What was seen
The child moved behind the chair when the adult approached, looked towards the floor and did not answer the question.
Speaking Charlotte's Language
What happened, what you noticed and what you think it means are connected - but they are not the same thing.
Say what you saw. Then make your interpretation visible. Then make the reasoning connecting them answerable.
You do not need the full systems architecture to write one careful note. Begin here.
Evidence or observation → interpretation → action. Concern can be serious without being written as though the observer directly witnessed the child's internal state.
The child moved behind the chair when the adult approached, looked towards the floor and did not answer the question.
I was concerned that the child may have felt frightened or uncertain.
I recorded the observation, made my concern explicit and sought further information.
An earlier version compressed the distinction between a child's lived reality and an adult's observation of it. This revision restores the fuller distinction between what occurred, the observer's account and the meaning later attributed to it.
This correction is visible because a project concerned with correction must model the standard it asks of others.
What occurred belongs to the field. Observation is an observer's account of what they perceived. Interpretation is the meaning attributed to what was perceived or otherwise known.
Professional judgement is necessary. The safeguard is not to remove judgement, but to keep consequential judgement visible enough to remain answerable to source, evidence, purpose, conditions, limits and correction.
The observer is not only the practitioner present with the child. A report writer, manager, panel member, reviewer or future professional reading an earlier record receives what survived previous observation, selection, representation, interpretation and compression.
At each consequential transition ask:
Role, purpose, authority, workload, time, organisational expectation, previous knowledge, emotional state and an existing direction of travel may affect what becomes noticeable, meaningful, recordable or actionable. This does not presume bias or improper motive. It asks the cleaner question: What is driving me here?
Selection does not end when an event is first recorded. It continues through retention, summary, report-writing, presentation, decision and institutional memory. At each significant transition ask what was selected, what was excluded and what materially changed.
Excellent reasoning cannot recover information that never reached the reasoning field.
The strength of a professional concern should not determine the shape of the evidence used to express it. Belief about the preferable outcome should remain distinguishable from representation of the evidence used to support it.
Concern may outweigh an observation. It should not reshape the observation.
Where materially necessary, successive records should preserve the route from evidence to interpretation to action. An earlier interpretation must not silently become later evidence merely because it entered the record. An action does not independently validate the interpretation that contributed to it.
Evidential discipline should not weaken merely because information supports or challenges an existing concern, trajectory or preferred conclusion.
Would I require the same evidential standard if this information pointed in the opposite direction?
Parity concerns discipline, not equal weight.
Historical records influence later attention, questioning, expectations, credibility, interpretation, selection and action. Historical knowledge must therefore retain provenance, temporal status, evidential status and corrigibility.
Repetition of one interpretation does not create independent corroboration.
A correction that does not travel is not a complete correction.
A correction that travels but cannot alter the knowledge it corrects has not completed its epistemic function. Correction does not require every preferred account to be accepted. It requires sufficiently reliable corrective information to be capable of changing consequential understanding where the evidence warrants it.
Institutions become stewards of consequential knowledge about human lives; they do not become owners of the realities that knowledge represents.
No record can contain a child.
An institution deserves confidence in what it says it knows when observation, representation, interpretation, inference, uncertainty, history, contestability and correction remain visible enough to be tested.
Sometimes the most trustworthy conclusion remains: We do not presently know.
This principle must not delay necessary safeguarding, medical, legal, statutory, supervisory, evidence-preservation or emergency action. Where action cannot wait, meet the duty first and preserve, as far as possible, the distinction between what was observed, what was inferred and why the action was necessary.
The child is not the observation. The observation is not the interpretation. The interpretation is not the child.