Memory is not a record

In March 1964, Kitty Genovese, a 28-year-old bar manager, was murdered near her home in Queens on her way back from work. Two weeks later, The New York Times reported that 38 of her neighbors had seen or heard the attack and failed to help.

The story caused national outrage over urban apathy. It inspired decades of research into what psychologists came to call the bystander effect, and it was taught in psychology classes for a generation.

It was also wrong. Later reporting found that far fewer people witnessed the attack than the story claimed, and that at least one neighbor did try to help. In 2016, the Times itself acknowledged that its original account had exaggerated the number of witnesses and what they perceived.¹

One of the most famous eyewitness stories of the twentieth century didn’t hold up. The ones on your floor might not either.

What the cameras show

Over our careers, we’ve investigated hundreds of incidents, from simple mistakes to melees involving dozens of people. In every case we can remember, the story told by the people involved differed from what the surveillance cameras recorded: who was there, when things happened, where, and how it unfolded.

Nobody had to be lying for that to be true. Memory isn’t a recording. It’s a reconstruction, and it gets rebuilt a little differently every time it’s told.

For a long time, eyewitness testimony was treated as the gold standard of evidence. Then DNA testing arrived. The Innocence Project has found that eyewitness misidentification played a role in roughly 70 percent of the wrongful convictions later overturned by DNA evidence in the United States.² Those witnesses weren’t careless. They were confident, sincere and wrong.

A confident witness and an accurate witness are not always the same person.

Would you bet a decision on it?

So ask yourself: do you really want your employment decisions to depend on secondhand reports and witness recollection?

Of course not. That means documenting the things that matter, all of them: workplace incidents, issues and events; training and certifications; meetings, notes and comments; checklists; and the routine items, like attendance and time-off requests, that become important the moment someone disputes them.

How to make a record that holds

• Pin down the facts. Confirm dates, times and everyone involved, including witnesses.
• Attach the evidence. Photos, video, receipts, signed forms: anything that doesn’t depend on memory.
• Stay objective. Describe what happened. Leave out guesses, labels and speculation.
• Have witnesses confirm. Ask the people involved to review the description of events, and note that they did.
• Do it now. Every hour that passes, memory drifts further from what the camera saw.

The narrative fills the vacuum

Without documentation, the story of what happened becomes an argument between lawyers, a contest of unreliable memories and mixed messages. Documentation won’t stop people from speculating. But it gives you a fixed point to bring the conversation back to, and it helps you defuse a false narrative before it costs you your reputation or your money.

Remember that in many employment disputes, the employer ends up carrying the burden of proof. Without a record made at the time, you’re trusting your fate to whoever remembers it most convincingly.

When something happens on your floor, log it while it’s fresh: the people involved, the responders, the witnesses, the photos. Memory fades. The record doesn’t.