透過您的圖書館登入
IP:18.191.234.62
  • 學位論文

陶容器作為研究分析單位之效益評估-以台中清水鎮社口遺址的陶器及其標準化研究為例

Vessel Lot as an Analytic Unit and the Evaluation of Its Efficiency-A Case study on the Pottery from She-Kou Site and its Standardization

指導教授 : 陳瑪玲

摘要


在考古學研究中,「陶片」與「陶容器」雖然都指涉過去人類所使用的陶製容器,卻分別暗示了器物「破碎」與「完整」這兩個相對概念。其中,陶容器才是過去人類認知和使用的單位,但在一般情況下,破碎的陶片更容易出現在考古遺址中,而許多完整陶容器的屬性如整體器形、口徑、高度等資訊,會因為破碎成陶片而跟著佚失,從而在考古學家進行陶片分析的時候,影響了分析策略、方法甚至分析結果。由於意識到這兩種不同單位在考古研究上可能造成的差異,有學者如Chilton(1994)開始提出應該先把零碎的陶片整合,利用屬性分析建構出陶容器單位,再以此進行考古學研究,Chilton等人雖然確實應用了這個方法和概念,卻沒有對所建構的陶容器單位在有效性方面作進一步的檢驗,也沒有去比較陶片和陶容器這兩個不同單位進行分析時的效益差別,因此這樣的概念是否能廣泛且有效的應用在各個考古遺址與材料上,無法從過去的研究中得到證明。 延續前述以陶容器為單位進行分析的想法,本論文主要的目的便是要對這個概念進行驗證,並且回答兩個基本的問題:1)陶容器單位能否有效建構、2)陶容器和陶片這兩種不同的單位用於分析在效益上有無差異?為了了解陶容器能否有效建構,筆者以社口遺址出土的陶質資料進行研究,先以經過修復組合成陶容器的樣本為例,對其中的陶片進行屬性登錄,尋找最適合用來進行陶容器單位建構的分群屬性組合;其次,將所選出的屬性組合利用統計的集群分析方法套用在未知群組的陶片樣本中,依照分群的結果建構出陶容器單位,並且根據樣本分布的狀況檢視陶容器單位的建構成果是否能被認定為有效。如果能有效的建構出陶容器單位,就可以進一步討論陶容器和陶片這兩種不同的單位分析成效上的差別,此處選擇以陶容器標準化製作的研究當作討論的議題,分別利用陶片和陶容器單位,討論社口遺址夾砂和泥質兩個不同群組的陶容器之間,標準化程度的差異,以便評估兩種不同單位用於考古學議題討論時所展現的效益與得到的成果。 在本研究所得到的結果方面,筆者認為雖然在集群分析的過程中,陶片分組的結果不可避免的會產生一些誤差,但是利用已知能拼湊組合的陶片樣本比對評估的結果,仍可以證明陶容器確實能被有效的建構;而在效益評比與標準化研究的層面,社口遺址的夾砂陶和泥質陶群組利用陶片和陶容器兩種單位進行分析時,都呈現出夾砂陶的變異程度小於泥質陶的趨勢,不過根據最後的統計檢定結果,以陶片為單位進行標準化比較研究的時候,這種趨勢無法明確的表現出來,只有當以建構出來的陶容器單位進行分析時,可以進一步證明夾砂陶的標準化程度高於泥質陶,筆者也從而推斷,在標準化議題的研究方面,以陶容器為單位進行分析確實能得到較好的結果,也進一步鼓勵我們能將這種概念帶入日後的研究之中。 除了在方法論上證明了陶容器建構的可行性與陶容器為單位進行分析的效益,本文對社口遺址陶容器進行標準化評估的結果,可以進一步就其背後的意涵進行討論。以本文的分析結果來看,夾砂陶和泥質陶之間標準化程度的差異。反映的並不是生產組織或專業化表現的差異,而是社口遺址人群對於兩種不同材質陶器的認知不同造成的結果,其中夾砂陶容器包含的器物類型單純,可能對應某種特殊的功能;泥質陶容器群組中的器物類型變異較大,顯示這類器物在人們認知、製作和使用等層面的限制或規範比較寬鬆,也因此造成兩群組標準化程度的差異。除了了解遺址內不同群組器物的標準化程度,筆者認為日後也能以本研究的結果為基礎,進一步的對社口遺址周圍性質接近的遺址進行陶容器單位的建構和分析,進一步比較區域內不同遺址之間的器物標準化、生產組織專業化等層次的議題。

並列摘要


In this thesis, I would like to answer two questions: 1) Can vessel lot be correctly constructed by attribute analysis? 2) Will it be more efficient to take the vessel lot rather than the sherd as a unit to discuss archeological topics? I take the She-Kou site as an example to answer these two questions. In archaeological studies, “vessel lot” and “sherd” both refer to the idea of pottery. Although the broken pottery is often excavated from archaeological sites in the form of sherd, in the real world, the unit that people cognize and use is “vessel lot” rather than “sherd”, and some attributes such as typology, diameter and height will also miss if we use sherd as an analytic unit. Being aware of this shortcoming, Chilton (1994) started to advocate taking vessel lot as an analytic unit to deal with archaeological subjects, She believes that we can integrate sherds into vessel lot by using attribute analysis. Although Chilton emphasizes the concept of vessel lot, we still need to consider its feasibility. Only when my two questions lead to positive conclusions can the concept of using vessel lot unit be applied into the research of other regions or archaeological sites. To answer the first question, I take the pottery materials from She-kou site to do some ceramic analysis. First, I use sherd data collected from restored vessel to run cluster analysis, and then try to find out which attributes are relatively important when we try to construct a vessel lot. After that, I use those attributes to run cluster analysis with all the scattered sherds. This analysis leads me to construct 22 different vessel lots. At last, I check sherd samples in each vessel lot. By the statistics tests and directly observation, I may conclude that with the appropriate attribute, vessel lot can be constructed by cluster analysis. As for the second question, I take standardization as an example to examine if it would lead to any difference by using “sherd” and “vessel” unit. In this part, I detect that both sherd and vessel lot reveal a trend that in the She-kou site, the group of tempered pottery is more homogeneous than the group of non-tempered pottery. Nevertheless, when we use sherd as a unit, this trend is hardly shown in the statistic test. On the contrary, if we use vessel lot as an analytic unit, this trend can be displayed clearly. The statistic test also confirms this result. In this case, I evaluate the efficiency of vessel lot unit, and the result is good enough to encourage the use of vessel lot as an analytic unit in archaeological study.

參考文獻


Arnold, Dean E. and A. Nieves
1992 Factors Affecting Standardization. In Ceramic Production and Distribution: An Integrated Approach. Ed. By G. Bey and C. Pool, pp.113-214, Westview Press, Boulder, Colorado
Arnold, Dean E
1993 The standardization hypothesis and ceramic mass production: Technological, compositional, and metric indexes of craft specialization at Tell Leilan, Syria. American Antiquity. Vol. 58(1):pp. 60-80
Chilton, Elizabeth S.

延伸閱讀