A few weeks before the release of my first book, a memoir about my mother’s murder, I had to take a polygraph exam. The two things were not in fact related, but that was easy to forget once I found myself strapped in a chair in a windowless room on the fourth floor of a federal building in El Paso, with some polygrapher I’d just met sitting behind me, asking questions.
Article continues below
I’d met my examiner, whom I’ll call Kevin, that morning. The federal scheduler had insisted on a 9:00 am appointment even though I lived four hours away, which meant I’d spent the previous night alone in a Motel 6 by the highway in El Paso, eating Del Taco and reflecting on the decisions that had led me to spending the night alone in a Motel 6 by the highway in El Paso, eating Del Taco. Technically, I was there because I’d applied for a job with Customs and Border Protection. But the truth seemed much more complicated than that.
I didn’t get much sleep, and showed up at the federal building early, dressed in what I imagined the government meant by comfortable clothing: black dress pants, plain white oxford, no tie. I looked like a banquet waiter. One other guy was in the waiting room when I walked in. As we sat there past the scheduled time of our appointments, we struck up a desultory conversation. Like me, he’d been in the hiring process for years, had driven down from Albuquerque the night before, and seemed nervous. He asked if I’d done any research on the polygraph. I said no, and asked him the same question. He said no. We were getting our first lies out of the way.
The lie detector was like any true story in America: the facts didn’t matter as long as a lot of people believed it.
The government’s guidelines had repeatedly stressed that we should not do any research before our polygraph. Their insistence struck me as odd: if the machine detects lies, why would it matter? I’d spent much of the previous week Googling polygraphs. I was in the middle of denying it to my new friend when the door opened and Kevin appeared.
He was around sixty, short, portly, bald, with a silly goatee and wire-rimmed glasses, wearing a baggy gray suit and a shirt in one of those colors I never can keep straight, puce or mauve or periwinkle, a little too festive for the occasion. Kevin was squinty and smug, with an air of hollow authority that reminded me of my middle-school principal. He didn’t even step into the waiting room, just swung the door open and shouted my full name. I stood and shook the hand he extended. He squeezed too hard and said I could call him by his first name, as if he were doing me some great favor.
Kevin led me down a drab hallway to a door on the left that led to my first disappointment. I’d been expecting the sort of tableau you see in cop movies, some dank cellar with a dangling bulb and a two-way mirror on the wall for me to stare defiantly into. Instead we entered a bare, sterile room with office chairs on either side of a desk. Wires ran from the computer on the desk across the room to a hard-backed chair festooned with cuffs and straps and sensors.
We sat in the normal chairs. Kevin leaned back in his, twirled a pencil, and said, “Let’s get started.” I knew this bit, the casual tone and performative warmth. I was a college professor, sort of, and I did this same bit on the first day of classes, trying to make my students trust me. In that first fluorescent moment, staring into Kevin’s beady eyes, I had a premonition: I was going to fail my polygraph exam.
*
Joan Didion once wrote that it’s easy to see the beginnings of things, and harder to see the ends. That has not been my experience. Ends are obvious: divorce, death, getting fired. Beginnings, on the other hand, seem subjective. If ends are facts, beginnings are truth: relative, random, subject to belief.
The story of my exam, for example, could begin two years before I met Kevin, when President Obama signed an act requiring polygraph screening for applicants to certain federal agencies. Or two years before that, when I started writing a memoir and began to understand what it means to tell the truth. Or it could begin with the polygraph itself, the kind of story America likes best: a simple one that isn’t all that true. The polygraph’s most commonly credited inventor is John Larson, an employee of the Berkeley Police Department, who developed a new device for interrogations in 1921. Larson was twenty-nine at the time, and, like me, a strange candidate for law enforcement: he might have been America’s first cop with a PhD.
His degree was in physiology, the science of the body’s systems. The then-prevailing scientific belief saw crime as biological, either hereditary or the result of a physical defect. Larson explored both possibilities. His undergraduate work tried to find familial patterns in fingerprints that predicted criminality. In grad school, Larson shifted his focus to thyroid deficiencies. The results were disappointing, so he turned to machines. Larson read an article about using blood pressure to detect deception, and decided to improve on its author’s technique by designing a machine that could do so more objectively. The polygraph was born.
The truer story is, as usual, more complicated. Larson’s machine was not so much an invention as it was an amalgam of existing devices. He didn’t believe it detected lies and didn’t call it a polygraph: Larson referred to his machine as an “emotion recorder.” His protégé and rival, Leonarde Keeler, would later come up with the term polygraph to help commercialize the device.
Polygraph organizations like to say the word means “many writings,” which is halfway true. It’s a neologist portmanteau of the Greek terms meaning exactly that, and the machine does indeed create many writings. But to claim the word means only and exactly one thing is to make the same mistake with language polygraphers habitually make with facts: believing that they’re static and absolute. Language, like truth, is neither. Words evolve and change over time and mean different things in different contexts. Polygraph has six different definitions, according to the Oxford English Dictionary—three of which predate the machine—and they range from a letter grouping in cryptography to a person imitating another to a writer of various works. (Which I guess makes this an essay about a polygraph taking a polygraph.)
More to the point, polygraph does not mean “lie detector.” Larson himself repudiated that term for the rest of his life. But that fact didn’t get in the way of a good story. Once the polygraph was adopted by police across America and heralded in the popular media, it took on a mythical new name: the lie detector. And as soon as the lie detector became famous, a bunch of men fought with each other for decades—mostly in their memoirs—over who was its true inventor. In his book The Truth Machine: A Social History of the Lie Detector, Geoffrey C. Bunn devotes an entire chapter to the question of who invented the device and offers enough credible candidates to field a baseball team: everyone from Carl Jung, who helped pioneer the field of psychology, to Étienne-Jules Marey, who did the same for cinematography. As Bunn puts it, with magisterial restraint, it was a “curious and notable fact that the lie detector’s principal actors mistrusted each other intensely.” Maybe they should have taken a polygraph to resolve the question.
*
My polygraph test was the final step in a process that had begun three years earlier, when I started applying for government jobs. At the time, I was living in San Francisco, teaching at Stanford, and was nearly finished with the book. While my job sounded impressive, it paid forty thousand dollars a year and was located in one of the most expensive areas of the country. I was in the second year of a two-year contract and had been trying unsuccessfully to get a real teaching job for years, something tenure-track, or at least a lecturer gig in a more affordable city, one with benefits and some semblance of job security.
When I was “on the market,” as they say, I applied to teaching jobs in Fairbanks, Alaska; Birmingham, Alabama; Spokane, Washington; Camden, New Jersey; Cullowhee, North Carolina; and on and on. I spent a lot of time back then browsing Wikipedia pages for cities I’d never been to, trying to convince myself I could live there. It was a moot point. Despite dozens of interviews, I couldn’t get a professor job. My memory has condensed the experience into one vivid example, when the writing faculty of a football powerhouse down South made me fly to Chicago, in January, on my own dime, so I could sit in their hotel suite sweating through my freshly pressed suit while they asked a series of oddly combative questions for an hour and a half, after which they never bothered to contact me again to tell me that I didn’t get the job I wouldn’t have taken if my life depended on it.
Somewhere along the way, I decided to give up on academia and find a less demoralizing line of work, something with better pay, more stability, maybe even a union. I applied to be a technical writer, an FBI agent, a cop. I didn’t have any luck with those, either. My brother suggested the Border Patrol. He was an agent, and so were a half-dozen of my childhood friends and my former baseball coach. They were always hiring. The starting pay was nearly double what I made at Stanford, with much better benefits, and I might be able to move back to Arizona.
Most of my old friends from home thought joining the Border Patrol was a good idea. I could be closer to family, make a living, maybe even buy a house. My social circle in San Francisco was another story. By then, I’d been in academia for more than a decade, surrounded by white liberals from wealthy backgrounds. When they heard I was trying to join the Border Patrol, my new friends all said the same thing: why?
I didn’t understand the question, and soon determined it was a matter of perspective. For people who grew up wealthy, or even middle-class—whatever that means—the proper career path seems to be to find a job that’s rewarding, fulfilling, whatever. I’d done every job I’d ever had for one reason, the same reason I was trying to join the Border Patrol: money. Did I want to drive around the border in an SUV detaining people? Of course not. But I didn’t want to help tech billionaires write their memoirs, either. At least the Border Patrol paid a living wage.
But the Border Patrol is racist, my friends said. And that might be true. But cries of racism rang hollow coming from people who worked in academia, one of the whitest industries in America. The Border Patrol is a hell of a lot more diverse than the average writing faculty.
Their underlying point was right, though: if I joined the Border Patrol, I’d be complicit. But I was already complicit. The longer I spent in higher ed, the more it seemed like an engine driving American inequality. At least the Border Patrol gave working-class people a path to economic mobility without being saddled by a lifetime of student loan debt. My brother had begun his career at the same time I’d started grad school. Nine years later, he made twice as much as I did, owned a house and a new Acura, had health insurance and a retirement account and the whole American dream. Meanwhile, in the three years since I applied to the Border Patrol, I’d taken another teaching job in Albuquerque, making forty-five grand, with shitty benefits and no long-term security. Besides, I still hadn’t decided to actually join the Border Patrol. I hadn’t been offered the job.
By the time I went to El Paso, I’d passed a four-hour written exam, a physical, a fitness test, an oral interview, and a background check so intensive that they’d talked to my coworkers, every neighbor in my apartment building, and my high school teachers. Once I passed the background check, I waited more than a year for the call from the polygraph office.
That call would be the first sign that the government and I had different ideas about truth. First, I got an email from a scheduler saying he’d been trying to reach me. I hadn’t received any messages, so I knew that wasn’t true, but had no way to prove it. I called the office and spoke to another man, who said he was a quality control agent, and that he’d call me back soon with a date and time for my appointment. He never did. Then I received a letter saying I’d been removed from consideration for refusing to take the polygraph. I called Mr. Quality Control. He accused me of lying, but grudgingly scheduled my appointment. So there I was, in a building full of liars, about to have mine detected.
*
Kevin started off by reminding me of the rules. One of them was that I could not discuss my test with anyone afterward. I’m sure he would have told me that I couldn’t write about it either, although I made sure not to ask. Clearly the government believed, despite all historical evidence to the contrary, in its power to control information. I believed my right to free speech was inalienable. In that sense, I guess this essay is a lie detector, too: we’re going to find out who was right.
Kevin ran me through some questions that might be on the test. Some were pedestrian, my name and address and so on. Others were bizarre: questions about bestiality, child porn, terrorism.
When he got to a question about my past drug use, Kevin’s tone changed. His smile fell and he held eye contact. The interrogation had begun. I told him the same well-rehearsed thing I’d told my background investigator: I experimented with drugs a few times in high school and college. It was a lie. Kevin seemed to detect it.
“What do you mean by few?” he asked.
“Not many.”
“Could it mean five?”
“I guess it could.”
“Don’t be a smartass.”
I looked around the room. It couldn’t be just me and Kevin. Surely there was a witness somewhere, a hidden camera, a recorder, anything to prove we actually said what his report would say we had. Later I would learn that federal polygraph protocol requires examiners to make audio recordings of exams, and that many examiners have been accused of ignoring that requirement, as well as an array of other sordid and unprofessional practices. I don’t remember Kevin saying anything about a recording, or seeing a device. But there’s no way to say for sure.
Kevin sighed elaborately and asked the maximum number of times I could have used illegal drugs. “Was it six? Eight?”
“Sure,” I said. “It was a long time ago.”
“Could it be ten?” he asked, with a smug little grin, and finally I caught on.
“No. Definitely not ten.” If Kevin wanted a fixed and certain truth, I’d give him one.
He wrote “6–8 times” on his notepad.
“I said it could have been that many, not that it was.”
Kevin said to let him do his job, and a shroud of dread descended. The test hadn’t even begun, and I’d already found myself at epistemological loggerheads with the federal government. Kevin thought he’d convinced me to tell the truth, which was that I’d done drugs between six and eight times in my life. I thought there was no truth to tell. I didn’t remember how many times it was. A few minutes earlier, Kevin had said there was no maximum threshold for drug use that would disqualify me—a bald-faced lie I had not yet detected—but I wasn’t about to say that the real number was closer to a hundred. So the lie became the truth: I’d done drugs six to eight times.
The polygraph works a lot like a memoir. It doesn’t find the truth, it creates it.
Almost immediately, I began to believe it. As I sat there watching Kevin scribble notes, a handful of specific drug experiences returned to me. The first time I got high and drunk at the same time, in eighth grade, and spent the night in my friend David’s bathroom with my head on the toilet seat while he tried to convince his mom that I had food poisoning. When Charlie taught me the trick where you blow the smoke through a paper towel tube with a fabric softener sheet rubber-banded over the end. The first line of meth I ever snorted, off a Bone Thugs CD case in Jeremy’s bedroom. The cooks at my first restaurant job going around at closing time on busy nights, handing out key bumps. The eight ball I went in on with two ex-con dishwashers at my second restaurant job, who told me they’d pay me back double in a week, but then Mike got stabbed and I realized I wasn’t cut out for that life. I never saw that money or the drugs, so maybe that one didn’t count.
But that was all during one relatively short period of my life, when I was a shithead teenager. I’d come so far since then. I was a college professor; technically a visiting assistant professor, but still. My memoir was about to come out. The sitting president at the time had admitted to using marijuana and cocaine in his memoir. Why should I be banned from a job for being a small-time delinquent twenty years ago? Remove those few wayward years and it was true enough that I’d only done drugs a few times in high school and college. As we say in the memoir business, it was my truth.
*
My new profession, memoirist, had a complicated relationship with truth, to put it mildly. Fake-memoir scandals have erupted more or less continuously as long as America has existed, from James Frey and his contemporaries, to the so-called autobiographies of Howard Hughes and Davy Crockett, to fantastical captivity narratives of the colonial era and dubious accounts of European explorers in the New World. Sometimes it seems like every notable American figure wrote a fictionalized memoir—even Wyatt Earp, the patron saint of my hometown. (Frontier Marshal. It’s a hoot.) I fudged plenty of facts myself, combining real people into composite characters, changing the order of events. Most memoirists do the same. The point of the genre isn’t accuracy or precision. The point is to tell a good story.
The same applies to the polygraph. There’s no real evidence for the machine’s accuracy. Its purpose is to monitor the body’s physical response to stimuli, but the body’s response to lying is indistinguishable from its response to any other stimulus. Even the telltale spike in the polygraph chart, itself largely a myth created by TV and movies, can indicate anything from a heart problem to sexual attraction. (Indeed, the machine’s inventors used it to detect both of those things. Larson married one of his first test subjects, and Keeler discovered a heart defect while testing the machine on himself.)
But from the early years of the machine to the present day, its proponents have told the same story of an infallible machine that detects lies. Polygraph organizations routinely estimate its accuracy at nearly 100 percent. Most of those estimates are invented out of thin air, and the few based on data suffer from an obvious sampling error. As early as 1939, Walter Summers—yet another purported inventor of a lie detector—pointed out the fundamental flaw on which all polygraph statistics are based: they “fail to relate the number of instances in which deception was actually practiced in a manner which eluded the examiner and the instrument.” You can’t detect the lies you can’t detect.
Independent studies suggest a polygraph exam is roughly as accurate as a coin flip, and that polygraph operators find as many as half of innocent subjects guilty. The scientific case against the polygraph is so compelling that the National Academy of Sciences, the American Psychological Association, the Congressional Office of Technology Assessment, and the United States Supreme Court have dismissed it as unreliable. A federal law forbids using the polygraph to screen applicants to private companies. For most jobs in America, an exam like mine would’ve been against the law.
In fact, the only people who seem to believe the polygraph is accurate are its operators. Every study I’ve found that supports the machine’s ability to detect deception was funded or performed by polygraphers. They’ve formed half a dozen different organizations dedicated to spreading the lie of the lie detector. Their websites are ironically similar to the polygraph itself: archaic, slipshod, rife with bias and bullshit. The International League of Polygraph Examiners calls the device’s invention “officially one of the greatest of all time,” and claims the accuracy of contemporary polygraphs is close to 100 percent. The American Polygraph Association, which claims to be the largest organization of polygraphers, has a section of its website devoted to Polygraph Validity Research.
It begins by stating the organization “believes that scientific evidence supports the validity of polygraph examinations.” The site includes a link to what seems to be the entire basis of that belief, the “Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques.” The document was prepared by a team of polygraphers, and it reads about how you’d expect. I made it far enough to learn a few astounding facts, including that until 2012, the American Polygraph Association did not require members to use methods supported by published research. In other words, for the first ninety-one years of its existence, polygraphers literally had no scientific standards. Luckily for me, they came up with some just in time for my exam.
The lie detector was like any true story in America: the facts didn’t matter as long as a lot of people believed it. And we wanted to believe it: the notion of a lie detector existed long before the polygraph did. Tellingly, it first appeared in fiction. Bunn traces its first known usage to Charles Walk’s 1909 detective novel The Yellow Circle, in which a character fantasizes about having a machine he calls a lie detector: “With its aid one can plumb the bottomless pits of a chap’s subconscious mind, and fathom all the mysteries of his subliminal ego.” In 1914, seven years before anyone claimed to have invented a polygraph, G. K. Chesterton mocked the notion of a lie detector in his mystery story The Mistake of the Machine, comparing it to the Dark Ages belief that blood would flow from a victim’s body if their murderer touched it. Bunn lists many other instances of lie detectors appearing in fiction long before anyone claimed to have invented one.
Meanwhile, the popular American media seemed obsessed with the idea of machines that could see inside our heads, hearts, and souls. Fourteen years before Larson’s polygraph debuted, the New York Times ran an article rhapsodizing about Jung’s electric psychometer. That “mysterious little machine” purported to detect emotions, not lies; still, the article foretold a future when it would be used to detect guilt, making criminal courts superfluous. Four years later, the same paper ran a two-page profile of the “big-hearted” men, Edward Johnstone and Henry Goddard, who’d been doing the “self-sacrificing work” of testing the psychometer on developmentally disabled children at an institution in New Jersey. Under Johnstone’s supervision, Goddard hooked kids up to a machine like Jung’s. Their intent was to study and eradicate “feeble-mindedness”; Johnstone was a noted member of the American eugenics movement. But the article breathlessly predicted a future in which the lie detector would replace the “impedimenta” of American justice, like judges and juries.
When the story started circulating that a cop with a PhD had invented a lie-detecting machine, a myth became a widely reported fact almost overnight. Within a year, the San Francisco Examiner claimed, “everyone has heard of the ‘lie detector.’” By then, polygraph results had already been banned from American courts, by a judge who was less enthused about the prospect of a machine replacing juries. It would be the first of many official dismissals of the polygraph. But it didn’t dampen the media’s fascination with the so-called lie detector—and, by extension, the American public’s.
That fascination was furthered by the polygraph’s sister inventions, psychology and cinematography, which were created by some of the same men at around the same time. In addition to his psychometer, Jung also helped create the field of modern psychology. Marey developed a different forerunner of the polygraph, as well as chronophotography, an important step in the development of cinema. Marston pioneered the idea of detecting lies based on physical responses, and later became a Hollywood censor and wrote a guide for aspiring screenwriters. His academic mentor, Hugo Münsterberg, helped lay the theoretical groundwork for the polygraph, and also published one of the earliest works of film theory.
The rise of psychology helped drive a cultural fascination with discovering the inner workings of the mind—especially the criminal mind—a desire the lie detector satisfied. Meanwhile, cinematography, the notion that we could document and preserve reality itself, had embarked on its ongoing project of destroying our cultural distinctions between fiction and fact.
Throughout its history, the polygraph has moved freely between the two realms. As it spread to police departments across the country, who used it to investigate real crimes, the machine also began to appear in advertisements and movies. The machine may have made its screen debut in the 1926 silent-film serial Officer 444, alongside Vollmer, who played an idealized version of himself, a criminologist using science against evil. By the 1930s, Marston was using the polygraph to screen-test Hollywood films, including Frankenstein—ironic, considering that Larson once compared his invention to the Monster—and to sell razor blades, gasoline, and cigarettes. In 1941, Marston invented Wonder Woman, a female superhero whose primary power was a Lasso of Truth, similar to a lie detector. Marston called the comic “psychological propaganda,” and it was hugely effective: within a year, Wonder Woman had her own comic book with a circulation of half a million. In 1946, Keeler starred in a noir movie alongside his version of the machine. The first TV show called Lie Detector debuted in the fifties; there have since been a handful of others, both scripted and reality, not to mention a continuous stream of polygraph appearances in film and television. While writing this essay, I watched lie detectors play prominent roles in two different TV shows. In one, the polygraph is accurate; in the other, it isn’t.
*
Kevin slid a blank sheet of paper across the desk and told me to draw the number five inside a black circle. I did, and slid it back. Kevin drew other numbers in other circles and said now he was going to hook me up to the machine.
He told me to take off my shoes and empty my pockets, then directed me to the chair. I sat on one sensor and put my feet on two others. Kevin wrapped two cords around my chest, slid a sphygmomanometer over my left arm, and stuck metal clamps on my right index and ring fingers. As he pumped up the cuff, Kevin asked if I was comfortable. He didn’t seem to be joking.
He sat behind his desk and said he was going to point to all the numbers on the paper and ask if I’d written them. I should say no every time, even for the one I’d written. He did. I did. He unhooked me, led me to the desk, and pointed to the lie on the computer screen. It looked like a lot of squiggly lines to me.
“Now we can take a break,” Kevin said. I looked at the clock, which wasn’t visible from the polygraph chair; we’d only been in the room for half an hour. Kevin smiled inscrutably. “Bathroom and water only. Be back in ten minutes.”
I wandered into the hallway, drank from a fountain, leaned against another beige wall, and tried to calm down. I was not then in a great place, psychologically speaking. I stayed up until sunrise a few nights a week, spent days on end inside my apartment, often went blank with anxiety in front of my classes, and was preoccupied by a vivid and persistent vision of myself swan-diving off my balcony. I would later be diagnosed with various conditions and embark upon a reasonably successful therapeutic journey, but right then, in that hallway, I was freaking the fuck out. My shirt clung wetly to my chest, where I could see my heart beating as if it was trying to escape, like the alien in Alien. If I had a heart attack in the chair, what would that look like on the polygraph readout?
I’d tried to learn techniques for managing anxiety. Most of them didn’t work— picture a beach, my ass—but a shrink I’d briefly seen had suggested imagining the worst possible outcome, and that seemed helpful. The idea was that you embrace the notion of failure and realize it wouldn’t be so bad, thereby relieving the pressure not to fail.
I tried it. What if I failed the poly? I’d go back to Albuquerque, keep teaching, apply for more jobs. This was my backup plan, which made me luckier than pretty much everyone else applying. Then again, that was not the worst-case situation. One problem with that exercise is that I could always come up with something worse. What if I got in a car accident on my tired four-hour drive and spent the rest of my pain-filled life alone in my rented house in Albuquerque? What if I passed the poly, got the job, and actually took it—got sent to some borderland armpit like Ajo or Wellton where I’d have to herd other human beings into the back of trucks? Woke up two or ten or thirty years down the road and didn’t recognize myself?
By the time Kevin came to get me, a few minutes sooner than the ten I’d been promised, I’d almost accepted my imminent failure. If my anxiety didn’t make me fail the test, something else would. I remembered something from my sorta-research about Catholics failing the polygraph at higher rates. Technically I was Catholic, baptized and confirmed, now lapsed, but that only made things worse. Maybe I’d ask Kevin how to become a polygrapher. How much training did it require? Did he enjoy it? How much money did he make? Later, I’d search around online and find out that the average polygraph examiner makes even less than I did at the time. Then again, the training only takes ten weeks, and there are actually jobs in that field. I’d been training for years to be a nonfiction professor and still had no idea what truth actually meant; maybe I should’ve just taken a polygraph course and become an official federal arbiter of facts, an asshole demigod like Kevin.
I shouldn’t be so hard on Kevin. Judging by his clothes and demeanor, he probably came from a similar background to mine. Maybe polygraphing was his version of teaching, a thing he did to pay the bills because it was better than his other options. Maybe he had a whole life to maintain, a family, a little house in some cul-de-sac on the West Side, two Toyotas and a swing set.
While Kevin strapped me back into the chair, I wondered what he told himself at night, trying to sleep, after watching applicants lose their best hope for a career to his machine. Kevin seemed like a smart guy, way too smart to believe in the simpleminded fantasy of a machine that detects lies. But that wasn’t his decision. It was his employer’s. And why is our government the only major employer in the world that uses polygraphs to screen prospective hires?
*
The answer to that question is based on a lie. Even the United States government isn’t dumb enough to believe the polygraph works. The machine’s real purpose is symbolic, as an icon of the power of the state. Law enforcement agencies don’t use the machine to detect lies. They use it to coerce confessions.
In its early days, the polygraph was considered a more humane version of the infamous “third degree,” the interrogation procedure it largely replaced, which involved beating the shit out of a suspect until they confessed. The third degree was itself a variation of another quintessentially American tactic, outright torture. The parallels between torture and the polygraph are obvious: the latter’s arcane parts and procedures, its use of restraints and stimuli, the gratuitous periods of waiting for what the subject knows is coming. The polygraph creates the very stress it’s designed to detect, then presents it as evidence of deception, which often leads its subject to confess. If the subject confesses, that confession effectively becomes the truth, whether it’s true or false or somewhere in between.
And the polygraph has a long history of coercing false confessions. It may begin with its maiden voyage in 1921, when Larson tested his new device on the residents of an all-female Berkeley dorm that had experienced a rash of petty thefts. Thanks in part to the polygraph, a suspect admitted to most of the thefts and withdrew from the university. But the crimes continued, and Larson himself doubted the veracity of her confession.
Not long after, Larson tested a man named Henry Wilkens who was accused of having his wife killed. The polygraph helped to exonerate Wilkens despite evidence of his guilt. After that, police began to doubt the polygraph’s utility, and some departments refused to use it. (The media had no such qualms: it continued to trumpet the infallibility of the “electric detective.”)
A year after the Wilkens case, a young Black man named James Frye retracted his confession to killing a Washington, DC, doctor, claiming it was coerced. Using his variation of the lie detector, Marston examined Frye and declared him innocent. But the judge prevented Marston from testifying as an expert at trial, and an appeals court upheld the ruling, instituting what became known as the Frye rule, which has largely prevented polygraph results from being admissible in American courts ever since.
But the machine remains useful for extracting confessions. And the conflation of confessions and truth is yet another lie, one that’s kept the polygraph alive for the last century as a peculiarly American delusion. Confessions are usually presumed to be true and treated as such in legal settings. But recent research suggests that false confessions are common, especially in the context of police interrogations.
Despite a growing body of evidence, including hundreds of exonerations based on DNA evidence, most people don’t believe in false confessions. A recent article in the Journal of the American Academy of Psychiatry and the Law explains why:
Most lay people believe in what has been referred to as the myth of psychological interrogation: that an innocent person will not falsely confess to police unless he is physically tortured or mentally ill…the myth of psychological interrogation persists because most people do not know what occurs during police interrogations, and because they wrongly assume that individuals do not act against their self-interest or engage in self-destructive behavior, such as falsely confessing to a crime that they did not commit.
The likelihood of a false confession increases when interrogators use certain tactics, especially elements of the so-called Reid Technique, a procedure created in the 1950s by John E. Reid. I didn’t know it at the time, but Kevin used elements of the Reid Technique in my test. He conducted it in a small, barely furnished, cold room; seated me in a hard, armless, straight-backed chair; and repeatedly encroached on my personal space.
Reid first used his technique (along with a polygraph) in 1955, to extract a confession from a man named Darrel Parker who was suspected of killing his wife. Parker was convicted and sentenced to life in prison. He served fifteen years before being released on appeal because Parker’s confession was ruled to have been coerced. Eighteen years after Parker’s release, a man on death row for other crimes confessed to the murder; he did so by showing his lawyers a passage of his memoirs that described the murder in detail, a passage the legal system apparently assumed to be true.
Neither those nor the Reid Technique’s subsequent high-profile failures, including the $2m settlement of a 2012 civil case by a wrongly convicted man named Juan Rivera, have prevented it from being adopted by police departments across America. The company founded in Reid’s name, John E. Reid and Associates, claims its technique is “the most widely used approach to question subjects in the world,” and recently registered a trademark on the term.
The Reid Technique™ involves a number of tactics, from creating an anxiety-inducing environment to a list of specific steps. According to Saul Kassin, perhaps the foremost American expert on false confessions, the purpose of those tactics is to “get suspects to incriminate themselves by increasing the anxiety associated with denial, plunging the subject into a state of despair and then minimizing the perceived consequences of confession.”
Like the polygraph, the Reid Technique isn’t designed to find the truth. Its purpose is to coerce confessions. Research suggests the Reid Technique may actually make interrogators worse at detecting truth. In an independent study, interrogators trained in the Reid Technique were less accurate, although “they were more confident and cited more reasons for their judgments.”
Myths exist for a reason, to explain collective phenomena, to explain us to ourselves.
The polygraph itself is not required for the Reid Technique, but it helps. Together, they have a long and checkered history of producing false confessions. In 2013, the Chicago Tribune found a pattern of false confessions obtained via polygraph exams and the Reid Technique, by examiners who routinely ignored accepted standards, including failing or refusing to record interrogations.
*
Kevin said the first battery of questions would cover my character, and asked if I had any questions. I did, but too many, and where to start? So I said no, and Kevin started the exam.
He asked me a battery of eight questions four times in different orders. By the time I typed notes on my phone after the test, I’d already forgotten one of them. The other seven were:
1. Have you misrepresented your past drug use?
2. Have you lied about participating in serious crimes?
3. Have you falsified info on forms?
4. Have you ever cheated to get ahead in your personal life?
5. Have you ever made disparaging comments about your supervisor?
6. Is the light on?
7. Have you taken a drink of water today?
Except for the last two, all of them seemed open to interpretation. For instance, I absolutely had misrepresented my past drug use, but only the number of times, not the drugs or the fact of doing them. And could my estimate be a lie when there’s no way to know the exact answer? What crimes are considered serious? What counts as cheating? Who gets to say? Has any employed person in America not made a single disparaging comment about a supervisor?
The first time through the questions, I tried to follow Kevin’s direction to answer quickly, yes or no, and to abide by his somewhat contradictory instructions to breathe normally while staying absolutely still. The second time through, my voice began to crack, and I swallowed.
“Stop!”
I turned my head to see who he was yelling at.
“Stay completely still!”
I turned back to the wall and tried to comply. Kevin kept yelling, asking combative and rhetorical questions: was I trying to beat the test, did I want to fail? I tried to calm myself by imagining something peaceful, although that was probably considered a countermeasure, and anyway, it didn’t work: I visualized ripping off the electrodes and punching Kevin. I tried the box-breathing technique I’d once learned from a veteran stepdad. That worked better. Possibly too well. Soon I caught myself nodding off.
That probably sounds like a lie. How could someone under that much stress be sleepy? Have you ever been interrogated? I don’t mean metaphorically, having a difficult conversation, confessing something to a spouse, parent, priest, boss. I mean actually interrogated, by a professional. No lawyer, no witnesses, nobody on your side. And he has a machine that says he’s right, not to mention the backing of the Department of Homeland Security, a vast and unaccountable agency built on the lie that policing and surveilling Americans will protect us from terrorism. Suddenly, this part of the Border Patrol application process made a grim kind of sense: I was getting a little taste of how an immigrant might feel. Except I deserved it. I’d signed up for this.
If you haven’t been in that situation, maybe you think, like I did before it was proven otherwise, that you’d be one of the exceptions. You’d beat the polygraph, like people do on TV. But that’s the thing: you can’t beat the polygraph, because the polygraph isn’t a lie detector, isn’t a test, isn’t even a machine. It’s a fact, part of a story power tells itself to justify its power. Maybe you can beat the machine— they don’t detect lies, so it’s not that hard—but you can’t beat an entire country that believes in it.
As Kevin went through the questions again, I sank into a fugue, part paranoia and part exhaustion, and lost track of time. Not what time it was—the whole idea of time. There was no past or future, only an endless present of sitting in that windowless room, strapped to a chair, wired to a machine, staring at a beige wall while a stranger I couldn’t see asked the same questions, over and over. I forgot why I was there, who I was, the truth and what it meant. At some point, I heard a sort of flutter, and my vision vibrated and jumped, as if someone had changed the reel. My mind floated up to the corner of the room and observed the proceedings from a cool remove. My memory of the rest of the test is from a vantage point outside of my body.
Jung defined this phenomenon as dissociation, the loss of a fixed and coherent identity. Reports of similar experiences were largely ignored in early psychology, but more recent studies suggest dissociation is fairly common, and can be triggered by drugs, trauma, stress, or nothing at all. Jung said dissociative states could prevent a subject from recalling important facts, among other things. “We talk about being able to control ourselves,” he wrote. “But self-control is a rare and remarkable virtue.”
The polygraph showed me I was neither rare nor remarkable. By the final time through the battery, I no longer knew what was true. For example: the first three times Kevin asked if I’d ever disparaged a supervisor, I’d rationalized, telling myself “disparaged” was a strong word. It means to regard as having little worth, and older definitions meant to dishonor or degrade; did Kevin know it came from the Old French disparagier, to marry unequally? Certainly I’d made fun of some bosses, and respected few, but I hadn’t disparaged my supervisors, per se. The fourth time he asked me the question, a crystalline memory popped into my detached head, a moment a few years before when I told my then-girlfriend that my then-boss was a fucking idiot.
“No,” I said. Kevin moved on.
Later I would learn that I wasn’t the first lie detector subject to report experiencing dissociation. I’m not even the first one to write about it. As an undergraduate at Harvard, Gertrude Stein worked in a laboratory run by Hugo Münsterberg, who came up with the earliest scientific rationale for lie detection. Stein’s first published work, an 1894 essay originally written for her sophomore composition class titled “In the Psychological Laboratory,” was an account of her experiences in the lab, including an instance of being connected to one of Münsterberg’s primitive predecessors of the polygraph. The essay’s third person narration and distinctive syntax are both harbingers of Stein’s future work, and her knowledge of the machine seems to have informed her later experiments in “automatic writing.” (It’s also worth noting that her autobiography has fictional elements.)
But her account interests me for other reasons. She describes the experience of being subjected to an exam in front of a group of students like so:
Strange fancies begin to crowd upon her, she feels that the silent pen is writing on and on forever. Her record is there she cannot escape it and the group about her begin to assume the shape of mocking fiends gloating over her imprisoned misery. Suddenly she starts, they have suddenly loosened a metronome directly behind her, to observe the effect, so now the morning’s work is over.
What it describes sounds like dissociation, or exactly what I felt when I was subjected to the polygraph.
The scientific literature suggests dissociation during polygraph exams is fairly common. In 1996, the polygrapher Donald J. Krapohl wrote an article for Polygraph, the official organ of the American Polygraph Association, that addresses the phenomenon. The article is typical of polygraphers’ attempts at justifying their profession in the sense that it’s paranoid, authoritarian, proto-fascist, and presents the opinions and experiences of a single polygraph operator as if they’re commandments carved into tablets. Krapohl begins with a blithe, moralistic tirade about “the phenomenon of mendacity” that “pervades every class and culture.” Lying, he claims, is endemic to certain types of people, having “served to defend or expand the interests of uncounted generations of monarchs, merchants, spouses, debtors, knaves, and saints.”
Kraphol attempts to codify four classes of countermeasures. The first, Physical Countermeasures, includes any instance in which a polygraph subject “use[s] movements in hopes of masking their reactions or misdirecting the examiner.” Of course, a polygraph subject might move for any number of reasons during an exam, including as a natural reaction to the very discomfort and stress it’s designed to cause; no examiner or machine can determine why. And one wonders why a test so supposedly accurate can be fooled by something as simple as flexing a muscle.
The second category, “Mental Countermeasures,” includes imagery, hypnosis, biofeedback training, placebos, and even personality. Notably, dissociation is considered a mental countermeasure. Here, again, is the rub: if the subject dissociates during a polygraph, who’s to say whether it’s intentional? Even Krapohl acknowledges that “the outward appearance of a dissociating subject is quite similar to that of a cooperating test subject.” In other words, nobody can tell if another person is dissociating, much less why. Even the person dissociating may not know; I certainly didn’t at the time. Studies suggest dissociation is often an unconscious response to intense stress of the sort the polygraph is designed to create in its subject. The polygraph works by stressing you out, but if the stress it causes in turn causes the subject to dissociate, they can be failed for trying to cheat.
*
Kevin gave me another break. I spent it in the waiting area, staring out over the rooftops of El Paso. The window faced east, so I couldn’t see the border, but the border is like the truth: you know when you’re close to it. After a few minutes, Kevin came and led me back into the interrogation room, where he sat me down and said I’d failed. The machine detected deception in my answers to either the drug question or the falsifying information question. I was amazed: he couldn’t even tell which question I’d lied about? And why hadn’t he detected my lie about disparaging my boss?
Kevin went on to say that I must have researched tactics to defeat the poly. My swallowing seemed to bother him to the point that he considered it evidence of deception. I said my mouth had been dry, but Kevin ignored me, telling stories about his brother who’d done drugs and other applicants who said they’d gotten high a thousand times. I didn’t know it at the time, but his psychological tactics were all elements of the Reid Technique: repeated, unwavering assertions of guilt; attempts to excuse or minimize the suspected crime; constant interruptions; professed sympathy; and, of course, outright lies.
Kevin said he was almost positive that my drug use wasn’t over the threshold. Earlier, he’d told me there was no threshold, and it dawned on me that Kevin wasn’t much of an interrogator without his machine. He kept fishing, accusing me of various lies, interrupting whenever I tried to deny, suggesting things I may have forgotten: didn’t I ever do any pills when I was a bartender? Did I really only do meth once or twice? He began pointing to the computer screen and picking up other deceptions. He detected possible lies in my academic record—I told him I’d gotten a 3.0 GPA at a state school, and why the hell would anybody lie about that?—and questioned whether I really had a master’s degree, even though I’d provided transcripts as part of my background investigation.
He accused, I denied, and it became the worst kind of male interaction, a matter of pride. He was lying, I was lying, everything we said was both true and false, depending how you looked at it. I wasn’t even hooked up to the machine anymore, so I didn’t understand why we were still talking. He’d already said that I failed. Couldn’t we call it a day?
As Kevin’s one-man theater dragged on, I thought about the long drive ahead of me, and began to understand why people make false confessions. It’s not because you don’t know the truth. It’s because the truth doesn’t matter. When the person across the table has all the power, what’s the point of arguing? It was pretty clear by then, a few hours into a test I’d already failed, that Kevin didn’t give a damn what was true. He wouldn’t be satisfied until I confessed. But fuck Kevin and his machine. I wasn’t confessing.
Kevin paused, and I thought we might finally be done until he asked about my mother’s murder: was there anything I hadn’t told him about that? It broke the spell. I dissociated in reverse, came fully into my body. My chest relaxed, my heart quieted, and I saw the situation clearly for the first time. Kevin was just some dickhead with a grift, doing a job based on lies for a government that practically invented them. Did I know anything about my mother’s murder? I’d spent five years writing about it. I was the world’s foremost authority on the subject. But I wasn’t telling Kevin. If he wanted to know, he could buy the fucking book.
I asked if I was free to leave. He shrugged. As I stood and went for the door, he asked if I would come back for another test with him if necessary. Sure, I said. I’d love to. It was the last thing I would ever say to Kevin, and I wanted to make sure it was a lie.
When I got home, I checked the internet forum and read a post by the guy who’d been sitting with me in the waiting room. He’d failed, too. His story was almost exactly the same as mine, except his test took twice as long. He must have tried harder than I did to tell the truth.
*
Three weeks later, my book came out, and I once again found myself answering questions. Whenever someone asked about truth, I’d say that I consulted the historical record when possible, but that the book was mostly based on memory. Sometimes, despite myself, that old cliche slipped out: it’s my truth. I’d think about the polygraph, Kevin’s endless questions, his assumption that a fixed, detectable truth existed in my memory.
Writing a book based on memory showed me it’s less a font of truth than a river of lies. We may tell ourselves stories in order to live, as Didion famously said, but nobody ever quotes the rest of that passage: “We live entirely, especially if we are writers, by the imposition of a narrative line upon disparate images, by the ‘ideas’ with which we have learned to freeze the shifting phantasmagoria which is our actual experience.” That sounds a lot like lying.
Even if we mean to tell the truth, the existing science suggests that memory is almost as unreliable as the polygraph. In 1885, the German psychologist Hermann Ebbinghaus did tests on himself and came up with his famed “forgetting curve,” a chart that showed we forget more than half of information within a few days. More recent studies have found that autobiographical memory—the deliberate recollection of facts, ideas, and experiences from one’s life—is not only inaccurate, but suggestible and frequently false. And factors such as depression and trauma, both psychological and physical, have been found to degrade autobiographical memory.
Then there’s the issue of stress, which also seems to have a range of effects on memory. Stress hormones like cortisol and adrenaline have been shown to aid in memory consolidation, which is, more or less, the process of storing recently learned information as memories for long-term recall. But those memories can change each time they’re remembered, through a process called reconsolidation: once accessed, the memory has to be rewritten, and it can be rewritten differently, revised just like a scene in a memoir. While stress may aid in memory consolidation, it has a profoundly negative effect on reconsolidation. Trauma has its own story to tell.
We’re all polygraphers, staring at the screens of our truth machines, proving ourselves right.
By the time I took the polygraph, my memory had already been rewritten. After five years of accessing and re-accessing memories of my mother and her death, I’d recently begun to understand how dangerous it is to write a book based primarily on memory. I don’t mean the truth: accuracy is overrated, not to mention impossible. The real danger is the sacrifice you have to make. By writing your memories, and rewriting them again and again, draft after draft, you replace them, erase them. By the time I finished my book, after revising every word half a dozen times, I didn’t remember my mother anymore. She was pages, scenes, sentences.
The polygraph works a lot like a memoir. It doesn’t find the truth, it creates it. First the exam makes you doubt or forget your memories. Then, by forcing you to re-access them again and again under stress, it literally rewrites them. Since my polygraph exam, I’ve believed that I did drugs between six and eight times before then, even though my rational mind knows that isn’t true. My experience of being polygraphed showed me that not only does the polygraph not detect lies, it manufactures them.
More than two million polygraph exams are given every year in America; it’s a two-billion-dollar industry. No other country in the world uses the poly to nearly the extent that we do, and most don’t use it at all. Why are we the only ones who build machines to detect the truth, and believe that they can, despite all evidence to the contrary? Why do we need to believe in a truth so simple it can be detected by a machine?
The lie detector is a myth. Everyone who doesn’t make money off of polygraphs agrees on that. But myths exist for a reason, to explain collective phenomena, to explain us to ourselves. Myths create a sense of community through shared belief. Judging by our obsession with lie detectors—and the fact that we continue to call them that, more than a century into this charade—the myth of a simple, detectable truth is one of the few beliefs most Americans do still share.
The lie detector came straight out of science fiction, and drifted into the realm of fact at the beginning of a century in which a succession of groundbreaking technologies would shatter and reshape our cultural conceptions of what was possible: Edison’s bulbs, Bell’s telephone, Ford’s mass-produced cars, the Wright Brothers’ airplane. By the time the polygraph came along, a credulous American public was used to tales of revolutionary, terrifying innovations that were going to change their lives forever. Some of those technologies were transforming reality itself.
Electricity was spreading across the country, quite literally changing the way people saw the world. So were telephone and radio access, which did the same for how we heard it. The polygraph’s siblings, cinema and psychology, were transforming the way people saw themselves. Meanwhile, Modernism—the artistic movement responding to those changes—was roiling the arts in every medium, revising our expressions of lived and imagined reality; the United States government banned James Joyce’s Ulysses the same year the polygraph debuted.
How it must have felt to be a human then. A man my age in 1921 might have remembered reading dime Westerns by candlelight as a boy, riding a horse to school, a time before voices—much less humans—could travel through the air. He’d seen the Progressive era rewrite the rules of society, including the victory of women’s suffrage the year before. He might have fought in the first World War. By the time the polygraph was invented, he may have owned a car or had access to a telephone, if he was wealthy. But he still got his news from newspapers or neighbors. He didn’t own a screen. Can we blame him for believing in a simple kind of truth that a new device could find?
A century later, I’m awash in new technologies, and deeply confused about what to believe. I was a young adult when smartphones arrived, in college when Facebook debuted, and dimly remember an early childhood before the internet changed everything. In the last few years, as of this writing, the COVID-19 pandemic has altered American life in ways nobody comprehends. Nobody trusts the media anymore, and appalling numbers of Americans refuse to believe in fundamental, verifiable facts. Meanwhile, conspiracies ricochet around the internet, gathering believers.
It’s hard not to see this moment as a parallel of the polygraph’s invention, another time when technology has done a number on whatever shared sense of truth America once had. That erosion has worsened our cultural and political divisions, which often hinge on what kind of truth we believe in: relative and constructed, or absolute and fixed. A century ago, the polygraph was born from the latter belief. But the notion of objective truth seems quaint and naïve in our era of fake news and algorithms, when the only truth that still exists is ours, a custom reality created for us and delivered to our devices. We’re all memoirists now, shouting our stories into the void. We’re all polygraphers, staring at the screens of our truth machines, proving ourselves right.
*
A few months after the exam, I got the official news: not only had I been found unsuitable for employment, I also had no right to appeal the decision and was barred from reapplying for a minimum of three years. The consequences of my polygraph exam were finally clear, the only real truth revealed in the whole process: I was never going to be a federal agent.
I wasn’t alone. I soon found out that Customs and Border Protection job applicants had a polygraph failure rate of 68.1 percent, more than double that of other law enforcement agencies. The CBP commissioner at the time said those statistics showed the polygraph was working and blamed the quality of the applicants. It struck me as a strange rhetorical strategy to suggest that his agency attracted applicants so much worse than those of any other law enforcement body in America, but what did I know: I was one of those applicants. One article described polygraph subjects being accused with no evidence of cheating on their wives and having cartel connections, in exams lasting eight hours or longer. Other law enforcement agencies called CBP’s conduct excessive, and Jeff Flake, then a Republican Senator from Arizona, suggested that operators were being forced to fail applicants to justify their own jobs.
A few months after my exam, I got an email officially notifying me of my failure. I replied to the email and asked for a copy of the polygraph report. The CBP representative told me I’d have to file a Freedom of Information Act request. I did. They ignored my request for four months, in violation of FOIA. When I threatened legal action, they denied my request for other reasons, both of which were lies.
I filed another FOIA request and was still waiting for a response when the Office of Personnel Management announced that it had suffered a data breach. The personnel files of millions of federal employees and applicants had been stolen by Chinese hackers. An investigation ensued, and the Inspector General accused OPM officials of lying about the hack. Multiple high-ranking officials, including the director of OPM, resigned in the aftermath.
Eventually I received a letter from the replacement director alerting me that I’d been affected by the hack. (In the letter, she admitted that her own background check had also been compromised, as if that was supposed to make me feel better.) The most sensitive imaginable document about my life—one that included all of my personal and financial information, as well as a complete history of my jobs, residences, and relationships—had been stolen. The government’s solution was to offer free credit monitoring to those affected.
As of this writing, it’s been more than a decade since my polygraph exam, and the government still hasn’t sent me a copy of the report. But at least now I know the truth is out there. It’s on a hard drive somewhere in China.
__________________________________
“The Memoirist and the Lie Detector” by Justin St. Germain appears in the latest issue of New England Review.
