Police procedures are usually written in relatively calm conditions. They have to be. Legislation, policy and operational guidelines need language that can be understood, taught, applied and later examined. Police officers exercise significant powers, and both the public and officers themselves benefit from clear boundaries around how those powers should be used.
The situations those rules eventually encounter are considerably less orderly.
An officer may arrive to find two people giving incompatible accounts, an intoxicated person whose behaviour is changing, a frightened child who will not speak, a witness who appears credible but later proves unreliable, or a person whose anger may represent either understandable frustration or the beginning of something dangerous. Information arrives unevenly. Some of it is wrong. Important information may not arrive at all.
The officer still has to do something.
This is one of the central tensions of policing. The organisation necessarily tries to create clarity around how officers should act, while officers themselves work in a human environment that repeatedly resists being made clear.
Rules can define boundaries. They can establish principles, responsibilities and expectations. What they cannot do is completely remove the need for judgement, because the grey area is often not in knowing the rule.
It is in deciding what kind of situation you are actually standing in.
The difficult part is often deciding which rule applies
A procedure can be perfectly clear once the relevant facts have been established. The problem is that police officers frequently have to act before that point.
A policy may describe what should happen when a threat reaches a particular level, when there are reasonable grounds for a particular action, when someone is considered vulnerable, or when certain risks are present. The officer must still interpret human behaviour and incomplete information to determine whether those conditions exist.
That is a very different task from simply remembering procedure.
A person who repeatedly reaches into a pocket may be nervous, distracted, intoxicated, concealing something or reaching for something dangerous. Someone refusing to follow instructions may be deliberately obstructive, frightened, confused or struggling to understand what is being asked of them. A person making an allegation may be entirely truthful while appearing inconsistent, or highly convincing while being untruthful.
Real situations rarely announce which interpretation is correct.
Experienced officers often become skilled at making these distinctions because they have encountered thousands of variations of human behaviour. They notice tone, context, changes in demeanour and details that are difficult to capture in a policy document. That experience can improve judgement considerably.
It cannot create certainty where certainty does not exist.
Professional judgement is therefore not an unfortunate gap left behind when procedures fail to be detailed enough. It is one of the things required to make procedures usable in the real world.
If every relevant feature of every police encounter could be specified beforehand, discretion would barely be necessary.
It is necessary precisely because they cannot.
Policing happens while the story is still being written
The uncertainty surrounding a police decision is easiest to appreciate while an incident is unfolding and easiest to forget once it has ended.
At the time, an officer may have partial information from communications, conflicting accounts from the people involved, limited opportunity to observe what occurred beforehand and no way of knowing with certainty what anyone will do next. Several interpretations may remain plausible simultaneously.
Then time passes.
Statements are taken. CCTV is found. Body-worn video is reviewed. Records are checked. Someone who seemed peripheral becomes important. Information that was unavailable at the scene becomes central to understanding what happened.
Eventually, there is something resembling a complete story.
The officer did not make the original decision inside that story.
They made it several chapters earlier.
That difference is psychologically important because once human beings know an outcome, it becomes remarkably difficult to reconstruct what genuine uncertainty felt like before it occurred. Psychologists describe this as hindsight bias. Knowledge of the ending changes the way we interpret the events leading towards it.
A person who subsequently becomes violent can appear, retrospectively, to have been showing obvious warning signs. If no violence occurs, identical behaviour may later look insufficient to have justified concern. Once an allegation is substantiated, the details supporting it can appear more conspicuous. If it proves false, inconsistencies that were ambiguous initially can suddenly seem glaring.
The ending reorganises the story.
Without deliberately guarding against that tendency, retrospective review can quietly turn something that was merely possible into something that appears to have been predictable.
The camera sees differently from the officer
Body-worn video has transformed the capacity to examine police encounters. It can provide valuable evidence, resolve disputes, identify inappropriate conduct and protect officers against allegations that are inconsistent with what actually occurred.
But a recording and an experience are not the same thing.
An officer experiences an incident continuously while processing several streams of information at once. They may be listening to a person in front of them while monitoring somebody nearby, hearing information over the radio, assessing the physical environment and thinking about what needs to happen next.
The reviewer has a different experience.
The footage can be paused. A particular movement can be watched repeatedly. Attention can be directed towards one person while temporarily ignoring everything else occurring in the frame. A sentence that disappeared into the complexity of the original interaction can be replayed until its significance becomes obvious.
The camera may have captured the detail perfectly.
That does not mean the officer experienced the detail with the same prominence.
This distinction does not diminish the evidentiary value of body-worn video. It helps us use it more intelligently. A recording can tell us what was visible or audible from a particular perspective; it cannot, by itself, reproduce the attentional demands of being the person making decisions inside the event.
The ability to examine four seconds for several minutes should improve our understanding of those four seconds, not create the illusion that the officer originally had several minutes in which to understand them.
A bad outcome and a bad decision are not the same thing
When an incident ends badly, there is an understandable desire to identify the point at which the outcome could have been prevented. Sometimes that process exposes poor judgement, inadequate preparation, a departure from procedure or a decision that genuinely should not have been made.
Accountability requires that possibility to remain open.
But outcome and decision quality are not interchangeable.
A reasonable decision can produce a bad outcome because risk can never be eliminated and human behaviour cannot be predicted perfectly. Conversely, a poor decision can occasionally produce an acceptable outcome because circumstances happen to be forgiving.
This distinction is familiar in many professions where decisions are made under uncertainty. A good decision is not one that guarantees a good outcome; no professional can meet that standard. It is one that was reasonable given the information, options, risks and constraints that existed when the decision was made.
Judging only from the outcome risks rewarding good fortune and condemning bad fortune.
That does not mean outcomes are irrelevant. A serious outcome may reveal something important about training, systems, policy or individual decision-making. It should prompt examination.
The crucial point is that the quality of a decision has to be evaluated from the decision point, not solely from the endpoint.
Discretion creates an unavoidable tension
Police organisations reasonably want consistency. Similar circumstances should not produce radically different responses simply because a different officer attends.
At the same time, good policing requires responsiveness to context.
Two family violence incidents may share the same broad classification while being substantially different in risk. Two intoxicated people behaving similarly may require different responses because of the environment, their history or the people around them. Two apparently minor disturbances may be heading in completely different directions.
Experienced policing involves recognising these differences rather than treating every superficially similar event as interchangeable.
That creates an unavoidable tension.
The more tightly an organisation specifies what must happen, the greater the consistency it may achieve, but the less room officers have to respond to unusual circumstances. The more discretion it allows, the greater the capacity for nuanced decision-making, but also the greater the possibility of variation, error and disagreement.
There is no policy sophisticated enough to make that tension disappear entirely.
It is inherent in trying to apply general rules to individual human situations.
This is why professional discretion should not be understood as freedom from rules. It is the reasoning required to apply rules when reality does not arrive in standard form.
The psychological attraction of certainty
There is another reason black-and-white explanations become appealing after difficult events.
Ambiguity is uncomfortable.
When something serious has happened, knowing that there was one identifiable mistake can sometimes feel psychologically easier than accepting that several reasonable decisions were available and none could guarantee the outcome.
A clear error restores a sense of order. It allows the event to be understood as preventable: if this had been done differently, that would not have happened.
Sometimes that conclusion is correct.
But there are also incidents where uncertainty was genuine, several outcomes were possible and the consequences could not have been completely controlled by the officer involved.
That can be harder for organisations to hold, and it can be harder for officers themselves.
Policing attracts and develops people who are accustomed to taking responsibility. When something goes wrong, that orientation can turn inward very quickly. An officer may replay the event repeatedly, identifying moments that now seem significant and imagining alternatives that feel increasingly obvious with each reconstruction.
Reflection is an essential part of professional development. It allows officers to recognise mistakes, improve judgement and consider what they might do differently in the future.
The problem begins when learning from the past becomes judging the past with knowledge that did not exist at the time.
An officer who now knows exactly which detail mattered can become highly critical of the earlier version of themselves who encountered that detail among dozens of others.
The psychologically fairer reconstruction preserves both perspectives: what is known now and what was reasonably knowable then.
Procedures cannot contain every situation
There is a natural organisational response when an unusual incident exposes uncertainty: clarify the procedure.
Often that is appropriate. Incidents reveal gaps, policies improve and organisations learn. Clearer guidance can prevent future errors.
But there is a limit to how much uncertainty can be designed out of human encounters.
A procedure necessarily reduces complexity. It identifies important variables, establishes boundaries and creates a framework for action. In doing so, it makes the world manageable enough for consistent professional practice.
The real world then adds the complexity back in.
People behave unexpectedly. Circumstances overlap. Risks compete. Information changes. A response that solves one problem creates another. Two legitimate principles point towards different actions.
This is not unique to policing. Medicine has clinical guidelines but still requires clinical judgement. Law has legislation and precedent but still requires interpretation. Psychology has professional standards but still requires practitioners to apply them to people whose lives rarely resemble textbook examples.
Professional systems need rules because judgement alone is insufficient, and they need judgement because rules alone are insufficient. Good practice exists in the relationship between the two.
Accountability should preserve the grey
None of this is an argument for lowering standards or protecting police decisions from scrutiny.
The opposite is true.
Meaningful accountability requires enough sophistication to distinguish between different kinds of failure. Deliberate misconduct is not the same as poor judgement. Poor judgement is not the same as an understandable error. An understandable error is not the same as a reasonable decision followed by an undesirable outcome.
Collapsing these distinctions does not produce stronger accountability. It produces less accurate accountability.
The same principle applies to organisational learning. If every adverse outcome is explained as an individual failure to follow the correct procedure, an organisation may miss the more useful lesson: that the procedure was difficult to apply, the available information was inadequate, competing expectations existed, or officers were being asked to exercise discretion without sufficient support for doing so.
Sometimes the officer needs to learn.
Sometimes the organisation does.
Sometimes both do.
And sometimes careful examination concludes that an undesirable outcome occurred despite a decision that was reasonable in the circumstances.
A mature system has to be capable of reaching that conclusion too.
The grey is part of the job
Police officers work in an unusual position between abstract rules and complicated human reality.
They are expected to be consistent without treating every situation as identical, decisive without pretending to possess certainty, and accountable for their judgement without knowing what will happen next. They must learn procedures thoroughly enough to act within them while developing enough professional judgement to recognise how those procedures apply when circumstances become complicated.
None of this means the rules are the problem.
The rules are essential.
But rules can only ever describe the world in categories, while officers encounter it as individual events involving particular people, incomplete information and circumstances that may never occur in exactly the same combination again.
That is where judgement lives.
The fairest examination of police decision-making therefore preserves the conditions under which the decision was actually made. It can use everything learned afterwards to understand the event and improve future practice without quietly giving that later knowledge to the officer retrospectively.
There will always be value in asking what could have been done better. There is equal value in remembering what was genuinely unclear.
Because the difficult part of professional judgement is rarely applying a rule once everything is known.
It is deciding what to do before everything is known.
