2.1 In the safe houses of principles and policy

The main rationale for advocating maximal use of the TL (the ‘monolingual principle’) is that for most students, the classroom is the only opportunity they have for exposure to the language. Hawkins (Reference Hawkins1987: 97–98) compares foreign language teaching to ‘gardening in the gale’ – the teacher plants seeds but these are constantly blown away between lessons – and so it is essential to maximize the learners’ exposure in the limited class time available. This argument is given additional strength by Krashen's (Reference Krashen1985) influential argument that languages are learnt most effectively when learners are exposed to lots of ‘comprehensible input’ in contexts of real communication. As Turnbull (Reference Turnbull2001: 532) puts it, in most foreign language contexts ‘the teacher is most often the sole linguistic model for the students and is therefore their main source of TL input’. A related reason is that by using the TL in the classroom not only as a ‘target’ to be learnt but also for other purposes, students are more likely to perceive it as a useful medium for communication and develop more positive motivation to learn it. As pointed out in Littlewood (Reference Littlewood1981: 45), ‘many learners are likely to remain unconvinced by our attempts to make them accept the foreign language as an effective means of satisfying their communicative needs, if we abandon it ourselves as soon as such needs arise in the immediate classroom situation’. A further reason commonly mentioned in the context of some methods (e.g. the direct method and audio-lingual method) has been that students need to ‘think in the TL’ and avoid interference from it, and that this can only occur if the TL and the L1 are kept as separate as possible. More recently, however, the notion that the two languages can (or should) be compartmentalized in this way has been questioned (e.g. by Cook Reference Cook2001; Butzkamm Reference Butzkamm2003).

For these and other reasons, since the grammar-translation approach was first challenged in the late 19th century, the monolingual principle has permeated every language teaching method that has found widespread official support. This emerges clearly from even a cursory look at surveys of recent developments such as those in Littlewood (Reference Littlewood and Spolsky1999), Larsen-Freeman (Reference Larsen-Freeman2000) and Richards & Rodgers (Reference Richards and Rodgers2001). Apart from a few ‘alternative’ methods such as Community Language Learning and Suggestopedia, there has been a general avoidance in methodological discussions (though not necessarily in classroom practice, of course) of techniques which use the students’ L1 and an emphasis on the importance of using the TL as the normal means of classroom communication.

At the level of national policy, the monolingual principle has been embodied in the guidelines of many countries. In Hong Kong, for example, the current English Language Curriculum Guide (Primary 1 – 6) (Curriculum Development Council 2004: 109) urges teachers to create ‘a language-rich environment [which] incorporates, for example, the use of English in all English lessons and beyond: teachers should teach English through English and encourage learners to interact with one another in English’. In modern foreign language teaching in the UK, the National Curriculum documents of the 1990s advocated total exclusion of the L1 but recent revisions have shown a ‘gradual shift in policy . . . to a measured inclusion of the mother tongue’ (Meiring & Norman Reference Meiring and Norman2002: 28) for planned purposes such as raising language awareness and helping learning through comparisons between the TL and L1. The Ministry of Education of Mainland China adopted a similar position in its experimental English syllabus for full-time high-school English classrooms published in 2000, stipulating that English teachers should use English ‘as much as possible’ in the English classroom, but recognizing that the L1 may be used for purposes ‘such as explaining or translating abstract English words and expressions, or special English structures’ (Ministry of Education of the PRC 2000). In South Korea, where there has traditionally been extensive use of the L1 in language teaching, the government has now adopted a policy of ‘Teaching English through English’ (TETE). The 1999 revision of the National Curriculum for English envisages a gradual increase of TETE to the point where English will be the only language used in elementary- and secondary-school English classes (see e.g. Jeon Reference Jeon2008; Kang Reference Kang2008; S.-Y. Kim Reference Kim2008).

In spite of these strong recommendations from the realms of principles and policy, we will see in the next section that many observational studies and surveys of classroom practice offer a different picture.

2.2 What proportion of TL and L1 do teachers use in their lessons?

Evidence from a variety of contexts shows that there is a wide discrepancy between official recommendations and the practice actually observed or reported in classrooms. For example, Turnbull (reported in Turnbull Reference Turnbull2001; Turnbull & Arnett Reference Turnbull and Arnett2002) analyzed the L1 and TL ‘functional units’ in the discourse of four teachers of French in Canadian secondary schools. He found that the use of L1 (English) ranged from 28% to 76%. S.-H. Kim & Elder (Reference Kim and Elder2005) examined the discourse of seven native-speaker secondary-school teachers of Japanese, Korean, German and French in New Zealand and found a range from 12% to 77% use of the students’ L1 (English). In South Korea, Liu et al. (Reference Liu, Ahn, Baek and Han2004) calculated the percentage of L1 (Korean) and TL words in the lessons of thirteen high-school teachers of English and found that their use of L1 ranged from only 10% (in a ‘model lesson’ intended to demonstrate TETE) to 90%. The overall average was 40% use of L1, but in view of the special nature of the data (the teachers supplied their own audio recordings), the researchers suggest that the teachers’ own estimate of 68% use of L1 may give a more reliable picture of their day-to-day practice. These and other studies support the conclusion reached by Macaro (Reference Macaro1997: 96) in summing up the results of several studies, including his own detailed investigations of TL and L1 use by secondary-school teachers in the UK, that exclusive or near-exclusive use of the TL ‘is rarely encountered in any learning context apart from [classrooms with mixed L1 learners]’.

The studies reported above are based on self-report or observation, neither of which gives an accurate picture of a teacher's actual, day-to-day behavior. To provide a third perspective, we asked 50 second-year tertiary students from Hong Kong (HK) and Mainland China (ML) to recall the percentage of classroom time their teachers had spent using the L1 (Cantonese or Putonghua) in junior-secondary-school English lessons. The results, presented in Table 1, are similar to those found in the studies described above.

Table 1

Percentage of lesson time using the L1 in HK and ML classrooms.

At one extreme, students recalled instruction which was almost exclusively TL-based; at the other, instruction which was heavily L1-based. As well as significant variation within each group, which may be due to factors such as those mentioned below (in section 2.3), there was significant variation between the HK and ML groups, suggesting variation between the cultures of teaching and learning. This emerges even more clearly when we consider the average overall percentage of L1 use that the students recalled: for the HK group, this was 20%; for the ML group, it was 64%. (One ML student wrote an explanatory note on her questionnaire: ‘In China we teach English through the medium of Chinese’.)

In her interviews with 59 modern foreign language teachers in the UK, Mitchell (Reference Mitchell1988: 28) noted some of the effects of this discrepancy between theory (or policy) and practice: ‘It appears that this is an issue about which methodologists have succeeded in inducing a sense of guilt in teachers, if nothing else; almost a third of this sample seemed almost to feel they were making an admission of unprofessional conduct, in “confessing” to low levels of FL use’. These teachers were in general agreement that the TL should be used ‘as much as possible’ (but not necessarily exclusively). This echoes Turnbull & Arnett's (Reference Turnbull and Arnett2002: 211) more general claim, mentioned above, about the ‘near consensus’ that maximum use of TL should be the goal. It also leads us to the question to be considered in the next section: When and why do teachers decide (or feel compelled) to use the L1?

2.3 When do teachers use the L1?

The 50 students from HK and ML mentioned above were also asked to recall the purposes for which their English teachers were most likely to use the L1 (Cantonese or Putonghua). From a list of nine purposes (drawn from a survey of the literature), they were asked to indicate on a four-point scale whether their teachers used the L1 ‘always’ (= 4 in the analysis), ‘sometimes’, ‘seldom’ or ‘never’ (= 1). There was also an option ‘I forget’ (which was chosen only four times in total). Table 2 lists the purposes in the order of mean for the whole group (N = 50). It also gives the separate means for the HK group and the ML group.

Table 2

Means of purposes for English teachers using the L1 in HK and ML classrooms.

The most common purposes, both overall and for each separate group, fall into the following three categories:

• • establishing constructive social relationships;

• • communicating complex meanings to ensure understanding and/or save time (‘explaining difficult grammar’ for the ML group, ‘giving the meaning of unknown words’ for the HK group);

• • maintaining control over the classroom environment.

The importance of these purposes is confirmed (in whole or in part) by other studies. For example, the teachers in Macaro's (Reference Macaro1997) study liked to use the L1 for relationship building, explaining difficult grammar, and disciplining. The teachers interviewed Mitchell (Reference Mitchell1988) also mentioned explaining (grammar and meanings) and disciplining as the two most common reasons for using L1; also important were providing background information and giving non-routine activity instructions. Like Mitchell, Liu et al. (Reference Liu, Ahn, Baek and Han2004) found that explaining difficult grammar and vocabulary was the most common reason for using L1. Qian (Reference Qian and Beaven2009) found that two primary-school teachers of English in Beijing switched to L1 for reasons which she classified as either ‘methodological’ (e.g. translating, clarifying, highlighting, efficiency) or ‘social’ (e.g. praising, encouraging, disapproving).

These three dimensions of pedagogical communication – establishing constructive relationships, ensuring understanding, and maintaining a disciplined environment – provide necessary scaffolding for classroom student learning and it may not be surprising if teachers have recourse to the L1 to establish them. On the other hand, since they are so essential, they also provide an ideal framework for TL communication and input, motivated by real communicative needs. A large percentage of teachers feel that they ought to use these opportunities for TL communication more. So again we may ask: What are the factors that explain why, so often, they do not do so? What are the constraints that often lead teachers to use the L1 even when they feel ‘guilty’ for doing so?

Many individual contextual factors are mentioned in the literature, e.g. pressure of exams and class size, but the factor that is mentioned most frequently is the students’ language proficiency. For some teachers interviewed by Mitchell (Reference Mitchell1988), for example, ‘the norm of FL use was too much for lower sets on “ability” grounds’ (p. 28), and similar views were expressed by teachers in Macaro's (Reference Macaro1997) study. The same constraint is mentioned in several studies of the introduction of TETE in South Korea (e.g. S.-Y. Kim Reference Kim2002, Reference Carless2008; Liu et al. Reference Liu, Ahn, Baek and Han2004; Jeon Reference Jeon2008; Kang Reference Kang2008). In many of these studies, some teachers have also mentioned lack of confidence in their own TL proficiency. However, native-speaker teachers too have often been found to use a high proportion of L1, e.g. in the studies by Polio & Duff (Reference Polio and Duff1994) and S.-H. Kim & Elder (Reference Kim and Elder2005).

Underlying these and other pragmatic factors which lead teachers to use the L1 is what Meiring & Norman (Reference Meiring and Norman2002: 32) describe as the ‘comforting effect’ and ‘psychological reassurance’ that learners receive from being able to relate learning to the familiar mother tongue. Meiring & Norman's comments about school-age learners are also echoed by the adult learners in Brooks-Lewis's (Reference Brooks-Lewis2009) study, who responded very positively to the inclusion of the L1 in their foreign language learning. Depriving students completely of this support by immersing them in a strange environment, where they feel disoriented and powerless, has been identified as one possible source of demotivation, especially for students with more limited proficiency (see e.g. Stables & Wikely Reference Stables and Wikely1999 in the UK; S.-Y. Kim Reference Kim2008 in Korea; also Crichton Reference Crichton2009 on how skilled teachers can overcome this).

3.1 Use of the L1 to achieve core goals

This domain focuses on specific techniques and tasks designed to help students internalize, comprehend and produce the TL.

3.2 Use of the L1 to achieve framework goals

We now enter the domain of ‘framework goals’ i.e. creating the affective and material conditions for learning. However, as indicated above, attending to these goals also provides a valuable context for input and meaningful communication.

4.1 The teacher's own determination and confidence

The term ‘day-to-day battle’ highlights a factor that emerges frequently in interviews with teachers: the importance of simple determination. There is constant temptation to abandon our ‘deliberate tactics’ and ‘judicious principles’ in favour of the easy way, which is of course always the L1. Soon we may open the L1 floodgates until ‘what one could call a near free for all’ (Turnbull Reference Turnbull2001: 535) results. To steer a course between this situation and the dogmatic banishment of the L1 is one of the major challenges in language teaching.

Pachler, Evans & Lawes (Reference Pachler, Evans and Lawes2007: 31) stress, in their advice to teachers in the UK, that establishing the TL as the main language ‘requires, particularly in the early stages of FL learning, complete commitment and consistency on the part of the teacher’ based on a ‘long-term view of language learning’. A teacher in her first year of teaching in Hong Kong described how she accepted the challenge. She persisted gently but firmly with her average proficiency students, answering them in English when they spoke to her in Cantonese, using conscious strategies to make herself clear, until English became the natural medium (Lee Yeelam, personal communication).

There are several reports that the increased experience which comes from a long-term view also leads teachers themselves to become more confident in using the TL. For example, an important finding in S.-Y. Kim's (Reference Kim2008) survey of Korean teachers of English, conducted after five years of TETE, is that ‘the more English they use, the less anxiety they feel’ (p. 67). Similarly, in Hong Kong, Lai (Reference Lai1996) found that a majority of pre-service students increased in confidence in using the TL as they gained in experience.

4.2 Communication strategies

Communicating successfully in the TL depends not only on the nature of the task and the message that has to be communicated, but also on the teacher's effective use of communication strategies. Mitchell (Reference Mitchell1988: 148) found in her classroom observations that most teachers used the L1 sometimes but ‘those [teachers] who were most successful in maintaining the norm of FL use for classroom communication also used a wide repertoire of other strategies for conveying meaning’. These included repetition, substituting an item with similar meaning, explaining in simpler terms, contrasting with items from a similar lexical set, exemplification and giving clues. Teachers may also gain valuable insights from studies of the modified speech which adults use when speaking to children (see e.g. Macaro Reference Macaro1997: 92–95 for examples) and research into the compensatory strategies which learners themselves often use to bridge gaps in linguistic knowledge (paraphrase, gestures, pictures and so on; see e.g. Tarone & Yule Reference Tarone and Yule1989: 109–113 for examples). Especially in early stages of experience, TL communication strategies are important enough to deserve attention as an aspect of initial lesson planning in order to meet anticipated difficulties.

4.3 ‘Starting simple’

Many earlier methods such as the Direct Method or the Structural–Situational Approach adopted techniques which made it comparatively easy to avoid the L1 right from the start. But with many techniques today (such as information-exchange tasks or role-plays), even if the task itself is simple, the instructions or explanations are sometimes so complex that using only the TL can lead to problems.

In contexts where the L1 is the normal medium of instruction in foreign language teaching but teachers aim to increase TL use, an important strategy is to use it first for tasks which are already very familiar to students (so that they have the relevant schemata to help comprehension) or which require only simple TL. In Korea, for example, S.-Y. Kim (Reference Kim2008: 68) advises teachers seeking to implement the TETE policy to explore ‘which types of tasks call for easy English’ in order to ‘gradually introduce teachers to the use of English in their classes’. Also in Korea, Lee (Reference Lee2007) asked teachers and students which tasks they found most suitable and accessible with TETE. The results suggest that teachers should start with tasks in which students ‘are already familiar with the content through previous exposure’ (p. 349), e.g. reviewing the previous class, summing up the class and checking comprehension. On the same principle, the TL can be used for classroom management first in routine situations which occur regularly and where the meanings are predictable. Some textbooks and handbooks include specific phrases which can be taught for such purposes (e.g. for English: Atkinson Reference Atkinson1993 and Slattery & Willis Reference Slattery and Willis2001; for modern foreign languages in the UK: Halliwell & Jones Reference Halliwell and Jones1991).