英语双及物小句的认知研究 (语言新专题研究系列)(txt+pdf+epub+mobi电子书下载)


发布时间:2020-07-30 18:15:41

点击下载

作者:田朝霞

出版社:世界图书出版公司

格式: AZW3, DOCX, EPUB, MOBI, PDF, TXT

英语双及物小句的认知研究 (语言新专题研究系列)

英语双及物小句的认知研究 (语言新专题研究系列)试读:

《语言科技文库》总序

李葆嘉

当代语言学已经进入了一个科学与技术的互补时代,信息处理水平成为衡量国家现代化水平的重要标志之一。知识世界的载体是语符系统,信息处理的根本对象是语言信息处理。与计算机的出现使得语言符号有可能成为数据处理对象相似,神经科学实验仪器设备的应用,使得在大脑神经层面探讨语言机制成为可能。这些无疑都引导语言研究走向科技化,“语言科技新思维”(李葆嘉2001)应运而生。所谓“语言科学”包括理论语言学、描写语言学、历史语言学、应用语言学等分支学科,所谓“语言技术”指语言研究的现代技术手段,包括语言信息处理、语音实验分析,以及语言的神经、心理和行为实验分析的技术手段等。就语言信息处理而言,又可以分为语料库研制技术、知识库研制技术、知识挖掘和抽取技术、句法信息处理技术、词汇信息处理技术、语音信息处理技术、语义信息处理技术、语用信息处理技术等。

2001年5月,南京师范大学文学院创办了史无前例的“语言科学及技术系”,率先迈出了从传统文科教育范型向现代科技教育范型转变的步伐。“十五”期间,南京师大“211工程”重点学科建设项目“语言信息处理与分领域语言研究的现代化”(陈小荷教授主持),以基础平台建设、资源建设和理论探索等为主,迈出了语言科技研究的一大步。“十一五”期间,南京师大文学院、外国语学院和国际文化教育学院联袂申报“211工程”三期重点学科建设项目。该项目以“语言科技”为引导,以“多学科交叉、跨院系整合、开放型营运”为理念,建设具有前瞻性、原创性、成长性的语言科技高级工作平台。以典型课题的工作原理为核心,进行资源开发和系统研制,拓展语音科技、二语习得的神经机制研究、言语能力受损儿童的语言能力研究等新方向。同时造就新一代学术领军人物和培养一批高层次复合型人才,以期形成一支高水平的交叉学科团队。该项目设计,体现了工作平台建设、理论创新、应用研究、人才培养、团队建设的学科发展一体化思路。其旨趣在于,加速语言研究从传统文科范型向现代科技范型的转变,以引领21世纪语言科技的新潮流。

作为新兴交叉学科项目,通过教育部组织的专家匿名评审,“语言科技创新及工作平台建设”(2008~2011)获批,总投入1000万元。总体而言,这一“语言科技创新”团队,分支学科齐全,专业知识互补。涵盖了理论语言学、计算语言学、语义科技、语音科技、实验方言学、历史语言学、神经语言学、二语习得研究、话语行为语言学等领域。这一期间,项目组成员获批的国家级基金项目达20多项。该项目理念之前瞻、实力之雄厚、工程之浩大、经费之保障,为学界瞩目。

2008年秋,本项目以南京师范大学语言科技研究所为实施单位正式启动。主要有三大任务:建设一个领先性的语言信息科技实验室、建立一个独创性的语言科技工作平台、撰著一套有特色的语言科技文库。

从实验室方案设计到设备招标采购,再到实验室用房改造,经过8个月的努力,2009年12月,语言信息科技实验室建成,为语言研究从传统范型向科技范型的转变提供了基本保障。该实验室划分为实验工作区、科研工作区和管理服务区。实验工作区建有语音实验与计算室、神经认知实验与计算室、课堂话语实录室三个专门实验室。科研工作区建有语义科技工作室、语音科技工作室、方言实验工作室、知识工程工作室Ⅰ(先秦词汇)、知识工程工作室Ⅱ(中古词汇)、知识工程工作室Ⅲ(敦煌俗语言文字)、语言习得神经机制工作室、语言习得中介机制工作室,以及参研工作室。管理区服务包括办公室、管理室、编辑室和交流室。出席“语言科技高层论坛暨语言信息科技实验室落成仪式”(2009年12月14日)的专家认为,该实验室体现了语言学跨学科研究的当代性和先进性,具有整体性、科技型、开放型三个特点,处于全国领先地位,是“语言科技新思维”的又一体现。同时认为,该实验室的科研工作涵盖了四个二级学科、四个博士学位点,有稳定明确的研究方向,有合理的设计规划和很好的科研基础;整体设计合理,功能齐备。以教育部重点实验室建设标准衡量,很多方面超过了指标。

语言科技工作平台是基于工作原理(课题定位—理论方法—技术路线—关键技术—评估方式)而建设的高级平台。一方面,从语言信息、语言知识和语言机制三个层面,围绕典型课题进行设备配置、资源建设和软件开发;一方面,将典型课题研究与工作平台建设融为一体,依据典型课题建设的子平台应具有解决同类课题的功能。

建设语言科技工作平台的目标是要实现语言研究手段的技术化和模型化,总体设计包括三个二级平台和八个子系统。

一、语言信息工作平台 1.语义科技工作系统(李葆嘉教授主持):基于词汇语义-句法语义的一体化研究思路,开发“人-机交互语义标注工具”,研制“深度语义标注信息库”;研制“幼儿(2~6)日常话语跟踪语料库”,完成幼儿语义系统和话语行为分析研究。2.语音科技工作系统(顾文涛教授主持):研制“多语言、多语境、多语用的语音语料库”,基于声学信号分析、感知实验和数学建模,完善语音韵律理论与相关技术应用。3.方言实验工作系统(刘俐李教授主持):完成“网络版汉语方言有声语料库”,拟定系统的可操作性语音、词汇、语法实验模型和研究方法,进一步完善新兴交叉学科“实验方言学”。

二、语言知识工作平台 1.先秦词汇统计与知识检索系统(陈小荷教授主持):研制“先秦文献语料库”、“专名知识库”、“汉语词汇档案库”等,开发先秦文献自动分词算法、古籍版本异文自动发现算法、同指专名检索软件工具等,完成“先秦汉语词汇统计与知识检索”。2.中古词汇统计与知识检索系统(董志翘教授主持):研制“中古文献语料库”、“专名知识库”、“中古汉语词汇档案库”等,开发中古文献自动分词和标注工具等,完成“中古汉语词汇统计与知识检索”。3.敦煌俗语言文字统计与检索系统(黄征教授主持):研制“敦煌文献资料库”、“敦煌文献俗词语档案库”,开发相应工具,完成“敦煌文献资料与知识检索”。

三、语言机制工作平台 1.二语习得的神经机制研究系统(倪传斌教授主持):研制“英语受蚀词汇库”等,基于行为学、脑成像和脑电三维度模型,进行中国人英语习得与磨蚀的神经机制研究,完成“基于神经机制的英语个性化学习分析系统”。2.二语习得的中介机制研究系统(肖奚强教授主持):研制“留学生汉语口语中介语语料库”,基于中介语理论、对比分析理论、偏误分析理论以及二语习得影响因素等,完成“留学生汉语习得的中介机制研究”。

这一工作平台,既是科技研究平台,也是人才培养平台,即一个现代化的科学研究和人才培养工作体系。

作为本项目的文本成果,《语言科技文库》包括计算语言学研究、语义语法学研究、汉语方言学研究、古代汉语学研究、语言教学与研究、语言新专题研究六个系列。其总体特征为:领域的开拓性、理论的原创性、选题的新颖性、方法的交叉性、考据的精审性、成果的应用性。在研究过程中,除了数据采集分析、资源建设和软件开发,更重要的还是要有新思路、新理论和新材料。陈小荷提出的先秦文献信息处理新方法,从先秦典籍注疏文献中挖掘出用于自动分词和词义消歧的知识,再注入已开发的古汉语分词和词性标注工具中去,所取得的先秦古籍版本异文自动发现、先秦词汇知识自动挖掘等成果均具开拓性。李葆嘉提出的语义语法学理论和话语行为理论,基于研制专用语料库或语义信息库和技术手段,开拓了语义网络建构、深度语义分析和话语行为研究等新的领域。刘俐李建构的实验方言学理论和方法,为方言学向现代科技方法的转型研究提供了新路,并取得了一系列新成果。黄征多年来从事敦煌文献及其俗词语文字研究,古代汉语学研究系列中的敦煌文献校录整理,以及敦煌写本字词考释、以古佚和疑伪经为中心的敦煌佛典词语和俗字研究、两汉声母系统研究等新见迭出。肖奚强基于汉语中介语语料库的二语习得研究,在对外汉语教学研究界已经产生了影响。钱玉莲的汉语介词与相应英语形式比较研究等专著各有亮色。倪传斌依据语言测试和认知实验等数据,从行为学、生理学和语言学三个层面分析影响中国英语学习者外语磨蚀的相关因素。刘宇红基于隐喻的理论探讨,对各类隐喻形式的结构、特性和解读规律进行了多视角的深入探讨。《语言科技文库》所收论著,由作者在2008年12月申报选题,2011年始逐步完稿。系列主编审读了书稿,主要就其学术价值、章节安排、内容关联、行文表述、图表绘制等方面,提出审阅意见。此后,作者们对书稿又进行了修改和润色。《语言科技文库》的作者,大多数是具有博士学位的年轻教师。对于我们这些20世纪80年代走进语言学研究领域的而言,出版论著可能已不足为道。然而,对于年轻学者而言,其论著的出版既是几年来研究的结晶,也是对其继续探索的促进。换而言之,“211工程”重点学科建设的目的之一,就是为年轻教师搭建一个可持续发展的科研和教学平台。学科带头人的主要任务之一就是提携后进。

尽管从根本上来说,科学或学术研究是一种个人的探索行为,然而复杂问题的研究,无疑需要群体协作。“学科建设”或团队合作模式,是20世纪90年代后期出现的一个新概念。这种模式涉及总体规划、多方协调,是需要付出精力和心血的。2008年,通过投票方式推举我担任该项目总负责时,就意识到自己成了一个“劳动班委”。2009年,前往安徽大学拜访黄德宽教授时,曾谈到“学科负责人的任务就是规划设计,争取项目经费和提供科研设备设施”,得到黄教授的赞许。2010年,申报江苏省高校哲学社会科学重点研究基地时,评审专家柳士镇教授提问的“作为一个交叉学科项目,各学科之间的协调是怎么考虑的,有什么做法”,可谓一语中的。作为后学,深知交叉研究之艰、学科整合之难。相关学科之间的整合协调需要借助行政机制,但凭借行政方式并非就能完成。当时的回答是,目前做到的是建成了一个可以合作研究的场所,至于学科之间的进一步沟通合作应有较长过程。有一点很明确,只有通过交叉项目,相应学科才能渗透,合作者才能逐步磨合。我们只是在一步步探索。

十一五期间的“211工程”建设项目即将完成,但是学科建设的任务并没有结束。2010年,“语言信息科技研究中心”被评审为江苏省高等学校哲学社会科学重点研究基地,为“语言科技”这一交叉领域注入了新的建设活力。重点研究基地建设,除了“跨院系整合、多学科交叉、开放型运行”理念,需要凸显“合作性攻关”。围绕交叉性项目,实施计算语言学、语音科技、神经语言学、语义科技等力量的联合攻关计划。只有通过全面开放以及和与国内外同行的合作交流,才有望建成具有影响的语言科技研究、人才培养和学术交流基地。

十年前,我(2001)曾写道:“语言科技”的内涵是以理论研究为指导,以描写研究为基础,以应用研究为枢纽,促使语言研究向计算机应用、认知科学和现代教育技术领域等延伸,沟通文理工相关学科以实现语言研究过程及其成果的技术化。“语言科技”的外延为语言工程科技、语言教育科技和语言研究科技。其中,“语言研究科技”是将语言研究活动与资源建设、软件开发相结合,其目标是实现语言学自身的科技化。还应包含语言实验、数据处理这些实验语音学、神经语言学研究的科技手段。

虽然语言学家不可能也不必要都转向语言计算或实验研究,尽管描写、考据和内省始终是最基本的方法,但是具有一定的语言科技意识却非常必要。语言学家只有了解有哪些可供利用的资源、软件或仪器,才能提高其研究深度、精度和效率。语言学家也只有了解到信息处理的语言研究需求,才有可能为之提供可资应用或参考的基础成果。“语言科技”是21世纪语言学研究的潮流。

此为出版缘起。是为总序。2011年8月谨识于南都

自序

记得六七年前,我开始准备这篇论文时,导师对我说过一句话:“你已在森林深处了。”许久,我弄不清这句话有何深意。在写作的过程中,我苦苦思索,心中一直忐忑不安……总觉得一篇论文的完成,就意味着闯出一条路子来了。

回首那几年的写作过程,果真犹如置身于森林深处。我曾经怀疑过,森林深处是否真有奥秘?追问森林探索的意义何在?又时而沉醉于森林深处的风景,时而不知是进是退,不知置身何处,路向何方?

现在论文要出版了,我觉得自己仍然深陷莽林。前面不是一条路,隐约几个岔路口。每前进一步,都将是森林深处的探索。

理论语言学离不开语料,认知语言学和功能语言学尤其关注实际使用中的语言。刚刚进入这个领域的时候,导师便建议我开始关注语料,最好对某一个小句类型展开系统研究。“那我就做双宾句吧。”我几乎不假思索地做了选择。当时只觉得好奇,如此常见的基本句式,为什么一直备受语言学界的关注?从二十世纪早期的Jespersen和Curme等语法学家到目前的认知句法,绝大多数语法理论都对其进行阐释,而阐述角度和得出的结论常常大相径庭。

两年中收集了几百页的英语双宾句语料,查遍了FrameNet中的所有动词框架,整理出其中与双及物动词相关的项目;同时,阅读许多英语双宾句的研究文献。然而,越整理,越迷惑,常常不知道是否该将一些语料归入其中,更不知道该如何排列,因为在编辑归类语料的过程中,必须依据一定的理论基础。究竟什么是双宾句呢?在比对文献时发现,原来就双宾句的界定问题,学者们仍是见仁见智。我这才意识到,自认为研究双宾句这么久了,原来根本就没有想过什么是双宾句?对其定义的视角又是什么?这时想起了导师的那句话,突然感到自己早已迷失在森林深处了……

在理论学习过程中亦是如此。从对理论语言学一窍不通,到开始对各派观点萌发兴趣。尤其是阅读到以双宾句为语料展开的不同研究,以及得出的不同结论时,甚至感到兴奋不已。感叹各家精妙深邃的理论探讨,渴望将它们都融入自己的研究之中。在激动地讲述英语双宾句的研究文献时,导师问我,是否清楚自己要做什么?研究双宾句的目的是什么?到底什么是双宾句?我猛然一惊,不知该如何作答。是啊,自己已经忘记了最初的研究目的,或者从来就没有真正清楚过。这才开始思考最基本的问题,原来熟知的“双宾句”和我要研究的“双及物小句”,因研究视角的差异,其在定义原则上竟如此不同!

森林探索仍在继续。有时因为新的发现而兴奋,有时因迷失方向而挫败。这或许就是理论研究的必经之路——从清晰到迷惑,再到豁然开朗,周而复始……我希望能跟上先行者,渴望有更多同行者一起在森林中探索。田朝霞剑桥大学2011年8月10日

程序

南京师范大学准备出版田朝霞的博士论文,我当然高兴,其原因有二,一是她是我的学生,二是论文属于理论语言学范畴。

不同的人对理论语言学的看法是不同的。持比较极端观点的人认为,理论语言学是一种“虚”学科,对社会经济发展意义不大,没有必要发展。持另一种极端观点的人则认为,理论语言学是引领学科。这两种说法都为偏颇之说,至今尚未有令人折服的理由和证据。虽然理论语言学的研究取得了瞩目的进展,但就其研究成果和现状而言,它的理论方法还无法引领其他学科建立自身学科的理论方法。但这并不说明第一种观点是正确的。我们有充分理由证明,理论语言学的研究对社会经济发展的意义目前是被大大地低估了。

虽然有些理论研究者认为,理论语言学研究不是为了具体应用,但语言学是横跨自然科学各学科以及人文科学各学科的桥梁,这是不争的事实。就自然科学而言,语言学和认知科学群的关系非常密切,其中包括计算语言学、神经语言学等。目前计算语言学的研究多半是建立在数学统计和匹配的基础上进行的,无论是识别还是理解,该领域的研究好的已经能够达到95%的准确率。为了提高准确率,该领域的研究者越来越注意到语言学理论模式的重要性。但目前语言学所参与的还只局限于对一些局部问题的解决上,其准确率仍然有3%~5%的缺陷难以补救。为此,有人开始用语言学理论模式来构建计算模式。大脑神经网络的研究是认知科学的又一个热门话题。有些人甚至认为,脑扫描技术将为语言学研究提供新的理论模式。当然,这种可能性是微乎其微的。原因有两个:一个是脑扫描技术和设备自身的局限性;另一个是作为一种论证手段,它必须以语言系统的理论模式为依据和研究对象,不合理的理论模式甚至会误导自身的研究及其研究成果的讨论和理论陈述的形成。很显然,这两个领域的研究非常需要理论语言学提供合理优质的理论模式。

语言是表达概念语义、表达思维结果的形式。目前思维乃至思想的合理研究和精致表述,较好的切入点仍是语言,因为思维进入语言表达后就进入了意识状态。思维以及文化的各种理论陈述、观点及其思辨都以语言表达为基础。因此,思维和文化的研究很大程度上基于语言义形关系的研究。对社会学、人类学、法学、管理学等理论观点的论证,更需要基于语篇展开。语言和思维、文化之间义形关系的研究,以及两者之间的理论表征,将为各学科理论论证提供有力的思辨以及基础。很显然,这些文科领域的研究同样需要理论语言学提供合理优质的理论模式。

可见,理论语言学是一种需要委以重任的基础理论学科。

由于各种原因,我国的理论语言学研究要比其他许多国家落后。更让人感到滑稽和无奈的是,许多大量被引的国内语言学研究成果中的许多观点,是二传过来的“舶来货”。有识之士间已经萌发出对“中国制造”的厌倦和对“中国创造”的向往。

双及物构式的研究,国内外已有很多成果了。田朝霞博士的论文既关注了他人的观点,又有一些自己的看法,并通过一些具体的例句进行论证。她对一些他人没有得出合理结论的部分,提出了自己的表述方案,并对其进行论证,这点是值得称赞的。至于这些表述中有多少是可以接受的,或可以推翻的,那就留给读者去做进一步研究和论证吧!程琪龙剑桥景苑2011年8月20日

Introduction

The faculty for language is often considered one of the most fascinating yet perplexing human abilities. Part of its true complexity turns out to be the form-meaning interaction in the actual use of language. Involved in the project to work out such puzzle are Cognitive Linguistics (CL) (Langacker 1987, 1991a, b; Taylor 2002; Croft 2001; Croft & Cruse 2004; Evans et al. 2006; Fillmore 1982; Goldberg 1995, Talmy 2000; etc.) and different approaches of functional linguistics (Halliday 1994; Dik 1989; Pinker 1989; Levin 1993; etc.). Neurocognitive linguistics (NCL) (Lamb 1999, 2004) in particular explicitly states that its major emphasis is on the operation of the linguistic system—production and comprehension.

The present exploration is concerned with the form-meaning interaction of the English ditransitive clause. It incorporates the spirit of neurocognitive linguistics (NCL), Cognitive Linguistics (CL) and functional approaches into the research, and adopts conceptual frame (CF) (CHENG 1995a, b, 2006a) in particular as its formal representation.

0.1 Aim and scope

0.1.1 The form of the ditransitive clause

The present research mainly covers the ditransitive clause in the English language. The ditransitive clause in this research relates to the ditransitive construction in earlier analyses, or namely the traditionally termed double object construction (DOC). Yet it differs from the DOC in its syntactic definition and in its coverage accordingly.

The DOC often refers to a construction in which a verb takes two NP objects as its complements, known as the "SVOO", or namely, the "Subject-Verb-Indirect Object (IO)-Direct Object (DO)" (cf. Jespersen 1940-49; Quirk et al. 1985; Van Ek & Robat 1985; Alexander et al. 1975; Stageberg & Oaks 2000; Huddleston 2002; Lockwood 2002; Crystal 1997; etc.). Indirect Object and Direct Object may have different labels as (Direct) Object (or the first object) and 0bject2 (or the second object) in recent theories such as lexical functional grammar (LFG) (cf. Bresnan 1982; Pinker 1989) and construction grammar (Goldberg 1995; 2006), but their grammatical relations remain the same. Clause (1) is considered a typical DOC in terms of its grammatical relations (1a).

The ditransitive clause in this research is delimited in terms of grammatical elements (or parts of speech) and their linear positions. Its syntactic structure is defined as NP1-VP-NP2-NP3 (1b), with NP and VP as Nominal Phrase and Verbal Phrase respectively. It should be noted that the NPs in the definition are not given grammatical relations such as Subject, Indirect Object and Direct Object.

A direct consequence of such a definition is manifested in that the present research has a wider coverage than the DOC, involving clauses in which the main verb has two NP complements. For example, the clauses in (2) are treated equally as ditransitive [1]clauses. What is more important, this research finds that the meanings of these sentences and the frequent type like (1) are conceptually related; they are all associated with a meaning of transfer of possession. This is what the present research is mainly concerned with.

(2)a. John asked Mary her age.b. Light me the gas before you go. (Kruisinga 1932; 190)c. I envy you your beauty. (Jespersen 1927; 295)d. The governor appointed him judge.e. Susan made Todd a good husband.

The ditransitive construction is often associated with the prepositional phrase with to or for as in (3) (Palmer & Blandford 1969; Quirk et al. 1985; Dekeyser et al. 1987; Huddleston 1988; etc.).

(3)a. John gave Mary a cake. / John gave a cake to Mary.b. John baked Mary a cake. / John baked a cake for Mary.[2]

The prepositional phrase is not the focus of this research; however, it is relevant in the sense that it is involved in the comprehension and production of the ditransitive clause. It is argued that the to-phrase shares a similar transfer situation with its counterpart ditransitive construction (as in (3a)), while the for-phrase (as in (3b)) is not directly related to the transfer situation designated by the clause (detailed in chapter 4). The at-phrase (as in (4)) is also touched upon for its interaction with the research object.

(4) She cast him a glance. / She cast a glance at him.

Finally, the passive voice of the ditransitive clause is beyond the scope of this research. Its complexity requires another research for a full explanation.

0.1.2 The meaning of the ditransitive clause

The meaning of clauses is defined in the spirit of neurocognitive linguistics (NCL) and Cognitive Linguistics (CL), and formally represented within the framework of conceptual frame (CF) (CHENG 2006a) (detailed in chapter 2). It is necessary to clarify two fundamental claims concerning the nature of meaning.i. Meaning is the conceptualization of the outside world.ii. Conceptual meanings of clauses activated in the linguistic

operation constitute a multidimensional network of conceptual

structure separate from yet connected to its syntactic structure.

The first principle relates to cognitive approaches to semantics, which are reflected in the writings of major proponents including Lakoff (1980, 1987), Langacker (1987, 1991b), Jackendoff (1990a, 2002), Fillmore (1982) Turner & Fauconnier (1998), Talmy (2000), Goldberg (1995), etc. The prime slogan for cognitive semantics is that "meanings are in the head" (Allwood & Gardenfors 1999: 21). Meaning is conceptualization (Langacker 1987, 1999b; Taylor 2002; Croft & Cruse 2004); it is the way that the mental system construes the outside world. Cognitive semantics is contrasted with the realistic or truth-conditional semantics in the sense that the truth condition of language is no longer a prime determinant of meaning. Instead, meaning is activated as a consequence of the mental process, and "words and other linguistic units are only 'prompts' for the construction of meaning" (Evans et al. 2006; cf. also Langacker 1987).

In the context of cognitive semantics, meaning is encyclopedic; encyclopedic knowledge is conceptual structure. The notion has two entailments: first, meaning derives from experience and conceptual structure in turn reflects human experience; second, in tandem with the first, encyclopedic knowledge should be dynamic both spatially and temporally. Spatially, the world is construed differently from different vantage points; temporally, situations are construed as dynamic processes. Cognitive approaches to grammar attempt to work out the conceptual structure of language.

Construction grammar (Goldberg 1995), following the tradition of frame semantics (Fillmore 1982) in which meanings are typically defined relative to some particular background frames or scenes, contends that "[S]imple clause constructions are associated directly with semantic structures which reflect scenes basic to human experience" (Goldberg 1995: 5). The ditransitive construction is associated with a dynamic scene of ex-perientially grounded gestalt; X causes Y to receive Z.

This research shares with Cognitive Linguistics most important assumptions concerning meaning. The point where conceptual frame (CF) diverges from construction grammar is the second principle that meaning is a multidimensional network related to syntactic structure. The notion entails two corollaries; conceptual structure is not tied to syntactic structure; conceptual structure is a multidimensional network.

In construction grammar (Goldberg 1995), conceptual structure (argument structure) is tied to syntactic structure in the sense that one argument corresponds to one grammatical relation. Argument structure is by nature the grammatical semantic structure. For example, Agent is paired with Subject, Recipient with Object, and Patient with 0bject2. This treatment is similar to case grammar (Gruber 1965; Fillmore 1968; Chafe 1970) and argument structure theories (Bresnan 1982; Pinker 1989; Levin 1993, 2004), in which the components of the two structures are connected by linking rules. The spirit they share is a one-to-one correspondence between arguments and grammatical relations.

It is argued that argument structure has a closer relationship to syntactic meanings than to conceptual meanings. Conceptual meanings are not tied to linguistic units. Conceptual structure and syntactic structure in this research belong to different linguistic subsystems—the conceptual subsystem and the syntactic subsystem, which interact with each other through realizations. To interpret it, conceptual components and grammatical elements are not connected to each other in a one-to-one correspondence. For example, the first NP (John) in John gave Mary a cake is realized as more than one participant (like Agent in argument structure) in its conceptual structures. It links to two participants; Causer and Actor (see details in Chapter 2). The comparison is illustrated as follows:

Subject (John)→Agent  (argument structure)

NP1 (John) = Causer + Actor  (the present research)

("→" means "links to" or "fuses with;" "=" means "is realized as;" "+" refers to a conceptual conflation between two participants)

Another difference of this research from construction grammar (and lexical functional grammar as well) is that, the conceptual structure in the former is a multidimensional network instead of a one-dimensional argument structure. For one thing, a clause often activates more than one basic conceptual structure. For example, John gave Mary a cake activates a causation structure and an action structure as well. For another, the multidimensional network of conceptual meanings activated by a clause is represented in a pre-activated frame covering more information than just basic conceptual structures.

First, the multidimensional network not only subsumes the conceptual structure directly evoked by the clause, that is, the conceptual process, but also involves situations at the initial and terminal point of the process. For example, John gave Mary a cake activates not only an action structure of giving and a causation structure designating the transfer of possession; it also activates what happened before the transfer and after the transfer. They constitute part of the language users' encyclopedic knowledge associated with their experience with the outside world.

Second, a multidimensional network captures the conceptual discrepancies of the variations of the transferred object (as a cake in John gave Mary a cake) that is associated with the ditransitive clause. This is a main feature that distinguishes this research from others. For example, the objects in each clause in (5) are conceptually different, identified as Thing (a cake), Identity (chairperson), Information (a story), Action (kiss) and Event (crushing a mountain). Such a difference leads to a significant discrepancy in the comprehension of the whole clause (see Chapter 3).

(5)a. John gave Mary a cake. (Thing)b. John appointed Mary chairperson. (Identity)c. John told Mary a story. (Information)d. John gave Mary a kiss. (Action)e. Crush me a mountain. (Event)

Third, the multidimensional network depicts the form-meaning interface, that is, the realization between conceptual structure and syntactic structure. This is vital because in this research the main purpose for exploring meaning is to reveal the interaction between conceptual structure and syntactic structure.

Pragmatic considerations

"Meanings are in the head" (Allwood & Gardenfors 1999: 21; also cf. Jackendoff 1990a). From this perspective, sentence meaning is speaker meaning. It is in this sense that the distinction between semantics and pragmatics is blurred in Cognitive Linguistics. It is claimed that pragmatics studies "the contribution of the context (that is linguistic and situational context) to the meaning" (Jaszczolt 2002: 1). Certainly this is important in the actual comprehension and production of language; however, the present research does not go that far. This research is mainly concerned with the meaning evoked by syntactic structure rather than the situational context. For example,

(6) John threw/brought Mary a cake.

The sentence directly evokes a causation structure entailing an orientation that Object (cake) is transferred to Possessive Goal (Mary). It does not necessarily mean that Mary got the cake in the truth condition. To know whether Mary got the cake, the language user often has to resort to pragmatic information—the situational context.

Besides, phonological factors, in particular where the nucleus falls in an intonation group—on Mary or on a cake (as in both John gave Mary a cake and John gave a cake to Mary), facilitate the actual comprehension of a sentence. Both pragmatic and phonological considerations play an important role in actual linguistic operation. However, as their complexity makes them chief topics for full [3]explorations, they are beyond the scope of this research.

0.1.3 The aim and the research perspective

The aim

试读结束[说明:试读内容隐藏了图片]

下载完整电子书


相关推荐

最新文章


© 2020 txtepub下载