Notes on Revision - Creative Writing for English Majors | CRW 3013, Study notes of Creative writing

Download Notes on Revision - Creative Writing for English Majors | CRW 3013 and more Study notes Creative writing in PDF only on Docsity! HONGKONGPAPERS IN LINGUISTICS AND LANGUAGE TEACHING 12 (APRIL 1989) ISSN 1015-2059 1 .1 . The Process Approach THE REVISION PROCESS IN ACADEMIC WRITING : FROM PEN & PAPER TO WORD PROCESSOR Stephen Chadwick and Nigel Bruce Language Centre University of Hong Kong Part 1 : INTRODUCTION Teachers of academic writing have in recent years been turning their attention increasingly to the process of writing, and exploring process-oriented approaches to writing instruction (Murray 1978, Perl 1979 & Taylor 1984) . Murray, for example, feels that the teaching profession's "normal obsession with product rather than process leads us towards dangerous misconceptions about the writing process," and that, "the process of discovery, of using language to find out what you are going to say, is a key part of the writing process" . This is not to underplay the importance of final products. Most writing eventually reaches a stage when the writer decides it is a 'final' product, whether that means readiness for publication, or simply for handing in to a teacher for assessment . The writer's aim is to reach a stage at which he is satisfied with his communication to the intended reader. Students of writing can only benefit from an approach which, instead of requiring a first draft as a final, unalterable product for assessment, encourages them to produce several drafts, with constructive feedback from tutors between drafts . Writing therefore goes through a kind of metamorphosis on its way to its final state, and it is important that students learn that this is a normal process even for a proficient and experienced writer. Central to this notion of writing being a process, is the importance of the revision, or rewriting, stage . " . . .writing is a discovery procedure which relies heavily on the power of revision to clarify and refine that discovery" (Taylor 1984) . It seems obvious that if students are to be able to clarify and refine what they want to say, then atomistic sentence-based or example-centred remedial instruction must be supplemented by the frequent revision of written drafts . "Teacher presentations of standard patterns of organization or discussions on how to support an argument certainly have their place, but there is no guarantee that the necessary skills will be transferred and that the students will be able to draw on the information when they actually need it . Showing students where their own arguments are weak or where their logic breaks down appears to be a more effective approach." (Taylor, 1984 - our italics) Revising their own texts allows students theopportunity to reflect exclusively on their own writing problems and makes any feedback or commentary from the instructor immediately relevant. In this regard, it is important to distinguish between first drafts and revised drafts. 1 .2 . 1 st Draft vs . Revised Draft It is very difficult to make a clear distinction between, or even to clearly define, the notions of a 1st draft and a revised draft . One difficulty in trying to make such a distinction is that a draft HONGKONG PAPERS IN LINGUISTICSANDLANGUAGETEACHING 12 (APRIL 1989) exists first in the mind, and is altered both before and during the actual act of writing, through deletions, substitutions and additions (Marder 1982) . Nevertheless, in the context of university assignments, which are often undertaken (at least at Hong Kong University) under great pressure from other academic claims on students' time and attention, we can, for the purposes of this paper, distinguish 1 st draft from revised draft in the following terms (assuming the pen-and-paper or type-written method as the norm) ; a) Since most students do not have time to craft a preliminary version, complete with on-going deletions, substitutions and additions, and then write out a neat, unblemished copy, a first draft is, in most respects, the same as a completed first attempt, intended for submission as under exam conditions . The limited amount of revision of the written text that does go on at this stage, before handing in the assignment, tends to be at the sentence, rather than discourse level. There may be some students who choose to revise and edit work and then recopy, but these would seem to be in the minority (see e.g. Foulds 1987) . Exactly how students go about preparing their 1 st draft is difficult to pin down other than by eliciting subjective writing protocols (e.g . Hayes & Flower, 1981) . The computer offers us the prospect of being able to register each editing intervention as part of a linear record ; we hope to be able to employ this technique in the near future l . A 1 st draft is generally uninfluenced by another opinion ; one can be 'too close' to a 1 st draft and unaware of one's audience . Sommers (1980) writes of student inability to "re-view their work again with different eyes" . b) A revised draft, on the other hand, is an attempt to "clarify and refine" and tends to be made after consultation with others, particularly tutors, thus increasing the sense of audience . Of course, students could well 'consult' themselves, through that inner dialogue which helps a writer to distance himself from a text and develop a greater sense of audience . This ideal, however, assumes that the students already possess the necessary metadiscoursal awareness to determine when they are not communicating successfully with their audience, and to be able to rectify any shortcomings . But, as Taylor (1984) puts it, "Lacking nativelike intuitions about vocabulary, syntax, tone, style, formality, and organizational patterns, students often cannot see problems in their own writing (Beach 1979, Perl 1979) and will need to rely extensively on positive, constructive feedback." Hence, since the goal of our instruction is to help students to develop these intuitions and self-questioning strategies so that effective revision can take place before submitting a draft for comments, we must explore ways to ensure interim feedback and guidance on an individual basis and not necessarily always from the classroom teacher. Eventually, one would hope to be able to rely on effective regular peer evaluation as an integral part of the process in the writing classroom The important thing is that in helping students to develop these strategies, feedback should be given between drafts, and not merely as an assessment of a final product. 1 .3 . Holistic Writing The emphasis in this paper is more on the contribution teachers can make to a move to a more learner-centred, process-oriented and holistic approach to the development of writing skills HONGKONG PAPERS IN UNGUISTICSANDLANGUAGE TEACHING 12 (APRIL 1989) We are faced then with an awkward dichotomy. On the one hand teachers feel rewriting to be essential if their students are going to develop a sense for what is good rhetorical organization, and on the other they are hesitant to ask for it because of these 'negative' side-effects of requiring students to rewrite sizeable chunks of text, or indeed entire assignments . It is therefore ironic that the increased sensitivity to the human factor in language teaching may unwittingly be preventing student writers from acquiring the necessary skills for holistic revision, only perpetuating what Sommers (1980) calls the "inability to see revision as a process" . 1 .7 . Pen-and-Paper vs. Word Processor However much one wishes to centre a course around the writing process, the conventional pen-and-paper, or even typewriter, medium obliges the writer to engage in complete re-writing, with all the issues of redundancy and demotivation that entails. The word-processor, on the other hand, offers teachers of writing the opportunity to focus student energy on creative writing at all times . It is in the light of our overall diagnosis of problems with 'pen-pushing' writing courses that the move to computer-assisted revision appears to be such a promising alternative and potential solution to a problem which, though apparently largely logistical, is at root one of conception. Are teachers, in their desire to effect improved writing competence, emphasizing the process or the product? The conventional approach, pen-and-paper or typed re-write, with all the accompanying drawbacks we have already mentioned, seems to us to impede both a process approach to writing improvement and a holistic view of text revision . On the basis of evidence presented by other researchers (Dalute, 1985 ; Phillips, 1986, Higgins & Johns, 1984) we hypothesized that using the Word Processor as a writing tool might encourage or enhance a process approach to writing instruction . We therefore undertook a pilot comparative study, with 2 groups of 1 st-year Engineering students at Hong Kong University, to explore to what extent the computer is, as Dalute describes it, " . . .the perfect writing tool for a process approach to writing". HONGKONGPAPERS IN LINGUISTICSAND LANGUAGE TEACHING 12 (APRIL 1989) 2.1 . Introduction a) Experimental Hypotheses b) Experimental Design A . writing performance Part 2 : THE STUDY Information was sought on 3 fronts : students' attitudes to each medium, their performance, and the processes they use in revising by each method . Our experimental hypotheses were that the use of a word processor has an effect on: B . student attitudes to writing and revising, and C . the process by which students revise their scripts We hoped that the results of the study would also enable us to provide some initial answers to a question posed by Martin Phillips (1986), namely : to what extent (will) the ways in which we propose to exploit the computer lead to positive benefits in terms of their impact on methodology?" This study took place between September 1987 and March 1988, at the University of Hong Kong's Language Centre . A control group of 13 first-year industrial and mechanical engineering students, and an experimental group comprising 12 first-year students from the civil and electrical engineering departments, both followed the Centre's 20-hour report writing course for engineers over a 10 week period . On entry to the course both groups were found to be comparable in terms of writing skills . The division between departments (industrial /mechanical and civil/electrical) was determined by timetable restrictions. Both groups had the same instructor so that course content and methods of instruction were, as far as possible, identical . The control group (hereafter referred to as "P group" - for pen-and-paper) were taught in a conventional classroom setting, doing all their writing by hand, whereas sessions for the experimental group ("C group" - for computer ) were timetabled in a computer laboratory at the University's Computer Centre, and did their writing on word processors . In order to monitor their performance, students sat pre- and post- course tests . Both groups wrote these tests by hand under examination conditions . Requiring the C group to complete the test on word processors might have been more appropriate, but would not have allowed a valid comparison of the two groups' results given the unpredictability of computer networks and the need for 'clockwork' timing in exam sessions. Indications of students' attitudes to writing were gained from pre- and post- course questionnaires . During the course students were asked to complete a number of short written assignments and three drafts of a long report . All of these assignments were written by the C group on word-processors . The written texts were analyzed in an attempt to measure any improvements in writing performance . The questionnaires, together with further analysis of the three drafts of the long report, also provided information on the strategies students employ in the process of writing. HONGKONG PAPERS IN UNGUISTICSAND LANGUAGE TEACHING 12 (APRIL 1888) 2 .2 . PERFORMANCE 2.2 .1 . Methods a) Pre-Course Tests All students wishing to follow an engineering course at the University must obtain a grade 'E' (the lowest Pass grade, the highest being 'A') on the Hong Kong Examination Authority's Use of English paper. This examination is designed to test general language proficiency. Since the results of the UE examination are simply given in the form of a grade on a scale of A to E, it was difficult to subject these grades to any statistical analysis . On entering the University, all engineering students sit the Language Centre's placement test . Based on their performance on this test, the weaker 50% of students are selected for the academic report-writing course. The placement test comprises a report-writing task and a C-test. Student responses on the C-test were marked on a right/wrong basis . The writing task required students to write a report from information provided in the form of graphs, notes and tables . The scripts were given a subjective impression mark by two separate markers on a 1 to 7 scale . The results from these two tests were subjected to statistical analysis using a simple t-test and a Mann- Whitney U-test to ascertain whether there was any measurable difference between the groups before the course began . b) End-of Course Projects Towards the end of the course all students worked on a problem-solving project during which they had to devise a scheme fulfilling the electrical energy requirements of a small fictitious island . Students worked in groups at the pre-writing stage, discussing options, doing calculations and making oral presentations to the rest of the class. Each group was asked to work as a team to come up with a plausible solution, but students then wrote reports outlining the scheme and justifying their arguments on their own . After writing the first draft of this report, the students received comments on the rhetorical features of their texts, as well as on the content . After a feedback session looking at common problems arising from the drafts, students were asked to rewrite the reports. This second draft was marked for spelling, lexical, stylistic and grammatical appropriateness, and students were asked to edit it to produce the third and final draft to be handed in for assessment . It was at this stage that the texts were graded . All of them were later typed on a computer (the different impressions made on the reader by handwriting/printout being an unwanted variable) and graded by two experienced markers who had taught the course but who did not know the students . A marking scheme was used which was designed to assess general writing proficiency and grades were again awarded on a 1 to 7 scale, according to success in meeting the rhetorical objectives of the course. c) Post-Course Achievement Test After completing the course all students sat an achievement test, designed to be a parallel test to the writing task on the placement test . The scripts were marked in the same way as in the placement test using the same two markers and a 1 to 7 scale. The markers were again asked to focus specifically on the rhetorical and linguistic skills taught on the course. Again, the results were subjected to statistical analysis, using a t-test and the Mann-Whitney U-test. HONGKONG PAPERS IN LINGUISTICS AND LANGUAGE TEACHING 12 (APRIL 1989) There seemed to be no major differences between the groups in terms of ease and enjoyment totals . However, when we look at the break-down of these totals, we see that the C group found writing assignments both easier and more enjoyable, while revising was easier though slightly less enjoyable . It is at the editing stage, however, that the C group's responses were most negative . The most striking difference between the groups was in how effective they felt their writing of the assignments was . 7 of the computer students (58%) felt their writing to be quite effectivelvery effective, whereas 12 of the P group (92%) felt it was okay/not very effective. The C group seemed to feel more positive about the effectiveness of the course in improving their ability to write assignments. If not required by their tutors to revise or edit their work, slightly more students in the C group thought they would revise organization anyway, while slightly more of the P group thought they would edit for grammar, spelling and vocabulary . The final question on the post-course questionnaire asked the open question, "Do you think the course has changed the way in which you write your assignments? ff so, in what way?" Almost all students (22/88%) felt the course had changed the way they wrote assignments. The P group concentrated their comments on the content of the course, feeling that it had improved their organization, revision skills etc . The C group wrote much longer comments, stressing the value of the computer as a writing tool . Of the 11 students who wrote something, 5 (45%) mentioned that the computer made revising, editing and rewriting easier . 5 also felt a computer printout improved the appearance of their work; as one student put it, "Usually my assignments are not tidy, but now by using the computer, it gives me the impression that my work is tidy and easier to be read." 4 students (36%) felt that it took them longer to write assignments on computers because they did not know how to type . Of these, 2 (18%) qualified this by saying that the time spent was worth it since the appearance of their work was improved . c) Staff-student consultation meeting This was attended by five students representing the six engineering writing classes. The C group representative said that his group liked writing and revising on computers, whereas the P group's representative said although his group felt revising was useful, they did not enjoy it. There was some discussion on the possible use of the computer to write assignments in future during which several students said they would like to be given the choice between writing on computers or by hand . 3 representatives went on to say that if given the choice they would still prefer to write by hand . Inability to type and slow typing speeds were cited as the reasons for the reluctance to use computers . 2.3 .3 . Discussion a) The Pre-Course Questionnaire Our main aim in giving this questionnaire to students before the course began was to gain some insight into the way students ordinarily tend to write assignments ; we felt this would, at least to some extent, offer a reflection on what they had previously been taught . It appears that the majority of students think a lot before writing, and very few write quickly with the intention of changing the draft afterwards. This could be due to the notion, gained through years of experience in secondary schools, that any draft handed in to the teacher is the final 10 HONGKONG PAPERS IN LINGUISTICSANDLANGUAGE TEACHING 12 (APRIL 1989) product . Students tend to have a sense of 'completeness' when they produce their first draft, and there is little notion of writing being a process . This is a notion that is difficult to 'un-learn' . Another result which seems to indicate an emphasis on the final product, is that while nearly all students felt it was important/essential to revise/edit their work, most thought it important to do this before receiving a teacher's comments. They are obviously unused to getting work back with comments which they are routinely expected to act upon . The sense of failure and punishment attached to a request to rewrite a text seems quite understandable in the light of these comments, since rewrites are not associated with the 'normal' writing process . b) The Post-Course Questionnaire It appears from these results that although there was no major difference in responses between the groups after the ten weeks, there was a move towards a more flexible approach by the C group, in the sense that they now had ease of revision in mind before they started writing . Both groups were told before writing that they could change/revise their work later, and the P group were told they need not rewrite the entire text but could "cut and paste" to facilitate the production of a revised draft. Nevertheless, the P group spent longer thinking before writing than the C group . This could well be because, even with the cut and paste facility, revision with pen- and-paper is still much more time-consuming and burdensome than with the word processor . The fact that more students in the C group said they spent quite a lotla lot of time revising the structure of whole paragraphs appears to have two possible explanations . Firstly, it may be that the C group simply took longer to accomplish the same type of revisions as the P group. This, however, seems unlikely. Even given that the P group could cut out sentences and re-arrange them, it is hard to believe that this is a faster method than using a computer to move text around on the screen . A more likely explanation seems to be that the ease with which the C group could manipulate paragraph structure meant they spent more time considering this aspect of their text revision . Several factors might account for why editing seemed to be easier for the P group . Firstly, several of the C-group students were totally unfamiliar with word-processing on computers . Secondly, those students who were familiar with word-processing were nevertheless unfamiliar with the software program the University employs (EC-Write. Thirdly, poor typing skills were cited by 2 (18%) students4 as the reason why they found working on the word-processor took longer - even for adding, deleting, substituting short phrases, single words or letters . These initial difficulties with computers are common to most beginners . Almost all who regularly use computers, however, see the difficulties with the mechanics of typing and so on as short-lived . At 20 hours, only half of which was actually spent on the word processor, the course did not allow students time to develop familiarity with the medium . We would hypothesize, therefore, that both ease and enjoyment would be enhanced as familiarity with the medium increases . Nevertheless, these initial difficulties do pose a problem in comparing writers using a totally familiar medium with others using one which is quite unfamiliar. Daiute (1985) discusses this issue at some length, commenting that nearly all studies are carried out when students are at this initial experimental stage. The interesting thing with our results is that the C group found revision easier, while the P group found editing easier . We do not know if the picture would remain the same if we administered another questionnaire after the students had been using the computers for several months . Our own experience tells us that after learning how to use spelling checkers, thesauruses, search-and-replace facilities, and how to move quickly around long texts using 'macro' commands (several complex commands condensed into one or two keystrokes), the computer users would also find editing easier on the computer. Nevertheless, as tertiary level writing teachers, we find this shift in emphasis away from surface level changes and on to macrostructural changes a welcome result of introducing the computer into writing classes, HONGKONG PAPERS IN LINGUISTICS ANDLANGUAGE TEACHING 12 (APRIL 1989) remedying an area of neglect in our students' previous writing instruction . One other problem we encountered was in interpreting student responses in the "effectiveness" sections (see Appendix 2) . In retrospect, we saw that while the wording of questions 18-20 seems to be fairly explicit, there may have been some ambiguity as to the precise meanings of questions 15-17 . Did students, for example, interpret "effective" as meaning that their end-products fulfilled the course goals effectively? Or that the computer was more effective as a writing instrument? We do not know the answer to this and are therefore reluctant to draw any conclusions from these responses . Responses to questions 18 and 19 appear to be the most intriguing and perhaps also the most significant . The C group seemed to think the course was more effective than the P group dic The fact that more students in the C group thought the course effective in improving the organization of their work, as well as being a result of previously mentioned factors, could also have to do with non-productive labour. Because we asked students to produce at least two (and i the case of the final project, three) drafts of their assignments, this factor may be an important one When students work on the word-processor, all their interventions produce meaningful results ; that is, they simply revise text, and need not rewrite. When writing using pen-and-paper, the student faces a dilemma : either he rewrites the whole, or a large portion, of his text, which will involve extensive copying, or else he decides to avoid this even if he sees certain limitations in the organization of the information as it stands . If he rewrites, he can hardly see the activity as "effective" . If he chooses not to, there may be a sense of frustration at leaving it in an unsatisfactory state . Perhaps it is this sense of frustration at being limited in what they can revise that influenced the notion of effectiveness for the P group . We discuss this idea further in section 3.2 . 2.4 . PROCESS 2.4.1 . Methods The three drafts of the long project were analyzed, using Faigley and Witte's (1981) taxonomy of revision changes as a basis . Since our main concern was whether there were any differences between the two groups in the way they revised (as opposed to edited) their texts, we confined ourselves to what Faigley and Witte call "Meaning" changes and ignored "Surface" changes (see 2.4.3a) . All meaning changes were recorded using their Addition, Deletion, Substitution, etc . table, both at the micro and macro structural levels5. In addition, we also recorded the number of interventions each student made. Interventions are defined here as the actual number of changes, as opposed to the number of sentence units which were changed . Therefore a major deletion of say, 6 sentences, would be recorded as 6 in the Faigley and Witte taxonomy, but as only one intervention . This would enable us to look not only at the total number of sentence units that were changed but also at their distributions within the texts. To make the analysis easier, the different strategies were colour-coded, using highlight pens . Since little evidence was found of Faigley & Witte's 'Consolidation' or'Distribution' strategies, we only marked Addition, Deletion, Substitution and Permutation changes, although Consolidation and Distribution strategies were looked for. Time did not allow us to examine the texts of all 25 students, so we chose to look at 6 student projects from each group. Students were ranked according to the grades given to them in the assessment and every alternate student in the rank order was chosen, giving a range of students from the most to the least proficient in each group. 12 HONGKONG PAPERS IN LINGUISTICS ANDLANGUAGE TEACHING 12 (APRIL 1988) Perhaps most interesting were the Permutation changes, especially those described above as 'global permutations' . Although these interventions were few in number (1 for the P group and 5 for the C group) and only occurred in 3 of the 12 scripts analyzed (1 P group and 2 C group scripts), they nevertheless amounted to a difference in actual sentence units changed of over 300% (4 for the P group and 13 for the C group), testifying to a sharpening of metadiscoursal awareness that we had not expected to find after such brief exposure to the word processor . The P group instance was a case of perceiving the redundancy of repeating in text what had been made eminently clear in a diagram . The writer retained only those propositions which drew attention to the most significant features of the scheme, placing them after the diagram . The global permutation changes made by the 2 C group students all featured the retention and re-location of text while revealing an awareness that it had been given an inappropriate rhetorical "value" . One of the 2 students realized that what he had called his 'Conclusion' was actually only adding qualification about potential drawbacks to his proposal, and so properly belonged in the section 'Recommended Scheme' . The other C group student, who had the most such interventions in his report (4), had first of all removed a rather philosophical introduction to his report, after some class discussion on the likely shared knowledge between the reporter and the targeted reader. He scaled down the frame of reference of the report, moving his 'Background to the Project' from p.3 to the opening section of the report . This writer had also begun his 'Feasibility' section outlining the purpose of the project ; he proceeded to move this and create a separate 'Objective of the Report' section . Where he had included a criticism of an unsuitable form of energy in his description of the different energy sources, he proceeded to move these comments to the 'Objective of the Report' section, narrowing the scope of the report by eliminating a priori unsuitable energy sources . Finally, this writer had listed his 'criteria' for selecting energy options in his 'Recommended Scheme' section, on p.6 of his report . He moved these 'criteria' 3 pages forward to a separate section just prior to the 'Feasibility' section, realizing that one cannot discuss the feasibility of an option without knowing what criteria one is judging the options by . In helping students to develop a sense of how well their writing fulfills its rhetorical goals, there seem to be clear advantages (over the Humanities-type essay) to working with a text format which encourages the assignment of information to the service of specific rhetorical macrofunctions (e.g . Background - Description of Energy Sources - Feasibility of Energy Sources - Recommended Scheme - Discussion - Conclusion) . Interpreting these data in terms of desirability of writing/revision strategy, we could say that the P method tends to encourage an 'all-or-nothing' approach to revision and, in the case of a report clearly segmented into sections, any 'holistic' approach would be confined to the level of a section, time and tolerance being the constraining factors . It is very difficult to insert single sentences, or small paragraphs into a larger text without adversely affecting its presentation (an important factor in reports), or opting for a total rewrite . Since they were given the possibility to make photocopies of their texts, cut them up and paste intact chunks of text onto their revised version, the P group students tended to think in terms of revising 'by section', opting to change those sections deemed least satisfactory . Where most of the text was acceptable, they decided to leave it alone. Conversely, where they felt unhappy with most of what they had written in a particular section, they chose to rewrite . It seemed to us that the focus was not so much on what should be changed but rather on what could be retained . Occasionally, however, a student would decide to rewrite a large section for the sake of making minor changes . In one extreme example, a student wanted to insert two sub-headings into the Introduction section of his report . He pasted a 20-line piece of paper over his old text having first recopied it exactly, except that his writing was slightly smaller to make room for his two additions . Interestingly, the 1 P group student who re- wrote his entire paper still produced a revision in which 50% of the text was unaltered, and therefore 50% of his time had been spent on redundant copying . One wonders how "effective" these (P group) students feel the revision stage of the writing process is in improving their work. 1 5 HONGKONG PAPERS IN LINGUISTICS AND LANGUAGE TEACHING 12 (APRIL 1989) The process analysis ultimately throws the spotlight back on attitudinal factors like punishm and re-writing apprehension. The C group seemed rather more willing to intervene in their texts . The relative ease which they could add, delete or re-locate text meant they were able to focus on the overall rhetorical structure of their reports . Freed from the restrictions imposed by the need to re-w they were more able to make a holistic appraisal of their texts. This would account for the la number of macrostructural additions, deletions and permutations for the C group, with the "c and-paste or total rewrite" technique accounting for the P group's much greater number of macrostructural substitutions . Other researchers, like Sommers, Perl, Raimes and Zamel, have told us that good w revise more at the 'rhetorical' level . The use of the word processor, at both the primary writir revision stage, would seem by that yardstick to be encouraging 'better' writing, with a greate awareness of the rhetorical function and patterns of information and of the accessibility of th ; information for the reader. Finally, a word about our methodology . One might well argue that if we had require( group students to rewrite the whole report they would have employed different revision strat( and would have intervened more frequently and more evenly throughout their texts, just as the group did . This is undoubtedly true, but it also proves an important point. Our awareness of writing apprehension and sense of punishment made us reluctant to require this from studen These reports were approximately 3 - 7 pages long, with several diagrams . Is a teacher justii requiring two rewrites from students, knowing that approximately half of the activity would sii be copying? We think not, and this may go a long way to explaining why so little instruction practice is really given in revision strategies during writing classes . HONGKONG PAPERS IN LINGUISTICS AND LANGUAGE TEACHING 12 (APRIL 1989) 3 .1 . Writing Q uality Part 3 : GENERAL DISCUSSION We have already stated in the Discussion on Performance earlier in this paper that our basic teaching instincts told us before the study began that we could not expect to see any dramatic effect on writing quality simply by introducing students to writing on the computer . Our reading of relevant literature strengthened that conviction . "With all our efforts there is still no hard evidence to indicate that the computer use in itself makes one a better writer" (Pierson & Leung, 1987) . "It would be absurd to say that word-processing leads to good writing" (Foulds, 1987) . "The few studies of writing quality have shown that writing on the computer is sometimes rated lower than writing done by the same people with traditional tools" (Daiute, 1985) . Those writers who were prepared to say that computers actually improved writing quality did not, unfortunately, do it very convincingly : "Word-processors . . . make on-screen editing so easy that they encourage the user to try out the effect of changing or adding words, changing the order of sentences or even paragraphs, and playing around with the aesthetic effect of different layouts . As a result they seem to be affecting the quality as well as the quantity of what is written on them" (Higgins & Johns, 1984 - our italics) . One cannot expect the mere opportunity to "play around" with layouts and "try out the effects" of changing words around in itself to improve the quality of students' writing . Writing programmes have to be designed around pragmatic principles so that the development of rhetorical, or'metadiscoursal', awareness in students becomes a prerequisite for any improvement in what we are calling the 'writing process' . Without such an awareness, in the words of Daiute (1985), "all the word-processing commands in the world won't help" . Not surprisingly, then, our results from the Performance section lead us to a rejection of our first experimental hypothesis . The computer did not directly affect writing quality, at least in the short term - which in this case means 10 weeks or 20 hours of instruction . Nevertheless, results from the Attitude and Process sections of the study tend to confirm our two other experimental hypotheses and cause us to agree with Higgins and Johns that writing on the computer will affect writing quality - but for quite different reasons . Despite the short time-span of the study and the fact that students were, for the most part, novices with word-processors, the computer seems to have had a positive effect on student attitudes to writing. This was especially evident with regard to the ease with which they felt they could write and revise and their sense of the effectiveness of a computer-based writing course. The computer certainly had an effect on the process the students went through in revising their work. It is these two aspects of the results - Attitude and Process - which we believe will improve writing quality in the long-term . If the computer changes the way students write and their attitudes towards writing, then these factors, together with the variety of writing aids the computer offers, will certainly affect the way teachers teach writing in the classroom 1 7 HONGKONGPAPERS IN LINGUISTICSAND LANGUAGE TEACHING 12 (APRIL 1889) b) Process : This shift of focus will facilitate the teaching of writing as a process cess. Since students can begin writing knowing all the time that revisions are both easy and painless, teachers for their part will routinely and frequently require revisions from students . They will be able to do this for the first time without any professional qualms about the punishment and non-productive labour factors . Revising will be seen not as a punishment for a poor final product, but as a natural stage in the writing process, starting with a consideration of the rhetorical macrostructure of one's text, and gradually moving through to the lexico-grammatical editing stage . c) Holistic Approach to Writing : Not only will teachers feel free to make comments on both rhetorical and linguistic features of student texts, but they will justifiably expect the students to act upon all of these in revising their work. This in turn will mean that students will develop the skill to see the text holistically, as an entity to be structured and restructured, and not merely as a series of sentences and paragraphs to be 'corrected' . d) The Writing Syllabus : We predict that writing syllabuses will change. Revision sessions are more likely to become a common element in the writing syllabus than was the case with pen-and-paper. Students will not have to do time-consuming rewriting at home without recourse to a tutor to clarify annotated comments, explain guidelines and make suggestions . Instead, revision sessions will be built into the course design and students can make constant use of the valuable resource their peers and tutors offer . As computers become more commonplace, they will also hopefully be able to book time on their institution's computer facilities to carry on their revision out of class. e) Individualization : Finally, the focal point of all writing classes for a student will be his own writing . One of the problems of group tuition is that students find it more difficult to perceive the relevance of a language point taught through someone else's work. It is not greatly motivating to have to continually abstract information about a linguistic system simply for its potential application to one's own writing . No doubt many teachers will argue that, while it is all very well to emphasize writing and revising as a process offering individualized feedback, the teacher will be veritably swamped with student assignments to comment on . Not only one, but two or even three drafts of the same paper! Of course, there may well be more work for the teacher in that area, but the computer offers many other facilities which would reduce the wastage of teacher time . The need for tedious repetition of the same points of grammar, for example, whether written on the students' papers or explained orally in the classroom, could be largely eliminated if interactive software, featuring both grammar and rhetorical structure instruction and exercises, were accessible to students on computers while they were writing assignments. On-line thesauruses and spelling checkers would also cut down on teachers' time. The time saved on these mundane, repetitious tasks would leave teachers freer to concentrate on the procedural and pragmatic dimensions of writing and revising. In the first chapter of Computers in Language Learning (1984), Higgins and Johns explain that their aim is not only to justifyusing computers, but also to discuss "the changes of attitude or approach that must accompany such use . The learners must be, to some extent, in tune with the medium before they can benefit from it ." 20 HONGKONGPAPERS IN LINGUISTICS ANDLANGUAGE TEACHING 12 (APRIL 1888) There is no doubt that the same changes of attitude and approach towards the use of computers in the writing classroom are essential on the part of teachers and educators if they are to provide students with the best possible environment in which to improve their writing skills . NOTES 1 . H. Pauwels, at the University of Antwerp (Prinsstraat 13, B-2000 Antwerpen), has used this technique with some success . 2 . After the Prague School : communicative dynamism can be described as that quality, or aggregate of qualities, in a text which impels a reader through that text, and which "pushes the communication forward" . (Firbas 1971 : 136) 3. "Rewriting is when playwriting really gets to be fun . . . In baseball you only get three swings and you're out . In rewriting, you get almost as many swings as you want and you know, sooner or later, you'll hit the ball" . Neil Simon, quoted in Murray (1978) . 4 . In response to the last, and only open, question : "Do you think the course has changed the way in which you write your assignments ? if so, in what way ?" 5. Faigley & Witte describe their categories most clearly in terms of 'Meaning-Preserving Surface Changes' . We quote : "Additions raise to the surface what can be inferred (you pay two dollars = > you pay a two-dollar entrance fee) . Deletions do the opposite so that a reader is forced to infer what had been explicit (several rustic looking restaurants = > several rustic restaurants) . Substitutions trade words or longer units that represent the same concept (out-of-the-way spots = > out-of-the-way places) . Permutations involve rearrangements or rearrangements with substitutions (springtime means to most people = > springtime, to most people, means) . Distributions occur when material in one segment is passed into more than one segment . A change where a writer revises what has been compressed into a single unit so that it falls into more than one unit is a distributional change (I figured after walking so far the least it could do would be to provide a relaxing dinner since I was hungry = > I figured the least it owed me was a good meal. All that walking made me hungry.) Consolidations do the opposite. Elements in two or more units are consolidated into one unit (And there you find Hamilton's Pool. It has cool green water surrounded by 50-foot cliffs and lush vegetation. = > And there you find Hamilton's Pool: cool green water surrounded by 50-foot cliffs and lush vegetation) . As the last example suggests, consolidations are the primary revision operation in sentence-combining exercises." (Faigley & Witte 1981, p.402) HONGKONGPAPERS IN LINGUISTICS AND LANGUAGE TEACHING 12 (APRIL 1989) REFERENCES Beach, R . 1979. 'The Effects of Between-draft Teacher Evaluation versus Student Self-evaluation on High School Students' Revising of Rough Drafts' . Research in the Teaching of English 13 : 111-119 . Daiute, C . 1985 . Writing & Computers . Reading, Mass. : Addison-Wesley. Daly,J . & Miller, M. 1975 . 'The Development of a Measure of Writing Apprehension' . Research in the Teachina of Enalish 9 . Deming, M.P . 1987 . The Effects of Word Processing on Basic College Writers' Revision Strategies Writing Apprehension and Writing Quality while Composing in the Expository Mode. Unpublished Ph.D . thesis, Georgia State Univ. Faigley, L. and Witte, S. 1981 . 'Analyzing Revision' . College Composition and Communication 32 : 400-414. Firbas, J . 1971 'On the Concept of Communicative Dynamism in the Theory of Functional Sentence Perspective' . Universitas Brunensis . Studia Minora A19 . Flower,L.& Hayes J. 1981 . 'A Cognitive Process Theory of Writing' . College Composition & Communication 32 : 365-387 . Foulds,D . 1987. 'The Word Processor in Language Teaching: New Friend or Old Foe?' . In Proceedings of the 1986 ILE International Seminar on Language Teacher Education , Hong Kong: ILE . Higgins, J . and Johns, T . 1984 . Computers in Language Learning . London : Collins. Kemmis, S., Atkin, R ., and Wright, E . 1977. 'How Do Students Learn?' Working Papers on computer-assisted learning : UNCAL evaluation studies . Occasional Publications . 5 . Centre for Applied Research in Education, University of East Anglia . Krashen,S . 1984. Writing : Research. Theory Applications Oxford : Pergamon . Marder,D . 1982 . 'Revision as Discovery and the Reduction of Entropy' . In R . Sudol (Ed .) Revising-New Essays for Teachers of Writina , ERIC Clearinghouse . Murray, D. 1978. 'Internal Revision : A Process of Discovery' . In Charles R. Cooper and Lee Odell (eds.), Research on Composing: Points of Departure . Urbana, Illinois : NCTE. Nold,E . 1982. 'Revising : Intentions and Conventions' . In R.Sudol (ed) Revising : New Essays for Teachers of Writing, ERIC Clearinghouse. Perl,S . 1979 . 'The Composing Process of Unskilled College Writers' . Research in the Teaching of Enalish 13 : 317-339 . Phillips,M . 1986 . 'CALL in its Educational Context' . In G .Leech & C . Candlin (Eds) Computers in Enalish Lanauaae Teachina & Research. New York: Longman . 22 HONGKONG PAPERS IN LINGUISTICSAND LANGUAGE TEACHING 12 (APRIL 1889) APPENDIX 2 0 $4 0 25 4) 4) A m b a) 0 0 b+ .r1 "rl 3 41 41 d' O 01 M O U N r) N a m o) r 0% a cr) co ra O U N N N CO a 0) co r .r a cq N cIQ co ko N %D a O U N M c4 o% a Co ri M N a N r; M f%. " En O.P -rl 9 -pa) N >r .: -rl I`I ,LI IT P: (d fo.ri fu N IT 0) !N cd .A ". (d O )4 (d 0) 0 W I~ p a) A4 O a .rl 0 r.., )-1 a) ,yw ~, "3 4J OIS tn E 0 IT 0 I'-l m 9 -rI .'j ("i -ri r~ En O -ri F-4 b 4.) -H T, +p H "rl -r4 > -rl a) b ~4 a)44 bw3 s4 O a) N 3 0 Lo ko r- m w to W z w ar H $4 E "rl V ld W w a w 0 w "rl 43 a) a 4l In M -p U) W H In r., r-l U) W -r'I 0 a) 0 - f 4 0 r-1 W U co r-O -H a)I Co M 44 N41 a% w pm rl0 }4 Aa >, a) 00 a) O R EN a) Dw4 zfZ4 WI W I I Izo aMLn a) v OI a) a) z M 4J r-l 'J w a (n A 0-14 fd 4-) .rt 0 ?, U 4.) -H O ?, (1) U) -n N 44 a) 'd >~ (d 44 :1 4 W W W av 0 a) a) a)W fn4) I--idIaa) s1 HONGKONG PAPERS IN LINGUISTICSANDLANGUAGE TEACHING 12 (APRIL 1989) APPENDIX 3A 26 a Ci ~f) ei O~ ri I Ln N O~ M I I 04 N .i IA w rl O H .41 M I U ".1 W N 4.) O Nb 4-1 A 114 CP °a v C7 '"{ N N ~--1 N n C1 \0 C'1 Ol ~" a , o w J i N H i a w 04 tT ~ HGC ~o w N M q O O U v a N ~ I O Lo cn kD e' -r v w o w w N O G) yJ in N M ri %0 I- n U Id HO 11 N 04 n M I 1 a N a I ) d H N V°a .~ a ,d w to (3i ~ a ! N N lf10Oa -, °' A o w co T in m a m w I -~v x w I a '° e N d x v °w a ed O V .1-4 14 lfl !n l0 N M Lo W U 41 u) r-1 r1 o+ O m 94 a w a o o a U 4~.) H G -p '~ E-4 D . "I Ql ~ W ~ N ~ t~ fA EG H ,'~,'HVR:O ~FCUGtiO HONGKONG PAPERS IN LINGUISTICSAND LANGUAGE TEACHING 12 (APRIL 1989) APPENDIX 3B Total Revision Changes in Second Drafts for the Computer and Pen-and-Paper Groups Pen-and-Paper Computer Group Group ANING CHANGES Total Number of Combined Revision Changes and Total Interventions in Second Drafts for Computer and Pen-and-Paper Groups C 27 MICROSTRUCTURE Changes Additions 21 29 2 Deletions 12 5 Substitutions 7 11 4 Permutations 3 9 5 Distributions 1 6 Consolidations I MACROSTRUCTURE Changes 1 Additions 33 62 2 Deletions 4 9 Substitutions 25 3 4 Permutations 4 13 S Distributions 6 Consolidations Pen-and-Paper Group Computer Group MICROSTRUCTURE Changes 43 55 MACROSTRUCTURE Changes 66 87 TOTAL Changes (Sentence units) 109 142 TOTAL INTERVENTIONS 65 91