From drill to discourse – handout & lecture


Wolfgang Butzkamm

From drill to discourse.

Exploiting the combinatorial power of language.

The lecture is an attempt to revitalise pattern practice methodology with concrete examples taken from various classrooms. Observations of how learners acquire languages naturally without pedagogical interventions provide a theoretical foundation. In the first part, I argue that double comprehension (understanding both functions and forms) is the most important single factor in language acquisition. I then present semi-communicative bilingual drills as an exercise type which facilitates pattern recognition, achieves fluency through oral repetition and focuses on meaning rather than on syntactical manipulation. Although the drills work with contextless sentences, these sentences can be processed as fragments of discourse and can lead right into communication, as documented in lesson transcripts. The problem of learning transfer from drill speech to real speech can thus be solved.1

This lecture has two parts. In the first part I shall talk about how we humans learn languages independent of teaching arrangements. These observations provide the theoretical point of departure for my teaching suggestions. First, analysis of natural language acquisition, second: teaching proposals.  Because we can only teach with confidence and clarity, if we understand how learners learn.

In language use and language learning, meaning is all-important, and comprehension is the key to learning. We begin to pick up the language when we identify bits of language and their meanings. Obviously, comprehensible input, usually defined as understandable messages, is the necessary condition for language acquisition. But it is not sufficient. Learners will crack the speech code only if they receive input that is comprehended at two levels. They must understand both what is meant – they must understand the message –  and how things are quite literally expressed, i.e. how the different meaning components are put together to produce the message. This is the principle of double comprehension.

In many cases both types of understanding can be conflated into one process, in others not. Children often get the meaning first before they understand the wording in detail. They initially acquire utterance wholes, fixed formulas (also called routines) which must be carved up until all their constituents and content elements can be used freely.

This learning process has been graphically demonstrated by Lilly Wong-Fillmore (1976), who observed five Mexican immigrant children in their Californian primary school, in their families and on the playground for a school year. Bit by bit the children began to break down their formulas and perceive a pattern with open slots in it allowing their language to become productive.

Lemme-see-it is at first one chunk, where my grandchild Noa would just say “gucken”; another one is I wannit, where Noa simply says  “haben”. Fillmore’s children started to break down these expressions into a fixed part (which is underlined) and a variable part:

Lemme see it the tweedle

I wannit the scissors

The structures eventually became variable in all their slots.  “Ich wieheißtdu Fathma“ is an example from German as a second language.

So children make the passage from formulas or chunks like Lemme-see-it to Let Robert see it; Let him do this etc. ; they begin to understand their internal grammar by extracting the words which they then use to build utterances of their own.

Here are examples of the pattern-finding process from L1 acquisition.

English children make mistakes such as it’s went or it’s played. French: Tu peux me taider? The model for this phrase is probably parental utterances such as Attends, je vais t’aider.  So for my grandchild Astor, who grows up in France, the verb is initially taider, not aider. He has not separated out verb and pronoun.

German:   Wenns du kommst. The force behind this are phrases like wenn’s geht, wenn’s regnet. But what about: Zerlaubst du’s?

This is a real puzzle. How does it come about? Clara & William Stern, who noted this down, suggest that the phrase comes from an incomplete  analysis of Papa hat’s erlaubt / Mama hat’s erlaubt.  I hope you‘ll never forget it. Children have to solve numerous riddles on their way to grammar. So let us keep our sense of wonder alive. Language acquisition is not an easy thing, not  just child’s play, it’s a miracle deeply embedded in our genes,  but a miracle which we slowly begin to understand.

Incidentally, at this point in first language acquisition, parents help their children in various ways. Here are two ways you are all familiar with. At the beginning, parents tend to avoid personal pronouns. „Now Mary has got the ball. Now Mummy has got the ball – instead of saying “you” and “I”. Mary and Mummy are  unambiguous, whereas the pronouns change their referents  and are more difficult to grasp.

And parents ask a lot of didactical questions such as What’s Mummy’s  name? What’s your little sister’s name? Caregivers give the child very finely tuned feedback, they restructure their own language, so that many parental utterances can be seen as mapping aids as well as segmentation aids that separate out, isolate and identify certain meaningful constituents and thus ease their children’s way into language. They make them hear certain words so that they stand out clearly – words which normally run together and blend together in a continuous stream: “Mamakommtgleich”.

All learners, not just children in natural acquisition situations, have problems in sorting out individual words and their distinct meanings, as we can see in the following examples.

A child learning English in kindergarten produced the sentence “I need three napples, please” ( as reported by Peltzer-Karpf, (2003)). He must have thought that “an apple” was actually “a napple”. The same kind of error was made centuries earlier when along with the Spanish fruit the Spanish word “naranja” was imported into England and wrongly understood as “an aranja”, which became “an orange”.  Analysis stopped half-way.

Similarly, classroom learners must break down utterances from their constituent parts in order to be able to recombine them meaningfully.  French beginners are usually taught the phrase “Je m’appelle Christophe”.  What Germans usually understand is “ich heiße Christophe”, which becomes a puzzle when they see it printed. So they should also know that the French actually say “Ich mich nenne Christophe”.  Again, double comprehension is needed. But puzzling this out costs mental resources. So why not clarify it right away by mirroring the phrase in German, as I just did?

Burmeister reports that some children in a bilingual kindergarten thought that “get your cups” meant “Trink was”, which clearly shows that understanding messages, getting the idea, getting the intention,  is only half the battle. So learners need help here, and we’ve seen that parents do help children to understand and tease apart (auftrennen) language patterns, which is just one way of scaffolding (abstützen) their learning processes. (LASS)

Teachers, for their part, can use mother tongue mirroring to scaffold foreign language learning. Mirroring the foreign construction in the native language is a natural strategy.  I remember an Australian boy who told me:

Whereas the French say “le petit poisson bleu”,” der kleine Fisch blaue”.

A Korean student of mine wrote: “A strategy I had chosen to learn a difficult structure was to compare it to Korean and then memorise a very simple sentence for illustration, for instance, what a good boy you are, where English word order is quite different from Korean.”

In my next example a pupil remembers a typical misunderstanding:  “Our teacher often demanded silence with the expression: [pikwait]. To me this was one word and I was absolutely proud when some day I learned the word „quiet“ and discovered its meaning. Although I had sensed what Herr X meant to say I could then correct the pronunciation in my mind because I had identified the isolated words.” Only from then on are sentences like be nice, be good,  be friendly within her reach. Or take the phrase See you tomorrow. German beginners who don’t see the phrase printed automatically assume this means bis morgen, which is literally until tomorrow. With this half analysis they can produce time phrases such as see you later, or see you on Monday etc., but it will prevent them from producing location phrases like see you at the gym, see you at the bus stop. Only a full analysis of the phrase such as provided by mother tongue mirroring  Seh euch morgen will do the job.  Double  comprehension is both necessary and sufficient.

Let’s change perspectives. Think of an English tourist who asks you: How do you say “What’s the time?” in German? You tell him: Just say: Wie spät ist es?  This works well from a communicative point of view. It’s the perfect equivalent, though not the only one. However, it’s good enough for tourists only. Language learners need to know more: *How late is it? That’s what the Germans say literally, which gives us the anatomy of the phrase, and the logic behind it.  That way, the German time phrase could become a recipe for many more sentences:

How old is it? –              Wie alt ist es?

How long is it –              Wie lang ist es?

How expensive is it? –    Wie teuer ist es? etc.

„Tes̜ekkür ederim“, in Turkish, means „thank you“. You’ve understood the message, which – as I’ve just said -  is a necessary condition for acquisition, because you can now use the phrase yourself.  But a formal, analytic understanding will take you much further: *”Thanks I make” or even better: “thanks make-I”. Because „I“, the personal pronoun, is expressed by the ending  -im. This kind of explanation, which I‘ve called mother tongue mirroring, is an elegant, plausible and highly satisfying way of clarifying foreign constructions. And yet it is conspicuously absent in our coursebooks, although it is easily understood and will eventually  help students to build more sentences along the same lines. They can analogise, improvise and risk sentences they’ve never heard before, which is the essence of language learning. That’s the point, the crux of the matter.

A Chinese student of mine (Tong Wu) reports:

“When we were in China I saw that double comprehension can indeed make a difference in a learner’s FL production. In those cases where my German friend only knew what a Chinese utterance meant, he could hardly be creative in terms of making new combinations out of what he just understood. In contrast, when he knew not only what it meant but also how it was constructed, he could easily create new expressions of his own to fit into different situations.”

And that’s precisely what children do. They want to say their own things. They  actually take risks, and sometimes go too far, and so they produce their well-known overextensions or overgeneralisations,

Hocher, vieler, die vielsten

Omnibussen, Anoraken; Onkels, Apfels

German past participles such as *aufgehebt, *ausgezieht, *ausgesteigt

plurals like mouses and foots,

past tense forms like sticked, bringed, putted, hitted.

All these forms which they can’t have retrieved preformed from memory, show that they are well on their way to grammar even if they overshoot in these cases.

Peter, between 2 and 3 years old, also produces his own forms with which he communicates successfully:


  • Das-zu-Dranmachen = Häkchen
  • Das-zu-Schmeißen = Luftballon
  • Das-zu-Bouillon-Reintun = Suppenkelle
  • Das-zu-Eier-Rausnehmen = Schaumlöffel…

Very useful, if you don’t  remember the names of things. – Similarly, my grandchild Olivia who grows up trilingually in France uses a mixed French-German sentence pattern which she can’t have heard before:


veux runter (want down)

veux Haus (want toy house)

veux anziehen (want to get dressed)

veux kuck (want to look)

veux Ilse (= she wants to talk to her aunt on the phone)


Language acquisition is innovative and creative.  It is not the acquisition of a growing repertoire of ready-made phrases or formulas with which tourists try to operate. Children not only imitate –initially this is all they can do – but then they go beyond what they’ve heard. They generate language, and through language, new ideas. This happens all the time, but we can only be sure that they don’t just reproduce what they’ve heard if they produce unconventional and ungrammatical language.

One last point before we come to teaching techniques: Children sort of  practise, or play with,  sentence patterns in non-communicative situations such as pre-sleep monologues. Witness the kind of unsolicited verbal play that Weir (1962, 109) recorded when her son was left alone in the dark before he went to sleep:

What colour

What colour blanket

What colour map

What colour glass

Here is a monologue from my own child: Papa pommt / Mama pommt / Auto pommt…

(pommt = kommt)


And, as I’ve just pointed out, playing with language is playing with ideas, as Natasha shows us who explores the counterfactual:


(Natasha is playing with Natasha; to herself, fast):

(pointing to her nose)                                         this is my foot

(pointing to her eyes)                                         this is my nose

(pointing to her foot)                                          this is my eyes

(pointing to her mouth)                                      this is my neck

(pointing to her bottom)                                     this my head

(pointing to her ankle)                                       this is my wrist…

Let me conclude this part by insisting that only with double comprehension can learners bring the basic and exclusively human property of language into play, its combinatorial power. It’s the core property of language, according to Chomsky, the core capacity which in our teaching methodology is referred to as the generative principle. In Humboldt’s famous words: We can make infinite use of finite means.

Denn sie (= die Sprache) steht ganz eigentlich einem unendlichen und wahrhaft gränzenlosen Gebiete, dem Inbegriff alles Denkbaren gegenüber. Sie muss daher von endlichen Mitteln einen unendlichen Gebrauch machen, und vermag dies durch die Identität der Gedanken- und Spracheerzeugenden Kraft. (Humboldt 1963, 477)


Notice the two aspects of the combinatorial power of language: the inexhaustibility of what is sayable and thinkable.  By manipulating the building blocks of language we produce new thoughts.  Language is not just for communication, it’s for thinking as well as for communication, it’s our thought organ.  This trick, so to speak, of combining and recombining, accounts for the vast expressive power of language. Is grammar the motor of thought? Does grammar make us smart?  Smarter than all the other beings on earth? Question mark!


What follows from this for teaching? What shall we do, as teachers?

The sentences pupils encounter in their basic texts such as dialogues, stories or songs must not remain encapsulated in those texts but must be varied and become productive sentence patterns. For instance, the line “What shall we do with a drunken sailor”, i.e.  the shall I / shall we construction  must not remained enshrined in the well-known sea shanty, but must be made available for other ideas such as:

What shall I do with my hair? What shall I do with my back (it hurts)?

What shall we do with our maths teacher (too much homework!)?

What shall I do with my wife? What shall do with my life?

We have thus opened up completely new dimensions and are miles away from drunken sailors.

In other words: A sentence must become a recipe for many more sentences, a germ cell for numerous other sentences. But watch out! Sentence variations must be experienced as sense variations, not just as syntactical manipulations, as Humboldt reminds us. We may safely assume that children who permutate sentences even in non-communicative situations are interested in the novel ideas which they generate, and not in syntax.

The problem is that pattern recognition, our innate instinct for analogy, comes only into play after a fair amount of concrete linguistic material has been learned (Tomasello 2002, 98).

So as teachers of 3hpw learners who don’t receive the massive language contacts of natural learners we cannot simply rely on the pattern-finding skills of children. This would be a sort of didactical Rousseauism, the naturalistic fallacy: Just make yourself understood and leave them alone with the language. No, we must find the right methodology to accelerate the learning process:

1. We must shorten the process of pattern recognition

2. We must practise a construction so it can take root and learners feel encouraged to risk something new on the analogy of what is familiar.

The solution I propose are semi-communicative bilingual pattern drills. They ought to be a cornerstone in our teaching methodology.  I shall spend the rest of this lecture to show how they work in practice.

How to proceed

1. Select a sentence that can be easily turned into a productive pattern from a basic text that has been studied carefully. It could be song, a story, a dialogue. The sentences are thus anchored in well-understood situations, but must now be freed from their embeddednes in a specific situational and linguistic context.

Here are two dialogues performed by my primary school children who I teach once a week, in the last lesson of the day: English is cool & Black eye sketch.

Wrong World is another such basic dialogue which the children have to act out: The very first sentence contains an important construction: Will you make me a sandwich?

2. Make sure the sentence is doubly understood. Learners should know what it means and how it means.

3. Begin with easy substitutions

  • Machst du mir mal bitte n Brot?
  • Machst du mir mal bitte zwei Brote?
  • Hilfst du mir mal bitte?
  • Hilfst du mal bitte deiner Mutter?…
  • Spielst du mal bitte mit deiner kleinen Schwester?
  • Wollt ihr mal bitte stille sein!

Now make your own sentences.


4. Every bilingual drill is turned over to the students and thus becomes monolingual: Now make your own sentences.  Don’t give too many prompts, but leave space for the learner, so that they can come up quickly with their own sentences: Will you please help your father / big sister/ friend  etc.

5. Use your voice, mimes and gestures to support meaning: „Wollt ihr mal bitte stille sein!“ That’s what makes oral MT cues effective.

Or take another sentence from the same dialogue:

  • Etwas stimmt nicht mit dieser Welt.
  • Etwas stimmt nicht mit meinem Computer.
  • Etwas stimmt nicht mit unserm Lehrer.
  • Etwas stimmt nicht mit ihm.
  • Etwas stimmt nicht mit meiner kleinen Schwester.

Now make your own sentences, English sentences.

6. So here is yet another point: We need to explore the communicative potential or reach of a given construction. So notice the mental leap from computer to teacher, i.e. from things to persons. Incidentally this little leap from things to persons is a big leap for retarded children. It seems that intellectually alert pupils make these semantic leaps by themselves and readily change topics whereas others keep within given domains, for instance animals or food items. But all learners must learn to generalise across various domains of experience. With older learners you could risk a sentence like there’s something wrong with our democracy. Notice the big shifts in terms of content in my next example, which is the line ‘All I want is a room somewhere’ from Eliza’s song in the musical My Fair Lady. Before starting, we make sure the learners understand both the message and the construction. We then elicit sentence variations from our class:

All I want is a nice cup of tea

All I want is a quiet class

All you want is love, but all you get is video.

The teacher’s idea is to practise the formal device of a ‘contact clause’ where the relative pronoun is left out. But in the minds of the pupils, these are variations on the theme of wishes and dreams, rather than a structure drill. The teacher asks himself how he can show his pupils through interesting substitution possibilities that this construction is suitable for their own needs of expression. His job is to probe the communicative radius of a construction, explore its semantic potential.

The exchangeable sentence elements become of greatest importance.

The next example is taken from the dialogue Home sweet home.

This is a wonderful opportunity for students to experience the function and form of present progressive constructions.

T: Ich mache einen Kuchen, bin dabei, einen Kuchen zu backen. In English?

S: bake a cake.

T: I‘m baking a cake.

S: I‘m baking a cake.

I‘m baking a cake.

T: Ich mache grad ne Pizza.

S: I‘m…

T: making

S: I‘m making a pizza.

T: Ich mache grade Sandwiches…

The example shows that repetition and easy substitutions are necessary for the learners to establish the specific sound structure of a construction and to get into the rhythm of it. Mistakes will be made, but will be ironed out, or practised away, through repetition of correct constructions. Part of language learning is habit formation, and this takes several successful repetitions, perhaps up to a dozen.  Repetitions are necessary for new constructions to take roots.

Naturally mother tongue cues don’t always succeed. It can happen that the mother-tongue words rather than the idea expressed functions as the mental trigger. In other words a pupil begins to translate, constructing an English sentence word for word analogous to the German wording. We need experience to effectively deal with or forestall interference errors. Most of the time errors can be prevented through appropriate cueing and sequencing or immediate prompts (schlichtes Vorsagen).

Aber ich mach doch grade meine Hausaufgaben. Use: doing.

  • But I‘m doing my homework.

Und ich spiele grad Geige. Aufpassen…die Geige!

  • And I‘m playing the violin.

The aim is of course for the pupils to operate at the content level and de-verbalise the mother-tongue cue, as conference interpreters do (deverbalisation hypothesis, see Butzkamm 1993, 57). In traditional monolingual pattern drills the pupil solves a formal problem: a word or word group has to be substituted at the right place. Bilingual drills run along different mental tracks. With mother tongue cues the pupil expresses an idea, as we do in normal speech. Where the idea comes from is of course important, so we always switch from teacher’s cues to sentences generated by the students themselves. The drill presents or exemplifies constructions instead of describing them, and simultaneously reveals their communicative range or radius. The rules are caught rather than taught.  No analytical and terminological overkill, as is often the case in the teaching of grammar..

We can help students to get into the rhythm of a construction and maximise language turnover through reciprocal pair work. Here is the dialogue  Head Boy / Head girl adapted from the Peanuts series.

In this short dialogue  the ‘ll construction occurs three times:

 I’ll help you / we’ll vote for you / I’ll do it and so it suggests itself for practice.  The teacher starts the drill as usual but then hands out a worksheet with German-English parallel columns. One partner gets the sheet and acts as the teacher who gives the mother tongue stimulus sentences.

Okay, ich mach’s. – Okay, I’ll do it.

Whenever his partner hesitates, he will prompt him with the correct answer.

Double comprehension again:  Because before the drill starts, the teacher explains: „We say: Ich mach’s. But the English say it differently: Ich werd’s machen.“ So the pupils notice the contrast. With MT cues, we don’t shy away from the contrast, we take the bull by the horns.

Habit formation again: Through repetition + variation the foreign construction becomes less foreign and begins to sound natural. Learners start developing  some kind of Ohrgefühl / Bauchgefühl for this construction. Only recently a Realschul-teacher pointed out to me, that good pupils could be trusted to write these exercises themselves, then be the teacher and practise with the class. A classical example of Learning by teaching (LdL).

With this activity learners can experience a sort of language explosion, because the number of sentences made available to them is rapidly increasing, and chances are that learners could eventually use some of them for personal communications.

I also suspect that bilingual drills are particularly useful if we can bring the FL and the  MT into sharp contrast: How long have you been doing this? Wie lange schon…? Learners must make the equation: the German wie-lange- schon-phrase is associated with the English have-been-doing construction.

The crowning glory of bilingual pattern drills is when the teacher succeeds in giving sentences, i.e. ideas that relate to problems of the day, in other words, when he can personalise, individualise or localise his sentence cues.

Here’s an example (grammar school, 2nd year English) where the teacher alludes to an impending general election in Germany in 2005: Schröder vs. Merkel. The class had been practising  somebody needs somebody or something.

Teacher                                                                      Student

Angie (Merkel) braucht Hilfe.Sie braucht Hilfe von ihren Freunden.Angie braucht Hilfe von den Wählern. Say: voters.Herr Schröder braucht auch Wähler.Sie alle brauchen unsere Stimmen. Say: votes.



Angie needs help.She needs help from her friends.Angie needs help from the voters.Herr Schröder needs voters, too. 

They all need our votes.

Everybody needs somebody to love.


And you could end up with: Jeder braucht einen zum Liebhaben.  Everybody needs somebody to love. It’s a pity I didn’t remember the phrase at the time.

This distinct focus on meaning would be impossible without L1 cues – which shows that the controversy about the use or non-use of the students’ native language is not to be solved with the banal advice to use it “judiciously”.

Results: from drill speech to real speech

Admittedly, a drill series cannot focus on meaning in its full force, because here language is not used in social encounters. So never have pattern drills been meant to replace truly communicative activities.  But can they prepare for them? Yes, they can.

The stage is set for a bit of real communication when the students are asked to make up their own sentences. When they do this, most of them are not performing language operations in a void. Some students may play it safe and give easy examples, but others will feel tempted to vie with the teacher, take risks and produce “loaded“ sentences. Be that as it may, the teacher can always ask a pupil: Is this sentence true for you? Or he can jump in directly, like the teacher in the following example. The class is practising the present continuous, and a pupil comes up with this sentence:

Pupil:     “My sister is doing a test in class 9b.”

Teacher. “Is your sister a pupil of this school?”

Pupil:      “Yes, she is.”

Teacher:   “What test is she sitting?”

Pupil:      “A maths test.”

Teacher:  ”So she is sitting a maths test right now? I hate maths. Do you like maths?” (Silke H.)

The teacher can always step out of grammar practice, he can do as if the pupil meant it seriously and can thus build small communicative islands in a sea of language practice.

Here is an example from a fifth form in their very first year English in a German secondary modern school of the 1980s (Hauptschule). The teacher (Stefan Eschbach) takes the sentence I’ve got a good idea from a previously introduced dialogue and starts with a bilingual pattern drill.


T:         Ich hab’ einen phantastischen Einfall.

P:         I’ve got a fantastic idea.

T:         Wir haben eine wunderbare Idee.

P:         We’ve got a wonderful idea.

T:         Er hat immer gute Ideen.

P:         He’s always got good ideas.

T:      Ehm… Ich hab’ ‘ne grüne Idee.

P:      I’ve got a green idea.

T:      Ich habe ‘ne blaue Idee.

P:      I’ve got a blue idea.


Stefan’s pupils had only a few adjectives available for substitutions. The teacher therefore makes a virtue out of necessity by presenting these comical sentences – just one of the means of shifting the focus from form onto sense. There is sense in nonsense!

Eventually he produces a loaded sentence and breaks out of the drill with a question:

T:      Ich habe einen dummen Lehrer. –  Okay, Jenny.

P:      I’ve got a silly teacher.

T:      Have you got a silly teacher?

P:      Yes.

T:      Ehm… Would you … ehm … be so kind as to tell me his name?

P:      Mr. Morrison.

PP:    Hahaha

T:      You’re laughing, hm?


Naturally the teacher receives an immediate answer which he was in fact expecting: he himself is called Mr. Morrison in his English lessons. This is a very brief communicative exchange, but we all start small, don’t we.

As a final step, the class can be instructed to write their own dialogues in groups, i.e. to change basic dialogues by using the sentence variations just practised. Remember the dialogue Home sweet home?

This is a dialogue written by pupils and based on Home sweet home. It was sent to me by someone who was then a trainee teacher with Marco Hoppe:

Belling? Yes, learners must take risks…

Here is another example of a dialogue written and performed by my pupils in the 1980s. But first the original Peanuts dialogue

  • Lucy: What‘s going on here?
  • Charlie Brown: I‘m helping Snoopy to bury a bone.
  • Lucy: Good grief!

Can‘t he do that himself?

  • Charlie Brown: He hates getting his hands dirty.


The teacher enters the classroom.

  • What‘s going on here?

We‘re playing football.

  •  Sorry, but who‘s playing football?

Peggy, Mary, Betty, Ann and I.

  •  Girls playing football? What are the boys doing?

They‘re playing with dolls.

Good grief!

And after a group presents their play, there is an opportunity for the class to ask questions, to comment on the play and even suggest how to improve it.  The teacher helps to clarify  what was perhaps unclear. This allows for some spontaneous messsage-oriented communication which is, ultimately, what we need. Because, as we all know, we learn how to communicate by communicating. So we are getting the results.

In my language teaching philosophy, however, the generative principle, which targets at the productive  power inherent in language  and puts it to use, is as important as the communicative principle. Sadly, it has been grossly neglected in mainstream thinking.  Teachers can harness these natural skills with semi-communicative pattern drills as I’ve just shown.  The drills proposed are grammar at work, grammar in action.  Yes, walk the walk from drill to communication and self-expression. Here is an example of pattern practice which my pupils made meaningful and enjoyable through intonation, mimes and gestures. I taught these phrases bilingually, but now the pupils no longer need the German: „No parroting“


Carl Sagan said: „When you are in love, you want to tell the world.“  Yes, and when you really have a message, you also want to tell the world. I feel I have a message to put across, because what I’ve just told you is far from the mainstream, and I feel passionately about this.

With regard to MT use in the L classroom, John Caldwell, Guy Cook and others think that „the way is open for a major paradigm shift in language teaching“ (Hall & Cook 2012, p. 299).  So let us do away with a MT taboo which is only self-crippling. Foreign language teaching must be based on a new foundation, the pupils‘ native language.  Millions of average language learners in average schools, and taught, on average, for 3-5 lessons per week, would be just a wee bit better off if teachers knew how to use the right kind of monolingual as well as bilingual techniques. Let me end by quoting from the epilogue of our book The bilingual reform. It is called „Capitalising on a priceless legacy“. This priceless legacy is our mother tongue, and FL teachers must make it their ally. Here are our concluding lines:

„Believe in the power of teaching. Experience the excitement of teaching. Teach with MT support. Teach with the wind beneath your wings.“

1. Sources

Wolfgang Butzkamm,  Lust zum Lehren, Lust zum Lernen (32012), chapter 9: „Richtig üben: das generative Prinzip“ (pp. 237 – 274)     

Wolfgang Butzkamm & John A. W. Caldwell, The bilingual reform. A paradigm shift in foreign language teaching (2009), chapter 6: „How to teach structures the bilingual way“  (pp. 120 – 141)

Wolfgang & Jürgen Butzkamm, Wie Kinder sprechen lernen (32008), chapter: „Das Jahr der Grammatik .“ (pp. 238 – 264)

Why Make Them Crawl If They Can Walk? Teaching with Mother Tongue Support  In: RELC Journal 42/3, (2011), 379-391.



From drill to discourse.

Exploiting the combinatorial power of language.

Wolfgang Butzkamm (

The lecture is an attempt to revitalise pattern practice methodology with concrete examples taken from various classrooms. Observations of how learners acquire languages naturally without pedagogical interventions provide a theoretical foundation. In the first part, I argue that double comprehension (understanding both functions and forms) is the most important single factor in language acquisition. I then present semi-communicative bilingual drills as an exercise type which facilitates pattern recognition, achieves fluency through oral repetition and focuses on meaning rather than on syntactical manipulation. Although the drills work with contextless sentences, these sentences can be processed as fragments of discourse and can lead right into communication, as documented in lesson transcripts. The problem of learning transfer from drill speech to real speech can thus be solved.

Natural language acquisition

  • Children hear complete utterances, not just individual words.
  • They understand messages, i.e. what is meant, and later produce messages.
  • But the input they receive must ultimately be comprehended at two levels: They must also detect / extract the recurring patterns behind the utterances, they must free the words. Why?
  • Because they must learn to recombine the meaning components in new ways  in order to say things they’ve never heard before.
  • They try new sentences & novel ideas and thus make infinite use of finite means.

Foreign language teaching

  • Double comprehension is the most important single factor in language acquisition and foreign language learning.
  • 3hours-per-week learners don’t get the amount and type of exposure necessary for both understanding the messages and finding the patterns all by themselves.
  • But they all have a naturally acquired language – a huge cognitive capital  accumulated over years – which is used to crack the code of foreign constructions.
  • Teachers must work with this capital rather than ignore it.
  • Sentences taken from well-known texts must be perceived as sentence patterns and become a recipe for many more sentences / ideas.
  • Bilingual pattern drills with a dual focus both on fluency and on content can be processed as fragments of discourse and lead right into communication.

Study questions and tasks

Do it yourself! Choose a productive pattern and devise your own bilingual structure drill along the lines laid down in the text. Try it out on your group. Can you adjust your sentences “to the moment”, to the needs and preferences of your fellow students?  Can you spontaneously involve them in short communicative interludes?

Compare bilingual drills with the standard monolingual exercises of schoolbooks. Discuss strengths and weaknesses.

Rivers notes in her learner’s diary (1979, 77): “It is very important to try out in new sentences what you have just learned, if not to others, at least in private talk to yourself”. Do you agree? Have you observed yourself doing this?

“Drill work before free work. This is perhaps the most important of the precepts to be observed” (Palmer 1921, 74). Discuss.

When is explicit grammatical knowledge needed, when not?  When do we need a rule along with grammatical terms?  Or is a mirrored version enough for learners to immediately see the point?  Give examples.

At the beginning the teacher analyzes the construction for the pupils and makes sure learners understand both function and form of the new construction. Should we rather rely on self-discovery?

How can we guard ourselves against persistent errors which echo the students’ own language (interference errors)?

The following objections could be raised by adherents of the communicative approach:1.Producing new sentences on one’s own is still far away from real conversation.

2. One should not practise with isolated, disconnected sentences because they are non-communicative. Discuss.