An approach to creative speaking activities in the young learners’ classroom

An approach to creative speaking activities in the young learners’ classroom

Carmen Becker* & Jana Roos**


The main focus of early foreign language learning in Europe is on the development of oral skills. In the classroom, speaking is usually reproductive and imitative and activities aim at the production of closely supported accurate output. Opportunities for young learners to experiment with the language and to use it productively outside of fixed dialogues are often rare. However, developing fluency and basic speaking competencies requires more than just producing memorized chunks of language. In order to progress in their acquisition of the target language and to become truly communicatively competent, learners also need to be able to use language spontaneously and creatively. This paper proposes an approach to creative speaking activities. It discusses communicative tasks and improvisation activities in relation to their potential to initiate language use that transcends formulae and reproduction. Finally, it develops a model of creative speaking that illustrates how young learners can be supported in gradually developing the skills for a more independent language use.

Keywords: communicative tasks, language acquisition, formulaic expressions, experimental language use, model of creative speaking

Citation: Education Inquiry (EDUI) 2016, 7, 27613,

Copyright: © 2016 Carmen Becker & Jana Roos. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License, allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.

Published: 4 March 2016

Correspondence to: Carmen Becker, TU Braunschweig, Englisches Seminar, Bienroder Weg 80, 38106 Braunschweig, Germany. Email:

*TU Braunschweig, English Seminar, Braunschweig, Germany. Email:

**Paderborn University, Department of English and American Studies, Paderborn, Germany. Email:

©Authors. ISSN 2000-4508, pp. 9–26


Throughout Europe, the main goal of foreign language learning in primary schools is the development of communicative competence, with an emphasis on the oral skills of listening and speaking (Enever 2011). In the classroom, young learners should be supported in developing a repertoire of vocabulary and fixed expressions to be used in role play and topic-based situations and that serve as a basis for further learning and language use. Especially at the beginning, speaking in this context is usually reproductive and imitative, giving young learners the possibility to practise language patterns and to master basic communicative situations such as, for example, presenting themselves with the help of simple rote-learned phrases. This goal is also reflected in English textbooks for these young learners, in which the majority of speaking activities aim at the production of guided accurate output (e.g. Becker, Gerngross and Puchta 2013; Hollbrügge and Kraaz 2006). Such an activity may take the form of a dialogue, for instance, for which the learners are provided with questions and answers that they can modify by inserting different lexical material into open slots in the given sentence structures. These types of activities can be very motivating, because they allow young learners to actively participate in the lesson and to interact successfully in the target language at an early point in the learning process.

Although it is generally accepted that interaction in the target language can facilitate the acquisition process, and there is increasing evidence that this is also the case for children’s language learning (Mackey, Kanganas and Oliver 2007, 288), opportunities for young learners to creatively experiment with the language and to go beyond the fixed dialogues are rare in the early foreign language classroom. This is also confirmed by Mitchell and Lee (2003), who compared learning cultures in British and Korean early foreign language classrooms. They came to the conclusion that learner interaction in both contexts was restricted to the use of prefabricated expressions, and that creativity “involved either selection of prefabricated exponents from material already practised, or oral ‘gap-filling’ (inserting one’s own lexical choice into a prefabricated grammar pattern)” (Mitchell and Lee 2003, 55).

As a consequence, the young learners’ chances to set up and test hypotheses and to gain first insights into the target language system are often limited. Research on the development of young learners’ spoken skills in the primary school context shows that the learners’ target language production is often limited to the use of formulaic sequences. In an extensive study of German primary school learners of English, Engel and Groot-Wilken (2007) found that after 2 years of instruction, the children could use prefabricated expressions in conversational interaction, but that they often had difficulties in constructing their own sentences (cf. also Engel 2009). In a combined cross-sectional and longitudinal study that examined the oral performance of young German learners of English at primary school, Lenzing (2013) comes to similar conclusions. Here, the children’s speech production was also characterised by the use of single words and formulaic sequences (cf. also Roos 2007). However, in her study, Lenzing was also able to show that a slow but gradual development towards less formulaic speech and more productive utterances took place after two years of instruction (Lenzing 2013, 362; cf. also Lenzing 2015). Referring to similar findings in their review of research on early foreign language learning in Europe, Edelenbos, Johnstone and Kubanek (2006) highlight “the value of helping children progress beyond prefabricated utterances” (Edelenbos, Johnstone and Kubanek 2006, 9).

As the development of fluency and basic speaking competencies requires more than just producing memorized chunks of language, learners need opportunities to use language spontaneously under what Thornbury (2005a, 13) calls ‘real operating conditions’, in order to progress in their acquisition of the target language and to become truly communicatively competent. In this context, Engel and Groot-Wilken (2007, 27) also point out that using the target language creatively may be an important motivational factor for the children: “In order to keep the joy, enthusiasm and ambition of young learners alive, it is important to provide them with more opportunities to discover and experiment with the language (…) (and) to give more attention to the creative and productive learning processes.” A key question is therefore how it is possible to integrate such opportunities into lessons for young learners. Here, the use of more open communicative activities seems to be a promising approach. Appropriate activities could include communicative tasks and improvisation activities, as they have the potential to initiate language use that transcends formulae and reproduction.

In this paper, we aim to contribute to the understanding of communicative competence in the young learners’ classroom by introducing a research-based model of creative speaking. In the first part, communicative tasks and improvisation activities are discussed as possible means to create opportunities for creative speaking in language lessons with young learners. At first, results from research into task-based work with young learners, as well as examples of their task-based language production, are examined. Then, improvisation activities are considered as a more open and flexible alternative to communicative tasks, which also provides various possibilities for children to work independently with the language material that is already available to them. In the second part of the paper, the model for a creative speaking approach is presented. It is based on the idea that it is important to gradually help young learners develop the necessary skills for more independent language use in the classroom. The model illustrates how activities such as tasks and improvisation activities can be approached stepwise in a way that allows children to rely and build on previously learned expressions and vocabulary and to use language beyond previously taught structures.

Using communicative tasks with young learners

A task-based approach has great potential when the goal is to engage learners in conversational interaction that allows them to creatively construct their own utterances. One definition of a ‘communicative task’ that summarises the main characteristics commonly attributed to tasks is the one offered by Ellis (2009, 223). He basically describes a communicative task as a meaning-focused activity, which involves a need to convey information and enables learners to use the linguistic means available to them in order to work towards a clearly defined outcome (Ellis 2003, 2009, 223). Thus, communicative tasks promote negotiation of meaning and communicative interaction in situations in which the focus is on task completion. Learners can benefit from the interaction that results from task-based work, because they are exposed to meaningful input and receive feedback on the language they produce as well as opportunities for producing modified output (Long 1996; Mackey 1999; Swain 1993).

Research on young learners’ task-based interaction

While the use of communicative tasks in language pedagogy and second language acquisition has been widely researched with older learners (for an overview see e.g. Ellis 2003; Lightbown and Spada 2013), there have been comparatively few studies with young learners (Pinter 2006). One reason may be that communicative tasks that create contexts for spontaneous speech are often thought to be too difficult for young beginning learners. However, research has shown that using tasks with children can be beneficial to their language development in many ways. Based on the observation that language practice in primary school contexts is often centred on pattern practice or memorizing prefabricated expressions, Pinter (2007) examined 10-year-old Hungarian learners of English who repeatedly worked with information-gap tasks. These tasks allowed the learners to “express their own meanings in a less restricted manner” (Pinter 2007, 189). The aim of the study was to observe changes in the learners’ performance and the gains task-based interaction with this age group can lead to. The repeated use of tasks not only resulted in a more fluent language use, it also led learners to respond to and to assist each other in the interaction, whereas initially they simply displayed their knowledge of English. These results provide evidence for the positive effects of using tasks with young learners. However, when comparing the task-based interactions of young learners to those of adults, Pinter (2006) also found that older learners handled the demands of a picture-differences task more effectively. They focused on problem-solving and completing the task, whereas the 10-year-old children who were observed mainly seemed to name items in their pictures, which did not prove equally effective. These differences may be attributed to the young learners’ cognitive development, but Pinter also points out the possibility that the children’s approach is an effect of familiar classroom practices, namely “the dominant classroom discourse of naming things that the children are so used to” (Pinter 2006, 626).

Mackey, Kanganas and Oliver (2007) also examined the benefits of task-based work with young learners. Based on the assumption that the feedback learners receive and pay attention to in task-based interaction can facilitate the acquisition process, they studied the relationship between task familiarity and the interactional feedback provided to young learners. The participants were 7- to 8-year-old children in a migrant ESL setting in Australia, who worked both with familiar and unfamiliar tasks. Both types of tasks lead to corrective feedback and interactional modifications, even though not always in a similar fashion, and thus to a kind of language use that is regarded as beneficial to the acquisition process. Of course, the migrant ESL setting in which the study was carried out is different from the European early foreign language learning contexts described in Pinter’s study or the other studies referred to in the first section of this chapter. Mackey, Kanganas and Oliver (2007, 288) themselves point out that in the existing studies on child interaction, contextual differences or individual learner variables may have led to mixed results. However, they nevertheless emphasise “the increasing evidence that interaction does play a role in children’s second language learning,” which is a key aspect that is shared by the different studies referred to here.

Overall, the research that is available at this point shows that young learners can benefit from task-based work in different ways. As communicative tasks are not yet a common component of textbooks for English in primary school, one possibility for teachers is to modify the activities that they offer, in order to integrate opportunities for a spontaneous and creative language use into lessons for young learners.

Tasks as opportunities for creative speaking

The following activities and excerpts from transcripts illustrate how a guided speaking activity can be modified and how turning it into a communicative task increases the learners’ possibilities of using the language more flexibly and creatively. Figure 1 shows a speaking activity from a textbook published in 2003 (Gerngross and Puchta 2003, 246). Here, learners look at a picture of a monster and have to describe it by using the given phrase “My monster has got ….” The learners are supposed to complete the sentence by filling different lexical material into the empty slot at the end, for example, “My monster has got three arms.” Thus, the learners can vary the language they use within narrowly defined limits. Ten years later, in 2013, a modified version of this activity was published in the edited version of the same textbook (Becker, Gerngross and Puchta 2013, 49). As Fig. 2 illustrates, the learners are still supposed to describe a monster and to do so by using the same sentence pattern as before. What is different, however, is that the activity has now turned into an information-gap task. Two learners have to work together and use the target language to exchange information in order to be able to find five differences. While the learners are still told how to describe the monster, they do more than simply reproduce a specific pattern and display language, as was the case in the first version of the activity (cf. Fig. 1). They now use the language in order to come to a shared understanding and a common solution, which adds a focus on meaning to the resulting interaction. This modification of the activity may reflect a general tendency in the ongoing development of English language education in primary schools, namely to provide young learners with more opportunities to interact in meaningful communicative situations (cf. e.g. Engel 2009).

Fig 1
Figure 1. Activity taken from the textbook Playway (Gerngross and Puchta 2003, 35), designed for German learners of English in primary school. (Reproduced with permission from Helbling.)

Fig 2
Figure 2. Activity taken from the textbook Playway (Becker, Gerngross and Puchta 2013), designed for German learners of English in primary school. (Reproduced with permission from Helbling.)

In their research on young learners’ language development, Roos (2007) and Lenzing (2013) used a modified and even more open version of the task presented above in order to elicit spontaneous speech data from 8- to 10-year-old German learners of English. This kind of data reflects “the learners’ capacity to utilise their interlanguage (IL) grammar under the time constraints of spontaneous oral language production” (Pienemann 1998, xvi). Therefore, it naturally contains errors and thus provides insights into the learners’ current stage of interlanguage development. In the studies, the format of an information-gap task was maintained, as the learners were each given a picture of a monster and had to find five differences between the pictures by talking to each other. In contrast to the previously described textbook activities, however, no predetermined language was given. Rather, the task allowed the children to rely entirely on their own linguistic repertoire. The following examples of learner interactions show how this may stimulate the creative and productive use of the language. In Excerpts 1 and 2, two pairs of learners in Grade 3 (8–9 years old; 1 year of instruction) communicate in order to solve the task:

S1 Have you monsters {ne*} ice cream?
S2 Yes.
S1 Have your monster blue?
S2 Yes. Is your monster (ehm) (/) (eh) {hat**} your monster green (eh) eggs? (#) ah ears.
S1 Yes.

S3 Do you three ears? (eh) two ears?
S4 Yes. (#) Have*n* you (eh) (###) Have*n* you (eh) (#) one nose?
S3 Yes. (#) Have you woolly hat?
S4 (Hm) no.

(/) = interruption; {} = German words/phrases; # = short pause; ### = long pause; * = an; ** = has

(The examples come from data collected for a cross-sectional study by Roos (2007); they were also included in a longitudinal study by Lenzing (2013).)

In Excerpt 1, S1 does not seem to be sure about the right choice of verbs to be used when asking questions. This leads to non-native-like utterances and even the use of a German verb form (hat – 3rd person singular of ‘have’). In addition, the use of the verb ‘have’ in a context that requires the verb ‘be’ and the fact that it is used repeatedly in a specific pattern may indicate that the former is part of an unanalysed chunk, which the learner uses to ask questions (cf. Lenzing 2013). Still, producing a number of errors does not keep the learner from being communicatively successful. Similarly, in Excerpt 2, the two learners experiment with the language material that is available to them in order to ask questions and finally find one difference between their pictures.

In Excerpt 3, more advanced learners in Grade 5 (10–11 years old; 2.5 years of instruction) work with the same task. What they say shows that the task can be used with learners at different competence levels, because the learners can rely on their own linguistic repertoire and can solve the task according to their individual level of proficiency.

S5 Has your thing on your picture one (um) (/) three ice (/)
S6 Balls.
S5 (…) balls on the ice?
S6 Yes. Have your monster (um) one nose?
S5 Yes. Has your monster (#) seven teeth in the (/) (in the) (#) mouth?
S6 Yes. (um) (#) Has your monster a {Bommel*} on the hat?
S5 No. (um)
S6 No?
S5 No. Okay. Has (/) (has) the ice balls there are orange, red, and yellow?
S6 No. They are green, orange, and yellow. {Grün**}.
(Unpublished data by Roos) * = pompom; ** = green

This excerpt shows that the learners support each other when solving the task: When S6 does not know how to say ‘scoops of ice cream’, S6 proposes the word ‘balls’ that S5 subsequently integrates into his utterance. The example also highlights the perception that tasks like the one used here can provide opportunities for negotiation of meaning. This is the case when learner C6 uses a clarification request (‘No?’) to make sure that the monster displayed on his partner’s picture does not have a pompom on its hat. The interactional moves of C6 indicate that he seems to be sensitive to the difficulties his partner is experiencing and proactively offers further support. He continues to facilitate meaning by adding a translation of the word ‘green’ – which closely resembles the German word – without being asked for it.

In summary, the examples above illustrate what happens if learners are not provided with fixed expressions, but have to rely on their existing linguistic repertoire. Here, “(e)vidence that children work naturally with rules and patterns comes from their creative productions of utterances” (Cameron 2001, 102). The data show many elements of authentic, natural interaction, such as repetitions, pauses and false starts that would not be found in the same way in a scripted dialogue. The kind of interaction illustrated above may often require a certain tolerance of ambiguity, but even though the learners are at a very basic level of competence and make several errors, they do not appear to have any difficulty in speaking to and understanding one another. In the classroom, the errors that the learners produce provide opportunities for corrective feedback, in ways that have been shown to drive second language acquisition forward. At the same time, the errors they produce can provide the teacher with information about the learners’ developing interlanguage.

The learners’ performances in this task demonstrate that more open communicative activities may push learners to work at the limit of their oral skills, but that, at the same time, this contributes to their creativity in finding new ways of expressing meaning. By making the most of the language that is available to them, they can explore the new language and extend their communicative competence. Another way to stimulate this kind of creative use of the target language in the young learners’ classroom is the use of more open-ended improvisation activities.

After examining the data on creative speaking tasks and identifying its potential for language development in the young learners’ classroom, open-ended creative drama activities were explored through classroom observations. The main purpose of those observations was to examine whether improvisation activities stimulated a creative use of the target language and whether they also had any potential for developing the young learners’ speaking skills.

Improvisation activities in the primary English classroom

Stinson and Winston (2011, 481) define improvisations as involving “spontaneous, active interactions that often simulate real-life events.” According to Phillips (1999, 6), improvisation and drama activities have a number of advantages for foreign language learning. One of the most important factors is that make-believe as a form of playful drama and dramatization is part of “children’s lives from an early age.” Phillips (1999, 6) notes that children begin naturally imagining and playing ‘being adults’ in concrete everyday situations at the age of three. Children play different roles, “they rehearse the language and the ‘script’ of the situation and experience the emotions involved, knowing that they can switch back to reality.” The natural desire to play and act can provide a springboard for speaking in the early language classroom. Thornbury (2005b, 96) stresses that real-life situations can be simulated through improvisations and a wider range of registers “practised than normally available in classroom talk as, for example, in situations involving interactions with total strangers.” Additionally, Zafeiriadou (2009, 6) emphasises that children who are an active part of an imaginative play situation and interact with each other usually use language communicatively. By regularly engaging in playful interactions in their first language, children have developed the competence to communicate verbally even with limited language. Therefore early foreign language learners already have a general competence in turn-taking but also in using non-verbal communication such as body movements, gestures and facial expressions, which can easily be transferred when engaging in drama activities in the foreign language, making drama activities a powerful tool for giving children the opportunity for communication and interaction in the early foreign language classroom (Phillips 1999, 6). Research has also shown that engaging in drama activities can have a considerable positive impact on language acquisition. Stinson and Winston (2011, 499) report that drama in foreign language classrooms can contribute “to a range of positive results including improved spontaneity, fluency, articulation and vocabulary.” They see the advantage of using drama activities as having a high impact on affective factors such as increasing student motivation and confidence (Stinson and Winston 2011, 482). Phillips (1999, 6) emphasises that open-ended improvisation activities allow learners to freely interpret a certain role and use the language repertoire available to them. Therefore “the same activity can be done at different levels at the same time, which means that all children can do it successfully and experience a high degree of autonomy which may increase the learners’ self-efficacy and motivation” (Phillips 1999, 6). Moreover, working towards a performance as a clear goal can have a very positive impact on the motivation of the learners (Phillips 1999, 6).

With regard to confidence, drama activities are seen as a means to lower students’ stress. “By taking on a role, children can escape from their everyday identity and lose their inhibitions. This is useful with children who are shy about speaking English, or don’t like joining in group activities” (Phillips 1999, 7) and can increase the oral participation and active involvement of inhibited learners. This potential is also underlined by Sağlamel and Kayaoğlu (2013), who found that drama activities can have a positive effect on the foreign language learner’s anxiety level.

The above-discussed research findings on drama and improvisation activities are also reflected in classroom practice in the early foreign language classroom in Germany. Here, drama activities are commonly used, and text types like mini-dialogues or action stories for listening and scripted speaking as part of follow-up activities are often integrated into textbooks (Becker, Gerngross and Puchta 2013). However, not many textbook seem to have integrated activities for non-scripted drama activities yet. One example among other language courses that contains improvisation tasks is the German edition of Playway (Becker, Gerngross and Puchta 2013), a language course for 8–10-year-old German learners of English as a foreign language at primary level. As an additional resource for learning, it contains posters (cf. Fig. 3) that can be used to introduce new words or to initiate improvisation activities for language acquisition. Becker, Gerngross and Puchta (2013) developed and trialled open-ended improvisation activities based on these posters with two classes of 8- to 10-year-old German learners of English after 1 year of instruction. In order to carry out the improvisation activity, the children formed groups and then chose a scene from the topic-based poster depicting various scenes in everyday situations. As a first step, the learners used the scene as a stimulus and staged it as a freeze frame. Then they spontaneously developed and improvised a dialogue matching their scene within a 5-minute time limit. Finally the children voluntarily presented their scenes to the class or produced a video recording with a flip-cam. The teacher saved the recordings in order to be able to give a personalized individual feedback. The learners could put copies of the scenes as work samples in the dossier of their portfolio, which gave them the opportunity to document their progress in speaking.

Fig 3
Figure 3. Family poster (Becker, Gerngross and Puchta 2013), designed for German learners of English in primary school. (Reproduced with permission from Helbling.)

The results from classroom observations cannot be generalized beyond the particular context and the conclusions drawn here should be considered as rather tentative. However, the data collected by Becker, Gerngross and Puchta (2013) show a number of interesting results that will need to be verified through further studies. The 5-minute time limit seems to encourage the children to focus on the task and to speak spontaneously, therefore contributing to their spontaneity as suggested by Stinson and Winston (2011). The primary learners seem to transfer their competencies from dramatizing in playful situations in their first language and are able to communicate non-verbally as described by Zafeiriadou (2009, 6). The observed beginning learners communicate with limited language, often speak in one-word sentences and often use onomatopoeia as well as gestures and facial expressions to convey meaning. More advanced children seem to make use of their individual language repertoire of memorized chunks of language or experiment with and generate new language by creatively combining and using chunks learned in a different context or making up new words such as propeller hat (cf. Fig. 4) by guessing intelligently and sleep shirt for ‘pyjamas’ by direct transfer from their L1.

Fig 4
Figure 4. The ‘propeller hat’ (Becker, Gerngross and Puchta 2013). (Reproduced with permission from Helbling.)

More advanced learners do not seem to rely on the use of facial expressions, gestures, variation in their voice and onomatopoeia to interact as much as beginning learners seem to do. As the children become more experienced with 5-minute improvisation activities, they seem to gain more confidence and increase their creativity. The following transcript gives an example of the language produced by the 9-year-old German learners. They are acting out a scene that is depicted on the poster above (cf. Fig. 3), in which a family is getting ready for a picnic:

C1 (grandmother) Let’s have a picnic!
C2 (boy shooting ball) Shoot!
C3 (grandfather) Ahhhh! I’m a super *goal-man
C4 (girl in red shirt) Ahhh.
C5 (mother) No. I don’t like it. *No play here.
(Becker, unpublished data)

The transcript shows that the learners use memorised chunks of language such as ‘Let’s have a picnic’ and ‘I don’t like it’, new word creations such as ‘*goal-man’ instead of ‘goalkeeper’ and onomatopoeia as in ‘ahhh’, but also start experimenting with language in ‘*No play here’ instead of ‘Don’t play here’.

The classroom observations suggest that improvisation activities share common features with communicative tasks as defined by Ellis (2003, 2009, 223). In improvisation activities as well as in tasks, learners need to focus on meaning; they convey information by using their individual linguistic repertoire and work towards a clearly defined outcome by spontaneously interacting with peers. In improvisations learners have a choice of what to say and how to say it. From a beginning level on, learners are therefore provided with opportunities for creative as well as productive language use and talk management as in turn-taking (cf. Thornbury 2005b, Zafeiriadou 2009). The classroom observations also seem to support Sağlamel and Kayaoğlu’s (2013) findings regarding anxiety levels. Putting oneself in an imaginary situation and speaking while pretending to be someone else as required in improvisations also seems to lower general inhibitions toward speaking. Of course, the activities that are selected always need to be adjusted to the individual learners’ needs and abilities. Since the observations do not reveal any use of repair strategies in improvisation activities, more data need to be collected and analysed to examine their potential for self-monitoring.

An approach to creative speaking activities – a model

In order to become truly communicatively competent, learners should be provided with manifold opportunities to become autonomous language users through activities that support their natural desire to interact with peers and allow them to make use of their “rich resources of imagination, creativity, curiosity, and playfulness” (Zafeiriadou 2009, 6). On the basis of the findings discussed above, a creative speaking approach was developed with the aim to promote productive language use in the young learners’ classroom. It is illustrated in the model in Fig. 5 and is described in the following sections.

Fig 5
Figure 5. An approach to creative speaking. (Becker and Roos 2015)

Level I: Reproductive language use

The model is organised across three levels. At the first level, reproductive language use, it integrates activities that foster reproductive language use, which is considered an important foundation of creative speaking. At this level, the learners use fixed expressions in order to be able to communicate successfully. Activities are guided as well as closely linked to the input given in class. They mainly promote imitation and therefore include saying rhymes and chants, singing songs and retelling stories or parts of stories and scripted acting and speaking of dialogues, role plays and sketches. They also include other speaking activities such as guided information-gap activities, where the language to be used is fully supplied. Guided activities are very motivating because they allow beginning learners to speak imitatively and to actively participate from a beginning level on. At the same time the imitative and repetitive character of the activities minimises the possibility of making mistakes, which may strengthen the learners’ self-confidence. Furthermore, the activities at the first level provide learners with a “means of gaining articulatory control over language” (Thornbury 2005b, 64), including pronunciation, intonation and sentence stress.

Activities that foster reproductive language use, foster noticing and help learners memorise vocabulary and chunks of language as well as structures and discourse gambits. Thornbury (2015b, 64) stresses that a combination of reproductive activities helps learners to increase their ‘fluency store’ by developing ‘islands of reliability’. Figure 6 shows a scripted opinion-gap activity, which can be assigned to Level I in the model. Here the learners talk about likes and dislikes and the language to be used is fully provided.

Fig 6
Figure 6. Fully scripted activity (Becker, Gerngross and Puchta 2013, 27). (Reproduced with permission from Helbling.)

Level II: creative language use

The next level, Creative Language Use, allows learners to ‘practise control’ (Thornbury 2005b, 63) over their individual language repertoires. Thornbury (2005b, 63) describes “practised control” as “demonstrating progressive control of a skill where the possibility of making mistakes is ever-present, but where support is always at hand.” The main goal of practising control is to support appropriation of the target language. According to Thornbury (2005b, 63), “[…] learning a skill is not simply a behaviour (like practice) or a mental process (like restructuring) […]. Central to the notion of a transfer of control is the idea that aspects of the skill are appropriated. Appropriation has “connotations of taking over the ownership of something, of ‘making something one’s own’” (Thornbury 2005b, 63).

In order to facilitate appropriation, the learners have to independently perform and creatively combine fixed expressions, but can still rely on support through the provision of partial scaffolding in form of phrases they are provided with. Examples of such activities include partly scripted guessing games and information-gap activities, story skeletons and gapped songs, chants or rhymes. In order to fill the gaps in the supportive framework, the learners will have to draw on their individual word store. This requirement adds challenge and an element of choice and self-determination, but is still highly predictable. It therefore encourages the learners’ independence. Successfully carrying out an activity with a reduced supportive framework may also promote the learners’ sense of self-efficacy and confidence (Puchta 2007, 1).

Figure 2 shows an example of a partly scripted picture differences task. The learners work in pairs. Both children exchange information in order to find differences. In order to carry out the activity, the learners can rely on the scaffolding in the form of the language structures given in the speech bubbles. At the same time, the activity allows the learners to retrieve words and chunks of language from their individual word store.

Level III: Creative and productive language use

Activities at Level III promote creative and productive language use and challenge learners to use the individual linguistic repertoire available to them in a meaningful context. This means that they are free to rely on rote-learned expressions, to creatively combine them or to use language totally creatively in order to find their own ways of expressing meaning. Possible activities include non-scripted information-gap activities such as picture differences tasks (cf. Fig. 3), opinion-gap tasks, non-scripted storytelling, role play and improvisation tasks (cf. Fig. 3). All activities at that level require that the learners “[…] marshal their newly acquired skills and deploy them unassisted” (Thornbury 2005b, 13). They also need to spontaneously interact with peers, retrieve appropriate language structures, cope with unpredictability, anticipate and plan ahead. Therefore, the learners are challenged to perform independently and can experience a very high degree of autonomy. Partly scripted activities from Level II can easily be modified by removing the support to make them suitable for Level III. The speech bubbles from the picture differences task described above (cf. Fig. 6) could, for example, be removed, which would allow the learners to operate independently.

The reduced support and freedom of language use at Level III inevitably leads to errors. In this context, however, it is important “to see errors as evidence of learners’ progress, in the sense that they show that learners are making creative attempts to use language beyond what they have been taught” (Nicholas, Lightbown and Spada 2001, 720). The tolerance for errors should therefore be high (Puchta 2007, 2). Thornbury (2015b, 111) emphasises that the learners need to be able to “experience autonomy” and experiment with language, but also need to be provided with effective and clear as well as discreet and sensitive feedback “for the improvement of the subsequent performance.” Feedback should therefore always be given after carrying out activities at the third level. Instead of an overt correction, which can be very demotivating and inhibiting, feedback that focuses on improvement may be very helpful. A feasible approach may be to record the learners while carrying out the task. In a feedback conference, learners and teacher watch the recording. The learners identify problems in the performance and areas they wish to improve. The teacher then makes suggestions for further improvement.


The primary EFL classroom offers various opportunities for young learners to work with the target language creatively and productively. In order to exploit children’s natural potential for the development of communicative competence, creating opportunities for an exploratory and independent language use should also be taken into account in textbook and curriculum development. The approach to creative speaking activities presented here illustrates how the way towards more autonomy in language use can be prepared and encouraged and how children can be systematically guided from stages of reproductive language use towards creative and productive speaking. The research available shows that tasks and improvisation activities are not too difficult to master for young learners, as it is often assumed, but have great potential to promote their communicative competence from the beginning on.


Becker, Carmen, Gerngross, Günter and Puchta, Herbert. (2013a). Playway 3 posterset. Innsbruck: Helbling and Klett.

Becker, Carmen, Gerngross, Günter and Puchta, Herbert. (2013b). Playway 4, teacher’s book. Innsbruck: Helbling and Klett.

Cameron, Lynne. (2001). Teaching languages to young learners. Cambridge: Cambridge University Press.

Edelenbos, Peter, Johnstone, Richard and Kubanek, Angelika. (2006). The main pedagogical principles underlying the teaching of languages to very young learners, languages for the children of Europe: Published research, good practice and main principles. European Commission. (Accessed 2016-01-10).

Enever, Janet. (2011). Policy. In ELLiE – Early language learning in Europe, Janet Enever (ed.), 23–43. British Council. (Accessed 2016-01-10).

Ellis, Rod. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.

Ellis, Rod. (2009). Task-based language teaching: sorting out the misunderstandings. International Journal of Applied Linguistics, 19(3), 221–246. Publisher Full Text

Engel, Gaby. (2009). EVENING – Konsequenzen für die Weiterentwicklung des Englischunterrichts in der Grundschule. In Englisch in der Primarstufe – Chancen und Herausforderungen. Evaluation und Erfahrungen aus der Praxis, Gaby Engel, Bernd Groot-Wilken and Eike Thürmann (eds.), 197–215. Berlin: Cornelsen Scriptor.

Engel, Gaby and Groot-Wilken, Bernd. (2007). English in primary schools in North Rhine-Westphalia (NRW). In Cidree report – English in primary education in the Netherlands, North-Rhine Westphalia (NRW) and Norway, Gaby Engel, Bas Trimbos, Ion Drew and Bernd Groot-Wilken (eds.), (Accessed 2015-01-20).

Gerngross, Günter and Puchta, Herbert. (2003). Playway 4 rainbow edition. Teacher’s book. Innsbruck: Helbling and Klett.

Hollbrügge, Birgit and Kraaz, Ulrike. (2006). Sunshine 4, pupil’s book. Berlin: Cornelsen.

Lenzing, Anke. (2013). The development of the grammatical system in early second language acquisition – The multiple constraints hypothesis. Amsterdam: John Benjamins.

Lenzing, Anke. (2015). Exploring regularities and dynamic systems in L2 development. Language Learning, 65(1), 89–122. Publisher Full Text

Lightbown, Patsy M. and Spada, Nina. (2013). How languages are learned. 4th ed. Oxford: Oxford University Press.

Long, Michael H. (1996). The role of the linguistic environment in second language acquisition. In Handbook of second language acquisition, Ritchie William and Bhatia Tej (eds.), 413–468. New York: Academic Press.

Mackey, Alison. (1999). Input, interaction and second language development: an empirical study of question formation in ESL. Studies in Second Language Acquisition, 21, 557–587. Publisher Full Text

Mackey, Alison, Kanganas, Alec Peter and Oliver, Rhonda. (2007). Task familiarity and interactional feedback in child ESL classrooms. Tesol Quarterly, 41 (2), 285–312. Publisher Full Text

Mitchell, Rosamond and Hye-Won Lee, Jenny. (2003). Sameness and difference in classroom learning cultures: interpretations of communicative pedagogy in the UK and Korea. Language Teaching Research, 7 (1), 35–63. Publisher Full Text

Nicholas, Howard, Lightbown, Patsy M and Spada, Nina. (2001). Recasts as feedback to language learners. Language Learning, 51 (4), 719–758. Publisher Full Text

Phillips, Sarah (1999). Drama with children, Oxford University Press, Oxford.

Pienemann, Manfred. (1998). Language processing and second language development. Processability theory. Amsterdam: John Benjamins.

Pinter, Annamaria. (2006). Verbal evidence of task-related strategies: child versus adult interactions. System, 34, 615–630. Publisher Full Text

Pinter, Annamaria. (2007). Some benefits of peer-peer interaction: 10-year-old children practising with a communication task. Language Teaching Research, 11(2), 189–207. Publisher Full Text

Puchta, Herbert. (2007). More than little parrots. Developing young learners’ speaking skills. (Accessed 2015-01-20).

Sağlamel, Hasan and Kayaoğlu, Mustafa N. (2013). Creative drama: a possible way to alleviate foreign language anxiety. RELC Journal, 44 (3), 377–394. Publisher Full Text

Roos, Jana. (2007). Spracherwerb und Sprachproduktion: Lernziele und Lernergebnisse im Englischunterricht der Grundschule. Tübingen: Narr.

Stinson, Madonna and Winston, Joe. (2011). Drama education and second language learning: a growing field of practice and research. Research in Drama Education. Journal of Applied Theatre and Performance, 16(4), 479–488. Publisher Full Text

Swain, Merrill. (1993). The output hypothesis: just speaking and writing aren’t enough. The Canadian Modern Languages Review, 50(1), 33–39.

Thornbury, Scott. (2005a). Awareness, appropriation and autonomy. English Language Teaching Professional, 40, 11–13.

Thornbury, Scott. (2005b). How to teach speaking. Harlow: Pearson Education Limited.

Zafeiriadou, Nelly. (2009). Drama in language teaching: a challenge for creative development. ISSUES, 23, 4–9.

About The Authors

Carmen Becker
TU Braunschweig, English Seminar, Braunschweig, Germany

Jana Roos
Paderborn University, Department of English and American Studies, Paderborn, Germany

Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Related Content