Gus Vollmer was lost in thought. It was February 1921, and the chief was sitting at his desk at police headquarters, sun-wrinkled face creased into a frown of concentration. He jotted notes on a yellow pad in his looping handwriting.
He’d spent the morning flicking through the latest issue of the Journal of the American Institute of Criminal Law and Criminology, which had finally published a paper he’d submitted more than six months earlier, about his plans to build a policing school in Berkeley. But something else had caught his eye.
Sandwiched between Vollmer’s article and one debating the benefits of hanging the mentally ill was a paper by a psychologist and lawyer called William Moulton Marston. Six years earlier, as a Harvard psychology undergraduate, Marston had conducted an experiment in which he claimed to be able to detect whether someone was telling the truth.
Inspired by an observation from his wife, Elizabeth – also a psychology student – who’d noticed how her blood pressure seemed to spike when she was upset or angry, Marston asked his fellow students to tell either a true story, written down on a card in front of them, or a false one which they’d made up themselves a few minutes earlier. He discovered that he could tell, based on blood pressure readings alone, which of his peers were lying.
Marston and his course mates were used to this kind of psychological prodding. Their tutor was Hugo Munsterberg, a German experimental psychologist who had pioneered the systematic use of physiological measurements as a proxy for studying emotion. In his Harvard lab, Munsterberg examined scores of students using word association games, measuring their reaction times, and their bodily responses to thoughts of joy, horror and patriotism.
The idea that the actions of the body could betray the mind has been around for centuries. In ancient China, suspects were made to chew uncooked rice: the grains were thought to stick in the dry, nervous mouths of the guilty. In 1730, the English author Daniel Defoe suggested that the racing pulse of potential pickpockets would give them away. ‘Guilt carries fear always about with it’, he wrote. ‘There is a tremor in the blood of a thief.’
Scientific investigations into this idea began in the 1850s. The Swiss psychiatrist Carl Jung used a galvanometer to track emotional changes in his patients – recording how the electrical conductance across the sweaty surface of the skin changed over time. In Russia, neuropsychologist Alexander Luria measured the reaction time and chattering fingertips of criminals under investigation.
In the late nineteenth century, the Italian criminologist Cesare Lombroso interrogated suspects using an air-filled glove to measure blood pressure. Another contraption required the subject to plunge their forearm into a tank of water – the more the level of the liquid fluctuated as they spoke, the greater the chance they were lying. Lombroso’s student Angelo Mosso designed a carefully balanced bed that would tip to one side as blood rushed to a suspect’s face in the course of a lie (although it was never built).
Marston’s insight about blood pressure and deception built on this work. Over the next few years, he conducted a handful of small-scale experiments – first with applicants to the army, and then with German prisoners in the aftermath of the First World War. But on both occasions, his superiors were left disappointed. Although Marston generally seemed to get the right answers, the manner of this precocious, handsome young man made them doubt his reliability.
The general impression they were left with was that the results had more to do with Marston as an individual than any scientific breakthrough he’d made. And he never seemed too interested in the hard work required to turn his discovery from an interesting observation into a practical, scientific tool.
It took him two years to even publish the results of his first experiments. Over 40 pages, he described in fine detail the blood pressure rise and fall in each of his subjects, and outlined what he saw as the tell-tale signs of deception. ‘The sudden sharp, short rises of blood pressure betray these substantial lies in an otherwise true story’, he wrote.
Vollmer read Marston’s latest article with growing interest. He’d heard about previous attempts to detect lying in the work of Hans Gross – things like chewing rice – and ‘decided they were crude, amateurish and impractical for his use’. But if blood pressure rises during lying, he thought, couldn’t that be registered on a machine? He turned the idea over in his head for a few minutes. Then he put down his notepad and reached for the telephone to call the desk sergeant, grasping the stick in one hand and pressing the bulb of the receiver to his ear with the other. ‘Send John Larson in to see me as soon as he reports for duty,’ he said.
That afternoon, Larson dressed in his dark patrolman’s uniform with the star over the heart and made his way into work for what he thought would be a standard evening on the downtown beat.
The sun was low in the sky when he arrived at City Hall, a sparkling three-storey building, built in 1909 and styled on the grand Hotel de Ville in the French town of Tours, with elegant white carvings and a clock tower which had been demanded by Berkeley’s discerning cultural elite. The budget didn’t quite stretch to a clock, though – even today, passers-by who glance up looking for the hour are met with a blank white circle. It’s timeless, in every sense.
Larson crossed the manicured lawn and ignored the city employees making their way up the grand main staircase. Instead, he walked around the left-hand side of the building to the south wing, opened a discreet side door and descended a flight of stairs into the basement headquarters of the Berkeley Police Department.
It was filled with the musty smell of sweat and old books. Exposed pipes ran along the ceiling, and windows at ground level filtered in a meagre amount of daylight. Dozens of officers worked under naked bulbs that hung from the ceiling.
Larson checked in with the desk sergeant and was ordered to report straight to the chief – not usually a good sign. For a moment, he stood frozen in something like panic. He thought he was in trouble.
About six weeks earlier, around the end of 1920, Larson had been on duty on the 4pm to midnight shift. Vollmer had worked out detailed route maps for each of his officers, but after making a few rounds of the ‘interwoven meshwork of dirty alleys’ most of them simply parked their car at a convenient intersection and watched the signal light.
Despite his usual diligence, Larson was no different. He was ‘loafing’ and watching the light when his old foe Henry Villa pulled up in his car and surprised Larson by suggesting they drive up to Barney’s Diner together for a malted milk.
They returned to their beats at about 11.45pm, and were back at the station by midnight to write their reports on a quiet night. But it hadn’t been a quiet night – at about 10.30pm, the Sunset Hardware Store in Larson’s territory had been cleaned out by a gang who smashed in the back window. He’d completely missed it, and Vollmer had seemed off with him ever since.
Larson picked his way past desks piled with equipment: one had ink pads for taking fingerprints; at another, police photographers diligently worked their way through rolls of film at trays of developing fluid.
Next, he passed through the open area where the officers shared desks, sitting at wide tables on chairs with wooden rollers. Eventually, he reached the chief’s office and his big flat-topped desk, which was neat and tidy save for a telephone, a reading lamp, a stack of books and journals, and a bowl of sweets for the children that Vollmer so adored. The walls were decorated with paintings of California landscapes.
Vollmer looked up and smiled when Larson entered. The chief favoured grey suits that matched the steeliness of his eyes, and his body was still lithe and strong. He had a way of holding himself that made him seem even taller than he actually was.
He had been following the rookie’s work on fingerprinting and hormones closely, and thought Larson would be the perfect man for the job he had in mind. ‘Sit down,’ he said, in his deep, strong voice. ‘I’m playing with an idea. I need your help. John, have you ever studied anything about detecting lies?’
The question set Larson’s mind racing – he’d never looked into the research on deception, but he was definitely intrigued, as well as relieved he wasn’t about to be disciplined. ‘That would be a tough assignment,’ he said, when Vollmer explained what he wanted to do. ‘Yes it will be,’ the chief replied. ‘I’d like to do some experimenting.’
Vollmer handed Larson a book by Cesare Lombroso, the Italian criminologist, and the journal with Marston’s article in it. The paper described a series of blood pressure tests conducted by Marston on 19 people in Boston – 16 women and three men – whose crimes ranged from shoplifting, to fornication, to being a ’stubborn child’.
The first case was of a 42-year-old white woman (each entry had the race neatly listed in brackets after the gender) and recovering drug user who was arrested after a hypodermic kit was found in her room. There was a brief record of the case given to Marston before the deception test (‘Defendant claims she is not now using drugs’), followed by his verdict based on the blood pressure readings (‘Innocent. Woman is not now using drugs’), and a follow-up explaining why he’d got it right (‘Medical examination showed increased weight, better all-round health, etc., which could not have existed were defendant now using drugs’).
Larson turned the pages with growing interest. He was captivated. Here was a discovery that could help solve the cases where there wasn’t enough physical evidence to bring anyone to justice, even with the new techniques Vollmer was pioneering. It could end some of the brutal tactics he’d encountered in his five months on the force, and bring honesty and justice into policing.
He thought he could do better than Marston, though. The physiology lab at Berkeley was full of measurement equipment he could use to get a more accurate view of deception. By the time he left Vollmer’s office that day, Larson had devised a plan for the experiments he could run, and – even better – Vollmer had reassigned him to work on the idea full-time. He would be a fully paid-up police officer, working solely in the lab – it was his dream role. No more late nights, no more hazing, no more roosters. ‘Remember, John,’ Vollmer had said, as he sent the delighted rookie on his way. ‘I want to be the first one tested.’
Over the next few weeks, Larson quickly set to work on improving Marston’s method – turning it from a fleeting measure on the dial of a blood pressure monitor into something more objective, which could potentially be used as evidence in court.
He combined Marston’s work with a couple of other books Vollmer had recommended – including one by Vittorio Benussi, an Austrian psychologist who had achieved 75 per cent accuracy in detecting deception by measuring people’s breathing rates.
Dr Robert Gesell, a physiology professor at Berkeley, had already developed a modified blood pressure cuff for tracking changes in flow and volume over time. Vollmer suggested that it might be adapted to detect lying. Larson used Gesell’s device as his base.
He spent long hours in the lab, attaching friends and a rotating cast of college students to various pieces of equipment, trying to determine whether the readings from each instrument correlated with lying. Larson confirmed Marston’s finding that when people lied, their breathing rate and pulse seemed to change in intensity and quicken. But to go further, he needed a way to record the body’s reactions so that they could be properly assessed after the fact.
In 1906, Dr James Mackenzie, a British cardiologist, had devised a machine that tracked blood pressure at the wrist and neck and translated it into two continuous lines drawn onto paper by pens that moved in concert with the subject’s body. Larson borrowed this idea for his own device, and he corralled the mechanic at the city garage into building a prototype – a grab bag of medical equipment attached to a wooden breadboard.
The machine had an ordinary blood pressure cuff that was wrapped around the upper arm, with a tube leading to a heavy rubber bulb, inside a larger glass one. As blood pumped through a subject’s veins, their arteries swelled, inflating the rubber bulb and changing the volume of air inside the glass one. A connected tube transferred this change to a delicate rubber tambour, which translated it into the movement of a thin stylus scratching a thin path on a slow-moving drum of smoked paper, white lines against black.
Larson added a pneumograph – two bands of elastic that wrapped around the chest to measure lung capacity and breathing rate, which were also recorded on the paper via a similar system of tubes and actuators. The finished machine was a mess – a Frankenstein device that was a mixture of all the research that had come before.
It looked, a reporter for The Examiner wrote later, like ‘a combination of a radio set, a stethoscope, a dentist’s drill, a gas stove, an aneroid barometer, a time ball, a wind gauge, and an Ingersoll watch’. But it provided, for the first time ever, a fixed record of a person’s blood pressure and the beat of their heart.
It was rough around the edges, Larson thought, but it ought to work. In public, he called it the ‘cardio-pneumo-psychograph’ – a label as unwieldy as that first prototype. Eventually, it would get a different name – the same moniker Mackenzie had given his cardiology device, a term derived from the Greek for ‘many writing’. Privately, Larson always thought of it as ‘the apparatus’. Everyone else called it the polygraph.
While the mechanic was tinkering with the machinery, Larson worked on the ‘software’ of lie detector tests from his ‘magpie den’ of an office. When Marston tested subjects, he had no way of taking continuous measurements of blood pressure, so he took readings once a minute while they told their stories. That was fine – but it made scientific validation of the results almost impossible.
Larson decided to reverse the procedure. In the experiment he devised, his ‘cardio-pneumo-psychograph’ would take continuous measurements from the body while the investigator asked specific, direct questions. That way, the answer to a certain query could be matched to its exact physiological response.
A month after his meeting with Vollmer, Larson was ready to show the apparatus to his boss. He’d promised the chief he’d be the first person to be tested, and in March 1921 he arranged a demonstration of the device at police headquarters.
He set up the lie detector in a small office in the basement. It had a square wooden block, like a chopping board, with a large iron column rising from it that held two thin styluses. These interacted with a horizontal scroll of charcoal-blackened paper, about a foot and a half long, which slowly rotated from right to left, propelled by two spinning drums. As it turned, the moving pens scratched away the black dust, leaving bone-white lines. Rubber tubes snaked away from the device towards an empty wooden chair, like the tentacles of some strange beast probing in the dark for its first victim.
Vollmer walked into the room with a group of sceptical officers in tow, including department veteran Jack Fisher – an intimidating presence who had served with the intelligence division in Siberia during the war. They were eager to see what the college boy had come up with.
Larson watched nervously as they scrutinised his work – the tangled mass of tubing and wires, the glass bulbs and rubber appendages. ‘It isn’t much to look at,’ he laughed. ‘But let’s see what it can do.’
But something wasn’t right. Vollmer had a playful look in his blue-grey eyes, and the beginnings of a smirk. Instead of taking the test as planned, the chief ordered Larson to strap himself into his own machine.
Once the rookie was hooked up, Vollmer – who was familiar with the basic premise of the test – started asking him questions, beginning with very simple ones. As Larson answered, the gathered officers peered over his shoulder at the lines on the chart.
Larson knew what was coming. Vollmer hadn’t brought up the theft at the hardware store, but the chief had ways of finding out what his officers got up to. Larson suspected Villa might have squealed.
So, when the question finally came – ‘Were you off your beat the night of the Sunset Hardware Store burglary?’ – he didn’t even have to answer. Larson’s body betrayed him before he could even speak; his physiological response so extreme that the needles of the machine shot off the piece of paper entirely. There was a stunned silence. ‘You don’t have to answer,’ crowed the chief. ‘We’ve got you hoisted by your own petard.’
Fisher, the veteran officer, went pale and offered to resign rather than be attached to ‘that God-damn thing’. Vollmer laughed. ‘I don’t need the machine for you.’
Next, it was Vollmer’s turn to be hooked to the device. Larson passed the cables around his chest and waist to measure his breathing, fitted the blood pressure cuff around his arm and pumped it up to the correct tightness with a rubber hand bulb. The strangeness of the situation amused him. ‘I never thought I’d ever try to catch you lying,’ he said.
‘I hope you do, John,’ replied Vollmer. ‘Just go ahead. See if you can catch me.’
For the first time, Larson ran through the preamble to the test that would become second nature over the decades to come. ‘Remember to answer yes or no to all questions,’ he told Vollmer.
‘No comments.’
Vollmer smiled. ‘The comments will come later.’
Larson started the test with simple questions to try and establish a baseline: ‘Did you eat lunch today?’ he asked.
‘Do you like to swim? Do you like ice cream?’
Vollmer answered ‘yes’ to each of these questions, and each time the pens on the polygraph traced a steady course – going up and down with the beat of his heart.
Larson continued, firing more questions at his boss. On two occasions – once, when he asked when Vollmer had gone to sleep, and again when he asked whether he liked roast beef – the needles made slightly more dramatic movements. The second time this happened, Vollmer stopped the test.
‘It works, John,’ he said. ‘I lied two times, about the time I went to bed, and my not liking roast beef and every time the needle shot up.’
He was ecstatic – he ordered Larson to test every officer in the department who could spare the time.
‘I really think, John, we are on the right track,’ he said. ‘But we’ve got to move slow. We’ve got to prove this machine works on an actual case.’
*
Excerpted from Tremors in the Blood: Murder, Obsession, and the Birth of the Lie Detector, by Amit Katwala. Copyright, 2023. Published by Crooked Lane Books. Reprinted with permission. All rights reserved.
