Skip to content ↓

June 2020 - Year 22 - Issue 3

ISSN 1755-9715

‘A Dove in Flight’ Agency in 21st Century Language Learning

Sharon Hartle is an associate professor in the Department of Foreign Languages and Literatures at Verona University. Specialized in ELT, she has worked for many years in the field of e-learning and specializes in multimedia material development for ELT in Blended Learning contexts. Email:



Learner agency is a term which has attracted considerable attention in academic circles. A rapid search for ‘learner agency’ on the Google Scholar search engine, in fact, returned 41,100 results (from 2016-2020) in under one second. What agency means for a learner will be considered below in greater depth. Beliefs about the learners role in the learning process can be placed alone a broad continuum. At one end there is the belief that the teacher’s role is to dispense knowledge and the learners’ a largely recpetive one. Larsen-Freeman ( 2019), in fact, makes the point that in language learning ‘perceptions of language learners as nonagentive persist. At the other end of the continuum are those who promote informal learning. A search for this term, in fact, on Google Scholar returned 128,000 results (from 2016-2020) in less than a second. Informal learning is generally linked to learning which takes place outside institutional settings and is increasingly linked with digital learning. (Godwin-Jones, 2018). Some even promote the idea that learners should be left to their own devices, taking complete responsibility for their learning (Mitra, 2013) or point out the fact that learning often takes place despite the teaching informally rather than because of it.(Holliday, 2013). If learning takes place informally, in fact, in such a way, the methodology being used by the teacher may be considered to be of little importance but the aim of this discussion is to argue the opposite, which is that precisely because learner agency is a crucial factor in learning then educators who are aware of this and are able to foster it, will be better able to mediate learning in their classes.


A Dove in Flight

In his chapter Control and Initiative: the dynamics of agency in the language classroom  (2013, p. 244), van Lier refers to Kant’s image of a dove flying through the air, that resists the pressure of the air against its wings, imagining that flying in ‘airless space’ would mean freer, faster movement. And yet if those currents and the air pressure were not there the dove would not be able to fly at all. That very air pressure or perceived constraint is, in fact, as van Lier says, ‘the resource that enables its flight’ (p. 244). In the 21st century, where access to information is for many, only a click away our learners may yearn to take flight in a ‘teacherless space’ but is this an effective approach to learning? Are learners able to take wing autonomously without the support of teachers or would such freedom mean that many would feel overwhelmed by the wealth of resources available and not even know where to start? This article looks at individual learner agency within two specific lecturer training sessions, where despite the lack of trainer expertise some learning may be said to have taken place thanks to the agency of the participants. If the trainers had been expert educators perhaps there might have been even more gain and the article goes on to explore some key ways in which support for learner initiative and agency may be provided. Successful learning requires agency, in fact, but that agency alone without guidance is not enough to foster effective learning.


Self direction and self resource centres

Agency is closely linked to a notion of self and is often related to notions of learner autonomy. Despite much debate as to what learner autonomy is a widely cited definition was provided by Holec in his report to the Council of Europe:

“Learner autonomy is the ability to take charge of one’s own learning have, and

to hold, the responsibility for all the decisions concerning all aspects of this learning…” (Holec, 1981, p. 3)

This is a tall order indeed for many learners, who are actually seeking guidance, and who may well wish to have charge of many aspects of their own learning but would like to be guided as to how to do that. Notions of learner autonomy, however, were instrumental in the popularity of self-access centres in language learning environments introduced in the second half of the last century, where institutions, for various reasons, introduced centres packed with resources for learners to chart their own learning process (Hartle, 2017). Many of these were doomed to failure, however, possibly because of a lack of expert guidance or possibly because many learners when given complete freedom simply do not know where to start and lack the framework or discipline for a sustained approach to language learning.

Informal learning in digitally mediated contexts

To return to the idea of a ‘teacherless space’ we are living in a time where communication and access to information is increasingly digitally mediated with widespread ease of access to digitally mediated knowledge, information, in an unprecedented way. This provides every single one of us with the opportunity to fly without constraints in that we can navigate the internet freely, exercising our own agency and choose what we wish to study and when we want to do so. In the light of this, together with other socio-economic changes, some consider existing education systems to be outdated, catering for the needs of the19th or 20th century industrialized societies (Robinson, 2010). The role of the individual as a self-directed agent in the learning process, but this time a digitally mediated one, has appeared once again in an increasing number of publications by those who believe in an informal, teacherless approach to education such as Salman Kahn (2011)  and Sugata Mitra ( 2013), to name only two, both of whom were widely supported by TED, and both of whom promote the notion of informal learning, where learners choose their own learning paths to solve learning problems. This is based in the notion of agency (Freeman, 2015; Van Lier, 2008) and of the importance of taking responsibility for one’s own learning as was also stressed by Holec in the definition above. The idea of being completely independent from teachers is also and at times stated explicitly. Kahn developed his popular Kahn Academy, in fact, not from the position of an educator but after developing YouTube videos to help his young relatives master mathematics and Mitra’s ‘Hole in the Wall’ experiments were introduced originally to see if under-priveleged Indian students could teach themselves to use computers, which was Mitra’s area of expertise at the time. He, in fact, set up computers on street corners in rural areas of India which lacked teachers, then simply left the children without directing them at all, in attempt to remove the teacher and to prove that children learn more effectively when their learning is self directed. He claimed that this approach was effective but Bennet (2015) pointed out the fact that many of the ‘holes in the wall’ had, just a few years later, been vandalized or were being used by older youths for purposes that are not linked to education. Another experiment which, like the self access centres, was perhaps not so effective as it initially seemed to be. Mitra, in fact, nowadays, provides some framework in his notion of self organised learning environments (SOLE) where questions are provided for children to find the answers to and they then report back to supportive ‘granny’ figures.


Does agency mean complete independence?

The children in Mitra’s context are almost completely self-directed and they learn by engaging in a process of discovery without guidance from an expert, but this may be a case of throwing the baby out with the bath water. Agency often seems to be confused with self-directed learning. Hattie and Yates ( 2014) point out the fact that the former is often conflated onto the latter, saying that there is a common belief that

“…people will learn better, with greater understanding, if they discover information for themselves, rather than being told the same information” (p. 77)

There is little basis, however, they say, suggesting that self-direction actually aids learning. This may well be true, in fact, and providing guidance to Mitra’s children may make their digital explorations even more fruitful. If you are given a choice of different sites to explore you can still choose which ones you are more interested in. Agency, in fact, in van Lier’s terms refers to much more than ‘do-it-yourself- learning discovery. The image of the dove is a powerful one because it reinforces the idea that a learner who is provided with expert guidance can make the choices he wishes to and learn what he or she wants or needs to and choice is a concept which is repeatedly associated with agency. Duff (Patricia, 2012, p. 417) describes agency as our capacity to choose, control our lives and work towards self and social transformation, where the control is provided by the element of choice. Another key concept which is related to agency is ‘relevance’, or, in the words of Lantolf and Thorne ‘the ability to assign relevance to things and events’.(James P. Lantolf & Thorne, 2006). If an activity is sensed to be meaningless, or irrelevant, in fact, or it is not clear why a learner is being asked to do it, then it will have less impact than an activity that is meaningful. Memorizing a series of unrelated words without a specific reason to do so might be an example of this. It is more likely to have greater impact if those words are related to a task that the learner needs to do. 

This ability to choose what we want to learn and what is relevant to us, in fact, is key in learning environments, not necessarily the fact of being self-directed. Individuals will exercise their agency in a whole variety of teaching contexts, and it is by no means clear that one approach is better than another, precisely because the classroom is a social space and the individual learners are motivated in different ways.


Agency at work in traditional learning contexts

Learners, in fact, will learn in various learning environments with teachers applying various teaching methodologies, and this is true not only of language learning contexts, but of many educational ones, so to illustrate this I would like to describe two examples of training sessions where I have recently participated as a learner. Neither of these contexts were organized in ways that I would consider to be optimized for learning but despite this learning took place.

The first of these two examples was a training session to about twenty lecturers on health and safety in the university. The presentation was a compulsory face to face traditional one, which followed a four-hour online course that had been completed previously. The presenter talked for three hours, including the coffee break, with a PowerPoint presentation as a visual aid. His language was both accessible and communicative but the subject matter was not particularly relevant to the real needs of lecturers present. He spent quite a long time, for instance, distinguishing between the notions of danger, risk and harm, all of which had already been covered in the online component of the course, and spent more time  highlighting the legislation, providing us with the relevant articles in the law. Observing the behavior of the participants was interesting as roughly half of them were distracted using their own computers to check email, create their own PowerPoint slides for their own lectures and so on. The only time people actually looked up and took an interest was when someone from the audience asked a question and the only slide that I remember was one that had an image of an office with several safety issues, which we were asked to identify, This anecdote shows that the agency of the participants was very much at work, and when the learners themselves took part in the discussion or were called on to act in doing something relevant to our own needs, such as identifying risk factors, we chose to participate.

Choosing to participate and interacting with the teacher actively then are both instances of agency at work within a teacher-centred framework. My personal intake was related to what I perceived to be relevant, which was where I could find the video I had to show my own students and what other training I needed to do. The relevance of the content, therefore, or lack of relevance determined whether I chose to invest mental energy into listening to and processing the information. The context of the digital course could be considered to be a prime example of a course where participants direct their own learning, because it consisted of a series of videos and readings with a final test, that could be navigated in any way the learner chose. It was, however, just as ineffective as the teacher centred session. The issue was not the learning approach, in fact, but the fact that the content was not directly relevant to the needs of the learners.

In the second example a small group of lecturers were given an hour-long training session on how to use new interactive whiteboards. The format was different from the example above because the trainer used demonstration rather than explanation. He showed the participants what could be done with each tool included in the software for the interactive whiteboard. This was still done in the style of a presentation with very limited participation from the learners, but the content was relevant. Everyone wanted to know how to use the whiteboard and the different tools. Very few participants were doing other work on their computers, and the attention levels were high. This session was short but there was a certain amount of cognitive overload, where too much information was being provided to process and recall, as so many different tools were being demonstrated. The methodology was also teacher centred, like the first example, but provided us with the expertise we needed to go away and practise using those tools independently. We were told that we could download the tools onto our own computers to experiment with them before using them in class. This training session, therefore, helped us to understand what we needed to and could do, and set the next step in our learning process, actually enabling us as learners to take the next step.

These two examples were teacher-centred but even so there was an inevitable component of learner agency, which is ultimately what makes the difference between effective and ineffective communication.


Helping the learners to take flight

If learning could take place in the two examples illustrated above, where the trainers were not, in fact educators and were not implementing strategies to support and foster learning, how much more effective can learning be in contexts where teachers actively take steps to encourage learner agency? Providing constraints in van Lier’s terms of ‘resources that enable flight’ does not mean abandoning learners to their own devices but means providing them with the tools they need to do the tasks they need to do.

Studying the grammar rules of comparative adjectives, for instance, as an end in itself would probably be counterproductive, but studying the same rules in order to carry out a task such as discussing which clothes to buy from on online store, would, for many learners be a different matter. Comparing clothes to buy is a relevant task  and studying the grammar rules of comparatives is one step along the way to being able to do this. Learners who are simply told to ‘choose some jeans to buy’would be faced with the final task without being provided with the steps to get there, and this is precisely where the expert of the teacher comes in. Hattie and Yates (op. cit. p. 73) stress the value of introducing progressive steps to enable learners to carry out a task and highlight them importance of structuring learning in such a way as to make it accessible to learners. They cite the work of Rosenshine (2012), who identifies a coherent set of functions with highly structured lessons such as:

  1. Initial review
  2. Formal presentation
  3. Guided practice
  4. Feedback
  5. Independent practice
  6. Follow-up review

Although this structure looks linear in practice it is not a question of following a recipe but of implementing these elements when required. The structure of the lesson, in this way, is not rigid, impeding flight, but can be adapted to the needs of the dove.

In the practical example of teaching comparative adjectives this may mean:

  1. Reviewing the grammar rules which have been introduced in a previous lesson and reinforcing knowledge where required; There may be other areas to clarify formally as well before moving on to practice;
  2. Providing exercises with pictures of clothing which may involve matching, gap-filling or freer production of the grammatical forms or may focus on pronunciation matters;
  3. Feedback can be provided overtly during monitoring of the exercises or covertly with notes being made for a feedback phase after exercises. Feedback in this case has the purpose not of identifying ‘problems’ as such but of identifying repair strategies or of providing information and strategies to learners which will help them progress to the next stage of the learning process and it may be delivered individually or done at group level;
  4. Independent practice means providing learners with the means to transfer these skills and practise in different ways, perhaps with larger tasks that involve a comparison component;
  5. Follow-up may involve reflection, self-assessment, further practice or review.

These stages can be moved around and split up so that the lesson becomes a cycle of learning stages aimed to help empower learners to use the language they are learning in a way which is meaningful and relevant for them.


Structure and process

The paragraph above provides us with guidelines to create a flexible structure for lessons but the process of what happens in a lesson is also important to consider. In the above-mentioned chapter van Lier also considers this aspect. This chapter, in fact, was written in a volume commemorating the work of Earl Stevick and his influence on language teaching. Stevick, as van Lier explains (op. cit, p. 242) considers the centrality of the teacher as being key, but it should not ‘conflict with the centrality of the learner’. Stevick, writing in 1980 (1980, p. 17), in fact, believes that the teacher can have nearly  100% of the control but the learners can have nearly 100% of the initiative. What this means is that the teacher may determine the framework of the lesson but or the initiative of the learners, in Stevick’s terms, means who says what, to whom and when. The focus here, then, is on speaking or contributing to a lesson, playing a ‘central, self-validating role’ (Stevick, op. cit. p. 17-19). In order for learners to be able to take such initiatives they need to be supported by teacher activity that enables them to have the confidence and the space to be able to discuss content and to explore their ideas about language as well as practising the language being presented.

Structure and process may superficially appear to be contrasting concepts but van Lier describes the two as being complimentary (op. cit, p. 246-247). In the two examples provided of training sessions for university lecturers above, there was, in fact, very little awareness by the trainers of how to mediate the communication in the classroom. They presented their content and believed that that was sufficient. This, however, enabled the participants to effectively ‘opt out’ of the lesson in the first case, where the content was non motivating or irrelevant and led to cognitive overload in the second case as there was not enough time to process the information. If the structure of the sessions had been flexible, group discussion moments could have been introduced, which does not simply mean trainer asking a question to the whole group, but small groups discussing issues with each other.

The concept of group work is by no means new but is still remarkably rare in classrooms around the globe. Hattie and Yates (op. cit., p. 45) cite the work of John Goodlad, who, in a 1994 study of over 1.017 classrooms in the USA reported that about 75% of class time was spent on teachers explaining, or asking yes/no questions to the whole class or single students, with teachers outtalking learners by a factor of three. Less than one percent of classroom time, in this study, was devoted to asking open questions that may require more complex or affective responses. Hattie and Yates underline the fact that this figure of 75% is widely cited in studies undertaken at that time. Learners in such contexts, then, are largely receptive if not passive, and yet, as Hattie and Yates emphasize (op. cit.,p. 47), ‘One of the major principles of learning is that a learner needs to make an active response to the source of learning.’

 Providing moments for experimentation and reflection in the training sessions would also have been useful. In the interactive whiteboard session, for instance, the software could have been installed on lecturers’ pcs in advance or at the beginning of the session to enable practical group work on specific tasks. Group work would also have given the instructor the chance to monitor the work being done and to supply constructive feedback designed to enable the learners to take the next step in the learning process.


In conclusion

Agency and self-direction are related to issues such as whether the learners can process everything the teacher would like them to, what they choose to focus on and what their intake is, as well as whether they will go on to experiment with the tools or language skills being illustrated. In the case of the university lecturers, those who are motivated to use such whiteboards in class will probably invest the time but those who do not will not. Other factors also come into play such as time constraints and practical considerations such as whether an interactive whiteboard can actually be used with a group of 300 students and whether it may be better with them to use a projector and a computer. The same is true of the learners. They may never need to use English outside the classroom and therefore may decide not to invest the energy into its mastery, but that is their decision and the teaching process will have at the very least provided them with knowledge and skills that they can choose to develop or not. Fostering learner agency does not mean abandoning learners to their own devices and such approaches may well be fundamentally flawed. Learner agency can be fostered by means of the expertise of the teacher, who can bring together the many participants, creating moments for initiative within the structure of the lesson and knowing how to manage the process of learning in such a way as to avoid issues of cognitive overload and minimal participation. Agency is about making choices about what you want and need to learn and how to get to the next stage on your journey. The learner, like the dove, in order to be able to reach his or her destination of choice needs support to learn how to fly.



Bennet, T. (2015). Sugata Mitra and the Hole in the Research. Retrieved from, last accessed 31.01.20

Freeman, D. (2015). Frozen in thought? How we think and what we do in ELT. UK: Iatefl Online. last accessed 31.01.20

Godwin-Jones, R. (2018). Chasing the butterfly effect: Informal language learning online as a complex system. Language Learning and Technology, 22(2), 8–27.

Hartle, S. (2017). The price of freedom: a few thoughts on learner autonomy. Retrieved from last accessed 31.01.20

Hattie, J., & Yates, G. (2014). Visible Learning and the Science of How We Learn. London and New York: Routledge.

Holec, H. (1981). Autonomy in foreign language learning. Oxford: Pergamon.

Holliday, A. (2013). ▶ TEC13 Day 02 Adrian Holliday Authenticity, Communities and Hidden Potentials. India. Retrieved from

Kahn, S. (2011). Salman Khan talk at TED 2011 (from Retrieved January 15, 2015, from last accessed 31.01.20

Lantolf, James P., & Thorne, S. L. (2006). Sociocultural Theory and the Genesis of Second Lanuguage Development. Oxford: Oxford University Press.

Larsen–Freeman, D. (2019). On Language Learner Agency: A Complex Dynamic Systems Theory Perspective. Modern Language Journal, 103, 61–79.

Mitra, S. (2013). Build a School in the Cloud. Retrieved from last accessed 31.01.20

Patricia, D. (2012). Identity, agency, and second language acquisition. In S. Gass & A. Mackey (Eds.), The Routledge Handbook of Second Language Acquisition (pp. 410–426). London: Routledge.

Robinson, K. (2010). Changing education paradigms. YouTube. Retrieved from last accessed 31.01.20

Rosenshine, B. (2012). Principles of instruction: Research based strategies that all teachers should know. American Educator, 36(1), 12–39.

Stevick, E. (1980). Teaching Language: A Way and Ways. Boston: Heinle & Heinle.

Van Lier, L. (2008). Agency in the classroom. In J.P. Lantolf & M. E. Poehner (Eds.), Sociocultural theory and the teaching of second languages (pp. 163–186). London: Equinox.

Van Lier, L. (2013). Control and initiative: the dynamics of agency in the language classroom. In J. Arnold & T. Murphey (Eds.), Meaningful Action (pp. 241–251). Cambridge: Cambridge University Press.


Please check the 21st Century Thinking Skills course at Pilgrims website.

Please check the How to Motivate Your Students course at Pilgrims website.

  • ‘A Dove in Flight’ Agency in 21st Century Language Learning
    Sharon Hartle, Italy

  • Technology in Education: How Technology Can Benefit Students and Educators
    Callum McPhillips,UK

  • The Webbing Sisters… Who Are they?
    Angela Maria Vitale, Italy

  • One Size Does Not Fit All
    Abeer Ali Okaz, Egypt