世界上很少地方近半世紀來的考古發現可與中國相比。在一九三○年代的殷墟發掘之後，一九六○年代以來更多、更豐盛的考古文化被發現或更為人所知。其中最引人矚目的是長江上游的廣漢三星堆文化與下游的良渚文化。殷墟商文化居於中原，且早已見於中國文獻記載，然而上述三星堆文化與良渚文化等多發現在中國邊緣地區，這些地區的上古時期在中國歷史記憶中是一片蠻荒。因此這些考古發現讓中國學術界十分震驚，甚至震驚了世界上所有對中國歷史有基本常識的人。這一篇論文，始於一個簡單但被忽略的問題：為何當真實的「過去」出土時，人們會對它感到如此震驚？顯然他們心中作為常識的「過去」，與出土的「過去」之間有相當差距。因此我們應問：過去人們對中國邊緣地區的歷史知識何時產生，為何及如何產生？真實的過去如何被遺忘？ 本文之目的有二：一為解答上述問題的「歷史知識考古」，另一則是透過此研究，我提出一些有關中國文獻歷史記憶之「文本分析」(textual analysis) 與中國民族之「邊緣／邊界研究」(border/frontier study) 的方法與理論觀點。在歷史知識考古方面，我說明由春秋戰國時期以來，在蜀地與吳地的華夏化過程中，此兩地之社會歷史記憶如何被失憶，以及，新的歷史記憶如何使得吳與蜀成為華夏之域，吳人與蜀人成為華夏之人，但皆處於華夏邊緣。這些歷史記憶稱，一個來自華夏的英雄祖先，遷徙或被分派到這些地方，成為本地統治者並帶來文明教化；在此之前，本地的「過去」是一片蠻荒。我稱此為「英雄徙邊記」歷史敘事。在「文本分析」與「邊界／邊緣研究」方面，我說明「英雄徙邊記」所蘊含的歷史心性 (historicity)、文類 (genre) 與其結構化情節 (schematic plot)，及這些文本結構、符號與「情境」之間的關係。

Few areas of the world can boast the volume of archaeological findings equal to that discovered in China in the past half century. Since the 1960s, following the great excavation of ancient remains of the Shang dynasty, a number of ancient civilizations have been discovered, and our understanding of the ancient world has improved a great deal. For instance, the Sanxingdui culture in the upper Yangtze River valley and the Liangzhu culture in the lower Yangtze River valley have attracted the attention of most scholars. Unlike the Shang civilization that was situated in the Central Plain and documented in ancient records, these two cultures were located far beyond Central Plain and in areas long consider uncivilized in the historical memory prevalent in early China. Therefore, the discovery of these ancient civilizations has astonished academic circles in China and even people all around the world who simply have a basic knowledge of Chinese history. This article starts with a simple but too often neglected question: why is the discovery of the truth about these civilizations so surprising? People were surprised at these archaeological findings because a discrepancy exists between the commonly acknowledged truths about “the past” and the realities of “the past” unearthed in peripheral areas. Therefore, we ought to ask when and how this historical perception of peripheral regions and its peoples formed, and when and how the actual past was forgotten.The purpose of this article is two-fold. First, I aim to unravel the “archaeology of historical knowledge”. Second, based on this research, I wish to present theories and methods for analyzing “texts” as historical memory passed down in oral and writings, and for border/frontier study on Chinese peoples. From the Spring and Autumn period to the Han dynasty, throughout the sinification of the Wu 吳 and the Shu 蜀 peoples, these two groups obtained a new historical memory and therefore became Huaxia on periphery. One stereo type of this new historical memory was: a hero migrated of had been dispatched to a far away land from Huaxia territory and he became genetic ruler of the land. This is what I term the “hero-want-to-frontier” historical narrative plot--in various hero-want-to-frontier narratives. I then go on to explain the relations between these “structures” and signs in narrative and their concomitant context of social reality.