Current Issues

in the Teaching
of Grammar:

An SLA
Perspective

ROD ELLIS
University of Auckland
Auckland, New Zealand

The Author:
He
published
Rodhas
Ellis
is a
widely
in the
field
professor
in the
of
SLA. His latest
Department
of
books are
Applied Language
Analyzing
Studies and
Language Learning
Linguistics at the
and Planning and
University of
Task Performance
Auckland, New
in a Second
Zealand.
Language.

What is this article

about?

EIGHT
QUESTIO
NS
THE

to be
addressed

8. Should grammar be
7.
Is there
a best way to
taught
in
separate
6.
Is there
any value
in
teach
grammar
for
lessons
or integrated
5.
Should
grammar
teaching
explicit
implicit
knowledge?
4.
Should
grammar
into
instruction
communicative
grammaticalbe intensive
instruction
be massed
3.
When
should
we teach
activities?
or
extensive?
1.
Should
we
teach
knowledge?
2.distributed?
What grammar
or
grammar?
grammar,
or should we
should
we teach?
simply create the
conditions by which
learners learn
naturally?

DEFINING

GRAMMA
R
TEACHING

TRADITIONAL VIEW
Grammar teaching
is viewed as the
presentation and
practice of
discrete
grammatical
structures.

Grammar teaching involves

any instructional technique that
draws learners attention to
some specific grammatical form
in such a way that it helps them
either to understand it
metalinguistically and/or process
it in comprehension and/or
production so that they can
internalize it.

Should we

TEACH
Grammar

GRAMMAR?

This question
was
motivated by
early
research into
naturalistic
acquisition.

Learners had
their own
built-in
syllabus for
learning
grammar.
- Corder,
1967

Grammar instruction played

no role in acquisition. Learners
would automatically proceed
along with their built-in
syllabus as long as they had
access to comprehensible
input and were sufficiently
motivated.
--Krashen, 1967

A number of empirical
studies were designed to:
b.. Examine
c.
Comparewhether
the
attempts
instructed
to teachof
a
Compare
thesuccess
order
ofofacquisition
and naturalistic
specific
grammatical
learners
structures
instructed
and naturalistic
learners
resulted in their acquisition
(e.g.,
Pica,
1983)
(e.g., White,
(Long,
Spada,
1983)
Lightbown,
& Ranta, 1991)

instruction
Instructed
The order
learners
was
of acquisition
no guarantee
generally
wasthat
the
achieved
learners
same for
would
higher
instructed
acquire
levels and
of
what they
naturalistic
grammatical
had
been taught.
competence than
learners

Thus, some researchers

concluded (e.g., Long,
1988) that teaching
grammar was beneficial
but that to be effective
grammar had to be taught
in a way that was
compatible with the natural
processes of acquisition.

Further, there
Subsequent
research,
is
such as Norris
evidence
that, and
contrary
Ortegas
to
Krashens
(2000)
(1993)
metaanalysis
continued
claims,
of 49
studies, hascontributes
instruction
borne out
theboth
to
overall
acquired
effectiveness
knowledge
as of
well as
grammar
learned
knowledge.
teaching.

There is also increasing

evidence that naturalistic
learning in the classroom
(e.g., in immersion
programmes) does not
typically result in high
levels of grammatical
competence (Genesee,
1987).

What

GRAMMAR

SHOULD WE
Teach?

TEACttH?

TWO SEPARATE
QUESTIONS:

Linguistics affords a
broad selection of
grammatical models
to choose from
including structural
grammars,
generative
grammars and
functional
grammars.

Traditionally, syllabuses have

been based on structural or
descriptive grammars.
Structural syllabuses
traditionally emphasised the
teaching of form over meaning
(e.g., Lado, 1970).

Modern syllabuses rightly

give more attention to
functions performed by
grammatical forms.

Establishing connections
between form and meaning
is a fundamental aspect of
language acquisition.
-- VanPatten, Williams, and Rott (2004)

Which

GRAMMATICA
L STRUCTURE
to Teach?

Two

POLAR
positions

FIRST
POSITION
KRASHENS MINIMALIST POSITION

Krashen (1982) argues that

grammar teaching should be
limited to a few simple and
portable rules such as 3rd
person-s and past tense-ed that
can be used to monitor output
from the acquired system.

FIRST
POSITION
KRASHENS MINIMALIST POSITION

However, there is now

ample evidence that many
learners are capable of
mastering a wide range of
explicit grammar rules.

FIRST
POSITION

KRASHENS MINIMALIST
POSITION

For example, Green and Hecht

(1992) found that university-level
students of English in Germany were
able to produce clear explanations
for 85% of the grammatical errors
they were asked to explain, while
overall the learners in their study
managed satisfactory explanations.

FIRST
POSITION
KRASHENS MINIMALIST POSITION

Macrory and Stone (2000)

reported that British
comprehensive school
students had a fairly good
explicit understanding of
the perfect tense in French.

FIRST
POSITION
KRASHENS MINIMALIST POSITION

Hu (2002) found that adult

Chinese learners of English
demonstrated correct
metalinguistic knowledge of
Prototypical rules of six English
structures but were less clear
about the peripheral rules for
these structures.

SECOND
POSITION

COMPREHENSIVE POSITION:

Teach the whole of the

grammar of the target
language

What then should

SELECTION
be based on?
inherent

LEARNING
difficulty

Learning difficulty can be based

on:
a.

The difficulty learners have in

understanding a grammatical
feature; and
b. To the difficulty they have in
internalizing a grammatical feature
so that they are able to use it
accurately in communication.

How then has

LEARNING
DIFFICULTY
been established?

frequency of
specific structures in the
input and their utility to
learners
factors such as the

have been invoked

(Mackey, 1976)

Two approaches to
delineate
cognitive difficulty:
a. Teach those forms that
differ from the learners first
language (L1).
b. Teach marked rather than
unmarked forms.

When should we

TEACH

GRAMMAR?

First, it is best to emphasize

the teaching of grammar in
the early stages of L2
acquisition.

Second, it is best to
emphasize meaning-focused
instruction to begin with and
introduce grammar teaching
later, when learners have
already begun to form their
interlanguages.

Behaviorist Theory:
error like sin needs to be
avoided at all costs
-- Brooks, 1960

Thus, it is necessary to ensure that

learners develop correct habits in the
first place.

N. Ellis (2005) has suggested

that learning necessarily
commences with an explicit
representation of linguistic
forms, which are then
developed through implicit
learning.

As for the second answer,

there
is ample evidence to show
that
learners can and do learn a
good deal of grammar without
being taught it.

Also, early interlanguage is

typically agrammatical. That is,
learners rely on a memorybased system of lexical
sequences, constructing
utterances either by accessing
ready-made chunks or by
simply concatenating lexical
items into simple strings.

Should Grammar
Teaching
BE MASSED
or DISTRIBUTED?

When grammar teaching

commences, we need to
consider whether it should
be concentrated into a short
period of time or spread over
a longer period.

Collins and colleagues report

their own study of three
intensive ESL programmes in
Canada,

One (the distributed

programme) taught over the
full 10 months of one school
year;
One (the massed programme)
concentrated into 5 months
but taught only to above
average students,

and the third (the massed plus

programme) concentrated
into 5 months, supplemented
without of class opportunities
to use English and taught to
students in mixed ability
levels.

The main findings:

The massed and especially
the massed-plus
outperformed the distributed
programme students on most
of the measure of learning

The study points to the need

for further research,
especially through studies
that compare massed and
distributed instructions
directed at specific
grammatical structures.

Should Grammar
Teaching
BE INTENSIVE
or EXTENSIVE?

Refers to instruction over a

sustained period of time
concerning a single
grammatical structure or
perhaps, a pair of contrasted
structures.

Refers to instruction
concerning a whole range of
structures within a short period
of time so that each structure
receives only minimal
attention in any one lesson.

Grammar teaching is typically

viewed as entailing intensive
instruction.
Example: The presentpractice-produce (PPP) model

It is assumed that with

sufficient opportunities for
practice, learners will
eventually succeed in
automatising the structures
they are taught.

As Ur says, the aim of

grammar practice is to get
students to learn the
structures so thoroughly that
they will be able to produce
them correctly on their own

Recent research (e.g., Spada &

Light bown, 1999) indicated that
even if learners are not ready to
learn the targeted structure,
intensive grammar teaching can
help them progress through the
sequence of stages involved in
the acquisition of that structure.

Earlier research showed that

learners do not always learn
what they are taught,
especially when learning is
measured in terms of
spontaneous production
(e.g., Kadia, 1987).

Intensive instruction also

helps learners to use
structures they have already
partially acquired more
accurately
(e.g., White, Spada,
Lightbown, & Ranta, 1991).

On the other hand, Extensive

grammar teaching can occur
within a learning activity, not
just as some kind of
postscript.

Cook (1989) has argued from

the perspective of universal
grammar that learners
require minimal evidence to
set a particular parameter for
the grammar they are
learning.

Loewen (2002) has shown

that even very brief episodes
of corrective feedback are
related to correctness on
subsequent tests.

Intensive instructive
instruction is time consuming
and thus, time will constrain
how many structures can be
addressed.

However, extensive grammar

teaching does not provide the
in-depth practice that some
structures may require before
they can be fully acquired.

Extensive instruction affords

the opportunity to attend to
large numbers of grammatical
structures. Also, many of the
structures will be addressed
repeatedly over a period of
time.

IS THERE ANY

VALUE

IN

TEACHING
EXPLICIT
GRAMMATICAL
KNOWLEDGE?

A distinction needs to be drawn

between explicit knowledge as
analyzed knowledge and as
metalinguistic explanation.

EXPLICIT
KNOWLEDG
E
consists of
the facts that
speakers of a
language

Analyzed
KNOWLEDG
E
entails a
conscious
awareness of
how a structural
feature works

Metalinguis
tic
EXPLANATI
ON
consists of
an
d
th
e
ab
ili
ty
to
u
n
de
rs
ta
n
d
ex
pl
an
at
io
ns
of
ru
le
s.

knowledge of
grammatical
metalanguage

Implicit
EXPLANATI
ON
is procedural,
is held unconsciously

a
n
d
c
a
n
o
n
l
y
b
e
v
e
r
b
a
li
z
e
d
if
it
i
s
m
a
d
e
i
m
p
li
ci
t.

Three separate questions to

consider:
1.
2.

3.

Is explicit knowledge of any value

in and of itself?
Is explicit knowledge of value in
facilitating the development of
implicit knowledge?
Is explicit knowledge best taught
deductively or inductively?

Is explicit knowledge of any

value in and of itself?

Studies by VanPatten and Oikennon (1996)

and Wong (2004) indicate that experimental
However,
Yuan
and
Ellis
showed
that
There
isthat
also
some
evidence
that
teaching
The
issue
related
to
the(2003)
question
Krashen
argues
that
learners
can
only use
groups
received
explicit
information
learners
grammatical
accuracy
improved
explicit
knowledge
by
itself
is not
effective.
explicit
knowledge
they
monitor,
concerns
the extent
to
which
learners
alone
performed
nowhen
differently
on
significantly
if and
they
had
time
for
on-line
which
required
that
they
are focused
on a in
interpretation
production
tests
than
are able
to use
their
explicit
knowledge
planning
while
performing
a narrative
task
form
and
havedid.
sufficient time
to access
the
control
group
actual
performance
knowledge.

Is explicit knowledge of value in

facilitating the development of
implicit knowledge?

Noninterface position (Krashen, 1981),

The
position
argues
that
explicit
explicit
Concerns
weak
what
interface
implicit
has become
position
knowledge
known
(Ellis,
are1993)
as the
and interface
thisand
conversion
occurs
by
priming
a
knowledge
becomes
implicit
knowledge
entirely
interface
claims
distinct
explicit
withknowledge
the
which
result
addresses
can
that
the
numberthat
ofhypothesis
key
acquisitional
processes,
inif
learners
have
the opportunity
for
explicit
role
convert
explicit
knowledge
into
knowledge
implicit
cannot
knowledge
plays
be in
converted
L2
if plentiful
the
particular
noticing
and
noticing
the
gap
communicative
practice.
into
acquisition.
implicit
knowledge.
learner
is ready
to acquire the targeted
(Schmidt,
1990).

Is explicit knowledge best

taught deductively or
inductively?

While Rosa and ONeill (1999) found no

significant
difference
in
In
Herron
inductive
andof
Tomosello
teaching,
learners
(1992)
found
areisfirst
a
A
studies
have
examined
the
Innumber
deductive
teaching,
aeffectiveness;
grammatical
The
question
assumes
that
there
value
Erlams
(2003)
own
revealed
a how
exposed
clear
advantage
to
exemplars
for study
inductive
of
the
grammatical
instruction;
relative
effectiveness
of
these
two then
structure
is
presented
initially
and
in
explicit
knowledge
and
addresses
significant
advantage
for
Robinson
(1996)
found
that
a group
deductive
structure
and
are
asked
tothe
arrive
at a
approaches
to
teaching
explicit
practiced
in
one
way
or
another.
best
to teach
it.
receiving
deductive
instruction
metalinguistic
approach
was
more
generalisation
effective
on their
knowledge.
The
results
have been
mixed.

Is
there
a
best
way

For

IMPLICIT

t is
i
,
n
o
i
t
s
e
u
q
e
h
t
r
e
w
s
n
a
To
the
y
f
i
t
n
e
d
i
o
t
necessary
or
f
s
n
o
i
t
p
o
l
a
instruction
:
r
a
m
m
a
r
g
teaching
d
e
s
a
b
t
u
p
1. In
d
e
s
a
b
n
o
i
t
c
2. Produ
instruction
i ve
t
c
e
r
r
o
c
f
o
3. Types
feedback

002b)
2
,
8
9
9
1
,
7
9
(Ellis 19

INPUT-BASED INSTRUCTION
It is based on a computational
model of L2 acquisition,
according to which
acquisition takes place as a
product of learners
comprehending and
processing input.

INPUT-BASED INSTRUCTION
VanPatten (1996, 2003) has
developed a version of the
input-based option that he
calls input processing
instruction.

INPUT-BASED INSTRUCTION
This is directed at helping
learners to overcome the
default processing strategies
that are a feature of
interlanguages

OUTPUT-BASED
INSTRUCTION

This can be found in both skillbuilding theory or in sociocultural

theory of L2 learning, according to
which learning arises out of social
interaction which scaffolds
learners attempts to produce
new grammatical structures.
--(Ohta, 2001)

CORRECTIVE FEEDBACKS
However,
(2000)
noted
This
type
of Murnoi
feedback
is more
EXPLICIT
FEEDBACK
Some
descriptive
studies
Input-based
feedback
models
Key
options:
IMPLICIT
FEEDBACK
Or,
as
inimplicit
this
contrived
example,
a
NNS:
Why
he
is
very
unhappy?
- Direct
that,
feedback
is
correction
or
metalinguistic
the correct
form
for
the learner.
compatible
with
the
focus-on-form
have
shown
that
outputrequest
clarification:
a. explanation.
Whether
feedback
iswhen it
NS:
Why for
isthe
he
very
unhappy?
probably
more
effective
approach
because
it
ensures
that
based
feedback
is
more
implicit
explicit
and;
There
are or
some
evidence
that
Occurs
when
the
corrective
of
is targeted
intensively
atforce
aexplicit
NNS:
Yeah
why
is
very
unhappy?
Output-based
feedback
elicits
learners
are
more
likely
to
stay
likely
to
lead
to
learners
feedback
is
more
effective
in
both
the
response
to
learner
error
is it
b.
Whether
preselected
the
feedback
than
when
is
input
NNS:
Why he
isform
very
unhappy?
--(Philp, 2003)
production
of
the
correct
form
eliciting
learners
immediate
correct
focused
on
meaning.
correcting
their
own
initial
masked,
for
example,
a
recast,
occurs
extensively
or
output
based. in incidental
NS:
Sorry?
use
ofreformulates
the structure
and
eliciting
from
the
learner
which
a
deviant
erroneous
utterances
in what
focus on form.
Inuse,
the for
latter,
subsequent
correct
example
in
NNS:
Why
is
he
very
unhappy?
utterance
correctingto
it form
while may be
explicit
attention
isa post-test
referred to as uptake.
keeping
the
same
meaning:
more effective.
(Carroll
& Swain 1993; Lyster, 2004).

In
Should
Separat Gramma
e
r
Lessons be
taught

FOCUS
ON
FORM
S

Refers
to instruction
This approach
involves
involving
structureteaching agrammar
in a
of-the-day
approach,
series of separate
where
lessons.the students
primary focus is on
form and where the
activities are directed
intensively at a single
grammatical structure.

FOCUS
ON
FORM
(FonF)

Long
(1988,
1991)
This
focus
be and
It
In
Focus
at entails
this
the
on
same
approach,
form
acan
focus
time
implies
on
they
Doughty
have
planned
, (2001)
where
a
meaning
attention
no
are
separate
engaged
with
to grammar
the
with
attention
argued
that
focused
taskrather
isand
required
to formstrongly
predetermined
lessons
understanding
but
arising
out
of
focus
on form
is best
to
occasions
for
theelicit
grammatical
grammar
producing
communicative
teaching
meaningful
structures
equipped
to into
promote
using
predetermined
activity.
will
integrated
messages.
also
be
intensive.
a
interlanguage
grammatical
structure,
curriculum consisting
of
development
because
the
as,
for example,tasks.
in
communicative
acquisition
of
implicit
Samuda (2001).
knowledge occurs as a
result of learners
attending to linguistic

FOCUS
ON
FORM
S

However,
other research
This approach
have
argued that
acknowledges
theavalue
focus-on-forms
approach
of teaching explicit
is
effective.
DeKeyser
knowledge
and
(1998),
for example, has
subsequently
argued
that grammatical
proceduralising
it by
structures
are learned
means of activities
gradually
the
(drills andthrough
tasks) that
automatisation
of
practice behaviours.
explicit
knowledge
Rather than
structures.

TYPE
1. Focus on
forms
2. Planned
focus on
form
3. Incidental
focus on
form

PRIMARY
FOCUS

DISTRIBUTI
ON

Form

Intensive

Meaning

Intensive

Meaning

Intensive

Ellis (2001, p.1)

CONCLUSION

Although there is now clear

conviction that a traditional
approach to teaching grammar
based on explicit explanations
and drill-like practice is
unlikely to result in the
acquisition of the implicit
knowledge needed for fluent
there
continues
to
be
a
and accurate communication
disagreement regarding what

The following are

the beliefs of the
author as regards
to
grammar teaching

7.
A
case
exists
for
teaching
8.
An
incidental
focus-onIn
the
case
of
a
focus6.
Use
should
be
made
of
9.
Corrective
feedback
is
1.The
grammar
taught
5.
Consideration
should
beto
4.
A
focus-on-forms
10.Grammar
In accordance
2.Teachers
should
3.
is
best with
taught
explicit
grammatical
form
approach
is as
of
special
on-forms
approach,
a as
both
input-based
and
important
for
learning
should
be
one
that
given
to
experimenting
approach
is
valid
long
these
beliefs,
grammar
endeavour
to
focus
on
learners
who
have
already
knowledge
as
a
means
of
value
because
it
affords
an
differentiated
approach
output-based
instructional
grammar.
It
is
best
emphasise
not
just
form
with
a
massed
rather
it
includes
an
opportunity
instruction
should
take
the
acquired
some
ability
tothan
use
those
grammatical
assisting
subsequent
opportunity
for
extensive
involving
sometimes
options.
conducted
using
aare
mixture
but
also
the
meanings
and
distributed
approach
to
for
learners
to
practice
form
of separate
grammar
the
language
rather
than
to
structures
that
known
acquisition
of
implicit
treatment
of
grammatical
deductive
and
sometimes
of
implicit
explicit
uses
different
teaching
grammar.
behaviour
in should
communicative
complete
beginners.
lessons
and
be
to
beof
problematic
to
knowledge.
Teaching
problems.
inductive
instruction
may
feedback
types
that
grammatical
structures..
However,
grammar
canare
be to
tasks.
integrated
into
learners
rather
than
try
explicit
knowledge
can
be
work
best.
taught
through
corrective
both
input
based
and
communicative
activities.
teach
the
whole
grammar.
incorporated into both a
feedback
as soon as
output
based.
focus-on-forms and a
learners begin to use the
focus-on-form approach.
language productively.

But what
is important
The
acquisition
of the is to
recognize what
options
areL2
grammatical
system
of an
available,
what
the and
is
a complex
process
theoretical
rationales
for
almost
certainly
can be
these
options
are,
and
what
assisted best by variety of
the
problems
are
with
these
approaches
rationales.

Thank you
for
listening!