UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL

IEXPRO

CHAPTER 3
INPUT, INTERACTION AND CONTEXT

3.1 INTRODUCTION
In the previous two units we considered language and the learner from the point of
view of internal processes. In this unit we will be broadening our view of interlanguage
development by taking into account the learner's linguistic interaction with the environment,
and how this can help or hinder L2 acquisition.
Firstly, we will be focusing on the role of input in learning an L2, because of all the
factors that play an important role in foreign language learning, input - the L2 that the
individual language learner is exposed to - is considered to be a fundamental element
in the process. Input alone, however, is not enough. We will also be looking at the role of
interaction with other speakers of the L2 in SLA, and the learner output which this
encourages. Finally we will be examining the importance of context in learning the L2,
specifically focusing on the difference between natural and classroom contexts, and the
role of instruction in helping a learner to acquire the L2.

3.2 INPUT AND LEARNING

Most studies which have examined the role of input in second language learning are in
agreement on the fundamental importance of this element in the process of learning. By
way of example we can cite works by Kasper (1986), Hyltenstam and Pienemann (1985),
Perdue (1984), Spolsky (1989), Scarcella and Higa (1981) and above all, Krashen (1985),
whose work we examined in the subject Second Language Acquisition. Kasper, for example,
points out that many studies influenced by Piaget's philosophy, agree in granting
context, input and negotiation an importance equivalent to the individual's internal
factors. On the other hand, Tollefson (1981) in his study on the variables that affect the
acquisition of an L2, decides to adopt a model that includes input, the individual, the
learning process and what has been learnt. But it is Krashen who goes a step further
with his input hypothesis as we will see.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
The first question is to try to define the features of input which promote learning. A
fundamental condition, according to several authors (for instance Krashen and Spolsky)
is that input has to be sufficient in terms of quantity, because, as Spolsky has put it,
The more time spent learning any aspect of a second language, the more will be
learned.
(1989:166)

Thus it would seem that the more the learner is exposed to the target language, the
more he/she will learn. This has obvious consequences for those of us who work as
English teachers in an EFL (as opposed to an ESL) environment: our learners tend to
have very limited exposure to L2 input. Two direct consequences that derive from this
limited access to the target language are the appearance of interferences of the mother
tongue on the one hand (Van Els et al., 1984) and the tendency towards fossilization on the
other (Perdue 1984). You will remember that we mentioned the latter implication of restricted
input on IL development in the first unit.
However, as the above-mentioned authors also state, a large quantity of input is not enough
to ensure successful SLA. We also have to take into account the kind of input the
individual is going to be in contact with. As far as the quality of input necessary for SLA
is concerned, Krashen gives a detailed explanation.
Krashen considers that human beings acquire language in only one possible way: by
understanding messages, or receiving what he calls comprehensible input. You will
remember from the subject Second Language Acquisition that Krashen defines
comprehensible input as i + 1, that is, input which is just beyond the learner's current
level of ability. By being exposed to comprehensible input, according to Krashen, a
learner can move from his/her current level (i), to the level i + 1, which is the next
level up in the natural order of acquisition. In other words, to progress in the knowledge
of the L2 it is necessary, according to Krashen, that the input the individual is exposed
to is at a slightly more advanced level than his/her present one. We do not intend to get
into a full-scale discussion of Krashen's theories here (for this, see, for instance, McLaughlin,
1987) but we do need to emphasise the obvious problem of defining both i and i + 1.
In other considerations of the role of input in SLA, some authors have considered the
possible importance that the frequency of certain forms in the target language can have
for input processing. Frequency, as Dulay, Burt and Krashen (1982) have put it, is the
number of times the individual hears or sees a specific target language form.
Theoretically, frequency would increase the probability of appearance of those structures
the individual is "ready" to process, increasing at the same time the possibilities for the
individual to process them.
However, studies carried out by Dulay and Burt (1974) based on the incidence of the
frequency of morphological and syntactical structures, show the complexity of this
phenomenon. While the production in learner output of certain structures and not others -in
spite of the same frequency of input- were observed, other structures that were not even
present in input were also detected in output. The problem with the results of this study
is that the specific stage of development of each individual learner's interlanguage was
not taken into account. Thus, despite the frequency of the forms, an individual learner
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
would only process those his/her interlanguage is ready to assimilate. Due to the
limitations of studies of this sort, the implications of the phenomenon of frequency in input
are still unclear and its effects difficult to delineate with precision. More recent studies,
however, such as that by Schmidt and Frota (1986) discussed in the last unit, do emphasise
the importance of frequency in input as an important variable in successful SLA.
Thus, it can be said with a reasonable amount of certainty that both the quality and frequency
of input, are fundamental features necessary for input to be assimilated by the individual
and turned into part of his/her interlanguage.

3.3 INPUT AND INTAKE

We will now focus on another question, posed by Corder (1981): that of the input that
is actually processed by the individual learner during a specific moment of the learning
process, namely, the intake. According to Corder, for input to be transformed into intake:
- The language data needs to be presented in a manner which will stimulate
its possible assimilation.
- The content of the language data needs to be compatible with the individual's
previous linguistic knowledge.
In short, it would appear that the individual needs exposure to a specific type of input if
language acquisition is to be facilitated. The question is, what should this input be like? We
will attempt to answer this question below.
Although, as we have seen, it is difficult to make an ideal characterisation of the
appropriate language data to include in input, we know at least one of the features
it has to include: the input has to be understood by the individual. Starting from the
comprehension of the message, the individual’s language acquisition machinery will start
incorporating new elements that are contained in the message. In other words, learners
need to understand the overall meaning of a message to then be able to focus more
closely on the actual linguistic forms used to convey that message. As Van Patten
concludes:
... only when input is easily understood can learners attend to form as part of the
intake process.
(1990:296)

Thus, despite the general rejection of some of Krashen's theories (such as his
acquisition/learning distinction), the overall thrust of his theory of comprehensible input
has received widespread approval. It should be noted in passing, however, that many
theorists also disagree with the specifics of this theory -for example, Krashen's claim
that input far outweighs output in terms of importance.
Another feature that affects input being transformed into intake and making learning
possible, is that it should be specifically addressed to the individual. That is to say,
personalised input (Seliger 1983) will be more effective than input that is not individually
addressed to the listener.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
A further characteristic of input that has been proposed to aid SLA has been proposed
by, among others, Schachter (1983). This is the advisability of simplifying input, the
argument here being that simplification makes it easier for the individual to recognise
the units of form present in the input, and to observe how they are combined and used.
Knibbeler (1989) even asserts that a simplified code is more useful than a native or near
native model. Research has shown that most EFL/ESL teachers do in fact modify their
speech automatically.
It has been pointed out that, mainly in the initial stages of the learning process, it
would be advisable to use concrete referents (what has been called the "here and
now" principle) to facilitate the understanding of the message for learners. What is more,
it is also important that input is received in a positive atmosphere in order to foster a
favourable affective disposition in the individual. This, according to Ellis (1985) will increase
the likelihood of input being transformed into intake.
Let us summarise this section. Research points to the following factors as having a direct
bearing on the transformation of input into intake:
• Comprehensible input, in terms of both linguistic forms and ideational content.

• Personalised utterances.

• Linguistically simplified language.

• The use of concrete referents.

• A positive learning environment.

3.4 INTERACTION
In the first part of this unit we have looked at the learner as if he/she were a passive
recipient of input. As we know, learning a language includes far more than merely
receiving input, as we also produce language. In this section we will be considering the role
of the individual during his/her contact with the target language, that is, the role of
interaction. The main debate as far as interaction is concerned is the extent to which
being a receptor is enough and the extent to which being a participant in conversation
is a necessary element for successful SLA. To consider this question we will be dividing
this section into two parts, which reflect the process which the individual goes through
during interaction: first we will look at learner output, then we will look at how this
may affect the interlocutor and the subsequent adjustments the learner may need to make
in further output -the process of interaction
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
3.5 OUTPUT
For some researchers (e.g. Krashen, Dulay and Burt, among others) an initial silent
phase, or silent period, when the individual starts his/her learning process seems to
favour the pace and quality of acquisition. That is, they propose an initial exposure to
input -with no time limited- which is merely receptive, before encouraging the learner to
try to produce messages in the second language. We will not discuss here the
implications of this silent period (see the treatment of methods in the subject
Methodological Approaches) because, although it is generally accepted that it can have
positive effects, many factors, such as the individual's personality, his/her communicative
needs in certain situations, and so on, can mitigate against an initial silent period. The
main question is, what happens afterwards?
It is here that we find some of the most severe criticisms of Krashen’s theories. According
to Krashen, producing the language is not necessary for acquisition - listening (to
comprehensible input) is enough. This opinion is contrary to Swain's, who asserts in her
Output Hypothesis that the most important element in the learning process is
comprehensible output: receiving and understanding messages is not enough- learners
have to produce them in order for acquisition to take place. Swain (1985) showed that
children who had received immersion education attained high levels of comprehension
which were not matched by a similarly high level of production. Swain suggests that the
chances of successful language acquisition are higher when the individual has the
opportunity of organising his/her output during interaction with an interlocutor.

Swain (1995: 125-126) makes the case for output thus:

First, it is hypothesized that output promotes “noticing". That is to say, in producing
the target language (vocally or subvocally) learners may notice a gap between what
they want to say and what they can say, leading them to recognize what they
do not know, or know only partially. In other words, under some circumstances, the
activity of producing the target language may prompt second language learners to
consciously recognize some of their linguistic problems...
A second way in which producing language may serve the language learning process
is through hypothesis testing. That is, producing output is one way of testing a
hypothesis about cornprehensibility or linguistic well-formedness...
Thirdly, as learners reflect upon their own target language use, their output serves
a metalinguistic function, enabling them to control and internalize linguistic
knowledge. My assumption at present is that there is theoretical justification for
considering a distinct metalinguistic function of output.
Output thus may be said to presuppose interaction. However, as we shall see, there is some
debate about exactly which type of interaction produces the kind of output which facilitates
language acquisition.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
3.6 NEGOTIATION
Related to the production of messages in a second language and the process of
acquisition, the situations in which the learner has to make his/her output
comprehensible are of special interest. This happens when the interlocutor indicates that
he/she does not understand the original message that the learner wanted to communicate.
Faced with the need to modify his/her output, the learner, it is hypothesised, will try
new structures and forms and will exploit his/her interlanguage to the full in the effort to be
understood. Moreover, the resulting output will serve the purpose of generating new
messages addressed to the learner, that is, further comprehensible input will be supplied.
This element of negotiation that appears with the conjunction of input and output leads us
to a fundamental part of the learning process: interaction.
Before examining interaction in more detail, it should be noted that, although many
studies stress the impossibility of acquiring a language without interaction, i.e. only through
received input (television, for example) there are isolated examples that contradict this
affirmation (e.g. One of the subjects in Naiman et al.’s 1978 study). What is clear is that
these “special" cases need to be subjected to further analysis for precise conclusions to be
drawn.
Nevertheless, despite the above caveat, the importance of interaction for the acquisition
of a second/foreign language is proposed nowadays by a great many studies. Several
theorists (for example Hatch 1983; Seliger 1983; Gass and Varonis 1989; Van Patten
1990, and Doughty 1991, among others) give interaction a predominant role in SLA,
and claim that its benefits can be listed as follows:
- Interaction allows the learner to practice his/her acquired linguistic knowledge
through output.

- The opportunities for trying to produce new morphological and syntactic

structures are increased, as are the possibilities for experimenting with and
hypothesising about language.

- Interaction generates input, procuring in this way more data about the target
language for the learner.

- Interaction facilitates the obtaining of personalised input, which increases the

individual’s attention to the message.

- Because of the communicative nature of interaction, it is adapted to the

learner's linguistic level, being modified or simplified when the understanding
of the messages requires it.

- Interaction enables the learner to obtain "negative" input (Schachter 1983):

whenever the attempt to communicate appears to be ineffective and the meaning
of the learner's message is unclear to the addressee, the learner necessarily
receives negative input which informs the speaker about the existing
communication problem. If this kind of input were not present, the result would
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
be the fossilisation of certain forms that do not belong to the target language
code.

- Interaction enables the acquisition of certain sociolinguistic and discursive

features inherent to communication.
In view of the importance attributed to interaction we are justified in asking to what
extent interaction in the target language affects the learning process in terms of both "quality"
of language and "speed" of acquisition. Although we are as yet far from having a
conclusive answer to this question the indications from the studies suggest that learners
who are involved in more interactions show more proficiency sooner in the L2, and they
have a more "mature" error profile - that is, their interlanguage shows less mother tongue
influence.

It must be pointed out though that more studies are needed if any conclusive evidence
is to be produced. In the meantime many researchers and many teachers proceed on the
assumption that interaction promotes acquisition. One voice urging caution in this field is
that of Pauline Foster. Foster (1998) suggests that some of the claims made for the
negotiation of meaning need to be looked at again. Her own study which, unlike some
of the seminal research in this particular field, was carried out in actual language
classrooms rather than under laboratory conditions, suggests that students do not in
fact engage in the negotiation of meaning in task execution where it was hypothesised
that they would. This, she suggests, has implications for the tasks we devise - a point
we will return to later in this unit.

3.7. CONTEXT
We have seen that interaction is widely considered a fundamental and necessary
feature of successful second language acquisition for the vast majority of learners.
We will now turn our attention to what it is that marks out the limits of interaction: the
context where the exposure to input takes place. The importance of context in
providing the social and cultural factors which may affect the learner have already been
covered in the subject Individual Factors, and we will not be reiterating these here,
although we will be making a distinction between formal and informal contexts.
A classification of the different settings where second language acquisition can take place
can be found, for example, in Berman (1984) who distinguishes between informal (natural)
and formal (mainly the classroom) settings. This is a distinction which all the researchers
who approach this topic take into account, as they try to enumerate the features of
each of these contexts and their possible advantages and disadvantages with the aim
of ascertaining which context is more favourable -and why- for second language learning.
Although what mainly concerns us as EFL teachers is the learning that takes place in
the classroom, we will also examine some of the features of natural contexts that affect
second language learning.

First of all, it would be useful to recap on the fundamental differences between both contexts.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO

TASK 3.1.
Consider the main differences between the learning that takes place in a formal (classroom)
context and a natural context. Use the grid below as a template for your answer.
FORMAL, CLASSROOM CONTEXTS NATURAL CONTEXTS
Input
Physical constraints
Output
Interaction

Before we look at the specific features of input that can be observed in the classroom,
we need to bear in mind, as we mentioned before, that the natural context can have an
important positive effect on the acquisition process. Dulay, Burt and Krashen note that
natural contact with the language:
triggers the subconscious acquisition of communication skills in that language.
(1982:15)
although they also admit that there are many factors that can limit its positive effects,
such as the limited possibilities of establishing communication with second language
speakers, difficulties in communication or absence of a silent period in the initial stages
of the learning process. However, in spite of the possible limitations that natural contexts
may have, it is obvious that both the possibilities of obtaining input and the opportunities
for interaction using the second language are highly superior; as such natural contexts
are those that offer the richest opportunities for successful language learning. However,
this does not mean that teaching in these contexts is not also necessary. As we will see,
there is evidence that points to the effectiveness of instruction in second language
learning in both natural and formal contexts.

3.7.1 THE FORMAL CONTEXT

Formal contexts, as we established above, refer to classroom-type situations where there
is an explicit focus on the conscious learning of rules and grammatical forms. Although
the communicative potential of formal contexts is very limited, they also present certain
advantages. For example, from a psychological point of view, Dulay, Burt and Krashen
(1982) have noted that formal instruction can have a positive effect on adults, in the
sense it makes it possible for them to check their progress in the learning process against
the 'objective' yardstick of grammatical forms covered by a syllabus.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
After reviewing the studies that have addressed this matter, Long (1988) concludes that
formal instruction has positive effects on the process of acquisition of a second language
both from the point of view of the pace or speed with which individuals learn the language
- also corroborated by Doughty (1991) - and their final competence. As regards this latter
point, it is even suggested that it would be impossible to reach a similar level to that
of the native speaker without instruction. However, certain reservations need to be
expressed on this point. We have seen, for example in previous subjects, that factors
such as the language spoken outside the classroom and the age of the individual, among
others, need to be taken into account when considering what makes for successful SLA.
Long also notes that instruction does not seem to affect the order of acquisition of the
different elements of the language, except for during limited periods of time and "in trivial
ways" (1988: 135).
A fundamental issue that affects the classroom context is that language is an instrument
for communication in the natural context, whilst in the formal one it becomes the
objective of learning (Seliger 1983): the content of communication and the form of the
language are the respective main focus in each context. In other words, the focus is
primarily on meaning in the natural context, while it is on linguistic form in the formal
context. However, the problem lies less in the formal context itself than in the way this
context is used. Spolsky (1989) mentions that formal learning can offer many opportunities
to observe and practise the new language, while researchers such as Krashen (1988),
Ellis (1985) and Seliger (1983) insist on the idea that the classroom could combine both
a formal and a natural focus, creating both a focus on the targeted language forms, and
"social context" where language is also used for communication.
The importance that the classroom can have in the learning process is highlighted by
Krashen (1988), who notes that if the second language is employed in a "realistic" way,
the classroom can tap into the advantages that natural contexts offer and make
learning and acquisition occur simultaneously. Furthermore, Seliger (1983) asserts that
language lessons offer more advantages to individuals than natural contexts because
in the former they feel "obliged" to use the second language, not only in a formal sense
but in order to communicate as well. Real communicative uses of language, such as
solving problems, making comments or chatting, demand the 'realistic' use of the
second language, combining in this way both a formal and a communicative dimension
in a single context.
It is precisely this aspect - the consideration of the classroom as a space for social
interaction - which polarises most views on the study of context. There are theorists,
such as d' Anglejan (1978) for example, who emphasise that the language produced in
the classroom, based on linguistic entities, can very rarely be likened to the communication
that takes place outside.
However, one problem with studies carried out in formal contexts is that the setting is
often an ESL environment. This has several implications: firstly, learners tend not to
share the same mother tongue, and thus have to resort to the second language to
interact (as opposed to the case of EFL contexts where there is a common mother
tongue), and secondly, outside the classroom, the ESL environment will offer almost
limitless communicative possibilities for the learner to engage in, in the target language.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
An interesting question is, therefore, what happens in EFL contexts such as ours, where
there is a common mother tongue, and very limited exposure, if any, to the target
language outside of the formal classroom context? It is clear that what happens in the
classroom is fundamental in understanding the process of learning in EFL contexts.
lf it is accepted that the role of interaction in the process of second language acquisition
is essential, then it would seem that SLA in an EFL context may well be doomed to
failure. Learners may decide to interact only in the L1 due to sharing a common mother
tongue, and even if they do decide to interact in the L2, meaningful interaction may in
any case be curbed by the activity types typical of the formal classroom. But despite
these not inconsiderable drawbacks, learners do learn the L2 in formal EFL classrooms,
although obviously to varying degrees of proficiency.
Thus the hypothesis that interaction in the classroom is the single most important variable
in formal contexts (Ellis 1985) must be interpreted cautiously, because, although it is
possible that the degree of participation in interactions in the target language is related
to "success" in learning, this could also be due to a series of other, interrelated factors.
Obvious candidates would be the degree of motivation a learner possessed, individual
aptitude and learning style, the specific social context and so on (see the subject
Individual Factors for a review of the main factors which influence SLA).

3.7.2 ANALYZING THE FORMAL CONTEXT: RESEARCH IN THE

CLASSROOM
Recent years have seen a great deal of research into what actually happens in the
formal classroom context. This move to observe, document and analyses classroom
behavior has been motivated by the desire to improve pedagogical practice. If we can
analyses what happens, and how this affects learning, we may be able to judge what is
effective (and what is ineffective) teaching, or so the reasoning goes.
The roles played by input and output - in other words, interaction - in promoting
successful learning have been examined through what has been called classroom
process research (Gaies 1983). Three main research areas have been focused on:
• Teacher talk;
• Interaction analysis;
• Discourse analysis.
We will recap and expand on each of them briefly.

[Link] TEACHER TALK

You will recall that the "special" language used by teachers in the classroom, called
teacher talk, has been considered a fundamental element for analysis in the formal
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
classroom context. This is hardly surprising, given the supposed importance of input and
interaction in the process of second language learning.

The importance of the type of language used by teachers has been studied by many
researchers, and the principles of analysis are very similar to those of foreigner talk studies
-which we will summarise later on-, as both are considered registers with specific formal
and interactive features. It is important to note that the studies carried out in the area of
teacher talk have been conducted both in the context of foreign language lessons (EFL
teaching) and other lessons where the second language was used as a means of
instruction (content teaching). The comparison between both situations did not bring up
any significant differences (Ellis 1985).

In short, what are the features of teacher talk? In the subject Classroom Management we
summarised some of the formal features of teacher talk, in which speech is modified or
adjusted in very specific ways. Here we will be taking a more general theoretical view of
teacher talk, and we will summarise the main features of teacher talk in point form
below.

- Teachers' speech adjustments are produced in a generalised way; they

are a common feature of both experienced and inexperienced teachers (Pica
and Long 1986).

- A teacher normally addresses many learners at the same time. The teacher tends
to modify his/her speech to adopt a "medium" register, that is, a register which
he/she considers to be comprehensible to the majority of learners. This means
that the individual learner is deprived of personalised input, which, as we saw
earlier enhances the possibilities of successful language learning.

- Related to the above point, we find that the elements of negotiation necessary
for the understanding of the message, and which are present in personalised,
one-to-one interactions, are limited in the classroom (Pica and Long 1986). This
could cause problems of understanding - with their obvious implications for
learning - in many individuals.

- Formal speech modifications by teachers can be detected with learners at all

language levels (Hakansson 1986), although the level of complexity of the
language used by the teacher will change according to the individual's level.
That is, the higher the level, the more complex the language used by the
teacher is (Gaies 1977).

- Generally speaking, speech modifications that are 'ungrammatical' are not

observed. Rather, less complexity in aspects such as sentence length,
subordination or expansions of the nominal phrase (Hakansson 1986) are to be
seen.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
[Link] INTERACTION ANALYSIS
lf, as we have seen, the features of language which teachers use can have an important
effect on the learning process, it has also been suggested that teachers are only one
part of the communicative situation. There is also the individual who is learning the target
language, who has a role as interlocutor. Thus an analysis that does not contemplate all
the participants in a communicative act, as well as the features of the resulting interaction
is, by definition, very limited.
Interaction analysis tries to broaden the focus of classroom process research by including
the learner in the equation. We examined the main models used in interaction analysis,
and identified some of the weaknesses involved in this area of research in a fair amount
of detail in Classroom Management, and it is not our intention to reiterate that
information again here. Suffice it to say that the attempts to analyse the uses of language
within the classroom context face the problem that, although they all can be grouped under
the common heading of interaction analysis, the variety of systems and categories used
make any comparison and hence any generalisation impossible.

[Link] DISCOURSE ANALYSIS

Another more focused attempt to analyse classroom contexts is through discourse analysis.
This, as with interaction analysis, takes into account all the participants in the interaction,
although it adds the new consideration of observing not only the language produced by
interlocutors but the way in which this language is combined to constitute more complex
units of discourse.
In terms of classroom discourse we can distinguish between four basic types of target
language use (McTear 1975):
1. Mechanical: there is no communicative exchange.

2. Meaningful: language is contextualised but there is no exchange of

real information.

3. Pseudo-communicative: there is some information exchange but in a

different way from what would happen outside the classroom, in natural contexts.

4. Real communicative: natural and spontaneous language use.

Despite the existence of McTear's four possible variations, the study of classroom
discourse has tended to focus on the forms that interactions normally take. Starting from
the consideration that most foreign language lessons are dominated by the figure of the
teacher, investigations such as the ones by Sinclair and Coulthard (1975), or Sinclair
and Brazil (1982), have noted that classroom discourse usually presents a profile
consisting of three different phases:
1. The teacher’s initiation of the exchange (1).
2. The student’s response (R).
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
3. The teacher’s feedback (F).

To Coulthard and Sinclair’s IRF sequence McTear (1975) adds the possibility of a fourth
phase: when the learner notices that the teacher's intervention suggests or requires another
response.

What, we might ask, are the implications of the organisation of discourse - centred on
the figure of the teacher - into these three (or four) phases for second language learning
in a classroom context?

TASK 3.2
What do you think the effects of IRF exchanges on SLA are? Briefly consider each of the
following areas in your answer:
- Input

- Output

- Negotiation

- The effectiveness of the feedback stage in IRF

Within the field of discourse analysis a fair amount of attention has been paid to the
role of feedback in the learning process. In accordance with the rather more theoretical
bias of this unit, we will now look in more depth at the role of feedback in SLA.

[Link] FEEDBACK AND SLA

According to many of the studies conducted in this area, the role of feedback in the learning
process is quite complex, and it is hotly disputed terrain. Both Chun et al. (1982) and
Seliger (1983) suggest that in informal contexts content is more important than form
and consequently there is little feedback in the form of error correction.
In the classroom, however, the situation is quite different because, as we have seen,
feedback is usually always present, playing in some cases a doubtful role as a
facilitating factor for learning. Feedback addressed to the individual, Seliger (1983) claims,
can promote the individual's revision of his/her interlanguage, by making him/her reformulate
his/her hypotheses about the L2. This begs the question of how useful it isfor the individual
to receive feedback which is not personalised and is addressed to the group.
Researchers such as Vigil and Oller (1976) have argued that feedback constitutes a
vital part of the learning process. They claim that the role of feedback is not merely to
provide positive or negative information about the target language, but that it represents
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
an affective support to conversation, and can be positive, neutral or negative, as it
provides the stimulus for learners' attempts to revise their production where necessary.
As we saw when we considered the role of output in an earlier section in this unit, this factor
has considerable importance in terms of promoting L2 acquisition. Seen in this light, the
value of feedback would seem to lie not only in its formal aspects, but also in its
motivational dimension.
Although there is a need to thoroughly investigate the effects of different types of feedback
on the whole process of language acquisition, it would appear that personalised feedback
which is directly addressed to the individual, offering affective support, is the most useful
and appropriate.

[Link] THE TASK

Another aspect which is important to consider in the classroom context is the role of the task
in channeling interaction. This is because the task determines both the type of language the
individual will receive and the language he/she will produce. It has been claimed that the
most appropriate task type for language learning is one that demands a mutual exchange
of information, and thus promotes negotiation. Young and Doughty (1987), for example,
insist on the fact that this exchange of information among individuals be 'genuine'. This
leads us to the obvious question of what 'genuine' tasks look like.
Studies claim that the tasks most likely to promote negotiation, interaction and hence,
acquisition, are the following:
• Information-gap tasks (Pica et al. 1989).
• Problem-solving tasks (Duff 1986).
• Debates (Duff 1986).
According to Pica and Doughty's study, one task type that is widely used in communicative
approaches -decision-making tasks- is considered to be a low generator of language
exchanges and hence inadvisable for acquisition and accordingly for practice in the
classroom.

3.7.3 THE NATURAL CONTEXT: FOREIGNER TALK

Going beyond formal contexts, a key question is to consider if and how the language
which learners are exposed to in the classroom is very different from the language
learners are exposed to in natural contexts. Even within natural contexts the kind of
language non-native individuals are exposed to varies considerably with respect to that
normally used by native speakers of the target language.
Several studies, such as ones by Long (1981) and Hatch (1983), confirm the existence
of speech modifications by native speakers in the use of the target language when the
interlocutor is not a native speaker. In other words, native speakers talk 'differently' to
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
foreigners. This different type of language, known as 'foreigner talk', can be said to be
the register used by the native speakers of a language when they address non-native
speakers. You may recall that it was briefly described in the subject Second Language
Acquisition.
Two main approaches to analysing the phenomenon of foreigner talk can be identified:
- The approach most widely used by researchers has been that of
examining specific examples of foreigner talk itself.

- A second approach has been through the field of discourse analysis,

which studies the conversations between native and non-native speakers.
This perspective is based on the fact that the input which learners receive
is not only determined by the native speaker but also by the learners
themselves, because their responses condition the new input they receive
from the native speaker.
In the latter approach, the emphasis is on the discourse both interlocutors construct. The
study of this kind of discourse can be made through the strategies used in conversations,
known as communication strategies.
As regards the first approach above (the direct study of manifestations of foreigner talk),
its implications are more limited because we lack data on the learner' s output in the
interaction, and thus on the elements of negotiation. Nevertheless, it can also provide
interesting insights into the input individuals receive
We will now turn to consider some of the most important features of foreigner talk.

[Link] FEATURES OF FOREIGNER TALK

First of all, the use of a different register by the native speaker, which is generally
characterised by simplification, starts from the desire to facilitate communication. That
is, the motivation behind the use of foreigner talk is that the non-native speaker understands
the native speaker. Thus, the basic function of foreigner talk is not to facilitate language
acquisition (Corder 1980) - as is the case with other registers such as teacher talk or
the language used by mothers when they talk to their children (motherese) - but
communication.
Secondly, it is also important to realise that this register is not always uniform but will
depend, to some extent, on certain other factors, such as:
• The non-native speakers’ disposition;

• The subject of conversation;

• The non-native speaker’s level of the target language;

• The non-native speaker’s age.

 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
It has also been noted that foreigner talk is sometimes characterised by a special
affective link between native and non-native speakers, also present in the language of
mothers and children. The problem here is that quite frequently the affective tone that
marks the native speaker's message is interpreted negatively by the non-native who
may feel insulted (Hatch 1983: 181) and may perceive the situation as patronising,
particularly when they consider that the simplifications used by the native speaker
exceed the strictly necessary. Especially humiliating -following Hatch again- is the
"affective" intonation so typical of mothers when talking to their children, used sometimes
to talk to foreigners without any communicative justification from the point of view of the
content of the message.
From a psycholinguistic point of view, Meisel (1980) notes that foreigner talk is a
manifestation of the universal strategies that are part of a speaker's linguistic competence.
That is to say, any speaker has the capacity to adapt his/her language production when
speaking to non-native speakers. The ways in which native speakers adjust their discourse
when they address non-native speakers take three fundamental forms (Ellis 1985):
- Regression:
The native speaker unconsciously goes back (or regresses) to different stages of
his/her own language acquisition till she/he finds an appropriate level for the non-
native being addressed.
- Adaptation:
The native speaker notices the forms of the linguistic system which are being used
by the non-native speaker and then uses the same forms.
- Negotiation:
The native speaker simplifies and clarifies the message in accordance with the
feedback obtained from the non-native speaker.
We have established that native speakers simplify their language when using foreigner talk.
This leads us to an interesting related question: How grammatically correct is foreigner talk?
Studies carried out in this area show rather contradictory results.
On the one hand, it is claimed that the language used by the native speaker, despite
being simplified, is characterised by its grammatical correctness, no matter who the
interlocutor is (e.g. Freed 1981). What changes depending on the interlocutor is the
syntactic complexity of these correct grammatical utterances. The most important factor
here seems to be the non-native speakers' level of second language proficiency. However,
Long (1983) points out that foreigner talk can be ungrammatical under the following
circumstances:
1. The learner’s level of proficiency in L2 -ungrammatical foreigner talk is more likely when
the learner’s proficiency is low.

2. The status of the native speaker -ungrammatical foreigner talk is more likely when
the native speaker is or thinks he or she is of higher status.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
3. The native speaker has prior experience of using foreigner talk but only of the "limited
kind" used to address non-native speakers of low proficiency.

4. The extent to which the conversation is spontaneous - ungrammatical foreigner talk

is less likely to be planned, formal discourse or in experimental situations.

(Ellis 1995:253)

The formal features of foreigner talk have been included in Appendix 3.1.

3.8 INPUT, INTERACTION AND TEACHING

We have seen that there are some important differences between natural and formal contexts
for a learner when learning a second language. To return to the classroom context, we will now
take a look at some of the suggestions that have been made to make up for the afore-
mentioned lack of opportunities for interaction and negotiation of input in the classroom.
Proposed solutions include a series of methodological changes such as promoting interaction
in more reduced groups where individuals can negotiate meaning (Pica and Long 1986).
Pica and Doughty (1985) also question the rush to include endless amounts of group work
in CLT, and problematise the widespread belief among EFL theorists and practitioners that
group work necessarily leads to the negotiation of meaning.

Another attempt to improve the situation in the classroom from the point of view of input,
is based on a suggestion of Edmonson's (1986). Edmonson stresses the high level of
ambiguity that characterises language classrooms. This ambiguity is frequently present in the
responses, in the form of feedback, that teachers give to students, who may be unclear about
what the teacher is signalling in feedback. Thus, Edmonson suggests, the teacher needs to try
to be clear and transparent in his/her speech, as feedback is one way of facilitating second
language learning. lf, as it has been pointed out, simplified input can facilitate the
understanding and acquisition of the target language it would be interesting to try to use it in
the classroom and observe its effects (Chaudron 1988; Hakansson 1986).

Another approach to providing learners with exposure to the L2 in the classroom involves
teaching other subjects (e.g. history, literature, science and so on) through the second language.
This is known as content teaching. Studies into the success of SLA in content teaching have
been fairly widespread, because content teaching presents certain similarities to natural
contexts, particularly as far as the elements of negotiation are concerned (Ellis 1985).
Hauptman et al. (1985), for example, found that affective factors played a major role in content
teaching. They point out, as examples of this, how much greater students' satisfaction is
when immersed in real communicative situations using the second language, which includes
solving problems in the second language, and how students are more prepared psychologically
to face situations that require the use of the second language.
Finally, it is important to bear in mind the relationship between a learner' s exposure to input
and interaction, and individual factors such as personality, cognitive style, attitude and
motivation, among others (see the subject Individual Factors for an overview of this area). In
many cases input in the L2 and opportunities for interaction are present in the learner's
environment, and it is the individual who needs to be able to seize this chance for
accelerating his/her own learning process. Let us briefly focus our attention on the types of
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
learners who are more likely to seize opportunities for exposure to the target language in the
classroom itself.

The hypothesis we are concerned with here is clearly expressed by Chaudron (1988:100).
Remember that in this unit we have already considered the fact that the teacher's input may be
a fundamental factor in aiding a learner's SLA in formal contexts. Therefore, the
counterpoint notion, as Chaudron puts it, is that:

... those learners who initiate interactions that result in speech directed to them will derive
more benefit from the input than if they are exposed to the input in a more vicarious
manner.
Seliger (1977) carried out a study in which he analysed the number of initiations of
interactions in the classroom by individual learners, in order to test this hypothesis.
Seliger distinguishes between two types of learners: those who generate more input
in the target language and those who generate less, high input generators and low
input generators respectively. He describes the characteristic behavior of each as follows:
Some students, the high input generators, are active/y involved and, like the child,
cause input to be directed at them by calling out, answering out of turn, and working
out answers to questions or drill cues directed to others. Such students can often be
observed talking to themselves during language drills. Still other students, the low input
generators, sit quietly but rarely and in some extreme cases never participate unless
specifically asked to do so.
(1977:266-7)
As might be expected, high input generators are the ones who, according to this study,
are more proficient in the target language in practically every aspect.
However, several subsequent studies (e.g. Day 1984; Strong 1983) have problematised
what initially seemed to be a straightforward relationship between high input generating
behaviour and successful SLA. Factors such as the learner's age, the social context for
learning and cultural factors were all seen to have an important effect on initiating
behaviour.
Chaudron summarises the situation in these words:
A direct relationship between learners’ initiating behaviors and development in
proficiency is not supported by the preceding studies. Yet it does appear that more
subtle distinctions in the factors involved, for example, functional types of initiations,
and social or age relationships, may influence the amount of initiation and interlocutors'
response to initiating behaviors. Cultural differences may furthermore influence the
extent to which particular learners engage in classroom interaction at all. With such
limitations, it seems inappropriate to regard initiating behaviors and the presumed input
generated from them as the ultimate factors responsible for language acquisition in
classrooms.
(1988:106)

Thus it would seem that the picture is a lot more complex than it appears at first glance.
 UNIVERSIDAD DE INNOVACIÓN PARA LA EXCELENCIA PROFESIONAL
IEXPRO
3.9 CONCLUSION
In this unit we have focused on the learner’s developing IL in a wider context: that of formal
versus natural contexts. We have looked at how input and interaction can aid successful
second language learning, although we have also noted that many of the studies into
input and interaction have been carried out in ESL environments. For this reason, we feel
that it would be useful for you to draw some conclusions from this unit, which relate
specifically to our EFL teaching context, in a final task.

TASK 3.3
Imagine that you need to write some advice for a new EFL teacher, telling him/her what
he/she should do in the classroom in order to improve his/her students' SLA through
input and interaction.

3.10 APPENDICES
[1] APPENDIX 3.1: The formal features of foreigner talk based on Hatch (1978; 1983)
and Long (1981; 1983).

3.11 SUGGESTIONS FOR FURTHER READING

[1] ELLIS, R. (1994): The Study of Second Language Acquisition. Oxford: Oxford
University Press. Chapter 13 & 14.

[2] ELLIS, R. (1997): Second Language Acquisition. Oxford: Oxford University Press.
Chapter 9.

[3] GASS, S. & MADDEN, C. (1985): Input in Second Language Acquisition.

Cambridge, Mass.: Newbury House.

[4] KRASHEN, S. (1985): The Input Hypothesis: Issues and implications. London:
Longman.