. . .
Vol 32 No 2, April - June 1994
Page 26

Forum2.jpg (7.95 Kb)


Rote, Rule, Role, Rule: From A to B and Back Again in Sutra-Chanting-Style Exercises
by David Kellogg

On the slopes of Yushan in Fuzhou, China, there is a Buddhist temple with an inscription commemorating the stay of a nineteenth- century reformer who spent some time there learning English. The couplet, which I no longer remember exactly, speaks of the harmonious medley of Buddhist chants and English dialogues breaking the brittle morning air, "With A and B riddling each other, and answering the Buddhist sutras."

Chanting texts is still very much part of EFL in China, surrounded by a halo of matinal asceticism. At its worst the practice consists of an unenlightening drone that deadens intonation and substitutes repetition for comprehension. But it can be a bracing morning warm-up to something more innovative and interactive. Perhaps, by going beyond the unvarying chant done in lockstep to the underlying rules and roles of common conversational situations, such EFL "sutras" may actually, as Buddhists claim, "cultivate character."

Attention to intonation

One obvious claim for chanting is that it can enhance intonation. There is an excellent and widely used set of cassettes by Carolyn Graham that stresses this aspect. Yet as Brazil et al. (1980) point out, intonation is strongly context-dependent, and what is learnt through repetition may be useless to inappropriate in interactive responses.

Interestingly (for me), the most successful of Graham's jazz chants are those that have a kind of sitcom straitjacket of situational presentation. These are contexts in which the roles are ironclad and the rules of their interaction highly inflexible - husband rushing a wife out the door, wife trying to drag husband out of bed, arguments about mothers-in-law, etc. In these stereotypical situations, the stresses are clear-cut and regular, which allows Graham to turn dialogues into rhythmic and repetitive chants.

Alongside - even through - certain intonation patterns, then, good chants can be used to introduce a very rough-and-ready "grammar" of certain conversations where moves are at least partially stereotyped and predictable. But how? And how to get beyond the stereotypes to more fluid roles and flexible responses?

A highly derivative effort

I began by simply trying to ape Graham's efforts, as a way of warming up a very cold class and weaning them from reading aloud in completely flat intonation. Here, for example, is the rather boisterous "interruptions sutra," for a class on various discussion techniques including interruption:

I'd like to talk to you today.

Sorry to break in, sorry to cut in.

Another thing that I'd like to say . . .

Sorry to cause you a little delay.

I'd like to make just two points more.

May I interrupt, may I have the floor?

Another thing we must all bear in mind. . . .

If I could just come right in here one more time.

This can be made very crudely interactive; the two halves of the class chant the first two lines simultaneously again and again. If the interrupters succeed in drowning out the speakers, they then take the speakers' part in the next two lines, while the speakers must try to interrupt, and so on.

Yes, that's all good fun, but of course it is not conversation; it is practically content free, and the "interaction" is a matter of sheer volume. (One can argue, of course, that much interaction between native speakers is like this - one thinks immediately of press conferences and presidential debates - but it is hard to argue that it needs to be taught.) So what next?

Greetings classified

Next, I made the chant itself less interactive, but chose a conversational function in which the choice of phrase was slightly more context-dependent. By that I mean that the form of the question depended on the desired reply, and the replies depended on the form - and content - of the previous utterance. Here is the "greetings sutra":

A. How's your job, how's your life, how's your wife been treating you?

B. Not too bad, not too good, I've got too much work to do.

A. Where've you been, where'd you go, tell me 'bout your holiday.

B. Stayed at home, stayed in bed, all the family stayed away.

A. Now what's up, now what's on, now what's happening with you?

B. Not a lot, nothing much, tell me what you've been up to...

( repeat from the beginning with sides switched )

Greetings are, of course, among the most stereotyped of exchanges. Yet we can see that casual greetings are crudely situation- and context-dependent, unlike formal, invariable greetings like "How do you do." For one thing, we have three very different types of questions here, easily distinguishable in their grammar and their intention. Clearly A is a person who knows something about B in each case, and is demonstrating polite concern for B without prying into details.

"How" questions are ideal for this purpose, since the answer does not need to go into detail. A vague adverb or adjective will do. (Though as in the example "I've got too much work to do," an obliging interlocutor may go into detail or answer indirectly by a wide variety of grammatical forms - the creative element of conversation will not be long denied!) Questions in perfectives and past tenses often refer to recent activities, and are answerable by statements in the past tense. Present-tense questions, on the other hand, are questions about current activities, and need to be answered by noun phrases, verbal nouns, and, of course, present tenses. Once these "rules" are abstracted, they can be creatively applied to roles. For example:

Make up "how" greetings for the following situations:

Your young niece is learning to play the accordion.

Your colleague recently returned from a geological survey of

Your old classmate stole your first girlfriend Lily Li in middle school. You haven't seen him for a long time. Suddenly, you meet him on the street.

Now make up responses. They should contain an adjective or adverb, and then some additional details.

Now get responses for your questions, and use your responses to answer your partner's questions.

Do the same with "where" and "what" questions.

Errors - inevitable and inscrutable

Mistakes are immediately possible, and, since this is an EFL classroom, virtually certain. Worse news, the rules violated are not at all clear-cut. To my ear "How's your life?" "Stayed in bed" is an incoherent exchange, reminiscent of the following unsatisfactory but frequent exchange I have with my students: "How was your class?" "Listening." On the other hand, "How's your life?" "I've got too much work to do" is coherent, but I am hard put to explain why. Tense is involved, but there is more to it.

We saw that "I've got too much work to do" in the above does not contain the expected curt adjective or adverb. It can be understood as an indirect reply. The problem is that in life all but the most perfunctory responses to greetings may be anomalous in this way. One way of encouraging the expression of specific and heartfelt concern is to be very specific and to be forthcoming with sentence and detail when one is merely asked for an adjectival or adverbial evaluation of one's experience. And the replies to such replies are still more unpredictable. What, actually, does one say when one asks "How are you?" and one receives an answer like "I'm about to die and I've got no one to bury me"?

The long goodbye

One interaction that escapes from this endless tree of branching possibilities is goodbyes. Consider this "so long sutra":

A. I've got to run, I've got to dash, I've got to move along.

B. Well, so do I, so I suppose that I'll be moving on.

A. I'll call you soon, I'll drop a line, I'll soon give you a ring.

B. Well, take good care and keep in touch, good luck with everything.

The assumption here (taken from page 100 in Blueprint [Abbs and Freebairn 1989]) is that goodbyes have the generalized structure: reason for leaving, agreement, further contact, and leave-taking phrase (with the proviso that in England the arrangements for further contact can be so formalistic and vacuous as to be practically phatic). Here, at least, the tree of possibilities converges, with the options narrowing rather than broadening at the final branching node. Sometimes, of course, it can move that way with dizzying speed; here is a sort of "so-short sutra":

A. Gotta go.

B. So do I.

A. See you 'round.

B. Yeah, bye.

And of course, one may skip whole stages, a possibility I'll get back to under "apologies" below.

The "so-long sutra" produced quite a good lesson, actually, with a special section on expressing obligation, another on modals and agreement ( So . . . I, Nor . . . I ), another one on future plans and suggestions ( I'll . . . , How about if I . . . ), and only the final part on an unproductive, uncreative, and wholly predictable, limited set of lexical exponents (leave-taking phrases).

Does a good chant yield a whole lesson plan, though?

On reflection the attempt to make the initial chant break down into specific stages of the lesson was misguided. There is no particular reason why the crude bit of discourse analysis done here should yield what Fanselow (1977) called "teachemes," (smallest possible meaningful teachable units?). For one thing, learners tend to get absorbed in the task at hand ("obligation," or "agreement," rather than relating the two) and forget that this is a subassembly of a larger conversation that they are assembling. For another, part of the problem is to teach learners to distinguish one functional "phase" from another; breaking it down for them defeats that purpose.

A better approach, I found, was to combine all the phases in a single presentation section. In order to make the necessary pronunciation-repetition exercises meaningful, I asked learners to sort a list of phrases according to these four functions: reasons for leaving, responses, arrangements, and leave-taking.

Now decide ( a ) how people signal that they want to end the conversation and give reasons for ending it, ( b ) how the other person responds in each case, ( c ) what arrangements they make for further contact, and ( d ) what leave-taking phrases they use. Mark the phrases a, b, c, and d according to their function.

___ Listen, I really have to/ought to be going now.

___ See you around.

___ I must get back to work.

___ Listen, why don't we meet for lunch?

___ I'll give you a ring.

___ Me too.

___ Take care.

___ So must I.

This is done in pairs and frontally checked, as necessary, by having people give examples of each category, so that they have to pronounce the exponent. A more "language-like" way is to have alternating sides of the classroom give examples that logically follow (e.g., "I've got some work to do." "Yeah, me too." "Listen, why don't we . . .") as in a volleyball game, where inappropriate replies (dropping the ball) give points to the other side. An advantage of this, of course, is that it leads directly into pair practice. ("Right, okay, now I want you to switch from volleyball to pingpong. Not doubles, these are singles matches.")

Less structured conversations: The apologies sutra

In more open-ended exchanges, the structure written into the chant begins to break down almost immediately. Take this "apologies sutra," which begins with a pattern used by Graham:

It's not good enough, it's not fast enough, I'm afraid it just won't do.

I'm terribly sorry, I'm awfully sorry, I apologize to you.

What's your story, what's your excuse, what do you have to say?

I've never done it before, you know. I'll never do it again that way.

Well, never mind, not to worry, I suppose it's really okay. I'll make it all up, I'll pay it all back, I just don't know what to say. . . .

Here the structure I intended was roughly six parts: complaint - apology - request for explanation - explanation - forgiveness - offer of restitution. Is this structure purely artificial, simply written into the initial chant and then recovered by the teacher through sleight-of-hand presentation? Yes and no. The categories in this chant are also recoverable from "authentic discourse" tapes - for example, I asked students to code the apologies in the tapescript of Leo Jones's Functions of English (1983, Unit 12).

On the other hand, the sequence of elements is often, in life (and also in the Functions tapescript), mixed up. Clearly, one can apologize preemptively, before a complaint; one can offer restitution and an explanation in the same breath, etc. Any such preempting of the structure, of course, completely alters what follows.

The key problem: The non-random but unpredictable reply

And that is the rub, and the nub of the problem. It is too easy, in a functional course, to teach questions. The responses, and the responses to the responses, get less and less predictable, and less and less teachable. Now of course, we can and do make language more teachable by making it less creative and interactive. But the creativity of language is precisely what is demanded by the specificity of the situations to which it must be appropriate. To make that creativity, that specificity, learnable, we need to go beyond rote or role, to certain very bendable and often breakable "rules" of interaction.

It seems to me that rote can be as good a starting point as any other for exploring the treacherous rules by which conversation unrolls. After all, in one sense, we restrict language in the very act of presenting it as a finite set of possibilities, and rote sutra-chanting is simply the most extreme example of this. Hopefully, learners will not stay in the situational straitjackets we have fashioned, however well they seem to fit.

Perhaps that is the rote-learning riddle of Yushan temple. Creativity, in the sense of being sensitive to situation specifics, is not something that teachers can present students with; to present language ("this is the way creative people greet in English") is to deny it. It is unpresentable, yet it is omnipresent, and it comes back to haunt A as soon as A must think of a reply to B.


  • Abbs, Brian and Ingrid Freebairn. 1989. Blueprint. London: Longman.
  • Brazil, David, M. Coulthard and C. Johns. 1980. Discourse intonation and language teaching. London: Longman.
  • Fanselow, J. F. 1977. Beyond RASHOMON - Conceptualizing and describing the teaching act. TESOL Quarterly, 11, 1, pp. 17_18.
  • Graham, Carolyn. 1978. Jazz chants. Oxford: Oxford University Press.
  • Jones, Leo. 1983. Functions of English. Cambridge: Cambridge University Press.

David Kellogg has taught in China continuously since 1984, except for 1990 1991, when he did a Master's in applied linguistics at the University of Essex. He is now teaching at Shaanxi Institute of Education in Xi'an.


Back to Top


Vol 32 No 2, April - June 1994
Page 26
. .

On October 1, 1999, the Bureau of Educational and Cultural Affairs will become part of the
U.S. Department of State. Bureau webpages are being updated accordingly. Thank you for your patience.