen Buddhism. Zen emphasized the guidance of teachers rather than the worship of a specific Buddha or bodhisattvas as a way to attain enlightenment. Zen spread rapidly since the late Unified Silla period and portraits of revered masters were accordingly produced and enshrined in temples across the country. Complimentary salutations found on these portraits are included in various literary anthologies.24 In Ikjejip (Anthology of Ikje) by Yi Je-hyeon (pennamed Ikje), a literati scholar at the end of Goryeo, is an article under the title, Salutation to the Portrait of Yi Guksa of Songgwangsa Temple. It paid tribute to the portrait painted by the order of the king. In Gajeongjip, an anthology by Yi Cheom, are writings in praise of a portrait of a revered priest of the Cheontaejong Sect. All these records indicate that the literati enjoyed writing salutations to portraits of revered Buddhist monks during the Goryeo period. This may be indicative of friendship between the literati and monks and the stylistic exchange between the portraits of Buddhist monks and the literati portrait paintings. Among the Goryeo portraits of masters extant today, noticeable are those of Bojo Guksa (1158~1210) and Jingak Guksa, (1178~1254) both enshrined in the Hall of National Preceptors at Songgwangsa Temple (Plate 22). (In January 1995, sixteen portraits of national preceptors, including the one of Cheongjin Guksa housed in this temple, were stolen). The portrait of Daegak Guksa (1055~1101) is enshrined at Seonamsa Temple in Suncheon, Jeolla-do Province and portrait of Gakjin Guksa (1270~1355) is housed at Baeggyangsa Temple in Jeolla-do Province. However, both works have lost much of their original value because of excessive restoration work. Few portraits of Buddhist monks have been preserved in their original state. Repeated restorations and copying have been necessary because of smoke damage from the incense and candles burned in Buddhist shrines and because of constant exposure to temple visitors. Nonetheless, the stylistic forms common at the time are still readily apparent in these portraits. ( Plate 22 ) Guksadang in Songgwangsa Temple, Suncheon, Jeollanam-do Portraits of Buddhist monks continued to be actively produced throughout the Joseon period despite the suppression of Buddhism by the Confucian rulers. There are several types depending on the subject’s pose. The typical example of the first style is the Portrait of Master Muhak (1327~1405), enshrined at Tongdosa Temple (Plate 23). The subject is portrayed full-length and seated in a chair as was the customary form of the late Goryeo period. The subject is seen from the side in a three-quarter view, holding a staff in one hand and gripping the arm of his chair with the other. His feet are placed on a low stool. The Buddhist robe is depicted in strongly contrasting complimentary colors and the contours of the master’s face are seen clearly probably because the subject did not wear a hat. The second type of monk portrait is exemplified by the paintings of the renowned Master Chaewol (Plate 24). The subject is also seen from a three-quarter view at full-length but he is seated on a cushion and holds a staff in his left hand. The right hand either holds Buddhist beads or rests on the subject’s knee. ( Plate 23 ) Portrait of Master Muhak, Uiyun, Joseon 1807; Ink and color on silk; 146.7x76.7cm; Tongdosa Shrine Museum ( Plate 24 ) Portrait of Buddhist Priest Jaewol Daesa, late Joseon; Ink and color on silk; 103x79cm; National Museum of Korea As pointed out before, few portraits of Buddhist monks have been preserved in their original state. Repeated restoration and copying have been necessary because of smoke damage from the incense and candles burned in Buddhist shrines and because of constant exposure to temple visitors. These portraits also suffer stylistically because they were generally painted in mountain temples secluded from the outside world. Thus, the artists were not influenced by the developments in poses, background, brushwork, and coloring techniques of the time. More recently, an increasing number of portraits of Buddhist monks produced in the late Joseon in the style of folk painting have been discovered. There are also group portraits of Buddhist monks. At Daeheungsa Temple and Jikjisa Temple are portraits that depict a number of Buddhist monks on single canvas (Plate 25). In China, literati scholars of the same society used to produce group portraits during the Ming and Qing periods but a group portrait of Buddhist monks were never produced elsewhere but in Korea. Group portraits of Buddhist monks are worth our attention because they are a genre unique to Korea. ( Plate 25 ) Group Portrait of Monks, Anonymous, late Joseon; Ink and color on silk; 112.3x227.4cm; Jikji Museum of Buddhist Arts III CHARACTERISTICS OF KOREAN PORTRAIT PAINTINGS As discussed, Korean portrait paintings are divided into six categories depending on the social status of the subjects. Each category has important social functions and meanings: Eojin as a symbol of everlasting prosperity of the dynasty; Gonsinsang to enhance the power of the monarchy and to prevent any disobedience; Girodosang to commemorate the respectful; Sadaebusang for scholars to be enshrined in seowon or yeongdang; portraits of women produced until discontinued under the prevailing Confucian ethics; and portraits of monks to be enshrined in Buddhist temples. This section summarizes the overall characteristics of Korean portraits. Most of Korean portraits (in a narrower sense the Joseon portraits because most of the portraits extant today were from Joseon) were produced to be enshrined and to be respected. The figures were drawn as symbolic objects to which tribute was to be given and almost all of the figures were high status male scholars who had worked for the dynasty. Portraits of women were rarely produced except in early Joseon because of the Confucianism doctrine that prohibits women from being seen and drawn by portraitists (because they were predominantly men). Not a single portrait of a child has ever been discovered. In China and Japan, most portraits are of grown-up men as was the case in Joseon. However, portraits of young women (Plate 26, 27) as well as old women (Plate 28) were produced in large numbers in both countries. In Japan, portraits of children who died young were often produced after their deaths to cherish them in memory (Plate 29) but no portraits of children were produced for enshrinement or ritual ceremonies in Korea. ( Plate 26 ) Portrait of a Concubine of Emperor Yongzheng, early 18th century China; 184x98cm; Palace Museum, Beijing ( Plate 27 ) A Woman Temple Prayer, 1582 Japan; Ink and color on silk; 75.6x31.9cm; Ryoanji Temple, Kyoto, Japan ( Plate 28 ) Tokugawa Ichihime, 1610, Japan; Ink and color on silk; Temple, Kyoto, Japan ( Plate 29 ) Portrait of Noble Ladies, Anonymous, late Qing China~early 20th century; Ink and color on silk; 156x98.2cm; Private Collection, Korea. The fact that most of Korean portraits were produced to pay respects for the deceased and not to appreciate their artistic beauty or to reflect the figures in the portraits is the most powerful element in shaping Korean portraiture. Firstly, most Korean portraits were hung on the shrine wall (Plate 30). Only the eldest son of the main family was to make a respectful bow before the portrait in the shrine and the other descendants were to bow outside the shrine. This customary rule was the same in every type of shrine ritual. Only significant scholars could enter the shrine to bow and the others were confined to the courtyard in front of the shrine. For this reason, Korean portraits were large enough for everyone to see them afar off. ( Plate 30 ) Portrait of Choe Chi-won, Housed in Museong Shrine Secondly, in Korean portraits, only one person is shown. There are no other elements that could distract viewers’ attention. It seems that Joseon portraitists knew how to make a portrait effective when it was to evoke the dignity of the portrayed figure. Most portraits in China are also only of one person. However, there were many cases in the Ming and Qing dynasties in which this rule was not followed. Family portraits that depicted a group of family members from different generations together (Plate 31) and group portraits of a teacher and students or friends (Plate 32) were common. Maids or servants were included in some portraits in addition to the main characters to emphasize the status of the main characters (Plate 33). In Portrait of Emperor Go-Daigo (reigned 1318~1333) of Japan, the emperor is portrayed with his subjects (Plate 34). ( Plate 31 ) Four Generations of the Family, China; Ink and color on paper; 164x86.4cm; Nanjing Museum, China ( Plate 32 ) Ten Elderly Gentlemen; Ink and color on paper; 37.2x135.4cm; Nanjing Museum, China ( Plate 33 ) Portrait of a Ming Scholar, Anonymous, repainted, Ming and Qing China; Ink and color on silk; 150.5x102cm; Private Collection, Korea. ( Plate 34 ) Emperor Go-Daigo, 14th century, Japan; Ink and color on silk; 131.8x77.3cm; Daitokuji Temple, Japan Joseon portraits depict only one scholar seated in a room alone without any background objects while many Chinese portraits depict a scholar relaxed and strolling with a servant or maid in a beautiful and scenic place (Plate 35). In Chinese portraits, various vessels or pieces of furniture such as a bookshelf and writing desk were added to show the social status, background, or tastes of the portrayed (Plate 36). ( Plate 35 ) Jiang Shun Fu, end of the 15th~early 16th centuries, Qing China; Ink and color on paper; 161x67cm; Nanjing Museum, China ( Plate 36 ) Seven Elderly Gentlemen, Lin Fuchang, Qing China; Ink and color on paper; 40.6x146cm; Nanjing Museum, China Korean portraits are not diverse in terms of style. They follow some patterns or formula without any dramatic pose or expression. In almost all portraits, the subject is seen from the side in a three-quarter view and very static pose (Plate 14). No dynamic pose is used in Korean portraiture. All Korean portraits depict the subject from the same angle and same gaze as seen in Portrait of Yi Jung-no (Plate 6), Song Si-yeol (Plate 14), and Choe Ik-hyeon (Plate 15). Portraits of Japan are different from those of Joseon as demonstrated in the Portrait of Shinran (1173~1262) (Plate 37), a monk who lived in Japan in the 13th century and the Portrait of Ikkyu Sojun (1304~1481) (Plate 38) in the collection of Tokyo National Museum. Japanese portraits express the spirit of the subject through the eyes. All Joseon literati portraits depict a scholar with his hands hidden in sleeves and folded in front of him, regardless of social status. In contrast, Chinese portraits from the 17th century and onwards increasingly depict subjects in many more stylish poses, such as one hand on the knee and the other hand holding a belt. Portraits in such stylish poses remained very popular until the end of Qing. With an active relationship with China during the reign of King Sukjong and onwards, Korean envoys to China had their portraits painted by Chinese painters and brought them back home to Korea, like the Portrait of Yi Gwang-jeong (1552~1627) (Plate 39). Through these paintings brought from China, the typical style of Chinese portraits was introduced to Korea characterized by a chair with a tiger skin underneath, frontal view, pose with one hand holding a belt and the other on the knee, and leather shoes on the shoe pedestal placed in “八” form. Among these features, a chair with a tiger skin underneath and leather shoes placed in “八” form were commonly used by Korean portraitists until late Joseon. The Chinese-style frontal view was used for a while but the traditional quarter view of Korea soon returned. The Chinese pose of hands was used by only a few painters and soon disappeared.25 These changes in portrait styles are valuable evidence that the preferences of Joseon people were different from the Chinese of the Ming and the Qing. It seems that Joseon literati did not like to look exaggerated or showy in their portraits. ( Plate 37 ) Shinran, 14th century, Japan; Ink and color on silk; 120x81.2cm; Nara National Museum, Japan ( Plate 38 ) Itkyu Sozun, latter half of the 15th century, Japan; Ink and color on paper; 43.8x26.6cm; Tokyo National Museum, Japan ( Plate 39 ) Portrait of Yi Gwang-jeong, Anonymous, 1602, China; Ink and color on silk, hanging scroll Such inclination is apparent in album portraiture for aesthetic appreciation. Even in album paintings (a genre in which relatively freer expression was possible), the subject is shown to be austere and dignified. For example, Gang Se-hwang (Plate 40) left a number of self-portraits and all these self-portraits are the same as the Portrait of Gang Se-hwang painted by a professional court painter (Plate 41) in terms of facial expression, angle, and posture as well as the tense look. The tense look is always there unchanged in all portraits of Gang Se-hwang. This characteristic is different from those of Japan and China. In his self-portrait, Okada Beisanzin (1744~1820) (Plate 42) captured an image of the relaxed scholar after becoming very drunk and comically expressed the state while Kitagawa Utamaro (1753~1806), a famous ukiyoe artist, expressed his humor by depicting himself in one of his own artworks (Plate 43). Ren Xiong (1823~1857), a Chinese literati painter, showed his boastful attitude toward the world in his self-portrait. All these examples of Japanese and Chinese self-portraits are in strong contrast with Korean’s. ( Plate 40 ) Self-portrait of Gang Se-hwang, 1782; Ink and color on silk, hanging scroll; 88.7x 51cm; Private collection ( Plate 41 ) Portrait of Gang Se-hwang, Yi Myeong-gi, 1783; Ink and color on silk, hanging scroll; 88.7x51cm; Private collection ( Plate 42 ) Self-portrait, Okada Beisanzin, Portraits of Scholar Painters, Art of Japan, Sibundo ( Plate 43 ) Self-portrait, Gitagawa Utamaro, Portraits of Scholar Painters, Art of Japan, Sibundo Thirdly, the fact that portraits in Joseon were to be used exclusively for ancestral rituals clearly dictated the approaches of the portrait painters in producing portraits. As portraits were meant to be used for ceremonial purpose and models for their descendants, painters should express not only the physical appearance but also the spirit of the subject. Of course, all portrait paintings of the East and the West aim to express the spirit of the subject. Even so, what the spirit implies or the meaning of the spirit differs from one society to another. For Korean portrait painters, the meaning of the spirit never did refer to the individual characteristics or inclination of their subjects. The portraits were designed to evoke in the viewer both positive memories and deep reverence for the subject and painters did their best to present their subjects in the best possible light. Joseon portraits depicted the subjects as the respectable figures that the viewers would want to remember and to worship rather than as they were at a certain moment or certain place. As we have a clear, foggy, rainy, and dark stormy day but the ideal day is a clear day, Joseon portrait painters thought that they should depict their subjects in the most dignified and decent appearance. To attain the ideal aspect, painters studied the social background as well as personality and moral character of their subjects. Is he a scholar or a military official? Is he a scholar living a secluded life? What is his social class? It was only natural, therefore, that the portraits of kings were meant to symbolize the supreme authority of the dynasty and painters did their best to portray the majesty of their ruler in audience with his loyal subjects. To serve the purpose, it is repeatedly stated in Seungjeongwon Ilgi (Dairy of the Royal Secretariat) that the frontal view was considered the most desirable. Meritorious subjects were most often portrayed as the embodiment of nobility and dignity and models for their descendants. As such, they were depicted in full-length, seated in a chair, and dressed in official robes. Paintings of literati emphasized the subjects’ intellectual character and they were seated, dressed in official robes or Confucian robes. What should be noted is that the so-called ‘desirable nature’ and ‘idealized traits’ are depicted all the same way in all types of portraits produced in Joseon regardless of whether they were portraits of kings, meritorious subjects, literati, or the elderly. As pointed out earlier, there were no differences in expressing the desirable nature of the subjects between literati bureaucrats or military officials, or between the king and subjects. Back in the Joseon period, the viewers wanted to see a tranquil and self-restrained gentleman in a portrait because such traits were regarded as the most virtuous of all. Therefore, both the portraits of Yi Jung-no who rendered distinguished military service (Plate 6) and Yi In-yeop (1656~1710) who was a typical literati-bureaucrat (Plate 44) were the same man in portraits meditating in tense and emotionless postures. In Joseon portraiture, there is no such dynamism and vigor as depicted in the Portrait of Ashikaga Dakauzi (1305~1398) (Plate 45). ( Plate 44 ) Portrait of Yi In-yeop, Anonymous, late Joseon; Ink and color on silk; 150.7x86.5cm; Gyeonggi Provincial Museum ( Plate 45 ) Portrait of Ashikaga Dakauzi, mid~14th century Japan; Ink and color on silk; Kyoto, Japan Fourthly, the descriptive style of Joseon portraiture is another important element making it unique. Throughout the Joseon period, portrait paintings evolved with the passage of time. In the early Joseon period, contours were drawn with thin double lines with the eyes, ears, mouth, and nose being defined with lines. Faces and folds of robes were not colored in early painting. In the mid Joseon period, the so-called five distinguished areas of the face were colored in very light red to denote highs and lows on the face (Plate 46) and folds of a robe were drawn with lines and minimal coloring around the folds, as can be seen from the details of the Portrait of Yi Jung-no (Plate 47). A painting technique to use repeated layers of brush strokes to give a dark feeling to hollow cheeks as well as to apply fewer brush strokes to give feelings of lightness to swollen areas of the face was used (Plate 48) in late Joseon. Painters studied physiognomy to better understand the general features of the human body. After studying the general structure of both the bones under the skin and big wrinkles on the skin (Plate 49), individual bone structures and wrinkles on the face were depicted to express the substantiality of the subject. ( Plate 46 ) Physiognomy example for painters ( Plate 47 ) Portrait of Yi Jung-no (in detail) ( Plate 48 ) Portrait of Choe Ik-hyeon (in detail), copied, Chaesansa Shrine ( Plate 49 ) Reference material for portrait painting Despite the changes discussed above, however, Joseon painters were devoted to depicting their subjects as realistically as possible. Verisimilitude was regarded as the key element in Korean portrait. Portraitists always worked with the conviction that even the most minor details such as pockmarks, freckles, and spots on the face should be depicted as they are. As already pointed out, the gaze of eyes is of the same angle as the face on Joseon portrait. If the face is in frontal view the eyes also look straight ahead. The eyelids carefully drawn in thin lines, the pupils painstakingly depicted to the finest detail, and the wrinkles carefully tinted in light red are the typical characteristics of Joseon portraits. The most representative example of Joseon’s realistic expression is the beards (Plate 50). Joseon painters regarded the beard as part of the face and each strand of the beard was rendered in fine lines of black and white applied alternately on skin-toned face. Such exquisite brush strokes for the beard cannot be found in portraits of any other countries in the world attesting to the fact that Joseon professional painters faithfully followed the principle of portrait painting: ‘It does not make a real portrait if only one strand of hair is incorrectly drawn.’ ( Plate 50 ) Portrait of Yi Chae (in detail), 1807; Ink and color on paper; 99.2x58cm; National Museum of Korea Joseon portrait painters did not attempt to show their subjects to be more than what they truly were. They strictly avoided any distortion or change for that reason. Nor did they pursue exaggeration by emphasizing noticeable features. Joseon painters had only one thing on their minds: how to depict their subjects as realistically as possible. Joseon portraits therefore look rigid without any variation in expression or any awkward element of personal touch. They are, for this reason, all similar in quality without any noticeable differences in depicting skills. The beauty of Korean portrait lies in this extreme realism and the expression exuding from it. This exhibition, therefore, should focus on revealing how our ancestors perceived portraits. When they appreciate and cherish a portrait, they regard the portrayed as the epiphany of the real person, not as a work of art. This perception of Joseon people on portraits is best understood in how they treated the portraits of the king. One record states that, in times of war or crisis, even the lowest ranking official strived to keep eojin safe from a disaster, and the king and the ministers of the dynasty used to cry desperately before the enshrined eojin of the deceased kings. When royal portraits were destroyed by fire, all the governing officials including the king himself, the princes, princesses, and the concubines were required to dress in white (the funerary color in Joseon) and lamented for three days as if the royal family members depicted had really passed away. When disasters like a fire, the falling of trees, and heavy rain or snow occurred around the place where eojin were housed, Joseon held rituals to comfort the portrayed.26 The rules and the customs in relation to eojin clearly demonstrate that how royal portraits were perceived by Joseon. Because of this perception of Joseon on portraits, whether they were members of the royal family or commoners, a greater number of traditional portraits have survived until today enabling us to analyze or appreciate the characteristics and the beauty of Korean portrait paintings.