Next Article in Journal
Community Cultural Wealth among a South Australian Muslim Population: A Reflective Evaluation of Strengthening Family Health and Wellbeing
Next Article in Special Issue
The Efforts of Government-Driven Reform of Both State and Personal Rites in Early Chosŏn: A Historical Shift from Spiritual Efficacy to Meritorious Practice
Previous Article in Journal
Collective Despair and a Time for Emergence: Proposing a Contemplacostal Spirituality
Previous Article in Special Issue
Authority over Death: Two Tales Related to the Lord of Mount Tai in Konjaku Monogatarishū
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Visits to the Palace of the Sea God in Ancient and Medieval Japan

Department of Languages, Literatures, and Cultures, University of Florida, Gainesville, FL 32611, USA
Religions 2024, 15(3), 350; https://doi.org/10.3390/rel15030350
Submission received: 26 December 2023 / Revised: 1 March 2024 / Accepted: 7 March 2024 / Published: 13 March 2024
(This article belongs to the Special Issue The Supernatural in East Asia)

Abstract

:
Visits to the palace of the sea god are a recurring theme in premodern Japanese narratives, and comparing these stories across time periods reveals shifting perceptions of the supernatural world. The earliest sources for narratives of travel to the palace of the sea god in Japan date from the eighth century, most notably in the stories of Luck of the Mountain and Urashima Tarō. In these stories, the descriptions of the sea god’s palace, the relationship of the sea god to the natural world, and even the location of the palace were tied to eighth-century understandings of kingship, weather, and geography. Later adaptations of these stories incorporate features of Buddhist geography, Chinese architecture, Buddhist narrative motifs, and even an alternative vision of supernatural time. These alterations occurred because the eighth-century narratives required adaptation to fit a new political, social, and religious reality. This comparison demonstrates that our vision of the supernatural world is inextricably tied to our understanding of the natural. When our fundamental grasp of the nature of reality changes, our imagination of the supernatural transforms in turn.

1. Introduction

The palace of the sea god appears early and often among the other worlds of premodern Japanese narratives. Fictional visitors arrive at the palace by design or by chance, and readers are treated to a fantastical description of supernatural beings and an otherworldly built environment. The structure of the story, in which a visitor goes to the palace, receives a boon, and returns to their homeland, is fundamentally the same across sources. Yet even for materials written in the same period, how the visitor arrives at the palace of the sea god and whether it is underwater or above ground differs. Moving into later periods, the type of structures described, the identity of the sea god, and the nature of the space itself have different descriptions. Because the narratives are variations on a single core story, comparison reveals shifting perceptions of the supernatural world in bold relief. These shifts in the character of the supernatural reflect changed understandings of the natural, and the variation demonstrates that our vision of the supernatural world is rooted in our grasp of the natural one.
The dichotomy of natural and supernatural that frames this article presents two immediate problems, both observed over a century ago. Firstly, in definition, both words are exceptionally slippery. Drown observed that “The road is indeed paved with ambiguities. Perhaps no word in common philosophical or theological use is more full of them than the word ‘nature’, and ergo, the word ‘supernatural’” (Drown 1913, p. 144). And though Drown endeavored to outline a comprehensive definition, here we must be satisfied with one created specifically for the purpose of writing this article. The second problem is that “natural” and “supernatural” are anachronistically applied to early Japan. As Durkheim pointed out, “the idea of the supernatural, as we understand it, dates only from to-day” (Durkheim 1915, p. 26). Durkheim argued that the concept of a natural order of things by which the supernatural distinguishes itself is a product of modern science.1 That which appears “supernatural” to us moderns was, for people in earlier times, “beautiful, rare or terrible spectacles, or causes of surprise and marvel (θαύματα, mirabilia, miracula); but they [early peoples] never saw in them [supernatural phenomena] glimpses of a mysterious world into which the reason cannot penetrate” (Durkheim 1915, p. 27). As such, I stress that “supernatural” is used in this article as an analytic category, created for our use and not meant to describe the ontology of early Japanese people.2
In this article, “supernatural” is defined as the non-quotidian and used to refer to the rare occasion of traveling to another world in early Japanese narratives. Attention to other worlds as they appear in literature has a long history in the academy, with special attention to the topic given in Patch (1950) and over Patch’s long career. Patch’s scholarship, like that of many early and mid-twentieth-century folklorists, was largely taxonomic, involving the collection and categorization of medieval narratives detailing travel to another world. The ultimate objective was the identification of common motifs and themes across both regions and time periods. More recent scholarship on other worlds in medieval literature such as Byrne (2016) is more restricted in period and more attentive to the narrative functions of other worlds. Rather than attempt a comprehensive definition of “otherworld”, Byrne focuses instead on generic markers such as boundaries and temporal distortions that signal the existence of an other world, a supernatural space, to the reader. Drawing on Byrne’s approach, analysis here will center on the palace of the sea god, a recurring motif that is clearly identified to the reader as a named otherworldly space through its relative inaccessibility, its unusual denizens, and its alternative temporality, among others. As the supernatural is defined by the natural, identifying the characteristics of this exceptional other world reveals the everyday worlds of early Japan.
The earliest sources for narratives of travel to the palace of the sea god in Japan date from the eighth century, referred to in Japanese studies as the ancient period. The most notable such stories are Luck of the Mountain and Urashima Tarō.3 These accounts appear in mytho-historical texts like the 712 Kojiki 古事記 (An Account of Ancient Matters) and the 720 Nihon shoki 日本書紀 (Chronicles of Japan), the eighth-century poetry collection Man’yōshū 万葉集 (Collection of Myriad Poems), and the gazetteers known as Fudoki 風土記 (Records of Wind and Earth).4 In these stories, the descriptions of the sea god’s palace, the relationship of the sea god to the natural world, and even the location of the palace were tied to eighth-century understandings of kingship, weather, and geography.
Medieval (12th–16th c.) adaptations of these stories incorporate features of Buddhist geography, Chinese architecture, Buddhist narrative motifs, and an alternative vision of supernatural time. The Hikohohodemi no mikoto emaki 彦火火出見尊絵巻 (Picture Scroll of Hiko-ho-ho-demi, hereafter Hikohohodemi), a medieval picture scroll, retells the story of Luck of the Mountain from Kojiki and Nihon shoki, but situates the story within a Buddhist worldview. Buddhist influence is even more pronounced in the Noh play Tama no i 玉の井 (The Jeweled Well), which abstracts a short part of the Luck of the Mountain story for stage performance. Finally, the folk story (otogi zōshi 御伽草子) of Urashima Tarō retells the narrative from eighth-century sources, but in a new medium and with a new destination for the hero, the palace of the Dragon King. The Urashima Tarō story was also recounted in court poetry during the Heian (794–1185) period and adapted into the Urashima Shrine (Urashima jinja) foundation story (engi 縁起). These intermediate adaptations provided valuable components for the story’s ultimate conversion into the folk story that is now its most widely known instantiation.
Existing research on these Japanese narratives tends towards two general approaches: folklore and mythology, and religious history. Scholars of myth and folklore have identified numerous parallels between narratives of travel across and under the sea such as the breaking of a taboo and marriage to a maiden from another world. Early comparative mythology persuasively linked these narratives to Indonesian, and later Micronesian and Polynesian, myths of supernatural travel.5 Comparison with other Japanese mythical narratives, such as the conquest of the Korean peninsula by Empress Jingū, introduces further connections like the birth of a child prince and supernatural powers over the sea. Folklorist Yanagita Kunio argued that stories of journeys to the palace of the sea god referred to visiting the nether world. Mythologist Mishina Shōei linked these narratives to a broader collection of stories that ultimately demonstrated Northeast Asian influence on Japanese myth.6 Akima (1993), the boldest treatment in English, built on Yanagita and Mishina to argue that these stories are historically structured narratives with travel and return from a foreign land paralleling death and resurrection. Two points observed by Akima: the inconsistent location of the palace of the sea god and Taoist influence on the Urashima story form the jumping-off point for this study. However, Akima does not perform a diachronic analysis of the source material.
A second approach, used in the history of religions, focuses on identifying the distinct features of medieval Japanese myths. One centralizing concept for this approach is “chūsei Nihongi” or “medieval chronicles”, a term coined by Itō Masayoshi in 1972. Nihongi traditionally referred to the text Nihon shoki, and in the medieval period, expanded to include many myths and legends claimed to originate with this text or derive from it. Often, however, there was no connection between these medieval myths and the 720 text Nihon shoki, at least based on the methods of evidentiary scholarship. The seeming spuriousness of references to Nihongi in medieval sources led early post-WWII Japanese scholars to regard medieval materials as erroneous.7 Itō (1972) argued that these medieval myths contained their own theoretical logic. Later scholars insisted, like Yamamoto (1998), that the medieval chronicles constitutes its own field of study. This later work largely concentrated on how Buddhism changed the nature of divinity (kami) in Japan.8 However, the thrust of this work was primarily theological in orientation, not geographical or cosmological.
Recently, scholarship such as Rambelli (2018) and Moerman (2021) addressed this lacuna by demonstrating that Buddhism introduced not only a new worldview to medieval Japan, but an actual new world. Moerman (2021) observed a set of features that defined the geography of the Buddhist world. Most important of these was the continent Jambudvīpa, one of four continents on each side of Mt. Sumeru, which stood at the center of the universe. Japan was located on the edge of Jambudvīpa, with this continent usually depicted centering on India. This geographical formation was not native to Japan, and as argued in Felt (2023), medieval Japanese attempts to reconcile Buddhist geography with Japanese myths required creative adaptation. Spatial continuity was a particular issue. For example, differences between Chinese and Japanese myths seemed to indicate that the sun and moon in China and Japan came from different sources, either the giant Pangu (China) or the kami Izanagi (Japan). One proposed interpretation was that there were in fact two suns and two moons, one for each realm. Yet citing the unity of Jambudvīpa, critics insisted that the sun and moon must be the same in both China and Japan. Put differently, in the Buddhist worldview, our lived reality constituted a single discrete space. Other worlds, of which there are no shortage in Mahayana Buddhism, were traversable only with supernatural help: Jizō fetching sinners from hell, Amida whisking the departed to the Pure Land, etc.
This article applies the geographical perspective and diachronic approach of recent religious historians to the seeming inconsistency of the location of the palace of the sea god identified by Akima. In ancient Japanese sources, the underwater world was not clearly under the surface of the ocean, and it was reachable via a defined route or passage. There is comparatively little description of the built environment or passage of time in the underwater world, and its ruler is identified as the god of the sea (Jp. Wata-tsu-mi 海神). In medieval depictions of these ancient narratives, the sea god begins to be identified as the Dragon King. The narratives incorporated Buddhist ideals of kingship and karmic retribution. Travel to these locations required supernatural intervention and became instantaneous, suggesting that the underwater realm was an alternative reality, not a contiguous realm. The built environment took on more pronounced otherworldly characteristics, and description of the unusual passage of time in the other world became a fixed narrative feature. These transformations resulted from the diffusion of a Buddhist-informed worldview and the perception that other worlds were physically other spaces not reachable by conventional means.

2. The Underwater World in Ancient Sources

The most famous example of a visit to the underwater world in eighth-century Japanese sources is the story of two brothers, the younger Luck of the Mountain, also called Ho-ori, Yama-sachi-hiko, and Hiko-ho-ho-demi, and the older Luck of the Sea, also called Ho-deri or Umi-sachi-hiko. The older brother had a talent for fishing, and the younger brother had a talent for hunting. On one occasion, the two brothers suggested exchanging their talents, and so the older brother gave the younger his fishhook, and the younger brother gave the older his bow and arrows. However, neither was successful in their new venture, and the younger brother lost the fishhook at sea. When pressed for its return, the younger brother forged replacement fishhooks, but the older would not accept them and demanded his original fishhook.
Forlorn, the younger brother wept by the seaside. A deity known as Old Man Shio-tsu-chi, meaning “spirit of the sea currents”, then came to the younger brother and hatched a plan to retrieve the original fishhook by sending the younger brother to the palace of the sea god, Wata-tsu-mi. There, the younger brother asked for aid in finding the original hook, and he also married the daughter of the sea god, Toyo-tama-hime, and received a pair of magical jewels that controlled the tide. Upon his return, the younger brother used the jewels to force his older brother into submission, and his children with the daughter of the sea god were the ancestors of the Japanese imperial line. Luck of the Mountain is the grandfather of the mythical Jinmu, first emperor of Japan. Luck of the Mountain’s natural dominance of the mountains combined with his mastery of the seas through marriage and cooperation with the sea god present the image of a ruler with power over both land and sea in his realm. The Luck of the Mountain story appears in one version in Kojiki and in five versions in Nihon shoki.9
The most unusual feature of the underwater world as it appears in eighth-century sources is that it is not consistently underwater. While its location is concretely identified, the prose of Kojiki suggests an above-ground location. When the god of sea currents meets the younger brother on the beach, he
… made a small boat of tightly woven bamboo, put the younger brother in the boat, and then instructed him, “When I push this boat out into the current, just go along with it. I believe there is a good and honored course for you. If you follow that route, there will be a palace that looks to be made of fish scales, the palace of the sea god”
The key features of this account are the tightly woven bamboo, which would prevent the boat from sinking, the role of the current, and the use of the words “honored course” 御路 and “route” 道. The tightly woven bamboo boat is clearly meant to float, not to sink. The sea god’s instructions to “go along with it” mean that the younger brother should not attempt to row or paddle or otherwise change the course of the boat, but they also suggest traveling atop the water by drifting along with the current. The “honored course” pays deference to the younger brother and recognizes his supremacy, foreshadowing his ultimate victory. “Course” and “route” also suggest established passages or thoroughfares, and “follow that route” combined with the notion of not paddling implies that the palace of the sea god is somewhere across the seas.
This above-ground location of the palace of the sea god is consistent with language used at the end of the Kojiki episode, when the daughter of the sea god, Toyo-tama-hime, comes ashore to give birth. She informs the younger brother of her pregnancy and asks him to build a hut on the beach where she can give birth. When she enters the hut, she instructs the younger brother not to look inside, but he breaks the taboo and sees that she has changed into a sea monster. Shamed, Toyo-tama-hime resolves never to see the younger brother again. She tells him, “I had intended to regularly traverse the sea route to come and go. However, you have seen my true form, and I am deeply embarrassed.” Then she blocked off the sea slope and returned” (Kōnoshi 1997, pp. 136–37). Here, the use of “traverse” 通 again suggests an established pathway, the “sea route” that leads from the palace of the dragon king to the central reed plain land where the younger brother dwells.
“Sea slope” 海坂 is more difficult to parse, but likely refers to a boundary or border between the world of the sea gods and our own. Elsewhere in Kojiki, the god Izanagi visits Yomi, the land of the dead, and on his return, he blocks off the “slope of Yomi” 黄泉坂 with a boulder, suggesting that 坂 refers to a boundary between natural and supernatural worlds (Kōnoshi 1997, pp. 47–49). Etymologically, the vernacular Japanese “saka” means slope and this is also the meaning of the Sinograph, but “saka” is probably related to “sakahi” (modern Jp. sakai 堺) for “border.” Kōnoshi (1986) has persuasively argued that in the case of Yomi, not only does the “slope” refer to a border, but Yomi itself, the land where Izanami goes when she dies, is not underground. As such, we can take “sea slope” to refer to the border between two worlds located on the surface of the Earth separated by the sea. The word “sea slope” also appears in Man’yōshū IX:1740: “When he rowed along/passing the sea slope/by chance/he rowed and met/he daughter/of the sea kami” (Satake et al. [1963] 2002, p. 217). The subject of the poem is Urashima Tarō; the repeated use of the verb “row” 漕 demands that the “sea slope” refer to a location on the surface of the ocean.
An underwater location for the palace of the sea god is clearly imagined in the main version of Nihon shoki and its four variants of this narrative.10 In the main version, when the god of the sea currents meets the younger brother on the beach, he made a “basket with no holes in it, put Hiko-ho-ho-demi inside the basket, and sank it into the sea” (Sakamoto et al. 1994–1995, vol. 1, pp. 158–94). Obviously “sank” 沈 must refer to a location under the sea, not across it, and the basket is presumably made airtight for the younger brother’s journey. This usage is consistent with Variants 1 and 3, in which the god of the sea currents “sank it [the basket]” 沈之 and Variant 3, in which the boat built by the god of the sea currents “sank of its own accord” 自然沈. In Variant 4, the younger brother “rides him [a shark] into the ocean” 乗彼入海, but it is not immediately apparent if this is meant down into the ocean or simply out to sea. Furthermore, in no version of Nihon shoki does the god of the sea currents say anything to the younger brother about a route or passage, nor does he give him any instructions about rowing or wayfinding.
After being submerged in the ocean, the younger brother immediately finds himself at the palace of the sea god, and there is no discussion of sea currents or being carried along. After Toyo-tama-hime gives birth, in Nihon shoki, she also “closes the sea route” 閉海途, implying some kind of passage between the world of the younger brother and the palace of the sea god, but the language is ambiguous as to the location of the palace itself. In order to preserve narrative consistency, the commentator in a recent edition of Nihon shoki, notes: “‘Sea route’ 海途 is glossed ‘umi no michi’; if it were ‘umi-ji’ or ‘umi-tsu-chi’ then it would refer to a passage on the ocean surface, but here the conception is of a path between the world above ground and a land within the ocean” (Kojima 1996, p. 161). In Variants 1, 3, and 4, Toyo-tama-hime does not close the sea route. Variant 4 suggests that the enmity between Toyo-tama-hime and the younger brother is the reason that creatures of the land and sea cannot pass into each other’s domains 此海陸不相通之縁也. That is to say, the reason that humans cannot survive underwater. Unlike Kojiki, in Nihon shoki, the palace of the sea god is clearly imagined as a world beneath the waves, not across them.
Luck of the Mountain is not the only visitor to an underwater world in eighth-century sources; Nihon shoki, Man’yōshū, and the gazetteer Tango no kuni fudoki 丹後国風土記 (hereafter Tango fudoki) all record the legend of the fisher Urashima Tarō, called “Urashima no ko.” In most versions of the story, Urashima caught a turtle at sea. The turtle turned into a beautiful woman, and Urashima traveled with the woman to another world. There he stayed for some time, and when he was ready to return, the woman gave him a box with instructions to never open it. Urashima returns to his hometown only to find that centuries have passed. He absent-mindedly opens the box he received, upon which he suddenly turns into an old man and realizes that the box contained the years that he had been gone. In eighth-century sources, the Nihon shoki version appears to be based on the Tango fudoki; Nihon shoki itself adds after a terse recounting of the story that “the story is in another volume,” presumably referring to Tango fudoki (Sakamoto et al. 1994–1995, vol. 3, p. 84).
The Man’yōshū version of the story is slightly different, and it demonstrates the same inconsistency about the location of the palace of the sea god as seen between Kojiki and Nihon shoki. The Man’yōshū version of the story appears in Book IX, poem 1740, a long-form poem or chōka 長歌, and poem 1741, a short-form poem or tanka 短歌. Both are about Urashima by the eighth-century poet Takahashi no Mushimaro (n.d.). The poem is too long to provide in full here, but three important features stand out in the Man’yōshū account. Firstly, as noted above, Urashima rows his boat across the border of the land of the sea, so his destination is ultimately atop the water, not underneath it. The Man’yōshū version of the story does not include a turtle; Urashima simply becomes so engrossed in his fishing that he continues rowing out to sea for an entire week. Secondly, after Urashima crosses the border, he meets “a daughter of the god of the sea”, and their ultimate destination ends up being the “divine palace of the god of the sea”. Thirdly, the location of the palace of the god of the sea is specified as “Tokoyo” 常代, where the two “neither aged nor died and stayed in this world for a long time.”
The Nihon shoki and Tango fudoki accounts of Urashima Tarō are similar on the first and third points, but not the second. Nihon shoki does not explain how Urashima Tarō reached his destination, only that he was riding in a boat fishing when he caught a giant turtle. Tango fudoki provides further detail: the turtle changes into a woman and explains that she is from a house of a sage of heaven, then Urashima rows the boat to her homeland. She herself describes coming to him by “riding the wind and waves”, and when they reach the border of the eternal land, she makes him close his eyes, then they arrive at an island “on the sea” (Uegaki 1997, pp. 473–79). In Nihon shoki, the pair arrive at Mt. Hōrai (Ch. Penglai) 蓬萊山; in Tango Fudoki, this is abbreviated to Mt. Hō 蓬山. Though the Sinographs indicate a different location than the everlasting land of Tokoyo, the vernacular gloss given for both is “the land of Tokoyo” (Tokoyo no kuni).11 That is to say, at least in the commentarial glosses, there is some conceptual overlap between Hōrai, the land of Taoist immortals, and Tokoyo. Tokoyo appears in several other places across eighth-century sources, but it is always portrayed as a land across the sea, not in the depths. For example, in Nihon shoki, the tiny kami Sukuna-bikona uses a millet stalk as a catapult to launch him across the sea to the land of Tokoyo. In the reign of Emperor Sujin, his vassal Tajima-mori travels by boat to Tokoyo in search of the fruit of everlasting life.
While Tokoyo appears multiple times in Kojiki and Nihon shoki, Hōrai is derived from Chinese sources. When and how the two lands became associated with each other in Japan is not clear, as the earliest eighth-century sources are ambiguous. However, Hōrai (as Penglai) has a rich literature in Chinese sources. In Japanese literature, it also features prominently in the Tale of the Bamboo Cutter, though in this narrative, it is depicted as unreachable. Tajima-mori’s quest for the fruit of everlasting life suggests that Tokoyo was connected to immortality, but on the other hand, elsewhere in Kojiki, it is depicted as a land of eternal night. This is also the strongest explanation for its nomenclature; “toko” meaning “always” and “yo” meaning “night”. Perhaps the safest explanation is simply that eighth-century Japanese had heard of three lands across the sea, Hōrai, Tokoyo, and the palace of the god of the sea, and conflated them.
These sources demonstrate that there were two imaginations of the palace of the sea god circulating in the eighth century. For Urashima Tarō, in the Man’yōshū, Nihon shoki, and Tango fudoki accounts, the couple travel to Tokoyo, a land that is across the sea that has a conceptual relationship to Hōrai through the characters used to render “Tokoyo”. However, neither Nihon shoki nor Tango fudoki indicate that the woman is associated with the god of the sea or that Urashima visits the palace of the god of the sea. This final point is given in the Man’yōshū account only. The above-ground location for the palace of the sea god in Man’yōshū accords with the description in Kojiki. However, Kojiki never suggests that the god of the sea lives in Tokoyo. In Nihon shoki, the palace of the sea god is under the sea, and based on Nihon shoki and Tango fudoki, the location visited by Luck of the Mountain is not the same location visited by Urashima Tarō.
While inconsistent on some of the details about the location of the palace and the relationship between the god of the sea and Hōrai, the descriptions of the palace itself and how the traveler arrived there are generally shared. The palace is described as “made of fish scales” in Kojiki (Kōnoshi 1997, p. 127). Nihon shoki is slightly more ostentatious; the exact words are provided in the next section but suffice to say in Nihon shoki the description is terse. The narratives do little to visually evoke the space, and this holds for the descriptions of Yomi, the Land of Ne, and Tokoyo in Kojiki and Nihon shoki as well. The lack of this description results in the purported “other world” not seeming particularly otherworldly. In a similar fashion, both Luck of the Mountain and Urashima Tarō travel to the palace of the sea god using conventional means. Even in Nihon shoki, which specifies that the palace was located underwater, the traveler arrives by riding the sea currents. In these narratives, all one needed to do to go to the other world was follow the way. Geographically and cosmologically, these other worlds were attached and accessible from our own, providing no god had intervened to block the way.

3. The Palace of the Sea God in Hikohohodemi No Mikoto Emaki

Adaptations to the story of Luck of the Mountain and Luck of the Sea are prominently visible in medieval accounts. The Hikohohodemi no mikoto emaki (hereafter Hikohohodemi), fuses the god of the sea with the Dragon King, depicts the palace of the sea god in Chinese fashion, and argues for an alternative ideal of kingship. Hikohohodemi was one of several picture scrolls commissioned by retired Emperor Go-Shirakawa at the end of the twelfth century. For Go-Shirakawa, these scrolls functioned primarily as propaganda that supported Go-Shirakawa’s claim to power during an era in which the ruling emperor, the retired emperor, and the prominent military clans Taira and Minamoto all jockeyed for supremacy. Go-Shirakawa commissioned scrolls such as the series Nenjū gyōji emaki 年中行事絵巻 (Picture Scrolls of Annual Events) and Hōgen sumahi zu emaki 保元相撲図絵巻 (Picture Scroll of the Sumo Festival from the Hōgen Era).12 While lost, the titles of these works suggest that Go-Shirakawa was interested in commemorating court events and rituals in which the emperor played an indispensable role, and by extension, implying that the emperor was the legitimate sovereign. Hikohohodemi, produced in the same studio as these other works, was written by the courtier Fujiwara no Norinaga (1109–n.d.), whose words fill the spaces between images in the scroll. Unfortunately, the original artist is unknown, and the original scroll is no longer extant, though later copies are generally assumed to be in the spirit of the original, especially regarding the text of the scroll and the composition of the picture scenes.
Norinaga changes the “sea god” to “Dragon King” throughout the story and uses “Dragon Palace” to refer to his palace. The picture scroll creates some level of suspense as the individual who sends the younger brother on his voyage does not tell him where he will be going, and it is only after the younger brother arrives, meets a series of women, and passes through a series of gates, that he encounters the owner of the palace, described as “within, there was someone he [the younger brother] thought was the master of the house.” Then, in the following scene, the two converse and Norinaga final identifies this figure: “the Dragon King listened, then summoned someone and issued orders to them” (Komatsu 1992a, pp. 104–6). From this point on, the figure, who had been identified as the “god of the sea” in Kojiki and Nihon shoki, is called only “Dragon King”.
Norinaga changes “sea god” to “Dragon King” under the influence of Buddhism. As is widely known, in East Asia, dragons are associated with rain and the powers of water, likely an influence from the Indian naga.13 The creatures live in pools, grottos, and other bodies of water. Also, like the naga, dragons function in literary narratives as either opponents of Buddhism who are tamed by virtuous monks, or as protectors of the Buddhist law and even the historical Buddha. In East Asia, this translated into a relationship between Buddhist specialists and rainmaking rituals. Monks, relying on the power of Buddhism, could call on the power of a dragon or dragon king to cause rain. Naturally, in agricultural societies, such power was of deep interest to the state, whose prosperity depended on regular rainfall. Norinaga, writing in the twelfth century when Buddhist institutions and ideas were widespread, makes a very natural substitution given that the sea kami was depicted, in Nihon shoki, living at the bottom of the ocean and possessing power over both the tides and the rains.
Norinaga’s shift to “Dragon King” reflects a shift in belief about the relationship of the ruler to Buddhist ritual practice. In the Nihon shoki, the sovereign serves as the most powerful of ritualists. For example, during a drought during the reign of Emperor Kōgyoku in 642 CE, first animals are sacrificed to the kami to no effect. Following this, sutra readings were performed at Buddhist temples, and a little rain fell. Then, finally, Kōgyoku prayed to each of the four directions, resulting in heavy rain (Sakamoto et al. 1994–1995, vol. 4, pp. 192–94). The ruler was the most powerful ritualist exercising power over the natural world. By the twelfth century, the rise in status of Buddhist institutions resulted in what Kuroda and Stone (1996) have called “mutual dependence of imperial law (ōbō) and Buddhist law (buppō)”. As Ruppert (2002) has argued, during this period, narratives relocating the dragon king of Anavatapta, biographies linking rainmaking rituals to Kūkai, founder of the Shingon sect in Japan, and the creation of a ritual regime among Shingon clerics, linked state interest and Buddhist ritual. Norinaga’s invocation of the Dragon King, who “wishes to see his daughter married to the prince of Japan”, recalls the ideal relationship between the ruler and the religious leadership the ruler patronized. For Go-Shirakawa, commissioner of Hikohohodemi, the invocation of the dragon king would remind religious institutions that no matter how grand the investments by aristocrats and warrior clans, the special relationship with the dragon king, effected by Shingon practitioners, was with the imperial line alone.
Norinaga also changed the plot of Hikohohodemi in order to exemplify the ideal relationship between sovereign and dragon king and imperial law and Buddhist law. In the original, Kojiki and Nihon shoki versions of the story, when the younger brother leaves the palace of the sea god, he returns to his own land, subjugates his brother, builds a birthing hut for Toyo-tama-hime, then has a falling out when he peeps inside as she gives birth. In Hikohohodemi, the younger brother receives the tide-controlling jewels from the Dragon King, travels back to Japan to subjugate his brother, then returns to the palace of the Dragon King. Then, the dragon king pronounces that the child should be born on land, builds a maternity hut on the beach, and brings his entire retinue with him for the birth. The younger brother peeks into the maternity hut during the birth, but there are no ill consequences, and in the image depicting the birth, Toyo-tama-hime is in human form. The breakup of the younger brother and Toyo-tama-hime, which in Kojiki and Nihon shoki explained the separation of the world of the sea god from the world of humans, has been completely removed.
Norinaga altered the plot because his vision of kingship was premised on cordial relations between the dragon king and sovereign, that is, a smooth interworking of imperial and Buddhist law. The original version of the story, in which Toyo-tama-hime resolves to never see the younger brother again, would not fit Norinaga’s vision. In the same vein, Norinaga makes the dragon king the primary agent of the story. The dragon king, not Toyo-tama-hime, asserts that the child must be born on land, and the dragon king, not the younger brother, builds the maternity hut.14 Norinaga converts the dragon king figure into an active character who participates in the safe continuation of the imperial succession.
In this vein, the most striking intervention by Norinaga on behalf of Go-Shirakawa is a closing scene that preaches the importance of offering regular tribute to the imperial court. The penultimate image of the scroll depicts the older brother, Luck of the Sea, with the facial hair of an old man and a young child pulling at his beard. The older brother appears to be ordering porters to prepare his tribute goods for shipment to the capital. The goods, melons and gourds, are arrayed before the older brother in his yard.15 Norinaga’s text explains, “The younger prince is now the emperor. As for the older prince, his estate is in the province of Yamato, in Yoshino district, and in accordance with his vow, he submits offerings every season” (Komatsu 1992a, p. 106). Then, the final image of the scroll shows the porter and his horse transporting the tribute across the rolling hills to the court of the younger brother. These final images are most striking because in the original versions of the story in Kojiki and Nihon shoki, while the older brother pledges to be the servant, or jester, depending on the version, of the younger brother, there is no discussion of tribute. Furthermore, as the older brother promises to serve at the gates of the palace of the younger, there is no association with Yoshino or suggestion that he was enfeoffed elsewhere in the country. Norinaga has invented a new ending for the story that reinforces fealty and resource transfer from provincial landholders to the imperial court.
The dragon king’s visit to the beach, depicted in stunning visual form in the picture scroll, illustrates a medieval vision of passage between worlds. In Kojiki, the younger brother reaches the palace of the sea god by riding the ocean currents, and in Nihon shoki, by sinking into the sea, either in a boat or basket provided by the god of the ocean currents. In Hikohohodemi, an unnamed old man meets the younger brother on the beach, and after hearing the younger brother’s story about the lost fishhook, tells the younger brother to get into a colander and close his eyes. In the space of an instant, the younger brother is transported to the palace of the dragon king. One possibility is that Norinaga misunderstood the passage in Nihon shoki in which the god of the ocean currents stopped up the holes in the basket; in Japanese, “basket” is a homophone with “eye”, and “stopped up” could also mean “cover, close” for eyes. However, the instantaneous transport suggests that the palace of the sea god was imagined, in Norinaga’s time, to be a truly other world, not reachable by any human means.
Similarly, when the dragon king comes to the beach for Toyo-tame-hime’s delivery, he “builds a bridge to the shore”. Kaneoka (2017) notes that analogous language appears in Wakasahiko Wakasahime daimyōjin himitsu engi 若狭彦若狭姫大明神秘密縁起, the temple origin tale for the Wakasa-hiko Wakasa-hime Shrine in Obama City, Fukui prefecture, in which Toyo-tama-hime crosses a bridge to reach Japan. Kaneoka argues that these cases are related to two factors. Firstly, the word bridge “hashi” changed in meaning. In ancient Japan, “hashi” referred to a vertical or horizontal connector and natural feature such as a log bridge or steppingstones. In later eras, “hashi’ referred to a built horizontal structure. To Kaneoka’s observations, we can add that for Norinaga and his patron Go-Shirakawa, it was important to emphasize that the Dragon King can be summoned to Japan to use his rainmaking powers. As such, it was appropriate for Hikohohodemi that the Dragon King had the power to create a bridge and that his domain was not, like in Kojiki and Nihon shoki, permanently cut off from Japan.
The narrative description of the palace of the god of the sea changes in Hikohohodemi, where the palace of the Dragon King is adorned with foreign treasures. In Kojiki, the palace looks like “fish scales”, suggesting a glimmer or sheen (Kōnoshi 1997, p. 127). In the main narrative of Nihon shoki, “the walls were neatly kept, and the roof of the tower gleamed like jade”(Sakamoto et al. 1994–1995, vol. 1, p. 160). In Variant 1, “the barbican was ostentatious and imposing, and the towers were magnificent” (Sakamoto et al. 1994–1995, vol. 1, p. 168). The Nihon shoki descriptions were certainly inspired by Chinese literature, but the description in Hikohohodemi outstrips all of the eighth-century sources: “When he [the younger brother] opened his eyes, he was speechless. The place was decorated with the seven treasures, adorned with lapis lazuli and crystal, and it shined so bright that he could not look upon it” (Komatsu 1992a, p. 104). The younger brother then goes into the palace, where there was “a middle gate. It was also adorned with lapis lazuli, and its glimmering sheen surpassed that of the first gate by ten times.” The younger brother passes through this gate as well, where inside he found “various palace buildings built in layers, each surpassing the middle gate by one-hundred times”. The clearest influence from Buddhism is the “seven jewels”, listed in the Lotus Sutra and the Infinite Life Sutra. These were known in Japan when Kojiki and Nihon shoki were compiled, as they appear in Man’yōshū V: 904. However, it was only in the medieval worldview that equated the sea kami with the Dragon King in which it made sense to apply them to the description of the palace building.
The long series of gates through which the younger brother passes is similarly absent in the original narrative, but a major feature of Hikohohodemi. Perhaps because of the picture scroll medium, the narrative devotes substantial attention to an area suited to visual presentation. Because the scroll is not an authentic twelfth-century original but rather a later copy, some circumspection is required when assessing the visual elements. However, as argued by Peter Glum, there are clearly two different architectural styles depicted in the picture scroll, and these are likely inherited from the original artwork. The second of these styles Glum (1988) notes is “modeled on Chinese prototypes, with heavy stone or tile pedestals, columns, and an intricate bracketing system of woodwork painted red, covered by heavy roof tiles” (Glum 1988, pp. 39–40). This architectural style is used for the palace of the Dragon King and matches depictions in later works that show events occurring in China. Again, some circumspection is warranted, but there is evidence that for twelfth-century Japanese, the idea of an “other world” had merged with that of the “foreign land”.
Four adaptations in Hikohohodemi to the ancient Luck of the Mountain narrative demonstrate Buddhist and foreign influence over the perception of the natural world. Firstly, the conversion of the sea kami into the Dragon King reflects a belief that kami associated with natural features had Buddhist equivalents. Given the syncretic Buddhist and kami worship traditions of medieval Japan, this dualism is not surprising on its own. However, it sets up a second, more striking adaptation to the nature of kingship, in which the legitimate ruler had an ideal relationship with the Dragon King and Buddhist ritual was firmly interlinked with imperial court ritual. Thirdly, the seeming instantaneous transport of Luck of the Mountain to the underwater realm and the bridge-building power of the Dragon King suggest that the creators of Hikohohodemi no longer conceived of geographic reality in terms of routes and passages, but rather supernatural transport. I address this change in more detail in the following section. Finally, the description of the palace itself suggests that within the Japanese medieval imagination, foreign lands were a reference point for depicting other worlds. Taken together, these adaptations demonstrate that for the Japanese elites who produced Hikohohodemi, the natural world was now governed by a new set of rules and rulers, and their powers and its geography permitted passage between realms that for ancient people had been permanently separated.

4. Other Medieval Accounts of the Palace of the Dragon King

Hikohohodemi was written in the late twelfth century, making it a very early medieval adaptation of the original story. Two related accounts of visits to the palace of the dragon king from the late medieval period are the Noh play Tama no i and the Shibukawa Urashima Tarō. The former was written by Kanze Nobumitsu (n.d.–1516), who wrote about thirty plays in his lifetime and was the great-nephew of the most famous Noh playwright Zeami Motokiyo (n.d.). The Shibukawa Urashima Tarō is part of a collection of 23 stories published by Shibukawa Seiemon (n.d.) around 1700 often called the “Companion Library” (otogi bunko 御伽文庫).16 Despite the late publication date, Miura (1990) notes that the original publication date of the 23 stories was probably earlier and that comparison between versions suggests that the Shibukawa version is in essence a late Muromachi-period (1336–1573) narrative. Following Miura’s lead, this paper will also consider the Shibukawa version as a late medieval source. Unlike Tama no i, the Urashima Tarō story had several different adaptations during the medieval period but given the number and variety of these and the disproportionate influence of the Shibukawa version, emphasis here will be given to the Shibukawa using the version in Ichiko (1958). For both the Shibukawa Urashima Tarō and Tama no i, Buddhist elements are more prevalent, and the “other world” becomes a discrete space with its own space/time.
Unsurprisingly given the three-hundred-year gap between the time of Hikohohodemi and the time of Tama no i, the plot of Nobumitsu’s Tama no i takes a very different shape than Norinaga’s Hikohohodemi. The accounts of Kojiki, Nihon shoki, and Hikohohodemi all give prominence to a water well on the grounds of the palace of the sea god/Dragon King. This well is where the younger brother and Toyo-tama-hime first meet, and it is the origin for the title of the Noh play Tama no i, or “the jeweled well”. Noh plays develop around the relationship of the shite, or main, and the waki, or supporting actor; often, the shite will change between Act 1 and Act 2 of the play. In Tama no i, the Luck of the Mountain plays the waki, Toyo-tama-hime plays the shite, and her sister Tama-yori-hime plays the zure or accompaniment in the first act. In Act 2, the Dragon King plays the shite in the second act with both Toyo-tama-hime and Tama-yori-hime as the zure.
The play opens with the younger brother announcing himself and a summary of the lost fishhook. The younger brother follows the instructions of the god of the sea currents, who is not depicted, arrives at the palace of the Dragon King, and meets Toyo-tama-hime and Tama-yori-hime at the jeweled well outside the palace gate. Inside the palace, the Dragon King promises to obtain the fishhook and to give the younger brother the two tide-controlling jewels. The younger brother marries Toyo-tama-hime. After three years, the younger brother asks about returning. Scallops and mussels throw a banquet for the couple, the two sisters dance, the Dragon King dances, then the younger brother and the two sisters ride a five-meter shark back to Japan, jewels in tow. In keeping with the conventions for Noh plays about kami, Nobumitsu celebrates the origins and story of Luck of the Mountain, who is accorded the status of a deity. The younger brother does not announce himself as connected with the imperial line, and the other characters praise him as the “descendant of the heavenly gods.”17 Since the play is ultimately meant to praise the younger brother and his marriage, demonstrated by the prominent addition of the clams and mussels, there is no breakup between the couple, and the play ends with the Dragon King returning to his palace.
Like Hikohohodemi, Tama no i associates the god of the sea with the Dragon King, but in the Noh play, the role of the Dragon King as a protector of the realm and the ideal relationship between the Dragon King and the emperor is more explicit. Nobumitsu usually refers to the character as “god of the sea,” but there are occasional references to the Dragon King such as the phrase “This is the Dragon Palace, palace of the god of the sea” (Nishino 1998, p. 479).18 In the second act, when the Dragon King appears to return the hook, he lays out the role of the two tide-controlling jewels and his daughter in terms of kingship and governance. “‘Since it may be that your older brother is still angry, I present to you the two jewels of the rising tide and the ebbing tide. Use them as you wish, and long may you govern your land’. Then he became the father-in-law of the imperial grandson and Toyo-tama-hime got pregnant”(ibid., p. 480). In Nobumitsu’s rendering, the Dragon King grants Luck of the Mountain power over water and unique connection to himself, the ultimate source of rainmaking. The Dragon King blesses Luck of the Mountain’s rule of his kingdom, and by identifying him as the “imperial grandson”, creates a pact with Luck of the Mountain while recognizing the legitimacy of the genealogical imperial line. Finally, the marriage and pregnancy to Toyo-tama-hime tie the Dragon King and the Japanese imperial line together by blood.
Geographically speaking, Tama no i suggests an underwater location for the palace of the sea god, and Nobumitsu’s primary aim in describing travel to the palace seems rather to reinforce the role of the Dragon King as the protector of Buddhism. The younger brother’s use of the phrase “home above” means that the palace is underwater, and there is no suggestion that this world is now inaccessible or unavailable as in Kojiki and Nihon shoki. Traveling to and from the palace of the Dragon King is undertaken just as one would travel to and from other Buddhist worlds, that is, via supernatural means, not by simply rowing off into the sunset. The younger brother arrives at the palace of the Dragon King through something akin to teleportation, coming to the palace “as if he had just set out upon the way” (ibid., p. 478). He returns, as in one Nihon shoki narrative, riding a shark. Before he leaves, he asks Toyo-tama-hime “As I have been here for three years, I should return to my home above. How might I be guided home?” She replies, “Do not worry! There are various sea vehicles who accompany the master of the palace of the sea god” (ibid., p. 480). The play on words underscores that the Dragon King, the protector of Buddhism, has the greater vehicle at his disposal. Goff (1991) notes that other Noh plays similarly reference wheels or vehicles in order to pair Buddhist metaphorical concepts with their literal equivalents. Nobumitsu uses the greater shark as the vehicle to deliver Luck of the Mountain.
One final change of note is to Toyo-tama-hime and the well from which she drinks, which are both noted for their purity and radiance. As Klein (1991) notes, the Nāga (Dragon) King’s daughter frequently appears in Noh plays as a role model for women, most notably in the play Ama, in which a woman dives to the palace of the Dragon King under the sea. In Tama no i, Toyo-tama-hime and her younger sister Tama-yori-bime are the daughters of the Dragon King. When they first appear on stage, they draw water from the well, which they repeatedly identify as “pure” and “clear.” That purity is contagious and spreads to the two women: “When we draw the water from this pure spring, our hearts become clear and we become young” and “When we draw this water of everlasting life, our hearts also become pure” (ibid., p. 478). Klein (1991) also notes the frequent relation of identity between women’s bodies and sacred icons in Noh plays, which we see in the second act. Upon presenting the two resplendent jewels to Luck of the Mountain, the two dance, clearly linking the two daughters to the pearls. Nobumitsu’s move to link the daughters and the jewels explains why the younger sister, Tama-yori-bime, is introduced so early in his version of the mythical narrative. In Kojiki and Nihon shoki, the younger sister first appears when Toyo-tama-hime comes to land to give birth, but Nobumitsu introduces her earlier in order to link the women with these objects. In Ama, the Dragon King possesses a sacred Buddhist jewel, and perhaps Nobumitsu also imagined these objects to be Buddhist treasures as well.
Like the Luck of the Mountain story, the Urashima Tarō story gradually incorporated more Buddhist elements in its later adaptations. Medieval versions of the story include but are not limited to citations in historical works, most importantly the historical works Fusō ryakki 扶桑略記 (Abbreviated Records of the Land of the Rising Sun) and Mizu kagami 水鏡 (The Water Mirror), references in Japanese poetry including Nihongi kyōen waka 日本紀竟宴和歌 (Nihongi Banquet Poetry) and Shūi wakashū 拾遺和歌集 (Collection of Gleanings), the picture scroll Urashima myōjin engi 浦島明神縁起 (Origin of the Urashima Myōjin Shrine), and the Shibukawa Urashima Tarō 浦島太郎. There is a substantial body of research on the historical development on the Urashima story and a full accounting of these primary and secondary sources is not possible here.19 This paper will focus on two points from the Shibukawa Urashima Tarō. Firstly, this version of the story is altered such that Urashima visits the palace of the Dragon King in repayment of an act of good karma. Secondly, the nature of the space and the passage of time within it are altered to reflect a medieval Japanese worldview in which the “other world” is a separate and discrete location from our own.
In the Shibukawa Urashima Tarō, the story itself is quite different, beginning with the name of the title character, who is now Urashima Tarō, not “Urashima no ko”.20 While fishing, Urashima catches a turtle, and knowing that turtles had exceptionally long lives, he releases it. The next day, when he goes out, he sees a woman in a boat saying that she wishes to repay the debt for setting the turtle free. Urashima joins her in the boat and rows for ten days, then they arrive at the palace of the Dragon King. The Shibukawa version is the first Urashima Tarō story to associate his destination with the Dragon King rather than the god of the sea. Urashima spends three years there, then returns home with the box and turns into an old man when he opens it. Finally, he transforms into a crane and flies to rejoin his wife, the turtle, as symbols of eternity, and he is venerated as the kami of the Urashima Myōjin Shrine.21
Among the many important changes to the Shibukawa version is the introduction of a karmic exchange. In the Shibukawa version, Urashima catches the turtle and reflects on its long lifespan, then decides to let the turtle go. The next day, the woman appears on the beach and asks him to take him to her homeland. Recall that in Nihon shoki and Tango fudoki, Urashima does not release the turtle, and in Man’yōshū, there is no turtle. The Shibukawa version has clearly adopted a motif of karmic retribution. Urashima catches the turtle, reflects that it would be inappropriate to keep it, releases it, and only then is his reward realized. The narrative pattern is common in Buddhist literature and existed in Japan from long before the Shibukawa version was written. Most notable is a story in the early ninth-century Nihon Ryōiki 日本霊異記 (Record of Miraculous Events in Japan), a collection of Buddhist anecdotal literature (setsuwa 説話), in which a priest buys four turtles and releases them. Later, when the priest is forced to jump into the ocean by pirates, the turtles save his life (Nakada 1995).
A major issue in the literature is why the Urashima story took so long to incorporate this Buddhist element when the karmic retribution literary paradigm was clearly present and widespread in Japan from centuries earlier. Miura (1990) suggests that this was because the story was written down, not transmitted orally, and so the story remained relatively fixed. However, by this logic, the Dragon King, also a Buddhist element, should not have been introduced in Hikohohodemi either; Nihon shoki was also written down, and both stories were recorded in the same text, Nihon shoki. Miura also notes that while Buddhist anecdotal literature, based on and strongly influenced by narratives from abroad, actively incorporated the karmic exchange motif, Japanese folklore only began to do so from the late medieval period, around the same time that the Shibukawa Urashima Tarō was written. This argument appears more promising, especially considering the differences in social class between Norinaga, author of Hikohohodemi, and the unknown original creator of the Shibukawa Urashima Tarō. Norinaga adapted imperial myth and was in direct employment by the emperor, suggesting that Buddhist influence marked medieval adaptations designed for elite consumption at a very early moment. For the Shibukawa Urashima Tarō, which was not limited to an elite audience and was widely reproduced, Buddhist elements appear later.
A second issue is the travel in the story, which has changed to accommodate a Buddhist geographic universe. In the Shibukawa version of Urashima, the protagonist travels to the palace of the Dragon King with the mysterious woman on board; presumably, she directed his travel and delivered the requisite supernatural aid. Another motif involves his falling asleep or closing his eyes, allowing the woman to transport him. This appears in some medieval versions of Urashima Tarō.22 Note that these elements do appear in some of the ancient versions. In Tango fudoki, Urashima closes his eyes, though only for a moment as they cross the border, not for a prolonged period of time as the couple teleports. In Nihon shoki and Tango fudoki, it also seems like the woman guides him to her home. However, in the Man’yōshū version of the story, Urashima arrives at the supernatural destination all on his own, by simply rowing for ten days. The Man’yōshū version accords with the geographical version of the world seen in the Kojiki narrative of Luck of the Mountain: a realm connected to other realms via routes that can be closed or shut off. Perhaps the versions of Urashima in Tango fudoki and Nihon shoki were influenced by Buddhism at a very early stage, but at any rate, by the medieval era and the Shibukawa version, a single world of Jambudvīpa with other worlds reachable via supernatural means was solidly in place.
The passage of time, or lack thereof, in Hōrai, a land of immortals, diffused into medieval Japanese other worlds generally, including the palace of the Dragon King. The description of the palace in Shibukawa Urashima Tarō surpasses all earlier versions of the Urashima story in depth and detail. When Urashima arrives, the maiden tells him that at this palace, the four seasons are simultaneously present in each of the four directions, along with their respective flowering plants and trees. Spring, in the east, had plum and cherry blossoms, willow trees swaying in the breeze, and the cries of the bush warbler. Summer in the south had deutzia, lotus blossoms, and waterfowl. Autumn in the west had colorful foliage and bush clover, and winter in the north had barren trees and accumulated snow. Numerous explanations for the overlapping seasons include demonstrating the mysteriousness of the other world and triggering Urashima’s longing for the capital (Hayashi 1987) and demonstrating the power of the inhabitant of the palace (Shimauchi 1988). Miura (1990) makes a strong case that this suspension of time was directly connected to the world of immortal ascetics that featured prominently in the earliest versions of the Urashima Tarō narrative. However, a similar depiction in which all four seasons are simultaneously present appears in the medieval folktale The Demon Shuten Dōji to describe the lair of a fearsome demon (Kimbrough 2018a, pp. 41–42). It would be more accurate to say that the idea of a timeless space of simultaneous seasons developed in medieval Japan as a characteristic feature of other worlds generally, not only those associated with Hōrai.
Tama no i and the Shibukawa Urashima Tarō each make two alterations to the ancient narratives that demonstrate Buddhist influence on the worldviews of their authors. For Tama no i, firstly, we again see the use of the Dragon King not only as the ruler of water, but as a guarantor of kingly legitimacy. It is striking that Nobumitsu appears to have incorporated this detail without referring to Hikohohodemi, suggesting a diffusion of the idea that the sovereign and the Dragon King have an ideal relationship. Secondly, the association of the two daughters of the Dragon King with Buddhist treasures creates links with other Noh plays that have far more pronounced Buddhist themes, especially Ama. For the Shibukawa Urashima Tarō, the introduction of a karmic exchange converts the story into Buddhist anecdotal literature. Secondly, the need for supernatural travel and the different passage of time in the realm of the Dragon King suggest that it is a discrete location not contiguous or commensurate with our own reality. Because the simultaneity of the seasons also appears in other medieval Japanese folk stories, it seems that the feature of alternative time characterized many other worlds. On the same note, the amalgamation of the palace of the Dragon King with the land of Taoist immortals suggests that perhaps by this time, other worlds had come to have some common aspects. As (Byrne 2016) argues, these features were critical for establishing other world travel narratives as a genre, making them identifiable and understandable to readers.

5. Conclusions

Comparison across periods reveals two different perceptions of other worlds in ancient and medieval Japan. In eighth-century tales, whether Luck of the Mountain or Urashima Tarō, the traveler arrives at a world that is essentially an extension of our own. Urashima Tarō rows across the horizon, and Luck of the Mountain rides in a basket or on a shark along an identified route or course. When Toyo-tama-hime leaves Luck of the Mountain, she closes off this route, much the same as Izanagi closes off the route to Yomi, the land of the dead, when divorcing his wife Izanami. Other routes, such as the “northern sea route,” also appear in Kojiki, and while there is no way to know where such a passage was intended to lead, eighth-century Japanese seemingly conceived of space in terms of physical travel. That included travel upwards and downwards, such as by the vertical “hashi” or the horizontal bridges that were almost certainly influenced by eighth-century public works projects to create permanent built structures.
From the medieval period, other worlds began appearing as actual alternate realities that could only be arrived at through supernatural intervention. In Hikohohodemi, Luck of the Mountain hops into the bright green basket provided by the god of the sea currents, then in the next illustration, he is on the beach of the sea god’s palace as if he had teleported. The basket itself is comically small, and when slung over the god’s shoulder looks more like a butterfly net. In some medieval versions of Urashima Tarō, he is told to close his eyes and keep them closed, or in which he falls asleep and awakes in the palace of the sea god. Other “other worlds” are not dissimilar in this regard; the warriors of The Demon Shuten Dōji climb through the mountains after divine intervention reveals a secret way to them, and the protagonist of The Tale of Tawara no Toda is led by the hand into the waters of Lake Biwa (Kimbrough 2018b).23 Because this space functions as a truly alternative reality, it can potentially have a different flow of time, as seen in the Shibukawa Urashima Tarō among others.
The reason that the medieval supernatural, including metaphysical elements and existence as a discrete and separated reality, was perhaps because in medieval Japan, reality was comprised of a single bounded world and an infinite number of other worlds accessible through supernatural means. The continent of Jambudvīpa contained all of the natural world, while an innumerable variety of other worlds introduced by Mahayana and esoteric Buddhist thought were connected but not accessible via conventional travel overland. Conversely, eighth-century sources like Kojiki suggest a variety of routes and borders leading to other worlds. Ancient Japanese sources envisioned the world as a collection of realms. Medieval other worlds were accessible via supernatural intervention, perhaps in response to a karmic act. That is to say, as the Buddhist cosmological and geographical imagination replaced the ancient Japanese perception of the natural, the supernatural world that existed beyond changed in turn to reflect a new social reality.24
This approach in this article tied together other worlds with supernatural phenomena in order to illustrate perspectives of the natural. The use of comparison between ancient and medieval sources meant severely restricting the content given the paucity of ancient sources. A broader survey of medieval tales involving travel to other worlds would certainly provide more detail and nuance. Another critical intersection not discussed here is medieval creation narratives, which, as Itō (2016) and others have argued, originate in Shingon Buddhist doctrine. These stories of how the world came into being, paired with perspectives from folk tales that provide a glimpse on what that world was imagined to look like, would link theological tenets with lived experience in a holistic treatment of medieval myth and Shinto.

Funding

This research received no external funding.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The author declares no conflict of interest.

Notes

1
Durkheim’s attention to the supernatural is part of a larger effort to define religion; Durkheim argues that the supernatural is not an effective descriptor for religion. Ironically, given the topic of this article, he uses Buddhism as his evidence that the sacred/profane distinction is a better foundation for defining religion than the supernatural.
2
Note that this approach differs from Durkheim, for whom categories were very real. Yet as Smith (1982) notes, in discussing religion, “Religion is solely the creation of the scholar’s study. It is created for the scholar’s analytic purposes by his imaginative acts of comparison and generalization. Religion has no independent existence apart from the academy”. I approach the supernatural in the same fashion.
3
The protagonist of the Urashima story was named “Urashima no ko” until the late medieval or early modern period, when his name changed to “Urashima Tarō”. Since he is best known by this latter name, I use it anachronistically in this paper.
4
On the composition and contents of Kojiki, Nihon shoki, and Man’yōshū, see Lurie (2011) and Duthie (2014). On Fudoki, see Palmer (2016) and Aoki (1997).
5
The earliest of these is Müller (1893); in Japanese, the exemplary treatments for comparative myths of Luck of the Mountain and links to the South Seas are Matsumoto (1928, 1971), Matsumura (1954–1958).
6
On Mishina and Korea, see Hirafuji (2004, 2013).
7
For example, the accompanying textual explanation (Jp. kaidai) for the Nihon koten bungaku zenshū version of Nihon shoki claims that the medieval chronicles are “based on empty logic and argumentation and without value for academic study of Nihon shoki”. See Sakamoto et al. (1994–1995, vol. 5).
8
On this point, see in particular Itō (2003, 2016).
9
For the myths from the Age of the Gods, Nihon shoki provides a main narrative followed by multiple supplementary or alternative versions of that narrative. The same story appears multiple times, sometimes with major differences, presumably due to the use of different source materials. The style probably meant to imitate Pei Songzhi’s (372–451) San guo zhi zhu 三国志注 (Annotations to Records of the Three Kingdoms). See Endō (2009) and Sakamoto (1991).
10
The first two volumes of Nihon shoki, sometimes referred to as the “Age of the Gods,” narrate creation to the birth of the mythological emperor, Jinmu. This account is customarily divided into eleven sections, which are the “Main Narrative.” After each section, several “variants” are included that provide a different account of the content from the main narrative of that section. Hiko-ho-ho-demi and his brother are born at the end of section 9 and his trip to the Palace of the Sea God is given in section 10. Along with the main narrative, there are four variants of section 10.
11
The vernacular glosses for Nihon shoki are not present on the oldest manuscript editions, though these manuscripts are also fragmentary. The vernacular reading of “Tokoyo” for this passage is attested from Heian sources, the oldest extant evidence of how Nihon shoki was read in premodern Japan. While there is no eighth-century manuscript of Nihon shoki that has this gloss, the Heian attestation strongly suggests that the conceptual link between Hōrai and Tokoyo was present from the eighth century. In the case of Fudoki, the vernacular gloss is almost certainly derived from that given to Nihon shoki in Heian materials.
12
On Go-Shirakawa, see Komatsu (1989). On the political significance of these picture scrolls, see Felt (2023).
13
A summary of notable studies on this topic is given in Ruppert (2002).
14
An anonymous reviewer suggested that this intervention reflects the aristocratic marriage practices of the time. McCullough suggests analyzing Heian marriages in terms of marital residences, houses and property, and children and family, all of which the Dragon King appears to be taking control of in the Hikohohodemi emaki account. See McCullough (1967).
15
Hotate (1986) argues that the visual grammar of these scenes recalls medieval tribute offerings of food and drink to the gods (Jp. mikuri 御厨).
16
On the emergence of the genre and the Shibukawa collection, see Ruch (1971).
17
Nobumitsu incorporates a number of antiquated words that appear in the Nihon shoki version of the story, and it is likely he referred to Nihon shoki or a commentary on the text. There is nothing suggesting that he referred to Hikohohodemi.
18
As in Hikohohodemi, the palace is also decorated with lapis lazuli, though the description is not nearly as evocative. Noh plays do not often include set construction, so the simplified description suits the medium; Nobumitsu may also have been considering the terse description in Nihon shoki.
19
On Urashima in Fusō ryakki and Mizu Kagami, see Brightwell (2020). Commentary on Urashima myōjin engi is given in Komatsu (1992a). Among the numerous surveys of Urashima Tarō in Japanese literature, Miura (1990, 1992), Mifune (2009), and Hayashi (2018) provide the most comprehensive diachronic treatment of the Urashima story.
20
Description of the Shibukawa Urashima Tarō here is taken from Urashima Tarō.
21
The shrine claims to be founded in 825 CE, ostensibly in relation to the date that Urashima returned from Hōrai. The shrine appears as Ura Shrine 宇良神社 in the 927 Engi shiki 延喜式 (Regulations and Laws of the Engi Era).
22
For example, see the description provided for the corresponding illustration in Komatsu (1992b).
23
Note that in The Tale of Tawara no Toda, the protagonist swims through the various cylinders that make up the cosmos as described in Buddhist text The Treasury of Abhidharma. Unfortunately, the medieval Luck of the Mountain and Urashima Tarō narratives do not make explicit reference to cosmology. Medieval commentaries like the Nihon shoki sanso demonstrate that the Buddhist cosmological universe depicted in The Treasury of Abhidharma was influential among Japanese scholars. See Kōnoshi (1999).
24
The transformation of the visual world under Buddhist influence in medieval Japan is amply demonstrated in Moerman (2021).

References

  1. Akima, Toshio. 1993. The Myth of the Goddess of the Undersea World and the Tale of Empress Jingū’s Subjugation of Silla. Japanese Journal of Religious Studies 20: 95–185. [Google Scholar] [CrossRef]
  2. Aoki, Michiko Yamaguchi. 1997. Records of Wind and Earth: A Translation of Fudoki, with Introduction and Commentaries. Ann Arbor: Association for Asian Studies. [Google Scholar]
  3. Brightwell, Erin. 2020. Reflecting the Past: Place, Language, and Principle in Japan’s Medieval Mirror Genre. Cambridge: Harvard University Asia Center. [Google Scholar]
  4. Byrne, Aisling. 2016. Otherworlds: Fantasy and History in Medieval Literature. Oxford: Oxford University Press. [Google Scholar]
  5. Drown, Edward S. 1913. What Is the Supernatural? The Harvard Theological Review 6: 143–55. [Google Scholar] [CrossRef]
  6. Durkheim, Emile. 1915. The Elementary Forms of the Religious Life. Translated by Joseph W. Swain. London: George Allen & Unwin Ltd. [Google Scholar]
  7. Duthie, Torquil. 2014. Man’yōshū and the Imperial Imagination in Early Japan. Leiden: Brill. [Google Scholar]
  8. Endō, Keita. 2009. Nihon shoki no bunchū: Denshō no fukusū sei kara. Hisutoria 214: 1–23. [Google Scholar]
  9. Felt, Matthieu. 2023. Meanings of Antiquity: Myth Interpretation in Premodern Japan. Cambridge: Harvard University Asia Center. [Google Scholar]
  10. Glum, Peter. 1988. The Hikohohodemi No Mikoto Emaki. Hamburg: Gesellschaft für Natur-und Völkerkunde Ostasiens e. V. [Google Scholar]
  11. Goff, Janet. 1991. Plays about Lady Rokujō. In Noh Drama and The Tale of Genji: The Art of Allusion in Fifteen Classical Plays. Princeton: Princeton University Press, pp. 125–49. [Google Scholar]
  12. Hayashi, Kōhei. 1987. Urashima to shiki: Urashima Tarō tanjō no shūhen. Tomakomai Komazawa tanki Daigaku kiyō 19: 1–35. [Google Scholar]
  13. Hayashi, Kōhei. 2018. Urashima Densetsu no Tenkai. Tokyo: Ōfū. [Google Scholar]
  14. Hirafuji, Kikuko. 2004. Shinwagaku to Nihon no Kamigami. Tokyo: Kōbundō. [Google Scholar]
  15. Hirafuji, Kikuko. 2013. Colonial Empire and Mythology Studies: Research on Japanese Myth in the Early Shōwa Period. In Kami Ways in Nationalist Territory: Shinto Studies in Prewar Japan and the West. Edited by Bernhard Scheid and Kate Wildman Nakai. Wien: Verlag der Österreichischen Akademie der Wissenschaften, pp. 75–107. [Google Scholar]
  16. Hotate, Michihisa. 1986. Hikohohodemi no mikoto emaki to mikuriteki sekai. In Kodai Kokka no Shihai to Kōzō. Edited by Tanaami Hiroshi. Tokyo: Tōkyōdō shuppan, pp. 360–94. [Google Scholar]
  17. Ichiko, Teiji. 1958. Urashima Tarō. In Otogizōshi. Nihon koten bungaku taikei 38. Tokyo: Iwanami shoten. [Google Scholar]
  18. Itō, Masayoshi. 1972. Chūsei Nihongi no rinkaku: Taiheiki ni okeru Urabe Kanekazu setsu o megutte. Bungaku 40: 28–48. [Google Scholar]
  19. Itō, Satoshi. 2003. The Medieval Period: The Kami Merge with Buddhism. In Shintō: A Short History. Translated by Mark Teeuwen, and John Breen. New York: RoutledgeCurzon, pp. 63–107. [Google Scholar]
  20. Itō, Satoshi. 2016. Shintō no Keisei to Chūsei Shinwa. Tokyo: Yoshikawa kōbunkan. [Google Scholar]
  21. Kaneoka, Rie. 2017. Ikai to no kōtsū: Umisachi Yamasachi shinwa o chūshin ni. Hikaku Nihon Bungaku Kyōiku Kenkyū Sentaa Kenkyū Nenpō 13: 18–25. [Google Scholar]
  22. Kimbrough, Keller, trans. 2018a. The Demon Shuten Dōji. In Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales. Keller Kimbrough, and Haruo Shirane, eds. New York: Columbia University Press, pp. 31–58. [Google Scholar]
  23. Kimbrough, Keller, trans. 2018b. The Tale of Tawara Toda. In Monsters, Animals, and Other Worlds: A Collection of Short Medieval Japanese Tales. Keller Kimbrough, and Haruo Shirane, eds. New York: Columbia University Press, pp. 72–99. [Google Scholar]
  24. Klein, Susan Blakeley. 1991. When the Moon Strikes the Bell: Desire and Enlightenment in the Noh Play Dojoji. The Journal of Japanese Studies 17: 291–322. [Google Scholar] [CrossRef]
  25. Kojima, Noriyuki. 1996. Nihon shoki 2. Shinpen Nihon koten bungaku zenshū 3. Tokyo: Shogakkan. [Google Scholar]
  26. Komatsu, Shigemi. 1989. Nihon Emaki Shūkō 1. Tokyo: Chūō kōronsha. [Google Scholar]
  27. Komatsu, Shigemi, ed. 1992a. Hikohohodemi no Mikoto Emaki. In Hikohohodemi no Mikoto Emaki, Urashima Myōjin Engi. Zoku Nihon no Emaki 19. Tokyo: Chūō kōron, pp. 1–46. [Google Scholar]
  28. Komatsu, Shigemi, ed. 1992b. Urashima myōjin engi. In Hikohohodemi no Mikoto Emaki, Urashima Myōjin Engi. Zoku Nihon no emaki 19. Tokyo: Chūō kōron, pp. 47–79. [Google Scholar]
  29. Kuroda, Toshio, and Jacqueline I. Stone. 1996. The Imperial Law and the Buddhist Law. Journal of Japanese Studies 23: 271–85. [Google Scholar] [CrossRef]
  30. Kōnoshi, Takamitsu. 1986. Kojiki no Sekaikan. Tokyo: Yoshikawa kōbunkan. [Google Scholar]
  31. Kōnoshi, Takamitsu, ed. 1997. Kojiki. Shinpen Nihon koten bungaku zenshū 1. Tokyo: Shogakkan. [Google Scholar]
  32. Kōnoshi, Takamitsu. 1999. Kojiki to Nihon Shoki: Tennō Shinwa no Rekishi. Tokyo: Kōdansha gendai shinsho. [Google Scholar]
  33. Lurie, David Barnett. 2011. Realms of Literacy: Early Japan and the History of Writing. Cambridge: Harvard University Asia Center. [Google Scholar]
  34. Matsumoto, Nobuhiro. 1928. Essai sur la Mythologie Japonaise. Paris: P. Geuthner. [Google Scholar]
  35. Matsumoto, Nobuhiro. 1971. Nihon Shinwa no Kenkyū. Tokyo: Heibonsha. [Google Scholar]
  36. Matsumura, Takeo. 1954–1958. Nihon Shinwa no Kenkyū. 4 vols. Tokyo: Baifūkan. [Google Scholar]
  37. McCullough, William H. 1967. Japanese Marriage Institutions in The Heian Period. Harvard Journal of Asiatic Studies 27: 103–67. [Google Scholar] [CrossRef]
  38. Mifune, Takayuki. 2009. Urashima Tarō no Nihon shi. Tokyo: Yoshikawa Kōbunkan. [Google Scholar]
  39. Miura, Sukeyuki. 1990. Urashima Tarō no Bungakushi: Ren’ai Shōsestsu no Hassei. Tokyo: Goryū shoin. [Google Scholar]
  40. Miura, Sukeyuki. 1992. Ryūgūjō: Urashima Tarō to hōraisan genso. Rekishi Tokuhon 37: 64–71. [Google Scholar]
  41. Moerman, D. Max. 2021. The Japanese Buddhist World Map. Honolulu: University of Hawai’i Press. [Google Scholar]
  42. Müller, Friedrich W. K. 1893. Mythe der Kei-Insulaner und Verwandtes. Zeitschrift für Ethnologie 25: 533–37. [Google Scholar]
  43. Nakada, Norio. 1995. Nihon Ryōiki. In Shinpen Nihon koten bungaku zenshū 10. Tokyo: Shogakkan. [Google Scholar]
  44. Nishino, Haruo. 1998. Tama no i. In Yōkyōku Hyakuban. Shin Nihon koten bungaku taikei 57. Tokyo: Iwanami shoten, pp. 477–81. [Google Scholar]
  45. Palmer, Edwina. 2016. Harima Fudoki: A Record of Ancient Japan Reinterpreted, Translated, Annotated, and with Commentary. Leiden: Brill. [Google Scholar]
  46. Patch, Howard R. 1950. The Other World According to Descriptions in Medieval Literature. Cambridge: Harvard University Press. [Google Scholar]
  47. Rambelli, Fabio, ed. 2018. The Sea and the Sacred in Japan. London: Bloomsbury Academic. [Google Scholar]
  48. Ruch, Barbara. 1971. Origins of The Companion Library: An Anthology of Medieval Japanese Stories. The Journal of Asian Studies 30: 593–610. [Google Scholar]
  49. Ruppert, Brian O. 2002. Buddhist Rainmaking in Early Japan: The Dragon King and the Ritual Careers of Esoteric Monks History of Religions. History of Religions 42: 143–74. [Google Scholar] [CrossRef]
  50. Sakamoto, Tarō. 1991. Six National Histories of Japan. Translated by John S. Brownlee. Vancouver: UBC Press. [Google Scholar]
  51. Sakamoto, Tarō, Ienaga Saburō, Inoue Mitsusada, and Ōno Susumu. 1994–1995. Nihon shoki. 5 vols. Tokyo: Iwanami bunko. [Google Scholar]
  52. Satake, Akihiro, Kinoshita Masatoshi, and Kojima Noriyuki. 2002. Man’yōshū Honbunhen. Tokyo: Hanawa shobō. First published 1963. [Google Scholar]
  53. Shimauchi, Keiji. 1988. Otogizōshi no Seishinshi. Tokyo: Perikansha. [Google Scholar]
  54. Smith, Jonathan Z. 1982. Imagining Religion: From Babylon to Jonestown. Chicago: University of Chicago Press. [Google Scholar]
  55. Uegaki, Setsuya. 1997. Fudoki. Shinpen Nihon Koten Bungaku Zenshū 5. Tokyo: Shogakkan. [Google Scholar]
  56. Yamamoto, Hiroko. 1998. Chūsei Shinwa. Tokyo: Iwanami shoten. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Felt, M. Visits to the Palace of the Sea God in Ancient and Medieval Japan. Religions 2024, 15, 350. https://doi.org/10.3390/rel15030350

AMA Style

Felt M. Visits to the Palace of the Sea God in Ancient and Medieval Japan. Religions. 2024; 15(3):350. https://doi.org/10.3390/rel15030350

Chicago/Turabian Style

Felt, Matthieu. 2024. "Visits to the Palace of the Sea God in Ancient and Medieval Japan" Religions 15, no. 3: 350. https://doi.org/10.3390/rel15030350

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop