本文探討為什麼在台灣災難之後總是重覆著相似的模式。筆者贊成李丁讚 (1997)所說的因為台灣尚未跳脫傳統主義的生活情境，行動模式受到未經反省的 習慣決定，因此難以有清楚的意識和溝通。不過，本文探討國姓鄉南港村的二 個例子並不缺乏李丁讚所言之第一步的情感建構階段，但是災難的「記憶」論述 與再現，催化並轉化感情，使得情感建構階段很快消失，既沒能跨越到社會交 互性論述，甚至成為官方合法性論述的動力。一個例子是「震爆點」上的九份二 山國家地震紀念園區；國家以及專業規劃團隊投入相當大的人力與金錢，彷彿是 國家直接在這個地景上銘刻它對921地震的記憶敘事。但是它從一開始就是龐大 的官方記憶，要轉化在地的傷痛記憶。另一個例子是「鹿神祭活動」，相對於「九 份二山國家921地震紀念園區」比較靜態、日常、官方、銘刻式的記憶，鹿神祭 在表面上以Connerton (1989)著名的區分，則具有動態、特殊、地方、表演式的 記憶。透過這二個例子，筆者要闡明的重點是，災難的重覆、災難論述的永恆 回歸，是因為遺忘，社會記不起來、記不清楚，或更糟的是「被扭曲」的記憶， 每次的敘說充滿了各式各樣的情感，然而這些情感似乎總像是第一次面臨災難， 沒有真正變成溝通，並且進而發揮行動上或學習上的效果。

This article studies why similar patterns of coping with disasters are repeated in Taiwan. To explore the issue, I take two cases in Nangang Village, Guoxing Township. The first case I examine is Jiufener Mountain National Earthquake Memorial Park, located at the epicenter of the 9/21 earthquake in the village. The state and the professional design teams put in place tremendous manpower and money to build the Park in a way that appears to inscribe onto the landscape the state’s grand narrative of the earthquake, which from the beginning aimed to convert local memories of suffering into an official memory. The second case is the “Deer Deity Festival” also held in the village after the earthquake. The Festival is distinguished from the Park, a fixed, routine, official, and inscriptive memorial-ization, by its dynamic, particular, local, and performative character, to use Con-nerton’s famous categorizations (1989). Through analysis of the two cases, I want to show that when a disaster strikes, discourses are easily transformed into official memory without the critical construction of social communication. When there is no learning and communication, a society may not act to change, and the patterns of coping with disaster are doomed to repeat.