朱丽师

（女，2016级博士，专业：汉语言文字学，研究方向：汉语史，导师：杨永龙）

　　摘 要

　　本文主要以汉语史中的复杂句为研究对象，探讨了不同句子类型间的整合模式。文章共分为四个部分，即绪论、复杂句的整合模式、小句整合模式的跨语言考察和结语。

　　绪论部分介绍了研究现状、选题意义、研究内容、研究方法、语料范围等。在研究现状部分，我们还对传统汉语和西方语言学中的句子分类作了详细对比；梳理了“并列关系>主次关系>主从关系”复杂句三分句间格局的形成，并提出了“并列关系>主次关系2>主次关系₁>主从关系”四分格局的主张；对本文涉及到的相关概念进行了介绍和界定。

　　第二章到第五章主要是以复杂句类型为分类标准，探讨汉语史中小句整合模式。通过多个细颗粒度的个案研究分别对小句宾语句的整合、小句状语句的整合（整合的起点为小句状语句、整合的终点为小句状语句）和联合复合句的整合进行深入考察，获得一些前人研究中未及注意的认识。

　　第二章主要讨论了小句宾语句的整合过程和机制。首先获得了关于整合条件的认识，即当主句动词是认知类非叙实动词时，才有与小句宾语发生整合的可能；在整合过程中，主句动词的虚化类型与情态动词、副词、话语标记等语法范畴相关。进而选取了“保证类”言语行为动词、“恐－怕”类动词、和“不见得”为个案考察对象，具体分析了它们由主从关系的“主句动词＋小句宾语”到简单句的整合。可以用Harris & Compbell（1995:172）概括这一整合过程：a.双小句变为单小句；b.母句中心谓语降格为情态成分c.小句宾语中的动词升格为中心动词。“莫道／说”“休道／说”类言说动词的语法化过程涉及到另一种整合模式，其原句法结构也是“主句动词＋小句宾语”，其整合结果“莫道NP₁／C₁，让步句C₂”是一个外层递进，内层包括让步句的结构式。

　　第三章讨论了小句状语句作为起点的整合过程和机制。在“小句状语＋主句”结构中，从句和主句都有可能发生降级。从句升级为主句，单动词主句降级为句内成分的情况如：主次关系“状语从句＋X好／好了／得了”整合为“简单句＋句末语气词”；主次关系“状语从句＋X得”整合为“须…… 才得”等形式的双情态词共现的简单句。从句发生非句化，降级为主句的句内情态成分的情况如：条件状语从句“弄／闹／搞不好，C₂”整合为表达认识情态的简单句，“弄／闹／搞不好”由条件句前件降级为句内情态成分，C₂由状语从句的主句变为独立简单句。

　　第四章讨论了小句状语句作为终点的整合和机制。选取“不是”“不拣／不拘”“比似”和“省得”为研究对象，对违实条件句“不是C₁，C₂”，无条件让步句“不拣／不拘C₁，C₂”，多种语义关系的条件句“比似C₁，C₂”和消极目的句“C₁，省得C₂”的句法演变过程进行了细致的考察。其中前三者均为双向依附主次关系，“C₁，省得C₂”是单向依附主次关系。

　　第五章简要讨论了联合复合句的整合过程和机制。“不单／不独C₁”从独立的简单句整合为联合复合句的分支小句，这一整合过程和“单／独”的语义辖域以及后续句C₂的有无密切相关。只有从句子整合的角度，才能很好地展示出这类词语获得句子衔接功能的过程。

　　结语部分从四方面总结了全文：

　　首先，借助既有文献对小句整合做了简要的跨语言的观察。西方学者主要从两个角度观察小句整合：从篇章到句法的整合，和复杂句内部的整合。介绍了西方学者对不同语言整合模式的考察，总体上看，这些研究均可概括为“并列关系>主次关系>主从关系”。归纳了小句整合的句法－语义表征，如语序、语调、名词性成分的关系管辖和特殊动词形式等。

　　其次，概括和梳理了小句整合和语法化的关系。小句整合是一个动态的句法演变过程，作为句子结构的语法化，是语法化体系的一个重要方面。通常讲到的小句整合与语法化的关系主要是指，句法结构和功能词之间有互动关系：在某种语法形式基础上发展出新的语法功能、从词汇项向语法项的演变（也即通常所说的“实词虚化”）过程中均可能引发小句整合，但不是所有的小句整合都伴随发生词汇项的语法化。

　　第三、概括出七个可能触发汉语句子整合的参项：论元的复杂度、小句主语的有无、后续句的有无、共享论元的有无、否定语义对句法降级的影响、句法自立度的增强和句子信息地位的高低。

　　第四、总结了汉语史中的小句整合模式：除了西方学者概括的“并列关系>双向依附主次关系>主从关系”这一小句整合斜坡外，汉语还有“双向依附主次关系>单向依附主次关系”、“简单句>双向依附主次关系”和“双向依附主次关系>简单句”等整合模式，并尝试对汉语小句整合模式体现的规律性原则进行了理论挖掘。

　　关键词：小句整合；小句宾语句；小句状语句；联合复合句；简单句；汉语史

　　Abstract

　　This dissertation focuses on the complex sentences in the historical Chinese language, and explores the modes of clausal combining with reference to different sentence types. It is composed of four sections, i.e. the introduction, the modes of combining concerning complex sentences, the combining of clauses from cross-linguistic perspectives, and the conclusion.

　　In the section of introduction, the dissertation introduces the accomplishments by the existing related researches, the manifold significance of the topic, the content and methodology of the present research, and the background information of language data. On introducing the existing relevant researches, the paper offers a detailed comparison among differing classifications of sentence types in both frameworks of Chinese and Western linguistics. Based on a brief introduction regarding the formation of the tripartite pattern of inter-clausal combinations, i.e. coordination> cosubordination> subordination, the dissertation further proposes a cline as ‘coordination> cosubordination2> cosubordination1> subordination’. This section also provides a set of definitions for the relevant conceptions and categories.

　　Chapters II to Chapter V dwell mainly upon the modes of combining of clauses in the historical Chinese with regard to the different types of complex sentences. On the basis of a number of finer-grained case descriptions, the dissertation offers in-depth inquiries into the combining concerning sentences containing clausal complements, those having adverbial clauses (as either start point or the endpoint of the combining process) and concerning coordinated compound sentences. Through these investigations the paper obtains a series of understandings that have hitherto been only inadequately notices or even unintended.

　　Chapter II deals mainly with the combining occurs in sentences containing clausal complements. This chapter first provides a finding concerning the conditions on which the combining of clauses is possible. To be concrete, if the main predicate verb is a non-factive verb, it is possible that the verb be integrated into the original clausal complement. Along with the combining, grammaticalization is likely to occur for the former matrix verb and it is related to the grammatical categories such as modal verbs, adverbs, discourse markers, etc. Based on this proposition the paper particularly focuses on several verbs of different semantic types, i.e. on the speech act verbs indicating baozheng (保证 lit. guarantee), the verbs of kong-pa (恐-怕 lit. fear-be.afraid), verbs and the commonly used discourse-based VP of doubt bujiande (不见得 lit. probably not; not necessary). The dissertation then analyzes the combining process from “matrix verb + clausal complement” into simple sentences. The paper then applies what is found in Harris & Compbell (1995:172) to summarize the combining process: A. Bi-clausal structure becomes a single clause; B. The main verb of the matrix sentence is downgraded into modality-indicating elements. C. Meanwhile the verb in the complement clause is upgraded to the level of the matrix verb. The grammaticalization of “modao/moshuo (莫道/莫说 lit. not to mention)” and “xiudao/xiushuo(休道/休说 lit. not to mention)” verbs associates itself with another mode of combining, with their original syntactic structure being also “main verb + complement clauses”. The result of the combining of “modao NP1/C1+C2 as a concession” is that a concession clause contained in the progressive clauses.

　　Chapter Ⅲ deals mainly with the combining occurs in sentences containing adverbial clauses which is the origin of the combining process. In the “adverbial clauses + main clause” structure, the subordinate clause and the main clause show the equal possibility to be downgraded. There is another pattern of the swift of syntactic level, i.e., when the subordinate clauses are upgraded to the main clauses and the former main clauses downgraded to the sentence-internal fragments. Such as the co-subordinate sentence “adverbial clause + Xhao/haole/dele(好/好了/得了 lit. good)” combined into “simple sentence (main clause) + sentence-final modal particle”; the cosubordinate sentence “adverbial clause + Xde(得 lit. good)” combined into a simple sentence in which the circumferential modal auxiliaries co-occur, i.e. “xu…caide(须……才得 lit. must…be necessary)”. The former subordinate clauses have been de-clausalized so as to become modal-indicating components for the main clauses. Such as the original conditional-consequence clausal structure of “ ‘nong/nao/gaobuhao，C2’ (‘弄/闹/搞不好,C2’ lit. ‘do-bad, C2’)” finally squeezes itself into a simple sentence expressing the epistemic modality. The conditional sentence downgrades into the intra-sentence modal element, and C2 changes from the status of a main clause to an independent simple sentence.

　　Chapter Ⅳ is mainly concerned with the combining occurs in sentences containing adverbial clauses which is the endpoint of the combining process. The paper selects “bushi(不是lit. not.be)” “bujian/buju (不拣/不拘 lit. not.limited.by not fettered by’)” “bisi (比似 lit. resemble)” and“shengde(省得 lit. in order not)” as the tokens for research and observes the syntactic evolution with regard to counterfactual conditional sentence “bushi C1, C2”, the unconditional concession sentence “bujian/buju C1, C2”, the multiply ambiguous sentence “bisi C1, C2” and the negatively purposive sentence “C1, shengde C2”. The first three are the cosubordination sentences showing bidirectional dependence, and “C1, shengde C2” mono-directional dependence.

　　Chapter Ⅴ focus the combining occurs in sentences containing compound sentences composed of coordination clauses. “budan/budu C1(不单/不独 lit. not only)” combines from an independent simple sentence into a pair of coordinants. This integration process is closely related to the semantic scope of “dan/du(单/独 lit. only)” and presence-absence of the subsequent sentence C2. Only from the perspective of sentence combining can the process of obtaining the cohesion function of such words be well demonstrated.

　　The section of conclusion provides summaries of the dissertation in four aspects:

　　Firstly, it offers across-linguistic perspective from which the functional and formal features regarding clause combining are typologized and revisited on the basis of existing literature including various reference grammars. Western scholars observe the combining of clauses mainly from two perspectives: the combining from text to syntax, and the that within complex sentences. By introducing western scholars’ investigation of combining modes in different language, these studies can be summarized so as to formulate a cline as ‘coordination – cosubordination – subordination’. This chapter also summarizes the syntactic-semantic representation of clause combining, such as word order, intonation, the relational government components of NPs and special verb forms.

　　Secondly, it generalizes what have been obtained by the previous chapters and gives a special attention upon the relationship between clause combining and grammaticalization. Clause combining, as the grammaticalization of sentence structure, is a dynamic syntactic evolution process and an important aspect of grammaticalization system. Generally speaking, the relationship between clause combining and grammaticalization mainly refers to the interaction between syntactic structure and functional words: both the development of new grammatical functions and the evolution from lexical items to grammatical items may lead to clause combining, but not all the process of clause combining is accompanied by the grammaticalization of lexical items.

　　Thirdly, the chapter reiterates the seven parameters that may trigger the clause combining effect in Chinese language: the complexity of arguments, the presence or absence of the clause subject, the presence or absence of subsequent sentences, the presence or absence of shared arguments, the influence of negative semantics on syntactic degradation, the enhancement of the syntactic independence and the status of sentence information.

　　Fourthly, it offers a general classification and conclusion of the modes of clause combining in the historical Chinese language. In addition to the canonical clause combining cline of ‘coordination > bidirectional cosubordination > subordination’ summarized by Western scholars, there also exists ‘bidirectional cosubordination > mono-directional cosubordination’, ‘simple sentence > bidirectional cosubordination’, ‘bidirectional cosubordination > simple sentence’ and other combining models. And the dissertation tries to theoretically explore the regularity principles embodied in the combining mode of Chinese clauses.

　　Key words: clause combining; complement clauses; adverbial clauses; coordinated compound sentences; simple sentences; historical Chinese language