The Post Office Horizon IT Scandal: Inquiry Reveals Shocking Testimonies

The core point: the scandal was also a failure of representation

The Horizon scandal is often described through technology, prosecutions and compensation. Those are essential. But the scandal also exposes a quieter institutional failure: the people raising concerns were too often isolated, disbelieved, managed or redirected rather than represented.

The National Federation of SubPostmasters was expected to speak for sub-postmasters. That made its relationship with the Post Office important. If a representative body is too close to the organisation its members must challenge, the risk is not merely reputational. The risk is that the first warnings are treated as noise.

George Thomson’s reported evidence matters because it appears to show how deeply some institutional actors continued to defend the Horizon system, even after the wider public record had shifted. The question for public trust is not whether one witness used strong language. It is whether the representative structure was capable of confronting uncomfortable evidence when it mattered.

What was said: defending Horizon and criticising the handling

The supplied draft records that George Thomson, former leader of the NFSP, defended the Horizon system and argued that the real scandal lay in the Post Office’s handling of the situation rather than the system itself. Public reporting of his evidence records that he denied betraying members, denied being too close to the Post Office, said he had been too trusting of information provided by the Post Office, and maintained that his position on Horizon had not changed after the NFSP’s deal with the Post Office.

That distinction matters. It is one thing to say the Post Office mishandled complaints, prosecutions and disclosure. It is another to maintain that Horizon was essentially robust when courts, campaigners and the inquiry record have focused heavily on bugs, errors, defects, unreliable data, governance failure and the human consequences of institutional denial.

Why it matters: members needed challenge, not reassurance

Sub-postmasters were not ordinary consumers making routine complaints. Many were small business operators facing unexplained shortfalls, demands for repayment, loss of contract, reputational damage, criminal investigation, prosecution, bankruptcy and severe personal consequences.

In that context, reassurance from institutional channels could be dangerous if it discouraged deeper investigation. A representative body did not need to prove every Horizon concern immediately. But it did need to test the pattern: repeated complaints, unexplained shortfalls, pressure to repay, prosecutions, settlement pressure, and the Post Office’s insistence on system reliability.

The public-interest concern is therefore structural. If members report a pattern of harm and the representative body remains aligned with the institution being challenged, members may be left without an effective early-warning system.

The representation test: what an effective body should show

The NFSP issue should not be reduced to a personal attack on one former official. The more useful question is institutional: what safeguards should a representative body have when its funding, access, policy influence or survival depend on a powerful counterpart?

Public trust requires a visible separation between partnership and representation. A representative body can work with an institution. It can negotiate, cooperate and protect the network. But it must also be able to disagree, gather contrary evidence, support individual members and escalate systemic risk.

The Horizon record: why “robust” language became untenable

Any current commentary must recognise the wider record. The Horizon scandal has moved beyond a dispute about a few branch-level accounting disagreements. It now includes group litigation, overturned convictions, a statutory public inquiry, compensation schemes, legislation and continuing scrutiny of accountability.

The Post Office Horizon IT Inquiry states that it was established to gather a clear account of the implementation and failings of the Horizon IT system over its lifetime. The Inquiry has also published Volume 1 of its Final Report, focused on human impact and redress, and has said further work continues on the remainder of the report.

Against that background, any defence of Horizon as “robust” must be handled carefully. The issue is not whether every Horizon transaction was wrong. The issue is that system reliability was treated as an institutional assumption while those challenging it faced severe consequences.

Public trust: when institutional loyalty becomes evidence blindness

The most damaging institutional failures often begin with loyalty that looks reasonable at the time. Staff trust internal experts. Leaders rely on familiar channels. Representative bodies protect negotiated relationships. Lawyers defend the client’s position. Technology suppliers defend the product. Each step can be explained. Together, they can produce evidence blindness.

The Thomson evidence matters because it sits within that broader pattern. The Inquiry was not merely asking whether one person was right or wrong about Horizon. It was examining how institutions responded when the people closest to the harm said the system was not behaving as it should.

The public-confidence failure is therefore not only historical. Any regulated, franchised or publicly significant system faces the same risk where complaints are individualised, representatives are financially dependent, and institutional reputation is treated as more important than warning signals.

Reform lessons: what the Horizon scandal should change

The lesson is not that representative bodies should never work with institutions. They often must. The lesson is that cooperation needs safeguards when member interests may conflict with institutional reputation.

Those safeguards should be practical. Funding arrangements should be transparent. Members should know whether their representative body is dependent on the organisation it may need to criticise. Serious complaints should be capable of independent escalation. Patterns should be recorded. Technical assurances should be tested. Whistleblowers and dissenting members should not be treated as threats to the brand.