When the System Fails

There is a response that comes reliably when police investigative failures are raised in connection with serious crime. It arrives in press statements, in official reviews, and in the testimony of senior officers to parliamentary committees. We did not have the tools then that we have now. You cannot judge the past by the standards of the present. This is sometimes true. It is also sometimes a shield.

The cases documented here span five decades and involve forces across England, Scotland and Greater Manchester. What they share is not a failure of technology. They share a failure of process, of rigour, and in some cases of honesty. In each case the consequence was not an administrative inconvenience. It was further harm to further victims, harm that might have been prevented.

Between 1975 and 1980, Peter Sutcliffe murdered thirteen women and attacked at least seven more across Yorkshire and Manchester. He was interviewed by police on nine separate occasions during the investigation and released each time. The investigation was paralysed by a hoax tape sent to George Oldfield, the assistant chief constable leading it, which detectives accepted as genuine for nearly two years. Resources were diverted to a man with a Sunderland accent. Sutcliffe spoke with a Yorkshire one.

The 1981 Byford Report identified that information held in the investigation's paper files was so disorganised that connections between Sutcliffe's nine interviews could not be made. The Sampson Report into the same investigation found that women reporting attacks who survived were not always taken seriously, and that prejudicial assumptions about which victims mattered shaped how the investigation prioritised its leads.

Black-cab driver John Worboys is believed to have drugged and assaulted more than 100 women in London between 2002 and 2008. The Metropolitan Police investigation failed at multiple points. Victims who reported were disbelieved. Officers told one woman that a black-cab driver would not commit such an offence. Forensic evidence was not collected. Cases were not linked.

Two of his victims successfully sued the Met under the Human Rights Act, with the Supreme Court ruling in 2018 that the police had violated Article 3 of the European Convention on Human Rights through their investigative failures. It is one of the few cases in which a police force has been held legally accountable for the consequences of failing to investigate.

Operation Midland (2014-2016) was a Metropolitan Police investigation into entirely fabricated allegations of a Westminster paedophile ring. Senior officers publicly described the false complainant's account as 'credible and true' before evidence had been tested. £2.5m was spent. Reputations were destroyed. The complainant was later convicted of perverting the course of justice.

In the same period, Stephen Port was murdering young gay men in Barking. Families reported their concerns. The Met dismissed the deaths as accidental overdoses. Port killed four men. The 2021 inquest jury found that police failures probably contributed to three of the deaths. The contrast between the resource lavished on a fabricated account and the indifference shown to genuine victims is the structural point.

In July 2024 Dame Vera Baird published her inquiry into the treatment of a woman who attended Pendleton police station to report domestic abuse and was instead arrested. The inquiry found GMP's conduct 'obviously wrong, in every way you can dream of.' The case is one VNS has documented because it is the clearest published example of institutional DARVO: a victim reporting harm being processed as the suspect.

• Failure to take initial reports seriously, especially from victims who do not fit the ideal-victim template
• Failure to link incidents that should have been linked, allowing a pattern to continue undetected
• Public statements made before evidence had been tested, then defended after evidence emerged
• Resource allocation that follows institutional reputation rather than victim risk
• Subsequent reviews that identified the failures with admirable clarity, and recommended fixes that the next inspection then found had not been implemented