The use of personal pronouns in the writing of argumentative essays by efl writers Introduction

Figure 1: Functions and examples of personal pronouns (adapted from Hyland, 2002, pp. 1100-1106)
This taxonomy was then used to examine the discourse functions of personal pronouns in the reports (student writing) and research articles (expert writing). The frequencies are shown in Figure 2:

	Student Reports		Research Articles
Function	Total		Total
	Raw	%	Raw	%
Stating a goal/ purpose	228	36	158	15
Explaining a procedure	199	31	400	38
Stating results/ claims	103	16	273	26
Expressing self-benefits	58	9	0	0
Elaborating an argument	49	8	220	21
Totals	637	100	1051	100

Figure 2: Discourse Functions of personal pronouns and their frequencies in student reports and research articles (based on Hyland, 2002, pp.1099-1100)

Figure 2 shows that the two most frequent uses in the student reports were stating a purpose and explaining a procedure. Hyland states these functions are less risky as they are simply organising the text or presenting their methodological procedure. This is supported by the interviews with students:

‘I’ is suitable for organizing the report, we are just saying about the research not about the ideas. It is only about the intention of the research and this is ok.
(economics student)

(Hyland, 2002, p.1101)

However, in contrast, the more risky functions, elaborating an argument and stating results and claims, which are seen as being more assertive, are used more frequently by the expert writers. This preference for author invisibility was made clear in the student interviews:

I don’t want to make myself important. Of course, it is my project and my result, but I am just ordinary student. Not an academic scholar with lots of knowledge and confident for myself.
(TESL student)

(Hyland, 2002, p.1105)

Another possible reason given is Ohta’s (1991) suggestion that the use of first person pronouns is largely unacceptable in Asian cultures where collective identity is more valued than individuality. This dichotomy between Western and Eastern culture is often stressed in the literature of applied linguistics but it also has been criticised as being too simplistic. As Kubota (1999, p.14) argues ‘The assumption underlying this approach is that there is a systematic, culturally determined way in which all members in a certain culture, think, behave and act’. Indeed Tang and John (1999) and Hyland (2002) have shown there is a continuum of usage which is not systematic and that student writers do use personal pronouns but in different ways to expert writers. Their research has indicated that students use ‘I’ but favour uses which possibly background their presence and are less ‘risky’. Also of particular relevance to this theme, is Natsukari (2012) who compared the argumentative essays of British and American undergraduates with their Japanese counterparts’ writing in English and discovered that the Japanese students overused ‘I’. It is too easy to state that students from a particular cultural background will use personal pronouns in the same way in their second language and the research underlines this. This variety of usage, which cannot simply be culturally determined, again supports the case for going to the source and asking the students as to why they make their particular, pragmatic choices.

Methodology
This paper examines the use of personal pronouns in the writing of argumentative essays by a group of seventeen Japanese university students over the course of a fourteen-week semester. This longitudinal case study examines the learner’s development and performance over a period of time, rather than just synchronically at one point in time. This element of time allowed me to collect data which reflects ‘natural changes in the learner’s behavior and knowledge, influenced by numerous possible factors, such as the environment, physical maturation, cognitive development, and schooling’ (Duff, 2008, p.41).
Data collection
In the first stage, I interviewed the participants to discover what experiences of writing the students had had, the kinds of writing they had done, and what they remember about the instruction they had previously received.
The second stage was a corpus of three argumentative essays written over the course of the fourteen-week semester by the seventeen participants. The final corpus of 44,674 words was then manually coded for personal pronouns. Also, to try to increase reliability a second coder independently coded the corpus.
Finally, in order to help discover the writers’ intentions and motivations behind their particular language choices, verbal protocol analysis (VPA) and retrospective, stimulated-recall interviews were conducted.