It is March 7, 1955. Montreal lies underneath the cover of late winter snow. It is cold—but not too cold—underneath your hospital gown. You await a potentially life-altering surgery in the world-famous Montreal Neurological Institute, just blocks away from McGill University. Now stop. Breathe deeply for a second. Listen to the scratch and scrabble of the electroencephalogram (EEG) in the background, the buzz of the sterile, white lights, and the clatter of scalpels behind your head. The anesthesiologist has hooked you up to a little vial full of nupercaine, a localized anesthetic. You stare blankly into the crowded room while the surgeon—an amiable man named Wilder Penfield—prepares to cut into your brain.

You are in your twenties but have lived with petit mal epilepsy your entire life. Dr. Penfield has promised you a solution: he will peel back your scalp, open up your cranium, and probe your brain with electrodes for signs of injury. He believes that your epilepsy is due to a head wound that you suffered in childhood—one that healed and distorted your grey matter in the process. Removing this lesion properly will ease, if not end, the seizures.

The procedure starts quietly, even tenderly, as Dr. Penfield chitchats with one of the nurses. You are nervous but try to still yourself for the task ahead. Dr. Penfield needs you lucid and attentive to find the epileptic focus. As he gently tests parts of your brain with the electrified metal prongs, he drops sterilized pieces of paper to mark the areas he has just touched. To the side, a Dictaphone records Dr. Penfield’s observations.

Most of the attempts at stimulation—the large majority of them near the central sulcus—produce relatively mundane results: a tremble of the hand, a twitch of the leg. Now and then, electrocution produces no visible effect. In moments such as these, Dr. Penfield feels tempted to glance over at his anesthesiologist to check if he, too, saw nothing. But then you speak up. “I heard a voice,” you say. (Or perhaps it was an image. A smell. A sensation.) With your consent, Dr. Penfield shocks the locus again.

Your vision and consciousness double for a while—you see and feel a memory of your sister with her child. Or one of your late mother as she combs her hair, of your best friend’s husband as he reads the newspaper, of your favorite parakeet as it hops in and out of your hand. This vision hovers in midair, superimposed on the surgical tent and the face of the bored anesthesiologist in front of you.

In speeches and articles to the neurological community, Penfield compares this phenomenon to a “filmstrip”—a homemade movie of your entire life, stored in the brain and elicited by electricity. And Lawrence Kubie, a psychoanalyst who collaborated with Penfield, refers to the procedure as “an electrically-stimulated Recherche du temps perdu,” a reference to Marcel Proust’s labyrinthine novel, which relies heavily on flashbacks and explores the theme of “involuntary” memory.

Others, such as the behaviorist psychologist Karl Lashley, the psychoanalyst Mortimer Ostow, and the cognitive psychologists Elizabeth and Geoffrey Loftus, were more critical. All of them harbored doubts about Penfield’s feat.

But the good doctor’s research raised the possibility of fully and accurately recovering memories, a fact that some argued could have tremendous implications. Psychoanalysts were especially excited by the potential this surgery held for the study of trauma and repressed memory. Kubie, for one, was confident that the Montreal procedure could help scientists understand “both the psychodynamics of repression, and also its neurophysiological substratum.” Years later, during a spate of child abuse and mistreatment cases, lawyers and their expert witnesses would also cite Penfield’s research to defend the trustworthiness of recovered memories as testimony before a court of law.

Journalists for TIME and other publications were less attentive to the therapeutic and legal dimensions, but were equally enthralled by the possibility of recreating “an entire segment of a person’s past life, complete with sight, sound, light, color, dimension, smell and the appropriate emotional feeling.” They compared Penfield’s filmstrip to state-of-the-art technologies—high-fidelity stereo, Kodak motion picture cameras, and sophisticated tape recorders—that, at the time, were just making their way into North American homes.

If Penfield’s procedure sounds like something straight out of science fiction—or, more specifically, like episode three of the first season of Black Mirror—then so be it. The point of this anecdote, which I have based largely on the descriptions of Penfield’s procedure in Katja Guenther’s article “Between Clinic and Experiment,” is not to defend the plausibility or even the desirability of this state of affairs.

Although neurologists who have studied the matter do not question the thoroughness and precision of Penfield’s observations, many of them point out that situations like the one described above are relatively anomalous. (By Elizabeth and Geoffrey Loftus’s count, only 3.5 percent of Penfield’s patients reported involuntary memories.) That does not mean that it is impossible to electrically “jog” someone’s memory. For the time being, however, Penfield’s process remains impractical and unreliable.

By the 1970s, Penfield’s work had fallen out of favor. As the historian Alison Winter notes in Memory: Fragments of a Modern History, Penfield’s research “seemed to stand in the way of a very different approach to mind and memory being promoted by the new field of cognitive psychology.” In this approach, Winter writes, “autobiographical memories were to be seen not as inert impressions but as dynamic, constantly changing entities.” As opposed to Penfield’s localized and static view of memory storage, neurologists and psychologists like Loftus came to believe that episodes from the past were scattered across the brain, displaced over the entire cerebral cortex.

As a historian and journalist, I find this latter view of memory rather persuasive. Perhaps this is because I have been conditioned by my own experiences with interviews and archives—because I see my métiers as herculean efforts to string together disparate impressions and documents of varying quality and fidelity, each deposited in different locales. More importantly, this view acknowledges the subjectivity involved in memory: it explains how we can reassemble the same episode in several different ways, how we can mix snippets from various places and times to create a composite whole.

But the historical methods that I learned in school are, in fact, relatively new. In the 1940s and 1950s, such prominent historians as Lucien Febvre and Marc Bloch aspired to a completeness and scientific impartiality in their histories similar to the completeness of Penfield’s “filmstrip.” They called this histoire totale, or total history, a name that betrays their project’s imperial ambition.

Alas, the first-generation Annales school’s dream is no more.

Over the past couple of decades—with the introduction of digitized archives and the proliferation of internet resources—a discipline that was once information-poor has become information-rich. As the historian Renée Sentilles observes in her article “Toiling in the Archives of Cyberspace,” there are paralyzing pains associated with this shift.

Historians, once trained to comb through every source in detail, instructed to master each pocket of material, are now advised to either radically narrow their scope—that is, become micro-historians—or strive for “functional competency” over a far larger body of texts—as countless global historians have tried to do.

In both cases, the underlying problem is the same: a new scholar cannot learn everything about the past that she hopes to narrate—cannot have an histoire totale—so she must pick and choose. (Or, in some cases, mix and match.) In the process, even micro-historians will probably have to call on material from many far-flung archives, just as Lashley once suggested happens when patients use an assortment of brain structures to remember an event.

One strategy for selecting historical documents is to follow a name or set of names across an archive. The French and Italian micro-historians who popularized this tactic believed that names were relatively stable identifiers that could then serve as a fil rouge, a guiding thread, for the narrative.

This, too, has parallels in Penfield’s research. After repeating the electrode test on several other patients, Penfield found that the few memories he could elicit were unimportant ones. He concluded that these recollections were in a deep freeze of sorts, hidden away from the neural pathways that were conventionally associated with memory. In a bold flourish, Penfield then hypothesized that these memories were subconscious tools—records of faces, voices, tastes, and smells—that could then be used to identify people and materials later on. And as identifiers, these little vignettes served more or less the same function that names did for French and Italian micro-historians.

(Today’s Internauts might most profitably compare these memories to the biometric recognition signatures that proliferate on social media—what Penfield had found, in other words, were proto-GIFs that the brain could use to recognize friends and acquaintances later on.)

When I take photographs, write in a journal, or otherwise attempt to document my life, I try to distinguish between two sorts of memories. I certainly want to preserve episodes that have a net impact on my sense of self, those which Penfield imagined were buried in parts of the brain that his electrodes could not access.

At the same time, however, I hope to keep track of fil rouge memories—names, numbers, and addresses—that will allow me to act on the narrative that I have strung together, to pursue the people and experiences that have been meaningful to me in the past.

And finally, I long to chuck into the trash all the redundancies that take up file space on my computer and shelf space on my bookshelf! As the Argentine writer Jorge Luis Borges points out in “Funes the Memorious,” a short story about a man who remembers everything, too much memory is painful. We occasionally have reason to forget, shred records, swipe our hard drives clean of eons upon eons of accumulated junk.

For the weekend scrapbooker, the ratio of trash to treasure is a personal choice. No matter how much of our material past we choose to “forget” through destruction, however, we can be sure of a couple constants. First: that each act of personal memorialization or forgetfulness is implicitly an “act of selfhood,” an attempt at defining one’s own identity. And second: that “acts of selfhood” are not necessarily “acts of life,” that is, depictions of lived reality.

In other words, because significant parts of our existence are subconscious, any attempt to capture “life” in the narrow boxes of the “self” is futile. We use pictures and diaries to tell ourselves stories about what kinds of people we are or aim to be. But when we sleep at night, there is so much more. We become organic, conflictive, incoherent—we contain and wrestle with our multitudes.

Many historians also struggle with the organic and contradictory nature of their sources. This is especially true of scholars working on the history of what were once considered “subaltern” subjects—marginal and irrecoverable people and events, lost or nearly lost to history and the law—who suddenly find themselves awash with new materials: newly unearthed court records, declassified documents, interviews conducted by Truth and Reconciliation Commissions, statistical time series, so on and so forth.

Despite the surplus of data, the dilemma before these historians has not changed. Do the new sources actually fill all the gaps in the historical record? Or do certain experiences remain—as the literary theorist Gayatri Spivak so powerfully emphasized—beyond recall?

Ultimately, I believe that humility is key. Unlike the psychoanalysts who drew on Penfield’s work, believing that they could “evoke unconscious memories” that had been repressed, historians should be wary of making oversized claims about their marginal subjects. Otherwise, they risk producing a newfangled “total history” that not even Clio would recognize.

Sergio Infante is an assistant editor at Foreign Affairs. Follow him on Twitter @SergioDInfante.