Each of us has an inner map of the world with rich connections between all the pieces. When we attempt to communicate with others, we’re limited by the words we use, which can’t possibly capture the richness of our internal map. As a result, we lose significant information as we try to communicate.

Back in the 1960’s, Noam Chomsky popularized the terms deep structure and shallow structure¹ to describe these two levels. As I attempt to share these ideas with you now, I’m converting what I know from my deep structure into the shallow structure represented in these words. You will then read the shallow structure and convert that into the deep structure in your own brain. Along the way, much will be lost.

In the process of converting from deep to shallow structure, we will modify the information in three specific ways. Some information will just be deleted or left out. Some other will be distorted in ways that may change the meaning or intent. Finally, some specific examples will be generalized to make the information more compact, again possibly changing the meaning.

Why we care about this in a coaching context is because these modifications will also limit options in our minds and that leads to us being stuck. By asking the right questions, we can make the client aware of where information has been deleted, distorted or generalized and can help them find more options in their own thinking².

These kinds of questions are themselves a model that can be used to change the clients model of the world. In effect, a meta-model.

The NLP Meta Model comes from modeling therapy leaders Virginia Satir (family therapy) and Fritz Perls (gestalt therapy) to find the language patterns that they used to help their clients unpack this modified information.

Let’s look at the specific kinds of statements we might hear.

Deletions

Important information is left out.

simple deletions

“I am anxious.”

Fails to provide any details. Why are you anxious? Is there a person, an event, or a situation that has led to this?

unspecified referential index

“We lost the game”

Uses a pronoun where it’s not clear who or what it references. Are you a member of the team that lost or were you just watching the game?

unspecified verbs

“She hurt me”

Uses a verb that is ambiguous. How exactly did she hurt you?

judgements

“It’s bad to be late”

How do you know it’s bad? Who says it’s bad?

comparisons

“This is a better sandwich”

Better than what? We can’t tell what this being compared to.

Generalizations

Our brains are exceptionally good at generalizing from a few specific examples. These generalizations can then limit the options we can see.

modal operators of necessity

“I must be on time” or “I should be on time”

Why, exactly?

modal operators of possibility

“I can go to that party” or “I may go to that party”

What would happen if you didn’t?

universals

“All cats are lazy.”

All of them? Which cats specifically?

Distortions

Information is twisted in a way that limits choice.

nominalizations

“I need to make a decision”

Have you decided? A nominalization is a verb (something dynamic) that has been converted into a noun (something static). Converting it back to a verb will help unpack what was really meant. In this case, a “decision” is static, while deciding is not.

mind reading

“If I’m late for work again then my boss will fire me”

How do you know this? Mind reading is when we assume we know what someone else is thinking, without any evidence that this is true.

cause and effect

“If it rains today then I’ll be unhappy”

So you’ll be happy if it doesn’t rain? How specifically is your mood linked to the rain?

complex equivalents

“He’s always late. He doesn’t like the job”

How exactly are those two things connected? Could there possibly be other reasons for being late? Complex equivalents link two things that shouldn’t be linked.

presuppositions

“I’ll leave when the cat is back inside”

This presupposes that 1) there is a cat and 2) the cat is outside. Those may not be valid assumptions but we don’t tend to question information provided in the form of a presupposition.

Conclusion

The first step to helping someone with the meta-model is to be able to recognize when they are missing information. Listen for the statements here and ask questions to help the client unpack what’s really happening. Just making the client aware that information is missing or incomplete can be a helpful step.