UNIT 2: Second/Foreign Language Acquisition Theories.

Different
Approaches. The Concept of Language. Error Analysis.

1. Foreign Language Teaching and Learning.

The study of the subject of Foreign Language Teaching (FLT) and Foreign Language
Learning (FLL) has developed increasingly. FLT was at one time thought to be
exclusively a matter of teaching techniques; it was felt that, if teaching was above a
certain minimum level of efficiency, learning would automatically follow. Teaching
was the active skill; learning, the passive one. Today, the active role of the learner is an
established principle. It is recognized that there are important individual differences
among learners, especially in personality and motivation, which can directly influence
the teaching outcome. In this view, people are seen to be largely responsible for their
own progress. Research is therefore now directed not only at the way teachers teach, but
also at the way learners learn.

The term “acquisition” is sometimes used to replace “learning” in this context; when the
emphasis is on the natural, unconscious way in which a learner can assimilate a foreign
language (as in bilingual contexts, or when using one of the “natural” approaches to
FLT). In several approaches, however, “acquisition” and “learning” are carefully
distinguished: the former is restricted to what takes pace in natural learning situations;
the latter to what takes place in classrooms when following a structured course with a
teacher.

2. Second Language Versus Foreign Language.

Several terminological distinctions are drawn within this field. A person's “mother
tongue” or “first language” (L1) is distinguished from any further languages that may be
acquired. The term “foreign language” is popularly used to refer to any language that is
not a native language in a country; and “second language” is also commonly used in this
way. But many linguists distinguish between “foreign” and “second” language use,
recognizing major differences in the learning aims, teaching methods, and achievement
levels involved. A Foreign Language (FL), in this more restricted sense, is a non-native
language taught in school that has no status as a routine medium of communication in
that country. A Second Language (SL) is a non-native language that is widely used for
purposes of communication, usually as a medium of education, government or business.
English, for example, has foreign language status in Japan but second language status in
Nigeria. The latter term is also used with reference to immigrants and indigenous groups
whose L1 is a minority language. In the USA, for example, English is a second
language for millions of immigrants from a wide range of language backgrounds as well
as for speakers of American Indian languages. In fact, the dividing line between ESL
and EFL fluctuates, and strict use of one term or the other may at times confuse social
and educational issues (Phillipson 1991).
 3. Foreign Language Acquisition.

We are far from completely understanding the language acquisition process. We are just
beginning to grapple with those aspects of the human neurological and biological make-
up which explain the child's ability to acquire language. Certainly it is clear that the
child is equipped from birth with the necessary neural prerequisites for language and
language use.

Our knowledge of the nature of human language tells us something about what the child
does and does not when acquiring a language:

- Children do not learn a language by storing all the words and all the sentences in
some giant mental dictionary. The list of words is finite, but no dictionary can
hold all the sentences, which are infinite in number.

- Children learn to understand sentences they have never heard before and to
construct sentences, most of which they have never produced before.

- Children must therefore learn ‘rules’ which permit them to use language
creatively. No one teaches them these rules. Their parents are no more aware of
the phonological, syntactic, morphological and semantic rules than the children
are. Children, then, seem to act like very efficient linguists equipped with a
perfect theory of language, who use this theory to build up the grammar and the
language they hear.

In addition to acquiring the complex rules of the grammar, that is, linguistic
competence, children must also learn the complex rules of the appropriate social use of
language, what certain scholars have called communicative competence. These include,
for example, the greetings which are to be used, the ‘taboo’ words, the polite forms of
address, the various styles which are appropriate to different situations, and so forth.

However, as a rule, when a child learns a first language, we may say that the child
learns a language under natural conditions. This learning situation differs greatly from
artificial ones, the most common one used in Foreign Language Acquisition being the
classroom.

We can distinguish four possible situations for language learning: on the one hand, first
language acquisition can occur either in the natural setting of the home and the
neighborhood, or rather, in the planned setting of a school. On the other hand, foreign
language acquisition can take place either naturally by immersion in the target language
community or in at school, where the situation is planned.

Steinberg claims that the planned condition of the classroom in which a foreign
language is learned may be said to include the following major differences with
reference to the natural condition:
 - Psychosocial demands of classroom (adjustment to group processes, classroom
discipline, and procedures; limited amount of attention).

- Pre-selected language data (teacher introduces pre-selected language:

spontaneity is limited, always following a planned curriculum).

- Grammatical rules presented (inductively or deductively; learners are to

understand, learn and apply abstract rules).

- Unreal limited situations (limited in scope and variety; simulation is implied).

- Educational aids and assignments (books, tapes, etc. may be used; working
assignments may be given).

Given such differences we may expect the development of Foreign Language

Acquisition to be different from that of First Language Acquisition.

Consequently, research on FLA focuses on the learner, and its goal is not to control or
predict the acquisition of forms of the language, but to shed light on those classroom
conditions that might further or hinder the learner’s appropriation of a foreign language.

Learning, for FLA, is not only a mental process, but also a social construct constrained
by the conventions of schooling. In classrooms, language learning is a socially mediated
process that originates and is ultimately fulfilled via the support system of the learning
environment or Language Acquisition Support System (Bruner 1985).

FLA research needs to consider the classroom as a place where people come together to
learn, where students and teachers work together and mutually construct a pattern of life
in their classrooms which has its own meaning. The classroom then is an active and
dynamic communicative environment in which both social and academic goals are
pursued (Green 1983). FLA in a classroom setting is, therefore, subject to the influence
of a multiplicity of variables that affect the learner’s acquisition of a foreign language
and the rate of this acquisition. What is needed is a way of conceptualizing the complex
social environment of the classroom in order to begin to investigate more seriously the
processes of teaching and learning other languages in a formal classroom setting.

4. Second Language Acquisition (SLA) Theories.

There has been no shortage of theorizing about Second Language Acquisition. This
abundance of approaches, theories, models and principles has even seemed superfluous
to many. Therefore, we will concentrate on four major theories: Behaviorism,
Cognitivism, the Monitor Model and Constructivism.

The first two theories are interesting from a historical point of view. They are primarily
First Language Acquisition models (the imitation and innateness approaches,
respectively), but they also helped the development of Second Language Acquisition
practice. The Monitor Model by Krashen has spawned some of the most interesting
discussions in the field for the past fifteen years, such as the difference between
acquisition and learning. Finally, we will discuss the constructivist theory, as this is the
psychological model which underlies our present Educational Reform.

4.1. Behaviorism and Second Language Acquisition.

A great deal of language learning and teaching in the 1950s and 1960s was influenced
by the tenets of Behaviorism. In this view, FLL is seen as a process of imitation and
reinforcement: learners attempt to copy what they hear, and by regular practice they
establish a set of acceptable habits in the new language.

Skinner stated that our performance as language users is the result of conditioning:
positive reinforcement of correct utterances and negative reinforcement of wrong
answers are the cause of our linguistic knowledge.

In the behaviorist model verbal behavior is conditioned or shaped until it coincides with
correct linguistic models. Properties of the L1 are thought to exercise an influence on
the course of L2 learning: learners “transfer” sounds, structures, and usages from one
language to the other. A widely used typology distinguishes two kinds of transfer:

1. Similarities between the two languages cause “positive transfer”: it proves

acceptable the use of L1 habits in the L2 setting (e.g. the assumption that the
subject goes before the verb satisfactorily transfers from English to French).

2. Differences cause “negative transfer”, generally known as “interference”: the

L1 habits cause errors in the L2 (e.g. the same assumption about subject-verb
order does not satisfactorily transfer into Welsh). Typical interference errors
include: I wait here since 3 hours (from French) and How long must my hand
in plaster stay? (from German). Problems of negative transfer are thought to
provide a major source of FLL difficulty. The main aim of behaviourist
teaching is thus to form new, correct linguistic habits through intensive
practice, eliminating interference errors in the process.

There are several problems presented by this account of FLL. Imitation alone does not
provide means of identifying the task facing learners, who are continually confronted
with the need to create and recognize novel utterances that go beyond the limitations of
the model sentences they may have practiced. Nor does imitation suffice as an
explanation of the way learners behave: not many of the errors that are theoretically
predicted by the differences between L1 and L2 in fact occur in the language of the
learners; and conversely, other errors are found that seem unrelated to the L1. In its
application to the field of SLA behaviourist theory led to the development of the
Contrastive Analysis Hypothesis by the American linguist Robert Lado (1957). This
hypothesis held that the prediction and repetitive practice of those language items which
were different from the first language and therefore likely to cause difficulty to the
learner would overcome interference and establish the habits of the new language
system.
When this theory adapted for the process of learning a foreign language the result was
the Audio-Lingual Method. This method consisted of unending series of drills, followed
by positive or negative reinforcement. It incorporated structural linguistic theory (use of
drills), and the language habit was formed by the constant repetition of correct
utterances which were reinforced by the teacher. Mistakes were immediately criticized
and error played no part in language learning. This method reached its height of
influence during the 1950s and 1960s, and it is still used in many parts of the world.

4.2. Cognitivism.

Cognitivism stems from Chomsky's reaction to Skinner’s book. Chomsky maintained

that language is not a form of behavior. Language is infinite. A person who knows a
language knows its grammar, which is a finite set of rules. Chomsky called the
grammatical knowledge competence, and the realization of these rules performance.

The Cognitive Code Approach theorists, e.g. Ausubel, were advocates of generative
grammar. It was based on the belief that language learning is a process which involves
active mental processes and not simply the forming of habits. It gave importance to the
learner’s active part in the process of using and learning language. Grammar was learnt
by means of both inductive and deductive processes. Most of these ideas have been
taken by the constructivist theory of learning.

Language is no longer seen as a unique and separate form of knowledge but as a

complex cognitive skill that can be described within the context of how people acquire
and store knowledge in general.

The main alternative to the behaviorist approach sees as central the role of cognitive
factors in language learning. In this view, learners are credited with using their
cognitive abilities in a creative way to work out hypotheses about the structure of the
FL. They construct rules, try them out, and alter them if they prove to be inadequate.

The learning of a language, in this account, proceeds in a series of transitional stages, as

learners acquire more knowledge of the L2. At each stage, they are in control of a
language system that is equivalent to neither the L1 nor the L2, an interlanguage.

4.3. The Monitor Model.

One of the most influential and indeed controversial theories of SLA based on the
Creative Construction model is that advanced by Stephen Krashen (1982). Krashen’s
view of SLA centers on five hypotheses which together form his Monitor Model.

4.3.1. The Acquisition-Learning Hypothesis.

Krashen’s identification of two different types of linguistic knowledge in SLA is the

central tenet of the Monitor Model. Acquisition, according to Krashen, is different from
learning in that it occurs subconsciously when the learner engages in communicative
interaction in the L2 where the focus is on meaning and there is no attention to form.
Learning occurs as a result of the conscious study of the L2 system focusing on formal
grammatical rules and the correction of errors.

According to Krashen, it is acquisition and not learning which enables the learner to
achieve competence in the second language. It is this distinction that leads him to
emphasize the need for communicative rather than formal instruction in the classroom
settings, and it is on this basis that, as he says, the language teacher should provide the
learner with situations in which he can acquire linguistic structures while taking part in
natural communication.

4.3.2. The Monitor Hypothesis.

The Monitor Hypothesis refers to the role of formal learning in this theory of SLA. The
monitor is the device that learners use to edit and reform their language production.
When a learner communicates in L2, she/he is using the knowledge of her/his acquired
system. Learnt knowledge will only be called upon if the learner attempts to correct or
modify in any way his/her linguistic production.

4.3.3. The Natural Order Hypothesis.

Formal grammatical structures are acquired by the learner in a predictable order, some
early and others later, irrespective of age, first language background and the apparent
grade of difficulty that the structures present.

4.3.4. The Input Hypothesis.

Acquisition takes place as a result of the learner having understood input that is just
beyond his or her current level of linguistic competence. Krashen refers to this as i + 1.

4.3.5. The Affective Filter Hypothesis.

Krashen claims that while comprehensible input is a necessary condition for language
acquisition to take place, it alone is not sufficient. The learner must be prepared to
absorb the input he receives. This will not occur if the learner is anxious or lacking in
motivation or self-confidence. This mental block is what Krashen calls the affective
filter.

All in all, we dare say the Monitor Model lacks empirical evidence and the ideas it
advocates for are rather more descriptive than scientific.

4.4. Constructivism.

Constructivism takes elements of both Cognitivism and the Monitor Model. It tries to
foster both acquisition and learning because our pupils can use language to create
learning. Effective learning must have several characteristics: relevant, meaningful,
comprehensible input.

Relevant learning means that the new input is based on our pupils’ needs. It is
meaningful because our pupils can put it in relation with previous knowledge and by
means of this process they understand the new input. This new input is comprehensible
when the gap between what they already know and the new input can be bridged by
means of both social and cognitive interaction. This process transforms input into
intake.

Constructivism is a mentalistic theory and thus considers as a main factor in learning the
hypothesizing and abstract analyzing of pupils. Errors are the result of the pupils’
efforts to adjust to the target system. Error is an integral part of the learning process.

A very important additional feature of the constructivist theory is the emphasis it places
on teaching our pupils learning to learn strategies. Learning to learn strategies
primarily concern is to focus our pupils’ attention on how they learn in addition to what
they learn.

4.5. Other Second Language Acquisition Theories.

4.5.1. Creative Construction Theory (Dulay and Burt, 1977).

Second language learners organize the language they hear and actively construct and
test out hypotheses as to how the L2 works. In this sense the learner’s first language,
rather than interfering negatively in this process, takes on a much more positive role,
representing a systematic attempt on the part of the learner to use his previous
knowledge in the learning of a new language. The transfer of rules from the mother
tongue, in the light of creative theory, is not seen as evidence of error and poor habit
formation but as one of the learner’s active strategies for interpreting and constructing
the new language system.

An important feature of the Creative Construction Theory is the suggestion that the
learner’s internal processing mechanisms operate on the second language input received
from the environment and are not dependent on the production of language by the
learner. In other words, the learner can successfully acquire language by understanding
samples of the target language through listening and reading. Comprehension of the
new language is vital for acquisition to take place. The learner’s oral or written
production is seen as a result of the learning process rather than the cause of learning.

4.5.2. The Interactionist View.

The interactionist position on first language acquisition is an attempt to reconcile the

behaviourist emphasis on the linguistic environment and the mentalist concern with a
special innate language faculty.
Language acquisition is a result of the relationship between the child’s internal capacity
for language and the environment in which the capacity is developed. Central to this
theory is the idea that the linguistic input received from the environment is modified and
adapted to the capability of the learner. This promotes comprehension and thus the
acquisition of the L2. Ideas such as foreigner talk and teacher talk –in order to negotiate
meaning– are derived from this assumption.

4.5.3. The Output Hypothesis.

Comprehensible input and the interaction hypotheses are not enough since they do not
attach any importance to a factor we can consider crucial for successful SLA:
comprehensible output.

Merrill Swain (1985) arrived at this conclusion after studies carried out with students in
bilingual immersion programmes in Canada. Her findings revealed that although
students enrolled in these programmes obtained levels of comprehension comparable to
that of native speakers their production skills were significantly inferior, especially
where grammatical competence was concerned.

Swain suggests, then, that comprehensible output is important for several reasons:

1. It provides opportunities to test out hypotheses using the language.

2. The learner must pay attention to the expression needed to convey an

intended meaning and by doing so improves his/her production skills.

5. Age Differences in Second Language Acquisition.

Older acquirers are faster in the early stages of second language acquisition because:

- They are better at obtaining comprehensible input as they have good

conversational management.

- They have superior knowledge of the world, which helps to make input
comprehensible.

- They can participate in conversation earlier, via use of first language syntax.

Younger acquires tend to attain higher levels of proficiency in second languages than
adults in the long run due to a lower affective filter.

6. Implications for Second Language Teaching.

The hypotheses about L2 acquisition predict that any successful L2 teaching programme
must have the following characteristics:

- It must supply input in the L2 that is comprehensible, interesting and relevant to

students. The goal is, thus, to transmit messages, not to practice grammar.
 - It must not force students to speak before they are ready and must be tolerant of
errors in early speech. We improve in grammatical accuracy by obtaining more
input, not by error correction.

- It must put grammar in its proper place. Some adults, and very few children, are
able to use conscious grammar rules to increase the grammatical accuracy of
their output; and even for these people, very strict conditions (time, focus on
form, and knowledge of the rule) need to be fulfilled before the conscious
knowledge of the grammar can be applied, given the monitor hypothesis
presented above.

7. Role of the First Language in Second Language Acquisition.

7.1. The Interlanguage Theory.

This new insight into how learners of an L2 build up a new language clearly showed
that people such as Lado and other behaviorist supporters had been too simplistic in
seeing L2 learning as restricted to a relationship between the L1 and the L2. The view
that began to become increasingly important at this point was that in the process of
acquiring a new language a learner is, in fact, using neither the L1 nor the L2, but
rather a third language system of his or her own.

This new language system that a learner constructs on his/her way to the mastery of the
target language has been referred to in a number of terms. Corder (1967) used the term
transitional competence to describe the system of rules that a learner has developed at a
particular stage, emphasizing its temporary nature as the learner progresses towards
competence in the TL. Nemser (1971) referred to it as approximative system and
described it as having some properties present in neither the L1 nor the L2. He draws
attention to structural aspects of the learner’s language, which gradually approaches
more or less closely, the full target language.

The term which has become the most widely accepted to refer to the independent
language system created by L2 learners is interlanguage (often abbreviated to IL)
introduced by Larry Selinker in 1972. He traced the origin of interlanguage in the
processes through which the mind acquires a second language. According to Selinker
this new system is different from both the L1 and the L2 systems but contains elements
from both. If we consider interlanguage within a continuum between the first language
system (the learner's initial knowledge) and the second language system (the learner’s
target), at any given point the learner is said to speak an ‘interlanguage’.

Selinker's interlanguage presents three principal features:

1. It is permeable because the rules that constitute the learner’s knowledge are
open to evolution at any given time.

2. It is dynamic because it is changing constantly, but rather slowly, thus the

idea of a continuum.
 3. It is systematic, because it is based on coherent rules which learners
construct and select in predictable ways.

Selinker also noted that two learning differs from L1 learning in that many learners
seldom reach the end of the interlanguage continuum. The learner’s developing
interlanguage ‘fossilizes’ at some point, that is, it still contains some rules different
from those of the L2 system and cannot be remedied by further instruction.

7.2. Error Analysis.

The interlanguage theory meant a complete change both for the kind of evidence to be
studied by SLA research and for the relationship established between the L1 and the L2.

Selinker's research methodology, included within a tradition associated with the work of
Corder (1967, 1971, 1981) centered on the description of the learner’s speech. Both
Selinker and Corder concluded that the incorrect sentences produced by L2 learners
were signs of their underlying interlanguage not of their poor control of the L2. This
view led to a new methodological system for studying some aspects of SLA known as
Error Analysis (EA), which approached L2 learning through a detailed analysis of the
learner’s L2 speech.

Error Analysis has two functions:

1. Investigating the language learning process (as a means to describe the

learner’s knowledge of the target language at any particular moment in his
learning career in order to relate this knowledge to the teaching he has been
receiving).

2. Guiding the remedial action we must take to correct any unsatisfactory state
of affairs for learner or teacher. (remedial teaching is necessary when we
detect a mismatch or disparity between the knowledge, skill, or ability of
someone and the demands that are made on him by the situation he finds
himself in).

From this perspective, EA began to develop as an alternative to Contrastive Analysis in

order to explain the majority of errors that had not been accounted for. CA had
concentrated on predicting the errors from L1 interference, whereas EA dealt with
errors by looking at the learners’ performance in relation to the L2.

It is interesting to note that although by the mid 1950’s it had already been
acknowledged within the field of first language acquisition that a child’s speech could
no longer be considered as a faulty version of the adult’s, it was not until the 1970’s that
this notion was also applied to the speech of second language learners. Corder’s views
of learning (1971) –drawn from the hypothesis– testing paradigm of first language
acquisition work when transferred to L2 learning suggest that L2 learners also progress
by actively constructing a series of hypotheses about the language they are learning
from the data they receive. This suggests that learner’s errors do not have to be
considered as signs of failure.

On the contrary, as Corder puts it, errors are a way the learner has of testing his
hypotheses about the nature of the language he is learning (Corder 1969). They are vital,
therefore to the L2 learning process since they constitute the clearest evidence for the
learner’s developing system and can also provide information about how learners
process the language data.

The information that EA provides about the learner’s interlanguage is of two types. The
first concerns the linguistic type of errors that L2 learners produce, linked to the
sequence of development the learner experiences; and the second type concerns the
psycholinguistic type of errors made by learners and is related to the kinds of strategies
they apply within their interlanguage. However, as Ellis (1985) writes, the first one is
not very helpful when it comes to understanding the learner's developmental system.
Error Analysis must necessarily present a very incomplete picture of SLA because it
focuses only on part of the language L2 learners produce. About the psycholinguistic
types of errors he claims that EA is on stronger ground, even though it still seems rather
difficult to code errors in terms of categories such as developmental or interference.

7.2.1. Definition of ‘Error’.

There does not seem to be total agreement about what exactly an error is. The terms
given to the different types of errors vary from one study to the other. One of the best
known definitions of errors is the one given by Corder (1967). In order to emphasize the
systematic nature of interlanguage rules Corder distinguishes between ‘lapses’, ‘errors’
and ‘mistakes’. He views lapses as errors due to lack of concentration or fatigue as
when the learners make slips, false starts or confusion of structures. As Corder points
out, lapses do not only occur among L2 learners but also among L1 speakers. The term
‘error’ is used to refer to regular patterns in the learner’s speech which consistently
differ from the target language model, a sign that a given rule has been completely
misunderstood. On the other hand, with the term ‘mistake’ he refers to memory lapses,
use of the wrong style, etc, but as in the case of lapses, mistakes are also produced by
native speakers and can often be corrected once the learner is aware of what he has
done. But the errors they make, from Corder’s perspective, are part of their current
system of interlanguage rules and therefore, not recognizable to the learners as such.

Corder’s distinction has been of great importance in SLA. The term mistake has been
attached to errors of performance, while the term error has been reserved to errors of
competence. Consequently, to evaluate the point of development which the learner’s
interlanguage has reached at any given time means to be able to distinguish between
errors that result from competence and mistakes that come from performance.
Nonetheless, nowadays the term ‘error’ has been widely accepted, whether it derives
from a lack of competence or from defective performance.
 7.2.2. Error Classification.

As already said, interlanguage theory attempted to find explanations for the errors in
language learners’ speech: errors must not be immediately punished and remedied; they
are signals that tell us that the language learner is making hypotheses about the target
language and trying to progress in his/her interlanguage.

As Littlewood suggests, when learners are actively constructing a system for the second
language, not all their incorrect notions are the result of transferring rules from the first
language.

If a structure in L1 differs from that of L2, negative transfer or interference will arise
(“This book likes me”). On the contrary, if a structure in both languages is the same,
there will be positive transfer or zero interference, and there will be no errors in L2
performance (e.g. Spanish and English plural marker “- s”).

Besides errors due to rules of transference from the mother tongue, sometimes called
“interlingual” errors, learners also produce many other errors which show that they are
processing the second language in their own terms. That is, many of their incorrect
notions, in fact, can be explained by direct reference to the target language. Errors of
this second type, which can be considered within the psycholinguistic type of errors
above mentioned are often called “intralingual” and are similar to those produced by the
child in the mother tongue. They suggest that the second language learner is employing
similar communication strategies:

- Overgeneralization of L2 rules: for example, when the learner tries to use L2

rules in ways which are not permitted, as in the example “What do he want?”.

- Transfer of training: as when a rule enters the learner’s system as a result of the
overuse of a word or expression by the teacher. For example, when insistence by
the teacher on the use of “he” discourages learners from using “she” in their own
speech.

- Simplification: When a learner uses all verbs in the present continuous tense; for
example, “I’m hearing him”.