The Philosophy that Never was

All quotes from Witkowski can be found in the online version of his article accessed here:
on December 10, 2010



If you go on doing what you're doing now,
You'll very likely go on getting what you're getting now
NLP-related presupposition

Visitors to this page may already be aware of a certain Donald Clark and his rants against "NLP".  It appears, from his various posts, blog entries, etc., that Mr Clark knows very little about NLP - and most of what he does know is quite simply wrong.
Perhaps not surprisingly, then, Mr Clark tends to give little or no evidence to support his claims beyond a short list of references to a few academic papers which he *presumably* has not read (since he unquestioningly recycles criticisms which are as inaccurate as his own comments).
In nearly every case the material referenced by Mr Clark has already been discussed on this website,such as those by Sharpley, Heap, Singer, etc.(see FAQ #28).

In December, 2010) Mr Clark announced on the TrainingZone forum that he had added another "witness" to his team, a Polish gentleman named Tomasz Witkowski.  I am still trying to discover what qualifications - formal or informal - Mr Witkowski has for evaluating any part of the field of NLP (FoNLP).  Despite a very polite enquiry on my part, on December 13, 2010, by March 2012 Mr Witkowski had not yet provided that information.

In fact Witkowski's article, subsequently published in the Polish Psychologicsal Bulletin (PPB) mirrored the shortcomings of many of the academic criticisms of NLP and the FoNLP.  It is somewhat remeniscent of a story about Orville and Wilbur Wright (creators of the first successful airplane):

It seems that for several days - despite the fact that travellers on the railway line close to the test field at Kittihawke had been able to watch the brothers successfully flying their plane (named "The Flyer") on several occasions - the local newspaper refused to feature this momentous story, or even send a reporter out to investigate, because reputable scientists had allegedly, albeit theoretically, proved that powered flight wasn't possible.

Stop Press

This article was written in the first week of January, 2011, after which I notified Mr Witkowski by e-mail (January 7, 2011) that the article existed - inviting him to visit the "hidden" version in order to let me know if there was anything that was inaccurate or unfair which he thought I should change before moving the article to the public area of my website.
To date (March 19, 2012) Mr Witkowski has still not replied to that invitation.
Some three months later Mr Witkowski issued what he calls "a challenge" to the NLP community on his website:

"Tomasz Witkowski challenges to a duel NLP-passengers from all over the world!"
(Translation by Google)

The " challenge" was not quite what it seemed, however.  That is to say, Mr Witkowski imposed a limitation on responses to his "challenge" - that responses must take the form of articles published in academic journals.  If Mr Witkowski was indeed looking for an honest and open discussion it seems to me we already have the basic requirements.  He has presented his article online, and I have posted a reply online.
If Mr Witkowski wishes to avoid the negative aspects of an online discussion group - like the one he has set up on his own website - any further comments could be restricted to those two venues.

I had hoped to submit a response to Witkowski for inclusion in the PPB, but various circumstances have prevented this, so for the meantime this web page, which has already been visited by hundreds of people from countries as far apart as Argentina and Australia, the USA and Italy, Britain and the Czech Republic, must terefore remain my one contribution to the "discussion".

*** The Short Version ***

(academic role at the time of publication)
Tomasz Witkowski:   Whilst his web site indicates that Mr Witkowski has a PhD (subject unknown), as indicated above, as of March 2012, 15 months after my first inquiry) Mr Witkowski is apparently unwilling to provide the basic details of his academic qualification(s) and role(s) at the time of publishing his article online or his qualifications for critiquing any aspect of NLP and or the FoNLP (field of NLP).

Critical Material:
Thirty-Five Years of Research on Neuro-Linguistic Programming.  NLP Research Database State of the Art or Pseudoscientific Decoration?, Tomasz Witkowski.  Published online at

Nature of criticisms:
The essence of Witkowski's entire article is summed up in just one quote on the final page of the main text:

The analysis of the NLP Research Data Base (state of the art) by all measures was like peeling an onion.  To reach its core, first I had to remove some useless layers, and once I arrived, I was close to tears.  Today, after 35 years of research devoted to the concept, NLP reminds one more of an unstable house built on the sand rather than an edifice founded on the empirical rock.
(Witkowski, 2010. p.65.  Emphasis added)

This quote is crucial to an understanding of Witkowski's entire article fails to serve as a coherent comment on any part of the FoNLP, see the section on Flaws, below.

Highly derivative.
One could create most of the article using nothing more than the reviews by Sharpley (1984, 1987) and Heap (1988).


The basic flaws in this article are:

Firstly, Witkowski clearly has no accurate idea what NLP or the FoNLP are about.
Though the phrases "NLP concept" and "the concept" occur some 22 times in the article, at no point does Witkowski explain what this "concept" is.  On the contrary, he makes a substantial number of inaccurate claims about "NLP" in general.  Despite his own comments to the contrary, he seems convinced that "NLP" is a form of therapy.

Secondly, as the quotation in the Criticism(s) section shows, the whole article is nothing more than a "sleight of pen", so to speak.  That is to say, Witkowski goes on at some length about a database of abstracts on the "NLP Community" website.  The abstracts are from articles related to the FoNLP (field of NLP) in some way, i.e. not necessarily directly.

Witkowski's fascination with this database seems to be based on his belief that by discrediting the database itself he will discredit "NLP".  Thus in the title and the body of the article he talks about the "NLP Research Data Base [sic].  State of the Art or Pseudoscientific Decoration?" and "All of this leaves me with an overwhelming impression that the analyzed base [sic] of scientific articles is treated just as theater decoration, being the background for the pseudoscientific farce, which NLP appears to be" (page 64).

But the name "NLP Community" does not signify that the database is part of some official NLP website.  Nor is the claim by various researchers that their work is scientific have anything to do with the question of whether NLP and/or the FoNLP are scientific.  This extrapolation is a figment of Mr Witkowski's imagination..

Like most academically-oriented critics of the FoNLP, the author of this report appears to have done no genuine research - that is, first hand experimentation - and instead has merely delivered yet another warmed-over rehash of previous, unexamined and unsound criticisms.

Even if Witkowski had analysed the contents of the NLP Community database effectively (which he hasn't) his findings would not reflect upon the validity of the various concepts, models and techniques included in the FoNLP.
Thus, far from providing an effective overview of the FoNLP, as Witkowski apparently believes, the article simply leads us on a "wild goose chase" to nowhere of any significance.

This example is particularly ineffectual in that the author is trying to present the contents of an independent database as proof that some (undefined) "NLP concept" is invalid.

*** End of Short Version ***

*** 'Director's Cut' ***

Mouth - Foot, Foot - Mouth, Aaaaagggghhhh!

The errors in this article - and there are many errors in this article - start in the first sentence of the abstract:

Huge popularity [sic] of Neuro-Linguistic Programming (NLP) therapies ....
(Witkowski, 2010. page 2)

In practice, NLP itself is a modelling technique, and nothing else.  Yet in the entire text (i.e. excluding the References section), the word "model" is mentioned only twice; and the word "modelling" (or "modeling") is not mentioned at all.
Instead we find, as is so often the case, the totally inaccurate assertion - again after Sharpley and Heap - that "Neuro-Linguistic Programming (NLP)" is a form of therapy - or maybe several therapies?

To claim that one is able to evaluate "NLP" when one doesn't actually know what it is, is quite a stretch.  To claim that one can accurately evaluate "NLP" whilst also claiming to know that it is something which it is not, pushes the credibility of the article right over the edge.

Just the One (Mrs Wembley)?

Witkowski's approach to the task he has set himself is fairly typical of critics of "NLP", a kind of literary fusillade in which he rapidly fires off claim after claim, with no discernible concern for their accuracy, possibly in the hope that none of his readers will notice the very poor marksmanship.
The main text of the article starts on page 3, where the author tries to justify his approach with this statement:

... despite that widespread presence of NLP [sic], none of the psychology textbooks I have heard of (Polish or English) presents an in-depth discussion of the concept [sic].  What's more, scientific authorities refrain from giving their opinions in this respect.  What is the scientific worth of the concept?  Only a thorough analysis of empirical research can lead to an answer to that question.
(Page 3)

As though to confirm our initial impression of the article, the author mistakenly describes "NLP" as a single concept, though he is certainly not referring to modelling.  But wait.  Despite initial appearances, as we go through this article we find that the article repeatedly confuses concepts, tenets, theories, assertions and so on.

Outline of the NLP Concept

(Witkowski's subheading)

In the last few lines of the same page we find the claim that:

[Bandler and Grinder] established that outstanding psychotherapists acted on the basis of implicit theories ...
(Page 3)

Absolutely not.  Bandler, Grinder, and Frank Pucelik (the co-creator Witkowski ignores, or doesn't know about), were categorically not interested in any theories Satir, Erickson, Perls or Farrelly might have had about what they were doing.  They were interested in the pure behaviour, without any theoretical overlay.  This is the defining factor that distinguishes non-analytical NLP modelling from any other style of modelling.  To miss this fact - as most critics of NLP do - is to work without knowing the very foundation of the entire field of NLP.

Next, Witkowski presents us with yet another absolutely basic misinterpretation of what underpin's whatever it is he thinks of as "NLP":

The central philosophy of NLP is summed up in the sentence “The Map is not the Territory” (see, e.g. Lankton, 1980, p. 7).
(Page 4)

In my copy of Mr Lankton's book page 7 is blank, but perhaps Mr Witkowski is referring to a foreign language translation.  So let's be generous and suppose that Mr Lankton did say this at some time or other.  Unfortunately this gets us nowhere.  Firstly because neither authentic NLP nor the FoNLP have any "central philosophy".  And secondly, because even if there was a "central philosophy of NLP" it would be inaccurate to treat Mr Lankton as an authoritative spokesman on the subject.
In fact, of course, "The map is not the territory" is just one of the "presuppositions" of the FoNLP.  That is to say, it is an epigram which is not necessarily true, though in some circumstances it can be of benefit to treat it "as if" it were true.
It is a "useful" idea - borrowed from Alfred Korzybski's General Semantics - that's all.

The Fixed PRS Error

And the author further undermines the credibility of the article with this next claim:

As [Bandler and Grinder] suggested, each of us processes the majority of information [about the world] using one primary representational system (PRS).
(Page 4)

In fact what Bandler and Grinder said was:

Our claim is that you are using all systems all the time.   In a particular context you will be aware of one system more than another.
(Bandler and Grinder, 1979. Page 36)

It is the sensory/representational system that a person is currently paying most attention to which is their PRS, and far from being static it will be dictated by a person's perceptions about the current context.  Again the article misses out this vital piece of the information, thus making it almost impossible to provide an accurate understanding of the PRS concept.

The Eye Accessing Cue Error

After misreporting the nature of PRSs, Witkowski continues to follow Sharpley's lead with a completely inaccurate description of the role of "eye accessing cues":

Another discovery of which the NLP originators were particularly proud of was to realize [sic] that access to the representational systems is possible through the so-called accessing cues that is precisely specified eye movements.  Careful observation of these movements should enable the NLP therapist to unequivocally identify the PRS of the patient, interlocutor etc., and, in consequence, facilitate matching their PRS.
(Page 4)

The first sentence is actually correct, but the second is entirely wrong.  The error, again drawn from Sharpley, directly contradicts what Bandler and Grinder actually said:

In order to identify which of the representational systems is the client's most highly valued one [i.e. their preferred representational system], the therapist needs only to pay attention to the predicates which the client uses to describe his experience.
(Bandler and Grinder, 1976.  Page 9)


The "representational system" is what's in consciousness, indicated by predicates.
(Bandler and Grinder, 1979.  Page 28)

The error lies in failing to recognise that different NLP-related techniques utilise knowledge of a person's representational systems in different ways (see Frogs into Princes, page 28).

But the errors don't stop there.  Witkowski continues:

All other hypotheses of the NLP system [sic] related to the arising of mental disorders, the type of therapy and communication, etc. stem from these basic assertions.
(Pages 4-5)

To be blunt, almost the entire sentence is untrue.

  • NLP is a single modelling technique.  Neither NLP nor the FoNLP is concerned with "hypotheses", only observations.
  • Neither NLP nor the FoNLP are related to "mental disorders" (this particular error appeared in a poll carried out by Norcross et al (2006), which is referred to at the end of Witkowski's article), and neither NLP nor the FoNLP can be used to determine what kind of therapy should be used to deal with "mental disorders".
  • Only the rather sweeping reference to the use of NLP-related techniques as an aid to communication is correct, as various NLP-related techniques can be used to enhance the effectiveness of most therapeutic techniques
  • The validity of the overall implication - that the PRS concept is the basis of "NLP" - should be considered in light of this fact:

    At a meeting with Richard Bandler in Santa Cruz, California, on July 9, 1986, the influence subcommittee ... was informed that PRS was no longer considered an important component [of the FoNLP].
    (Druckman and Swets, 1988.  Page 140)

In basic terms, a primary use of predicate matching - based on a person's PRS - was creating/enhancing and maintaining rapport.  As the FoNLP developed (in the late 1970s), other techniques that could be used for this purpose were added, hence the lessening of the importance of PRSs and predicate matching in particular.

The FoNLP and "Science"

Neither Bandler nor Grinder have had much time for academic psychologists, whom they regard as being locked into a mindset where "should" takes precedence over "does".  That is to say - as is perfectly demonstrated in Witkowski's article - the focus is not on what works (which has practical value) but on what "should" work, according to some theory or set of statistics (which often, as in Witkowski's article, turns out to be of limited value at most or, in most cases, positively obstructive).
Witkowski attributes this attitude specifically to Bandler (page 5), apparently unaware that (a) psychology is not a science in the way that neurology, chemistry, biology, etc. are sciences, (b) Grinder has agreed with Bandler since 1974 at the latest, and (c) going by comments made on various NLP-related online chat groups this disdain is shared by a majority of the members of the NLP community in English-speaking countries.

After all, why worry about someone else's theories about techniques which you are already successfully using?


Selection of Material for Analysis

(Witkowski's subheadings)

In this section the author explains how he selected an online database OF ABSTRACTS which he claims is "on the web pages of NLP Community [sic]".  It is not clear whether Mr Witkowski understands that "NLP Community" is the name of the website - not a description of some kind of official status.  On the contrary (again from comments made - and not made - on various NLP-related chat groups) it seems that few English-speaking NLPers have heard of this site, and even fewer have ever visited it.

Witkowski further states that he used the database because:

In Poland this base is recommended by e.g. Polski Instytut NLP (The Polish Institute for NLP) whose founder and chairman – Benedykt Peczko – personally suggested it to me ..."
(Page 6)

I will not bother to comment on this since the whole question of where Witkowski found this information is more or less irrelevant.  For reasons that will become crystal clear a little later in this evaluation.

Quantitative Analysis

(Witkowski's subheading)

At first glance there might not seem to be any point in discussing this next section in any detail either, other than to evaluate Witkowski's claims as to the significance of the journals which featured the articles he is citing.  Indeed, this portion of the article is highly ironical in that the more Witkowski tries to justify his approach, the more he undermines the validity of the article. : As in this next statement:

[My first operation] was to select the most reliable studies for further analysis.  To this end, I evaluated them based on the criterion of whether the magazine in which the given articles were published is noted on the Master Journal List of the Institute for Scientific Information in Philadelphia.
(Page 7)

And then, in a rather strange codicil:

This operation does not require justification in more detail.
(Page 7)

In reality this point is very definitely in need of further discussion, because it comes very close to the most fundamental flaw in Witkowski's article - and those of several other academic criticisms I have evaluated in the FAQ #28 Project.  And maybe Witkowski himself understood this, at some level, since he immediately goes on to try to provide a justification for his action:

Although there are many doubts raised to this list, magazines from the Master Journal List are much less likely to have published unreliable articles than others.
(Page 7)

In other words, Mr Witkowski is claiming that the status of any given journal on the ISI list is some kind of guarantee of the accuracy of the contents of any article published in that journal.  But there is absolutely no justification for that assumption, especially in relation to articles on subjects that neither editors nor the relevant referees are genuinely well-informed about - such as NLP and/or the FoNLP.  As I will now demonstrate:

If we accept Mr Witkowski's argument, then according to the Journal of Counselling Psychology, which the Polish Ministry of Science and Higher Education has given a "weighting" of 15 (the higher the better):

  • In 1981 they published a peer-reviewed experimental report submitted by William Falzett, which stated that his research "supported the contention that matching of the client's PRS with counselor predicates can enhance the atmosphere of trust in the relationship".  So the predicate matching technique was supposedly effective;
  • In 1984 they published Sharpley's first review, also peer-reviewed, which said that if PRSs existed at all they were unstable and very difficult to detect.  So that's at best a "highly doubtful" rating for the predicate matching technique;
  • In 1986 they published Einspruch and Forman's critique of Sharpley's article and a further review, also peer-reviewed/  According to this article, Sharpley had failed to take account of some very basic flaws in the original research he reviewed - like the possibility of results being skewed as a consequence of the "observer effect", and the predicate matching technique was back in favour;
  • But in 1987 they published Sharpley's second review, peer-reviewed ... etc., in which he claimed he would rebut Einspruch and Forman's allegations.  So bang went the predicate matching technique, again.

In practice, then, Mr Witkowski's "hypothesis of reliability" is itself completely unreliable.  And not just at the lower ratings.  He also cites articles from 6 journals with a weighting of 24.  They include:

  • An article by Coe and Sharcoff (1985) in the International Journal of Clinical and Experimental Hypnosis which is invalid because the researchers were examining a claim that the co-creators of the FoNLP had never made;
  • An article by Matthews, Kirsch and Mosher (1985) in the Journal of Abnormal Psychology which is invalid because it investigates the "NLP double hypnotic induction", which doesn't exist, i.e. whilst there is certainly a technique called the "double hypnotic induction", it is not an NLP-related technique;
  • Witkowski claims that an article by Marcel Kinsbourne (1974), in Neuropsycholgia, "is exceptional as it tested the hypotheses concerning eye movements" (page 12).  In fact Kinsbourne was definitely not testing any NLP-related claims or observations.  He was examining the possibility that "eye and head turning indicates cerebral lateralization".  In other words, it was aimed at studying a possible link head and/or eye movements as an indication of physiological brain functionality rather than the results of unspecified brain activity.

A Mountain Out of a Molehill

And there's another reason to take especial notice of this section - because it illustrates the major flaw in the article.

To be precise, despite several pages of text relating to a variety of statistics about the reports he has chosen to include in his analysis, Witkowski consistently overlooks, or ignores, a key question that I've already touched on.  To be precise:

  • At no point in the article does Witkowski consider the question: Were the various articles based on an accurate understanding of NLP-related concepts, techniques and claims?  Because unless they were, the status of the journal any article appeared in, and the total number of articles in any of Witkowski's three categories - supportive, inconclusive/mixed result and non-supportive - tells us nothing about authentic NLP-related concepts and/or techniques and/or claims.

Which is a very real problem.  I have not yet read all of the articles cited, but the articles that I have read seldom if ever show a full/accurate understanding of the authentic FoNLP.  I think this includes approx. three-quarters of Witkowski's 33 studies (in the order they are mentioned in his article) but since he does not provide an explicit list I cannot be sure:

  1. Thomason, Arbuckle & Cady (1980)
  2. Farmer, Rooney & Cunningham (1985)
  3. Poffel & Cross (1985)
  4. Burke et al (2003)
  5. Gumm, Walker & Day (1982)
  6. Coe & Scharcoff (1985)
  7. Fromme and Daniell (1984)
  8. Elich, Thompson & Miller (1985)
  9. Graunke & Roberts (1985)
  10. Sharpley (1984)
  11. Sharpley (1987)
  12. Dowd & Pety (1982)
  13. Dowd & Hingst, (1983)
  14. Ellickson (1983)
  15. Krugman, Kirsch & Wickless (1985)
  16. Matthews, Kirsch & Mosher (1985)
  17. Mercier & Johnson (1984)
  18. Hammer (1983)
  19. Buckner & Mera (1987)
  20. Wertheim, Habib & Cumming (1986)
  21. Durand, Wetzel & Hansen (1989)
  22. Wilbur & Wilbur (1987)
  23. Heap (1988)
  24. Dorn, Brunson, Bradford and Atwater (1983)
  25. Norcross, Koocher & Garofalo (2006)

Graunke & Roberts and Hammer's studies both produced results which confirmed a claim made by Bandler and Grinder but which the researchers mistakenly assumed had rebutted a genuine claim.

Since Witkowski fails to give a complete, clear list of the 33 studies he says he analysed it isn't possible to tell whether this article was on that list or not.

Despite Witkowski's claims:

This high number of magazines may be treated as indirect verification of the reliability of the gathered empirical evidence.
(Page 9)


I put the selected sample through a quantitative analysis, as a result of which three categories of studies emerged:
1.  thirty-three empirical articles, which tested the tenets of the concept and/or the tenets-derived hypothesis.
(Page 9)

In practice both of these claims are plain wrong, and I'm not sure the second claim even makes sense since Witkowski's classifications simply match those described by Sharpley in 1984.  So what contribution did Witkowski's highly dubious "quantitative analysis" make?
In practice it seems that all we really have here is yet more prima facie evidence that Witkowski, along with a substantial number of his selected researchers, had/have failed to address what NLP and the FoNLP are about.

Spot the Target

On pages 10-11 of his article Witkowski discusses his methodology, but to little useful effect.

As mentioned near the start of this evaluation, there is a presupposition often used by NLPers that goes: "The map is not the territory", or "don't eat the menu".  But that is clearly exactly what Witkowski is doing in this article.
Though it is entitled NLP Research Database - State of the Art or Pseudoscientific Decoration?, by the halfway point in the article Witkowski makes it clear that he is treating the database (or "base", as he repeatedly calls it) as though it were the FoNLP in its entirety.  It seems that Witkowski genuinely believes that if he can find fault with this database then he has demonstrated the existence of some major flaw in the FoNLP itself.  Thus, in the Conclusions section of the article he writes:

The analysis of the Neuro-Linguistic Programming Research Data Base [sic] (state of the art) by all measures was like peeling an onion.  To reach its core [sic], I first had to remove some useless layers, and once I arrived, I felt like crying.  Today, after 35 years of research devoted to the concept, NLP [sic] reminds one more of an unstable house built on the sand rather than an edifice founded on empirically based rock.
(Page 21)

This may look like a crushing indictment, but in reality it doesn't even make much sense.

  • A database is simply a repository for information which, in most modern databases, can be thought of as though it were arranged in a series of inter-related lists or "tables"
  • Therefore there is no "core" in the sense Witkowski seems to be using the word, and there is no database operation that equates to "remov[ing] useless layers"
  • Witkowski makes two implicit assumptions:
    1. That the FoNLP - which is concerned with subjective experience - should be amenable to a form of investigation which is applicable to objective facts;
    2. That the way a database on a single site is managed somehow reflects on the credibility of the set of concepts and techniques that make up the FoNLP as a whole.
    Neither of these assumptions is logically coherent.

Qualitative Analysis

(Witkowski's subheading)

Here, again, all we get - apart from something close to a rehash of Sharpley's invalid review articles, is more evidence that Witkowski has failed to check the claims allegedly being investigated against genuine NLP-related claims.  On page 15, for example, we read:

With respect to the category of non-supportive articles, the majority of studies concerned the basic NLP tenets [sic].  Several works were devoted only to tests of the eye movement hypothesis (Thomason, Arbuckle & Cady, 1980; Farmer, Rooney & Cunningham, 1985; Poffel & Cross, 1985; Burke et al., 2003).  They all provided unequivocally negative results.  The preferred modality was researched into by Gumm, Walker and Day(1982), and also by Coe and Scharcoff (1985).  In both cases the results did not support the neuro-linguistic programming theory.
(Page 13)

Most of this statement is true.  But the bits that aren't true are the bits that invalidate Witkowski's claims.
Because, as indicated in the previous list, none of these investigations addressed genuine claims made for the FoNLP, and therefore could not possibly have supported or not supported any "neuro-linguistic programming theory", even if the FoNLP were based on theory rather than observations.  For example:

  • Gumm, Walker and Day(1982) investigated the spurious idea that a person's PRS could be determined by following eye-accessing cues, or self-report, or tracking their use of predicates;
  • If Witkowski has used only the relevant Abstract to evaluate the Thomason, Arbuckle & Cady, 1980 result, then it is worth pointing out that the relevant abstract - in its entirety - reads: "This study did not support the eye movements hypothesis".
    It seems quite incredible that anyone would treat this as sufficient information to evaluate the reliability of the research;
  • Coe and Scharcoff (1985) based their work on not one but two errors.  They started with the erroneous idea that NLPers supposed that: "... most people have a primary representational system for dealing with the world" (what were the rest of us supposed to have?  And when did "preferred" become "primary"?).  They then followed the same error as Gumm, Walker and Day regarding how a person's "primary" representational system could be determined.  (And remember, this was in a journal with a rating of 24);
  • Burke et al's 2003 study is another example of an allegedly negative result that actually confirms authentic claims made about the FoNLP.  The abstract includes this observation: "Instead of a universal pattern, as suggested by the neurolinguistic [sic] programming hypothesis, this study yielded subject-specific idiosyncratic eye-movements across all modalities".  But there has never been an authentic claim that any universal pattern existed.
    On the contrary, Bandler and Grinder - when discussing eye accessing cues - advised trainees: "You will find people who are organized in odd ways.  But even somebody who is organized in a totally different way will be systematic; their eye movements will be systematic for them". (Frogs into Princes, 1979.  Page 27.  Italics as in original text).

This citing of studies that failed to investigate genuine claims made for NLP-related concepts and techniques continues through to page 16, still without a single question as to the accuracy of the guidelines used by the researchers.  It seems that like Sharpley, Witkowski regards adherence to convention - such as having a control group - as being far more important than the question of whether researchers are actually doing what they think/say they are doing.


(Witkowski's subheading)

As if to again emphasize that last point, in the very first sentence of this section Witkowski states that:

Among the studies classified into the category of the NLP tenets supportive, there is none, which would indicate in unequivocal terms the existence of different representational systems.
(Page 16)

Let me write this slowly so that there can be no mistake -

There are five representational systems (in NLP-related jargon), which, according to Bandler and Grinder, equate to the five senses enjoyed by most human beings:


Skin, viscera, etc.


(Tactile and Emotional)


Visual cortex
Broca's area, Wernicke's area, etc.
Somatosensory cortex, etc.
Olfactory cortex
Gustatory cortex

Is there really some doubt as to whether a person with all their faculties has five senses, and that these five senses are used to collect and process information?  Unless there is, the assertion that there is no evidence "which would indicate in unequivocal terms the existence of different representational systems" - borrowed from Sharpley, if I recall correctly - is a statement for which, to coin a phrase, there isn't one shred of supporting evidence.

The Higher the Fewer?

The next "error" appears when we are told that:

The studies classified into the non-supportive category are marked by a much higher methodological level.
(Page 17)

This harks back to my previous comment about which characteristics Witkowski deems important in reports of research which allegedly investigates some aspect of the FoNLP.

We have now seen that a majority of Witkowski's evidence comes from reports and reviews which do not address genuine claims made about authentic NLP-related concepts and/or techniques.  Yet Witkowski still holds the research in high esteem because it is, he alleges, marked by a much higher methodological level.

In what sense, we might reasonably ask, is carrying out research without adequate preparation or knowledge, and thereby obtaining meaningless results, the mark of "a much higher methodological level"?

Fairly and Objectively Investigated?

Further evidence that Witkowski's own article is based on inadequate research/knowledge comes at the end of page 17:

Heap (1988) analyzed 63 empirical studies and came to the conclusion that the assertions of NLP writers concerning representational systems have been objectively and fairly investigated and found to be lacking.
(Pages 17-18)

Of course Witkowski quotes this fact (yes, that is indeed what Heap wrote, in 1988).  What he either ignores or simply doesn't know about is an article on the subject entitled Neuro-Linguistic Programming, Time for an informed review, which appeared in Dr. Heap's own yearly magazine, The Skeptical Intelligencer (Vol. 11, 2008), which rebutted many of Heap's claims.  Or as Dr Heap himself put it:

[The author of the article] has, metaphorically speaking, rolled up his sleeves and given me a thorough beating in his paper
(Heap, 2009.  Accessed at, December 16, 2010)

This is not to say that Heap has suddenly switched sides, but rather that he has - unlike so many critics - been willing to give a contrary point of view a fair hearing.  It might be to the point for Witkowski to look at some of the contrary evidence - for example in the various evaluations in the FAQ #28 Project - instead of merely regurgitating decades old material without adding anything of any consequence.

PRSs to the Fore, Again!

Perhaps predictably, Witkowski's article now returns to the "detecting someone's PRS" scenario.  Though there's really no 'perhaps' about it, because the research being discussed is so incredibly limited in range.  This time he cites Dorn, Brunson, Bradford and Atwater's (1983) review of the research, saying that they "concluded from their review of the literature that there was no demonstrably reliable method of assessing the hypothesized PRS.  Yet as we've already seen, Bandler and Grinder had twice, in published works, made it clear that all that one needed to do to discover a person's PRS was "listen to the predicates a person uses to describe their experience".
In other words, a person's current PRS is whichever sensory mode their use of predicates last indicated - in the current situation.
How can that possibly be an unreliable method?

Witkowski Calling Base

And another reprise - this time a little under two pages about what Witkowski seems to regard as faults in the Beilefeld "base" - which of course is completely irrelevant to an evaluation of NLP (the modelling process) and/or the FoNLP (the various concepts and techniques connected with NLP).

Witkowski makes a brave, though pointless attempt to forge such a link when he writes:

The base is commonly invoked by NLP followers and indicated as evidence for the existence of solid empirical grounds of their preferred concept. ...
(Page 18)

This may be true in Poland - though Witkowski notably fails to provide any evidence to support his claim - but I personally cannot think of anyone that has invoked this database in any discussion or online article I know - other than critics of whatever they think of as "NLP".
This observation is possibly further supported by Witkowski himself since he continues with the allegation:

It is most likely that most of them have never looked through the base.
(Page 18)

Witkowski tries to loop this back to prove that the database is merely "the background for the pseudoscientific farce that NLP appears to be."  But how does his analysis of the database demonstrate anything about the elements within NLP or the overall FoNLP?  This is a non sequitur for which Witkowski provides no viable evidence.  Whatever the people running the database in Beilefeld do tells us nothing about NLP, the FoNLP or even the NLP-related community in general.  If this fact wasn't taken into account when Witkowski was preparing his article then his article is even further adrift than the contents of my evaluation may have suggested.

Indeed, the next claim, like most of the rest of his allegations, has been around for over a quarter of a century, though it is effectively no more than a thinly disguised attempt to present an opinion as a fact:

It [the allegedly pseudoscientific nature of NLP] is primarily revealed in the language - full of borrowing from science or expressions referring to it, devoid of any meanings whatsoever.
(Page 19)

Harsh words, especially since Witkowski only provides one example, and that, like so many of Witkowski's allegations, is yet another blatant error based on ignorance of the facts.

It is seen already [sic] in the very name - neuro-linguistic programming - which is a cruel deception.  At the neuronal level it provides no explanation at all and has nothing in common with academic linguistics or programming.
(Page 19)

Wrong, wrong and wrong.  And in an overall sense - wrong again:

  • What we know now, but didn't realize prior to the major advances in the study of the brain over the last 2-3 decades, is the degree of plasticity in the brain, right down at the synaptic level (a single neuron may share synapses with up to 10,000, or even more, other neurons).  Our personal experiences are constantly modelling and re-modelling the networks of connections within the brain.  And insofar as NLP and the FoNLP are concerned with perceptions and experience they are directly concerned with activity at the neuronal level;
  • At the time that he started working with Bandler and Pucelik, John Grinder was a non-tenured professor of linguistics at the University of California, Santa Cruz.  Since gaining his doctorate in linguistics, Grinder had worked with one of the leading linguists of the day, Noam Chomsky, and had co-authored a college level textbook on Transformational Grammar.  The suggestion that Grinder did not bring his skills and knowledge as a linguist to the development of NLP and the FoNLP is simply not credible.  Though it might be interesting to see Witkowski try to justify this claim;
  • The notion of the human brain being like some kind of computer and therefore in some sense programmable was very popular in the 1960s and 1970s.  Nor has use of the metaphor died out even now.  Like so many of Witkowski's claims this has been said many times before. and makes no more sense now than it did then;
  • But the biggest mistake - again almost certainly borrowed from elsewhere - is the claim that the name "neuro-linguistic programming" was an attempt to make "NLP" look scientific.
    I suspect that the people who say this are missing two important points:
    1. The fact that the term neurolinguistics, though it had been used informally beforehand, was only adopted formally when it was used by Harry Whitaker, who founded the Journal of Neurolinguistics in 1985.
    2. The term "neuro-linguistic", on the other hand, was coined by Alfred Korzybski, the creator of General Semantics, long before it was adopted by the co-creators of the FoNLP.  So far I have traced the public use of the term back to 1935 - half a century before the first publication of the Journal of Neurolinguistics, and approximately 40 years before Richard Bandler decided to combine it with the word "programming" as the name for the study of communication that he, Grinder and Pucelik were working on.  (Korzybski's definitive book on General Semantics - Science and Sanity - is included in the bibliography of Bandler and Grinder's first book on NLP and the FoNLP, The Structure of Magic I (1975).

    And by the way, "neuro-linguistic" (with hyphen) refers to the interaction between our brains and the language we hear and use.
    It is not connected with aphasiology, or linguistic consequences of damage to the brain, etc.  Thus the assumed connection is purely a matter of erroneous perception and not a matter of fact.

In the Beginning is the End

I started this evaluation by saying that the first error in Witkowski's article was in the first sentence of the Abstract.  And sure enough it turns out that the last error is in the final sentence.

In fact, the last three pages of the article are as far off beam as any other part of the article:

If You Go on Doing What They Did Then ...

One may even come to believe that my analysis was a vain effort after all.  It yielded the same conclusions as the ones arrived at by Sharpley (1984, 1987), Heap (1988) and others.
(Page 19)

This is a somewhat puzzling observation.

  • Witkowski has based pretty much his entire article on those of Sharpley (1984, 1987) and Heap (1988) - but ignoring Heap's follow-up article published in 2008;
  • At least half of his research examples were used by Sharpley and/or Heap in the 1980s;
  • Those additional examples that we have tracked so far are invalid for the same reasons as the earlier studies;
  • Although Witkowski includes both The Structure of Magic II (1976) and Frogs into Princes (1979) in his "References", he has somehow managed to overlook the vital piece of information that there is only one way to determine a person's preferred representational system, which is clearly stated in both books;
  • Witkowski has repeated Sharpley's error (re-stated in an e-mail to myself in 2009) that having an article published in a peer-reviewed magazine ensures that the contents are correct, and
  • Has totally ignored the possibility that there might be (and in fact are) crucial errors in many (maybe all) of the research examples cited in his article.

Under those circumstances why on earth would we expect Witkowski to get any results that differ from those of Sharpley and Heap, over thirty years ago?

The Blind Leading the Blind

The next observation shows, beyond any possible doubt, that Witkowski is still applying the same totally uncritical acceptance to recent research that Sharpley and Heap applied to the studies they reviewed:

Instead we find NLP on the list of discredited therapies.  Norcross, Koocher and Garofalo (2006) sought to establish consensus on discredited psychological treatments and assessments using Delphi methodology.
(Page 21)

This study, together with its 2009 follow-up constitute - in my opinion - two of the most incompetent pieces of research carried out by professional researchers that I have ever come across.

I have written up both polls in this evaluation.  For the moment I would just like to note a few things that Witkowski has failed to mention:

  1. The fact that Norcross et al made no attempt to describe the "treatments" that the people being polled were asked to rate, so there was no check that everyone was rating the same thing;
  2. The fact that the so-called "NLP-therapy" offered in the poll doesn't even exist;
  3. The fact that despite this limitation, 29% of the self-rated "expert" respondents in the first of two rounds of polling claimed that they knew enough about this hypothetical technique to rate it for creditability!
  4. The fact that the first stage poll was sent to 290 "experts.  Only 138 people replied, and 37 of those replies were unusable ("for a multitude of reasons").
    Of the 101 people who received the second poll, only 85 replied.
    And of the 85, 24.1% said that they didn't know enough to rate the so-called "Neuro-Linguistic Programming (NLP) ... treatment".
    That left just 64.5 experts who actually gave a rating to the non-existent "treatment" in the second poll (and no, I don't know how Norcross et al got that half an expert, I'm just reporting their published figures);
  5. The fact that the "Delphi Methodology" features built-in peer pressure between the first and second rounds;
  6. The fact that even though it didn't exist, the "NLP" treatment wasn't rated as "discredited".
    In the first round it was given a rating of 3.57 (3 = "possibly discredited", 4 = "probably discredited"), and in round 2, after peer pressure had been applied, a rating of 3.87.

Readers might think it interesting that Witkowski either didn't know about, or chose to ignore, all of these rather basic and obvious anomalies.
Is that really what one would call a "scientific" approach?

Start With an Error - End With an Error

In the last lines of the Conclusions section of his article, Witkowski writes:

In 1988 Heap passed a verdict on NLP. As the title of his article indicated, an interim one.  In the conclusions he wrote:
If it turns out to be the case that these therapeutic procedures are indeed as rapid and powerful as is claimed, no one will rejoice more than the present author.  If however these claims fare no better than the ones already investigated then the final verdict on NLP will be a harsh one indeed (p. 276).

I am fully convinced that we have gathered enough evidence to announce this harsh verdict already now [sic].
(Page 21)

Which takes us back to the previous point about the crucial lack of originality in this article.

Apart from the occasional mention of a piece of research from some time later than the 1980s, and the tedious and essentially pointless discussion of notional rating of journals, this entire article could have been put together using nothing more than Sharpley and Heap's reviews and the abstracts from a couple of dozen of the relevant pieces of original research.  Thus, in the example below, you will find a paragraph from Dr Michael Heap's 1989 chapter (page 268) on the left - and an unattributed paragraph from the online version of Witkowski's article (page 4) on the right:

Perhaps the central philosophy of NLP is most aptly summed up in the sentence 'The Map is not the Territory' (see, e.g. Lankton, 1980, p. 7).

The central philosophy of NLP is summed up in the sentence “The Map is not the Territory” (see, e.g. Lankton, 1980, p. 7).

That is, each of us operates on the basis of our internal representation of the world (our 'map') and not the world itself (the 'territory').  The maps that we create may be limited in many ways, impoverished, distorted and inflexible.

That means, each of us operates on the basis of our internal representation of the world (the “map”) and not the world itself (the “territory”). The maps that we create are mostly limited and distorted.

The choices that we thus make available to ourselves are restricted, and our transactions with the world will accordingly be frustrating and difficult.

(Witkowski skips over this portion of Dr Heap's text)

It is therefore the therapist’s task to understand and operate on the basis of the client’s map of the territory ...

The therapist’s task is to understand and operate on the basis of the client’s map of the territory.

... in order to assist the client to overcome these restrictions and thus provide him with more choices.

(Witkowski's paragraph ends at the close of the previous sentence)


To represent this article as being in any way an advance on the decades-old material it is based on is simply being economical with the truth.  Witkowski has provided no persuasive evidence that Heap's views - or Sharpley's, or anyone else's - were anything more than under-researched and consequently ill-informed - as, indeed, is Witkowski's own article.


Bandler, R., & Grinder, J. (1978/1979). Frogs into Princes.  Moab, UT: Real People Press.†

Burke, D. T., Meleger, A., Schneider, J. C., Snyder, J., Dorvlo, A. S., & Al-Adawi, S. (2003).  Eye-movements and ongoing task processing. Perceptual and Motor Skills, 96, 1330-1338.

Coe, W. C., & Scharcoff, J. A. (1985).  An empirical evaluation of the neurolinguistic programming model.  International Journal of Clinical and Experimental Hypnosis, 33, 310-318.

Dorn, F. J., Brunson, B. I., Bradford, I., & Atwater, M. (1983).  Assessment of primary representational systems with neurolinguistic programming: Examination of preliminary literature.  American Mental Health Counselors Association Journal, 5, 161-168.

Dowd, T. E., & Hingst, A. G. (1983).  Matching therapists' predicates: an in vivo test of effectiveness.  Perceptual and Motor Skills, 57, 207-210.

Dowd, T. E., & Pety, J. (1982).  Effect of counselor predicate matching on perceived social influence and client satisfaction.  Journal of Counseling Psychology, 29, 206-209.

Elich. M., Thompson, R. W., & Miller, L. (1985).  Mental imagery as revealed by eye movements and spoken predicates: a test of Neurolinguistic Programming.  Journal of Counseling Psychology, 32, 622-625.

Ellickson, J. L. (1983).  Representational systems and eye movements in an interview.  Journal of Counseling Psychology, 30, 339-345.

Fromme, D. K., & Daniell, J. (1984).  Neurolinguistic Programming examined: imagery, sensory mode, and communication.  Journal of Counseling Psychology, 31.

Grinder, J. and Bandler, R. (1976).  The Structure of Magic II.  Science and Behavior Books, Palo Alto.

Graunke, B., & Roberts, T. K. (1985).  Neurolinguistic programming: The impact of imagery tasks on sensory predicate usage.  Journal of Counseling Psychology, 32, 525-530.

Hammer, A. (1983).  Matching perceptual predicates: Effect on perceived empathy in a counseling analogue.  Journal of Counseling Psychology, 30, 172-179.

Heap. M. (1988) Neurolinguistic programming: An interim verdict. In M. Heap (Ed.) Hypnosis: Current Clinical, Experimental and Forensic Practices. London: Croom Helm, pp 268-280.

Novella, S. (2007).  Neurolinguistic Programming and other Nonsense on Neurologic Blog site

Sharpley, C. F. (1984).  Predicate matching in NLP: a review of research on the preferred representational systemJournal of Counseling Psychology, 31, 238-248.

Sharpley, C. F. (1987).  Research findings on Neurolinguistic Programming: nonsupportive data or untestable theory?.  Journal of Counseling Psychology, 34, 103-107.

Witkowski, T. (2010 ?).  Thirty-Five Years of Research on Neuro-Linguistic Programming.  NLP Research Database State of the Art or Pseudoscientific Decoration?.  Published online and accessed at on December 14, 2010.

† The book Frogs into Princes is the edited transcript of a training seminar which most likely took place in Colorado in 1978.


Andy Bradbury can be contacted at: