In this paper I criticize the studies of Ooki Nakamura, the authority in this field (sections 2 - 3), and describe the mortuary system of the Kamegaoka culture, which was a culture of the Tohoku region in the Final Jomon period characterized by distinctive, high-quality ceramics and other artifacts (sections 4 - 5 (1)). I argue that the mortuary system was influenced by the settlement system, which was largely determined by the topography of the sites of the Kamegaoka culture (section 5 (2)). Finally, I discuss whether the mortuary system of the Kamegaoka culture was that of a transegalitarian society (section 6).
I develop three main criticisms of Nakamura's work: the criteria for distinguishing graves, the criteria for distin-guishing pit-burial clusters, and the criteria for identifying transegalitarian societies. Although Nakamura's criteria for distinguishing graves reflect various levels of certainty, he treated them all equally. In contrast, I propose to give criterion points according to degree of certainty. Following this, we should attempt to distinguish graves from pits in the next procedure. Does a pit have the same characteristics as the criteria for distinguishing graves? How many points in total do they come to according to a list of criteria giving points based on certainty? Do the total points exceed those of graves (5 points)?
Nakamura treats all pits in a group of pits that were dug in a comparatively short period of time as graves if one or several pits in the group had the same characteristics as the criteria for distinguishing graves. While I sympathize with Nakamura's hope to advance mortuary archaeology beyond the limitations of the data, I believe this approach is too rough and attempt to consider more objective evaluations. My approach centers on the "probability" of mortuary inter-pretations. I count the points of each pit in the group that was treated as a grave by Nakamura according to the previ-ous list of criteria. The total points are termed "A". As I argue that pits with more than 5 points can be classified as graves, if all pits in the group are graves, the total is the number of pits multiplied by 5 points. This total is termed "B". A÷B thus reflects the "probability" of mortuary interpretation.
I point out that there were dual mortuary systems in Kamegaoka society that were divided by dual settlement sys-tems caused by lifestyle differences between open and mountainous regions. I conclude that we cannot find sufficient evidence of a transegalitarian society in the mortuary system of the Kamegaoka culture.

View full abstract

From the mid-Nara period, provincial Kokubunji temples were built all over Japan. Most of the roof-tiles required for these temples were produced in special workshops administered by the provincial governors (kokushi). In previous research, however, I have shown that when Kokubunji construction began to decline, and with it the demand for roof-tiles, each province differed as to whether the existing workshops were maintained or else were shut down and craftsmen brought in from elsewhere when tiles were required for temple repairs and so forth. In this paper I primarily attempt to analyze the different histories of Kokubunji tile workshops by comparing provinces where tiles were produced in association with ceramics with those where they were made separately.
Omi, Musashi, Kamino, Shimono, Sado, Mutsu and Dewa are provinces where ceramics were produced in the same kilns as tiles in the Kokubunji roof-tile workshops. As a result of an examination of both the design and the fabrication techniques of roof tiles from these provinces, it was shown that, with a few exceptions, the same traditions of design and fabrication were maintained from the start of production through until the repair phase in the 9 th century. From this it is argued that the same groups of craftsmen continued to be engaged in roof-tile production at Kokubunji tile workshops that combined the production of ceramics.
From the 9 th century, provincial combined tile and ceramic kilns were mainly organized around the requirements of ceramic production. With the reduced demand for roof tiles with the decline in Kokubunji construction, roof tile craft workers could only continue their work through depending on ceramics.

View full abstract

In recent years, within the field of urban history a reevaluation of the history of the city in Japan has begun. This reevaluation has proposed that the ancient capitals (palaces) and early modern castle towns should be seen as "traditional cities" in contrast to modern urban centers. Against this background, this paper uses results from actual archaeological excavations to analyze an urban form peculiar to Japan that cannot be classified as a capital or a castle town: the medieval "religious city". It is argued that the "religious cities" or "Koto type" medieval temple settlements of Omi district should be seen as a type of medieval city that, together with the Sengoku (Warring States) and Nobunaga-Hideyoshi type castle towns, fused and evolved into the early modern castle town of the Tokugawa period.
Using the perspective of "urban archaeology", the temple settlements of medieval Omi are analyzed using three of the ten characteristics of cities proposed by V.G. Childe: (1) a concentrated population, (2) the presence of craftsmen and other specialists who do not directly produce food, and (4) the presence of monuments or public facilities-in this case straight roads. As a result, it was possible to identify a type of settlement with a group of defining features including planned, straight roads built down from the mountainside with artificial terraces on both sides. These settlements were called"Koto type" medieval temple settlements and were interpreted as "religious cities". Based primarily on the results of excavations at the Binmanji site in Shiga Prefecture, four developmental stages were identified using changes in mountain beliefs and temples and their surrounding settlements. The establishment of straight roads corresponds to the type II-a of the three directional model previously proposed by the author.
Of the four stages, stage III is argued to be that of the typical "Koto type" medieval temple settlement. In order to examine the historical significance of these settlements, this paper looks at their formation, development, and spread to other regions. As a result, it is proposed that the technology and philosophy behind these urban settlements may have been used in the temple compound towns of the Hokuriku, the medieval castles of Omi, and even at Ando castle. From the examples of Sasaki Rokkaku's Kannonji castle and Kyogoku's Joheiji castle, it can be suggested that these settlements were established at almost the same time as the warrior class began building mountainside sacred precincts known as "Sanjo Goten".
This paper proposes the "religious city" as one type of medieval town in Japanese urban history. The "Koto type" medieval temple settlement is of particular importance as one distinctive "religious city" because it can be hypothesized that these settlements had a strong influence on the urban plans of not just the castles of the Warring States era, but also the defensive layout of Ando castle and the urban plan of the Yoshizaki temple compound town, both of which were connected to the later castle towns of the early modern period.

View full abstract

The authors have previously published basic studies on deep pots decorated with masks and figures in issues 1 (1994) and 8 (1999) of this journal. These earlier works dealt with the classification, distribution and function of these vessels. In this paper, we publish additional materials that have been discovered since our last work.
In 1994, there were 443 examples of these vessels, but this total had reached 601 in 1999 and stands at 750 at present. New discoveries have been made at an average of about 30 a year, but examining recent finds there has not been a major difference in the trend toward new finds and we believe that we can bring our basic descriptive research to an end. Although there was no change in the distribution of these vessels from southwest Hokkaido to Gifu Prefecture, several previous blank areas were filled in Akita, Toyama and other areas. Moreover, we have clarified that the distribution of these vessels corresponds with that of the deciduous forest zone.
New finds from the Early Jomon phase have increased, but there is no change in the fact that the most typical examples come from the first half of the Middle Jomon. The Early Jomon was the phase when post-glacial warming advanced and the current vegetation zones of the Japanese archipelago were established. The overlap between the distribution of deep pots decorated with masks and figures and the highly seasonal deciduous forests is of great importance in understanding the function of these decorated vessels. This strongly suggests that Jomon religion was formed in an attempt to rejuvenate the powers of nature, which were weakened in the winter, through female figures whose fertility compensated for death. We have begun to move beyond description to the systematic analysis of these decorated vessels.

View full abstract

Standing pillar rituals in which a pillar is erected at a ritual location are known from ethnographic examples all over the world. In Japan, the Suwa Taisha Shrine Onbashira-sai festival is a well-known example of such a ritual. The origins of this pillar festival are often said to lie in the Jomon period, but this has not been proven and research on this topic is poorly developed. Furthermore, the use of landscape theory in Jomon research to explain these rituals in the same way as stone circles is problematic. This article analyzes the origin and development of standing pillar rituals in the Japanese archipelago by examining their distribution and locational features. As a result, it was concluded that there were three lineages of these rituals and that they have not simply continued unchanged through to the present.
Following this, a study was made of archaeological and historical materials relating to the distribution, locational environment and morphology of standing pillar rituals through human history. Regions considered were the Eurasian continent, ancient Egypt, and North and South America. Related examples of beliefs about sacred trees are also discussed and materials from ancient China are used to argue that sacred trees and standing pillars have the same meanings. Ethnographic examples are used to show that prior to the 19th century, this link between pillar rituals and sacred tree beliefs was universal. A combination of archaeological, historic and ethnographic information was used to suggest a three-stage developmental scheme and to demonstrate theoretically that the roots of pillar rituals can be found in sacred tree beliefs. Roots in ceremonies related to death and rebirth are also universal and it is concluded that the Jomon-type standing pillar rituals of the Japanese archipelago are typical examples of such ceremonies.
It can be suggested that sacred tree beliefs were born from the forest and tree worship of the beginning of the Neolithic and that standing pillar rituals combined human death and rebirth within the rotating structure of nature. Following that, these rituals were incorporated into social and religious systems as part of the historical development of each region and their objectives and form diversified.

View full abstract

A cluster of 23 side-chamber tombs was discovered at the Gongju Tanji-ri site on the Korean peninsula. Preservation was extremely good and the tomb structure was clear. The grave goods and human skeletal remains were fully preserved. This site is an extremely important find for understanding not just the structural characteristics and chronology of side-chamber tombs on the Korean peninsula, but also the problems, long discussed in Japanese archaeology, of the origin of the side-chamber tombs of the Japanese archipelago and ancient Korea-Japan relations as a whole.
A rough date for the side-chamber tomb construction can be obtained from the typological features of pottery and other objects used as grave goods. From the typology of bowls with fitted covers (futatsuki) and tripods, it can be concluded that side-chamber tombs were built during the early half of the Paekche Yushin phase (late 5 th century).
The structural characteristics of the Gongju Tanji-ri side-chamber tombs are similar to the early side-chamber tombs of the Japanese archipelago which are distributed across northern Kyushu at sites such as Takenami, Yukuhashi City, Fukuoka Prefecture and Uenoharu, Oita Prefecture. These date from the late 5 th to the early 6 th centuries and have quite a few Peninsula-type grave goods. These are important materials for lifting the veil that has so far surrounded the origins of the side-chamber tombs of the archipelago. The excavation of the Gongju Tanji-ri site gives us the opportunity to consider further the possibility that side-chamber tombs originated in the Paekche region.

View full abstract

Located on the left bank of the lower reaches of the Kinokawa River in northern Wakayama Prefecture, Dainichiyama Mound 35 is part of the nationally-designated Iwase Senzuka kofun cluster that extends over the Iwase mountains. From its size and location, this mound had traditionally been seen as one of the main kofun of the cluster. Although the mound is preserved in a designated locality, many details remain unclear from previous research, which includes the plans of the mound and stone chamber and the haniwa surface collection made by Kansai University.
With the aim of preserving and utilizing the kofun cluster, from Heisei 15 Wakayama Prefecture began preservation and repair work at the Iwase Senzuka kofun cluster national historical site, and excavations were conducted at Dainichiyama Mound 35 as part of this work. These excavations clarified various aspects of the mound, including the structure of the central burial and outer mound, the mound size, and its shape. It was also shown that there were projections on the east-west constriction. Excavations on the eastern projection found many haniwa and Sue ware placed in an area surrounded by a row of circular haniwa. Analysis of artifacts during fiscal Heisei 16 confirmed representational haniwa including houses, lids, swords, human figures, birds and horses. Sue vessel forms included kame pots, pedestalled takatsuki, and stands. The representational haniwa include a gliding bird, a bird with a long beak that is probably a crane, a horse haniwa mud guard with armor slats, and the hip roof of a house with pillars sup-porting the ridgepole beam. These types of haniwa are extremely rare, with few similar examples in western Japan. Moreover, this is the first example of a bird-shaped haniwa depicting flight.
The excavations demonstrated that the mound has three levels, but it is not certain if these are three step-like con-struction levels or whether the bottom level should be seen as a foundation platform associated with the mound. There are many points which await further excavation, but the author tends to support the second interpretation based on the currently available evidence. A comparison of tombs of the same age in the Iwase Senzuka cluster with the recently-excavated Imashirozuka kofun is used to support this interpretation and related problems are also discussed.
Dainichiyama Mound 35 may be the largest keyhole-shaped tomb in the Iwase Senzuka kofun cluster, which dates from the first half of the 6th century. The haniwa cluster on the projections is of a quality and quantity that is noteworthy even in western Japan. Further excavations and analysis may well change the picture of Dainichiyama Mound 35 presented in this paper, contributing further to research on Kofun tomb rituals and the local history of the Wakayama region.

View full abstract

The Okishio castle site was designated a nationally designated historic site in fiscal Heisei 10 and given the official name of the Akamatsu castle site. Following this, excavations and surveys at the site were carried out by Yumemae-cho Board of Education, Hyogo Prefecture.
The castle was built on Mt. Shiroyama and is known to have been used by the Akamatsu family, the shugo (provincial constables) of Harima. Traditionally the castle is said to have been built in the first year of the Bunmei era (1469) and to have been used by the five generations of the Akamatsu following Akamatsu Masanori.
The ruins of the castle have around 70 enclosures and, with features covering an area of 600m eastwest and 400m north-south, they make up the biggest castle site in Harima. The geomorphology of the mountain castle can be described as forming two summits: enclosure I-1 which had a military function and the residential enclosure II-1 in the western enclosure cluster. In terms of function, these two areas had mutually complementary functions.
During the archaeological excavations, towers were found in the central enclosure and the gate area in enclosure I-I . The presence of these two towers shows that defensive functions were developed and provide proof that this enclosure was the military court (tsume-no-maru).
It was confirmed that many full-scale residential enclosures with gates and foundation stones were constructed within the western enclosure cluster (clusters II-V), particularly in the enclosure II-1 on the mountain summit. Our ex-cavations demonstrated that construction of these residential enclosures was concentrated at the end of the Warring States period during the time from the Eiroku to the early Tensho eras (1557-1581). In other words, it was shown that the date at which the mountain castle was enlarged and became a full-scale residence was 100 years earlier than was traditionally assumed. At the end of the Warring States period, the political power of the Akamatsu is said to have declined during the time of Yoshisuke and Norifusa, but the construction of this castle leads to the opposite conclusion. In this way, the surveys and excavations at the Okishio castle site have brought new perspectives to research on the Akamatsu.
Within each enclosure cluster, the central cluster (such as II-1) was always accessed by corridor 2 . These areas had high-status garden and stone-built facilities. Entrance-shaped landforms were associated with both the western and eastern ends of corridor 2, but they were closed off from the other enclosure clusters making a sort of central space. From this it was surmised that within the western enclosure cluster there was a status difference between the main enclosures centered on II-1 and the lower status enclosure courts. This demonstrates that this area was where the provincial constable displayed his status and ritual authority.

View full abstract