On Forgetting Names — Letters from the Floor

The moment is familiar to anyone who has spent any length of time in business. A person walks up to you in the lobby of a hotel, or the corridor of a conference center, or the doorway of an office. They greet you by name. They are pleased to see you. They are picking up a conversation you had nine months ago in another city, and they remember the details: your daughter’s college search, the book you recommended, the project you mentioned was about to ship. They are warm and specific. You are mortified, because you have no idea who they are.

There is no graceful way out of this moment. The bluff is recognizable to anyone over the age of about thirty, which is to say, recognizable to the person you are bluffing. You make eye contact for slightly too long, hoping the name will come. You ask a softening question, “how have you been,” that buys you nothing. You make it through ninety seconds, and then someone else mercifully arrives and saves you, and you spend the next half hour trying to think of who that person was, and you spend the half hour after that wondering what is wrong with you.

The honest answer, mostly, is that nothing is wrong with you. The forgetting is not a moral failure. It is a specific and very ordinary feature of how human memory works, and the feature has a name in the cognitive psychology literature: the Baker/baker paradox.

The paradox was demonstrated most cleanly in a series of experiments conducted by the British psychologist Gillian Cohen in 1990. Show one group of people a photograph of a man and tell them his last name is Baker. Show another group of people the same photograph and tell them the man is a baker. A few days later, bring everyone back and ask them what they remember about the man. The group that was told he was a baker remembers, in significantly greater numbers, that he was a baker. The group that was told his name was Baker more often draws a blank. The word is identical. The fate of the word in memory is not.

The mechanism is, in retrospect, straightforward. When you learn that a stranger is a baker, the word arrives in your head connected to a web of other knowledge: bread, flour, ovens, early mornings, the smell of a bakery, the visual image of a person in a white apron. Memory researchers call these associations semantic hooks. The word “baker” gets stored in a network of related concepts, and any one of those concepts, encountered later, can pull the word back into consciousness. When you learn that a stranger’s name is Baker, the same word arrives at your hippocampus stripped of all those associations. It is a bare sound. There is nothing for it to attach to. The cognitive psychologists Deborah Burke and Donald MacKay, working in the early 1990s, developed a detailed account of why proper names are particularly vulnerable to this kind of retrieval failure: they sit, in the architecture of memory, on what Burke and MacKay called weak links to the rest of the mind. Other facts about a person, what they do, where they live, whether they have children, are integrated into a wider knowledge structure. The name is essentially alone. When you reach for it, there is nothing in the surrounding network to help you find it.

This is why you can remember everything else about the person standing in front of you and not their name. The everything-else is connected to other things. The name connects only to the person, and the person is the thing you are trying to identify in the first place.

What is remarkable, given how reliable this difficulty is, is that humans have been working on the problem for at least twenty-five centuries.

In the fifth century BCE, in Thessaly, a Greek lyric poet named Simonides of Ceos was hired to perform at a banquet thrown by the nobleman Scopas. After his recitation, Simonides was called outside on some pretext. While he was out, the roof of the banqueting hall collapsed and killed everyone inside. The bodies were so badly crushed that the dead were unrecognizable to their families. Simonides, asked to help, realized that he could remember exactly where each guest had been seated. By walking the wreckage in his mind, he was able to identify every victim by their position.

This is the story Cicero tells in De Oratore, four centuries after the fact, and it is the founding legend of the discipline that came to be called the art of memory. Cicero, and Quintilian after him, attribute to Simonides the invention of what they called the method of loci: the practice of placing items to be remembered at specific locations in an imagined space, usually a familiar building, so that one could walk through the space mentally and pick up each item in turn. The method survives in modern training as the memory palace. It is the same technique that today’s competitive memorizers use to memorize the order of decks of cards and thousand-digit strings of pi. It is also, in its application to names, an explicit cognitive workaround for the Baker/baker paradox. To remember that a woman you have just met is named Olivia, you do not simply repeat the name to yourself. You attach it to an image: Olivia eating blue ice cream, say. It is the technique that memory athletes like the three-time world champion Alex Mullen describe in detail, and the stranger the image, the more reliably it sticks. You are, in essence, manufacturing the semantic hooks that the bare name did not arrive with. You are doing for the proper noun what the brain does effortlessly for the common one.

The art of memory was central to classical and medieval rhetoric. Cicero used it to memorize his speeches. Augustine, who left a remarkable description of memory in his Confessions, treated the inner architecture of remembered things as one of the great theaters of the mind. Frances Yates, in her 1966 study The Art of Memory, traced the technique’s elaborate development through the Renaissance, when it became entangled with hermetic philosophy and the imagined cosmologies of figures like Giordano Bruno, whose memory palaces grew so baroque that they functioned as something closer to encyclopedias of the universe than as practical aids to recall.

The technique receded, but it never disappeared. It received a second life in the 21st century with the publication of Joshua Foer’s Moonwalking with Einstein (2011), which traced Foer’s own journey from journalist covering the U.S. Memory Championship to winner of the U.S. Memory Championship one year later, having trained himself in the same techniques Simonides invented. The book made it broadly known that competitive memorizers are, in almost every case, people of ordinary memory who have learned an unusually rigorous discipline. The neuroscientist Charan Ranganath, in his 2024 book Why We Remember, makes the same point in less anecdotal form: human memory is not a recording device. It is a reconstructive system, evolved to remember what was useful to remember, which was rarely abstract labels. The fact that we can train it to do more than its defaults is one of the more impressive features of the human cognitive endowment. The fact that we mostly do not is one of the less impressive.

What this means, practically, is that the people who are very good at remembering names are almost never blessed with superior memory. They have, at most, slightly better-than-average attention at the moment of introduction, and they have developed habits that compensate for what the rest of us do badly. They repeat the name back: “It’s good to meet you, Maria.” They use it twice more in the first two minutes of the conversation, when the name is still fresh and reinforcement is cheap. They look for an association, sometimes consciously, sometimes by reflex. They are, in short, doing some version of what Simonides would have recognized. They are giving the bare name something to attach to.

There is a particular kind of person in any room who does this consistently, and they tend to be the people whose careers are built on remembering. The salesperson who has worked the same territory for fifteen years and still remembers the spouses and children of customers she sees once a quarter. The political fundraiser who can move through a donor reception calling every guest by name without ever sounding like she is performing the trick. The longtime concierge at a hotel who recognizes a guest from a stay three years ago. None of them, when you ask, will claim to have an unusual memory. They will tell you instead that they pay attention, and that they have small habits, and that the habits add up.

The small social violence of forgetting a name is the violence of having failed, in some small way, to honor the person standing in front of you. The forgetting is not, in itself, a moral fact. It is a cognitive one, and an unavoidable one, and most of the people we forget will forgive us if we ask them, gently, to remind us. But what the forgetting registers, accurately, is the absence of the small attention that would have prevented it. The remembering, when other people do it for us, registers exactly the same thing in reverse. To be remembered by name, by someone whose business it was to remember many names, is to be told without words: you mattered enough to attach to something.

This is the part of the equation that is worth taking seriously, because most of us spend a lot of time hoping to be remembered and very little time doing the work of remembering. The work is not glamorous. It is the work of repeating a name back at introduction, of looking for the association, of writing it down on the back of the card before the conversation ends. The techniques are not new. They are, in fact, almost the oldest techniques there are. The art of memory has been part of the equipment of educated people for two and a half millennia. We have mostly forgotten it, which is fitting, and slightly funny, and entirely the point.

Further Reading