|Abstract: ||多年以來,片語式動詞的學習對於英語為外國語（EFL）的學習者而言, 一直是一個關鍵性的議題。研究表示（Talmy, 1998, 2000）,片語式動詞存在於衛星框架式語言（satellite-framed language）當中, 像是英語以及德語。這類語言會藉由一個主要動詞,並附帶一個介副詞,來表示動作的方向。然而,像是西班牙語及法語,則屬於動詞框架式語言（verb-framed language）,因為這類語言只需要一個單一的動詞,便能表達動作以及其方向這兩個訊息。在台灣,英語為外國語的學習者主要以中文為母語。根據認知語言學家,像是Talmy（2000）以及Slobin（2004）指出,中文可歸類於衛星框架式語言;然而,近年來有越來越多的語言學家相信（Chen and Guo, 2008; Niguchi, 2011）,中文應該被劃分到一個新的屬性當中,也就是等量框架式語言（equipollently-framed language）。不管是衛星框架式語言或是等量框架式語言,就動詞的結構來說,中文並非完全與英文迴異。因此,可以假設的是,對於英語中使用介副詞（如in, out以及on）表達動作的方向的概念,台灣的英語學習者比起日本的英語學習者（母語為動詞框架式語言）應具有較良好的理解能力才是（Yasuda, 2010）。即便語言的屬性不同於英語,藉由明確地教授存在於人類的經驗中,以身體對空間及方向概念為基礎所延伸出來的方向性譬喻,此一認知為導向的教學（cognitive instruction）對片語式動詞的學習是有所助益的。既使對於日語為母語的成人英語學習者來說,此認知為導向的教學法具有相當正面的效果（Yasuda, 2010）,但對台灣擁有初級英語程度的青少年來說,效果似乎不甚理想（Yang and Hsieh, 2010）。因此,本研究即是探討,以認知為導向的教學法是否能幫助擁有中等英語程度的台灣青年學習片語式動詞。共計168名來自台灣北部某所大學的學生參與本實驗的預試及主試,在這兩次的實驗中,受測者皆被分為認知組（CI group）與非認知組（Non-CI group）。兩組非認知組皆授與傳統導向教法,而兩組認知組則接受認知導向教法。非認知組與認知組皆以中文教授,差別在於非認知組使用的教材是中文翻譯文本,而認知組則是使用圖像基模以及多義詞網路。本研究中所使用的片語式動詞結合了由美國當代語料庫所截取的74個最常使用的動詞,以及最高頻的三個方向性介副詞：out, up以及off,所產生出的204種組合。在研究者進行了初步的難度測試之後,篩選出92個片語式動詞,並以之替實驗教材與實驗考題建立資料庫。這些片語式動詞包含15種意義（out具有6個多義, up具有5個多義,而off具有4個多義）,而每一個意義皆平均地由基礎意義和延伸意義組合而成。在預試當中,36個片語式動詞以兩堂各三十分鐘的課程授與完畢;然而,在主試中,75個片語式動詞以二十三週的課程教授（一周一堂,一堂課40分鐘）。除了在預試的兩個組別僅接受選擇題型的前測與後測之外,主試的兩個組別皆參與了選擇題型與填空題型的前測,後測以及延遲後測。預試的結果顯示,以認知為導向的教法在短期學習的角度看來,並未優於傳統教法,且同樣的結果也一再重複於主試中的選擇題型。但是,就長期學習以及延伸意義學習此兩種角度來看,主試中的填空題型顯示,認知組的表現優於非認知組。簡單來說,本研究發現傳統導向教法對短期學習有某種程度的幫助,但就前測到延遲後測的結果看來,認知導向的教法卻能幫助長期的學習。|
Mastering phrasal verbs (PVs) has been a critical issue for learners in EFL environment for many years. PVs are often observed in satellite-framed language (S-language), such as English and German, which expresses the path of a motion by a main verb accompanied by a particle, while languages such as Spanish and French are referred to as verb-framed language (V-language), since their verbs alone convey both path and motion information (Talmy, 1988,2000). Mandarin Chinese, which is spoken by the majority of EFL learners in Taiwan, as some cognitive linguists argue, belongs to S-language (Talmy, 2000, Slobin, 2004), although in recent years, more and more linguists believe Chinese should be categorized into a separate typology that is referred to as an equipollently-framed language (Chen and Guo 2008, Noguchi, 2011), where the concepts of manner and path are expressed in a series of verbs following the main motion verb. Be it a S- or E-language, Mandarin Chinese is not that entirely different from English in terms of motion construction. Arguably, Taiwanese learners should have relatively less difficulties in understanding the path trajectories depicted by various particles such as in, out, on, etc. in English than, let say, Japanese EFL learners whose first language, Japanese, is a V-language (Yasuda, 2010). Even languages of different typology from English could benefit from the cognitive instruction (CI) in learning PVs by explicitly teaching the fact that they converge on orientational metaphors that are extended from spatial orientations grounded in the experiences of human body. This was true for adult subjects in a Japanese EFL study (Yasuda, 2010), but not so for Taiwanese teenagers of basic English level who did not respond positively to this approach (Yang and Hsieh, 2010). In view of this, this present study is designed to investigate whether CI benefited the learning of PVs among young adults of intermediate proficiency in Taiwan.168 subjects in a university in northern Taiwan were recruited for a pilot study and a main study, and the participants in both studies were divided into two groups (CI and Non-CI group). The two Non-CI groups received a traditional approach, which in Taiwanese context is providing idiomatic Chinese translation, while the two CI groups were given CI with image schemas and polysemy networks linking all the senses of tested PVs. The PVs chosen for the study consist of 74 most frequent verbs in COCA (http://corpus.byu.edu/coca/) combining with three most frequent orientational particles, out, up, and off, yielding 204 combinations. After a familiarity and degree of difficulty test, these items were reduced to 92 PVs of 15 different senses (6 of out, 5 of up, and 4 of off) with a comparable distribution of basic senses and extended senses in usage to form an item bank for teaching and testing. In the pilot phase 36 PVs were taught over two sessions (30mins x 2), whereas in the main study 75 PVs were taught over a period of 23 weeks (one session per week and 40mins. per session). Except for the groups in the pilot study which only received comprehension pre-test and post-test, the groups in the main study were given comprehension and production pre-tests, post- and delayed post-tests. The results from the pilot study indicate that CI did not show any advantage over translation instruction in the short-term retention in the comprehension tasks. Such results were somewhat repeated in the main study in the comprehension tasks as well. However, in the productive tasks in the main study, CI group performed better than Non-CI group, particular in the long-term extension and extended senses. To conclude, although the translation approach still shows certain efficacy in this study, it is only limited to short-term retention. CI indeed helped learners attain better learning outcome than translation approach in long-term retention especially from pre-test to delayed post-test.