by Ted Chiang
The earliest birthday I remember is my fourth; I remember blowing out the candles on my cake, the thrill of tearing the wrapping paper off the presents. There’s no video of the event, but there are snapshots in the family album, and they are consistent with what I remember. In fact, I suspect I no longer remember the day itself. It’s more likely that I manufactured the memory when I was first shown the snapshots and over time, I’ve imbued it with the emotion I imagine I felt that day. Little by little, over repeated instances of recall, I’ve created a happy memory for myself.
Another of my earliest memories is of playing on the living room rug, pushing toy cars around, while my grandmother worked at her sewing machine; she would occasionally turn and smile warmly at me. There are no photos of that moment, so I know the recollection is mine and mine alone. It is a lovely, idyllic memory. Would I want to be presented with actual footage of that afternoon? No; absolutely not.
Regarding the role of truth in autobiography, the critic Roy Pascal wrote, “On the one side are the truths of fact, on the other the truth of the writer’s feeling, and where the two coincide cannot be decided by any outside authority in advance.” Our memories are private autobiographies, and that afternoon with my grandmother features prominently in mine because of the feelings associated with it. What if video footage revealed that my grandmother’s smile was in fact perfunctory, that she was actually frustrated because her sewing wasn’t going well? What’s important to me about that memory is the happiness I associated with it, and I wouldn’t want that jeopardized.
It seemed to me that continuous video of my entire childhood would be full of facts but devoid of feeling, simply because cameras couldn’t capture the emotional dimension of events. As far as the camera was concerned, that afternoon with my grandmother would be indistinguishable from a hundred others. And if I’d grown up with access to all the video footage, there’d have been no way for me to assign more emotional weight to any particular day, no nucleus around which nostalgia could accrete.
And what will the consequences be when people can claim to remember their infancy? I could readily imagine a situation where, if you ask a young person what her earliest memory is, she will simply look baffled; after all, she has video dating back to the day of her birth. The inability to remember the first few years of one’s life—what psychologists call childhood amnesia—might soon be a thing of the past. No more would parents tell their children anecdotes beginning with the words “You don’t remember this because you were just a toddler when it happened.” It’ll be as if childhood amnesia is a characteristic of humanity’s childhood, and in ouroboric fashion, our youth will vanish from our memories.
Part of me wanted to stop this, to protect children’s ability to see the beginning of their lives filtered through gauze, to keep those origin stories from being replaced by cold, desaturated video. But maybe they will feel just as warmly about their lossless digital memories as I do of my imperfect, organic memories.
People are made of stories. Our memories are not the impartial accumulation of every second we’ve lived; they’re the narrative that we assembled out of selected moments. Which is why, even when we’ve experienced the same events as other individuals, we never constructed identical narratives: the criteria used for selecting moments were different for each of us, and a reflection of our personalities. Each of us noticed the details that caught our attention and remembered what was important to us, and the narratives we built shaped our personalities in turn.
But, I wondered, if everyone remembered everything, would our differences get shaved away? What would happen to our sense of selves? It seemed to me that a perfect memory couldn’t be a narrative any more than unedited security-cam footage could be a feature film.
• • •
When Jijingi was twenty, an officer from the administration came to the village to speak with Sabe. He had brought with him a young Tiv man who had attended the mission school in Katsina-Ala. The administration wanted to have a written record of all the disputes brought before the tribal courts, so they were assigning each chief one of these youths to act as a scribe. Sabe had Jijingi come forward, and to the officer he said, “I know you don’t have enough scribes for all of Tivland. Jijingi here has learned to write; he can act as our scribe, and you can send your boy to another village.” The officer tested Jijingi’s ability to write, but Moseby had taught him well, and eventually the officer agreed to have him be Sabe’s scribe.
After the officer had left, Jijingi asked Sabe why he hadn’t wanted the boy from Katsina-Ala to be his scribe.
“No one who comes from the mission school can be trusted,” said Sabe.
“Why not? Did the Europeans make them liars?”
“They’re partly to blame, but so are we. When the Europeans collected boys for the mission school years ago, most elders gave them the ones they wanted to get rid of, the layabouts and malcontents. Now those boys have returned, and they feel no kinship with anyone. They wield their knowledge of writing like a long gun; they demand their chiefs find them wives, or else they’ll write lies about them and have the Europeans depose them.”
Jijingi knew a boy who was always complaining and looking for ways to avoid work; it would be a disaster if someone like him had power over Sabe. “Can’t you tell the Europeans about this?”
“Many have,” Sabe answered. “It was Maisho of the Kwande clan who warned me about the scribes; they were installed in Kwande villages first. Maisho was fortunate that the Europeans believed him instead of his scribe’s lies, but he knows of other chiefs who were not so lucky; the Europeans often believe paper over people. I don’t wish to take the chance.” He looked at Jijingi seriously. “You are my kin, Jijingi, and kin to everyone in this village. I trust you to write down what I say.”
“Yes, Sabe.”
Tribal court was held every month, from morning until late afternoon for three days in a row, and it always attracted an audience, sometimes one so large that Sabe had to demand everyone sit to allow the breeze to reach the center of the circle. Jijingi sat next to Sabe and recorded the details of each dispute in a book the officer had left. It was a good job; he was paid out of the fees collected from the disputants, and he was given not just a chair but a small table too, which he could use for writing even when court wasn’t in session. The complaints Sabe heard were varied—one might be about a stolen bicycle, another might be about whether a man was responsible for his neighbor’s crops failing—but most had to do with wives. For one such dispute, Jijingi wrote down the following:
Umem’s wife Girgi has run away from home and gone back to her kin. Her kinsman Anongo has tried to convince her to stay with her husband, but Girgi refuses, and there is no more Anongo can do. Umem demands the return of the £11 he paid as bridewealth. Anongo says he has no money at the moment, and moreover that he was only paid £6.
Sabe requested witnesses for both sides. Anongo says he has witnesses, but they have gone on a trip. Umem produces a witness, who is sworn in. He testifies that he himself counted the £11 that Umem paid to Anongo.
Sabe asks Girgi to return to her husband and be a good wife, but she says she has had all that she can stand of him. Sabe instructs Anongo to repay Umem £11, the first payment to be in three months when his crops are saleable. Anongo agrees.
It was the final dispute of the day, by which time Sabe was clearly tired. “Selling vegetables to pay back bridewealth,” he said afterwards, shaking his head. “This wouldn’t have happened when I was a boy.”
Jijingi knew what he meant. In the past, the elders said, you conducted exchanges with similar items: if you wanted a goat, you could trade chickens for it; if you wanted to marry a woman, you promised one of your kinswomen to her family. Then the Europeans said they would no longer accept vegetables as payment for taxes, insisting that it be paid in coin. Before long, everything could be exchanged for money; you could use it to buy everything from a calabash to a wife. The elders consider
ed it absurd.
“The old ways are vanishing,” agreed Jijingi. He didn’t say that young people preferred things this way, because the Europeans had also decreed that bridewealth could only be paid if the woman consented to the marriage. In the past, a young woman might be promised to an old man with leprous hands and rotting teeth, and have no choice but to marry him. Now a woman could marry the man she favored, as long as he could afford to pay the bridewealth. Jijingi himself was saving money to marry.
Moseby came to watch sometimes, but he found the proceedings confusing, and often asked Jijingi questions afterwards.
“For example, there was the dispute between Umem and Anongo over how much bridewealth was owed. Why was only the witness sworn in?” asked Moseby.
“To ensure that he said precisely what happened.”
“But if Umem and Anongo were sworn in, that would have ensured they said precisely what happened too. Anongo was able to lie because he was not sworn in.”
“Anongo didn’t lie,” said Jijingi. “He said what he considered right, just as Umem did.”
“But what Anongo said wasn’t the same as what the witness said.”
“But that doesn’t mean he was lying.” Then Jijingi remembered something about the European language, and understood Moseby’s confusion. “Our language has two words for what in your language is called ‘true.’ There is what’s right, mimi, and what’s precise, vough. In a dispute the principals say what they consider right; they speak mimi. The witnesses, however, are sworn to say precisely what happened; they speak vough. When Sabe has heard what happened can he decide what action is mimi for everyone. But it’s not lying if the principals don’t speak vough, as long as they speak mimi.”
Moseby clearly disapproved. “In the land I come from, everyone who testifies in court must swear to speak vough, even the principals.”
Jijingi didn’t see the point of that, but all he said was, “Every tribe has its own customs.”
“Yes, customs may vary, but the truth is the truth; it doesn’t change from one person to another. And remember what the Bible says: the truth shall set you free.”
“I remember,” said Jijingi. Moseby had said that it was knowing God’s truth that had made the Europeans so successful. There was no denying their wealth or power, but who knew what was the cause?
• • •
In order to write about Remem, it was only fair that I try it out myself. The problem was that I didn’t have a lifelog for it to index; typically I only activated my personal cam when I was conducting an interview or covering an event. But I’ve certainly spent time in the presence of people who kept lifelogs, and I could make use of what they’d recorded. While all lifelogging software has privacy controls in place, most people also grant basic sharing rights: if your actions were recorded in their lifelog, you have access to the footage in which you’re present. So I launched an agent to assemble a partial lifelog from the footage others had recorded, using my GPS history as the basis for the query. Over the course of a week, my request propagated through social networks and public video archives, and I was rewarded with snippets of video ranging from a few seconds in length to a few hours: not just security-cam footage but excerpts from the lifelogs of friends, acquaintances, and even complete strangers.
The resulting lifelog was of course highly fragmentary compared to what I would have had if I’d been recording video myself, and the footage was all from a third-person perspective rather than the first-person that most lifelogs have, but Remem was able to work with that. I expected that coverage would be thickest in the later years, simply due to the increasing popularity of lifelogs. It was somewhat to my surprise, then, that when I looked at a graph of the coverage, I found a bump in the coverage over a decade ago. Nicole had been keeping a lifelog since she was a teenager, so an unexpectedly large segment of my domestic life was present.
I was initially a bit uncertain of how to test Remem, since I obviously couldn’t ask it to bring up video of an event I didn’t remember. I figured I’d start out with something I did remember. I subvocalized, “The time Vince told me about his trip to Palau.”
My retinal projector displayed a window in the lower left corner of my field of vision: I’m having lunch with my friends Vincent and Jeremy. Vincent didn’t maintain a lifelog either, so the footage was from Jeremy’s point of view. I listened to Vincent rave about scuba diving for a minute.
Next I tried something that I only vaguely remembered. “The dinner banquet when I sat between Deborah and Lyle.” I didn’t remember who else was sitting at the table, and wondered if Remem could help me identify them.
Sure enough, Deborah had been recording that evening, and with her video I was able to use a recognition agent to identity everyone sitting across from us.
After those initial successes, I had a run of failures; not surprising, considering the gaps in the lifelog. But over the course of an hour-long trip survey of past events, Remem’s performance was generally impressive.
Finally it seemed time for me to try Remem on some memories that were more emotionally freighted. My relationship with Nicole felt strong enough now for me to safely revisit the fights we’d had when she was young. I figured I’d start with the argument I remembered clearly, and work backwards from there.
I subvocalized, “The time Nicole yelled at me ‘you’re the reason she left.’”
The window displays the kitchen of the house we lived in when Nicole was growing up. The footage is from Nicole’s point of view, and I’m standing in front of the stove. It’s obvious we’re fighting.
“You’re the reason she left. You can leave too, for all I care. I sure as hell would be better off without you.”
The words were just as I remembered them, but it wasn’t Nicole saying them.
It was me.
My first thought was that it must be a fake, that Nicole had edited the video to put her words into my mouth. She must have noticed my request for access to her lifelog footage, and concocted this to teach me a lesson. Or perhaps it was a film she had created to show her friends, to reinforce the stories she told about me. But why was she still so angry at me, that she would do such a thing? Hadn’t we gotten past this?
I started skimming through the video, looking for inconsistencies that would indicate where the edited footage had been spliced in. The subsequent footage showed Nicole running out of the house, just as I remembered, so there wouldn’t be signs of inconsistency there. I rewound the video and started watching the preceding argument.
Initially I was angry as I watched, angry at Nicole for going to such lengths to create this lie, because the preceding footage was all consistent with me being the one who yelled at her. Then some of what I was saying in the video began to sound queasily familiar: complaining about being called to her school again because she’d gotten into trouble, accusing her of spending time with the wrong crowd. But this wasn’t the context in which I’d said those things, was it? I had been voicing my concern, not berating her. Nicole must have adapted things I’d said elsewhere to make her slanderous video more plausible. That was the only explanation, right?
I asked Remem to examine the video’s watermark, and it reported the video was unmodified. I saw that Remem had suggested a correction in my search terms: where I had said “the time Nicole yelled at me,” it offered “the time I yelled at Nicole.” The correction must have been displayed at the same time as the initial search result, but I hadn’t noticed. I shut down Remem in disgust, furious at the product. I was about to search for information on forging a digital watermark to prove this video was faked, but I stopped myself, recognizing it as an act of desperation.
I would have testified, hand on a stack of Bibles or using any oath required of me, that it was Nicole who’d accused me of being the reason her mother left us. My recollection of that argument was as clear as any memory I had, but that wasn’t the only reason I found the video hard to believe; it was also my knowle
dge that—whatever my faults or imperfections—I was never the kind of father who could say such a thing to his child.
Yet here was digital video proving that I had been exactly that kind of father. And while I wasn’t that man anymore, I couldn’t deny that I was continuous with him.
Even more telling was the fact that for many years I had successfully hidden the truth from myself. Earlier I said that the details we choose to remember are a reflection of our personalities. What did it say about me that I put those words in Nicole’s mouth instead of mine?
I remembered that argument as being a turning point for me. I had imagined a narrative of redemption and self-improvement in which I was the heroic single father, rising to meet the challenge. But the reality was…what? How much of what had happened since then could I take credit for?
I restarted Remem and began looking at video of Nicole’s graduation from college. That was an event I had recorded myself, so I had footage of Nicole’s face, and she seemed genuinely happy in my presence. Was she hiding her true feelings so well that I couldn’t detect them? Or, if our relationship had actually improved, how had that happened? I had obviously been a much worse father fourteen years ago than I’d thought; it would be tempting to conclude I had come farther to reach where I currently was, but I couldn’t trust my perceptions anymore. Did Nicole even have positive feelings about me now?
I wasn’t going to try using Remem to answer this question; I needed to go to the source. I called Nicole and left a message saying I wanted to talk to her, and asking if I could come over to her apartment that evening.