PAIVA, V.L.M.O.Autonomy in second language acquisiton . SHARE: An Electronic Magazine by Omar Villarreal and Marina Kirac N. 146, ano 6, May 6th 2005.




Vera Lúcia Menezes de Oliveira e Paiva (UFMG/CNPq)


We should not only use the brains we have, but all that we can borrow.  (Woodrow Wilson)



In this paper, after reviewing some concepts of autonomy in second language acquisition literature, I discuss the construct on the light of the chaos theory. In order to provide empirical evidence for my assumptions, a corpus of 80 language learning histories, collected in Brazil, were examined and some of them were intertwined with the theoretical discussion.


 The concept


Autonomy is a concept that was first addressed in the foreign language (FL) teaching field when the communicative approach emerged. Before that there was no room for autonomy, given that the teacher used to control all the learning activities and the students’ rights were limited to the choice of the school[2], although young learners had no voice on that either. The materials were usually devised to control and guide the students who had no options on the selection of structures and vocabulary or the order they ought to be learned[3]. The teacher’s roles were to control the students and be sure that all the steps previously planned for them were being followed. The very concept of language did not leave space for autonomous mental processes[4] as language was conceived as a linear and close system which would be better learned if the students were exposed to a range of structures ordered from the least difficult to the most complex ones.

Besides the rigid structural linguistic principles underlying the teaching approaches, the role of the learners were also minimized as their different learning styles and preferred learning strategies were fully ignored and even fought against. The old approaches were teacher-centered and no autonomous learning attempt was expected. Translating lists of disconnected words or nonsense sentences; memorizing dialogues at home in order to please the teacher cannot be considered as instances of autonomy. One can say that the students who do their best at home are autonomous learners, but that does not imply that they have exercised their freedom to choose what and how to learn.

Nevertheless, shy examples of autonomy might have happened among those learners regardless of their teachers’ orientations and teaching methods. They might have used a dictionary to learn words not included in the material; they might have tried to build up other sentences which would be meaningful in their own reality, etc. In sum, they might have tried to make language work for them.

I myself started learning English by means of the Direct Method[5]. I cannot forget my first teacher saying that translation was forbidden and that we were supposed to think in English. Although that was a basic assumption in her words, there was nothing in the classroom activities that would trigger the aimed “thinking in English”. We would just repeat words and some isolated sentences without any communicative purpose. Translating was a key strategy sinfully used and a bilingual dictionary was a forbidden treasure.

I was already graduated when I decided to learn French and, again, no autonomy was encouraged. The method was the audio-visual one. The teacher would make us repeat the dialogues ad nauseum and  we were supposed to memorize them without  reading any word at all. The only aids were the pictures because the authors believed that listening should precede reading. The whole book was made up of pictures only and we were deprived from the written forms of the language we were striving to learn.

Using my autonomous instinct, I bought the teacher’s manual in order to have access to the tape scripts, but my teacher did not suspect I had done that. That was the only way I found to memorize the dialogues, as the pictures were not enough for me to retrieve the sentences associated with them. In a certain way, I was disputing the control of my own learning with the method and the teacher, which matches Wisniewska’s definition of autonomy: “Learner autonomy can be described as the ability to take control of one's own learning in order to maximize its full potential” (1998, p. 24).

My autonomy, in that case, led me to the opposite direction prescribed by the principles underlying the audio-visual method because I used the written forms to help me during the listening and practicing period[6], which was considered harmful to one’s acquisition. The more controlled the learners were the better because there would be fewer opportunities for them to make mistakes. As language learning was, at that time, a matter of habit formation, mistakes should be avoided at all cost in order to avoid imperfect linguistic behavior.

The teacher herself was not autonomous as she was also supposed to follow all the ordered steps prescribed by the method. Even the amount of vocabulary to be learned was limited and there was no space for the teacher to be creative.

In the seventies, with the emergence of a new concept of language – language as communication – and emphasis on the cognitive processes, autonomy appeared as a central feature in FL teaching. The communicative approach opened up the door for more autonomous learners, although many factors, which will be discussed later, can still prevent autonomous learning experiences.


Defining autonomy

Defining autonomy is not an easy task, mainly because there are very few contexts where the learners can actually exercise their autonomy. Seldom are the students totally free from the interference of external factors that can work as obstacles for the desired autonomy. Studying alone, for example, is not necessarily a synonym for autonomy because, as reminded by Dickinson (1987), much of the decision making and management of learning may be built into materials. For him, there are degrees of autonomy ranging from self-directed to externally directed choices involving the following factors: decision to learn; method of learning; pace; when/where; materials; monitoring; internal assessment and external assessment. I consider that other factors such as learners’ characteristics; teachers; technology; power relations; educational legislation; and other socio, cultural, economical and political aspects may also interfere in the autonomy process.

Before discussing some of those aspects, I would like to review some concepts of autonomy.  Let me start with what autonomy is not in the words of Little (1991). For him autonomy is not synonymous of self-instruction, neither something teachers do to their learners, although he does not think autonomy means that the teacher is redundant. Little reminds us that autonomy is not “a single easily described behavior” as it “can manifest in very many different ways” (p.3-4), that is, there are degrees of autonomy.

One of the most familiar definitions of autonomy is the one by Holec (1981:3). According to him, autonomy is “the ability to take charge of one’s learning”. Although it touches a central aspect of the phenomenon, I consider it somewhat naïve because we cannot think of one’s own responsibility without taking into account other factors that do interfere in the learning process. The same can be said of the definition presented by Little (1991), which reads that "autonomy is a capacity of self-direction. This capacity is exercised in the planning, monitoring and evaluation of learning activities, and necessarily embraces both the content and the process of learning” (p.4). His concept includes some metacognitive strategies that deal with different phases of the learning process – planning, monitoring and evaluation – which are central aspects in an autonomous learning process. Both authors’ definitions focus the autonomous learners as human beings free from external constraints.

Dickinson (1987) goes to the same direction and states that “an autonomous learner is one who is totally responsible for making and implementing all of the decisions concerned with his own learning” (p.9). Moreover, he emphasizes this view by adding that he sees autonomy as “a mode of learning – one in which the individual is responsible for all the decisions connected with her learning, and undertakes the implementation of these decisions” (p.27). His concept of autonomy is, in my opinion, somewhat utopian if one considers that hardly ever will language learners be able to make and implement all the decisions concerning their learning. They will, at least, depend on material written by somebody else who on their turn have already decided about the content and the linguistic and methodological principles underlying the material. Nevertheless, it seems that both Little’s and Dickinson’s definitions can apply to the highest degree of autonomy, the one which enables the learner to choose what to learn, how and when, without the constraints of any formal educational context. This idea is also present in Crabbe’s (1993) ideological argument:the individual has the right to be free to exercise his or her own choices as in other areas, and not become a victim (even an unwitting one) of choices made by social institutions” (p. 443).

That argument reminds me of the etymological meaning of the word – the “right of self-government”, as registered by the Oxford Dictionary of English Etymology and the Online Etymology Dictionary which offers the following entry: “autonomy - 1623, from Gk. autonomia, from autonomos "independent," from auto- "self" (comb. form) + nomos "custom, law." Autonomous is recorded from 1800.” [ a8etym.htm] (October 7th 2003)

This is also the way philosophy sees autonomy. “To be autonomous is to be a law to oneself; autonomous agents are self-governing agents.” (Buss, 2002).

Candy (1989) is another one to point out the menace formal education can represent to the learners’ freedom to make their own choices. For him “autonomy is an innate capacity of the individual which may be suppressed or distorted by institutional education”[7] (p.101).

Young (1986) and Pennycook (1997) follow similar line of thought, although they do not mention formal education. For Young (1986), in Pennycook (1997), autonomy means “authoring one’s own world without being subject to the will of others” (p. 35) and for Pennycook (1997) it is “the struggle to become the author of one’s own world, to be able to create one’s own meaning, to pursue cultural alternatives amid the cultural politics of everyday life” (p.39).

Freire (1997), in his book on autonomy, does not define what autonomy is, but one can infer that he understands autonomy as the learner’s capacity and freedom to construct and reconstruct the taught knowledge. Although the concept of freedom is still an important issue, Freire does not disregard the importance of the teacher whose role, in his view, is not to transmit knowledge, but to create possibilities for his/her own production or construction of knowledge.

Littlewood (1996) presents a similar definition:

We can define an autonomous person as one who has an independent capacity to make and carry out the choices which govern his or her actions. This capacity depends on two main components: ability and willingness. Thus, a person may have the ability to make independent choices but feel no willingness  to do so (e.g. because such behaviour is not perceived as appropriate to his or her role in a particular situation) Conversely, a person may be willing to exercise independent choices but not have the ability to do so (p. 428).


Willingness, for Littlewood (1996), is made up of motivation and confidence and ability is a mixture of knowledge and skills (p.431). Sheerin (1997) also points out that “it is important to distinguish between disposition and ability because a learner may be disposed to be independent in an activity such as setting objectives, but lack the technical ability…”(p. 57)

According to Littlewood, one may exhibit three types of autonomy: autonomy as a communicator (using the language creatively with appropriate communicative strategies), as a learner (engaging in independent learning using appropriate learning strategies) and as a person (expressing personal meanings and creating personal learning contexts).  I would add that, nowadays, it is important to include also autonomy as IT (Information Technology) user – ability to use the technology and solve technological problems – mainly in FL learning context where the Internet is an important tool in the learning process. Warschauer (2002) goes further and defends that “the concept of autonomy must be extended to apply not only to self-directed use of language and today’s technology but also to the ability to develop, explore, evaluate and adapt new technologies as it evolves” (p. 457).

            Most applied linguists emphasize only the second type of autonomy. Coterall (1995), for instance, defines autonomy as “the extent which learners demonstrate the ability to use a set of tactics for taking control of their learning" (p.195). Crabbe (1993) also focus the learner and says that the psychological argument – we learn better when we are in charge of our own learning – is the most appealing. Dickinson (1995), on the same track, describes “autonomy as both an attitude towards learning and a capacity for independent learning” (p.167). He quotes Wang and Peverly (1986) who concluded that


independent or autonomous learners are those who have the capacity for being active and independent in the learning process; they can identify goals, formulate their own goals, and can change goals to suit their own learning needs and interests; they are able to use learning strategies, and to monitor their own learning. (p. 167),


Similarly, Little (2003), in spite of acknowledging that "learner autonomy is a problematic term because it is widely confused with self-instruction”, states that


there is nevertheless broad agreement that autonomous learners understand the purpose of their learning programme, explicitly accept responsibility for their learning, share in the setting of learning goals, take initiatives in planning and executing learning activities, and regularly review their learning and evaluate its effectiveness (cf. Holec 1981, Little 1991). In other words, there is a consensus that the practice of learner autonomy requires insight, a positive attitude, a capacity for reflection, and a readiness to be proactive in self-management and in interaction with others.


Little (2003) adds to his concept the idea of “autonomy as communicator”, as he includes in his definition the social aspect of language learning and the interaction with others as part of the acquisition process.

            Freire (1970,1997), Young, (1986), Pennycook (1997) and Benson (1997) defend the idea of autonomy as a person, that is, autonomy as a right, implying control of one’s own learning process and respect for the learner’s identity. This critical view of autonomy aims at social transformation, freedom to think and act in order to become the author of one’s own world.

            Benson (1997) suggests “that there are, in fact, three major versions of learner autonomy for language learning (technical, psychological and political" (p.18) which he correlates with positivism, constructivism and critical theory.

The technically autonomous learners are the ones who are equipped with the necessary skills and techniques which enable them to learn a language without the constraints of a formal institution and without a teacher. The psychological version defines autonomy as a capacity for being responsible for one’s own learning, and the political version focus on the "control over the content and process of one’s own learning (Benson, 1997, p.25).”

Summing up, we can say that the concepts of autonomy focus on one or more of the following points:


ü      innate capacity

ü      a set of skills that can be learned (Benson, 1997)

ü      responsibility for one’s own learning.

ü      control over content and process, self-direction, self-management

ü      right/freedom to make one’s own choices and to construct one’s own learning


I dare say that none of the definitions listed above acknowledges the obstacles one can face when trying to be autonomous, as will be shown in the development of this paper. Kerka (1999) gathers some authors who recognize the interference of external factors on the learners’ autonomy. According to the author:


In order for self-directed learning to achieve its emancipatory potential, certain political conditions must be in place (Brookfield 1993, p. 227). Organizational culture may limit learner control over the educational environment. Marginalized or low-income groups may have limited access to learning resources (Merriam and Caffarella 1999). Gray (1999) proposes that the Internet may be one of the most powerful and important self-directed learning tools in existence (p. 120). The Internet does have the liberating potential to deliver new modes of learning; overcome resource, time, and place barriers; and equalize learning opportunities. However, gender and income imbalances still exist among users, vested interests may act to exert control over what is transmitted and who has access to it, and instructional deficiencies in online learning have yet to be adequately addressed (ibid.).


            So far, we can conclude that autonomy is not only a matter of one’s own responsibility for one’s learning and it is not “provided” by the approach or tolerated by the teacher. It is something much more complex.

Some Applied Linguistics researchers have acknowledged this complexity. Benson (1997) asserts that “the  concept of autonomy as it has developed in the field of language learning is complex and multifaceted” (p. 29) and Sheerin (1997) observes that “learner independence is a complex construct, a cluster of dispositions and abilities to undertake certain activities” (p.57). Breen and Mann (1997) advocate that we may seek to develop the ideals of autonomy in practice but, being alert to their complexities, are better able to struggle, with the constraints upon them” (p.133). Nicolaides and Fernandes (2002) agree that

autonomy seems to be an idea much more complex than one can think at first. We do not know much about its relation to the learning context and, even in a more favorable context, we do not know the most feasible way to implement it.


 Silva (2003), also talking about autonomy, reminds us that “the social systems are complex organizations in which the social and the individual are supposed to coexist” (p.74).

Having the complexity of the concept in mind and also some constraints which interfere in one’s autonomy, I will propose the following definition: Autonomy is a complex socio-cognitive system, manifested in different degrees of   independence and control of one’s own learning process, involving capacities, abilities, attitudes, willingness, decision making, choices, planning, actions, and assessment either as a language learner or as a communicator inside or outside the classroom. As a complex system it is dynamic, chaotic, unpredictable, non-linear, adaptative, open, self-organizing, and sensitive to initial conditions and feedback.

I consider autonomy a socio-cognitive system because it involves not only the individual mental states and processes, but also the social dimension if we view language as communication and not as a set of linguistic structures only. In order to learn a language, one can also use the language and develop autonomy as a communicator (See Littlewood, 1996). The different degrees of independence and control will vary according to the individual characteristics and the socio-political context.

Taking Sinclair (1997) and Karlsson (1997) as a starting point, I would like to present a summary of the different aspects of autonomy as discussed so far.

1.      Autonomy involves a capacity either innate or learned;

2.      Autonomy involves self-confidence and motivation;

3.      Autonomy involves the use of individual learning strategies.

4.      Autonomy is a process which manifests itself in different degrees;

5.      The degrees of autonomy are not stable and can vary depending on internal and external conditions;

6.       Autonomy depends on the learner’s willingness to take responsibility for their own learning;

7.      Autonomy requires awareness of the learning process;

8.      Autonomy is closely related to metacognitive strategies: planning/making decisions, monitoring, and evaluating;

9.      Autonomy has both individual and social dimensions;

10.  The teacher can help the learner to be autonomous both inside and outside the classroom;

11.  Autonomy inevitably involves a change in power relationships;

12.  The promotion of learner autonomy must take into account psychological, technical, social and political dimensions.


Autonomy as a complex system


            Not only is autonomy a complex system, but so are education and second language acquisition. Lorensen (2002) argues that

Education is an uncertain endeavor. Not only is it difficult to exactly predict what will happen in the class each day, it is nearly impossible to ascertain what the best course of education for any given person or class may be. The reasons for this are simple. Education is connected to the rest (sic) universe and as such is fully subject to the chaos that naturally exists in reality.


Larsen-Freeman (1997) had already noticed that “there are many striking similarities between the new science of chaos/complexity and second language acquisition (SLA)” and demonstrates “how the study of complex nonlinear systems casts several enduring SLA conundrums in a new light.”(p. 141). On the same direction, Finch (2002) understands that complexity theory is “offering a new description of the learning environment and providing further justification for the promotion of autonomy in language learning[8]”.

As Benson (1997) claims, autonomy is a complex and multifaceted concept (p.29). It consists of a large number of elements, which makes it difficult to be comprehensibly described by a single definition. Complexity is then the first characteristic of such a system which is also called a dynamic system. A complex system is not a state, but a process and each component of the system belongs to an environment build up by the interactions among its parts. Nothing is fixed, on the contrary, there is a constant movement of action and reaction and changes happen over time. Such system is also called chaotic. Chaos, according to Lorenz (1995), “is a standard term for non-periodic behavior” (p. 20). He explains that “in systems that are now called chaotic, most initial states are followed by non periodic behavior, and only a special few lead to periodicity” (p. 20). In complex or chaotic systems, there are periods of inertia and periods of creativity. The same way, as autonomy is a chaotic process, one can experiment periods of more or less independence and control. Only a few learners are autonomous all the time.

On reviewing the twelve aspects of autonomy, listed above, one can easily recognize some features that are natural characteristics of any complex or dynamic system, such as the idea of process in opposition to state, instability, variability, and adaptability. Thus, I would like to suggest that autonomy is a complex system nested in another complex system, the SLA system[9]. The components of a complex system are themselves complex systems. I dare say that autonomy is essentially part of SLA because it is responsible for an essential feature of that complex system – the self-organization. The learners’ autonomy may self-organize acquisition as the cognitive processes and some learning choices depend on the learners, even when they are under the pressure of highly controlled educational environment.

In order to find out how students approach language learning and if we can find evidence for autonomy, we have been collecting language learning histories[10]. In our corpus, we find lots of evidence of autonomous learning even when the learner is submitted to external control. Let us see some examples[11]:


My teacher always asked us to translate the texts and also the vocabulary exercises at home. I must admit that I acquired a certain knowledge about the language, as well as vocabulary in those times, but I got to the conclusion that if I did not studied by myself I would not learn so much. Since I always liked studying languages, especially English, I used to have a different hobby: I used to read my bilingual dictionary every time I could and also translate the lyrics of songs that I liked in order to increase my vocabulary. I decided to watch films with subtitles instead of seeing those dubbed ones. I confess that now I hate dubbed films, whatever is the original language. I also used to listen to the songs in order to get the pronunciation of some words, what did not work very well because of the different pronunciation some words have in songs. I only took English classes regarding conversation when I attended to English I classes in 2000 []


In the example above, the leaner felt that the school experience was not enough for him to acquire the language and developed his own strategies. Although one can question the efficiency of reading a bilingual dictionary, that student took control of his learning, planned what to do and got used to listening to authentic language by watching movies without subtitles. His autonomous behavior helped him become aware of some features of spoken discourse, such as pronunciation variation.

            The following example shows an unusual experience.

My first contact with English happened in 1987, when I was eleven years old. It was an English course in my neighborhood. Actually it was just an introductory course, really focused on basic English. The classroom activities followed a traditional method, by using non authentic materials, and teacher centered all the time. Then I went to high school, where English classes are simply awful. Every year the same subjects were taught to us, such as verb to be, negative forms, interrogative forms etc. However, the sport I have been practicing from that period so far is full of English words and expressions, what made me more interested in English. In fact skateboard has been a ‘catapult’ to my English learning process. It is common to meet native English speakers in skateboard contests, so I had to communicate with them in order to comment the contest, or even about my turn in it, for instance. This first steps where then, related to communicative learning process, since real use of language was required in order to communicate. Slangs and jargons were used all the time, and I did not know what exactly they meant, but I could get their meaning through the context we were in. After that, my interest have increased in many aspects of English, such as music, art and sports, what is just the continuity of the process that I began with when I was a child. []


In this history, we can find an undergraduate student, a prospective English language teacher. As so, he is able to use some academic jargon in order to reflect upon his language history. We can see the tension between the high school teacher’s concept of language, grammar structure, and the leaner awareness that one learn a language by using it, by communicating. His chance to use the language was not something planned, it was much more a product of his desire to communicate with his skate partners. If in the first example we have a student displaying a certain degree of autonomy as a language learner, in the second one, we have another narrator highlighting his ability as an autonomous communicator, one who can use effective strategies to successfully interact and develop his acquisition process.

Autonomy is a property of a complex system, in our case SLA, because it changes for reasons that are, usually, entirely internal to itself, such as willingness to learn in a more independent way. Autonomy is thus a key feature in SLA and, as we could see in the two examples, it triggers the learning process.

The complex systems are also, dynamic, non-linear, unpredictable, open, adaptative, self-organizing, fractal, and sensitive to initial conditions and to feedback (Gleik,1987; Lewin,1992; Lorenz,1995).

A dynamic system continuously changes over time, often as the result of feedback, and adapts itself to the new environment, learning from its experience. The changes are non-linear as the effect is not necessarily proportional to the cause. They are chaotic because the system is apparently disordered, although there is an underlying order in this apparent disorder. Nothing is determined or predicable. Just a small change in the initial conditions can drastically change the long-term behavior of a system. Kirshbaum (2002) explains that


the unpredictability that is thus inherent in the natural evolution of complex systems then can yield results that are totally unpredictable based on knowledge of the original conditions. Such unpredictable results are called emergent properties. Emergent properties thus show how complex systems are inherently creative ones.


The systems are open as new elements can enter or leave the system, and any element in the system influences and is influenced by quite a few others. McGroarty (1998), for example, acknowledges the constraints from the educational system in language learning. According to him:


The objectives, goals, and activities associated with language teaching [similarly] constrain opportunities to learn, because they adhere to conventions arising from educational and institutional history rather from the contemporary experience of learners and teachers (p.613).


My hypothesis is that, in an educational context, those and other elements can work either for or against autonomy. Consider, for instance, the following: leaner; teacher; institution; material; socio and political contexts; legislation; technology.

Caixa de texto: The learner
Caixa de texto: The teachers
Caixa de texto:      Socio, political, and
    economical context
Caixa de texto: Educational   
Caixa de texto: Technology
Caixa de texto: The school context
Caixa de texto: The input











                                 Fig. 1. Autonomy system representation


         In the autonomy system representation above, I have explicitly pointed out some elements I consider crucial when one thinks of autonomy. I am aware that it is only a graphical representation to make some aspects explicit, but they are not separate entities and one interrelates with the other. Teachers and learners aspects, for instance, are part of the school context which, on the other hand, is also part of the social context.

I do not consider autonomy as a personal trait only. It may be an innate characteristic, but it can be also repressed or fostered by interior and exterior conditions. Let us examine each of those elements of the autonomy system.


The learner

As far as the learner is concerned, the following factors might interfere positively or negatively in one’s autonomy: personality; capacity, abilities; intelligences[12], learning style; attitude; learning strategies, motivation; willingness to learn; willingness to communicate; critical sense; culture; beliefs; age; freedom; independence; metacognitive strategies; language affiliation[13]; confidence; responsibility; and previous experiences. [14]

I am not going to discuss all these factors, but just present some narratives which exemplify willingness to learn, a central component of autonomy, according to Littlewood (1996), and metacognitive strategies, which I consider of paramount importance to support learners’ autonomy.

 In the following excerpt, we can see a good example of an autonomous learner as far as willingness to learn is concerned.

Before starting studying English in the public school, I tried to learn English by myself at the age of 10. I loved songs sang in English, but I had to find out what the lyrics were telling me. Accordingly, I used a small dictionary several times and made an effort to join the sentences with the purpose of comprehending every song that I used to like. I also tried hard to copy the singer’s pronunciation of words and that helped me on identifying the same words in different songs.

I was in the 7th grade when I have my first formal English class. Although I was so excited about really studying English, I got somehow disappointed when I was told that the teacher would work just with grammar, reading and a little writing. Luckily, the teacher aimed to do more. She developed interesting projects, worked with pronunciation, used a lot of games, songs and videos. Her classes were nice, but something was missing for me. I did not want just to repeat words and sentences. I wanted to really speak English. The teacher used to ask us our opinion about a subject but comments were always in Portuguese. I wanted more. 

I continued studying by myself and at school, until I got a wonderful opportunity: I started working as a receptionist in private language school. I could study for free as I was an employee there. I took a three years course there and, in my opinion, it had a lot to do with the communicative approach. Classes were very communicative and student-centered; varied materials were applied; there was a lot of interaction - we were usually encouraged to express our point of view and give personal exemplification; (…)


The willingness to learn English appeared at the age of 10 and she tried to study the language alone. She used songs and experienced different cognitive strategies to try to understand the messages of the songs. Her desire to speak English did not find response in the formal school and the narrator compensated for that by being responsible for her own learning ( a sign of the system adaptability) until she got the opportunity to find a language school where she could develop her oral skills. It seems that the need to study alone decreased as she found a learning environment which matched her needs.

            The next language history extract is similar to the previous one, but the author makes it clear that he continued to be autonomous.

My learning history begins when I was thirteen years old. It was the first time I had contact with English, because only Spanish was offered in the 5th and 6th grades in the school where I studied. Therefore, when I was in the 7th grade and I was supposed to learn English, I got scared, because that language sounded so weird and I could not understand a word, while most of my classmates already knew it a lot. In this year, I was too shy and I could not participate in the class, since I was afraid of mistaking and sounding ridiculous to my classmates who knew at least a little English.

However, the next year. I changed my position. Somehow that weird language started to get my attention and I realized I had some facility to learn it. From this moment on, I entered in an English course and had good results. Then I decided to studied at (name of a language school is mentioned) and again I was successful. Actually, I think that these courses were a tool for me to develop my skills, but a bit part of my learning processes depended on me. I say it because I’ve always been a very shy person and to afraid of speaking in public. Thus, I just could improve my English, mainly my oral skills, by studying on my own, through songs, movies and cartoons.


In this extract, we can see that the narrator is aware of his innate capacity to learn a language. Despite the negative influence of his first experience when he could not follow his classmates, a suddenly change in his attitude (“I changed my position”) and his consciousness about his learning capacity urged him to find a better place to learn. Even though, he went on control of his own learning. His shyness and the fear of speaking in public made him look for alternative strategies to develop his oral skills.

            This example can also illustrate another feature of the chaos theory – the sensitive dependence on the initial conditions. Lorenz (2001) redefines “a chaotic system as one that is sensitively dependent on interior changes in the initial conditions[15] (p.24).” He states that

Sensitivity to exterior changes will not by itself imply chaos. Concurrently, we may wish to modify our idea as to what constitutes a single dynamical system, and decide that, if we have altered the value of any virtual constant, we have replaced our system by another system. In that case chaos, as just redefined, will be equivalent to sensitive dependence on changes that are made within one and the same system.


In the last example, it is clear that an inner change in the learner’s attitude made all the difference in his learning process. Fear and shyness were overcome by his motivation, willingness to learn, attitude, beliefs and own objectives which could not be achieved during junior high school. His changing position replaced his old learning system for a more autonomous one.

            The next two excerpts show us that, even when the narrators acknowledge the efficiency of the formal experience, some of them still perform some autonomous acts.


In 1998 I began to study in an English course. The course helped improve my accuracy but didn’t teach me communicative competencies. The material has emphasized grammar exercises and didn’t had much about formal and informal language, oral and written English and cultural context. It was up to the teacher: some of them have explained while another just have followed the book.

I have my own methodologies as listen to a lot of music in English, watch movies and TV in English. At University I discovered a great way to learn: reading. Literature makes me think about English. Also it helps me to improve vocabulary and to learn the language use.


The learner in the first example, had the capacity to evaluate the course and his needs (metacognitive strategy) and find out the best way to learn the language. He also had the insight that reading literary texts could be a good way to improve vocabulary and to learn more about language use.

I think that the structural method was quite good for me because I had the common sense and autonomy to look for other resources outside the classroom, not depending on my teacher to teach me everything I was supposed to know.


The second example was chosen so as to show how some students are consciously autonomous. In this example, the learner is aware of the limitations of the methodology used by the teacher, although she recognizes its positive features. She evaluates her needs (metacognitive strategy) and assumes she is the one responsible for her own learning and that she cannot depend only on the teacher. She knows she can also look for resources, use her own strategies and build up her own agenda for her learning process.

Finally, we have a third example of a highly autonomous learner, both as a communicator and as a learner. After his high school, he decided to go on learning by his own.


(…) In 1989, however (yes, folks, I am almost as old as a dinosaur, but I still do not bite, ok?!), I started learning English by myself, reading the magazine “Speak Up”, attending a distance short term course and exchanging letters with people from different countries. The problem was that I only practiced reading and writing; no listening, no talking at all, and such a procedure brought me some problems, some limitations, later on, when I decided to attend a “normal” course at a private English school. []


            This history demonstrates that being autonomous is not only a matter of being responsible for one’s learning because, depending on the context, it is not easy to find opportunities to develop oral skills. The students long for “communities of practice”, which, according to Murphy (forthcoming) are the social construction of “communities of practice to which they belong and to which they aspire.”  These context limitations present no solution for the learner other than enrolling in a private language schools.


The teacher

The teacher plays an important role on the learner’s autonomy development. The teacher might be qualified or non-qualified; authoritative; supportive; an advisor; a knower; a researcher; a facilitator, a consultant, a personal tutor, a helper, a counselor,  a controller, a coach, a negotiator, and in FL contexts, a good or not-so-good language model as, many times, the teacher is the only competent FL speaker the learner has contact with.

In our corpus of 80 language learning histories, when teachers are mentioned, it happens in three different ways. Most narrators just describe what the teachers did in the classroom, some praise their good teachers and others criticize their teachers. As all the narrators are prospective teachers and have already studied Applied Linguistics, they are aware of the new trends in language teaching and usually complain that they have undergone teacher-centered experiences, as one can see in the following example:

My first contact with the English language was at year seven here in Brazil. The classes were very much teacher centered and based on the PPP method. It was very boring and I didn't have any interest in learning the language.


There are also the ones who observe that their teachers do not show any autonomy and remain over-reliant on the textbook.

In high school I had a teacher called Beth, who did not do anything else but follow the book by the rules, which was reading the text, and memorizing some specific words for the quiz.


The classes were mostly the same and most of my teachers wouldn't change a comma in the lessons (they would really follow the teacher's guide).


It is not uncommon to find complaints about the teachers’ attitudes in comparison to more rewarding experiences. In the following example, the student talks about two different teachers. One who did not share the stage with the students and others who empowered the students with tasks to develop their communicative skills.


She was kind of rude with us and she thought that she was the best, that she never commited a mistake; she was the center of the class. She did not motivate us and her activities were mainly to study grammar points. 

(…) The teachers were great and they tried to motivate us as much as possible. The teachers taught us to interpret the social meaning of the choice of linguistic varieties and to use language with the appropriate social meaning for the different communication situations. They also used to teach us to understand some aspects of a culture ( people´s beliefs and values) and the main purpose of their classes was to give us the competence to be able to communicate effectively in English and not the only purpose to have a grammatical competence.



The good teacher in these students’ opinions seems to be the one who helps them develop their autonomy as communicators. This idea is repeated in the following excerpt:

(…) when i was in high school, i had good classes of english...
the teachers used to incentivate the students to speak and communicate (specially by doing pair work activities).


Teachers can inhibit students’ attempt to speak or make the students feel comfortable as described in the next examples:

My teachers used to correct me when I made my mistakes of pronunciation when I read a text; thus I didn’t want to speak in the classes.


My teacher used to lend me extra books for reading just by pleasure and I practiced a lot of my English talking to myself, but I could not talk to somebody else, because I was always afraid of make mistakes in pronunciation and very insecure.

At the university, things changed completely. The professor explained us that we also had to collect materials, practice lot of listening activities besides the ones we had to practice in class. The interaction in class motivated me and most of the students to talk a lot and the more we used to speak, the more we learned in terms of vocabulary, pronunciation, grammar, etc. She also advised us to leave grammar activities to do at home and bring doubts to class.


            One can conclude that teachers may influence the student’s development of autonomy, but even when they do not play the expected roles, an inner chaos might urge the students to take decisions  to increase their learning processes.


The input

            Several questions arise when one think of the importance of the input for autonomous learning. What kind of input is available for learners? Are there good textbooks? What are the other kinds of material available? Are students exposed to authentic communicative situations? Is there a self-access center and what does it offer to the learner?

            When self-access[16] centers emerged, it seemed that the ideal conditions for autonomy had been achieved. As Sheerin (1997:55) reminds us, “[o]ne of the main reasons for setting-up self-access facilities is to cater for learners’ individual needs. Individual learners have particular weaknesses which they may wish to work on alone or in small groups with similar needs.” Nevertheless, as pointed out by Kelly (1996) "Creating a self-access center does not in itself enable learners to become self-directed. Learners need to undergo a considerable transformation of their beliefs about language and their role as learners to be able to undertake independent learning effectively"( p.93-94).


In our language histories there is no mention to self-access centers as most of the schools do not have one. Nevertheless, students do refer to input sources other than the textbook, as we can see in the examples below:


I remember reading many things in English: from shampoos labels to whole books. I have over twenty relatives living in the US nowadays, and they’d send me many things: books, magazines, candies etc. It sure has motivated me, being curious the way I am, to understand whatever was written on these things. []


the material also got more “real” because we started to use a lot of newspaper, music, magazines, films, etc.




At this University, I had an experience to meet foreign students, talked to them and practiced my English.



Textbooks are continuously mentioned in many narratives as something students liked or disliked, but there is no reference to material that challenges them to make choices, an essential condition for autonomy. A textbook, such as the ones described in the narratives below, does not seem to contribute to one’s autonomy.

The educational books demanded accuracy in the answers and I didn’t have any context situations. The activities was always very repetitives.



The teachers used to plan all classes according to the book adopted for all the school which contained only structural exercises.



Nevertheless, students also say favorable things about their textbooks and mention reading other books. See the following excerpt:

The book we used was completely based on audiolingualism approach, but I can say that it increased more vocabulary, my knowledge in grammar; moreover, my reading and writing abilities developed since I had to write and interpret a lot of texts.  (…) My teacher used to lend me extra books for reading just by pleasure and I practiced a lot of my English talking to myself, but I could not talk to somebody else, because I was always afraid of make mistakes in pronunciation and very insecure.




If books do not attend the students’ needs, on the other hand, due to our cultural and economic dependence on the United States, privileged students can be in contact with authentic language through Cable TV, the Internet, movies and songs. As they do not have interaction with an English speaking community, they attempt do compensate for that by means of mass media. Almost all the learners, in their histories, report that they look for the language in use by listening to songs, watching movies, reading magazines.


The context

The context may foster autonomy or hinder it. There are macro and micro contexts ranging from the political, economical and macro social contexts to the micro social and educational ones such as the school, the classroom, including the teacher and the classmates.

Many questions might be asked. In which country is the language being learned? What are the political relationship among this country and the English speaking ones? Do learners have access to English speakers with whom to interact? Can they easily travel to foreign countries to practice the language? Is there any political or economical dependence relationship? Are there hard feelings or prejudice against the English speaking people? Are books and other materials easily imported? Does everyone have access to foreign language learning independent of their social class? Are there any similarities between the native and the foreign languages?

In 1993, when I visited China, for instance, I was informed that Chinese people had to ask for permission to the government in order to set up a satellite dish. Contact with English speaking TV programs was then not so easy as it was in other countries. In Brazil, textbooks for Portuguese, History, Geography, etc, are freely distributed to poor students, but FL materials are not classified as a priority. Learning a foreign language is, in fact, a commodity for higher classes, although it is an obligatory subject in high school curricula. Those are some of the examples of how political and economical context can present obstacles to autonomy.

Some institutional context features which might interfere in the learning process are: the pedagogical project, the size of the classes, the financial support for updating materials and equipments, and the investment in teachers’ continuing education.

In the Brazilian context there is a strong belief that foreign languages are not learned in high schools. In fact, most high school institutions focus only grammar and translation and, sometimes, reading. The oral skills are usually ignored. It is common sense in Brazil that if one wants to learn a language, one must go to a private language school. When we read some students’ histories, we realize that those private language schools may have an important role in one’s learning, but they are not the only factor because not all of them are acknowledged as ideal schools.

In one of the histories, the student says she had studied at three different language schools, and we can see in the selected excerpts that only one seemed to have met her communicative needs.

(…) I had a lot of Communication practice and I must say it was where I most learned, because I was supposed to talk all the time and it was not random talk: it involved a lot of real life situations, picture analysis, picture comparison, role-plays, focus on communication.

(…) Teachers spoke all the time, there were no new activities and almost all the classes concerned with grammar. It seems that they were afraid of doing something more daring, something new, because they were traditional and had a good number of enrolled students.

(…) I didn’t really studied English there, I studied how to take the test, although I learned countless vocabulary.

[ ]


            The schools can foster learners’ autonomy by offering them resource centers,  good libraries, and computer assisted language activities. The philosophical and educational principles which underlie the school pedagogic project may either leave space for autonomy or posit obstacles for more autonomous learners.

            The following example shows one student complaining about an authoritarian school.

My trajectory into English territory started many years ago while I was following 7th grade class at a public school. The class was full, about 50 students in it. Because of militarism ideology or another stupid reason the boys and girls were separated in different classrooms and even corridors. So it is easy to imagine a large group of boys in plenty energy confined to a small room and even worse, restricted to a small and uncomfortable desk. Despite talking a lot, receiving hard punishment for small things and having no rights we had no voice to complain or say nothing against anything. We had to accept the rules as they were.  []


In Brazil, as in many foreign language contexts, there are no free language courses and imported material is expensive. As we can see in the following examples, poor students face economical obstacles, but some of them appeal to their creativity and autonomy to overcome them.


Unfortunately, the lack of money was always an obstacle for having access to different idioms. []


It is interesting to see that autonomy can be the result of a non-favorable context as the one described by our next narrator.

I have NEVER had formal instructions in English before enter the college. I studied in a school where English was taught from “7ª série” on. But it was a public school and there was NO available English teachers at the time. The school staff kept telling us: “We are going to find you an English teacher, but while this does not happen, you are going to have “religion” classes to replace the English ones”. I heard that discourse the “7ª, 8ª séries”. When I started high school I thought this problem would be solved. But it was not. Hence I had no formal instructions before the college. When I decided to try “vestibular” I borrow one set of books and tapes (from CURSOS DE IDIOMAS GLOBO - CIG) and I studied by myself. The English test in “vestibular” for me was EXTREMELY hard. The things I got from CIG was not enough to cover the kind of test required in “vestibular”. Well, but I passed vestibular and enter the college. It was in 1999. []


Both students in the preceding excerpts belong to poor social layers. The former did not afford a private language course and the latter attended a high school which substituted the English classes for religious ones probably because there was no English teacher available. Our narrator borrowed some material and managed to learn enough to pass the university entrance examination.

Poor people usually do not travel and have no contact with foreigners[17]. Brazil is a continental country and seldom does anyone have the opportunity to interact with foreigners. The Internet is still not available in all public schools and underprivileged students cannot afford a personal computer linked to the Internet.

            On the other hand, we have histories of prospective teachers whose initial conditions made all the difference as we can see in the next example. The student’s mother was herself an English teacher and our narrator had the chance to get in touch with the language since she was a young child. Her environment not only offered her someone who spoke the language, but also a lot of material, opportunity to travel, and enrollment in an English course for children.

My English learning experience is quite different since I started having contact with the language when I was very young, something like two/ three years old, that is because my mother is an English teacher, so she started teaching me songs, poems, verses, prayers, etc, in English. It was so exciting! I felt like me and my mother had this secret code language that only us could speak. For sure this early learning had a great role in my future motivation to learn more and more about that “code”. My house was always full of English books and English materials in general, so when I was around 7, I started reading books and “teaching myself” with a didactic book called “Steps”.

 I joined an English school when I was 9 years old, it was called “ New Way”, and it had a fantastic environment, teacher were very well trained and we had a total emphasis in communication, both oral and written, I studied there till I was 18. This studies were essential to give me a communicative competence, specially the functional and socialinguistic ones, since they made me aware of the language and its structure. When I was 11 I went to a trip in U.S, it was a great deal for me as I could see that I was really able to communicate with natives. I still remember how excited I was because I could ask for a map in Epcot Center!  []


Another one had the chance to travel and decided to learn English when she was seven.

I started studying English when I was 7 years old. I had traveled to Disney World and when I came back, I told my mother I wanted to learn English. I’ve always studied at the same English school, (name of the language school is mentioned here). I study there until today, but now I’m taking a course for teachers.


The next narrator had the chance to interact with Americans and Brazilians who speak English.

I´ve never been in a classroom to learn English but I had hundreds of teachers. Virtually every American or Brazilian who knew more than me and with whom I came into contact was my teacher. I asked questions all the time and had a bilingual dictionary in my jacket pocket at all times. I also kept a list of words which I had difficulty remembering so that I wouldn´t have to look them up again.

My exposure to the language was pretty much the way Communicative Approach teachers try to expose students in the classroom: natural settings, real situations and everyday language and seldom using translation (especially after moving to Tulsa where I had no contact with Brazilians). 

Just as the Communicative Approach preaches I learned everything, from grammar to idioms and phrasal verbs, using them in real life settings. 



            As we could see, the context is also complex and dynamic and continuously changes over time. Different students react differently to the context constraints and adapt themselves looking for alternatives to supply what school has denied them.


Educational policies


Educational policies are another aspect to be considered. In Brazil, the national educational legislation dated back in 1986 has acknowledged the importance of autonomy in two different aspects: first, by giving the students the right to have their previous experiences incorporated into their curriculum, and second, by accepting distance education as a legal experience. Students who already master certain contents or who decide to learn them independently can apply for a test and get the right of not attending the respective courses.

Everything is interconnected, and distance education was probably recognized as a legal experience in the last Brazilian educational law due to the Internet technology. The new technology solved problems of space and time for instruction, feedback and interaction.

            Other important aspects are the school curriculum, the academic rules and evaluation systems. Autonomy has more chances to happen if the school context offers flexible curriculum, which offers the students the opportunity to choose what to learn among a wide range of different courses. Autonomy is also likely to be stimulated if evaluation system is open enough to embrace alternative assessment as portfolios and if academic rules value individual experiences, such as exchange programs and contact with proficient speakers in tandem learning programs or key pal interactions.




The use of technology can also contribute to foster autonomy. Although we know that good teachers can provide good courses with simple material, nobody would deny that technology can increase learning opportunities. Printed material, photocopies, dictionaries, visual aids, radio, cable TV, movies, songs, newspapers and magazines, videos, computers, internet tools (chat, forum, newsgroups, learning platforms, etc), software, online resources, digital corpora, DVDs, CD-roms, tape-recorders, translating machines, and language labs are examples of cultural artifacts which can empower learners in their attempt to be autonomous.


The benefits of technology integration are best realized when learning is not just the process of transferring facts from one person to another, but when the teachers’ goals is to empower students as thinkers and problem solvers. Technology provides an excellent platform – a conceptual environment – where children can collect information in multiple formats and then organize, visualize, link, and discover relationships among facts and events. Students can use the same technologies to communicate their idea to others, to argue and critique their perspectives, to persuade and teach others, and to add greater levels of understanding to their growing knowledge. (Sandholtz, Ringstaff, Dwyer, 1997, p. 176)


The Internet has been a good vehicle for communication, but we know that not everyone, in many parts of the world, is connected to the web. Students who have Internet access at home can exercise their autonomy to use additional materials and also to seek for interaction opportunities.

On-line courses are also an excellent context to develop autonomy. The way students approach this experience can show their degree of autonomy. The following utterances, from some English teachers in an on-line experience are good examples of different degrees of autonomy:


The only thing I know about computers is to send and open e-mails. That's why I am terrified about your discipline.


I'm very interested about this online course. I don't know much about computers. I hope I enjoy this course!


I will do my best to deal with the computer, because I almost know nothing about it. I'm sure this course's being on line will help me a lot. It will take me some time to feel comfortable dealing with it, but for sure I'll get it.


The examples were ordered in such a way as to show how students’ feelings range from terror, to hope and certainty. They also show that given the initial conditions (in this case, an online course), reactions are not proportional to the cause and that students react in different ways. In our examples, willingness, fear and self-confidence showed up as possible facts which might interfere in one’s autonomy.




I would like to return to my initial quotation “We should not only use the brains we have, but all that we can borrow (Woodrow Wilson)”. In this sense, I think that autonomy in language learning ideal contexts should be regarded as distributed autonomy, that is, a learner and his willingness for autonomy, sharing their achievements with other learners and borrowing theirs; teachers who are themselves autonomous and who offer the learners some choices concerning the learning activities and who accept their rights to question and  to suggest changes in the route of the course; schools which are flexible enough to accept innovative experiences and which allow teachers and learners to be the authors of the educational process; technology which provides artifacts for teachers and learners to exercise their autonomy as persons, learners, communicators, and technology users; and, finally, a fair social, political and economical system which gives every learner good learning opportunities and every teacher good teaching conditions.

No learner is omnipotent. Learners have their autonomy limited by several constraints as discussed in this paper. In formal contexts, autonomy cannot be seen as individualization, but as a possibility of sharing potentialities, as distributed autonomy. Teachers’ role would include tolerance to avoid conflict with more autonomous learners and stimulating them to share their knowledge with their classmates.

A teacher who recognizes his students’ autonomy must be prepared for a different kind of learning environment – less hierarchical, with more distributed power and more distributed autonomy – where the most creative students are the strange attractors which yield a balance between centralized management and distributed autonomy.

As Benson and Voller (1997) put it,


autonomous modes of learning imply a re-evaluation of the roles of both learner and teacher, the relationship between them, and the relationship of both to institutions of learning. These roles and relationships can be complex and are not reducible to simple expectations of behaviour or distribution of power (p.93).


The use of the Internet has brought a new dynamic and decentralized learning context. The advancement of information technology has created worlds of distributed intelligence where students are interconnected with other students with different degrees of autonomy and all of them have access to countless resources.

To give support to this discussion, I would like the readers to examine one last language learning history:


I started learning English when I was very young. I think I was 6 or 7 years old. I remember my father went to England and stayed there for three months studying English. When he was back, he started teaching us (my sister and me) words and some simple sentences like: give me water please. I remember we came to memorize over a thousand words. My mother sometimes made quizzes with us giving lollypops as prize for the one that knew all the words she asked us. She used to give us the word in Portuguese and we had to say it in English. I remember we had fun with this. Then at the age of 8 I was put in an English class with an American teacher named Henry that passed lots of time teaching the wright pronunciation of words like: world and three. At that time we live in a little town.Then we moved to Rio de Janeiro and my parents put me in (name of a language school is mentioned). I had a hard time there because there was too much grammar structures and I was supposed to study a lot which I didn't do. I hated the course and my classmates. I think I was the worse student of my class and the teacher used to look at me with a sympathetic expression but she wouldn't do anything. At the end of the year I was going to repeat the course because my grades were very bad. Then my mother said to me: this is an expensive course and we are able to maintain only one of you ( me or my sister) studying. If you fail I will not keep you there. I was so afraid of that, which I studied a lot. A cousin recorded the lesson for me and I listened to them a hundred times. I did the final test and almost took 10 in it. I think that was the time when , as Vera Menezes says in her text (Fractal Model), the organizing of chaos happened in my mind. After that experience English became something fun for things and me to study again were a lot easier from that time on. At the age of 12 the whole family went to Bolivia where I studied in a Canadian school where we had English classes every day. Each class was divided in three levels at the time of the English class and I was in the advanced one. I had no fun studying because the classes were based on the grammar and we had to memorize lots of unusual words. From all that I only remember a verb: to corroborate. When I got back to Brazil I studied in (name of a language school is mentioned) were we used to do lots of drills and controlled activities using diapositives to tell stories and then change them a bit. The time I was there I felt my English improved a lot. I stayed there for two years. Then I stopped studying English. My maintenance of the language was done through music and movies. I made some trips abroad when I had to use English and that is all.



This history reveals how unpredictable autonomy can be. The narrator’s context provided initial conditions favorable for autonomy and success. She was put in contact with English since she was very young, her father and mother knew the language and was enrolled in an English Language School when she was eight. Nevertheless, moving to Rio de Janeiro and attending a famous school brought chaos to her life and she did not feel like studying English. The edge of chaos was her mother’s threat. An inner change made her more autonomous and with the help of a cousin, reinforcing the idea of distributed autonomy, she got some recorded material and managed to get good grades. After chaos a new order appeared and learning English became a more pleasant task.

            This phenomenon is well explained by the chaos theory, as discussed by Waldrop (1992):

(…)  complex systems have somehow acquired the ability to bring order and chaos into a special kind of balance. This balance point – often called the edge of chaos – is where the components of a system never quite lock into place, and yet never quite dissolve into turbulence, either. The edge of chaos is where life has enough stability to sustain itself and enough creativity to deserve the name of life (p.11).


            In the narrative above and in many others, we could find evidence for autonomy as a complex system. There are periods of inertia and periods of creativity. Learning and autonomy are not linear processes and learners exhibit different degrees of independence. The dynamicity of the learning process and the interference of different aspects of the system bring chaos and changes happen as the result of feedback from the macro and micro contexts.

Finally, I would like to go back to the twelve aspects of autonomy which closed the first part of this paper and review them on the light of our narratives.

We have no evidence to say that autonomy is an innate capacity (1), although we cannot deny that it may exist. On the other hand, we have enough examples of autonomy as the result of adaptability to different situations, that is, as a learned capacity (2). In most language learning histories, we could easily see that self-confidence, motivation (3) gave the students the necessary affective support for them to choose their own learning strategies. It was also clear that the autonomy degree varies (4) and that some students are more willing than others to take responsibility for their own learning. It is also clear that autonomy depends on internal changes and external conditions (5). Internal changes, which can be named as the edge of chaos, can give birth to autonomy and external conditions, either favorable or unfavorable, can also lead the student to a more autonomous behavior mainly when the students are highly willing to take control of their own learning (6).

One aspect which called my attention throughout the histories was the awareness of what learning a language is (7) shared by most of the students. For them learning a language is using the language. It was a recurrent feeling that some teachers were not providing enough authentic input and also that they needed real situations to use the language. In the absence of a community of practice (see Murphey, Jin and Li-Chi, forthcoming), most of the narrators reported that they had appealed to mass media – movies, songs, videos – and, in less extent, to interaction through the Internet. This capacity of evaluating the learning process and the decisions students took were evidence of the importance of metacognitive strategies (8) for autonomous learners.

It became also clear that autonomy, although some narrators had reported that they talked to themselves, is not a matter of individualization, as the social dimensions (9) of learning were also implicit in many histories. Students reported the help of relatives, classmates, traveling experiences, and the importance of cultural production imported from English speaking countries. Teachers are also mentioned as catalysts for autonomy (10) and it happens in two opposite situations. Teachers motivate students to be autonomous by lending material, suggesting strategies, advising, giving choices, etc. Paradoxically, teachers are also catalysts of autonomy when they are not able to fulfill students’ expectations. Some of our narrators took charge of their learning process because they did not want to be limited to grammar and translation.

We can conclude that autonomy does involve a change in power relationship (11) and that autonomy must be considered in terms of psychological, technical, social and political dimensions. Our corpus of language learning histories presented many different experiences, showing that teachers do not have control over their students. Some students reveal they are able to make decisions and guide their own learning process. The learners also report circumstances where they had suffered social, economical and political constraints, although some of them were able to overcome the unfair obstacles.

            As  Waldrop (1992) puts it,

(…) these complex, self-organizing systems are adaptative, in that they don’t just passively respond to events the way a rock might roll around in an earthquake. They actively try to turn whatever happens to their advantage. Thus, the human brain constantly organizes and reorganizes its billions of neural connections so as to learn from experience (sometimes, anyway). (p.12)


The language learning histories reveal how the narrators adapted themselves to different situations. Unfortunately they did not realize that they had the right to demand more of the schools and accepted curricula which did not match their needs.

The learners do not perceive the school as a right, but as an uncontestable authoritarian entity. Fortunately, some learners undergo the disorder of chaos and look for experiences which bridge the gap imposed by formal education. Instead of passively accepting the limited curricula offered by schools, they develop their own strategies, they exercise their autonomy and become authors of their own learning histories.






Benson, P & Voller, P. (Eds.). (1997) Autonomy & Independence in language learning. London and New York: Longman.

Benson, P. (1997) The philosophy and politics of learner autonomy. In Benson, P & Voller, P. (1997) (Eds.) Autonomy & Independence in language learning. (pp.18-34) London and New York: Longman.

Blin. F. (1999). CALL and the Development of Autonomy. In Debski, R; Levy, M. (Eds.) WORLDCALL: Global Perspectives on Computer-assisted Language Learning. (pp.133-147). The Netherlands: Sweets & Zeitlinger.

Breen, M & Mann, S. (1997)  Shooting arrows at the Sun: Perspectives on a pedagogy for Autonomy. In Benson, P & Voller, P. (Eds.) Autonomy & Independence in language learning. (pp. 132-149) London and New York: Longman.

Brookfield, S. (1986) Understanding and Facilitating Adult Learning. Milton Keynes: Open University Press.

Buss, S. (2002). Personal Autonomy. In Zalta, E. N. (Ed.). The Stanford Encyclopedia of PhilosophyI (Winter 2002 Edition). Available: win2002/entries/personal-autonomy/]

Candy, P.C. (1989).Constructivism and the Study of Self-direction in Adult Learning. Studies in the Education of Adults, 21, 95-116.

Coterall, S. (1995). Readiness for autonomy: investigating learner beliefs. System,  23 (2.), 195-205.

Crabbe, D. (1993).Fostering Autonomy from within the Classroom: the Teacher’s Responsibility. System, 21 (4), 443-52.

Dickinson, L. (1991). Self-instruction in Language Learning. Cambridge: Cambridge University Press.

Dickinson, L. (1995). Autonomy and Motivation: a Literature Review. System, 23 (2), 165-174.

Finch, A. (2002). Autonomy: Where Are We? Where Are We Going? JALT CUE-SIG Proceedings, pp15-42. Available:

Freire, P. (1970). Pedagogia do Oprimido. Rio de Janeiro: Paz e Terra.

Freire, P. (1997). Pedagogia da Autonomia. Rio de Janeiro: Paz e Terra.

Gardner, Howard. (1993). Frames of Mind: The Theory of Multiple Intelligences. New York: Basic.

Gleick, J. (1987). Chaos: Making a new science. New Yok: Penguin Books.

Holec, H. (1981). Autonomy and Foreign Language Learning. Oxford: Pergamon. (First published [1979], Strasbourg: Council of Europe.)

Karlsson, L., F. (1997). From Here to Autonomy. A Helsinki University Language Centre Autonomous Learning Project. Helsinki: Helsinki University Press. Available:

Kelly, R. Language counseling for learner autonomy: the skilled helper in self-access language learning. In: Pemberton, R, et al. (Eds.). (1996). Taking control: autonomy in language learning. Hong Kong: Hong Kong University Press.

Kerka, S (1999). Self-Directed Learning: Myths and Realities. N. 3. Available:   16/10/2003

Kirshbaum, D. (2002) Introduction to Complex System. Available:

Larsen-Freeman, D. (1997). Chaos/complexity science and second language acquisition. Applied Linguistics. Oxford : Oxford University Press,  2 (18),141-165.

Lewin, T. (1992). Complexity: life at the edge of chaos. New York: Macmillan.

Little, D., & Brammerts, H. (Eds.). (1996). A Guide to Language Learning in Tandem via the Internet. Dublin: Trinity College, Centre for Language and Communication Studies.

Little, D. (1991). Learner Autonomy: Definitions, Issues and Problems 1. Dublin: Authentic.

Little, D. (2003). Learner autonomy and second/foreign language learning. Available:

Littlewood, W.(1996) “Autonomy”: an Anatomy and a Framework. System, 24, (4), 427-435.

Lorenz, E. N. (2001). The essence of chaos. Seattle: The University of Whashington Press.

Lorenzen, M. (2002). Chaos Theory and Education. Available:

McGroarty, M.(1998). Constructive and Constructivist Challenges for Applied Linguistics. Language Learning, 48 (4), 591-622.

Murphey, T.; Chen, J.; & Chen, L. (In press). Learners’ constructions of identities and imagined communities. (pp.83-100) In P. Benson & D. Nunan, (eds.). Experiences of Language Learning. Cambridge: Cambridge University Press.

Mynard, J. & Sorflaten, R.  Independent Learning in Your Classroom. Available: http://

Nicolaides, C & Fernandes, V. (2002) Autonomia no ensino de LE: uma questão cultural. In Paiva, V.L.M.O., DUTRA, D; MELLO, H.R. (Eds). Anais do VI Congresso Brasileiro de Lingüística Aplicada. ALAB. [CD-ROM]

Oxford, R. L. (2001) Language Learning Strategies. In: Carter, R.; Nunan, D. (Eds.) The Cambridge Guide to Teaching to Speakers of Other Languages. (pp.166-172). Cambridge: Cambridge University Press.

Paiva, V.L.M.O. (2002). Caleidoscópio: Fractais de uma oficina de ensino aprendizagem. Unpublished memorial, Universidade Federal de Minas Gerais, Brazil.

Pennycook, A. (1997) CulturalAlternatives and Autonomy. In Benson, P & Voller, P. (Eds.) Autonomy & Independence in language learning. (pp. 35-53) London and New York: Longman.

Sandholtz, J.H., Ringstaff, C.Dwyer, D.C. (1997). Teaching with technology: creating student-centered classrooms. New York and London: Teachers College Press, Columbia University.

Sheerin, S. (1997). An Exploration of the relationship between self-access and independent learning. In Benson, P & Voller, P. (Eds.) Autonomy & Independence in language learning. (pp.54-65) London and New York: Longman.

Silva, D.F. (2003). Propostas para a educação lingüística em ambiente informatizado. Calidoscópio, 1 (1), 69-87

Sinclair, B. (1997) Learner Autonomy: the Cross Cultural Question. In: IATEFL Issues. N.139. Available: Archives/ Texts/39Sinclair.html

Waldrop, M. M. (1993). Complexity: the emerging science at the edge of order and chaos. New York: Touchstone.

Wang, M. C. & Peverly, S. T. (1986) The Self-instructive process in classroom learning contexts. Contemporay Educational Psychology, (11), 370-404

Warschauer, M. (2002). A developmental perspective on technology in language education. TESOL Quarterly, 36(3), 453-475.

Wisniewska, I. (1998). Designing Materials for Teacher Autonomy. Forum, 36 (2), 24.



[1] I am thankful to Adail Rodrigues Junior and Júnia Braga for their qualified comments and also to the undergraduate students (Elisa, Fernanda, Lindiane and Lucas), members of my research group, who were the first readers of this paper.

[2] In my hometown, in the sixties, there were only two famous English Language Schools, one supported by the British Council and the other by the American Embassy. The former favored grammar and the latter conversation. The choice for one of them could be a sign either of the preferred skills or of the preferred accent.

[3] It seems that fewer experiences so far have given the learners the right to choose what to learn.

[4] However, it does not mean that autonomous mental processes did not happen. In fact, the ones who devised the methodologies were not aware of them.

[5] The Direct Method was developed as a reaction to Grammar Translation Method. Meaning was communicated only through the target language by means of associations of words with the classroom context, pictures and gestures.

[6] It was assumed by the audiovisual method that students should only listen and repeat during a period of 8 to 10 weeks. Reading and writing would be introduced when the basic phonetic structures had been thoroughly acquired.

[7] It is worth pointing out that other authors, as Karlsson et al (1997), believe autonomy is a capacity that can be learned.

[8]Although part 2.1 of his paper is named Complexity theory and autonomy, he does not refer to the theory to discuss what autonomy is.

[9] In Paiva (2002), I advocate that SLA is a complex system and that several theories of language acquisition are, in fact, describing different aspects of the same system.

[10] The corpus is published on the web in our project homepage:

[11] The histories are reproduced without any editing.

[12] The term “intelligences” is used in the plural, taking into account Gardner’s (1993) concept of multiple intelligences.

[13] By language affiliation, I mean the feelings the language awakens in the learner. The leaner my love or hate the language and feel positive or negative feelings. They may also look at the language and people who speak it through stereotypical lenses.

[14]  I am aware that there are many other factors, even ones we cannot think of due to the nature of the complex systems.

[15] Initial conditions, according to Lorenz (2001: 9) “need not be the ones that existed when a system was created. Often they are the conditions at the beginning of an experiment or a computation, but they may also be the ones at the beginning of any stretch of time that interests an investigator, so that one person’s initial conditions may be another’s midstream or final conditions.”

[16]  Sheerin (1997) explains that the “term ‘self-access’ refers to learning material and organizational systems (designed for direct access by users)” ( p. 54).

[17] It is worth mentioning that some poor Brazilians go to the United States to work as maids, drivers, waiters, etc, and that a few of them become English Teachers when they come back. The same happens with privileged young people who have the opportunity to stay abroad in exchange programs.