See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/308904899

The role of output & feedback in second language acquisition: A classroom-

based study of grammar acquisition by adult English language learners

Article  in  Eesti ja Soome-Ugri Keeleteaduse Ajakiri · June 2011

DOI: 10.12697/jeful.2011.2.2.01

CITATIONS READS

6 8,506

1 author:

Ewa Donesch-Jezo
Andrzej Frycz Modrzewski Krakow University
13 PUBLICATIONS   14 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Ewa Donesch-Jezo on 06 October 2016.

The user has requested enhancement of the downloaded file.

 ESUKA – JEFUL 2011, 2 – 2: 9 – 28

THE ROLE OF OUTPUT AND FEEDBACK IN

SECOND LANGUAGE ACQUISITION:
A CLASSROOM-BASED STUDY OF GRAMMAR
ACQUISITION BY ADULT ENGLISH
LANGUAGE LEARNERS

Ewa Donesch-Jezo
Jagiellonian University

Abstract. Second language acquisition (SLA) has been a con-

cern of many teachers and researchers since the early 1960s.
One of the issues related to SLA has been finding the techniques
which effectively focus the learners’ attention on the target form.
A number of theories and methods have been advocated for this
purpose, ranging from implicit options to more explicit ones. Al-
though each of the suggested methods has its own advantages,
they have, so far, proved insufficient to get learners to be able
to notice the gap between their own interlanguage forms and the
target language forms. Swain (1995, 1998) has argued that apart
from providing the learners with comprehensible input, compre-
hensible output also plays a beneficial role in L2 acquisition. By
encouraging the learners to speak or write in the L2 and provid-
ing them with the opportunities to do so, the learners can notice
that they are not able to say what they want to say in the target
language. On the basis of Swain’s output hypothesis, we assumed
that encouraging adult learners (university students) to produce
target-like output would promote their achievement of the gram-
matical competence necessary for producing academic tests. The
purpose of this article is to present the evidence from a class-
room-based, small-scale study of the effect of output on learner
acquisition of L2 modal verbs, adjectives and adverbs convey-
ing the meanings of uncertainty, all of which are parts of speech
that are important metadiscourse items. The results of the present
study suggest that an approach in which students are encouraged
to produce comprehensible output, combined with their being
provided with learning reinforcement ensured by appropriate
feedback, can be an effective source of establishing long-lasting
grammatical accuracy in the students’ target language.

Keywords: L2 acquisition, output, feedback, grammatical accu-

racy, metadiscourse, consciousness raising
10 Ewa Donesch-Jezo

1. Introduction

In order to teach a second/foreign language effectively,

every teacher should be aware how it is acquired by the learners.
Our knowledge about how learners acquire a second language
(also referred to as L2) has developed extensively in the past
40-50 years. Its development has been related to the growth of
media technology and the world being seen as a “global village”
and also to the introduction of the World Wide Web, all of which
have led to a change of attitude to the learning of an L2. Learn-
ing a foreign language has no longer been seen as being a kind of
pastime or hobby, but rather as a means of obtaining education
and/or a better job. Therefore, the understanding of the process
of second language acquisition and the methods necessary to
adopt for a maximization of its effectiveness have been in the
centre of attention of a number of researchers and teachers alike
(Ellis 1997: 3). In this context, the meaning of the term second
language acquisition (SLA) needs careful explanation. Second
language acquisition refers to the process of the learning of a
non-native language after the native language has been acquired.
Thus a second language may mean a third or fourth language,
which is not necessarily contrasted with a foreign language, and
this language may be learned in a formal and systematic way (in
a classroom) or in an informal and unsystematic way (outside
a classroom) or in a combination of both (Mitchell and Myles
2004: 5).
This search for the most effective methodology for SLA
has resulted in a growing body of empirical studies and theories
based on them, and these have given rise to the implementation
of various pedagogical methods advocated by researchers. These
methods range from implicit ones, that is, ones without any con-
scious instruction on the language system, such as input flooding
(Krashen 1985), and input enhancement (Lightbown and Spada
1990, Sharwood-Smith 1993, White 1991, 1998) to explicit ones,
based on providing conscious instruction on the form, meaning
and function of grammatical structures, which include direct
rule explanations (DeKeyser 1998, 2007), consciousness-raising
procedures (Sharwood-Smith 1981, 1993), and metalinguistic ex-
planatory feedback (Pica et al. 1987, Gass and Mackey 2007).
In all current theories on second language learning,
comprehensible input is seen as being the sine qua non of L2
 The role of output and feedback in second language acquisition 11

acquisition. It means that the language which the learner is ex-

posed to (e.g. in the texts he or she is provided with for reading
or listening) is a fundamental component of the learning process,
as it contains and provides all the necessary evidence from which
learners can form linguistic hypotheses (VanPatten and Williams
2007: 177).
The most influential propagator of the crucial role of input
in SLA was Stephen Krashen (1982, 1985, 1998). The basic claim
of Krashen’s “Input Hypothesis” (1985) was that the provision
and availability of input which is comprehensible to the learner,
and which is just beyond the learner’s current second language
competence (i+1), is the only necessary condition for language
learning – provided the input is of interest and is relevant to the
learner and is consciously paid attention to by him or her. In or-
der to increase the perceptual salience of the linguistic forms in
the input, some researchers (e.g. Sharwood-Smith 1981, Doughty
1991, White 1998) proposed typographical enhancement of these
forms through the use of italics, bolding, enlargement and un-
derlining. They argued that this type of enhancement directed
learners’ attention to the target forms which, according to these
researchers, in addition to providing the learner with extensive
reading and listening material, would promote acquisition of the
target forms.
Another component which is required for successful sec-
ond language learning is output. Merrill Swain (1985, 1995, 2005)
has argued that comprehensible output (i.e. second language pro-
duction) also plays an important role in L2 acquisition since it
ensures mental grammatical processing and is the most effective
stimulus for the development of the learner’s interlanguage. Her
research into immersion classes taught with content-based sec-
ond language instruction revealed that despite extended exposure
to a target language input, the immersion students’ productive
ability was far from that of native speakers’. This investigation
led Swain to a formulation of her “Output Hypothesis” (1985)
in which she put forward the idea that what the learners were
lacking were sufficient opportunities for second language pro-
duction. Swain argued that it is only during the production of
the second language (in speech or writing), that the learners can
notice that they are not able to say what they want to say in the
target language. As Swain stated: “Output may stimulate learn-
ers to move from the semantic, open-ended non-deterministic,
12 Ewa Donesch-Jezo

strategic processing prevalent in comprehension to the complete

grammatical processing needed for accurate production. Output,
thus, would seem to have a potentially significant role in the de-
velopment of syntax and morphology” (Swain 1995: 128).
One indispensable condition is necessarily present if we
are to guarantee the facilitating role of output in L2 acquisition,
according to Swain’s claim, and that is the “forcing” of the learner
to produce language. It means that by motivating the students to
carry out special learning tasks requiring interaction and by giv-
ing them instant feedback, the teacher is encouraging language
production to play the main role.
This paper presents the evidence from a classroom-based,
small-scale study of the effect of output on the acquisition of L2,
with special reference to modal auxiliary verbs, adjectives and
adverbs conveying the meaning of uncertainty. These linguistic
items, commonly occurring in academic discourse in English,
especially in its written form such as research papers, are compo-
nents of metadiscourse. Metadiscourse is the term which is given
for those lexico-grammatical items by which writers of research
papers present both themselves and their arguments, and also it
is the language by which they convey their judgements and opin-
ions in a way that opens a dialogue with other researchers who
may have alternative opinions. Metadiscourse allows the writer
to organize the prose as a coherent piece of text and serves to
reveal the writer’s relationship to the message presented and also
to the reader (Hyland 2004: 109). Modal auxiliary verbs (e.g.
may/might, can/could, would, should), adjectives of uncertainty
(e.g. likely, possible, probable) and adverbs of uncertainty (e.g.
perhaps, possibly, probably) are the “hedges” which withhold
the complete commitment of the writer to a presented research,
weakening in this way the force of his or her argument in the
discourse, and leaving an open space for the reader’s response
(Hyland 2004, 2005). The knowledge of metadiscourse items by
university students will allow them to express their arguments in
written and spoken academic texts with appropriate caution and
due circumspection, which are the discursive features required
by the conventions of Anglo-American academic communities
(Hyland 2005: 92-93).
The aim of the present study was to answer the following
questions: (1) Which of the three investigated methods for the
presentation of the previously mentioned grammatical structures
 The role of output and feedback in second language acquisition 13

is the most effective for adult learners (in this case university
students in medicine)? (2) Does the encouragement of learners
to produce target-like output (by carrying out special activities)
promote the development of the grammatical and rhetorical com-
petence necessary for the learner to be able to create academic
discourse of a quality conforming to the recognized academic
conventions?
By conducting the above-mentioned comparative study
the author of this paper tried, firstly, to find the best method for
the acquiring of certain linguistic items by university students
and, secondly, to check the effectiveness of output production (as
argued by Swain) on SLA by adult learners. The first part of this
paper provides information about the role of output and feedback
in SLA, and this forms the theoretical foundation for the empiri-
cal study, described in the second part of this paper.

2. Potential roles of output and feedback in second

language learning

Swain’s (1985) output hypothesis was formulated as a

reaction to Krashen’s (1985) input hypothesis and as a reaction
against what Swain saw as the inefficacy of the use of compre-
hensible input alone in the development of learners’ linguistic
competence in the immersion schools in Canada. As Swain wrote
“... the output hypothesis claims that the act of producing language
(speaking and writing) constitutes, under certain circumstances,
part of the process of second language learning” (Swain 2005:
471).
Overall, five basic functions of output in L2 acquisition
were suggested by Swain (1985, 1995, 2005):
1. The first function she described was the noticing func-
tion, also referred to as the consciousness-raising role. What she
meant by this was that when trying to produce the second lan-
guage, learners notice a gap in their knowledge, and as a result of
this, they consciously identify their linguistic errors and lacks of
knowledge. This process can stimulate the cognitive generation
of new linguistic knowledge or the consolidation of the previ-
ously possessed knowledge. In other words, language production
involves the using of previously met language items in ways that
they have not been used before.
14 Ewa Donesch-Jezo

2. Secondly, she noticed a hypothesis-testing function of

output. She observed that the learner’s language production is
confirmed or modified on the basis of the learner’s interaction
and the feedback that he or she receives. This hypothesis test-
ing is extremely important in interaction when learners negotiate,
either with each other or with their teacher, in an attempt to con-
vey meaning. A fruitful negotiation may lead to a semantic or
morphosyntactic modification of output. When feedback is un-
available, the learner checks his or her hypothesis on the basis
of the internalized knowledge that he or she possesses about the
language.
3. The third function of output that she discerned was a
metalinguistic one. Metalanguage is used by learners when they
collaboratively try to solve linguistic problems and, as a result,
extend these learners’ knowledge about the forms and rules of
a language. The solving of grammar problems through a study
of comprehensible output, with grammatical structures being the
targeted topic of conversation, involves a lot of talk about lan-
guage. This talk contributes to language learning. By encouraging
learners to use metalanguage when they encounter linguistic
problems during the performance of learning tasks, the teacher is
able not only to observe “working” hypotheses about the foreign
language, but he or she can also obtain information on various
learning strategies which are being used by the learners.
4. Fourthly, Swain saw output as having a function in the
development of fluency. She argued that the production of output
promotes automatization of language use. Through frequent tar-
get language production learners can gain fluency and confidence
in L2 use.
5. The fifth function of output, as seen by Swain, was that
it enabled learners to move from a semantic to a syntactic use of
language. Language use gets learners to move from a stage of
comprehension (a semantic use of the language) to a syntactic
use of language (when the learner is paying attention to linguistic
form/s). For example, the learner noticing that his or her utter-
ance has not been understood by the interlocutor, reformulates it
to comply with target language standards.
Comprehensible output production is usually inseparably
linked with feedback, which is a kind of interaction providing
learners with error correction and with metalinguistic informa-
tion, facilitating improvement of the accuracy of L2 production.
 The role of output and feedback in second language acquisition 15

In the last 20 years, building on the research of Anderson (1982)

and Vygotsky (1978), feedback has been regarded in SLA as be-
ing a crucial factor in the encouragement and consolidation of
learning.
Feedback is an interaction that makes the learner aware of
his or her incorrect use of language, and provides the model for
a correct L2 use. Explicit feedback (Birdsong 1989, Gass 1988,
White 1991) may take the form of direct correction, e.g. telling
the learner that he or she is using the wrong word or the wrong
grammatical structure, and following this up with metalinguistic
instruction and explanation in how to use the word correctly.
Implicit feedback (Doughty and Varela 1998, Oliver 1995)
is a form of indirect correction which includes strategies such as:

• Confirmation checks – the use of expressions to elicit a con-

firmation that the learner’s sentence has been correctly heard
or understood (e.g. Is that what you mean?).
• Clarification requests – the use of expressions to elicit a clari-
fication of the utterance (e.g. What did you say?).
• Comprehension checks – the use of expressions to check
that an utterance has been understood (e.g. Did you under-
stand?).
• Recasts – the rephrasing of an incorrect utterance using the
correct form while still maintaining the original meaning.

Feedback focuses itself on the problematic aspects of

the learners’ interlanguage and gives them the opportunity to
concentrate on their language production and comprehension,
which, in turn, helps them to notice any gaps in their knowledge.
As Schmidt and Frota suggest “a second language learner will
begin to acquire the target-like form if, and only if, it is present
in comprehended input and ‘noticed’ in the normal sense of the
word, that is, consciously” (1986: 311).

3. Study of the role of output and feedback in L2

learning

While there is general agreement that accuracy is an im-

portant goal of L2 teaching, especially in the case of students
at the tertiary educational level, it is, however, not clear what
16 Ewa Donesch-Jezo

kind of classroom procedures should be used for the teaching

of particular linguistic forms to particular students. Keeping this
in mind teachers should constantly be trying to compare the ef-
fectiveness of various methods and techniques to find the most
suitable means of presentation and practicing of L2 features for
their students.

3.1. The purpose of the study and study design

In order to find the most effective classroom procedures

for enhancing the development of accuracy in the use of some
features of metadiscourse (modal auxiliary verbs and some ad-
jectives and adverbs), the study was designed to compare and
assess three methods that could be used for the promoting of the
acquisition of the target structures. These methods were: (1) ex-
plicit instruction provision (DeKeyser 1998, Doughty 1991, Ellis
1991, Lightbown and Pienemann 1993), (2) input enhancement
and implicit feedback provision (Lightbown and Spada 1990,
White 1991, 1998), and (3) the forcing of learners to produce
target-like output (Kowal and Swain 1994, 1997, Swain 1985,
1995, 1998, 2005).
The study described here was motivated by the following
factors:

• the fact that the students were having, and had had, problems
with the use of modal auxiliary verbs conveying the meaning
of probability and logical assumption (e.g. Chronic disease
self management program can improve health status ...;The
impact of specific interventions could be overestimated; Dis-
ease management programmes may not achieve expected
effects ...), especially with modal verbs followed by a per-
fect infinitive “modal+have+past participle” (e.g. The report
suggested that the therapy might have affected findings ...;
Individuals could have been prescribed more than one drug
from each category ...; There might have been some inaccu-
racy in our estimation of effectiveness ...);
• the realization of the importance for the students to be able to
recognize that this type of linguistic items is used by authors
of research papers in the field of medicine to distance them-
selves (hedge) from their claims presented in the paper;
 The role of output and feedback in second language acquisition 17

• the knowledge of the relationship between the form, meaning

and function of modal auxiliary verbs as well as adjectives
and adverbs of uncertainty is necessary for a correct produc-
tion of academic texts;
• the students’ production of meaningful output enables them
to notice the gap in their current use of interlanguage;
• the students’ output production gives them an opportunity
to discuss, analyse and experiment with the use of new and
known L2 structures.

3.2. Material and method

The study was carried out over a 4-week period in a regu-

lar 2-hour-per-week schedule. The study group included a total
of 45 students in the 3rd year of medicine (their mother tongue
was Polish). They were the participants of three classes of their
obligatory EMP (English for Medical Purposes) course in the
Medical College at Jagiellonian University in Krakow. The stu-
dents represented a B2 level of proficiency in English. These
three classes, which constituted three study groups (A, B and C),
comprised 14-16 persons each. The linguistic features in focus
were elements of metadiscourse whose function in academic dis-
course was problematic for the students.
The students of all three study groups, working in pairs,
were provided with the language corpus, which consisted of
selected samples of authentic research articles in the field of
medicine that had been published in international scientific
journals. These articles had been retrieved online from the uni-
versity library, and were presented to the students together with
a set of language-learning tasks prepared by the teacher. As both
“Introduction” and “Discussion” sections of research articles
are sections which are densely packed with metadiscourse items
(Salager-Meyer 1994), these two sections constituted the main
portion of the language input.
There were three different treatment conditions designed
for each group:
Group A received the input in a form in which the tar-
get metadiscourse elements (hedges) were typographically
enhanced by underlining and highlighting them with colour
marker-pens. No explicit grammar rule was given, but the stu-
18 Ewa Donesch-Jezo

dents were provided with positive and negative feedback by the

teacher, in the form of error correction and metalinguistic ex-
planations.
Group B was provided with the input enhanced in the
same manner as for Group A. Explicit rule explanation on the
form and function of the target linguistic items was given to
the students prior to reading the text and to doing the language-
learning tasks. The feedback obtained from the teacher was in
the form of error correction, no metalinguistic explanations
were provided during the students’ reading the text and doing
the tasks that were assigned to them after the reading had been
finished.
Group C received non-enhanced input, and before reading
the text and doing the tasks the students performed a dictogloss
(overall four dictogloss sessions). In the dictogloss activity
(Swain 1998, Wajnryb 1989), a short, specially adapted research
article, dense with target forms was read twice to the students.
When it was read for the first time, the students were asked to lis-
ten to the passage. During the second reading the students were
allowed to take notes to help them reconstruct the text. Then the
students worked in pairs or small groups to reconstruct a dic-
togloss text (for 25–30 minutes) from their shared notes. The
students’ performance of the task was observed by the teacher,
who approached each pair and provided them with corrective and
explicit metalinguistic feedback. The purpose of the teacher’s
feedback was to draw attention to particular grammatical forms
and to heighten the students’ awareness about the structural and
functional aspects of hedges and other linguistic items that were
necessary for carrying out dictogloss and language-learning tasks.
Finally, the reconstructed texts were compared between the pairs,
and also with the original dictogloss passage, in a whole-class
context.
It should also be noted that before hearing the dictogloss
passage (Group C), and before reading the text and doing the
language-learning tasks (Groups A and B), the unknown vocabu-
lary was explained by the teacher and the topic of the text was
discussed by the students through a brainstorming activity.
The table below shows the required tasks of the learners
and the teacher for every step of the dictogloss activity.
 The role of output and feedback in second language acquisition 19

Table 1. Steps for dictogloss activity (source: Nation and

Newton 2009: 68)

Step Students Teacher

1. Preparation Vocabulary study to prepare for the test.
Discuss the topic (predict vocabulary and
content etc.)
2. Listening for Listen to the whole Reads the text at nor-
meaning text mal speed
3. Listening and note- Take notes listing key Reads again at nor-
taking words mal speed
4. Text reconstruction Work in groups to Helps groups
in groups reconstruct an ap-
proximation of the Offers guidance
text from notes
5. Text comparison Compare group ver- Facilitates class com-
between groups sions of the text. Pay parison of versions
attention to points from different groups.
of usage that emerge Facilitates discussion
from the discussion and correction of
errors

It is worth noticing that steps 4 and 5 direct the learners’

attention to language form (i.e. word forms, word order, gram-
mar rules etc.) after having been exposed to the meaning-focused
listening in steps 2 and 3.
The set of language-learning tasks, which all the three
groups performed during and after reading the text, included the
following activities (adapted from Channel 1994, Hyland 2005,
Salager-Meyer 1994):

• Ranking the sentences in terms of the strength of the writer’s

claim;
• Choosing the hedges for filling the gaps to weaken the au-
thor’s claim;
• Identifying the examples of hedges in the text, and assigning
meaning to them;
• Distinguishing between statements in the text which report
facts and those which express opinions;
• Adding or removing all hedges from the text and noticing the
effect this has on it;
• Gap-filling exercise – inserting an appropriate metadiscourse
20 Ewa Donesch-Jezo

item in the gapped text;

• Transforming a spoken text, such as a lecture, into an essay,
including the indicated hedging devices;
• Changing a non-academic text from a popular science
magazine into an academic text for a scientific journal, and
discussing what changes should be made;
• Translating a text in Polish into English for a similar aca-
demic community, and comparing the use of hedges in the
two languages;
• Conducting an analysis of a self-chosen research article from
the point of view of the hedges used in them and their rhetori-
cal function in the text;
• Writing an essay entitled “The relation between smoking and
lung cancer”.

In order to assess the long-term effectiveness of the three

treatment procedures aiming to help the students gain accuracy
in using hedges, all the groups underwent a pre-test prior to the
experiment, an immediate post-test just after finishing the ex-
periment, and a remote post-test 2 months after the experiment.
The pre-test comprised the following tasks: gap-filling, multiple
choice questions and the ranking of metadiscourse items accord-
ing to their strength. The post-tests involved a set of tasks such as
multiple-choice, gap-filling, correction of sentences, translation
from Polish into English, and the ranking of sentences according
to the strength of the assertiveness of the presented information
in them.

3.3. Results

The effectiveness of these three approaches to teaching

modal verbs and some adjectives and adverbs was assessed by
two tests: an immediate post-test (immediately after finishing the
teaching session) and a remote post-test 2 months later.
On immediate tests based on multiple-choice and gap fill-
ing tasks groups A and C performed similarly, but on remote
tests, especially those requiring manipulation with the texts such
as conversion of a non-academic text into an academic text in
which hedges should be used to present the authors’ claims in a
cautious way, group C outperformed both groups A and B.
 The role of output and feedback in second language acquisition 21

The results presented in Figure 1 show the total score

obtained in the tasks performed at these tests, expressed in the
percentage of the maximum score.

Figure 1. Results of the tests.

Pre-test: Group A: 62%; Group B: 64%, Group C: 60%
Immediate post-test: Group A: 91%; Group B: 78%, Group C: 95%
Remote post-test: Group A: 86%; Group B: 74%, Group C: 93%

4. Discussion

The contribution of output to second language acquisi-

tion is still a subject for debate since it has not been sufficiently
documented in the literature concerning the whole subject of L2
acquisition (Shehadeh 2002). The research that has been carried
out into the role of “pushed” output in second language vocabu-
lary acquisition seems to show benefits more clearly than in the
case of second language grammar acquisition. For example,
De la Fuente (2002), Ellis and He (1999), Swain and Lapkin
(1995, 1998) have observed a direct positive relation between
the learner’s frequency of exposure to output and second lan-
22 Ewa Donesch-Jezo

guage vocabulary development. When teaching vocabulary to

students who were at a low level of proficiency in the language,
Ellis and He (1999) found that the group of learners who were
required to give instructions and explanations to an interlocutor
outperformed at subsequent post-tests the groups who were not
required to produce language but only performed learning activi-
ties, such as matching words with pictures, matching words with
definitions, labelling pictures, etc. Similarly, De la Fuente’s study
(2002) has shown that an “output” group received higher scores
at post-tests than a “non-output” group, as far as production of
vocabulary was concerned. However, when reception of vocabu-
lary was assessed, all groups achieved a similar level.
The results of the investigations into the contribution of a
“forced” output to the development of grammar accuracy in second
language learners have not been unequivocal. Nobuyoshi and Ellis
(1993) investigated to see whether there had been any improve-
ment in the grammatical accuracy of second language learners’
utterances in a small-scale study of the learning of the English
past tense. These researchers noticed that encouraging English
language learners to modify their output by means of clarification
requests led to increased accuracy in the using of these past tense
forms. Izumi and Bigelow (2000) studied the influence of “forced”
output in the learning of third conditional clauses in English. The
experimental groups, who had to generate written texts similar to
those they had been presented with as their input, considerably im-
proved the accuracy of their writing during the experiment, but at
post-tests, which focused on the use of target grammar structures,
both experimental and control groups performed similarly. Ana-
logical results were also obtained by Izumi et al. (1999) in their
study about the acquisition of relative clauses. The comparison of
the results obtained by the experimental “output” group with the
“non-output” group showed no differences between the studied
groups. Although the output task was also a production task, it did
not promote acquisition of the target form.
The results obtained from our experiment seem to be in-
consistent with the results obtained by Izumi and Bigelow (2000),
and by Izumi et al. (1999) since in the immediate post-test, the
“output” group (Group C) and one of the “non-output” groups
(Group A) obtained similar mean scores, but on the remote post-
test the “output” group (Group C) quite noticeably outperformed
both “non-output” groups.
 The role of output and feedback in second language acquisition 23

Although some researchers, e.g. Ellis (2001), Mitchell

and Myles (2004), Shehadeh (2002), state that the benefits of
“pushed” output remain somewhat elusive and hard to demonst-
rate since there is still a severe lack of data proving that learner’s
“pushed” output has any effect on second language learning, the
results of our study seem to show a beneficial role for “pushed”
output in achieving a greater long-term accuracy in the use of
modal verbs, adjectives and adverbs that convey meanings of un-
certainty in the expressing of opinions and in the presenting of
a certain authorial distance from the proposition. Our observa-
tion of the performance of the students’ text reconstruction tasks
showed that at first the students were more concerned with pro-
ducing a coherent text than with attention to the target forms.
After the teacher had directed their attention to the target forms,
the students started to notice them and use these forms in the re-
constructed text more often. It should be noted that the students
in all our three study groups were on the intermediate level of
L2 proficiency, and the target forms had not been totally new
for at least some of them. However, the pre-test showed that
these target forms had not been adequately mastered by them,
especially when they were required to use them in a sociolinguis-
tically appropriate manner. Although our study was conducted on
a small scale, it provides some evidence supporting the role of
output in SLA, this being an especially important role in the ef-
fective development of long-term accuracy in the use of grammar
forms in texts generated by them according to accepted academic
standards.

5. Conclusion

The results of the above-presented study suggest that the pro-

cedure of “forcing” students to produce meaningful output can be
an effective source of long-lasting grammatical and rhetorical accu-
racy. It means that dictogloss sessions during which students were
required to produce comprehensible output and check their hypoth-
eses about the L2 may be useful in heightening students’ awareness
of grammatical and rhetorical aspects of research articles in the
context of teaching ESP. Performing dictogloss activities com-
bined with the teacher’s explicit feedback provision, students were
engaged in target language production, which focused their atten-
24 Ewa Donesch-Jezo

tion not only on vocabulary, morphology and syntactic structures,

but also on rhetorical conventions used in an academic context.
Moreover, the results of this study seem to indicate that supplying
metalinguistic explanatory feedback during the students’ task per-
formance (Groups A and C) facilitates the learning of grammatical
structures better than providing grammar rules prior to carrying out
learning activities with no explicit feedback (Group B).
The results of the above-described investigation seem to
indicate the potential role of the combination of three factors in
the promotion of long-term second language grammar compe-
tence. These factors are:
1. Comprehensible output production stimulated by dic-
togloss, which allows students to check their hypotheses about
the L2, notice the gap in their current linguistic knowledge, and
acquire fluency in target language production;
2. Explicit awareness-raising tasks, which develop the
grammatical and rhetorical competence necessary for language
production;
3. Teacher’s explicit feedback during dictogloss activity
and language-learning tasks, which provided error correction and
metalinguistic explanations on the structure and function of the
linguistic items in question.
Since the number of students investigated was insufficient
to provide convincing quantitative evidence, it is necessary to
carry out a study including a larger number of participants. The
results from this type of research would give ESP teachers infor-
mation which could be utilized in their task of the preparation of
effective language presentation and language-learning activities.
The benefits from this would be not only the raising of students’
linguistic and rhetorical consciousness, but also the developing
of their fluency and confidence in the use of the target language.

Address:
Ewa Donesch-Jezo
Department of Foreign Languages
Jagiellonian University
Os. Kolorowe 22/37
31-940 Krakow
Poland
E-mail: [email protected]
 The role of output and feedback in second language acquisition 25

References
Anderson, John R. (1982) “Acquisition of cognitive skill”. Psychological
Review 89, 369–406.
Birdsong, David (1989) Metalinguistic performance and interlinguistic compe-
tence. New York: Springer.
Channel, Joanna (1994) Vague language. Oxford: Oxford University Press.
De la Fuente, Marıa José (2002) “Negotiation and oral acquisition of L2
vocabulary”. Studies in Second Language Acquisition 24, 1, 81–112.
DeKeyser, Robert (1998) “Beyond focus on form: cognitive perspectives on
learning and practicing second language grammar”. In Catherine
Doughty and Jessica Williams, eds. Focus on form in classroom second
language acquisition, 42–63. New York: Cambridge University Press.
DeKeyser, Robert (2007) “Skill acquisition theory”. In Bill VanPatten and
Jessica Williams, eds. Theories in second language acquisition: an
introduction, 97–113. Mahwah, NJ: Erlbaum.
Doughty, Catherine (1991) “Second language instruction does make a differ-
ence: evidence from an empirical study on SL relativization”. Studies
in Second Language Acquisition 13, 4, 431–469.
Doughty, Catherine and Elizabeth Varela (1998) “Communicative focus on
form”. In Catherine Doughty and Jessica Williams, eds. Focus on form
in classroom second language acquisition, 114–139. Cambridge: Cam-
bridge University Press.
Ellis, Rod (1991) “Grammar teaching – practice or consciousness-raising?”
In Rod Ellis, ed. Second language acquisition and second language
pedagogy, 232–241. Clevedon, Avon: Multilingual Matters.
Ellis, Rod (1997) Second language acquisition. Oxford: Oxford University Press.
Ellis, Rod (2001) “Methodological options in grammar teaching materials”.
In Eli Hinkel and Sandra Fotos, eds. New perspectives on grammar
teaching in second language classrooms. Mahwah, NJ: Lawrence Erl-
baum.
Ellis, Rod and Xien He (1999) “The roles of modified input and output in the
incidental acquisition of word meanings”. Studies in Second Language
Acquisition 21, 2, 285–301.
Gass, Susan M. (1988) “Integrating research areas: a framework for second
language studies”. Applied Linquistics 9, 2, 198–217.
Gass, Susan M. and Alison Mackey (2007). “Input, interaction and output in
second language learning”. In Bill VanPatten and Jessica Williams, eds.
Theories in second language acquisition: an introduction, 175–200.
Mahwah, NJ: Erlbaum.
Hyland, Ken (2004) Disciplinary discourses: social interactions in academic
writing. 2nd ed. Michigan: The University of Michigan Press.
26 Ewa Donesch-Jezo

Hyland, Ken (2005) Metadiscourse: exploring interaction in writing. London:

Continuum.
Izumi, Shinichi and Martha Bigelow (2000) “Does output promote noticing and
second language acquisition?” TESOL Quarterly 34, 2, 329–378.
Izumi, Shinichi, Martha Bigelow, Miho Fujiwara, and Sarah Fearnow (1999)
“Testing the output hypothesis: effects of output on noticing and second
language acquisition”. Studies in Second Language Acquisition 21, 3,
421–452.
Kowal, Maria and Merrill Swain (1994) “Using collaborative language pro-
duction tasks to promote students’ language awareness”. Language
Awareness 3, 2, 73–93.
Kowal, Maria and Merrill Swain (1997) “From semantic to syntactic process-
ing: how can we promote metalinguistic awareness in the French
immersion classroom?” In Robert Keith Johnson and Merrill Swain,
eds. Immersion education: international perspectives, 284–309. New
York: Cambridge University Press.
Krashen, Stephen D. (1982) Principles and practice in second language acqui-
sition. London: Pergamon.
Krashen, Stephen D. (1985) The input hypothesis: issues and implications.
New York: Longman.
Krashen, Stephen D. (1998) “Comprehensible output?” System 26, 175–182.
Lightbown, Patsy M. and Manfred Pienemann (1993) “Comments on Stephen
D. Krashen’s ‘Teaching issues: formal grammar instruction’”. TESOL
Quarterly, 27, 4, 717–722.
Lightbown, Patsy M. and Nina Spada (1990) “Focus on form and corrective
feedback in communicative language teaching: effects on second
language learning”. Studies in Second Language Acquisition 12, 4,
429–448.
Mitchell, Rosamond and Florence Myles (2004) Second language learning
theories. 2nd ed. London: Arnold.
Nation, I.S.P. and Jonathan Newton (2009) Teaching ESL/EFL Listening and
Speaking. New York, London: Routledge.
Nobuyoshi, Junko and Rod Ellis (1993) “Focused communication tasks and
second language acquisition”. ELT Journal 47, 3, 203–210.
Oliver, Rhonda (1995) “Negative feedback in child NS-NNS conversation”.
Studies in Second Language Acquisition 17, 4, 459–481.
Pica, Teresa, Richard Young, and Catherine Doughty (1987) “The impact of
interaction on comprehension”. TESOL Quarterly 21, 4, 737–758.
Salager-Meyer, Françoise (1994) “Hedges and contextual communicative
function in medical English: written discourse”. Journal of English for
Specific Purposes 13, 2, 149–170.
 The role of output and feedback in second language acquisition 27

Schmidt, Richard and Sylvia Frota (1986) “Developing basic conversational

ability in a second language: a case study of an adult learner of Portu-
guese”. In Richard R. Day, ed. Talking to learn: conversation in second
language acquisition, 237–326. Rowley, MA: Newbury House.
Sharwood-Smith, Michael (1981) “Consciousness-raising and second language
acquisition theory”. Applied Linguistics 2, 2, 159–168.
Sharwood-Smith, Michael (1993) “Input enhancement in instructed SLA: theo-
retical bases”. Studies in Second Language Acquisition 15, 2, 165–179.
Shehadeh, Ali (2002) “Comprehensible output, from occurrence to acquisi-
tion: an agenda for acquisitional research”. Language Learning 52, 3,
597–647.
Swain, Merrill (1985) “Communicative competence: some roles of comprehen-
sible input and comprehensible output in its development”. In Susan M.
Gass and Carolyn G. Madden, eds. Input in second language acquisi-
tion, 235–253. Rowley, MA: Newbury House.
Swain, Merrill (1995) “Three functions of output in second language learning”.
In Guy Cook and Barbara Seidlhofer, eds. Principle and practice in ap-
plied linguistics, 125–144. Oxford: Oxford University Press.
Swain, Merrill (1998) “Focus on form through conscious reflection”. In Cathe-
rine Doughty and Jessica Williams, eds. Focus on form in classroom
second language acquisition, 64–82. Cambridge: Cambridge Univer-
sity Press.
Swain, Merrill (2005) “The output hypothesis: theory and research”. In Eli
Heinkel, ed. Handbook of research in second language teaching and
learning, 471–483. Mahwah, NJ: Lawrence Erlbaum Associates.
Swain, Merrill and Sharon Lapkin (1995) “Problems in output and the cogni-
tive processes they generate: a step towards second language learning”.
Applied Linguistics 16, 3, 371–391.
Swain, Merrill and Sharon Lapkin (1998) “Interaction and second language
learning: two adolescent French immersion students working together”.
Modern Language Journal 82, 3, 320–337.
VanPatten, Bill and Jessica Williams (2007) Theories in second language
acquisition. London: Lawrence Erlbaum Associates, Publishers.
Vygotsky, Lev S. (1978) Mind in society: the development of higher psycho-
logical processes. Cambridge, MA: Harvard University Press.
Wajnryb, Ruth (1989) “Dicto-gloss: a text-based communicative approach to
teaching and learning grammar”. English Teaching Forum 27, 4, 16–29.
White, Joanna (1998) “Getting the learners’ attention: a typographical input
enhancement study”. In Catherine Doughty and Jessica Williams, eds.
Focus on form in classroom second language acquisition, 85–113.
Cambridge: Cambridge University Press.
 28 Ewa Donesch-Jezo

White, Lydia (1991) “Adverb placement in second language acquisition: some

effects of positive and negative evidence in the classroom”. Second
Language Research 7, 2, 133–161.

Kokkuvõte. Ewa Donesch-Jezo: Väljundi ja tagasiside roll teise

keele omandamisel: keeletunnipõhine uurimus täiskasvanud ing-
lise keele õppijate grammatika omandamisest. Alates 1960. aastate
algusest on teise keele omandamist uurinud väga paljud õpetajad ja
teadlased. Üks teise keele omandamisega seotud teemadest on olnud
meetodi leidmine, mis suunaks õppija keskendama oma tähelepanu
sihtkeele vormile. Selle saavutamiseks on välja pakutud mitmeid teoo-
riad ja meetodeid, alustades kaudsematest ja lõpetades konkreetsemate
võimalustega. Kuigi igal väljapakutud meetodil on oma eelised, ei ole
neist senini piisanud, et panna õppijaid märkama lõhet nende enda
vahekeele vormide ja sihtkeele vormide vahel. Swain (1995, 1998) on
väitnud, et peale selle, et õppijatele antakse arusaadav sisend, täidab
teise keele omandamisel tähtsat rolli ka arusaadav väljund. Kui julgus-
tada õppijaid rääkima ja kirjutama teises keeles ning luua neile selleks
võimalusi, märkavad nad, et ei ole suutelised ütlema seda, mida nad
sihtkeeles öelda sooviks. Swaini hüpoteesi põhjal väljundi kohta eelda-
takse, et kui julgustada täiskasvanud õppijaid (üliõpilasi) moodustama
sihtkeele moodi väljundeid, soodustab see akadeemiliste testide tegemi-
seks vajalike grammatiliste oskuste omandamist. Uurimustöö eesmärk
on keeletunnis läbi viidud väiksemahulise uuringu abil näidata väljundi
mõju õppijale teise keele ebakindlust väljendavate modaaltegusõnade,
omadussõnade ja määrsõnade omandamisel. Kõik need sõnaliigid on
olulised metadiskursuse osad. Uuringu tulemused näitasid, et selline
lähenemine, kus õppijaid julgustati tekitama arusaadavat väljundit koos
õpitu kinnistamise eest hoolitsemisega, mis tagati sobiva tagasisidega,
võib olla tõhus viis kestva grammatilise täpsuse saavutamiseks õppijate
sihtkeeles.

Märksõnad: teise keele omandamine, tagasiside, grammatiline täpsus,

metadiskursus, teadlikkuse tõstmine

View publication stats