Skip to navigation – Site map

HomeNuméros55English Pronunciation and the Spe...

English Pronunciation and the Spelling-Sound Code: What Priorities for Teachers of EFL?

Susan Moore Mauroux
p. 175-191

Abstracts

The article aims first to identify significant learner challenges regarding the relationship between written-spoken English which can be usefully addressed by raising awareness of the specificities – and regularities – of the English spelling-sound code. The role of teaching exams is then discussed, analyzing how specialist competence is tested and how far this relates to learner issues, and a few strategies to bridge the theory-application gap are outlined. Priorities identified concern vowel reduction where the absence of any letter-sound correspondence is an issue, and the articulation and/or perception of sounds, especially certain vowels (/ɪ/, /i:/ and /aɪ/; /ɒ/, /əʊ/ and /ɔ:/), often associated with confusing spellings (<i> and <o>). For these, it is argued that awareness of how the English spelling-sound code works should enable learners to improve both production and comprehension. The detailed analysis of the phonology papers of the French teaching exam (agrégation externe d’anglais) based on papers and reports from 2000 to 2020 focuses on three elements: morphophonology where vowel quality is determined by certain endings; IPA transcription/spelling-sound which tests pronunciation and knowledge of the code; and phonetic variation involving connected speech. The study shows that questions regularly focus on aspects of the system which could be usefully transferred to learners.

Top of page

Full text

  • 1  Aslam Sipra explains that confusion is triggered because the relationship between English spelling (...)

1One of the paradoxes of mastering spoken English is that in the learning environment, even when the main objective is speaking, teachers rely to some extent on written sources: oral task-based activities usually involve written instructions as do tests, and the written mode is the one used for reference purposes like, for instance, memorizing the lexicon (Duchet, 2018: 5). As Schottman (2012: 77) reminds us: “For the foreign language learner, the question is […] how to pronounce [a word] knowing how it is spelled.” Yet this is far from easy and written English is itself often identified as being a source of error for spoken English because learners tend to use their own L1 code to decode L2 (Aslam Sipra, 2013: 1211; Bouvet, 2021; Fournier, 2010: ix; Howe & Moore Mauroux, 2021: 1; Moore Mauroux, 2010: 20). However, English has a very different system and learners need to understand its specificities to be able to interpret it.

2This approach implies a degree of theoretical knowledge, which begs two related but distinct questions that are central to this article: what do English teachers need to know as specialists of the field; and what do teachers need to know to teach L2 English? These questions lead us to a third one: how far do the theoretical principles acquired during higher education relate to learner priorities in the classroom?

  • 2  The Agrégation externe d’anglais, hereafter referred to as AEA, is a French recruitment exam for t (...)

3I will first identify the challenges that spoken English presents for L1 French learners, particularly regarding the spelling-to-sound system, and discuss how far an understanding of this code may help them with both perception and production. I will then analyse how spelling-sound issues are addressed in a teacher recruitment exam, the agrégation externe d’anglais (AEA)2, looking at the phonology subsection of the linguistics papers from 2000 to 2020, together with the jury reports, considering how far this theoretical knowledge relates to specialists and/or to learners.

The English Spelling-Sound Code: Identifying the Challenges

English Rhythm and the Spelling-Sound Code

4Many researchers consider that the rhythm of English is the key to spoken English (Fournier, 2010; Howe & Moore, 2021; Huart, 2010; Moore Mauroux, 2010; Schottman, 2012) and spelling-sound issues can only be addressed once this is understood. Indeed, rules for spelling-sound correspondences only apply to stressed syllables; unstressed syllables, as well as grammatical words, frequently undergo vowel reduction, thus erasing any spelling-sound connection. This is the first challenge for L2 learners to appreciate: what they hear will not always be what they expect from written English.

  • 3  Many of its idiosyncrasies are rooted in the history of the language (Viel, 2003; Schottman, 2012)

5If English spelling still appears to be somewhat unpredictable3, researchers today generally agree that there is a coherent, if complex, system of rules that enables speakers to predict pronunciation. From Guierre (1987) to Duchet (2018) whose Code de l’anglais oral was first published in 1993, on to works that appeared soon after phonology was introduced in the AEA (Deschamps et al., 2004; Viel, 2003), they all give a very comprehensive overview of the English spelling-sound system, including detailed information regarding rules, subrules and exceptions. They address a specialist readership, especially candidates for teaching exams, where extensive knowledge of the system is required from a theoretical perspective but also with regard to their mastery of the spoken language, which is a prerequisite for teaching.

6These issues are referred to in recent official instructions for education (Bulletin Officiel, 2019, annexe I:4; annexe II: 4) which specify that teachers should highlight regularities and spelling-sound correspondences in class. I shall now consider which appear to be most relevant for French L1 learners of English.

The Sounds of English: Sound Similarities, Spelling Correspondences and Sources of Confusion

7Some sounds of English represent articulatory challenges. As Flege (1987: 48) explains: “Languages differ both in terms of the number of contrastive sound units they possess, and according to how those units are realized articulatorily.” He identifies three categories of phonemes: those that are identical in L1 and L2, those that are similar which are generally articulated according to L1 articulatory principles, and those that are new.

8There are several consonant phonemes that are either similar (but different) or entirely new. Most represent only slight comprehension or intelligibility issues. Nonetheless, producing the English /r/ sound as a French uvular [ʁ] will inevitably identify the speaker as French, as will inappropriate /h/ insertion (Moore Mauroux, 2010: 3) or h-dropping (which may affect communication). Highly proficient users (C2) will aim to master slight phonetic differences between French and English (for example, /t/ and /d/ articulated as alveolars in English and dentals in French).

  • 4  The CEFRL defines production objectives as follows: intelligibility for A2-B1 learners, expressivi (...)
  • 5  These two varieties are the articulatory reference models for teachers. However, see also Schottma (...)
  • 6  The correspondence does not work from spelling to sound as <th> can be pronounced /t/ in both SBE and </th> (...)

9The new sounds /θ/ and /ð/ are a challenge for French learners even if mispronouncing them will not greatly affect intelligibility4. They represent a specificity of spoken English in both Southern British English and General American (henceforth SBE and GA), the two reference varieties5 for teaching exams, but French learners tend to pronounce them /s/ and /z/ which, like uvular [ʁ], indicates the speaker’s origin. There is, however, a direct spelling-sound correspondence in SBE and GA as /θ/ and /ð/ are always spelt with the digraph <th>6, unlike French where it is pronounced /t/ so knowing the code means learners should know how to pronounce them. I shall concentrate on how the code works for SBE and GA as standard references, but learners will hear different realisations and significant differences need to be highlighted if they affect comprehension.

  • 7  For a detailed analysis of the spelling-sound code, see especially Deschamps et al., 2004; Duchet, (...)

10The vowel system of English functions very differently from French as it uses a lax/tense (or short/long) distinction that is unfamiliar to L1 French speakers resulting in both production and perception issues. The vowel system also includes diphthongs, which, although unfamiliar to L1 French speakers, do not all present the same articulatory challenges. However, once learners have a few simple keys to the spelling-sound code7, the distinction between lax vowels and tense vowels (or diphthongs) is often predictable from the spelling. I shall now consider the pairs or groups of vowels that appear to be most problematic for French learners, using examples inspired from Huart (2010: 85-6) and Moore Mauroux (2010: 5).

/ɪ/ (ship, sit); /i:/ (sheep); /aɪ/ (site)

  • 8  Even if there seem to be limits to this theory in terms of age and/or exposure to the model (Isbel (...)

Differentiating these sounds relies on the lax/tense (diphthong) distinction which French learners have difficulty perceiving and producing, corroborating Flege’s findings (1987: 48) which show a clear correlation between perceiving sounds and being able to produce them8. Yet there are regular correspondences between the type of vowel and its spelling. Providing learners are aware of these distinctions, contextual spelling-sound rules, such as the examples given below, should enable them to predict the type of vowel they need to produce and allow them to decode what they hear. This is particularly relevant for lax (short) vowels where there are few irregularities.
a) Checked vowel contexts <’VC#>, <’VC2> correspond to lax vowels (ship, sit, sitting).
b) Free vowel contexts <’V#>, <’VCe> (silent <e>, also known as “magic E”) (site), and digraphs, <VV> (sheep) correspond to tense vowels or diphthongs.
NB. The following conventions are used: <…>: orthographic transcription (spelling); ‘: stress mark; V: graphic vowel; C: graphic consonant; #: morphological boundary.

  • 9  The basic checked value for <i> is /ɪ/, the free value is /aɪ/.

Unlike /ɪ/, the diphthong /aɪ/ presents little articulatory difficulty, except that it is a new sound, but words spelt <i> will be mispronounced if learners use their L1 spelling-sound code. The sound produced by many French learners for both of these words is close to /i:/, so sit and site then rhyme with the French word gite. Yet the same reading rule applies here to distinguish sit (lax) from site (a diphthong9). Errors like this often occur on transparent words and obviously affect intelligibility.

/e/ and /eɪ/ (sells/sales)

This pair is equally problematic but the confusion between the two sounds is due to syllable structure rather than any real difficulty in articulation. As most French speakers have an allophonic alternation between the sound in an open syllable (‘V#, as in French thé) and the sound in the checked context of a following consonant (‘VC#, as in French thème), they are likely to transfer this allophonic rule to English, thus producing a sound similar to /e/ instead of the required English diphthong which frequently occurs in the sound context /VC#/. Yet the vowel sound can easily be predicted from the spelling: a lax vowel /e/ for sells (<’VC2>), a diphthong in sales <’VCe>.

/æ/ and /ʌ/ (ran/run)

  • 10  See Huart’s work (2010: 99-105) on irregular verbs and pronunciation, including not only vowel alt (...)

Vowel sounds in minimal pairs like ran/run or sang/sung differentiate grammatical categories10 as well as lexical items like hat/hut or bag/bug and as such, are very relevant issues. Both are lax vowels but seem difficult for French learners to differentiate and to articulate, probably because there are no equivalent vowels in French. Huart (2010: 85) suggests that learners approximate /æ/ to the French open vowel in words like patte, all spelt with the letter <a> but this sound is in fact closer to /ʌ/ (which is never spelt <a>). The interference between the two sound systems as well as the spelling-sound codes is confusing for French learners and requires work on sound perception. However, once learners have learnt to differentiate them, the direct correspondence between letter and sound (/æ/ <a>; /ʌ/ <u>) should make their pronunciation easy to predict.

/ɒ/ and /əʊ/ and /ɔ:/ (cod, code, cord)

These three vowel phonemes are probably the most complicated ones in the English vowel system, presenting both articulatory and perception challenges for L1 French speakers. Moreover, like the letter <i>, the letter <o> confuses French learners as it corresponds in French to other sounds and so, as Huart (2010: 86) explains, interaction with the French system is a source of error. There is a real issue in terms of intelligibility and understanding, whatever the native variety, since there is an opposition between different types of vowels (but not the same ones).

11Spelling is a reliable indication of pronunciation for pairs corresponding to /ɒ/ and /əʊ/ in SBE, like hop/hoping (derived from hope) or cod/code, and can also be used to distinguish /əʊ/ and /ɔ:/ in code/cord, although this requires an understanding of the effect of post-vocalic r (r-modification). But if the /ɔ:/ <o+r> correspondence is a regular value, /ɔ:/ also corresponds to the spellings <au> and <aw>. A further source of confusion is that the letter <o> corresponds to a number of other pronunciations, most significantly a large subclass of often very common words pronounced /ʌ/ (mother, son, money), but it also appears in a number of common digraphs, as for example <ou> and <ow> pronounced /aʊ/ and the very exceptional <-ough> which has a number of pronunciations.

12Having provided a diagnostic analysis of particularly relevant learner challenges that an understanding of the spelling-sound code should help resolve, I will now turn to the way these issues are addressed in the phonology subsection of the linguistics paper of the AEA. Reference will be made both to the papers themselves and to the jury reports that provide valuable insight into the way candidates have coped with the various questions over the years. I shall consider what challenges are addressed, how far the focus is on regularities, and whether the questions have evolved over the years as regards the approach to spelling-sound issues. I shall also discuss how far the questions relate to the challenges identified in the first section and which are relevant for EFL teachers at all levels.

The AEA Teaching Exam and Phonology

  • 11AEA Expression orale reports point to recurrent problems such as rising intonation (giving the imp (...)

13A recruitment exam sets out to test that candidates are sufficiently competent in their field; it also needs to be selective as the number of places is limited. This is particularly true of the very competitive AEA which aims to recruit highly competent specialists of English. Since the introduction of phonology in a distinct linguistics paper in 2000, candidates have had to master spoken English as well as to understand and be able to explain the theory behind it. These two objectives are interdependent (Gardelle, 2012: 66) as understanding the system should enable candidates to improve their own pronunciation11; it should also allow them, as teachers, to address the pronunciation difficulties of their students more efficiently.

14Based on a written paper, the questions have always revolved around a few crucial issues: IPA transcription, word stress patterns and stress hierarchy, the spelling-sound code, intonation, and connected speech processes. Candidates need to master certain codes, all of which are relevant for spelling-sound questions: IPA symbols to represent pronunciation in written form, a graphic code to demonstrate spelling-sound correspondences, a numeric code for stress patterns showing both stressed and unstressed syllables.

  • 12  Specific word stress questions, however, will not be studied in detail here.

15The present study focuses on questions referring to three aspects of the system: word stress12, phonetic variation (including GA/SBE variation, strong/weak forms, phonetic processes, both word internally and in discourse) and IPA spelling-sound issues. In all, 597 items were analysed, excluding the text for IPA transcription (question 1) which will nonetheless be included in the discussion. Some 30 items correspond to short phrases concerning connected speech processes across word boundaries. The chart in Figure 1 shows the balance between them.

Fig. 1: Distribution of AEA question categories: 2000-2020.

Fig. 1: Distribution of AEA question categories: 2000-2020.
  • 13  If traditionally phonology has been mainly associated with the speaking competence, it also involv (...)

For each of these categories, the study aims to determine how far explanations focus on the regularities or irregularities of the system, and to what extent the questions are relevant for teaching L2 from both production and perception perspectives13. Note that some items overlap, testing more than one aspect in the same question (word stress and IPA spelling-sound issues, for example).

Word Stress: Morphophonology and Vowel Quality

  • 14  Until 2007, two questions per paper were devoted to word stress, with a minimum of six words per p (...)
  • 15  It is also significant when discussing intonation patterns and connected speech phenomena.
  • 16  <a> 6 times, <e>, <i> and <u> once each.

16Word stress and sound prediction are interconnected since vowel correspondences only apply to stressed syllables. This was explicit in the earlier papers where questions involved a combination of IPA transcription, explanations for word stress placement and comments on the quality of the stressed vowel. Today questions on word stress are distinct from spelling-sound questions and there are fewer of them14 but knowledge of word stress placement is an implicit requirement for IPA transcription15—and correct pronunciation. Indeed, candidates need to identify the status of syllables as stressed (where rules apply) or unstressed (potentially with a reduced vowel) and several papers from 2008 onwards have included questions where the underlined vowel is indeed unstressed and reduced16 thereby highlighting a major challenge for learners.

  • 17  See Deschamps et al. for details (2004: 198-203).

17An understanding of morphophonology helps predict both stress placement and the quality of the stressed vowel. The following chart in Figure 2 below shows the proportion of explanations for vowel quality that rely on rules involving word endings17 (e.g. ION, -iC, Luick, -ate).

Fig. 2: Morphophonology/other explanations for vowel quality.

Fig. 2: Morphophonology/other explanations for vowel quality.

These rules will clearly help candidates improve their own capacity to predict pronunciation but they involve quite advanced theoretical knowledge, which will be less easily transferrable to secondary school learners, depending on their level.

  • 18  For inseparable prefixes, stress can also mark the category with nouns stressed on the “prefix”, v (...)

18Likewise, the distinction between separable and inseparable prefixes needs to be made as the pronunciation of the “prefix” with a full or a reduced vowel depends largely on its stressed/ unstressed18 status. Such words frequently appear in the phonology paper, either in specific questions or within the text for transcription (question 1), and are regularly cited as problematic in reports. Yet there has only been one paper where explanations for the pronunciation of the “prefix” was required and the report (Moore Mauroux, 2018: 79) confirms how tricky candidates found it. Perhaps this is an area where a greater awareness of theoretical considerations would be beneficial.

IPA Transcription and Spelling → Sound

19Vowels
The IPA spelling-sound category (in Figure 1 above) represents 312 items of which 290 are vowels. Figure 3 recapitulates the distribution between the vowel letters that appear in the stressed syllable and/or as an underlined letter for analysis.

Fig. 3: IPA transcription and explanations for vowel quality: number of items per vowel letter.

Fig. 3: IPA transcription and explanations for vowel quality: number of items per vowel letter.

The questions set out to test candidates’ knowledge of spelling-sound regularities and exceptions. The following illustration (Figure 4 below) shows the distribution for monographic vowels for the types of justifications required. Reduced vowels correspond to unstressed vowels; regular rules refer to predictions based on checked/free vowel contexts (modified or not by <r>); morphophonology to word ending rules (as specified in Figure 2) which are particularly efficient for <e> and <i> accounting for, respectively, 48% and 35% of “regular” items. Subrules refer to regular exceptions to predictions based on checked/free vowel contexts; and finally, a few items are exceptional.

Fig. 4: Rules and subrules for vowel quality of monographic vowels.

Fig. 4: Rules and subrules for vowel quality of monographic vowels.

Between 75% and 87% of words spelt <e>, <i>, <u> can be predicted from regular rules or depend on word ending rules. Over 50% of words spelt <a> and <o>, tested in almost every paper, can be explained by the same rules. For the others, knowledge of sometimes quite complex subrules is required: the effect of post-vocalic <l> or of prevocalic <w> before <a>, or “ASK” words, representing one of the major differences between GA and SBE, for example. The <w+a> subrule (21 items) predicts that the vowel is pronounced with the sound /ɒ/ (SBE). It applies to many common words (what, want, watch) and as such is clearly significant for teaching, not just as specialist knowledge; some are rarer words such as squalor and swath.

20As previously stated, the letter <o> is potentially misleading with a large subclass of words spelt <o> and pronounced /ʌ/. Papers include many items (17) corresponding to this subclass thereby raising awareness of this potential pitfall; they also include a number of items corresponding to exceptional pronunciations of this spelling for some very common words: don’t, won’t, gone, woman, women, again significant for teaching as learners understandably tend to mispronounce them.

  • 19  Reports remind candidates that they should learn pronunciations corresponding to this very unpredi (...)

21Vowel digraphs appear less often in the papers except for <ou> which appears at regular intervals testing candidates on individual items and even on series of words spelt <ou> (in 2003 and 2008). Questions focus on items corresponding to irregular and even exceptional pronunciations which generally concern common words (could, should, young, country) including some spelt <ough19> (enough, though, through). This contrasts with the digraph <ea>, for example, which has appeared just 5 times, three of which testing several items (2003, 2005 and again in 2016). The required justifications are very varied, including r-modification, but there has been little specific reference to the significant subclass corresponding to many common words (bread, head…) where <ea> is pronounced /e/.

22If the papers promote an understanding of rules corresponding to regularities especially for monographic vowels, the treatment of digraphs is less systematic. Yet, digraphs are clearly a source of confusion for learners, as for example the <ea> → /e/ subclass and the high-frequency digraphs <aw>, <au> (Deschamps et al., 2004: 175). Likewise, questions tend to focus on regularities or on exceptional pronunciations of common words, but there is rarely any specific reference to their regular/ exceptional status (with the exception of the 2016 paper). Raising awareness of their status would be helpful for teachers who, according to official instructions for phonology objectives, are required to draw attention to regularities (Bulletin Officiel, 2019).

23Consonants
Questions related to consonants often involve IPA traps where the spelling does not correspond to the symbol. Apart from words spelt <VV+gh>, questions include misleading spellings such as doubt or plumber including silent letters. Other questions involve words where different pronunciations of letters distinguish syntactic categories, as in abuse where <s> pronounced /s/ is a noun, /z/ a verb, thereby linking phonology and syntax. Likewise, questions on <n> may require reference to prefix status and to syllable structure (whether or not to pronounce the following <g>). Yet questions on the pronunciation of <n> or <s> also involve the phonetic process of assimilation which is one aspect of phonetic variation.

Phonetic Variation

24Interest in phonetic variation has increased over the years as shown in the graph in Figure 5 below that represents the three categories in Figure 1 year by year.

Fig. 5: AEA questions from 2000-2020 by category.

Fig. 5: AEA questions from 2000-2020 by category.

25Phonetic variation affects spelling-sound questions which can be seen from a different perspective when analyzed in context. Pronunciation is influenced in connected speech (across word boundaries) but also within words, it can vary according to different varieties of English, and words can take on different forms (weak or strong) in discourse. This is clearly an issue for comprehension.

  • 20  Questions on <n> requiring an understanding of assimilation appeared in several of the first papers. </n> (...)

26Reports insist that connected speech processes which have appeared in papers from 200020 are important for both candidates and teachers. The papers have regularly included specific questions, sometimes requiring an analysis of several words with the flexional ending <-ed> (2009 and 2014). Explanations for the pronunciation of this ending require an understanding of how progressive assimilation of voicing functions. It is often incorrectly transcribed by candidates—and mispronounced by learners—on account of the spelling which encourages them to insert a syllable systematically. Like <-s>, these endings are almost always present in the text for transcription.

27Exercises on <-ed> in secondary education often dwell on the voiced/voiceless distinction (a significant theoretical consideration) rather than the addition—or not—of a syllable, which is the main problem for communication. The same approach can be applied to the flexional ending <-s> (tested in the 2020 paper). However, the problem for learners is different, as they will tend not to pronounce it at all leading to potential grammatical errors. Yet, as Huart points out (2010: 127), in connected speech, it is quite tricky to distinguish between these different realisations in many phonetic contexts.

  • 21  They concerned linking phenomena such as linking-r but also /w/ or /j/ insertion and /h/ elision.

28Connected speech phenomena were also tested in the very first papers21 and have been the object of a specific question since 2012. Some processes involve articulatory distinctions (aspiration of initial consonants) which are relevant for production for advanced learners, others involve processes like assimilation and elision, which can affect comprehension at any level, and understanding them should enable teachers to predict some of their students’ issues.

  • 22  According to AEA Expression orale reports, coherence is a real issue for many candidates.

29Variation between GA and SBE, first introduced in 2008, has appeared in all papers since 2017. Candidates are expected to know the significant differences between these reference varieties for two reasons: to improve coherence in their chosen variety22, and to have greater awareness of potential comprehension and production issues for their students who are exposed to many varieties of English.

30As regards weak/strong forms, reducing grammatical words where appropriate will enable advanced learners, a fortiori, teachers, to acquire fluidity and authenticity, but comprehension is the main issue for A2-B2 learners. AEA papers have always tested an understanding of this aspect of English rhythm in the text for transcription, requiring candidates to “use weak forms where appropriate” but reports state that it still represents a major challenge and recent papers have required explanations in specific questions.

31If teaching recruitment exams can focus on learner difficulties such as this one, this will have a positive impact on how well future teachers are prepared to deal with the difficulties of learners in the classroom because it raises awareness of the issues at stake. And indeed, if some items require quite advanced knowledge, my analysis of AEA papers shows that spelling-sound questions tend to test regularities or exceptional correspondences on high-frequency words, ensuring candidates have a solid understanding of the spoken English system as a whole.

Phonology: From Teaching Exams to Continuing Education

32Teaching exams so far have tested speaking competence and theoretical knowledge of the system which are acquired in higher education and while preparing for the teaching exam. Transferring this competence to the classroom is one of the objectives of continuing education and raises two main questions: first, what is most relevant for learners (these priorities were discussed in the first section); second, how can theoretical principles be used to help learners improve their spoken English?

33Devised for a continuing education training session in the Académie de Limoges, Howe proposes a methodological approach and suggests a number of strategies that may be used together with possible activities for the class, adaptable to all levels, as shown in the following recapitulative table (Howe: 2021).

Fig. 6: Working on sounds: strategies and activities from Howe, 2021 (My translation).

Teaching strategies

Possible activities

– l encourage students to observe written-sound regularities

– l use your body as well as your voice

– l analyse students’ errors and understand the source of error

– l guide students so they produce sounds correctly

– l propose exercises that link sound and sense

– l identify potential sources of confusion or ambiguity in documents

– l encourage students to consult dictionaries and other useful references

Pronounce it like a pro
Select 2 or 3 sounds that you want to work on in the module and get the class to repeat them all together.

Create sound bags
Create word lists with the selected sound to work on sound perception and discrimination as well as memorisation.

Focus on minimal pairs
Link sound and sense.

Use IPA flashcards

Link sounds to simple and familiar words that will be used as the sound model throughout the year.

Spot the odd one out
Create a series of 3 words related to the module with one that is different (sound or spelling, depending on the level).

Tongue twisters
Chainspeaking or speaking with a different tempo.

The methodological framework proposed includes recurrent activities, some of which involve the whole class (chainspeaking and “all together now” repetition), others to be done individually or in groups (listening, observation, classification). Key concepts are systematically associated with gestures or graphic representations to help learners assimilate them. For example, to show that reduced vowels correspond to loss of identity, the vowel can be coloured grey. A retracting hand movement towards the body (open to closing) indicates a reduced vowel whereas the prominence of a stressed syllable can be shown by an opening movement of the hand towards the class. Sounds can be represented by showing the sound model on a flashcard with a large pictogram and the corresponding IPA symbol. The flashcards (Howe: 2021) do not include spellings, just pictures or symbols which were chosen because they were familiar: a picture of a cake is the sound model for for /eɪ/; the number 4 is the sound model for /ɔ:/.

  • 23  See Howe & Moore (2021) for a discussion of how concepts regarding prosody may be integrated.

34A number of principles underlie the strategies. Like all linguistic objectives, phonology activities should help learners with the intermediary or final tasks that are set. Likewise, as the system is interconnected, phonology can usefully be linked to lexical or syntactic considerations or to pragmatic objectives. For sounds for example, activities can include words that are selected according to the lexical field of the module and which include the sound (or spelling) to be worked on. The objectives of activities is first to ensure that the key words for the module are understood and pronounced correctly, avoiding potential confusion or ambiguity, whether they appear in documents or are just useful to talk about the subject. The choice of words may also involve other objectives23, depending on students’ levels:
– Drawing attention to regularities in spelling-sound correspondences (the L2 code) which can then be reused later when working on other sounds/spellings;
– Ensuring that the pronunciation of frequently used words is assimilated, whether they are regular (paid, glad), belong to a subclass (nothing, what) or are exceptional (country, women, said);
– Raising awareness of spelling-sound traps, especially on transparent words or words borrowed from English and used in French (identity, cottage).
Sorting activities like “sound bags” and “odd one out” can be linked to both perception and production competences and will enable students to observe regularities as well as potentially surprising pronunciations (particularly in high-frequency words). Activities of this type, if proposed on a regular basis, can also be worked on at home and should enable learners to improve both their spoken English and their listening competence.

35Continuing education training sessions aim to bridge the gap between phonological theory and practical applications, and help qualified teachers transfer their theoretical competence to the classroom, using their experience with learners at various levels to develop appropriate tools and activities and help their students improve oral skills.

Conclusion

36The central premise of this article is that understanding the spoken English system enables teachers to address students’ pronunciation difficulties more efficiently because it helps them predict and understand potential errors and so focus on the most relevant challenges. Indeed, setting up activities such as the ones suggested in Howe (2021) requires a degree of theoretical knowledge which can be acquired in higher education, whilst preparing for teaching exams, and beyond, in continuing education sessions.

37Yet, there is often a gap between theory and the practical needs of the classroom since specialists and learners have different perspectives. Specialists of English learn rules, subrules and lists of exceptions to help them master pronunciation issues, irrespective of the frequency of the words, their complexity and their (ir)regularities, aiming as far as possible to reproduce a near-native model. For learners from A2-B2 level, as official instructions state (Bulletin Officiel, 2019), intelligibility is the main objective. For them, spelling-sound issues are elsewhere: they first need to perceive sounds and to produce them so as to be understood; focus will be on simpler words, high-frequency words and on regular spelling-sound correspondences for English, especially where these enable them to avoid the errors triggered by using their L1 code.

38From 2022, phonology questions will be included in the other national teaching examination, the “CAPES”, including both segmental and suprasegmental aspects of the system that are relevant for teaching. Like AEA, this will surely encourage candidates to develop their knowledge of the spoken English system and improve their own speaking competence. Moreover, the way future teachers are tested will influence not only the candidates’ preparation for the exam but also how prepared they are to help their future students. Indeed, as well as testing theoretical competence, raising awareness of the most significant learner challenges seems to be particularly relevant for candidates who often lack experience in the classroom.

Top of page

Bibliography

ASLAM SIPRA, M. (2013): “Impact of English orthography on L2 acquisition”, English Language Teaching 6.3, p. 116-124.

BOUVET, P. (2021): La Phonologie de l’anglais : à l’école, au collège, au lycée, Paris, Belin Éducation, 504 p.

Bulletin Officiel spécial n° 1 du 22 janvier 2019, annexe I, https://www.education.gouv.fr/bo/19/Special1/MENE1901585A.htm (accessed 20 December 2021).

Bulletin Officiel spécial n° 1 du 22 janvier 2019, annexe II, https://www.education.gouv.fr/bo/19/Special1/MENE1901585A.htm (accessed 20 December 2021).

Cadre européen commun de référence pour les langues : apprendre, enseigner, évaluer. Volume complémentaire avec de nouveaux descripteurs, février 2018, CAPES, sujets zéro (2021): https://saesfrance.org/capes-externe-2022-sujets-zero/ (accessed 26 July 2021).

Conseil de l’Europe, https://eduscol.education.fr/langues-vivantes/actualites/actualites/article/volume-complementaire-du-cecr-avec-de-nouveaux-descripteurs.html/ (accessed 25 May 2021).

DESCHAMPS, A., DUCHET, J.-L., FOURNIER, J-M. & O’NEIL, M. (2004): English phonology and graphophonemics, Paris, Ophrys, 225 p.

DUCHET, J.-L. (2018): Code de l’anglais oral, 3e édition, Paris, Ophrys, 144 p.

FLEGE, J.E. (1987): “The production of new and similar phones in a foreign language: Evidence for the effect of equivalence classification”, Journal of Phonetics 15, p. 47-65.

FOURNIER, J.-M. (2010): Manuel d’anglais oral, Paris, Ophrys, 251 p.

GARDELLE, L. (2012): “Phonologie”, in CHARLOT, C., Rapport du jury de l’agrégation externe d’anglais, session 2018, p. 62-67.

GLAIN, O. (2019): “Phonologie”, in BORT, F., Rapport du jury de l’agrégation externe d’anglais, session 2018, p. 87-106.

GUIERRE, L. (1987): Règles et exercices de prononciation anglaise, Paris, A. Colin-Longman, 128 p.

HOWE, L. (2021): Padlet : Les compétences phonologiques des élèves : https://padlet.com/lila123howe1/phonologie (accessed 4 August 2021).

HOWE, L. & MOORE MAUROUX, S. (2021): “Les mots du système : la prosodie de l’anglais chez l’apprenant francophone”, Recherche et pratiques pédagogiques en langues de spécialité, 40.1, 2021. https://doi.org/10.4000/apliut.8557.

HUART, R. (2010): Nouvelle Grammaire de l’anglais oral, Paris, Ophrys, 258 p.

ISBELL, D. (2016): “The Perception-Production Link in L2 Phonology”, MSU Working Papers in SLS 7, p. 57-67.

MOORE MAUROUX, S. (2018): “Phonologie”, in BORT, F., Rapport du jury de l’agrégation externe d’anglais, session 2018, p. 70-86.

MOORE MAUROUX, S. (2010): “Quelques outils méthodologiques pour aider l’étudiant francophone face à la prononciation de l’anglais”, Cahiers de l’APLIUT 29.3, p. 36-49. https://doi.org/10.4000/apliut.595.

Rapports des jurys de l’agrégation externe d’anglais, sessions 2000-202024, https://saesfrance.org/concours/agregations/agregation-externe/agregation-externe/rapports-du-jury-de-lagregation-externe/ (accessed 25 May 2021).

SCHOTTMAN, W. (2012): Advanced English Pronunciation, Paris, Ophrys, 162 p.

VIEL, M. (2003): Manuel de Phonologie, Paris, A. Colin, 138 p.

Top of page

Notes

1  Aslam Sipra explains that confusion is triggered because the relationship between English spelling and pronunciation is not direct, unlike in Urdu, the basis for his study.

2  The Agrégation externe d’anglais, hereafter referred to as AEA, is a French recruitment exam for teachers that has tested phonological theory since 2000. The CAPES, another secondary education recruitment exam, is to include questions on phonology from 2022.

3  Many of its idiosyncrasies are rooted in the history of the language (Viel, 2003; Schottman, 2012).

4  The CEFRL defines production objectives as follows: intelligibility for A2-B1 learners, expressivity for B2, fluidity from C1, complete mastery at C2. French official instructions insist that intelligibility, rather than the native speaker model, is the aim at A2-B2 level (Bulletin Officiel, 2019), but learners are still expected to produce phonemes with progressively more accuracy.

5  These two varieties are the articulatory reference models for teachers. However, see also Schottman (2012: vi) who provides a number of different models for the exercises in her book. Indeed, if reports from all teaching exams (CAPES, agrégation interne, AEA) insist on the importance for teachers of having a coherent model of English, it need not be one of the reference models.

6  The correspondence does not work from spelling to sound as <th> can be pronounced /t/ in both SBE and GA but this concerns only a small number of words, often proper names like Thomas or Thames.

7  For a detailed analysis of the spelling-sound code, see especially Deschamps et al., 2004; Duchet, 2018; Fournier, 2010; Guierre, 1987; Schottman, 2012; Viel, 2003.

8  Even if there seem to be limits to this theory in terms of age and/or exposure to the model (Isbell, 2016), it still provides a useful framework for work on both articulation and sound perception.

9  The basic checked value for <i> is /ɪ/, the free value is /aɪ/.

10  See Huart’s work (2010: 99-105) on irregular verbs and pronunciation, including not only vowel alternation but geminate consonants as in write/written that correspond to distinctive sounds.

11AEA Expression orale reports point to recurrent problems such as rising intonation (giving the impression of incertitude), misplaced stress especially on pseudo-prefixes, and difficulties with certain vowels (such as the ones reviewed in the first part of this article).

12  Specific word stress questions, however, will not be studied in detail here.

13  If traditionally phonology has been mainly associated with the speaking competence, it also involves comprehension issues. Reports have for many years insisted on the importance of listening to spoken English in various media to prepare for the phonology paper (which is also useful for the oral compréhension-restitution exam).

14  Until 2007, two questions per paper were devoted to word stress, with a minimum of six words per paper, some proposing as many as fourteen! Since 2010, there has been a specific word stress question focusing on just three to five words for which candidates need to indicate the stress patterns (now systematically given in numeric form) and give detailed explanations.

15  It is also significant when discussing intonation patterns and connected speech phenomena.

16  <a> 6 times, <e>, <i> and <u> once each.

17  See Deschamps et al. for details (2004: 198-203).

18  For inseparable prefixes, stress can also mark the category with nouns stressed on the “prefix”, verbs on the stem with reduction on the “prefix”. See Duchet (2018: 24-35).

19  Reports remind candidates that they should learn pronunciations corresponding to this very unpredictable spelling.

20  Questions on <n> requiring an understanding of assimilation appeared in several of the first papers. Recently, questions have involved a variety of phonetic processes recapitulated in an AEA report (Glain, 2019: 100).

21  They concerned linking phenomena such as linking-r but also /w/ or /j/ insertion and /h/ elision.

22  According to AEA Expression orale reports, coherence is a real issue for many candidates.

23  See Howe & Moore (2021) for a discussion of how concepts regarding prosody may be integrated.

24  The SAES archives start in 2010. My thanks to the colleagues who provided me with earlier papers and reports.

Top of page

List of illustrations

Title Fig. 1: Distribution of AEA question categories: 2000-2020.
URL http://journals.openedition.org/ranam/docannexe/image/474/img-1.png
File image/png, 23k
Title Fig. 2: Morphophonology/other explanations for vowel quality.
URL http://journals.openedition.org/ranam/docannexe/image/474/img-2.png
File image/png, 24k
Title Fig. 3: IPA transcription and explanations for vowel quality: number of items per vowel letter.
URL http://journals.openedition.org/ranam/docannexe/image/474/img-3.png
File image/png, 31k
Title Fig. 4: Rules and subrules for vowel quality of monographic vowels.
URL http://journals.openedition.org/ranam/docannexe/image/474/img-4.png
File image/png, 22k
Title Fig. 5: AEA questions from 2000-2020 by category.
URL http://journals.openedition.org/ranam/docannexe/image/474/img-5.png
File image/png, 77k
Top of page

References

Bibliographical reference

Susan Moore Mauroux, “English Pronunciation and the Spelling-Sound Code: What Priorities for Teachers of EFL?”Recherches anglaises et nord-américaines, 55 | 2022, 175-191.

Electronic reference

Susan Moore Mauroux, “English Pronunciation and the Spelling-Sound Code: What Priorities for Teachers of EFL?”Recherches anglaises et nord-américaines [Online], 55 | 2022, Online since 01 February 2024, connection on 22 June 2025. URL: http://journals.openedition.org/ranam/474; DOI: https://doi.org/10.4000/ranam.474

Top of page

About the author

Susan Moore Mauroux

Université de Limoges, EA 3648, CeReS, Centre de recherches en sémiotique

Top of page

Copyright

CC-BY-NC-SA-4.0

The text only may be used under licence CC BY-NC-SA 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search