LIBRARY
View detailsComplete Pattern Library
All frameworks, all workbooks, all tools. One practitioner’s library.
04 — COMMUNICATION
Language Patterns
The specific words and structures people use reveal — and reinforce — their internal models of reality.
“’I have to’ and ’I choose to’ describe the same action from entirely different internal positions. The language someone uses tells you where they stand relative to their own experience.”
How it works
The structure underneath
Language is not a neutral reporting system. It actively shapes perception. The Milton Model (from Milton Erickson’s therapeutic language) uses deliberately vague language to allow the listener’s unconscious mind to fill in meaning. The Meta Model (from Bandler and Grinder’s early work) does the opposite — it uses precise questions to recover the specific experience behind generalised language.
Every sentence deletes, distorts, or generalises experience. ’They never listen to me’ deletes who specifically, distorts frequency (never?), and generalises across all contexts. These aren’t errors — they’re the natural compression of experience into language. But they also maintain the problem by preventing specific examination.
Predicates reveal representational preferences. ’I see what you mean,’ ’That sounds right,’ and ’I feel like that’s correct’ are not interchangeable — they indicate which sensory system the speaker is processing through. Matching predicates builds rapport at a level below conscious awareness.
The practitioner’s task is dual: to hear the structure in other people’s language, and to use their own language with precision and intention. Every question either opens or closes a conversational space.
Reframing: context and content reframes with examples
NLP language patterns are recurring structures in how people phrase things — what a sentence assumes, leaves out or frames — that shape how a listener understands it. The full pattern lists live in their own notes: the Meta Model for precision questions and the Milton Model for permissive, artfully vague language. This page goes deeper on two patterns that sit between them and turn up in almost every useful conversation: reframes and presuppositions.
A reframe changes the meaning of an experience without changing the experience. NLP distinguishes two kinds, both drawn from the early modelling of therapists such as Virginia Satir and Milton Erickson.
Context reframes
A context reframe answers the question “Where would this behaviour be useful?” It suits complaints of the form “I’m too…” or “He’s too…”, because almost every behaviour has a context in which it is an asset.
“My team lead is so stubborn, nothing gets changed once it’s agreed.”
“Where would you want someone who holds a line once it’s agreed? Say, during a client negotiation, or when scope starts to creep?”
Content reframes
A content reframe, sometimes called a meaning reframe, keeps the context and asks “What else could this mean?” It suits complex equivalences, sentences where one event is taken to mean another.
“The client asked for three rounds of changes. They obviously don’t trust us.”
“Or they’re invested enough to keep going instead of walking away. Which would they have done if they didn’t trust you?”
A reframe works only if it is true enough for the listener to accept. Offered as a slogan it sounds dismissive; offered as a genuine question it widens the person’s view. Watch the response: a pause, a change in breathing or a slower “Hm, maybe” suggests the new meaning has landed. An immediate “Yes, but…” tells you it has not, and that you have moved too fast.
Presuppositions in everyday language
A presupposition is what must already be true for a sentence to make sense. It is never stated directly, which is why it is persuasive and why it is worth hearing. “When did you stop missing deadlines?” presupposes that you used to miss them. “Which of these options would you like to start with?” presupposes that you will start with one.
| Form | Example | What it takes for granted |
|---|---|---|
| Time words (before, after, when, while) | “Before we sign off the budget, let’s check the timeline.” | The budget will be signed off. |
| Ordinals (first, next, another) | “What’s the first change you’d make?” | There will be more than one change. |
| Either/or | “Shall we meet Tuesday or Thursday?” | We will meet. |
| Awareness verbs (realise, notice, know) | “Have you noticed how much calmer the team is?” | The team is calmer. |
| Change-of-state words (stop, still, again, yet) | “You haven’t solved it yet.” | You will solve it. |
In conversation, listen for the presuppositions in the other person’s problem statement. “Why do I always sabotage good opportunities?” presupposes that it happens always, that it is sabotage and that the opportunities were good. Each assumption is a place to ask a Meta Model question before you attempt any answer. “Always? When was the last time it didn’t happen?” “What makes it sabotage, rather than a reasonable choice at the time?” Answering the question as asked would accept all three presuppositions and, quietly, strengthen them. Questioning them first is often the whole intervention.
Using language patterns well: practice and limits
- Hear before you use. Spend a week noticing presuppositions and reframes in meetings and emails before you try to construct them. Recognition comes first.
- Match the pattern to the moment. Meta Model questions suit vague problem statements. Milton-style language suits moments when a person needs room to find their own answer. Using the wrong one feels either intrusive or evasive.
- Keep it ethical. Presuppositions that close down choice (“You’ll obviously want the premium package”) are manipulative and tend to damage trust when noticed. Use them to open options, not to remove them.
- Treat predicate matching lightly. Mirroring sensory words can help a conversation feel easier, but controlled studies have not supported the idea of a fixed preferred representational system. Listen for it as a cue, not as a diagnosis.
Free quiz · 15 questions · ~6 min
Take the free quizTest your knowledge: NLP Practitioner Knowledge Quiz
Test your grasp of the core NLP Practitioner toolkit: rapport, sensory language, anchoring, precision questions, outcomes and reframes.
In practice
Where it applies
- Therapeutic conversation — hearing the structure of a problem in how someone describes it, not just what they describe
- Influence and persuasion — matching the linguistic patterns of your audience to increase receptivity
- Written communication — constructing language that opens rather than closes cognitive space
- Coaching — using precise language to challenge limiting frames without triggering defensiveness
Connections
Related frameworks
Next
Continue the series
Language Patterns: questions people ask
What are NLP language patterns?
NLP language patterns are recurring structures in how people speak that reveal, or influence, how they are thinking. The two main collections are the Meta Model, a set of questions that recover detail lost in generalised language, and the Milton Model, a set of deliberately general phrasings adapted from Milton Erickson’s hypnotic work. Reframes and presuppositions draw on both.
What is the difference between the Milton Model and the Meta Model?
They are mirror images. The Meta Model, published by Bandler and Grinder in “The Structure of Magic” (1975), uses specific questions to move from vague language back to concrete experience. The Milton Model, from their 1975 study of Erickson, uses vague language so the listener fills in their own meaning. See the Meta Model and Milton Model notes for full lists.
What is a reframe in NLP?
A reframe changes the meaning someone gives to an event without changing the event. A context reframe asks where a behaviour would be useful; a content reframe asks what else an event could mean. Reframes work when the listener finds the new meaning plausible, so they are best offered as questions rather than corrections.
Do NLP language patterns actually work?
The evidence base for NLP as a whole is limited and contested, and some claims, such as fixed preferred sensory systems, have not held up in controlled studies. The language patterns themselves are better understood as practical listening and questioning tools. Asking a precise question or noticing a hidden assumption is useful whatever theory you attach to it.
Put it to work
View all 13 tools