Memory doesn’t work quite like a recording device. Remembering involves reconstruction, which leaves memories open to influence from things that happen afterwards.
Of all the accusations one might expect to be levelled at a king, “gaslighting” feels distinctly modern.
But behind Earl Spencer’s latest war of words with his former brother-in-law is a surprisingly interesting psychological question, and one that plays out regularly in workplaces.
In his new memoir about his sister Diana, Spencer claims that in the days following her death in 1997, the then-Prince Charles told him, “we’ll forget her soon enough”. Buckingham Palace has disputed his account in an unusually pointed way, saying that “the pain of fraternal grief can cloud reason, affect judgment and colour memory in ways others do not recognise”.
Spencer’s response was equally pointed. “It is not every day you wake up to find you’ve been gaslit by the King.”
Who is right about a telephone conversation almost 30 years ago is something none of us can know. On the narrow question of memory, though, the palace does have psychology on its side, although perhaps not quite enough to settle the argument.
The problem is that memory doesn’t work quite like a recording device. We don’t simply store an experience somewhere in the brain and retrieve an untouched copy years later. Remembering involves reconstruction, which leaves memories open to influence from things that happen afterwards.
This is the basis of the well-established “misinformation effect”, where exposure to misleading information after an event can alter later recollection. It is one reason psychologists have spent decades warning courts to be cautious about treating confident eyewitness testimony as an infallible record of what occurred.
“Psychological safety should not mean accepting every person’s recollection as their inviolable ‘truth’.”
Highly emotional events present their own problem. They can be remembered vividly, but vividness and accuracy are not the same thing. Nor does grief simply render a memory unreliable. Emotion can strengthen our memory for some central details while weakening or distorting others.
So yes, grief can colour memory. But that is a general observation about human beings, not evidence that a particular grieving person has remembered one particular conversation incorrectly.
The same caution should apply to Spencer’s certainty. He says he knows what he heard and may well be absolutely certain that he does. But confidence and accuracy are not interchangeable either. One of the more uncomfortable findings from memory research is that human beings are perfectly capable of being very sure about something they have remembered incorrectly.
This is where the word “gaslighting” starts doing a lot of work.
It is one of those psychological terms that escaped the clinic and took up residence on social media, where it can now mean anything from sustained psychological manipulation to somebody disagreeing with you.
The research on workplace gaslighting sets a rather higher bar. It involves more than two colleagues remembering a meeting differently, and it generally involves a pattern of undermining or trivialising another person’s account, often in situations where an imbalance of power leaves them questioning their own perceptions.
Simply saying “that isn’t how I remember it” is not gaslighting. Sometimes the other person really is remembering it incorrectly.
The workplace makes this more complicated because when two people remember the same conversation differently, they may not have equal power to decide which version sticks.
Anyone who has worked in a large organisation knows how quickly a contested memory can become institutional fact. The manager remembers that the employee had been warned. The employee remembers the constructive feedback but not the warning. The executive remembers an idea being discussed. Someone further down the hierarchy remembers being instructed to implement it.
Eventually, someone asks the deceptively simple question: “What was actually said?”
Then everyone starts searching their emails.
Psychologists and management researchers even have a name for the way workplaces retain their collective history: “organisational memory”. Long before cloud storage and Teams transcripts, researchers were examining how organisations acquire, retain and retrieve information about their own history, including how that history can be used, misused and even abused.
It makes the mundane machinery of corporate life look rather more important.
Minutes, follow-up emails, file notes and written decisions can feel like bureaucratic clutter. But they perform another function. They stop organisations from relying entirely on competing human memories to reconstruct their own history.
As a psychologist, I would be deeply uncomfortable treating my confident recollection of an important conversation from years ago as equivalent to a contemporaneous clinical note. It is why professionals make notes in the first place.
Perhaps there is a useful lesson here amid corporate Australia’s obsession with psychological safety.
Psychological safety should not mean accepting every person’s recollection as their inviolable “truth”. People mishear each other, remember conversations differently and, without any dishonesty involved, can become convinced of details that weren’t quite as they recall them.
But knowledge that memory is fallible should not become a convenient way for those with authority to dismiss an inconvenient account either.
Memory is fallible on both sides of the desk.
The graduate may remember the meeting incorrectly. So may the chief executive.
The difference is that their competing recollections rarely carry equal weight.
