Some "inconvenient truths" of pronunciation teaching

2015 is coming to an end and once again, I wanted to thank you for following, reading and/or criticising this blog. 

This post is a very brief reflection on some of the many lessons I have learned this year, out of reading good books, interacting with my experienced colleagues, but also, and more importantly, after considering the feedback and success or failure of my best teachers, my students. I don't want to elaborate on these ideas at length, since there is a lot to say about each, but at least I wanted to list them here today. Who knows, I might write a few posts on these thoughts in 2016.

These are some of the "inconvenient truths" that I believe every EFL teacher needs to embrace when it comes to pronunciation teaching:

• Pronunciation is a "messy" matter. Things may definitely go awry, discipline-wise, when we do pronunciation work. Pronunciation tasks may turn out to be "ridiculous" for students, and they will most certainly be "noisy".  The development of these activities requires a lot of control and attention on the part of the teacher, a lot of moving around the classroom to monitor what students are doing; keeping your eyes open for possible peer bullying or ridiculing, and yes, it may be tiring. We may need to decide very carefully on our timing (the last few minutes before the bell goes?) if we have a difficult group, we may need to change seating arrangements, or establish a "safe" routine. We, of course, also need to make sure our students understand why they need to work on their sounds and intonation in the first place. All in all, pronunciation "chaos" is an issue we have to anticipate, but it should not scare us away from integrating pronunciation to our lessons regularly.
• Non-native teachers of English can be good models and instructors. This is something I heard Robin Walker state many times. I think we should never underestimate the process that each of us, non-native speakers of English, has actually been through to build our accents. We can use all this knowledge of our L1 and of what it takes to learn a foreign accent to our and our students' advantage. At times I get a bit cross at teachers systematising sounds only by introducing videos or animations of native speakers producing the sounds. I think that our presence and constant input in the classroom can be a great asset to our students as well, and we can demonstrate the production of the sound ourselves, "live", in many different ways (see my "tips and tricks" post), apart from the introduction of native speaker voices (or other voices, if you are using other L2 speakers as models, which you can read about in the ELFPron blog). There is also a corollary: perception and production may go hand in hand, though we need to ensure exposure to a number of different accents and voices, while perhaps selecting one variety for perception (already a tough issue!).
• Student success depends a lot on psycho-social aspects. As I listed on my post on pronunciation goals, and on my review of the book "Pronunciation Myths", motivation, feelings about the foreign language, rapport with the teacher and fellow students,  they all have a major bearing on student performance. This is something that needs to be acknowledged from the very start, and the more we get to know about students' expectations and views on this, the better. Students need to appropriate the process and the new articulatory moves to make sure there is progress. This is a tough and very complex issue, and I personally need to work it out still.
• Pronunciation is a matter of the mind...and the body (and the heart, given the previous bullet point...). This is quite obvious, you might think. But indeed, we need to remember the implications of working with people's bodies. We are, metaphorically speaking, the "fitness instructors" of our students' speech muscles and organs, perhaps their "physiotherapists", in a way. We teach our students how to become aware of what their organs do, and how to make changes to their speech with different exercises, different "dance" or "fitness" routines to develop their proprioception abilities (You can read a lot about this on Adrian Underhill's blog, by the way!). Plus, giving students feedback on their performance is not the same as making corrections on a written composition, and this requires great care and kindness; it's a skill I personally have been trying to learn for years, and I have not always succeeded in.
• Proper pronunciation teaching resembles the work of a tailor or a craftsman. I have recently realised there is no "one-size-fits-all" approach to pronunciation. Part of the systematisation process can be carried out with the group as a whole, but the ultimate fine-tuning processes need to be developed on an individual basis, and this requires getting to know our learners individually, testing different approaches and strategies, giving lots of personalised feedback, and this, means investing quite some of our own time, yes. And it is also a craft for us to learn; I have been doing this for years and still have a lot of questions, I keep discovering a lot of new ways to approach things, and I pretty much suspect that as times change, practices will have to change. So all in all, pronunciation teaching and learning make up a very personal process of discovery, of constant reflection, and as such, it only works if we help our students become aware of the inner workings of their speaking body, help them create different mental images of what the foreign language and its accents sound like, and invite them to ask themselves the right questions to be able to keep up with the times.

And some of my beliefs, omnipresent in this blog:

• Pronunciation can be seamlessly integrated to other areas, content and skills in the ELT classroom.
• Pronunciation work can be fun.
• We need to teach intonation, and do so communicatively, and in context.
• Teaching pronunciation requires, as with other content areas of language, some serious research and study on the part of the teacher.

Finally, two new discoveries I've been trying to develop into coherent thoughts:

• Success in a foreign language pronunciation is partly dependent on our ability to be mindful.  (A podcast-y reflection on this coming up very soon)
• Learning a foreign accent is in a way a process of appropiation. The big challenge is finding and loving our own L2 voice. (Some Harry Potter-inspired reflections on this here)

In summary, I think pronunciation teaching is by far, a greater learning experience for the teacher than it might be for the learner in the long run. I honestly thrive in all the lessons I keep learning, and I hope I can still continue asking myself questions to improve on my practices. And to keep blogging, of course (BTW, it does feel like this...)

Image from: https://missglayiii.files.wordpress.com/2011/07/hp4_17.jpg

I want to wish you all a lovely start of 2016. 'kiːp 'kɑːm ən prə'naʊns 'ɒn

Season's Greetings

Review #3, Part 2: "Pronunciation Myths", by Linda Grant.

On my previous post I briefly reviewed the Prologue and the first chapter of the Kindle book version of Pronunciation Myths: Applying Second Language Research to Classroom Teaching. (BTW, I am trying to come to grips with the way of citing Kindle books....the lack of pagination can be maddening! So I hope you will excuse the lack of page numbers for the quotes. I will, in the future, try to adopt the approach in the APA guidelines (that is...counting paragraphs and all that jazz!): http://blog.apastyle.org/apastyle/2009/09/how-do-i-cite-a-kindle.html)

This time, I will be discussing the research reviewed in the book to debunk Myths 2, 3 and 4. These chapters were written by Beth Zielinski & Lynda Yates, John Field, and Judy Gilbert, respectively. 

(Warning: It is quite a long post!)

***

Myth 2: Pronunciation Instruction is not appropriate for beginning-level learners - Beth Zielinski and Lynda Yates.

I was particularly interested in this chapter, since I remember with great shame my very first class in Italian, and my first lesson in French. I was already a graduate teacher of English and I was in awe at both my new teachers, who spent the first half hour or more discussing the sounds of the languages I was about to embark on learning. The same thing happened to my colleagues learning Portuguese. Why is it that most teachers overlook this very important first introduction to the pronunciation of English language during the first lesson? Isn't this the perfect moment to plant the seed of curiosity and to start drawing students' awareness of the noticeable differences between their L1 (Spanish, in my case) and English?

The very first point the authors make is that pronunciation does indeed affect both intelligibility and confidence to speak, as previous research has found, at least in a context where English is spoken outside the classroom. It is claimed that beginnning learners do care about pronunciation, and that the later students are introduced to it, the more limited their improvement will be. Zielinski and Yates favour both the introduction to metaphonological knowledge and awareness, and incidental teaching and correction. They underline the fact that instruction also enhances on other, less "assessable" skills, such as confidence and understanding, and this was found in the action research projects in Springwall (2002).

The authors of this chapter propose following a systematic approach to instruction following four stages: a) listening and awareness, b) control, c) practice, d) extension.

I want to highlight the essential remark made in this chapter, which is that pronunciation is a spoken skill, and however obvious this may sound, we at times forget about this when we carry out reading aloud tasks and select texts to that end. Lexico-grammatical and discourse-semantic choices differ greatly in the spoken-written mode continuum, as those of us who do Systemic Functional Linguistic work very well know.

A few other reminders are made on this chapter, regarding the use of different "modalities" (i.e.: learning styles); provide targeted feedback, immediately or in a delayed form, depending on the task at hand; have one pronunciation goal in every lesson, in combination with other areas and skills; introduce a variety of activities. The authors suggest that readers consult the AMEP Research Centre in Australia website for ideas.

***

Myth 3: Pronunciation Teaching has to establish in the minds of language learners a set of distinct consonant and vowel sounds - John Field

John Field addresses a very interesting topic, and begins with a number of thought-provoking questions: "a) What is it that we expect learners to store in their minds during ear training? (...) b) How confident are we that storing a phoneme in the mind in some way enables the learner to recognise it when it is heard again in connected speech?". Apart from recognition, Field also questions how much of this "storing" of sounds actually contributes to production as well.

The research review reports on some truths we sometimes forget about: phonemes do not have a "standard" form because of the multiplicity of allophones they may be realised with, and because of all the phenomena involved in co-articulation. This is why training via minimal pairs does not always work as a form of enabling the later recall of phonemes. A clear implication of this is the need to expose learners to the sound in different contexts and in different word positions.

The rest of the chapter unfolds under the scrutiny of two theories that present assumptions in the ESL world:

"A- Although phonemes vary a great deal, any example of a given phoneme has characteristics that distinguish it from all others."

Field notes a first very important truth "the raw speech reaching the listener's ear is not a string of phonemes but a series of acoustic cues that the listener has to match to phonemes". The author reports studies carried out by Liberman in Yale in 1957 that attempted to somehow find "quintessential" (my own interpretation of the study) characterising features of English phonemes, but did not find features which were unique or exclusive to each. Plus, they found that speed had an effect on the perception of phonemes. The perception of consonants was found to be categorical, whereas perceiving vowels requires a distinction between three formants. There are no "constant" values to identify vowels, as we know. So this first theory in a way appears to be quite weak, since as Field concludes, "there is no simple one-to-one match between a group of acoustic cues and a phoneme".

"B - We store a set of 'ideal' phonemes in the mind, against which to match what we hear."

This assumption is related to the belief that our mind cancels out unusual features in the perception of sounds so as to match them to the closest "ideal" phoneme around. This view is related to  the notion of "prototype" (Rosch, 1975) or "template" against which all other manifestations of a word or phoneme are contrasted, and this makes for an economical view of the work the mind carries out in perception. However, Field argues that research by neuroscientists has proven that the mind is capable of more complex forms of storing, and that individual variation of phonemes due to coarticulation also acts counter to the "template" rule. 

(BTW, these two very interesting questions are currently being addressed by Richard Cauldwell in his "Listening Cherries" blog, and in his book "Phonology for Listening (2013)".)

Therefore, three alternative theories are put forward, which claim that a) the phoneme is not the unit of representation we use when we listen; b) the phoneme is just one of the clues used, perhaps not the most important; c) our minds store many variants of the phoneme heard in different contexts and voices.

Regarding a), Field reports studies that have found that listeners need to access the word, the "demi-syllable" (Dupoux, 1993) or the syllable in order to identify the number of phonemes. In this respect, the author believes that teachers should focus on high-frequency syllables as units of perception.

As to theory b), Field notes that phenomena such as ambisyllabicity shows the complexity existing in the recognition of syllable boundaries, and extends the list of cues employed by listeners to include phonemes, words and lexical chunks, and these are supplemented by contextual and co-textual elements. Research by Marlsen-Wilson 1975 has revealed that apparently we "process speech as we hear it at a delay of about a quarter of a second" and that we are actively making hypotheses as to what we believe we have heard, fine-tuning our perception unit by unit. Studies on L2 perception have found that learners "do not trust their perceptual skills in relation to smaller units of language", and they use a more "constant" unit, the word. However, as we know, words also vary, as Cauldwell (2013) states, by adopting different "shapes" in the stream of speech. So Field believes that perception should be trained at the level of the syntatic and lexical chunk as well.

The remarks made by Field regarding the last theory are quite fascinating. It seems that we store in our minds a large number of versions and examples of phonemes and word shapes, and we activate our memory of these items. The teaching implications that Field mentions include the need to expose learners to different realisations and accents, and reviews studies by Pisoni, Lively and Logan (1994) in which this kind of variable ear training was found to be effective. 

Field suggests that teachers should work at the level of the syllable, by working on common syllable patterns and recognition of syllable boundaries, as well as for stress. Later, instruction should be driven into larger units, by proposing "word activation" tasks to predict lexical items to be encountered, as well as "gap fill" activities with typical lexical chunks and their reductions and variation (Field introduces an interesting appendix on this). Focus on function words also acts as a good training exercise to identify changes in speech. The author also proposes exposing learners to many accents, voices and contexts following a gradual progression, in order to create "memories" of different sound and word shapes.

I found this chapter particularly fascinating and thought-provoking. I think it also bears great implications for production, since at times our students attempt to produce these "ideal" phonemes in contexts that in fact require a number of variations due to co-articulation. Even though we may teach allophonic and phonemic rules for different processes, students at times only produce strings of "ideal" phonemes, as if they were putting together a puzzle of individual sound shapes instead of making all the changes and adaptations that the surrounding environment of neighbouring sound requires. A lot to continue pondering on...

***

Myth 4: Intonation is hard to teach - Judy Gilbert

Those of you who know me will have guessed that this is one of the chapters that drove me into buying the book. There is such a huge gap in intonation teaching (for a huge number of reasons, some of those listed in the chapter), and there is so much that could be done in this respect....Anyway.

Judy Gilbert begins with the retelling of a few anecdotes regarding misunderstandings and intonation. She believes that one of the problems with intonation teaching is that it is filled with "too many abstract concepts.....that tend to blur the specific teaching point". Studies by Kang and Pickering (2011) on comprehensibility highlight the role of intonation for listeners to "confirm if an item is new or one that they are already aware of, to track important information, and to predict when one topic is ending and another is beginning" (2011:6). A stronger focus on prosody in the language lesson is also recommended by Derwing and Rossiter (2003) in order to make students' English more understandable. 

Gilbert reminds her readers that the "basic signals of rhythm and melody specific to one's first language are generally learned by the time a child is one year old". This affects the way we may produce our L2 intonation, and the way we perceive other people's intonation, which could lead to cultural misunderstandings, as well as communication problems, especially when it comes to inaccurate stress choices.

The author discusses some of the reasons behind teachers' uneasiness when it comes to intonation teaching, such as the lack of "friendliness" of intonational descriptions. The first "discouraging approach" is the set of technical rules based on grammar; the second relates to "technical rules based on pitch levels" drawn by phoneticians who cannot establish an agreement when it comes to "meanings" . Gilbert also nots that in these models, the different manifestations of pitch levels which are at times presented as absolutes, instead of as gradient. The third approach seen as "uninviting" is what Gilbert calls "subjective rules based on intuitions about attitude", which is "anxiety-inducing" (Roach 1991:165) Gilbert makes a point of the fact that attitudes are "culturally or contextually dependent", and that most attitudinal descriptions of intonation have applied conflicting values to different pitch levels and emotions. It is also establised that attitude is never expressed by intonation alone.

Gilbert moves on to present different proposals for the teaching of intonation. The first suggestion is to draw students' awareness as to how their L1 and English differ in terms of information delivery, since English uses specific prosodic signals (change in pich, lengthening and "extra clarity" of the vowel in question), and other languages may perhaps employ other non-prosodic features, such as word order. Another suggestion includes the teaching of intonation as a priority, done contextually in dialogues. The next idea is related to the teaching of "listener-friendly" intonation, intonation that helps the listener "follow" what the speaker is saying in the belief that "time spent helping students concentrate on the major rhythmic and melodic signals of English is more importnat than any other efforts to improve intelligibility". However, Gilbert warns teachers that learners may not believe the teacher when it comes to the introduction of intonation, and that they may feel silly trying intonation tasks, which in a way implies that the teaching of prosody is also an act of "persuasion" (and I agree entirely!). Creating activities that may focus on miscommunication or ambiguity, and which require immpediate communicative feedback, may help to debunk this view. Gilbert reminds her readers that one of the clearer benefits of intonation teaching include the fact that suprasegmentals fit nicely into communicative settings. 

Next, Gilbert makes a point of the features which she considers to be "main", and "most teachable". These include forms of highlighting new information and the separation of the message into thought groups. Gilbert puts forward her model of the "Pyramid of the musical signalling system", which goes from a peak vowel to a stressed syllable, followed by a focus word embedded in a thought group. The attention given to the production the peak vowel in the stressed syllable of the most important word in the thought group is an essential component of communication, and thus makes the delivery of the mssage more "listener-friendly".

Prosody Pyramid - Gilbert (2008)

Gilbert also suggests a holistic presentation of intonation, using template sentences through "quality repetition" so as to "get a clear image of what the flow" should feel like and thus not sacrifice fluency. The author believes that students should get an "acoustic impression" of a short chunk as a whole, and then focus on their individual components through repetition. The presentation of the template should be done through listening first, many times, before actually producing it, respecting the rhtyhm and intonation of the thought group (or larger unit) selected. Students can then be invited to analyse the inner workings of the tought group which has now been internalised.

I particularly liked Gilbert's defence of repetition as a strategy to "produce a long-term memory resource that they (students) can access when they need to remember how it went", since "cells that fire together wire together" (Shatz, 1992:65). The value of group practice to "overcome individual pyschological inhibitions" is also highlighted. 

The inclusion of other styles, such as kynesthetic reinforcement, or the use of gadgets (such as a kazoo or a wide rubber band) to introduce pitch patterns, stress or lengthening is an interesting addition to the teacher's kit that Gilbert proposes.

This is a great chapter, full of ideas, but the most fantastic inclusion to my mind is this quote, drawn from an unnamed teacher trainee Gilbert once met: "teaching pronunciation without prosody is like teaching ballroom dancing, only the students must practice standing still, without a partner, and without music".

I believe that the points made by Gilbert are definitely true. However, I need to point out that when it comes to intonation teaching, the "native speaker gut" is very useful, especially to give feedback or make associations between intonation and "meanings" or "interpretation". But the truth is that us, non-native teachers of English, struggle when trying to understand how the intonation of English works, and many of the theories reviewed and slightly criticised in this chapter have actually proven useful to many of us to get an idea of what patterns to use. In fact, I think that the development of metaphonological awareness and constructs when it comes to understanding -and later teaching- the intonation of English really works, if done conscientiously, and within the study of pragmatics and discourse.

***

Thank you for bearing with me till the end of this long post. Part 3 of this review will discuss the remaining "myths" in this fabulous collection by Linda Grant. See you soon!