駕馭混亂語篇：口譯學生逐步口譯之連貫策略

Abstract

本研究旨在探討不同訓練階段（碩一、碩二、通過專業考）的口譯學習者，在逐步口譯模式下如何應對結構鬆散、語意混亂的語篇，並透過口譯策略改善譯文的條理與連貫性。儘管過往研究已探討口譯中的銜接與連貫，但對於結構鬆散、語意跳躍等真實情境中的常見狀況，尚缺乏系統性的實證探討。為填補此研究缺口，本研究邀請十二位受試者進行實驗，口譯素材使用兩篇由研究者設計的演講，其中刻意加入離題、語序顛倒、轉折詞不足等語篇不連貫特徵。本研究採用混合研究方法，量性部分的評分項目分為「訊息正確性」、「表達流暢度」及兩項連貫性評分指標——「結構邏輯性」與「段落連接清晰度」；質性部分則透過實驗後的刺激回想訪談，深入了解受試者策略使用背後的決策過程。分析結果指出，所有訓練階段的受試者在譯文中皆普遍運用了三類連貫策略：重組、顯化與選擇性省略，且策略使用頻率與連貫性或整體表現分數之間，並無顯著相關。表現較佳的受試者往往針對語篇的混亂程度作出取捨，當內容跳躍或缺乏邏輯銜接時，會主動進行重組或補充；若判斷內容仍可被聽者理解，則傾向不過度干預，避免徒增認知負擔。本研究發現，連貫性的表現與訓練階段關聯不大，策略的運用亦非一套制式可套用的技巧。是否使用某種策略或使用頻率，並不能反映表現優劣。真正影響連貫性的關鍵，在於口譯員能否根據語篇特性與聽者需求，做出合適的判斷，讓口譯產出結構清楚、表達流暢。

This study investigates how trainee interpreters at different training levels manage incoherent source discourse during consecutive interpreting, with a focus on the coherence strategies they employ under cognitively demanding conditions. While prior research has addressed cohesion and coherence in interpreting, few studies have examined how interpreters respond to disorganized, spontaneous speech—a common challenge in real-world settings. Twelve participants, representing three stages of interpreter training (first-year, second-year, and post-professional exam), were tasked with interpreting two specially designed speeches containing deliberate incoherence (e.g., digressions, disrupted sequences, and weak transitions). A mixed-methods approach was adopted: quantitatively, participants were assessed on accuracy, delivery, and two coherence-related criteria (logical progression and clarity of transitions); qualitatively, post-task interviews explored their perceptions of task difficulty and reasoning behind strategy use. The analysis identified three main coherence-enhancing strategies in participant output: Restructuring, Explicitation, and Selective Omission. These were employed across all training levels, but frequency of use did not consistently predict higher coherence or overall performance scores. Correlation analyses showed weak and non-significant links between strategy counts and coherence ratings. Instead, interpreters who achieved better coherence tended to intervene selectively—prioritizing structural clarity or flow when the discourse breakdown posed genuine challenges to comprehension, and avoiding unnecessary effort when the payoff was limited. Rather than treating strategy use as a fixed skill tied to experience level, the findings suggest that coherence in interpreting is shaped by individual judgment—specifically, how interpreters assess when and how to intervene based on the discourse context. The study contributes to interpreter training by offering a situational perspective on coherence-building and highlighting the importance of flexible, goal-oriented decision-making.