APPLIED LINGUISTICS

UNIT 1

The term Applied Linguistics seems to be of quite recent coinage. It emerged in the late 1940’s and 1950’s both in
America and Britain to refer to the incipient field of the study of the teaching and learning of foreign languages. The
1980’s saw a widening of the term to include topics that are language-related beyond linguistics itself. At this point it
is necessary to distinguish linguistics from applied linguistics. Davies and Elder (2004: 11) point to the fact that while
“linguistics is primarily concerned with language itself and with language problems in so far as they provide evidence
for better language description or for a linguistic theory, applied linguistics is interested in language problems for
what they reveal about the role of language in people’s daily lives and whether intervention is either possible or
desirable.” (2004: 12) Applied linguistics has evolved as independent from linguistics as its only source, and is often
problem-oriented towards the analysis and resolution of language-related social problems that people find in the
real world (teachers, academics, lawyers, policy developers, business clients, to mention only a few) across a wide
range of settings. Also, in a preface to his An Introduction to Applied Linguistics: from Practice to Theory(1999: vii),
Davies explains his approach to applied linguistics as “a theorizing approach to practical experience in the language
professions, notably, but not exclusively, those concerned with language learning and teaching. It is concerned with
the problems, the processes, the mechanisms and the purpose of language in use.”

1.2 ATTEMPTING A DEFINITION

Applied linguistics may be defined, then, in a simplified sense, as a science that mediates between theory and
practice, with a view to the resolution of language-related problems in various contexts. According to Widdowson
(2005: 12), in attempting a definition, two things are taken into account: first, that it deals with problems in the real
world in which language is involved, and second, that, and as a necessary consequence of the first, it is
interdisciplinary in nature.

As to the variety of issues related to the field, Kaplan (2002) lists several foci of study for the discipline:

1- Language teaching and learning, recently concentrating on topics like information processing, patterns of
interaction and language awareness. Also, teacher development issues are being addressed, such as how to engage
teachers in action research and give them training in evaluating their practices.
2- The role of critical studies, such as critical discourse analysis, critical pedagogy and ethics.
3- Language use in academic and professional settings, where language can act as a positive mechanism or as an
obstacle in achieving results.
4- Descriptive analyses of language in real settings, and their possible applications in corpus linguistics, genre analysis
and register variation.
5- Multilingualism in schools, communities, work settings or language policies.
6- Language testing and assessment issues.
7- The role of applied linguistics as mediator between research and practice.

It becomes apparent from the list above that the scope of applied linguistics is difficult to determine. Also, as the
resolution of real world problems is a complex task, applied linguistics can be effective if it remains interdisciplinary.
However, as in most sciences, there is disagreement. Widdowson (2005: 25) is clear about his position with regards
to the danger posited by interdisciplinarity, in that there may be a separation between the domain experience of the
real world and the theoretical discipline knowledge: “interdisciplinarity…cannot actually provide such a service since
it operates on a level of abstraction remote from the actualities as experienced by the folk. If we are to engage with
real-world issues, we need to develop a methodological approach that mediates between these two orders of reality
of discipline and domain” in which none is unilaterally imposed.

1.3 DICHOTOMIES RELATED TO THE DISCIPLINE

In trying to understand the environment of applied linguistics, it is necessary to consider its source and target, the
former referring to what applied linguistics draws on, the latter to what AL is being applied to. According to Davies
(1999: vii), there is a two-way relationship: “Like any other discipline, applied linguistics draws on theories from
related disciplines with which it explores the professional experience of its practitioners and which in turn are
themselves illuminated by that experience.” He considers both source and target as dependent on the view one
adheres to, either a restricted view in which applied linguistics only applies theoretical linguistics and therefore there
is little connection with real life problems, or a wider one in which the discipline would have to concern itself with
whatever is related to language. Neither position seems to be realistic: “Linguistics…must play an important role in
applied linguistics but by no means the only role. Applied linguistics must also draw on psychology, sociology,
education, measurement theory and so on.” (Davies 1999:3) Regarding the target and scope of the science, there are
two poles and in-between positions, as to whether applied linguistics should focus on language teaching with a
broad meaning, i.e. to inform fields like speech therapy, translation and language planning, or, on the other hand, it
should have a much wider application. This position, however, should present limitations so as to avoid creating a
science of everything (Davies 1999). Yet another distinction must be made, that of the Applied-Linguistics versus the
Linguistics-Applied approach to language problems. Applied linguistics “looks outward, beyond language in an
attempt to explain, perhaps even ameliorate social problems ‘in the real world’” (Davies 2004:11), and uses
information sources from fields that stand outside linguistics (Ellis 1997). On the other hand, linguistics applied
makes use of language data to extend knowledge about language concerned with theories about language.
However, this distinction is not always clear-cut, and it seems to be one not of areas of interest but of orientation of
research.In sum, applied linguistics is too complex a field to offer easy definitions. Its complexity in terms of
components, orientations and outcomes is also a source of various theories and hypotheses that cause considerable
debate among its practitioners. However, the dominant concern for applied linguistics seems to be that of language
teaching and learning, in that more applied linguists devote their attention to this field than to any other (Davies
1999: 63). As Littlewood (2004: 502) states “the study of second language learning is an immensely rich and varied
enterprise. Most participants…still see its ultimate justification in terms of the desire to improve learning and
teaching. In this respect the study of second language learning is an important branch of the overall field of applied
linguistics”.

UNIT 2

Apart from SLA, other linguistic disciplines offer valuable contributions to language teaching, though at a macro-
level. Few research orientations in the field of SLA seem to isolate language from its use in its study. Most recent
work “entails a certain notion of language: language as social, as meaningful, and as always embedded in a social
context and history. Language is not an isolated phenomenon; language is deeply social, intertwined with social
processes and interaction” (Wodak, in Wodak 2000). It is clear, then, that language teaching cannot be restricted to
learning to use the target language excluded from its social context. Thus the crucial role of discourse analysis, and
its related discipline, critical discourse analysis, hence CDA. This idea is in keeping with Widdowson’s (2004: viii):
“Whereas I had thought of language teaching as the main area of practical concern which discourse analysis could be
relevant to, CDA had a much more ambitious and much more significant agenda. Its concern was to educate people
more broadly in the abuse of power by linguistic means, to reveal how language is used for deception and distortion
and the fostering of prejudice.” However, not all scholars go along with such a negative view. Wodak (2000: 2)
explains that critical “means distinguishing complexity and denying easy, dichotomous explanations. It means
making contradictions transparent.” Language teaching should be done within the frame of language education, so
as to empower learners to become aware of the uses allowed by language, both native and second languages alike.
In this respect, language teaching becomes an enabling tool for life, thus the relevance of CDA in this module.

2.1 CDA: A DEFINITION

CDA’s roots lie in classical rhetoric, text linguistics, sociolinguistics, applied linguistics and pragmatics (Wodak 2002),
fields that recognise both the value of interaction and the social creation of texts. Wodak (2000: 186) provides a
clear definition: “CDA is not a homogeneous theory with a set of clear and defined tools; rather, it is a research
program with many facets and numerous different theoretical and methodological approaches”. CDA may be
considered one of the strands of applied linguistics in that it does not restrict itself to descriptions of problems but
may suggest solutions. CDA is mainly associated with analysis of texts produced by the media and politics, but it also
has a strong link with organisational discourse, gender, racism, political correctness and language awareness, to
mention just a few. “I assume that critical research … should be concerned with social issues that are at the core of
the imminent changes taking place in current societies. Through CDA, we are able to make social interaction
transparent and understandable to [the citizens] who have little or no access to the elites involved in decision
making.” (Wodak 2000: 191) Van Dijk agrees with the position above: CDA is not so much a direction, school, or
specialization next to the many other ‘approaches’ in discourse studies. Rather, it aims to offer a different ‘mode’ or
‘perspective’ of theorizing, analysis, and application throughout the whole field. We may find a more or less critical
perspective in such diverse areas as pragmatics, conversation analysis, narrative analysis, rhetoric, stylistics,
sociolinguistics, ethnography, or media analysis, among others. (van Dijk 2007: 352) Thus, according to the aims of
research, the methodologies used by the different scholars who apply CDA differ greatly.

As regards the use of the term critical, Rogers (2004) offers three explanations: first, CDA is associated with the study
of power relations, in a dialectical manner between discourse and its social context. Second, it is concerned with the
description, interpretation and explanation of the relationship between language form and function. Finally, CDA
explicitly addresses social problems and analyses how discourse operates in the construction of those problems. This
action-oriented stance is usually referred to as critical language awareness. Within the frame of CDA, discourse
assumes that language is always social, and its analysis happens above the level of sentence or clause. In Rogers’
words (2004: 356), “discourse both reflects and constructs the social world and is referred to as constitutive,
dialectic, and dialogic. Discourse is never just a product, but a set of consumptive, productive, distributive, and
reproductive processes that is in relation to the social world.” Finally, as regards analysis, Rogers (2004) refers to the
fact that, even though there are central principles that unite all analysis carried out in the field of CDA, there is also
much dissent, especially as far as methodology is concerned. For instance, some analyses are more
languageoriented, while others focus more on the context in which discourses arise. In sum, it becomes clear that
CDA implies an interdisciplinary approach, as the social problems it deals with are complex issues that cannot be
analysed within a single field. Furthermore, critical discourse analysis is inevitably rooted in a socio-political situation
researchers can hardly isolate from, which implies close involvement on the part of researchers with dominated
groups.

2.2 MAIN PRINCIPLES OF CDA

In spite of the variety of types of CDA, there is a common core to all. The overall conceptual and theoretical
frameworks focus on the way discourse reproduces social dominance; that is, “CDA focuses on the ways discourse
structures enact, confirm, legitimate, reproduce, or challenge relations of power and dominance in society.” (van
Dijk 2007: 353) 7 Fairclough and Wodak (in van Dijk 2007: 353) summarize the main tenets of CDA as follows: 1. CDA
addresses social problems 2. Power relations are discursive 3. Discourse constitutes society and culture 4. Discourse
does ideological work 5. Discourse is historical 6. The link between text and society is mediated 7. Discourse analysis
is interpretative and explanatory 8. Discourse is a form of social action. Item four above refers to ideology. This is a
term that is recurrent in CDA, and is closely linked with the notion of power. Fairclough (1989: 2) defines ideology as
“‘common-sense’ assumptions which are implicit in the conventions according to which people interact linguistically,
and of which people are generally not consciously aware.” For example, Fairclough describes a typical consultation
between doctors and patients, in which the doctor’s authority and hierarchy are treated as natural, and therefore it
is natural for him/her to make decisions and control the course of the consultation, while the patient is expected to
comply and cooperate. Ideologies are closely linked to power, as the nature of the ideological assumptions present
in particular conventions, and therefore the nature of those conventions, depend on the power relations that
underlie the conventions. Ideologies are a means of legitimizing existing social relations and differences of power,
simply through the recurrence of ordinary, familiar ways of behaving which take these relations and power
differences for granted. Ideologies are closely linked to language, because using language is the commonest form of
social behaviour, and the form of social behaviour where we rely most on ‘common-sense’ assumptions. (Fairclough
1989: 2) 8 Practices which seem to be universal and based on “common sense” have become naturalised, and can be
traced to have an origin in dominant groups. As regards item six in the list above, it is necessary to clarify the
distinction between the terms text and discourse, both pervasive in CDA. Although they are sometimes used
interchangeably, some authors make a distinction. Fairclough (1989: 24) views text as a product rather than a
process, and discourse as the “whole process of social interaction of which a text is just a part.” This definition is in
keeping with the one proposed by Systemic Functional Linguistics (on which much of discourse analysis is based),
which claims that text is “any passage, spoken or written, of whatever length, that does form a ‘unified whole’”
(Halliday and Hasan in Eggins 2004: 23-24) A text is unified thanks to the interaction of cohesion and coherence, the
result of which is an instance of linguistic resources in a meaningful way within a situational and cultural context.
(Eggins 2004: 24) On the other hand, SFL describes discourse technically as the level of meaning above the lexico-
grammar, that is, the level concerned with relations of meaning across a text. The following section will focus on
some general basic concepts that relate discourse, cognition and society (van Dijk 2007).

2.2.1 Macro vs. micro

“Language use, discourse, verbal interaction, and communication belong to the micro level of the social order.
Power, dominance, and inequality between social groups are typically terms that belong to a macro level of
analysis.” In everyday interaction, these two levels co-exist. Van Dijk (2007: 354) offers the following list of ways in
which the gap between the macro and the micro levels can be bridged: 1 Members–groups: Language users engage
in discourse as members of (several) social groups, organizations, or institutions; and conversely, groups thus may
act "by" their members. 2 Actions–process: Social acts of individual actors are thus constituent parts of group actions
and social processes, such as legislation, newsmaking, or the reproduction 9 of racism. 3 Context–social structure:
Situations of discursive interaction are similarly part or constitutive of social structure; for example, a press
conference may be a typical practice of organizations and media institutions. That is, "local" and more "global"
contexts are closely related, and both exercise constraints on discourse. 4 Personal and social cognition: Language
users as social actors have both personal and social cognition: personal memories, knowledge and opinions, as well
as those shared with members of the group or culture as a whole. Both types of cognition influence interaction and
discourse of individual members, whereas shared "social representations" govern the collective actions of a group.

2.2.2 Power as control

A central notion in CDA is that of power, specifically the social power of groups and institutions. Van Dijk (2007: 354-
355) defines power in terms of control: groups have (more or less) power if they are able to (more or less) control
the acts and minds of (members of) other groups. This ability presupposes a power base of privileged access to
scarce social resources, such as force, money, status, fame, knowledge, information, ‘culture’, or indeed various
forms of public discourse and communication. Power is seldom absolute, as groups may exercise more or less
control over other groups, sometimes only in specific situations. Dominated groups may also “resist, condone,
comply with or legitimate such power, even find it ‘natural’. The power of dominant groups may be present in laws,
rules, habits, or general consensus, in everyday actions that are taken for granted. In this respect, the exercise of
power is not always obviously coercive, which may make it more dangerous to the dominated groups. 10 Persuasion
and manipulation are two key concepts in that action, which is controlled by our minds, can be modified by
influencing people’s minds. This is directly related to access to specific forms of discourse (politics, the media,
science, etc.), which is, then, a resource of power. In short, groups that control influential discourse have more
chances to control the minds and actions of others. Control of public discourse Ordinary people can have active
control over discourse, as in everyday talk with family and friends. They can also be passive recipients of talk or text
from their bosses, teachers, etc. On the other hand, more powerful social groups and institutions, and their leaders,
have access to and control over public discourse and discourse types, such as doctors, teachers, lawyers, journalists,
etc. Regarding communicative events, context and the structure of text and talk should be taken into account.
Context is defined as “the mentally represented structure of those properties of the social situation that are relevant
for the production or comprehension of discourse” (van Dijk 2007: 356). It involves setting (time and place), ongoing
actions (discourses and genres), participants and their mental representations (knowledge, opinions, goals, attitudes
and ideologies). “Controlling context involves control over one or more of these categories, e.g. determining the
definition of the communicative situation, deciding on time and place of the communicative event, or on which
participants may or must be present, and in which roles, or what knowledge or opinions they should (not) have, and
which social actions may or must be accomplished by discourse.” (2007: 356) In connection with the structure of text
and talk, powerful groups, or their members, can decide on the discourse genres or speech acts of an occasion. For
example, a teacher can ask a student to read a story instead of re-telling it. In more critical situations, control over
the structure of text and talk is evidenced when some people are excluded from producing discourse. Also relevant
is who controls the topics and topic change, as when, for example, teachers decide what to talk about or when to
stop talking about a certain topic. Virtually all structures of context, text and talk may be more or less controlled by
powerful speakers, who may abuse their power at the expense of the less powerful participants. Mind control
Controlling people’s minds is another way of reproducing dominance. This can be achieved in a number of ways.
First, people tend to accept beliefs, knowledge and opinions from what they consider trustworthy sources. Second,
in some situations, participants are obliged to be recipients of discourse, as in a classroom situation. Third,
sometimes, there are no public discourses or media that can provide alternative options. Fourth, some recipients
may not have the necessary knowledge to challenge the discourses they are exposed to. Discourse structure can also
exert mind control. At a global level, topics may influence what people see as the most important information. Also,
argumentation can be persuasive in that it usually involves social opinions that are hidden and taken for granted. At
a local level, presuppositions, necessary for communication, may make discourse harder to challenge, as they
communicate beliefs implicitly. This means that dominant groups may control public discourse and therefore the
minds of the public. Public discourse is, then “involved in dominance (power abuse) and in the production and
reproduction of social inequality.” (van Dijk 2007: 358) However, control over discourse has its limits, as the
complexity of beliefs makes it difficult to predict which features of a specific text will have which effects on the
minds of recipients. (van Dijk 2007)

2.3 AREAS OF CDA

The following are some of the areas that have been researched under the label of CDA. This section will also be
based on van Dijk’s analysis (2007). For each area there are two chapters attached; one looks at the corresponding
issue from a more global point of view, while the other focuses more on language forms.

2.3.1 Gender: Studies of gender inequality are an exception in that most studies in the field have not been done
within the scope of CDA, although much of it deals with inequality and domination. See the accompanying chapters
for a survey and description of the field’s evolution (Kendall and Tannen 2007), and for the asymmetrical
representations of women and men, and how they can be sexist (Shân Wareing 2004).

2.3.2 The media: The discourse of the media has media given rise to a vast body of research, which has revealed an
often biased presentation of information. Much of the work on this area has focused on systemic functional
grammar, as “events and actions may be described with syntactic variations that are a function of the underlying
involvement of actors… by leaving agency and responsibility implicit” (van Dijk 2007: 359) Thus, linguistic elements
such as transitivity and lexical structure appear as highly powerful tools. See Cotter (2007) and Thornborrow (2004).

2.3.3 Political discourse: In van Dijk’s words (2007: 360), political discourse plays a central role in the “enactment,
reproduction, and legitimization of power and domination”, so work in this area has also been prolific. Political
information is usually obtained through discourse, sometimes through the media. Political action is also carried out
by means of discourse, so the relation between these areas is inextricable. See Jones and Peccei (2004) and Wilson
(2007)

2.3.4 Ethnicity: The study of the influence of discourse in the perpetuation of ethnic inequality traditionally focused
on ethnocentric and racist representations in the mass media, literature and films. These representations reflect the
images of the Other in the discourses of European travellers, explorers and merchants among others, based on an
exotic impression on the one hand, and on opinions that considered the Other an inferior being. This influenced
public opinion, and led to generalised social representation. In van Dijk’s words (2007: 362), “racism (including
antisemitism, xenophobia, and related forms of resentment against "racially" or ethnically defined Others) is a
complex system of social and political inequality that is also reproduced by discourse in general, and by elite
discourses in particular”.

2.4 LANGUAGE AWARENESS

The language classroom is not alien to social relations of power, so it is the teacher’s task to become aware of
his/her language practices and develop students’ awareness of the power of language in the reproduction or
transformation of inequality. In Fairclough’s words (1989: 1), “consciousness is the first step towards emancipation.”
Learners should become conscious of the fact that language is not simply a tool for expressing ideas and carrying out
tasks. It also, and inevitably, expresses, constitutes and reproduces social identities and relations, including those of
power. (1989) Of course, the situation of learners of English as a foreign language (as opposed to second language) is
different from that of immigrants, who are sometimes “disadvantaged sections of the society, whose experiences of
domination and racism are particularly sharp.” (1989: 235) Nevertheless, critical language awareness should be part
of the language curriculum in that it may enable learners to discover how domination works in a modern society,
“through ‘consent’ rather than by ‘coercion’, through ideology, and through language.” (1989: 233). This kind of
knowledge will hopefully be transferred to the analysis of the mother tongue, therefore opening up new ways of
looking at language in general. Fairclough (1992), claims that we are living in a period of intense social change, and
that language is, albeit not evidently, extremely important within those changes.

First, changes are related to the way in which power and social control are exercised, from more explicit to more
implicit ways. This implies that language is becoming more important in sustaining and reproducing power relations.

Second, language practices are changing. For example, in the service sector, there are many people who have to
communicate with clients, so specific communicative skills have become of utmost importance, and employees are
specifically trained to satisfy the communicative demands of their jobs. Also, there is a tendency towards the use of
a conversational style in interactions between, for instance, professionals and clients, and shop assistants and
customers. This suggests that changes in language practices are reflecting changes in social relationships and social
identities, which, in turn, may reflect a democratisation of professional domains. However, a conversational style
may provide “a strategy for exercising power in more subtle and implicit ways.” (Fairclough 1992: 5)

Third, language itself is becoming a target for change. Some institutions are conceptualising language in terms of
techniques or skills which are designed to improve their effectiveness, in various domains, and more or less
independently of context. This leads to the redesigning of language practices and training of personnel. (Fairclough
1992: 5) In sum, since power relations are exercised implicitly in language, and language practices are more
consciously controlled, language education cannot fail to provide learners with critical language awareness.
Fairclough (1989: 242-243) suggests the following cycle for language education in schools (it will be slightly adapted
here to suit the needs of our situation):

1. Reflection on experience: students are asked to reflect on their own discourse and experience, and to share their
reflections with the class. E.g.: learners reflect on the differences between speech and writing.

2. Systematising experience: the teacher shows students how to express the reflections systematically. E.g.: present
an account of the functional differences between speech and writing, the social prestige of various uses of writing,
and the distribution of access to prestigious uses.

3. Explanation: the knowledge acquired through the two previous stages is further reflected upon and analysed by
the class, and social explanations are sought. E.g.: reflect on the social reasons for access to prestigious uses of
writing, the language they are written in, etc.

4. Developing practice: the awareness resulting from the previous stages is used to develop purposeful discourse.
E.g.: a writing activity that breaks the conventions for writing, for instance writing about topics which are not usually
written about, writing in a language not normally used for those purposes, etc. It is important to bear in mind that
critical language awareness is best applied to types of discourse that have real significance in students’ own lives and
experience. Also, people tend to be more open to critical approaches when they are engaged in social struggle.

Finally, for most people, focusing on language may make little sense at the beginning, so its relevance will have to be
demonstrated. (Fairclough 1989: 245) However, time invested in enabling learners to become critically language
aware is never wasted, as it may empower them to become critical thinkers and actors in a society that is constantly
changing. Being aware of the opportunities offered by language, and of the restrictions it may impose, is a necessity,
both for language professionals and for learners. Language teachers, first, second or foreign alike, should be
committed to raising their students’ consciousness. This may help them recognise the power and freedom that lie in
knowledge, and the responsibility they imply.

UNIT 3

Second Language Acquisition is the branch of applied linguistics that looks into language learning once the native
language, or languages, have been acquired. Before going deeper into the subject, some terms need to be clarified.
As in most research in the field, L2 refers to any language a learner learns after having incorporated his/her L1, no
matter how many languages he/she speaks. Also, no distinction is made in this module between an L2 and a FL, the
former acting as a cover term for a language learnt while living in the country where the target language is spoken
and for a language learnt where there is no contact with the target language community. In cases where the
distinction is necessary, it will be made. The concepts of learning and acquisition are also taken as synonymous,
unless a difference is specified. Finally, some words need to be said about the term language. In early SLA work, it
usually meant grammar and vocabulary, while in current work the distinction between knowledge and use is
important. Also, after the introduction of the concept of Communicative Competence, the concept of language
includes not only linguistic knowledge, but also discourse, pragmatic, sociolinguistic and socio-cultural knowledge.

3.1 BRIEF HISTORY OF SLA

Interest in second language learning is not new. Although SLA began to gain standing as an independent field during
the 1970’s, work had been done in L2 learning via linguistics or first language acquisition research (Cook 1993: 8).
From that moment, SLA research has increased dramatically, and has branched out into many subfields, each
following its own theoretical framework. Many researchers are still pedagogically motivated, while, on the other
hand, much research has departed from practical purposes and has become highly theoretical. Again, it is impossible
and purposeless to draw clear lines between these two extremes, because, as Littlewood (2004: 502) clearly states,
“such ‘non-applied’ research is also likely to improve the basis for making practical decisions”. However, he warns as
to the dangers of teachers trying to apply theoretical research to their teaching practices without due consideration.
The first approaches to the study of L2 learning were derived from general learning principles, within the field of
behaviourist psychology, which dominated the scene between the 1940’s and 1950’s. The first two approaches that
will be discussed in this section (Cook 1993) are based on phrase structure syntax, which analyses sentences by
segmenting them into smaller and smaller units, until they cannot be segmented any further. These approaches are
consistent with the behaviourist view of learning that prevailed at that time, and which viewed learning as the
building of a system of habits acquired through stimulus-response. Uriel Weinreich focused on how two language
systems relate to each other in a person who is bilingual, i.e. somebody who uses two languages alternatively,
without any further specification as to the definition of bilingualism. Weinreich proposes two key concepts: first,
interference, “instances of deviation from the norm of either language which occur in the speech of bilinguals as a
result of their familiarity with more than one language” (in Cook 1993:8). Interference may be present both in
bilinguals’ knowledge of the language and in their actual speech, and across all areas of language. The second key
concept in Weinreich’s view is that of the link between the two languages in an individual’s mind, in terms of how
concepts and words are related. This second key concept has been found flawed in that it restricts the focus to
vocabulary, leaving out other aspects of language. On the other hand, the notion of interference is recurrent in SLA
research.
While Weinreich’s work concentrates on knowledge of language, Robert Lado’s has a pedagogical approach. He
based his research on Contrastive Analysis, a detailed comparison of L1 and L2. The basic notion in Lado’s work is
that of transfer. “Individuals tend to transfer the forms and meanings, and the distribution of forms and meanings of
their native language and culture to the foreign language and culture” (in Cook 1993: 11). This predicts that transfer
will be negative when there is no correspondence between the two languages, while it will be positive when
language items are the same. Language teaching should, therefore, concentrate on those areas in which both
languages differ more, which are expected to cause most trouble. Lado’s work has been criticised on the grounds
that many of the difficulties predicted by it do not actually come true and, at the same time, many of learners’
problems are not predicted by Contrastive Analysis. However, the notion of transfer, just like that of interference, is
still found in current work on SLA. At the end of the 1950’s, Noam Chomsky’s work on the acquisition of an L1 began
to undermine the then prevailing behaviourist approach to the study of language acquisition. Chomsky posited that
stimulus-response learning does not explain the creative aspect of language, i.e. the possibility humans have of
creating sentences they have never heard before. He also considered that the child’s L1 is a developing system in its
own right, not a defective version of adult speech. This concept is known as the independent grammars assumption.

In later work, Chomsky introduced the Language Acquisition Device (LAD), also referred to as the “black box”, a
device in the mind which is specific to language (thus originating the mentalist view of language learning), that
analyses the input it encounters to construct a “generative grammar”. Comparing the input children receive with
their speech production, it is possible to deduce how the process of acquisition develops. In this view, children alter
their grammars by testing their hypotheses until a final one is found, which corresponds to adult competence.
Nowadays, the notion of the LAD has been incorporated into the wider Universal Grammar theory. On the other
hand, the validity of hypothesis-testing in L1 learning is disputed because of the fact that children seldom get the
necessary negative feedback that would allow them to test their hypotheses. Nevertheless, the independent
grammars assumption and the hypothesis-testing model are central elements of SLA.

The early 1970’s saw the beginnings of SLA as an independent field, as well as its branching out into various sub-
fields. One facilitating factor was the recognition that an L2 learner at times, contrarily to what Contrastive Analysis
predicted, uses neither the L1 nor the L2. The independent grammars assumption therefore applies also to L2
learners, who develop an approximate system that gradually nears the target language. The term interlanguage (IL)
was introduced by Larry Selinker to refer to the learner’s independent language system. It differed from the
approximate system in that the former, according to Selinker, seldom reached target language standards. Selinker
claims that IL depends on five central processes, which are part of a general psychological structure: transfer (L1
features are projected onto the L2), overgeneralisation of L2 rules, transfer of training (sometimes a teacher´s
overuse of a language feature may discourage learners from the use of other features), strategies of L2 learning (e.g.
simplification of learners’ structures) and communication strategies (e.g. learners’ omission of redundant
grammatical items). Criticism to Selinker’s IL mentions the fact that he is not clear about whether IL refers to the
learner’s knowledge of the L2 or to its actual use. The same applies to the five processes held responsible for IL.
Nevertheless, the notion of interlanguage was a major contribution to the study of SLA, and has been further
developed by later researchers.

At the same time, Pit Corder introduced a methodology for studying SLA known as Error Analysis. Corder claims that
mistakes, both in L1 and L2, are not really mistakes, but evidence of an internal grammar. Corder also claims that
errors are a means of testing hypotheses, in accordance with Selinker’s IL. There are two methodological problems,
recognised by Corder himself. First, it may be extremely difficult to determine whether a mistake comes from
competence (error) or performance (mistake). The second problem concerns the nature of the error. It is not always
possible to discover the learner’s intended meaning from his/her speech, and as this is a subjective process, it is
prone to failure.

By this time, early SLA theories have abandoned habit-formation and the behaviourist school which supported it, in
favour of the mentalist school which propounded hypothesis-testing as a means of building an interim grammar. The
1980’s saw SLA gather momentum as many researchers followed in previous researchers’ footsteps, while others
took different paths. Much work emanated from various languagerelated disciplines, broadening the field as to
goals, views and methodology. At present, SLA offers a vast array of perspectives, from the most theoretical to the
most practical. Such a variety of approaches ensures debate and disagreement, reflecting the dynamic nature of the
field.

3.2 COMPONENTS OF LANGUAGE LEARNING

Different current approaches to SLA seek to explain language acquisition from different perspectives, based on
different theoretical stances. So far, no definite answers have been advanced. This section explores a variety of
elements that have been identified as crucial factors in the process of acquisition. The following table, taken in part
from Ellis (1985: 276), provides a framework for examining the components of SLA.

3.2.1 Situational factors

Situational factors relate to the context of SLA, i.e. the environment in which learning takes place, and affect both
the nature of the input and the strategies used by learners. Contexts of SLA can be broadly characterised as either
naturalistic or instructed. Naturalistic learning happens within a non-instructional community, in which the learner is
exposed to the target language at work or in social interaction, or at school where instruction is directed at native
speakers. This is typically the case of immigrants. On the other hand, instructed learning normally happens in a
language classroom, although with the new technologies, new ways of learning are becoming more accessible,
allowing for autonomous learning. This polarity is mediated by in-between possibilities, like formal L2 instruction
within the target language community and immersion courses both abroad and at home. According to Ellis (1985:
16), “One of the key issues in SLA research is the extent to which the process of SLA is similar or different in the two
environments.” Many people consider learning in a naturalistic context more effective. According to Lightbown and
Spada (1993), this may be so because of the fact that most successful learners have had exposure to the target
language outside the classroom. Contrarily, other researchers, based on studies, make a claim in favour of
instruction, which they see as “potentially effective, provided it is relevant to the learners’ needs.” (Doughty 2005)
All learning contexts are significant in that they offer different learning opportunities. The following comparison of
natural acquisition and communicative instruction is based on Lightbown and Spada (1993). (Traditional, grammar-
based instruction, in which the focus is on learning the target language instead of its communicative use will be left
out of the comparison since it has been shown not to be effective.)

It is evident that both input and interaction are likely to be different according to context. Both, the type of input
learners receive and the kind of interaction in which they engage affect the negotiation of meaning to a great extent.
As will be seen later, negotiation of meaning is considered to be a crucial element in the acquisition process. As
regards discourse types, both contexts may provide the same or similar types, but in different degrees. As Ellis (1985:
152) summarises, “Considerable differences between natural and classroom environments arise, particularly when
the focus is on form in language lessons. These differences are not absolute; they vary in degree according to the
type of classroom and the type of teaching.”

Situational factors also include social ones. From a variationist perspective, Tarone (2010: 54) studies how “social
setting systematically influences both the kind of second language (L2) input learners receive and their cognitive
processing of it; the speech production of L2 learners; and even, upon occasion, the stages in which learner language
(or interlanguage) forms are acquired.” In other words, contextual variables such as “the identity and role of
interlocutors, topic, and task, as well as contextual linguistic forms,” have a systematic impact on the “learner’s
perception, production, and acquisition of specific aspects of the second language system.” Social settings influence
the input provided in two basic ways. First, learners are likely to have more contact with the particular variety of the
target language used by their social setting. For example, learners in an immersion programme are likely to have
closer contact with academic genres. On the other hand, learners in a classroom situation will lack exposure to
vernacular varieties. Second, the amount of input adjustment provided will vary according to the setting. Tarone
(2010), in accordance with Lightbown and Spada, reports that adjusted input is less likely to be offered to learners in
naturalistic contexts, while it is more available in classroom situations.
Also, social settings influence negotiation of meaning, i.e. conversations involving interlocutors in trying to overcome
communication problems, by selecting salient topics, topic shifting, checking of comprehension, requests for
clarification, slower pace, repetition, etc. Negotiation of meaning can be focused on form or meaning. Tarone (2010)
reports on how learners with a lower-proficiency level tend not to negotiate meaning with higher-proficiency
learners, as the latter are not usually willing to offer explanations. On the other hand, there is more negotiation
when the lower-proficiency learners are the senders of the message. “Social relationships between learners strongly
impacted key cognitive processes involved in the negotiation of meaning…, disregard for an interlocutor who is less
socially dominant or significant to the learner may also cause a learner to ignore or discount that interlocutor’s
corrective feedback on their L2 form.” This point is related to the fact that error detection does not depend only on
psycholinguistic factors, such as attention, but also on the social context, represented by the accuracy demands of
the task, who the listeners are, or whether there is an audience, to name a few. Also, error correction is always
value-laden, so noticing the mistake does not always result in uptake, since the corrector (native speaker, teacher or
peer) has to be accepted by the learner as having the right to give the correction.

Social settings also make an impact on learners’ language production. Research shows that learners adjust their
interlanguage to the forms used by their interlocutors, according to whether there is identification with the
interlocutors or not. This issue is more clearly seen in naturalistic settings, and is explored by Accommodation Theory
(which, due to space restrictions will not be dealt with here). Nevertheless, the mentioned phenomenon does occur
in classroom settings, albeit to a lesser degree. According to Tarone (2010), “In sociocultural theory, new IL forms
originate in collaborative dialogue with supportive others and gradually get internalized”. In other words, new IL
forms may develop in collaborative dialogue in a relaxed setting, a process referred to as scaffolding. For a more
detailed account of contexts of SLA and their influence on learning processes, see Ellis (1994).

3.2.2 Input

Situational factors go hand in hand with input. Also external to the learner, input is of vital importance in SLA, as
learning depends directly on it. Lightbown and Spada (1993: 122) define input as “The language which the learner is
exposed to (either written or spoken) in the environment.” This positive evidence serves as the data which the
learner uses to learn the target language. However, not all input is processed by the learner, as it may have not been
understood, or attended to. The input that is understood and attended to, and therefore processed, is referred to as
intake. This distinction is a fundamental one, as it is intake that leads to learning. SLA cannot ignore L1 acquisition.
Lightbown and Spada (1993: 16) refer to three different general accounts of L1 acquisition which, despite not being
contemporary with one another, offer explanations for different aspects of children’s language development. The
behaviourist position, which posits that children learn by imitating their interlocutors and by receiving feedback on
their utterances, which in turn reinforces or corrects them, may explain how word meanings and some language
routines are learned. The innatist, or nativist, position minimises the role of input by considering it a mere trigger
which activates the internal mechanisms that human beings are endowed with. According to this view, children are
born with some kind of “innately specified knowledge” (Gass 2005: 176) that helps in grammar formation, as the
input they receive does not provide all the information that is necessary for “the extraction of abstractions” (2005:
175). Finally, the interactionist view, which claims that language acquisition is the result of the interplay between the
learner’s mental abilities and the linguistic environment, may explain “how children relate form and meaning…, how
they interact in conversations, and how they use language appropriately” (Lightbown and Spada 1993:16). Thus, in
this view, language acquisition is the result of the interaction of input factors and innate mechanisms.

Early studies of L1 acquisition have focused on the input received by children, sometimes called motherese or
caretaker speech. Child-directed speech is adapted to be made more comprehensible, and it changes according to
the developmental stages of children. It usually contains shorter utterances, few subordinate and co-ordinate
constructions, tutorial questions (i.e. questions to which the interlocutor already knows the answer), and a high level
of redundancy. There are also adjustments in pitch, intonation and rhythm (Ellis 1985: 130). As regards the functions
of motherese, mothers do not pay much attention to the correctness of their children’s speech, while they do
concentrate on the social appropriateness and meaning. Regarding the effects of simplified input, the available
evidence suggests that the route1 of acquisition is not altered by differences in the linguistic environment. On the
other hand, the way mothers talk to their children has effects on the rate2 of learning. However, according to Ellis
(1985: 131), “the key features of the input appear to be interactional rather than formal. That is, it is the mother’s
choice of discourse function (e.g. commands rather than questions) and the devices she uses to sustain the
conversation (e.g. requests for clarifications, expansions, acknowledgements) which provide the right kind of data to
foster development. Research into this type of modified speech led to a new consideration of the role of input, as
more than simply a factor that triggered innate mechanisms, which in turn led to an interactionist view of language
development.

Modified input is not restricted to child-directed language. Within the sphere of SLA, input is present in natural and
instructed settings. As regards natural linguistic environments, there are two areas of special interest: foreigner talk
(simplified talk used by native speakers (NS) to address non-native speakers (NNS)), and the discourse involving
conversations between NS and NNS. Foreigner talk is to be seen as dynamic, in that it changes according to
situational factors, such as “the topic of conversation, the age of the participants…, and, in particular, the proficiency
of the learners.” (Ellis 1985: 133) Foreigner talk is similar to motherese in that both contain simplifications within the
grammar of the language. However, foreigner talk can also contain some ungrammatical speech, if the NNS has a
very low proficiency in the L2, or if the NS considers himself/herself to be of a higher status. Discourse studies have
shown that input is not determined only by the native speaker. The feedback the non-native speaker provides helps
to delineate the nature of the subsequent input provided by the NS. This is particularly noticeable in interaction
between NS’s and older learners, as it is more likely to have instances of negotiation of meaning, by means of
requests for clarification, echoing, repair strategies, and recasts (a corrected version of a NNS’ incorrect utterance).
In sum, although the basic function of foreigner talk is to facilitate communication, it may indirectly provide a
teaching/learning opportunity, which is central to L2 acquisition.

Similar constructs are involved in instructed settings, as the type of language used by the teacher and the type of
interactions that take place in the classroom are conducive to learning. Ellis (1985: 143), for example believes in the
“rejection of language teaching method as the principal determinant of successful learning.” Focusing on classroom
interaction as the major factor affecting SLA in instructed settings led to research into teacher talk and into the
discourse generated in the classroom. Teacher talk is characterised by having formal adjustments in syntax (shorter
utterances), in pronunciation (more accurate, standard pronunciation with lower-level learners), and in lexis (using
more general words). In contrast with foreigner talk, teacher talk does not contain ungrammatical speech. Teacher
talk involves the use of interactional adjustments, some resembling those in motherese, like repetition, prompting,
and expansions. In the classroom, tutorial questions are much more frequent than in naturalistic settings. Research
reported by Ellis (1985: 145) notes that while comprehension checks are more frequent in the classroom,
confirmation checks and requests for clarification are not. The analysis of the discourse produced in classrooms has
shown that, in teacher-oriented teaching, discourse typically consists of three stages, in which the teacher initiates
interaction, the learner responds, and the teacher gives feedback. Apart from reinforcing the teacher’s role as the
sole manager of the learning process, this type of distorted input may not be conducive to language learning in all
levels, as it does not give learners experience in, for example, how to initiate interaction.

Ellis reports on four types of language use that have been identified in the classroom (Ellis 1985:147-148): 1.
mechanical, where no exchange of meaning is involved; 2. meaningful, where language use is contextualized but still
no real information is conveyed; 3. pseudo-communicative, where information is exchanged, but in a way that would
be unlikely to occur outside the classroom; and 4. real communication, which consists of spontaneous natural
speech. This description of discourse in the classroom is complemented by Ellis’ framework (1985: 148), in which he
describes three pedagogical goals: (1) core goals, which relate to the explicit pedagogic purpose of the lesson (e.g. to
teach specific aspects of the L2, …); (2) framework goals, which relate to the organization requirements of the lesson
(e.g. giving out materials, managing pupil behaviour); and (3) social goals, involving the use of language for more
personal purposes (e.g. imparting private information, quarrelling).
He also distinguishes types of address, that is, who functions as speaker, listener and hearer. These aspects combine
to produce a wide variety of classroom interactions, with different patterns. For example, interactions with a
framework goal usually include many directives, to which the learners may respond non-verbally. The frequency of
this type of interaction, which may be profitable for lower level learners, may vary according to the type of
classroom and level of students. Thus, attention to interaction forms becomes of uppermost importance. Although
much of the current research into SLA shows that there may be a natural route for acquisition, the interplay
between input and interaction can affect it in several ways. According to Ellis (1985), one of these is by providing
learners with formulaic speech. Ready-made chunks appear in routinised interactions, and may serve immediate
communication purposes. Also, they are raw material for analysis into component parts, which is a vital part of the
learning process. Frequency of occurrence may also affect the route of acquisition, as learners are likely to learn first
items from the language they are most frequently exposed to. A third issue concerning input is the availability of
comprehensible input. In Krashen’s view, stated in his SLA theory, sometimes known as the Input Hypothesis
(Krashen 1982), learners need to receive input that contains samples of the language which, according to the natural
order, are due to be acquired next. This is what he calls i + 1. In order for learners to understand the input, it is a
prerequisite that they are focused on meaning, not on form. According to Long (in Ellis 1985: 157), input is rendered
comprehensible by the fact that learners use the linguistic and extralinguistic contexts and their general knowledge
to interpret language which they do not actually know. Also, during interaction, some adjustments are usually made
to aid comprehension. If these conditions are met, acquisition will proceed automatically. However, there is
disagreement as to the validity of Krashen’s position. Ellis (1985: 158) explains that SLA can take place “without two-
way communication”, as when input is encountered when reading or watching television or films. In this case, there
are no interactional modifications. Another problem with Krashen’s view is that “interactional modifications do not
always result in comprehensible input” (1985: 158). Finally, Krashen’s theory fails to recognise the role of output.
While output is considered a vital element in language acquisition (see section on output), the only role allotted to it
by Krashen is that of generator of more input (Krashen 1982: 60). Yet another problem is that not all input that is
understood will be processed by the learner. Intake, a necessary condition for SLA will therefore not always be
present. Little is known about how learners select from the input they receive, but it is believed that factors that may
mould intake are motivation, the internal processing mechanisms, and features of the input.

Input may also affect the rate of acquisition. The studies reported by Ellis (1985: 160-161) into the effects of input
and interaction in classroom environments have led him to suggest that the following features are likely to aid rapid
development: 1. A high quantity of input directed at the learner. 2. The learner’s perceived need to communicate in
the L2. 3. Independent control of the propositional content by the learner (e.g. control over topic choice). 4.
Adherence to the ‘here-and-now’ principle, at least initially. 5. The performance of a range of speech acts by both
native speaker/teacher and the learner (i.e. the learner needs the opportunity to listen to and to produce language
used to perform different language functions.) 6. Exposure to a high quantity of directives. 7. Exposure to a high
quantity of ‘extending’ utterances (e.g. requests for clarification and confirmation, paraphrases and expansions.) 8.
Opportunities for uninhibited ‘practice’ (which may provide opportunities to experiment using ‘new’ forms). In sum,
it is clear that input is one of the vital elements in SLA. However, what is not so clear is the amount of responsibility
that it has, as opposed to the internal mechanisms. As Hatch (in Ellis 1985: 162) states, “While social interaction may
give the learner the ‘best’ data to work with, the brain in turn must work out a fitting and relevant model of that
input.”

3.2.3 Learner differences

In contrast with situational factors and input, which are external to the learner, there is another construct that
interacts with input and learner processes: learner differences. This area of SLA has been motivated by the need to
explain the striking variation in learning outcomes, especially as regards rate of learning and levels of achievement.
Individual factors have proved to be difficult to define and classify. According to Ellis (1985), this is due to the fact
that qualities like aptitude or motivation cannot be directly observed. Also, individual factors interact with one
another affecting language learning. Those factors are in fact clusters of behaviours, and there is no consensus
among researchers as to their definitions. However, Ellis proposes a distinction between personal and general
factors. The first are “highly idiosyncratic features of each individual’s approach to learning a L2” (1985: 100), while
the second are characteristics of all learners. General factors are of two types: modifiable (likely to change during the
learning process) and unmodifiable. Both personal and general factors have social, cognitive and affective aspects.
These aspects are all present in the mentioned factors, in different degrees. Cognitive factors are related to the
problem-solving strategies used by the learner. Social aspects concern the relationship a learner has with native
speakers of the target language and with speakers of his/her own language. Finally, affective factors are relative to
the emotional responses caused by the learning process.

3.2.3.1 Personal factors

Group dynamics

Group dynamics, important in classroom settings, relate competitiveness and anxiety experienced by learners. Bailey
(in Ellis 1985: 101-102) presents a model of how the learner’s self-image is affected by comparison with other
learners, resulting in a successful or unsuccessful self-image. If the learner has a positive self-image, then learning
will be enhanced. On the contrary, if the image is a negative one, this may result in facilitating or debilitating anxiety.
In the case of the former, the learner will try to improve L2. In the case of the latter, learning may be impaired or
even abandoned.

Attitudes to the teacher and course materials

In Ellis’ review of the topic (1985), learners appear to have very different views as to what makes a good teacher.
Some prefer teaching to be structured or predictable, others dislike having to follow somebody else’s teaching plan.
Something similar applies to coursebooks. In general, adult learners prefer a variety of materials to a rigid use of the
coursebook.

Individual learning techniques

Different learners make use of a wide variety of learning techniques. They may be classified into techniques for
studying the L2, such as preparing and memorising vocabulary lists, learning words in context and reading to
reinforce learning, and for obtaining L2 input, as in seeking opportunities for communication with native speakers, or
for exposure to the target language through, for example, films.

3.2.3.2 General factors

This section will refer to the factors that have received most attention in SLA research.

Personality

Although personality is considered to be a crucial variable in success in language learning, the available research
does not provide conclusive results. According to Ellis, there is fairly substantial support for the claim that
extroverted learners “will do better in acquiring basic interpersonal communication skills” (1994: 520), mainly by
obtaining more input. However, “the effects of extroversion/introversion may be situation-dependent”. He
concludes that this may be so because personality “becomes a major factor only in the acquisition of communicative
competence.” (1985: 121)

Intelligence and aptitude

There is no general consensus as to whether intelligence and language aptitude are separate constructs or are two
aspects of a single general language faculty. Even so, both have been found to have an influence on L2 learning,
especially when studied in the context of classroom learning. For those scholars who consider intelligence separate
from aptitude, intelligence refers to a general reasoning ability, which underlies our use of academic skills. It does
not refer to the knowledge in our minds, but to our ability to learn. According to the studies reported by Ellis (1985:
111), intelligence does not seem to be a “mayor determinant of L1 acquisition”, so it is probably not so important in
SLA in naturalistic contexts. As regards instructed SLA, intelligence may influence the acquisition of some skills
associated with formal study, like reading, grammar, vocabulary and free writing, while its relation with the
development of oral fluency skills is much less certain. Also, the influence of intelligence is restricted to the rate and
success of SLA, as there is no evidence that it affects the route of acquisition. On the other hand, language aptitude
has been found to be a better predictor of L2 learning than intelligence. Aptitude refers to “specific cognitive
qualities needed for SLA” (Ellis 1985: 111), and following Carroll’s research (in Ellis 1994:496) four factors are
identified: 1. Phonemic coding ability (the ability to code foreign sounds in a way that they can be remembered
later). This ability is seen as related to the ability to spell and to handle sound-symbol relationships. 2. Grammatical
sensitivity (the ability to recognize the grammatical functions of words in sentences). 3. Inductive language learning
ability (the ability to identify patterns of correspondence and relationships involving form and meaning). 4. Rote
learning ability (the ability to form and remember associations between stimuli). This ability is hypothesized to be
involved in vocabulary learning. It should be noted that aptitude has been studied mostly in relation to the linguistic
aspect of language learning, as opposed to the development of interpersonal communication. Also, aptitude is not a
prerequisite for SLA, but a capacity that “enhances the rate and ease of learning.” (Ellis 1994: 495) In this respect,
there is no evidence that aptitude has any effect on the route of SLA, while, on the other hand, it can be expected to
influence the rate of development and have effects on ultimate success in SLA.

Cognitive and learning styles

The concept of learning styles comes from general psychology. On the whole, cognitive and learning styles are
relevant to SLA as, according to Dörnyei and Skehan, different learning styles “may be equally valid and
advantageous” (2005: 450), and it is possible for different styles to make contributions to different domains. Also, as
they appear to be less fixed than other factors (like aptitude), learners may adapt their styles to meet the needs of
particular situations. Ellis (1985: 114) defines cognitive styles as the way in which people “perceive, conceptualize,
organize and recall information.” Dörnyei and Skehan make a distinction between cognitive and learning styles: “The
former can be defined as a predisposition to process information in a characteristic manner while the latter can be
defined as a typical preference for approaching learning in general. The former, in other words, is more restricted to
information-processing preferences, while the latter embraces all aspects of learning.” (2005: 451) As regards
cognitive styles, a number of distinctions have been made in cognitive psychology, but one that has attracted much
attention in SLA is that of the contrast between field independence (FI) and field dependence (FD). The following
table shows the main characteristics of FD and FI cognitive styles:

As can be hypothesised from the table above, field independents may prefer to study alone, and benefit from
analysing linguistic material, while field dependents may work well in groups and profit more from communicative
language use. There are, however, some points to clarify. First, the distinction should not be taken as comprising two
polarities, but poles on a cline, with individuals varying in their predispositions. Second, according to some research
reported by Ellis (1985: 115), the effects of cognitive style may be related to age, in that field independence is
“facilitative in the case of late adolescence but not before.” Also, researchers disagree as to the usefulness of the
FD/FI distinction in that there have been problems both as to the definitions and measurements of the constructs. In
all, the results are not at all conclusive, but the research area is considered to be promising. Other approaches to the
study of learning styles include sensory preference. Reid (in Dörnyei and Skehan 2005) distinguishes four perceptual
learning modalities: 1. Visual (e.g. reading and studying charts) 2. Auditory (preference towards listening) 3.
Kinaesthetic (involving physical response) 4. Tactile (using their hands, as in building models) According to this
model, learners benefit from tasks that allow them to use their preferred sensory styles. Various other approaches
are being studied in relation to their effects on SLA. For more info, see Dörnyei and Skehan (2005) . As a general
conclusion, it can be said that learners vary a great deal in their preferred styles for L2 learning, and that there is no
“best” style. In Ellis’ words (1994: 508), it may be that learners who “display flexibility are those who are most
successful, but there is no real evidence yet for such a conclusion.” Dörnyei and Skehan (2005: 454) conclude that
the concepts of cognitive and learning styles have not been clearly defined in the literature, and are sometimes
conflated with other factors. However, they argue that, while they may not “deserve high research priority, … they
have not been eliminated as potentially relevant second language linked measures.”
Motivation

The concept of motivation, which has attracted much attention within the field of SLA as being considered one
accurate predictor of L2 learning success, is not without difficulties. One of them is related to the conceptualisation
of the construct, another to the fact that motivation cannot be directly observed, and therefore it has to be inferred
from people’s behaviours (Ellis 1985: 116). Dörnyei and Skehan (2005) describe motivation as concerning “the
direction and magnitude of human behavior, or, more specifically (i) the choice of a particular action, (ii) the
persistence with it, and (iii) the effort expended on it. In broad terms, motivation is responsible for why people
decide to do something, how long they are willing to sustain the activity, and how hard they are going to pursue it.”
Given the array of potential determinants of motivation, different approaches to it focus on different factors on
which to base their studies. As a result, none of the current positions offer a comprehensive view of all the factors
that affect motivation. One of the central approaches to motivation is that of Gardner, oriented towards the roles of
attitudes and motivation in SLA (in Dörnyei and Skehan 2005). His model includes the distinction between integrative
and instrumental orientations towards L2 learning. The first reflects a positive disposition towards the L2 group and
a desire to identify with its culture. The second relates to functional goals, such as passing an examination, getting a
better job, facilitating the study of other subjects through the medium of the L2, etc. Although these two concepts
are widely known in the L2 field, there is still a wider concept in Gardner’s theory, the integrative motive, which is
made up of three main components: “(i) integrativeness, subsuming integrative orientation, interest in foreign
languages, and attitudes toward the L2 community; (ii) attitudes toward the learning situation, comprising attitudes
toward the teacher and the course; and (iii) motivation, which according to Gardner is made up of motivational
intensity, desire to learn the language, and attitudes toward learning the language.” (in Dörnyei and Skehan 2005:
458) Related to Gardner´s integrative motive is the notion of intrinsic motivation. This is understood as an inherent
interest in the learning process and tasks. This concept is of utmost importance in SLA as it is possible to manipulate
it in the classroom, in an attempt to arouse and sustain students’ interest. Ellis (1994) reports on several ways to
enhance learners’ motivation: by providing opportunities for communication, by learners’ becoming self-directed
(able to choose their learning objectives and ways of achieving this, and to evaluate their own progress), by
presenting students with tasks that pose a reasonable challenge, by providing opportunities for group work, and
variety. Another influential approach is that proposed by Heckhausen and Kuhl, known as Action Control Theory (in
Dörnyei and Skehan 2005: 461), which differentiates between the predecisional and the postdecisional phases of
motivation. The former refers to “the pre-actional stage of deliberation associated with planning, goal setting, and
intention formation” while the latter is about “influences that come into force when action has started and therefore
concern motivational maintenance and control, perseverance, and overcoming various internal obstacles to action.”

From a neurobiological perspective, Schumann’s approach tries to link the study of language with cognitive science
(in Dörnyei and Skehan 2005: 462). The key issue is stimulus appraisal, which occurs in the brain along five
dimensions: novelty (degree of unexpectedness/familiarity); pleasantness (attractiveness); goal/need significance
(whether the stimulus is instrumental in satisfying needs or achieving goals); coping potential (whether the individual
expects to be able to cope with the event); and selfand social image (whether the event is compatible with social
norms and the individual's self-concept). Thus, stimulus appraisal can be seen as a key process underlying executive
motivation. From the points mentioned above there can be little doubt that motivation is a powerful factor in SLA.
According to Ellis (1985: 119), its effects can be seen on both the rate and success of L2 learning. What is not clear is
whether it is motivation that produces successful learning, or vice versa. Whatever the directionality, the key issue is
that, as motivation is more susceptible to change than other personal factors, teachers have a greater role in
fostering it. This can be achieved by a careful selection of activities that are relevant to the learners’ interests and
needs, and that are in accordance with their level, so as to prevent anxiety (if the task is too demanding) and
boredom (if the task does not pose a challenge), and by giving learners enough autonomy so that they can gain more
control over their own learning process.

Age
Age has been the most considered individual factor in SLA, and it is still the most controversial. Although age does
not present any difficulties as to measuring, it can hardly be separated from other factors that inevitably interact
with it, such as context of learning, length of stay (in the case of immigrants), amount of instruction, age of onset,
and ultimate attainment, to mention a few.

In SLA research, it is widely agreed that in a naturalistic context, early starters tend to attain high levels of language
competence. In contrast, learners who start later in life, especially after the end of adolescence, show a great
variability in their levels of linguistic attainment (Ortega 2011: 176). Strozer (1994: 130) is clear about the effects of
age: all normal children are totally successful at acquiring the language of their communities, while most adults who
try do not succeed in developing a mastery of a single foreign language. This sharp disparity … at first may strike us
as paradoxical (the greatest success is achieved by the least developed organisms, which are in fact less capable at
most things than adult organisms). The current, and unresolved debate, concerns the explanations for the commonly
agreed on fact expressed above. One of the most influential biological explanations is the Critical Period Hypothesis
(CPH), first advanced in the late 1960´s, and mostly associated with Eric Lenneberg. The hypothesis holds that
“language acquisition must occur before the onset of puberty in order for language to develop fully.” (Johnson &
Newport 1989:77) Lenneberg proposed a neurological mechanism which might be responsible for the change in
learning abilities: once the brain has achieved adult values by puberty, it loses the plasticity needed for language
acquisition. This points to a sharp decline in the ability to acquire an L1. However, Johnson and Newport (1989)
mention two problems with Lenneberg’s formulation of the critical period hypothesis. First, previous studies show
that the decline in the ability to learn an L1 is not in fact as sharp as claimed. Second, L1 acquisition by
postpubescent learners, albeit lower in ultimate attainment, is not altogether impossible. In conclusion, an extreme
interpretation of the hypothesis should be ruled out. Johnson and Newport’s seminal study investigated whether the
CPH applies to L2 as well. It focuses on ultimate command of the grammar of the L2 with respect to the age of
exposure to that language. Subjects on this study were Chinese and Korean immigrants to the US, who learned
English as a second language. They varied in the age at which they moved to the US, and had lived there for at least 3
years.

In order to make the distinction between L1 and L2 clear, the researchers outline two versions of the critical period
hypothesis. While the two versions make the same predictions for L1 acquisition, they differ as regards L2 learning.
The exercise hypothesis (1989: 109) states that “Early in life, humans have a superior capacity for acquiring
languages. If the capacity is not exercised during this time, it will disappear or decline with maturation. If the
capacity is exercised, however, further language learning abilities will remain intact throughout life.” This version
predicts that L2 learning should be the same in children and adults, or perhaps even superior in adult learners, as
they already possess language skills from their L1. The maturational state hypothesis, on the other hand, claims that
the superior capacity for learning languages “disappears or declines with maturation” (1989: 109), and therefore
predicts that children are better at acquiring any language, L1 or L2.

The results of the study support the maturational state hypothesis: “Human beings appear to have a special capacity
for acquiring language in childhood, regardless of whether the language is their first or second.” (1989: 109) As
regards the decline in performance, the study did not find a sudden drop in performance at the end of the critical
period. Instead, it showed a gradual decline from about age seven on until adulthood, with a marked drop around
puberty. After puberty, performance did not continue to decline with increasing age, but showed important
individual differences. From these findings, it can be concluded that learning an L2 after puberty is not impossible, in
spite of some deficiencies. Also, for adult learners, age is not a predictor of performance. The study also takes into
account variables other than age, in order to investigate whether experiential or attitudinal factors can affect the
effects obtained for age of acquisition. As regards the experiential variable, it was found that length of exposure did
not alter performance significantly beyond the first few years of exposure (in an immersion context). Attitudinal
factors, such as motivation, identification and self-consciousness, were also unable to explain away the effects of
age.
In sum, Johnson and Newport’s work seems to prove their claim that the critical period does exist, although not in an
extreme interpretation (some researchers refer to it as a sensitive period), and that it applies to both first and
second languages. Also, the existence of a critical period does not rule out the benefits of exercise, allowing for the
possibility of language acquisition after puberty, in spite of its wide variety as regards outcomes. Thus, the two
versions of the hypothesis are not mutually exclusive. Hyltenstam and Abrahamsson (2005: 420) classify current
explanations for age effects into three areas. According to these authors’ review of the literature on the topic,
biological explanations have addressed the brain's steady loss of flexibility or plasticity. Even though little is known
about the exact cerebral mechanisms that are responsible for differential outcomes of language learning at different
phases of life, there is enough independent knowledge of changes in the brain taking place during the time when
language acquisition outcomes differ systematically to be suggestive of hypothesized relations between the two.
Cerebral plasticity is related to the ability of neurons to make new and varied connections depending on the
stimulus, and also to the strengthening of those connections. Other biological explanations include metabolic
changes in the brain around puberty, thickening of the corpus callosum and the process of lateralisation, that is, “the
neurological capacity for understanding and producing language, which initially involves both hemispheres of the
brain,” and which “is slowly concentrated in the left hemisphere for most people.” (Ellis 1985: 107) As far as
social/psychological explanations for the effects of age are concerned, these factors are believed to have some
influence on the process of L2 acquisition. However, their impact is not as significant as that of age itself. For
instance, “there is no direct evidence that children would be inherently more motivated to learn the L2, or that they
receive more input than adults… In addition, several empirical studies … have shown that motivational factors
cannot account for the decrease in ultimate attainment with increasing” age of onset (Hyltenstam and Abrahamsson
(2005:421-422). In the case of adult learners, while age is still the dominant factor affecting outcomes, the variability
between highly successful learners and other L2 learners may be considered the result of (a combination of) non-
maturational variables such as motivation, affective/attitudinal factors, input, type of instruction, verbal analytical
ability, metalinguistic awareness, and language aptitude.

Finally, Hyltenstam and Abrahamsson offer cognitive explanations, which seem to enjoy general consensus as
regards the fact that cognitive factors must have an influence on the process of L2 learning, although there is
disagreement as to how they play their part. According to their review (2005: 423), general problem-solving
mechanisms, typical of older learners, may “counteract the normal ‘direct’ processing of target language input” in
children. This view is similar to explanations linked to the UG (Universal Grammar)(see below) position, which can be
viewed from different perspectives. The fundamental difference hypothesis claims that adults are different from
children in that they no longer have access to the language acquisition device (LAD), and therefore have to rely on
general problem-solving mechanisms. The competition hypothesis claims that adults have continued access to UG,
and that the LAD competes with general problem-solving processes, which eventually succeed. The less is more
hypothesisstates that children’s limitations on cognitive capacity allow them to “focus on and store component
pieces of the input, while adults unsuccessfully try to analyze complex chunks of input simultaneously.”
Nevertheless, these cognitive explanations are not without controversy, as they would predict basic differences in
learning processes between children and adults as regards the order and sequence of acquisition, while the available
evidence points to the fact that age does not affect the route of acquisition. In other words, children, adolescents
and adults follow the same stages of development, irrespective of their age. On the other hand, rate and success
appear to be strongly influenced by the maturational factor. Ellis (1985: 106) gives the following summary of the
effects of age: 1. Starting age does not affect the route of SLA. Although there may be differences in the acquisitional
order, these are not the result of age.

2. Starting age affects the rate of learning. Where grammar and vocabulary are concerned, adolescent learners do
better than either children or adults, when the length of exposure is held constant. 3. Both number of years of
exposure and starting age affect the level of success. The number of years’ exposure contributes greatly to the
overall communication fluency of the learners, but starting age determines the levels of accuracy achieved,
particularly in pronunciation. As regards the different areas of language acquisition, they do not seem to be affected
alike. Strozer (1994: 160) reports on adult acquisition of accent-free phonology, widely accepted as the first loss
brought about by age. Research shows that the critical period “relates more to the peripherals of production and
perception than to the central core of language… since there seems to be no critical period for at least one aspect of
language acquisition, namely, the growth of the vocabulary, which is a lifelong process.” In sum, maturation seems
to be a major determinant for language acquisition, while nonmaturational factors seem to account for the
variability between highly successful and nonexceptional L2 learners of the same starting age. As regards
social/psychological factors, they seem to become more important with age, as they can compensate for the
negative effects of maturation. Finally, it is necessary to look at instruction in relation to the influence of age on the
SLA process. In a naturalistic setting, younger learners usually have more overall time to learn. They also have more
opportunities to use the language without the strong pressure to speak correctly and fluently that they may
experience in a classroom setting. On the other hand, adolescent and adults may develop a sense of frustration
when they find themselves in situations in which they are expected to express complex ideas and language.
Contrarily, in the case of instructed L2 learning, according to Lightbown and Spada (1993: 50), “it may be more
efficient to begin second language teaching later. In research on school learners receiving a few hours of instruction
per week, learners who start later (for example, at age 10, 11, or 12) catch up very quickly with those who began
earlier.” However, they warn of the fact that “One or two hours a week – even for seven or eight years – will not
produce very advanced second language speakers.” In spite of this, there may be an advantage in an early start with
appropriate teaching and a sufficient amount of time. Some characteristics need to be taken into account when
teaching children, such as their need to use their bodies when learning, their need to develop a strong emotional
attachment with their teachers, and their short attention span, which calls for variety of activities. According to a
study reported in Johnstone (2002: 12), when children are around 6, they feel positive about learning a new
language when they enjoy their classroom activities. At around 9, children notice that they are learning, and feel
pleasure from this. As regards older learners, they benefit from the fact that they can make use of already existing
metalinguistic awareness from their L1. Also, they are more experienced in the negotiation of meaning, and
therefore will get better input.

Johnstone (2002: 13-14) provides a list of learner characteristics together with advantages and disadvantages for
different ages. “Given appropriate teaching conditions for learning, younger learners may possess the following
advantages over older beginners … ▪ they are likely to find it easier to acquire a good command of the sound system
of the language, not only the pronunciation of individual sounds but also patterns of intonation; ▪ they are likely to
be less ‘language anxious” than many older learners and hence be more able to absorb language rather than block it
out; ▪ they are likely to have more time available overall. If young beginners at age 5 are compared with older
beginners at age 10 then after one year the older group are likely to be ahead. However, if both groups are
compared at (say) age 14, the younger beginners stand a better chance of being ahead, in part because of the
greater amount of time available overall; ▪ an earlier start enables productive links to be made between first and
additional languages, which can have important benefits for a child’s language awareness and literacy; ▪ a range of
acquisitional processes can come into play, e.g. largely intuitive processes at an early age, complemented by more
analytical processes later. This potentially allows the additional language to become more deeply embedded in the
person; ▪ there can be a positive influence on children’s general educational development (e.g. cognitive, emotional,
cultural) and on the formation of a multilingual and intercultural identity.” On the other hand, “Older learners may
possess some or all of the following advantages over younger beginners: ▪ they may be able to plot their new
language to concepts about the world which they already possess from their first language. ▪ they may be more
experience in handling the discourse of conversations and other language activities, and thus may be more adept at
gaining feedback from native speakers or teachers and in negotiating meaning…; ▪ they are likely to have acquired a
wider range of strategies for learning, e.g. notetaking, use of reference materials, searching for underlying pattern.
This, allied to their established literacy in their first language, may help them become more efficient learners; ▪ they
may have a clearer sense of why they are learning an additional language and may therefore be able to work
purposefully towards objectives of their own choosing.” To conclude, and taking into account the teaching practice,
knowledge about the effects of age should provide a basis for the design of teaching materials and methodologies,as
well as language policies. Teachers can be better equipped for evaluating materials, techniques and practices in
relation to their students and their needs. In Johnstone’s words (2002: 21), “given suitable teaching, motivation and
support, it is possible to make a success of language learning at any age and stage, though older learners are less
likely to approximate to the levels of a native speaker.”

UNIT 4

4.1 Learner processes

Another complex element in second language acquisition is learner processes. What goes on inside the “black box”
has been a topic of long-standing debate, and has triggered several sometimes conflicting approaches. The main
disagreement seems to be about the nature of language learning, that is, whether it is biologically determined and
language specific, or whether language is developed within general cognitive abilities. The construct of learner
processes comprises three elements that are in constant interplay: L2 knowledge, language processing and other
knowledge. It is necessary to keep in mind that learner processes are affected by situational factors, input and
individual differences, and that, these, in turn are affected by learner processes.

4.1.1. Other knowledge

Schemata

The first component to be discussed in this section will be knowledge of the world. In order to understand language,
humans rely not only on their knowledge of language, either L1 or L2, but also on their knowledge of the world. This
is mostly automatic and unaware. Relevant to this area is the concept of knowledge schemata. According to Cook
(1989: 69), these are “mental representations of typical situations,…used in discourse processing to predict the
contents of the particular situation which the discourse describes. The idea is that the mind, stimulated by key words
or phrases in the text, or by the context, activates a knowledge schema, and uses it to make sense of the discourse.”
Schemata involve the use of pre-existing knowledge of the world (that is, stereotypical situations, sequences of
events, roles, relationships, etc) in the comprehension of interaction. Apart from aiding comprehension, schemata
allow communication to be economical, in that information that is shared by the interlocutors will be taken for
granted and, therefore, will not need to be fully explicit. Schemata are complex data structures, many of which may
be activated at the same time. Also, the mind is capable of building new ones as well as of discarding old ones.
Communication may fail if people make false assumptions as to shared schemata, and in the case of people from
different cultures and languages, there may be misjudgements and mismatches. As can be seen, language learning is
not alone in the process, and it can be aided by knowledge which does not belong in the realm of language.

L1

The other fundamental component of this section on other knowledge is the L1, and its influence on SLA, mainly as
regards transfer. Not surprisingly, the problems related to cross-linguistic influence are various and complex. In
consequence, there is no conclusive theory of transfer. The role of L1 was first studied within the behaviourist
position, and was linked to the errors produced by learners. According to this view, errors were the result of
interference from the habits of the L1. Weinreich studied patterns of negative transfer, or what he called
interference, as well as positive transfer, that is, “the facilitating influences that may arise from cross-linguistic
similarities.” (Odlin 2005: 334) Contrastive Analysis was developed as an attempt to predict areas in which learners
would encounter difficulty. However, the claims made by Contrastive Analysis were not supported by evidence, since
many of the errors predicted by it did not occur, while others, not predicted, did actually arise. The role of the L1 was
reconsidered, and the L1 is now viewed as “a resource of knowledge which learners will use both consciously and
subconsciously to help them sift the L2 data in the input and to perform as best as they can in the L2.” (Ellis 1985:
40)The use of this resource depends on linguistic factors, such as the formal and pragmatic features of both L1 and
L2, and on psycholinguistic and sociolinguistic factors, such as the learner’s stage of development and type of
language use. Although Ellis (1985: 40) claims that the influence of the L1 is most evident in the phonological aspect
of the target language, studies reported by Odlin (2005) show that there are problems with this view, in that
frequency of occurrence and individual perceptions of the similarities between the native and target language may
affect predictions of interference.

One factor which Weinreich and Lado already considered important, and which is still considered so, is that of
language distance, that is, what learners think the difference between their native language and the target language
is. Learners have the possibility of relying on their L1 for help. “Having such opportunities, however, does not
guarantee that any particular learner will do the necessary looking or come to the right conclusion about just how
congruent a crosslinguistic correspondence is.” (Odlin 2005: 337) In other words, language distance depends on
individual perceptions of interlingual similarities, and judgements may not always be accurate. Moreover, there may
be other constraints, such as general cognitive capacities, like perception and memory.

Language acquisition is not independent of social factors, thus the notion of cultural transfer. This includes a variety
of factors, from literacy and discourse conventions, to conceptualisations and pragmatics. Transfer, both positive
and negative, are intuitively obvious as regards writing conventions. On the other hand, conceptual and pragmatic
processes of transfer are not so evident. Odlin (2005: 351) defines conceptual transfer as the mapping of concepts
underlying the L1 words onto L2 linguistic labels, regardless of the possible differences in the semantic boundaries
between the corresponding words. Conceptual transfer entails more than just the mapping of concepts onto new
words, as is evident in the coding of concepts in, for example, idioms, as correspondence is not always absolute.
Sometimes, conceptualisations are different, as in the mass-count distinction. Pragmatic transfer can also be at work,
as, for example, in how affect is coded in language. In Odlin’s words (2005: 352) “Learning to speak a second
language normally requires learning a repertoire at least somewhat different in how affect is coded, and difficulties
may arise either from ‘being at a loss for words’ or from using a pragmatic routine acceptable in the native language
but not in the target language.” Learning new pragmatic routines may cause an identity problem, and therefore, an
affective constraint.

In sum, schemata and the L1 exert considerable influence on learner processes, which, in turn, influence the process
of SLA. Formal language instruction can help by raising learners’ awareness of these facts, which are usually not
obvious to students. Instruction can help by guiding learners as to the process of transfer, so that their L1 becomes
more of a facilitating element.

4.1.2. Language processing

This section will explore some of the internal processes that allow for language acquisition. Different researchers
describe what goes on inside the “black box” differently, focusing on various aspects according to the theoretical
point of view they subscribe to. However, this is an oversimplification as, even within similar backgrounds, there are
different foci of attention. The concepts introduced in this section are key concepts in language processing, and will
provide the basis for the description of SLA models below. Ellis (1994: 415) proposes a classification of the available
explanations for the internal mechanisms that are responsible for SLA into two strands: those that focus on the
linguistic nature of acquisition, and those that focus on its cognitive nature. The former discuss L2 acquisition in
terms of linguistic rules, principles and universals, and lies within the field of linguistics. The latter concentrates on
processes, strategies and operations, and is therefore within the scope of psychology.

Linguistic description

UG

The theory of Universal Grammar (UG), both a description of grammar and a model of language acquisition, derives
from Chomsky’s view that language learning is best explained in terms of an independent language faculty,
sometimes called Language Acquisition Device (LAD), which is independent of the general cognitive mechanisms
responsible for intellectual development. The LAD is genetically endowed with principles about grammar, which
allow the child to build his/her L1 by means of hypothesis testing. This position minimises the role of input, in that
the role ascribed to it is simply that of the trigger that activates the process of acquisition. The fact that the data
available from the input is not sufficient to enable the child to acquire certain rules, together with the fact that
children seldom receive negative feedback are two of the main reasons for the explanation of the innateness of UG.
This claim is referred to as the Poverty of the Stimulus argument. UG is a system of linguistic universals that are
present in all human languages. In Ellis’s discussion, (1994: 430) these universals consist of universal principles and
language-specific parameters, which constrain the hypotheses learners will make. The first are highly abstract
properties of grammar which underlie the grammatical rules of all languages. The second are principles that vary
from language to language. Another key concept within UG is markedness, and is related to core and peripheral
features. Core features are governed by UG, and are unmarked, and therefore require minimal evidence to be
acquired. Peripheral features are marked, and will require much more evidence to be acquired.

A number of researchers have argued that the logical problem of language acquisition also applies to SLA, that is, L2
learners “possess knowledge of the L2 that they could not have acquired from the input and which must, therefore,
have existed within their own minds.” (Ellis 1994: 432) Thus, the poverty of the stimulus argument applies to L2
learning too, although SLA may be more dependent on the nature of the evidence. In conclusion, although the
theory of UG is a quite well developed one, it does not yet offer a conclusive explanation of language processing. (UG
will be taken up below, in the section on theories of acquisition, and will be developed a little further.)

Cognitive descriptions

On the cognitive side, research has concentrated on constructs like processes, strategies and operations. This
subsection will explore some of the internal processes that account for what learners do with input data.

Declarative/procedural memory

Work within the neurosciences has produced a model that claims that “the distinction between the mental lexicon
and the mental grammar in language is tied to the distinction between declarative and procedural memory….That is,
lexical memory depends largely on the declarative memory system, whereas aspects of grammar depend on the
procedural memory system.” (Ullman 2004: 233) These two memory systems depend on different parts of the brain.
Declarative memory is associated with the learning, representation and use of facts and events, and at least part of
this knowledge can be consciously retrieved. It is “knowing what”. Procedural memory, on the other hand, is an
implicit memory system, as it is not usually available to conscious access. It is instrumental in the learning of new,
and the control of established habits and skills; it is “knowing how”. Learning in this system is gradual and slow, as
compared to the fast learning produced by declarative memory. The relevance of this model to SLA lies on the fact
that, according to its supporters, the brain system underlying declarative memory also underlies the mental lexicon,
that is, it stores word meanings, sounds and categories; it stores both free and bound morphemes, irregular forms,
idioms, etc. On the other hand, procedural knowledge is assumed to play a crucial role in the combination of stored
forms and abstract representations into complex structures. These two systems are hypothesised to interact, in that
the “rapid lexical/declarative storage of sequences of lexical forms should provide a database from which
grammatical rules can gradually and implicitly be abstracted by the procedural memory system.” (Ullman 2004) Also,
“Access to a stored representation which has similar mappings to one which could be composed compositionally by
the procedural system (e.g. an irregular versus a regular past-tense form of the same verb) would block completion
of the latter computation.” The declarative/procedural model can provide an insight into the acquisition of the
different elements of language, as well as give the process a biological underpinning.

Explicit/implicit learning

These two concepts are relevant to the study of SLA as regards their effectiveness in the learning process. There is
disagreement, though, as to the definitions of the concepts, especially as regards the conceptualisations of terms
used in definitions. Basically, implicit learning takes place without consciousness or awareness, while explicit learning
involves intentionality. However, DeKeyser (2005: 241) explains that “subjects in experiments on implicit learning
usually have the intention of learning something, even though they may learn something different from what they
intended to learn.” Also, there are different positions as to the existence of an interface between explicit and explicit
knowledge. In DeKeyser’s words, (2005: 242) “Even though implicitly acquired knowledge tends to remain implicit,
and explicitly acquired knowledge tends to remain explicit, explicitly learned knowledge can become implicit in the
sense that learners can lose awareness of its structure over time, and learners can become aware of the structure of
implicit knowledge when attempting to access it”. This is not without debate, as there are different opinions as to
the existence of an interface between explicit knowledge and language use, and the current empirical evidence is
not conclusive. DeKeyser (2005: 250-251) summarises the positions as follows:

1. a non-interface position, usually associated with Krashen, which states that explicit learning can never become
implicit (“acquired”) knowledge.

2. an opposite view, which claims that the gap between explicit learning and use can be bridged by means of
practice, at least for some rules.

3. two intermediate positions, favoured by supporters of focus on form (i.e. learners direct their conscious attention
to a language feature while trying to carry out a communicative activity. This is in contrast with focus on forms, i.e.
focus on the formal elements of the language, without attention to meaning.) Within this intermediate position, one
is for making learners explicitly aware of a structure, but the focus on form technique used is not necessarily explicit,
as with input enhancement (input which is manipulated by a teacher to make some features more salient, so as to
facilitate acquisition). Another intermediate position is that advocated by Ellis (1997: 123), who claims that explicit
learning can “help the learner to notice features in the input that would otherwise be ignored.” Also, explicit
knowledge may lead the learner to notice the gap between his/her performance and input.

As regards SLA, the review of current research supplied by DeKeyser (2005:251) suggests that there may be a
positive role for attention to form, either through the explicit teaching of grammar and explicit error correction, or
through more indirect means such as input enhancement. Moreover, research suggests that it is possible that
learners lose their awareness of rules after large amounts of communicative use and automatisation of the rules. “At
that point they not only have procedural knowledge that is functionally equivalent to implicitly acquired knowledge,
but even implicit knowledge in the narrow sense of knowledge without awareness.” As can be seen from the
preceding section, the notions of explicit/implicit, declarative/procedural are not totally independent of one
another. The same applies to the following pair.

Automatic /controlled processes

These concepts come from an approach to SLA that views language learning as a special case of complex skill
acquisition. According to Segalowitz (2005:292) automaticity involves fast processing, and is related to the need for
attention and effort in the acquisition of skills. As one’s skill gradually increases, and moves from non-automatic to
automatic, performance seems to be more efficient, that is, faster, more accurate and more stable, and the amount
of attention and effort needed seems to decrease. As regards language learning, increased performance efficiency
contributes to fluency. Automaticity involves fast processing and effortless production; it is independent of the
amount of information being processed, and is unaware of the process. In contrast, non-automatic or controlled
processes are slow, likely to be inhibited by interfering events, they depend on the information load and involve
awareness.

Segalowitz (2005:304) explains some of the reasons why automaticity can benefit SLA: because automatic processing
consumes fewer attentional resources than does controlled processing, the more automatic performance becomes
the more attentional resources there are left over for other purposes. Thus, for example, if one can handle the
phonology and syntax of a second language automatically, then more attention can be paid to processing semantic,
pragmatic, and sociolinguistic levels of communication. A second reason to favor automaticity is that once a
mechanism becomes automatic it will process information very quickly and accurately, being immune to
interference from other sources of information. This in itself improves the quality of performance. Third, there are
strong reasons for associating automaticity with important … aspects of fluency…. To the extent that fluency
represents the ability to speak or read quickly, accurately, and without undue hesitation, then automatic execution
of certain aspects of L2 performance such as pronunciation, grammatical processing, and word recognition would, by
definition, promote fluency.

The pedagogical implications relate to the way in which automaticity can be developed. Massive practice is one of
the suggestions, though in instructed settings it is no so easily available. Nevertheless, properly organised practice,
especially in the areas of vocabulary and in the learning of chunks, may facilitate the acquisition of syntactic
patterns, and result in more fluent language use.

Attention and memory

Some approaches to SLA place more emphasis on attention than others. Innatist views minimise the role of attention
to a mere trigger of innate knowledge (Schmidt 2001: 6). However, in cognitive accounts, attention and memory are
essential, as language acquisition is largely determined by what learners pay attention to and notice in the input.
Robinson (2005: 472) describes attention as “the process that encodes language input, keeps it active in working and
short-term memory, and retrieves it from long-term memory.” In Schmidt’s account, (Schmidt 2001) attention refers
to several mechanisms, which include alertness, orientation, detection within selective attention, facilitation and
inhibition. Orienting refers to allotting attentional resources to sensory stimuli, which can be modulated by the
“alertness or executive attentional system”. These two subsystems enhance the possibility of detection, the
cognitive registration of stimuli, necessary for processing. Detection may be without awareness, or with awareness,
i.e. noticing. Finally, the attention system may inhibit the processing of irrelevant information, thus facilitating the
processing of relevant information without interference.

According to Schmidt (2001), attention is limited, and is related to two general information processing mechanisms,
namely, controlled and automatic processes. He also advances a variant, which involves specific resource pools for
processing, as required by the tasks (perceptual vs. response), codes of processing (spatial vs. verbal), and modality
(auditory vs. visual). Within this view, activities that demand attention can be carried out simultaneously more easily
if they call upon different resource pools. Another characteristic of attention is that it is selective. Since attention is
limited, it must be strategically directed. It is argued that in SLA, attention is first directed towards meaning, and only
then to the formal aspects of language. A third feature of attention is that it is subject to voluntary control. This is
particularly relevant to teaching, since one of its functions is to help learners focus their attention. However, there is
also a type of involuntary attention, which may interfere with what is being attended to, which is data-driven, while
voluntary attention is “directed to outside events by inner intentions” (Schmidt 2001:14). Moreover, attention is
necessary for the control of action. This is relative to controlled/automatic processes, in that the more automatic a
process is, the less attention it requires. Finally, it is argued that there is very little learning without attention, as
unattended stimuli remains in short-term memory for only a few seconds. As Schmidt states, (2001: 23)

Because we know that attention can be involuntarily attracted to stimuli, it cannot be claimed that learners must
intentionally focus their attention on each particular aspect of the L2 input in order to learn it. Even if it is true that in
order to learn anything one must attend to it, that does not entail that it is necessary to have either the intention to
attend or the intention to learn. On the other hand, we know that preparatory attention and voluntary orienting
vastly improve encoding…and since many features of L2 input are likely to be infrequent, non-salient and
communicatively redundant, intentionally focused attention may be a practical (though not theoretical) necessity for
successful language learning.

The role of memory in language acquisition is mainly connected to topics such as implicit learning, the effect of
short-term memory on SLA, and the accessibility of information in stored in long-term memory. Short-term memory
(STM) has a limited capacity, decays fast, and is more easily changed. On the other hand, long-term memory (LTM)
has unlimited capacity, stored information persists over time, and is not easily modified. Robinson (2005: 487)
reports that most memory researchers claim that short-term memory is “that part of long-term memory in a
currently heightened state of activation, and further, that awareness and working memory are isomorphic, and
correspond to the contents of short-term memory which are within the focus of attention”. Whether STM is or is not
a part of LTM, information that has been detected can enter STM and access information already stored in LTM
without awareness. For new information to enter LTM, and be learned, it has to first enter STM and be rehearsed
there. Rehearsal can be of two types, maintenance rehearsal, which requires data driven, instancebased processing,
and elaborative rehearsal, requiring conceptually driven, schema-based processing. Noticing is then, the result of
rehearsal, which may send information from STM to LTM. To conclude, both attention and memory are important
constructs in SLA. A fuller account of them is out of the scope of this work, but their relevance will be taken up later
in the discussion of learning theories.

Strategies

Strategies are internal processes that account for how learners deal with input data and how they use their
resources in the production of messages. Arriving at a definition of the concept of strategy is not a simple task, as
different researchers define them differently. Also, there is no consensus as to their classification, as some view
them as general strategies, while others consider there are specific language learning strategies. Another source of
disagreement relates to whether language is acquired implicitly, without conscious effort. If this is the case, there
would be no place for strategies in language acquisition. In Cook’s words (Cook 1993:136), “The concept of learning
strategies…goes against the belief that language knowledge differs from other forms of knowledge and that second
language acquisition therefore differs from the acquisition of other forms of knowledge”. What is clear is that while
other learner processes may pass unnoticed, the use of strategies springs from learners themselves, and therefore
give learners choice over their actions. Cook (1993: 113) classifies strategies into learning and communication
strategies. The first refer to the learners’ attempt to learn an L2, while the second refer to their attempts to solve
communication problems.

Learning strategies

O’Malley and Chamot (in Cook 1993: 113) define learning strategies as “the special thoughts or bahaviours that
individuals use to help them comprehend, learn, or retain new information”. They sub-divide learning strategies into
metacognitive, cognitive and social-mediation strategies. (The lists of examples provided are by no means
exhaustive.)

a. Metacognitive strategies
They allow for planning, monitoring or evaluating the success of a learning activity. They are strategies about
learning. Here are some examples:
●selective attention: deciding in advance to pay attention to specific parts of the input;
●self-management: trying to maximise the conditions for learning (e.g. sitting near the teacher);
●advance preparation: planning the language needed for an upcoming task;
●self-monitoring: checking one´s performance as one speaks;
●self-evaluation: checking one´s performance against one´s own standards.

b. Cognitive strategies

They manipulate the input so as to enhance learning. They operate directly on language. For example, repetition,
resourcing (using language reference materials), translation, note-taking, elaboration (relating new information to
already existing concepts), summarising.

c. Social mediation strategies

Also called social/affective strategies, they involve interaction with other people, or control over one’s
affective states. Some examples are cooperation (working with fellow-students), questions for clarification,
and self-talk (boosting one’s confidence to do a task more successfully).

Communication strategies

The study of communication strategies has tended to focus on the strategies used by learners to overcome
communication problems. Cook (1993) presents two camps that have studied communication strategies:
sociolinguistic researchers, who look at strategies in terms of social interaction, and psycholinguistic researchers,
who view strategies as psychological processes.

The sociolinguistic perspective, represented by Tarone (in Cook 1993: 120-122), defines communication strategies as
“mutual attempts of two interlocutors to agree on a meaning in situations where requisite meaning structures do
not seem to be shared”. These are some examples:

● Avoidance: the learner avoids the communication problem by avoiding the topic, or abandoning the message.
● Paraphrase: the learner uses circumlocution, approximation or coins a word to compensate for his/her lack of an
L2 word.
● Conscious transfer: transfer from L1 by means of literal translation or language switch.
● Appeal for assistance: a direct appeal for help.
● Mime: using the body to convey meaning.

The psycholinguistic view focuses on the individual’s response to a communication problem, rather than on a joint
response. This stance defines communication strategies as “potentially conscious plans for solving what to an
individual presents itself as a problem in reaching a particular communicative goal” (Faerch and Kasper, in Cook
1993: 122-124). These researchers claim that learners have two options: achievement and avoidance.
a. Achievement strategies These strategies help the learner to solve his/her problem. They may be of two types:
Non-cooperative strategies:

● L1/L3 strategies: code-switching or foreignising;

● Interlanguage strategies: based on the evolving IL, such as substituting items, generalisation, description,
exemplification or word-coining;
● Non-linguistic strategies: miming, imitation.

Cooperative startegies:
● Direct appeal for help.

b. Avoidance strategies The learner either avoids a linguistic form (phonological, morphological or grammatical), or
avoids a language function by, for instance, abandoning a topic.
As can be seen, the study of strategies has concentrated on different aspects of their use, as there is disagreement as
to whether they are conscious and intentional or subconscious, or as to whether they have a direct or indirect effect
on SLA. Ellis (1994: 532) advances the following list of characteristics as a kind of summary: 1. Strategies refer to
both general approaches and specific actions or techniques used to learn an L2. 2. Strategies are problem-orientated
– the learner deploys a strategy to overcome some particular learning problem. 3. Learners are generally aware of
the strategies they use and can identify what they consist of if they are asked to pay attention to what they are
doing/thinking. 4. Strategies involve linguistic behaviour (such as requesting the name of an object) and non-
linguistic (such as pointing at an object as to be told its name). 5. Linguistic strategies can be performed in the L1 and
in the L2. 6. Some strategies are behavioural while others are mental. Thus some strategies are observable, while
others are not. 7. In the main, strategies contribute indirectly to learning by providing learners with data about the
L2 which they can process. However, some strategies may also contribute directly (for example, memorization
strategies directed at specific lexical items or grammatical rules). 8. Strategy use varies considerably as a result of
both the kind of tasks the learner is engaged in and individual learner preferences.

This last point is related to studies into strategy use. Several factors seem to affect the use of strategies. As regards
age, for example, younger learners tend to use simpler strategies than older learners. According to Ellis (1994),
language aptitude does not seem to be strongly related to strategy use, and while it is likely that there is a relation
between strategy use and learning style, there is little research to testify to it. As regards motivation, Ellis (1994)
refers to studies that found that highly motivated learners use more learning strategies, and that the learners’ goals
are likely to determine the choice of strategy. Personality types also affect strategy use, as it is expected that
extrovert learners will make more use of social/affective strategies. However, these links have not been
demonstrated yet.
Situational factors also affect strategy use. Studies report that social strategies are rare in classroom settings, except
for “requests for clarification”. Task type can also influence the use of strategies. Ellis (1994: 544) reports on studies
that showed that students tend to use more strategies for vocabulary learning tasks and oral drills, while the lowest
frequencies were for listening comprehension, inferencing, making oral presentations and engaging in operational
communication. In short, strategy use is influenced by a variety of factors, both internal and external to the learner.
It is also amenable to change through training, therefore teachers can guide learners into discovering which
strategies suit their styles and task types better.

Related to the study of learning strategies are those on the “good language learner”, that is, learners who achieve
high levels of competence. Stevick (in Ellis 1994: 546-550) identifies an “overall pattern”, in that successful learners
have a concern for language form, they search for meaning in the L2 data they are exposed to, and try to engage in
real communication. They also engage actively in language learning, as they prefer to manage their own learning
process, of which they are aware. Finally, good language learners are flexible in their strategy use, as they can
choose them in accordance with task requirements. A downside of a very successful use of strategies is that it may
interfere with acquisition. It is not infrequent to find a student who is so skillful in making up for his/her lack of
linguistic knowledge that he/she will not need to form or test hypotheses, thus making learning more difficult. Also,
it is not clear yet which aspects of SLA are benefited by learning and communication strategies, namely both lexis
and grammar, or just lexis. Anyway, they can be a powerful aid in the process of language development, and as they
are sometimes conscious, and give learners power over their leaning processes.

4.1.3. L2 knowledge

The concept of Interlanguage (IL) was developed in the late 1960’s and early 1970’s as researchers found out that
learners’ errors were not random, but systematic (White 2005: 17). IL is systematic in that, at all stages of
development, learners operate according to an internalised mental grammar. IL is, then, a natural language system
which is independent of both the learner’s L1 and the target language (Ellis 1985: 299), and which may show some
features of the L1 and some characteristics which are very general and tend to occur in most IL systems (Lightbown
and Spada 1993: 55). It should be clear, then, that interlanguage is not an impoverished version of the target
language, but a language on its own. Learners follow similar routes in their L2 acquisition, irrespective of factors such
as their L1, age, learning situations, or individual differences, although there are minor differences as regards the
order in which some features are learnt. Interlanguages are very dynamic and evolve as learners receive input and
revise their hypotheses and make their L2 system more and more complex.

The analysis of learners’ errors has contributed to the understanding of SLA, and has shown that L2 learners pass
through certain sequences of development which are similar to the ones that children go through when they are
learning their L1. Research has revealed the existence of developmental errors, which reflect the learners’
internalised grammar of the target language, and the existence of developmental sequences, which are predictable.
It is important to bear in mind that learners do not necessarily leave one stage before entering another, as at any
given point learners may use language forms that are typical of various stages. According to Lightbown and Spada
(1993: 67),

It is perhaps better to think of a stage being characterised by the emergence and increasing frequency of a particular
form rather than by the disappearance of an earlier one. Even when a more advanced stage comes to dominate in a
learner’s speech, conditions of stress or complexity in a communicative interaction can cause the learner to ‘slip’
back to an earlier stage.

Of course, the development of target-like rules is gradual, with overlaps, and there is variability. For example, some
learners may take longer to pass certain stages. Spanish-speaking learners, not surprisingly, tend to spend longer on
the first stage. Also, learning settings may affect the rate of development, as language instruction may accelerate the
speed with which some learners pass through some stages. Although research on developmental sequences has
concentrated only on a few structures, the results suggest that the sequence of acquisition may be built-in in the
learner’s mind. Littlewood (2004: 530) adds that the same sequences have been found in the spontaneous speech of
classroom learners who have been taught the correct target forms. This implies that the “internal syllabus often
overrides the external syllabus which the teacher or coursebook tries to impose.” Some explanations are suggested.
For example, one form is more likely to be learnt if it is supported by transfer. Also, a form may be acquired more
rapidly if it is more frequent or more salient in the input. Littlewood invokes Pienemann with his explanation that
some structures may be governed by psycholinguistic processing constraints, “so that the achievement of one stage
is a prerequisite for achieving the next” (2004: 531). Given the fact that IL contains errors, and that, as stated before,
at any stage of development, the learner’s IL will contain several competing rules, another characteristic of the
interlanguage system is variability. As separating L2 knowledge from its use is highly complicated, variability will be
dealt with below, together with output.

4.2. Output

The final component of the process of language acquisition in this approach is output. Traditionally, output was seen
as a way of practising already existing knowledge, allowing for fluency. Krashen (1982), in his Input Hypothesis,
claimed that the only positive role for output is to allow learners to get more input. However, most researchers
nowadays argue that output is a requisite for L2 acquisition. Output may be taken as the external manifestation of a
learner’s interlanguage. As mentioned before, output and IL are inextricably linked and, therefore, it is difficult to
consider them apart. One characteristic of IL, and of output as well, is variability. Variability is evident when learners
use two or more linguistic forms for realising a grammatical structure. However, this variability is not always random.
There are two kinds of variation: systematic and non-systematic, or free.

In Ellis’s (1997) view, the choice of alternate forms in systematic variation is determined by the following factors: 1.
Linguistic context: one linguistic form may trigger the use of another form. For example, learners may use full ‘be’,
sometimes contracted, and sometimes omit it entirely, target-like variants being more common with pronoun
subjects, ‘be’ being omitted with noun subjects. 2. Situational context: learners tend to use more correct forms when
the situation requires the use of more formal language. 3. Psycholinguistic context: whether learners have the
opportunity to plan their production. Learners tend to use more accurate forms when they can plan their output.

Non-systematic variation, on the other hand, is the use of alternate forms in free variation. This is more common in
the initial stages. According to Ellis (1997: 29), learners begin by acquiring a single form, which they use for a variety
of functions. Later, they acquire other forms and use them interchangeably with the initial one. Then, they start
using the forms systematically, target-like forms in planned discourse, until they become consistent. From a
variationist point of view (see below for a description of this approach), Tarone (in Ellis 1985: 83) presents her
description of the effects of the situational context as a continuum of IL styles. At one end of the continuum is the
vernacular style, which is called upon when the learner is not attending to his speech. This is the style that is both
most natural and most systematic. At the other end of the continuum is the careful style, which is most clearly
evident in tasks that require the learner to make a grammatical judgement… The careful style is called upon when
the learner is attending closely to his speech.

In between there are other styles which also vary according to the degree of attention required by the language
tasks learners are engaged in. The vernacular style is the most stable and consistent, as it is not subject to the
influence of other knowledge sources. When learners are engaged in unplanned communication, they tend to fall
back on language features learnt in the early stages of acquisition. On the other hand, the forms present in a careful
style are likely to be highly variable, according to the degree of monitoring. This distinction is fundamental in
instructed learning settings, since students’ performance will vary according to the style required by the tasks they
engage in. Thus, when they are performing communicative tasks without preparation, they will tend to use the
vernacular style, which reflects the current state of a learner’s interlanguage. When planning time is available,
learners will tend to use a more careful style, and monitor their speech. This will allow for correctness, and also for
attention to be paid to their speech, which, in turn, may foster noticing and consequent language development.

According to Ellis (1985: 94), also following a variationist perspective, variability may play a role in language
development. In relation to systematic variability, language development consists of the “gradual extension of
regularities from formal to progressively informal styles on the one hand, and from simple to increasingly complex
linguistic contexts on the other. In this sense, then, SLA involves a gradual reduction in the degree of variability as
non-target language variants are eliminated in a steadily growing range of environments. This reduction represents a
shift to the norms of the careful style.” Another role ascribed to variation is psycholinguistic in nature, and is related
to learners’ capacity to attend to formal features in their speech. As the “careful forms are used, they are practised,
with the result that they become more automatic, require less attention, and so are available for use in
interlanguage styles nearer the vernacular end of the continuum. Thus styles of communication initially associated
with planned discourse are eventually accessible in unplanned discourse, provided that sufficient practice has taken
place.” (Ellis 1985: 95)

On the other hand, free variability plays a different role, which is related to the fact that it is the major source of
instability. According to the economy principle of linguistic organisation, the presence of two forms in free variation
is not efficient. The economy principle makes sure that the new forms are “either integrated into the system by
putting them to work to distinguish meanings, or that they are eliminated.” (Ellis 1985: 95) Following this principle,
learners will progressively sort out forms into functional slots in order to solve free variability. Summing up, language
development, seen from this perspective, involves two processes: first, re-arranging form-function relations to
maximise effectiveness, thus making free variation gradually become systematic. Second, moving forms from careful
styles towards the vernacular style, which is the primary objective of most, if not of all, learners.

The output hypothesis, developed by Merrill Swain (1995: 125), in direct opposition to Krashen’s view, claims that
“producing language serves second language acquisition in several ways.” Output makes learners process language
more deeply, and play a more active role in their learning process as well. Learners may then move from the
semantic processing typical of comprehension to the grammar processing needed for accurate production. This
potential role of output in fostering the development of syntax and morphology underlies the three functions
ascribed to it by the output hypothesis.

1. Noticing: is a consciousness-raising function: “in producing the target language, learners may encounter a
linguistic problem leading them to notice what they do not know, or know only partially.” (Swain 1995: 129)
Thus, producing the target language may make learners aware of a language item they need to learn or
consolidate.

2. Hypothesis testing: errors in learners’ IL reveal hypothesis they hold about the target language. When learners
engage in communication and need to negotiate meaning, according to a study reported by Swain, over one-third of
their utterances are modified either semantically, syntactically or morphologically. Even though there is no proof yet
that the modified utterances are retained in the learner’s IL, the assumption is that the process contributes to
acquisition, as this modified output may be considered the leading-edge of a learner’s IL.

3. Conscious reflection: The third function of output is a metalinguistic one, that is, using language to reflect on it, by
means of tasks that, while being communicative, focus on language form. An example of such a task is that of
collaborative work, in which learners are led to reflect on their production as they attempt to create meaning.
Conscious reflection is also important in that it creates the conditions that are necessary for the other two functions
to take place.

To conclude, the value of understanding the nature of output should not be underestimated. From the teacher’s
point of view, output is more than the external realisation of the learners’ interlanguage; it is also a means for
language development. From the learners’ viewpoint, it is also necessary for them to understand the nature of their
production, so as to avoid the anxiety that may be brought about by their errors, which may result in their losing
motivation. Also, it is important for learners to be able to detect changes in their output, and to become aware of
their development. Output may serve as a starting point for further work, in that it provides teachers with data to
work on. From the learners’ point of view, provided they are conscious of their linguistic production, output supplies
them with a measure of their achievement, and with opportunities for further development.