Pilgrims HomeContentsEditorialMarjor ArticleJokesShort ArticleIdeas from the CorporaLesson OutlinesStudent VoicesPublicationsAn Old ExercisePilgrims Course OutlineReaders LettersPrevious EditionsLindstromberg ColumnTeacher Resource Books Preview

Copyright Information

Would you like to receive publication updates from HLT? You can by joining the free mailing list today.


Humanising Language Teaching
Year 5; Issue 2; March 03

Major Article

Why I won't say good-bye to the Lexical Approach

Hanna Kryszewska, BC, University of Gdansk, Poland.

If this article interest you, Pilgrims offers courses in this area. Click here for more information.

  1. Background
  2. What's a chunk?
  3. Chunking for beginners
  4. Good habits. Learner Training and Teacher Training
  5. Exam preparation
  6. Methodology
  7. Conclusions


I am a non-native speaker of English. English is a foreign language to me and I learned it in the late sixties and seventies with the help of my non-native teachers who believed that if we, their students, learned a lot of new words and mastered the intricacies of English grammar we would speak perfect English. In those days learning English behind the Iron Curtain, with no access to real language we almost believed that our teachers were right. However, contacts with native and non-native speakers who had acquired the language rather than learned it , taught me something new. I often spoke 'in English' but didn't speak English. I could communicate but the language I produced was different from what a native speaker would say. To give an example, on the phone in response to Can I speak to Hania? I would say Hania here, instead of Speaking. On the other hand I could notice that a learner who had been exposed to native speaker English would usually say the right things, even if his/her knowledge of grammar and vocabulary were poorer than mine. Their English seemed more real and right. Later in my adult life I had to learn a lot of chunks or routines and I find I am still learning them. I sometimes wish I had learned my chunks much earlier in life and I feel I have wasted a lot of time. As a teacher I want to help my students to learn about chunks and how to use the right chunks as early as possible. By doing so in a planned and conscious way I can introduce an element of acquisition into the process of learning a language. It is what I would call conscious acquisition. As a language teacher I am very practical and chunks and the Lexical Approach are a way of improving my teaching and not just part of an academic dispute. They make sense in the classroom as they help students perform better in real life as well as in exam situations.

What's a chunk?

Last year when talking to a group of teachers about the Lexical Approach, the first thing I did was to introduce the idea of chunks. I went through Michel Lewis' types of chunks ( The Lexical Approach , LTP, 1993) including the updated versions which I learned about at various conferences. The teachers listened carefully and after some time asked So what is a chunk? I replied A chunk is a chunk. And they said How can you define a chunk? I said It can be big, or it can be small. So they said So what's special about it? I said Well, it's a bit of language. So they said What's a bit? I was beginning to feel like Alice in Wonderland. I sought the help of a dictionary Chunk: a thick solid piece that has been broken off smth., e.g. a chunk of cheese. I read it out. But it got us nowhere. I could see the same question written on the teachers' faces. And then the word cheese inspired me. You have to think about food. A chunk is a big or a small bit, can be different size but it always makes sense and it a whole. Take a piece of meat: a bite makes sense, a chop makes sense, a joint makes sense, etc . These chunks are different in size but they are always a complete whole and make sense to the butcher, the chef or the hungry teenager. I did not get carried away and resisted the temptation to compare language to a butcher's shop. However, I realized that I had managed to prepare the ground for the work on chunks with this group. Practising teachers with their practical minds set on classroom applications need to know why they should know about chunks and why chunks will improve their teaching. So we left behind academic dispute, Nattinger, DeCarico, Willis, Lewis et al. and entered the classroom.

Chunking for beginners

It is often believed that that the concept of chunks is useful with more adult learners at higher levels. You need sophistication and good command of the language. In my teaching experience I have found that this is not necessarily true. One of the first things that my young learners aged 11-12 at elementary level learned was that they can communicate fluently and effectively without using full sentences.
What is wrong with full sentences?, you might ask. Full sentences are a real trap when learners at elementary level try to communicate. They make a lot of mistakes, in fact, they make lots of attempts (mistakes in structures and language the learners haven't been taught yet) rather than errors or slips ( mistakes in structures they have been already taught). We teachers can control the language if it is some form of controlled or semi-controlled practice; we lose control when it comes to production and communication. The danger of making a lot of mistakes (attempts) in the language our learners use to communicate with is that these faulty attempts become fossilized and, later, very hard to eradicate. A learner who for a certain period of time happily and successfully communicated using sentences like Yesterday I go to the cinema., or If the weather will be nice I will go for a walk., will one day find it very hard to switch into: Yesterday I went to the cinema., or If the weather is nice I will go for a walk.,. One of the reasons being that old habits die hard. It will also take the teacher a long time to undo the negative effects the learner's ingenuity and creativity in using the language in free communication. Hence the teachers might avoid genuine communication at lower levels for fear of the learners' mistakes. This can lead to allowing the students to use the target language only in artificial, limited and structured activities and exchanges, which is not very motivating.

Chunks are a perfect solution to the problem of personalization and freer expression with learners at very low levels. I find I can have good conversations with my low level learners of any age when we use chunks. Imagine a conversation with an elementary student about his/her last weekend when Simple Past is a structure that has not been taught yet. Seems impossible, and yet a conversation like the one below is part of my standard practice and many of my lessons start like, say this:

  • Maciek. Your weekend… Tell me.
  • Oh, nice. Very nice.
  • Your grandmother's house?
  • Yes, in Malbork.
  • A long walk?
  • Yes, to the castle.
  • And later?
  • Computer games.
  • How long?
  • 2 hours. My limit. You know.

But what happens next Maciek (11) takes the initiative and the conversation takes a new turn:

  • And you.
  • Oh typical.
  • Much work?
  • Well, some.
  • And then?
  • A film on TV.
  • What title?
  • Don't remember. About the war.
  • No walk?
  • No, too lazy.

When you analyze the conversation you can see that:

  1. communication took place
  2. both speakers, the teacher and the learner, enjoyed equal status
  3. the learner made no crucial grammar mistakes
  4. intonation was very important
  5. the length of chunks was of varying length, from single word chunks to routines
  6. the learner took the initiative
  7. the learner both made statements and asked questions.

What I find particularly hard about these conversations that as a teacher I have to restrain myself and stick to chunks, in other words grade my language down. The learners quickly adjust to the idea of chunks and it helps their self esteem. They can have a perfect complete conversation and the conversation moves swiftly. They see that with the little language they have got they can communicate. A little is a lot, as Gattegno said. However, if by mistake I as a teacher or simply a more competent language user I get carried away and used a full sentence, say, in the Past tense, the learners would want to do the same and start making attempts. Conversations and communication through chunks are also well received by low level adult learners. What they find attractive is that when communicating they do not often make mistakes. An adult learner will often delay engaging in conversations until he/she is sure of an error-free utterance. This may unnecessarily delay authentic communication.

Good habits. Learner Training and Teacher Training

Recognizing and learning the right chunks is part of learner training and helps to form good learning habits. It is part of vocabulary learning techniques or, a term which seems more appropriate, lexis learning techniques. These techniques do not involve pure mnemonics. At their core lies what is sometimes called collocational competence. The students need to develop the skill to select, store and retrieve chunks. Learners need to learn from the very start that words appear in a certain context or environment of words, and any attempt to ignore or rebel against it results in the learners' speaking in English instead of speaking English. In my experience there are four main areas to work on. Firstly, when the students come across a new word, say take they need to write down in their exercise book or remember the whole chunk in which the word take appears, e.g. take a rest. Secondly they need to be aware that the meaning of the word take is different is different in You are tired. I take it and He was taken to court.

Thirdly similar chunks can have different meaning although the from is the same, for example the chunk take it for granted. , as in: I take it for granted and He takes his wife for granted. These two very similar chunks in form differ in meaning and usage. Fourthly similar but not identical forms of chunks are misleading and students find it hard to recognize that small differences in form, lead to differences in meaning and usage. If they fail to do so, it leads to mistakes, for example lose weight versus lose a weight, or make toast versus make a toast. A student must be taught that they need to learn chunks with great precision and not 'approximately'. An additional factor is that students must be aware that what they recognize as a chunk might not be one, for example used to is a misleading chunk. There is either used to + inf or to be used to + gerund.

In a recent test that I gave my new class they were to use chunks in a meaningful context. One of the chunks was bear with smb. The majority of the sentences the students produced used the chunk with the meaning tolerate. They failed to notice the difference between bear smb. and bear with smb. meaning be patient, despite some work on the problem of chunking. They still have a long way to go. These students are proficiency level and they still need to be taught good habits of language learning and chunks. In Teaching Collocation ( ed. Michel Lewis 2000 LTP) Jimmie Hill makes an interesting observation. According to Hill, progressing from level to level in language learning, especially at higher levels, does not involve just learning new words and new grammar. In fact, many teachers agree that at proficiency level (Cambridge Esol exam: Certificate of Proficiency in English) there is hardly any new grammar, most of the structures will have been taught at the advanced level ( Cambridge Esol exam: Certificate in Advanced English). That is reflected in grammar books published for these two levels together, for example, Advanced Language Practice (Michael Vince, Heinemann) contains grammar exercises described as appropriate for both the levels without making any distinction between advanced (CAE) and proficiency (CPE) levels. Basically this book sends out the message that there is no new grammar at proficiency level. But what learners can do is learn more chunks. In Grammar and Vocabulary for Cambridge Advanced and Proficiency ( Side and Wellman, Longman 1999) the authors combine grammar and vocabulary/lexical input and practice. These authors send out a clear message: learning sophisticated vocabulary, like one hundred different verbs describing the ways of walking or looking at something does not make the learner a proficiency level student. Something my teachers used to believe. I couldn't agree more with Hill that progress at higher levels means to a great extent learning 'old' well known words in new chunks and contexts, for example make breakfast, make the bed, make money, make do, kiss and make up, etc. We need to look for materials which do that consistently and consciously.

Students need to recognize and be able isolate a chunk and then be able to use it correctly themselves. Asking students to select chunks from a text they are working on in class is a very good activity for learner training. The teacher, however, must check that the students select the chunks correctly. The best test is ask the students to select chunks from a text, and then have them use thse chunks in new contexts created by the students. It is a test on two levels: recognition and then correct application. Another way to make the students aware of the phenomenon of chunking is an activity I learned from Piotr Steinbrich recently. He suggests that instead of asking the learners to find chunks of language in a text, you ask the students to find words which are not part of a chunk. The result is that the learners see that words do not usually exist in isolation. To be fair, single words feature as the head entries in dictionaries. However, in any respectable, good modern dictionary there are many examples of chunks and sentences which clarify the use of the given word in a chunk and then follows a sentence or a number of sentences that further clarify the usage. It is only in the cheap and old fashioned bilingual dictionaries ( which some students unfortunately still use and what is worse, actually like) that, for example, the word take is translated into a number of verbs without further explanation as to usage. The problem of chunking and looking words up in dictionaries is a very important part of learner training: If you look up the word, look for its meaning in the chunk, its place in a chunk and then learn the whole chunk. When the students see for themselves that words function in all kinds of chunks, they are more convinced about the usefulness of chunking. The benefit is that chunks speed up the learning process, students learn ready phrases and use them correctly, aware of differences in form, meaning and usage.

Phrasal verbs are hard to teach and learn, they are also tricky to use. The reason is that the course books don't teach them in chunks. There is, for example make up and meanings to apply make up, to invent, to reconcile. It is far too abstract. The only way I personally could learn and then use phrasal verbs was remembering whole chunks and contexts: She made up for the party., She didn't want to tell her mother the truth, so she made up a story., They quarrelled and then made up. As usual. Very easy. All you need to do next is use them in similar contexts.

The teacher must be aware of the right ways of selecting the chunks he or she teaches to their students. Recently a teacher was telling me he taught his learners the idiom cat's whiskers meaning be important. This is wasted time and effort. How will the students use the idiom? They may produce sentences like The President is cat's whiskers in the country. , This exam is cat's whiskers for me. Or even I am cat's whiskers. The chunk the teacher gave the students was incomplete hence misleading. The chunk is he/she thinks he/she is the cat's whiskers. Another example is the chunk: If I were you. In fact Jonathan Marks makes interesting observations about the role of intonation and stress pattern in helping to recognize chunks (IATEFL Poland 1997). We, teachers, tend to believe that there is a chunk is If I were you but in fact it is If I were you I'd. This is confirmed by stress and intonation. All these examples prove that we teachers may need to revise our understanding of what a chunk is and in fact learn together with our students.

Exam preparation

So far we have talked about teaching general English classes. Exam classes are special, they have a more concrete focus and require preparing students for dealing with specific exam tasks. The role of chunks in some exam tasks is discussed by Peter Hargreaves in Teaching Collocation (ed. Michael Lewis). In my exam classes when preparing my students for the Cambridge Esol exams (FCE, CAE and CPE), especially for Paper 3 called English in Use or Use of English, I find chunking and collocation are the basis on which some tasks are written and how distracters are generated. Making the students aware of that helps them to perform better and understand why the given answer is correct . I would like to use a few examples all taken from the FCE, CAE and CPE Handbooks (CUP). The FCE multiple close test is lexical but more often than not it tests the student's knowledge of collocations.

The London Tea Trade Centre is a centre of an industry of…….(1) importance in the ……. (2) lives of the British. 1 A high --- B wide --- C great --- D large 2 A common --- B typical --- C everyday --- D usual

The way to choose the correct answer is to recognize the two chunks : of great importance and everyday lives. Students often argue that high or large are correct answers and use their bilingual dictionaries to prove they mean the same as great. So here we teachers face the possibility of students claiming the dictionary is 'wrong'. The learners may not believe it, especially if we are a non-native teacher. A teacher who is a native speaker of the language may get away with: We don't say that. However, any teacher versus a dictionary wins when saying: This is a popular, high frequency chunk., and then can support the statement with an example from a monolingual dictionary based on the Corpus analysis of English language.

Let us consider the error correction task in the English in Use Paper (: FCE Handbook). Find the mistake and spot the word which does not belong there.

  1. One of my father's friends who has a small travel agency, and
  2. She has been very kindly given me a holiday job. It was
  3. difficult in the beginning because I tend to find it very hard to
  4. get up in the morning. My boss is very keen on some punctuality

When we look at the way the mistakes are created we can see these are not random additions. Analyzing the mistakes in lines 41. 42 and 44 we can see that the addition of one word in a chunk creates the mistake. However, the fact that there is a similar chunk may be confusing. This other correct chunk exists but is inappropriate in this context. In line 41 who is incorrect but there is a possible and similar chunk which is correct: One of my father's friends who has a small travel agency is a very nice man., in line 42 the mistake is created by the addition of the word been but there is a possible passive construction She has been very kind to me… and finally in line 44 the mistake is the word some but you can say: My boss is very keen on some sports. The way to spot the mistake/ the extra word is to spot which word does not belong to the chunk in the text, however, it does belong to a very similar chunk which is correct in other linguistic contexts.

In the CAE Paper 3 there is one a task which is difficult both to do and to teach for. It is called Register Transfer. You have to read a text in one register and then complete another text in a different register expressing the same meaning. You are allowed to use maximum two words and you are not allowed to lift from the text. Again its main principle is spotting chunks in informal register and coming up with equivalent chunks meaning something similar in the formal register, or vice versa. So, for example,:

(Informal /Original/ Fragment )
Write to all club members to make them feel at home and to give them the latest details about all the activities coming up soon.

(Formal/ To be completed/Fragment)
Dear Club members,
We have an exciting few months ahead of us. The purpose of this letter is to……… (1) you on our plans for………(2).

The answers are:

1. update
2. the future

When the students do the task they have to read the original text, read the gapped text and find which chunks in the original text they have to paraphrase. It is almost detective work. So in our example they have to spot the underlined chunks:

Write to all club members to make them feel at home and to give them the latest details about all the activities coming up soon., and then 'translate them into

Dear Club members, We have an exciting few months ahead of us. The purpose of this letter is to update you on our plans for the future.

The majority of students have big problems with this task. Perhaps it is because sometimes the students are more familiar with formal chunks and less with informal ones, or vice versa. In order to succeed in this task the candidate must be equally familiar with chunks in both registers. Very often it is to do with the learner's exposure to chunks in these two registers. It is an area to consider when we prepare for this task.

The revised CPE exam brings another task involving chunks called: Gapped Sentences. There are three sentences from which one word has been removed. For all the three sentences it is the same word used as the same part of speech.

Melanie practised her lines each day after school, getting increasingly nervous as the date of her audition ………… even closer.

Although Tim had been in the lead for most of the race, as they reached the final bend, Graham ……… level and threatened to overtake him.

On an impulse, Laura………… all the money out of her bank account and went to London, intending to spend every last penny of it.

The word is: drew, and in order to find it the candidates need to know a number of the uses of the verb to draw: the date drew closer, smb. drew level with smb. and drew money form the bank. The way the chunks are chosen seems to be the principle that there is a chunk students learned at lower levels ( pre-intermediate, intermediate – sentence C), at higher levels (upper intermediate, advanced – sentence A) and at very advanced level (proficiency - sentence B).They are chunks containing the same word but they are chunks of different frequency. One thing is clear, however, and that it is what Hill observed. Reaching proficiency level is meeting old words in new chunks.

I have given examples from Paper 3 in the Cambridge Esol exams. However the role of chunks is involved in testing receptive skills and productive skills. For example, in the Reading paper in some tasks students need to find the correct answer by spotting certain wording in the question and find similar meaning worded differently in the text, for example in multiple matching or multiple choice questions. What is more they need to do it fast because there is a time limit. In the Listening paper students may have to listen for a chunk and write it down or its part on the answer sheet. Productive skills involve a lot of chunking too, however I will not go into further detail in this paper.


My adventure with chunks started in the mid nineties after I heard Michael Lewis give a presentation at IATEFL Poland. In that presentation he introduced the concept of chunking and at the same time admitted that although there is a new linguistic description of the language there does not seem any new methodology or new special activities springing from this description. I took it as a challenge and a few years later together with Paul Davis, I submitted a proposal for a resource book. Publishers were unenthusiastic and the book is still in manuscript form. Penny Ur, one of the people who read the manuscript, voiced a question that I found very important and crucial, a question to which I am beginning to find an answer. The question was ( I quote from memory): Does new language description have to result in new methodology? From what we can see, despite the fact that chunks and corpus analysis of the language have been with us well over a decade there has been no major breakthrough in teaching. The only explanation I can see is that in ELT language teaching we have been strongly influenced by developments in psychology which have in various ways contributed to new trends, approaches or insights into language teaching in various forms, such as Audiolingualism, Cognitive Constructivism, Humanism, Suggestopaedia, Teaching Through Multiple Intelligences, just to name a few. This is different from Continental European pedagogy which is deeply rooted in philosophy. What we experience now is that linguistics provides the data about the language and we teachers expect a successful methodology which implements the findings in the classroom. Teachers do recognize the interesting facts revealed about the language, however, they see no way or hardly any way they can integrate the findings in their own teaching. So perhaps we are looking for the wrong solution.

An alternative is to reconcile the existing models of teaching with the Lexical Approach. In fact this has begun to happen. New editions of course books have an added element which is described in the blurb as includes elements of the Lexical Approach. But this may be just adding few more exercises on collocations and the buzz word to attract more teachers and students, because Lexical Approach is undeniably a 'buzz word' and the flavour of the last 10 years. We may envisage a scenario that all that extensive research and expense will lead to is an occasional exercise on collocations and chunks in a standard language class. And yet despite the lack of success with publishers we can see that teacher training courses on the Lexical Approach and Spoken Grammar (see: Pilgrims Language Courses) are becoming increasingly popular among teachers. At the same time teachers always ask about the place of the Lexical Approach and Spoken Grammar in classrooms and they ask for practical solutions.

In fact, there are two models of teaching which perfectly encompass both the trends. These are: Jim Scrivener's ARC model (Learning Teaching, Heinemann 1994) and Task Based Teaching (A Task Based Approach, Sheila Estaire, Javier Zanon, Macmillan 1994 and A Framework for Task Based Learning, (Jane Willis, Longman 1996,)). In Scrivener's ARC model ( Authentic Use, Restricted Use and Clarification) there is room for corpus analysis, chunks and spoken grammar in the Authentic Use aspect of the model, there is room for student and course book corpus in the Restricted Use aspect and there is time for additional exercises, consolidation and practice in Clarification in the third aspect of the model. As for Task Based Learning there is extensive room for the Lexical Approach and Spoken Grammar in the later stages which involve exposure, analysis and reflection on the unit, ideas for recycling content of unit within future units and ideas for improving effectiveness of learning ( Estaire et al.) and exposure, consciousness raising activities to identify and process specific language features, bringing other useful words, phrases and patterns to students' attention, practice words, phrases and patterns from the analysis activities, enter useful language items in their language notebooks (Willis).

Presentation of the Lexical Approach within the two models mentioned above makes sense to practising teachers. From my teacher training experience I communicate my ideas and classroom experience when I embed the Lexical Approach in this frame. If I present dry academic facts and then do some activities the teachers do not see how and when they can use them in class. I have found that for the teachers I worked with, publications on chunks and grammar like Macmillan English Dictionary – Workbook ( Adrian Underhill, Macmillan 2002) are more accessible and practical than Exploring Grammar in Context (Carter, Hughes and McCarthy, CUP). This is not to say one publication is better than the other. All I want to say is that practising teachers can see that some ways of going about the corpus analysis of the language and chunks fit in better with classroom practice and day to day teaching. After all we need to respond to the teachers' and learners' need needs and help them find ways of using the brilliant publications and data available as eye openers. Perhaps the key lies in recommendations and suggestions on how we use these publications.


The Lexical Approach, Corpus Analysis of the language and Spoken Grammar are here to stay, at least for me. I can see how much my teaching and my students have benefited from them. I feel that I am beginning to find my own day to day answer to the problem. Every day brings a new idea on how to deal with chunks or yet another realization that my students' problems in language learning spring from their lack of recognition or understanding of the concept of chunks. True, if native speakers learn their language through chunks it may not necessarily mean that this is the way a learner needs to learn a foreign language. Besides, how many learners model themselves on native speakers and would like to speak like them? But if we watch the expert closely, an expert who learned most of the things he/she needed to in their childhood in a natural way, we may help our learners perform better and offer new insights and take a few shortcuts in learning and in turn speed up the process. The Lexical Approach may have to struggle a bit more to find its place in language teaching, however, I hope that a voice like mine will help it find its place in language teaching it deserves.

Carter, Hughes and McCarthy, Exploring Grammar in Context, CUP
Estaire,S., Zanon,J.,, A Task Based Approach, Macmillan 1994
Hargreaves, P., Teaching Collocation (ed. Michael Lewis) LTP, 2000
Hill, J., Teaching Collocation ( ed. Michel Lewis) LTP, 2000
Lewis, M., The Lexical Approach , LTP, 1993
Marks, J., presentation at IATEFL Poland Conference 1997, LσdŸ, Poland
Scrivener, J., Learning Teaching, Heinemann, 1994
Side, R., Wellman, G., Grammar and Vocabulary for Cambridge Advanced and Proficiency, Longman, 1999
Steinbrich, P. private communication
UCLES Handbooks for FCE, CAE and CPE exams, CUP
Underhill, A., Macmillan English Dictionary – Workbook, Macmillan 2002
Vince, M., Advanced Language Practice, Heinemann/Macmillan
Willis, J., A Framework for Task Based Learning, Longman, 1996

Back to the top