A Critique of NLP

The 1988 book, Enhancing Human Performance, published by the National Academy of Science’s Division of Behavioral Social Sciences and Education, edited by Daniel Druckman and John Swets, contains the most comprehensive examination of Neuro-Linguistic Programming. It concludes that the evidence for NLP’s efficacy is weak. Although the critique is over 20 years old, many of the reasons they cite for the lack of positive empirical data are still valid.

NLP is criticized for lacking internal consistency. The terms are occasionally loose, and at different times and among different practitioners, the importance of various constructs shift. In addition, say Druckman and Swets, many of the NLP constructs, such as the statistical brain or the holographic brain, are metaphors disguised as scientific fact. These same criticisms, of course, have been aimed at Freud’s concepts of id, ego and superego.

NLP was founded and developed by observing what works and then developing models and hypotheses from those observations. This is not a model that lends itself to traditional, controlled design research, especially when observations of different clients and different practitioners can contradict each other. Its founders, Bandler and Grinder, were not interested in proving the efficacy of NLP through traditional methodology and indeed often seemed disdainful of such research. They were more content to demonstrate its effectiveness, believing that demonstration was sufficient proof. While much of NLP has been content to rest its efficacy claims upon the notion that “what works, works,” this is not a test that gives it respectability among researchers and the traditional establishment. With its loosely defined terms and lack of traditional empirical proof, NLP has acquired the reputation of a fringe movement, with all the negative connotations which that carries.

As an example of the lack of empirical research, Druckman and Swets say there is “no direct support cited for the NLP-postulated relation between eye gaze direction and representational system [italics theirs].” (This is also a criticism leveled at EMDR—a technique, incidentally, that many NLP people claim was expropriated from Bandler and Grinder without attribution). This, and other suppositions that support NLP, insist Druckman and and Swets, are “merely a series of concatenated anecdotes and facts that lead to no particular conclusion,” and the experiments that demonstrate its efficacy are mostly case studies.

After Grinder and Bandler had a severe falling out in 1981, the movement splintered, which added to the lack of internal consistency. It is difficult, if not impossible, to study a treatment which lacks not only consistent definition of its terms but consistent training among its practitioners. According to Steve Andreas, in an unpublished article:

Since there is no credentialing or licensing organization, or any other way of maintaining integrity and consistency, training in “NLP” is at least as varied as the wider field of psychotherapy…. NLP developed far from the halls of academia, and there are no controls on the use of the name, or criteria for who can advertise themselves as trainers. Anyone can offer NLP trainings, whether they have any qualifications or not, and some people have done so on the basis of having read a book or attended a weekend workshop. While this freedom has allowed for a lot of fruitful experimentation and development, it has also opened the floodgates for all sorts of incompetence, misuse and exploitation. Any technology that works can be used to harm as well as to heal, and NLP is no exception.

Much of the NLP research has been done by practitioners who, quite simply, did not know how, or didn’t care, to set up solid study designs. According to Andreas, who was trained as a scientist in his pre-NLP and pre-psychology days, most of the studies failed to select appropriate aspects of NLP to research, create an appropriate linguistic/behavioral protocol for gathering data, or control for confounding variables. After reviewing approximately 20 studies of NLP, Druckman and Swets concluded that the studies which found positive effects were not well-controlled or well-operationalized designs. Some studies for example, claimed to find NLP effective by measuring high degrees of therapist-client empathy which do not indicate therapeutic effectiveness. (Again, Druckman and Swets were writing in the days before other research found significant correlations between therapist-client alliance and positive therapy outcomes, but their critique still stands: one cannot prove effectiveness of outcomes by measuring something else).

Druckman and Swets went to considerable lengths to fairly review NLP. Their subcommittee even conducted interviews with one of its founders and with several of its trainers and trainees. Although some trainees reported negative experiences, the majority were “satisfied trainees who believe that NLP has improved their communication skills and made them more effective in exercising interpersonal influence.” Although this suggests that NLP is effective in increasing the self-confidence and subjective well-being of trainees, say, Druckman and Swets, “subjective feelings of change are not necessarily a reliable guide to increased performance.”

They concluded that, “There is little or no empirical evidence to support either NLP assumptions or NLP effectiveness.”

References

Andreas, Steve (unpublished). What Ever Happened to Neuro-Linguistic Programming?

Druckman, D and Swets, JA, eds (1988). Enhancing Human Performance: Issues, Theories and Techniques, Washington, DC: National Academies Press

 
Top of Page