著者

網 伸也

出版者

国立歴史民俗博物館

雑誌

国立歴史民俗博物館研究報告 (ISSN:02867400)

巻号頁・発行日

vol.134, pp.111-136, 2007-03

---

古代都城において「京」の空間に方形街区が形成されるのは天武朝以降であり、藤原京(新益京)には計画的な条坊街区が造営された。そして、平城京以後の諸宮では、「京」における条坊の存在が既成事実として議論されてきた。しかし、「京」は王権の所在地として周辺地域から視的あるいは理念的に区別される空間であり、方形街区としての条坊の有無は本質的に「京」の必要条件とはならない。実際に、奈良時代における「京」の概念には条坊街区の存在はあまり考慮されておらず、宮を中心に広がる特別な政治領域を「京」として捉えていたことがわかる。そして、宮城を取り囲む「京」に街区が形成される場合にも、計画的に条坊街区が造営される場合と、必要に応じて街区が造営されていく場合が想定できる。ここでは、まず都城成立期である藤原京の考察を行い、日本の古代都城がいかにして確立していったかを明らかにし、平城京をはじめとする奈良時代の「京」の実態分析を行った。その結果、古代都城の構造には、全体の京域条坊プランを計画的に設定し宮城もその計画線の中に収めていくタイプ(計画線閉合型)と、まず宮の造営を行い必要に応じて京域の条坊を施工していくタイプ(中軸線開放型)があることが判明した。厳密にいえば、全体の方形地割計画線を設定する前者のタイプは藤原京と平城京だけであり、その構造原理は形を変えて平安京にも引き継がれたと想定できる。その他の都城は宮の造営が先行し、宮の造営中軸線あるいは東西計画線を基準にして京域街区が形成された。長岡京も宮城の造営がまず先行して行われており、その京域にできるだけ計画的条坊を施工しようとした特殊な都城であったため、構造的矛盾を孕む結果となってしまったと考えられる。桓武天皇の再度にわたる平安京遷都は、特殊な長岡京造営の中で実現することができなかった計画的都城の完成をめざして行われたと考えられるのである。The adoption of a block street pattern for the space of the "capital" ("kyo") in ancient walled cities began during the Temmu era when Fujiwara-kyo (Aramashi-kyo) was built following a planned grid pattern. The existence of a grid pattern as a fait accompli for capitals with palaces from the time of Heijo-kyo onwards has been the subject of debate. However, as the location of imperial power, the capital was a space that was visually and conceptually separated from the surrounding area so that in essence the presence of a grid pattern for the streets was not an absolute requirement to make it a capital. In fact, little consideration was taken of the existence of a grid pattern for streets within the concept of a capital during the Nara period. We know that a special political sector that expanded outward from a palace at the center was viewed as constituting a capital. We may also assume that even when streets were built in a capital with an enclosed palace there were instances when a planned grid pattern was adopted and others in which streets were built as the need arose.This paper considers Fujiwara-kyo which was built at a time when walled cities were first established in Japan. It discusses the process by which ancient walled cities were built and investigates the capitals of the Nara period, of which Heijo-kyo was the first. The findings reveal that either one of two structures was adopted for these ancient walled cities. One entailed a planned grid pattern for the entire capital with a palace incorporated in the plan (planned closed type), while in the other type the palace was built first and a grid pattern implemented for the area of the capital as the need arose (central axis, open type). Strictly speaking, the first type with its grid pattern covering the entire area applies to only the Fujiwara and Heijo capitals. However, following a change to its underlying principle, this structure was also most likely adopted for the Heian capital. In other walled cities the construction of the palace occurred first, whereupon the streets were created based on the palace located along a central axis, or on a line projecting east to west. In the case of Nagaoka-kyo too, the palace was built first, and because an attempt had been made to build a unique walled city following a planned grid pattern wherever possible, the result included some structural contradictions. It is possible that Emperor Kammu shifted the capital to Heian for a second time with the aim of completing the planned walled city that he was unable to achieve when building Nagaoka-kyo.