Why active language shapes outcomes in workplace sexual misconduct cases
In workplace investigations, we talk about neutrality as though it exists in a vacuum. Policies emphasise independence. Investigators are trained not to assume guilt. Reports are carefully caveated with alleged, reported, claimed.
And yet one of the most powerful influences on how an investigation is experienced, and on how rigorous it actually is, goes almost entirely unexamined.
Language.
Not the formal outcome. Not the process map. The words we choose.
Because language is not neutral. It never has been.
Passive and active language in workplace sexual misconduct cases
Take a phrasing you will find in investigation reports across the country.
"The complainant was allegedly touched inappropriately."
Grammatically correct. Common. Seemingly neutral.
But it has removed the subject of the action. It centres the experience without ever naming who is said to have created it.
Now compare:
"The respondent allegedly touched the complainant inappropriately."
Same legal position. Same neutrality. Same allegation, unproven. But a completely different sentence, because it restores agency. It places responsibility back where it is alleged to belong, with the alleged actor, without making any finding of fact whatsoever.
This is not semantics. It is about how everyone reading that report interprets responsibility, safety and control.
How trauma affects the way people recall events
There is a substantial body of research into how people process and recall highly stressful or threatening experiences.
Rather than being encoded as neat, linear narratives, these experiences tend to be remembered in ways that prioritise sensory detail, emotion, and moments of perceived threat. Recall is often not chronological. People move between details, feelings and fragments as they describe what happened to them.
This is not dysfunction. It is what human memory does when safety is compromised.
In practice, it looks like this:
- Accounts that do not follow a strict timeline
- Details emerging gradually, or out of order
- Language that becomes careful or tentative in places
Here is where the two things meet. When an investigation report then leans heavily on passive construction, it mirrors that same lack of definition back at the reader. The account was fragmented because of trauma. The report is now vague because of grammar. And the two become indistinguishable to anyone reading it afterwards.
The complainant's memory is doing exactly what memory does under threat. The report has no such excuse.
Passive language, ambiguity and self-blame
There is a well-established pattern in the research on sexual violence. Where responsibility is unclear, people fill in the gaps themselves.
Too often, that process turns inward.
This is not a mysterious phenomenon. It has names. The just-world hypothesis describes our need to believe that outcomes are deserved, because a world in which harm arrives randomly is unbearable to contemplate. Defensive attribution describes the way we unconsciously assign some responsibility to the person harmed, precisely because it lets us believe it could not happen to us. Both operate in the reader. Both operate in the person who was harmed.
Language either interrupts those instincts or feeds them.
When our words, spoken or written, repeatedly say:
- "She was subjected to"
- "The complainant experienced"
- "It is alleged that she was"
the person at the centre of the report becomes the grammatical subject of every sentence, while the alleged behaviour floats free of any actor at all.
Read enough of those sentences and a shape forms. Something happened to her. Nobody appears to have done it.
For anyone navigating a workplace sexual misconduct investigation, the outcome is not the only thing that determines how they experience it. Even where an allegation is not upheld, being clearly heard matters. Seeing an actor and an action accurately articulated, and responsibility not quietly dissolved out of the sentence, makes a material difference to whether a person leaves the process intact.
Where language removes the alleged actor, it reinforces uncertainty and internal questioning. It also, in our experience, generates precisely the complaints of procedural unfairness and bias that organisations most want to avoid.
Where language is clear, within an independent and balanced process, people are more likely to feel seen, believed and taken seriously.
That distinction has nothing to do with influencing outcomes. It has everything to do with how those outcomes are experienced, and whether they hold.
Active language does not assume guilt. It enables a defence.
The most common objection to all of this is that active language implies the respondent did it.
It does not. And the reason it does not is more interesting than a simple denial.
There is a clear distinction between describing an allegation with clarity and making a finding of fact. Saying the respondent allegedly did something determines nothing. It maintains clarity about who is said to have done what.
But here is the part that is almost always missed.
A vague allegation cannot be answered.
"You are alleged to have been involved in an incident" gives a respondent nothing to respond to. There is no act to accept, deny, contextualise or explain. It invites a defence built on atmosphere rather than fact.
"You are alleged to have placed your hand on the complainant's waist, in the stockroom, on 3 March" can be answered. The respondent knows exactly what is being put to them. They can accept it, deny it, or explain it. They can bring evidence.
Passive, agentless language is not a courtesy to the accused. It is an obstacle to their defence, and it is a gift to any solicitor later arguing that the process was unfair.
Active language serves both parties. It also produces a better report.
The rigour argument, which is the one that should worry you
We have talked about how passive language affects the people in the investigation. There is a harder point to make about what it does to the investigator.
Passive construction is where an investigator hides from a decision.
"Was allegedly touched" allows you to write an entire paragraph without ever committing to who is said to have done what. It feels like caution. It is frequently avoidance. And an investigator who has not been forced by their own sentences to name an actor and an action has not yet done the analytical work the case requires.
Ambiguity in the writing is very often ambiguity in the thinking. It travels forward into the findings, into the rationale, and into whatever a tribunal makes of it two years later.
Active language is not a stylistic preference. It is a discipline that forces clarity of thought at the exact point where clarity is most difficult and most necessary.
The report is the artefact. Treat it accordingly.
In HR, trauma-informed practice is usually discussed in terms of behaviour. How we conduct interviews. How we respond to distress. How we create space and offer choice.
All essential. And all of it can be undone by the document that follows.
The written output of an investigation is rarely examined through a trauma-informed lens, and yet for the people involved it becomes the lasting record of what happened to them.
It is what decision-makers read. It is what may be revisited in a grievance, an appeal, or a tribunal. It is what defines, formally and permanently, how their experience is understood by their employer.
If that document consistently strips agency from the alleged perpetrator, it does more than describe the process badly. It quietly reshapes what the process concluded.
Resisting re-traumatisation is one of the core principles of trauma-informed practice. It does not stop when the interview ends. It applies to every sentence you write afterwards, in a document the person may read many times, for years.
Practical shifts you can make immediately
None of this requires overhauling your investigation framework. It requires intentional language.
| Instead of | Write |
|---|---|
| "The complainant was subjected to unwanted contact" | "The respondent is alleged to have made unwanted contact with the complainant" |
| "She experienced inappropriate comments" | "The respondent allegedly made inappropriate comments to the complainant" |
| "It is alleged that she was touched" | "The respondent allegedly touched the complainant" |
| "An incident occurred on 3 March" | "The respondent allegedly [specific act] on 3 March" |
Three rules that cover almost every case:
- Every allegation names an actor and an action. If your sentence has an action but no actor, rewrite it.
- Keep alleged doing its job. It carries the whole legal weight. You do not need passive voice as well.
- Be specific enough that the respondent could answer it. If they could not, the allegation is not yet properly framed.
These are small shifts. They reduce ambiguity, reduce the risk of unintentional victim blaming, and produce a more precise record.
They also protect the organisation. Investigations that feel unclear or invalidating are more likely to be challenged, more likely to escalate, and more likely to erode trust in HR and in leadership. Processes that demonstrate clarity, consistency and awareness of impact strengthen the credibility of their own findings and reduce legal and reputational exposure.
This is not overcorrection. It is doing the work well.
Neutrality is not the absence of an actor
Neutrality is the gold standard in investigations, and rightly so.
But neutrality does not mean removing responsibility from the sentence. It does not mean writing so carefully that nobody appears to have done anything at all.
Active language does not compromise independence. It strengthens it.
Clarity is not bias. And accountability, even when it is only alleged, should never quietly disappear from the narrative.
Where this sits in practice
We teach the subtle power of language in this kind of casework as part of our Investigative Practice for Sexual Misconduct course, alongside trauma-informed record keeping and note taking. It is also central to Victimisation and Victim Blaming, where we examine the psychological theories that drive these instincts in all of us.
Where organisations need the work done rather than taught, we conduct independent investigations into sexual harassment, assault and misconduct.
Support
If anything in this article has affected you, Rape Crisis England and Wales run a free 24/7 support line on 0808 500 2222.
Talk to us
If you are handling a complex or sensitive investigation, or you want your team to write reports that hold up, we would welcome the conversation.
Email: help@onesourcehr.co.uk Telephone: 01709 460500