Robert Dilts Logical Levels - a critique

Dr Mike Armour

I am always delighted to receive submissions for publishing in CorporateCoach. Mike Armour submitted these comments (see below) in response to my video on YouTube.

Robert Dilts developed his Neurological Levels of Change model from a theoretical structure described by Gregory Bateson in his book Steps to an Ecology of Mind.

Bateson's concept was that learning takes place on different levels and the Dilts' model appears to map across directly.

Bateson's levels are shown here with Dilts' levels in brackets: -

• A context for learning (environment)


• Learning (behaviour)


• Learning to learn (skills and competencies)


• Believing that learning is vital to life (beliefs and values)


• Being a learner (identity)


• Contributing to others with your learning (purpose)

Robert Dilts Logical Levels of Change

I find this model very useful. I use it in facilitation and coaching as an audit tool for collecting information and also for building models of self and others (e.g. in negotiation), or present state and future state. In its original context it is used to identify appropriate levels for intervention or discussion.

However, I know that some people – including John Grinder, apparently – dispute that it is technically a model at all.

For me, it is just a tool that is useful.

Here is what Mike Armour says: -

Like you, I borrow heavily from NLP in my work with clients, and am certified at the practitioner, advanced practitioner, and trainer levels. Early in the process of certification I became interested in Robert Dilts work, and have drawn on it regularly, particularly his logical levels model and his Disney patterns.

I thought I might offer some observations with you about that model.

Let me frame what I'm about to say with a general observation about NLP. Many of the approaches that have developed in NLP over the years contain elements that contradict basic presuppositions of the field itself. Just to cite an example, the intervention models developed by the Andreases sometimes call for purposefully creating parts to achieve certain strategies, while one of the primary presuppositions of NLP is that the mind does not need parts to operate.

There is a similar lack of precision and consistency, it seems to me, in Dilts' model. And because of this imprecision, I have modified it in my own use.

The basic problem is that Dilts calls his model "logical levels", taken right out of the language of Gregory Bateson. But the Dilts' levels do not fit the basic structure of logical levels in Bateson, where each level fully encompasses everything in the level below it. Or to put it another way, every successive level upward in Bateson is a greater abstraction of the levels below it. The levels in Dilts' model do not follow that structure. So Dilts' model, by being called Logical Levels, creates an inconsistency in the broad literature of NLP, because we then end up with two different models called Logical Levels, with fundamentally different underlying structures.

But beyond that, NLP is clearly about neurological sorting. And my other concern is that Dilts' model is not cleanly sorted from a neurological standpoint. This is particularly the case at the values/beliefs level. Values, by NLP definitions, function in the affective realm. Beliefs function in the cognitive domain. Putting them side by side in the same level makes for a weaker model, in my judgment, than if we treat values and beliefs each as a distinctive level and explore their dynamic separately.

Another shortcoming of Dilts' model, I believe, is that it does not have a place for attitudes. While attitudes are driven by values and beliefs, they are a distinctly different category of internal processing. From my assessment, they more nearly belong to the affective domain than the cognitive domain and therefore should not be seen as just another form of belief. And unlike beliefs, which operate in the Internal Processing/Representation segment of the Mercedes Model ('think - feel - do', editor.), attitudes express themselves primarily as Internal States.

In my own work, therefore, I've modified Dilts' model to create a level called Attitudes just above what I call Skills and Abilities and which most people call Capabilities. Further, I've separated Beliefs out as a separate level and married it to something else that neither Dilts nor NLP in general has paid much attention to, and that's Conviction. While convictions are a type of belief, there is an existential difference in believing something and being absolutely convinced of it. Since both belief and conviction are out of the cognitive realm, I typically speak of "Beliefs and Convictions" together when I describe this level.

I think one of the greatest challenges is finding a proper name for the top level in Dilts. Using Spirituality at this level does not seem to go very well in business settings. Many have substituted Purpose as a name for this level, which is workable in most settings, but perhaps not comprehensive enough unless it is defined in terms very similar to Viktor Frankl's use of the word. So far I've not been able to identify an English term that encompasses all that this level embraces. I've ended up with the somewhat awkward term "Transcendent Connection," which I then have to explain as that cause, relationship, or ideal which transcends our own personal existence and to which we are so committed that we will sacrifice everything for it. This definition then embraces things rather removed from spirituality, including patriotism and family, as well as the transcendent spiritual realm.

Dr. Mike Armour is the founder of
Strategic Leadership Development International,
based in Dallas, Texas.

blog comments powered by Disqus