A Study on Korean Voice System and its Application to Acquisition and Teaching
This study attempt to redefine and classify Korean voice system to help leaners of Korean as a second language.
Traditionally, voice is understood with a prototypical c...
A Study on Korean Voice System and its Application to Acquisition and Teaching
This study attempt to redefine and classify Korean voice system to help leaners of Korean as a second language.
Traditionally, voice is understood with a prototypical concept of active-passive opposition (Shibatani 2004). However, this previous perspective on the voice system has many limitations to explain voice phenomena.
In cognitive linguistics and functional-typological perspectives, voice was defined as ‘a grammatical category which represents correspondences between syntactic structure and semantic relation that the predicate responds to the main argument(the subject) in a sentence, related to the participants’ participating ways in the various event types of actions’ (Shibatani 2006).
Based on the Usage-based model, it is believed that language structure emerges from its use, and the language reflects personal experience accumulated by the conceptualization process. In the view of the constructional approach, construction itself also has its particular meaning distinguished from the words, and voice can be classified into different constructions. Some voice constructions are associated with morphological markers, which can be polysemous, and these poly-functional markers can be identified using different sentence constructions (Goldberg 1995). And the similar concepts can be adjacent in the conceptual space and coded the same forms. Semantic maps represent functional domains of grammatical forms on the conceptual space and semantic boundaries between functions. Using semantic map model, polysemous voice markers can be distinguished by different voice categories. Prototypical transitive construction can be changed covertly from active voice to passive voice. However, some types of transitive constructions cannot. Even it is possible these types to be classified as not prototypical active voice but as other categories of voice.
Therefore, Korean voice categories can be classified by semantic map boundaries, different sentence constructions, and restrictions in voice alternation. Based on these criteria, this study classified the Korean voice into active, causative, passive, reflexive, reciprocal, potential, spontaneous, body action middle, and emotional middle, and they can be represented within voice continuum. Each voice category reflects the different image-schema of various event types.
Active voice is a construction whose most prominent participant is simultaneously the most energetic one in the sentence. Causative voice is a construction with the causality of performed actions that a causer caused a causee results in change in state or in action, and this is expressed with special voice marker in a verb phrase. Passive voice is a construction that the patient participant, affected by the event, is appeared the most prominent subject in a sentence and marked in verb with a specific voice marker. Reflexive voice is a construction which the patient correspondence with the agent oneself in a transitive event. Reciprocal voice is a construction that participants typically affect each other in the events. Potential voice is a construction indicating the degree of the property of the theme. Spontaneous voice is a construction with the action scenes which the agent is omitted and the patient does an action of the event automatically. Body action middle voice represents an event which naturally affects body of the actor. Emotional middle voice is a construction that represents a experiencer’s metal state.
Korean voice categories form a continuum and represent the shared characteristics through the similarity between constructions. The [N-i V] construction is used to represent 1-participant active, spontaneous, body action middle, and emotional middle voice. The construction [N-i N-ul V] is shared by 2-participant active, reflexive, body action middle and emotional middle voice. Also, the construction [N-un N-i V] represents potential and emotional middle. Korean morphological voice marker ‘-i-, -hi-, -li-, -ki-’ are commonly used in causative, passive, potential, and spontaneous voice. While reciprocal voice is represented by the construction [N-i N-wa V].
As there is a correspondence between the construal of a scene and the sentence structure, second language learners of Korean must learn how Korean native speakers perceive the construction. In this study, how Korean second language learners and native speakers process the active and the passive were examined with the behavioral study and ERP study using the picture-sentence verification task. For observation of acquisition of various voice categories, a selective blank task and an acceptability judgment task were used. First, in behavioral response of the judgment for the active and passive sentences, each group processed the passive sentence more slowly and inaccurately than the active sentences. As a result, Korean native speakers showed faster and more accurate processing of the nominative passive (“oppa-ka cap-hy-eyo.” My brother is caught.) than the dative passive (“enni-eygye cap-hy-eyo.” Caught by my sister.). The Chinese learners of Korean, on the other hand, processed the dative passive faster and more accurately than the nominative passive. This result is consistent with the results from the ERP study of the English-speaking heritage learners of Korean.
It is possible that second language learners of Korean process the passive sentences as a structure rather than as a construal for a scene. They used word order as a important cue for sentence processing as same as in their mother tongue (Chinese or English). However, it took more time for Korean native speakers to process the dative passive, which are used for a pragmatic purpose that highlight again the degraded agent in the context where the patient is salient in the typical nominative passive. The ERP experiments showed the similar results. The results of Korean native speakers showed the N400 component for dative passive which represents the semantic processing. However English-speaking heritage learners showed the ELAN (Early Left Anterior Negativity) component which represents the rapid automatic processing. In addition, in the experiment of acquisition of various voice categories, second language learners of Korean showed different patterns from those of Korean children and adults. Learners felt difficulty in learning causative, passive, potential and spontaneous voice, and among them especially the accuracy for potential voice was low.
Therefore, if L2 learners do not understand how Koreans perceive the construal of events, they will make voice construction errors. Therefor, there were limitations in traditional teaching method using active-passive voice alternation. To be more effective teaching Korean voice, it is necessary to make the functions of each voice construction more explicitly noticeable for the learners (Littlemore 2009). Based on the results of this research, a cognitive teaching approach can be adapted to help learners to recognize the characteristics of voice in relation to a scene, and to construct voice constructions by noticing characteristics of them. Therefore, a new teaching model, which includes the process of role discovering in a scene recognition, paying attention to the form in the situation, and constructing the target construction, so called DNC (Discovering-Noticing-Constructing), was proposed.
，韩语论文，韩语论文范文 |
