Institut numerique

1.2.2. Suprasegmental difficulties and recurrent errors

Although most French EFL learners do not realize it, they have a number of problems with
English rhythm and prosody (Mortreux, 2008), all the more as they often prefer to practise
vowel production, thus producing full vowels only and not realizing vowel reduction
naturally. Burgess and Spencer (2000) used questionnaires that they gave to EFL teachers and
found that stress, rhythm, intonation and vowel reduction were all mentioned as major areas
of difficulty experienced by learners. They remark that this is “all the more interesting as
many pronunciation materials have tended to focus primarily on segmental features” (idem:
197). If one reasons that misproducing rhythm, for example, is “like being out of beat in
music” (Lemmens, 2010: iii), it is easily understandable that suprasegmental errors are just as
important as segmental errors. The prosody of a language should be seen as its basic
structure. The major difficulty of acquisition lies in the fact that two different languages have
differences at the suprasegmental level, and English and French are no exceptions. French
learners automatically reproduce L1 prosodic features, and the resulting errors have a
devastating effect on intelligibility, just as segmental errors do.

In this subsection, the extent to which suprasegmental features constitute a source of
difficulty for French EFL learners is analyzed. The main suprasegmental components of the
English phonological system – namely intonation, stress, and rhythm – are looked at. Also,
the way they may become problematic to French learners, and therefore the more or less
important consequences they have on communication and foreign-accentedness even, are
studied.

Intonation

The suprasegmental feature of English phonology that can be said to be the least problematic
to French EFL learners is intonation. Roach (2009: 3) defines it as “the use of the pitch of the
voice to convey meaning”. Halliday’s (1967) analysis of intonation is quite relevant and has
been taken up many times in the literature. The author divides it into “the three T’s”:
Tonality (the chunking of speech into intonational phrases, or tone-units), Tonicity (nucleus
placement), and Tone (mainly, but not only: fall, rise, and fall-rise). There are some
similarities between the French and English intonational systems, especially concerning
tones and tone meanings. In a nutshell, a rising tone indicates incompleteness, non-finality,
sometimes friendliness and positivity, while a falling tone usually means completeness,
finality, and seriousness (Cruttenden, 1997; Deschamps et al., 2004; Wells, 2006). Nonetheless,
French learners do have a problem with the realization of the English fall-rise tone, probably
out of embarrassment to produce such a different tone from those of their L1. As far as
tonality is concerned, both French and English have “tone-units”, or intonational phrases
(abbreviated IP). The only problem that can be mentioned, though, is the placing of tone-unit
boundaries (| and || for longer pauses) where punctuation is absent. Commas usually align
with tone-unit boundaries, but there can be a tone-unit boundary where there is no
punctuation, e.g. after a long subject or in sentences like: Would you like tea | or coffee?.

The most problematic component of intonation for French EFL learners is tonicity, or the
placement of the nucleus (also called nuclear stress, nuclear accent, tonic accent, or primary
accent), i.e. which word/syllable receives main prominence. The nucleus represents the focus
domain of the intonational phrase, where the information can be new or contrastive.
Vallduví (1991, cited in Rasier & Hiligsmann, 2007: 49) categorizes the accentuation of
Germanic languages as “plastic”, which means that prominence serves to show information
focus, while Romance languages have “non-plastic” accentuation. Contrary to English, the
most prominent syllable in a French IP is the last syllable, regardless of the word. That is
why Vaissière (2002: 11) points out: “In French, focusing, topicalisation and the theme-rheme
distinction are all related to word order and phrasing (there is morpho-syntactically marked
focus), not to differences in prominence”. If French relies on morpho-syntactic devices to
mark information that is in focus, it is understandable that French EFL learners misplace
nuclei when they speak English. The nucleus in the latter language can even be the first
syllable of an IP; such a sentence as C’est moi qui l’ai fait (literally: it is I who did it) is the direct
equivalent of I did it, with oral emphasis, i.e. the nucleus, on I. Besides the recurrent error of
stressing given information instead of new, it is just as common to hear a French speaker
misplace the nucleus in so-called “event sentences”, where the tonicity is unexpected. For
example, in The phone’s ringing(5), the place of the nucleus is unexpected, but possibly
explained by the fact that the event is seen as a whole, and therefore the noun bears the
nucleus (Deschamps et al., 2004; Wells, 2006). As is specified in Rasier and Hiligsmann (2007),
it is easier for a speaker of a language with “plastic” accentuation like English, to produce a
language with non-plastic accentuation like French, than the other way round, hence the
difficulties that French learners have with English tonicity.

Mennen (2006: 1) alleges that “impressions based on intonation may lead to ill-founded
stereotypes about national or linguistic groups”. The contribution of intonation to foreignaccentedness
and the intelligibility of a message is indeed undeniable. As was said above, the
placement of the nuclear syllable is one of the most significant elements in the realization of
English intonation. While any syllable in French can be prominent and bear the nucleus
provided it is in phrase-final position, in English it is first and foremost bound to word stress
patterns. That is why in order to make sure that native speakers will understand the message
and produce correct tonicity, it is necessary to be aware of the notion of lexical stress:
“Prerequisite for the description of intonation, we have to know which syllables are stressed
in words so that we then know which syllables are potentially accentable in utterances”
(Cruttenden, 1997: 15). Di Cristo (2004: 88) says that nuclear accent is in fact at the interface
of the notions of intonation and lexical stress. Consequently, the latter can be regarded as a
basis for intonation, and the problems that French speakers have with it are of paramount
importance in the understanding of the intonational difficulties.

Lexical stress

English and French have totally different views of stress, hence the frequency of errors.
Vaissière (2002: 6) describes how French speakers perceive stress: “The notion of (lexical)
stress is indeed very elusive for French natives. They only discover the existence of that
unnatural and unnecessary complication when they have to learn a foreign language”. Still,
French learners must be aware of the existence of lexical stress in English, as it is a very
different feature from their L1, but also a very important feature for intelligibility. As a
matter of fact, it is one of the few prosodic features of English pronunciation that are taught
in French secondary schools, albeit still rarely. That is probably because lexical stress is
directly linked to some segmental features – e.g. reduced forms of function words and the
schwa. The way the French system differs from the English one can be summarized by
Henry, Bonneau and Colotte’s (2007) remark:

The French lexical accent is essentially correlated to a lengthening of the last
syllable of the word. Thus French learners will tend to keep this lengthening to
English realizations even on unstressed syllables. […] The English lexical accent
is strongly marked on an acoustical point of view whereas the French one is
relatively weak. […] English lexical accent is characterized by a pitch
modification, an increase of intensity and a lengthening of the vocalic nucleus of
the stressed syllable. (1595)

As is specified here, the transfer of the L1 pattern on the L2 production is almost systematic
and unconscious with French speakers, who simply assign equal stress and weight to all
syllables when they speak English. These mis- (or non-)realizations of English stresses can be
illustrated by the widespread overuse of the English word people in French, with the
restricted meaning of “celebrity(ies)”. In this word, apart from the gallicization going as far
as using un people (“a celebrity”) and des people(s) (“celebrities”), the influence of the French
prosodic system made the word be pronounced as /piˈpoel/, or even /piˈpɔl/(6), in which the
vocalic reduction – i.e. to a schwa /ə/ or a syllabic consonant /l̩/ – in the second syllable has
been replaced by a typically French full vowel. Similarly, in polysyllabic words ending in
-age (e.g. village, sausage), French speakers very often use a tense vowel /eɪ/ and stress the
ending, which at the same time is possibly due to the influence of the word age /ˈeɪdʒ/. This
example illustrates how such a suprasegmental feature as lexical stress may be at the origin
of segmental errors. Hodges (2006) illustrates the difficulties that French learners come
across through the series of words derived from ˈdemocrat: demoˈcratic, deˈmocracy. Even
though all three words are closely related, both semantically and morphologically, several
stress rules (e.g. the stress-imposing ending -ic, and the Greek origin of the components)
force the lexical stress to fall on a certain syllable, and that is something that French learners
do not understand easily.

Comprehensibility can be affected by errors involving lexical stress (McNerney &
Mendelsohn, 1992). With stress-alternating pairs, e.g. ˈpresent vs. preˈsent, the primary stress is
on the first syllable if the disyllabic word is an noun or an adjective, but it is on the second
syllable if the word is a verb. When the two words are closely related (e.g. ˈabsent vs. to ab
ˈsent, an ˈinsult vs. to inˈsult), understandability cannot be overly affected by misplacement of
the stress. However, when the two words are only homographs, but in no way related (e.g.
present), a native English speaker might have to think a little before realizing what word was
intended. Similarly, when a compound stress, such as in ˈEnglish ˌteacher (= a teacher who
teaches English), can involve a confusion with a simple phrase stress pattern (adjective +
noun), such as ˈEnglish ˈteacher (= a teacher who is English), French learners make errors that
have an impact on understandability. As a consequence, native speakers might simply stop
communication by dint of mental corrections. It is therefore crucial that teachers should
teach the correct stress pattern of a word immediately when the word is first learned, as is
suggested by Roach (2009: 76): “it would be easier to go back to the idea of learning the stress
for each word individually”.

Even beyond the correct understanding of a word, the primary and secondary stresses of
English words contribute to the overall rhythm of the language. If they are not realized
properly, the whole rhythm is spoiled, English melody is broken, and communication can
become even harder.

Rhythm

Abercrombie (1967: 96) remarks that “all human speech possesses rhythm”. The rhythm of a
language is mainly constituted by the way the language uses stresses (sometimes called
rhythmic beats) and accents (or pitch prominences). Bertrán (1999: 126) reminds the reader
that “in linguistics, the word rhythm is a metaphor, borrowed from music”. If a music had no
rhythm, then it could not be called “music”, and that is exactly the same for the rhythm of a
language. If it is not correctly produced, native listeners will not recognize their language.
That is why the respect of a language’s rhythm is crucial in the learning of the L2
pronunciation.

According to Cruttenden (1997), English rhythm has three degrees of stress/accent: (a)
primary stress/accent – called “nucleus” above – is the principal pitch prominence; (b)
secondary stress/accent is a subsidiary pitch prominence, and is often called the “onset”, i.e.
the first stressed syllable of an IP; (c) tertiary stress (not “accent” this time, as it is not a pitch
prominence) corresponds to the rhythmic stresses of the IP, that is, in the head or tail. In
Wells’s (2006: 229) terms, this third type of stress is said to be “downgraded” in rapid, casual
speech, according to the “rule of three”; the rhythmic stresses between the onset accent and
the nuclear accent are pronounced rapidly along with the other unstressed syllables.
However, it is necessary for French EFL learners to know about the traditional alternation
between stressed and unstressed syllables, especially because it is also linked to segmentals
(i.e. weak forms and strong forms) that French speakers are not familiar with (Mortreux,
2008).

The notion of “foot” given by Halliday (1967: 12) is defined as the component of English
rhythm. Contrary to the foot in poetry, here the foot is a unit of rhythm that consists of an
ictus – one stressed syllable –, and a remiss – the following unstressed syllable(s) before the
next stressed syllable. The rhythm of English is produced by a succession of feet, i.e. of
stressed and unstressed syllables (Abercrombie, 1967: 36). Then, what is a source of difficulty
for French learners is to know what to stress (content words: nouns, verbs, adverbs,
adjective, and demonstratives, question words, etc.), and what is unstressed (function words:
pronouns, articles, conjunctions, etc.). The French system is indeed very different, as has
already been seen with lexical stress. Ploquin (2009: 94) explains that “French differs from
Latin and other Romance languages in that its stress domain is the phrase rather than the
word”. Therefore, when learning English as a foreign language, French learners must
become aware of the difference between the rhythm of their L1 and that of the target
language. The interference of the L1 as far as rhythm is concerned is such that Hahn (2004)
believes that it cannot be avoided. While rhythm is among the earliest things that are
acquired by infants, it is one of the most difficult things for adults to modify when they learn
a foreign language.

Very often, suprasegmental errors lead to segmental errors, e.g. when an error of stress
assignment prevents vowel reduction from occurring naturally (cf. words like village).
Rhythm and prosody are the basic structure of a language, but they are also among the most
difficult features to acquire for an L2 learner. The big differences between the English and
French prosodic systems are at the origin of the difficulties that French EFL learners come
across. The difference between the two languages, and the resulting errors of production, can
be illustrated by the stress-timing and syllable-timing theory.

5 The underlining shows the syllable bearing the nucleus.
6 The word people is now present in monolingual and bilingual dictionaries . e.g. it is in the 2010
French-English Robert & Collins dictionary, translated as gcelebrityh and transcribed /pip.l/. The
pronunciation /pi.pol/, with a closed /o/, has also been heard by some French TV presenters.

Page suivante : 1.2.3. French vs. English: syllable- and stress-timing theory

Retour au menu : EXPERIMENTAL RESEARCH INTO THE ACQUISITION OF ENGLISH RHYTHM AND PROSODY BY FRENCH LEARNERS