Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016

ISSN: 2548-5865

PRINCIPLES OF SECOND LANGUAGE ACQUISITION IN CHILDREN

Haerazi (1)
([email protected](1))
Faculty of Education for Language and Art (FPBS)
Mataram Institute of Teacher Training and Education (IKIP)

ABSTRACT
To understand the principles of second language acquisition, we could adopt a
variety of perspective. Research on second language acquisition (SLA) by
children and adults is characterized by many different subfields and perspectives,
both cognitive and social in orientation. Although children feature as participants
in this research, it is relatively rare to find reviews or overviews of SLA that deal
specifically with child SLA although there are a few important exceptions. This
general lack of focus on children‟s SLA is somewhat surprising, considering that
data from children as first language learners have often provided a basis and
impetus for SLA theorizing. Among the best-known first language studies to
prove influential was Brown‟s seminal work showing a predictable order of
morpheme acquisition by children under the age of three. Many early years
settings now welcome children and families from different cultures who use
languages other than English. Young children who are starting to learn English as
an additional language may also be attending a nursery school, pre-school, day
nursery or child-minder perhaps for the first time. They will bring with them
many skills and experiences from their home culture and will be both anxious and
excited about their new situation. A good foundation for learning English as an
additional language is embedded in quality early years practice. To know more
about the principle of second language acquisition in children, this paper will
present some issues related with it such as the nature and the role of language
learning and the logical problem in language learning.

Keywords: The Principles of Language Learning and Second Language

Acquisition

INTRODUCTION children are usually not influenced by

One hypothesis holds that caregivers‟ speech style. Second,
children learn language by imitating children continually produce novel
what adults say, by trying to repeat utterances in two senses. For one
what they hear. However, several thing, they hear a finite number of
facts, showing that there is no sentences, but they come to be able to
necessary similarity between produce and understand indefinitely
linguistic input and linguistic output, many sentences, including vast
militate against this hypothesis. First, numbers they have never heard and
studies of parents‟ speech suggest that therefore cannot be imitating. For

1
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

another thing, children produce surprising, considering that data from

utterances that they cannot have heard children as first language learners
before, because the adult speakers in have often provided a basis and
their environment do not produce impetus for SLA theorizing. Among
them. For these reasons, we have to the best-known first language studies
know the principles of second to prove influential was Brown‟s
language acquisition. The principles seminal work showing a predictable
of second language acquisition were order of morpheme acquisition by
made as foundations to teach second children under the age of three
language in language teaching and (Jenefer Philp, Alison Mackey, and
learning. Rhonda Oliver, 2008: 03).
To understand the principles of Many early years settings now
second language acquisition, we welcome children and families from
could adopt a variety of perspective. different cultures who use languages
Research on second language other than English. Young children
acquisition (SLA) by children and who are starting to learn English as an
adults is characterized by many additional language may also be
different subfields and perspectives, attending a nursery school, pre-
both cognitive and social in school, day nursery or child-minder
orientation. Although children feature perhaps for the first time. They will
as participants in this research, it is bring with them many skills and
relatively rare to find reviews or experiences from their home culture
overviews of SLA that deal and will be both anxious and excited
specifically with child SLA although about their new situation. A good
there are a few important exceptions. foundation for learning English as an
This general lack of focus on additional language is embedded in
children‟s SLA is somewhat quality early years practice.

DISCUSSION an unconscious sort. That is, very

The Nature of Language Learning young children learn how to form
Fundamental to understanding of particular grammatical structures,
the nature of SLA is an understanding such as relative clauses. They also
of what it is that needs to be learned. learn that relative clauses often have
A facile answer is that a second modifying function, but in a
language learner needs to learn the conscious sense they do not know
„grammar‟ of the language target, but that it is a relative clause and could
what is meant by this? What is presumably not state relative clauses
language, how can we characterize are used for.
the knowledge that humans have of Muriel Seville Troike (2006: 12)
language?. All humans acquire a gave us illustration that much of your
language in the first few years of life. own first language acquisition was
The knowledge acquired is largely of completed before you ever came to

2
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

school, and this development hear and respond to two (or more)
normally takes place without any languages in their environment, the
conscious effort. By the age of six result will be simultaneous
months an infant has produced all of multilingualism (multiple L1s
the vowel sounds and most of the acquired by about three years of age).
consonant sounds of any language in As noted in the first chapter,
the world, including some that do not simultaneous multilingualism is not
occur in the language(s) their parents within the usual scope of study in
speak. If children hear English SLA, which focuses on sequential
spoken around them, they will learn multilingualism(L2s acquired after
to discriminate among those sounds L1).
that make a difference in the meaning Our understanding of (and
of English words (the phonemes), and speculation about) how children
they will learn to disregard those that accomplish the early mastery of L1(s)
do not. If the children hear Spanish has changed radically in the past fifty
spoken around them, they will learn years or so, primarily owing to
to discriminate among some sounds developments in linguistics and
the English speaker learns to ignore, psychology. It was once suggested
as between the flapped r in pero „but‟ that first language acquisition is in
and the trilled rr in perro „dog,‟ and large part the result of children‟s
to disregard some differences that are natural desire to please their doting
not distinctive in Spanish, but vital to parents, who wait impatiently for
English word-meaning, as the sh and them to utter a recognizable word.
ch of share and chair. Yet the offspring of even relatively
On average children have indifferent parents successfully
mastered most of the distinctive acquire language at about the same
sounds of their first language before rate. Others argued that children‟s
they are three years old, and an language acquisition is purposive,
awareness of basic discourse patterns that they develop language because of
such as conversational turn-taking their urge to communicate their wants
appear at an even earlier age. and needs to the people who take care
Children control most of the basic of them. This has not proven to be an
first language grammatical patterns adequate explanation, however, since
before they are five or six, although within children‟s limited sphere of
complex grammatical patterns activity, communicative needs seem
continue to develop through the to be largely satisfied by gesture and
school years. such non-speech sounds as squeals,
The same natural and generally whines, grunts, and cries.
effortless learning processes take Perhaps the most widely held
place when there is significant view by the middle of the twentieth
exposure to more than one language century was that children learn
in early childhood. If young children language by imitation (the

3
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

stimulusresponsetheory). While it is explained as imitations at all, since

true that much of children‟s initial they can never have heard them
language learning can be attributed to before. The two following terms
their imitation of sounds and words mostly influenced in language
around them, many of their utterances learning.
are quite original and cannot be

The role of natural ability doing so if they had to “learn” them

Humans are born with a natural in the usualsense of that word.
ability or innate capacity to learn Children‟s ability to create new
language. Such a predisposition must utterances is remarkable,and their
be assumed in order to explain ability to recognize when a string of
several facts: (1) Children begin to common words doesnot constitute a
learn their L1 at the same age, and in grammatical sentence in the language
much the same way, whether it is is even more so. Forexample, children
English, Bengali, Korean, Swahili, or acquiring English L1 can recognize
any other language in the world. (b) early on that Cookiesme giveis
Children master the basic ungrammatical. They have never been
phonological and grammatical told, surely, that the particulargroup
operations in their L1 by the age of of words is not an English sentence,
about five or six, as noted above, but they somehowknow, nevertheless.
regardless of what the language is. (c) If a child had to consciously learn the
Children can understand and create set of abstractprinciples that indicate
novel utterances; they are not limited which sequences of words are
to repeating what they have heard, possible sentencesin their language as
and indeed the utterances that opposed to those that are not, only the
children produce are often smartestwould learn to talk, and it
systematically different from those of would take them many more years
the adults around them. (d) There is a than it actuallydoes. This is part of
cut-off age for L1 acquisition, beyond “the logical problem of language
which it can never be complete. (e) acquisition,”which is discussed
Acquisition of L1 is not simply a further below.
facet of general intelligence. The role of social experience
In viewing the natural ability to Not all of L1 acquisition can be
acquire language in terms of innate attributed to innate ability, for
capacity, we are saying that part of language specific learning also plays
language structure is a crucial role. Even if the universal
genetically“given” to every human properties of language are
child. All languages are incredibly preprogrammed in children, they
complex systemswhich no children must learn all of those features which
could possibly master in their early distinguish their L1 from all other
years to thedegree they succeed in possible human languages. Children

4
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

will never acquire such language- many children do not receive this
specific knowledge unless that type of input and still develop
language is used with them and language at essentially the same rate.
around them, and they will learn to Sources of L1 input and
use only the language(s) used around interaction vary depending on cultural
them, no matter what their linguistic and social factors. Mothers‟ talk is
heritage. American-born children of often assumed to be the most
Korean or Greek ancestry will never important source of early language
learn the language of their input to children, but fathers or older
grandparents if only English siblings have major childrearing
surrounds them, for instance, and they responsibilities in many societies and
will find their ancestral language just may be the dominant source of input,
as hard to learn as any other English and wealthier social classes in many
speakers do if they attempt to learn it cultures delegate most of the
as an adult. Appropriate social childrearing responsibilities to
experience, including L1 input and nannies or servants. The relative
interaction, is thus a necessary importance of input from other young
condition for acquisition. children also varies in different
Intentional L1 teaching to young cultures, as does the importance of
children is not necessary and indeed social institutions such as nursery
may have little effect. Some parents schools.
“correct” their children‟s immature L1 Versus L2 Learning
pronunciation and grammar but most This brief comparison of L1 and
do not, and there is no noticeable L2 learning is divided into three
change in rate of acquisition among phases. The first is the initial state,
children who receive such instruction. which many linguists and
Some adults simplify both grammar psychologists believe includes the
and word choice, adding more underlying knowledge about language
complex structures as the child does, structures and principles that is in
but adults‟ notion of “simplicity” learners‟ heads at the very start of L1
does not correspond to the actual or L2 acquisition. The second phase,
sequence in language acquisition. the intermediate states, covers all
Some adults imitate children‟s stages of basic language
language production, and in this development. This includes the
imitation, they sometimes provide maturational changes which take
expansions of children‟s structures place in what I have called “child
(such as saying Yes, that’s abig, grammar,” and the L2 developmental
brown dog in response to the child sequence which is known as learner
saying That dog). The expansion may language (also interlanguage or IL).
play a role in developing children‟s For this phase, we will compare
ability to understand new forms, but it processes of L1 and L2 development,
cannot be considered necessary since and then compare the conditions

5
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

which are necessary or which learners‟ ability to judge which L2

facilitate language learning. The third utterances are not possible (such as
phase is the final state, which is the the Cookies me give example
outcome of L1 and L2 learning. mentioned above), an aspect of
Initial state children‟s L1 competence which is
While the initial state in attributed to innate capacity. Many
children‟s minds for L1 almost surely who take a social perspective tend to
is an innate capacity to learn reject such judgments of (un)
language, it is not at all certain grammaticality as convincing
whether or not suchnatural ability is evidence because they result from
part of the initial state in older artificial tasks which do not include
learners for L2 acquisition(hence the actual circumstances of L2
“?” in 2.2). Some linguists and interpretation and use. Many who
psychologists believe thatthe genetic take a psychological perspective in
predisposition which children have turn reject socially constituted
from birth to learn languageremains evidence (such as natural language
with them throughout life, and then production) because the many
differences in thefinal outcomes of L1 variables which go along with actual
and L2 learning are attributable to social usage cannot be controlled for
other factors.Others believe that some experimental investigation. So,
aspects of the innate capacity which although the question of the extent to
childrenhave for L1 remain in force which innate capacity for language
for acquisition of subsequent acquisition remains available in SLA
languages, butthat some aspects of is a very interesting and important
this natural ability are lost with one, it is likely to remain unresolved
advancing age. Stillothers believe that for some years to come.
no innate capacity for language There is complete agreement,
acquisition remainsbeyond childhood, however, that since L2 acquisition
and then subsequent languages are follows L1 acquisition, a major
learned by meanswhich are more akin component of the initial state for L2
to how older learners acquire other learning must be prior knowledge of
domains ofknowledge, such as L1. This entails knowledge of how
mathematics or history. language (in general) works, as well
Because it is impossible for us to as a myriad of language-specific
observe mental capacity for language features which are only partially
learning directly, the different beliefs relevant for production of the new L2.
are based largely on theoretical This prior knowledge of L1 is
assumptions and are tested by indirect responsible for the transfer from L1
methods which individuals who come to L2 during second language
from different disciplinary development, which we will consider
perspectives may not agree on. For as part of the second phase of L1
example, many linguists rely on versus L2 learning.

6
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

L2 learners also already possess Processes

real-world knowledge in their initial Development, as we have seen, is
state for language acquisition which a spontaneous and largely
young children lack at the point they unconscious process in L1 child
begin learning their L1. This has grammar, where it is closely
come with cognitive development and correlated with cognitive maturation.
with experience by virtue of being As noted above, as children mature,
older. The initial state for L2 learning so do their language abilities. In
also includes knowledge of means for contrast, the development of learner
accomplishing such interactional language (or inter-language) for L2
functions as requesting, commanding, learners occurs at an age when
promising, and apologizing, which cognitive maturity cannot be
have developed in conjunction with considered a significant factor; L2
L1 acquisition but are not present in learners have already reached a level
the L1 initial state. of maturity where they can
The initial state of L1 learning understand and produce complex
thus is composed solely of an innate utterances in their L1, and level of
capacity for language acquisition maturity is not language-specific.
which may or may not continue to be Processes other than maturation must
available for L2, or may be available be involved to explain development
only in some limited ways. The initial in SLA. Just as we cannot directly
state for L2 learning, on the other observe mental capacity, we cannot
hand, has resources of L1 directly observe developmental
competence, world knowledge, and processes, but we can infer from the
established skills for interaction, utterances which learners understand
which can be both an asset and an and produce at different stages what
impediment. processes are possibly taking place.
Intermediate state This addresses the fundamental how
Both L1 and L2 learners go question of SLA, which we will
through intermediate states as they explore from different perspectives in
progress from their initial to their the chapters which follow. While
final state linguistic systems. There is answers to this question vary, there is
similarity in that the development of general agreement that cross-
both L1 and L2 is largely systematic, linguistic influence, or transfer of
including predictable sequencing of prior knowledge from L1 to L2, is
many phenomena within each and one of the processes that is involved
some similarity of sequencing across in inter-language development. Two
languages, and in the fact that L1 and major types of transfer which occur
L2 learners both play a creative role are: Positive transfer, when an L1
in their own language development structure or rule is used in an L2
and do not merely mimic what they utterance and that use is appropriate
have heard or been taught. or “correct” in the L2; and Negative

7
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

transfer (or interference), when an interaction generally facilitates SLA,

L1 structure or rule is used in an L2 it is not a necessary condition. It is
utterance and that use is inappropriate possible for some individuals to reach
and considered an “error.” a fairly high level of proficiency in
Cross-linguistic influence occurs L2 even if they have input only from
in all levels of IL: vocabulary, such generally non-reciprocal sources
pronunciation, grammar, and all other as radio, television, or written text.
aspects of language structure and use. The role of input and interaction in
Positive transfer facilitates L2 SLA is also discussed in next chapter.
learning because an L1 structure or Facilitating condition
rule that also works for L2 means that While L1 learning by children
a new one doesn‟t have to be learned. occurs without instruction, and while
For example, a word that has the rate of L1 development is not
essentially the same form and significantly influenced by correction
meaning in both languages can of immature forms or by degree of
transfer appropriately from L1 to L2: motivation to speak, both rate and
e.g. exterior „outside‟ is a word in ultimate level of development in L2
both Spanish and English can be facilitated or inhibited by
(pronounced differently, but with the many social and individual factors.
same spelling and meaning). Negative Identifying and explaining facilitating
transfer of L1 features can often be conditions essentially addresses the
inferred from forms in the second fundamental why question of SLA:
language which are unlike any that why are some L2 learners more
are likely to be produced by a native successful than others? Some of the
speaker of the L2, or are an conditions which will be explored in
integration of elements which would chapters that follow are: (1)
not occur in monolingual speech. feedback, including correction of L2
Necessary condition learners‟ errors; (2) aptitude,
Language input to the learner is including memory capacity and
absolutely necessary for either L1 or analytic ability; (3) motivation, or
L2 learning to take place. Children need and desire to learn; and (4)
additionally require interaction with instruction, or explicit teaching in
other people for L1 learning to occur. school settings.
In contrast, while reciprocal social
Final state into adulthood, the basic phonological
The final state is the outcome of and grammatical systems of whatever
L1 or L2 learning. The final state of language(s) children hear around
L1 development by definition is them are essentially established by
native linguistic competence. While the age of about five or six years (as
vocabulary learning and cultivation of we have already noted), along with
specialized registers (such as formal vocabulary knowledge and interaction
academic written style) may continue skills that are adequate for fulfilling

8
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

communicative functions. This is a requiring no extraordinary aptitude or

universal human achievement, effort.
The processes of Language This is well known that English
Learning in Children learners over regularize irregular past
Language Learning through tense verbs and say for example goed
Imitation
instead of went and singed instead of
One hypothesis holds that
sang, although they have never heard
children learn language by imitating
these forms, because adults do not use
what adults say, by trying to repeat
them. In the same vein Guasti,
what they hear. However, several
Thornton, and Wexler (1995) have
facts, showing that there is no
found that English-speaking children
necessary similarity between
aged 4-5 years produce negative
linguistic input and linguistic output,
questions with the form in (15). No
militate against this hypothesis. First,
adult utters such sentences; thus,
studies of parents' speech suggest that
children cannot have learned them by
children are usually not influenced by
imitation.
caregivers' speech style. Newport,
Gleitman, and Gleitman (via Maria Language Learning through
Guasti, 2008: 11) have shown that a Reinforcement
high proportion of parents' utterances Behaviorist psychologists have
are questions (What do you want?) claimed that language is learned
and commands (Get the toy car!) and through the mechanism of reinforcing
only 25% are simple declaratives. By the contingent association between
contrast, simple declaratives are the stimulus and response, the same
first kind of sentence that children general-purpose mechanism that is
mostly produce. Second, children invoked to explain other learning
continually produce novel utterances, processes in animals and in humans.
in two senses. For one thing, they According to this view, children learn
hear a finite number of sentences, but language because they are positively
they come to be able to produce and reinforced when they produce correct
understand indefinitely many verbal expressions, negatively
sentences, including vast numbers reinforced when they make errors.
they have never heard and therefore Although the learning-through-
cannot be imitating. For another thing reinforcement hypothesis is simple, it
(and this is the most compelling cannot explain how humans acquire
evidence against the acquisition- language and cannot characterize
through-imitation hypothesis), human linguistic competence, as
children produce utterances that they Chomsky (1959) details in his review
cannot have heard before, because the of Skinner 1957. First, like the
adult speakers in their environment acquisition-through-imitation
do not produce them. hypothesis, it cannot explain the fact
that children acquire competence over

9
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

an indefinite number of sentences: associations between input (stimulus)

they understand and produce and output (response) patterns.
sentences they have never heard and Interactions among these units
produced before, that is, for which no give rise to behavior that simulates,
reinforcement was provided. Second, sometimes very accurately and
parents generally pay attention to precisely, actual human behavior. A
what children say and not how they network consists at least of input and
say it. If a child asks a question, the output units connected by modifiable
adult will hardly check for its weighted links. During the learning
grammatical correctness, but will phase the network is presented with
simply answer. examples of both input and output.
Language Learning through Given an input, the network is
Association modifiable the weights of its
connections so as to produce the
Another hypothesis about how
correct output. After learning, the
language acquisition occurs is
network can generalize to new stimuli
expressed by an approach called
provided they belong to the same
connectionism, neural networks, or
class of stimuli used in the training
parallel distributed processing. At the
phase. Notice that in these models
outset it is worth noting, as does
neither nodes nor links correspond to
Marcus (2001), that the term
linguistic categories or rules. These
connectionism is ambiguous.
are represented in the network by
Generally it is associated with the
various patterns of activation among
idea that brain circuits do not support
links. Here we will briefly look at
the representation of symbols and
some linguistic phenomena
rules; connectionist models are thus
connectionists have sought to account
usually opposed to models in which
for, noting simply that many intricate
symbols are manipulated. However,
aspects of language acquisition and of
in addition to symbol- and rule-free
human linguistic competence still
models, there exist connectionist
await explanation within a
models whose goal is to explain how
connectionist approach.
symbolic manipulations can be
implemented in a neural substrate. Universal Grammar
The remarks that follow apply to
Universal Grammar (UG)
models that aim at eliminating continues the tradition which
symbols and rules. Connectionist Chomsky introduced in his earlier
models or artificial neural networks work. Two concepts in particular are
are inspired by a coarse metaphor of still of central importance: (1) what
the brain, in that they consist of needs to be accounted for in language
several interconnected neuronlike
acquisition is linguisticcompetence,
processing units modified by learning or speaker-hearers‟ underlying
knowledge of language. This is

10
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

distinguished from linguistic fragmentary linguistic experience”

performance, or speaker-hearers‟ (Chomsky 2002:8). In other words,
actual use of language in specific while children‟s acquisition of the
instances. (2) Such knowledge of specific language that is spoken by
language goes beyond what could be their parents and others in their social
learned from the input people receive. setting requires input in that
This is the logical problem of language, the acquisition task is
languagelearning, or the poverty-of- possible (and almost invariably
the stimulus argument. successful) because of children‟s
built-in capacity. One of the most
Chomsky and his followers
important issues in a UG approach to
have claimed since the 1950s that the
the study of SLA has been whether
nature of speaker-hearers‟
this innate resource is still available
competence in their native language
to individuals who are acquiring
can be accounted for only by innate
additional languages beyond the age
knowledge that the human species is
of early childhood.
genetically endowed with. They argue
that children (at least) come to the The Logical Problem of Language
task of acquiring a specific language Learning
already possessing general knowledge How is it possible for children
of what all languages have in to achieve the final state of L1
common, including constraints on development with general ease and
how any natural language can be complete success, given the
structured. This innate knowledge is complexity of the linguistic system
in what Chomsky calls the language which they acquire and their
faculty, which is “a component of the immature cognitive capacity at the
human mind, physically represented age they do so? This question forms
in the brain and part of the biological the logical problem of
endowment of the species” (Chomsky
languagelearning. The “problem” as
2002:1). What all languages have in it has been formulated by linguists
common is Universal Grammar. relates most importantly to syntactic
If a language faculty indeed phenomena. As noted in the
exists, it is a potential solution to the preceding section, most linguists and
“logical problem” because its psychologists assume this
existence would mean that children achievement must be attributed to
already have a rich system of innate and spontaneous language-
linguistic knowledge which they learning constructs and/or processes.
bring to the task of L1 learning. They The notion that innate linguistic
wouldn‟t need to learn this underlying knowledge must underlie language
system, but only build upon it “on the acquisition was prominently espoused
basis of other inner resources by Noam Chomsky (1957, 1965),
activated by a limited and who subsequently formulated a

11
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

theory of UniversalGrammar which to at least partially

has been very influential in SLA comprehend.Unlike many L2
theory and research (to be discussed learners, children almost never
in Chapter 3). This view has been receive any explicitinstruction in L1
supported by arguments\such as the during the early years when
following: acquisition takes place, andthey
seldom receive any negative
Children’s knowledge of language
evidence, or correction (and often fail
goes beyond what could be learned
torecognize it when they do).
from the input they receive
This is essentially the poverty- Constraints and principles cannot
of-the-stimulus argument. According be learned
tothis argument, children often hear Children‟s access to general
incomplete or ungrammatical constraints and principles which
utterancesalong with grammatical govern language could account for
input, and yet they are somehow able the relatively short time it takes for
to filterthe language they hear so that the L1 grammar to emerge, and for
the ungrammatical input is not the fact that it does so systematically
incorporatedinto their L1 system. and without any “wild” divergences.
Further, children are commonly This could be so because innate
recipients ofsimplified input from principles lead children to organize
adults, which does not include data the input they receive only in certain
for all of thecomplexities which are ways and not others. In addition to
within their linguistic competence. In the lack of negative evidence
addition,children hear only a finite mentioned above, constraints and
subset of possible grammatical principles cannot be learned in part
sentences, andyet they are able to because children acquire a first
abstract general principles and language at an age when such
constraints whichallow them to abstractions are beyond their
interpret and produce an infinite comprehension; constraints and
number of sentenceswhich they have principles are thus outside the realm
never heard before. Even more of learning processes which are
remarkable, children‟s related to general intelligence.
linguisticcompetence includes According to Jackendoff (1997:
knowledge of which sentences are not 5) approaches this capacity in
possible,although input does not children as a “paradox of language
provide them with this information: acquisition”: If general-purpose
i.e.input “underdetermines” the intelligence were sufficient to extract
grammar that develops. Almost all L1 the principles of mental grammar,
linguisticinput to children is positive
linguists (or psychologists or
evidence, or actual utterances byother computer scientists), at least some of
speakers which the children are able whom have more than adequate

12
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

general intelligence, would have frequently say things like goed, mans,
discovered the principles long ago. mouses, and sheeps, even though it is
The fact that we are all still searching highly unlikely that any adult around
and arguing, while every normal child them ever produced such forms in
manages to extract the principles front of them. We also know that
unaided, suggests that the normal children do not learn language simply
child is using something other than by imitation because they do not
general-purpose intelligence. imitate adult language well when
asked to do so.
Universal patterns of development
cannot be explained bylanguage-
specific input
Linguistic input always consists
of the sounds, words, phrases,
sentences, and other surface-level
units of a specific human language.
However, in spite of the surface
differences in input (to the point that
people who are speaking different
languages can‟t understand one
another), there are similar patterns in
child acquisition of any language in
the world. The extent of this
similarity suggests that language
universals are not only constructs
derived from sophisticated theories
and analyses by linguists, but also
innate representations in every young
child‟s mind.
For a long time, people thought
that children learned language by
imitating those around them. More
recent points of view claim that
children have an innate language
ability. There are three major
arguments supporting this notion.
First of all, children often say things
that adults do not. This is especially
true of children's tendency to use
regular patterns to form plurals or
past tenses on words that would have
irregular formation. Children

13
Journal of English Language Teaching Volume 3 Nomor 1, Februari 2016
ISSN: 2548-5865

REFERENCES

Kay, Crosse. (2007). Introducing English as an Additional Language to Young

Children: A Practical Handbook. Paul Chapman Publishing.

Lightbown, P. M., & Spada, N. (1999). How Languages are Learned (Second
Edi). Oxford: Oxford University Press.

Bialystok, E., & Hakuta, K. (1994). In Other Words: The Science and Psychology
of Second-LanguageAcquisition. New York: Basic Books.

Gass, S.M., & Selinker, L. (2008). Second Language Acquisition: An Introductory

Course. New York: Routledge Taylor & Francis Group.

Troike, M.S. (2005). Introducing Second Language Acquisition. Cambridge:

Cambridge University Press.

14