國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
14
CHAPTER 2 LITERATURE REVIEW
An inquiry into personal pronoun use in students‘ texts needs to draw on some theoretical constructs, empirical results and pedagogical applications that have been established for the past few decades. The aim of this chapter, therefore, is to review the role of personal pronouns in the interactive writing paradigm, and also to present a conceptual framework in which personal pronouns are grounded. First, the view of writing as a form of interaction is provided, focusing on two essential
features—writer‘s voice and audience awareness. Then, the pragmatic functions of personal pronouns will be closely reviewed, followed by empirical studies on personal pronoun use in academic and school writing, oral discourses, and other genres. Next, the argumentative genre required of the study is introduced, and the measurements of writing quality are reviewed. Finally, the theoretical construct of Systemic Functional Grammar, from which the metadiscourse theory is developed, will be presented. The chapter will conclude with the approach the present study adopts.
Writing as a Form of Interaction
For the past few decades, writing paradigm has shifted from text-oriented approach to writer-oriented approach and currently to social constructivism approach (Matsuda, 2003). The traditional text-oriented approach in the 1960s focuses on the forms and written products. In the 1970s, influenced by L1 writing research on composing process, a writer-oriented approach was developed. It describes writing as an expressive and cognitive process, with an emphasis on the cycling procedures of revising, editing, and feedback given both by the teacher and peers (Flower & Hayes,
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
1981; Frodesen & Holten, 2003; Grabe & Kaplan, 1996; Johns, 2003). A more recent trend sees writing as a social activity situated in a culturally defined context. The transitions to the social stage of writing in 1980s and discourse community stage in 1990s point to the fact that there is an increasing stress on discourse communities and the role of social construction, especially in academic and professional contexts (Grabe & Kaplan, 1996; Johns, 1990). The social context defines the meaning and purpose of writing, and thereby confines the writing conventions a writer has to follow. The writer, with his intentions to convey, is expected to maneuver the
messages in a conventionally-accepted fashion and more importantly, to build rapport in response to the needs of the reader in the defined discourse (Hyland, 2002a). The appropriateness and effectiveness of messages are hence determined by the extent to which the writer balances his purposes with the reader‘s potential responses and cultural frames.
Simply put, writing is not just the expression of the writer‘s personal experiences and propositions; it also performs an interpersonal function in that it maintains the expected relationship between the writer and the reader, and even with the members in a broader cultural and discourse community. In other words, writing is an act of negotiation between the writer and the reader for achieving a certain
common communicative purpose. It is seen as a dialogic nature of interaction—a form of social communication (Beaugrande and Dressler, 1981; Carrell, 1987;
Thompson, 2001).
Writer’s Voice
The concept of ―textual interaction‖ (Kim, 2009) is demonstrated by two major features in writing: (1) writer‘s voice and (2) audience awareness (Hyland, 2005b).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Voice, although diversely defined, is strongly associated with the writer‘s essential inner self and personal identity (Matsuda & Tardy, 2007). Ramanathan & Atkinson (1999) further link voice to the ―ideology of individualism,‖ and define the notion as a
―linguistic behavior which is clear, overt, expressive and even assertive and
demonstrative‖ (p.48), indicating the writer‘s unique and distinctive personal authority.
By explicitly and clearly voicing his opinions and evaluations, the writer constructs credible representation of himself and his arguments, making himself recognized by members of his discourse community (Hyland, 2002b).
Authorial voice in writing has been mainly expressed through the linguistic markers of stance (Breeze, 2007), which convey how the writer relates himself to the content message, both personally and socially, and how he emotionally interacts with his audience—distantly or intimately (Reilly, et al., 2005). For example, personal pronoun I highlights the writer‘s presence in discourse. The use of hedging device (e.g.
perhaps, someone, anything) can decrease the writer‘s responsibility for the truth value of claims and display his hesitation, uncertainty or indirectness (Crismore &
Vande Kopple, 1988; Crompton, 1997; Hinkel, 2005; Salager-Meyer, 1994).
Conversely, intensifiers or emphatics (e.g. will, must, clearly, always) can imply certainty or emphasize the force of accompanying propositions (Brown & Levinson, 1987; Hinkel, 2005; Lau, 2004; Levinson, 1987). Attitude markers (e.g. agree, prefer, essential, unfortunately) reflect the writer‘s affective attitude to propositions (Hyland, 2005a). These rhetorical devices altogether express the writer‘s personal voice with a strengthening or weakening force (Dafouz-Milne, 2008).
Audience Awareness
Of equal importance attached to effective interactive writing is the notion of audience awareness—how the writer connects to his reader by recognizing his
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
presence and engaging him in discourse (Ramanathan & Kaplan, 1996). In fact, the writer‘s self-presentation hinges greatly on his awareness of audience. The
recognition of audience presence marks the distinction of what Bereiter and Scardamalia (as cited in Grabe & Kaplan, 1996) term ―knowledge telling‖ and
―knowledge transforming‖ in their writing process model. Less mature writers tend to focus on the topic alone and strive for self-expression. The ideas that they express are retrieved from their long-term memory and transferred directly into the written text.
However, expert writers take both their writing purposes and audience into account.
The ideas retrieved from memory are transformed by their efforts to resolve the conflict between their ideas and the rhetorical goal. Therefore, they are more capable of creating an internal image of the reader, responding to the audience‘s expectations and objections, and taking adaptive moves to present more persuasive arguments (Zainuddin & Moore, 2003).
As with authorial voice, audience awareness in the text can be overtly signaled by some linguistic forms, such as reader pronoun you, questions and directives (Hyland, 2005b, 2005c). Pronoun you explicitly marks the reader‘s presence, and questions have a direct appeal in bringing the reader into a dialogue. The writer could either challenge the reader into thinking about the topic or encourage the reader to accept the direction the text is taking (Thompson, 2001). Questions can also function as a distancing and hedging technique or serve to refute other authors or theories (Webber, 1994, p.266). Directives, also contributing to the dialogic dimension, instruct the reader what to see and are often accompanied by obligation modals (Hyland, 2005b).
The Role of Voice and Audience in Writing Classroom
Both writer‘s voice and audience awareness are recognized as significant
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
indicators of good writing (Carvalho, 2002; Cheng, 2005; Krause & O‘Brien, 1999;
Thompson, 2001). An overt marking of writer‘s voice and individuality facilitates the writer‘s credibility, and appealing to the reader‘s views and emotions contributes to the construction of mutual bond (Crismore, et al., 1993). Hines‘ study (2004) clearly points out good-quality papers tend to have a higher degree of effective use of voice.
For example, the personal voice of I enables the writer to show the relevance of the propositions to his personal opinions and attitudes, and we includes the reader as a member of a discourse community and as a friend, making the writer sound closer and intimate. On the other hand, Hyland (2001) posits that successful academic writing in English incorporates a good awareness of audience. By appealing to the reader‘s empathy, well-beings, concerns, and values, the writer attaches an interpersonal tone to the text, and the overall quality can thus be improved (Hines, 2004).
It has been argued that self-expressive voice and audience-related strategies should be explicitly modeled and taught. As a consequence, many American universities are currently teaching students to address the audience and to express their own voice in writing classroom in order to sound persuasive to the intended reader (Hines, 2004; Ramanathan & Atkinson, 1999). Unfortunately, the two notions are culture-bound, and they may not translate themselves automatically into L2 writing (Ramanathan, & Kaplan, 1996; Zainuddin & Moore‘s, 2003). Most EFL students, especially those from collectivistic cultures, do not usually write to express themselves but to become integrated into a scholarly community most of the time.
They tend to say what they believe will not disturb the group or threaten the positive faces of their peers (Ramanathan & Atkinson, 1999). Given this, EFL students often find it difficult to create a new self—a confident, assertive and distinctive one. They do not know when to intrude their personal assertions during the interactive process (Duenas, 2007; Hyland, 2002b). Nor do they have a clear sense of audience and the
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
way to recognize the reader‘s counter-positions (Cheng & Steffensen, 1996). For example, the majority of Hong Kong L2 students in Krause & O‘Brien‘s (1999) study failed to have a dialogic talk with the reader due to their false assumption of cultural commonality with the reader.
In response to the problem, Ramanathan & Kaplan (1996) propose a discipline-oriented approach, arguing that if students are aware of the writing conventions that are unique in their chosen disciplines, they will be capable of expressing their voice and attending to the reader‘s needs more appropriately. Other researchers (e.g. Berkenkotter, 1984; Hays, et al., 1988; Hyland, 2005a; Midgette et al., 2008) also claim that a carefully designed and sequential instruction could sensitize learners to their own voice and the audience they are addressing, which would improve the overall quality of the essay writing.
Personal Pronouns Overview
As has been reviewed above, writer‘s voice and audience awareness can be both overtly reflected in personal pronouns used in texts. Personal pronouns are defined as
―items used to refer to the speaker-writer (I), the addressee (you) and other person and objects whose references are presumed to be clear from the context (he, she, it, they)‖
(Hell, et al., 2005, p.242). They are central to face-to-face interaction, and able to help the writer state personal opinions, acknowledge claims, and guide the reader through the arguments (Harwood, 2007). Thompson and Thetela (1995, p.108) argue that personal pronouns are employed as ―projected roles‖ which function as the textual personae of the intended writer and reader. For example, the first and second person pronouns reflect how the reader is conceptualized by the writer and the degree of
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
rapport the writer intends to establish with the reader (Kim, 2009). The implied social meanings of self-projection and reader-conception vary in accordance with the
discourse contexts where they occur.
Although there is evidence that personal pronouns in general are sparingly used in academic writing due to the dominant values of critical objectivity or scientific neutrality (Biber, et al., 1999), there is growing acknowledgement that personal pronouns play an important role in constructing the relations with the reader and the research community, especially in soft disciplines such as humanities and social sciences (Breeze, 2007; Kuo, 1999). As a matter of fact, ―personal pronouns are at the intersection of the grammatical and pragmatic subsystems of language‖ (Rounds, 1987, p. 14). Apart from their cohesive function in text discourse, personal pronouns have multiple semantic references and are polypragmatic, especially in the
construction of interpersonal relationship (Fetzer & Bull, 2008; Lau, 2004; Rounds, 1987). When the writer seeks to represent himself, he is also defining the others in a close or distant manner, and revealing the cultural and discourse community he belongs to. Implicit in this view is that personal pronouns embody the assumptions of the writer made about himself, the other(s), the cultural context, and discourse
community.
Pennycook (1994) makes the point that from discursive perspectives, all personal pronouns are ―political‖ in that they are strategically contrived to represent either the writer or the others. The choice of a specific personal pronoun reflects the writer‘s egocentricity or solidarity, sympathy or indifference, involvement or
detachment in the discourse (Muhlhausler & Harre, 1990; Wales, 1996). In other words, personal pronouns can be used to indicate authority or rejection but also express modesty or acceptance, like the two sides of a coin. The selection of personal pronouns is not determined so much by grammatical concerns as by the
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
sociolinguistic and pragmatic/rhetorical considerations.
Approaches to Analyzing Personal Pronouns
Personal pronouns have been examined from several approaches: (1) grammatical approach, (2) textual/endophoric approach, (3) deictic/discourse
approach, and (4) pragmatic/sociolinguistic approach (Tian, 2001a; Wales, 1996). The grammatical approach is a linguistic one; for example, we refers to more than two first persons, and he names a third person singular. The textual approach identifies
personal pronouns anaphorically and cataphorically for the textual cohesion. Other research views personal pronouns from deixis—a reference point whose interpretation is determined by the context of the utterance, i.e. time or space, for example, the first person in self-reference, the second person in addressee-reference, and the third person in other-reference. The pragmatic approach stems from interactional
pragmatics and sees personal pronouns as interpersonal markers. A pragmatic use of personal pronouns can convey the interactants‘ power, egocentricity, objectivity or solidarity, make generalizations, and realize other interpersonal functions.
Although personal pronouns are typically referential and deictic—a canonical and unmarked use, they may be non-referential or impersonal, where no specific person is identified. Kitagawa & Lehrer (1990) assert that impersonal pronouns (e.g.
you and we), often interchangeable with everyone or one, can be used to describe structural knowledge and universal events in everyday life. The shifts between referential and impersonal uses can be regarded as rhetorical or pragmatic strategies for fulfilling some specific communicative functions as the discourse flows.
Currently, there are increasing interests in the pragmatic approach to personal pronouns. Research has attempted to explore the discourse functions performed by types of personal pronouns in various disciplines and genres, especially academic
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
research papers and scientific writing (e.g. Fortanet, 2004; Hyland 2001a, 2001b;
Inigo-Mora, 2004; Kuo, 1999; Luzon, 2009; Tang and John, 1999; Vassileva, 1998).
For example, Hyland (2001b) points out that self-mention in research articles highlights the authors‘ contribution in a field. Kuo (1999) identifies some major discourse functions of we in scientific journals, such as explaining what is done, stating a goal, showing results, justifying a proposition, etc.
Some studies (e.g. Biber, et al., 1999; Dafouz, et al., 2007, Flottum, et al., 2006, Harwood, 2005b) have further explored the discourse stance as expressed by personal pronouns with their collocated linguistic forms. It is argued that personal pronouns cannot be studied alone; the co-text (co-occurring linguistic features) that
accompanies them have to be investigated simultaneously. Reilly, et al. (2005) contend that it is the constellation of forms, including the morphological, lexical, syntactic, and discourse levels, that dynamically but subtly characterizes the writer‘
stance. In addition to research on pragmatic use of English pronouns, other studies (e.g. Kuo, 2002, 2003; Lin, 1993; Martinez, 2005) have compared personal pronouns in different languages in terms of frequency distribution and discourse functions.
Pragmatic Functions of Personal Pronouns
As mentioned earlier, personal pronouns are polyvalent and potentially fulfill a wide range of communicative functions. In the following, the various pragmatic meanings of different personal pronouns will be reviewed in depth:
First Person Singular Pronoun I
Self-mention pronoun I is the most common resource to represent authorial self and denote an ego. By placing himself in the deictic center, the writer can express his very subjective, personalized views and affective attitudes. He can also establish his credibility and confidence by including his personal experiences (Baumgarten &
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
House, 2009; Berman, 2004; Kuo, 1999; Muhlhausler & Harre, 1990). I, being a writer-oriented perspective, is specific in reference and displays the writer‘s strong commitment to his claims and responsibility for the arguments (Hyland, 2002b;
Muhlhausler & Harre, 1990). Moreover, I enables the writer to intrude himself into the discourse whenever desirable so that he can elaborate his argument, strengthen his argument, or promote his contributions. While expressing self, the writer also gets the reader involved in the discourse (Berman, 2004) by constructing the others as they (Pennycook, 1994), and contextualizing the framework for the reader to follow (Hyland, 2001a; 2001b; 2002b; Kuo, 1999).
In addition to its referential function, I can be used as an impersonal pronoun in discourse situation when structural knowledge and general truth are involved
(Kitagawa and Lehrer, 1990). Impersonal I is quite limited in its distribution, mostly in a hypothetical context where the writer offers himself as role model, describing how the world he presents woks. For example, in the sentence ―Thus, in order to be able to take the subway in New York I simply need a ‗taking a subway‘ script or frame, if I have one‖ (ibid., p.741-742), I is used in an impersonal sense, which can be
replaced by formal one.
Although I often acts as a ―foregrounding‖ device to place the writer as an explicit, active creator and guider, it may also function as a hedge, or a politeness marker, to avoid arousing the writer‘s strong and probably opposing arguments (Myers, 1989). For instance, the hedging phrases ―I think‖ and ―I feel‖ leave space to the reader for interpretation of the claim (Hardwood, 2005b, 2007). The purpose of downplaying is to attain objectivity and also maintain a harmonious relationship between the writer and the reader (Hyland, 2005c).
First Person Pronoun Plural We
The other self-mention pronoun is plural we. We in English can be employed to
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
refer to the writer, one or more than one reader, non-participants and the general public (Muhlhausler & Harre, 1990; Quirk et al., 1985). Due to its potentially diverse references, the meaning of we is often vague and can vary in the same context (Biber, et al., 1999).
We implies both senses of exclusion and inclusion, and can be used to solicit rejection or solidarity (Myers, 1992; Pennycook, 1994). Exclusive we constructs a we/you or we/they dichotomy (Harwood, 2007), which distinguishes the writer‘s significant contributions and knowledge authority over the others‘ (Harwood, 2006, Wales, 1996). It is, therefore, a signal of rejection. In contrast, inclusive we explicitly embraces a wider audience and more importantly sends a signal of communality and consensuality. It can denote the joint effort of the writer and the reader in text
construction and secures agreement from the reader (Hinkel, 1999; Hyland, 2002a, 2005a; Kim, 2009; Wales, 1996). Also, as a sign of closeness and intimacy, we can also set up a dialogue with the reader and thus enhance the reader‘s involvement (Fetzer & Bull, 2008; Hines, 2004).
Seen differently, inclusive we, being a form of politeness marker, can act as persuaders as well as face-mitigators (Fortanet, 2004; Harwood, 2005; Hyland, 2001a, 2005b; Myers, 1989; Rounds, 1987). It tones down the face-threatening act (FTA) of self-citation and reduces personal contribution on account of its vagueness and fuzziness (Chen, 2010; Hyland, 2001b; Thompson, 2001). The writer can thus ―avoid drawing attention to himself/herself, and the writing becomes somewhat more
impersonal‖ (Biber, et al., 1999, p. 330). The phrase ―we believe,‖ for example, functions as a hedge on the writer‘s claim and therefore reduces the egocentricity of I.
Moreover, we is often collocated with emphatics, attitude markers, and modals of obligation to soften the directive force (Hyland, 2002a).
Finally, we can be used to organize the text and to advertise the writer‘s claims
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
(Harwood, 2005a). The discourse structure function assists the writer in guiding the reader through an argument and towards a preferred interpretation of a phenomenon in a shared journey of exploration (Hyland, 2001a).
In brief, we appears to be most prominently used in academic discourse, speech, advertising, and politics (Fortanet, 2004; Kim, 2009; Kuo, 1999; Myers, 1989;
Muhlhausler & Hares, 1990; Wales, 1996). The meaning of we is often vague and has to be determined by the context in which it occurs. But whenever we is used, the double assumption of authority and commonality is always conveyed (Pennycook, 1994). It basically fulfills several major functions: strengthening the writer‘s authority, reducing the writer‘s responsibility, displaying the writer‘s modesty, increasing the
Muhlhausler & Hares, 1990; Wales, 1996). The meaning of we is often vague and has to be determined by the context in which it occurs. But whenever we is used, the double assumption of authority and commonality is always conveyed (Pennycook, 1994). It basically fulfills several major functions: strengthening the writer‘s authority, reducing the writer‘s responsibility, displaying the writer‘s modesty, increasing the