How to Teach Speaking

Task File

What speakers know

l^
'13 Speech genres Page (The first Classifr the following speech genres according to the criteria in the table. one has been done for you).

purpose

participation non-interactive

planning
planned

airport announcements
universitY lecture

transactional

telephoning a friend radio itrlenriew

hY".t<r4 i*'tr u{}. 'rc'.ol-,rgt{

*"rn*{.r-{ rrft\\
frnr-^ndi,ro--nl ,r\t-f rrf,'.,-f r.\

nnff^tern{.',t', ,f' nln^.-l lr n n.^n .'r,r,;] ,i i +ur* irdernrtt*r

i'rt-er^rCi.rt
/r" ,n\.,

TV weather forecast

^n -ird-err{4',, tr

asking street directions

speech of thanks

.,xtr-'rrr*i,r

\,'t,',.-,*"'\ /,,^ti rr-,irt;r',.

'1r'ri.{

n.nir.f,oltt

,.ntr1trd':

,-nr,

*,s-rJlr:-

:**{

- B

The followins extract from the play Tea Party,by Harold Pinter, attempts to replicate spoken lu.rguig.. How successfirl is it, do.youthink? What feafures of spoken grammar * does it display, and which ones are missing?

Spoken grammar and vocabulary Page-2O-- . .

"",ro."bnlu.!

rlr*e,,r-,'.. ,-\

-rri

John: (choosing a-c3!3.) These are good. Tom: What arey?

Diana: (choosing a bridg'Jq

These look nice'

L,,<t rgblois: you look wnderful. Mrs Disson: Absolutely wonderful. Doesn't she, Peter? v J t Disley: Marvellous. ' I .toit, What do you think of your grandsons? -rath TheY've grown uP, haven't theY? Lois: of course, we knew them when they were that high, didn't we, Tom? Father: So did we. *4 Tom: Yes. Willy: Big lads now, aren't they, these two? John: Cake, GrannY' ' Mother: No, l've had one. John: Have two.
Father: l'll have one.
Mother: He's had one.

Father: l'll have two.

I C

narratves Page 14 H.r. i, an example of a,'uthentic

SPoken
are rePresentative

spoken.narrative, taken from The Language.of it with the Kedgeree story on page2. What Compie Con,(.,e;saon by Fi Prldtram. extent do you think these rwo narratives what features do they have in common?-To

oftheir

genre?

ScoftThornburyHowtoTeachspeakingoPearsonEducationLimited2005
PHOTOCOPIABLE

135

I t I

l"
I I

I
I

I
I

kng in other language

crossed

I
t
I I I
Grace
a o
o

an ocean

lost my tongue _, ' "from the root of the old one

a new one bas s?run{

Nichols

,i\t'

Dff ereges between

Ll and L2 speaking

Commgnication strategies

What L2 speakers need to know Availability for use: implications for teaching

I 1'{x

t I
] I I I
F

Differences between Lf and L2 speaking

The description, offered in the preceding chapters, of how rpok.n.n.y i, achieved, has made no distinction between the speaking of a first language (L1) and the speaking of another language (L2).We have discussed the skills and kinds of knowledge involved in achieving fluency as if these were absolute qualities that all speakers share. Clearly this is not the case. Even among L1 speakers there can be wide variations in the degree of fluency that individuals demonstrate. These differences are exacerbated when it comes to speaking in a language different from your own. The inevitabie lack of fluency involved is a source of frustration and even embarrassment, as attested by these quotes from learners of English, in response to the question: Wbicb aspects of your EnglisU.O: *" most uant t1" im?ro.ve? ' 'My weak points in English is speaking and listening. I suppose I am not so bad at reading and writing, but especially, my speaking is awful. I want to
improve my speaking ability as once I had.'

t
$
"

' 'l would like to improve my spoken English and my pronunciation. I think have terrible Russian accent. Therefore l'm very shy to speak.' 'This is the problem, I have been learning English long, but I can't speak,
understand the conversation but I can't answer imrnediately as I like.'

!'
t-

'The problem is to speak English with other people face to face. I can't find words. I always use the same sentences.'

'l know I need to practice my speaking a lot. During all my life, I have been
doing grammar and reading, but nobody has taught me how to speak English. I think that this skill is always forgotten when someone teaches English.' 'sometimes I use English in my work and this is always a painful moment for me in which my heart is in my boots and I despairing search the words.'

I'
I
I

27

i
{
#ili-r: !:inlnt--

How to Teach SPeaking

fb, pracrice is identified as an important contributing fa_ctor,,T "f,;;p";ilifi rp:"{"g-fh'ure. And by practice is meant, not practice of gra.,im^.-and voiabul"rjr, b.,, practice of intera*ive speaking itself. The comiined effel,or,"i.r.l?diiln.i.,

These quores identifr pT. fa*ors th-at can contribute to a rack of L2 fluency, and in particular !.y hw a u.L inhibit face_ to-face inrera*ion, quite independently h;;-;;.mmatical and rexical knowledge a speaker hui. shortJg."f

a
-

oi"uioil"o;

t
-

into their existing language qrlgw]gdge, .".ffi;J,' accessed, "r;rl"r that it.is nor yer eairy retrievabre. ThJ process orurr"nging.iffir--r, o, retrieving the word is not yet automatic. They then feellikJthe Jtud.rrt *ho 'despairing searchfes for] ihe words'.
of speed. Furthermore, pressue to be accural

,rr.ly exrensive or as estabshed as thir knowleag. ,r ", like the student who said, 'r can't find words. I always use "r. the same sentences,. on the orher hand,,the problem may be rr than the unavailability of that knowledge. It hs not ,"rn.r..rl, integrated

i th. ,^*.iirr," ,hil.*f,h";e amending to their interlocurors, adjusting tTleir message accordingly, and the m an age rn e n t of convers a tin al ru rns. Tfi . ;kix; ;;:il"#*::f,::';: ll are essentially rhe same and shourd, i" th.ory, b; ;;"rf#;ie from the speaker's first language into the ,..on. what is significantry different is, of course,-the language itself L2 speakers, knowledge of the L2, including it, uo."b.ilr,, ;;:-;:;*";i'i,

".;;;, *oliJ-;ifit i"rrroo*. How then does L2 ,p.ulcng.di"ff.r from fr ,p""gii;;.rn,, of the stages of menral processing invlved, th.t. ir p-'"; J; ,nui a;rr.r.n.. at all' Like L1 speakerq L) speakers also.p-roduce spe.ch through a process of concep ru azing, the n form ura ng, an d' fi r ;iiil"i-g, iirrt.g wh ic h ^llt rime they are also sel'monitorine.
ri

It is this lack of genuine speaking opportunitiEs *hi|,"rr -g."-** fb;*"^.,y studenrs, feeling thar, howev.r m.rch u"a *."uui"ry'",n* i.*, they are insufficiently preparS! for spking tn th.

"rJ;;;;; "[ t, "".ra?"i^iil;;:: relegated to the chat s.age at the beginning .na Jil;;r.
of interactive real-time
tark,

;;";f :; orten'simpry_y"y,oi;;h;;;,';;;i:.T,i?',^t-iiff1:t"1.["i:TilI ' expressions' If speaking-ar-9.rar is deart with] it i, with only at the leve.r of pronunciation.
with
Frequentl training

what can be done abour this? The commenr that'this sk'l [i.e. speakingr is always forgotten when someone tethes E;g]iJ';;';r;;.tAxi;j teaching methods (apart from the most b""ki"j^;;9;ii?i'rpr^rc.,g, but less as a sk-i'in irs own right thvi as a m:ans E.,,.n in relatively commun-ication-orented *.thodos.I;;. "fr;;;i;;!o__"..

,,1,X3,ruTf;:.;Jn

acute

""" or"*i.ty *i'." r, ."-., to ,p."ki.,j

is

rack

"n.r'."rt in the

skill

fearures, is

t
t
C
G

C
C C
C
C

rh.ir"i;;

"il'k #*#J;. b..o-.

The process may be complicated. by a tendency to formurate the urterance ^ in the L1 and then first 'transrate'it inio the L2,wthan obvious cost in terms

Com-.rF
stE
G

.rror, ryay mean rhat the self-moniort#or[:f'f prolonged again with.a negative .ffeci in t&-r of nu.".y. ii.r; extended mentar deliberations that qpeakers can enter rnto afe welr captured in this description-or his stude:,: H"-;h,;t'M.a ; ;ril;ustralian !y professor of economics in Japan:
28

nT,?j:ftk'#

j
C
I i
,i

.,

..:@.l,

.t*:tii{t*+r;:'a'.'!:,-t..t-

-- -...i.:-.'

Speaking in another language

Talking with them has been a trial of patience as I watch their faces work like computer screens. Inside, their brains are composing sentences, searching for the most appropriate word, then running the draft past their mind's eye for grammatical mistakes. Finally, the sentence is allowed out. I reply. They look uncertain, sometimes ask fbr a re-run, before their facial screen goes blank while a new sentence is undr construction. They seem terrified of making a mistake, which is no way to become fluent' Yet their knowledge of formal grammar is far greater-than Australian undergraduates and they have

extensivevocabularies.,

In more scientific terms, ih.r. srudents are having trouble distributing their attentional capacityrbetween planning and articulation, not to mention the added deman{"f coping *ith n...iinput. Also, their anxiety is causing :xcessive self-mqitoringr tiey are what-the researcher Stephen Kashen termed monitof overusers. Ofcourse, nqt all L2 speakers agonize to this extent. In fact, some speakers
adopt a completely different strategy, preferring-to use (and possibly overuse) the tde language they have at their linguistic'fingertips', so to speak, rather than constru"ct ovel ttrrrr,..., from s"cratch. No1 ail-Japan.t. l.^tnets ate as portrayed above; Pico Iyer describes another Japanese speaker who has acllieved communicative fluency using minimal means:
Sachiko-san was as unabashed and unruly in her embrace of English as inOst of her compatriots were reticent and shy. ... She was happy to plunge ahead without a second thought for grammaL scattering meanings and ambiguities as she went. Plurals were made singular; articles were dropped, verbs were rarely inflected, and word order was exploded - often, in fact, she seemed

to be making Japanese sentences with

a few English words thrown in. Often, moreover, to vex the misunderstandings further, she spoke both languages at once ... .

'

Rather than 'computing' each utterance using the relatively- slow, albeit more accurate, rul-basd system, Sachiko-san seems to be drawing on a store of memorized words nd chunks. And even when sachiko's memory fails, all is not lost she simply resorts to Japanese. She knows how to make the test use of all availabli rLro.rt..t and as at her command a vatiety of

*'ff,'::fi:i3ff,:il;r9:lfiff

',11..1,'"''::::'i'11'f:1'HHJ::"i1il? ::

communicate in a second language is called strategic.comPetence'

Communication Strategic competence is achieved bymeans ofwhat are called communication strategies rt "t"[-i"r. Sorne commonly encountered communication strategies are: ,

'

circmlocution: such as I get a red in m1 bead to mean shy word coinage: such as 'uegetarianist for rtegetarian 't,ih m gtltg,-!. Spanish word una foreignizin[ a word: (meaning Jfil. fot papers) into the English-sounding a carpet

carPeta

29

t
How to Teach SPeaking

I
approximation: using an alternative, related word, such
as

' ' ' ' '

using taork table

for

zporkbench

using an all-purpose word, such as stztff, thing, maAe, do language switch: using the L1 word or expression (also called code-

switching)
paralinguistics: using gesture, mime, and so on, to convey the intended
meanlng appealing for help, e.g. by leaving an utterance incomplete, as in:
Speaker 1: The taxi driver get angry he os his, erm, how you say?
Speaker 2:

t
G

temper

Speaker 1: he lose his temper and

j_re shout me

Of course, the speaker might deoie that the message

is simply not achievable,

by whatever means, and adopt:'what is called an avoidance strategy, such as abandoning the message altogether or replacing the original message with one that is less ambitious. r

Another type of strategy, called a discourse strategy, is the wholesale borrowing by the speaker of segments of other speakers' ufterances, often in the form ofunanalysed units, as in this instance:
Speaker 1: When did you last see your brother? Speaker 2; Last see your brother six years ago .

A related discourse
Speaker

strateg'y is the repetition of one's own previous utterance:

f
F

1 : The woman hear a noise ... Speaker 2: What kind of noise? Speaker 1:The woman hear a noise, loud one ...

..-

f
C

Such strategies are similar to the production strategies used by proficient in that they help 'buy'valuable processing time and thus maintain the illusion of fluency. It should be obvious that a repertoire of communication and discourse strategies can prove very usefirl for learners in that it allows them to achieve a degree of communicative effectiveness beyond their immediate nguistic means. However, researchers are in nvo minds about the long-term benefits of such strategies. While they may provide learners with an initial conversational 'foothold', they may also lead to the premature closing down ofthe learner's developing language system (or interlanguage) - a process that is sometimes called fossilization. Certain learners seem to become dependent on their strategic competence at the expense of their overall linguistic competence. Sachiko - in the example quoted above - succeeds at being highly communicative, but at what future cost? There appears to be a trade-off between early fluency and iater interlanguage development. Sachiko's Engsh was probably a product of the conditions in which
speakers (see page 7), she used it. It had develooed in face-to-face encounters. with litde time for careful planning or self-rnonit<ring. Nor, perhaps, did she possess the kind of

,p."f
C
C

F F F
C
C
C

30

{?,."'u,

,!+-.ai!:.:-a

Speaking in another language

learned grammar knowledge necessary in order to fine-tune her uttrances, even if time had been available to do so. She had no choice, therefore, but to depend on words, including the liberal use of ready-made chunks, rarher than on grammar. As we saw in Chapter 2, the conditions in which speech occurs exert a

powerful influence on its qualiry; in terms of its fluenc its accuracy, and its complexiry. So, depending sn the kinds of conditions their speaking is subject to,learners are likely to adopt different coping strategies. Some, like Sachiko, who are using their L2 in reaJ-life encounters, with litde or no chance for caeful planning or monitoring, may opt for relying on words rather than grammar tb get their meaning across. This, however, may have harmfirl effects on-their long-term interlanguage development. Others, with more time on tfr hands, will settle for rnir" anall'tic, grammar-based
approach, but-they will pay the price in fluency. Ideally, ofiburse,learners will find a balance between speed and planning, between fast acce\s and slow analysis. And, in the end, the kind ofspeaking they achieve shoufd be the one that is most suiied to their individual nleds. Vherl accuracy is less a prioriry as in Sachiko's case, a non-analic strategy may workbest. On the other hand,where the long-term goals of the leaner involve speaking with precision, a jump-start into fluency may be counterpre.ductive. However, a lifetime spent studying grammar is no guarantee that speaking will come naturall either. As the student (on page 27) complained 'During all my life,I have been doing grammar and reading, but nobody has taught me how to speak English'. Kowledge that is not'available for use'is knowledge that is dead on the page. And this raises two fundamental questions facing teachers of speaking:

' '
What L2 speakers need

What knowledge is required for speaking? How can this knowledge be made available for

use?

We'Il consider each of these questions in turn.

In the previous chapter, we reviewed and sted the kinds of knowledge that
proficient speakers draw on when speaking. Let's revisit that list, and evaluate

to know

how the nguistic aspects of speaker knowledge apply to second language

speaking. (By definition, extralinguistic knowledge, such as knowledge

oftopic,

context, and famiarity with the other speakers need not concern us, since this will be either present or absent irrespective of the language spoken.)

t
*

ial

Sociocultural knowledge The value of teaching socioculrural knowledge, i.e. the culturally embedded rules of social behaviour, is debatable. Many of these so-called rules are based on flims often hearsay, evidence. And they can tend to reinforce stereoq?es, to the point of caricature. The notion that all British speakers of English talk mainly about the weather, do not suffer conversational silences, and say sorry a)J. the ti1ne, is about as well-founded as the idea that they also wear bowler hats and carry furled umbrellas. Moreover, for many learners nowadays such 'rules'may be irrelevant since they will be learning English as an International Language (EIL) rather

ts
all

ch

*''

3f

I
How to Teach Speaking

I
= *

than the English that is used in, sa Birmingham or Baltimore. What is more important than learning local sbciocultural customs might be to develop intercultural competence - that is, the ability to manage crosscultural encounters irrespective of the culture of the language being used, and taking into account that difference and ambiguiry are inhereni in all communication. Simply knowing how to ask Hozt do yu do that here? may be more usefil than i list of 'dos"and don'rs'.
Genre knowledge

I I
I
I

Genre knowledge includes knowing .how different speech evenrs are structured, and this will be particulaily relevant to learners whose specific purposes for learning English include mastering spoken genres of a more formal tpe, such as giving br4iness presentations or academic lectures. For more day-to-day commuplcation, such as service encounters or casual conversation, the genres arg;.likely to be either easily transferable from the leaner's L1 or so loosely defined as to be difficult to teach in any formal
Sense

anwvav.

This does not mean that genres should be ignored or that features of language should be introduced out of their generic context. On the contrary, because genres are recognizable across culhrres, they serve as a usefii way of providing learners with new language in a familiar frame. The question Anything eke? w' make more sense when it is embedded in a shopping dialogue than if it were presented in isolation. What learners probably d {rot need, though, is to be taught the generic structure itself Teaching a learner that you greet shopkeepers on entering a store, and then wait to be
asked what you want, may be somewhat condescending. Likewise, teaching learners that speakers take turns in conversation is tantamount to teaching

I I I

tht boola have pages. What the learners need, more than thi generic structure of the interaction, are specific ways of realizing particular interactional moves. In other words, they need speech-act knowledge.
readers

L2

Speech acts
Just as learners need to know how specific discourse moves are realized, they also need to know the ways specific speech acts (also called functions) ar!

I I I

typically encoded. For example, the following ways of offering advice or

suggestions are common:

I I

l'd ... (if I were you)

You'd bett'er..,. lf you want my advice, you

to ... Why don't you ...

You ought
,.
i

I t

On the bther hand, the following.ways are less common in informal spoken
English:
I advise vou to ... My advice to yowould be ..,.' What I suggesi ii ... I have a sLJggestion ...

I
32

rE==i:-;::4:-:

- *.-'.:. 1,;t::*.i;,,:-.-.

.-i,.; -.

Speaking in another language

While.these, all perfectly possible from a grammatical point of or rafery occur:

Why do you not ... ? I have some advice ... My suggestion to you would be ... lf you want my suggestion, ...

vieq

never

This

suggests

that learners .in.rot necessarily inruit the way that speech

acts are customarily realizd,'nor the way that they are reahzed in spolien, as opposed to written, Ength. There is a good case, therefore, for the explicit

teaching of these forms. Apart from anhing, they are typica-lly reaJtzed in short, memorabl,e/formulas, and therefore can be learned and stored as extended lexical items, much in the way that generations of tourists have used phraseboo.k$ to get by with.

Registelt
will also need to know how to aCapt these speech-act formulas for different siruations, according to such contxt variables as the status of the
I;earners

person they are talking to. Exposure to different registers of speech, plus directed attention to the ways in which spoken language is made more or less formal, should be sufficient, at least for general English purposes, to sensitize learners to this area. Role-plays (see page 96) are probably one of the best ways of practising different constellations of register variables, such as the differences that social status makes. Discourse Discourse knowledge, as noted in the last chapter, involves using grammar and vocabulary in order to connect speaking turns and to signal speaker

! I I
I ! ! I I I
I

intentions. Discourse knowledge also assumes an understanding of how speaking turns are managed - knowing that, for example, talk is collaboratively constructed through the taking and yielding of turns. However, since this is a universal featue of spoken interaction, it is not something learners need to be taught. They simply need to know how these turn-management moves zre realized in the secbnd language, through the use, primarily, of discourse markers. The list of discourse markers in Chapter 1 (page 9) would serve as a useful starting point in this area. Grammar We noted in the preceding chapters that grammar knowledge for speaking purposes consists largely of those grammar systems that favour rapid, realtime speech production. Since spontaneous speech is produced in clauselength units rather than sentence-length ones) a sentence grammar will be of limited usefulness for speaking. It is sentence grammar, however, that has always been the main focus of language teaching. Learners are taught to manipulate relatively lengthy and complex constructions that are more typical of written than of spoken language. To take one example: the conditional uould is traditionally first taught as an element of the second

I
t

33

a
.e:E*i"=,-.-=*.
---,

. .

^-

How to Teach Speaking

-a
zuould-claus-e (zf b,ad th.e time,I utould study barder; if Id bad he me,I uauld ! haae studied barder),rarher than being tauht rn l.rn"r,t of zoould-clatses "ranalyses on their own (I zaould neaer eat borse meat).But of corpora of spoken

and third conditional constructions, which consist of an if-clause and

English show rhat uould-clauses occur four time more often withut an lf clause than with one. Likewis, learners are taught grammar items-without a clear distinction being made_befween spoken and written grammar. of course, there is a great deal ofoverlap, but there are certain structures that are much iess frequ;nt in speech than in writing, such as reported.speech, subordinate clauses, relative clauses, and the passive. on the ther hand, some features of spoken syntax (such as heads and tails, and ellipsiq - see page 2r),get little or no artention at all in many mainstream ELI corses. speaking would probably need to include . t:9.. grammar for inforryd ; the following items:
associated

t t t t

' ' ' . ' '

a command

of present and'past simple, and the abiry to use the latter to

sequence narratives.

famarity with the use of the continuous and perfect aspect forms of verbs, both to frame and background information in narratiies, as in I zuas coming out of tbe^supermarket ... it d been raining ... . a knowledge of the most frequently occurrin[ modal and semi-modal verbs (i.e. can, tlill, uould, bave to, going to, ,rr t). ,h: abty to formulate quertioni, especially yes/no- but also
to-questions.
some basic conjun ctions (and,
so,

C C
C

but) in order to string together sequences

be said

ofclausal and non-clausal units. one, or rwo all-purpose quoting expressions, of the
saxQ... ryPe.

... and tlten I

F F

f
C

Vocabulary

we noted in

words the learners will need, the most useful criterion seleltion is pr*ablyfequen:/. A working knowledge ofthe 1,500 most frequent words in English would srand a learner in god stead. Even the to 200 most common words will provide the learneiwith a lot of conversarional mileage,
since they include:

the last chapter that native speakers employ over 2,500 words to cover 950/o of their needs. Learners can probably-griby on a lor fewer, T1yb. half that number, especially for the prrrpos.i o?."rrrrl conversation. obviousl for_more speciazed purposes .h business negotiations or academic soeaking, they will ned more. Short^r of knowing eiactly which

"for

F
F

'

all the common question forming words, such as

zubose

zu/tere,

...?

uhy, uhen,

hozu,

22) such as this, tbat,

34

modal auxiliary verbs: zoould, zoill, can, may, might, sbould etc. "11 all the pronouns, such as it, I, me, 1tou, tbey, us, nd te possessive forms such as mj, your, hers, tbeir the demonstrative pro-nouns and other common deictic devices (see page
here, tltere, nozu, tben

,lr.

e
G

f
J

t
[!'. - .:. .;:.='.,":,,,t.--,,.,r,
3

t
r
:

Speaking in another language

I I I I I I

' . ' . ' '

all the common prepositions, such as in, on, near,fro., 4fttri betueen the full range of spoken discourse markers (see page 9), such as zuell,
so,

oh,

but, and, right, nota

common backchannel expressions, such as really, no, zabat, and boztt ... as in hou atuful! hoqD rDonderful! common iequencing andiinking words, such as then,frst, so, end, 0r, next common ways of adding emphasis, such as really, very, just, sa common ways of hedging (i.e. reducing assertiveness), such as actually,
guite, rather, sort (of) all-purpose words, such s thing, things, place, tirne, zuay, make and do

i
I I I t I t
'.t

Most learners'dictionaries, such as the Longman Dictionary of Contemporary Engluh (LDOCE),now higtrlight high frequency words and even indicate their relative frequncy in speech and in writing. Here, for example, is the bet \n the LDOCE: entryfor th.

betlp!

v past tense and past partliple be[, present participte

lberl

;-

The formula 51 in the box means lwnrp,,.- ' "r'..,.:..1-"".,,.*o4,*, .r 1nn .-,^-l^ . . .' | | , "ii, .: .'"r' ,, i':':: words rl-J:'r;lrf ':rr:;Ji-i::i::r':'r::: that bet is in the top 1000 lmrd* 50 of spoken English. In fact, betis so i -t . .o.,i-o' i' sioken Engsh that i*5tr"iiT?t'i"T"SHf Y#Jfffi*"'ff:

'g1al9$l-P"g'tIn the following efiract (from data that was used in the preparation of the
coursebookN tural EnglD, in which a Polish learner is describinga shopping experience, the words that do not fall in the top 2000 words band in spoken English, according to the LDOCE, are underned. The words that are in the 1000-2000 band are in italics. The words in the 1-1000 band are unmarked.

I deserves

'' .ment: a speclar comment:

it

]S

I
I

:ls

t, 4
n.

Al lt happened I think two years ago, I went to a shop. lt was Saturday. I usually do my shopp ing on Saturday. 5o I went to a shop to buy shoes, and

*
dS

It
)'t

v'
a,

I =rc t
t:

lwenttothat particular shop in which lfound my pair of shoes... B: Expensive? A: Yeah, quite expensive. B: How much? . A: About forty to f-fty pounds, something like that. So I went there, it was full of people and ltried on the shoes that I liked, so I decided to buy them. So I bought them. I went home after that, but it was almost the end of the day, the shoppng day, so it wasn't left a long time for the shops to close, so when I went home and decided to try on the shoes again, I saw that in the bag were two left shoes. 5o I had, well, it was quite an expensive pair of slroes, so I tried to go back to the shop and exchange them so although I knew that they will exchange them, I was a bit worried. But I was late and the shop was closed already and I had to go on ... next day on Sunday to get the proper pdir of shoes. B: Did you manage to get it?
A: Yes, finally.

'

'

il

*
il

35

.t

iE

How to Teach Speaking

f
The learner has told her storyusing onlywords within the-2000 top frequency range(apar-tfromfour),andthevastmajorityofthewords -92o/o'of thetotal, words used - are in the top 1000. Th student (who was in an advanced class) manages to be .colmufcatlvely effective^using only a mited range of words. (By the way, the fact that saturday and.sundly, igfrrrl and,fftl, are relatively infrequent may seem counternruitive, especially girnth^tih.r. are words that are taught very early in a langrag.^.o..,rr. ih. f".t is that they belong to frequendy occurringiets - i:e. ayiof the week, and numbers - but the frequency of occurrenceJis distribute across the mmbers within the set. This suggests there is a case' for teaching not just frequent words .that but frequendy occurring sers of words) . The point, made eylier, that let forms part of commonry used phrases, alerts us to the fact that, as irnfortant as individual words'are, so too are chunks. In the last chapteE w..9 3aw how a mental store of *.*orr.d chunks is essential if.fluency is to-6e achieved. In fact, kngw-ing that bet is very freqrlelt is of little use unlegs the learner knows that it forirs the nucleus of the following high frequency chunks:
I bet

F F F

F F

F F

l'll bet
you bet wanna bet? don't bet on it your best bet (is ...)
my bet (is...)
a safe bet

E E

literal sense, asin I bet on a borse in the frfth ,ace. Moreover, the advantage of learningiih. formul"ic chunks associated with high-frequency words is that many oflhese form common syntactic strings, such as be + going + to + verb as it! going to rain. so, by leanins the chuik o the learner gets the string - and the glamlnar - ,for fre'.

The problem is that there are a great many more combinations of words than there are individual words. which chunks are likely to be of use to learners? Until more informarion about the frequency of chnks becomes available, a rule of thumb might be to focus on the chunks that are associated with the most common words in the.language. The frequency of these common words may- ole a lot to thefact that they form.at least one element of commonly used chunks. It's a safe be.t, for example, that the word betis moreoften .rs. in its idiomatic combinations (I be.t, you bet, zuanna bet?) thanit is in its more

F F

G
Av

ror

Phonology Finall we come to.phonology, an area which is perhaps the least amenabre to conscious control at the moment of articulation. Mst adult learners will b9o"y, to varying degiees, the influence of their first languaqe pronrrnciation 'wh9n r.p...#p a second.lango?gg,and this.need not b"e so long ""ptut..n as intelligibiliry is not threarened. Intelligibiliry however,'is in the ear of

i-b

F F F

e
36

Speaking in another language

the beholder. That is to

say, what might be intelligible to one listener is not necessarily intelligible to another. Native speakers, for example, frequently identify the non-native-like use of stress, rhhm, and intonation as being a greater bar to intelligibiliry and a stronger marker of accent, than the way indidual vowel and consonant

sounds are pronounced.

This is particularly acute when lack of fluency

segments speech into very short runs, as in this example, where a Japanese speaker is discussing her plan-to show picrures of moden Japan to US school children. Stressed words are printed in capital letters, and pause lengths are shown in brackets. in secgpds:

... not only WORDS t (.4) | can 5HOW I (.4) the (.4) PICTURES z (.6) HELPED (.8) STUDENTS o un- HELP- (.5) STUDENTS , (.2) UNDERSTAND the (.4)

JAPANESE CULTUfrT
/-l
1'

As Ann Wenerstrom comments, by speaking in such short bursts, with each wod giv6n almost equal emphasis,'the ef[ect is to obscure the main point of the discourse because every word seems to be singled out as worthy
of comment'.

Of couise, native speaker judgments are irrelevant if speakers are speaking English as an International Language. The researcher JenniferJenkins examined the main causes of communication breakdown when non-native speakers are talking to each other and, on this basis, identified the following areas of pronunciation as being crucial for intelligibility:

' . ' .

certain'core'consonant sounds (but not vowels) the contrast between long and short vowels (as in bit and heat) consonant clusters, especially those at the beginning of words, such
\n product

as

2r

sentence stress, r.e. the correct placement of stress in an utterance, especially contrastive stress (sei my COUSIN, not m1 sister)

If this st seems relatively short, it is nevertheless consistent with the view we have taken so far - that fluent spoken English is not simply a function of a wide-ranging knowledge of grammar, an extensive vocabulary and a native-like pronunciation. In fact, as this chapter has tried to demonstrate,
fluency may be achievable with relatively minimal means.

Availability for use: implications for teaching

So far we have attempted to answer the first of the two key questions posed on page 31, i.e. what knowledge is required for speaking? Now we can turn to the second question: how can this knowledge be made available for use? Essentiall to ensure availability for use, there are three Processes involved. Learners need to be made aware of features of the target knowledge-base, they need to integrate these features into their existing knowledge-base, and they need to develop the capacity to mobilize these features under realtime conditions. Depending on the view of learning that is adopted, these processes are named, ddscribed, and rated differently. There have been at least three theories oflanguage learning that are relevant to the teaching of

37

I I

How to Teach Speaking

speaking: behaviourist, cognitivist, and sociocultural theory and we shall briefly review each in turn. According to behaviourism,language learning is essentially the formation of good language'habits'through repeated reinforcement. In its popularized form, audiolingualism, the three stages oflearningwere called presentation, practice, and production (PPP). The three-step PPP process was aimed at developing automatic habits largely through classroom processes of modelling, repetition, and controlled practice. PPP was apped originally to the teaching of grammar, but, by extensiori it has been used to structue the teaching of language skills as well, including speaking. A rypical teaching

sequence

might involve listening tq, and imitating, a taped dialogue,

followed by repetition of features of the dialogue, and then performance of the dialogue in class. ./ A cognitivist account of langtage learning rejects the behaviourist view of
the learners as emptyvessels ggiting to be lled, and instead credits them

with

an information processing {apacity, analogous to computers. According to this view, the learning of a complex skill, ke speaking, is seen as a movement from controlled to automatic processing. Initiall conscious attention (or awareness-r"lting) is apped to the learning of the individual stages (or rules) of a procedure that, through repeated activation, ae chunked into a single manageable'program'. This is integrated into existing knowledge, a stage which will involve some restructuring of the user's nguistic system, and is then readily available for use, with minimal attentional control on the part of the user. This is the stage known as autonomy. ^ In teaching terms, cognitivi"st theory replaced th; PPP model with one that progresses from awareness-raising, through proceduralization, to autonomy. -In fact, it is only the first stage that is significandy different, in terms of'classroom practice. Awareness-raising impes an explicil focus on the rules of the system, whereas strict audiolingual practice insisted on simply imitating models without any explicit attention being given to the rules that generated them.

learning of a first and a second language - is mediated through social and cultural activity. To achieve autonomy in a skill, the learner first needs to experience other-regulation, that is, the mediation of a 'better other', whether parent, peer, or teacher. This typically takes the form of assisted performance, whereby the teacher interacts with the learner to provide a supportive framework (or scaffold) within which the learners can extend their present competence. Through this shared activity, new knowledge is jointly constructed until the learners are in a position to appropriate it - i.e. to make it their own - at which stage the scaffolding can be gradually dismantled. Learners are now able to function independently in a state of self-regulation. A good example of this is the way an older hild will teach a younger one the rules of a game, by both talking and walking it through, until the younger one has got the hang of it.

The cognitivist model prioritizes mental functions over social ones. Sociocultural theory on the other hand, situates the learning process firmly in its social context. According to this view, all learning - including the

Co

38

t I

I I t

T
rd

-:-it,i|:a*--,.:.'+;+.r,

.- =-..- ,'..- -...--.,:.- j.;---.-,..i*.,,1#,".+--,.!.34jn;;!ri4i*t!l]?s*'r.:

,!i

Speaking in another language

I
;

Learning, according to the sociocultural view, is fundamentally a social phenomenn, requiring both actiry and interactivity.In classroom terms, it lakes place in cycs of aisisted perfoimance , in which iearning is collaborative, co-constructed, and scaffolded. For example, learners may set about solving

a problem in small groups, during which the teacher intervenes when

3d

t I oi
il

to

to provide suggeitions or even to model the targeted behaviour. AI tree iheories hve elements in common, especially when these are translated into classroom procedures.The following table attempts to display the relation between diffient elements of each model:

n...rt"ty

I oi

behaviourist theory presentation, modell ing

practce
t

cognitivist theory
awareness-raising

sociocultural theory

other-regulation appropriation self-regulation

proceduralization, restructuring automaticity, autonomY

productiory,i

t
Itl

to

or

I
t
;

:ll,

t
*

to

tis

$
he

il

These surfce similarities, however, shouldrit be allowed to disguise the fact that each theory reflects z very different conception of the mind. The behaviourist mind i simply a brain, pushed, pulled, and moulded by forces beyond its control. The c-ognitivist mind is a computerized black b-ox, busily processing input into o.rtpnt. The sociocultural mind is a netwo-ik, I joint construct"of'the discouri. .o*-rrniry through which it is distributed. Each metaphor for the mind clearly has differen-t implications in terms of of language learning in particular. Nevertheless, each theory learning, "nd incorpolat., ,t"g."*.h roughly equates with awareness, whereby the " something .*. R"a each theory attemPts to explain l.arn, encounters" how this knowledge is integrated, or appropriated, into the learner's existing sysrems. Ria n""Uy, ach -theory aicepts that at least some of this .r.* ko*l.dge becomes ".,ril"ble for use: it is automated and the learner is autonomous'

Conclusions

t
I l'

he

to

In this chapter we have looked at speaking from the point of view of the learner, coping with the challenge of speaking in a second (or other) language. Essentially, the difficulties that the learner-speaker faces break down into two main areas: . knowledge factors: the learner doesn't yet know aspects of the skills factors: the learner's knowledge s not sufficiently automated to ensure fluencY. As a result, there maY also be: . affective factors, such as lack of confidence or self-consciousness, which might inhibit fluencY. Learners compensate for their insufficient knowledge of the language system by using communication strategies, and they compensate for lack of fluency by using discourse strategies' Over-reliance on'such strategies, however, could lead to premature fossilization of the learner's interlanguage. Fossilization may also

ed
nc'l

language that enable Production'

it

F'

of

= 'h

I t

39

E
* {
result from a preference for a lexical mode of processing, as opposed to a more grammatical one. In terms of the knowledge base that enables speech, learners need:

How to Teach Speaking

. .

a core grammar.
a core vocabulary

of at least 1000 high-frequency items.

:t',ffi ;'Hfr :*$fJ#li'J#t1,,!"i".1"*,li,,chas

requesting or inviting). immediate situation. ,:t In order to activate these.&nowledge areas and make them available for use in fluent,face-to-face talk, the learning process needs

'f
;
A
i.e.

' mastery of those features of pronuntiation that inhibit intelligbility. Also important is that speakers remember to take into account context factors, including the cultural context and the context of the

@u

:" il'Ht jj j"fr:*'ff

awareness.

;:i i,ii; ll?',ll:l

J,"i?i'

""

ed

e _b a

e, i e

to integrate these features into ther existing knowledge-base,

appropriation. to develop the capacity to mobilze these features under real-time conditions and unassisted, i.e. autonomy.

l f
C

Looking

ahead

the chapter that follows we will look at the first of these areas awareness - and suggest ways that learners' awareness of the features of spoken language can be optimized. In the two subsequent chapters, we will consider activities that target appropriation and autonomy.
In

t
]

f
]

I
I I
t
I
;
40
I I

t
;

'r-