Adapting and Writing Language Lessons/Chapter 4

by Earl W. Stevick

CHAPTER 4

WRITING ADAPTABLE MATERIALS

THE PROBLEM

If the blind lead the blind, shall
not both fall into the pit?

Luke 6 :39

The preceding chapter suggested ways of adapting and using language materials that are already at hand. Whoever rejects what is at hand and writes his own lessons assumes a double responsibility: to produce something that really is better for his purposes than what existed before, and to reduce the likelihood that those who come after him will feel that they in turn have to write their own courses. Chapter 4 applies to the writing of new materials many of the same principles that guided adaptation in Chapter 3 and its appendices.

Writing lessons for seldom-taught languages and writing them for commonly-taught ones differ in much more than the names of the languages. Students in seldom-taught languages are likely to be more mature, and many have already in mind some imminent and very definite use to which they hope to put the language. Yet in the development of language study materials, writers are only rarely qualified to speak for the public among whom the prospective students are to live and work. Sometimes, the materials-writing 'team' consists of one person, who is biologically coextensive with the only student. It is for this situation that we have Gudschinsky's How to Learn an Unwritten Language (1967: cf. also Ward [1937] and Bloomfield [1942]). Most frequently, however, the materials writers directly represent neither the student nor his future audience. All too often, they have begun and completed their work without even seeing either the students (because they have not yet been enrolled) or the audience (because they are remote either geographically or socially or both). The plight of such writers is implicit in the following quotation from the introduction to one textbook:

This grammar is designed for a highly heterogeneous audience, composed primarily of the following groups: (1) area specialists interested in language- or culture-studies of all parts of tropical Africa: (2) ethnographers focussing on the social structure of the Edo-speaking peoples: (3) historians working directly or indirectly on the Benin project sponsored by Ibadan University; (4) linguists more concerned with analytical procedures than with specific languages or language-groups: (5) missionaries in the Benin area who wish to reach their parishioners more immediately than they can in English: and (6) Bini-speaking teachers and writers who seek a more exact understanding of their own language than conventional training in English grammar offers them. As though this assemblage were not already diverse enough, the grammar is intended secondarily for any and all foreign visitors to Benin Province who may find a description of one of the Edo languages useful or interesting. The author's hope is that this volume may have something to offer to each of the abovementioned audiences, though he fully appreciates the very real possibility that it may fall between scholarly stools in such a way as to leave all of its prospective audiences unsatisfied.

Wescott, A Bini Grammar, page 1

Except where materials are being prepared in the midst of an ongoing training course, consultation with students is obviously impossible. (Consultation with students must take place during adaptation, and it is largely to enhance the status of adaptation that we have placed that chapter ahead of this one.) Getting preliminary information from spokesmen for anticipated audiences is not impossible, however, and writers should be willing to spend some time and considerable effort in assembling it. At the same time, they must remember that the potential audience for a student going to Greece is not just undifferentiated 'Greeks.' Potential audiences for different trainees interested in the same country will be partly alike and partly different, even within a series of programs sponsored by a single sending agency.

All this uncertainity, instead of filling writers with diffidence and godly fear, seems to send them forth to sin all the more bravely. The result has been publication of much that is idiosyncratic, some of which is good, some of which is useful to others, and some of which is neither. In the 1960's, when money for such enterprises was more freely available, this profusion was either inspiring, amusing or annoying, depending on one's point of view. In the 1970's, however, we can no longer afford to invest many thousands of dollars in a course which embodies the theories, needs or prejudices of its own writers, but which may then be rejected by most other prospective users. This is particularly true for the less-frequently taught languages.

The problem, then, is how to minimize the likelihood that a set of materials will be rejected by new programs operating with different aims, different kinds of students, different theoretical convictions, and different prejudices. The solution that is proposed here depends on building into the materials a number of clearly-defined options relating to the choice of material, its possible replacement, and the ways in which it may be used. On the other hand, too much flexibility may be just as disastrous as too little. For those who want to follow, materials must give firm guidance; for those who want to tamper, there must be clear indications of how to select, rearrange, and complement without destroying.

The goal which we have set for this chapter is more delicate than for any of the other chapters. We are attempting to tell others how they can do something (write adaptable materials) without telling them how they must do it. To put the same dilemma in another way, if suggestions are to be helpful they must be fairly concrete, yet the whole purpose of these suggestions will be defeated if they are taken as prescriptions for yet one more ultimate format. This chapter therefore presents a somewhat idealized scheme for materials development, which will be partly exemplified and partly contradicted in the appendices which follow it. Like Jabberwocky, it is supposed to fill the reader's head with indefinite ideas; unlike Jabberwocky, it is supposed to help the reader to produce very definite ideas when he applies it to any specific problem in materials development.

A WAY

One of the most noteworthy (and least noted) attempts to view the writing of materials for seldom-taught languages is John Francis' Projection (1969). Using Francis' analysis as a point of departure, we may say that the writing team must provide for three 'functions' (specification, presentation, articulation) on each of two 'scales' (coarse-grained, fine-grained). The flow chart (Fig. 1) shows how these are related to one another:

Fig. 1

Coarse-grained Specification. This is the responsibility of a qualified spokesman for the potential audience. Just who is qualified for this role depends on who is being trained for what. The spokesman may be a group of community representatives, or a ministry of agriculture, or a Peace Corps country director. For the Thai materials described in Appendix G, the principal spokesmen were two returned Peace Corps Volunteers, one of whom had spent two years in malaria control and one of whom had spent the same amount of time in leprosy work. In Appendix H, coarse-grained specification was provided by a director of area studies in a stateside training program. In Appendix I, it was provided by teachers, but by teachers who had had observed at first hand the way their students might need to use the language.

Coarse-grained specification answers questions in each of the three 'dimensions' of Chapter 3:

1. Socio-cultural. ('What will be the trainee's position relative to speakers of the language -- and bearers of the culture -- in which he is interested?') This information may be given in the form of a careful prose description. The source of the information itself may be discussion, questionnaires, surveys, or some combination of these.

2. Topical. ('What kinds of messages will the trainee need to handle?') This information is best given in the form of a list of problems or tasks. Some of these may involve social situations of interest to a broad spectrum of trainees, while others will be within special fields of interest or technical specialization.

3. Linguistic proficiency. (' How well will the student need to understand, speak, read and write the language?')

This information as it comes from spokesmen for the future audience will not be in technical linguistic terms, but will be stated functionally, perhaps somewhat as follows:

The trainee must learn to participate effectively in all general conversation; he must be able to discuss leprosy control and health education with reasonable ease; comprehension must be quite complete for a normal rate of speech; his vocabulary must be broad enough so that he rarely has to grope for a word; his accent may still be foreign, but his control of the grammar must be good enough so that his errors never interfere with understanding, and rarely disturb a native speaker.

The above statement, which is based on an official description of the Foreign Service Institute's widely-circulated S-3 rating, represents a rather high goal. Much less demanding would be:

The trainee must learn to ask and answer questions on topics that are very familiar to him; he may not understand even simple sentences unless they are repeated at a slower rate than normal speech; he may make frequent errors of pronunciation and grammar, and his speaking vocabulary may be inadequate to express anything but the most elementary needs, but he must be able to make himself understood to a native speaker who is used to dealing with foreigners; he must be able to take care of his need for food, street and road directions, matters of personal hygiene and cleanliness, including laundry, tell time, and handle basic courtesy requirements.

This, of course, describes only a very rudimentary (actually, pre-rudimentary) control, labelled S-l on the Foreign Service Institute's scale.

In between these two is S-2:

The trainee must learn to handle with confidence, though perhaps not with facility, most social situations including introductions and casual conversations about his work, family and autobiographical information; the same applies to limited business requirements, in that he must be able to handle routine matters even though he may need help with complications; he should be able to understand most conversations on non-technical subjects and have a speaking vocabulary sufficient to express himself simply with some circumlocutions; his accent must be intelligible; his control of the grammar, even though it may not be thorough or confident, should enable him to handle elementary constructions quite accurately.

The above descriptions are intended to be suggestive, but are not recommended for adoption.

Anyone who tries to use them in any real situation will want to be much more specific in some respects, but in many situations they will be inappropriate even as bases for amplification.

Coarse-grained specification comes before all other activities, whether the project is a conventional one of writing materials for distant and future students, or whether a lone Peace Corps Volunteer is getting ready to find his/her own way through a hitherto unwritten language of the African savanna. Lack of such specification leaves the team vulnerable to 'materials-writer's malaise,' the symptoms of which are evident in the quotation on p. 132. As we have said, it is the prospective materials developer who asks the questions: the answers come from outside the language teaching community. This, then, is the first of a series of interfaces.

Fine-grained Specification. Fine-grained specification is the domain -- and the only domain -- over which the outside specialists hold unchallenged hegemony. Given that a trainee will be operating within some general setting, an anthropologist or other cross-cultural specialist is needed to preside over the drawing up of a 'role model' (Wight and Hammons, 1970), which lists the kinds of people with whom the trainee will interact, and also shows how the culture preconditions his relationships with each of them. Given that a trainee will be expected to help others learn to drill wells or raise chickens, or that he will have to arrange for getting his laundry done, someone with authoritative knowledge must provide details of each of these matters. Given that the trainee should have a particular level of competence in a particular socio-cultural setting, the professional linguistic scientist can provide lists of verb tenses, noun cases, stylistic levels, clause types, and grammatical relationships that are indispensable. The items in each list (sociocultural, topical, linguistic) may be marked to show relative frequency, importance and/or difficulty.

Here is the reason for separating coarse from fine specification. To have let the poultry raiser, anthropologist, and linguistic scientist into the picture too soon would have led to disproportionate influence of their theoretical preoccupations and past experiences, and a disastrous loss in validity relative to the interests of the future audience. To allow them to remain in the picture after fine specification has been completed is to invite those same professional preoccupations to distort the teachability of the end product. But to ignore the specialists altogether would be to stumble through the dark toward a distant candle, or to build a house following only a floor plan.

A very rough, but full-scale example of the grammatical part of fine specification for the linguistic aspect of an S-2 on the FSI rating scale is the following, for Brazilian portuguese. This is primarily a list of contrasts that a student must learn to control as he speaks the language. The three main headings in the list are Sentence Patterns, Verbs, Substantives and Other Matters.

Sentence Patterns

Affirmative VS. negative statements.

Statements vs. yes-no questions.

Yes-no questions vs. either-or questions.

Content questions with:

Que?	What? (adj.)
O que?	What? (pronoun)
Who?
Quando?	When?
Quanto?	How much?
Onde?	Where?
Como?	How?
Qual?	Which?
Porque?	Why?
[English equivalents are only approximate.]

Short answers vs. long or yes-no answers.

Exclamations that emphasize:

Noun

Verb

Adjective

Verbs

Contrast between singular and plural, in the same endings that reflect person and tense.

Contrasts between first and third persons.

[In Brazilian Portuguese, special second-person forms are very little used. The third person forms are used instead.]

Contrast among the most indispensable tenses:

Present indicative.

Preterite.
Periphrastic future.
Periphrastic progressive.

Infinitive.

And limited exposure to:

Imperfective.

Present subjunctive.
Past subjunctive.
Future subjunctive.
Personal infinitive.

Future.

The above forms for the three 'regular' conjugations and for:

	ser	be	querer	want
	estar	be	poder	be able
	fazer	do	saber	know
	dar	give	ver	see
	ter	have	vir	come
	ir	go	trazer	bring
	por	put	ha	there is

Substantives and Other Matters

Gender and number in:

Definite articles.

Indefinite articles.
Object pronouns.
Adjectives.
Demonstratives.
Possessive pronouns.

Demonstratives.

Distinctions among three series represented by:

	este	this
	ese	that
	aquele	that

Pronominalization:

Subject pronouns.

Pronoun objects of prepositions.
Pronoun objects of verbs (direct and indirect).

Possessive pronouns.

Prepositions appropriate with assorted verbs and adjectives.

Contractions.

It should be emphasized that the above example of 'fine specification' is to be taken seriously only with regard to its size, and not in its details, even for Portuguese. For other languages, a comparable specification of major grammar points might vary greatly in its content, but probably somewhat less in its length.

The output of the fine specification function may be pictured as a three-dimensional matrix:

To recapitulate:

1. Specification begins outside the area of languageteaching, and relates to it facts from other areas: culture, law, work, requirements of the sending organization.
2. Given a particular set of external conditions, specification is relatively inflexible; that is, it does not depend on the preferences of the materials developer or of the prospective users.
3. Specification takes the form of a set of lists.
4. Linguistic scientists, anthropologists, poultry raisers, and other specialists from outside the area of language teaching are particularly useful in preparing these lists.

Coarse-grained presentation. Here, the data which the writing team elicited from the public, and which were cast by the specialists into the form of detailed lists, must finally be put onto paper and/or film and/or tape. Control has passed into the hands of the language teaching specialist, and he must choose among a wide array of formats, methods and approaches. It is at this point of choice that the present proposal differs from the practice of almost all materials developers. Most writers take it for granted that they are called on to layout for the student some path which he is to follow, and which will lead to the desired goal. The path may consist of conventional lessons or a self-instructional program or a combination of live and canned instruction, and a self-instructional program may be linear, branching, or cyclical. Any fixed set of materials, however, carries within it the seeds of its own rejection: irrelevant content, inappro priate length, or uncongenial format. Furthermore, it fails to tap the enthusiasm that comes when the users of a course feel that something of themselves is invested in its creation. This is one reason why some pedagogical monstrosities have produced good results, and why some well-constructed materials have fallen flat.

One way to go at coarse-grained presentation is the following:

HOW TO WRITE A $5000 (CHEAP!) SARKHANESE COURSE

1. Prepare a sketch of the language. Make it short enough so that an interested layman can read it at one sitting, and clear enough so that he won't get up and leave it. Make it long enough so that a student can relate to it most of the grammatical features that he finds in ordinary written or spoken texts, but don't try to make it exhaustive. Write from the point of view of the student, not of the linguistic scientist. (That is why the sketch is part of 'coarse presentation' rather than 'fine specification'!) If the sketch is well-written, it will also give to the student a convenient bird's-eye view of the language, to which he will be able to refer his own detailed experiences as they accumulate.

This kind of sketch is the subject of Chapter 5 ('Learner's synopses'), and is illustrated in Appendices M (p. 235), N (p.258), and 0 (p. 284).

2. Present within a small, lively and non-committal vocabulary, and in as foolproof a way as you can manage, the main points of grammar. These will be the same points that the linguistic scientist listed in his contribution to 'fine specification,' and that entered into the sketch in step 1 (above). Present them in at least two ways:

a. Identify each point and give simple directions for demonstrating it with minimum dependence on the student's native language or on other knowledge of the target language. This is the 'enactive' mode of presentation (Chapter 3, p. 61 ).

b. Give a brief, clear explanation of the structural item. Make this explanation as independent as possible from the explanations of the other points, or from the sketch that you prepared in step 1, but include cross-references to the sketch, and/or some more comprehensive treatment of the grammar. Use charts and diagrams if you think they will help. These are the 'symbolic' and 'iconic' modes.

Put each point on a separate sheet of paper (or 5"x8" card). This will make it easy for you or others to rearrange them. Appendix K, p. 220, shows grammar points of Swahili that have been treated as suggested above. Some excellent examples for English are found in Harold Palmer's little book on The Teaching of Oral English.

Just the output of this one step, if arranged in some appropriate order, would form a sparse set of lessons. The Swahili materials in Appendix K have been used that way several times. In the teaching of Eskimo, too, S. T. Mallon (1970) reports:

One hundred and twenty three-by-five inch cards were prepared, one to a lesson. On the face of each was written a phrase or sentence illustrating the structure for that lesson. No other formal lesson plan was prepared. (Last year in Ottawa the principal had written out a series of 120 fully detailed lesson plans: on arrival in Rankin Inlet he discarded them as being too restrictive and inflexible.) Instead of written lessons plans the instructor preferred to rely on his own teaching experience, on the material at hand, and on the spontaneous reactions of informants and students. The instructor would enter the class with a preconceived notion of how to conduct the lesson, but prepared to adapt.

3. Present, in as foolproof a way as you can manage, the main question-types and virtually all of the interrogative words, with sample answers. This is part of the 'grammar,' of course, but its main purpose is to enable the student to explore the vocabulary of the language for topics that he is interested in. One way of presenting and exploiting questions is the 'Cummings device,' discussed in Chapter 6 and exemplified in Appendices p(p. 331), Q (P. 337 ), R (P. 346), G (P.154) and elsewhere.

4. Stop. Recognize that the course is incomplete. It is incomplete for two reasons, but also for a third and a fourth:

a. All students will need many more words than you have included so far. (But they will differ as to just what words they do need.)

b. All students will need much more practice with the grammar than you have provided for. (But they will differ as to how often, how long, and how they should practice.)

But also:

c. What is in the course has no connection with anything that really matters to the student. Words are connected to words (either Sarkhanese or English) and patterns are connected to patterns, but there is no feel or motion, no touch, no smell, no flavor and no joy. There are no people yet, only teachers and students. There is no flesh, but only dry bones.

d. You may have taught the student to speak the language a little, but you have not taught him to learn it for himself.

5. Although steps 1-4 will produce only an incomplete course, yet what they do produce will be useful to almost anyone who undertakes to teach or learn Sarkhanese. If there is still money in the budget (and there should be), begin to complete the course for one reason, but also for a second:

a. You can give to teachers and students something to use in their work together.

But also:

b. You can give to teachers and students an example of how they can complete the course for themselves.

As you begin to complete the course, follow your own convictions and the needs of some moment. Aim more at effectiveness than at permanence. It is more important that your lessons should work than that they should last (or sell!). You may decide:

a. that the most foolproof way to present the essential structures (step 2, above) is through a fixed, se1fcontained 'program' which depends as little as possible on teachers, and as much as possible on books, tapes, and visual aids that you yourself will devise. This is is the route taken by Spanish programmatic Course (Appendix D), and by programmed self-instruction in general. Or you may decide:

b. that the most foolproof way to present those same structures is through examples, explanations, and teachersupervised drills. Here are the audiolingual courses (e.g. Appendix A ), and also the pattern-practice courses (e.g. Appendix C). Or you may decide:

c. that the most foolproof way to present the structures is through a series of activities in which talk agrees with action, action agrees with talk, and both go on together. The 'total response' experiments of Asher (1965), the 'Situational Reinforcement' of Eugene Hall, and the 'Silent way' of Gattegno (1970) all emphasize this principle, though in quite different ways. Or you may decide:

d. that there is some other way better than any of these.

You may also decide:

e. that the 'dry bones' of structure should be stacked near the beginning of the course; or

f. that they should be scattered throughout the course as a whole.

6. Before you begin each lesson, list a number of things that the student will be able to do in Sarkhanese at the end of it. These may take any of several forms.

a. things to learn through eliciting further information from instructors or from fellow students;

b. games (including free conversation!) that are fun in themselves;

c. role playing situations that the student can imagine himself being in someday;

d. printed or taped information in which the student is interested, and which is not available to him otherwise;

e. doing things together (e.g. trading postage stamps,gardening, assembling a bicycle) that involve language.

Have these objectives, or 'pay-offs,' in mind as you write the lesson, and put them on the last page when you have finished. Aim at a lesson that the student can finish in 1-4 class hours.

Examples of payoffs are found in Chapter 3; p. 54-57; Appendix R, pp. 361-364, Appendix G, p. 184f and elsewhere.

7. Assemble structure points (step 2) and Cummings devices(step 3) that seem appropriate for the objectives of step 6. Put each on a separate sheet of paper -- not just a separate page. Combine the Cummings devices into an exchange sequence^[1] something like the following:

What is this?

It's a book.

Where is the book?

It's on the table.

Is the book red, or blue?

It's blue.

(This sort of stuff is called a 'dialog' in some textbooks. The difference is that a exchange sequence so emphasizes lightness and transparency (p.47f) that it is credible only in a language classroom.) Put each exchange sequence on a separate sheet of paper.

If you think it expedient to do so, go on and write some genuine, lifelike dialog that incorporates the contents of the exchange sequence but goes beyond it. Either kind of dialog has certain advantages: it provides a change of pace from the very short Cummings devices or the situationa11y disjointed drills (see Newmark and Reibel, 1968, p.149 )i it provides a kind of transition from them to the connected discourse that the students will have to produce as they 'apply' their Sarkhanese at the end of the 1esson; it provides a vehicle for introducing set expressions, sentence connectors, and other items that do not lend themselves to drills or Cummings devices.

8. Write whatever drill materials seem necessary. Put each drill on a separate sheet of paper, double-spaced, with plenty of white space around it. (For examples, see Appendix G, pp.165-182, and Appendix I, pp. 206-214).

9. Leave room for additions and changes, and show that you expect them. That, of course, was the reason for doub1espacing the items in steps 7 and 8, and especially for putting each one on a separate sheet of paper.

If you have followed the instructions in steps 1-9, your lessons will be clusters of available items that support one another, rather than fixed sequences of activity. The users of your lessons will be able to modify or replace any item. Whatever items they choose, they can use in any of several orders: one will want to begin with memorization of a dialog or an exchange sequence; another will want to build up to the dialog or exchange sequence through drills and Cummings devices, and then memorize it; a third will want to eschew dialog memorization altogether. The same class may handle one such 'cluster' in one way, and another in another way.

At this point, you have completed a Sarkhanese course that is a least minimally usable, and that is at the same time maximally adaptable. The options that you have left for users of your course are more numerous and also more obvious than those provided in most language-teaching materials. (For the sake of those who do not want options, you can always arrange your 'clusters' in some linear order and number them serially.)

Although we have written this section of Chapter 4 in terms of a non-existent language, Sarkhanese, it is not merely a programmatic statement. Full-scale materials have been written in this way and classroom tested for Spanish (Teacher Corps),Mauritian Creole and Thai (peace corps). The same system has been tested on a smaller scale for Portuguese, French and Swahili. Appendix J describes how the cluster format was used in the Spanish materials, and Appendix G gives examples for Thai.

Fine-grained Presentation. But the work of the writers is not yet ended. In addition to general procedures, they should suggest a number of superficial variations of technique which will be sufficient either to reduce or increase the pressure, as the need arises. Examples are the change from fixed to random order in calling on students; change of pace; racing against the clock; exchange of roles between student and teacher. The essential difference between these variations of technique and the steps in a procedure is that the latter are relatively fixed, while the timing and ordering of the former depend on clues that come out of the moment-to-moment behavior of a particular class.

Coarse-grained Articulation. What Francis calls the 'articulation function' is easy for writers to overlook or take for granted, yet conscious attention to it can contribute greatly toward teachability. The articulation function consists of two 'routines': a 'criterion routine' by which one decides that it is time to move on to something else, and a 'selection routine' by which one decides what that something else is to be. The writers should make very explicit suggestions for the 'articulation' of each part of each cluster, particularly with regard to the criterion routine. These suggestions might be in some such form as: 'Continue with this drill until the students can complete it in 40 seconds or less, but in no case longer than 7 minutes.' 'Do this role play on at least two different occasions. Be sure that each student has had a chance to take both parts. Do not spend more than 20 minutes on the first occasion, or 15 minutes on he second.'

Fine-grained Articulation. This consists of the decisions that the individual teacher makes as he teaches. It governs the choice of material from the lists of content, social roles, and linguistic features (Fine Specification), and also governs the choice of minor variations of technique (Fine presentation). Among them, these choices determine what actually happens -- in Francis' terms, 'the course.' The course, in this sense, is what the student encounters.

SUMMARY

On p. 135, we pictured the writing of language textbooks as a flow chart, and the process as a linear one. Seen in another way, the same activities we have described in this chapter are concentric: each successive procedure establishes a nucleus around which to fit what may be produced by later procedures. Writers of lessons may provide one layer of inner structure, or many. It would be a mistake, however, for them to assume that they can supply the final outer layer; only the users of the lessons can do that.

The remaining chapters of this book are about devices that we have mentioned as particularly useful in adapting materials or in making them adaptable: learners' synopses, Cummings devices, one kind of sample of language in use, and routine drills.

1. This term arises out of discussions with Carol Flamm, and is approximately equivalent to what Eugene Hall has called a 'response sequence.'