Two Versions of Task-Based Language Teaching

This is a written report of a talk I gave to members of the SLB last month. There’s quite a lot here; it’s a bit “staccato” because it’s based on the slides I used in the talk; and it doesn’t do justice to Mike Long’s version of TBLT. Good, positive start, eh!

Both the syllabuses presented here are alternatives to a coursebook-driven syllabus, so let’s begin with a quick reminder of why alternatives are so sorely needed.

All over the world, coursebooks are the most widely-used tool for organising classroom-based ELT today; indeed, it’s not an exaggeration to say that the global ELT industry today is coursebook-driven. The question of syllabus design is resolved at a stroke by the decision to use Headway or English File, or something similar. Teachers use the coursebook to lead students through a sequence of lessons which are based on the presentation and practice of discrete items of English grammar and lexis. The approach consists of first presenting and explaining bits of the language, thus instilling declarative knowledge (i.e., conscious knowledge about English), and then moving on to practice those bits of language, on the assumption that, through such practice, declarative knowledge  is converted into procedural knowledge (i.e., conscious knowledge that , becomes unconscious knowledge of how to use English in real communication). But the assumption is false; actually, it involves three false assumptions.

False Assumption 1: Declarative knowledge converts to procedural knowledge through classroom practice. It does not. Knowing that the past tense of has is had doesn’t mean that with a bit of classroom practice you can use had fluently and correctly in real-time communication. L2 learning isn’t like studying geography or biology, where declarative knowledge is primary, it’s more like learning to swim. .  .

False Assumption 2: SLA is a process of mastering, one by one, accumulating structural items. It is not. All the items are inextricably inter-related, and there’s no evidence of items being learned one at a time, or of any item being learned linearly. As Long (2015) says: “The assumption that learners can move from zero knowledge to mastery of negation, the present tense, subject- verb agreement, conditionals, relative clauses, or whatever, one at a time, and move on to the next item in the list, is a fantasy”.

False Assumption 3: Learners learn what they’re taught when they’re taught it. They don’t.  L2 learners follow their own developmental route, a series of interlocking linguistic systems called “interlanguages”. The route is not affected by L1, learning context, or teaching method. Teaching affects the rate but not the route of SLA.

I’ve already expounded on all this, (see here for a summary of the arguments) so let’s move to Long’s TBLT.

Mike Long’s Task Based Language Teaching (TBLT) is very different from other forms of TBLT, because it’s based on identifying Target Tasks from which Pedagogical Tasks are derived. All that follows is taken from Long’s (2015) excellent, highly recommended book, SLA and TBLT. 

Unlike other ELT scholars, Long defines a task using the ordinary meaning of the word to refer to “the hundred and one things people do in everyday life, at work, at play, and in between”.

A task is a piece of work undertaken for oneself or for others, freely or for some reward.

Examples of tasks are:

filling out a form, buying a pair of shoes, making an airline reservation, borrowing a library book, taking a driving test, typing a letter, weighing a patient, sorting letters, making a hotel reservation, finding a street destination and helping someone cross a road.

Target Tasks

Target Tasks are what students need to be able to do in the L2. Long insists that the best sources of information on target tasks are “Domain Experts”, or “insiders”: professionals (practicing lawyers, accountants, doctors , bankers, academics, …) tradespeople, Human Resource managers, foremen, and so on. Linked to these domain experts are authentic documents that they use in their respective fields. So you have to talk to the experts and ask them for the documents and texts that they use most.

There is also some useful published literature in our own field, and I would particularly recommend Brown’s (2009) book Foreign and second language needs analysis and the Foreign Language  Program Evaluation Project, published by the University  of Hawaii (2011). see here: .

A third useful source is government publications. Long points out that the U.S. Dictionary of Occupational Titles gives Task-based descriptions of 12,000 occupations, and his book makes special reference to a Soldiers Manual of Common Tasks.   

Examples of Target Tasks

One example of a target task is provided by Gilabert (2005) English for Catalan journalists. Examining the overall task “Journalist interviews a source”, Gilabert identified these tasks:

  • Deciding on the decision-making process
  • Contacting the source
  • Documenting the interview
    • gathering different information sources (previous interviews, documents, Internet items, etc.)
    • selecting materials for questions
    • producing a set of questions
  • Arranging the interview
  • Dong the interview
  • Translating the transcript or interview for publication

(That’s my brief summary of Long’s more complete discussion. In “Documenting the interview”, I’ve added sub-sections, but in fact Gilabert identifies sub-tasks for most of the main tasks.)

Another example of a target task is Negotiate a Police Traffic Stop as described by Long 2015, (ps. 279-286).  This is an interesting and revealing account of a project carried out by CASA de Maryland’s ESL programme in 2009. Recently arrived Latino and African migrant workers in the USA often got stopped by the police while driving to work. O’Connell (2012) set about gathering genuine sources of target discourse and, through cooperation with four police departments, he was given scripts of police stops, recorded various actual stops himself, and inverviewed police officers who made traffic stops. From his data, O’Conell produced a scematic represenation of a traffic stop

From there, O’Connell devised prototypical dialogues, and, building on this work, Long subsequently suggested a module in the course consisting of six Pedadogical Tasks (PTs).

PT1: Intro (Schema Building)

PT2: The Real Thing. Recording of the stop

PT3: What Happened?

PT4: Reading Along

PT5: Role Play 1

PT6: Role Play 2: The Exit Task.

An important element of Long’s TBLT is the transition from Target Tasks to Pedagogical Tasks. Here are the Steps:

(Long, 2015, p.204: Figure 8.1 Steps & Processes in TBLT syllabus design)

Sequencing Pedagogical Tasks

There is then the question of how to sequence the pedagogical tasks. Long discusses criteria for making decisions about sequencing which include

  • Valency: the communicative value, reach or coverage of the task
  • Frequency
  • Learnability: what students are capable of processing
  • Complexity and Difficulty.

These are interesting matters, and I really can’t do justice to them here; I urge you to read Long ‘s discussion of them.

Analyzing Target Discourse

Long devotes a complete chapter to this key step in his TBLT process. It consists of collecting target discourse samples through recordings and written texts, and involves the kind of work O’Connell did with the target task “Negotiate a Police Traffic Stop”. Once the samples have been collected, they have to be analysed. Long (2015, 187-203) gives 5 cases of target discourse analysis. He concludes:

The primary focus of language analysis in TBLT is the dynamic qualities of target discourse – how language is used to accomplish tasks – not simply the linguistic features of static texts. Once target tasks have been identified, samples of language use surrounding their accomplishment are mined for anything likely to facilitate task performance. However, the focus of an Analysis of Discourse is not on achieving an exhaustive, generative model of the kind sought in a true Discourse Analysis, but, again, on dimensions considered important for successful task completion.

Samples of genuine language use are thus another key ingredient. Long says that analyses of such samples reveal many common properties that are rarely modeled in commercially produced pedagogic materials. Language use, even in relatively formal situations, is often more colloquial than that found in coursebooks, usually because, unlike coursebook models, it is context-embedded. Conversations are also typically less self-contained and more open-ended than contrived textbook “dialogs.”

Those characteristics, in turn, allow use of other devices rarely found in pedagogic models, notably, extensive ellipsis and intertextuality. Finally, even fairly simple, constrained tasks, such as service encounters, whose focus is usually thought of first and foremost in terms of referential communication, are facilitated by the use of “small talk”.

Materials: Input Simplification and Elaboration

Long next considers task-based materials, and here, his use of “Input simplification and elaboration” is the key. Having pointed out the problems found in the artifical discourse found in textbooks, and the problems surrounding the use of so-called “authentic texts”, Long argues for a special treatment of the texts selected, involving simplification and elaboration. Again, I can’t myself give any satisfactory quick summary of how Long suggests materials are confected in this way (and, again, I refer you to the clear explanation and examples  found in Chapter 9), but  here’s one example of an elaborated text.

Based on the schematic representation of moves in a police traffic stop, stripping away idiosyncratic moves and topics, incorporating language from the genuine originals, and given differences in the two events, O’Connell produced 2 prototypicial dialogues for a traffic stop;  one resulting in issuance of a warning, the other resulting in issuance of a fine. This is the first dialogue.

Elaborated Dialogue 1

The warning officer: Good evening ma’am. I’m Officer Smith with the Pleasantville Police Department. Can I see your license and registration please?

Driver: Sure, here they are. (Gives officer documents)

Officer: Okay. And is this still your current address?

Driver: Yes, it is.

Officer: Okay. Now, I stopped you because one of your brake lights, your rear left brake light, is out. Were you aware of that?

Driver: No, no I wasn’t.

Officer: Okay. Please stay in the car and I’ll be back with you in a minute. (5–10 minutes pass)

Officer: Okay, ma’am. I’m giving you a warning on the brake light. (Gives driver warning) You need to get that fixed, though, as it’s a violation that you could be cited for. Okay?

Driver: Yes, sir, thank you.

Officer: And here’s your license and registration. (Gives driver documents)

Driver: Thank you.

Officer: Have a safe evening, and be careful pulling out here.

The 2 elaborated texts were recorded and the oral and written texts were part of the materials used in the 6 pedagogical tasks that were mentioned above.


Stages in Long’s TBLT

  1. Needs analysis to identify target tasks
  2. Classify into target task types
  3. Derive Pedagogic tasks
  4. Sequence to FonF task-based syllabus
  5. Implement with appropriate methodology & pedagogy
  6. Task-based, criterion-referenced, performance assessment
  7. Evaluate programme

Long’s TBLT is, in my opinion the best described and best justified syllabus currently available in ELT. But it’s very demanding on the course designer and teacher. The design of Target Tasks and Pedagogical tasks, the analysis of target discourse, and the elaboration of materials that use Long’s ideas of input simplification and elaboration, requires a lot of time and expertise. We’ll return to it at the end, but for now, let’s look at a Process TBLT syllabus.

Process Syllabuses

Breen (1987) summarised the differences between what he refers to as Product and Process syllabuses (Long follows Wilkins (1976) and calls them Synthetic and Analytic)

A Process Syllabus is based on this rationale:

  • Authentic communication between learners involves the genuine need to share meaning and to negotiate about things that actually matter and require action on a learner’s part.
  • Meta-communication and shared decision-making are necessary conditions of language learning in any classroom.
  • The Process Syllabus empowers learners and stresses the vital role of the teacher.

A Process TBLT Syllabus

I tentatively propose a Process TBLT syllabus based on this rationale:

  • Problem-solving tasks generate learner interaction, real communication, the negotiation of meaning, rich comprehensible input and output.
  • There is a focus on the processes of learner participation in discourse.
  • Tasks are sequenced on the basis of addressing learner problems as they arise, thereby overcoming sequencing limitations of conventional syllabus design criteria
  • Learners work on all parts of the sylllabus, including input and materials design

Methodological Principles of Process TBLT syllabus

  • Promote “Learning by Doing” (Real-world tasks)
  • Provide rich input (Realistic target language use)
  • Focus on Form (not FoFs)
  • Provide Negative Feedback (Recasts +)
  • Involve learners in decision-making
  • Respect Interlanguage Development.

Needs Analysis and Tasks 

Needs Analysis consists of:

  • Pre-course questionnaires
  • Interviews
  •  Planning sessions during course


  • Meaning is primary
  • Focus on outcomes
  • A real-world relationship

We distinguish between Macro and Micro Tasks

Macro tasks: Solve a well-defined problem. My suggestion is that a macro task, involving 6 to 15 hours class time forms the framework for micro tasks, and that it is framed as a problem. Examples:  

  1. How can we re-organising the banking sector, post 2008, in Country X so as to avoid a repetition of the 2008 collapse, provide individuals with cheap, efficient, reliable services, etc. etc..?
  2. How do new sophisticated statistics software packages affect house / car / life / …. insurance in Country X?
  3. How can parents deal with teenagers’ use of mobile phones?
  4. The Reinvention of Danone: planning for continued growth.
  5. How can we ensure the continuation of Newspaper X in Country X?
  6. What is the best model for primary & secondary education: Finland, Poland, or UK?
  7. How can the problems of water scarcity in Country X be tackled?
  8. How can traffic problems in City X be tackled?
  9. How can discrimination (race, gender, age, ..) in Industry X or Sector X in Country X be tackled?
  10. How can tourism in Location X be promoted?

Micro tasks: Each Macro Task is broken down into a series of Micro Tasks. Here is a suggested sequence, not definitive.

  • Understand the problem
  • Suggest Tentative Solutions
  • Gather information
  • Present information
  • Analyse and Assess information
  • Test Tentative Solutions
  • Propose Solution
  • Discuss Solution
  • Make decision
  • Report


The teacher takes charge of the first section of the course. At the end of Section 1, there is a Feedback / Planning session. Students fill in 2 short questionnaires and then put together a plan for Section 2. They tell the teacher what topic or topics they want to work on, how they feel about help with grammar, vocab., etc., and how they want to work. The teachers uses their feedback and their plan to design Section 2 of the course. The teacher then leads students through Section 2. At the end of the section, there is a new Feedback / Planning session, the students have learned from the first 2 sections a bit more about how to use the planning session to their best advantage, so they do a better job of planning Section 3, the teacher puts the plan together, and on it goes. A 100 hour course will consist of 8 to 10 sections.

An Example: General Business Course for adults

  • Type of Student: Adult
  • Number of Students: 12
  • Level: Mid-Intermediate (CEFR: B2).
  • Course Duration: 100 hours; 6 hours a week.
  • Main Objectives: Improve English for business purposes. Priority: oral communication.

First 12 hours: Section 1

Before the course starts, students fill in a NA and are interviewed. The teacher designs and leads the first section of the course by leading work on a suitable Macro task, chosen on the basis of data gathered from NA and interviews. To carry out the Micro Tasks, materials from a Materials Bank are used: worksheets, web-based material, videos, oral and written texts, articles, newspaper reports, …

While carrying out the tasks, the teacher attends to grammar through negative feedback and focus on form; attends to vocabulary building and lexical chunks in vocab. sessions, includes some written work in class, sets homework of various types, establishes a website for the class, provides a mixture of group work, pair work, whole class work, and in general gives students a taste of what’s possible.

First Feedback / Planning Session

Tool 1: Feedback Sheet

1= very bad   10 = excellent

General feeling about course so far:                      1 2 3  4  5  6  7  8  9  10

My participation:                                                       1  2  3  4  5  6  7  8  9  10

My progress:                                                               1  2  3  4  5  6  7  8  9  10

Teacher:                                                                       1  2  3  4  5  6  7  8  9  10

Activities:                                                                     1  2  3  4  5  6  7  8  9  10

Use of time:                                                                1  2  3  4  5  6  7  8  9  10

Level of difficulty:                                                      1  2  3  4  5  6  7  8  9  10


Best parts of the classes:


Worst parts of the classes:


Too much / too little time was spent on:


General Comments:


Tool 2: Planning Sheet

What should be the Topic/s for the next Section of the course?


Class Work: What proportion of the time should we work

  • individually,
  • in pairs,
  • in small groups,
  • as a whole group?

Activities   Name activities yu want to do. Be as detailed as possible.  

  • Listening (video / audio / etc.)
  • Reading (texts, internet searches, etc.)
  • Writing (e-mails, reports, etc.)
  • Speaking (discussions, meetings, stories, presentations, etc.)
  • Grammar work
  • Vocabulary work
  • Pronunciation work

What other recommendations do you have?

Prepare a report giving your recommendations for Section 2 of the course.


Students are given the 2 questionnaires which they fill in individually.

They then get into groups of 4 to discuss their answers, reach consensus on the plan for Section 2, and prepare a report to give to the whole class. During this first planning session, it’s important for the teacher to encourage everybody to make specific, focused suggestions. My experience using this type of syllabus is that students will tend to say “We liked Section 1 well enough, you carry on, just let’s have a bit more of this and a bit less of that.” You have to insist that they give more input than this.

The whole class meets to hear reports from each group. The teacher records this meeting. The teacher listens to the feedback comments and makes no attempt to defend himself/herself against any criticism. The groups then report on their plans for Section 2, after which the class discusses the reports together and reaches  final decisions. The teacher promises to present the plan for Section 2 in the next class.

At the next class, the teacher presents the plan for Section 2. Exactly how much this reflects the students’ plan will depend on the context, but in any case it’s the teacher’s job to explain the plan, and to make sure it sufficiently reflects the students’ suggestions. Then Section 2 begins.

In Section 2, the teacher leads students through one or more Macro tasks, chosen in line with the students’ decisions on topic. The Micro Tasks involve activities involving the 4 skills, and are again chosen to reflect the students’ comments on where they want to concentrate. The materials for these activities come from a Materials Bank, and it’s obvious that this Materials Bank plays a very important part in the delivery of this type of course. In the SLB cooperative, we’re currently working on materials that are coded according to topic, skill, level, grammar point, vocabulary area, group / pair / whole class work, etc., etc., so that if members want to do a syllabus of this type, they can quickly assemble the needed Micro Tasks which make up the Macro Task.

 At the end of Section 2, the 2nd Feedback/Planning session is held. Students fill in the same questionnaires and go through the same group and whole class discussion phases. They do it better this time; and they’ll do it better the 3rd time.

Variations in Process TBLT

The basic idea is that the syllabus is divided into sections, and each section does macro problem-solving task(s). Each section preceded by a planning session. The variable elements are:

    • Number of Sections
    • Content of Planning Session (how much students decide)
    • Materials: (from Materials Bank to Coursebook)
    • Nature of tasks
    • Nature of vocab. and grammar work


The Process TBLT syllabus is very sketchy and open to lots more criticism. The NA is open to all the criticsms Long makes of it, and so are the tasks themselves; but on the other hand, students engage in messy work of learning and directly affect decisions. It’s also very adaptable.

Both versions avoid the false assumptions made by product syllabuses, both are learner-centred, and both are likely to be more rewarding for all concerned than coursebook-driven ELT.

This skeleton outline indicates 2 Very different views of TBLT. Long’s TBLT is, in my opinion, much better all round. It’s better thought through, more principled, better informed, more thorough, more focused. But is it practical? I fear too many will think that it isn’t, and fail to appreciate that it makes a lot of sense for all those teachers who have to put any kind of ESP course together or who, working as autonomous teachers, can design courses specially for their clients.

Breen, M. (1987) Contemporary Paradigms in Syllabus Design. Part I   Language Teaching Volume 20, Issue 2

Breen, M. (1987) Contemporary Paradigms in Syllabus Design. Part 2   Language Teaching Volume 20, Issue 3

Breen, M., & Littlejohn, A. (Eds.). (2000). Classroom decision-making negotiation and process syllabus in practice. Cambridge: Cambridge University Press.

Brown, J. (2009) Foreign and second language needs analysis. In Doughty, C. and Long, M.,   Handbook of Language Teaching Wley

Long. M. (2015) SLA and TBLT. Wiley.

University of Hawaii (2011) Foreign Language  Program Evaluation Project.). see here: .

Wilkins. D. (1976) Notional Syllabuses. OUP.

Willis, D. and Willis, J. (2007) Doing TBLT. Oxford University Press

19 thoughts on “Two Versions of Task-Based Language Teaching

  1. Nice to see a TBLT syllabus fleshed out. I do have a few queries.

    1. Can TBLT be reconciled with the demands of courses designed to prepare ss for exams such as FCE, IELTS, TOEFL, etc? Since these exams are major players on the EFL stage, I’d imagine any syllabus that can prepare ss for them would stand a much better chance of establishing itself in the industry.

    2. Long’s TBLT syllabus, at least, looks to be grounded in Vygotsky’s sociocultural theory, which itself gave rise to Concept Based Learning. The difference is, I think, that CBL works from analysis of a concept, rather than towards completion of a task. Further, CBL would concern itself with all the elements involved in the process (from a practical, rather than linguistic POV) rather than task-completion specific ones. Does a TBLT syllabus account for communicative interactions that aren’t obviously task based (such as a purely casual conversation, or watching and understanding a TV soap) and if not, is there anything to be said for assimilating CBL into a TBLT syllabus?


    1. Hi Robert. I’ll butt in here. I would say casual conversation is a task, but others do and have disagreed with me. I suppose I see the task as building rapport or maintaining a relationship. If you can’t keep your end of a casual conversation up, this simply cannot happen.

      As for the standardised tests, I would say you have a synthetic task: pass the test. Sub-tasks are: talk about a picture, discuss a social problem and choose a suitable solution, etc.


    2. Hi Robert,
      Q.1 Exam preparation courses are certainly task-based – even by Long’s exacting specification. The target tasks are plainly the exam components themselves and the Pedagogical tasks are what the teacher thinks the students need to get ready. The trouble is that the exams themselves are often very bad. If students wants to learn English rather than pass an exam, their needs aren’t catered for well in an exam prep. course.
      2. Long recognises the value of scaffolding, and so of course do many who use a PPP approach. CBL is another story. The communication that goes on about and around the tasks (as opposed to doing them)is very important (and I think Long and I would agree that as much as possible of this should be done in the target language). I don’t know enough about CBL (and the little I’ve read didn’t impress me much) to know if it could be assimilating into a TBLT syllabus.


      1. Yes, unfortunately the only meritable literature on CBL I’ve been able to find is an attempt to teach Spanish aspect and a combination of CBL with speech-act theory in ELTJ (2015), which is what I had in mind while reading your post.


  2. Interesting post, Geoff and it was good to see you elaborate on your Breen-derived syllabus.

    I don’t think it needs to be either/or with these as long as one is cherry picking in a principled way.

    Long’s NA is a really good idea but the DA is quite impractical. One problem I have even with an ESP class I teach is that I don’t have access to practitioner-patient discourse. Due to this, I have to use a corpus to help make decisions about what to cover and how to cover it after conducting a NA. I use texts occasionally for this class but often, out of necessity, it’s a Dogme/TBLT hybrid, using learner language as the primary text and a bit of teacher intuition regarding phrasing, pragmatics, semantics and such.

    My FonF work is entirely reactive, although I will take in materials to cater to problematic language the next lesson if a preliminary FonF looks like it is nearly taken on but needs a bit more practice.

    One thing other people might want to check out is Japan Game Lab ( ) and James York’s Kotoba Rollers framework, a game-based TBLT derivation.

    Cheers. Marc


    1. Hi Marc,

      I suppose one can cherry pick from Long’s TBLT “in a principled way”, but it’s important to realise that one major principle of his version of TBLT is to get away from Needs Analysis done by teachers or applied linguists, and to base the syllabus on Target Tasks which are defined by asking domain experts. Another related point is that he rejects the usual definition of a task (which your second comment seems to endorse) as an activity which (a) has an outcome attainable only by the interaction among participants, (b) has some mechanism for structuring and sequencing interaction, and (c) has a focus on meaning exchange.

      As for DA, (or AD – Analysis of Discourse, Long’s less stringent analysis of samples) it’s not always impractical, but if you’re talking about English for doctors, there’s obviously a problem about gathering samples of practitioner-patient discourse, although no doubt it can be overcome.

      I’m sure your Dogme/TBLT hybrid has its merits, and it sounds like a refreshing change from using a coursebook. It has little to do with Long’s TBLT, or with the “Process TBLT” syllabus I proposed, but never mind – it just goes to show that there are LOTS of alternatives!

      Liked by 1 person

      1. Thanks Geoff, though I’m not sure I endorse tasks that *require* participant interaction, and I’m fairly laissez-faire about sequencing interaction. I always felt that I was doing some kind of Long-based TBLT (following the methodological principles) but filling gaps in and around the DA/AD. I’ll have to find a bit of time to look at the big book. Cheers.


  3. A comprehensive overview of the superb presentation you delivered at the SLB event. Thanks Geoff! (We’re still sweeping up scales from the floor – the kind that fall from eyes, that is …)


    1. Thanks Neil. Really, “you had to be there”! The difference between the report and the event itself is SO enormous. The good vibes, the “bien rollo” as we say here, and the delight of being part of an energetic, diverse, inquisitive local group cut like a shining blade thru the gloom of contemporary ELT practice. Everybody: Join a local group!

      Liked by 1 person

  4. Geoff, a few brief points on your piece and some reader comments.

    First, an update. For help with identifying target tasks in the occupational sector, there is now a considerably revised, free, on-line version of the Dictionary of Occupational Titles. Search for O*NET (Occupational Information Network). The asterisk is part of the name. O*NET covers far fewer occupations that the DOT as yet, but is better in some ways, and easy to use. It provides a handy preliminary source for understanding what an occupation involves — the target tasks.

    Having identified the target tasks for a group of learners using O*NET and other sources and methods, the second step in a needs analysis (NA) is the collection and analysis of genuine samples of target discourse, i.e., the language used by native speakers to accomplish those tasks successfully. This is not as onerous as is sometimes assumed, but in any case, is necessary if LT is to become a serious undertaking.

    We presumably want to make sure we are exposing learners to realistic samples of the tasks and language they will have to deal with — something English for Nebulous Purposes (ENP) commercial textbooks are spectacularly bad at, whatever the publishers claim. Doing so has always seemed to me to be no more or less than matching the expectations we have that purveyors of other services (physicians, lawyers, nurses, architects, engineers, etc.) will provide what we need, and not simply dish out the same product to everyone. Physicians, for example, decide on a course of treatment only after identifying what ails us. That typically involves running tests and arriving at a medical diagnosis — roughly, the medical equivalent of a needs analysis. It is encouraging to see that the field as a whole has increasingly been taking responsibility for the quality of its work. For example, an increasing use of NA, and steady improvements in NA methodology, over the past 30 years is documented in Serafini, Lake and Long (2015).

    You do not have to be an advocate of TBLT to believe relevance to students’ needs is important. Writing in TESOL Quarterly over 30 years ago, Auerbach and Burgess (1985) pointed out how damaging it can be for learners to be fobbed off with unrealistic dialogues typical of most commercial textbooks — even textbooks marketed as being designed for the very group of learners in question. All the studies I am aware of since then that have compared genuine language samples with LT materials supposedly designed to prepare learners for the same tasks, but on the basis of the author’s intuitions, have found the textbook models to be highly misleading, sometimes risibly so, even when the target discourse pertains to everyday “public” tasks with which most of us are familiar (or think we are). See, e.g., work on restaurant reservations (Granena, 2008), doctor-patient communication (Cathcart, 1989), and service encounters in coffee shops (Bartlett, 2005). Studies have been published on the language of business meetings, sales transactions in duty-free shops, and on other kinds of service encounters, as well as a vast amount of work on English for academic purposes. Bartlett (2005), incidentally, is another case, like the work of Stephen O’Connell on police traffic stops that you summarized, where the researcher (an MA student at the time) demonstrates how to move from the analysis of genuine samples of target discourse to prototypical models usable in task-based materials.

    A legitimate concern often raised about NA in general, and analysis of target discourse in particular, is the workload involved. The answer is that NAs certainly involve some front-end heavy lifting, but are not rocket science, pay huge dividends, and do not need to be re-done every time a course is offered! They are best conducted at the program level, by specialist needs analysts and/or trained teachers given release time for the purpose, and well in advance of a course’s opening day — by which time it is far too late to find out what students need and select and/or write the required materials.

    NAs are always desirable, but are especially valuable in large, tertiary programs with stable student throughput, e.g., University EFL programs. The work done this year will be usable next because we already know the students will be much the same. They are best carried out by a cadre of trained, experienced teachers, used to working cooperatively and, preferably, in control of their own workplace, whereupon pride in one’s work comes naturally.

    In my experience, NAs are usually interesting and fun, and provide the crucial input to designing the pedagogic materials teachers and students need, but which commercial textbook writers and publishers are loathe to provide. They tend to prefer one-size-fits-all textbooks that can be sold to unsuspecting masses via their “international” lists.

    Auerbach, E. R., & Burgess, D. (1985). The hidden curriculum of survival ESL. TESOL Quarterly 19, 3, 475-495.
    Bartlett, N. D. (2005). A double shot 2% mocha latte, please, with whip: Service encounters in two coffee shops and at a coffee cart. In Long, M. H. (ed.), Second language needs analysis (pp. 305-343). Cambridge: Cambridge University Press.
    Cathcart, R. L. (1989). Authentic English and the survival English curriculum. TESOL Quarterly 23, 1, 105-126.
    Granena, G. (2008). Elaboration and simplification in scripted and genuine telephone service encounters. International Review of Applied Linguistics 46, 2, 137-166.
    Serafini, E. J., Lake, J. B., & Long, M. H. (2015). Methodological improvements in identifying specialized learner needs. English for Specific Purposes 40, 11-26.

    Liked by 4 people

  5. Hi everybody,
    “NAs are always desirable, but are especially valuable in large, tertiary programs with stable student throughput, e.g., University EFL programs.” Like so, as a variation of NA, we have started to gather disciplinary papers students are supposed to read in their professional studies and over time we´ll work out a very specific corpus. This will help us to identify the most critical vocabulary that we plan to integrate in language instruction. For now, students are going to do an online recognition exercise, yes I think I know this word (in a phrase), or nope–never seen, to give us an idea where students are. Still very much experimenting, but we hope that language instruction will become more relevant.


  6. To continue the discussion, Geoff mentioned the idea of creating Macro tasks, and then
    focussing on micro tasks within this.

    I think another possible solution is to create “environments”. That is, immersive situations
    within which all the required tasks can be situated – and which can hold the learners’s interest over a whole course.

    One example here would be the “Widgets” task-based series, but there are other examples. I taught a course on Logistics where groups of learners had to create their own logistics company from scratch and over the course they had to define their product, produce a marketing plan, take part in a trade fair, and conduct business with other groups. Assessment was based on a presentation they gave summarizing their work, evaluating their progress, as well as on the documentation they produced during the course (eg invoices, flyers etc.). It worked well.

    I’m sure there are lots of teachers doing good individual task-based work – the problem is linking these individual efforts together in some way.


    1. Hi Paul,

      Thanks for this; the logistics course sounds good. Neil McMillan and I are working on a SLB project where we try to implement Long’s (2015) TBLT design. It starts with task as the unit of needs analysis – to find out what real world tasks the learners will have to perform. Long notes that “there are ready-made occupational descriptions in many sections such as federal state and local government, business, unions, education, and the military”. It’s important to stress that this is a job for those with work experience, not for teachers who talk to the clients for half an hour. Once you have a good idea of the tasks – all the things that the learners actually have to do – you then design the pedagogical tasks. In this way, the series of tasks that make up the course fit together well.

      As you say, the trick is to link up teachers who have designed courses like this so that ideas, designs and materials can be shared.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s