The Globe's Robyn Doolittle has followed her series on sexual assault allegations, Unfounded, with classic journalistic doggedness. She continues to ask how police forces are responding to the 20-month investigation that found that one out of every five sex-assault allegations is being dismissed as "unfounded."

But it was her description of what "unfounded" meant that was criticized by a philosophy professor emeritus from McMaster University.

This is her description: " 'unfounded' – a term that means the investigating officer does not believe a crime occurred."

Story continues below advertisement

David Hitchcock wrote to ask for a correction and said The Globe and Mail should not mislead its readers: "That is not what the term means. According to Statistics Canada's Crime Reporting Manual Version 1.0, which guides the reporting by all police forces in Canada: 'Unfounded' means that the police investigation has established that a crime did not happen or was not attempted."

"There is a difference between an investigation's establishing something and an investigating officer's believing something. There is also a difference between a crime's not having occurred and a crime's not having happened or been attempted. Please publish a correction in your usual place on page 2." He added in a separate letter: "On the Statistics Canada definition, the investigation needs something approaching proof for a report of a crime to be called 'unfounded.' Mere belief that no crime occurred, without proof, should not be enough."

The Globe and Mail corrects "significant factual errors" – more than 500 every year. Names, numbers, incorrect facts etc. and many corrections are made thanks to the vigilant readers like Prof. Hitchcock.

While Prof. Hitchcock is correct in stating the official definition, I don't think a correction is needed or, well, correct. That official definition is that the police have "established that a crime did not happen …"

But can they establish that something did not happen? I think it is clear reading this series that one police officer reviewing evidence cannot always establish that nothing happened. Officers might believe nothing happened or that there isn't enough evidence, but clearly they cannot say absolutely that nothing happened.

In all cases, but especially cases of sex assault, the crime might not be reported for some time, there could be a question around consent or there might be little factual evidence. I argued with the professor that the police weigh the evidence along with what they believe is the credibility of those involved. And in fact, the decision on whether a crime has in fact occurred is made only by a judge or jury and not the police. And even the courts have shown they can be wrong.

Police and Crown attorneys decide whether they believe there is enough evidence to proceed. The police use their judgment, evidence and, yes, their belief to come to a conclusion and the police can be and have been wrong.

Story continues below advertisement

Ms. Doolittle noted that "if a case ends up unfounded it is because a human being decided it was unfounded and as The Globe's series has shown, that determination hasn't always been correct. In response to The Globe's investigation, nearly a third of the country's police services have announced they plan to review sex-assault cases that were closed as unfounded. In fact, more than 10,000 cases will be reviewed. Most of those audits are ongoing and the results are unknown, but already there have been many cases where police services have announced that cases were incorrectly dismissed as unfounded – including the case of Ava Williams, the main subject in the first story. In that case, the detective believed the allegation was unfounded. After The Globe inquired about the case, the service concluded he was wrong."

This feels like a test in my first year philosophy class from years gone by and I would be interested to hear what you philosophers think about this argument.

In deciding what will be corrected, I rely on the two key words in the policy description: "significant factual" errors. To me, the above issue was neither a significant nor factual error.

Another reader wrote to me on Monday. He was correct, but after making his point he agreed that it was minor. It was a reference to Brexit and the story referred to "Vote Leave leader Nigel Farage." So the reader is right that Mr. Farage would be better and correctly described as the UKIP leader.

Still I wondered if it was significant enough to warrant a correction. Mr. Farage was a certainly a leader within the Vote Leave campaign and an active campaigner although he wasn't "the" leader. I let the reader know that I thought no correction was needed because while it was factually inaccurate, it wasn't significant and he agreed that "other readers are unlikely to care or notice."

A third reader (one of many who note possible errors) let us know about an error that was both significant and factual. It was in a Moment in Time feature on the sentencing of Julius and Ethel Rosenberg. This is part of the correction: "[The article] incorrectly stated that Julius Rosenberg admitted passing secrets to the Soviets. He did not."