Why Introduce Meta-Levels to Modeling?

Chapter 3 from the book:

Advanced Modeling Using Meta-Levels
by L. Michael Hall, Ph.D

Continuing The Study of the
Structure of Subjective Experience

With the first extensive work on modeling in NLP (NLP: The Study of the Structure of Subjective Experience, 1980), the co-originators of this field provided an enriched version of the TOTE model of Miller, Gallanter, and Pribram (1960 Plans and the Structure of Behavior). And that had updated the original Stimulus-- Response model (S-R) of Behaviorism. Ultimately these modeling formats have provided us a flow chart of human neuro-linguistic behaviors and responses from originating stimulus (trigger) to final experience (S-R).

So what, if anything, do we find missing in this model that we need to add?

We primarily find meta-levels missing in these modeling formats. Overall, they operate, and we use them, in a linear and horizontal fashion. I say "primarily," because the NLP Strategy Model certainly did include "the meta response," although we find even the meta-responses formatted in diagrams as just another linear step in the model.

-- In terms of modeling, this has predisposed strategy analysis to working with processes in linear ways and to neglecting higher meta-levels. Numerous exceptions, of course, exist to this generalization. Among those who made modeling their business, individuals like Wyatt Woodsmall have incorporated the higher level frames of values, beliefs, meta-programs, etc. in their modeling. Others, like John McWhirter, have also incorporated the importance of the why questions into their modeling processes as they have recognized the importance of exploring the reasons, motivations, values, and beliefs that drive the strategy.

To model these higher meta-level facets of subjective experience and enrich the modeling model itself -- we have to expand the model of modeling so that it gives much more prominence to meta-levels. This means several things. Minimally, it means--

The Realm of the "Whys?"

In NLP, we do generally do not ask the "why" question. As a model of models and a model of modeling -- we have historically focused primarily on how things work. This has directed our attention to process, not on content, and especially not on "the why of content." Thus we ask lots and lots of process questions that elicit the structure of the subjective experience:

"How do you do this?"

"How do you know you feel joyful or depressed?"

"If I wanted to replicate this experience of effective leadership, what do I have to think and how do I do this thinking, feeling, choosing, etc.?"

Accordingly, Bandler and Grinder began this field by stressing (and over-stressing) the absolute importance of process and structure. As they did, they had to distance themselves from the very strong and pervasive influence of content and the "why" question.

"Why do you feel depressed?"

"Why do you get yourself into these states?"

"Why can't you seem to get out of these moods?"

"Why do you always sabotage yourself when things go well?"

Why We Created an Inhibition Against Asking "Why?"

An important reason governed the prohibition against the why question. After all, what happens when you ask the why question with regard to a problem state? It invites you to begin to explain that problem state , to come up with reasons for it, reasons that support it, and reasons that justify it. The why question also tends to send our brains backwards -- backwards to the history from which the pattern or problem arose. Accordingly, it presupposes that "if I could just understand where this came from, its origins, then I could overcome it and put it behind me."

Traditional psychology from the days of Sigmund Freud had long operated on these assumptions. "Conscious understanding about the source of problems inherently cures." "Insight per se is curative." "You can't get over a problem until you understand its causes."

Understanding causes and sources can sometimes play a therapeutic role in our minds-and-emotions. If in response to the why, we go, "Oh, so that's why I have thought and felt like this!" "Oh, that explains the thoughts, feelings, and experiences I have had! I now see how I drew some false and inadequate conclusions about myself, others, and that I did so based on inadequate information."

Yet more typically, the exploration of the why tends to create more and more of a problem-focus and, more often than not, increasingly solidifies a person in his or her problem. Neuro-linguistically this makes sense. After all, when we ask why? we invite a person to explain and justify their experience. This drives them inside (a transderivational search, TDS) process to their history and model of the world. And in doing this, it locates them even more solidly inside the very frames-of-reference that created the problem.

Further, brains seem to have this peculiar habit of inventing answers. Plant a question in consciousness -- and your brain will go to work to construct some answer to it. It will do this even for unanswerable questions. Unless you teach your brain to question some questions, your brain will naturally operate from the assumption that "there is an answer to this question and I must find or create it." Accordingly, this makes some questions especially dangerous and toxic.

"Why am I so stupid?"

"Why don't things go smoother for me?"

"Why did I marry such a jerk?"

"Why do I always sabotage myself?"

Ask those questions for a brain untrained in questioning frames and meta-frames, and it will go about the job of providing answers. As it does, however, it unknowingly accepts all the assumptions in the questions while it focuses on simply explaining things! [By the way, this demonstrates the importance of building and running our meta-brain with some real useful meta-awarenesses like, "I don't have to, nor should I, accept every question at face value. I can question questions before I even begin to answer them! I can decide which questions will elicit useful information."]

The modeling approach of NLP therefore began with a whole set of opposite assumptions.

We don't need to understand the source or origin of a problem to solve it.

We only need to understand how it works and its internal structure.

Understanding how we do the behavior that doesn't work allows us to stop repeating the limiting behavior.

We can overcome limiting patterns by focusing more fully on what we want (our desired outcome).

The more we cue our brain about our desired outcome -- the more we program ourselves to become solution-focused.

Flexibility of response in the here-and-now enables us to access the resources that we need in moving toward the future.

With these emphases, NLP from the beginning took a very negative attitude toward why? questions. Yet as it did, some unfortunate consequences arose. What limitations has the why? inhibition generated?

It has created an basic dislike of any search for past causation.

It has blinded us from discerning the existence of different kinds of "whys,"

It has prevented us from modeling the meta-levels of why?

The Numerous Kinds of "Whys?"

With regard to the why? question, we can identify several different kinds of "whys."

The Why of Causation/Source.

"Why do you act (feel, think) that way?"

The Why of Explanation.

"Why do you judge yourself so harshly?"

The Why of Teleology/Outcome. (Final effects, desired outcomes)

"Why do you do that?" (i.e. What do you seek to accomplish in doing that? For what purpose?"

The Why of Value/Importance. (Values, frames-of-references, beliefs)

"Why do you do that?" (i.e. What value does this hold for you?")
"Why do you find this important and significant?"

An Aversion to the Concept of Causation

The inhibition of the why? question has also created a general aversion to exploration of past "causes." With some NLPers that I have known, this has seemed to create a state of intolerance about listening or pacing a person's problem state. They have taken the why? inhibition so literally and seriously (not exactly "the spirit of NLP"!), that they become distraught when they even begin to work with someone's story that involves very much historical content. They want to rush forward to giving out solutions with interventions of cure before they even begin to pace the person's model of the world.

By way of contrast we do see a more balanced approach in some of the Time-Lines processes and especially the Re-Imprinting Process of Robert Dilts. These NLP patterns certainly take a more balanced and thoughtful approach to "past" sources of difficulties and pains (i.e. past beliefs, decisions, experiences, etc.). It even uses the TDS process for tracking down earlier occurrences of a mapping problems around "self," worth, dignity, purpose, destiny, etc.

Getting to "The Whys?" By Incorporating Meta-States

In the next chapter (Ch. 4), I will explore the meaning and significance of logical levels in NLP and to the Meta-States Model (Hall, 1995). Later, I will devote an entire chapter to that model (Ch. 9). This model allows us to incorporate the "whys" of teleology/ outcome, explanation, causation, and value/importance into our modeling.

Interesting enough, this approach corresponds to the most recent models in the field of linguistics. There Fauiconnier (1987) developed the idea of mental spaces which Lakoff (1987) and Langacker (1987, 1991) has used in their development of Cognitive Linguistics. Here, as they move further and further away from the defunct model of Chomsky's (1956, 1965) old Transformational Grammar, they posit a linguistics and grammar based on "embodiment" about how people actually think and represent information using mental spaces. As Lakoff and others increasingly discovered the inadequacies and anomalies of the old model, they broke away from Transformation/Generative Grammar and emphasized new factors: the role of metaphor, metonym, and image in conceptualization, prototype categories (instead of formal, classical Aristotelian categories), cognitive domains of knowledge, etc.

What does all of this mean?

It highlights the NLP principle of frames and frames-of-frames as the contexts within which "meaning" arises. In other words, meta-levels. We have always said that meaning depends upon context. "Contexts determine and govern meaning." In other words, we need to know more than just the mere representation of data -- whether the expert we wish to model coded it in visual, auditory, or kinesthetic, or even the submodality qualities of the representation. We have to also know about the meta-context within which such representations occurs. We need to know the domain of knowledge -- the beliefs, values, understandings, etc. This takes us upward into the domain of meta-level (or meta-state) processes. Hence, Figure 3:1.

I first happened upon these meta-levels in modeling as I researched the topic of resilience. After getting a good sense of "the territory" of the concept, I sought to identify the strategy of numerous individuals who manifested the ability of resiliently bouncing back from seeming tragedies. As I unpacked their strategies and sought to understand them -- I kept finding myself having to go to meta-levels to comprehend how their neuro-linguistics worked. As a result, this lead to me to the discovery of Meta-States.

These resilient ones didn't just have thoughts-and-feelings that induced them into a primary state of "resilience." They had a complex system of thoughts-and-feelings about thoughts-and-feelings, that is mind-body states about other mind-body states. Recognizing these meta-levels of states (that we typically label "values," "beliefs," frames of references, conceptual categories, mental contexts, abstract domains, etc.), I also understood why a mere replication of the thoughts-and-feelings at the primary level (the first level structure of the strategy) would probably not work if directly installed in someone -- without the meta-level structures.

Figure 3:1

Meta-Levels Beliefs








Primary Level




Internal Representations in
a particular sequence

About --


To not understand why a person does what he or she does -- the values, beliefs, and understandings that drive the behavior (whether a skill, expertise, expression of personal genius, etc.), prevents us from obtaining a full picture of the experience. Certainly we need to know what a person represents, how they code it in their VAK representational modalities and submodalities, and what meta-programs they use in sorting the things they pay attention to. No doubt about that.

We also need to understand the meta-level frames and frames-of-frames (contexts and contexts-of-contexts) within which a person lives, breathes, thinks, and operates. To do that we have to go meta. We have to go meta to the person's meta-levels and there we have to elicit meta-level distinctions about their frames of values, beliefs, assumptions, etc.

When we do this -- we will work with their "logical levels" of abstractions and gain information into how their psycho-logics operates.

Note: If you would like more information about the book NEURO-LINGUISTIC PROGRAMMING - GOING META: Advanced Modeling Using Meta-Levels, contact L. Michael Hall, Ph.D. at: .


About Us

NS Community


NS Writings



Site Map

(c)1999 L. Michael Hall. All rights reserved.