「標誌性」與「典型類型性」對了解形態系統發展的效用

On the Utility of the Concepts of Markedness and Prototypes in Understanding the Development of Morphological Systems

研究一個語言或語群的形態系統的歷史發展，有時會面臨一種情況：有兩個或兩個以上的語意範疇呈現同樣的標誌。在這種情況之下，研究者必須要了解哪一種用法是最古老的，以及探究其最可能的發展途徑。此對有古老文獻的語言來說，問題不大，可是對沒有文字的語言，通常就比較困難。本文旨在探討可以輔佐我們解決上述困惑的兩種理論概念：標誌性（markedness）與典型性（prototypes）。標誌性的理論概念，可以輔佐我們知道某一個形態形式的哪一種用法在不同語言中較普遍。而典型性的理論概念卻能輔佐我們知道哪一種語意範疇（或語意範疇中的份子）在心理上較顯著，在語意上較基本。本文在闡述標誌性與典型性的理論概念作為探究語意標誌最可能發展途徑的工具的同時，提出以下藏緬語中的兩種同形標誌現象為論證的例子：獨龍語裡反身、中間和狀態化的同形標誌現象以及在許多藏緬語中，同一個格助詞用來標記幾種不同語意角色的同形標誌現象。本文探討的結果顯示：獨龍語的反身、中間和狀態化的標誌，最初可能只有標記反身狀況的作用，後來逐漸由語意的引伸擴散到中間狀況。由於中間態的語意特點，又被引伸到狀態化的作用。另一方面，在比較145個藏緬語格助詞的同形標誌現象的結果顯示，這些格助詞的可能發展如下：從格＞工具格＞方法狀語＞施事格＞前行或使役從屬連詞；隨同格＞工具格；處所格＞與格＞受事格＞目的、時間或條件連詞。

In attempting to understand the history of the morphology of a language or group of languages, we occasionally face a problem of isomorphy, where two or more semantic categories evince the same formal marking. We then must decide which use of that particular form of marking is the oldest, and also determine the possible source and path of development of the marking. In languages with written documents of great time depth this is often not a problem, but in unwritten languages it can be quite difficult. This paper discusses two tools that can be used for this purpose: the concepts of markedness and prototypes. Markedness can tell us which use of the morphological form is cross-linguistically common, while prototype theory can tell us which semantic categories (or members within a semantic category) are more psychologically salient or semantically basic. Using two relatively uncontroversial examples, this paper suggests that aside from the standard types of evidence used for determining the probable path of development of a form of marking, particularly in cases of isomorphy we can use the concepts of markedness and grammatical prototypes to help us determine the most probable path of development of isomorphic forms. In the first part of the paper the concepts of markedness and prototypes used are explained, then two examples from Tibeto-Burman languages are presented: the isomorphy of the reflexive, middle, and ''stativizing'' markers in Dulong, and the cross-linguistic phenomenon within Tibeto-Burman where a single case marker is used to mark a number of different semantic roles. Using the concepts of markedness and prototypes, it is shown that the isomorphy of the reflexive, middle and ''stativizing'' markers in Dulong/Rawang is most likely the result of a marker originally having only a reflexive use being extended to cover middle situations, and then, because of the nature of middles, being further extended to the use as a ''stativizer,'' and that the patterns of isomorphy of case marking in the 145 Tibeto-Burman languages surveyed probably developed along the following clines: ablative > instrumental > manner adverbial > agentive > anterior or causal clause subordinator; comitative > instrumental; locative > dative > patient > purposive, temporal, or conditional clausal subordinator.