2.1 Learners can and should be taught how to listen not just tested on their listening

Research provides evidence that learners of MFL in England perceive listening as one of the most difficult skills in which to improve, particularly beyond the immediate beginner stage (Graham Reference Graham2006). Problems noted among learners, even those in their sixth year of language learning, include poor monitoring of understanding, poor application of background knowledge to help overcome problems of comprehension; problems in speech segmentation and recognising familiar vocabulary in the speech stream. In Graham et al. (Reference Graham, Santos and Vanderplank2011) we found that over a six-month period, during a time when they were exposed to much more challenging listening material than they had experienced beforehand, learners displayed stability in how they used strategies and that differences between proficient and less proficient listeners, regarding strategy use, persisted over time. In a more recent study with younger learners (aged 13–14), Macaro (Reference Macaro2014) found they relied almost totally on just one strategy, listening out for cognate words, when trying to understand what their teacher was saying to them in French classes.

Several authors (such as Field Reference Field2008; Goh Reference Goh and Harwood2010) have called for more teaching of listening as a skill in its own right, rather than something which teachers assume will develop of its own accord. Listening strategy instruction is one approach to the teaching of listening. There is some variability in outcomes from studies exploring such instruction (see Macaro et al. Reference Macaro, Graham, Vanderplank, Cohen and Macaro2007) but I would argue that more recent investigations provide firmer evidence of its benefits. These studies include those that investigate (a) approaches combining explicit strategy instruction and a metacognitive element (e.g. Harris Reference Harris2007, who studied 13–14 year olds in England; Graham & Macaro Reference Graham and Macaro2008, working with 16–17 year olds in the same context); and (b) more implicit approaches where the development of metacognitive awareness and/or metacognitive strategies is the main objective (e.g. Goh & Taib Reference Goh and Taib2006; Vandergrift & Tafaghodtari Reference Vandergrift and Tafaghodtari2010). Both explicit and implicit approaches are underpinned by the notion that practice and contact with spoken language alone are insufficient to improve how well learners listen. They also reflect a growing emphasis on the development of metacognition in relation to listening (see also below).

The extent to which the methods underlying the generally positive outcomes of such studies have reached classroom practice seems, from my perspective, to be limited, a view that is also expressed by other authors across a range of contexts who comment on teachers doing listening activities or tests with their learners rather than teaching them how to listen. This ‘doing’ or ‘testing’ approach dates back a long while in England; thus Chambers (Reference Chambers1996) outlines what he sees as the typical approach to listening in foreign language classrooms in that context: learners listen to a text and complete an accompanying exercise, which is then corrected. Listening becomes principally a test of comprehension, typically based on an exercise from a textbook, which, as an examination of such materials indicates, usually requires the understanding of specific details rather than the global meaning of the text (thus probably depriving learners of a sense of achievement if they rarely feel they understand more than a few snippets of class listening materials). This constitutes what Field (Reference Field2008) calls the ‘Comprehension Approach’ and which, he argues, does little to develop effective listening skills. Working as a teacher educator, a couple of years ago I asked my 26 student teachers about the practice they had observed during their ten-week practicum. Twenty-two (or 85%) of the students had rarely or never observed teachers showing learners how to improve their listening skills.

Seeking further evidence of what teachers believe and do with regard to the explicit teaching of listening, we carried out a study of language teachers in England involving questionnaires given to 115 teachers with a range of years of teaching experience (the findings from those questionnaires are reported in Graham et al. Reference Graham, Santos and Francis-Brophy2014). We then conducted observations followed by interviews with 13 of these teachers. In addition, we analysed a number of commonly used textbooks for the type of listening activities presented and what support they gave teachers for listening instruction (a full analysis of these data sources is given in Graham & Santos Reference Graham and Santos2015). While there was a strong level of agreement from the closed questionnaire responses of teachers with the idea that it is possible to teach learners how to ‘listen more effectively’, such agreement was then contradicted by other aspects of teachers’ replies. When we asked them through an open-ended item to outline the four most important things that they did when conducting listening activities, and to give reasons, there was very little sign of any actual teaching of listening or teaching of listening strategies. Most respondents outlined steps taken to make sure learners were clear about task completion procedures (e.g. where and how they should record their answers) rather than any practices adopted to focus on improving listening as a skill. Justification for practices adopted was mainly concerned with helping learners get the right answers to questions set, behaving in the expected manner, and meeting assessment requirements.

The subsequent observations and interviews largely confirmed the questionnaire responses: that teachers did listening activities with their classes but rarely taught listening. The purpose of such activities was rarely seen as helping learners to improve their listening; instead, other purposes such as preparing learners for assessments, for developing some other skill or body of knowledge (e.g. of verb endings; preparing learners for speaking tasks) were identified. Siegel (Reference Siegel2014) argues that teachers do not to teach listening because they lack the pedagogical knowledge to do so, a view which I tend to share and which seems to be borne out by a comment from one of the teachers we interviewed about the best way to teach listening:

The hesitancy from this teacher suggests a certain resignation (‘there is only so much you can do’) as well as uncertainty about what might be useful ways to teach listening.

Teachers’ lack of pedagogical knowledge may arise from lack of input on listening in their training; it may also be exacerbated by a lack of guidance in the materials they use. In addition, more than may be the case for other skill areas, teachers may turn to commercially produced materials for listening, because of the difficulties involved in creating their own. Our textbook analysis found little support for teachers, with scant focus on the teaching of listening strategies, and indeed the tasks set offered little scope for doing so. Activities mainly consisted of locating very specific, local information from generally quite short texts with little or no redundancy, and with very little or no potentially more challenging vocabulary. Any advice given for using materials with learners was mainly of a practical, procedural nature. Furthermore, teachers responding to our questionnaire confirmed that they relied heavily on textbooks as sources of listening material and activities, something to which the school inspectorate in England (Ofsted 2011) attributes learners’ weaknesses in listening.

2.2 The development of metacognitive strategies/metacognitive awareness through peer discussion and reflection is helpful

The previous section highlighted that studies have reported improvement in listening attainment through approaches that seek to develop learners’ metacognitive awareness and use of metacognitive strategies. For example, working with quite young learners, i.e. 11 years old, Goh & Taib (Reference Goh and Taib2006) found that combining traditional listening activities with post-listening reflections and discussions of metacognitive strategies for listening led to improved confidence among learners, especially for the less proficient ones. Similarly, Vandergrift & Tafaghodtari (Reference Vandergrift and Tafaghodtari2010) record improved listening scores for lower proficiency learners (although not for those of higher listening proficiency) as a result of a metacognitive instruction programme for listening, in which there was no explicit teaching of named strategies but ‘guided practice on the listening process as a whole’, developing ‘implicit knowledge about L2 listening through task performance’ (487‒488). The programme took the form of a ‘pedagogical cycle’ (Vandergrift & Tafaghodtari Reference Vandergrift and Tafaghodtari2010: 472) involving prediction of what might be heard, verification or checking of these predictions and peer discussion of what was heard, reflection on strategies employed and future goal-setting. Learners of all proficiency levels showed greater metacognitive awareness and sense of control over their listening after the intervention. The authors comment that the peer discussion element was particularly helpful, aiding the development of strategies for planning, evaluation and problem-solving.

Giving learners opportunities to reflect on, and talk about how they listen and how they arrive at understanding might seem to be a relatively uncomplicated thing to do in the classroom, and as the study by Goh & Taib (Reference Goh and Taib2006) shows, can be achieved alongside working on traditional listening activities and with relatively young learners. It does not need to even involve much teacher input as Cross (Reference Cross2010) found, in a study in which adult learners of English increased their metacognitive awareness merely by discussing listening activities in pairs, comparing solutions to problems with listening, and writing about their strategies in a diary format. Nevertheless, there seems to be few signs that this metacognitive development occurs frequently in classrooms, although more in some contexts than in others. Siegel (Reference Siegel2014), observing 30 listening lessons from ten teachers of university level learners of English in Japan, recorded 24 instances of teachers encouraging metacognitive strategies for listening, which he defines as ‘planning, monitoring, and evaluating listening as well as problem-solving’ (p. 25). This is a relatively high number of occurrences compared with what our research in classrooms in England reports, possibly because of the difference in age range of the learners under investigation and in their levels of language proficiency. Siegel (Reference Siegel2014) does also point out, however, that many of the 24 instances involved attention to task procedures, which arguably are not part of metacognitive development. It is also not clear that Siegel's teachers engaged learners in metacognitive reflection any more than seems to be the case for teachers in England. Such teachers, responding to the Graham et al. (Reference Graham, Santos and Francis-Brophy2014) questionnaire, referred to post-listening discussions with learners that centred almost totally on what answers they had given to comprehension questions. In the lessons observed in the same study, we saw minimal discussion of the listening process, of how listening ‘problems’ or challenges might be approached, either before, during or after learners heard the passage, with teachers tending to discuss assessment-related or procedural issues instead. There was little sign of developing a problem-solving approach to listening. In one of the post-observation interviews, a teacher indicated that she avoided asking learners about how they had tried to understand what they had heard, claiming that they disliked such questions: ‘If I ask them that question they lose the will to live’. Another explained that she had tried to engage learners in more metacognitive discussions about listening, but with little success, because learners were ill-equipped for such discussions: ‘Even though I ask them a lot “how do you feel about the listening?” they just say . . . their vocab to describe it is limited [to]: “It's hard, it's too fast”. They use the same words all the time.’ This might suggest that such learners need more opportunities to get used to this kind of reflection, perhaps through approaches other than whole-class discussions. For example, this might involve keeping a journal of their listening reflections and the strategies they have tried out, or paired discussions about listening. Our questionnaire indicated that tools like these are hardly used by teachers in England, perhaps because they are unaware of what benefits they might have, or because they are more focused on assessing whether learners have met the rather narrowly conceived expected levels of attainment in listening stipulated by national frameworks I outlined earlier.

2.3 Attention to the development of bottom-up and top-down strategies is important

In an exploration of what differentiates learners who gain high overall scores in listening exams from those gaining low scores, Tsui & Fullilove (Reference Tsui and Fullilove1998) found that the former were better able than the latter to verify predictions made on the basis of prior or background knowledge by attending to later in-text information (i.e. applying bottom-up and top-down strategies in appropriate combinations). Other research (e.g. Vandergrift Reference Vandergrift1998) emphasises the importance of such combinations. In my reading of the research literature, learners seem to have more problems with bottom-up aspects of listening (e.g. speech segmentation, recognising known words in connected speech), as noted in a number of studies across a range of contexts including UK learners of French and German, and Chinese and Arabic learners of English (e.g. Goh Reference Goh2000; Hasan Reference Hasan2000; Graham Reference Graham2006). In addition, it has been argued that listening strategy interventions that include a bottom-up element as well as attention to top-down and metacognitive development are likely to be more successful (Graham & Macaro Reference Graham and Macaro2008). Indeed, Vandergrift & Tafaghodtari (Reference Vandergrift and Tafaghodtari2010) surmise that their intervention might have had more impact on learners of higher listening proficiency levels had it included a bottom-up aspect.

While textbooks designed for learners of English often seem to include attention to bottom-up features through activities such as minimal pairs and dictation, our review of courses written for learners of French, German and Spanish in England indicates much less attention. Books aimed at beginners might include short passages for learners to listen to and repeat, but without any explicit focus on phonology, on grapheme/phoneme correspondences, intonation patterns, etc. In our questionnaire, while teachers seemed very aware that one of the main causes of difficulty in listening for their learners was identifying word/sentence boundaries, they reported using very few classroom practices that might address such difficulties, such as focusing on grapheme-phoneme correspondences or other activities suggested by Field (Reference Field2008). It may be that textbook authors are more concerned with providing teachers with listening activities that allow them to assess what level of proficiency their learners have reached, or whether they know the vocabulary that will be assessed later in the textbook unit, rather than activities that develop these bottom-up skills. In addition, in a context where, generally, high school learners receive only two hours of language instruction a week, teachers may feel they can ill-afford to devote time to such skill development rather than to covering the syllabus and getting to the end of the textbook.

3.1 Prediction and pre-listening

As discussed earlier, Vandergrift's pedagogical cycle (see, for example, Reference Vandergrift2004, Reference Vandergrift2007) has been shown in a number of studies to be potentially beneficial to learners. An important part of that cycle involves learners in making predictions, before listening, about what they might hear, and then verifying and discussing those predictions. What we might call ‘pre-listening’ also occurs in a body of research regarding the role of so-called Advance Organisers, whereby the teacher pre-activates learners’ linguistic and non-linguistic knowledge by providing, for example, mind-maps, vocabulary lists, asking learners to engage in brainstorming activities related to expected content of what is going to be heard. By and large, studies of the provision of such Advance Organisers suggest that they improve learners’ ability to understand what they hear (e.g. Herron et al. Reference Herron, Cole, York and Linden1998; Jafari & Hashim Reference Jafari and Hashim2012). However, it is not clear whether this improved understanding has any long-term benefits for listening development.

Nevertheless, prediction and pre-listening activities seem to have found their way into the languages classroom in England. In our 2014 questionnaire, around 48% of teachers said they often or always asked learners to predict vocabulary that might occur in the listening passage; even more (78%) reminded learners of vocabulary linked to the topic in hand. These figures for ‘pre-listening’ discussion were far higher than those for post-listening discussion, and there was little evidence of learners and teachers monitoring or checking any predictions made. Siegel (Reference Siegel2014) likewise reports that in the tertiary-level classes he observed the setting up of predictions was much more frequent than any checking of those predictions after listening. Teachers in England also seem to direct learners’ attention more to individual items of vocabulary than to broader themes that might rise in a listening passage, possibly because they see it as a surer way of helping learners obtain ‘correct answers’ to accompanying comprehension questions. In an earlier study, reported in a number of papers (e.g. Graham, Santos & Vanderplank Reference Graham, Santos and Vanderplank2010), older learners of French were also reportedly using prediction very frequently before listening. This often took the form of noting down nouns that might be heard in the up-coming passage and might relate to multiple-choice items accompanying the listening passage. There was some evidence that learners had been advised by their teachers to follow this approach (Graham et al. Reference Graham, Santos and Vanderplank2011). Unfortunately, in many instances, prediction proved to be unhelpful; for example, by leading learners to imagine hearing the predicted word even if it did not occur, and then drawing erroneous conclusions about the passage as a whole; or focusing so much on trying to hear the predicted items that the overall sense of the passage was ignored. Problems also arose because learners failed to verify whether their predictions were correct or not, possibly because they had never been taught how to.

3.2 Learners find L2 listening difficult and it causes them anxiety

Language-related anxiety has received less research attention in recent years than was the case in the 1970s–1990s; however, the findings of such studies may have resonated clearly with teachers, with Arnold (Reference Arnold2000) commenting that studies into language learning anxiety have confirmed what teachers know ‘intuitively’ (p. 777). Earlier studies (e.g. Horwitz, Horwitz & Cope Reference Horwitz, Horwitz and Cope1986) found that aural comprehension was a source of much anxiety, with students, for example, worrying that they needed to understand every word that they heard in order to be able to comprehend. More recently, Bekleyen (Reference Bekleyen2009) identifies listening as a source of anxiety for student teachers in their own language development.

Teachers might make at least two responses to these research findings, it seems to me. First, they might look for ways to give learners the tools to cope with the problems they experience in listening. In other words, teaching them how to improve their listening by means of some of the methods in Section 2. Alternatively, however, they might decide to try to remove the source of difficulty in the first place, by simplifying the material and tasks that learners are given. In the context of England, the second approach seems to be the one most frequently taken by teachers, materials writers and curriculum designers. It is arguably the least helpful reaction to the difficulties learners experience in listening, because of the potential impact on their motivation and language learning development if they only ever encounter carefully scripted spoken extracts using a narrow range of language. Thus, within the National Curriculum for learners in the 11–14 age group, changes were implemented in 2007 that reduced the expectations of learners regarding their ability to ‘cope with language spoken at near normal speed in everyday circumstances that has little or no interference or hesitancy’ (QCA 2007). Prior to 2007, learners of average attainment were expected to be able to do this, but post-2007 that requirement was moved into the level descriptor for learners working a long way beyond normal expectations. In other words, there is no longer the expectation that most learners would be able to cope with any language approaching authentic speech. Furthermore, from 2007 to 2015, in the GCSE (examination taken at age 16), this lowering of expectations was matched by a reduction in the marks available for the listening component of the examination – from 25% to 20%. Observing teaching in language classrooms and talking to teachers leads me to conclude that these changes in national expectations for listening have resulted in less attention being paid to listening in England, and also lower expectations of what learners would be capable of understanding. From 2018 the GCSE will set higher expectations for listening, and it will be interesting to see what impact this has on how teachers approach it in the classroom.

There was evidence of lower expectations in the interviews we conducted with teachers. Several spoke of the anxiety learners experienced around listening and how they tried to ease this for them, through steps such as making sure passages only contained known language. Whether teachers, in offering easier, less authentic tasks and trying to shield learners from difficulties, really are responding to research findings about listening being an anxiety-inducing activity, is, however, a moot point. They may be equally motivated by the knowledge that language learning in England is an unpopular option, and believe that offering less difficult work will make it more popular. As I have argued elsewhere, however (e.g. Macaro et al. Reference Macaro, Graham and Woore2015), giving learners the tools with which to face any listening challenges they experience seems a more positive way to improve their motivation for language study. Of course, listening tasks and materials need to be accessible to learners, but in my view this does not mean they have to be without challenge.