Summary “Introducing Second Language
Acquisition”
Chapter 1: Introduction
Second language acquisition means the learning of another language after the first language
(L1) has been acquired. Opinions vary about what might be considered the earliest age from
which second language learning would be differentiated from simultaneous language
acquisition or bilingual first language acquisition. In general, however, second language
acquisition describes learning another language after the early years of childhood.
Note that a distinction between “second” language acquisition and “foreign” language
acquisition is an important one in some contexts. In such cases, “second language acquisition”
applies to circumstances in which the language learned (target language) is generally the
dominant language used in the learner’s environment, while “foreign language acquisition”
indicates that the learner lives in the L1 environment and contact with target language
speakers is not widely available. In this regard, in the context of learning English, a frequent
abbreviation is “ESL” to refer to “English as a second language” and “EFL” for “English as a
foreign language”.
Chapter 2: First Language Acquisition
From sound to word
Considerable research has gone into examining the L1 acquisition process and much of this
information reveals that infants appear to come into the world equipped to acquire the
language they are exposed to in their environment. Linguists often use the term “prewired” to
describe this state of readiness. In fact, many linguists argue that innate structures are the only
reasonable explanation for the rapidity of development and universality of stages that
characterize first language acquisition.
Substantial evidence supports the idea of a genetic predisposition for language. Preferences
for certain sounds of very young infants can be measured using the high amplitude sucking
(HAS) technique. This technique is used to study infant perceptual abilities, through the
recording of an infant’s sucking rate as a measure of its attention to various stimuli. Results:
- Newborns prefer speech sounds to non-speech sounds;
- Young infants prefer looking at the human face and prefer gazing at mouth movements that
move in synchrony with the speech produced by those movements;
- Young children from many different cultures and languages of origin are able to perceive a
multitude of sound differences, even those not occurring in the language of their environment,
an ability known as “sound/auditory discrimination” while adults are often unable to
differentiate those same sounds if they are not used in the native language. This ability
disappears when the sounds are not used in the native language however. It is thought that this
winnowing out of unnecessary perceptive distinctions allows the child to reserve mental space
for those contrasts that are important in his or her language.
Another argument that children come “prewired” for languages is the fact that despite
significant geographic and cultural differences, babies all over the world appear to go through
similar linguistic stages and reach linguistic milestones at similar ages. The first recognizable
1
,pre-linguistic stage is that of “babbling”, when the infant begins to produce a certain number
of distinct sounds, usually taking the form of a consonant-vowel sequence. An early phase of
babbling is sometimes referred to as reduplicated babbling since the consonant-vowel
sequences tend to be repeated (da-da-da). Later on, nonreduplicated babbling begins to
predominate since infants begin to vary the CV sequences they produce (ba-ga-da). Some
correspondence has been found between babbling and later language: the proportion of certain
vocalizations is related to the proportion of those sounds used in the child’s language
environment.
At the age of 1 year, infants start to produce their first words. This “first-word stage” is
sometimes called the “holophrastic” stage, derived from the Greek words “holo” (one) and
“phrastic”(phrase or sentence), referring to the idea that the single words appear to substitute
for the thought conveyed in a full sentence.
A common occurrence in early word learning is the child’s using a word beyond its usual
sphere of reference, known as overextension: e.g. using ‘dog’ to refer to all animals with four
legs. Underextension, when a word is used less broadly than its true domain of reference also
occurs, but this is harder to capture.
From word to sentence
From about 18 months, children begin to put two words together in the same intonational
phrase unit and enter the so called “two-words stage”. Some children go through a “word
spurt”, where new words spring up in the child’s vocabulary on an almost daily basis. Another
important stage occurs when children begin to link together more than two words (two-,
three-, four- or more word long units), and enter the so called “telegraphic stage”. The label
telegraphic is used to reflect the fact that these strings tend to omit function words, and largely
consist of content words.
One method used to examine L1 speech is the system of mean length of utterance (MLU): a
system to measure early linguistic development by calculating the average number of
morphemes, the meaning-bearing units of language, per utterance. Studies found that
children acquire morphemes in similar order. Frequency of the forms in the input, the
language of the child’s environment, did not appear to affect their order of acquisition. Other
studies have found important evidence that children are able to generalize rules to items they
have never been exposed to, revealing that children are able to do more than simply produce a
rote imitation of utterances they are exposed to in the environment. Studies also reveal that
children reorganize their growing grammatical knowledge in systematic ways. By the age of
five or six, complex syntactic constructions and virtually the entire phonological repertoire of
their language are well in place in most children.
Theoretical views
One fundamental distinction underlying various theoretical views about L1 acquisition
revolves around the extent to which language is viewed as basically the result of innate
processes (nativism), and the extent to which environmental factors are considered as
primarily responsible (empiricism). Four views:
Behaviourist view: empiricist view where the child is a passive recipient, subjected
to environmental influences. In this view, language was considered as “verbal
behaviour”, and only what was observable and measurable was accepted as a means
2
, to evaluate language acquisition. Behaviourists explained vocabulary comprehension
through “classical conditioning” or the pairing of a stimulus and a response.
Productive vocabulary is explained by “operant conditioning”: when the child utters a
word that produces the desired effect, then the child is more likely to produce that
word and, in contrast, words that do not trigger hoped-for responses tend to disappear.
This approach also anticipates a role for direct imitation: imitation is considered to be
self-reinforcing, and allows a shortcut so that tedious shaping of each verbal response
is not necessary.
!! Researchers found little evidence for this theory and the view does not explain why
children form utterances that are not heard in the environment.
Universal Grammar: nativist view which suggests that humans possess a “language
faculty”, i.e. a universal set of underlying principles, called UG. UG allows children
to form hypotheses about language when they are exposed to a finite set of examples
from their environment. Without UG, language learning would be impossible because
the input data are insufficiently rich to allow acquisition to occur (= “poverty of the
stimulus”). The theory also specifies that language is unique to humans; other
species’ communication systems are fundamentally different from human language.
Evidence for this theory comes from the fact that all newborns can discriminate
between all human sounds and that children go through the same stages of
development.
!! Too little attention is given to the role of the environment and the crucial role
played by interaction between the child and his/her social network.
Interactionist approach: this approach gives explicit acknowledgement to the
contribution of both innate structures and the role of the environment. The social
interactionist view argues that while there is substantial evidence that innate structures
allow for language acquisition, the role of the environment is more important than
acting as a simple trigger for development, as proposed in UG. Social interactionists
give importance to the interplay between linguistic structures, cognitive abilities, and
the social and linguistic environment. A crucial aspect of this environment is the
speech adaptations to which children are exposed, called child-directed speech
(CDS). Examples of this are shorter utterances, repetition, a heavy reliance on
questions etc. Many researchers suggest that CDS may make it easier for the child to
pick out the important segments in speech.
!! This theory has been criticized for making assumptions that remain untested, e.g.
regarding CDS.
Emergentism - connectionist viewpoint: in the emergentist perspective, language
development is no longer seen as a process of acquiring abstract rules, but as the
emergence of language abilities in real time. This suggests that language emerges out
of a complex network of interconnections between neurons. One predominant
emergentist-inspired model is connectionism, which proposes that language is
learned through exposure to language in the environment, the input. This exposure
allows the construction of associations among units. When the associations are strong
enough, the units and patterns become permanently acquired.
!! It is not clear how connectionism can account for the effect of non-linguistic,
contextual information, such as tone of voice, gestures etc in language acquisition.
3
, One important difference between B and E is that B assumes that children learn languages by
imitation. You repeat what you hear in the input and if people are happy about your repetition,
you get a reward (e.g. mum's smile). So language acquisition is seen as a process of reward
contingencies. Both UG and E approaches posit that children cannot learn from imitation. One
obvious counterexample is that they make mistakes they have never heard in the input: e.g.
goed, bringed.
The main difference between UG (generative grammar) and usage-based (emergentist)
approaches is that proponents of UG believe children can only learn a language because they
have innate knowledge, whereas UB/E approaches claim that learning is possible through
experience. In UG approaches input is only necessary to set your parameters in LAD in the
correct position (e.g. does my language have articles?). And in UB (emergentist) approaches
input is the central learning source. Connectionism is just one type of usage-based (emer)
approaches. It highlights learning as a process of making neural connections. For instance, we
connect the sound 'cat' with an image of a four-legged animal.
First language vs second language acquisition
Differences:
(1) Age: all L1 learners are exposed to their first language in the earliest stages of life at a
time when many other developmental processes are just beginning to occur. Language is an
intrinsic component of the child’s overall cognitive and social development. The child’s
developing language accompanies the unfolding cognitive abilities; certain cognitive
attainments coincide with language markers associated with those cognitive stages, e.g.
object permanence: the awareness that objects don’t vanish when they are no longer visible.
Infants are incapable of advanced reflective thinking and planning. L2 learners on the other
hand learn the target language beyond infancy and early childhood. Their cognitive
development allows them metalinguistic awareness, meaning that they are able to reflect on
language as a tool for thought or learning. Due to this awareness, L2 learners come to the
language learning task equipped in a very different way from the L1 acquirer.
(2) The L2 learner already has a linguistic system. Learning L1 is necessary to satisfy basic
needs. For a L2 learner, one linguistic repertoire is already in place to ensure efficient
communication to satisfy basic needs. This also lead to affective and emotional differences:
while the child’s crucial first events and emotions are accompanied by communication in the
L1, for L2 learners those deeply embedded feelings are already linked with their native
language. L2 learners also deal with a phenomenon called transfer or interference that
occurs when aspects of the L1 are used in the L2. This interference can both hinder (interfere)
and help in the learning process (where structures across languages are similar). From the
perspective of the social context, expectations are very different in the two cases: low for
young L1 learners and high for adult L2 learners.
(3) The context: L1 learners usually learns his first language at home, as well as through
interactions with caregivers outside the home. L2 learners usually learn the target language in
an instructional setting, such as school. Different teaching methods can result in different
outcomes among L2 learners.
(4) Time: L1 learners are commonly exposed to a significant amount of that language, while
L2 learners may be limited to as little as an hour once a week during their classes for example.
Also, CDS is not common for L2 learners.
4
