透過量詞短語學習分格:人工語言的學習者如何標記直接賓語

The Partitive through Quantification in an Artificial Language: How Learners Mark the Direct Object of a Transitive Phrase

霍芝龍(Jesse Holmes)

指導教授 ： 林若芙

國立清華大學/人文社會學院/語言學研究所/碩士(2021年)

摘要

由於芬蘭語的分格/賓格格位系統在世界上實屬罕見，許多語言學家曾經推測過漸生題旨(incremental theme)、量化(quantification)、完成性(telicity)在形成這種系統的過程扮演什麼角色。過去有些行為研究探討過，在學習另一種表達完結性的語言系統時，第一語言有什麼樣的影響，並發現第一語言的影響在學習者達到精通的程度時會消失。本研究欲探討分格/賓格格位系統的發展，以及其演變過程和量化、完結性、動詞完整性(verbal integrity；由漸生題旨和數量敏感性衍生的概念)，基於前人的研究，我們設計了一套人工語法，透過名詞量化詞組，間接地教漢語母語者類似芬蘭語的分格/賓格格位系統。在受試者習得人工語法後，我們觀察受試者在翻漢語簡單句子到人工語言時，如何標記直接賓語。根據受試者的反應，我們發現到幾個明顯會影響標記直覺的因素：完結性、動詞完整性、時態。結果顯示，受試者仰賴完結性和動詞完整性來決定標記。然而，人工語言的學習時間、程度、格位標記的量化用法只能預測完結性的仰賴程度。簡言之，受試者愈熟練格位標記量化用法，仰賴完結性的可能性就愈高。本研究結果為芬蘭語分格的歷史發展提供了可能的解釋：分格的非完結性用法和量化用法有關聯，且必須從個別動詞所決定的實際情況抽象化。本研究亦示範如何以人工語言測試歷史發展的假說，並啟發未來更多有關完結性的行為研究。

關鍵字

完結性 ； 分格 ； 量化 ； 人工語言 ； 語言學習

並列摘要

The partitive-accusative case system of Finnish is cross-linguistically unusual, and many linguists have speculated how such a system might arise with regards to incremental themes, quantification, and telicity. Several behavior studies have investigated the effects of first languages (L1s) on learning different systems of encoding telicity, showing that L1 effects disappear with high-enough proficiency. Accordingly, to investigate the development of a partitive-accusative case system as well as the relation between quantification, telicity, and verbal integrity (relating to incremental theme and quantity-sensitivity), an artificial language was used to teach native Mandarin speakers a partitive- accusative case system similar to Finnish, indirectly through nominal quantification. After the artificial language was acquired, we observed how the participants marked the direct objects of transitive verbs when translating simple sentences from Mandarin into the artificial language. Based on their responses, we then identified what factors influenced the participants' intuitions the most: telicity (telic vs. atelic), verbal integrity (incremental vs. holistic), and/or tense (non-past vs. past). The results showed significant reliance on telicity as well as verbal integrity; however, only reliance on telicity was predicted by proficiency, acquisition time with the artificial language, and the quantification uses of the case marking. In short, the higher the subject’s proficiency in the language and association of quantification with the case marking, the more likely they were to rely on telicity. The results of this study offer insight into the potential development of the partitive case in Finnish, namely, how the atelic use of the partitive requires the association of quantification and abstraction away from verb-dependent usage. This study also hopes to act as an example of how artificial languages can be used to test historical hypotheses, and as an inspiration for future behavioral studies on telicity.

並列關鍵字

telicity ； partitive case ； quantification ； artificial language ； language learning

參考文獻

Brooks, P. J., & Kempe, V. (2013). Individual Differences in adult foreign language learning: The mediating effect of metalinguistic awareness. Memory and Cognition, 41, 281–296.

Cowan, J. W. (1997). The Complete Lojban Language. The Logical Language Group, Inc. Demonte, V., & McNally, L. (Eds.). (2012). Telicity, change, and state: A cross-categorial view of

event structure (Vol. 39). Oxford University Press.

Dowty, David R. 1991. Thematic proto-roles and argument selection. Language 67, 547–619. Ettlinger M, Morgan-Short K, Faretta-Stutenberg M, and Wong P C (2016) The relationship between

artificial and second language learning. Cognitive Science 40, 822–847.