In the 1950s and 1960s, Dr. Milton H. Erickson was achieving results with patients that defied conventional psychiatric understanding. People who had been told they could never change their behavior, recover from phobias, or address deep psychological trauma were walking out of his office transformed — sometimes after a single session. His approach was unusual: he rarely confronted resistance directly, almost never gave direct commands, and used language in a way that was simultaneously warm, engaging, and deeply hypnotic. What exactly was he doing?

When Richard Bandler and John Grinder analyzed Erickson's transcripts and recordings in the early 1970s, they discovered a precise set of linguistic patterns — structures that appeared consistently in his therapeutic language and that had very specific effects on listeners' unconscious processing. They named this system the Milton Model, after its source. Today, the Milton Model is considered one of the most sophisticated communication frameworks in NLP — a toolkit for crafting language that bypasses conscious resistance, creates deep resonance, and opens space for lasting change. This article covers all 14 core patterns with examples, practice scripts, and guidance on ethical application. For an introduction to how NLP coaching works, see our article on why NLP coaching works.

Who Was Milton Erickson?

Milton H. Erickson (1901–1980) was an American psychiatrist and psychologist who became the world's foremost practitioner of medical hypnosis. What made Erickson extraordinary was not just his effectiveness, but the nature of his approach. Unlike classical hypnosis, which relied on direct authoritarian suggestion ("You will feel relaxed"), Erickson used indirect, permissive, naturalistic language that worked with the patient's resistance rather than against it.

Erickson believed deeply that each patient possessed the resources necessary for change — the therapist's role was to create the conditions for the patient to access those resources, not to impose a solution. This belief is now one of the core NLP presuppositions. His conversational, storytelling approach seemed deceptively simple — but analysis revealed extraordinary linguistic sophistication operating beneath the surface. He had developed, largely intuitively, an entire grammar of influence.

Milton Model vs Meta Model: The Inverse Relationship

Understanding the Milton Model requires first understanding where it sits in relation to the NLP Meta Model. The Meta Model — developed from Bandler and Grinder's study of Fritz Perls and Virginia Satir — uses precise, challenging questions to challenge and clarify vague language. It moves toward specificity. When someone says "People never understand me," the Meta Model response is: "Which people specifically? Never in what situation? How do you know they don't understand?" The goal is to recover the specific, concrete experience behind the generalization.

The Milton Model is the deliberate mirror image. It intentionally creates vague, open-ended language that allows the listener to fill in meaning from their own experience. The more artfully vague the language, the more each listener personalizes its meaning. This is why Erickson's therapeutic stories resonated so differently with each patient — each person was unconsciously hearing their own story. For a deeper comparison of language patterns, see our guide to NLP language patterns for influence.

The 14 Core Milton Model Patterns

1. Nominalizations

Turning verbs into abstract nouns

Process words (verbs and adjectives) are converted into nouns, removing the action and making the concept abstract and internally defined. Because nominalizations have no specific sensory content, each listener fills them with their own meaning.

Example: "Your understanding of this creates the freedom to make new decisions." (understanding, freedom, decisions are all nominalized processes)

2. Unspecified Verbs

Verbs without a defined how

Verbs that do not specify the precise process they describe. The listener's unconscious supplies the specific action in their own way.

Example: "You can learn from this in ways that help you grow." (learn how? help how? grow in what way?)

3. Universal Quantifiers

All, every, always, never

Words that generalize to all instances. In the Milton Model, these are used to create powerful, all-encompassing positive generalizations that feel intuitively true.

Example: "Every experience you've ever had has contributed to who you are now." "All the resources you need are within you."

4. Modal Operators of Possibility and Necessity

Can, could, will, must, should

Words that frame what is possible, necessary, or inevitable. Possibility operators expand the listener's sense of what is available. Necessity operators create a sense of movement and direction.

Example: "You can discover something valuable here." "It's possible that you already know what you need."

5. Presuppositions

Embedded assumptions that are taken for granted

Presuppositions are perhaps the most powerful Milton Model pattern. They embed assumptions that the listener must accept in order to process the sentence — the assumption slips past conscious analysis.

Example: "As you continue to relax, you might notice..." (presupposes you are already relaxing) "I'm curious when you'll start to feel that shift..." (presupposes the shift will happen; the question is only about timing)

6. Pacing and Leading

Matching experience then directing it

Pacing means making statements that match the listener's current verifiable experience. Leading means following the pacing statements with a suggestion that moves them toward a desired state. Because the pacing establishes truth, the leading statement inherits its credibility.

Example: "You're reading this now [pace], and breathing [pace], and you might be wondering [pace], what it would feel like to have that sense of quiet certainty [lead]..."

7. Tag Questions

Questions appended to statements to soften resistance

Tag questions redirect the listener's attention away from evaluating the main statement to answering the tag — effectively inviting agreement or consideration with less resistance.

Example: "It would be interesting to explore what's possible, wouldn't it?" "You already know more than you think you do, don't you?"

8. Conversational Postulates

Questions that imply actions

Questions that are technically yes/no questions but pragmatically function as requests or suggestions. The listener responds to the implied meaning rather than the literal question.

Example: "Can you think of a time when you felt completely confident?" (pragmatically invites the listener to recall such a time, not merely to answer yes/no)

9. Embedded Commands

Suggestions embedded within larger sentences

Commands embedded within questions, statements, or stories so they do not register as direct commands at the conscious level. The unconscious mind processes them as instructions while the conscious mind focuses on the surrounding content.

Example: "I don't know if you're ready to feel completely at ease now, but..." "Some people find that when they read something like this, they begin to feel hopeful."

10. Double Binds

Offering choices where all options lead to the desired outcome

A double bind presents the listener with a choice, both options of which lead toward the desired outcome. The illusion of choice reduces resistance while ensuring movement in the right direction.

Example: "I don't know whether you'll notice a shift during this session or after it..." "Would you like to explore that now or in a few minutes?"

11. Lost Performative

Value judgments without a stated source

Statements that assert a value or truth without identifying who is making the judgment. This removes the evaluator and makes the statement feel like a general truth.

Example: "It's good to take time for yourself." "It's important to trust your instincts." (Who says? No source is identified — the listener accepts it as general truth.)

12. Cause and Effect

Implying causal links between events

Language that implies a causal relationship between a stimulus and a response, anchoring a desired state to an ongoing, verifiable experience.

Example: "The more you focus on your breathing, the more relaxed you become." "As you hear these words, you can allow new possibilities to emerge."

13. Mind Reading

Claiming to know the listener's internal experience

Statements that presume to know what the listener is thinking or feeling. Used carefully, these create a sense of being deeply understood. If accurate, they build profound rapport; if slightly inaccurate, the listener naturally adjusts to make them true.

Example: "You're wondering whether this will work for you..." "A part of you already knows the answer..."

14. Selectional Restriction Violation

Attributing human properties to non-human entities or processes

Speaking as if an inanimate process or part of the body has human qualities. This pattern engages unconscious processing by using metaphorical language that bypasses rational analysis.

Example: "Let your unconscious mind work on this in the background..." "Allow your wise inner self to guide the next step..."

A Complete Milton Model Practice Script

Opening a Coaching Session — Milton Model Style

"As you settle in [pace] and breathe naturally [pace], you might begin to notice a growing sense of curiosity [embedded suggestion + nominalization] about what's possible [conversational postulate] for you today. And you don't have to do anything special — just allow [embedded command] your mind to become receptive [nominalization] in its own way [double bind — it will be receptive regardless of how]. Some people find [lost performative framing] that the moment they give themselves permission to explore [cause-effect presupposition], something valuable naturally emerges... wouldn't it be interesting [tag question] to discover what that is for you today?"

Ethical Use of the Milton Model

The power of Milton Model language patterns creates an ethical responsibility. In coaching and therapeutic contexts, these patterns are used with the explicit intention of helping the client access their own resources and move toward their stated goals — they are employed transparently as part of the coaching process. In leadership and teaching, they help create resonance, inspire possibility, and facilitate learning.

The Ethical Line in Language Influence

The ethical question for any influence technique is: Does this serve the other person's genuine interests and autonomy, or does it override them? Using Milton Model language to help a coaching client access confidence, overcome a fear, or connect with their own values is ethical. Using it to manipulate someone into a decision that serves only your interests is not. Responsible NLP practitioners hold this distinction as foundational.

Integrating Milton and Meta Models in Practice

Master NLP practitioners move fluidly between the Meta Model and Milton Model depending on what a client needs in any given moment. The Meta Model is used to clarify, challenge, and specify — to gather information, break down limiting beliefs, and identify the precise structure of a problem. The Milton Model is used to create possibility, bypass resistance, deliver suggestions, and facilitate change at an unconscious level. Knowing when to shift between them — from the detective precision of the Meta Model to the artful ambiguity of the Milton Model — is one of the hallmarks of advanced NLP practice.

If you are new to NLP language patterns, a good starting point is developing your ability to recognize Milton Model patterns when you encounter them in everyday communication — in advertising, political speeches, skilled teachers, and great storytellers. You will begin to hear them everywhere. For practical NLP exercises to develop these skills in your daily life, see our guide to daily NLP practice. And for the full context of how these language skills fit into a complete NLP coaching framework, visit our NLP journey guide.

📚 Recommended Resources — Amazon.ca

As an Amazon Associate we earn from qualifying purchases at no extra cost to you.