ACADEMIC YEAR 2013-2014
YEAR ONE: INTRODUCTION TO LANGUAGE LEARNING AND TEACHING
#1: THEORIES OF FIRST LANGUAGE ACQUISITION
(extracted from: Majer, J. 2010.  First language acquisition . In Lewandowska-Tomaszczyk, B. (Ed.), New
Ways to Language. A
ódz
: A
ódz
University Press. 317-334)
1. Child-directed speech  input for first language acquisition
Predisposition to acquire language is sufficient for speech to emerge on condition that a normally developing
child is actually exposed to samples of natural linguistic data and given opportunities to use language for
communicative purposes. Studies of the so-called  wolf children  infants badly treated by adults, e.g. kept
in dark closets and isolated from a speaking community  indicate that unless INPUT (i.e. the linguistic data
in the form of speech directed at the child and from which he/she can acquire language) is granted in the
early years of life, the linguistic system, particularly grammar, may simply never be acquired successfully.
However unfortunate such cases may be, they nevertheless enable researchers to assess the extent to which
language is innate and capable of developing despite the lack of adult input.
As pointed out by Field (2004), extended duration of the DEPRIVATION of linguistic input is believed to
coincide with the timing for BRAIN LATERALIZATION (typically shortened to LATERALIZATION), i.e. the view
popular in neurolinguistics that one part the brain, typically the left hemisphere, assumes a responsibility for
language. The author argues that:
Children deprived of language go on to develop full linguistic competence if they are brought into
society before the age of about eight or nine; those who are rescued at a later stage may develop an
extensive vocabulary but often manifest an incomplete system of syntax. Hence a conclusion that
lateralization was closely associated with the acquisition of a first language and that, if acquisition
were to be fully successful, it had to occur during the period of flexibility (p. 80).
A famous example of extreme deprivation of linguistic data is the case of Genie, a child who was not
exposed to language while she was growing up. Locked away in an attic, she was very seldom spoken to for
the first thirteen years of her life. When she was finally set free from her imprisonment in 1970, Genie was
unable to speak and never learned language properly, despite the efforts by linguists and therapists who tried
to teach her. As pointed out by Pinker (1994), she was able to associate the meaning of certain objects with
words, but could not put these words together to formulate comprehensible, grammatical sentences. The lack
of access to linguistic input before the age of puberty had caused the girl s innate capacity for language
acquisition to deteriorate. Because Genie evidently missed the opportunity predetermined by the biological
timetable to acquire language fully, her case has been used as a strong argument for the CRITICAL PERIOD
HYPOTHESIS.
Fortunately, however, nearly all babies are raised in circumstances which provide no impediments to
language acquisition and so they begin to talk spontaneously. That is no doubt owing to the distinctive
quantitative and qualitative characteristics of language input in the speech used by parents, older siblings,
babysitters, etc. This modified register, referred to as  baby talk ,  mother talk ,  motherese ,  caretaker
speech , CAREGIVER TALK or, perhaps more appropriately, CHILD-DIRECTED SPEECH, is characterized by a
number of formal adjustments when compared to the language directed at more mature speakers. For
example, it is highly unlikely for the following series of questions to occur in normal adult native-speaker
discourse (after Snow 1977):
(mother to little daughter) What else have you got in your face?
Where s your nose?
Where s you nose?
Ann s nose? (1)
1
How do caregivers modify their speech when addressing babies? Firstly, utterances are relatively short
and grammatically simpler, with e.g. reductions of inflections or avoidance of the first and second person
pronouns (as in Mummy s coming). Secondly, adults use many repetitions, tutorial questions (i.e. the kind
that caregivers know answers to) and tag questions. Thirdly, while caregivers avoid abstract or difficult
words, they show a preference for diminutives ( baby morphemes , as in doggie), its own lexical variants
(such as beddy byes), and unique lexicon items (i.e. words reduced to canonical forms typically found in
children s output, e.g. rabbit > wabbit). Finally, adults also make significant phonological adjustments such
as clearer enunciation and exaggerated intonational contours, higher pitch and increased volume, or slower
speech rate and longer pauses which are supposed to ensure greater salience of the major constituents in a
sentence. All these simplifying, clarifying and expressive phenomena, combined with facial expressions and
other non-vocal modalities, tune the input to the perceptive sensitivity of the child and are thus powerful
clues to comprehension.
Parents and other caretakers are also known to modify the propositional content of their speech in order to
help babies understand what is being said. The most obvious characteristics of such message adjustments are
reference to the  here-and-now deictics and provision of contextual clues. Typically, a caregiver refers to
the things that are going on at the moment of speaking and that are within the perceptual range of the child s
senses, i.e. what can be seen, heard, felt, smelled or tasted directly. Besides, a great deal of repetition and
rephrasing is provided. In this way child-directed speech ensures a great deal of REDUNDANCY, which is
indispensable for the infant to decode what is being said when linguistic resources aiding information
processing are still limited.
A question is in order, namely whether the social register known as caregiver talk can be treated as an
explicit instructional mode. Do parents, babysitters, etc. make a conscious effort to teach children language?
Or does this form of input and interaction serve different functions? First of all, if we accept that babies are
born with innate language faculty, then they do not need to be taught language. Nor could they really
understand explicit instruction, given their cognitive immaturity. In actual fact, it is not until they are three to
four years old, i.e. when the essentials of grammar have already been acquired, that they begin to
comprehend simple metalinguistic comments. Besides, most parents would not be in a position to act as
language teachers and explain the complexities of the system to their children.
To be sure, child-directed speech may indeed have certain characteristics of TEACHER TALK (i.e. the
modified register used by instructors to promote communication with learners), while children may be
engaged in echoing and imitation, i.e. forms of practice. Yet because they are not aware of being in the
process of language acquisition and because their motivation for learning stems from a desire to understand
and to express themselves, caregiver-child interactions are different pedagogical situations from classroom
procedures, however informal the latter may be.
How do we find out about the form and content of such interactions? Insights can be obtained through
direct observation and analysis of written records such as diaries. It follows from those studies that mothers,
fathers and other caretakers seem to pay much less attention to the formal correctness of their children s
speech than might be surmised. Furthermore, rather than concentrate on the inadequacies of grammar, lexis
or pronunciation, they are quick to correct mistakes in point of fact. For the same reason, they respond if
their offsprings utterances are not appropriate socially. On the other hand, children s non-target output
features such as phonological reductions and substitutions, word coinages or morphological and structural
overgeneralizations seem to be easily accepted by adults and they may even be  borrowed into the family
discourse to accompany the unique lexicon and hypochoristics (affectionate nicknames and diminutives)
typical for caregiver talk. However, as the child grows and participates in discourse as a fully-fledged
speaker, there is progressively less modified input in the adult speech.
Recapitulating, caregivers adapt their input in systematic ways when talking to children. Field (2004)
argues that  many of these modifications potentially assist the child in the bootstrapping process of
identifying words and recognizing phrase boundaries, or in making matches between words and objects in
the real world (p. 134). But the fundamental aims of child-directed speech are to provide affection,
socialize the infant and shape its behaviour. That is why NEGATIVE COGNITIVE FEEDBACK  such as
correction  is often overridden by POSITIVE AFFECTIVE FEEDBACK. Accordingly, the instructional function,
which is carried out implicitly and indirectly, is secondary to clear communication and socialization. All of
2
these characteristics make caregiver talk a special variety of adult speech. As it turns out, the language
acquisition process needs precisely that kind of register as input.
2. Theoretical approaches to first language acquisition
There exist several conflicting theoretical approaches to the growth of competence in L1 in little children.
For example, language acquisition is believed to be the following different concepts:
·ð a process of habit formation (behaviourism);
·ð the effect of the activation by input of an innate language faculty which is fully developed at birth
(mentalism/nativism);
·ð a process which must be completed within a critical period (the innateness hypothesis);
·ð the effect of the child s desire to interact with its caregivers (interactionism);
·ð a process of tracing patterns in diversified input owing to general cognitive predisposition
(cognitivism).
These major positions, as well as some other approaches related to the mainstream theories, are dealt with
briefly in the remainder part of this handout.
2.1. Early behaviourism
Some early 20th century theorists explained all learning in terms of a certain form of CONDITIONING. The
best-known representative of that position was the Russian psychologist Ivan P. Pavlov (1849-1936), who
used dogs while working on conditioned reflexes and who developed the classical conditioning theory with
its S-R (STIMULUS-RESPONSE) dyad. In the first half of the 20th century, the American psychologist John B.
Watson (1878-1958) demonstrated how phobias could arise out of a normally innocuous stimulus (like a
white rabbit). His research before World War I marks the official launching of BEHAVIOURISM as the new
paradigm for the major portion of American experimental psychology.
The behaviourist theory of learning is firmly based on the empiricist view that it is possible to study only
what can be sensed directly. Therefore, rather than speculate about the internal operations going on in the
mind, researchers representing this movement in psychology confine themselves to external manifestations
of behaviour. Byram (2000) points out that central notions in behaviourism  were linked to a positivist view
of the scientific method, which accepts only phenomena that are observable and measurable as worthy of
serious attention (p. 74).
2.2 Modern behaviourism
The behaviourist approach to first language acquisition is best associated with the works of Burrhus
Frederic Skinner (1904-1990). In his most important publication, entitled Verbal Behavior and published in
1957, he put forward the view that learning was the result of environmental rather than genetic factors.
Introducing the notion of OPERANTS, i.e. the range of behaviours that organisms perform or are capable of
performing, he also elaborated on the theory of CLASSICAL CONDITIONING by emphasizing the role of
positive REINFORCEMENT  concrete or abstract reward that a learner gets upon a successful completion of a
task. Skinner s principles came to be called OPERANT CONDITIONING, in which a formation of the association
between a stimulus and a response becomes established because it is strengthened through reinforcement or
REWARD. As underlined by Byram (2000: 74),  reinforcement of stimulus and response patterns leads to
repetitious behaviour or habits, which are to some extent predictable, based upon previous experience .
Summing up, in modern behaviourism learning a language is conceived to be a process of habit formation
based on the triadic system of S-R-R (STIMULUS-RESPONSE-REINFORCEMENT). Since developmental change
is triggered by reward contingencies, no genetically transmitted, innate information is necessary for speech
to emerge in a child. On the other hand, crucial roles in L1 acquisition are ascribed to imitation, repetition,
positive reinforcement and correction.
3
2.3 The mentalist/nativist position
Skinner s theory was devastated in a review of his book published in 1959 by the eminent American linguist
Noam Chomsky, who represented a trend opposed to empiricist theories of language acquisition called
MENTALISM. He put forth an idea of a linguistic faculty called LANGUAGE ACQUISITION DEVICE (LAD)
which includes innately specified structural properties constraining the possible forms human language could
assume. As explicated by Goodluck (1991),  the child has, as it were, blueprints for all the possible types of
language in her head [& ]. In the course of language development, she settles on the particular grammar of
the language surrounding her (p. 140). And as further added by Jordan (2004: 132),  this native endowment
[& ] can be seen as a  black box : children receive a certain amount of language input from their
environment which is processed in some way by the LAD so that they end up with their linguistic
competence . Because the hypothesized predisposition to acquire linguistic structure is believed to be
present at birth, the theory of language acquisition in question is also referred to as NATIVISM. The
fundamental claim of that approach is that the child s mind has content (i.e. knowledge in the form of
concepts and propositions) independent of experience (Valian 2009: 15).
According to the mentalist/nativist position, all children are born with an innate grammar that enables
them to generate and utter sentences they have never heard before (Jordan 2004: 124). The role of linguistic
input is to activate the parameters adequate for a particular language pattern from among the syntactic
universals contained in LAD. Nativists claim that without some kind of innate learning mechanism it would
be difficult to account for the remarkable ease and speed at which infants acquire such a complex system, if
one considers that  they are easily distracted, have great difficulty with tasks that require conscious
memorization, and have very limited working memory (Field 2004: 150).
Another strong mentalist/nativist claim in favour of LAD is the POVERTY OF THE STIMULUS argument.
Indeed, children do not seem to induce rules of grammar directly from the input they hear since they are
exposed to all the characteristics of natural connected speech which contain little language that could serve
as a model for acquisition. For example, when adults around a child interact, their turns might overlap or
they might interrupt one another. What is more, they may change their mind in mid-sentence, never finish
what they were planning to express originally, or even stop altogether. Thus, according to Chomsky (1965),
the input described above as caregiver talk can be considered  degenerate . This is because, as pointed out
by Field (2004), it  exemplifies only a limited range of the possible sentences of the language. The child is
exposed to a range of speakers, with different voices, intonation patterns and accents. Finally, the input
provides examples of language performance when the child s goal is to develop competence (p. 186).
Jordan (2004) writes that:
On the basis of degenerate input children produce language which is far more complex and rule-
based than could be expected, and which is very similar to that of other adult native speakers of the
same language variety, at an age when they have difficulty grasping abstract concepts. That their
production is rule-based and not mere imitation as the behaviourist view held, is shown by the fact
that they frequently invent well-formed utterances of their own (p. 125).
A question is in order: how could the infant build a grammar relying solely on such impoverished data?
The poverty of the stimulus argument inspired Noam Chomsky to put forward an idea that was an
elaboration on the earlier concept of LAD, namely UNIVERSAL GRAMMAR (UG)  a common structural basis
for all languages which contains a limited set of rules for organizing a linguistic system. UG is believed to
reduce the number of hypotheses about L1 structure formed by the child, as well as confine the incidence of
potential errors. According to Goodluck (1991: 144), the child analyses input  into strings of words via a
sentence processing mechanism which is made up of (i) existing rules of the language being acquired and/or
(ii) principles of UG.
Since building a competence is based upon very sparse linguistic resources, i.e. with no lexicon on which
to base the identification of the sound sequences found in the input and with minimal clues as to the location
of word boundaries in connected speech, it has been suggested (e.g. by Field 2004) that the child must rely
on  some kind of technique which gives it a head start  just as straps can help one to pull on a pair of boots
(p. 38). The same author adds that  a theory of syntactic bootstrapping postulates that infants reach
conclusions about words on the basis of their inflections and other grammatical properties: thus the child
4
learns that the difference between It s sib and It s a sib serves to distinguish real-world entities that are mass
from those that are count (2004: 63). Yet however accurately the BOOTSTRAPPING metaphor (referred to
earlier in this course; cf. Handout #1) may reflect the task of L1 acquisition, it still does not indicate whether
LAD or UG is a unique, autonomous cognitive module, independent from human intelligence, or perhaps
part of general cognitive problem-solving faculty (Jordan 2004: 149). This dilemma remains a fundamental
question posed before modern PSYCHOLINGUISTICS  a domain shared by linguistics and psychology aiming
to study the relationship between language and the human mind.
2.4 The innateness hypothesis
A view closely related to the mentalist/nativist position is the INNATENESS HYPOTHESIS. It, too, suggests that
language development is biologically determined and that it is enough for a child to have access to linguistic
input in order to trigger the process of L1 acquisition. However, what broadens the concept of nativism as
described in the preceding section, according to the theory of the neurobiology of language propounded by
Eric Lenneberg (1967), there is a fixed timetable for L1 acquisition. The researcher in question claims that
LATENT LANGUAGE STRUCTURE (the counterpart of LAD in Chomsky s mentalist/nativist approach), like
other biological functions, operates successfully only when it is activated at the right time  a time referred to
as the CRITICAL PERIOD.
Lenneberg s innateness hypothesis can be supported by three powerful arguments, viz.:
·ð A systematic sequence of stages  i.e. a predictable order in which a child develops his/her linguistic
abilities; for example, babbling always emerges before the one-word stage, which in turn typically
occurs before the onset of telegraphic speech, etc.
·ð A progression independent of external stimuli  i.e.  a biologically timed programme that is not
dependent on exposure to speech (Goodluck 1991: 141). Not only does babbling occur
physiologically in deaf infants, but also six-week-old babies are known to be able to discriminate
between speech sounds that are categorized as by adults distinct linguistic signs.
·ð A critical period  early and middle childhood appears to be a decisive age, after which L1
acquisition can be impaired since certain linguistic abilities may no longer be attainable. Evidence is
provided not only by case studies of the aforementioned  wolf children , but also by the commonly
observed decreasing capability of post-pubertal students to learn L2 as successfully as L1.
The first two of the above arguments refer to the concepts to be described later in this course in connection
with the stages of first language acquisition (see Handout #2), while the third has important implications for
second language acquisition and foreign language learning.
2.5. The cognitivist approach
The mentalist/nativist and innatist positions appear to be in opposition not only to empiricist traditions in L1
acquisition research, but also to the COGNITIVIST APPROACH, whose most eminent representative is thought
to be the Swiss psychologist Jean Piaget. Whereas the former theories tend to ignore cognitive prerequisites
for language development by concentrating first and foremost on linguistic forms, the latter view attaches
considerably more significance in the constitution of the human mind to processes and factors such as
perception, meaning, thought, memory and emotion.
In Piaget s (1957) theory of child development, there is no place for the modularity of the human mind.
Nor is there any reason to assume the existence of an innate acquisition faculty or any other specialized
learning device. Like in a few approaches reviewed above (cf. interactionism), the infant is believed to form
his/her own concepts through interaction with the environment (Jordan 2004: 140), which is also the
triggering force behind the growth of linguistic competence. Piaget holds that the acquisition of the mother
tongue unfolds as children build a sense of identity in reference to the environment.
Piaget speaks of four stages of general cognitive development through which people evolve between birth
and late adolescence with processes and patterns changing systematically. The said stages are believed to be
5
invariant, that is impossible to skip or reorder, though normal children may show some variability in terms of
the precise age at which they reach each consecutive stage. The four stages of child development according
to Piaget are as follows:
·ð the SENSORI-MOTOR STAGE  from birth to 2 years. Initially, children s activities are wholly
reflexive, as objects are differentiated on the basis of grasping and sucking. The reflexes, directed
towards the child s own body, evolve into complex cognitive schemata as the child develops.
Language acquisition is underway, while mental structures, which at first are largely concerned with
the cognition of concrete objects, are gradually replaced by the perception of the world in terms of
mental images towards the end of the stage.
·ð the PRE-OPERATIONAL STAGE  from 2 to 7 years. This stage is marked by the development of
representational skills in the area of mental imaginary, i.e. the ability to use words as symbols to
represent objects in the world. Thinking remains egocentric, and so children tend to perceive the
world from their own perspective, and they are still not fully aware that there may be alternative
points of view. Typically, by the age of 4 the essentials of the process of language acquisition are
already mastered.
·ð the CONCRETE OPERATIONAL STAGE  from 7 to 11 years. Thinking is decentred and problem
solving is less confined by egocentrism, though abstract thinking is still not possible. Children s
reasoning process becomes logical as they develop the ability to apply logical thought to concrete
problems. They also learn the mastery of classes, relations, and numbers. No longer being fixed to
one dimension or one point of perception, they manipulate the concepts of colour, space and shape.
Furthermore, they are able to exchange thoughts between themselves and through social interactions
they learn that their own concepts can be either denied or verified.
·ð the FORMAL OPERATIONAL STAGE  from 11 years onwards. Older children develop the ability to
deal with hypothetical situations and to monitor their own thinking. At this stage cognitive structures
attain the peak of development, thanks to which children are able to apply logic to problems.
Thinking and cognition become more abstract and children are able to form hypotheses and follow
an argument. Past this stage, there seem to be no further structural improvements in the cognitive
apparatus.
Recapitulating, according to Piaget, human cognition progresses rapidly until the age of 16 and then
slows down when an individual reaches maturity. The most important implication of the cognitivist approach
for our considerations is that the crucial stage in human development, i.e. the transition between concrete and
formal operations, coincides with puberty, when a child begins to think in logical, abstract and symbolic
terms. This claim appears to be another strong cognitive argument behind a critical period for language
acquisition. Besides, as the biological timetable in the cognitivist approach reaches beyond infancy into later
childhood and early adolescence, the implications are also relevant for L2 acquisition.
References and suggestions for further study
Byram, M. 2000. Routledge Encyclopedia of Language Teaching and Learning. London: Taylor & Francis Routledge.
Chomsky, N. 1959.  Review of B. F. Skinner's Verbal Behavior . Language 25: 26-58.
Chomsky, N. 1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.
Clark, Eve. V. 2009. First Language Acquisition. Second Edition. Cambridge: Cambridge University Press.
Field, J. 2004. Psycholinguistics. The Key Concepts. London: Routledge.
Fernández, E. M. & Smith Cairns, H. 2011. Fundamentals of Psycholinguistics. Chichester: Wiley-Blackwell.
Goodluck, H. 1991. Language Acquisition. A Linguistic Introduction. Oxford: Blackwell.
Goodluck, H. 2011.  First language acquisition . WIREs Cogn Sci 2: 47-54.
Jordan, G. 2004. Theory Construction in Second Language Acquisition. Amsterdam: John Benjamins.
Lenneberg, E. 1967. Biological Foundations of Language. New York: Wiley.
Peccei, J. S. 2006. Child Language. A Resource Book for Students. London: Routledge.
Piaget, J. 1957. Logic and Psychology. New York: Basic Books.
Pinker, S. 1994. The Language Instinct. New York: W. Morrow & Co.
Skinner, B. F. 1957. Verbal Behavior. Appleton-Century-Crofts: New York.
Snow, C. E. 1977.  The development of conversation between mothers and babies . Journal of Child Language 4: 1-22.
Valian, V. 2009.  Innateness and learnability . In Bavin, E. L. (Ed.), The Cambridge Handbook of Child Language. Cambridge:
Cambridge University Press. 15-34.
6