華語口語請求言語行為-以「任務型小組活動」採集法為例

Abstract

請求言語行為是日常交際中最常見的言語行為之一，參照Blum-Kulka, House & Kasper（1989）的請求言語行為序列中可分為起始行為、中心行為和輔助行為三個部分。過去進行請求言語行為相關研究時，學者普遍傾向採用語篇補全測驗（Discourse completion tasks，簡稱為DCTs）方式進行問卷調查並採集語料。作答過程中受試者往往傾向以簡短、常識性質的形式應答，因此不易觀察到自然真實的言語表現情況；再者，通過書寫作答形式採集口語現象相關語料，其調查結果不一定能真實反映出受試者在實際的交際情境中慣用之言語行為表現。然而，蒐集真實的自然口語語料實屬不易，以田野調查法為例，需取得受試者的同意以進行長時間的近身觀察，方能採集到真實的口語語料。因此，本研究為蒐集接近真實且便於取得的華語請求言語行為口語語料，設計一套「任務型小組活動」，希望透過活動進行的過程能夠採集到較貼近日常生活且真實反映母語者在實際交際情境中的半自然語料。 本研究首先通過前人文獻整理，確立華語請求言語行為的定義和策略類型框架；此外，本研究所欲觀察的易影響請求策略選用的變因有四：包括交際雙方之間的社會距離、年齡輩份差距、性別差異以及請求本身的難易度級別。因此，透過任務型小組活動（Task-based Activity）進行口語語料蒐集時，研究者便事先設計36項不同的請求事項作為活動使用的任務籤，其中包含了「具體物品」、「具體事項」、「抽象事物」、「訊息索取」四種類別，又分成簡易、中等、困難等不同的難易度級別；而受試者篩選方面，四人為一組，每組組員之間皆存在社會距離之親疏、年齡輩份差距和性別等請求變因。讓組內的每位受試者輪流擔任任務執行者，並以隨機抽籤方式決定任務籤內容及達成該請求的對象，藉此觀察不同社會關係之間的交際雙方在提出不同難易度級別的請求時所選用的策略表現差異。 本研究以Blum-Kulka, House & Kasper（1989）的請求言語行為策略分類框架為基準參照，搭配本研究採集之語料中的策略表現，建立出一套本研究的華語請求策略分類進行語料分析。首先，研究者分別針對起始、中心和輔助行為三種序列結構，以前述四種請求變因作為分項進行數據整理與分析，觀察各項變因對說話者在提出請求時的策略選用影響，並彙整出整體語料中所傾向選用的請求策略。在起始行為方面，華語請求言語行為最常使用「稱呼語」作為起始行為策略，佔60.1%（172/286）；中心行為方面，則傾向使用「規約性間接策略」作為中心行為策略，佔52.5%（189/360），其中又以「規約性間接策略」中的「準備性詢問」的選用頻率最高，佔49.2%（177/360）；輔助行為方面，說話者傾向選用「說明原因」作為輔助行為策略，佔41.2%（91/221）；「詢問前的試探」次之，佔38.5%（85/221）；而「降低難度」的選用比例又次之，佔16.3%（36/221）。 根據本研究口語語料觀察，前述四種變因均會影響說話者對請求言語行為策略的選用，只是每個變因所影響的序列表現不盡一致，需將請求各序列進行細項對比，才能夠觀察得出語料中各請求變因在請求言語行為真正影響的部分。

Request is one of the most common speech acts in daily communications and interactions. The nature of the speech act of request has the possibility of humiliating the participants of the interaction thus it imposes certain level of burden to the speaker, which is called face-threatening acts (FTA) by Brown and Levinson's politeness theory. Therefore, when Chinese learners try to use the speech act of request in the appropriate way, they should take the social relationships between both sides of communication, the context of speech, and the degree of difficulty for request into account in order to choose the proper request strategies. According to Blum-Kulka, House & Kasper (1989)’s research, the sequences of request can be classified into three parts, including head acts, and two optional parts—alerters and supportive moves. When studying “request”, previous studies generally tend to use the Discourse completion tasks (DCTs) to collect data. But it turns out that the subjects in DCTs tend to respond in a short, common sense form, so it’s not easy to observe the natural speech performance. On the other hand, whether it’s suitable to use the form of writing to collect oral data is doubtful. Therefore, he results of the survey of DCTs can’t be used to reflect the speech performance of speaker in the actual situation of natural communication. However, this study doesn't only talk about Chinese request in oral situations, but also focuses on how to collect natural oral data and which approach is the most proper way in this study. Before the data collection for this study, there is a definition of the speech act of request, the strategic frameworks of Chinese request based on previous research results. Furthermore, the comparison and assessment of different kinds of previous research methods is also included. In addition, there are four research variants be set up, which are the differences of social distance, age, gender and the degree of difficulty. According to the research needs, this study decides to use “task-based activity” to collect the semi-natural oral data of request. To set up the four research variants above, these variables are added to the subject's conditional setting for the task-based activity in this study. As a result, there is the analysis and discussion of data collection reveals that conventionally indirect strategy is the most frequent strategy of head acts in Chinese request, which accounts for 52.5%(189/360), and social factors would decide on the choice of an appropriate form of request strategies. In terms of alerters, “address form” is the most frequent strategy in Chinese request which accounts for 60.1%(172/286), and it’s also relatively complex and indicating mutual social relationships in certain language environment. Lastly, in terms of supportive acts, the frequency of “Cause explanation” is quite high, which accounts for 41.2%(91/221).