ABSTRACT
This paper focuses on Marked Theme in spoken Chinese. Previous studies have explored the functions of Marked Themes within clauses, but not many researchers have observed this phenomenon from a discourse semantics perspective. Combining the framework of discourse semantics with the use of corpus study, this paper first explores the occurrence of such structures in conversations and then examines their discourse functions. Data from the corpus show that, in line with findings from corpus studies of written texts, the occurrence of Marked Themes is quite low. The inspection of their discourse functions leads to the following preliminary findings: (1) in sequences of activities, Marked Themes often function as lexical cohesive devices to build up taxonomical relations across utterances; (2) from the view point of identification, Marked Themes function as anaphoric devices to help keep track of the Participants in question; (3) from the viewpoint of periodicity, while most Marked Themes convey given information, especially contrastive information, a considerable amount of Marked Themes function as HyperTheme to start new phases of discourse.
1. Introduction
This paper focuses on one element which appears in the initial position of the clause in Chinese and may or may not be related to the Process of the Actor + Process + Goal configuration or the Predicator of the Subject + Predicator configuration. Elements filling this slot bear various labels. Systemic functional linguists named them according to their textual functions within clauses, using terms ranging from Absolute Theme (Matthiessen 1995; Caffarel 2006), Contextual Topical Theme (Contextual Theme for short, Li 2007; Fang 2008), to Marked Topical Theme (Li 2007; Fang 2019). In traditional Chinese linguistics and other approaches, including structural, cognitive functional and generative approaches, they are called “Main Subject” (referring to nominal group, in contrast with the clause Subject, Chao [1968] 2011; Zhu 1982, Zhu 1999; Yuan 1996; Shen 2012), Topic (Li and Thompson 1976, 1981), Adjunct (referring to nominal groups or prepositional phrases indicating time, place, condition, etc. Yuan 1996) or “Non-Subject Topic” (Chen and Wang 2010).
There has been a wealth of previous research on Themes in Chinese from a systemic functional perspective, including comprehensive overviews of the THEME system in Chinese clauses (Zhang and Fang 1994; Li 2007; Fang 2019), and observations on the distribution of Marked Themes (or Non-Subject Topics) in written texts, novels (Li 2007), short stories (Chen and Wang 2010), and academic articles (Yu 2002), etc. However, there have been fewer studies on the choice of Marked Themes in spoken texts, in dialogues or conversations. Li and Thompson (1981) maintain that “topic,” their term for this element, is a discourse rather than sentence notion in that such elements are often closely related to the preceding and following discourse. Matthiessen (1995) and Li (2007) also maintain that such elements are often related to the clause cohesively through reference or lexical cohesion. Fang (2008) finds that one of the functions of Marked Theme is to link clauses in a Theme-Rheme chain. However, except for the endeavors which probe this issue from a pragmatic perspective (LaPolla 1995, LaPolla 2009; Xu and Liu 2007, etc.), detailed explorations of the functions of such elements at discourse level are lacking. How often do Marked Themes occur in everyday conversation? What are the functions of such elements in discourses? How do they relate to preceding or following discourse? These will be the questions we aim to address in this study.
Here we use the term “Marked Theme” as a superordinate term to include all three kinds of Themes identified in previous studies: Absolute Theme, Contextual Topical Theme, and Marked Topical Theme. Based on a corpus of Chinese conversations, we first calculate the occurrence of Marked Themes in turns and moves, and then examine the functions of these Marked Themes in discourse using the discourse semantics framework developed by Martin (1992) and Martin and Rose ([2003] 2007). This framework fits the present study in that some Marked Themes – Absolute Theme and Preposed Theme in particular, when viewed at clause level, often appear to be dangling element, “independent of the verb” (Li and Thompson 1976, 463), not conforming to any sentence-internal constraints (except in the case of inverted Goal), yet they play an important role in connecting current utterances to previous ones.
The rest of the paper is organized as follows: Section 2 discusses the notions of Theme and Marked Theme in Chinese, and Section 3 briefly introduces the framework of discourse semantics. Then, Section 4 describes the corpus used in the study and the research method taken to answer the research questions, while Section 5 discusses the functions of Marked Themes in conversational data from the corpus. Finally, Section 6 comes to some conclusions and discusses future research directions.
2. Theme and Marked Theme in Chinese
In systemic functional linguistics (SFL for short), the clause is viewed as encoding three distinct types of meanings or metafunctions: ideational, construing people’s experience of the world around and inside them; interpersonal, enacting the world of social roles and relations; and textual, organizing ideational and interpersonal meanings as a semiotic world of information organized as text. These metafunctions are realized by different types of structures that map onto one another in the clause.
Ideational function is further differentiated into two subsystems: experiential and logical. The experiential function is realized by the TRANSITIVITY system. The clause, realizing the ideational metafunction, represents our experiences as various processes of doing (material), thinking (mental), or being (relational), etc. The unmarked functional configuration of representation in an English clause is Actor + Process + Goal, where Actor is the element which “the speaker portrays as the one who does the deed” (Halliday [1985] 1994, 34); Process is the doing, happening and being, etc; Goal is the “participant being affected or impacted by the involvement of the Actor in the Process” (Matthiessen, Teruya, and Lam 2010, 108). A figure is “a configuration of elements – a process, participants involved in it and attendant circumstances” (Matthiessen, Teruya, and Lam 2010, 97). The logical metafunction manifests itself in the formation of group and phrase complexes and clause complexes. In Chinese, “the nucleus of a figure consists of the process plus a participant directly involved in it” (Li 2007, 38). Sometimes one or two additional participants may get involved in the process. The unmarked functional configuration (for material process) of a Chinese clause consists of (Actor) + Process + (Circumstance).
The interpersonal metafunction is realized by the clause as an exchange, encoding the symbolic exchange between the speaker and listener (Halliday [1985] 1994, 34). The grammatical resources for realizing this metafunction is the MOOD system. The unmarked functional configuration in English is Subject + Finite. In Chinese, Finite is realized implicitly; therefore, the unmarked functional configuration is Subject + (Adjunct) + Predicator + (Complement). Fang (2008, 90) maintains that “from the point of view of its interpersonal function, the clause divides into a Predicator, Subject, Complement and Adjuncts – an Interpersonal Structure.” In both English and Chinese, the Subject is “the element the speaker makes responsible for the validity of what he is saying” (Halliday [1985] 1994, 34).
The textual metafunction is realized by the clause as a message, a quantum of information in the flow of information as a text unfolds (Halliday 1994: 34, Halliday and Matthiessen 2014). This metafunction is realized through the THEME system which is the resource for “manipulating the contextualization of the clause,” the resource for “setting up a local context for each clause in a text” Matthiessen (1995, 531). The unmarked functional configuration in both English and Chinese is Theme + Rheme. Theme is the point of departure, the local context of the message, and the background against which the speaker presents new information. The Rheme is the remainder of message where the Theme is developed.
Following SFL framework, Fang (2008, 90) formulated a clause in Chinese as “the combination of a single experiential structure, a single interpersonal structure and a single thematic structure” as shown below:
In English, a verb or verbal group is the most important element to delineate a clause. Thompson (2014, 17) defines a clause as “any stretch of language centered around a verbal group.” However, in Chinese, verbs or verbal groups are not the most important elements in defining a clause. While most clauses have a verb or a verbal group as their Predicator, some clauses may not have a verbal element. Under such circumstances, the Predicator, which may be realized by a verb, a noun or an adjective, functions as the defining element to form a clause. To see whether a stretch of language is a clause or not, we need to see whether there is a Predicator. McDonald (1998, 349) points out that in Chinese clauses, the Predicator is “precisely that element that gets ‘tossed’ back-and-forth in dialogic exchange.” Therefore, he proposes a distinction between major clauses (with Predicator, or one where a Process/Predicator can be added) and minor clauses (without Predicator, or one where a Process/Predicator cannot be added) (1998, 60). This distinction will be followed in this study.
2.1. Theme in Chinese
Theme occurs at the initial position of a clause and orients the clause in its local context. Theme is delimited with reference to the TRANSITIVITY structure and MOOD structure. In experiential structure, the guiding principle of thematic structure is that the Theme contains a single experiential function, whether Participant or Circumstance (most commonly), or Process. This experiential element, in its textual function, is referred to as the Topical Theme (Halliday and Matthiessen 2014). Chinese Themes can be divided into simple Theme and multiple Theme. Simple Theme is realized by one component in the experiential structure, thus it is also called Experiential Theme or Topical Theme. Multiple Themes include Textual Theme and Interpersonal Theme which typically occur before the Experiential Theme in clauses. Examples (1) and (2) below show the simple Themes realized by a nominal group (as in (1)) and an embedded clause (as in (2)). Simple Themes can also be realized by an embedded clause in declarative clauses or a verb in imperative clauses. In this article, unless otherwise specified, examples are drawn from our own corpus of spoken Chinese (PKUC). In each example, the first line displays the Chinese characters, the second line gives romanised version using the standard Chinese pinyin, with tones omitted. The third line is the word-for-word translation or glossing. The fourth line provides a functional analysis in the clause. The fifth line is the English translation. Themes in question are underlined, with reference to their experiential functions. Abbreviations for grammatical words follow the conventions of Li (2007) and Fang (2008), such as MEAS for measure, ASP for aspect marker, TPART for topic particle, POSS for possessive marker, NEG for negation, VAVD for verbal adverb, etc.
(1) 植物学 也 挺 有意思。[PKUC005]
Zhiwuxue ye ting youyisi
botany also very interesting
Theme/Carrier Rheme/Attribute
“Botany is also very interesting.”
(2) 用 信用卡 担保 就 可以。[PKUC033]
Yong xinyongka danbao jiu keyi
use credit card guarantee just fine
Theme/Carrier Rheme/Attribute
“To guarantee with a credit card will be fine.”
In (1) 植物学 (zhiwuxue, “botany”) is both the Carrier and Theme. In (2) the embedded clause 用信用卡担保 (yong xinyongka danbao, “to guarantee with a credit card”) functions as the Carrier and Topical Theme.
Multiple themes occur when elements other than the components of the transitivity structure proceed the Topical Theme. These elements may be grouped into two categories: Textural and Interpersonal Themes. Textual Themes are realized by conjunction like 然后 (ranhou, “then”) in (3) or continuatives like 总之 (zongzhi, “in conclusion”) whose function is to link the clause with previous clauses. Interpersonal Themes are realized by modal/comment Adjuncts or vocatives in Chinese clauses. In (4) the interpersonal Theme is realized by the modal Adjunct 可能 (keneng, “maybe,” “probably”).
(3) 然后 我 就 打算 每一周 节约 100 块 钱。[PKUC033]
Ranhou wo jiu dasuan meiyizhou jieyue 100 kuai qian
then I ADV plan every week save 100 MEAS money
Textual Topical/Senser
“Then I plan to save 100 yuan every week.”
(4) 可能 大部分 同学 还是 高三 补 的 英语。[PKUC047]
Keneng dabufen tongxue haishi gaosan bu de yingyu
probably most students still senior three make up for POSS English
Interpersonal Topical/Actor
“Probably most students improved their English in senior three.”
This study includes Themes in both clause simplexes and clause complexes. By clause simplex, we mean a single clause which consists of one experiential structure, one textual structure, and one interpersonal structure (cf. Matthiessen 1995, 189). A clause simplex can be either a free clause or a bound one. If it is free, it serves as an independent clause by itself. If bound, it relates to the main clause hypotactically. As our interest in this study is on the contributions of Marked Theme in the flow of information, only Themes in free clause in clause simplexes, Unmarked, Elliptical, Marked, were considered and counted. Themes in bound clauses, which often do not contribute to the development of thematic progression in a text, were not counted. Examples (1), (2) and (4) above belong to free clause simplexes, and clause (3) is a clause complex. Clause complexes, or complex clauses, consist of more than two simplex clauses which are linked together. When clause complexes occurred within one utterance, only Themes in independent clause were counted. Themes in dependent clauses coming before the independent clause on which they depend were not counted. In such cases, we follow the conventions of Thompson (2014, 159) and treat the whole dependent clause as Theme. Themes in dependent clauses coming after the independent clause they depend on or in the projected clauses like example (3) were not counted either. When clause simplexes, globally belonging to a complex clause that is jointly constructed by different speakers, occur separately in different utterances across turns, their Themes were counted independently. In example (19) T23 (i) below, where the second clause simplex within a clause complex structure ((所以) … … 因为, (suoyi) … yinwei, “the reason why … because”) occurs by itself in a new turn, its Theme, though coming after a typical causal conjunction for cause-effect clause complex, is counted, as the presumed main clause has to be retrieved far back across several turns to the question raised in T20.
Apart from the textual and interpersonal Themes, some Chinese clauses allow one more element to occur before the Actor + Process + Goal or Subject + Predicator configuration, a nominal group or a prepositional phrase. Some of them fall into the category of Contextual Topical Theme and some into the category of the Marked Topical Themes in Li (2007) and Fang (2008, 2019). In this study, the term Marked Theme will be used as a hypernym to cover two sub-categories of Themes which are not conflated with the Subject, the Contextual Themes and Marked Topical Themes.
2.2. Marked Themes in Chinese
Distinction between Marked and Unmarked Theme in English was made in relation to their functions in the MOOD structure. In Chinese, both mood function and transitivity functions are taken into consideration as mood is not realized by word order but rather by mood particles. In declarative clauses, the typical order is Subject ^ (Adjunct) Predicator ^ (Complement) (^ means followed by). When Subject/Actor is conflated and functions as Theme, it is Unmarked Theme, a default choice that speakers most frequently select, as in (1). Li maintains that “about 90% of the thematic prominence is assigned to the participant (of experiential metafunction) which is conflated with the Subject (of the interpersonal metafunction) of the clause” (2007, 164). In imperative clause, the Unmarked Theme will be the verbal group which comes at the clause-initial position. In yes/no and WH-interrogatives, the default structure is Subject + yes/no/WH- + Predicator + mood particle (吗ma, 吧ba, etc.). In exclamative clauses, the typical word order is also Subject + (Adjunct) + Predicator + assessment particle (呀ya). Hence, the Subject/Actor in both interrogative and exclamative clauses will be the Unmarked Theme. However, word order in Chinese clauses can be very flexible, especially in spoken language. Li (2007, 164) claimed that “almost every functional component in the experiential metafunction can take the clause-initial position and can therefore be given thematic prominence.” When other elements instead of the Subject/Actor occur at the clause-initial position, we have Marked Theme.
Marked Theme in Chinese has attracted considerable interest among researchers in the past three decades. A clear distinction is made between Contextual Topical Theme and Marked Topical Theme. Li (2007) and Fang (2008) identified two types of Contextual Theme, Absolute Theme and Preposed Theme. Absolute Theme stands “outside the experiential structure of the clause altogether” (Halliday and McDonald 2004, 321),provides “the textual ‘subject matter’ that plays a role in the textual metafunction only” and “does not serve any interpersonal or ideational roles in the clause and is thus not structurally integrated with these metafunctions through conflation” (Matthiessen 1995, 552). Caffarel (2006, 197) thinks that this element only has a textual function and “does not map onto a transitivity or interpersonal function.” Matthiessen’s Absolute Theme covers both Absolute Theme and Preposed Theme in Li’s and Fang’s Contextual Theme. It also includes the cases where the subject matter is introduced by a preposition such as about, as for, regarding, etc. In Li’s (2007) and Fang’s (2008, 2019) THEME systems, such cases are recategorized into Marked Topical Theme which are realized by Circumstances of Matter. Marked Topical Theme refers to Themes realized by one of the experiential components rather than the Actor/Subject conflation which has been moved to the initial position of the clause and is given thematic prominence. In this study, the categorizations of Contextual Topical Theme (Contextual Theme for short) and Marked Topical Theme of Li (2007) and Fang (2008, 2019) will be maintained as two subtypes of Marked Themes.
2.2.1. Contextual Topical Theme
Contextual Topical Theme refers to the functional element which is realized by a nominal group or prepositional phrase that occurs in the clause-initial position, before any component of the transitivity structure, and provides setting for the whole clause. Contextual Topical Theme is further divided into two subcategories, Absolute Theme and Preposed Theme. Absolute Themes occur before the transitivity structure or the Subject + Predicator configuration. Li (2007, 180–181) points out that this element is related to the rest of the clause by lexical cohesion and often exhibits a part-whole relationship with the Experiential Theme, as in (5). Chafe (1976), Matthiessen (1995) and Chen (1996) note that this element often provides the “subject matter” or domain for the clause, as shown in (6) and (7).
(5) 这个 女孩 眼睛 很 大。(Li and Thompson 1981, 92)
Zhege nvhai yanjing hen da
this girl eyes very big
Theme (Marked) Theme/Carrier Rheme/Attribute
“This girl, (her) eyes very big.”
(6) 物价 纽约 最 贵。 (Chen 1996)
Wujia niuyue zui gui
price New York most expensive
Theme (Marked) Theme/Carrier Rheme/Attribute
“(As for) prices of things, New York is the most expensive place for shopping.”
(7) 那 种 豆子, 一 斤 三十 块 钱。(Pan and Hu 2008, 1976)
Na zhong douzi yi jin sanshi kuai qian
That CL beans one MEAS thirty MEAS money
Theme (Marked) Theme/Token Rheme/Value
“One catty of that kind of beans is thirty dollars.”
Clauses with Absolute Theme can be analyzed at two layers: within the experiential structures and at the whole clause level. Within the experiential structure, the first experiential element will be the Theme, i.e., eyes in (5), New York in (6) one catty in (7) and leaves in (8).
(8) 这 棵 树 叶子 大。 (Chao [1968] 2011, 120)
Zhei ke shu yezi da,
this MEAS tree leaves big
Subject/Carrier Attribute
Theme2 Rheme2
Marked Theme1 Rheme1
“This tree has big leaves and few flowers.”
At the whole clause level, the first element in (8) will be Marked Theme and the clause 叶子大 (yezi da, “leaves are big”) Rheme. The absolute Themes in clauses like (5), (6), (7), and (8) are also known as Topics (Li and Thompson 1976, 1981), Chinese-style topics (Chafe 1976) or Non-subject Topics (Chen and Wang 2010). They stand outside the clause structure but usually set a frame or provide a setting for the event described in the clause. This element is what distinguishes Chinese from many other languages. As the concept of Topic plays a fundamental role in forming Chinese clauses Li and Thompson (1976), Li and Thompson (1981) characterized Chinese as topic-prominent language.
The other type of Contextual Theme is Preposed Theme. Preposed Theme, according to Li (2007, 180), is identical to one of the elements in the clause. It has a relationship of co-referentiality with a particular interpersonal or experiential component and may be realized by a nominal group which may be picked up anaphorically by a pronoun as in (9) or by another nominal group in the clause structure as shown in (10).
(9) 那 个 人 她 就 不是 很 信赖 民宿。[PKUC033]
Nei ge ren ta jiu bushi hen xinlai minsu
that MEAS person she ADV NEG very trust homestay
Preposed Theme Topical/Senser Phenomenon
“That girl, she just didn’t trust homestay.”
(10) 通情达理, 有求必应, 大家 都 喜欢 这 种 人。(Fang 2008, 106)
Tongqingdali, youqiubiying, dajia dou xihuan zhe zhong ren
reasonable ready to help everybody VADV like this CL person
Preposed Theme Topical/Senser Phenomenon
“(Someone who is) reasonable and ready to help: everybody likes such a person.”
In (9) 她 (ta, she) refers to the girl just mentioned. In (10), the two idiomatic phrases which are used to refer to a kind of person are resumed by a nominal group at the end of the clause functioning as Phenomenon/Complement. One common feature shared between Absolute Theme and Preposed Theme is that they have no experiential or interpersonal role to play by themselves in the clause structure. For this reason, they are also known as dangling Topics (Pan and Hu 2008). This feature distinguishes them from the Marked Topical Themes.
2.2.2. Marked Topical Theme
Marked Topical Theme refers to Themes realized by non-Subject participants in the representational structure, i.e., by Circumstances (Adjunct), non-Subject participants (Complement), or process (Predicator). Themes in examples (11), (12), and (13) are Marked Topical Themes frequently occurring in spoken Chinese.
(11) 球类运动 我 都 挺 喜欢 的。[PKUC002]
qiuleiyundong wo dou ting xihuan de
ball games I ADV ADV like MPART
Marked Theme/Complement/Phenomenon
“As for ball games, I like them all.”
(12) 衬衫 我 买 了 三 件。 (Chen and Wang 2010, 321)
Chenshan wo mai le san jian
Shirt I buy ASP three MEAS
Marked Theme/Complement/Goal
“Those shirts, I bought three of them.”
(13) 主治医师, 他 早 就 是 了。(Xu & Liu, 2007, 157)
Zhuzhiyishi ta zao jiu shi le
Primary care physician he early already be ASP
Marked Theme/Complement/Identifier
“The position of primary care physician he took on years ago already.”
In (11), the Phenomenon in the mental clause has been moved to the clause-initial position and given thematic prominence. In (12) the Goal and in (13) the Identifier occur before the Subject + Predicator, hence these count as Marked Topical Themes. Such elements are also called inverted Object or object-fronting by Chinese linguists (Yuan 1996).
Circumstances of time, location1 or matter can also be thematized as Marked Topical Theme. The function of the nominal groups in (14) and (15) is to set a temporal or spatial domain or scope for the event of the clause. In (14) the nominal group 青壮年的时候 (qingzhuangnian de shihou, “at a young age”) provides a time setting for the clause. In (15) the nominal group 我们那边 (women neibian, “our place”, “my hometown”) provides a location setting for the description of snow. The prepositional phrase in (16) states the matter that people are concerned about.
(14) 青壮年 的 时候 一定 不会 在 这边。[PKUC003]
Qingzhuangnian de shihou yiding buhui zai zhebian
young POSS time definitely NEG be here
Marked Theme/Adjunct (time) Process
“At a young age, (people) will definitely not be here.”
(15) 我们 那边 雪 很 厚。[PKUC014]
Women neibian xue hen hou
we there snow very thick
Marked Theme/Adjunct (location) Carrier Attribute
“In my hometown, the snow was very thick.”
(16) 对于 波斯湾 的 战争, 我 们 都 很 关心。(Fang 2008, 103)
Duiyu bosiwan de zhanzheng women dou hen guanxin
PREP Gulf War we ADV VADV concern
Marked Theme/Adjunct (matter) Senser
“As for the Gulf War, we are all very concerned about it.”
Two special cases found in our corpus which previous studies have not mentioned are the Marked Themes of Epithet and Contingency. Epithet plays a functional role in nominal groups, as modifier of the Head. However, in conversations, they can be picked up by the speaker to form class-member relations, as shown in (17).
(17) 红色 的 它 有 两 本 红色 的。[PKUC005]
Hongse de ta you liang ben hongse de
Red POSS it have two MEAS red POSS
Preposed Theme/Epithet Epithet
“As for the red, there are two red books,”
In this episode, the two speakers are recalling a series of encyclopedias they had both read in their childhood, and listing the subjects introduced in the series. In this utterance, the speaker is contrasting “red” with “green” just mentioned by the other speaker. Across the utterances, color terms are used as nominal groups in which the Epithet remains and the Head noun 书 (shu, “book”) is omitted.
Circumstances of contingency “specify an element on which the actualization of the process depends” (Halliday and Matthiessen 2014, 323). There are three subtypes of contingency: condition; concession or frustrated cause; and default or negative condition. The three subtypes are often realized by clauses in English and Chinese. However, in conversations, some elliptical forms can also function as conditional, frustrated cause or negative condition, as in (18).
(18) 本科 的话 其实 没 的 选 的。[PKUC005]
Benke dehua qishi mei de xuan de
undergraduate TPART in fact no POSS select POSS
Marked Theme Predicator
“If you are an undergraduate, you cannot choose which language to study.”
In this episode, the two students are talking about whether it is possible for undergraduates to select particular languages as their major, and the speaker is drawing a contrast between what is available to undergraduates and postgraduates. Here the noun 本科 (benke, “undergraduate”) functions as an conditional, specifying the condition under which one cannot make their own choice.
In many Chinese clauses, a Subject which is implied in the context can be omitted, like “undergraduate” in example (18). It never occurs that a marked Theme of a clause simplex or of the main clause in a clause complex is omitted. Therefore, we treat all the elliptical cases where Subjects are not present as tokens of elliptical (unmarked) Theme.
Both Contextual Topical Theme and Marked Topical Theme have a cohesive function in that they link the clause-initial element with other elements of the experiential structure of the clause. Fang (2008) summarized four functions of Topical Themes in Chinese clauses: 1) aboutness function, as in (9) to (13); 2) clause-linking function in a Theme-Rheme chain; 3) providing a frame of time or location, as in (14) and (15); and 4) restricting a domain or range, as in (6) and (8). The four functions apply to both Contextual Theme and Topical Theme.
The connection between Marked Themes and other components of experiential configuration has been thoroughly discussed. However, not enough attention has been paid to their functions in larger discourse segments, e.g., between clauses complexes or across utterances in conversations. The scope of reference of Contextual Topical Theme and Marked Topical Theme often go beyond the clause boundaries into the wider context of discourse. The relationship between the Contextual Theme and other elements in the experiential structure sometimes may not be fully revealed without looking at the context they occur, such as 我们那边 (women neibian, “our place,” “my hometown”) in (15). In systemic functional linguistics, Martin (1992) and Martin and Rose ([2003] 2007) developed a new perspective of looking at the function of resources at the discourse level to see the roles played by certain resources in connecting processes into sequences of activities, connecting participants into identity chains, and connecting small waves of information into bigger waves of HyperTheme and HyperNew, etc. In this study, we will take this perspective to firstly observe the occurrences of Marked Themes in spoken Chinese and then examine their functions beyond the clause.
3. Discourse semantics perspective
The theoretical framework of discourse semantics (Martin 1992; Martin and Rose ([2003] 2007) is informed by systemic functional linguistics and aims at accounting for continuity of meaning beyond the clause. Up till now, six systems have been identified for making meaning across the text: namely, IDEATION, CONJUNCTION, APPRAISAL, NEGOTIATION, IDENTIFICATION, and PERIODICITY. The system of IDEATION consists of resources which realize the field of discourse, focusing on activities and activity sequences. Field is one of the three variables of register, the other two being tenor and mode. Field of discourse refers to the type of social action. Tenor refers to the relationship between speaker and listener. Mode refers to the symbolic organization of discourse (Matthiessen, Teruya, and Lam 2010, 176). The system of CONJUNCTION deals with interconnections between activities or arguments. The system of APPRRAISAL is concerned with the attitudes negotiated in a text. The system of NEGOTIATION is concerned with the speech roles taken and assigned by the speakers, the moves they take and the organization of social interaction as exchange structure. The system of IDENTIFICATION focuses on the tracking of participants through a text. And finally, the system of PERIODICITY deals with the information organization of a text. As the role of Topical Themes is comparatively prominent in the systems of IDEATION, IDENTIFICATION and PERIODICITY, we will briefly introduce these three systems in more detail.
The system of IDEATION is concerned with the experiences encoded in a discourse, what kinds of activities participants are involved in and how these activities are developed into sequences. It deals with two major categories of relations: nuclear relations at clause level; and taxonomical relations in sequences of activities. Nuclear relations refer to the lexical relations between processes, people, things, places, and qualities within each clause. Taxonomical relations refer to the relations between activities and include relations of synonymy, antonymy, contrast, class-member, and co-class, as well as patronymic relations, as shown in Table 1. While at clause level Experiential Themes can be conflated with Participants, Processes, or Circumstances, beyond the clause level, we can use the system of IDEATION to show how Experiential Themes and Contextual Topical Themes are connected with other Participants or Circumstances in sequences of activities. This analysis will be developed in Section 4 below.
The system of IDENTIFICATION is concerned with the textual resources used to track the people, places or things which have been introduced in a discourse. Marked Themes always occur at the clause-initial position and function to connect the current clause with what comes before or after.
PERIODICITY looks at the information flow of discourse. It is concerned with how a clause is organized in a way to prepare its reader/hearer for something new to come, and how this new element shifts into a given to pave the way for something newer to occur in the upcoming discourse (Martin and Rose [2003] 2007). At the clause-initial position, Marked Themes are expected to convey given information. Will this really be the case? How are they connected with other elements in the discourse and what are their contributions to the rhythmic organization of a conversation as a coherent discourse?
4. Marked Themes in the Corpus
The research questions can be reformulated as follows:
How often do Marked Themes occur in natural conversations?
What are their functions in conversations?
To answer these questions, a corpus-based investigation was undertaken. The corpus used in this study is the Peking University Corpus of Spoken Chinese (PKUC henceforth) our team constructed over the past 6 years. The part of the corpus we used here contains audio recordings and transcriptions of 20 conversations each lasting about 5 min. The total time of the audio recordings is 80.95 min. There are altogether 1,381 turns, 2,635 moves,2 and total 33,739 Chinese characters. In most cases, a move is coextensive with a clause. In elliptical utterances, a segment of a clause or a single word, for instance an interjection, may also constitute a move. If a clause complex is uttered in one breath without pause, it will be considered as one move. Most of the conversations occur in relation to daily activities by postgraduate students on campus. A simplified version of the transcription conventions of Eggins and Slade (1997) for extended utterances was adopted and moves within one turn will be marked as (i), (ii) or (iii). The occurrences of Marked Theme and the percentages of these elements among moves and turns are given in Table 2.
Number of occurrences and percentages of Marked Themes.
| Occurrence | Percentage in moves | Percentage in turns | |
|---|---|---|---|
| Contextual Topical Theme | 28 | 1.06% | 2.03% |
| Marked Topical Themes | 149 | 5.65% | 10.79% |
| Marked Themes in Total | 177 | 6.72% | 12.82% |
Data from the calculation show that the occurrences of Marked Themes in the 20 conversations of the corpus are quite low, accounting for only 6.72% of all the moves and 12.82% of all turns. Unmarked Themes occur 1018 times, accounting for 38.63%. Elliptical clauses in which the Unmarked Themes are omitted occur 915 times, making up 34.72%. Minor clauses without a Predicator occur 525 times, making up the remaining 19.92%. To testify whether this is a special case, we compared the current data with two sets of data compiled for previous studies. The first comes from a corpus study of Chinese short stories by Chen and Gao (2010), who collected 60 short stories from 1919 to 1996 and counted the occurrences of Marked Themes (following Chafe 1976, they used the term “Chinese-style topics”) in sentences. Out of a total of 18,754 sentences, Marked Themes occur 645 times, accounting for 3.44% among all the sentences. The other study was conducted by Gao (2018), who compared the occurrences of unmarked Theme and Marked Themes (following Chen and Wang 2010, she used the term “Non-Subject Topics”) in five conversations. She found that among the total number of 706 moves, Subjects occur 360 times, making up 50.99% while Marked Themes occur 29 times, making up 4.11%. Elliptical clauses in which Subjects are omitted occur 231, making up 32.72%. Minor clauses which are realized by a single nominal group, an interjection, or a response marker like 对(dui, yeah) occur 86 times, making up 12.18%.
The distributions of the subcategories of Marked Themes in the present study are given in Table 3. Among the two major categories, the occurrence of Marked Topical Theme is much higher than that of the Contextual Topical Theme. Among Contextual Topical Themes, the occurrence of Absolute Themes is much higher than that of Preposed Themes. Within Marked Topical Themes, Circumstances occur more frequently than Inverted Goal. Among the three subtypes of Circumstances, Circumstances of time occur more frequently than location and contingency.
Distributions of Marked Themes in conversations.
| Marked Theme | Subcategory | Occurrence | Subtotal | Percentage |
|---|---|---|---|---|
| Contextual Topical Theme | Absolute | 23 | 23 | 12.99% |
| Preposed | 5 | 2.82% | ||
| Subject | 3 | |||
| Complement | 1 | |||
| Epithet | 1 | |||
| Marked Topical Theme | Circumstance | 110 | 62.15% | |
| Time | 82 | |||
| Location | 20 | |||
| Contingency | 8 | |||
| Inverted Goala | 39 | 39 | 22.03% | |
| Total | 177 | 177 | 100% |
A preliminary conclusion from this corpus study is that although the percentages of Marked Themes are somewhat higher in conversations than in written texts, they are still quite low in comparison with the occurrences of Unmarked Themes which are conflated with Subject/Actor. Chinese language has been categorized as topic-prominent language in that both the Subject and the Marked Themes (“Chinese-style topic” in Chafe 1976, “non-subject topic” in Chen and Wang 2010) are regarded as topic (Li and Thompson 1981; Shen 2012, etc.). It is on these grounds that Chinese clause structure is viewed as Topic + Comment structure. Data from the present study, as well as the two previous corpus-based studies, pose a challenge to the canonical typology that Chinese belongs to topic-prominent languages so far as we clearly distinguish Subject from non-subject-Topics – Marked Theme in this study. Therefore, we can say that our data from real-time interaction support the claim that Chinese is still mainly Subject-Predicator (or subject-predicate in traditional grammar) or Subject-Verb-Object (SVO) language (“one where word order is determined by or determines the grammatical relations ‘subject’ and ‘object,’” LaPolla 2009, 9). Data of minor clauses support the claims made by Chao (1959; [1968] 2011) and Shen (2012) that minor clauses occur very frequently in spoken Chinese.
When we look at the context in which Marked Themes occur, we realize that the functions of such elements go far beyond the clauses they occur in. Whenever a speaker selects a Marked Theme, they have a good reason for doing so, with the particular context imposing constraints on such choices. In the next section, we examine the functions of Marked Themes in discourse.
5. Functions of Marked Themes in conversations
When discussing the constraints imposed by context on the element which may be chosen as Theme in a clause, Fang (2008) shows how Marked Topical Theme may be used to create patterns of thematic progression in dialogues. In this section, we will demonstrate how in the ongoing conversations, Marked Themes can be used as a way of connecting the current activity with the preceding ones, to keep track of the Participants introduced before, and to organize the information of the turns and across turns within phases3 of discourse.
5.1. Sequences of activities – taxonomical relations
At clause level, social activities are construed as figures which comprise participants, process, and circumstances. “The activity is the social and/or semiotic process that the interactants in the context are engaged in.” (Matthiessen, Teruya, and Lam 2010, 95). Halliday ([1985] 1994) divided social processes into six categories, material, mental, relational, behavioral, verbal, and existential. The core elements in the processes involve the process (doing, sensing, being, behaving, saying, and existing), the participants (Actor, Goal, Senser, Phenomenon, Carrier, Attribute, etc.) and Circumstances. Above the clause level, activities are construed as sequences as text unfolds. In conversations, when talking about daily activities – doings, happenings, beings, etc., speakers may use Marked Theme to connect what they are to say with what has just been said in the preceding utterances. Marked Themes can be simple repetitions whereby the speaker picks up one element from previous utterance and makes it the Marked Theme for the current utterance, as in (19). From now on, relevant elements in conversations, like the repeated ones in (19), are underlined and in boldface.
(19) PKUC00144
T20 B: (i) 那 你 小 时候 基本上 就是, 玩 雪 玩 得 比较
Na ni xiao shihou jibenshang jiushi wan xue wan de bijiao
then you young time basically ADV play snow play ASP comparatively
Circumstance/Time
多 吧？
duo ba
more MPART
“When you were young, you must have had a lot of fun with snow.”
T21 A: (i) 啊, 对, 因为5 小 的 时候, 你 知道 么6,
A, dui, yinwei xiao de shihou, ni zhidao me
ah yes because young POSS time, you know MPART
Marked Theme/Time
‘Ah, yes, because when I was young, you know’
(ii) 小 的 时候 我 竟然 喜欢 冬天。
Xiao de shihou wo jingran xihuan dongtian
Young POSS time I actually like winter
Marked Theme/Time
‘When I was young, I actually loved winter.’
T22 B: (i) 啊 哈！
A ha
ah ha
‘Ah ha!’
T23 A: (i) 因为 小 的 时候, 那样 上学 的话 就 可以
Yinwei xiao de shihou, nayang shangxue de hua jiu keyi
because young POSS time, thus go to school TPART ADV MPART
Marked Theme/Time
打呲溜滑,
daciliuhua
skating
‘Because when I was young, we could skate to school,’
(ii) 就 不用 走路。
Jiu buyong zoulu
ADV NEG walk
‘instead of walking.’
In T21 (i) and T23 (i), the conjunction 因为 (yinwei, “because”) could be interpreted in two ways: as a causal conjunction to connect a subsequent clause with a prior one – for example, in T23 (i), making “I actually loved winter” a main clause and the subsequent clause “because when I was young” dependent on it; alternatively it could also be interpreted as a discourse marker as in T21 (i) and T23 (i) whose function is to connect the subsequent move with one in a previous turn (Zhang 2008). In this case, it seemed to us that the predominant function of 因为 (yinwei, “because”) was to continue the speaker’s turn, and hence the two clauses in T21 (i) and T23 (i) were not taken as subordinate clauses and both instances of 因为 (yinwei, “because”) were analysed as Textual Themes. The nominal group 小的时候 (xiao de shihou, “at a young age”, “when young”) in B’s question in T20 is a Circumstance of time. This Circumstance is repeated as a Marked Themes three times in A’s two turns, T21 and T23, as the time frame for the descriptions of her childhood experiences. By repeating this Circumstance of time in T21 (i), she tried to keep her talk focused on this time frame. In T21 (ii) she explained in general why she had had so much fun with snow when she was young. By keeping this Circumstance of time as Marked Theme again in T23 (i), she gave a more detailed explanation for why she loved winter. Across the moves, speaker A kept her descriptions within the same time frame.
Marked Themes can also be used to build up hyponymic relations across utterances, like those ones in (20) below. At clause level, 球类运动 (qiuleiyundong, “ballgames”) in T4 functions as the preposed Phenomenon of 喜欢 (xihuan, “like”).
(20) PKUC002
T1 VINCENT: (i) 有 什么 喜欢 的 体育运动 吗？
You shenme xihuan de tiyuyundong ma
have what like POSS sports MPART
“Do you like any sports?”
T2 JULIE: (i) 比如 说 … … 呃, 篮球, 乒乓球, 羽毛球 … …
Biru shuo e lanqiu pingpangqiu yumaoqiu
for instance eh basketball table tennis badminton
“For instance, eh, basketball, table tennis, badminton, … ”
T3 VINCENT: (i) 就 看 嘛 … …
Jiu kan ma
ADV look MPART
“Well take this into account ….”
T4 JULIE: (i) 就是 … … 呃, 怎么 说 呢？
Jiushi e zenme shuo ne
ADV MPART how say MPART
“That is … oh, how should I put it?”
(ii) 球类运动 我 都 挺 喜欢 的。
Qiuleiyundong wo dou ting xihuan de
ball games I ADV ADV like MPART
Marked Theme/Phenomenon
“As for ball games, I like (them) all.”
When in (20) T1 Vincent asked Julie what sports she liked, Julie listed several games, including basketball, table tennis, badminton in turn. When Vincent started his turn with an elliptical clause 就看嘛 (jiu kan ma, “Well take this into account”), Julie immediately used her turn to sum up what she had listed in T2 and used “ballgames” to refer both to the individual games she had introduced in her turn and to narrow down the scope of “sports” in Vincent’s question. The inverted Phenomenon/Complement is simultaneously a hypernym of the games Julie listed in her T2 and a hyponym of “sports” in Vincent’s question. Anaphorically it connects the speaker’s current utterance with her own list in T2 and to responding to the addressee’s previous question in T1.
Marked Themes can also be used to form contrast across utterances, as in (21). This is one of the most prominent features of Marked Themes (or Topics) recognized in the literature (Li and Thompson 1981; Xu and Liu 2007).
(21) PKUC003
T31 VINCENT:
(i) 青壮年 的 时候 一定 不 会 在 这边。
Qingzhuangnian de shihou yiding bu hui zai zheibian
young adults POSS time definitely NEG MADV be here
Marked Theme/Circumstance: Time
“Young adults won’t stay here.”
T32 AMELIA: (i) 对 对。
dui dui
yes yes
(Yeah.)
T33 VINCENT: (i) 老 了 可以 去 养老。
Lao le keyi qu yanglao
old ASP MADV go retire
Marked Theme/Circumstance: Time
“When old people can retire there.”
In (21), the two students were talking about the populations of Amelia’s hometown, Xiangyang, Hubei province, China. The two agreed that most young people had left for big cities, and only old people were living there. 老了 (lao le, “having become old”) in T33 is an elliptical form of 老的时候 (lao de shihou, “when getting old,” “at old age”). The two phrases 年轻的时候 (nianqing de shihou, “when young,” “at a young age,” “young adults”) and 老了 (lao le, “having become old,” “in old age”) form a contrast of time, coupling the two utterances in a parallel pattern. There is debate over whether to view Circumstances of time and location as Marked Theme (Topic) or as Unmarked Theme in Chinese clauses. Up till now, most Chinese linguists agree that what are analysed here as Circumstances of time and location in clause-initial position are thematized elements which have been given thematic prominence (Li and Thompson 1981; Zhang and Fang 1994; Yuan 1996; Li 2007), and thus should be viewed as Marked Themes. The two contrasting clauses can be rephrased in normal word order as 人们年轻的时候不会在这儿 (renmen nianqing de shihou buhui zai zheer, “people at young age will not be here”) and人们老了以后可以回来养老 (renmen laole yihou keyi huilai yanglao, “people in their old age can come back here to retire”).
Repetition, hyponym, and contrast are three subtypes of taxonomical relations frequently deployed by speakers to link sequences of activities across utterances. Eighty-six out of the total of 177 Marked Themes in the corpus data are used to build up taxonomical relations across utterances, among which lexical repetition and hyponymic relations occur 72 times, making 40.68% of the total and contrast of time and contingency occur 14 times, accounting for 7.91% of the total.
5.2. Identity chains
The functions of Marked Theme in constructing identity chains go together with its functions in building up taxonomic relations. When used in repetition, synonymy, antonymy, contrast, co-class and co-part, Marked Themes help keep track of Participants introduced previously in the text. In (22), the two speakers are talking about two stadiums on campus. When Vincent asked Julie which stadium she had preferred, Julie said that she had only played once in Khoo Teck Puat. Then, Julie wanted to know which one Vincent thought was better. Vincent briefly commented that both were okay but at the May 4th Stadium people could not borrow badminton rackets. Julie continues the topic by starting with “Khoo Teck Puat” as Marked Theme.
(22) PKUC002
T22 JULIE:
(i) 我 只 去 过 邱德拔,
Wo zhi qu guo qiudeba,
I ADV go ASP Khoo Teck Puat
‘I have only been to Khoo Teck Puat,’
(ii) 五四 没有 去 过。
wusi meiyou qu guo
May 4th NEG go ASP
Marked Theme/Scope/Complement
‘the May 4th Stadium I haven’t been to.’
T23 VINCENT:
(i) 我 当时 为了 带队 训练,
Wo dangshi weile daidui xunlian,
I then for lead team training
我 这 两 个 球馆 的 卡 都 订 了 一 张。
wo zhe liang ge qiuguan de ka dou ding le yi zhang
I this two MEAS stadium POSS card both bought ASP one MEAS
‘To lead my team for training, I bought a card for each of them.’
T24 JULIE: (i) 你 对 它们 俩 有 什么 评价？
Ni dui tamen lia you shenme pingjia
you PREP them two have what comment
‘What do you think of the two?’
T25 VINCENT: (i) 呃 … … 还 好 吧 … …
E hai hao ba
eh ADV good MPART
‘Eh … both are ok.’
(ii) 就是 五四 不 能 借 拍子。
Jiushi wusi bu neng jie paizi
ADV the May 4th NEG MADV borrow racket
Marked Theme/Location
‘It’s just that at the May 4th you cannot borrow (badminton) rackets.’
…
T28 JULIE: (i) 邱德拔 其实 我 去 玩 的 那 唯一 一 次 就是 … …
Qiudeba qishi wo qu wan de na weiyi yi ci jiushi
Khoo Teck Puat actually I go play POSS that only one MEAS ADV
Marked Theme/Location
‘Khoo Teck Puat actually I have only been (there) once, that is … ’
(ii) 哦 不, 其实 是 两 次。
E bu qishi shi liang ci
eh NEG actually be two MEAS
‘Oh, no, actually twice.’
As can be seen in (22) above, in T22 (ii) Julie introduced 五四 (wusi, “the May 4th Stadium”) in Marked Theme (MT) position to form a contrast with her mention of 邱德拔 (Khoo Teck Puat) at the end of the preceding clause in T22 (i). Following that, in T24 an anaphoric pronoun它们俩 (ta men lia, “the two of them”) is used to keep track of the identities of the two stadiums. Then Marked Themes are used in T25 and T28 referring back to the two stadiums by their respective names. This clearly shows how an identity chain of the two stadiums are formed sequentially as the conversation proceeds:
Two nominal groups: 1Khoo + MT: 2Wusi —>
anaphoric pronoun: tamenlia ‘the two of them’—>
MT: nominal group referring to 2Wusi —>
MT: nominal group referring to 1 Khoo
As seen in the above sequence of identity chains, the Marked Themes in (22) serve to 1) make a contrast with the proceeding element in the discourse; 2) relate the element to one of those in the proceeding discourse; 3) refer back to a topic that has been discussed earlier. The Marked Themes function here to distribute the names and other expressions referring to them in a way that allow interlocutors to keep track of them through the discourse.
5.3. Marked Theme as Hyper-Theme
Marked Themes mostly convey given information, as their referents are retrievable either in the previous utterance, in the speech context (as with the personal or demonstrative pronouns) or in the shared knowledge base of the participants.
Marked Themes expressing repetition, hyponymy or contrast, while building up taxonomic relations, convey given information which provides a background for new information to be added. However, Absolute Themes, Preposed Themes and Circumstances of Time and Location, and Inverted Goal can also be used to introduce new settings for new activities, as (23) shows.
(23) PKUC010
T5 L: (ii) 他们 叫 对方 的 时候 会 叫 比如 许 先生。
Tamen jiao duifang de shihou hui jiao biru xu xiansheng
They call other POSS time MPART call for instance Xu mister
‘When they address each other, they would use for example Mr./Master Xu.’
T6 C: (i) 诶, 我们 那边 是 … …
Ai, women neibian shi
ah, we there be
‘Eh, in my home town,
(ii) 那个 老 婆婆, 老 奶奶 丈夫 去世 很久 之后,
neige lao popo lao nainai zhangfu qushi henjiu zhihou
that elderly grandma old grannie husband died long after
Preposed Theme/Value/Complement
long after an elderly woman’s husband had passed away’
(iii) 我们 也 会 用 一 个 男性 的 称呼 来 称呼 她,
women ye hui yong yi ge nanxing de chenghu lai chenghu ta
we also MADV use one MEAS male POSS address term come address her
Value/Complement
‘We would still use a male’s address term to address her.’
This is a beginning of a conversation on address terms used for elderly women. One special tradition in China is to address a woman with an honorific address term for man, like 许先生 (xu xiansheng, “Mr./Master Xu”). This tradition started from as early as the beginning of the 20th century during the Republican period when educated men and women fought for the equal status of women in opposition to the feudal tradition. Even nowadays, people still use 先生 (xiansheng, “Master”) following the family name to address elderly women to show the greatest respect, especially when the woman is a distinguished scholar herself, like the famous writer and translator Yang Jiang, wife of famous writer Qian Zhongshu.
At the beginning of her turn, C in (23) recalled a tradition in her hometown where elderly widows were addressed using the address terms of their late husbands. The generic “elderly grandma” is a Preposed Theme which is resumed by the anaphorical pronoun 她 (ta, “her”) as Complement near the end of the clause. Before this turn, the two speakers were talking about a movie The Golden Age in which the couple addressed each other with 先生 (xiansheng, “Master”). With the Preposed Theme, C shifts the setting from the movie to her hometown where a similar tradition is still retained among elderly people, thus starting a new phase. A phase, according to Martin and Rose ([2003] 2007, 102) is a text segment where consistency of field, tenor and mode is maintained, “within each phase we would expect activities to be related, as members of a wider set of activities, or as sub-parts of larger activities.” In conversations, a phase may be coextensive with an utterance, and sometimes may extend beyond one speaker’s utterance, depending on whether the same subject matter is being discussed. The Theme ^ Rheme structure in a phase is reformulated as HyperTheme ^ HyperNew. A HyperTheme corresponds to a topic sentence in written text, and a phase is usually, but not always, coextensive with a paragraph.
Like asides in drama, Absolute Themes and Preposed Themes serve as the point of departure for episodes whose function is to explain new topics the speakers have just brought in, or to introduce a new phase. A majority of Absolute Themes (14 out of 23), Preposed Themes (3 out of 5) and a significant proportion of inverted Goals (12 out of 39) function as HyperThemes of new phases, making up 16.38% of the Marked Themes. Circumstances of time and location, when setting frames for clauses, can also serve as new settings for new phases. Eleven out of 82 Circumstances of time and 15 out of 20 Circumstances of location are deployed to introduce new phases of activities in conversations. Altogether 55 out of 177 are used as HyperThemes, making up 31.07% of the total Marked Themes. However, out of the 177 Marked Themes, none of them turn out to be the MacroTheme of the whole conversation. Evidence from this corpus of daily conversations would suggest that Marked Themes mainly function in the local utterances or across utterances, but not as global topics.
6. Conclusion
Most Marked Themes, Absolute and Preposed Themes, in particular, when looked at from the structuring of the clause, appear to be outside the experiential or interpersonal structure, i.e., not directly involved as Participants in processes of activities, or related to the Predicator of the Subject + Predicator configuration. However, when examined at discourse level, they play importance cohesive roles in different discourse semantic systems.
Our analysis of the data from the corpus supports the claim made by previous corpse-based studies (Chen and Gao 2010) that Chinese may not be as topic-prominent as presumed to be so far as we clearly distinguish Unmarked Theme (Theme/Subject conflation) from Marked Theme (Chafe’s Chinese-style topic). Our study indicates that the frequency of Marked Themes across all moves in daily conversations, though somewhat higher than that in written texts, is still quite low, at less than 10%.
With regard to the motivation for the use of Marked Theme, we found that it has to do with the necessity of cohesion in conversations. Unlike Marked Themes in written discourse which often function to scaffold discontinuity, i.e., to “signal new phases in a discourse: a new setting in time or a shift in major participants” (Martin and Rose [2003] 2007, 192), Marked Themes which are simultaneously used to build up taxonomical relations are significant cohesive devices used to connect sequences of activities across moves and turns. Across different turns in utterances, lexical items functioning as Marked Themes perform key functions in the construction of taxonomical relations in sequences of activities, linking Participants and Circumstances in ongoing events. In the system of IDENTIFICATION, Marked Themes may also be deployed to track the identities of the participants or places simultaneously with their roles in sequences of activities, thus building up identity chains across utterances. In the system of PERIODICITY, Contextual Themes, including Absolute Themes and Preposed Themes and Marked Topical Themes, including in particular inverted Goals and Circumstance of location, frequently function as HyperTheme of an extended stretch of conversation, opening new phases to provide background information or settings for new events.
In future, we hope to extend our study to a wider range of daily conversations in Chinese with the aim of finding distinctive patterns in the use of Marked Themes in different registers. We will also plan to carry out a comparative study on the use and functions of Marked Themes across different languages within similar registers.
About the authors
Yanmei Gao is associate professor at Peking University, Beijing, China. Her research interests include systemic functional linguistics, lexical semantics and discourse semantics. Recently she is working on stance-taking and exchange structures in face-to-face interaction across languages.
Guoyan Lyu is associate professor in the School of Foreign Studies at Beijing Information Science & Technology University. Guoyan is interested in exploring the functional dimensions of language and syntax-semantics relations in world languages.
Notes
- 1.
Li and Thompson (1981, 94) think that time phrases and locative phrases occurring in clause-initial position should also be considered as Topics – Marked Topical Theme in this study.
- 2.
Following Martin (1992) and Eggins and Slade (1997), the move is defined with reference to clause. In most cases, a move corresponds with a clause. Matthiessen (1995, 434) considered the move as the “basic dialogue unit,” “the contribution an interactant makes to the development of the dialogue.” He thinks that a move “selects in the system of SPEECH FUNCTION for a type of interact, where the speaker adopts a speech functional role and assigns the addressee a complementary role.”
- 3.
Phase is a term adopted by Martin and Rose ([2003] 2007) and Rose (2006) as an intermediate semantic unit between stages and sequences. The boundaries of a phase are delimited by the consistency of field, tenor and mode. For more detailed discussion, see Martin and Rose ([2003] 2007) and Gao (2012).
- 4.
Following the transcription conventions of Eggins and Slade (1997), the number before the speaker is turn; here T3 refers to the third turn in the episode. Names in capital are the pseudo names of interlocutors. (i), (ii), etc., mark the moves in each turn. Each move marks a single unit of utterance, in most cases corresponding to a clause in English.
- 5.
The function of the causal conjunction 因为 (yinwei) in conversations has been debated over the past two decades (Biq 1995; Shen 2003; Song and Tao 2008, 2009). While still used as a causal connective, it has developed some new functions beyond connecting two clauses within clause complexes. Across utterances, occurring at the beginning of a new turn or in mid-turn, it functions as a discourse marker whose main function is to link the current utterance with one’s previous utterance which has been cut short by other interlocutors.
- 6.
Acknowledgments
The authors are grateful to the anonymous reviewers for their critical comments.
Disclosure statement
No potential conflict of interest was reported by the author(s).
References
Biq Y.-O. 1995 . “ Chinese Causal Sequencing and Yinwei in Conversation and Press Reportage .” Annual Meeting of the Berkeley Linguistics Society 21 ( 2 ): 47 – 60 . doi: 10.3765/bls.v21i2.1374 . Search in Google Scholar
Caffarel A. 2006 . A Systemic Functional Grammar of French: From Grammar to Discourse . London : Continuum . Search in Google Scholar
Chafe W. 1976 . “ Giveness, Contrastiveness, Definiteness, Subject, Topic, and Point of View .” In Subject and Topic , edited by C. N. Li , 25 – 55 . New York : Academic Press . Search in Google Scholar
Chao Y. R. [1968] 2011 . A Grammar of Spoken Chinese . Reprint . Beijing : Commercial Press . Search in Google Scholar
Chao Y. R. 1959 . “ How Chinese Logic Operates .” Anthropological Linguistics 1 ( 1 ): 1 – 8 . Search in Google Scholar
Chen P. 1996 . “ Pragmatic Interpretations of Structural Topics and Relativization in Chinese .” Journal of Pragmatics 26 ( 3 ): 389 – 406 . doi: 10.1016/0378-2166(95)00042-9 . Search in Google Scholar
Eggins S. , and D.Slade . 1997 . Analysing Casual Conversation . London : Cassell . Search in Google Scholar
Fang Y. 2008 . “ A Study of Topical Theme in Chinese: An SFL Perspective .” In Meaning in Context: Implementing Intelligent Application of Language Studies , edited by J. J. Webster , 84 – 114 . London : Continuum . Search in Google Scholar
Fang Y. 2019 . “ On Topical Theme in Chinese .” Contemporary Rhetoric 2 : 11 – 27 . Search in Google Scholar
Gao Y. 2012 . “ Defining ‘Phase’ in Discourse Semantics .” Contemporary Foreign Languages Studies 3 : 137 – 149 . Search in Google Scholar
Gao Y. 2018 . “ Dialogic Resonance and Topic Structure .” The Third International Conference on Interactional Linguistics and Chinese Language Studies , 23 08 2014 . Beijing : Beijing Language and Culture University . Search in Google Scholar
Chen, G. and J.Wang . 2010 . “ Unmarked Non-subject Topics in Chinese .” Chinese Teaching in the World 24 ( 3 ): 310 – 323 . Search in Google Scholar
Halliday M. A. K. [1985] 1994 . An Introduction to Functional Grammar . 2nd ed. London : Arnold . Search in Google Scholar
Halliday M. A. K. , and E.McDonald . 2004 . “ Metafunctional Profile of the Grammar of Chinese .” In Language Typology: A Functional Perspective , edited by A. Caffarel , J. R.Martin, and C. M. I. M.Matthiessen , 305 – 396 . Amsterdam : Benjamins . 10.1075/cilt.253.08halSearch in Google Scholar
Halliday M. A. K. , and C. M. I. M.Matthiessen . 2014 . Halliday’s Introduction to Functional Grammar . London : Routledge . 10.4324/9780203783771Search in Google Scholar
Jing. C. , and Y.Gao . 2000 . “ Is Chinese Topic-prominent Language? ” Foreign Languages and Their Teaching 5 : 11 – 14 . Search in Google Scholar
LaPolla R. 1995 . “ Pragmatic Relations and Word Order in Chinese ”. In Word Order in Discourse , edited by Pamela Downing and Michael Noonan, 299 – 331 . Amsterdam : John Benjamins . 10.1075/tsl.30.11lapSearch in Google Scholar
LaPolla R. 2009 . “ Chinese as a Topic-Comment (Not Topic-Prominent and Not SVO) Language .” In Studies of Chinese Linguistics: Functional Approaches , edited by J. Xing , 9 – 22 . Hong Kong : Hong Kong University Press . Search in Google Scholar
Li C. N. , and S.Thompson . 1976 . “ Subject and Topic: A Typology of Language .” In Subject and Topic , edited by C. N. Li , 457 – 489 . New York : Academic Press . Search in Google Scholar
Li C. N. , and S. A.Thompson . 1981 . Mandarin Chinese: A Functional Reference Grammar . Berkeley : University of California Press . Search in Google Scholar
Li E. S.-H. 2007 . A Systemic Functional Grammar of Chinese: A Text-Based Analysis . London : Continuum . Search in Google Scholar
Martin J. R. 1992 . English Text: System and Structure . Amsterdam : John Benjamins . 10.1075/z.59Search in Google Scholar
Martin J. R. , and D.Rose . [2003] 2007 . Working with Discourse: Meaning beyond the Clause . 2nd ed. London : Continuum . Search in Google Scholar
Matthiessen C. M. I. M. 1995 . Lexicogrammatical Cartography: English Systems . Tokyo : International Language Sciences Publishers . Search in Google Scholar
Matthiessen C. M. I. M. , K.Teruya, and M.Lam . 2010 . Key Terms in Systemic Functional Linguistics . London : Continuum . Search in Google Scholar
McDonald E. 1998 . “ Clause and Verbal Group Systems in Chinese: A Text-based Functional Approach .” PhD thesis . Sydney : Macquarie University . Search in Google Scholar
Östman J. 1981 . You Know. A Discourse-Functional Approach . Amsterdam : John Benjamins . Search in Google Scholar
Pan H. , and J. Hu. 2008 . “ A Semantic-pragmatic Interface Account of (Dangling) Topics in Mandarin Chinese .” Journal of Pragmatics 40 ( 11 ): 1966 – 1981 . doi: 10.1016/j.pragma.2008.03.005 . Search in Google Scholar
Rose D . 2006 . “ Reading Genre: A New Wave of Analysis .” Linguistics and The Human Sciences, 2 ( 2 ): 185 –204 . doi:10.1558/lhs.v2i2.185. 10.1558/lhs.v2i2.185Search in Google Scholar
Schourup L. C. 1985 . Common Discourse Particles in English Conversation . New York : Garland . Search in Google Scholar
Shen J. 2003 . “ Fuju Sanyu “Xing, Zhi, Yan” (Compound Sentences in Three Conceptual Domains: Acting, Knowing, and Uttering) .” Zhongguo Yuwen 3 : 195 – 204 . Search in Google Scholar
Shen J. 2012 . “ On Minor Sentence and Flowing Sentences in Chinese: In Commemoration of the 120th Birthday of Yuen Rem Chao .” Studies of the Chinese Language 5 : 403 – 415 . Search in Google Scholar
Song Z. , and H.Tao . 2008 . “ Hanyu Yinguo Fuju Shunxu De Huayu Fenxi Yu Bijiao (Discourse Analysis and Comparison of Causal Clauses in Chinese and English) .” Hanyu Xuebao 4 : 61 – 71 . Search in Google Scholar
Song Z. , and H.Tao . 2009 . “ A Unified Account of Causal Clause Sequences in Mandarin Chinese and Its Implications .” Studies in Language 33 ( 1 ): 69 – 102 . doi: 10.1075/sl.33.1.04son . Search in Google Scholar
Thompson G. 2014 . Introducing Functional Grammar . London : Routledge . 10.4324/9780203785270Search in Google Scholar
Xu L. , and D.Liu . 2007 . Topic: Structural and Functional Analysis . Shanghai : Shanghai Education Press . Search in Google Scholar
Yu W. 2002 . “ A Comparative Study of Marked Themes in Academic Genres in Chinese and English .” Foreign Languages and Their Teaching 154 ( 1 ): 8 – 18 . Search in Google Scholar
Yuan Y. 1996 . “ Topicalization and Grammatical Relations .” Studies of the Chinese Language 4 : 241 – 254 . Search in Google Scholar
Zhang B. , and M.Fang . 1994 . “ The Thematic Structure of Spoken Chinese .” Journal of Peking University (Philosophy and Social Sciences) 2 : 66 – 75 . Search in Google Scholar
Zhang C. 2008 . “ Huayubiaoji Ni Zhidao Ma (Discourse Marker ‘You Know’) .” Journal of Harbin University 29 ( 11 ): 85 – 88 . Search in Google Scholar
Zhu D. 1982 . Course Notes on Grammar . Beijing : Commercial Press . Search in Google Scholar
Zhu D. 1999 . Questions and Answers on Grammar . Beijing : Commercial Press . Search in Google Scholar
© 2020 Informa UK Limited, trading as Taylor & Francis Group
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
