Institut numerique

2.2.1. Segmentals and L2 acquisition

As was said in the previous section, the influence of the L1 phonological system starts very
early in newborn infants. In the first years of life, while the L1 influence develops rapidly but
is still quite recent, the acquisition of an L2 remains easy for these early learners, contrary to
late learners like teenagers or adults. Celce-Murcia, Brinton and Goodwin (1996) explain:

It is undoubtedly the case that adults will acquire the phonological system of a
second language in a manner different from that of their first language, given
that the acquisition of the new sounds in the second language must be integrated
into already existing neural networks. (16)

Thereby, Flege (1992) underlines the obvious difference between the acquisition of L1 sounds
and the acquisition of L2 sounds. While L1 acquirers are newborns and have no other
linguistic influence – hence their universal capacity previously mentioned –, late L2 learners
already possess a whole phonetic system based on the L1, as the L1 influence has kept
growing over time (Corder, 1967: 163). Therefore, L2 production errors are inevitable. In fact,
late learners tend to analyze L2 phonemes in terms of the L1 phonetic inventory quasisystematically,
and that triggered off the emergence of some famous theories in the field of
L2 phonology acquisition.

The Critical Period Hypothesis, put forward by Lenneberg (1967, cited in e.g. Celce-
Murcia, Brinton & Goodwin, 1996), is well-known and often debated. According to it, when
someone has passed a certain age, it is no longer possible for him/her to attain native-like
pronunciation of a language. This is redolent of Corder’s (1967: 163) supporting comment:
“[…] the learning of the mother-tongue is part of the whole maturational process of the child,
whilst learning a second language normally begins only after the maturational process is
largely complete”. More and more studies affirm that the critical period occurs around the
age of six. At the onset of puberty, that is, the alleged end of the critical period, the huge
influence of the L1 existing phonemic categories, but also brain lateralization – or the loss of
plasticity of the brain – prevent the prepubescent from achieving complete mastery of a
language at the phonological level. The fact that the productions of L2 sounds by early
learners are better if compared with those by late learners is said to be evidence of the
Critical Period Hypothesis (Flege, 1992).

The Contrastive Analysis Hypothesis (CA) proposed by Lado (1957, cited in e.g. Pica, 1994:
52) also inspired many other theories of L2 acquisition. It holds that the elements in the L2
that are similar to those in the L1 are simple for the learner to acquire, whereas the elements
that are different are harder to acquire. Thereby, L2 acquisition involves a degree of crosslinguistic
similarity, and the L2 sounds are filtered through the L1 system. In the cases of
dissimilar structures, the phenomenon of interference occurs, that is, the influence of the
learner’s linguistic knowledge on the acquisition of the L2 (Cebrian, 2003: 2). In Rasier and
Hiligsmann (2007), this is referred to as “negative transfer”. Conversely, when the L1
influence (i.e. transfer) enhances L2 acquisition, then it is “positive transfer”. The term
“interlanguage” has been coined to refer to the ever-evolving language system that L2
learners mentally create. It consists of transfers from L1 to L2, input from the L2, and
“universals”, i.e. forms that are present neither in the L1, nor in the L2 (Vergun, 2006: 11). If a
negative transfer “fossilizes”, the future correction of the error becomes almost impossible.
The Contrastive Analysis hypothesis was often disfavoured, however, because it fails to
predict which particular L2 sounds are easy to acquire, and which are difficult and likely to
become fossilized.

Similarly, Kuhl’s (1991) Perceptual Magnet Effect theory (PME), already mentioned in 2.1.1.
about the influence of the L1 on the child’s perceptual capacities, implies that the sound
prototypes that are formed and based on the L1 interfere with adult learners’ ability to
perceive L2 contrasts. The underlying concept is that the L1 predetermines the perception –
and thereby production – of the L2. The L1 prototypes act as perceptual magnets that attract
not only other members of the same category, as was explained in 2.1.1., but they also attract
L2 sounds. This theory was motivated by the fact that infants are universal listeners and can
discriminate any phonetic contrasts, whereas adults have difficulty in discriminating
contrasts that are not in their L1. Age is therefore an important factor in language acquisition,
and a young age seems better for acquiring a second/foreign language. Baker and
Trofimovich’s (2005: 3) experimental study also confirms that “in late bilinguals, L2
perception and production are often influenced by the L1 at least in the beginning stages of
L2 learning”.

This theory is in total agreement with Flege’s (1992, for example) Speech Learning Model
(SLM) of second language sound acquisition. According to it, non-native phonemes are
classified in terms of L1 phonemes on the basis of similarity. If an L2 sound is similar to an L1
sound, then the learner will classify it in an already existing phonetic category that was
developed during L1 acquisition, and the acquisition of the actual L2 sound will not be easy.
On the contrary, if an L2 sound does not resemble any L1 sound and is considered as new,
then it is acquired more easily because the difference is more obvious. This automatic
process is referred to as the Equivalence Classification Hypothesis (Flege, 1992: 572). L1 and
L2 speech sounds thus interact through two distinct mechanisms (Flege, Schirru & MacKay,
2003): phonetic category assimilation – the formation of a new category is blocked as long as
the L2 sound is identified, or “equated”, with an L1 sound –, and phonetic category
dissimilation – a new phonetic category is established for the L2 sound. If an L2 sound
happens to be identical to an L1 sound, no new phonetic category needs to be created, and the
sound is produced correctly. Contrary to Lado and the Contrastive Analysis Hypothesis,
Flege (1992) mentions a possible factor that enables one to determine if an L1 sound and an
L2 sound will be perceived as similar or not; if they are represented by the same symbol in
the International Phonetic Alphabet, then they might be classified as similar, and vice versa.
Flege’s model confirms that the earlier the L2 is learned, the better perception and
production skills will be. Yet, it has some limitations, because it focuses on the production of
vowels especially.

Very close to Flege’s SLM is Best’s (1995) Perceptual Assimilation Model (PAM), briefly
mentioned in 2.1.3. This theory centres on the discrimination of non-native phonemes at the
level of perception, as opposed to the SLM. The major point is that non-native sounds may be
perceptually assimilated in three ways. They can be: categorized exemplar of some native
phoneme – the similarity between the L1 sound and L2 sound is very strong; uncategorized
consonant or vowel that is similar to two or more L1 phonemes; non-assimilable non-speech
sound, with no detected similarity to any native phonemes. As an illustration of this theory,
Best’s team conducted experiments with English speakers’ perception of Zulu and Tigrinya
contrasts. The listeners who participated perceptually assimilated and discriminated the
non-native consonants with respect to their phonetic similarity to native contrasts, in
accordance with predictions from the PAM.

Best, MacRoberts and Goodell (2001: 776) underline that Best’s, Flege’s, and Kuhl’s models
“all presume that adults’ discrimination of non-native speech contrasts is systematically
related to their having acquired a native speech system”. The difficulties encountered by L2
learners are indeed linked with the proximity between L1 sounds and L2 sounds. As far as
empirical studies are concerned, the production of vowels has especially been at the core of
the studies on the acquisition of segments (Carlotti, 2007). Ploquin’s (2009) experiment, for
one, investigated the perception of English vowels by French L2 learners, with a
discrimination task involving pairs like He want fish and He won’t fish. The difficulty in
recognizing English phonemic contrasts was confirmed. In addition, Flege (1992) points out
the lack of studies on the link between the perception and the production of an L2, so that it
is not always clear whether difficulties find their source at a motoric level or at a perceptual
level.

A parallel between the aforementioned theories and Lenneberg’s Critical Period
Hypothesis can be drawn. According to the latter, brain lateralization before puberty makes
it impossible for late learners to achieve native-like L2 production. The various models that
were looked at above seem to confirm that the acquisition of L2 segments is hard, all the
more as the L1 influence is stronger and stronger with age. L2 sounds are constantly
analyzed according to the L1 existing phonetic system. Nonetheless, some researchers like
Birdsong (2003) attest that native-likeness is possible even in late learners. Thorén (2008)
noticed that some adults have acquired native-like pronunciation, and some learners who
started learning an L2 before the age of six display a foreign accent. In the experiment
conducted by Bongaerts et al. (1997), very good Dutch-speaking learners of English, i.e. late
L2 learners, and a control group of native English speakers, were recorded and rated by
linguistically inexperienced native English speakers on a 5-point scale. Some of the L2
learners actually received the same scores as native speakers. Therefore, the authors conclude
that it is not impossible to achieve authentic, native-like accent. Still, the experiment was
limited to Dutch speakers, and perceptual skills were not tested.

Whether late learners can attain native-likeness or not, the influence of the L1 phonetic
inventory on the perception and production of L2 sounds is undeniable. Furthermore, as is
assumed in many theories, young acquirers form mental phonetic categories that are based
on the L1, hence the problems that late learners have with the acquisition of the L2
phonology. Although these theories focus on the perception and production of L2 phonemes,
and especially vowels, Thorén (2008: 20) emphasizes that the “category building”
phenomenon could also apply to prosodic contrasts and categories. Yet, very few studies
have investigated prosodic features in the L2 acquisition process.

Retour au menu : Experimental research into the acquisition of English rhythm and prosody by French learners