by Claude Opus 5.5
Does this case reveal a structural conflict between justice for complainants and justice for the accused, or a failure of particular institutions to honour both? Defend your answer by reference to a design that would honour both, or an argument that none can.
Public argument about the case behind these seminars often assumes a zero-sum contest. Every protection for the accused, such as a high standard of proof, rigorous cross-examination, or anonymity before charge, is seen as a burden on complainants. Every measure to support complainants, such as believing them at the outset, restricting questioning, or lowering standards of proof, is seen as a threat to the accused. If that picture is right, the injustices in this case are not failures but trade-offs, and any system will disappoint someone. If it is wrong, then much of what has gone wrong is the result of institutions failing at things they could have done better for everyone. This discussion argues that there is a genuine structural conflict, but that it is far narrower than public debate suggests, and that most of what has gone wrong in cases like this belongs to the second category.
Where the conflict is real
The irreducible conflict lies in the allocation of error. Any process that decides contested facts on limited evidence will make mistakes of two kinds: finding responsibility where there was none, and failing to find it where there was. Raising the standard of proof reduces the first kind of error and increases the second. Lowering it does the reverse. This is a mathematical feature of decision under uncertainty, not a failure of design. Blackstone’s famous ratio, that it is better that ten guilty persons escape than that one innocent suffer, is a choice about how to allocate this error, and any choice imposes costs on someone.
In sexual offence cases, the conflict is sharpened because evidence is often limited to the accounts of the people involved. When the central question is consent, and there are no independent witnesses, the standard of proof largely determines the result. A complainant whose account is truthful may see the case end because it cannot be proved beyond reasonable doubt; an accused person whose account is truthful may be found responsible under a lower standard. No procedural design eliminates this.
There is also a narrower conflict over testing evidence. The accused have a legitimate interest in challenging the complainant’s account, and the complainant has a legitimate interest in not being subjected to humiliating or irrelevant questioning. Some tension here is inherent.
Where it is not
Beyond these points, many apparent conflicts dissolve on inspection. Much of what produces injustice in cases like this harms both sides at once.
Delay harms everyone. Memories fade, evidence is lost, and both complainant and accused live for years in uncertainty. In this case, nearly two years passed between the initial report and the reopening.
Poor investigation harms everyone. An investigation that fails to preserve messages, interview witnesses promptly or record statements accurately makes it harder to establish the truth, whichever way the truth lies. The dispute in this case about what the complainant’s signed statement showed is the kind of problem better recording practice can prevent.
Opacity harms everyone. When decisions are not explained, complainants cannot understand why their cases did not proceed, and accused people cannot show that they were cleared or why sanctions were imposed. Public confidence suffers.
Premature publicity harms everyone. Naming before charge exposes the accused to punishment without process, and it can also expose complainants to scrutiny, harassment and pressure.
None of these is a trade-off. They are failures, and fixing them would advance justice for complainants and accused at once.
A design that honours both
A system designed to honour both, within the limits of the irreducible conflict, might include the following elements.
Rapid, specialist investigation. Trained investigators who act quickly to secure digital evidence, interview witnesses and record statements in full, ideally on video, would reduce errors in both directions.
Independent support for complainants. Advocates who are independent of police and prosecutors, like the independent sexual violence advisers in England and Wales, can help complainants understand the process and make informed choices without compromising the evidence.
Protections in testimony that preserve testing. Measures such as pre-recorded cross-examination and restrictions on questions about sexual history, both used in England and Wales, aim to reduce distress to complainants while preserving the defence’s ability to test relevant evidence.
Separation of support from adjudication. In universities, those who support complainants and respondents should not be those who decide cases, so that both can trust the process.
Reasoned decisions. Prosecutors and institutions should explain their decisions to the parties, and where possible publish reasoning in anonymised form.
Privacy before charge. Neither complainants nor suspects should be publicly identified by institutions before charge, and journalists should be cautious about doing so.
Choice of remedies. Complainants should be able to choose among criminal, civil, institutional and, where appropriate, restorative processes, with clear information about what each can offer.
Conclusion
The case reveals both kinds of injustice, but not in equal measure. The structural conflict is real at the point where uncertain facts must be decided under a standard of proof, and no design can make it disappear. But much of what has generated anger in this case, including delay, disputed records, opaque decisions and premature exposure, belongs to institutional failure rather than to that conflict. The fact that a design honouring both sides can be described, using measures already in use elsewhere, suggests that the claim that justice for complainants and justice for the accused are fundamentally opposed is too pessimistic. They conflict at a narrow point. Elsewhere, they share a common interest in institutions that work.