How evidence is selected, interpreted and validated — and what two very different military accountability processes reveal about institutional scrutiny.Introduction — The Authority of an Investigative ConclusionAn investigation can fail without anyone destroying evidence, falsifying a document or expressly ordering a cover-up.
Sometimes the failure is procedural. A photograph is not examined. A witness's words are misunderstood. A contemporaneous record is considered unnecessary. A question about police conduct becomes a question about property rights. An independent reviewer accepts the resulting conclusion because, on the evidence that was considered, it appears reasonable.
Eventually, the institution announces an outcome.
What remains less visible is the process by which that outcome became possible.
On 11 November 2026, legal commentator Joshua Rozenberg examined evidence emerging from the Afghanistan Inquiry concerning
Operation Northmoor, the Royal Military Police investigation into allegations of unlawful killings by British special forces in Afghanistan.
His account raises troubling questions about the relationship between an investigation, the evidence it chooses to examine, the advice of independent reviewers and the institutional decision to bring proceedings to an end.
I have been examining a different process: the investigation of my complaint concerning RAF Police conduct at Northwood Headquarters in August 2025, the subsequent statutory review by the Service Police Complaints Commissioner, and a further complaint about the methodology employed by the original investigator.
The underlying allegations are fundamentally different. One concerns possible unlawful killings during military operations. The other concerns the treatment of a civilian at a military establishment and the conduct of military police.
They should not be equated in gravity, legal framework or consequences.
But their investigative structures invite comparison.
Both raise questions about how evidence is selected, how institutional accounts are tested, what independent review actually examines and whether an oversight process can validate an outcome without adequately interrogating the methodology that produced it.
The subject of this article is therefore not the guilt or innocence of any soldier or police officer.
It is the architecture of investigative accountability.
1. Afghanistan: When an Investigation Ends Before the Evidence Is TestedThe allegations and Operation NorthmoorOperation Northmoor was established to investigate allegations of unlawful killings by British forces in Afghanistan.
The allegations now being examined by the Afghanistan Inquiry include claims that Afghan men were killed in circumstances inconsistent with the accounts subsequently provided of military operations.
Some allegations concern the treatment of men of fighting age and whether individuals were assessed as threats on the basis of their actual conduct or broader assumptions about their status.
These are allegations, not established findings of unlawful killing.
Rozenberg's commentary focuses on evidence given by Lieutenant Colonel John Harvey, who led Operation Northmoor, about the investigation's handling of material potentially relevant to determining what had occurred.
Particularly significant were photographs of deceased Afghan men, including images that appeared to raise questions about the circumstances of their deaths.
The photographs had potential evidential value in either direction.
Expert examination might have identified features inconsistent with the accounts of armed encounters. Equally, it might have supported those accounts or demonstrated that the images could not reliably establish the circumstances of death.
That is precisely why forensic examination mattered.
The investigative task was not to establish that the soldiers were guilty. It was to test the available evidence sufficiently to determine whether criminal allegations could be substantiated or excluded.
The investigation: evidence that was not examinedAccording to Rozenberg's account of Harvey's testimony, the photographs were never submitted for the relevant expert examination.
Harvey accepted that such examination might have clarified the injuries and potentially exonerated the soldiers.
The reported justification for not obtaining that evidence came, at least in part, from an independent review team, which advised against pursuing the examination.
This creates a methodological difficulty.
An investigator may reasonably decide that a proposed enquiry is disproportionate, speculative or incapable of producing material evidence. Criminal investigations are not required to pursue every imaginable possibility.
But that judgment requires an evidential foundation.
Where photographs appear capable of resolving a material factual dispute, a decision that expert examination would be futile requires particular scrutiny.
Otherwise, the investigation risks reaching a conclusion about the usefulness of evidence before obtaining the expertise necessary to evaluate it.
This can be described as
evidential foreclosure: a potentially relevant avenue of enquiry is closed before its evidential value has been properly established.
The resulting absence of evidence is then liable to be treated as an objective feature of the case rather than a consequence of the investigation's own decisions.
"Independent review" and investigative judgmentThe second issue concerns the role of the independent review team.
Independent oversight is ordinarily presented as a safeguard against investigative error.
External reviewers may identify overlooked evidence, challenge assumptions, recommend further enquiries and test whether conclusions are properly supported.
But independence is not established merely by appointing a separate reviewing body.
It must be demonstrated through the review's actual function.
Rozenberg reports Harvey's evidence that he accepted conclusions he considered wrong because the review team had the support of his superior, the Provost Marshal.
The account also raises questions about communications between the Provost Marshal and senior special forces leadership while the investigation was ongoing.
These matters do not, without further evidence, establish improper interference or deliberate suppression.
They do raise an important structural question:
was the independent review team testing the investigation, or was its authority being used to constrain the investigators' professional judgment?The distinction is fundamental.
A review can improve investigative independence by challenging the original investigators. It can also undermine that independence if its recommendations become effectively determinative without adequate evidential justification.
The relevant measure is functional independence, not institutional description.
Investigation plus review: the outcomeOperation Northmoor closed in 2019 without prosecution referrals.
Rozenberg reports that Harvey accepted that reasonable and proportionate lines of enquiry had remained outstanding.
If that evidence is accepted, the problem is not simply that an investigation ended without charges.
Many proper criminal investigations end without prosecution because the evidence does not justify further action.
The problem is whether the evidence was sufficiently investigated before the decision was made.
A failure to prosecute is not itself evidence of investigative misconduct.
But a decision not to prosecute, reached after potentially material evidence has been left unexamined, presents a different question.
Was the evidential insufficiency unavoidable, or did the investigative process contribute to producing it?
The legal dimensionThe potential legal implications extend beyond investigative best practice.
Where applicable, Article 2 of the European Convention on Human Rights imposes a procedural obligation to conduct an effective investigation into deaths potentially involving unlawful state action.
An effective investigation must be capable of establishing the relevant circumstances and identifying those responsible where wrongdoing occurred. Independence and the pursuit of reasonable investigative steps are central considerations.
The application of Article 2 to particular deaths during overseas military operations requires a separate jurisdictional analysis. It cannot simply be assumed.
International humanitarian law and the relevant domestic criminal law may also impose duties concerning the investigation of suspected war crimes.
However, a deficient investigation is not automatically itself a war crime, and Rozenberg's account does not establish the criminal liability of any individual.
The narrower and more defensible conclusion is that the reported testimony raises substantial questions about investigative adequacy, independence and the evidential basis for closure.
It also exposes a problem of wider institutional importance:
What happens when insufficient evidence becomes the justification for closing an investigation, without first establishing whether that insufficiency resulted from investigative omissions?2. Northwood: When an Investigation Becomes the Subject of a ComplaintA civilian complaint in a military environmentOn 26 August 2025, I attended Northwood Headquarters, where an access restriction concerning me was already recorded within the military security system.
Access was refused.
The lawfulness of that restriction is the subject of separate proceedings and is not the principal concern of this article.
The relevant events began after the refusal, when RAF Police became involved.
My complaint concerned the officers' subsequent conduct, including the explanation of their authority, the manner in which they dealt with me, the proportionality of their actions, potential discriminatory treatment and the information communicated during the escalation to Hertfordshire Police.
The legal basis for refusing entry to a military establishment does not necessarily determine every question concerning the conduct of police personnel after entry has been refused.
A lawful access decision would not, by itself, establish that every subsequent police communication was accurate, necessary or proportionate.
The original investigation: what question was answered?The complaint was investigated by DSCC Professional Standards, with WO2 Watts conducting the original investigation.
The outcome, issued on 20 January 2026, did not uphold the agreed heads of complaint.
I subsequently challenged the investigative methodology.
My principal concern was that the investigation had increasingly treated the underlying access restriction, property rights, trespass and related matters as determinative of complaints about subsequent police conduct.
This is an allegation of
issue substitution.
An investigation may be factually accurate about one issue and nevertheless fail to resolve another.
For example, establishing that a person had no permission to enter a military establishment may justify an instruction to leave. It does not necessarily answer whether an officer accurately described that person's status to civilian police, whether criminal suspicion was properly represented, or whether the conduct complained of was proportionate.
The proper investigative question must therefore be defined before the evidence is assessed.
Otherwise, the investigator risks constructing a persuasive answer to a question that was not the substance of the complaint.
Evidence attribution: when a complainant's words change meaningA second concern involved the interpretation of my own representations.
The statutory review subsequently acknowledged two misunderstandings.
In one instance, the original investigation understood a reference to somebody having lied as an admission
that I had lied.
In another, a statement recognising the officers' role in implementing an existing access restriction was treated as a broader concession concerning the lawfulness of their conduct.
These distinctions are material to investigative method.
A complainant's statement is evidence of what the complainant said. An investigator's interpretation of that statement is a separate proposition.
When the two are conflated, the investigative record can begin to contain apparent admissions that were never actually made.
This does not necessarily involve deliberate falsification.
An investigator may misunderstand ambiguous language in good faith. I do not believe she misunderstood as I have the transcript of our interview, which clearly records what I have actually said. If she misinterpreted what I have actually said, it raises the question of her professional suitability for the role of an investigator. But let's proceed further with the analysis.
But the absence of dishonesty does not establish the absence of evidential consequences.
A genuine misunderstanding can still affect an assessment of credibility, consistency or the merits of a complaint.
The proper question is not confined to whether the investigator was biased.
It is whether the factual conclusions remained reliable after the attribution errors were identified.
Review Wye: acknowledgement without reversalOn 30 April 2026, Jonathan Rees KC, acting as Service Police Complaints Commissioner, issued
Review Wye under Regulation 64.
His professional background is particularly relevant to the questions examined in this article. Before assuming the Commissioner's role in November 2025, Rees had served for five years as Director of Service Prosecutions, heading the Service Prosecuting Authority responsible for prosecuting criminal offences within the UK Armed Forces, including serious offences arising from military operations. He was therefore no stranger to the evidential standards, investigative responsibilities and institutional independence required within the military justice system. Indeed, in March 2026, giving evidence to Parliament, he drew expressly on his experience as Director of Service Prosecutions when discussing the treatment of victims and the effectiveness of Service Police investigations. His appointment as Commissioner placed him in a different institutional position: no longer responsible for prosecutorial decisions, but for independent oversight of the police complaints system. This makes his approach to
Review Wye especially significant. The question is not whether he possessed sufficient professional experience to recognise investigative defects, but how that experience was applied when assessing an investigation in which factual misunderstandings were acknowledged and additional evidential enquiries were considered unnecessary.
The review considered whether the original investigation had been fair, balanced and sufficiently thorough, and whether the complaint outcomes were reasonable and proportionate.
It was not a fresh investigation.
In paragraphs 61–63, Rees acknowledged that the original investigators appeared to have misunderstood the two representations.
He considered the misunderstandings genuine, found that they did not provide cogent evidence of bias, and concluded that they had not affected the complaint outcomes.
This is an important qualification.
The review did not ignore the errors. It expressly addressed them and found them immaterial.
The unresolved methodological question is whether the explanation adequately demonstrated that the original conclusions were unaffected.
An oversight body can reasonably conclude that an error is harmless. But the strength of that conclusion depends on an analysis of the error's relationship to the original reasoning.
Merely identifying an error as innocent does not determine whether it was material.
2.5 Contemporaneous records and the limits of enquiry
Review Wye also considered my criticism that the original investigation had not obtained relevant contemporaneous operational records.
These included potential records from the Visitor Management System, NPSS and associated incident-reporting systems.
At paragraph 67, Rees considered that the existing witness statements provided clear evidence concerning the relevant security notifications and police interaction.
He questioned what additional records would have contributed, given the statements already available and the substantial agreement about the events.
That is a legitimate question for a reviewer to ask.
But it also reveals a methodological defect.
Witness statements and contemporaneous system records do not necessarily perform the same evidential function.
A witness may accurately recall an interaction while being mistaken about the origin, timing or wording of the information on which they acted.
Operational records may establish matters that recollection cannot: when a notification was created, what instruction it contained, who entered it, and what information was available at a particular moment.
The existence of witness statements therefore does not, without further analysis, establish that independent records would have been redundant.
The question is whether the omitted material was reasonably capable of resolving a disputed issue within the complaint.
The civilian-police briefingOne particularly significant aspect of Review Wye is its account of what RAF Police communicated to Hertfordshire Police.
At paragraph 37(xliii), the review records body-worn-video evidence that the civilian officers were told, among other things, that Commander Pastor and I were under investigation.
The account also includes information about the access restriction, the circumstances of our attempted visit and the status of Northwood as a prohibited place.
These statements matter because they were not merely internal assessments.
They were operational communications from military police
to civilian police.
A claim that someone is under investigation may carry implications materially different from a statement that the person is subject to an administrative access restriction.
The relevant questions include what investigation was being referred to, whether it concerned both individuals, what evidence supported the description, and whether the information was accurately communicated.
Review Wye correctly distinguished the conduct of Hertfordshire Police from the subject of its own review.
But that jurisdictional distinction does not automatically remove RAF Police communications
to Hertfordshire Police from consideration.
The issue is whether those communications were within the scope of the original complaint and, if so, whether their accuracy and justification were adequately investigated.
The discriminatory contextReview Wye also records that the female RAF Police officer knew, before the incident, of rumours circulating within MARCOM that I was believed to be a Russian spy.
The officer denied that this knowledge had caused her to act with bias.
At paragraph 80, Rees recognised that the original complaint response was inaccurate insofar as it suggested that the officer had not known of the rumour.
He nevertheless found insufficient evidence to uphold the discrimination complaint.
The existence of a rumour does not prove discriminatory treatment.
Nor does the officer's prior knowledge establish that the rumour influenced her conduct.
But the investigative question is whether the potential influence of such information was tested against all available circumstantial evidence.
Discriminatory assumptions may operate without an explicit instruction and without a decision-maker consciously recognising their influence.
That possibility must be evaluated objectively, while avoiding the opposite error of treating awareness of a stereotype as proof of discriminatory motivation.
The wider documentary record also includes the Ministry of Defence's position, in separate judicial-review proceedings, that my place of birth was relevant to the access decision.
Place of birth is not itself a separately enumerated protected characteristic under section 9 of the Equality Act 2010, although it may be relevant evidence in a claim concerning nationality or national origins.
These matters raise questions about institutional assumptions and information provenance. They do not, individually or collectively, establish that the RAF Police officers discriminated against me.
Investigation plus review: the outcomeReview Wye upheld the original complaint outcome.
I subsequently brought judicial-review proceedings challenging the lawfulness of the statutory review.
A further complaint, SPCC 2026/129, concerns the methodology of the original investigator, WO2 Watts. It is distinct from SPCC 2026/128, which concerns the original RAF Police officers.
The Watts complaint raises allegations about evidence attribution, evidential handling, the adequacy of enquiries, police-facing information and issue substitution.
In September 2026, the SPCC considered the further complaints under Regulation 16 and determined that investigation by the Commissioner was unnecessary, referring the matters onward under the statutory process.
The available correspondence also raises questions about the weight given to Review Wye and the SPCC's position in the pending judicial review.
A decision that Commissioner-led investigation is unnecessary does not establish that no other investigation or complaint handling can occur.
Nor does reliance on an earlier review automatically constitute unlawful predetermination.
But the sequence presents a further accountability question:
When the methodology of an investigator becomes the subject of a new complaint, how independently can that methodology be assessed if the earlier review upholding the investigation is itself relied upon in deciding how the new complaint should be handled?That question takes us beyond the merits of a single complaint.
It concerns the capacity of an oversight system to examine its own previous conclusions.
Two Investigations, One Institutional Problem?What can legitimately be compared?Operation Northmoor and the Northwood complaints process are not legally interchangeable.
Northmoor was a criminal investigation into allegations of unlawful killing during armed conflict.
The Northwood process concerned police conduct towards a civilian and the subsequent statutory review of a professional-standards investigation.
Different legal obligations, investigative powers, evidential thresholds and institutional arrangements apply.
Nevertheless, both processes expose potential vulnerabilities in the relationship between investigation and oversight.
The comparison is not between the alleged harms.
It is between the methods by which institutions establish facts and decide whether their own personnel have acted improperly.
Evidential foreclosureIn Northmoor, the reported concern is that potentially significant photographs were not subjected to expert examination before the investigation ended.
In Northwood, the concern is that contemporaneous operational records were regarded as unnecessary because witness statements were already available.
The evidential questions differ, but the methodological issue is comparable.
An investigation must determine not only what the available evidence establishes, but whether additional reasonable enquiries could materially alter that assessment.
Where the decision not to investigate further rests on assumptions about what the missing evidence would show, the process risks becoming circular.
Evidence is not obtained because it is considered unnecessary.
It is then unavailable to challenge the conclusion that it was unnecessary.
Evidential transformation: when an investigation generates its own factual premisesA second and potentially more subtle problem arises when the investigative process changes the meaning of the evidence it receives.
In the Northwood investigation, two of my representations were misunderstood.
The resulting interpretations were incorporated into the official investigative account before the statutory reviewer acknowledged the errors.
This illustrates how an evidential premise can arise within the investigation itself.
The process begins with a primary statement.
An investigator interprets that statement.
The interpretation becomes part of an official record.
The official record then informs the assessment of the complaint.
Finally, an independent reviewer considers the investigation and determines whether the outcome should stand.
At each stage, the proposition may acquire additional institutional authority.
Yet the original source remains unchanged.
The problem is not necessarily that an investigator knowingly invents a fact.
It is that the distinction between source evidence and investigative inference can disappear as information moves through successive institutional processes.
This phenomenon may be described as
investigator-generated factual premises.
The term identifies a methodological risk, not a finding of fabrication.
It is especially significant where an apparent admission, concession or contradiction influences an assessment of credibility.
A later reviewer may identify the original error and still conclude that the outcome was unaffected.
That may be correct. Somehow I do not think so. And I definitely did not see the evidence nor reasoning supporting that conclusion.
But it should be demonstrated through an examination of the reasoning, rather than assumed from the apparent good faith of the investigator.
"Independent review" or institutional validation?The third common issue concerns what independent review actually accomplishes.
In Northmoor, according to the reported testimony, independent reviewers advised against an investigative step that might have produced important forensic evidence.
In Northwood, the statutory reviewer endorsed the original investigation despite acknowledging certain errors and rejecting the need for further enquiries.
These are not identical decisions.
But both raise the question of how a reviewer evaluates the evidential foundation of an investigation.
There is a fundamental difference between reviewing whether a conclusion appears reasonable on the existing record and examining whether the record was sufficiently complete and reliable to justify that conclusion.
The former tests the coherence of an outcome.
The latter tests the process by which the outcome was produced.
A rigorous oversight mechanism may need to do both, within the limits of its statutory function.
Otherwise, a review can become a form of institutional validation: a process that gives additional authority to an outcome without sufficiently examining whether the underlying investigation was capable of producing a reliable answer.
When the review becomes part of the problemThe subsequent complaint against WO2 Watts introduces another dimension.
An original investigation is completed.
A statutory review upholds its outcome.
The review itself becomes the subject of judicial proceedings.
A new complaint then challenges the methodology of the original investigator.
The institution must decide how to handle that complaint while its earlier review remains contested.
This creates a potential feedback problem.
If the previous review is treated as substantially resolving the very methodological questions now complained of, the new complaint may struggle to obtain an assessment independent of the first cycle.
Conversely, the institution may legitimately consider the earlier findings while ensuring that genuinely new allegations and evidence are examined on their merits.
The legal question is not whether an institution must disregard everything it has previously decided.
It is whether its procedures preserve a meaningful opportunity to identify and correct error.
An effective accountability system should be capable of saying not only that its original decision was justified, but also that it was wrong.
The legal significance of investigative methodologyThese questions engage different legal standards in the two cases.
In the Afghanistan context, the issues may concern the procedural duty to investigate potentially unlawful deaths, criminal investigative standards and obligations under international humanitarian law.
In the Northwood context, the questions concern the statutory complaints framework, the proper scope of the Regulation 64 review, relevant considerations, rationality, procedural fairness and the adequacy of reasons.
An investigative omission does not automatically amount to a legal error.
A reviewer may lawfully conclude that a particular enquiry would add nothing material.
A factual misunderstanding may be harmless.
An earlier decision may legitimately inform a later one.
But those conclusions require defensible reasoning.
The common legal concern is whether the relevant decision-maker applied the correct function to a sufficiently reliable evidential record.
The common methodological concern is whether the investigation's own choices prevented the institution from recognising facts that might have altered the outcome.
The Culture of Investigation: From War Zones to Civilian LifeMilitary accountability is often discussed in the context of armed conflict.
War crimes investigations attract public attention because the underlying allegations involve the gravest possible consequences of state power.
But the institutional habits that shape an investigation are not confined to a battlefield.
They concern ordinary questions of professional judgment: which witness is believed, which document is requested, which inconsistency is examined, which allegation is considered relevant, and how a reviewing authority responds when errors become apparent.
Those habits also matter when military institutions exercise authority over civilians.
Institutional culture is visible in ordinary proceduresA military institution's commitment to legality should not be assessed exclusively by reference to its conduct during armed conflict.
It should also be visible in the way it handles complaints, explains decisions, maintains accurate records and responds to evidence that challenges its own personnel.
The point is not that a defective civilian complaints investigation predicts unlawful conduct in war.
No such causal relationship is established by the two cases examined here.
The more modest proposition is that investigative integrity is an institutional capability.
The ability to test inconvenient evidence, distinguish fact from inference and correct internal error should be present wherever an institution exercises coercive or administrative power.
An organisation does not require a war zone to demonstrate whether it possesses those capabilities.
The danger of institutional labelsA related issue concerns the way institutions classify individuals.
In the Afghanistan allegations, questions arise about whether men of fighting age were treated as threats without adequate individualised assessment.
At Northwood, the documentary record raises questions about the relationship between place of birth, rumours concerning supposed espionage, security classifications and the subsequent description of a civilian to police officers.
The circumstances and legal questions are entirely different.
Neither comparison establishes that racial prejudice caused the alleged Afghan killings or the Northwood incident.
But both invite scrutiny of the process by which an assumption about a person's identity or status can become operationally significant.
A classification may begin as intelligence, hearsay, an administrative entry or an investigator's interpretation.
Once recorded within an institutional system, it may influence subsequent decisions.
Other officials may then treat the existence of the record as evidence that the underlying classification was justified.
This creates a risk of information acquiring authority through repetition rather than verification.
The essential safeguard is traceability.
Who made the original assertion? What evidence supported it? What degree of confidence was attached to it? Who subsequently relied upon it? Was it ever independently tested?
These are not merely data-management questions.
They are questions about procedural justice.
Independence is a method, not a labelThe term
independent review carries considerable institutional authority.
It suggests distance from the original decision-maker and the possibility of impartial correction.
But independence cannot be established solely by organisational separation or professional seniority.
A reviewer must also be willing and able to question the evidential assumptions on which the original investigation rests.
That does not mean every review must become a reinvestigation.
It means the scope of review must be adequate to its purpose.
If the purpose includes assessing whether an investigation was sufficiently thorough, the reviewer must be capable of recognising when the absence of evidence reflects an omitted enquiry rather than the absence of relevant facts.
If the purpose includes evaluating the reasonableness of an outcome, the reviewer must distinguish between a conclusion supported by reliable evidence and a conclusion that appears coherent only because contrary evidence was never examined.
And where an error is acknowledged, the institution should be able to explain why the error was or was not material.
These are elementary requirements of reliable institutional reasoning.
They are also the foundation of public trust.
Who investigates the investigators?The Afghanistan Inquiry provides an opportunity to examine the decisions made during Operation Northmoor, including the role of external reviewers and the reasons why particular evidence was not tested.
The continuing Northwood complaints process raises a different but related question: whether a specialist military complaints system can meaningfully examine allegations about the methodology of an investigation whose outcome it has already upheld.
Neither process should be prejudged.
The value of independent scrutiny lies precisely in its ability to establish what occurred, including where an allegation of institutional failure proves unfounded.
But an accountability system must remain capable of reaching an uncomfortable conclusion about its own earlier work.
Otherwise, each successive procedural layer risks adding authority without adding knowledge.
An investigation produces a finding.
A review endorses the finding.
A later decision relies on the review.
Over time, the original evidential weakness may become increasingly difficult to identify because it is surrounded by apparently authoritative decisions.
That is not necessarily a conspiracy. It is worse.
It is an ordinary consequence of institutional processes that are better at preserving conclusions than revisiting their factual foundations.
And that is precisely why investigative methodology deserves public scrutiny.
Conclusion — The Question Behind the OutcomeThe most important question raised by Rozenberg's account of Operation Northmoor is not simply why the investigation ended without prosecutions.
It is whether the evidence was adequately tested before that outcome was reached.
The most important question raised by my experience of the Northwood complaints process is not simply why the original complaint was not upheld.
It is whether the investigation and subsequent review adequately examined the evidence, the actual substance of the complaint and the consequences of acknowledged factual misunderstandings.
Both questions concern the integrity of the process by which an institution establishes what it believes to be true.
The danger is not limited to deliberate suppression or dishonesty.
It also lies in ordinary methodological choices that narrow the evidence, transform its meaning or give premature authority to a conclusion.
A genuinely independent accountability system must be able to distinguish between an investigation that found no evidence of wrongdoing and an investigation that did not adequately examine the evidence capable of establishing whether wrongdoing occurred.
The true test of independent review is not whether it can defend an investigative conclusion. It is whether it can recognise when that conclusion was produced by a defective investigative process — and correct it.That standard matters in Afghanistan.
It matters at Northwood.
And it matters wherever state institutions exercise power over individuals whose ability to challenge the official record is limited.
On 26 August 2025 I was lucky not to be shot dead at Northwood gate.
Author's note and sourcesJanina White is a solicitor in England and Wales. She is the complainant in the Northwood matters discussed above and the claimant in related proceedings. The analysis reflects her professional interpretation of the documentary record and distinguishes her allegations from findings made by the relevant authorities.
The Afghanistan discussion draws on Joshua Rozenberg's commentary published on 11 November 2026 concerning evidence before the Afghanistan Inquiry. The underlying allegations remain subject to the Inquiry's examination.
The Northwood discussion draws principally on
Review Wye, SPCC/2026/93, dated 30 April 2026, particularly paragraphs 37(xliii), 58–68 and 76–84; the original DSCC Professional Standards complaint outcome dated 20 January 2026; and the subsequent complaint and review chronology concerning SPCC 2026/128 and SPCC 2026/129.
The lawfulness of Review Wye is disputed in judicial-review proceedings,
R (White) v Service Police Complaints Commissioner, AC-2026-LON-003290. No final judicial determination of the alleged review defects is asserted here.
The article does not allege that the persons involved in the Northwood complaints committed war crimes, nor that the two cases involve misconduct of equivalent gravity.