Teaching Grammar Searchable
Teaching Grammar Searchable
5. Introduction:
Teaching Grammar
While some practitioners and theorists have regarded it as axiomatic that grammar
teaching is central to effective language teaching, others have seen at best a minimal
role for grammar teaching, and a variety of positions between these two extremes have
been proposed. This chapter considers the theoretical arguments and empirical
evidence cited in support of the various approaches to the teaching of grammar which
have been advocated over the years. The chapter considers the main arguments in
relation to the necessity of teaching grammar, and discusses the relative merits of
proactive and reactive grammar teaching, input-based and output-based approaches,
and implicit and explicit teaching. The chapter also outlines the factors we need to
consider when deciding on an appropriate approach to teaching grammar, and
concludes, along with Ellis (2006), that what is fundamentally needed is a broad view of
what constitutes grammar teaching and a context-sensitive approach to the issue.
1. Introduction
For those outside the field of language teaching, the degree of controversy which
surrounds the teaching of grammar probably comes as a surprise given that, for many,
learning a language is almost synonymous with learning grammar. The uncertainty
around grammar teaching is, however, emphasised by Borg (1999: 157): "In ELT,
grammar teaching clearly constitutes [...] an ill-defined domain: the role of formal
instruction itself has been a perennial area of debate, and more than 20 years of research
have failed to yield firm guidelines for grammar teaching methodology". There is
certainly no doubt that grammar teaching continues to excite intense and even fractious
debate (see for example the exchange between Ellis (2006b) and Sheen (2006) in the
pages of TESOL Quarterly). The aim of this chapter is to add some definition to what
Borg (1999: 160) describes as the "ill-defined domain" of grammar teaching. This will
not, however, take the form of definitive answers to questions about grammar teaching:
as we shall see, language teaching research is some way from being able to offer
conclusive evidence in favour of particular approaches or positions. What I hope to
offer, however, is a set of options which will be useful in developing an informed and
principled approach to grammar teaching, and a set of factors which need to be
considered in considering those options. The chapter begins by addressing the
fundamental question of whether we need to teach grammar at all. The chapter then
discusses a number of different approaches to teaching grammar. The approaches are
discussed separately for the sake of clarity, but, as will become evident, the suggestion
is not that they are mutually exclusive, despite what some of their more extreme
Teaching Grammar
120
121
Ivor Timmis
advocates may say. Finally, the factors which might influence the approach or particular
combination of approaches to teaching grammar we might adopt are discussed.
123
Ivor Timmis
Teaching Grammar
In terms of depth of treatment, Ellis (2006: 94) notes the advantages of intensive
grammar teaching involving practice: "[...] recent research (e.g. Spada/Ltghtbown
1999) indicates 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".
that what they term "integrated form-focused instruction" with young learners led to
more successful acquisition of certain forms than purely meaning based instruction with
no focus at all on grammar. Spada and Lightbown (2008: 191) cite a number of other
studies which, they claim, offer support to "the hypothesis that attention to language
form within the context of communicative practice can lead to progress in learners'
language development. Although this progress has been observed in the short term for
most studies, long-term improvement has also been reported [...]".
122
7. Input-based Approaches
A further choice facing teachers is that between input-based approaches and outputbased approaches. As we noted above, output-based approaches tend to be associated
with proactive approaches, but this is not a necessary equation: input-based approaches
Teaching Grammar
124
125
Ivor Timmis
can also be proactive. Input-based approaches, as Ellis (2006; 98) summarises "seek to
draw learners' attention to the targeted structure(s) in one or more ways: simply by
contriving for numerous exemplars of the structure(s) to be present in the input
materials, by highlighting the target structure(s) in some way (e.g. by using bold or
italics in written texts), or by means of interpretation tasks [...] directed at drawing
learners' attention to form-meaning mappings". The theoretical rationale for inputbased approaches rests on a computational model of language acquisition in which
acquisition occurs as a result of processing input (Ellis 2006). It could also be argued
that an element of the rationale for input-based approaches lies in the negative evidence
that intensive practice of a structure does not generally lead to immediate automatic
control in spontaneous production. An area of primary interest in input-based
approaches is the relationship between 'input' - the language the learner is exposed to and 'intake' - that part of the input which becomes part of the learner's potential
productive repertoire. Schmidt (1990) argued that the crucial link between input and
intake was the process of "noticing", whereby learners become conscious of the gap
between what they are able to produce and the second language input. In other words,
learners need to pay conscious attention to form in input in order for it to become
available as output. Approaches which emphasise noticing from input have a good deal
in common with "consciousness-raising" (Willis 1996) and "language awareness
approaches" (Bolitho et al. 2003): there is an emphasis on drawing students' attention to
a contextualised target feature and asking them to 'discover' its form, meaning and use
inductively. Students are not necessarily required to produce the target form.
8. Output-based Approaches
Output-based approaches are of two distinct kinds: those in which output is
fundamentally about practising a particular target feature or features intensively, and
those in which the output is seen as a general stimulus to acquisition where different
learners might actually acquire or notice different features in the course of the same
activity. The case for output as practice of a particular feature reflects the view of
language as essentially a skill. The argument (DeKeyser 2003) is that learners can,
through intensive meaningful practice, develop automatic control of structures which
have initially been learned explicitly. It is important to emphasise here that the practice
needs to be meaningful - practice is not just a matter of quantity, but also of quality. As
Swan (2005: 383) notes, it is the simplistic equation of 'practice' with the behaviourist
drilling which characterised the audiolingual method that has perhaps led to an
indiscriminate condemnation of practice: "The fact that systematic practice is associated
with 'discredited' behaviourist theory, and with a short-lived fashion for exclusively
mechanical structure-drilling which perhaps only achieved 'false automatization' (Ellis
2003: 105), has led many scholars to dismiss its use as irrelevant to acquisition".
Discrete item focused practice is not, however, the only kind of output which has
been advocated for grammatical development. There is an argument that interaction of
the right kind can itself contribute to grammatical development (Long 1991) and an
argument that activities which reflect the nature and role of grammar in communication
can promote acquisition (Thornbury 2001; Cullen 2008). Social interaction can itself
126
Teaching Grammar
Ivor Timmis
Level
Opinion is divided as to whether beginner programmes should feature intensive
treatment of basic grammatical structures or whether they should accelerate learners to a
communicative threshold through vocabulary teaching and communicative practice
(Ellis 2006). It seems reasonable to suppose that learners' aims should play a role here
and that there will be more of a place for intensive grammar teaching in longer term
programmes. There is also a case, I would argue, that as learners move up the levels
there will be a stronger tendency to use reactive approaches as the learners are more
likely to have encountered a particular item previously and perhaps even to have
received explicit instruction on the point.
127
Learners' Expectations
Borg cites an interesting argument put forward by one of his research subjects.
This respondent argued that he taught grammar, not because he believed it made much
difference to his learners' communicative ability, but because meeting their
expectations in this regard motivated them more for other aspects of the course. Willis
(2003) makes a similar point in relation to grammar practice, emphasising that it seems
to meet a psychological need. While a programme based only on meeting learners'
expectation would produce stasis, there would seem to be a case for taking account of
these psychological needs.
Learner Abilities
Spada and Lightbown (2008) suggest that learners' literacy in their first language
and knowledge of grammatical terminology will influence how far we can adopt
explicit approaches. We also noted above the case that analytical learners may be better
able to cope with inductive approaches.
12. Conclusion
In conclusion, it can be seen, as Borg (1999: 160) writes, that grammar teaching is
"a complex decision-making process, rather than the unthinking application of a best
method". One of the factors that makes the decision-making process complex is the
range of options available if one takes a broad view of what constitutes grammar
teaching such as that proposed by Ellis (2006: 84): "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". In deciding on an
appropriate option, teachers then have to consider a range of factors, as we have seen
above, in relation to the learners and the structure to be taught. A further complicating
factor is the state of the research evidence. As Sheen (2003: 227) puts it, "our
knowledge of the nature of the process of classroom SLA is so limited that
theoretically-driven advocacies are not sufficient to justify unquestioned acceptance"
(my italics). The empirical evidence, on the other hand is inconclusive. It neither
sanctions a "back to basics" approach, as Spada and Lightbown (2008) note, nor, as
Swan (2007) argues, does it suggest that such approaches should be outlawed. As Swan
(2007) points out, proactive explicit approaches are often caricatured, but there is no
reason why they should involve endless repetition and meaningless drills. What the
research evidence does sanction, I would argue, is 'principled eclecticism'. Teachers
need to be critically informed about research findings and able to use this information to
guide their choice of approaches. The challenge is essentially to match the approach
with the type of learner and the type of language structure. It is a challenge which is
considerably easier to state than to meet, but that is the nature of language teaching and
therein lies its enduring interest.
Bibliography
Bley-Vroman, Robert (1989): What is the Logical Problem of Foreign Language
Learning? In: Gass, Susan/Schacter, Jacquelyn (eds.): Linguistic Perspectives on
Second Language Acquisition. New York: CUP, 41-68.
Borg, Simon (1999): Teachers' Theories in Grammar Teaching. In: ELT Journal 53(3),
157-167.
Cameron, Lynne (2001): Teaching Languages to Young Learners. Cambridge: CUP.
Cullen, Richard (2008): Teaching Grammar as a Liberating Force. In: ELT Journal
62(3), 221-230.
DeKeyser, Robert (1998): Beyond Focus on Form: Cognitive Perspectives on Learning
and Practicing Second Language Grammar. In: Doughty, Catherine/Williams,
Jessica (eds.): Focus on Form in Classroom Second Language Acquisition.
Cambridge: CUP, 42-63.
DeKeyser, Robert (2003): Implicit and Explicit Learning. In: Doughty, Catherine/Long,
Michael (eds.): The Handbook of Second Language Acquisition. Maiden, MA:
Blackwell, 313-348.
Ellis, Rod (1997): Second Language Acquisition. Oxford: OUP.
Ellis, Rod (2003): Task-based Learning and Teaching. Oxford: OUP.
Ellis, Rod (2006a): Current Issues in the Teaching of Grammar: An SLA Perspective.
In: TESOL Quarterly 40(1), 83-107.
Ellis, Rod (2006b): The Author Replies: A Balanced Perspective: A Reply to Ron
Sheen. In: TESOL Quarterly 40(4), 833-837.
129
Teaching Grammar
Ivor Timmis
128
Rules,
Patterns
and
Words:
in
ELT.
ML ^
G r a m m a r JVsn-liinfi A p p r o a c h e s
.
Proactive Grammar Teaching
-> PPP (Presentation-Practice-Production): grammar is presented by the teacher, then practised
and produced by the learners (a traditional approach)
Reactive Grammar Teaching
Long (1982): "focus on form" (proactive planned form of instruction) vs. "focus on forms"
-> grammatical items are dealt with when they create difficulty in a communicative activity
Planned Reactive Grammar Teaching
-> the teacher anticipates what students will have difficulties with and plans tasks accordingly,
yet focusing primarily on meaning
-> represents a third "alternative" approach (Spada/Lightbown 2008)
Input-based Approaches
-> theory: computational model of language acquisition -> acquisition occurs as a result of
processing input (Ellis 2006); the teacher seeks to provide rich input
Output-based Approaches
theory: learners can develop control of structures which have been learnt explicitly through
intensive meaningful practice (DeKeyser 2003); die teacher provides productive activities
-> grammaticisation tasks which ask learners to 'add' grammar to a lexical skeleton
Explicit Versus Implicit Teaching
explicit knowledge = conscious knowledge of grammar (learnable and verbalisable)
implicit knowledge = procedural (can only be verbalised if it is made explicit)
1. Introduction
Comfort Grammar assumes that learners aged 15+ and upwards have experienced
proactive grammar teaching within a school system, which has given them a grounding
in the forms and structures of English, but left them with further work to do.
Unpublished research by MA students in Klagenfurt, Baden and Salzburg (BIFIE:
2007-2010) working to produce empirical data on the language levels of Austrian
school-leavers (18+) has indicated Common European Framework of Reference
(CEFR) levels of Bl to B2, a lower level than commonly assumed. Inaccurate tense use
being one element of this performance, Comfort Grammar aims to gradually eradicate
the errors arising in live production caused by the dual burdens of mother tongue
interference and tense construction overload. It does not seek to cover all error types nor
to make the standard grammatical explanations for tense use more palatable by, for
example, using humorous examples (Bosewitz/Kleinschroth 2002) or by attempting to
enter learners' minds through an organic, colour-coded combination of text and image
(Warr 2003). In part, it mirrors the process undergone by Emmerson (2002, 2006) in
reducing the original Business Grammar Builder (implicitly already a reduction of the
full canon) to the 'Essential' version, where essential is taken to mean greater focus on
business examples rather than distilled into a shorter version.
132
Anthony Hall
133
English tense use has the distinct difficulty that it incorporates the notion of "personal
attitude", as part of the verb's meaning. Therefore learners must be made aware that
their choice of tense has functional and attitudinal overtones. It is not just a descriptor of
time.
The order in which the tense choices below are presented reflects both the
regularity with which the learner is faced with the particular choice, and,
correspondingly, when an error occurs, the probable frequency with which their
instructor needs to refer to it in their corrective feedback. This does not mean that they
need necessarily be taught in this order, and certainly not that they should be taught en
bloc.
Tell a story
Past
Discuss something
Present
Anthony Hall
134
The learner needs to know that at any given point in the conversation their choice of
tense is their decision. They must decide what they want to do. Is it to tell a story, or is
it to discuss something? Making a diagram of this choice reinforces the reality of that
choice, and effectively of the power they have as one of two people in a conversation.
The diagram (Figure 1) is deliberately simple so that it can be drawn quickly whenever
necessary. It is accompanied orally by the additional information that a) you use the past
tense to begin the story, and stay in the past tense until the story is complete; and b) to
discuss something, you use the present tense.
Figure 2 shows the next step in which the function of the present perfect tense is
added to the diagram.
Tell a story
Discuss something
Past
Present
4
Present Perfect
Figure 2: The bridging function of the Present Perfect
Again, the drawing is accompanied by an explanation: The present perfect tense is a
bridge between the two functions and times. It allows the speaker to include
experience, facts, or effects from the past within their discussion or elaboration.
Managing this separation and mixing of the tenses in open, free conversation is not
easy. The tense system of the learner's first language is happy to interfere. Rather than
binary, contrastive exercises, extended opportunities for free dialogue must be given in
which the learners exercise their understanding. Initially, the stronger member of a pair
will dominate the choice of topic and tense. This is due in part to the weaker member
remaining the more passive partner but also to the Residual Speaker Syndrome, which
posits that once a speaker is in control of the dialogue it requires some effort for them to
relinquish it, since, as the last person to have spoken, responsibility for filling the
silence of a pause falls back on them. However, the weaker member will eventually
decide they have the confidence to help in contributing to the course of the dialogue
(Krashen 1981). Error-free performance is not the aim; rather the willingness to provide
spoken material that is personal and meaningful. Written assignments can also be used
to give learners the opportunity to write their own sentences or mini-stories.
Finally, in the third step, the other tenses are added to the diagram (Figure 3). The
allocation of the majority of the tenses to the discussion function is again a
simplification, but highly motivating. An explanation can be given that the Past Perfect
relates to the Past in the same way as the Present Perfect to the Present. The use of the
term Conditionals may seem surprising, as many linguists would not classify
conditionals as a "tense". Comfort Grammar uses the term nevertheless since it is a
form which expresses the notion that two ideas or activities are in a relationship with
each other, i.e. one is conditional upon the other. Secondly, the term fits the function of
discussion or deliberation exactly. Many learners give this area of grammar the name
"if-clauses", a conceptual simplification which will be discussed further in chapter five.
TELL A STORY
DISCUSS SOMETHING
Past
Past Perfect
135
Anthony Hall
interpretation for the attitudinal aspect to be revealed. In "The Difference Box", Newby
(1989: 88) gives the explanation that with activity verbs the - i n g form is nearly always
used, with the simple form only being used on rare occasions to draw particular
attention to the length of the activity (author's translation and juxtaposition). This is the
reverse of the Comfort Grammar view that the simple tense presents the unmarked case.
For learners who have been able to convert grammatical samples, explanation and
practice into consistently accurate production, Comfort Grammar may be superfluous or
at best an add-on. For those who switch off at the thought of grammar it is a relief that
its explanations are brief. For the large group of learners between these two poles it is a
positive experience to be offered a skeletal, summary plan or map which they can
follow. They are using grammar to discuss alternatives, and to enrich their quality of
expression. This bears comparison with the way they learnt their first language, and
learners respond appreciatively to the notion that self-generated dialogue is a powerful
medium for improvement, especially for those who use the present tense to cover
present, past and future.
occurred" (2002: 63ff) represents a more complex task than summing the choices onto a
single scale from 0 to 100. Placing the tenses and their perceived attitudes on this scale
is a productive pedagogical task, encouraging debate and interaction, which in turn
blurs the boundaries between input-based and output-based approaches. I am indebted
to Prof. D. Gabrovsek, Ljubljana University, for introducing me to the concept of clines
of meaning at a staff seminar in Klagenfurt in 2007. Applying the concept to functional
semantic options was my own idea. Newby also distinguishes three basic positions, with
the German terms, Ungewissheit (Weak Attitude) and Uberzeugung (Strong Attitude)
offering a stronger conceptual polarisation around the central Neutrale Haltung (Neutral
Attitude) (1989: 93-97).
136
X
0%to20%
firm intention
X
40% to 60%
The Added Meanings outlined below are thus a response to the unasked question:
"How firm or concrete is the plan?" Their position on the cline may be elicited,
presented explicitly, or deduced. Students should justify their decisions. For the EFL
classroom this means moving away from textbook-guided teaching with its emphasis on
storylines and characters, on role play and unconvincing texts, and on right and wrong
answers, gap-filling, and other puzzle-style grammar practice. In its place an ad hoc
discussion is held about differences in meaning - in the sense of Sinclair's grammar of
meaning - with samples evolving out of a dialogue, and practice occurring through
continuation and completion at the learners' level of performance within a context of
their own making.
When talking about a future action the speaker expresses, through his or her tense
selection, an attitude concerning the certainty of that action. In Comfort Grammar terms
the future clearly falls under the category of discussing rather than telling a story, but
what is its particular contribution? The functional description proposed here is that of
informing, of telling the story before it happens. We select from the different future
forms the one that best expresses our understanding of, reasons for, and wishes and
beliefs about what is going to happen in the shorter or longer term. This includes saying
what we can or cannot do and what may or may not be possible. This puts our
conversation partner(s) in the picture, and they can respond by mirroring our view or by
modifying it. Thus, they may choose to confirm our assessment by using the same tense
form, or they may take a different view, and hence opt for a different future form. The
range of choices is presented as a cline of definiteness, as shown in Figure 4.
vague wish or belief
137
fixed arrangement
X
9 0 % to 100%
"We'll do it (next week)". Without a time marker, the simple 'will' future form
constitutes an offer, and with the time marker a promise, and we assume the speaker is
being genuine, hence the 90%+ certainty rating. However, offers and promises do fail to
materialise, because we cannot fully control the future. Illness, commitments, the
weather, the economy, mobile phones, these all conspire to reduce the certainty that the
job will get done.
"Don't worry, this time next year we'll be eating caviar". This use of the 'will
future' places the time focus clearly in the future, though without substance, in order to
reassure. The use of the continuous form has the further effect of adding intensity by
transporting the listener 'live' to the action in that distant future.
Anthony Hall
In a further activity suitable for the E F L classroom, we can first interpret the pragmatics
of the tense selected, and then use the cline as a prompt to propose alternative responses
to those given.
138
(i)
"I have an appointment at ten o'clock" (= pragmatically, "I can't see you
before midday"). Response: "OK, so I'll see you at one".
(ii) "The plane arrives at four in the afternoon" (= pragmatically, "can you pick
me up?"). Response: "Oh dear, I don't finish work till five".
(iii) "When is she coming next?" (Straight request for information). Response:
"We'll arrange that at the first meeting" (Effectively, "I don't know, but I will
do tomorrow". Thus the exchange postpones the point of time at which
knowledge or certainty will be achieved.)
The function of informing and the cline of definiteness imply involvement and
dependence on what is still to happen, with the aim of managing its effect on the present
or the future. If we adopt the perspective that the future is "what has not happened yet"
(Ward/Woods 2007: 36) we might be more tentative and talk of clines of likelihood or
probability. This in turn would lead to the inclusion of verbs such as to doubt, believe,
hope, trust, understand, etc., of adverbial qualifications (possibly, probably), of
negatives and of modals and of verbal structures after likely and unlikely. Altogether
that would probably overload the cline, reducing the clarity of and separation between
the three meanings and consequently its usefulness to learners.
5. Practising Conditionals
Connecting Thinking to (Un)likelihood
If tense use in a single verb sentence has its pitfalls, then a conditional sentence
which links two ideas, and contains two verbs, must be twice as tricky. No wonder then
that learners place "if-sentences" high on their list of unfathomable English grammar. In
Comfort Grammar this association of " i f with previous learning difficulty is broken by
listing it as a vertical image (see Figure 5), with three other words which introduce
conditions, namely once, when, and whenever. For "comfort" reasons (brevity,
memorability, and avoidance of the additional complexity introduced by the negative),
if not, unless, as long as, every time etc. are not included in the list at this point.
The teacher can then explain that each of the four link terms adds its own semantic
perspective on how the two parts of the sentence are linked, as shown:
once
when
if
whenever
a time restriction
a general truth
all other cases of likelihood
repeated event
139
140
Anthony Hall
141
7. Conclusion
This article has introduced a methodology for EFL and ESP which addresses both
the what and the how of foreign language teaching, while making the learners partners
in the process. The purpose of Comfort Grammar is thus three-fold. For those learners
effectively without or with only rudimentary grammatical knowledge it offers a concise,
summarising, alternative route to understanding verb use in English. For other learners
who have already been taught grammar to varying levels of expertise it offers a
different, simplifying perspective from which to reappraise their existing skill and
upgrade their language performance. Finally, for the classroom teacher or private
language trainer engaged in open-ended, dialogic teaching it is a case-based means of
responding supportively to error, ambiguity, and ambivalence in their learners'
language expression.
Bibliography
Aitken, Rosemary (2002): Teaching Tenses. Brighton: ELB Publishing.
Bosewitz, Rene/Kleinschroth, Robert (2002): Joke Your Way Through English
Grammar. Berlin: Rowohlt.
Byrne, Donn (1976): Teaching Oral English. London: Longman.
Emmerson, Paul (2002): Business Grammar Builder. Oxford: Macmillan Education.
Emmerson, Paul (2004): Email English. Oxford: Macmillan Education.
Emmerson, Paul (2006): Essential Business Grammar Builder. Oxford: Macmillan
Education.
Hall, Anthony (1994): Writing Computer Science Texts in English. In: Gyorkos, Joszef/
Krisper, Marjan/Mayr, Heinrich (eds.): RE '94 The Third Conference on Reengineering of Information Systems. Ljubljana: SRC, 85-96.
Hall, Anthony (2007): Who Knows Donovan? Developing Learner Dialogue in ESP
Environments. In: Delanoy, Werner/Helbig, Jorg/James, Allan (eds.): Towards a
Dialogic Anglistics. Wien: LIT, 193-206.
Hall, Anthony (2010): Information Structure in Writing: A Teacher's Look Across
Languages. In: Doleschal, Ursula/Mertlitsch, Carmen (eds.): Wissenschaftliches
Schreiben 2: Writing Across the Curriculum. Klagenfurt: Sisyphus.
Havranek, Gertraud (2002): When is Corrective Feedback Most Likely to Succeed? In:
International Journal of Educational Research 37, 255-270.
Krashen, Stephen (1981): Second Language Acquisition and Second Language
Learning. Oxford: Pergamon.
Newby, David (1989): Grammar for Communication. Wien: Osterreichischer
Bundesverlag.
Sinclair, John/Fox, Gwyneth (1990): Collins Cobuild English Grammar. London:
Harper Collins.
Sinclair, John (2004): Trust the Text. Abingdon: Routledge.
Summers, Della/Bullon Stephen (2003): Longman Dictionary of Contemporary English.
Harlow: Pearson Education Limited
Thornbury, Scott (2005): Beyond the Sentence: Introducing Discourse Analysis. Oxford:
Macmillan Education.