On June 23, the Associated Press issued its second retraction on the “mass grave” hoax. Below is an excerpt.

Revelations this month that nuns had buried nearly 800 infants and young children in unmarked graves at an Irish orphanage during the last century caused stark headlines and stirred strong emotions and calls for investigation. Since then, however, a more sober picture has emerged that exposes how many of those headlines were wrong.

The case of the Tuam “mother and baby home” offers a study in how exaggeration can multiply in the news media, embellishing occurrences that should have been gripping enough on their own.

The key fact is that a researcher, Catherine Corless, spent years seeking records of all the children who died in the orphanage in County Galway during its years of operation from 1925 to 1961. She found 797 death records – and only one record that one of the youngsters had been buried alongside relatives in a Catholic cemetery.

The rest, Corless surmised, were likely interred in unmarked graves on the orphanage grounds, including in a disused septic tank.

The reports of unmarked graves shouldn’t have come as a surprise to the Irish public, who for decades have known that some of the 10 defunct “mother and baby homes,” which chiefly housed the children of unwed mothers, held grave sites filled with forgotten dead.

The religious orders’ use of unmarked graves reflected the crippling poverty of the time, the infancy of most of the victims, and the lack of plots in cemeteries corresponding to the children’s fractured families.

Until recent weeks, nobody had put a precise number on the fatalities at Tuam. Corless spent months – and more than 3,000 euros ($4,000) of her own money – buying copies of death certificates and organizing them.

Her list of the dead shows that nearly 80 percent were younger than 1; two died within 10 minutes of birth and never received first names. Ninety-one died in the 1920s, 247 in the 1930s, 388 in the 1940s, 70 in the 1950s, and one more child in 1960. The most common causes were flu, measles, pneumonia, tuberculosis and whooping cough. Contrary to the allegations of widespread starvation highlighted in some reports, only 18 children were recorded as suffering from severe malnutrition.

While publicly available records are incomplete, sporadic inspection reports indicate that the orphanage’s population exceeded 250 throughout the worst years of child mortality, when overcrowding would have encouraged the spread of infection.

When Corless published her findings on a Facebook campaign page, and Irish media noticed, she speculated to reporters that the resting place of most, if not all, could be inside a disused septic tank on the site. By the time Irish and British tabloids went to print in early June, that speculation had become a certainty, the word “disused” had disappeared, and U.S. newspapers picked up the report, inserting more errors, including one that claimed the researcher had found all 796 remains in a septic tank.

The Associated Press was among the media organizations that covered Corless and her findings, repeating incorrect Irish news reports that suggested the babies who died had never been baptized and that Catholic Church teaching guided priests not to baptize the babies of unwed mothers or give to them Christian burials.

The reports of denial of baptism later were contradicted by the Tuam Archdiocese, which found a registry showing that the home had baptized more than 2,000 babies. The AP issued a corrective story on Friday after discovering its errors.

Brendan O’Neill, editor of the London-based online magazine Spiked, said journalists worldwide “got a whiff of Corless’s findings and turned them into the stuff of nightmares.” He noted that several top newspapers in the United States stated that 800 baby skeletons had been found in a septic tank, and that commentators fueled by a “Twitter mob” mentality compared the deaths to Nazi-era genocide.

The Irish Times in Dublin interviewed Corless about why she thought the former septic tank could have become a bone repository. She explained that her assertion was based on the study of old site maps and the 40-year-old recollections of two local men who, as boys, had found an underground chamber on the site containing skeletons. It had sounded to her like the tank could be the location.

But the newspaper spotted discrepancies in Corless’ maps, and found records showing that the actual septic tank remained in use until the late 1930s, which meant it could not have been used as a burial spot. Other analysts pointed out that the decommissioned septic tank would be too small to hold many bodies. And the two men who had reported seeing skeletons in 1975 said, on reflection, that they doubted more than 20 were inside the concreted hole.

Ireland last week announced it would open a judge-led investigation into the care of children in Tuam and nine other defunct facilities, and the handling of their remains. Whether the fact-finding effort will include excavations at any of the former homes or DNA analysis of remains has yet to be decided.

In an editorial, the Irish Times said Ireland was suffering “self-induced amnesia” given that historians already had documented “staggeringly high mortality rates in some mother and child homes.” It noted that Tuam’s mortality rate appeared lower than others, and predicted the upcoming inquiry into the entire system would be painful.

“Learning from the past can be a disturbing process,” it said. “It involves an examination of failures and the acceptance of hurtful conclusions. It means making amends for past societal wrongs. It should establish why certain things happened, rather than heap blame on those who implemented policy.”