Beyond imitation – Five L2 writing teaching techniques that work, yet few Modern Language teachers use

Please note: this post was co-authored by Steve Smith of 


  1. Introduction – Why modern language teachers need to re-evaluate their attitudes and strategies about teaching writing

In these writers’ experience, many foreign classrooms instructors’ attitudes and strategies about teaching writing are more product- than process-oriented. By this we mean that explicit writing instruction – when it does occur – tends to rely mostly on

(1) ‘imitation’ – the provision of lists of model phrases/sentences which in the best scenarios are ‘drilled in’ through gap-fill practice;

(2) explicit grammar instruction which is rarely contextualized in whole-discourse practice (e.g. essay writing);

(3) learning from feedback on written output (usually a creative piece of narrative or discursive essay) – which usually occurs through annotations on margins, lists of targets or, in the best scenarios, one-to-one conferences. Feedback usually tackles all the deficit areas through one set of corrections / conference session;

(4) essay writing practice – often from day one, in the belief that practice makes perfect.

Some teachers train their students to compose in their native language first and  then  translate the L1 output thereby generated into the L2. This cumbersome process, however, must be utterly discouraged by teachers if they want students to attain some degree of fluency in target language writing and become more ‘spontaneous’ writers. Moreover, as Kobayashi and Rinnert (1992) found, this approach does indeed lead to more complex L2 output, but also lead to making many more errors than composing directly in the L2, for obvious reasons: the sentences language learners create when composing in their native language are usually too cognitively challenging and linguistically complex for their existing levels of L2 proficiency. Hence, the translations are bound to be inaccurate. Finally, it is this kind of approach which encourages less resilient and committed students to ‘google-translate’.

None of the above focuses the learners explicitly on the process of writing intended as the transformation of concepts or ideas (or ‘propositions’ as cognitive pyschologists call them) into words and syntactic structures.

One scenario in which the process of writing is indeed focused on in the UK modern language classroom refers to the teaching of higher meta-components of essay composition, e.g. planning, prioritizing, organizing and evaluating ideas. In other words: content production and organization. To our knowledge, however, even this practice is not as frequent and systematic as it should be.

Another strategy is to engage students in L2 reading in the belief that the language items in the articles or narratives they read in class or as assignments will be internalized and eventually resurface in their written pieces. This is not an erroneous assumption if students do read frequently and extensively in the target language.

Our take on the above is that these approaches usually work with the more talented, committed, self-reliant and highly metacognizant language learners, especially those who are highly proficient writers in their first language. But what is the student is not a gifted L1 writer; does not memorize the lists of connectives, model phrases and key terms her teacher diligently prepared for her whilst writing her essay; does not read extensively outside the classroom; does not spend more than a few minutes’ time – as most students do – processing her teacher’s feedback and rarely refers to the targets set for her? How do we expect such students to improve their essay writing?

  1. Strategies suggested in previous posts

In previous posts Gianfranco tackled the issue by suggesting that:

(1) teachers practice writing instruction which addresses different communicative and discourse functions/skills as discrete items in each lesson or set of lessons. The assignments set would engage students in intensive practice of those functions. So, for example, if the functions is ‘explaining’, instructors would teach a series of lessons on the relevant discourse markers (e.g. because, due to the fact that, etc.) – contextualized in the topic at hand – and provide in- and out-of-the classroom practice in those discourse markers. This would still be ‘imitation’, though, if the teachers simply provided lists and asked the learners to ‘get on with it’ and fill in a cloze text or make-up random sentences. It would focus on the process, however, if the students were asked to analyze the use of the discourse markers under study in model L2 texts; to work out different ways to convey the message contained in a sentence through using a range of discourse markers without significantly altering the meaning.

(2) teachers do not throw students in the deep end by asking them to write essay after essay from day one; but rather, that classroom and out-of-the-classroom activities focus on micro-writing, i.e. the process of writing an introduction/conclusion or developing one of the ideas brainstormed in the idea-generation phase into a paragraph.

(3) teacher feedback focus not simply on providing a correct L2 alternative to the student erroneous output (product-based feedback), but attempt to address the cognitive causes of learner deficits by collaboratively investigating the processes that underlie those deficits (process-based feedback). This entails that feedback on a piece of writing may be provided over several sessions each session focusing on different deficits identified in student output (e.g. one session on the relevance of some of the concepts selected by the students ; one on the organization of the essays; one on sentence level errors).

(4) parallel texts be used in order to raise learner awareness of the differences between L2 and L1 writing across a number of dimensions of the text. If done from the very early days of instruction, this kind of work can dispel the assumptions held by many language learners that the L2 is but a literal, word for word translation of the L1.

  1. The writing process as transformation

In this post we shall tackle the issue from a different angle: we shall focus on one aspect of L2 writer proficiency development which is often neglected by modern language teachers: the development of linguistic variety (both in terms of vocabulary and grammar structures), clarity, concision and, most importantly, syntactic maturity (i.e. the ability to produce complex sentences). The rationale for choosing these aspects of writer development is motivated by the fact that, as Phillips (1996) rightly notes:

Currently, theorists regard writing not as a product but as a continuous process of arranging and re-arranging words and syntactic structures until a writer finds the ones which best communicate the desired idea or message.

Syntactic maturity is based on the principle that mature writers tend to use more transformations in their writing and therefore write with more syntactic complexity. William Strong says “that syntactic growth (in terms of increased sentence length, depth of modification, and subordination) is a natural and inexorable feature of normal language development ” (1986). In “An I-Search Perspective on Language/Composition Research” he identifies three indices of syntactic growth:

(1) increased noun modification by means of adjectives, relative clauses, and phrases;

(2) increased nominalization in clausal, infinitive, and gerund constructions;

and (3) increased depth of modification through embedding.

In the twenty-first century class, more than ever, teachers need to identify methods for teaching writing which provide students with choice and flexibility (both lexical and structural). Why ‘more than ever’? Because in this day and age, the ‘cut and paste’ attitude to processing and sharing knowledge is rampant. Hence, our learners need to be equipped with the cognitive and linguistic tools to transform whatever knowledge they process into their own words, effectively, not merely to avoid plagiarism, but also because transformation involves higher order thinking and consequently deeper learning and greater ownership over the information being communicated.

If we, as teachers, accept this premise, then the predominantly imitative / model-based approach to writing currently in use in most UK modern language classrooms needs to be replaced by or at least supplemented with a more dynamic approach which explicitly promotes and nurtures syntactic complexity by actively engaging the learners in more than mere imitation – i.e. the sheer application of a pre-packaged model; an approach, that is, that explicitly encourages the student writer to use the model phrases/sentences provided by teachers, L2 texts or reference materials in a transformational, creative and risk-taking fashion.

4.Beyond imitation

3.1 – Sentence-combining techniques

One set of techniques that does push writing instruction well beyond the boundaries of sheer imitation and has a highly successful track-record -evidenced by scores of L1 and L2 writing research studies – is Sentence combining, defined by Phillips (1996) as

A technique of putting strings of sentence kernels together in a variety of ways so that completed sentences possess greater syntactic maturity.

In her seminal review of L1 sentence combining studies Phillips (1996) concludes that

Most of the experiments on sentence combining relate sentence combining and cumulative sentence exercises to gains in syntactic maturity.

Mounting evidence indicates that L2 student writing, too, benefits from intensive sentence combining instruction (Cooper and Morain, 1980; Enginarlar, 1994; Riazi, 2002; Juffs et al., 2014).

3.1.1 Signaled combining.

Two sentences are provided and specific instructions for sentence construction are provided. Here is an example I have used with a pre-intermediate class

I have a sister (who)

Her name is Marie

The result would be :

I have a sister who is called Marie

Signaled combining is useful when one wants to drill in a particular grammar structure or connectives in a controlled linguistic environment.

3.1.2 Open sentence combining.

In this approach,the students are not cued. For example, the kernels below

I have a sister

My sister is called Marie

She is friendly, pleasant and helpful

I argue with her from time to time

she is too talkative

could be combined as

my sister, who is called Marie, is very friendly, pleasant and helpful but from time to time I argue with her because she is too talkative

As Mellon (cited in Daiker, 1985) notes, open combining has the advantage to allow the students to learn a variety of ways ‘to transform sentences, make linguistic choices, experiment with structures and discern which sentences produce the most effective results in written language.

3.1.3 The cumulative sentence (page 13)

This approach has Robert Marzano, Joseph Lawlor, Terry Phelps, Nancy Swanson, and Dennis Packard amongst its strongest advocates. The concept of the cumulative sentence evolved from Christensen’s belief that written composition is an additive process in which a writer begins with a major idea and then adds to it so that the reader can grasp the meaning.

The cumulative sentence, says Christensen, “is the opposite of the periodic sentence. . . . It is dynamic rather than static, representing the mind thinking. The main clause exhausts the mere fact of the idea. . . . The additions stay with the main idea”. A cumulative sentence contains a main clause and several modifying clauses. Here is an example:

she came to our house

she came yesterday

she was dressed in black

she was accompanied by her brother

her brother looked sad

Could be combined as :

she came to our home yesterday, dressed in black, accompanied by her brother who looked sad

As Phillips points out, cumulative sentences encourage students to vary their output, add metaphoric descriptions, rephrase confusing periodic sentences into clearer ones and eliminate redundant elements.

 3.1.4 Whole-discourse exercises

These are more challenging but more useful if we are trying to forge effective essay writers as they do not confine syntactic transformation and manipulation to stand-alone sentences but contextualize them in the development of a concept or set of concepts. Whole discourse exercises build on the previous techniques by presenting the students with various sets of sentence kernels (Gianfranco usually uses 5 or six sets); the task: to create a sentence out of each set and then group the resulting sentences cohesively into a meaningful and logically arranged paragraph.

Mellon (1985) says that whole-discourse exercises have two benefits. The first is that by freeing students from concern with content, whole-discourse exercises help students improve their syntactic manipulations. The second is that whole discourse exercises help students improve writing both within and between sentences.

The way Gianfranco goes about creating whole discourse exercises is by decombining a paragraph from a textbook and asking the students to recombine it. Students enjoy it and learn a lot of vocabulary in the process, too.

3.1.5 Decombining

Decombining can be used as a starting point for any of the recombining activities described above, not simply for Whole-discourse exercises. In the absence of sentence combining exercises in published MFL materials, teachers can make their own by decombining sentences found in the coursebooks or L2 sources available to them.

However, decombining is a great learning activity for students, too, as by deconstructing texts they become more aware of the writing process, especially when they are required to analyze the choices made by the author. What Gianfranco normally does, is to ask the students to decombine a text in a given lesson and ask them to go back to it two or three lessons later and have a go at recombining it –without having the original in front of them, obviously.

  1. Paraphrasing

Paraphrasing is an important skill to have and one which requires transformation. It develops students’ vocabulary by forcing them to use synonyms; their grammar/syntax by often having to drastically alter the sentence structure (e.g. from active to passive voice); it may even encourage the use of metaphors, imagery, analogy and other rhetorical figures in more adventurous learners.

One very fruitful activity Gianfranco carries out with his A2 students is to ask them to paraphrase sentences which sound ambiguous or even obscure in an attempt to enhance their clarity (he provides the sentence in the L2 with the intended meaning that the author failed to express effectively next to it).

  1. Summarizing and Shrinking

Upper intermediate student-writers often lack concision. Summarizing is a very effective way to get students to learn to be concise, especially if they are given a word limit and are not allowed to repeat more than a very limited number of the language items included in the original text.

Shrinking, one of Gianfranco’s favourite writing activities, pushes the summarizing challenge a notch further by requiring the students to concentrate the meaning of a paragraph into a single sentence. A word or even character limit can be imposed, here, too. In the past Gianfranco has used Twitter for this activity – forbidding any word abbreviations/contractions or verb ellipses.

6. Tips

  1. Do not spend too much classroom time on these activities – after a few lessons in which you would have modelled how to go about these activities (e.g. using think-aloud techniques) assign these activities as homework;
  2. Distributed better than massed practice – Better do a little bit of the above every lesson, contextualized in the topic at hand, possibly after practising the vocabulary you will include in the sentence. Unless you have highly motivated or highly needy students, do not spend a whole lesson doing this – students may find it tedious. A lot of these activities make for excellent plenaries.
  3. Avoid cognitive overload – Unless you are working with very proficient L2 writers, do not use too much unfamiliar language in the to-be-combined/paraphrased/shrunk texts.
  4. Make it fun and/or competitive – sentence combining/paraphrasing and shrinking on MWBs or on Twitter under time conditions can easily be made fun and competitive.
  5. Match to ability – whereas all of the above can be used with any of your upper intermediate learners, only the easiest forms of sentence combining and paraphrasing are suitable for your intermediate learners (e.g. signaled combining)
  6. Extensive modelling – Do provide a lot of modelling before engaging the students into the more open ended of the above activities. There will be students who will find these activities very challenging. Since these are likely to be those who need this kind of practice the most, prepping them adequately so as to enhance their chances to succeed is paramount.

7. Concluding remarks

Much written instruction in UK high schools occurs through the imitation of models, feedback on writing practice and explicit grammar instruction. However, not much explicit and systematic effort is made to develop variety, clarity, concision and syntactic maturity, the ability to produce sentences that are longer, contain complex subordination and deeper modification. Yet, the attainment of these four goals is a must if we aim to forge effective writers and communicators in general.

Another reason to focus on the development of L2-learner ability to transform and manipulate language effectively refers to the fact that a lot of 21st-century-student learning occurs through the digital medium, mostly on the Internet. Hence, today’s language learners need, more than ever, to be able to transform whatever L2 knowledge they find on internet based sources into their own words not only to avoid plagiarism but also to make it their own.

Sentence combining, Paraphrasing, Summarizing and Shrinking hold the potential to enhance these areas of L2 learner writing proficiency. The effectiveness of the sentence combining techniques discussed above is supported by a vast body of research evidence. As for the other three we could not locate any substantial research evidence. However, they worked well for us – as language learners – and for many of our students over the years.

We believe that if students received systematic practice in activities of this kind from their pre-intermediate days all the way to GCSE (intermediate to upper intermediate level), the notoriously huge gap between learner writing proficiency at the end of KS4 (14-16 years old) and the level of competence needed at KS5 (16 to 18 years old) would be significantly reduced.

More on this topic can be found in ‘The Language Teacher Toolkit’ , the book Steve Smith and I co-authored, published on


10 thoughts on “Beyond imitation – Five L2 writing teaching techniques that work, yet few Modern Language teachers use

  1. […] The researchers investigated the effect of grammar instruction involving sentence combining tasks on the essay writing of 130 third quarter students of French. The subjects were divided into two groups: the experimental group received 60 to 150 minutes instruction per week through sentence combining exercises whilst the control group was taught ‘traditionally’ through workbook exercises. The experimental group outperformed the control group on seven of the nine measures of syntactic complexity adopted. Although the study did not look at the overall quality of the informants’ essays but only at the syntactic complexity, its findings are very interesting and has encouraged me to incorporate sentence combining tasks more regularly in my teaching strategies. Here is an discussion of the merits of sentence combining instruction and how it can be implemented […]

    Liked by 1 person

Leave a Reply to Gianfranco Conti, Phd (Lang. Ed.), MA (TEFL) Cancel reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s