The Sins* of the Fathers

by Carmen Bostic St. Clair and John Grinder

In a rather remarkable burst of activity and in a relatively short space of time (1973 - 1979), Richard Bandler and John Grinder created a series of models that included three of the most highly regarded psychotherapists/psychiatrists in the English speaking world: Fritz Perls, Virginia Satir and Milton H. Erickson. This collaboration between Grinder and Bandler literally created the discipline known as Neuro-Linguistic Programming (NLP). All this is uncontroversial and has achieved a status approaching mythology.

The core activity that characterized the activities of these two men, the co-creators of the technology known as NLP, was modeling - a complex set of activities that can be understood in its final analysis as the ability to map tacit knowledge (behavioral competency) onto an explicit, transferable representation. This modeling expresses itself in a collection of patterns known in NLP circles as a model. By whatever accidents of personal history (see Whispering in the Wind by Carmen Bostic St. Clair and John Grinder, especially the section entitled Personal Antecedents of NLP, pages 120 - 138, where Grinder offers his representation of these accidents) Grinder and Bandler had developed a set of skills that allowed them to succeed in mastering significant portions of the patterning of these highly recognized therapeutic communicators and of codifying these patterns.

While a presentation of the intricacies of modeling, NLP’s core activity and the method used by Bandler and Grinder in the joint studies that established NLP, are well beyond the scope of this short article, the reader is invited to a presentation both of various aspects of modeling and vivid l.//and highly specific descriptions of some of the contexts of discovery which came from this historical era (Whispering in the Wind, the section entitled Contexts of Discovery, pages 140 - 197 and especially pages 179 - 197, see www.nlpwhisperinginthewind.com).

Looking back from the perceptual position of 2002, especially with the advantage of having created a second code for NLP application, it is clear that there are certain flaws in the coding of what has come to be known as the classic code (roughly, the set of patterns coded by Grinder and Bandler during their collaboration 1973 - 1979). Our purpose in this article is to briefly identify what these flaws are and to propose a simple strategy for correcting them.

We find it prudent to open this portion of the article with a highly personal statement by one of the co-creators of NLP:

Personal Statement by John Grinder

The creation of Neuro-Linguistic Programming (NLP) represents a superb example of collaboration. I could not have created NLP by myself nor do I believe, could have Richard Bandler. Each of us brought specific talents and capabilities to the endeavor, not the least of which was the ability to work as a team. For some six years, we worked side by side as researchers, provoking, supporting, challenging and amusing one another in our efforts to codify excellence in terms that made it available to the rest of the world.

Both as individuals and as a team, we followed the strategy of Acting As If impeccably and offered one another continuing challenges, stimulation and feedback as we developed the representations of the patterns that presently define the Classic Code in NLP. While it may be possible to distinguish partially the initial strengths of each of us, there was a deep cross-training that occurred in our collaboration through which we learned from one another how to carry out the extraordinary feats that have set the historical standard for NLP practice - both at the level of modeling as well as in its applications. I therefore recognize with pleasure the essential historical contribution of Richard Bandler as the co-creator alongside myself of the technology of NLP, and I specifically offer him even now my congratulations and best wishes in his continuing work.

Those readers in search of a model of excellent collaboration will do well to step past the present state of affairs between us and focus on the work accomplished by the two of us in the period 1973 through 1979.

Whispering in the Wind by Carmen Bostic St. Clair and John Grinder, page 120

As I (JG) hope the above statement makes clear, I am quite proud of the patterning that Bandler and I accomplished together. We were breaking entirely new ground and managed to make a significant contribution. Nevertheless, as part of this trial and error experimentation, we made decisions, especially in the coding phase of our collaboration that in retrospect require correction. We will proceed by offering and then analyzing a prototypic anchoring format - a generalization over the set of anchoring patterns coded and presented in the classic code and in wide use in applications in NLP:

Prototypic Classic Anchoring NLP Pattern

1. Identification (consciously) by the client of the change to be made (present state)

2. Identification (consciously) of the difference the client desires - this can take the form of identifying the desired state or the resource the client wishes to apply to the present state to change it or simply the specific behavior that the client desires to experience in the context in which he or she wants the change to occur

3. Accessing of both the present and the desired states/resource (typically each are anchored) - the sequence of accessing and anchoring depends on the perceived needs of the client and the style of the agent of change and is, in general, not a critical ordering

4. Making the connection (e.g. integrating, sequencing, stacking, chaining, future pacing…) between the present state and the desired state or resource or new behavior, typically through the manipulation of anchors.

5. Test the work for effectiveness (anchors, in the street…)

A moment’s thought will reveal that this generalized (or prototypic) format covers formats as diverse as collapsing anchors, change personal history, time lines… Thus the remarks that follow are perfectly general and applicable to all anchoring formats that we are familiar with.

Consider what is being proposed here; and more interestingly, what is missing. From the above format, it is clear that the conscious mind of the client is being assigned a series of tasks:

the selection of the change to be made (present state)

the selection of the desired state, desired resource or specific behavior to be positioned through anchoring in the context in which the client desires the change to occur

As we commented in Whispering (page 214)

At some point in the anchoring format and completely consistent with the ethics of NLP application (which requires that the NLP practitioner confine his or her manipulations to the process level and leave the content entirely to the client), the NLP practitioner will ask the client to decide what the desired state (goal, objective) for the change work will be. Notice that this is a call for the client to make a conscious decision.

At some point further on in the format and equally consistent with the ethics of NLP application, the practitioner will ask the client to decide what behavior or state or resource he or she would like to implement to replace the undesirable behavior. Once again, this decision is one made consciously by the client.

These are important decisions and it is unfortunate in the extreme that the classic code assigns the responsibility for these decisions to the client’s conscious mind - precisely the part of the client least competent to make such decisions

More telling is the complete absence of any explicit involvement of the unconscious mind in any portion of the format. Given the efficacy and ecological quality of the patterns made explicit by Grinder and Bandler in their modeling of Dr. Milton H. Erickson, this is somewhat startling.

Now allow us to reassure the reader that the absence of any explicit involvement of the unconscious mind is not to be confused with the absence of actual involvement of the unconscious mind. The absence of any explicit involvement is an issue of adequate coding while it is difficult to imagine any significant change occurring without the actual active involvement of the unconscious mind.

Said differently, the involvement of the unconscious mind was always a critical part of the therapeutic encounter for Grinder and Bandler. How specifically, you ask? Let us count the ways!

Nearly all the intermediate objectives in the process of change such as rapport (ensuring the unconscious is attentive), so-called gathering information (map manipulations), determination and utilization of the available representational systems and their repetitive sequences (strategies, for those readers indoctrinated in NLP terminology)… are best accomplished non-verbally. The ongoing physiological responses to the actions of the agent of change are the reference point around which the simultaneous manipulation of states occurs - please note that this happens for both parties: the client and the agent of change. Anchoring is, after all, a symmetrical relationship.

Even in publications expressly patterning verbal productions (The Structure of Magic, for example) there are numerous warnings to the reader about the importance of non-verbal communication. In parallel with the verbal component of communication, there are complex non-verbal messages on offer. Thus the emphasis on the importance of calibration as a fundamental skill set - consider the set of spontaneous idiomatic signals available even in a “normal” conversation. And most of all beware of blind pilots!

The flaw then is in the coding rather than in the behavior. In particular we can identify the failure to capture and make explicit a crucial aspect of the interaction between client and agent of change. However, given that NLP is the modeling technology par excellence for complex human behavior, one would expect that such a crucial assignment of responsibility would be explicitly directed to the resources in the client that are most capable of performing them - in this case, the unconscious mind as opposed to the conscious mind. Further the agent of change’s relationship with the client’s unconscious as well as the relationship between the conscious and unconscious processes in the client surely deserves explicit representation.

We turn our attention to the process of coding and its consequences. In Whispering in the Wind, we point out that there are no known algorithms for mapping from a complex set of behaviors (of excellence in NLP) onto an explicit representation or model. This portion of the modeling experience remains at present an art form deeply embedded in the world of heuristics. Further, it is not difficult to demonstrate that given a comprehensive record of some highly valued behavior (the performance of a genius, say as in NLP modeling) and well specified criteria for success (efficacy - the patterning actually gets transferred - and efficiency - the transfer occurs in a relatively efficient manner), there still remains multiple possible models of this behavior of excellence; each of which meet well-specified criteria. This can be understood as an excellent demonstration of multiple perceptual positions or multiple descriptions. However, our point here is different.

Let us characterize the puzzle as a punctuation issue. Let us assume that the modeler has already met the criteria of having assimilated the patterning of the model unconsciously - with all conscious filters (all mappings (characterized as f 2 in Whispering) subsequent to the 4-tuple or First Access suspended). This is a more precise way of stating that Bandler and I set aside all conscious attempts to understand what we were imitating until we had achieved mastery of the patterning we were pursuing. Further, let us assume that the modeler has demonstrated the consistent ability to elicit from the world the same set of responses the model does with roughly the same quality and within the same time frame. How is the modeler to punctuate these now assimilated complex behaviors to arrive at an explicit model of the patterning of excellence - one that meets the criteria of efficacy and efficiency? Or equivalently, how is the modeler to select some portions of the behavior and represent them and to completely ignore others? What is the optimal level of specificity of the model?... Any modeler will recognize that the ability to answer these questions through action will determine in large part the success or failure of the modeling endeavor.

As we pointed out above, the punctuation imposed by Bandler and Grinder in the original studies in genius took a couple of turns that we suggest constitute errors of coding. The argument proceeds from consequence. In the classic code, Grinder and Bandler made certain decisions in punctuating the sequences of behavior displayed by their various models. We focus on the absence of any coding of the involvement of the unconscious mind. From the point of view of the client (and apparently for many NLP practitioners) the process pattern used in the intervention simply does not identify the unconscious mind explicitly as an active agent in this process. However as we commented above, no significant change will occur without the active involvement of the unconscious mind. The result is mystification. The punctuation fails to identify and make explicit the active involvement of the unconscious. One highly unfortunate consequence is that when the client attempts self-application - a step toward achieving independence from the agent of change - a goal that surely is deeply embedded in the ethical practice of NLP - the procedure fails. Often the client will arrive at any number of false conclusions:

I need the agent of change to make effective change

I am too inexperienced/stupid/untrained… to accomplish these changes by myself

One of the consequences of such mystification is that clients fail to achieve the independence they seek.

Consider the issue from the perceptual position of a well-intentioned would-be practitioner of NLP: sometimes pattern X from the classic code works like a dream and at other times, it works not at all. And most damaging, the would-be practitioner has no clue as to what the difference that makes the difference is. Since nowhere in the format or pattern is the explicit involvement of the unconscious mind specified nor how specifically it is to be involved.

These then constitute coding errors which in their enthusiasm at the time, Grinder and Bandler committed as they worked to create the fundamental patterns of change application during their collaboration. All these flaws proceed from the initial punctuation by the modelers. Grinder and Bandler so assumed that ability to establish and maintain a high state of rapport with the unconscious of the client and to monitor the idiomotor signals of the client in order to detect acceptance or rejection of the various behaviors that were being proposed and acted upon, that they failed to make this critical aspect of the patterning explicit. - with the resultant mystification we described above in both clients and would-be practitioners.

Punctuating the change encounter without identifying the appropriate assignment of responsibility for making the decision regarding desired state, resource or new behavior and most tellingly without making explicit the involvement of the unconscious results is a mystification of the entire process.

Fortunately addressing this flaw, once identified, is rather simple. We propose that these coding flaws can be most easily corrected if the following guidelines are respected.

1. there is a re-assignment of responsibilities such that the unconscious mind is actively and explicitly involved in the decision regarding the selection of desired state, new resource and/or preferred behavior.

2. there is an explicit way to involve the unconscious mind in these decisions

There are multiple ways that these could be accomplished. Perhaps the simplest is to insist on establishing of a set of involuntary signals that (ideally) the client has access to which allow him or her to present any decision to the unconscious mind for ratification. Thus, the practitioner could follow the same classic formats as before but insert a verification procedure by which the client uses the involuntary signals to verify that the unconscious accepts (or rejects) the decision being made consciously, step by step. In fact, historically, through calibration skills, this is precisely what typically occurs in a well-conducted change session, with the agent of change intuitively accepting the responsibility for monitoring the non-verbal signals (the naturalistic version of the involuntary signal system) for acceptance or rejection of each move in the dance of change. When discussed at all, this is usually understood to be an ongoing congruency check. The point is that as long as the agent of change takes this role and responsibility and neither the agent or the client is explicit that this monitoring ongoing congruency check is being conducted, the client will remain mystified (and possibly the agent of change as well) as to how well the change patterning works in the presence of the agent of change and how poorly things proceed when self-application is attempted.

A signal system, verified to be involuntary - such as the class of signal systems developed in 6 step reframing - would serve well here. There are, of course, stronger and weaker versions of this solution to correcting this coding flaw. One might request of the unconscious the actual taking of these important decisions (offering responses visually and auditory - both verified by the involuntary kinesthetic signals) rather than having the conscious mind make such decisions and present them to the unconscious mind for acceptance or rejection. Either strong or weak versions of this reform would make explicit the involvement of unconscious mind in the process of change. In this sense, then, while Grinder and Bandler attended most carefully to the idiomotor signals naturally occurring in the exchange between them and clients, they failed to identify this as a specific requirement in their coding of the patterning of the change process.

We would be remiss not to encourage readers to develop and explore the myriad ways that this explicit involvement of the unconscious in making these decisions. As an example, in the new code (developed initially by John Grinder and Judith Delozier in the mid-80s and further developed by John Grinder and Carmen Bostic St. Clair in the 90s), the preferred method is the development of a know-nothing state. The client is supervised in developing a high performance state (typically through the agency of a new code game) that is then connected (through a version of future pacing - an extremely effective deployment of anchoring) with the context in which the client desires that the change occur. Indeed, the visual and auditory stimuli that define the context of application are the re-activating anchors. At no time does the client attempt to consciously formulate what differences (neither the desired state, the new resource nor the preferred behavior) they desire to occur in that context. Thus at the end of the session, the client knows something important has shifted but typically has no conscious access to the specific differences that are available. They are literally in a know-nothing state with respect to the changes made. This know-nothing state will resolve into specific behavior only when the context of application is present - that is, when client next re-enters the context and without any conscious effort on her or his part, the auditory and visual stimuli re-activate the high performance state and clients find themselves performing in new and creatively effective ways. Since the changes occur at the level of state (the high performance state replacing the client’s previous response state), there is a strong tendency for the client’s behavior to continue to vary in the context selected as various aspects of that context shift. This is clearly a generative approach to making change.

This brings us to one additional design flaw in the classic code: the absence of effective contextualization. While at the time of the collaboration between Bandler and Grinder, there was little explicit attention paid to framing and the preparation and management of context. The preferred way of knowing whether the choices being exercised by the client was, as detailed above, the ability to run an ongoing calibration for congruity. If the client selected an inappropriate new behavior, the agent of change would detect an objection from the unconscious in the form of some idiomotor signal.

The required distinction is that of distinguishing between 1st and 2nd order changes. Briefly, a second order change is required whenever any one or more of the following three markers are present:

1. an addiction

2. a physiological symptom

3 a behavior with significant secondary gain involved

The first two of these criteria are well-defined; the third requires much development (see Whispering in the Wind, section The Breakthrough Pattern pages 198 - 227 for a fuller discussion). All changes that are not second order changes are first order changes - the complement set of the set of second order changes.

Let’s take as an example a man who has a drinking disorder - an alcoholic - or to people who desire to lose weight. It can be usefully applied to any addiction. In the typical case, an investigation of the client’s past would reveal that he has succeeded in stopping drinking for limited periods of time but then returns to the bottle. If we were to make explicit what the payoffs - secondary benefits or secondary gains - of this behavior are, we would discover one or more of the following:

he drinks to relax

he drinks to escape the pressures of everyday life

he drinks to achieve a state of sociability

…

Suppose that we focus on the positive intention of achieving access to a state of relaxation. This positive intention is the name of a set - namely, the set of all behaviors that offer the client access to a state of relaxation. This set will, by definition, always include the original behavior.

Ways of Achieving a State of relaxation

b1, b2, b3,……………, bi, bi+1,…………, bi+j (alcoholism),……………………,bn

In other words, within the set of ways to achieve states of relaxation, we find a large number of behaviors, b1 (sports), b2 (reading), b3 (meditation), bi (drugs), bi+1 (yoga), bi+j (alcoholism), bn-1 (breathing exercises), bn (community service)... Once we have specified (partially at least) what the members of the set are, the change task is greatly simplified: simply select three or more behaviors from the set to replace the behavior in question - in this case, alcoholism.

In a classic addiction case, such as alcoholism, there is typically more than a single payoff or secondary gain involved. The practitioner is cautioned then to divide the change work into a series of sessions, one for each of the positive intentions and their associated payoffs. Thus, the application of this step leads naturally to the generation of a series of sets, each defined by each of the positive intentions behind the behavior to be changed.

Once again, without the active explicit engagement of the unconscious mind, there is little likelihood of a successful change. Thus the judicious use of the positive intention behind the behavior to be shifted (excessive drinking) as the context from within which the new behaviors will emerge is a powerful way of organizing the resources both of the client (the unconscious resources typically) and of the agent of change as well.

The know-nothing state from the new code represents another way of creating the effective contextualization - the use of the entire set of visual and auditory stimuli that define the context as the re-activating triggers that allows the unconscious to select the new experiences appropriate in the identified context.

Our intention in this short presentation has been to call the NLP application community’s attention to a set of coding flaws made by Grinder and Bandler in their initial collaboration. We have suggested a number of ways to correct these punctuations. In particular, we strongly urge practitioners of NLP application to consider the consequences of the these coding errors and select their own preferred ways of correcting them by insisting on the re-assignment of the responsibilities for decision making, the explicit involvement of the unconscious mind and the use of contextualization to achieve a certain precision within the process of change. Correcting these flaws will simultaneously remove the mystification from these processes and facilitate our clients achieving independence of us - always a worthy goal as part of the process of change. As this presentation has been quite brief, we invite readers interested in exploring these issues in greater depth to have a look at Whispering in the Wind (www.nlpwhisperinginthewind.com)

You can see further extracts of Whispering in the Wind and participate in an online discussion about the issues raised in this article by visiting www.nlpwhisperinginthewind.com

* We are indebted to Jeisyn Murphy (www.Got-NLP.com) for pointing out to us that one of the original meanings of the term sin was in the context of Greek archery where it simply meant that the archer has missed the target. This is perfectly in accord with our intentions behind our usage here. Both Jeisyn Murphy and Michael Carroll (michael@nlp-academy.com) offered comments that were of value to me in writing this article.