How to Teach Grammar

teach
mar?

m The case against grammar

e Grammar and methods
El Grammar now
'iii Basic principles for grammar teaching

Attitudes to 1n 1622 a certain Joseph \Vebbc, schoolmaster and textbook 1vriter, wrote:
grammar 'No man c<Ln run speedily to the mark of language that is shackled .. with
grammar precepts.' He maintained that grammar could be picked up
through simply communicating: 'By exercise of reading, writing, and
speaking aU things belonging to Gnunrnar, will without labour, and
whether we ·will or no, thrust themselves upon us.'
Vv'ebbe was one of the earliest educators to question the value of grammar
instruction, but certainly not the last. In fact, no other issue has so
preoccupied theorists and practitioners :l.S the grammar debate, and the
history of language tc<:~ching is essentially the history of the claims and
counterclaims for and against the teaching of grammar. Differences in
attitude to the role of grammar underpin differences between methods)
between teacher~;, and between learners. It is a subject that eyeryone
involved in language teaching and learning has an opinion on. And these
opinions arc often strongly and uncompromisingly stated. Here, for
example, are a number of recent ~taternents on rhc subject:
'There. is no doubt that (l knmdedg;e- itnplici.t or expLicit- of
grammatical rules is essential {or the mastery of a language.'
(Penny Ur, a teacher tntiner, ;u1d author of Gmmmar Pmrtice Acti1.1itie.r)
'The effects of grarmnar teaching ... appear to be per~pheral and fr;1.gile.'
(Stephen Kraehcn, an ini1ucntial, if controversial, applied linguist)
'A sound knowledge of grammar is essential if pupils are going to use
Eng-li~b crc:1tln-h·.'
(1l;m I-·Iutchinsc;n, a courschod:. writer)
'Grammar is not \·cry important: The nujority of langu~>gcs haYe a very
compkx gramm:H. English has little gmmrnar and consequently it is
not n-ry impurt;lnt to understand it.'
(From the publicity of a London bngu:1gc school)

14
 2 * 1/Vhy teach grammar?

'Grammar is not the basis ofhnguage acquisiti.on, and the balance of

linguistic research clearly invalidates any view to the contrary.'
(I\-·1ichae1 Lewis, a popular \>vTiter on teaching methods)
Since so little is known (still!) about how languages are acquired, this book
wiJl try to avoid taking an entrenched position on th'.: iss\.\e. R·athcr, by
sifting the arguments for and against, it is hoped that readers will be in ;l
better position to make up their own minck Let's first look at the c<J.se for
grammar.

The case for There are many arguments for putting grammar in tlw foreground in second
grammar language teaching. Here are seven of them:

The sentence-machine argument

Part of the process of langu<~ge learning must be what is sometimes called
item-learning- that is the memorisation of individual items such as words
and phrases. However, there is a limit to the number of items a person can
both retain and retrieve. Even travellers' phrr~:oe books have limited
usefulness- good for a three-week hdiday, but there comes a point where
we need to learn some patterns or rules to enable us to genenne new
sentences. That is to say, grammar. Grammar, after all, is a description of
the regularlties in a langu',lge, and knowledge of these rcgul-.lritics provides
the learner with the means to generate a potentially enormous number of
original sentences. The number of possible new sentences is consHained
only by the vocabuL1ry at the learner's command ;md his or her cre;nivity.
Grammar is a kind of 'sentence-making machine'. It fOllows thM the
teaching of grammar offers the learner the means for potentially limitless
linguistic creativity.

The fine-tuning argument

As we saw in Chapter 1, the purpose of grammar seems to be to dlow for
greater subtlety of meaning than a merely lexical system em cater for. \Vhilc
it is possible to get a lot of communic:ltivc mileage out of simply stringing
words and phrases together, there comes a point where 'J\1e Tarzan, you
.Jane'-type language fails to deliver, both in terms of intelligibility and in
terms of appropriacy. This is particularly the case for wrinen language,
which generally needs to be more explicit than spoken language. For
example, the following errors are likely to conl~tse the reader:
last Monday night I was boring in my house.
After speaking a lot time with him! thought that him attracted me.
We took a wrong plane and when I saw it was very later because the plane
took up.
Five years ago I would want to go to India but in that time anybody of my
friends didn't want to go.
The teaching of grammar, it is argued, scJTes as a corrccti\·c ag~1insr the kind
of Jmbiguity rcpre~cntcd ln the~e cxamt,lc5.

15
How to Teach Grammar

The fossilisation argument

It is possible for highly motivated learners with a particular aptitude for
languages to achieve amazing levels of proficiency without any formal study.
But more often 'pick it up as you go along' learners reach a language
plateau beyond which it is very difficult to progress. To put it technic.clly,
their linguistic com~Jetence fOssiliscs. Research suggests that learners who
receive no instruction seem to be at risk of fossilising sooner than those
who do receive instruction. Of course, this doesn't necessarily mean raking
formal lessons ·- the grammar study can be self-directed, as in this case
(from Christopher Isherwood's autobiographical novel Christopher and his
Kind):
Humphrey said suddenly, 'You speak German so well- tell me, why don't
you cn~r usc the subjunctive mood?' Christopher had to admit that he
didn't kno\v how to. In the davs when he had studied German, he had
left the subjunctive to be dc;llt with later, since it wasn't absolutely
esscnti<ll and he was in a hurry. By this time he could hop through the
language without its aid, like an agile man with only one leg. But now
Christopher set himself to master the subjunctive. Very soon, he had
done so. Proud of this accomplishment, he began showing off whenever
he talked: 'had it not been for him, I should never have asked nwselfwh<tt
I would do if they were to ... etc, etc.' Humphrey was much ar~1uscd.

The advancc-organiser argument

Grammar instruction might also have a delayed effect. The researcher
Richard Schmidt kept a diary of his experience learning Portuguese in
Brazil. Initially he had enrotled in formal language classes where then~ was
a heavy emphasis on grammar. \Vlwn he subsequently left these classes to
tra1·cl in Brazil his Portuguese made good progress, a f,tct he <1ttributed to
the usc he was making of it. However, as he interacted narumlly with
Brazilians he >vas aware that certain features of the talk - certain
grammatint! items- ~ccmed to catch his attention. He noticed them. It so
happened that thc~e items were also items he had studied in his classes.
\Vhat's more, being more noticeable, these items seemed to stick. Schmidt
concluded that noticing is a prerequisite for acquisition. The grammar
teaching he had rccci\'ed prc\·iously, \vhilc insutlicicnt in itself to turn him
into a llueut Portuguese speaker, had primed him to notice what might
otherwise have gone unnoticed, and hence had indirectly influenced his
learning. It h~td :tctcd as a kind of advance organiser tOr his later acquisition
of the langu·,tgc.

'The discrete item aq,.,rument

Language - <\11)' bnguage ·-seen from 'outside', can seem to be a gig~mtic,
shapclc% m~tss, pn:senting an insuperable ch;lllengc f()r the learner. Because
grammar cunslsts of an apparently finite set of rules, it can help to reduce
the ;tpp·.nent enormity uf the language learning task for both tt:achcrs and
'tudl·nts. By tidying language up and org;tnising it into neat cuegories
(~ometimcs c;dlcd discrete items), gr~tm:n;uians make langu<tgc digestible.

16
 2 " VVhy t12ach grammCJr?

A discrete item is any unit of the grammar system that is sunlcicntly

narrowly defined to form the fOcus of a lesson or an exercise: e.g. the pre.(O/f
continuous, the definite article, possessive pronouns. l~'r/Js, on the other hand,
or smtences are not categories that are sufficiently discrete t(n· teaching
purposes, since they allow fOr further sub-categories. Each discrete item can
be isolated from the language that normally em·elops it. It em then be
slotted into a syllabus of other discrete items, and targeted fOr indiYichul
attention and testing. Other ways of packaging language for teaching
purposes arc less easily organised into a syllabus. For example, commt.mi-
c~1tive functions, such as a.rl:.ing finJours, making requr.rts, expressinl!, rt'grets,
and text type categories, such as narratives, iw:lrurtiom, phone COIIVa.wtiom,
are often thought to be too large and unruly for the purposes of lesson
design.

The rule-of-lav.' argument

lt follows from the discrete-item argument that, since grammar is a system
of learnable rules, it lends itself to a view of teaching and learning !mown ;1S
transmission. A transmission view sees the role of education as the transfer
of a body of knowledge (typically in the form of bets and rules) from those
that have the lmowledge to those that do not. Such a \'iew is typically
associated with the kind of institutionalised learning where rules, order, and
discipline are highly valued. The need for rules, order and discipline is
particularly acute in large classes of unruly and unmotivated teenagers- a
situation that many teachers of English are confronted with daily In this
sort of situation grammar offers the te:Kher a structured system that can be
taught and tested in methodical steps. The ~tlternativc - allowing learners
simply to experience the b.nguage through commtmicnion- may simply be
out of the question.

The learner expectations argument (1)

Regardless of the theoretical and ideological arguments for or ~1gain-;t
grammar teaching, many learners come to language ~.:lasses with f:t.irly f-L-.:cd
expectations as to what they \vill do there. These expectations may derive
from previous classroom experience of language learning. They may also
deri\'e from experience of classrooms in general where (traditionally, at least)
teaching is of the transmission kind mentioned aboYc. On the other h;lnd,
their expectations that teaching will be grammar-fixused may ~tem from
ti·ustration experienced at trying to pick up a second langu~1gc in a non··
classroom setting, such as through self-study, or through immersion ln thc
target language culture. Such students may luvc enrolled in Lmgu;\ge cbssc~
specifically to ensure th:H the learning e:-:peri<.'n<.::e is made more e!Yici<.'tH ;md
systematic. The teacher who ignores this expectation b_\· erlCOLii";ltjng
learners simply to experience language is likely to fru~tratc and ;d!c:dte
them.

17
How to Teach Grammar

The case Just as arguments lw\'e been marshalled in favmn of grammar teaching,
against likewise several cases have been made against it. Here are the main ones:
grammar
The lmowledge-how argument
I know what is involved in riding a bike: keeping your bahnce, pedalling,
steering by means of the handlebars and so on. This does not mean to say
that I know how to ride a bike. The same analogy applies to language
learning. h can be ·vinved as a body of knowledge - such as vocab1.lhtry
and grammar. Or it can be viewed as a skill (or a complex set of skills). If
you take the language-is-skill point of view1 then it follows that, like bike
riding, you learn it by doing it, not by studying it. Learning-by-doing is
what is called experiential learning. Much of the bad press associated with
intellectual approaches to language learning - through the learning of
copious grammar rules, for example - stems from the failure on the part
of the learner to translate rules into skills. It is a failure that accounts for this
observation by Jerome K. Jerome, writing in Three 1Hrn on tbe Bummcl about
a typicd English schoolboy's French:
I-Ie may be able to tell the time, or make a few guarded obsen'<1tions
concerning the weather. i\io doubt he could repeat a goodly number of
irregular verbs by heart ... [But] when the proud parent takes his son to
Dieppc merely to discover that the lad does not know enough to call a
cab, he abuses not the system but the innocent victim.
Proponents of the 'knowledge-how' view might argue that 1vhat the boy
needed was not so much grammar as classroom experience that simulated
the kind of conditions in which he would eventually use his French.

The communication argument

There i.s more to knmving a language than \mowing its grammar. 1t is one
thing to know that Do you dtink? is a present simple question. It is another
thing to know that it can function as an offer. This simple observation is at
the heart of what is now called the Communicative Approach, or
Communicative Language Teaching (CLT). From the 1970s on, theorists
have been arguing that grammatical knowledge (linguistic competence) is
merely one component of \Nhat they call communicative competence.
Cornmunicativc competence invoh'es knowing how to use the grammar and
Yocabulary of the language to achieve communicative goals, and knowing
how to do this in a soci<lll)' appropriate way.
Two schools of thought emerged as to the best means of achieving the
objectives o( this communicative approach. Both schools placed a high
premium on putting the language to communicati\'c use. But they differed
as to when you should do this. The first- or shallow-end approach -might
be C.\'Ml.mcd up as the Yiew that you learn a bnguage in order to usc it. Thnt
i~: learn the rules and then apply them in life-like communication. The
more radical line, hmve\·cr, is that you use a language in order to learn it.
Proponents of this deep-end approach take an experiential view oflcarning:
ym1 bun to communicate by com_m\lnicating. They argue rhat, b:· means of
;lctiYitie~ thM engage the learner in life-like communication, the gpmmar

18
 2 " Vv'hy teach sJrarnmar?

will be ;Kguired virtually unconsciously. Studying the rules of gr<~mm<~r is

~imply a waste of valuable time.

The acquisition argument

The f1ct that we all learned our first language witlJout being: taught
grammar rules has not escaped theorists. If it works for the j-Jr~t, why
shouldn't it work for the second? This ls an argument th:\t has been around
at least sinceJo~eph \Vebbe's day (see page 14). It received a new impetus in
the 1970s through the ·work of the applied linguist Stephen Krashen.
Krashen rnakes tbe distinction between learning, on the one hand, and
<lcquisition, on the other. Learning, according 'w Krashcn, results from
tOrmal instruction, typically in grammar, and is of limited usc for real
COlTln>unic.o,tion. Acquisitian, howeser, is a natural process: it i>< the proctss
by which the first language is picked up, :tnd by which other languages are
picked up solely through cont;KJ- with speakers of those langlwges.
Acquisition occurs (according to Kr<,,.hen) when the k'.·;>.rner is cxpmed to
the right input in a strc~s-fi:ee environment so that innate learning capacities
are triggered. Success in a second language is due to acquisition, not
learning, he ·.trgues. :\lorco\'ef, he cLcirns that learnt knowledge can nc\'er
become acquired lmowlcdge.
Krashcn's tl1eory had an important ini1uencc on language teaching
practices, especially "\'l'itb teachers who were disenchanted with the 'drill-
and-repeat' type methodology that prn·ailcd in the 1950s and 1960s.
Rejection of the formal srudy of grarnmar is central to Krashen's ';\atural
Approach'.

The nahual order argument

Krashen's acquisition/learning hrpothcsis drew heavily on studies that
suggest there is a natural order of acquisition of grammatical items,
irrespectiYe of the order in which they <11'C tm1ght (5ee page 10). This Y)C'>'>'
derives partly from the work of the linguist Noam Chomsky. Chomsky
argues that humans are 'hard~wircd' to learn languages: that is, there ;trc
univcrs·al principles of grammar that we arc born with. The idea of an imy,ltc
universal grammar helps explain simil:uitics in the developmental order in
first bnguage acquisition as well as in second language acquisition. It
explains why EngLish children, Th·,li tecw,>gcrs and Saudi addts ;ell go
through a I no likt jis/1 stage before progressing to l ,/on'! ii!._·,'.fi.r/1. It <tlso
suggests that attempts to subvert the natur;cl order by sticking rigidly to a
tmditional grammar sylhbus :md insisting on immediate accuracy arc
foredoomed, In short, the natural order argument insi~ts rh:H :1 ru;tbouk
gr~1mmar is not, nor can ever become, a mental grammar.

The lexical chunks arf:,rtHncnt

\Vc have already noted the fan rh:lt language learning eccms 1"0 '1m·oln: an
clement of item~learning. Vocabulary learning i~ largely item-learning. Sn
too is the retention of whole phrases. idioms, sc>cial t(mnubc ctl'. in the f( 1 rrn
of -what <~re sonJetirrn;c; c;llkd chunks ol 1-.<n):;u"-ge. Chunb are l:s~er th;m
\\'ords hut often les~ than ~entcnces.l-Icrc :ne ,,_;me C\>llliTl(lll r;.;,lnl)'k":

19
How to Teilch Grarn1:1ar

excuse rne?
so far so good
what on earth?
have a nice day
be that as it mqy
if you ask me
not on your life
here you are

Acquiring chunks of language not only saves the learner phnning time in
the cut-and-thrust of real interaction, but seems to play a role in language
development too, It has been argued that many of the expressions that
yot~ng children pick up, like all-gone, or gimme (as in gimmc the bail), are
learned as chunks and only later unpacked into their cornponcnt parts.
Once unpacked, new combinations, such as gi'ue bcr t/.Jc ball start to emerge.
It has been argued that this process of analysing previously stored chunks
plays an important rok in first language acquisition.
HcnN much of second Ltng-uage acquisition involves item-learning a:>
opposed to rule-learning is still an open question. ~evcrthcless, in recent
years there has been a growing recognition of the importance of word- and
dJUnk-Jc:lfnlng, such that some writersllave proposed a lex.ical approach ro
tc<tching, in contmst to the traditional emphasis on sentence grammar.
Among other things, a lexicaJ approach promotes the learning of fiec1uently
used and fairly formulaic expressions (Ha·ve JOII C'i.H!r been"' ? Would)'oU !ike
il. f) rather than the study of rather abstract grammatical categories such
as the present perfect or conditionals.

The learner expectations argtlmt:nt (2)

\Vhile many learners come to language classes in the expectation that at
least some of the time they will be studying the grammar of the language,
there uc many others who may <liready haYe had years of grammar study
at schon! and are urgently in need of a chance to put this knowledge to
work ~1cstionnaires of adult students in general English courses almost
irmtri;tblr idcnti(r 'conn:rsation' as a high priority, and t·hese statcmems
(from l.oolcing ot Language C/as.rrooms, Carnbridge UniYersity Press) by a
mnge of EFL students studying in Britain arc typi..:al:
'fn Ccnnany there's more homework, gnunmar exercises, and things like
th:H. I I ere, l think you\·e got more chance to speak and therefore learn
rhc bngu·,l_t~c.'
'Sumclimcs, speaking and things like that help a lot, because if you don't
~pe·,1k English, and ju~t do writing exercises, it's no good.'
'I like lLt\·ing- com·cr~ations because, yes, gmmmcl!" is impurtant, but it's
not rnuc:h fun . '
The btrner expcct~ttion argmncnt cuts both ways: some learners demand
gramm:u, others ju~t want t·o talk. It's tl1c te:JChcr's job to respond scJHiti\·cly
to these cxpcctatlons, to proYide a balance where possible, and C\·cn to
ncgoti:tt\? ~~ comtHt)l11i~e.

20
 2 ~ Why teach grarmnaf'

BefOre attempting to bring the grammar debate up to d.1te, and to draw

some conclusions hom recent research evidence, it may pay to briefly sketch
in the way attitudes to grammar teaching have influenced the ebb Jnd t1ow
of different teaching methods.

Grammar and In the last century the architects of language teaching methods have been
methods preoccupied with two basic design decisions concerning gramrnar:
• Should the method adhere to a grammM syllabus?
• Should the rules of grammar be made explicit?
The various \vays they answered these questions help distinguish the
different methods ±i-om each other. \Vhat fOllow~ is a potted history of
methods in the light of their approach to these issues.

Grammar-Translation, as its n:tme suggests, took grammar as the starting

point fOr instruction. Grammar..'fransbtion courses followed a gnmrnar
syllabus and lessons typically began with an explicit statement of the rule,
followed by exercises involving translation into and out of the mother
tongue.

The Direct l\1ethod, which emerged in the mid- to late-nineteenth ccntmy',

challenged the way that Gramrnar-Transhnion fOcused exclusively on the
written language. By claiming to be a 'natural' method, the Direct I\-fethod
prioritl.scd oral skills, and, \Vhi.le following a syllabus of gramnur structure's,
rejected explicit grammar teaching. The learners, it was supposed, picked up
the grammar in much the same way as children pick up the grammar of
their mother tongue, simply by being immersed in hnguage.

Audiolingualism, a largely North American invention, stayed faithful to

the Direct I'vlethod belief in the primacy of speech, but W;1S even more strict
in its rejection ofgramnur teaching. Audiolingualism derived its theoretical
base from behaviourist psychology, which considered language ~ts simply a
form of behaviour, ro be learned through the f\xrnation of correct habits_
B:abit formation was a process in which the application of rules plarcd no
part. The Audiolingunl syllabus con~isted of a graded list of sentence
patterns, which, although not necessarily labelled as such, were grammatictl
in origin. These patterns formed the basis of pattern-practice drills, the
distinguishing feature of Audiolingual dassroom pctctice.
0Jo,l!l1 Chomsl")''s claim, in the late 1950s, that language ability is nut
habituated beh;wiour but an innate human capacity, prompted a re<1~,e~s­
ment of drill-and-repeat type reaching practice~. The \·iew tktt 11'C :ue
equipped at birth for bnguage •tcquisition led, as we saw on pase 19, ru
Krashen's belief that hmn~d instruction wa~ unnecessary. His :'\\1tural
Approach does awa\' with both a gramm;u svlbbus :md explicit rulc-gil'ing.
Instead, learners are ..exposed to lar'ge doses oi' comprehensible input. 'Inn.l~l'
pi'<-Kesscs convert this input into output, in time. Llke the Direct .\lc:lJ,·,J,
the 0;;ltllral Approach artcmprs to repl';cw.' the con~1'1tinn~ of tlr:;t bng'-u~;e
xquisition. Gr<Hnm:H, :lCC\Hding to rl1is sc~·n;uio, ;, irn:lcl':lnt.

21
How :o Teach Grammar

The development, in the 1970s, of Communicative Language Teaching

(CLT) was motivated by developments in the new science of socio-
linguistics, and the belief that communicative competence consists of more
than simply the kno1.vkdge of the rules of grammar (see above, page 18).
Nevertheless, CLf, in its shallow-end version at least, did not reject gram-
mar teaching out of hand. In fact, grammar was still the rnain component
of the syllabus qf CLT comscs, even if it \Vas dressed up in functional
labels: asking tbe -way, talking about yo11rselj; makingfiilure plans etc. Explicit
attention t·o grammar rules was not incompatible with communicative
practice, either. Chomsky, after all, had claimed that language was rule-
governed, and this seemed to suggest to theorists that explicit rule-giving
may have a place after all. This belief was around at about the time that
CIT was being developed, and was readily absorbed into it. Grammar
rules reappeared in coursebooks, and grammar teaching re-emerged
in classrooms, often, it must be said, at the expense of communicative
practice.
Deep-end CLT, on the other hand, rejected both grammar~based
syllabuses and grammar instruction. A leading proponent of this view was
XS. Prabhu, a teacher of English in southern India. In his Bangalore
Project, he attempted to replicate natural acquisition processes by having
students work through a syllabus of tasks for which no formal grammar
instruction was supposedly needed nor provided. Successful completion
of the task ~ for example, following a map ~ was the lesson objective,
rather than successful application of a rule of grammar. The Bangalore
Project was the predecessor of what is now known as task-based learning.
Task-based learning has more recently relaxed its approach to grammar,
largely through recognition of the nlue of a focus on form (see below,
page 24).
To summarise the story so far: to the first of the questions posed above
(S'bou!d t/.1e method odbere to a grammatim/ sy!!ab11.1?) most approaches to
language teaching up until the 1970s have answered firmly Yes. The actual
f(Jn11 of the syllabus di±Yered considerably from method to method, but,
until such organising categories as functions or tasks were proposed, sylla-
buses were essentially grammar-based.
On the question of the explicitness of rule teaching there is a dear divide
between those methods that seek to mirror the processes of first language
acquisition~ such <lS the Direct l\'lethod and the Natural Approach~ and
those such as Grammar-Translation - that see second language acquisi-
tion as a more intellc-:rual process. The former methods reject grammar
inotruction, while the latter accept a role for conscious rule-learning.
Finally, even in methods where rules are made explicit, there may be a
different emphasis with regard to the way the learner arrives at these rules.
In some approaches, such as Grammar-Translation, the rules arc simply
presented to the learner, who then goes on to apply them through the study
and manipulation of examples (a deductiYe approach: see Chapter 3).
Other approaches, including rhe shallow-end fi:m11 of the communicati\·e
approach, often n:guirc the learners first to study examples and work the
rules PUt for thc1meh-cs (an inductiYe 8j,proach: see Chapter 4).

22
 2 ~ Why teach grammar?

At the risk of over-simplif):ing matters, the following chart indicates the

relative importance these methods attach to the teaching of gramm~u:

zero heavy grammar

grammar emphasis
c----.L--~--~---L._____j___J_~~----~

t
Natural Approach
t
Audiolingualism
t
Shallow-end
t
Grammar-
Deep-end CLT Direct Method CI:f Translation

::irammar now VVhat, then, is the status of grammar now? \Vhat is common practice with
regard to the teaching of grammar, and what directions for future practice
are suggested by recent and current research?
Firstly, it is important to establish the fact that 'grammar teaching' can
mean different things to ditTerent people. It may mean simply teaching to
a grammar syllabus but otherwise not making :my refCrence to grammar
in the classroom at all (as was the case with A.udiolingualism). On the
other hand it may mean teaching to a communicative syllabus (e.g. of
functions or of tasks) but dealing with grammar questions that arise in the
course of doing communicative activities. This is sometimes called covert
grarnmar teaching. .iviore typically, grammar teaching means teaching to a
grammar syllabus and explicitly presenting the rules of grammar, using
grammar terminology. This is known as overt grammar teaching.
Lately, a good deal has been written about a grammar revivaL There is a
\Videspread belief that, with the introduction of Communicative Language
Teaching, attention to grammar was eclipsed by an emphasis on experiential
learning and purely communicative goals. This is only partly true: syllabuses
did appear in the 1970s that appeared to marginalise grammar in favour of
functions. But, as \Vas pointed out in the previous section, a closer look at k·\
these syllabuses shows that they often had a strong grap1mar basis. And a
glance at so-called communicativ~! coursebooks confirms that grammar
explanations are much more conspicuous now than they were, say, in the
heyday of either the Direct .I\'lcthod or Audiolingualism.
The view that CLT deposed grammar may also stem from a tendency to
equate grammar with accuracy. It is true that, in comparison with
Audiolingualism, CIT has tended to place more weight on being intelligible.
than on being correct. Such an emphasis need not be at the expense of
attention to the rules of grammar, howcwr.. Relaxing on accuracy simply
acknowledges the fact that the rules of grammar take a long time to establi~h '"
themselves, and that, in the meantime, the lertrners' wish to communicHl'
should not be needlessly frustrated.
It is also true that the deep-end version of Cl.T, as promoted by Prahhu
(see page 22), was hostile to explicit grammar teaching. But this was
relatively short-lived, and, while of enormous interest fwm a theoretical
perspective, lt seems to have had little or no infltiencc on global ch~oro(~m
pra.cticc. If grammar cn·r went away. it was only very briefly and Illlt \u:· far.

23
Hov.,, to leacr1 Grarnrmr

The sense that we are experiencing a grammar revival has been

underlined by the emergence of two influential theoretical concepts:
focus on form
consciousness- raising
Both concepts owe ~omcthing to the \mrk of Stephen Krashen, even if only
as a reaction to his claim that cLtssroorn teaching is a vvaste of time. You \vill
remember that Krashen distinguishes betvveen acquisition and learning.
;J Grammar te;Khing- that is, attention to the f(mns of the language -lies in
the domain of !earning and, says Krashcn, has little or no influence on
language acquisition. l\1ore recently, research suggests that without some
attention to form, learners run the risk of fossilisation. A focus on form
docs not necessarily mean a return to dri.U-and-repeat t:l'e methods of
teaching. ?\'or does it mean the use of an off-the-shelf grammar syllabus. A
fOcus on _fOrm may simplv mean correcting a mistake. In this sense, a fOcus
on form is compatible wi.th a taSk-based approach.
Related to the notion of focus on form is the notion of consciousness-
raising. Krashen argued that acquisition is a largely unconscious process.
All that is needed to tt·igger it are large doses of comprehensible input.
Other theorists have argued that the learner's role is perhaps less passive
than Kra~hen implies, and that acquisition involves conscious processes, of
which the most flmdamcntal is attention. \'Ve have seen how Schmidt (see
page 16) concluded that noticing spoken language items in Brazil helped
his Portuguese. It f(lllows that helping learners attend to language items
may help them acquire them. Pointing out tCatures of the grammatico1l
system is thus a f(_mn of consciousness-raising. It may not lead directly and
instantly to the acquisition of the item in question. But it may nevertheless
trigger a train of mental processes that in time will result in accurate ;wd
\tppropri·,tte production.
1t might seem that we have come full circle, and that grammrtr
consciousness-raising is simply a smart term f(Jr what was once caLled
grammar presentation. But presentation is usually paired with practice,
implying immediate- and accurate- output. Consciousness-raising, on the
other hand, docs not nccess;trily entail production: it may simply exist at the
level of undcr~tanding. And remembering, In fact, put simply, that's wbt
rai~cd consciomness is: the state of remembering, having understood
something.
"I'o sum up: if the tea('her uses techniques that direct the learner's a1ten-
tion to f{mn, and if the teacher proYides acti\'ities that promote awareness
of grammar, learning seems to resuk \Ve need, therctOre, to add to the
pro-gr;lmmar position the arguments for a focus on form and for
consciousness-raising, 'fogethcr they comprise the paying-attention-
to-f(lrm argument. That is tn say, bnning seems to be enhanced when
the learner'~ attention is dircncd ro setting the fOrms right, and when the
learner's attention is directed to features of the grammatical s\·stcm.
These would seem to tip the babncc in f.n.;l\Jr uf granm;~1r. \Vhilc the
'anti-gr~mmnr' position is strongly and e\'Cn fiercely argued, it tends tu
depctHl on one basic assumpti,m, th<lt is, that the procesSl'S uf second

24
 2 ~ Why teach gra1nrnart

language acquisition mirror those of first language acquisition. This is an

assumption that is hotly debated. \Vhile there are certainly cases of adult
learners who have reached near-n:Hi\'e levels of protlclcncy in a second
language simply through immersion in the second language culture, these
tend to be exceptions rather than the rule. On the other hand, there ar.c
compelling ;uguments to support the view that without attention to form,
including grammatical form, the ktrner is unlikely to progress beyond the
most basic level of communication.
But this doesn't mean that grammar should be the go~1l of teaching,
nor that a fOcus on fonTt ;tlone is sufficient. The goal of the communi-
cHive movement - communicative competence - embraces more than
just grammar, and implies a focus on meaning as welL lr may be that
communicative competence is best achieved through communicating,
through making meanings, and that grammar is a way of tidying these
meanings up. If so, the teacher's energies should be directed mainly at
providing opportunities ±()r authentic language use, employing grammar
as a resource rather than as an end in itself. As Leibniz is supposed to
have said: 'A langu<tge is acquired through practice; it is merely perfected
through grammar.'

Basic principles \Ve hn\·e looked at the arguments for and against incorporating grammar
for grammar into language teaching, and concluded that, on balance, there is a convincing
teaching case fOr a role for grammar. The remainder of the book will explore hmv this
role can be realised in the classroom. It will be usd\.d at this stage to draw
up some basic rules of thumb for gmmmar teaching-.. rules of tht!mb which
Yvill scr\'c as the criteria f(lr evaluating the practical approaches that follO\v.

TheE-Factor: Efficiency"' econon1y) e.ase 1t\nd efficacy

Given that dealing with gramm<lr is only a part of a teacher's ;Ktivitics, and
given that classroom time is very limited, it would seem imperative that
whatever grammar teaching is done is done as efficiently as possible. H~ as
has been suggested, the teacher's energies should be at least partly directed
at getting learners to communicate, prolonged attention to grammar is
diHlcult ro jttstif)'. Likewise, ir ~l gr,unmar activity requires a great deal of
time to set up or a lot of m;tterials, is it the most ctlicient deplorment of the
teacher's limited time, energ-y and resources? \Vhen comidcring <l!1 acti\-ity
for the presentation or practice of gr;tmrnar the first question to a~k, is: lie;-;_::
~:Oicimt i.r it? Etllciency, in turn, can be broken down into rhree f:h.:tor~:
economy, ease, and efficacy.
\Vhen presenting grammar, a sound rule of rhumb is: the shorter the
better. It has been sho\\'n th<tt economy is a key f;t<.:tor in the tmining ut'
technical skills: when learning how to drive a c1r or opcr:Hc :1 computer,
;1 little prior te:u..:hing seems to be more cffe..:tin: than a lot. The nl<JfC
the instructor piles on instructions, the more confused the trainee is likdy·
to bccume. The ~',unc would >cctn to appty in hnb!:uc1.~t lCcH:hins: he
economical.

25
How to Teach Grammar

Be economical, too, in terms of planning and resources. The ease factor

recognises the fact that most teachers lead busy lives, have many classes, and
simply cannot afford to sacrifice valuable free time preparing elaborate
classroom materials. Of course, the inyestmcnt of time and energy in the
preparation of materials is often accompanied by a commitment on the
part of the teacher to making them work. But, realistically, painstaking
preparation is not always going to be possible. Generally speaking, the easier
an activity is to set: up, the better it is.
Finally, and most importantly: will it work? That is to say, what is its
efficacy? This fiKtor is the least easy to evaluate. \Ve have to operate more
on hunch than on hard data. Learning, like language, resists measurement.
Of course, there are tests, and these can provide feedback to the teacher on
the eHlcacy of the teaching/learning process. Nevertheless, testing is
notorimlsly problematic (see Chapter 9 for a discussion on this). Moreover,
there is much greater scepticism nowadays as to the extent that teaching
causes learning. This need not undermine our faith in the classroom as a
good place for language learning. VVe now know a lot more about what
constitute the best conditions for learning. If teachers can't directly cause
learning, they can at least provide the optimal conditions for it.
As we have seen (page 24), a prerequisite for learning is attention. So the
efilcacy of a grammar activity can be partly measured by the degree of
attention it arouses. This means trying to exclude from the focus of the
learner's attention any distracting or irrelevant details. Attention without
understanding, however, is probably a waste of time, so efficacy will in part
depend on the amount and quality of contextual information, explanation
and checking. Finally, understanding without memory would seem to be
equally ineffectiye, and so the efficacy of a presentation will also depend on
how memorable it is.
None of these c:ondttlons, however, \Viti be sutlicient if there i.s a hck of
motivation and, in the absence of some external motivational factor (for
example, an examination, or the anticipation of opportunities to use the
language), it is the teacher's job to choose tasks and materials that engage
the learners. Tasks and materials that are involving, that are relevant to their
needs, that have an achievable outcome, and that have an element of
challenge while providing the necessary support, are more likely to be
motivating than those that do not have these qualities.
Efficiency, then, can be defined as the optimal setting of three related
factors: economy, case, and efficacy. To put it simply: are the time and
resources spent on preparing and executing a grammar task justified in
terms of its probabk learning outcome?

The A-factor: Appropl'iacy

:'\o class oflearners is the same: not onlv are their needs, interests, le\'el and
goals going to vary, but their beliefs, attitudes and values will be diftCrent
too. Thus, an activity that works for one group of learners- i.e. that fulfils
the E-factor criteria- is not necessarily going to work for another. It may
simply not be appropriate. Hence, any classroom actiYity must be evaluated
not only according to criteria of eftlciency, but also of appropriacy. Factors
to consider when determining appropriacy include:

26
 2 & Why teach grammar?

the age of the learners

their level
the size of the group
the constitution of the group, e,g. monolingual or multilingual
\vhat theil' needs are, e.g. to pa:.s a public exan1ination
the learners' interests
the available materials and resources
the learners' previous learning experience and hence present expectations
any cultural factors that might affect attitude~, e.g. their perception of the
role and status of the teacher
the educational context, e.g. private school or state school, at home or
abroad
Activities that fail to take the above factors into account are unlikely to
work The age of the learners is very import::mt. Research suggests that
children are more disposed to language learning actiYities that incline
towards acqui.sition rather than towards learnin[~· That is, they <tre better at
picking up language implicitly, rather than learning it as a system of explicit
rules. Adult learners, on the other hand, may do better at activities which
involve analysis and memorisation.
Cultural factors, too, will determine the success of classroom activities.
Recently there have been a number of writers who have queried the
appropriacy of indiscritTtinately and uncriticaHy applyi.ng methodologies in
contexts for which they were never designed. Communicative Language
Teaching (CLT) has been a particular target of these criticisms. CIT values,
among other things, learner-centredness, that is, giving the learners more
responsibility and involvement in the learning process. This is often
achieved through discovery learning activities (for example, where learners
work out rules themselves) ?.nd through group work as opposed to the
traditional teacher-fronted lesson. CLT also takes a relatively relaxed
attitude tmvards accuracy, in the belief that meaning takes precedence over
form. Finally, CLT has inherited the humanist view that language is an
expression of personal meaning, rather than an expression of a common
culture. Such notions, it is arg<Jed, deriw from \"cry \\'estern beliefs about
educ<~tion ;md hnguage. Its critics argue thil.t CLT is ·,m inappropri<lte
methodology in those cultural contexts where rhe teacher is regarded as a
fount of wisdom, and where accuracy is valued more highly than fluency.
Of course, no learning situation is static, and, with the right combination
of consultation, negotiation, and learner training, CYen the most entrenched
attitudes are susceptible to change. The teacher is therefore encouraged to
be both alhenturous as \Vell as critical, when cor(:o.idering the a.ctivities in the
chapters that foll~w.

27
 Ilow to T'"ach Gramrn,Jr

Conclusions In answer to the question 'Why teach grammar?' the following

reasons were advanced:
the sentence~machine argument
the fine-tuning argument
the fossilisation argument
the advance··organiser argument
the discrete item argument
the rule-of-law argument
the learner expecMftions argument
There are some compelling reasons why not to teach grammar:
the 'knowledge-how' argument
the communication argument
the acquisition argument
the natural order argument
the lexical chunks argument
the learner expectations argument
To the arguments in favour should be added two more recent insights
from second language acquisition research. These are the notions of
focus on 'form and of grammar consciousness-raising. Together they
comprise:
,; the paying-attention-to-form argument
On balance, the evidence suggests that there is a good case for a role
'for grammar-focused teaching.
Grammar presentation and practice activities should be evaluated
according to:
" how efficient they are (theE-factor)
" how appropriate they are (the A-factor)
The efficiency of an activity is gauged by determining:
its economy~ how time-efficient is it?
., its ease ~ how easy is it to set up?
" its e'fficacy ~is it consistent with good learning principles?
The appropriacy of an activity takes into account:
., learners' needs and interests
• learners' attitudes and expectations
It is these twin aims~ efficiency and appropriacy- which underscore
the description and evaluation of the techniques outlined in the 1·est
o'f the book.

Looking ahead The two chapters that follow look at contrasting ways that grammar
can be presented. The first of these is concerned with deductive
approaches, where the starting point is the grammar rule. The second
looks at inductive approaches, where the starting point is language
data.

28