Article from

First presented at The Developing Group 6 Oct 2012

What’s in a Name?: The effect of labelling on experience

James Lawley

The name you can say isn't the real name.
(Tao Te Ching)

Penny Tompkins and I are taking a course, ‘Model Thinking’. We’ve noticed that just about every aspect of the course is labelled. Each module, subject, model, formula, methodology has a name. We started musing on just how much the name or label influences our thinking about the topic. For example, in economics there is a model of how people make decisions called the ‘rational agent model’. It is based on the assumption that people optimise, i.e. they make the best, most profitable decision possible. It took decades before economists like Daniel Kahneman and Amos Tversky tried something radical - they investigated what people actually do. To everyone’s surprise ‘rational’ behaviour is rare, even among economists! Thus Behavioural Economics was born and scientists have been discovering more and more ‘cognitive biases’ ever since. At the last count, listed almost two hundred.[1]

Let’s look a little deeper into these labels. Once a decision-making process is called ‘rational’ it presupposes everything else is other-than-rational. By comparison, if what most people do most of the time is not rational, can it legitimately be called a cognitive 'bias’? Bias is not a neutral term. It comes with the pejorative connotations of ‘prejudice’, ‘distortion’ or ‘deviation from the norm’. And few can resist the impulse to want to ‘correct’ thinking labelled in that way.

Lets go back for a moment to ‘rational’. The Meta-Model question: ‘Rational compared to what?’ is helpful here. The answer is mathematical purity. Based on an exceedingly small and very defined part of the world, mathematics can prove there is an optimum solution. This is the ‘positivist stance’ which maintains there is an ‘objective solution’ to a problem.

But this approach is doomed to be an inadequate model of behaviour because we humans  take into account the wider context, the immediate situation, our history, our expectations, our feelings, other people’s behaviour, feedback, etc. etc. Complexities personal to the individual which no mathematical model can take into account.

Having said that, although the rational agent model may be inadequate it can still be useful.

Compare the positivist stance of the rational agent model to the ‘constructivist stance’ inherent in one of the Presuppositions of NLP: “People make the best choices available to them, given their model of the world”.[2] From this perspective, people's thinking is not ‘irrational’ or ‘bias’ – quite the opposite. Each decision makes sense from within the individual's model of the world. If the decisions doesn’t make sense to us that just means we don’t understand enough about the factors that influenced their decision-making.

Interestingly, the format of the Coursera quizzes compounds the ‘objective’, ‘rational’ and ‘right’ way of thinking because there is always a “correct” solution to every question. As I discovered, even spelling a mathematicians name wrong by one letter is judged  “incorrect” (by the computer program that marks the test).

The same kind of thinking can be applied to the ‘normal’ distribution (also known, more neutrally, as the Bell-shaped curve). Everything else must be other-than-normal. But when we look at the real world we find many, if not most systems, are anything but ‘normal’.

Surely it is no coincidence that problems that can be solved by normal distributions and rational logic are simple, mathematically speaking. Mathematicians naturally picked the low-hanging fruit first. But the way they labelled that fruit constrains the way we thing about the more complex fruit on higher branches.

How much language shapes our thought has been the subject of many debates over the years. The so-called Sapir-Whorf hypothesis holds the strong view that language determines our thinking: if we don’t have a word for a concept, we cannot think about it. The weak view, called linguistic relativity, maintains language influences or structures our thinking but does not determine it.  

At first, cognitive and socio-linguists, then psychologists, and more recently neuroscientists have over the last 20 years produced substantial evidence that our language and our thinking are inseparably linked – often in surprising ways. These findings have contributed to a new conceptualising of cognition, as dynamical, embodied, extended, distributed and situated (DEEDS for short).[3]

To conclude this section, all language can be regarded as a form of labelling. Labels can be a stepping stone to learning and they can be a straight-jacket of our own making. Labels such as 'rational', 'optimal’, 'normal' and ‘bias’, bias us to think in certain ways – whether we are aware of it or not. We can counter these effects but it takes conscious intention, training our attention, and often the help of external aids.

When we consider that all words are labels, we realise every word slants our thinking in some way. Given we have to label to communicate verbally, and since most of us, most of the time, have no idea how we label, or the effects of one label over another, we thought it would make an interesting topic for a Developing Group day.

Below we begin to scratch the surface of this rich topic by looking at labelling from a number of perspectives.


Metaphor is an indirect way of labelling. It doesn’t say X is called Y, it says X is not Y but it is like Y – even if the ‘like’ is silent.

Since the early days of the cognitive linguistics revolution in the 1980s, evidence for the role of metaphor in shaping thought has grown from a trickle to a torrent. Yet some of the most compelling evidence for the effect of metaphor was already available from the field of witness testimony. For example, Elizabeth Loftus and her colleagues asked subjects who had watched a video of a traffic incident: "About how fast were the cars going when they X into each other?". Significantly higher estimates of speed were given when 'X' was smashed, compared to collided, bumped or hit.[4]  Similar discrepancies have been found when subjects were asked to estimate people’s height depending on whether the word tall or short was used.

My favourite example of scientists not thinking through the effects of a sticky name is when they started to use the term 'junk’ DNA for the vast majority of DNA that didn’t appear to do anything useful. As soon as I heard the term I said “Nature doesn’t do ‘junk’. It just means we don’t understand what this DNA does yet.” Our understanding of DNA has since expanded and it has recently been confirmed that much of this ‘junk’ is vital for ‘regulating’ genes. I wonder whether the ‘junk’ label had any effect on delaying these discoveries. It certainly fooled lots of people into thinking scientists had a complete map of the human genome.


An interesting and often unconscious way of labelling akin to metaphor is metonymy. Perhaps the most common type of metonymy uses part of a thing or process as the name for the whole thing or process, e.g. “Downing Street issued a statement”. Here the street the Prime Minister lives in stands for the entire office of the head of government. Once you ‘get’ metonymy you’ll find examples of it all over the place. For an example close to home, we need look no further than ‘clean’. It started out as a metaphor and is more and more being used metonymically.

In terms of the labelling effect, by directing our attention to a part, metonymy can reduce our appreciation of the whole.


A particularly interesting kind of labelling of experience we have studied is meta-comments. These verbal and nonverbal self-reflexive expressions comment on what has just experienced. Therapy and coaching clients often use meta-comments to signal when something significant has just happened. Recognising people’s meta-comment opens up a little used window into how they experience and respond to their own perceptions.


All labels direct attention in a certain way. We are constantly  assessing our experience against our values and preferences. Some labels come with a presupposed valence. ‘Creativity’ for example is generally regarded as positive, while ‘worry’ is considered negative. Other words are more more centrally located on our personal and cultural evaluation scales. These labels are more or less neutral.

If you examine any of the words in the standard Clean Language question set, you’ll find they have a neutral valence.[8] This is no accident. Apart from the obvious desire to not contaminate, David Grove reformed an idea of Freud’s. Psychoanalysts were trained to be blank slates onto which their patients could project and transfer their repressed urges. Grove took a sideways step. Instead of projecting on to him, he created contexts – “psychoactive psychescapes” – where clients projected on to themselves.

Just because we can name a phenomenon doesn’t mean we understand it or that we have explained it. Gregory Bateson called the tendency to mistake a name for an explanation the “dormative principle”.  Bateson explains by telling of story (naturally): Asked why opium puts people to sleep, a character in Molière’s Le Malade Imaginaire replies “Because it contains a dormative principle”.[10] The impressive label seems to explain something, and we can nod wisely, but it is tautological. Here is a more up to date example:
Client: I have a saboteur that prevents me from achieving anything.
Fac:    And how do you know you have a saboteur?
Client: I must have because I don’t achieve anything.

A label can have a positive value in medical diagnosis. Many people report feeling better, or at least relief when a doctor has given them a name for their illness. This effect occurs even when the name is simply a description of the symptoms and adds no further explanation, e.g.
Patient: Doctor, I have this problem with my skin
Doctor:  Ah, you have dermatitis.
Dermatitis is a fancy name for “inflammation of the skin”. The patient already knew their skin was inflamed, but being able to call it Dermatitis seems to add something. I’m not attempting to diminish this effect since it can be of real psychological benefit. For a start it implies the patient doesn’t have something worse. However the same effect can occur even when the illness is about as bad as it can get. [Does this phenomenon have a name?]

Equally, labelling can limit. Calling HIV/AIDS a “gay disease” had deadly consequences. But so does saying HIV/AIDS is “curable”. Researchers have started to use “curable” as a label for what was considered an “incurable” disease. This is worrying some health professional who think it may undermine their “prevention is better than a lifetime on drugs” message and lead to a rise in infection and mortality rates.

The Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, has over 300 categories of mental ‘disorders’. Many of these diagnostic categories didn’t exist until recently and some that were prevalent, like hysteria, no longer exist. Does this say more about our approach to labelling than it does about mental conditions? Thomas Szasz went as far so to argue that ‘mental illness’ was not a scientific description of a human condition but a label for a myth.[11]

In small ways we are constantly diagnosing, categorising and labelling our own experience. Sometimes our (mis)diagnosis or (mis)categorisation doesn’t serve us. Richard Bandler pointed out that (mis)labelling the ‘unfamiliar’ as ‘uncomfortable’ can prevent people from learning.

The business of labels

Labelling has become big business since it has been relabeled ‘branding’.

On a smaller scale, advertising has always used labels creatively to sell products and services. Sometimes that creativity becomes unscrupulous.

A joint investigation by BBC Panorama and the British Medical Council concluded that there is “a striking lack of evidence” for the performance-enhancing claims made for sports drinks, sports supplements and sports shoes.[12] A drink can make us go “faster, stronger, for longer” (and yes, we are talking about Lucozade Sport and not a different kind of performance enhancement) or a running shoe can “minimise injury, optimise comfort and maximise speed". The study concluded that the positive effects claimed for these products had no more value than those conferred by water, unprocessed food and bare feet. And there was some suggestion that over-relying on these so-called sports enhancing products may do more harm than good.

I have written about The Dieters Paradox which describes how labels that suggest health and well-being about a food mislead us into thinking that it is ‘good for us’ and therefore we can eat more of it – even when the opposite is true. I was recently fooled in just this way when I noticed that a single portion of the “all natural” and “organic” Greek yogurt I had purchased contained more than 50% of my recommended daily saturated fat intake.


Once a client has placed themselves where they need to be, the first thing we aim to do in Symbolic Modelling is to facilitate them to identify and then label the relevant bits of their experience (which we call ‘symbols’). Proper nouns, pronouns and common nouns maybe grammar’s official names, but verbs, adjectives and prepositions are labels as well. David Grove taught me that no matter what part of speech is used, every verbal or nonverbal description a person uses is naming that aspect of their experience. And that name can then be used to reference that experience – whether it is grammatical or not.

One of the most interesting examples of how labelling changes our experience is nominalisation – turning a process into a thing by using a noun in place of a verb, e.g. referring to ‘a thought’ rather than ‘thinking’. It is common for people’s mental representations of a noun to be static while processes are represented with movement. Nominalisation fixes phenomena. While it is often easier to talk about a thing rather than a process, it can be at the cost of objectifying something dynamic. Multiple nominalisations also have the effect of inducing trance – an experience well known to students attending university lectures.

When Gregory Bateson read the draft of The Structure of Magic I, he famously asked Richard Bandler and John Grinder: Is ‘I’ a nominalisation?

Separating adjectives from nouns

In 1998 David Grove made a discovery:
"I was looking up the word 'adjective' and found that one particular dictionary had the root [as] ‘to throw at’ or ‘to lay to’. So an adjective is a word which is thrown at a noun or laid to it. In that moment I realised that the conjoining of an adjective to a noun is not that stable. If the noun is thrown at or laid to, then I ought to be able to throw it back from whence it came, or I can fire it, I can lay it off, so that an adjective is not necessarily welded or riveted to that noun."[13]
For example, if a client says ‘I come from an alcoholic family’ the adjective ‘alcoholic’ has at some time been thrown against the noun ‘family’. David figured there was likely a time before the family had been labelled alcoholic. He devised a clever "pulling back" question for releasing or separating the noun from the tyranny of the adjective:
And what kind of [noun] was that [noun] before [it] was [adjective]?

And what kind of family was that family before it was alcoholic?

In "I was a terrible boy", the “boy” has been labelled “terrible”. If we presume there was a time before the boy was terrible, he would have had a different ‘characterlogical adjective’:[14]
Fac:    And what kind of boy was that boy before he was terrible?
Client: A free spirit.

In his two volumes of Six Blind Elephants, Steve Andreas shows just how pervasive and fundamental categorisation and hence labelling is to our way of being.[15]

One of our simplest and yet most useful categorisation models is REPROCess (Resource-Explanation-Problem-Remedy-Outcome-Change).  Originally we called the ‘remedy’ category ‘solution’. We found that some people has such strong preconceptions about this word that they couldn’t see the distinction we were making. After a couple of years trying to explain that the label ‘solution’ is irrelevant, it’s being able to distinguish the class of experience that’s important, we gave in to feedback and changed the label to ‘remedy’. We’ve never had that problem since.


Eugene Gendlin, the originator of Focusing, appreciated the value of finding just the right label for a particular kind of experience. He realised that we often get a vague sense we know something but cannot articulate it, even to ourselves. This is not an emotion, it is a “felt sense” experienced by the body.

While there is always more to a felt sense than can be expressed verbally, Genlin recognised the value of allowing the sense to become tangible and to finding a word or phrase that resonated with it. Accurately naming a felt sense is often accompanied a "felt shift". The person begins to move to a place of clarity and freedom – with insight, understanding and action often following close behind.

The parallels with Symbolic Modelling should be obvious, and that is in part because Penny and I admired Genlin’s research and methodology. David Grove certainly knew of Genlin and I’d be surprised if David’s early work with metaphor wasn’t also influenced by him.

Labels that are not directly translatable

Comparing labelling across languages produces some fascinating anomalies. Oliver Burkeman cites examples of emotions for which there is no direct translation into English:  

The Portuguese ‘saudades’ refers to a particular kind of longing, and the Korean ‘han’ is a form of collectively felt resentment in the face of injustice, blended with lamentation. But the sense of cosiness embodied by [the Danish word] ‘hygge’ is especially interesting because something like it occurs again and again in non-English languages: German ‘Gemütlichkeit’ is similar, as is Czech ‘pohoda’ and Dutch ‘gezelligheid’.[17]

Perhaps less obvious are variations in labels for colours. Different languages divide up the colour spectrum in different ways. According to John Lyons:
Word-for-word translations of colour-terms across languages is frequently impossible. For example, there is no word in French that covers exactly what ‘brown’ does in English; there is no single word in Russian, Spanish or Italian that corresponds to ‘blue’; no single word in Hungarian that corresponds to ‘red’.[18]

Apparently, some languages only have three colour-terms. What effect does that have on experience I wonder?

Spiritual States

Labelling an experience creates a strange situation. By labelling we create a separation between us and the named; between the perceiver and the perceived. Moreover, once we have labelled an experience we have not only identified it, we have in some way identified with it. By labelling, the perceiver creates a relationship with the perceived (see our PPRC model). Once we know the name for something it is a non-trivial matter to change our mind and re-label it. Therefore the relationship created is a kind of attachment.

Many spiritual traditions suggest that ‘un-attaching’ or ‘dis-identifying’ from our thoughts has value. Ken Wilber points out that before we can ‘dis-identify’ with anything we first have to identify with it. Identifying thus is a step towards transcending. Not by excluding the dis-identified aspect but by including it. Having transcended and included we are now in a position to identify it anew from a higher-level perspective. This in summary is Wilber’s 1-2-1 model of development: identify with, dis-identify from, identify at a higher level.[19]

Other spiritual practices recommend experiencing without labeling. Just being in the presence of, without internal dialogue, and without awareness of the name for our experience. Carlos Castaneda’s Don Juan called this “stopping the world”.[20]

Beyond that, the peak of the spiritual mountain is the nondual state where there is no separation at all between perceiver and perceived – or anything else for that matter.

Postscript 29 June 2013

“Studies suggest a sort of linguistic Heisenberg principle: as soon as you label a concept,  you change how people perceive it. It’s difficult to imagine a truly neutral label.” Adam Alter, The Power of Names, 3 June 2013.

Postscript 5 Dec 2015

Article in the Guardian newspaper:
"Experts say doctors over-keen on labels after woman died from refusing treatment"


1 For me the most interesting finding from behavioural economics, is not that we have ‘cognitive biases’ but that even when we are aware we are operating out of these assumptions, for the most part we continue to make decisions and act in the same way. See my blog, The Illusion of Validity

2 Robert Dilts, Strategies of Genius, Vol III, Meta Publications, 1995.

3 See my blogs: DEEDS putting minds back together and More Good DEEDS.

4 Loftus, E. F., & Zanni, G (1975). Eyewitness testimony: The influence of the wording of a question. Bulletin of the Psychonomic Society, 5, 86-88.

5 Metonymy and Part-Whole Relationships was the subject of a Developing Group paper, Aug 2002.

6 Penny Tompkins and James Lawley, ‘If only God would give us a sign’: The Role of Meta-Comments, Acuity, Vol.1, No.1, November 2010.

7 See Clean Evaluative Interviewing, a Developing Group paper, Dec 2011.

8 Of course there will be exceptions – and not just among clients. The LinkedIn Clean Language Facilitators in Business group has been discussing, Why does the question ‘what would you like to have happen?’ have the word 'have' in it?. Some facilitators react adversely to the word “have”. 

9 See Becausation,
a Developing Group paper, Apr 2006.

10 Gregory, Bateson, Mind and Nature: A Necessary Unity, Bantam, 1980, p. 90.

11 Thomas Szasz, The Myth of Mental Illness: Foundations of a Theory of Personal Conduct, 1961.

12 Panorama: The Truth About Sports Products, BBC One, 19 July 2012.

13 'Adjective' derives from the Latin verb adjicere, from ad- ‘toward’ + jacere ‘throw’. Quote from 'Conservation of Conversation' a workshop led by David Grove and Wendy Sullivan, 24-25 May 2006. 

For more on characterlogical adjectives see Judith DeLozier’s article ‘Mastery New Coding and Systemic-NLP’, NLP World, Volume 2, No. 1, March 1995 at

15 Steve Andreas Six Blind Elephants: Understanding Ourselves and Each Other, Volumes I & II, (Real People Press, 2006) delves deeply into the principles and practice of “perceptual scope and category”.  For an extract about Symbolic Modelling and Clean Language see

16 Eugene Gendlin, Focusing, Bantam Books; 2nd (revised) edition, 1982.

17 Oliver Burkeman, Are some emotions untranslatable?, The Guardian,  Saturday 3 October 2009.

18 John Lyons, Language and Linguistics, Cambridge University Press, 1981, p. 313. There’s a nice blog including a fascinating clip from BBC Horizon about how the Himba tribe describe colour at:

19 Ken Wilber, Integral Psychology: Consciousness, Spirit, Psychology, Therapy, Shambhala, 2000.

20 Carlos Castaneda, Journey to Ixtlan: The Lessons of Don Juan, Simon & Schuster, 1972.


James LawleyJames Lawley is a UKCP registered psychotherapist, coach in business, and certified NLP trainer, and professional modeller. He is a co-developer of Symbolic Modelling and co-author (with Penny Tompkins) of Metaphors in Mind: Transformation through Symbolic Modelling. For a more detailed  biography see about us and his blog.


All information on this web site (unless otherwise stated) is Copyright © 1997- Penny Tompkins and James Lawley of The Developing Company. All rights reserved. You may reproduce and disseminate any of our copyrighted information for personal use only providing the original source is clearly identified. If you wish to use the material for any other reason please contact:

Penny Tompkins and James Lawley

The Developing Company, PO Box 349, LISBURN, BT28 1WZ, United Kingdom
Tel./Fax. 0845 3 31 35 31 * International: +44 845 3 31 35 31
email: info }at{