<i>Shikinen Sengu</i>and the Ontology of Architecture in Japan

> There is nothing to see, and they won't let you see it. Basil Chamberlain If we compare the architecture of Western civilization to a museum, Japanese architecture [is like] a theater. Toyo Ito Visitors to Ise Jingu in Japan may pick up a free brochure with a caption history of the shrine, photographs of important rites under way, descriptions of the site's principal features, and a plan view drawing. This last catches the eye right away. Paired with each sanctuary compound, or goshoden, is an open space, the kodenchi, whose dimensions duplicate the goshoden's (Figure 1). Curious philosophers who look into the pairing of goshoden to kodenchi will find themselves rethinking the ontology of architecture. Plan view detail of the naiku at Ise Jingu. Right: the current sanctuary or goshoden. Left: the kodenchi, the empty site of the next goshoden, to be completed in 2013. It might seem that no topic in aesthetics need less attend to cultural variation than the ontology of art. Ontology is after all a branch of metaphysics and metaphysics abstracts from cultural contingency. However, contemporary work in the ontology of art ties ontology to culture in a way that ought to encourage cross‐cultural studies in the ontology of art. Philosophers have for some time studied theories of art in the context of cultural variation.1 A theory of art states—perhaps in the form of a definition—the features that make items works of art. Candidate art‐making features include having significant form, affording an experience with marked aesthetic character, or occupying a specified role in an institution.2 Armed with such a theory, we may ask three questions. Do members of any given culture make works of art? Do they have a concept of art—do they group together as a kind the items that have the art‐making features? Do they know the theory of art—do they know that artworks have the art‐making features?3 An ontology of art is not a theory of art. The world arguably includes material objects, properties, events, and abstracta like sets and types; and an ontology of art places artworks in one or more of these categories. Thus it does not imply a theory of art: the statement that some works of art are material objects does not settle what makes any material object a work of art. Nor does a theory of art imply an ontology of art: the claim that all and only artworks have significant form leaves open whether artworks are material objects, events, or abstracta—any of these may have or lack significant form. Finding that members of a given culture make art, have a concept of art, and know the theory of art leaves open the ontology of art in the culture. In crafting ontologies of art, philosophers take a cue from the tacit ontological conceptions implicit in appreciative practices.4 They assume that an adequate ontology of art must be consistent with and help illuminate those practices. Sometimes, an ontology of art leads us to revise our appreciative practices; but if, upon reflection, our appreciative practices are sound, then they furnish materials for a good ontology of art. As David Davies puts the principle, a work of art is an item belonging to the ontological category whose members have the properties that appreciative practice, on reflection, requires us to attribute to the work. For example, Nelson Goodman observes that appreciations of a work and forgeries of it always differ in content, but whereas some artworks can be forged (such as paintings), others cannot be forged (such as poems).5 He therefore distinguishes single‐instance works, whose duplicates are never instances, from multiple‐instance works, which can be copied. Single‐instance works are material particulars or token events and multiple‐instance works are compliance classes. Although Goodman's ontology of art is not widely accepted, his methodology is entirely orthodox in the sense that it is intended to fit our appreciative practices.6 If a phenomenon is a cultural invariant, then nothing can be learned about it by studying it in different cultures (except, of course, that it is a cultural invariant). Perhaps sound art appreciation is a cultural invariant. If it is, then the ontology it implies is also invariant, since anything implied by an invariant is also invariant. The ontology of art in one culture will generalize to all others, and we need not look beyond our own shores for a full ontology of art. It follows that the relevance of cross‐cultural studies to the ontology of art depends on whether sound practices of art appreciation vary culturally. According to pluralist ontologies of art, no ontological category contains all artworks. If some are material objects, then some are not—perhaps they are token events or abstracta. Works are assigned to ontological categories so as to accommodate sound appreciative practices. Since the practices are usually taken to govern groups of works, each group is placed in an ontological category. The question is then how to group works. Initially at least, it makes sense simply to group works into art forms. Thus paintings are material objects, musical works are abstract objects, and so on. What determines ontological category is, roughly, art form. If art forms are cultural invariants, then pluralist ontologies of art need not attend to cultural variation; but art forms are unlikely to be cultural invariants. Nothing rules out their varying within or across cultures in ways that impact the ontology of art. Peter Kivy and Jerrold Levinson famously hold that musical works are abstracta, but both are careful to qualify their views as applying to European art music in the grand tradition from Bach to Stravinsky.7 In the same spirit, Stephen Davies develops a pluralist ontology of musical works, including Javanese gamelan.8 What goes for music may go for other art forms. According to monist ontologies of art, all works of art belong to the same ontological category. This implies that appreciative practices are invariant enough to embed a uniform ontology of art. For example, David Davies identifies the work of art with a token event, namely, the artist's act of specifying a focus of appreciation. The focus of appreciation comprises a content, a vehicle for expressing the content, and a set of understandings by means of which the vehicle expresses the content.9 The vehicle may be an object, a performance, or something else, but the work itself is the artist's act of specifying the focus of appreciation—it is not the focus that gets specified. Thus all works of art from all times and places belong to the same ontological category: all are token events. Even so, Davies's monism about works allows for pluralism about vehicles for expressing content. A painting and a sonata are both, according to Davies, acts of specifying a focus of appreciation, but in each case the focus may have different elements. In the case of the painting, the focus might comprise a material object plus a set of understandings through which material objects bear content. In the case of the sonata, the focus might comprise an abstractum plus a set of understandings through which abstracta bear content. Again, works in different art forms have different vehicles or different content‐determining understandings. Art forms are not likely to be cultural invariants, but Davies's monism also accommodates cultural variation in art forms. The following three propositions are mutually consistent. Every work of art is an act of specifying a focus of appreciation (monism is true). The vehicular component of the focus of appreciation varies culturally. The understandings by means of which the vehicular component bears content vary culturally. Differences in sound appreciative practices from culture to culture can imply all three propositions together. In sum, cross‐cultural studies are relevant to the ontology of art as long as sound practices of art appreciation vary enough that they imply different ontologies. Pluralist ontologies of art represent this variation as a variation in the ontological categories of works. Monist ontologies of art represent it as a variation in the elements that make up foci of appreciation specified by artworks. However, the question remains whether sound practices of art appreciation do in fact vary in ways that imply different ontologies. Ise Jingu, located on the coast about one‐hundred kilometers southeast of Kyoto, is among Japan's most important and most visited Shinto shrines. It is also one of the oldest, though, as we shall see, the shikinen sengu makes dating a tricky business. Each Shinto shrine is consecrated to the worship of one or more kami, and Ise Jingu is primarily dedicated to Amaterasu Omikami, the sun goddess, from whom the Japanese emperors claimed descent. Legend tells that it was founded in the first century BCE by Emperor Suinin, though historical evidence indicates a founding in the fifth or sixth centuries CE, around the time when the nearby court at Yamato was establishing itself as the imperial family. As this family's shrine and hence indirectly a national shrine, it has served since then to represent and promote the authority of the emperor and to focus Japanese nationalism.10 Many Japanese consider that they should visit the shrine at least once in their lifetime and more than 6 million do visit every year. The shrine occupies two parcels of land six kilometers apart. The smaller geku (or outer shrine) comprises a cluster of about two dozen buildings, the most important being a sanctuary for Toyoke Omikami, the goddess of clothing, food, and housing—purveyor to Amaterasu. The naiku (or inner shrine) comprises a roughly equal number of structures set in a forest of 5,500 hectares. The structures are modest in scale and are built in a plain style of unfinished Japanese cypress with thatched miscanthus roofing, in obvious resistance to Buddhist and Chinese influence. One of these structures, the goshoden, houses Amaterasu herself in the form of a mirror. Remarkably, the goshoden (and all other sanctified structures, including the Uji Bridge) has been rebuilt about every twenty years (with occasional breaks) since the eighth century CE in a cycle called the shikinen sengu. Moreover, the rebuilding is not exactly in situ. As Figure 1 shows, the goshoden sits next to a vacant lot, the kodenchi, where the goshoden is rebuilt. Goshoden and kodenchi in this way swap sites with each sengu cycle. The sengu is a major material and ritual undertaking.11 The work takes about eight years, from the initial preparation of the kodenchi grounds to the final transfer of the kami. Timber from more than 15,000 trees is required, along with 900 meters of silk, 3.75 kilograms of gold, and 260 kilograms of lacquer. Ancient techniques and tools of ancient design are used. News reports put the cost of the 2013 sengu at 55 billion yen. In addition, every stage of building is marked by rites, thirty‐five in all, some modest, some involving more than one hundred thousand participants, and others elaborate and esoteric. One might seek out the meaning of the sengu in its rationale, but there is no widely accepted explanation of the practice. According to one theological explanation, Shinto requires that kami be lodged in fresh and beautiful shrines; according to another, the sengu articulates Shinto doctrines of purity and renewal.12 According to a broader cultural explanation, the sengu embodies mujukan, an awareness of the transience of life, nature, and artifacts.13 The close relationship between Ise Jingu and the imperial household suggests a political explanation, that in undertaking the enormous, recurring expense of the sengu, the household renewed its unwavering commitment to the shrine and thus its national authority.14 Architectural explanations have also been proposed: the sengu ensures that the skills needed to maintain the shrine are passed down from one generation of artisans to the next, or it maintains the purity of building materials and preserves ancient architectural forms.15 However plausible or implausible these explanations, the sengu is a fact. Moreover, the sengu is no mere curiosity: it arguably distills one strand of architectural practice in Japan. This can be seen in other sacred architecture, in vernacular building, and also in a distinctively Japanese brand of modernism. More than thirty other shrines are regularly rebuilt. Kasuga Taisha in Nara is rebuilt every thirty years, Kamo Mioya Jinja in Kyoto every fifty years, and Nukisaki Jinja in Gumma Prefecture every thirteen years, for instance. Many more are rebuilt on an irregular schedule—Izumo Taisha in Shimane Prefecture, which rivals Ise Jingu in age and status, has been rebuilt twenty‐five times. Buddhist monasteries and temples are also rebuilt as needed—for example, Kofukuji in Nara has now been taken down to its foundations and is slated for completion in 2010. Traditional Japanese vernacular architecture characteristically features wood, paper, and other relatively ephemeral building materials, which are allowed to weather, and traditional houses are rebuilt every few decades. Thus tourists in Japan are struck by the absence of old town centers—Kyoto's Gion district being a well‐known exception. Almost all the historic sites that attract visitors have been rebuilt several times. Kinkakuji, which is perhaps Kyoto's best‐known site, was burnt to the ground in 1950, rebuilt in 1955, and regilded in 1987—but nobody regards this as remarkable or as diminishing the site's value. Modernist architecture is seen in Japan as extending tradition. This conception of Japanese modernism out of a to traditional architecture following The traditional architecture, Ise Jingu, with after the and the of traditional architecture with The most of this are an by the of Japanese architecture, in of Japanese and Shinto Ise and The with a in which the tells how his for the of Japan's most sacred way after the to a of the and of so as one architectural practice in Japan was up with and not be of the context of architecture, a that is in the of Japanese architecture to the elements of tradition in traditional architecture to architecture. traditional and was to the of the the same Japanese modernism was to a in Japanese architecture to architectural form for the building this strand of Japanese modernism are the of the and who architectural structures as of elements in and out in of and is the of the his with a one in houses by and within years of their The by Toyo Japan's best‐known was put up for a are to look as are buildings, which are of suggests several explanations of this architecture of land in make the of building cost to land among the in the so that of is In is more than and in an The sengu is also a architecture and in the tradition of building and a culture of than that of As of rebuilding with can or the who that there is nothing to see at Ise with appreciative tools to European architecture that fit the strand of Japanese architecture to which Ise Jingu need not all Japanese architectural appreciation takes the same is the of at the shikinen sengu suggests that some Japanese architecture might a different appreciative that the age to a building is a in its an appreciation of a building as old have a different content it as are two Ise Jingu is more than one thousand years Ise Jingu is no more than twenty years let to the goshoden at the are and a ontology of architecture implies that is and is However, one that the naiku and geku are and that shrines are old and They are old they are with the that there at least as as They are they are rebuilt every twenty such means that both and are and are both only there is an to An to the ontology must the The ontology that a between and a of architectural appreciation in the of a building are not of a work. see a of the is not to see the and to see a of the as it was in BCE is not to see the cannot architecture by means of a to Thus architectural works with multiple‐instance works like to any of the is to the According to the a building is a material object as sense objects like and It is up of each roughly to a of and it into when the are is a of some It the of some and it does not the of all its However, it is whether it the of all its no that its when it was is in this way some well‐known If a building is a material object as sense and then is Ise Jingu is up of that together no more than twenty years Although it sits next to the where a different building it is not the of that The is that no building the of all its there was a time when both and a material object cannot since the sengu does not the of each of the goshoden no there is no need to the well‐known An ontology of Ise Jingu that is consistent with both and must in such a way that they are also consistent with each In addition, it should fit Ise Jingu as an of the strand of Japanese architectural practice. architecture a for one to the Although they there are multiple‐instance works of architecture. An is and for what it is, one must it as a work that has Although the are material objects, the work itself is an it to be a as a one might that Ise Jingu is a multiple‐instance work whose are material objects built on of ground about every twenty The is consistent with both and Ise current is no more than twenty years old and in that sense is Since the is at least as old as its first the is more than one thousand years and in that sense is What is old is the and what is is its current instance. However, Ise Jingu is different from works like and the that it has a different The properties of a are the properties that its have The properties that of have are material and features specified in the of must fit the the features of Ise Jingu in a of it are to it. Although there is evidence that the current does roughly the a that there have been design the centuries not be taken to that the shrine long to and was by an is, the sengu allows for in material and from one rebuilding to the next and the evidence is that there have been the properties of and its are not properties of Ise Jingu. What then are the properties of the One from the that the shrine to an architectural practice that includes other traditional Japanese architecture and the distinctively Japanese modernism by such as Ito and A of in this strand of architectural practice in Japan is that they weather, and in the (or make to example, by Although they do in fact their is the of building does not the is of the and the of the old goshoden does not Ise Jingu. the is a material object to material and design is a to design a of Ise Jingu is that it of building, and Ise Jingu on therefore the shrine from a strand of architectural practice in Japan. According to the Ise Jingu is a work whose are material objects to material and design the shrine a architectural practice, as Japanese and it then its features are this one might the that Ise Jingu is a multiple‐instance work. are not material objects to design they are events of about twenty in which a is and This with and also the of the shrine in a architectural practice. Even so the Ise Jingu from other works to belong to the same architectural practice. and are not multiple‐instance works. They are events. Thus the implies a structures like Ise Jingu are multiple‐instance types; structures that are not rebuilt are token events. One might the If means that there are two appreciative one for works like the and for works like Ise Jingu, then the view that the shrine to the architectural practice must be an is to seek an ontology that and and also accommodates the ontological of an architectural practice that includes Ise Jingu and the is an to the multiple‐instance in one strand of architectural practice in Japan are material they are all token events. Sometimes, a building is the history of a material object from its fresh through and it is taken Sometimes, the includes one or more of If is a material object, then what tourists visit is a of the built by in If it is a token that includes then what tourists see is what the In the case of Ise Jingu, the is and and it is up of a of and on sites in a the sengu. The of this is that it Ise Jingu, traditional Japanese architecture, and distinctively Japanese modernism to a Thus Ise Jingu is not an it is a case of a architectural practice. This is exactly how it is in the on architecture in Japan. the same there are between ephemeral works like and works of rebuilding like Ise Jingu. This those only three years, in Ise Jingu, is through the two events differ in and only one does not to a in both are token events. ontologies might and and the of the strand of architectural practice that takes in Ise Jingu, traditional vernacular architecture, and Japanese modernism. For that it be to take that strand of Japanese architectural appreciation as that Ise Jingu is a token Ise Jingu does of architecture to look at a of ontological than into when they their to the and with the that some architectural works are to be more like than like one might that the of appreciation is If it is not sound, no ontology it implies any the one might ask whether the European of architecture appreciation is it is by to the Japanese A is cultural there is more than one ontology of architecture sound architecture appreciation varies across

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.