The Critical Factors of Learner Engagement and Self-Regulation
Applied Linguistics 2014: 35/4: 418–440 � Oxford University Press 2014 doi:10.1093/applin/amu012 Advance Access published on 4 June 2014
Exceptional Outcomes in L2 Phonology: The Critical Factors of Learner Engagement and Self-Regulation
1 ALENE MOYER
1 School of Languages, Literatures and Cultures, College of Arts and Humanities,
University of Maryland
E-mail: moyera@umd.edu
A number of studies attest to the late language learner’s ability to attain native-
like outcomes in morphology and syntax, with accent often the only linguistic
hint of their non-native status. Nevertheless, some do end up sounding native-
like despite a late start. This article explores possible explanations for ’excep-
tional’ outcomes in L2 phonology, specifically, whether such learners’ abilities
are due to innate talent, a metacognitive learning approach, a certain social-
psychological orientation, or specific kinds of experience. Various learners
profiles are compared, an argument is made for learner engagement and
self-regulation, and areas for future research are outlined.
INTRODUCTION
It is no exaggeration to say that those beyond early childhood who aim to
master a new language begin at a vastly different starting point than those who
begin at birth. The second language acquisition (SLA) literature is replete with
theories and hypotheses about why this is so, ranging from neuro-cognitive to
social to psychological explanations including first language (L1) interference,
affective ‘filters’ of one sort or another, the decreasing accessibility of an innate
language acquisition device, social and cultural barriers to assimilation, etc.
(see Bley-Vroman 1989). What is certain is that at least one language is already
in place as a knowledge base, which can imply greater metalinguistic aware-
ness, yet may also be detrimental insofar as L1 cues and patterns are already
salient (see Hansen 2004 for second language, or L2; Kuhl et al. 2008 and
Strange and Shafer 2008 for L1). The first language(s) may limit what the
learner notices in L2, and what she or he is therefore able to emulate at the
level of performance.
According to Selinker (1972), just 5–10 percent of adult language learners
can expect to reach a native-like level, but even this low threshold may be
somewhat ambitious for phonological fluency. Nevertheless, some late lear-
ners do attain a level that can be described as native, or native-like, for some
series of perception-based and/or production-based tasks (e.g. Ioup et al. 1994;
Bongaerts et al. 1995; Moyer 1999). This fact begs two questions that have long
fascinated SLA scholars: (i) What makes a successful language learner? 1
(ii) Why
Downloaded from https://academic.oup.com/applij/article-abstract/35/4/418/2887856 by University of California, Santa Cruz user on 04 January 2018
does phonology uniquely challenge so-called ‘late’ language learners in comparison
with other aspects of language? This article merges both questions by examining
several reasons why some late learners are particularly successful in the realm
of accent.
Scovel (1988) famously asserted that age effects in L2 phonological acquisi-
tion are directly related to neuro-muscular or perceptual skill development,
rather than affective factors. His argument was based on two important prem-
ises: (i) phonology uniquely relies on neuro-muscular faculties for both per-
ception and production; (ii) affective factors could not reasonably restrict
phonology, yet have no effect on other aspects of language ability. Indeed,
shifts in neuro-muscular flexibility and or cognitive mechanisms have long
been assumed responsible for the relative difficulty of learning a new sound
system given that phonology relies on both speech-motor control and audi-
tory-perceptual neural networks. On the other hand, phonology also holds a
unique connection to one’s sense of self, or identity, and therefore speaks to
more than just neuro-cognitive and neuro-muscular constraints. Moreover, it
is undeniable that target language experience shapes one’s approach to acqui-
sition over the long term, and thus the likelihood of native-like attainment.
Evidence confirms correlations between accent ratings and a host of individual
factors, among them: length of residence (LOR) in the target language country,
age of onset/first exposure, and both quantity and quality of experience in the
target language, not to mention motivation and attitudes (e.g. Purcell and
Suter 1980; Thompson 1991; Bongaerts et al. 1995; Elliott 1995; Flege and
Liu 2001; Diaz-Campos 2004; see also Moyer 2013).
In sum, numerous cognitive, social, and psychological factors, both intrinsic
and extrinsic in nature, point towards a possible understanding of exceptional
outcomes. In a sense then, the phenomenon of exceptionality signifies a nexus
for the two dominant paradigms of SLA: a decidedly cognitive or psycholin-
guistic approach on the one hand, and on the other hand, a largely sociolin-
guistic perspective focused on the ‘whole person’. This article argues that
the mysteries of exceptional learning, so rare in L2 phonology, cannot be
explained by either one or the other, but resides at the intersection of both
realms. What can explain the fact that some L2 learners, despite a
late start, end up sounding native-like? Are we to understand them as ‘phono-
logical geniuses’ with extraordinary, innate talents? Alternatively,
do they have special ways of utilizing input, or can they somehow access
linguistic resources in unusual ways? What accounts for their extraordinary
success?
With these questions in mind, I first describe what is generally implied by
‘exceptionality’ in L2 phonology, then present case studies which suggest a
number of common characteristics of their approach to language learning. In
so doing, the relevance of both self-regulation and engagement with the target
language become clear. I conclude by suggesting that the fascination with
some as-yet-determined special talent obscures the need for an integrated
examination of the cognitive, social, and experiential factors that co-vary
A. MOYER 419
Downloaded from https://academic.oup.com/applij/article-abstract/35/4/418/2887856 by University of California, Santa Cruz user on 04 January 2018
with age. The research on exceptionality calls for a dynamic view of learner
engagement with the target language over time in order to understand the
ways that exceptional learners make the most of the available input, and take a
flexible approach, responding to the circumstances at hand.
EXCEPTIONALITY IN L2 PHONOLOGY
To clarify, ‘exceptional’ refers to those who defy the Critical Period Hypothesis
(Lenneberg 1967); they sound native-like even though their exposure to the
target language comes after age 9–10 years (the critical period for phonology is
arguably even earlier, but 9–10 years is a relatively common yardstick in the
research, in keeping with Lenneberg’s original hypothesis). So, which specific
skills or skill sets are implied when we talk about exceptionality, or native-like-
ness, in phonology? By and large, we mean the ability to perceive and/or
produce new sounds like a native speaker would, verified through relevant
tasks which are often isolated or decontextualized (see Levis and Moyer 2014).
Kuhl’s 2007 study on American and Japanese adults confirmed that this is
challenging owing to L1 category salience. Her American listeners could ac-
curately pinpoint the acoustic differences between /r/ and /l/ while her
Japanese listeners could not owing to this contrast’s absence in Japanese.
Further distinctions based on subtle features like vowel quality, aspiration,
and voice onset time (e.g. the difference between /I/ and /E/ or /d/ and /t/) can also be difficult to detect if they are irrelevant in L1. This is likely more
difficult when L1 and L2 features overlap, but are not quite the same, as Flege
and Hillenbrand (1987) have shown for the English vs. French versions of the
phoneme /u/. Instruction and experience can bridge this gap for both produc-
tion and perception (e.g. Flege and Hillenbrand 1987; Rojczyk 2011), even
long-term (e.g. Sereno and Wang 2007), but mastery eludes most L2 learners,
it seems, and even the untrained ear can detect the difference between native
and non-native speech. As shown in Major (2007), listeners completely un-
familiar with the language in question can accurately separate native controls
from non-native speakers, which suggests that there is something unique, and
highly salient, about a non-native accent.
Accent is not just a matter of phonetic or segmental precision. To sound
‘native-like’ the learner must control a number of different features that op-
erate in conjunction with one another, including tempo, rhythm, pause, junc-
ture, pitch patterns, and intonation. Pickering and Baker (2014) confirm that
judgments of accentedness rely on sentence stress (prominence), pause place-
ment patterns, speech rate, and tone choice. (They also point out, however,
that such judgments are prone to listener background variables such as native/
non-native status and attitudes towards the speakers’ presumed backgrounds.)
While tests of such ability are limited to isolated words or phrases, as noted,
some do include a complex range of tasks including spontaneous speech,
which allows for greater confidence in deeming a given learner as ‘exceptional’
(see Moyer 2013). Few such cases have been examined in depth, however.
420 EXCEPTIONAL OUTCOMES IN L2 PHONOLOGY
Downloaded from https://academic.oup.com/applij/article-abstract/35/4/418/2887856 by University of California, Santa Cruz user on 04 January 2018
