What was Ch’ilsŏngdae 칠성대 and why is it controversial?

I am still investigating this myself, but this turns up every now and then and I think I ought to clarify my view on the matter. Ch’ilsŏngdae is controversial in academic discussions as hard evidence is no longer extant. Despite the fact that its actual location is still debated, one can find seven stylized memorials scattered around the old city. I should note that these current signs that mark the alleged location of Ch’ilsŏngdae are likely incorrect (and of course a terrible waste of Cheju taxpayer money!) and that we actually do not know what the complex looked like exactly.

What was Ch’ilsŏngdae 칠성대 (The Seven Stars ritual complex)?

First off, “Ch’ilsŏngdae” seems to ahve been a complex of 7 altars arranged in the form of the Big Dipper asterism in which Samsŏnghyŏl 삼성혈 served as the North Star Polaris 북극성. While its primary function was likely ritual, if one were to look at a map of old Cheju City from 1914, one would notice that settlement concentrations seemed to follow the pattern of the Big Dipper. While Ch’ilsŏngt’ong 칠성통 street was the handle, the more densely populated area south of Kwandŏkjŏng 관덕정 was the scoop (see Nemeth David 1984 and Kim T’aeil 2012). It may seem far-fetched that a civilization as tiny as T’amna could conceive of something so complex, but not a few small island cultures had highly sophisticated geographical and astronomical knowledge (consider, for example, the remarkably accurate stick charts of Marshall Islands peoples). That being said, one should also be cautious in assuming the precise location of the “handle” altars in Ch’ilsŏngt’ong because Ch’ilsŏngt’ong street was in fact modified during the Japanese colonial period.

Mun Mubyŏng’s (2012) reading of the Ch’ilsŏng Ponp’uri demonstrates that the myth of the Ch’ilsŏng snake deities correspond to seven sites within Cheju City as well as seven sites within a Cheju traditional house. The Ch’ilsŏng Ponp’uri appears to recall a time when Ch’ilsŏngdae existed. Although Mun Mubyŏng hypothesizes that the Ch’ilsŏngje was originally a major ritual festival for T’amna as a whole, there is no clear description of such in what few classical records survive today.

The significance of the Big Dipper asterism is hardly unique to Cheju though the island’s tradition of associating it with snake deities certainly is. Seven Stars deities feature in Chinese Buddhist-Daoist tradition. On the Korean mainland, one can find a Ch’ilsŏng altar in Buddhist temples as well as particular prayer ceremonies to the Seven Stars deities. David Nemeth (1984, 1987) observes that the geomantic alignment with the Seven Stars was considered a source of power for the ancient Chinese Han Dynasty (206 BCE – 220 CE) emperors. Ch’ilsŏng worship in Korea was the result of a dynamic synthesis of local adaptations and Chinese Buddhist-Daoist elements. The fact that one can easily find Polaris with the aid of the Big Dipper asterism is also what makes it such a significant group of stars all across the world. All things considered, one can understand why some intellectuals in Cheju find it baffling that development planners choose to emphasize (if not outright exaggerate) this one aspect – shared commonly amongst Cheju’s neighboring cultures – out of many other aspects that are actually unique to Cheju.

Did Ch’ilsŏngdae exist?

We do have some documentary evidence that Ch’ilsŏngdae existed somewhere within Cheju Ŭpsŏng 제주읍성 (the original Cheju City). Scholars researching the matter cite the 15th century Tongguk Yŏji Sŭngnam 동국여지승람, Yi Wŏnjin’s 1653 T’amnaji 탐라지, the Chŭngbo T’amnaji 증보 탐라지 written in the 19th century, and Kim Sŏgik’s 1924 P’ahallok 파한록. Kim Sŏgik’s work and the Chŭngbo T’amnaji are of particular importance to us because these two records are some of the last glimpses we have of Ch’ilsŏngdae. Both works indicate a complex of seven sites within Cheju’s city walls. A 1926 newspaper article and photograph (also posted on those problematic Ch’ilsŏngdae signs) shows that rituals were still performed at Ch’ilsŏngdae in the first quarter of the 20th century. The complex disappeared perhaps sometime in the 1930s.

But this is where things get complicated. Although we do have documentary evidence and even a newspaper photograph, the locations of all seven sites is difficult to confirm. The different sources, even the work of Kim Sŏgik, who saw the actual Ch’ilsŏngdae, is vague. Given the lack of other supporting documentary sources on Cheju City’s layout, tracking the seven places listed in the extant record makes pinpointing their sites easier said than done. A recent attempt to use advanced GIS techniques to find a possible location for Ch’ilsŏngdae in relation to Samsŏnghyŏl has indicated that the current seven signposts seem to be far off the mark. On the other hand, these seven new points in recent research also show that interpretations of the extant documentary sources may need to be reexamined as they appear to contradict what has been written.

What of the Ch’ilsŏngdae restoration project?

In this matter, I have to express my agreement with the opinion that the attempt to rehabilitate and reconstruct Ch’ilsŏngdae is very problematic. We can say that there was a “Ch’ilsŏngdae” and that it was in old Cheju City but we cannot say for certain the exact positions of the seven altars. Until the matter can be resolved with more solid evidence – such as additional documentary sources, personal testimonies from the colonial period, and archeological finds – attempting to reconstruct Ch’ilsŏngdae based on speculation seems to be a wasteful endeavor.


“Paths of the Lost Kingdom” Walking Tour: Part 3

8. Ch’ilsŏngt’ong 칠성통:
Ch’ilsŏngt’ong is as old as it is new. What we see in this area today is the result of gentrification for the purposes of Cheju’s ever-expanding tourism. Ch’ilsŏngt’ong has seen the vicissitudes of Cheju City’s changes. From the Japanese colonial period (1910-1945) to the early 1980s, it was the city’s commercial downtown and once comparable to the bustling City Hall area of today. During both the period of Japanese rule and military dictator Park Chung Hee’s 박정희 reign (1960-1979), Ch’ilsŏngt’ong’s formerly winding olle-like pathways were straightened and widened to reflect the modernist city planning of Tokyo and Seoul. Straightened city streets also can serve a symbolic purpose in that they serve to remind local residents not only their dependency on economic systems tied to metropolitan centers but also the power and capability of ruling elites. Japanese shops and housing once were concentrated around here during the colonial period. Two of Cheju’s earlier theaters were also built along this road. Some of Ch’ilsŏngt’ong’s oldest shops that have survived the changes are the jewelry shops. This area experienced a serious downturn with Cheju City’s haphazard development in the 1980s, especially with the centers of activity moved toward Sin-Cheju and a continued failure to maintain infrastructure, but has since experienced a major commercial revival. Results of Ch’ilsŏngt’ong’s revival are mixed – on the one hand, commerce has returned to the area but on the other hand Cheju’s brand of gentrification in which unique locales are effaced to attract corporate (and nowadays Chinese) investment for short-term returns is posing a serious problem.

In ancient times, there was a complex of seven altar or shrine-like sites called “Ch’ilsŏngdae.” This is noted in several Chosŏn Dynasty and early twentieth century texts including the 1530 Sinjŭng Tongguk Yŏji Sŭngnam 신증동국여지승람, the 1653 T’amnaji 탐라지, and the 1924 P’ahallok 파한록. From a 1926 colonial photograph, one can see that rituals for Ch’ilsŏngdae were performed even to the beginning of the twentieth century. The sites associated with Ch’ilsŏngdae were arranged in the form of the Puktu Ch’ilsŏng 북두칠성 asterism, literally “the Seven Stars of the Northern Ladle” or what is known to us in the West as “the Big Dipper,” a group of stars part of the larger Ursa Major (Big Bear) constellation. The Big Dipper asterism has worldwide prominence due to its traditional usefulness for navigation – from the Big Dipper one can locate Polaris, the North Star. In East Asian astrology, the Big Dipper’s cosmological importance is also due to its unusual consistency in its apparent orbit around Polaris throughout the year (Nemeth, 1987: 164). At the height of each season, the ladle of the dipper seems to point a cardinal direction vis-à-vis its position in relation to Polaris. “Ch’ilsŏng” also has a special microcosmic significance in Cheju traditional spiritual practices in the worship of snake and household deities associated with these seven stars.

The Ch’ilsŏng Ponp’uri 칠성본풀이 narrates the strange story of Ch’ilsŏng’agi 칠성아기. According to the myth, an aging couple prayed at a temple for a child and the goddess Ch’ilsŏng’agi was the result. When both of her parents had to return to heaven to serve in the celestial government, they left her in the care of their maidservant. Ch’ilsŏng’agi, curious about the outside world, pestered her maidservant to allow her to go out. She got lost while venturing out and in the fields a spiritual master seized her seven times. Her body became like silk though she retained her face and she was placed under the horse-mounting block of her house. When her parents came back and discovered what had transpired, they put Ch’ilsŏng’agi inside a crate and cast it into the sea. The box reached the town of Hamdŏk 함덕 where seven women divers discovered it and found seven snakes within. Terrified of what they had found, the seven women divers contracted spirit sickness and had a ritual exorcism performed. But because they appeased the snake spirits, they eventually became very prosperous. The story also continues to say that some generations later, when the descendants of these women divers failed to perform a ritual at the appropriate time, they were forced to flee Hamdŏk and into Cheju City. But because Cheju City’s gates were locked, they had to sneak in via small holes. They emerged from the holes as snakes and these snake deities took up seven positions within the walled city. Hence the seven snakes became guardians of seven points of the house and seven points of Cheju City.

Folklorist Mun Mubyŏng 문무병 (2012) suggested that the now mostly domestic “Ch’ilsŏngje” 칠성제 rites are survivals of T’amna’s ancient state ritual. As hinted in the Ch’ilsŏng mythology, while Ch’ilsŏng deities, children of the goddess Ch’ilsŏng’agi, took up positions around domestic space as protectors they also took up positions around ancient Cheju City. Within the space of a traditional Cheju house, one Ch’ilsŏng deity takes the position between the kitchen and main elevated-floor room (sangbang 상방), one presides over the rear door of the main house (ankŏri 안거리) as the Patch’ilsŏng 밭칠성, one protects the storage room (kop’ang 고팡) as the Anch’ilsŏng 안칠성, and the other four are scattered around the corners of the house. As the domestic Ch’ilsŏng deities are protectors of the household, they traditionally were entreated whenever people performed ancestral rites. Ch’ilsŏngdae’s rituals were perhaps done with a similar intent but at a much larger scale for protecting the community. Aside from the spiritual purposes, Cheju’s shamanic rituals also doubled as festival-like events.

The actual locations of the “Ch’ilsŏngdae” sites are unknown and continue to remain a matter of debate to this date. Although historical records all appear to agree that such a complex existed and the two T’amnaji texts appear to give almost identical locations (Kim, 2012: 157), the exact locations are not consistent. Kim Sŏgik, who directly witnessed Ch’ilsŏngdae’s eventual destruction by the Japanese colonizers, offers convincing testimony on the locations, but even his 1924 record is vague. The clearest description we have is that of the position behind Kwandŏkjŏng 관덕정. More recently architecture professor Kim T’aeil 김태일, using descriptions of Ch’ilsŏngdae in Chosŏn Dynasty records such as the Chŭngbo T’amnaji 증보 탐라지 and Kim Sŏgik’s 김석익 1924 P’ahallok and GIS analysis proposed a possible arrangement in which the asterism’s ladle extends into the Ch’ilsŏngt’ong area while the scoop opens southeast below Kwandŏkjŏng. Following this proposition, from the point of the ladle’s tip, one could trace a line to Samsŏnghyŏl 삼성혈, which would represent Polaris. In 1984, American geographer David J. Nemeth at Toledo University suggests that the present Cheju Mokgwanaji 제주목관아지, the former provincial government complex, is possibly within the dipper portion where the concentration of geomantic-astrological energy is strongest. Nemeth noted the prevalence of Ch’ilsŏng-based geomancy in East Asia as far back as the Chinese Han Dynasty period. Nemeth’s proposition would also allow for one to trace the tip of the ladle to Samsŏnghyŏl, but his earlier estimations are possibly innacurate given that the locations noted in the P’ahallok are not near the present Mokgwanaji site. Archeological discoveries from civilizations all over the world from Stonehenge to the pyramids of Giza have demonstrated that ancient peoples were very well aware of astral positions out of necessity (and the convenient fact that their night sky view was far clearer than it is today), thus T’amna people were likely no less competent in astral observations. On the other hand, because of the multiple possibilities for configurations and the vague nature of extant records, Ch’ilsŏngdae will continue to puzzle scholars.

9. Mugŭnsŏng 무근성:
Mugŭnsŏng is one of the possible locations of the old T’amna fortress, T’amnasŏng 탐라성, as mentioned in the Chosŏn Dynasty Chŭngbo T’amnaji record where it is also called “Kojusŏng” 고주성 and “Chinsŏng-tong” 진성통. One way to interpret its name is that it refers literally to the “old fortress,” which would indicate an older prior settlement. If the T’amnaji’s estimations are correct, then that would mean that the previous Tamna capital formed around Pyŏngmunch’ŏn 병문천 stream (the Chosŏn Dynasty walled town of Cheju-ŭpsŏng 제주읍성 formed by Sanjich’ŏn 산지천 to the east). Place names are not always accurate means to determine historical locations as there are several places around Cheju that have a name meaning “old fortress,” but the vicinity of Mugŭnsŏng nonetheless has a long history. In both the Chosŏn Dynasty and Japanese colonial period, officials and more well-to-do people resided in Mugŭnsŏng. Near Kwandŏkjŏng on a small alleyway street is a small Japanese-style ryokan where officials once stayed during the colonial period. Even today, Mugŭnsŏng retains some of the older-style of winding narrow alleyways that once characterized much of Cheju City as well as a handful of surviving thatched-roof structures. Narrow streets and packed clustered neighborhoods called “kol” 골 were features of old Cheju City. The streets were widened during the Japanese period and in the 1960s-1970s, but with the exception of the major roads of Kwandŏk-ro 관덕로, Chung’ang-ro 중앙로, Puksŏng-ro 북성로, and T’apdong-ro 탑동로, street patterns in Cheju seem to perpetuate much older antecedents – the majority of the streets in a 1914 map of old Cheju City are still consistent today.

Cheju’s particular style of settlements and housing developed over the course of millennia of experience with the island’s geographical particularities. The traditional use of mud for thatched homes were intended to serve as a form of interior climate control – in the winter earthen walls could better retain heat while in the summer they remained relatively cool. Cheju’s style of thatched roof and rock walls were also made specifically with the dangers of fierce winds in mind. The center of Cheju City itself was a short distance away from the sea. Settlements moved further inland during the early Chosŏn Dynasty because of the constant threat from pirate raids, but even during the medieval period they kept some distance from the shoreline. The issue of T’apdong Plaza, which is ravaged whenever the island is hit with a substantial typhoon, illustrates one major reason why this is the case. Islanders were perfectly aware of what a typhoon was capable of doing, especially when it came to storm surges, something that artificially reclaimed land and breakwaters are ill-equipped to resist. Prior to T’apdong Plaza’s construction in the 1990s, the coastline of old Cheju City consisted of basalt rock formations that served as a natural barrier against waves and storm surges.

At least until the 12th century, T’amna had an unusually ambiguous relationship with the Koryŏ Dynasty. T’amna was considered subordinate to the Koryŏ kingdom but nevertheless developed in a different manner. While some suggest that T’amna was already subsumed into Koryŏ by the beginning of the dynasty, others claim that T’amna persisted until the Chosŏn court under King T’aejong 태종왕 finally abolished T’amna native titles in the early fifteenth century. Historian Roh Myoung-ho 노명호 (2005), however, suggests that Tamna instead appears to have served as a kind of vassal state and became fully a district of the larger Koryŏ kingdom only following its 1105 designation as T’amna Prefecture. In a recorded event in the 2nd year of King Hyŏnjong 현종왕 (1011), T’amna is said to have requested that the Koryŏ king regard T’amna in the same manner of other higher-level districts of the Koryŏ kingdom – chu 주 (provinces) and hyŏn 현(counties). At the same time T’amna operated as a foreign country. In the Koryŏsa and Koryŏsa Chŏllyo records, one can note that historians appear to have continued to regard T’amna as a foreign state until the 12th century. In any case, T’amna was subordinate to Koryŏ as it periodically dispatched tribute goods and sent requests for royal investiture. In the year 1044 in the Koryŏsa, T’amna’s Sŏngju (king of T’amna) sent a request to the Koryŏ court to select the future ruler when it was apparent that the prince would have no heirs. Other hints that T’amna was a vassal state were records of Koryŏ dispatching its own officials known as “Kudangsa” 구당사 to observe T’amna affairs. What did come to T’amna via this relationship were access to both the land and maritime trade networks in which the Koryŏ capital Kaegyŏng 개경 (today’s Kaesŏng 개성) was connected, some measure of prestige for T’amna rulers, and cultural imports such as Buddhism. While T’amna sent representatives to attend Koryŏ’s P’algwanhoe event, Buddhism in T’amna was directly sponsored by the Koryŏ court in the form of significant temples such as Mansusa 만수사 and Haeryunsa 해륜사, which flanked Jeju City’s old center. This trend continued also during the period of Mongol rule from the late 13th to late 14th century.

10. Pak Family Thatched House 박 씨 초가

At around three hundred years of age, the Pak family thatched house is one of the oldest buildings in Cheju City. The person living here is currently the eighth generation caretaker. The chief reason for this house to have such remarkable longevity despite all the change occurring around it is because of the family’s deep religious faith in geomancy. The house is said to be at a perfectly auspicious site. While the interior has undergone modernization, the general shape, form, and layout of the house retains Cheju traditional elements.

Unlike the Korean mainland, in Cheju tradition the “front gate” is not actually the front gate of the house but the front door of the Ankŏri, the main house (Kim HJ, 2007: 182). The most important ritual date for Munjŏnje, the door god rite, is the 15th day of the first lunar month, but the Munjŏn deity 문전 is often acknowledged in domestic rituals. Whenever any ritual is performed, offerings are given to Munjŏn. Another unique feature of Cheju’s mythology is that the kitchen god “Chowang” 조왕 is identified as female. As a daily practice, women of Jeju households once made daily offerings to Chowang by placing a bowl by the kitchen stove.

The story of Munjŏn and his mother Chowang is elaborated in the Munjŏn Ponp’uri 문전본풀이. Multiple versions exist, but the following is one in which the deities correspond to the order of the traditional Cheju house. Two deities, Nam Sŏnbi 남선비 and Yŏsan Puin 여산부인 had seven sons. Because the family was deeply impoverished, Nam Sŏnbi engaged in rice trade abroad. Nam Sŏnbi was seduced by Noiljedegwi’il’s 노일제데괴일 daughter, who bamboozled him into squandering his earnings. Yŏsan Puin searched for her husband and dispelled the enchantment put upon him by feeding him rice. Kwi’il’s daughter, unwilling to relinquish control over Nam Sŏnbi, drowned Yŏsan Puin when she pretended to treat her to a bath. She used her powers to assume the form of Yŏsan Puin and returned with Nam Sŏnbi. All save for the youngest of the seven sons were deceived by Kwi’il’s daughter. The youngest son exposed Kwi’il’s daughter’s plot to kill them. In shame, Kwi’il’s daughter hung herself in the toilet and she became the toilet goddess. The disgraced Nam Sŏnbi committed suicide at the chŏngnang of the house compound and became the Chŏngjumok-chŏngsalji 정주목정살지 deity. To revive their slain mother, the seven sons traveled to the Flower Fields of the Western Heaven to retrieve the flower of reincarnation. The mother Yŏsan Puin was reincarnated as the hearth goddess Chowang. The youngest son who exposed Kw’il’s daughter became the Munjŏn deity while the five elder sons became the Obang T’osin 오방토신 (Earth Deities of the Five Directions) and the sixth son became the protector deity of the rear gate.

11. Kwandŏkjŏng 관덕정 / Cheju Mokgwanaji (Government Complex) 제주목관아지:
Kwandŏkjŏng is said to be the oldest extant structure in Cheju City. Although other structures of a similar function existed at the site of Cheju Mokgwanaji during the Koryŏ Dynasty and T’amna Period. In 1416, the Chosŏn court under King T’aejong, reorganized Cheju Island’s administrative districts into Chejumok 제주목, Taejŏng-hyŏn 대정현, and Chŏngŭi-hyŏn 정의현. Kwandŏkjŏng, which literally means “the Pavilion of Observing Virtue,” was first built in 1448 during the reign of King Sejong 세종왕 with the intention of serving as a command post for drilling military forces. The two words “observing virtue” related to archery and the Confucian philosophy of archery – in order for an archer to perfectly strike a target the archer must have perfect discipline, a concept that also translates into conceptions of proper conduct. The presence of Kwandŏkjŏng and the adjoining government complex called “Mokgwanaji” were manifestations of the Seoul government’s power in – and power over – Cheju.

T’amna’s decline under the combined pressures of outside forces and possible mismanagement by its own native elites was long and slow in coming. Whether or not this began as early as the beginning of Koryŏ is debatable, but by the mid-Koryŏ Dynasty the status of T’amna vis-à-vis the mainland had fallen far indeed. In 1229, the name “Cheju” appears in historical records when the Koryŏsa mentions Chinese merchant castaways reaching the shores of Cheju. But this name seemed to have been interchangeable with T’amna and it is actually not until more than six decades later that the name appears more frequent. The nature of things changed in 1295 when both the Koryŏ and Mongol Yuan courts wrangled over direct control of the island.

The exact time of T’amna’s demise as an autonomous kingdom remains unknown but there is no doubt that by the time of the Chosŏn Dynasty 조선왕조 (1392-1910) Cheju Island was a de facto colony. In 1404, during the reign of King T’aejong, following the request of Cheju’s native elites in 1402, the Chosŏn court changed the former Sŏngju and Wangja titles into Chwadojigwan (左都知管) and Udojigwan (右都知管), the Provincial Managing Officer of the Left and the Provincial Managing Officer of the Right. In the early part of the Chosŏn Dynasty, Cheju appeared to have maintained much of the old T’amna structures and it did not seem that native elites cared particularly much for Seoul court titles while they were more interested in maintaining their local clout (Go 2007: 165). Chosŏn records note that local elites abused their power over commoners, but this could also be a pretext for the central authority’s intent to eventually break their power and establish some semblance of centralized rule. Since the founding of the Chosŏn Dynasty, central authorities had sought to break the influence of local elites known as “hyangni” 향리. In 1445, during the reign of King Sejong, Cheju’s native titles were finally abolished.

The vicissitudes of T’amna can serve as crucial lessons for Cheju today. Cheju’s complicated and troubled relationship with the Korean mainland has been long in the making as the island was at first a tributary vassal state and eventually became a quasi-colony. While we cannot be certain if mainlander accounts of the misdeeds of Cheju’s local elites can be trusted (as people of the Korean elite yangban class had it in their interest to denigrate local leaders and hyangni), their twilight years in the early Chosŏn Dynasty could serve as a crucial warning to Cheju’s current elites. T’amna remains relevant to us today as its traditions and history allow us to make sense of how Cheju as a culture and society came to be.



Duncan, John. 2000. The origins of the Chosŏn Dynasty. Seattle: University of Washington Press.

Ko Changsŏk. 2007. T’amnaguk sidaesa. Seoul: Seoul Munhwasa.


Hyŏn Yongchun. 1996. Chejudo sinhwa. Seoul: Sŏmundang.

Jiu Tangshu.

Cho Tongil. 1997. Tong Asia kubi sŏsasi-ŭi yangsang-kwa pyŏnch’ŏn. Seoul: Munhwagwajisŏngsa.

Chosŏn wangjo sillok.

Kim Iru. 2004. “Koryŏ sidae T’amna jumindŭl-ŭi kŏju jiyŏk-kwa haesang hwaldong,” Hanguksa hakbo, 18: 9-35.

Kim T’aeil, Kang Mun’gyu, Kim T’aehwan, et al. 2012. Chŏnhwan’gi-ŭi Cheju tosi kŏnch’uk. Seoul: Tosŏ ch’ulp’an pogosa.

Mun Mubyŏng. 2012. “T’amna komunhwa-wa Ch’ilsŏng sinang,” Minjok mihak, 11(2): 15-50.

Nemeth, David J. 1987. Architecture of ideology: Neo-Confucian imprinting on Cheju Island, Korea, Berkeley: University of California Press.

Pettid, Michael J. 2001. “Vengeful gods and shrewd men: response to the loss of sovereignty on Cheju Island,” East Asian history, 22: 171-186.

Yi Wŏnjin. 2002. Yŏkju T’amnaji. Kim Chanhŭp, et al, editors. Seoul: Purŭn Yŏksa.