Vol. 1. No. 4 A-3 June 1995
Return to Table of Contents Return to Main Page

What the "Process Approach" Means to Practising Teachers of Second Language Writing Skills

Tim Caudery
University of Aarhus, Denmark


The article reports on a survey conducted on TESL-L aimed at discovering whether ESL teachers have similar concepts of the "process approach" to writing, or whether the concept has now evolved in different ways in different places. The survey results show that teachers actually have strongly differing ideas as to what process writing is. Such divergence may be typical of teaching approaches which have been in use for some time; but in this instance the changes may have been accentuated because the process approach was originally developed in and for the L1 classroom, and has been adapted for L2 teaching. It is suggested that some of the heat which characterized the initial debate on process writing may have cooled a little, making the time ripe for fresh discussion of what we have learned about teaching writing in L2 and possible ways forward in the future.

The article also comments briefly on the process of conducting surveys on an electronic network.

Twenty years ago or so, researchers and teachers of writing to native speakers of English were beginning to explore the processes that went on in the creation of written text. They discovered that writing was a highly complex process, made up of various subprocesses that occurred not one after another in a strict linear sequence, but cyclically and in varying patterns. They also discovered differences in the writing processes of "expert" and non-expert writers. For writing teachers dissatisfied with "traditional" American approaches to teaching writing, these findings suggested a way forward. Instead of concentrating on the writing that students produced and making critical comments on it, they could aim to help students write better by aiding them in the actual process of writing, by finding the source of their problems in creating good written texts and enabling them to overcome those difficulties. Such ideas also accorded well with the thoughts being expressed in the late seventies in numerous articles (e.g., Murray, 1980) by Donald Murray, a professional writer and an influential [-1-] teacher of writing. His ideas were based not on research into the writing processes of others, but on his reflections on his own writing processes. He placed emphasis on the importance of a series of drafts in the writing process as the writer gradually discovered through writing what it was that s/he wanted to say.

Thus, a "process approach" to teaching writing was born. The key factor was that teaching focussed on the writing process rather than the final product. But this was in the first instance a teaching approach, not a teaching method; the pedagogical methods and means whereby a teacher was to help students adopt more successful writing processes were not laid down or even very clearly implied by most of those who did empirical research into writing processes.

In second language teaching, a certain amount of writing skills teaching had always gone on - to meet the needs of students who were to take international language examinations, for example. But for much of this century, the ESL classroom had been dominated by methods aimed at acquiring spoken language skills--that is to say, at least, this has been so in those teaching situations where pedagogy was not still firmly rooted in grammar/translation techniques. In the early seventies, communicative teaching methodology and work on functional/notional syllabuses directed our attention more firmly towards the specific needs of the individual learner. These needs were viewed not only in terms of particular language items but also of particular types of communication, and the resulting realization that different learners actually had different requirements with respect to language skills meant that new attention was given to, among other things, the teaching of writing. In this context, the process approach arrived on the scene at a very opportune moment--for in second language teaching, the problem was not so much that traditional methods of teaching writing had proved inadequate as that there had previously existed no coherent, theory-based approach at all for teaching writing in a second language. And so, slowly at first but with gathering momentum, the process approach to writing teaching has been widely adopted in the second language classroom.

The process approach has always been controversial, however. In part, this has been for practical reasons; process teaching often requires more in the way of input from teachers and students alike, and the degree of individualization involved can also present organizational problems, leading to disruption of "normal" teaching patterns. There has been debate over whether the focus of the teaching is always appropriate for the students concerned; for example, for students preparing for language examinations there is an obvious conflict between the extended composing processes encouraged by the process approach and the single-draft writing usually necessary in an examination (Horowitz, 1986). The fact that the process approach found its justification in the mother-tongue [-2-] classroom through opposition to an existing methodology has always tinged the approach with an adversarial flavour; justification for using a process approach always seems to start not with "here's a wonderful new way of teaching writing" but rather with "old- fashioned approaches to writing teaching are wrong. "The essence of the process approach has even been reduced to a slogan, a battle- cry, which stresses this confrontational aspect: "Process, not product." And now, opponents of the process approach are beginning to gather under a new banner, that of the "genre approach."

The reader will, I hope, forgive this superficial and over- generalized introduction. It is not my purpose in this article to describe the history of the process approach in detail, but rather to begin to examine its current position in ESL teaching. At this point, some fifteen years or so into the history of the process approach in a second language teaching context, and with new ideas on the teaching of writing being introduced, I feel that it is useful to take stock and see what the process approach now means to the language teaching profession. What have we learned from it? Is it being used as well as it could be? Where do we go from here?

As teaching approaches become more widespread, various things happen to them. One trend is diversification; different people interpret ideas in different ways. This is particularly likely to be the case where we are talking, as with process writing, about a general approach to teaching rather than a closely-defined teaching method. Thus, many different types of teaching may go on under the same general label. Another trend common with teaching approaches is simplification; as ideas spread from one teacher to another, it is the strongest and most distinctive elements of the original approach that tend to survive. Simplification may be accompanied by distortion of the original idea--or evolution, to suggest a more positive name for the process. Linked to this trend is a process of selection; while some teachers may use, say, a particular teaching method in its "pure" form, others come to incorporate bits and pieces of it into their teaching, so that elements of a rather special teaching approach become incorporated into "mainstream" teaching. Have some of these things occurred in the case of the process approach?

As a first step in this process of weighing up the current position of the process approach, and in evaluating what we have learned from it and what it can offer for the future, I attempted in November 1993 to make a survey of teachers' views by using the electronic network for ESL teachers, TESL-L. This article reports the results of that survey. First, however, I would like to discuss the use of electronic mail as a medium for survey research--a topic which, I would suggest, is particularly appropriate for an electronic journal distributed on the Internet. [-3-]

Conducting surveys on electronic networks

Internet networks offer an exciting new medium for survey work, but it is certainly not appropriate for surveys of all types. The medium has certain disadvantages in comparison with other means of making surveys; on the other hand, it also offers features which are quite unique and which can provide feedback of a type which is not easy to obtain by other means.

Given the size of TESL-L, the most obvious advantage of using the network for surveys is the number of teachers who can be reached. When the survey I describe was conducted, the number of subscribers to TESL-L was something over two thousand; at the time of writing, the number has risen to over six thousand. There is no other inexpensive way of inviting so large a number of people with a common interest to participate in a survey. However, it must be remembered that there is no reason to assume that subscribers to TESL-L form a representative sample of the world's population of ESL teachers; on the contrary, there are clearly many teachers who have neither the facilities nor the time to participate in an electronic discussion network, and thus there are some types of ESL teacher who are scarcely represented at all in the TESL-L membership, or who are seriously under-represented in proportion to their number (for example, teachers in certain countries). Despite the rapidly increasing numbers of subscribers, then, it will probably be many years before it can be taken for granted that an electronic survey will reach at least a reasonable number of teachers in all types of teaching situation. A survey of TESL-L members is not, then, a representative survey of the TESL profession.

A further problem in establishing the constituency one is surveying through an electronic network is that one cannot even estimate the size of the survey population, let alone its nature. The survey population size certainly cannot be regarded as being the total membership of TESL-L. Even if a survey is addressed to all members of the net, one cannot know how many people actually read the message asking for participation. In the case of the survey I conducted, however, not all the subscribers to TESL-L would have been even potential respondents, since the survey was addressed to teachers of writing skills, and obviously only quite a small minority of ESL teachers fall into this category. Many other surveys would face the same problem. It is thus quite meaningless to talk about a "response rate" to a network questionnaire in percentage terms, since one has no real idea of the number of potential respondents.

Nevertheless, the actual number of responses I received-- 19--seems very small in proportion to the probable number of recipients of the questionnaire. It seems likely to me that the vast majority of TESL-L subscribers are "passive" members of the network who may [-4-] read postings, but who rarely or never respond to them. This conclusion is based not only on the number of responses to my own survey, but also on reports of other information-collecting exercises which appear on the net from time to time, where again numbers of people responding seem low. The number of people who contribute regularly to net discussions is also to be counted in scores, or at most hundreds. For the great majority of subscribers, then, the "TESL-L habit" seems to involve reading, but rarely or never writing. Consequently, it seems likely that a questionnaire survey conducted over the network is likely to obtain a much smaller percentage of responses than one carried out by, for example, mailing questionnaires directly to selected persons.

What use is it, then, that a survey goes out to so many potential respondents, if relatively few choose to actually complete it? The answer is to be found, I think, in the quality of the responses obtained. The responses received to my own survey were thoughtful and full, and provided considerable insight into the opinions of individual teachers. If one can interest people sufficiently to take the time to complete a survey at all, then my own experience suggests that they will be willing to provide information in some depth.

As a means of discovering what percentage of teachers use red pens for marking, or seat their students in a particular pattern, or wear white socks, the Internet is likely to be, to put it bluntly, a dead loss. No statistically valid statements can be made as a result of surveys conducted through the medium. One can say that a certain percentage of the respondents said x or y, but since the respondents are entirely self-selected, on grounds which are unknown, from an unknown number of potential respondents, who are drawn from a population which is unrepresentative of ESL teachers in general, one has no idea at all whether the views of a majority of respondents reflect the views of a majority of teachers. But as a means of getting a sense of some of the ideas that are "around," "in the air," the Internet seems invaluable. The convenience it offers as a means of communication, where answers can be sent at the touch of a few buttons, and the general informality of network communications, are I believe factors which encourage respondents who are interested in a survey to respond openly and in some detail. The fact that responses come in within a few days or even hours of the questionnaire being sent out is a great boon to the researcher. Furthermore, surveys can easily be followed up if necessary with supplementary questions to those who did respond; this could help counteract the problem that it is not possible to test a "pilot" version of the survey form in advance.

On the basis of my own experience, then, I would suggest the following guidelines for Internet surveys:

  1. Use surveys to find out what people are thinking. Don't attempt to discover what proportion of teachers do something, or have a particular view. This means that the questionnaires used will contain mostly open-ended questions rather than "tick the box" items, unlike the forms used for many other types of survey.[-5-]

  2. People may send off a survey response simply to do a researcher a favour, but they are more likely to respond if there is some extra motivation for them to do so-- perhaps because completing the survey is interesting in itself. Bear this in mind when designing surveys.

  3. If possible, send the complete survey to the whole network. If people have to contact the researcher to ask for a survey form, then this cuts the number of likely respondents still further.

  4. But--if the survey form is to be sent out to the whole network, then net courtesy demands that it should be kept very short.

The "process approach" survey

On November 23, 1993, I posted the following message to TESL-L.

     To: Everyone involved in teaching L2 writing skills.

     In connection with a research project, I'm trying to find out
     what people now understand by the concept of a "process
     approach" to  teaching L2 writing. I've a feeling that it's
     come to mean rather  different things to different people.
     Could you help me by responding *very briefly* to this ultra-
     short questionnaire? Please reply to me personally at
     ENGTC@HUM.AAU.DK. If the questionnaire should happen to inspire
     you to make a comment on the network, could you save it until
     next week so that you don't influence other people's responses?
     Naturally I'll report back on my findings in due course.

     Many thanks in advance for helping with this.


     Tim Caudery, University of Aarhus, Denmark.


  1. Do you use, wholly or in part, a "process approach" (or whatever precise name you like to give it!) to teaching writing?

  2. If "yes," how would you define the characteristics of your "process approach" to teaching writing? (I'm not trying to get people to write extended descriptions, or crystal-clear definitions--so please don't sweat blood over formulating this!)[-6-]

    If "no," why not? (Again, no need to answer this at length-- I'm simply interested in knowing what broad characteristics you regard the approach as having which make it unsuitable for your purposes.)

The "research project" in question was a Ph.D. project on the value of making multiple drafts, and the information obtained from this survey was used in connection with that project. However, the survey threw up results which seemed very interesting, and worth reporting separately. A brief summary of these results was sent out on the TESL-L net shortly after the survey was completed; this article presents them more fully.

Question 1 was not an attempt to discover what proportion of teachers used a process approach; it was simply there so that I knew from which viewpoint respondents were writing. Of the 18 usable responses to the survey, 15 were from process approach users, and 1 came from a respondent who used at least elements of the process approach, without giving a categorical yes or no to Question 1. There were 2 respondents who gave a clear negative answer to Question 1. Obviously, teachers who did not use the process approach but had no very strong views on it were unlikely to reply to the survey, so the proportion of process approach users among the respondents was no surprise and gives no indication whatsoever of the actual proportion of writing teachers who use the process approach.

Question 2 was designed to find out what people focussed on when asked to describe or define the process approach. By discovering what aspects of the approach most immediately came to mind, I hoped to build up a picture of the image that the process approach now had for teachers. As predicted, there were considerable differences between teachers in what they regarded as being the most important elements of the approach, and I take this as evidence that, as the process approach has become more widely used, there has been divergence in interpretation of just what it is. There were also surprising omissions in the material--aspects of the process approach which no-one mentioned. While this could be coincidence, it could also be evidence that certain aspects of the process approach have tended to disappear into the background in teachers' thinking. Conversely, there were elements of the approach which were foregrounded with greater frequency than one might have expected. In all interpretations, however, I was constrained by the lack of any sound statistical basis for analysis, so the conclusions drawn must be seen merely as "straws in the wind."

Emotive and attitudinal associations

In the past, the process approach always seems to have brought out strong emotive feelings in teachers. Piper (1989) writes that [-7-]" [t]he heat which is generated by argument about a "process" approach suggests strong personal commitments to one side or the other and by extension a lack of hard evidence with which to convince the other side." The survey shows that strong feelings do still enter into discussion of process writing, but not for everyone.

Perhaps such feelings were most evident among the two respondents who did not use the process approach. In response to Question 1, for example, Respondent (R) 14 wrote "No, non, nyet, iie, la, and ALL other versions of "no" I can think of!!!!" while R1 contented himself with an electronic shudder: "No no no no no no." R1 associated the process approach with other, "touchy-feely," attitudes to teaching which he disliked, because of the general views held by the teacher trainer who had introduced him to process writing.

Among proponents of the approach, R3 might, perhaps, be said to show signs of an emotive commitment when she wrote "Metaphorically it is moving, trans-everything, going from one place to another and every point or decision along the way is as important to understand and know and master in a metacognitive way, in a self-conscious way as the product is." R17 placed importance on the ideological aspects of process writing:

I think of a process approach as a way to let students work on successive drafts of a piece of work before it receives its final evaluation. When I was in school, it seemed that most of what I wrote was immediately marked and that the schools operated as meritocracies, that is, separating people on the basis of ability or achievement. As a teacher I feel I have to help everyone and give them the opportunity to learn through their writing, not just be continually judged like horses at a finish line.

Three respondents mentioned in their definitions in some way or another that the process approach represented a contrast with approaches focusing on product, which suggests that the "process, not product" confrontational element is still significant. On the other hand, two respondents specifically mentioned that they believed that product as well as process was important.

The majority of respondents who used the process approach betrayed little emotive or ideological feeling towards it. Of course, respondents were asked to define what they understood by the process approach, not to justify its use, and perhaps some of them would have been quite militant if the question had been phrased in another way. On the other hand, there were a number of respondents who, while stating that they used a process approach, were emphatically not beating a "process approach" war-drum. Four respondents carefully qualified their affirmative responses to [-8-] Question 1; for example, R13 wrote "Yes, I think I use something like what's called a process approach to the teaching of writing," and R15 stated that "I guess I use a process approach to writing at least some of the time."

Two respondents specifically mentioned problems they perceived with the approach. Distancing himself from the "radical processors," R16 wrote:

To most of my colleagues, process means prewriting activities, brainstorming, group work, etc. Then students are supposed to write multiple drafts of personal essays, revisions and final a product. However, I often get the impression that the process is the goal ... I have taken the approach that the process is only a means to an end.

R7 carefully separated what she regarded as essential elements of the process approach from ideological elements which have come to be associated with it:

While collaboration, peer review, author "ownership" of the writing and promoting the idea that the writer should say what he/she wants to say are often associated with process writing, I don't think they are NECESSARILY part of a process writing approach. Rather, they are part of a philosophy of communication, teaching, and learning that is compatible with a process approach to writing.

The questionnaire responses suggest to me that the emotive, ideological and confrontational aspects of the advocacy of process writing techniques may be waning as the approach becomes more widely used, and as more teachers decide to adopt elements of the approach some of the time rather than embracing it in its entirety. This is likely to promote a shift in emphasis in the debate on writing techniques towards questions of what is effective in terms of learning outcomes and efficient and practical in different teaching situations.

Definitions of the process approach

The questionnaire asked respondents to give brief definitions of the process approach. It was intended that this element of the questionnaire, asking as it did for "quick responses," would work rather like a word association test, producing an indication of the concepts that were uppermost in teachers' minds when they thought of process writing. I expected that such concepts might include the research origins of the approach, the classroom activities associated with it, and the learning outcomes that it produced.

None of the respondents who used the process approach specifically mentioned research into writing processes. R7 did [-9-] mention " important underlying assumptions of the approach" which included the assumption "that the writer's ideas develop and change while writing." Six "process approach user" respondents linked their definitions in part to a concept of what writing is, thus implying to a greater or lesser degree that the process approach does reflect (and thus does originate in) what we know about the writing process itself. However, there appeared to be a majority view of "the writing process" as now being something which is more or less established, i.e. not a matter which requires any further research to demonstrate its characteristics. For example, R6 wrote: "I have the students go through the process one step at a time.... I see process writing as what a writer should be doing anyway even if it weren't required by the teacher" (my emphasis). Other respondents also talked about "the" process of writing; for instance "the process of writing, i.e. pre-writing, drafting, and finally editing" (R11; my emphasis). The uncertainties--and with them sense of discovery--about the writing process that characterized the 80s now seem behind us in the minds of many teachers.

However, it certainly does not appear that there is unanimity on the subject of what the writing process is. For while many respondents appeared to view writing primarily in terms of a sequence of "macro" stages, a smaller number implied that they regarded it slightly differently. R12 reminded us "that there is not one process. Some students need more revisions than others and some can get by with just outlining." The idea that writers use different processes was echoed by R18: "I want [my students] to become aware that writing is a messy process (for most people), and I want to help them in understanding how their own process works." But it was left to R14, a non-user of the process approach, to point out that writing research has showed up the great variation that occurs at the "micro" level in the writing process. Interestingly, this respondent felt that the process approach, as it is now taught, has become a methodology which actually contradicts what research originally revealed:

[W]riting is NOT a linear or "lockstep" progression of stages--watch an L2 adult doing some word processing for a while--the RECURSIVENESS is *phenomenal*--*RECORD* this activity (via macros/whatever)--and CORRELATE with "think-aloud" protocol analysis--then LOOK at the results-- you'll NEVER use a process approach again.

Finally, on this matter of different views of what the writing process is, I found it curious that no respondents made any mention of the fact that as ESL teachers, we are dealing with the process of writing in a second language. I will return to this point later. [-10-]

For the practising teacher, what goes on in the classroom is perhaps likely to be uppermost in the mind, and therefore it is probably not surprising that the "classroom activities" theme was predominant in the responses. For example, R10 wrote "[the process approach] involves writing in several stages: brainstorming, diagramming, outlining, and multiple drafts combined with peer and teacher commentary. In ideal circumstances, it also involves individual student-teacher conferences." R16 described the process approach not in terms of activities for individual writing projects but of the entire syllabus: "I follow a strict process approach.... [W]e spend the first week on pre-writing activities, the second week on Planning and Organizing, the third on Rough drafts, etc until we complete the process about halfway through the term, leaving time to apply this process to different projects." Almost all the respondents referred to classroom activities in their definitions; ten people defined the process approach largely in terms of classroom activities, and of these, four defined the process approach exclusively in these terms. The activities referred to related to the "macro" stages of the writing process. For some respondents, teaching writing appears to have developed into a fairly set pattern of activities in a stage by stage sequence. While the general sequence is established, the actual activities used may vary, as R6 explicitly stated: "Sometimes we do brainstorming as a class on a specific topic and everyone follows that topic up. Other times I have students work in small groups and do all the pre- writing activities together until they actually come to the first draft part. Other times it is all individual."

Seven people included fairly specific comments on learning outcomes in their definitions, and a couple of others made remarks that showed less directly that they had learning outcomes in mind. While we may assume that most respondents would probably have as their ultimate aim helping students to write better, most people who actually wrote about aims mentioned more immediate objectives, and here there were considerable differences in emphasis. R13, for example, wrote that "the emphasis in instruction is less on producing the perfect product and more on becoming aware of the various composing options"--a view clearly in line with the "process, not product" concept. Other respondents, however, placed stress on the fact that the product was also important; four people mentioned explicitly that the aim of the approach is to produce good writing, or improved writing skills. R8, after describing various classroom activities she uses, stated that "We do all this with the objective of writing a clear, well developed, logically organized, and grammatically sound final product." While three respondents stressed the need for students to become more aware of the writing process, two others emphasized the value of the process approach in helping students acquire good writing habits--and this idea was I think implied in the responses of others who stressed a stage- by- stage sequential approach to writing. Finally, there were respond- ents who noted other objectives. R13, already quoted above, [-11-] went on to mention introducing students to "the joys of playing with text," thus introducing a motivation theme which was suggested rather than explicitly explored in several other responses--for example, in R18's statement that she tried to "set things up so that students will feel in control of what they write." The concept of using "brainstorming" activities, and of encouraging thinking through process writing activities, were mentioned by several people in some shape or form.

Other issues emerged from the survey. One was practicality; R1, who did not use the process approach, gave as one reason for this that "[the approach] seems impractical for me since my classes are too large to be dealing with students' papers that are in many different stages of preparation." Evaluation was an issue for two process approach users, in addition to R17 (see quotation in the section on emotive and attitudinal associations above). R11 wrote bluntly: "One of the problems with this approach is evaluation"; R4 indicated his solution to this: "[W]hen I evaluate, I evaluate how well they have planned, the extent to which they have tried to revise, etc. And the final evaluation in the course is based on portfolios." Such an approach to evaluation might not be acceptable to all, since it seems to imply that a well-written draft which does not need much revision is less highly regarded than a poorly-written draft which is extensively revised. Finally, three respondents specifically mentioned audience in their definitions: "[We] consider questions like audience expectations" (R13), "I do everything I can to find a real audience for real papers.... the point is that the process should lead to a real outcome" (R12), "[The students] publish their writing in some form or another. Publishing could be a reading, a posting on a regular bulletin board or on an electronic bulletin board. They can send their writing electronically or through snail mail to other people to read."

Generalizing from the survey responses

It is clear, even from the small number of responses to this survey, that "the process approach" is by no means a unitary concept now, if indeed it ever was. Though the definitions given often had elements in common, there were also strong differences in emphasis, and even some completely contradictory ideas.

While some respondents held fast to the "process, not product" ideal of earlier years, explicitly down-playing the importance of final product and emphasizing the focus on the process, at least as many respondents seem to have adopted a "process AND product" approach, stressing that the writing process is a means to an end, and consequently that process- based teaching should have that end firmly in sight. People who emphasized the importance of the writing process often seemed concerned with wider educational aims, mentioning the benefits of writing as a means of promoting thinking [-12-]and thinking processes, or linking the process approach to educational philosophies. Those who mentioned the importance of product seemed in some instances to take a much more pragmatic, immediate-goal driven view. This division may to some extent have been a reflection of the widely differing types of teaching situation in which L2 teachers work, and the differing roles that they perceive themselves as having.

Another contrast in the responses was between those who focussed on the reasons for using a process approach, whether in terms of learning outcomes or educational philosophy, and those who concentrated mainly on what went on in the classroom, usually in terms of the stages which students were taken through in producing their writing. I do not intend to suggest that the respondents who wrote primarily about classroom activities would not have been able to give reasons for using the process approach; on the contrary, I am sure they could have furnished excellent reasons for their choice, if they had been specifically asked to do so. But I would suggest that the fact that so many respondents saw the approach primarily in terms of activities does indicate that the reasons for using process writing are frequently regarded by practitioners as being more or less self-evident. As any teaching approach slips from first youth into middle age, the basic assumptions behind it are often thought about and questioned less and less frequently by those who use it. That may now be happening with the process approach, and it is perhaps not an altogether healthy thing.

Practicality was an issue raised by some respondents. While some wrote in terms of the difficulties implied by a "full-scale" process approach, there was evidence that others had solved the practicality problems by simplification. Some respondents used only elements of a process approach, or used a process approach for some of the time only. In other instances it seemed that the whole class was moved through a writing process step by step, and sometimes in step with each other; individual variation in writing processes, and in particularly the micro-processes of composing, often appeared to be ignored. This approach to teaching writing is specifically criticized by some advocates of the process approach in the L1 classroom; Applebee, for example, writes that "The process approach to writing instruction has been inadequately and improperly conceptualized, as a series of activities or steps in the writing process. ... [P]rocesses are trivialized when they are divorced from the purposes they serve" (1984, p. 188). Similarly Rose, in his collective critique of twenty books for the L1 writing classroom, wrote that "In the midst of good sense I found the kinds of arbitrary prescriptions that some students could read as rigid rules. ... [D]iscussions of the composing process, outlining, revision, and modes of discourse, though tempered with qualifiers and alternatives, were often presented in a way that implied a rigidity, a fixedness that is simply not borne out in the way that writers, mediocre to talented, write" (1981, p. 67). Purves and Purves (1986, p. 175) warned that:[-13-]

[T]he currently fashionable term process brings with it some unfortunate connotations. When Emig (1971) first discussed the ways by which students wrote, she used the term composing processes, implying that there might be several sorts of processes. Her caution has been abandoned by subsequent scholars, who use the singular, process, and suggest that there is but one process operating in all writers in all sorts of context. Further, process bears the denotation that it is purely linear and highly predictable, almost mechanical. Yet Emig and many other [sic] have suggested quite the opposite.

Some people might criticize some of the questionnaire respondents for apparently having diverged from a "pure" process approach in precisely the ways warned against here. Others might claim that adaptations of the process approach are necessary and inevitable if ESL students are to examine writing processes at all. In many instances L2 writing teachers are more constrained than their L1 counterparts; there may, for example, be less time available for teaching writing, both in terms of lessons per week and overall course length. This may have encouraged and even enforced a simplification of the process approach, or the pragmatic habit of using "bits and pieces" of the approach. After all, mainstream TESL tends to be eclectic rather than dogmatic, frequently absorbing and adapting ideas in forms that teachers regard as usable in their teaching situations.


The responses I received to my questionnaire were complex, thoughtful, thought-provoking, and in many instances detailed and lengthy. They were also very varied, and trying to find common threads in them involved a great deal of "reading between the lines." I may have already done too much over- interpretation of the responses in my discussion above. To move on from that discussion to a conclusion is even more dangerous, and I do so with some trepidation.

I have already mentioned that I was struck by the fact that not one of the respondents made any mention of the fact that they were giving their views on process writing in the L2 classroom. It is clear that there are similarities between writing in L1 and L2, but there are also differences, most obvious among them being the constraints imposed by imperfect knowledge of the language code involved. Another difference, which may be equally important, is that L2 writers may have already developed successful writing processes--in L1. The work of Arndt (1987) and Hall (1990) suggests similarities in L1 and L2 writing processes of individuals, while Cumming (1988; 1989) demonstrated that the quality of writing [-14-] products in L1 and L2 were related, though the quality of the L2 product was independently affected by knowledge of L2. This evidence suggests that a) writing processes developed in L1 are transferable, for good or ill, to L2; and b) that poor L2 writing is not necessarily a sign that L1 writing processes are inadequate, but may simply be due to inadequate knowledge of L2. A poor L2 writer, then, may or may not need help to develop and improve writing processes. For some L2 writers, even though they may be producing poor products, a process approach to teaching may be irrelevant, or may need considerable adaptation. Furthermore, the fact that there are slight differences in the process that may be appropriate for writing in a language where one has native speaker knowledge and writing in a language where one has less complete knowledge suggests again that the process approach may need some adaptation for the second language classroom.

Amazingly, however, relatively little seems to have been done to develop a process approach which is specifically oriented towards L2 writing. If, for example, one examines the process writing theory outlined in the "Introduction" to White and Arndt (1991, pp. 1-10), a major book on process writing theory and practice for L2 teachers, one finds nothing (apart from the samples of student writing) which marks the book out as being intended specifically for the L2 teacher as opposed to the L1 teacher. I suggest that more could be done now towards elaborating a teaching approach and methodologies specifically for teaching L2 writing, drawing on what we have learned from writing process research and the process approach in the classroom.

The survey, I believe, suggests that the time for this may be ripe. Passions surrounding the process approach seem on the wane, suggesting that debate can be more rational than perhaps might have been the case earlier. A tradition for pragmatically adapting the process approach to the teaching circumstances of different classrooms seems to be gathering strength. On the other hand, rediscovery and reexamination of some of the research roots of the process approach may be necessary at this time to prevent writing teaching eventually becoming stultified and over-simplified as nothing more than a set sequence of teaching stages which are too inflexible to take account of the needs and existing skills of the individual.

The process approach may have a new lease of life ahead of it in the second language classroom in a renewed and reshaped form. Or the next trend in teaching writing may not be recognizably a "process approach," but may nevertheless draw on what that approach has taught us. The survey reported in this text does not show what the future will be; but I do think it suggests that change may be in the air. [-15-]


Applebee, A. N. (1984). Conclusion. In Applebee, Arthur N. (ed.), Contexts for learning to write. (pp. 183-9). Norwood, NJ: Ablex Publishing Corporation.

Arndt, V. (1987). Six writers in search of texts: A protocol-based study of L1 and L2 writing. ELT Journal 41(4), 257-67.

Cumming, A. H. (1988). Writing expertise and second language proficiency in ESL writing performance. Unpublished doctoral dissertation, University of Toronto, Ontario.

Cumming, A. H. (1989). Writing expertise and second-language proficiency. Language Learning 39(1), 81-141.

Emig, J. (1971). The composing processes of twelfth graders. NCTE Research Report no. 13.X. Urbana, IL: National Council of Teachers of English.

Horowitz, Daniel. 1986. Process, not product: Less than meets the eye. TESOL Quarterly 20(1), 141-4.

Murray, D.M. (1980). Writing as process: How writing finds its own meaning. In T.R. Donovan and B.W. McClelland (eds), Eight approaches to teaching composition (pp. 3-20). Urbana, IL: National Council of Teachers of English.

Piper, A. (1989). Writing instruction and the development of ESL writing skills: Is there a relationship? System 17(2), 211-22.

Purves, A. C. and Purves, W. C. (1986). Viewpoints: Cultures, text models, and the activity of writing. Research in the Teaching of English 20(2), 174-97.

Rose, M. (1981). Sophisticated, ineffective books--the dismantling of process in composition texts. College Composition and Communication 32, 65-74.

White, R. and Arndt, V. (1991). Process writing. London and New York: Longman.


Return to Table of Contents Return to Top Return to Main Page
© Copyright rests with authors. Please cite TESL-EJ appropriately.

Editor's Note: Dashed numbers in square brackets indicate the end of each page in the paginated ASCII version of this article, which is the definitive edition. Please use these page numbers when citing this work.