作者：李晓红
出版社：上海交通大学出版社
格式: AZW3, DOCX, EPUB, MOBI, PDF, TXT
双语语料库界面下英汉语义韵对比研究试读：
前言
语义韵概念的提出无疑是语料库语言学领域跨世纪的重要发现与主要研究课题。后弗斯学派的语料库语言学者在研究词语搭配时发现了超越单个词语界限的态度意义，这一发现为词语搭配研究和短语学研究提供了新的理论建构和工作范式。在近三十年的发展中，语义韵概念的探索经历了由单语视角向双语视角的转向，为对比短语学研究提供了新的方法与路径。
在语料库语言学领域和短语学研究中，约翰·辛克莱（John Sinclair）提出的“成语原则”（idiom principle）和“短语趋势”（phraseological tendency）是探索意义单位过程中最具变革意义和深远影响的发现，也是语料库语言学的理论基石。语义韵是辛克莱继而提出的扩展意义单位模型中的重要概念，是由节点词、搭配、类联接、语义趋向等要素共同选择而最终指向的态度意义。根据扩展意义单位模型，跨语言视角的语义韵研究主要通过调查双语词语的意义和功能对应来确立双语对等。然而，迄今为止，跨语言视角的语义韵研究仍以英语和其他西语的对比为主，基于双语语料库证据，严谨、系统地探讨英汉词语的语义趋向和语义韵如何构建双语扩展词汇单位的研究是一项既必要又艰巨的工作。
本书秉承辛克莱关于扩展意义单位的理论框架，在英汉双语视角下探讨并揭示两种语言表达语义韵——隐含态度意义——的特征异同。在研究方法上，采用自下而上的语料库语言学研究范式，借助平行语料库和对比语料库中客观、丰富的语料证据，发掘和描述新的语言现象，概括新的语言型式。英汉、汉英双向平行语料库能够提供大量的双语对应证据，有助于拟定词语层级的对应。对比语料库——现代汉语语料库和英国国家语料库——用于发现以节点词为核心的短语层级意义单位，即包含搭配、类联接、语义趋向和语义韵各组合要素的扩展词汇单位。借助对比语料库中客观、丰富的语料证据能够发现凭直觉和经验无法发觉的语言聚合规律，揭示双语扩展词汇单位在形式、意义和功能层面的对应特征，最终确立最高程度的双语对等。本书的研究过程以数据为基础，采用了相互对应率（MC）、林奈双名法（Linnaean-style binomial notation）等测量和描述手段对语言证据进行数据处理和描述。在探讨语义韵过程中，本书提出了语义韵常模、语义韵力度、主导语义趋向等描述参数，系统、详尽地讨论了双语界面下语义韵及语义趋向的相互作用和共选规律。
研究发现，英汉语义韵表达既有共同特征，又呈现特有趋势。共同特征包括两方面：第一，节点词的内涵义无法直接表明扩展词汇单位的语义韵。单个词项蕴涵着语义韵潜势，即与特定的词汇-语法环境共选并表达积极、消极或中性态度意义的可能性，词语与语境相互共选时，语义韵潜势将被触发和具体化。语义韵同词语内涵义的区别在于语义韵是集复现搭配、特定句法结构及语义一致性的统一体。第二，节点词的主导语义趋向通常与语义韵保持和谐一致；有些语义范畴的节点词其主导语义趋向则可能与语义韵冲突对立。在解读语义韵过程中，可选语义趋向具有三种作用：态度指示语、态度强化语和语义携带语。距节点词越远，可选语义趋向的频数越低，相应的搭配和类联接规律越不明显，因此非显著性可选语义趋向的位置预示着扩展词汇单位的语义边界。
英汉语义韵表达体现三方面的异同特征。第一，语义韵力度差异。初似对应词呈冲突对立的语义韵及较高的语义韵力度，这说明以二者为核心的扩展词汇单位分别表达的典型态度意义具有语言特殊性或文化特殊性。第二，语义韵空缺。违背语义韵常模的个例使用表明，在目标语中出现的语义韵差异及语义韵非典型性现象皆由文化差异、意识形态差异和价值观差异所致。第三，类联接高度不一致。两种语言在表达语义韵时体现高度不一致的类联接选择趋向。
研究的另一重要发现是，要确立适宜的双语型式对等，语义韵的一致对应或高度相似是关键所在。通过考察扩展词汇单位在语义趋向和语义韵层面的特征异同，能够建立不同程度的双语对等。双语扩展词汇单位的语义和功能高度相似，二者可确立为高度型式对等；双语扩展词汇单位中的一方允许语义韵偏离，二者为部分型式对等；语义韵冲突对立的扩展词汇单位则为不对等型式。因此，语义韵对等是双语词汇型式最高程度的对等。比较和探讨扩展词汇单位的语义韵对等为语言对比研究、翻译研究、语言教学及双语词典编撰开辟了新路径，为调查和确立双语对等提供崭新的研究视角。
语义韵概念在国外学界正处于多视角的理论阐释和更科学的方法探索阶段，而在国内学界仍是比较新的学术概念，近年间国内学者也开始越来越多地关注这一语言现象。语义韵的概念和研究方法尚在逐步完善，本书的描述和探讨难免有不尽准确之处，恳请广大专家同仁不吝指正。
拙著付梓之际，心存无限感激。
由衷感谢北京航空航天大学卫乃兴教授。他的悉心培养和专业指导是我顺利完成上海交通大学语言学博士研究的坚实基础。在博士研究和本书撰写过程中，他及时给予富有真知灼见的指导，耐心审阅本书的几次草稿，锱铢必较地指正文中谬误。卫乃兴教授严谨治学、孜孜不怠的学术风范是我完成学业和这本拙著的源源动力。
本书还得到上海交通大学王同顺教授、上海交通大学胡全生教授、上海复旦大学曲卫国教授、华东理工大学邵志洪教授、华东师范大学邹为诚教授的指点与帮助，深表谢意。
上海交通大学语言文字研究所的学友们多年来与我一起探讨，给予我不尽的帮助与鼓励；上海电力学院外国语学院的领导和同事给予我大力支持与帮助，由衷谢意难能细表。
寒窗苦读离不开家人默默地关爱和支持。我为长期以来不能在父母身边尽孝充满愧疚，更为父母无怨无悔地帮我照看幼子无限感激。爱人涂文强给予我莫大的支持与鼓励，陪伴我克服障碍，完成拙著。若无家人奉献相助，何谈今日之所得！
上海交通大学出版社的编辑为本书的出版花费大量时间和精力，给予了宝贵支持。谨向他们表示诚挚的感谢。李晓红2015年7月Chapter 1Introduction1.1Background of the study
In the past two decades the concept and phenomenon of semantic prosody have assumed a place of central importance in corpus linguistics; particularly, it has been brought into close integration with the study of attitudinal meaning and pragmatics. The concept of semantic prosody arises from post-Firthian corpus linguistics and is widely elaborated in phraseological studies. In effect, semantic prosody has increasingly been given considerable attention in corpus linguistics and beyond. Studies of semantic prosody have widened the horizons in the foreland field of linguistic exploration and enriched the methodology and contents of linguistic description (卫乃兴 2011a).
Discussions, questions and debates about the nature of semantic prosody, its identification and linguistic/non-linguistic implications lie at the heart of previous studies. In the literature, the term “semantic prosody” embraces different interpretations and is assigned with a broad range of features; however, the most common and acceptable interpretation can be paraphrased as: semantic prosody is semantically a property of meaning prevailing in the context as a result of lexico-grammatical co-selection, and pragmatically the hidden attitude and communicative purpose of a speaker/writer (Louw 1993, Sinclair 1996, 卫乃兴 2011a).
In Sinclair’s (1996) Model of Extended Units of Meaning (henceforth MEUM), one of the most pioneering works for subsequent theoretical and empirical research, semantic prosody is considered as an obligatory property of an extended lexical unit. Semantic preference, or semantic association (Hoey 2005), reveals the relation between a word and a set of semantic categories, closely connected with the concept of collocation. The proposal of semantic preference leads us to a new understanding of collocation, because what we begin to note is not only “word-to-word collocation” but “word-to-class” or “word-to-semantic-class-of-word collocation” (Bublitz 1998：16). Studies of semantic prosody and semantic preference have opened a brand new perspective to explore how the Idiom Principle or Phraseological Tendency (Sinclair 1996) operates on the syntagmatic axis and how the co-selection between lexis and grammar is oriented by communicative functions in languages, genres and language varieties (卫乃兴 2011a).
In the literature, semantic prosody and semantic preference have been explored most extensively by Sinclair (1996), Stubbs (2001), Partington (2004) and Bednarek (2008), among many others. The two notions have been widely and fully explored in a monolingual context, mainly in English. There are, however, only a few researchers (Partington 1998, Berber-Sardinha 2000, Tognini-Bonelli 2002, Dam-Jensen and Zethsen 2007) showing keen interest in contrasting semantic prosody between English and other European languages. Still less work has been carried out in a systematic way to study semantic preference and semantic prosody within a contrastive context involving English and a non-European language (notable exceptions include Xiao and McEnery 2006, 卫乃兴 2011b, 李晓红 and 卫乃兴 2012).
Recently, cross-language studies of semantic prosody have been undergoing a shift of focus from identifying equivalents of individual words to establishing equivalence of lexical patternings concerning similarities in form, meaning and function. In pursuit of equivalence at word level, Berber-Sardinha (2000) and Xiao and McEnery (2006) both conclude that collocational behaviour and semantic prosodies of presumed equivalents are unpredictable across two languages, being quite similar in some cases and quite different in others. In Partington’s (1998) comparison between English and Italian, the different lexico-grammatical environments displayed by “look-alikes” or “false friends” between the two languages lead to a conclusion that perfect equivalents are rather few. Based on the MEUM, Tognini-Bonelli’s (2002) corpus-driven study stresses that comparable units of meaning in English and Italian are identified as “functionally complete units of meaning” only when they correspond at the level of semantic prosody. Three different empirical methods suggested by 卫乃兴 (2011b) are revealing in the establishment of corresponding phraseological units across languages. The study emphasizes that lexical and phraseological equivalents between English and Chinese can be recognized when correspondences are fulfilled in terms of form, meaning and function.
The contrastive studies of semantic prosody are of both theoretical and methodological importance. However, once we go further into the multi-faceted nature of semantic prosody and the contrastive approaches adopted so far for identifying equivalence, some issues inevitably arise, particularly from the contrastive perspective. Firstly, the fuzzy distinction between semantic prosody and connotation, as well as the one between semantic prosody and semantic preference, always confronts the concept with various criticisms and even a denial of its existence (Whitsitt 2005). Secondly, with regard to the description of semantic prosody and semantic preference, there exists a dispute about whether the binary distinction between “positive” and “negative” is over-simplification (Hunston 2007, Bednarek 2008) or fundamental (Morley and Partington 2009). Thirdly, within the contrastive framework the importance of semantic prosody in identifying equivalents across languages has begun to receive increasing attention; however, the role of semantic preference in this respect has not been given due weight. Since semantic preference “contributes powerfully to building ［...］ prosody” (Partington 2004：151), it is worthwhile to further explore how equivalence can be set up in terms of semantic preference and semantic prosody.1.2Objectives and significance of the study
Introducing semantic prosody into cross-language studies is a practice much younger than the discovery of the phenomenon. In view of the strong and weak points in previous studies of semantic prosody, the research objective of the present study is twofold. First, to explore and unveil the common and language-specific properties of semantic prosody in the cross-linguistic context. We start with groups of presumed translation pairs across Chinese and English as shown in a bi-directional parallel corpus and further explore the semantic and pragmatic profiles of each pair in two comparable corpora. Comparisons of similarities will include the relationship between semantic prosody and connotation, the role of semantic preference in contributing to a prosody and the process of identifying and producing a prosody. Differences will be discussed in terms of prosodic strength, prosodic clash and structural patterning. Second, to explore the factors that contribute to producing and identifying semantic prosody by setting up extended lexical units across the two languages. To this end, MEUM is to be enriched from the contrastive perspective. In uncovering similar and dissimilar features of semantic prosody, extended lexical units in the two languages will be recognised and then contrasted in terms of form, meaning and function. More weight will be given to the extent to which each pair of equivalents corresponds at different levels. It is presumed that equivalence at the level of larger units rather than single words may possibly be figured out, which will shed light on translation studies, language teaching and bilingual lexicography.
Corpus data offers the opportunity to corroborate or improve theoretical models, but detailed corpus-based studies of semantic prosody between English and non-European languages are still few and far between. In the academia at home, preliminary contrastive studies have been conducted to explore semantic prosody mainly from the aspect of collocation, to describe it by observing the triple-division principle, or to identify word-level equivalents evidenced in examples that have been extensively discussed by former analysts. Another common practice of previous studies is that the cross-linguistic equivalents are typically posited on the basis of the analysts’ language experience or with reference to information in some well-known bilingual dictionaries, without consulting parallel corpora for recurrent translation equivalents (Tognini-Bonelli 2002).
Confronted with these issues head on, this study is intended as a contribution to the systematic description of semantic prosody across English and Chinese, with both theoretical and methodological significance.
By focusing upon semantic and pragmatic profiles of extended lexical units, the study is to closely examine common features of semantic prosody shared by English and Chinese in order to reveal the multi-faceted nature of semantic prosody. Its fuzzy relations with connotation and semantic preference are to be disentangled in the process, which will justify semantic prosody as a unitary theoretical concept. Furthermore, the study comes up with a binomial notation and the ratio of prosodic strength as two useful means to explore semantic prosody. The former, being the qualitative analysis, concentrates on the specific description of semantic prosody, whereas the latter makes it possible to have the phenomenon assessed in a quantitative way. By means of this, the study witnesses the fine details of distinctions in expressing semantic prosody across the two languages. The language-specific tendencies are essentially better reflections of cultural and ideological differences. Another important theoretical significance is that the study broadens the concept of equivalence by emphasizing the importance of equivalence at the level of semantic prosody. This provides a new perspective for establishing translation equivalence in translation studies, bilingual lexicography as well as EFL teaching and learning.
So far, MEUM has proved to be a powerful mechanism in the monolingual context to show how components of meaning co-select in service of conveying an attitude or communication purpose. The present study draws upon insights from MEUM but focuses on examining whether the model can be employed as an effective one in the contrastive context. In the process, MEUM will be used to identify from two general corpora the recurrent lexical units showing regularities in semantic preference and semantic prosody. MEUM constitutes the important part of a five-step comparative procedure, where the focus is enlarged to look at similarities or dissimilarities between the extended lexical units recognized by MEUM. The aim is to find out whether the larger lexical units are used as single choices of similar semantic prosody in English and Chinese. If the proposed method works well, the five-step comparative procedure containing MEUM throughout the analysis will prove to be a useful and appropriate method for establishing equivalence at the level of phrasal units.1.3Organization of the thesis
This thesis is composed of eight chapters. Following the present introductory chapter, Chapter 2 provides a brief review of studies on semantic prosody in monolingual and cross-lingual contexts. In discussing monolingual studies, instead of providing a coherent chronological development on the subject, an attempt is made to generalize different theoretical viewpoints particularly in terms of the way semantic prosody has been defined and characterized. The purpose of reviewing cross-lingual studies is to show how semantic prosody can be taken into account for identifying translation equivalents across languages; on the other hand, it should be emphasized that contrastive studies on semantic prosody need a more systematic method of comparing prosodies across languages and a fresh understanding of equivalence.
Chapter 3 introduces the corpus-based methodology. Definitions and new parametres of semantic preference and semantic prosody are provided, and other relevant terms used are re-clarified. After a brief description of corpora and corpus tools, a five-step analytical procedure adopted throughout the subsequent chapters will be elaborated on.
Chapter 4 initially sketches out the prima facie translation pairs to be investigated in the study. This is followed by a case study so as to exemplify the process of identifying semantic prosody based on MEUM. Finally, the representation and explanation of overall data provide a general view of the relations between semantic prosody, semantic preference and connotative polarity.
Chapter 5 reports on the similar characteristics of semantic prosody between English and Chinese—in particular the relationships between semantic prosody, connotation and semantic preference. Initially, data from the two comparable corpora show that either in Chinese or in English semantic prosody has little to do with the overt or covert connotation of a node in question, but it is much anchored in an extended lexical unit. This attribute of context-dependency distinguishes the notion of semantic prosody from that of connotation. What follows is a discussion of the relationship between semantic preference and semantic prosody. Through exploring the role of predominant and optional preferences in contributing to the reading of a prosody, similarities are observed between the two languages. It is then argued that universally the way we identify a prosody is in the reversed order of producing it.
Chapter 6 continues the contrast, concentrating on the dissimilarities between the two languages. Prosodic strengths, turning the abstract notion of semantic prosody into a measurable one, manifest straightforwardly different tendencies of delivering attitudinal meanings in English and Chinese. Prosodic clashes, reflected in the recurrent use of particular phrasings in one language but rare use or ever absence in another, reveal that prosodies are culture-specific. It is also argued that in expressing a prosody English and Chinese exhibit stark differences in structural patternings.
In Chapter 7 attention is turned to the establishment of patterning equivalence between the two languages. The main part of this chapter contains discussions of numerous examples, with three kinds of patterning equivalence set up by using the criteria of semantic prosody and semantic preference. In the final part an attempt is made to explain how semantic prosody helps in distinguishing synonymous equivalents.
Chapter 8 summarizes the major findings of the study, addresses the possible implications and points out the limitations and perspectives for further research in the area.Chapter 2A Review of the Literature and the Theoretical BackgroundIn this chapter, we provide a brief review of the literature on semantic prosody, starting from groundbreaking works in the monolingual studies on the subject and moving on to the revealing studies in the contrastive linguistic context. The subsequent sections are devoted to a discussion of different viewpoints of semantic prosody.2.1Major views of semantic prosody in the monolingual context
Sinclair, although not explicitly using the term “semantic prosody”, first describes the phenomenon that “many uses of words and phrases show a tendency to occur in a certain semantic environment” (1991：112). Later analysts of semantic prosody have picked up different aspects of Sinclair’s ideas, and a number of different but related understandings of semantic prosody have sprung up. Generally, there are four major interpretations of semantic prosody: contagious viewpoint, connotational viewpoint, functional viewpoint and discoursal viewpoint.2.1.1 The contagious view
Louw writes two articles on semantic prosody. The term “semantic prosody” gains its currency in “Irony in the Text or Insincerity in the Writer?” (1993), in which Louw’s first definition of semantic prosody is known as the “contagion of meaning”. His interpretation of semantic prosody embraces three key points.
First, Louw defines semantic prosody as “a consistent aura of meaning with which a form is imbued by its collocates” (1993：157). According to him, the transfer of meaning is unidirectional from the collocates to the node form. It is necessary to point out that what is transferred between the co-occurring words is not the meaning of a single form, but similar semantic features shared by habitual collocates of the same semantic set. This “semantic consistency” is capable of colouring the node form and creating a specific semantic atmosphere over the whole context. Louw also states that a prosody is an inherent and inseparable part of a form.
Second, Louw points out that semantic prosodies remained hidden from our perception and inaccessible to human intuition and that they can be properly and systematically assessed only with the assistance of corpora and computational method. In assessing the phenomenon, Louw has always supported the dichotomy of semantic prosodies as negative and positive. “The prosodies based on very frequent forms can bifurcate into ‘good’ and ‘bad’” (1993：171). His examples include utterly, days are, BENT on and symptomatic of, all of which are shown to express unfavourable prosodies. Louw later adds that “negative semantic prosodies are much more frequent than positive ones” (2000：52).
Third, Louw is particularly interested in the relationship between semantic prosody and irony. He assumes that when a given form was consistently associated with a set of desirables or undesirables, any apparently unexpected collocate would be perceived as a deviation from the well-established prosodic trend and the instance should be interpreted as ironic. By observing symptomatic of, Louw perceives another kind of prosodic conflict which is quite involuntary, with no ironic intention at all. This departure from the normal prosodic profile serves as an indicator of a speaker/writer’s real attitude, and it exposes the insincerity and inconsistency between what s/he utters and what s/he thinks. In brief, irony can be seen as the deliberate creation of an instantiation against the prevalent prosodic trend in the context; while insincerity is the involuntary prosodic conflict which discloses the “speaker’s real attitude even where s/he is at pains to conceal it” (1993：157).
The “contagious view” is based upon a parallel with Firth’s discussions of prosody in phonological terms. Sounds transcend segmental boundaries and in the same way meanings flow to a node
试读结束[说明：试读内容隐藏了图片]