Volume 27, No 1
Article 文章
Abstract 摘要
The ancestry of the Chinese language is still a topic of vigorous debate. (See Wang, ed.,1995). In my own research, I find the immediate affiliation with Tibeto-Burman to be the most probable, i.e., that Chinese is a Sino-Tibetan language. Beyond that horizon, I accept the more distant affiliation of Chinese (and Sino-Tibetan in general) with Sino-Caucasian (or Dene-Caucasian: see Starostin, 1995). However, Dene-Caucasian properly consists of more language families than just Sino-Tibetan, Caucasian, and Yeniseian, the languages cited by Starostin. This paper brings in additional evidence from the other Dene-Caucasian languages, namely Basque, Burushaski, and Na-Dene. It will be shown that this supplemental evidence sheds additional light on the deep ancestry of the Chinese language.
汉语语源的问题一直是热烈争论的题目。（参看王士元1995）在研究中，我发现汉语和藏缅语有直接关系，及汉语是汉藏语中的一种语言。我还同意在汉藏语以外，汉-高加索语族（或者Dene-高加索语族，参看Starostin 1995）是汉语(一般地讲，包括和汉藏语系语言)更远的来源。但是，Dene-高加索语族的内容不只限于Starostin所提到的汉语语系，高加索语系和 Yeniseian 语系。本文根据 Dene-高加索语族中另外几种语言的材料，即巴斯克语，Buru-shaski 语和 Na-Dene 语的材料，为汉语具有更远古的语源关系提供了进一步的线索。
Abstract 摘要
This study investigates Chao’s tonal representation (1956, 1968) and its quantification using acoustic data of Taiwan Mandarin. By positing transforming formulae for Chao’s two versions (equi-exponential and non-equi-exponential intervals), we have shown that the tonal relationship in Taiwan Mandarin is non-equi-exponential on the five-point scale. The resulting tonal values are different from the prescriptive ones based on Beijing Mandarin–55, 35, 214, and 51. Instead, the four tones in Taiwan Mandarin are 44, 323, 312, and 42. Different from the prescribed system, Taiwan Mandarin demonstrates a narrower four-way distinction, lower tonal heights, and more conservative tonal contours. The new tonal values show invariance across levels (in isolated and non-isolated syllables) and positions (word-initial, medial, and final). It is thought that Chao has posited an equi-exponential version for tone languages in general and the modulated version is tailor-made for Mandarin. The discrepancy between the prescriptive version and the result of this study might be attributed to both diachronic changes and synchronic dialectal influences.
本研究以台湾地区国语之声学语料，藉由重新建构的公式，探讨赵氏（1956,1968）两个版本（等对数间距与非等对数间距）的声调系统及其量化的可能性。研究结果显示，台湾地区国语在五段声调尺上所呈现出来的乃是非等对数间距，同时，调值亦与以北京国语为本的制式调值──55、35、214及51──大不相同。本研究发现，台湾地区国语四声的调值为44、323、312及42。与制式调值相比，台湾地区国语声调为一四段式系统，且其调高偏低、调幅较窄。本研究所计算出的新调值所呈现的稳定性极高，跨越不同层次（单音及连音）及位置（字首、字中、字尾）。因此，我们认为，赵氏所建构的两种声调尺版本，其中等对数间距声调尺乃是为声调语言作一概括性的描述，而非等对数间距声调尺乃是为国语所特别设置的。至于造成制氏调值与本研究的结果不同的原因可能基于历史变迁，亦有可能是受断代方言的影响。
Abstract 摘要
The aim of this paper is to reevaluate the role of the Minnan element in Baba Malay (BM). Spoken by the Straits-born Chinese of Melaka and Singapore, BM has been described as a creole based on Malay with a Minnan substrate (Lim 1981, Holm 1989). However Pakir (1986) regards it as a dialect of Malay, and in Grimes (1992) BM is described as a Malay dialect with some borrowings from Hokkien. In this paper we attempt to resolve this question, showing that the Minnan element in BM represents a case of substrate influence rather than borrowing. Inspired by the framework of Thomason & Kaufman (1988), we believe that the nature of a language is a function of the history of its speakers. The historical context of BM, and the evidence given by a careful analysis of its structure, clearly argue in favor of treating the language as a creole. We then consider the structural influence of Hokkien on BM in terms of shift and substrate influence, focusing on syntactic features of BM including possessive and passive constructions and the TMA system.
本文章是研究巴巴马来话–Babamalay–的历史和结构。BM是早年从福建移民来的闽南人和马来西亚当地的女人通婚后形成的一种语言。作者提出的问题是：BM话是不是一种从马来话和闽南话发展出来的混合语言？本文试图说明闽南话对BM有很重要的影响，特别是在语法方面。因为BM表现出很清楚的混合特性，我们不认为 BM 是一种马来的方言，我们要了解的是它的闽南底层。
Abstract 摘要
This is a study of the grammar of Mandarin demonstratives based on actual spoken discourse data. Traditionally, it is assumed that spatial distinctions are primary for demonstrative uses. For example, zhe is used when the referent is close to the speaker, while na appears when the referent is distant from the speaker. I show that in actual discourse demonstratives operate in a much more complicated fashion and that the traditional view does not account well for the interactional data examined. Using such notions as indexical ground and the dynamic nature of indexical ground proposed by discourse researchers, I suggest that the non-concrete dimensions of indexical ground must be introduced to explain the grammar of demonstratives. Specifically, I propose that factors such as the shift of discourse structures, discourse properties of the focused referent, the building of text, the speaker’s assumption about the hearer, and the speaker’s attitude toward the referent, play an important role in determining the use of Mandarin demonstratives.
传统上认为，指代词(这、那以及由这， 那构成的指代结构， 如这个、那个、这里、哪里等)的使用是跟所指对象与说话人的实际距离的远近有关的。虽然这种“距离说” 能解释一般的情形， 但并不能说明事物的全部。我们考察了实际录音的言谈材料后 发现，远指与近指指代词的选择跟很多传统上没有意识到的因素有关，比如言谈结构的变化（对话还是叙事），所指的话语特征（有指还是无指），所指是否跟上文有关，说话人对听话人的知识判断，以及说话人对名词所指的态度等。这些都是决定谈话中用“这”还是“那”的有关话语因素，而这些因素显然超出了“距离说”的解释范围。尽管我们的研究还很初步，但它可以用来说明这样一个道理，研究语法不能不看实际的言谈材料。换句话说，研究话语活动并不简单地是去研究某条抽象“语法”如何被实际运用，而且也是在揭示真正的语法规律。
Abstract 摘要
This paper suggests a periodization scheme for northwest Chinese dialect history, from the immediate post-Han period to the present. For the purposes of the discussion, the northwest area is divided into two major sub-regions, i.e. the Guanzhong plain of southern Shaanxi and the Gansu corridor. The scheme is of necessity built around available textual materials from successive historical periods, emphasizing those stages which appear most clearly in the sources. It takes as its basis phonology rather than syntax or lexicon, since the latter two areas are not well represented in pre-moderm texts. The suggested periodization consists of seven stages, i.e. Stage I. Pre-Old Northwest Chinese (ca. 280 A.D.), Stage II. Old Northwest Chinese (ca. 400), Stage III. Sui-Tang Chang’an (580-650), Stage IV. Mid-Tang Chang’an and Mid-Tang Shazhou (700-800), Stage V. Late Tang Chang’an and Shazhou proper (800-1000), Stage VI. Post-Shazhou (twelfth century), and Stage VII. the Modern Period.
本文介绍西北方言历史的演变过程，并以此历史过程加以断代。 说讨论的年代， 由汉末三国开始，至现代为止。 地域包括陕西关中与甘肃走廊。 断代工作着重于资料丰富的时期。由于词汇和句法方面的材料缺乏， 乃以声韵演变为断代之标准。 所划分出来的历史阶段总之如下：一，前古代西北话阶段（公元280年左右）二，古代西北话阶段（公元400年左右）三，隋唐长安阶段（公元580-650年）四，盛唐长安与前沙州阶段（公元700-800年）五，晚唐长安与沙州阶段 (公元800-1000年) 六，后沙州阶段（公元1200年左右）七，现代方言阶段。
Abstract 摘要
The paper is a study of the change and variation in the syllable-initial and syllable-final consonants in Hong Kong Cantonese since the 19th century. Historical records, recent sociolinguistic data, the speech of young adult speakers of present-day Hong Kong Cantonese, and experimental phonetic data show that (a) the syllable-initial stops, glides, lateral liquid, non-coronal fricatives, and bilabial nasal have been stable; (b) there were two series of syllable-initial coronal affricates and two series of syllable-initial coronal fricatives in Cantonese, and both series in each case have merged to become a single series; (c) articulatorily these syllable-initial sibilants are lamino-alveolar or lamino-denti-alveolar sounds; (d) the historical syllable-initial [n-] has unconditionally merged with [l-]; (e) the historical syllable-initial [-] has split into []-initial and zero-initial, and the historcal zero-initial has split into []-initial and zero-initial as well; and (f) both alveolarization and velarization have taken place in the syllable-final stops [-t, -k] and nasals [-n, -], and the prevailing direction of shift is from back to front, that is, alveolarization.
本文研究香港粤语自十九世纪以来，声母及韵尾辅音的历史变化与变异。根据历史纪录、最近社会语言学资料、说今日香港粤语之年青大学生的话语、及实验语音资料，显示：（1）粤语的塞音、介音、闪音、唇齿擦音、喉擦音、及双唇鼻音声母，在历史过程中是比较稳定的；（2）早期粤语中两类塞擦音声母及两类擦音声母，分别合并成一类塞擦音声母及一类擦音声母；（3）这一类的塞擦音及擦音声母都是舌叶/齿龈声母，或舌叶/齿- 齿龈声母；（4）早期齿龈鼻音声母 ［n-］已在无条件下变成闪音声母 ［1-］；(5)早期舌根鼻音声母 ［-］及零声母，分别同时分裂成舌根鼻音声母［-］及零声母两类； （6）早期塞音韵尾 [-t, -k] 与鼻音韵尾 [-n, -] 都分别齿龈化及舌根化，而主要变化的方向是由后向前，即齿龈化。
Review 书评
Report 报告
Announcement 消息
Memoriam 缅怀
Volume 27, No 2
Article 文章
Abstract 摘要
In this article, we show that the parallelism between Mandarin gei ditransitive constructions and their English counterparts is misleading. We argue here that the gei that occurs next to a verb is part of a complex predicate, and that the gei phrase that occurs after the DO (direct object) is part of a SVC (Serial Verb Construction). The theoretical implications of this account of gei and some diachronical ramifications are also given.
本文证明汉语带「给」的双及物句式与英语的双宾句式仅是表面雷同。我们主张动词后的「给」是复合动词的一部份，而在直接宾语后由「给」带领的词组则是序列动词结构 (SVC, Serial Verb Construction) 的一部份。除了论据外，我们也提出了这个分析在理论上的重要性和历史语法中可能的佐证。
Abstract 摘要
Previous studies of Sino-Korean claim that the phonological system of modern Sino-Korean was derived from Middle Chinese. This conventional view is often misinterpreted to mean that Sino-Korean was never influenced by phonological changes in Chinese except for the Middle Chinese period. This study, however, demonstrates that Sino-Korean has had constant contact with Chinese throughout its history by identifying traces of Old Chinese, Middle Chinese, Old Mandarin, and Modern Mandarin all mingled in characters with multiple readings in modern Sino-Korean.
过去有关韩国汉字音的研究认为韩国汉字音与中国中古音最接近。这种看法导致一种误解：中古时期是韩国汉字音与中国音接触的唯一时期。本文先分析韩国现代汉字音里常用的六十多个多音字和异读字的读音，从中找出韩国汉字音里所存在的中国上古音与中古音以及近代汉语与现代汉语读音的痕迹。由此证明韩中两国语音接触的历史应提早到上古时期，历史上韩国汉字音一直受到中国语音变化的影响。
Abstract 摘要
The research reported here is intended as a contribution to an understanding of the adverbial clauses, used by speakers in spontaneous communication. This study aims at exploring adverbial clauses in spoken Mandarin conversations on the basis of quantitative analysis. There are four-hour conversation corpus in this research. The adverbial clauses in the data were divided into (i) preposed clauses to their modified material across continuing intonation, (ii) postposed clauses to their modified material across continuing intonation, and (iii) postposed clauses to their modified material with final intonation (rising question intonation or final falling intonation). After an inspection of the data, the results suggest that the temporal and conditional clauses are favored to occur before their modified material; and the causal clauses, after their modified material. The data also show that causal clauses are fundamentally different from temporal and conditional clauses in their use. In conversation, they are well-situated for appearing after the material they modify to introduce more information for understanding. Generally, the prototypical use of an adverbial clause is to prepose it before the material they link in Chinese spoken discourse, except the causal clause.
近来一些中外学者如 Du Bois (1985, 1987) 及Huang (1994, 1995, 1996) 等的研究说明言谈中不断而多样的语用需要促成了语法的规律化，即语法的形成是基于语言使用的规律化(routinization)。基于此信念，本研究旨在探讨汉语副词子句的语用功能及其常见的讯息顺序结构 (preferred information sequence)。论文前半部份着重于口语会话中表时间、条件、原因等副词子句的分布及用法。论文后半部份将口语语料的研究结果与书面语中的副词子句的分布做一比较。本研究中所使用的口语会话语料共有四小时，其中包含三小时不正式的日常生活会话及一小时的收音机广播节目的访谈(interviews)及叩应(call-ins)对话。以 Sacks, Schegloff, and Jefferson (1974) 的会话分析 (conversation analysis)，即发言机会转接模式 (turn-taking system)为理论架构。并依照Ford (1993)分类英语会话中副词子句的方法将此中文语料库中的副词子句依其主要修饰子句的相对位置及其语调调型 (intonation contour)分为以下三类：(i)出现在主要子句前的副词子句；(ii)出现在主要子句后的副词子句，其主要子句是一继续的语调 (continuing intonation)；(iii)出现在主要子句后的副词子句，而其主要子句是一结束的语调 (ending intonation)，以观察在不同位置与不同语调的副词子句的言谈功能。大致上，本研究所显示的结果与研究英语言谈与篇章的结果(如Ramsay, 1987; Ford, 1993)相同：出现在主要子句前的副词子句为引述下文之功用，然出现在主要子句后的副词子句则为补充解释前面的句子。另一方面，研究结果发现：表时间、条件与让步子句倾向于出现在主要（被修饰）子句之前，但表原因的子句在一般日常会话中大多出现于主要子句后，在收音机广播的访谈及叩应语料则出现于主要子句前后各占一半。此外，绝大部份出现在主要子句后的表原因子句，其所修饰的子句是一结束语调；此表示说话者在说此类子句之前，已经把本来意识焦点中想说的话说完了。换言之，大部份的表原因子句与其主要子句之间的关系极为松散，几乎是独立存在的子句。这表示在语法上表原因的连接词其实是个对等连接词 (coordinate conjunction)。可见表原因的子句之用法与其它类的副词子句极为不同，它们在会话中常出现于带有结束语调之主要子句后，此乃口语的特色。说话者偏好于使用表原因子句来补充说明其前句话，或是回答对方的问题，以消除听话者的疑惑，具有沟通的功能。论文最后从说话者的语言认知因素及各类副词子句如表时间、条件、原因子句的语意特色来解释中文会话中各类副词子句常见的讯息顺序结构。
Abstract 摘要
In Mandarin, the disposal marker is 把 pa3 and the passive marker is 被 pei4. These markers do not derive from the same lexical source. In some localities of Hunan, however, these two types of markers may derive from the same lexical source. Moreover, there are aberrant disposal markers which do not derive from 把 pa3 and passive markers which do not derive from 被 pei4. According to my investigations in 107 localities in Hunan, twenty forms can be employed as disposal markers and thirty five can be employed as passive markers. In this paper, the distribution of disposal and passive markers across dialects in Hunan will be presented. The path of grammaticalization of these two types of markers will be revealed by the analysis of their etymology in the Hunan dialects. There are clear cut distinctions between the two types of markers in the Hunan dialects geographically. In the Mandarin-speaking area, the main pattern is that the disposal ‘to give’ contrasts with passive ‘to suffer’; in the Xiang-speaking area, the main usage is that the disposal and the passive markers derive from a pair of synonyms meaning ‘to give’; in the Gan and Kejia speaking area, the main pattern is that the disposal marker ‘to take’ contrasts with the passive marker ‘to give’; and in the southern area, where both a Mandarin dialect and unclassified local dialects co-exist, the main usage is that the disposal and the passive markers derive from the same form ‘to give’. I propose that not only are the lexical sources of the passive/disposal markers in the Hunan dialects different from Mandarin, but also that the path of grammaticalization is also different. The first step of grammaticalization of disposal/passive markers meaning ‘to give’ occurred with a re-analysis that changed the verb ‘to give’ into a preposition indicating relationship between a subject and an object in a double-object construction.
本文旨在通过湖南方言中表处置和被动词的介词词源的研究，探讨这两类介词的语法化进程. 在我所调查的107个方言点中，有20个介词形式可以用来表处置，35个介词形式可以用来表被动。表处置与表被动的介词的词源以及地理分布可粗略地描写如下：在湘北湘西的官话区，表处置的介词来源于表“给予”意义的动词，表被动的介词来源于表“遭受”意义的动词；在湘南中部的湘方言区，表处置和表被动的介词来自一组表“给予”意义的同义词；在湘东的赣客方言区，表处置的介词来源于表“拿取”意义的动词，表被动的介词来自表“给予”意义的动词；在湘南的双语区，表被动和表处置的介词共用一个词形式，这个词形式来自表“给予”的动词。我认为湘南保留了南方方言中较古老的表处置和被动的形式。表“给予”的动词发展为表被动的介词第一步是这个动词在双宾句式中虚化为介词。而这个介词的处置义是后起的。
Abstract 摘要
In the Hakka dialect, there are more than 40 nouns with gendered suffixes. These gendered nouns are widespread in the daily language including the names of animals, body parts and inanimate objects. Gendered nouns are also found in other neighboring dialects, but to a much lesser extent and they share little homology with these nouns. The occurrence of these nouns is a very interesting phenomenon and their significance is discussed.
在客方言中﹐有超过四十个代性别词缀的名词﹐这些带性别词缀的名词，也广泛地分布于其它邻近方言中﹐但数量远少于客语﹐而且同源性也不高。这些名词的出现很值得注意﹐而我在本文中﹐也将它们与其它方言和少数民族语言比较。
Abstract 摘要
This article discusses the theoretical and methodological issues involved in the practice of periodization. Theoretically, it clarifies the nature of periodization, which is a type of diachronic classification. In order to understand this concept better, discussions are offered to show the differences and similarities between periodization and other diachronic classifications. Methodologically, this article shows that periodization may be carried out by employing numerical methods. The aim of the numerical approach suggested here is to rid periodization of its traditionally subjective nature and to provide objective and stable classifications. For the purpose of demonstration, the data contained in Wang Li’s Phonological History of Chinese (1985) are utilized and calculated.
本文讨论语言分期中所涉及的理论和方法上的问题。在理论上，本文指出语言分期属于历史语言分类范畴，并分析了语言分期和其它历史分类的异同。在方法上，本文建议运用更为客观的方法来作语言分期的基础。运用数理方法的主要目的在于屏除在语言分期中的主观因素，进而建立更为客观和稳定的分期。为了讨论具体方法的目的，本文利用并统计了王力先生《汉语语音史》一书中的资料。
ZH-HK 
EN 
ZH-CN