In everyday speech, it is very common to come across such phrases as it’s not what you said
but the way you said it, or a situation in which a mother disapproves of her child’s tone (Wells,
2006). These simple instances seem to provide some evidence that not only phonemes, but
the global structure of a spoken language, and more particularly its prosody, have a crucial
role to play. When it comes to the acquisition of a foreign language, suprasegmental
features(1), i.e. literally the features “above the segments” such as stress, rhythm, and
intonation, thus play a major part in pronunciation skills, intelligibility, and foreignaccentedness.
In this respect, Mennen (2006) contends:
Just as poor pronunciation [of individual sounds] can make a foreign language
learner very difficult to understand, poor prosodic and intonational skills can
have an equally devastating effect on communication and can make conversation
frustrating and unpleasant for both learners and their listeners. (3)
It is true in both French and English that communication can easily be broken, e.g., if the
intonation of such-or-such sentence is not realized properly – an intended polite request
pronounced with a low pitch and a falling tone can be perceived as an abrupt command.
Similarly, if rhythm and stress patterns are not respected as they are when one speaks a
foreign language, the intelligibility and comprehensibility of the message will be affected. In
fact, just like rhythm and beats in music – hence the borrowing of the words in phonology
(Bertrán, 1999) –, rhythm and stresses in language constitute a basis for accurate
pronunciation. By the same token, Busà (2008: 118) believes that “focusing on stress, rhythm
and intonation can help learners to improve their overall pronunciation, and to sound more
natural, and can lead to more comprehensible speech as well as better understanding of
other people’s speech”. However, non-native speakers may not be quite aware of it and tend
to focus on the production of vowels instead, especially if their native language differs
considerably as far as rhythm and prosody are concerned.
The present study was motivated by the observation that French learners of English as a
Foreign Language (abbreviated EFL) very often fail to be properly understood by native
English speakers in spite of a sometimes impeccable pronunciation of segments (e.g.
Birdsong, 2003). When French speakers attempt to speak English, most of the time they use a
typically French rhythmic pattern, rather than take into account the English alternation
between stressed and unstressed syllables and the vowel reduction phenomenon that
naturally results from that. They also have difficulty with what Halliday (1967) calls tonicity,
or the placing of the nuclear accent (also called nucleus, tonic, or tonic accent), which might
cause a wrong focus on new or contrastive information, or simply a stronger, unpleasant
foreign accent. The obvious differences between English and French phonologies are at the
origin of the problems that French speakers have with English pronunciation. Not only the
stress and accent systems, but more generally the prosodic structures of the two languages
share hardly any characteristics. Moreover, contrary to written English, one of the difficulties
that arise about spoken English is that it may not be so easily understood by French learners,
and even any foreign learners, who might first hear nothing but a single stream of sounds.
Oral language does not seem, at first, to be segmented like written language and its clear-cut
punctuation, and learners cannot rely on such things as orthographic similarities to
understand a word. In this respect, French speakers do not notice the differences in stress
patterns, rhythm, etc. Instead, they might only point that the English language has a
different, particular – or peculiar – melody. That is why the role of foreign language teachers
is to help learners overcome their difficulties and avoid recurrent errors.
In French secondary schools, where foreign languages and especially English are
compulsory, there has been apparent priority given to the teaching of grammar and
vocabulary. Teachers correct their students’ lexical or grammatical mistakes, but not so often
mispronunciation, apart from the unexpected irregular pronunciation of certain words such
as recipe /’res?pi/ or psychiatry /sa?’ka??tri/(2) (personal interview with two English teachers,
March 2011). Ploquin (2009: 40) goes as far as saying that pronunciation teaching “is seen as
some form of gilding” in language classes, which means that foreign language teachers
intentionally dismiss the teaching of pronunciation as a superficial element. In addition,
when pronunciation is taught, the articulation of individual sounds is emphasized, mainly
with exercises on the distinction between lax vowels and tense vowels, at the expense of
prosodic features. Still, one could argue that the role of suprasegmentals is just as important,
all the more as a French learner of English is not naturally aware of the great differences
between his/her native language and the target language.
As is specified in Derwing and Munro (2005: 386), “it is widely accepted that
suprasegmentals are very important to intelligibility, but as yet few studies support this
belief”. Separate studies on English and French phonologies, including an interest in
prosody for a few decades, are not lacking. However, cross-linguistic studies on these two
different systems are scarce. Vaissière (2002: 1) mentions “the lack of a comparable way of
describing prosody in different languages”; giving the obvious prosodic differences between
the two languages, it is hard, in the author’s view, to analyze the two prosodic systems with
the same tools. Very few experiments have been carried out on the acquisition of English
phonology by French learners, and even fewer on the acquisition of suprasegmentals (e.g.
Tortel, 2009). Accordingly, one of the central objectives of this project is to elaborate a pilot
study, leading to a future full-scale one that will enable us to have evidence of the
importance of prosody with respect to segments. A parallel between the acquisition of
English segmentals and suprasegmentals by French EFL learners should be drawn so as to
arrive at better EFL teaching methods.
The procedure of the pilot experiment that we mean to conduct is as follows: by use of
carefully-created stimuli, and after a first control recording of their pronunciation of English,
French speakers will be divided into two separate groups with a different focus each. One
group will receive a standard training on English segments, and the other group will focus
on rhythm and prosody. Then, after a second recording, native English speakers and experts
in English phonology will rate the productions in a blind evaluation, and it will become
possible to have a comparative insight into the effect of each training on the production skills
of the subjects. The latter will be evaluated and compared to determine which aspect of
English pronunciation has the better effect on their production, but also to witness the
evolution within the groups and the efficiency of a prosodic training in comparison with a
segmental training. In the scope of this work, it is the elaboration of the stimuli and of the
whole experimental protocol that is especially focused on, with an emphasis placed on read
speech. With the results of the pilot experiment, the way will be open to future between- and
within-groups experiments that will take into account the effect of longer trainings on
perceptual skills, read speech, and spontaneous speech of a larger number of participants.
Apart from the direct goal of the experiment, this research also involves the field of teaching
and didactics. Future findings will bring into light which aspects of English pronunciation
teachers should insist on rather than simply giving more importance to vowels and taking
no, or at least little, heed of stresses, accents, rhythm, and intonation.
Therefore, the central hypothesis of this study is that prosody, i.e. suprasegmental features
of English as a Foreign Language, are as important as, indeed more important than,
segmental features in speech intelligibility, communication, and foreign-accentedness. Being
understandable and understood by native English speakers is not necessarily a question of
articulatory phonetics. Rather, producing rhythm correctly and placing stresses and accents
where they should be may prove to have more weight in intelligibility and communication, if
not to be sufficient. The underlying claim is that the teaching of rhythm and prosody in EFL
class contexts should be re-evaluated, regarded as a primary aspect of English pronunciation,
and no longer underestimated. The hypothesis will be verified through the elaboration and
results of the experiment after reviewing the existing literature on the same research domain.
In Chapter 1, English pronunciation from the point of view of French EFL learners is
examined. It is indeed necessary to single out learners’ pronunciation problems before
applying future findings to the field of teaching. The chapter includes the place of the
teaching of pronunciation among the teachings of grammar and vocabulary, as well as a
parallel between the teaching of segments and that of suprasegments when pronunciation
happens to be present. Then, a descriptive overview of the most common segmental and
suprasegmental difficulties that French speakers come up against will serve to analyze some
typical production errors, and the extent to which they might lead to unintelligibility and
stronger foreign-accentedness. Together with a brief word on the syllable-timing/stresstiming
distinction, that will help us understand why French learners become discouraged
when it comes to learning English pronunciation, and why the suprasegmental aspect of
spoken English is overlooked by teachers. At the same time, the account of the most
predictable errors by French EFL learners will go hand-in-hand with the creation of the
stimuli of the experiment, detailed in Chapter 3.
In order to elaborate the experiment, it is necessary to have a clear view of previous and
ongoing studies on the acquisition of English phonology. That is dealt with in Chapter 2. One
section is devoted to the acquisition of English as a First Language (abbreviated L1), with a
particular focus on the acquisition of segments compared with the acquisition of prosody in
the first years of life. The development of both perception and production capacities in
infants is looked at. The other section is on Second and Foreign Language (L2) Acquisition(3).
However, given the pivot of the present work, the acquisition of phonology in the context of
Foreign Language Acquisition is especially emphasized. The whole section comprises the
acquisition of L2 segmentals, the acquisition of L2 suprasegmentals, and an account of the
few comparative studies on their respective roles. The parallel between L1 and L2 acquisition
will serve not only to have an insight into the processes of acquisition of segments and
suprasegments, but also to compare L1 acquisition with L2 acquisition, leading to the
question of the order of importance between segments and suprasegments in L2 teaching.
The resulting overview will enable us to enhance how our project fits in the literature, and
consequently to accentuate its interest and originality.
In Chapter 3, the objective and procedure of the pilot experiment are explained. Above all,
the elaboration of the stimuli and trainings is detailed, and a description of all participants –
subjects and listeners – is provided. Chapter 4 dwells on the results of the experiment, and
therefore concludes on the possibility or not to use the experiment as a basis for future
research in the context of the acquisition of English as a Foreign Language by French
speakers.
Finally, Chapter 5 is a general conclusion on the whole experimental research, and it
provides an answer to the central hypothesis. It also highlights the limitations of the study
and gives perspectives for further work.