Chapter 1: Learning an additional language: Looking for universal characteristics
Introduction
This chapter begins with a whole series of studies and discussions, called second language acquisition (SLA). SLA researchers have often studied foreign language classrooms or the second language learning of children â learners and contexts that may seem very distant from the adult ESOL learners in your classrooms. However, these studies have been the source of significant theories that form part of the foundation of current research into additional language learning. In addition, much of the research and discussion about ESOL learning today assumes a knowledge of concepts that have arisen from attempts to identify and explain universal characteristics of second language acquisition. Thus in the first section of this chapter we briefly describe some of these key concepts, in order to help you to understand current research and theory.
In the second section we look at three examples of research into additional language learning that have focused on language classrooms. They have been chosen for three reasons. First, they focus on learning in classrooms, and most ESOL provision supports language learning in this context. Second, they illustrate the different approaches to classroom research discussed by Michael Breen in the second reading of this chapter. Finally, they have been chosen to help us appreciate the difficulties of tracing such a complex process.
Claims to knowledge in this area usually attract debate and critique. Much of this arises because of the need to cross traditional research boundaries in order to try to understand additional and foreign language learning. To understand the acquisition process involves, among other factors, knowledge about mental processes (cognition), about the learning context and social relations, and about how language actually operates in social practice. Thus studies by researchers expert in the way the mind works (that is, in cognitive processes such as remembering) are accused of neglecting the social situation of the learner, and those who have studied classroom relationships, or the language experiences of migrant workers, are criticized for neglecting specific cognitive processes. It is very difficult for specific research projects to cover all aspects of language acquisition.
Therefore, many questions about this language learning process remain unanswered. However, second language acquisition studies help us to understand the complexity of the process and the variables involved, so we start this chapter with some of the key concepts that have emerged both from this research activity and theoretical perspectives (for a detailed overview of SLA studies see Ellis, 2001; Firth and Wagner, 1997; Long, 1988; Mitchell and Myles, 2001).
Key concepts of SLA
This section aims to give you a brief introduction to some of the terms and ideas that you will meet when reading about SLA research and theory.
Interlanguage
Much of the earlier research in SLA was inspired by advances in knowledge of how children acquire their first language. In these first language acquisition studies investigations of the speech of young children showed that children developed their understanding and use of grammatical structures, such as negatives and interrogatives, in a particular order. It was argued that deviant grammatical forms produced by children (such as âI goed to the parkâ) were evidence of a subconscious but active mental learning process whereby children were producing speech on the basis of their current knowledge of the rules. In this example the child has overgeneralized the rule about forming the past tense by applying it incorrectly to an irregular verb.
Consequently, researchers set out to study the language produced by learners of additional and foreign languages to see if there was any similarity between first and second language acquisition. Obviously, all additional language learners produce incorrect forms when they speak, partly because of their lack of knowledge of the grammar of the new language, and partly because their knowledge of their first language can lead them to hypothesize incorrectly about similarities in structure. However, would a study of learner language over time show systematic and patterned changes in the errors made that could indicate an active universal process, despite differences caused by first language interference?
At first, these studies did indeed indicate the possibility of there being an order in the acquisition of certain structural features universal to all learners of second languages. But this was an enormous task to carry out, and convincing evidence has not emerged.
What these studies of learner language did do, however, was to establish the concept of a dynamic and systematic process of language learning, and the language produced by learners as they learn was given the name interlanguage (often abbreviated to IL) to reflect this process. The errors made by learners as they developed their knowledge of the new language were characteristic of inter-language and seen as part of this active process. This way of thinking about errors opposed the previously dominant behaviourist thinking about second language acquisition. For behaviourists, learner errors were due to poor learning or bad habits, because learning was solely a process of repetition, imitation and memorization. Whereas, for these interlanguage theorists, errors were evidence of an active internal learning process: âLearners work their way through a number of developmental stages, from very primitive and deviant versions of the L2, to progressively more elaborate and target-like versionsâ (Mitchell and Myles, 2001, p. 18, emphasis in original). Thus the word interlanguage was coined to show that learners are actively journeying from one language to another in the learning process, and that their errors are part of this journey.
Comprehensible input
Making comparisons between first and second language acquisition led some researchers to hypothesize that additional language learners needed conditions for learning similar to those of first language acquisition. Young children gradually acquire the full range of grammatical structures through immersion in the language but without being formally taught. It was therefore argued that additional language learners also need to be exposed to language that is meaningful to them in order to learn a language. This meaningful language was called comprehensible input. The theories of Stephen Krashen on the role of input were both the most extreme and the most well known (Krashen, 1985). For Krashen, input which is at the right level of difficulty for the learner is all that is needed for interlanguage development and change. He points to the successes achieved by immersion education for children (for example, Canadian English-speaking children are taught school subjects in French) as instances of learning through comprehensible input. In this type of education there is little formal second language teaching.
Output
There have been many critiques of Krashenâs language learning hypotheses (e.g. McLaughlin, 1987; Skehan, 2001; and see both readings for this unit). Some of the criticism has centred on the evidence from SLA research into immersion education. Although earlier studies of additional language learning through immersion showed that childrenâs additional language level was much higher than those who learned the language in formal language classrooms, these earlier studies focused on learnersâ abilities to comprehend the target language. Later studies of Canadian immersion programmes showed a gap between the childrenâs understanding of the target language and their production of the language (Swain and Lapkin, 1982). This led Merrill Swain to argue that the act of speaking â output â was also a necessary condition for interlanguage development (Swain, 1995). She hypothesized that having to interact helped learners to make the input more comprehensible, and also compelled learners to pay attention to the role of grammar in the input and to try out hypotheses gained in the learning process.
Cognitive processes: information processing and restructuring
Output is also considered essential to language learning by those researchers who have focused on trying to understand the mental processes involved in language learning (see e.g. Johnson, 1996; Skehan, 2001). Cognitive psychologists have attempted to explore what happens when we listen to or read something in the additional language, and when we speak or write with it. This is generally referred to as information processing. They stress that both comprehension and production are complex multi-level tasks. Speaking, for example, involves:
- Deciding on the content of the verbal message. This draws on the speakerâs knowledge of the topic and of the interactional conventions in which she is participating.
- Formulating the content. This draws on her current knowledge of vocabulary and grammar.
- Preparing how to articulate the content. This draws on her knowledge of the sound system.
- Actually producing the sounds, plus appropriate stress and intonation.
Throughout these phases the speaker also continually monitors her selection for appropriateness and accuracy. Over time, and given the opportunities, the learner is able to reduce the amount of processing time she needs to speak through the creation of automatic links between the different processes.
Changes in a learnerâs grammatical or lexical knowledge are called restructuring because new knowledge can cause the learner to revise her previously learned knowledge. Such restructuring can have an effect on information processing. Although new linguistic knowledge may result in greater accuracy or complexity of speech, the new knowledge can also seem to slow the learner down if this new item causes change within the existing knowledge store.
Acquisition versus learning
Krashen argued that learning an additional language was parallel to learning a first language in that it was a subconscious process and that formal teaching of structure did not help. Here is an example of this first language acquisition process that focuses on the acquisition of English language negatives. Young children start off by adding the word No to the beginning of utterances (âNo comb hairâ) and then gradually change to embedding no within the clause, before the verb (âDaddy no comb hairâ). The correct use of negative modal verbs (e.g. canât) come later. Children are not usually aware of making these changes, or of how they came about, and resist correction by adults. Thus Krashen talks of acquisition processes rather than conscious learning. This view challenges the usefulness of the teaching of grammatical structures in language classrooms.
However, direct comparisons between older children and adults learning an additional language and young children learning a first language do not take into account the huge differences in knowledge of the world and variety of learning experiences between these groups of learners, let alone the fact that young children are developing their cognitive processes alongside their linguistic abilities. Furthermore, it has since been argued that learning an additional language involves the conscious process of noticing. This is a process whereby the learner consciously notices a difference or âgapâ between her additional language use (or interlanguage) and that of the text she is reading or the person she is talking to. This provides an opportunity for a change in her knowledge of the language (restructuring). Noticing is said to be an action that comes from the learner when, and if, she is ready. This learning may be implicit, in that the learner may not stop in the middle of an interaction and reflect on this new knowledge. If questioned, however, the learner can explain her change in language production (see Ellis (2001) for more information on noticing).
It is argued that knowledge of the form of an additional language, usually gained through the language classroom, can support this process of noticing, and therefore, contrary to Krashenâs theories, formal learning can play a significant role in the process.
Socialization
More recently, some researchers have highlighted the social nature of additional language learning, and suggest that we should talk about socialization rather than acquisition, because learners need to learn about the social and cultural conventions of language use as well as the structures and vocabulary. This emphasis has come from studies of minority language workersâ language use in naturally occurring situations in the target culture (Bremer et al., 1996). These studies show that minority language workers develop the majority language through having to use it to pursue their daily lives, for example, in the workplace, or while looking for work. (We discuss such research in more detail in Chapters 2 and 5.)
An example of such learning is given by Celia Roberts in the box below. It is the closing moments of an interview between a job counsellor (T) and Marcello (M), an Italian worker in Germany who is seeking employment:
- 1 M : Wir muss vergessen <laughs>
We have to forget - 2 T : Ja + gut + dann hatten wir die saache fĂźr heut
OK good so weâre through for today - 3 Und wenn sie also in zukunft noch fragen haben kommen sie bei mir vorbei ja
And if you have any questions in future youâll look in OK - 4 M : Ja
Yes - 5 T : <Rufen sie an> ok <leans, back, speaks quietly, looks at door, stands up>
Give me a call OK - 6 M : So und jetzt muss ich gehen
So and now I must go - 7 T : <Ja>
- 8 M : < > <both laugh>
- 9 T : Wiedersehen
Bye - 10 M : Wiedersehen danke
Bye thank you
Transcript conventions
+ short pause
< > additional comments on way of speaking, etc.
[ ] overlap
(xxx) inaudible or omitted word
At one level, this could be construed as a simple case of pragmatic failure. Marcello fails to understand the pre-closing signals of T including âjaâ, âgutâ and âdann hatten wir die sache fĂźr hurtâ and advice for the future. It is only with the nonverbal cues that Marcello realizes that they are in the middle of leave taking. His interpretive difficulty is not surprising since as Scarcella (1982) has argued, conventional features such as greetings are acquired before pre-closing. But this sequence is also an unusually explicit moment of language socialization when at line 6 Marcello topicalizes the act of departure. This is more than just a matter of picking up on some pre-closing signals, and it is worth mentioning here that the crucial nonverbal signals which are part of the interactive environment are rarely considered in linguistic pragmatics.
(C. Roberts, in Candlin and Mercer, 2001: 111)
At line 6 Marcello actually makes explicit his new understanding of the meaning of both linguistic and nonverbal signals that the interview is over. Roberts calls this process an apprenticeship model, where the learner learns by doing, by participating in social interactions. However, she also points out that this model does not take into account social factors that affect learning, such as class and identity. These are discussed in Chapter 2.
Although these concepts have been introduced here in roughly the order they were developed, they do not represent a neat set of âstepsâ in the process of understanding how we learn another language. Specific research projects tend to draw on only one or two dimensions of this field of work, depending on their interests and the types of research methodologies they are using.
In the following section we move on to look at three small pieces of research that have all focused on learning in the language classroom, but have asked different questions and so have used different methods of investigation. These studies illustrate the critique of SLA classroom research offered by Breen in Reading 2, and so are also intended to help you evaluate his arguments. I suggest you read this section before reading Breen. The first piece of research approximates to what Breen calls âthe classroom as experimental laboratoryâ and the second study is an example of his category of âthe classroom as discourseâ. The third piece of research attempts to bring together the methods of these two different approaches in order to try to investigate both cognitive processes and the social nature of learning. This attempt to chart the relationships between the mental processing of new language and the social relations of a class is essential to understanding language learning in classrooms, according to Breen.
Research into language learning in the classroom: some examples
As may be seen in the above section on acquisition versus learning, the role of the language classroom in additional language learning has been both argued for and argued against. Since ESOL teachers have little choice but to support additional language learning from within the classroom, we will now look at three studies that attempt to investigate the thorny question of the relationship between teaching and learning. The third study is described in detail in the first reading of this chapter, so it is summarized only briefly in this section.
Tracing uptake
The first piece of research we shall look at was carried out in a foreign language classroom. It is included here because it addresses the important question of the relationship between classroom practice and student learning. The research was undertaken in one class by Assia Slimani in Algeria (Slimani, 2001). Her aim was to record what learners claimed to have learned from a particular lesson, which she called uptake, and then to analyse how these claims related to classroom interaction. Thus the research relies on the rec...