Self-Modelling

Definition

While a Clean Language questioner constructs their own external "model" of what the speaker is saying, the speaker simultaneously does something different and more valuable: they self-model, teasing apart their own previously "joined up" experience into pieces they can examine consciously. Because the questions add almost no outside content, self-modelling happens using only the speaker's own material, which is why it can surface insights the speaker did not consciously have and the questioner never intended to produce.

In the Book

Chapter 7 distinguishes the questioner's modelling (building a diagram from what someone else says, held lightly as an approximation, not "the truth") from the speaker's self-modelling. It illustrates self-modelling with Zannie Rose's client, who came to a session about something unrelated, mentioned feeling "beaten with a rod" in certain conversations, and — a few Clean questions later — realized the "rod" was the same as her recurrent backache and reported it had stopped hurting, with no conscious goal of resolving it. Judy's coaching client Catherine arrived aware only of having many career options and left with a self-generated metaphor of stepping stones across a tranquil stream, giving her a concrete next step without needing a full plan. The book flags "Oh — I'm doing it again!" as the tell-tale sign a speaker has caught themselves self-modelling in real time.

Why It Matters

This concept locates the mechanism of change inside the person being questioned rather than inside the questioner's expertise or interpretation — the facilitator's job is to keep the channel clean enough that self-modelling can happen, not to supply the insight. It generalizes to any practice built around asking rather than telling (coaching, Socratic teaching, therapeutic interviewing): the value isn't the answer you'd have given, it's the structure the other party discovers by being made to articulate their own experience without your content mixed in.