ESL TeachingTraining

NLP and ELT: any connections?

This is the text of a lecture I gave as part of a series on the history of EFL teaching. It is written to be spoken so the references are not all academically formatted and there may be one or two points which don’t make sense because of the context. I have embedded the slides which accompanied the talk.

2005 TESOL Conference NLP

Neuro Linguistic Programming: Genuine Tool or Gimmick?

    Steve Hirschhorn
    Head of Language Teaching and Training Unit
    St. Mary's University College


I begin with a piece of advice which comes from NLP to those who are sceptical. I offer it without comment.

For the Sceptics?

“You don’t have to believe. If you don’t, pretend to believe”.(Unattributed web ref)

I’m not going to start by taking NLP by the horns, instead I’m going to try to contextualise it first. I am assuming that you are here because you are interested in NLP as a classroom tool, an aid, a support or to add something to your teaching. You are here because you want to do more and you want your students to benefit more from what you do. You may know a lot about NLP already and you may know very little. What you probably do know is that NLP has something to say about the ‘how’ of human interaction and little or nothing to say directly about what to teach or when. NLP was not specifically designed for teachers but for human beings – it will not comment on your syllabus but might comment on the interaction between you and your students and between them. Françoise Cormon (1986) notes:

“language learning is at its best when teachers teach the students, not the syllabus.”

This comment indicates that we would do well to focus on who our students are and how they do what they do rather than the next item in some arbitrarily organised structural, functional or lexical sequence.

So on the basis that NLP might be able to offer teachers support to the human systems which operate in every classroom, it may be worth a look.

Teacher support seems to comprise two distinct types: that which offers practical ‘how to’ advice on teaching pronunciation or creating an activity or a lesson plan and that purporting to offer more psycho-pedagogic or psycho-didactic advice.

NLP falls into the second category. It offers, amongst other things, to enhance your communications – to make them more effective. It offers to put you in touch with your learners by increasing your ability as a communicator and as an observer of human communications.

NLP in Humanistic Education

With such an aim in mind it seems to me that we can plant NLP amongst the considerable foliage in the garden of Humanistic Techniques or Approaches. These may vary in type and style, consider for example a pole with Suggestopedia at one end and Silent Way at the other, but all have certain fundamental precepts in common:

Humanistic Education Means:
  • the centrality of the human spirit,
  • focus on the learning environment,
  • the importance of equality,
  • and the essential nature of communication.

I think NLP purports to interrogate all of these tenets to a greater or lesser extent.

Perhaps you would all agree that it seems natural and ‘right’ to wish to maximise classroom communications and thus enhance the affective environment. Am I right in assuming that most teachers these days understand and promote the idea that learning is more likely to occur if students and teacher are at ease with each other, with the physical environment and with the learning materials? There seems to be general agreement that under those conditions, communication is more likely to be positive and useful. So communication is at the heart of teaching as evidenced by the 70s push towards Communicative Language Learning and Teaching. While I don’t want to conflate the two meanings of communication here: classroom communications which we value as part of an affective environment and communicativeness which is a recognition that language is not a system of rules, there is an undeniable connection.

Communicative Learning and Teaching is still on our agenda and if we combine it with another current (though not new) movement: ‘eclecticism’, we may say that we are duty bound to examine anything which is likely to augment our effectiveness as communicating teachers or facilitators or ameliorate our students’ experience as learners.

NLP holds the promise to assist us in this.

This is especially true if we take account of our minimal influence in the classroom.

Let me explain what I mean: we may think we determine what our learners leave the classroom with, that our communications, our sophisticated communicative activities and our awareness of learning styles combine to ensure that learners go away with what’s on our lesson plan.

All of this seems logical enough and yet Rod Ellis’ response to an interviewer in 1993 should dissuade anyone foolhardy enough to imagine that we are in some way a major determining feature of our language classrooms. Asked what the teacher now knows after almost half a century of Second Language Acquisition research, he replied:

Minimal Influence...

“.. ultimately, what is learned is controlled by the learner and not the teacher, not the text books and not the syllabus.”(Ellis, R. 1993)

This tells us that what we teach is of less importance than to whom and under what conditions learning takes place. It tells us in no uncertain terms that in respect of what is actually taken out of the classroom, we have little impact on our learners.

Ellis offers indirect support for the notion that we should be focussing our energies on creating the best conditions for learning / acquisition to take place. We should be examining the environment and context in which we teach and working to create the optimal conditions. The less influence we have, the more we need to make it count.

Empathy for Students

Showing empathy for our students is something which most of us probably try to do since we intrinsically understand the value of such behaviour. If we don't, I suggest that we can't learn to do it since it is outside our terms of reference. Likewise the unnatural or mechanistic application of NLP tools will result in failure because it will be incongruent and therefore apparent.

If we go to Carl Rogers (arguably the modern father of Humanistic Education) we can find the source of much genuine humanistic thinking.

Essential Attitudes...

“... realness or genuineness.”

“... more likely to be effective..” (Rogers, C. 1983)

"Perhaps the most basic of …. essential attitudes is realness or genuineness. When the facilitator is a real person being what he is, entering into a relationship with the learner without presenting a front or façade, he is much more likely to be effective."


Rogers is noting that falsifying mood, reaction or interaction is ineffective as a device and I would add that it is also offensive. He goes on:

Essential Attitudes...

“... realness is the most important of the attitudes...

...more constructive to be real than to be pseudo-empathic...” (Ibid.)

"Some teachers raise the question, `But what if I am not feeling empathic, do not, at this moment, prize or accept or like my students. What then?' My response is that realness is the most important of the attitudes mentioned. It is not accidental that this attitude was described first. So if one has little understanding of the student's inner world, and a dislike for his students or their behaviour, it is almost certainly more constructive to be real than to be pseudo-empathic, or to put on a façade of caring." (Ibid.113)

Rogers is being absolutely clear that even if we don't like our students, it's better to be genuine about it than create a caring mask which of course is transparent to another human being. I am not suggesting that it is normal or even acceptable to dislike our students but all of us who have spent more than a few weeks in a classroom know that we react with our students just as we do with any other group of people with whom we have long term contact; most are distributed along a continuum from 'extremely fond of' to 'thoroughly dislike'.

Rogers was appalled by often simplistic caricatures of his therapeutic and educational innovations.

'Doing Humanism'

Humanism is about attitude and philosophical belief systems, not techniques. One doesn't 'do' a humanistic lesson or ‘adopt’ a humanistic technique.


“Our appetite for technique has led us to try some of their (the Humanistic Approaches) novel and colourful classroom practices, without serious attention to the values, attitudes and awarenesses that inform those practices.” (Underhill, A. 1989)

Underhill (1989:1) makes the point clearly:

"Our appetite for technique has led us to try some of their (the Humanistic Approaches) novel and colourful classroom practices, without serious attention to the values, attitudes and awarenesses that inform those practices."

Perhaps after all we don’t really need ‘bolt-ons’ since they take no account of context – maybe we should be looking for something more holistic – something which either creates context or is at least aware of context.

Now, having laid out the dangers of picking up atomistic techniques, let’s move on to examine the specific field of NLP, bearing in mind that it too offers techniques which can be dislocated from their context.

To quote NLP Master Trainer Rex Steven Sikes:

"NLP is not something one does to someone else, it is not the techniques, it is not only the mind or the body. NLP, for me, is about how the NLP operator lives his/her life in balance. NLP can be one of the aspects of having a wonderful glorious experience.
Mastering NLP means understanding the concepts behind the techniques, and being able to operate from that level. If you reach that point, the techniques only serve as a safety net: something you can go back to as a resource if everything else fails.”

If we pick out the potentially cringe-making bits of that we are left with a clear enough message:

The Deeper Picture...

NLP is not about techniques. It’s about understanding the underlying concepts of the techniques. (Sikes, R.S. web ref)

Description and Definition

It’s time then to try to define NLP: NLP stands for "Neuro-Linguistic Programming". This must mean that brain function in language use can be in some way ordered or programmed in order to achieve something.

A common definition of NLP is this one:

The study of the structure of subjective experience

which some people complete by adding:

and anything that can be derived from it

Here’s another from Joseph O’Connor a well-known NLP author:

NLP explores how your thoughts (neuro) are affected by words (linguistic) leading to action (programming).

And another from Robert Dilts:

NLP is whatever works

Some might comment that this last one is unhelpful to someone trying to understand NLP.

And another:

NLP claims to help people change by teaching them to program their brains

O’Connor also notes that

“You cannot pin NLP down to a single definition

But I think we need to, after all if we can use words to institute change in people, we can surely find the words to describe the process.

From its inception in the mid 70s, the NLP creators began by observing others and combining their own previous studies to create models of excellence, John Grinder in the field of linguistics and Richard Bandler in psychology. Bandler and Grinder began by studying the highly successful therapeutic techniques of Milton Erikson in Hypnotherapy, Fritz Perles in Gestalt and Virginia Satir in Family Therapy. All seemed to have something in common in the way they worked, and that something was a communicative device which appeared to be in tune with their clients: what became known in NLP as the Primary Representational System or PRS. It was noticed that brilliant therapists seemed to achieve a high level of concord with their patients by, to some degree, reflecting their speech patterns and non-verbal communication.

NLP claims that each of us has a PRS, which means that we process information in specific modes: visual, auditory, kinaesthetic, olfactory or gustatory. According to NLP theory, a person's PRS can be determined by the kinds of words they use or by the direction of their eye movements during processing. Supposedly, a therapist will have a better rapport with a client if they have a matching PRS. Having established ‘where’ one’s interlocutor is, one can respond accordingly, to keep them there or even to shift them somewhere more conducive to change.

Therefore, according to NLP theory, I will have a better chance of passing a message or being understood if I can link my own communications with yours. Perhaps I can use the analogy of dancing (which I don’t actually do): where a couple who move in synchrony are more likely to achieve their aim – whatever that is in dancing!

Of course teachers work with groups, and individuals within a group will work differently. The fact that we are working with more than one person at a time means that it must be difficult to link up with all of them simultaneously. Perhaps then our flexibility in being able to communicate using a variety of channels is important; this is known as Meta-Programme in NLP.

Meta-Programming and Transformational Grammar

Bandler and Grinder proposed that a system of asking questions and gathering information based on Chomskian Transformational Grammar could be combined with information extracted from minimal non-verbal cues to indicate very specific kinds of thought processing.

We’ll look briefly at both of these devices – firstly TG.

The NLP Meta-Model is based on Chomskian Transformation Grammar. TG is now about 40 years old, has undergone many mutations and is far from universally accepted.

Now here’s a tricky bit – I don’t know how many of you are Chomsky experts and I don’t know how many of you know little about his work. I’m veering towards the former but think it might be worth reiterating as briefly as possible since it bears on NLP directly.

This is Chomsky’s TG in a minute flat!

Somehow humans produce language. TG says we do it by assembling phrases from a resource. That resource is (or was called) Deep Structure and is the underlying derivative form of Surface Structure: the utterances we produce.

TG revisited...

Surface Structure (formed language)

Deep Structure (raw data)

Deep Structure is the working material which is turned into well-formed utterances by language speakers based on their semantic, syntactic and phonological competence.

An appropriate example for language teachers might be this pair:

Surface Structure (formed language)

John’s eaten the fish

The fish has been eaten by John



vp - np - Q - det -adj - p - neg

Deep Structure (raw data)

‘John’s eaten the fish.’

‘The fish has been eaten by John.’

Now both of these are well-formed and both, according to Transformational Grammar, come from the same deep structure. But the order of assembly is different and we, as language teachers also consciously note that there is a subtle difference in meaning. Indeed we probably spend time with our learners getting them to handle the passive voice with confidence.

For Deep Structure to emerge as Surface Structure a number of possible transformations can occur. Bearing in mind that Bandler and Grinder were initially developing NLP as a therapeutic tool or set of tools, we must contextualise the Meta-Model in a therapeutic setting.

These transformations then from Deep to Surface Structure depend firstly on what a particular language defines as well-formed but secondly, and importantly for the therapist, they depend on the way that the client can or wishes to represent his/her world. We may have said: ‘John’s eaten something,’ ‘someone’s eaten the fish’, ‘someone’s eaten something’, ‘the fish has been eaten by John’, ‘the fish has been eaten’ and each of these would be a choice made by the speaker and based on the context of the utterance. As would: ‘something’s been eaten’ and even ‘something’s been eaten by someone’. Unlikely, I know but well-formed in the TG sense

Some possible transformations..

  • John’s eaten the fish (basic)
  • John’s eaten something (generalisation)
  • Someone’s eaten the fish (generalisation)
  • Someone’s eaten something (dbl generalisation)

  • The fish has been eaten by John
  • The fish has been eaten (deletion)
  • Something’s been eaten (gen & del)

It’s not hard to see that ‘someone’s eaten the fish’ has generalised the actor – the cause of the fish’s disappearance. Now an NLP operator might decide to examine whether or not that generalisation was significant or not – perhaps someone is being protected; ‘my brother’s eaten the fish and I don’t want him to get into trouble’, or perhaps someone is so threatening that they have been removed from the scene; ‘my brother’s eaten the fish and if I tell you, he’ll beat me again’.

So who’s eaten it?

My brother’s eaten the fish and I don’t want him to get into trouble.

my brother’s eaten the fish and if I tell you, he’ll beat me again.

Or the client may have deleted something on the way from Deep to Surface either consciously or unconsciously – out of choice or due to the absence of choice. One of the precepts of Eriksonian psychotherapy, which undoubtedly influenced NLP, is that our behaviour is always the best choice available to us at a given time. The fewer choices we perceive, the more restricted we are. So a transformation might indicate that the client is restricted in their choice of expression due to some pressing element of the wider context. The next question the therapist might need to ask him or herself would be ‘What is restricting this person’s choices?’

Generalisation and Deletion are two possible transformations and I’ve used them just as an example of how NLP employs the principles of TG to gain insight.

Put very simply indeed: NLP uses TG as a window on the mind.

Non-verbal Cues

Let’s move on to the other part of the window, the non-verbal cues.

These include eye movements, certain gestures, breathing patterns, voice tone changes and even very subtle cues such as pupil dilation and skin colour changes.

Perhaps the most well-known and even controversial of the NLP cues is that of eye movements. This is what NLP proposes:

We have 5 channels at our disposal with which to process information; these are:

  • Visual =V
  • Auditory= A
  • Kinaesthetic = K
  • Gustatory = G and
  • Olfactory =O

I say ‘at our disposal’ but in effect we each have a preferred channel, that’s the Primary Representation System; which you’ll recall is said to be a subliminal expression of how our brains are processing.

According to NLP theory the way we move our eyes while processing can indicate our preferred channel. If we can identify the channel our interlocutor is working in, we can create rapport by joining them.

So let’s run a simple example and see what happens.

I have removed this activity since it had relevance in the context of a live presentation but as a document being read, it does not.

Ok, so that’s not supposed to be an experiment, just a bit of fun to break up my monologue. But it’s not hard to see how many variables we are dealing with here: firstly

And indeed the countless variables might be seen as hampering research into NLP but also into any other teaching and learning system.

So we’ve looked briefly at two of the tools which NLP operators use to gather information but who says that all this true? Have we got any evidence?

Research into NLP

A sceptic might comment that it would be nice to have some empirical support for the claims commonly made by NLP but I have found the comment we started with many times in NLP literature:

“You don’t have to believe. If you don’t, pretend to believe”.

In other words suspend your critical belief systems and see what happens. Personally I can do that with almost anything, especially bank statements but it’s not a course of action which most in the scientific community readily agree to. Although having said that, there are many examples of scientific endeavour which have had to do just that before reaching a conclusion. In terms of evidence or even research we find ourselves in pretty deep water unless we suspend belief.

NLP is not extensively reviewed in our own (ELT) literature:

NLP in ELT Literature

  • ELTJ: 0
  • ETP: 6
  • No mention of NLP in any major works dealing with ELT methodology

Puchta has contributed a chapter to Jane Arnold’s 1999 Affect in Language Learning. Richards and Rogers’ 1986 Approaches and Methods in Language Teaching deals with Suggestopedia, Silent Way, TPR and other approaches but does not mention NLP.

Richard and Renandya 2002 Methodology in Language Teaching, had plenty of space to include NLP but didn’t.

Williams and Burden 1997 Psychology for Language Teachers makes no mention of NLP – though it purports to offer an overview of ‘recent developments’ in the field of educational psychology. I could go on…

NLP is only mentioned in ELTJ in the review sections and bibliographies. In other words no-one has had an article dealing directly with NLP printed for that journal.

Keith Morrow (the Editor) told me:

Since 2002, we have had two submissions dealing with NLP (out of 500+ on the database). Neither of these made it to publication for reasons unconnected with the topic.

ETP fares a little better; since 1996 they have published 6 articles dealing with NLP or closely related topics, one by Judith on the use of Metaphor. I haven’t counted Jim Wingate’s series, since we’ve already mentioned it.

In researching this talk, I reviewed 57 abstracts from academic studies relating to NLP. Broadly speaking, these dealt with:

Of these, the result are as follows:


To my mind this highlights one of the several difficulties with NLP which is the fundamental lack of scientifically accepted supporting evidence. I suppose if all the results of such research agreed that there was no evidence in support of an NLP hypothesis, then life would be simpler. We could say: NLP? It doesn’t work or even “that aspect of NLP doesn’t stand up to scrutiny”. But here we have a situation in which similar kinds of research is undertaken with significantly different results. Some researchers have shown that there’s a link between how someone is processing and the movements of their eyes. Others have shown that there isn’t and yet others have shown that there might be! Clearly, the scientific community doesn’t like this state of affairs and one manifestation of their frustration is, I think, to declare that NLP uses language sloppily or inaccurately. If terms appear to be misused or treated as interchangeable, then we are hampered in our attempts to understand or test hypotheses. Now this may be true from a particular point of view, but one could argue that NLP is (or was) a new field and therefore has the prerogative to adopt and adapt terminology. Unfortunately, much of the terminology it uses to describe its procedures has been borrowed from linguistics or psychology and then tweaked so it means something ever so slightly different.

Terminology shifts..

cognitive stylesrepresentational systems
hemispheric involvementmultiple intelligences

Cognitive styles becomes representational systems and hemispheric involvement becomes multiple intelligences. Sometimes theories and hypotheses such as schema theory and the affective filter hypothesis are popularised to snappy approximations of their original form in sayings like the map becomes the territory or whether you think you can or whether you think you can't, you're probably right. These are not designed for the scientific community, they’re designed with popular science in mind.

I think another objection which arrives from the academic community is that NLP seems rather smug with itself. It doesn’t appear to value the rigour with which scientists generally comport themselves in experimental and research situations. Perhaps this is because NLP from the inside out is more art than science – perhaps it is art which uses some science words. Perhaps it is a hybrid of art and science. Certainly there is no reason in my view why something which can’t be empirically proven should be ignored or vilified. The very fact that some research projects have shown positive results should be enough to encourage an expansion of such projects and to find ways of ensuring ever greater rigour.

Indeed, if we look at our own field, some might be tempted to suggest that our partnership with conventional science, with its empirically proven conclusions, has taken us as far as it can. SLA should have a dialogue with ELT methodology but we certainly haven’t been offered or been able to offer anything coherent for some years. Perhaps the last major flutterings of excitement were created by Michael Lewis’ Lexical Approach 14 years ago, based at least in part on an SLA hypothesis, (Pawley and Syder 1983). Corpus Linguistics, interesting from a theoretical standpoint and now with some practical application, has not revolutionised our work and most of us, I think, probably use a combination of theory, experience and hunch to do our jobs. If I’m right then we need to be receptive to all the available data - to do anything else would be wasteful - and we need to look at it without prejudice but with the professional caution born of previous attempts at the ‘super method’.

Problems with NLP

Stever Robbins, perhaps one of the most coherent writers on NLP (and not to be confused with Tony Robbins) notes that there are 4 major problems with NLP from an experimental point of view. I’ll try to summarise them here:

  1. “NLP takes skill.” Research programmes may not have facility to find the highly skilled people who can reliably conduct their experiments.
  2. “There's no quality control in NLP.” The results of a therapy or an experiment will depend on the quality of the operator.
  3. “It's hard to isolate small chunks of NLP to test.” This goes back in a way to where I started this morning, it’s dangerous to use or to try to test isolated elements of NLP.
  4. “People who appear to have the same symptoms may in fact have different cognitive structures” and so will react differently to the same NLP interventions.

Robbins is pinpointing variables which must by the nature of things remain variable – you can’t guarantee the skill of the operator unless of course you always use the same ones – clearly impractical.

So there seem to be inherent problems with trying to measure something which is in itself so fluid, flexible and variable.

Robbins recounts how he tried to set up a research project with a ‘scientist’, this is what he says:

“Alas, the four or five times I've asked for help designing a study which would let us know if NLP is even worth studying, only one ‘real scientist’ responded.

Our dialog lasted one round of e-mail:

Scientist: We find out if NLP is valid by comparing a group of people treated with NLP technology with people treated with accepted, ‘valid’ treatments.

Stever: Central to NLP is the skill level of the therapist. A lousy NLP therapist will not get results. The model and skill set require a well-defined, but difficult level of competence. [Like any complex skill.]

How do we make sure that the group of people treated with NLP are treated by competent NLPers? After all, since NLP isn't an accepted field, it's been nearly impossible to get a good credentialing system in place, so just going on people's word ‘I'm good’ isn't going to give high quality results.

The same problem also applies to the ‘traditional’ therapists in the control group.

There was no response from ‘scientist’.”

Let’s pop back to the definition of NLP from earlier:

The study of the structure of subjective experience and anything that can be derived from it

It seems to me that one might ask if subjective experience has definable structure and if it does, can it be studied? One of the further criticisms of NLP commonly made by academia is that it is dealing with intangibles which by their nature cannot be measured and therefore cannot be assessed or tested.

So we have something highly variable which itself works with intangibles. It’s not promising territory for scientific investigation.

But perhaps I am being too generous and supportive of NLP.

NLP as a Money Making Scheme

Some say that NLP is little more than a moneymaking scheme. It has certainly attracted vast amounts of money and it hasn’t helped its own image by offering atomistic selling techniques to push more vacuum cleaners. One only has to do a web search to find NLP courses, some at considerable cost, which suggest the promise of a stress-free life, a better career, higher sales figures, more loving relationships and so on.

Here are some examples which I downloaded in about 5 minutes. I’ve stripped out the chat and left titles (in red) and some objectives (in blue).

Basically NLP as an institution promises to sort out pretty much all aspects of your life.

NLP in Management:
NLP in Relationships: deepens the relationship
enabling both people
NLP & Yourself: the most valuable and far-reaching applications
NLP & Therapy: ideal add-on
NLP & Sales: effectively identify and fulfil needs
NLP & your Career: give yourself the edge
NLP & Sport: an excellent performance enhancing set of tools
Influencing Skills influence with integrity
Inspirational Leadership
inspire others to follow
Business Effectiveness Skills
Effective self-management
Managing Continuous Change
managing and responding to external change
NLP Business Communication Skills
sound foundation of mutual understanding and tolerance of difference and diversity
Influencing through Language with NLP challenge beliefs and rules
Empowering others through NLP development of joint visions
invariably achieve more
Peak Performance
higher level of awareness and effectiveness.
NLP Business Practitioner Certification
easier to achieve
emotional resources fully available
Change limiting beliefs
asking key questions
new skills with modelling
improve your performance
Create new options
Develop awareness
body language to establish and maintain good working relationships

Read the minimal cues
Shift viewpoints
The Principles For Change
change the way you experience and interact
help you get your communication across easily
take control
Well Formed Outcomes - mastering your goals
Setting goals Enabling others
The Power of Influence
Rapid rapport building
Utilising body language
Building rapport
Understanding how we represent the world
understanding another person's thinking
Taking control of your brain
use your brain more powerfully
how to change limiting beliefs
change the way you feel
The Power Of Language
use language masterfully
gather vital information
transform, influence and negotiate
the key to sales mastery
Creating powerful associations - Anchoring
Create resourceful anchors
Learn about strategies - the thing that creates our behaviour
Elicit and utilise
buying strategy
change the less useful ones
Congruence & wholeness
resolve internal conflict
Feel totally aligned

Questionable Ethics?

And then a couple of even more worrying links I found on the web:

Ethical Questions

““Find out the sneaky way to use NLP to ‘fly beneath the radar’ of your listeners' perception””

““How to use NLP covertly””

So the commercialisation of NLP is commonly cited as revealing a scam. And there is an ethical question too which is a bigger debate than I can approach here. But let us just note that a set of principles or ideas can be dislocated from their ethical origins by unscrupulous operators in almost any field of endeavour.

More Objections

Some of the objections we have examined are characterised by responses I got after sending out an initial email probe to gauge the interest in a day on NLP.

I received the following from a well-known author who writes on the psychology of language and psycholinguistics:

“Afraid I'm on the side of the academics and/or mainstream educators. In psycholinguistic terms, I can testify that it is based on very shaky foundations: enormous generalisations about (e.g.) left brain / right brain and no serious attempt to take on board very basic findings in (e.g.) studies of attention or working memory. I've also read just enough to come across some very specious linguistic analysis using grandsounding (and sometimes inaccurate) terminology to impress non-linguists. Frankly, I think it's the kind of thing that gets ELT a reputation for soggy thinking.

I'd suggest that it probably already gets more publicity than it deserves, and that a respectable institution like yours should be very wary of giving it more.

Sorry to sound so negative - but I've always believed in informed thinking rather than hunches. I'm afraid a day on this topic would attract me rather less than an evangelical prayer meeting with George W Bush.”

I wrote back suggesting that he get off the fence!

A couple of other academics with whom I am in touch and I know to have feelings about NLP, wouldn’t commit themselves to a comment either.

Here’s another response, this time from an ELT professional (both reproduced with permission by the way):

“I guess it's just personal teaching styles - my approach is very much emphasis on vocab, no frills (CD-Roms, endless photocopies, etc), get students doing/discussing interesting things etc, so my influences are Michael Lewis, Krashen, Dave & Jane Willis, Scott Thornbury. In my experience it's pretty hard getting anyone (teachers or students) to move away from a grammar-based syllabus, so I think getting them to accept an even more radical approach is a bit of a non-starter.

…… I'm always pushing methods that I believe in and that have worked for me in class, and in that respect NLP is not something I've used or would encourage (every journal seems to have been filled with articles on NLP recently, so I have read a bit about it), and I would guess that for many others in the profession discussing these kinds of ideas on the fringe of the mainstream is a bit of an irrelevance. Of course, if it works for Jane Revell et al., then great.”

A bit of an irrelevance? What is the mainstream these days? I don’t think we can talk about the mainstream any more and haven’t been able to since the passing of pure Direct Method. The Communicative Approach hasn’t manifested as a definable mainstream methodology, it’s many things to many people. My guess is that we could cite any commonly known ELT technique right here and now and find that there exists a divergence of opinion amongst us, and rightly so. It’s true that there is general agreement on principles or a dominant discourse in western ELT but if we compare the current situation with the height of Audio Lingualism or Direct Method, we are fragmented in thought and deed.

Training Course Exam Boards, for example, Cambridge and Trinity must clearly recognise that there is no current consensus on mainstream methodology since they both cling to the raft of tradition and standardisation in their judgement of what constitutes a good lesson and a good teacher. I recognise that new teachers need a solid base (although that could be interpreted in a number of ways) but at the Diploma level surely the exam boards should be leading rather than trailing.



  • NLP is controversial.
  • It varies in form from the serious to the flippant.
  • It promises much but can’t deliver consistently under experimental conditions.
  • It is sidelined by many in the academic community.
  • NLP is generally unwilling or unable to subject itself to serious external examination.

On the other hand...

  • Just because something is there doesn’t mean we have to prove it’s there.
  • In ELT we already work largely without empirical evidence.
  • It promises much but can’t deliver consistently under experimental conditions.
  • We need to know about it in order to decide whether it’s for us.
  • Operated ‘correctly’ NLP provides its own context.

So to summarise, NLP is based on some pretty old science but then so is the internal combustion engine. TG has now been superseded and The Structure of Magic was published 30 years ago but these are not reasonable grounds for dismissing NLP. In the final analysis, like any other techniques, approach or methodology, NLP will work or not for an individual teacher with an individual group because it is right for them and not because it is scientifically proven or not.

There is little else which has aroused so much attention, both positive and negative, in the field of psychology and psychotherapy over the intervening years. There is little else which has consistently captured popular interest over that period – Eric Berne’s Transactional Analysis almost came close and was used as a basis for Harris’ Games People Play. But that’s about as far as TA went in terms of general interest. There are of course plenty of groups which study and practice TA but those can’t be compared to the astonishing success and ubiquity of NLP.

I can’t do better than to leave you with a comment from a personal communication with Stever Robbins:

“NLP is a tough nut to crack. I've been involved in it for 22 years now, and am 100% convinced that it's extraordinarily powerful compared to most other forms of therapy. That said, it's nowhere near the miracle cure that many claim it to be.”


Aspy, D. & Roebuck,F.Kids don't learn from people they don't like1977 Human Resources Development Press
Cormon, F.Humanistic Activities and Teacher Motivation1986 ELTJ 40/4 OUP
Ellis, R.Talking Shop1993ELTJ 47/1 OUP
Naiman, N. et. al.The Good Language Learner1978 Ontario Institute for Language Education
Rogers, C.Empathic: an unappreciated way of being. In A Way of being1980 Houghton Mifflin
Rogers, C.Freedom to Learn for the 80s1983 Macmillan/Merrill
Rubin, JWhat the good language learner can teach us1975 TESOL Quarterly
Spolsky, B.Conditions for Second Language Learning1989 OUP
Underhill, AProcess in Humanistic Education1989 ELTJ 43/4 OUP


Feel free to contact me about this article by email:

The Case Against a ‘Clinical Supervision’ Model in Language Teaching