It was well after midnight when David Levine, a junior at Stanford, left his job in a physics lab and walked home alone across the mostly deserted campus early on the morning of Sept. 11, 1973.

He was outside Meyer Undergraduate Library when someone set upon him, in what police described at the time as a “swift and vicious attack.” The assailant stabbed Levine more than a dozen times, delivering many of the blows to his back after he had fallen the ground. The 20-year-old never had a chance to fight back.

Nearly 46 years later, Levine’s killing is the lone enduring mystery in a string of four brutal murders of young people on and around the Stanford campus in 1973 and 1974. The cases went unsolved for decades, until DNA evidence recently allowed investigators to identify suspects in three of them.

Now, authorities say, only Levine’s murder remains an enigma, with no suspects arrested or identified and no apparent witnesses to the crime.

But cold case investigators say they are newly optimistic that advances in DNA technology might finally crack the case.

“Evidence that may have been reviewed in the past didn’t have the same potential that it does now,” said Sgt. Shannon Catalano, of the Santa Clara County Sheriff’s Office, who is involved in the investigation of Levine’s killing.

The Sheriff’s Office and the Santa Clara County District Attorney’s Office are beginning a new review of police reports and evidence in the case, and will work with the county’s crime lab to “make sure that everything that could conceivably have suspect DNA is being tested with the latest technology,” said Matt Braker, the prosecutor who leads the cold case unit.

No motive has ever been determined for the killing — Levine’s wallet was still in his back pocket, a watch still on his wrist when a jogger found him dead on the pavement around 3 a.m., according to this news organization’s coverage at the time. And no one could think of anyone who would want to hurt the physics major who loved to talk politics in his dorm and so impressed Stanford researchers that they hired him for a lab position reserved for the top undergraduates.

Catalano said there isn’t any sign that Levine’s killing was related to the three other murders in the area, all of which targeted young women.

At the time, investigators had a theory that Levine’s killing might have been part of one of the Bay Area’s most infamous crime sprees: the racially motivated “Zebra Murders” that terrified San Francisco around the same time.

The murders, in which black assailants targeted white victims at random, were carried out by a group known as the “Death Angels,” who may have been responsible for dozens of killings around the Bay Area in the 1970s.

But the connection to the Levine case was thin: It was based on the apparent randomness of Levine’s murder and a witness who claimed to have seen black men in the area near the time he was killed, said Sgt. Noe Cortez, who began re-investigating the Levine case and the other Stanford killings in 2016 when he worked in the sheriff’s cold case unit. A link between Levine’s murder and the Zebra Murders has never been confirmed.

“That’s where it went cold,” Cortez said. “They weren’t able to connect any of the individuals they looked at through any forensics from back then.”

‘A brilliant person’

The son of a Cornell University professor, Levine grew up in Ithaca, New York, and appeared to be flourishing at Stanford.

He played chess at the Stanford Coffee House and won a seat in the university’s student government association; in the coming school year, he was set to serve on a campus research committee, then graduate a year early. In an undated photo provided by Stanford, he appears neat and clean-cut, wearing horn-rimmed glasses, a light suit and a dark tie.

Levine’s family in Ithaca declined an interview request from this news organization. In a brief phone conversation, his mother, Lima Levine, said, “We think that this is a matter for law enforcement, and we have confidence in their process.”

Those who knew Levine at Stanford described him as a young man full of potential.

John P. Wikswo, a physics graduate student who asked Levine to work on a research project he ran in the 1970s, said he hired the young man because he was struck by his “intelligence, intensity — a desire to learn, a willingness to work.”

Long before the days of personal computers, Wikswo said, a “fearless” Levine told him he could build by hand the computer system Wikswo needed, one that could process data from research aimed at developing a better way to diagnose heart disease.

Wikswo, now a professor at Vanderbilt University in Nashville, said Levine’s death has weighed on him. He was killed during the break between the summer and fall quarters. Most students had gone home, but Levine stayed on campus to keep working in the lab.

One of the last things Levine did was write a letter to Wikswo, who was vacationing with his family in Virginia, updating him on his work on the project. Wikswo received the letter two days after he learned of Levine’s death.

“He was a brilliant person,” Wikswo said.

Search for DNA

The emerging use of genealogy databases — which allow law enforcement to compare crime scene DNA samples against DNA submitted by the public to genetic testing sites — is one reason investigators have new hope in the Levine case.

The method has become increasingly popular among law enforcement since it was used to identify Joseph James DeAngelo as the suspected Golden State Killer last year, although it also has raised privacy concerns, prompting some DNA testing companies to inform customers that their DNA could be used by police.

Authorities say the databases provided a break that helped them solve two of the four Stanford killings. DNA taken from the scene where 21-year-old Stanford graduate Leslie Perlov was found strangled to death in the hills near campus, seven months before Levine’s murder, led detectives last fall to a Hayward man they have since charged with Perlov’s killing.

In May, further testing tied the same man, John Arthur Getreu, to the killing of 21-year-old Janet Taylor, who was strangled to death and left in a ditch on Sand Hill Road in the spring of 1974, according to investigators. He is now facing murder charges in Santa Clara and San Mateo counties, where he has pleaded not guilty.

DNA was also integral in solving the fall 1974 killing of Arlis Perry, a 19-year-old who was murdered inside Stanford Memorial Church, though in that case authorities said the DNA was used to confirm a long-time suspect as the killer. The man, former security guard Stephen Crawford, took his own life as authorities closed in on him last June.

But detectives can’t test DNA they don’t have, and Cortez said detectives up to this point have not been able to identify a sample of DNA from the attacker in the Levine case.

Catalano said the investigation now will focus on going back over police reports and crime scene evidence to find “what has the potential to yield something new, like a new DNA profile.”

She and Braker also cautioned that finding a DNA sample isn’t a guarantee that the case will be solved.

“You have to have DNA left at the scene, it has to be of a certain quantity and quality,” Braker said. “There are a lot of things that have to fall into place for that to be able to get you to your suspect.”

Sense of loss

Soon after his murder, co-workers in Levine’s lab described him to this newspaper as a quiet and hard-working student who seemed to keep to himself. But George Schnurle, who lived down the hall from Levine, said in a recent interview that wasn’t his reputation at Stanford’s Naranja dorm. Schnurle remembered him as a fixture of common-room debates, where the stridently progressive Levine loved to dive deep into conversations about politics and physics.

“He was very engaged in conversation,” Schnurle said, “but it wasn’t small talk.”

Levine’s death, he said, shattered his sense of the Stanford campus as a serene and secure place. After reading over the last year about how DNA testing has solved the other long-cold Stanford cases, Schnurle said he has become “hopefully optimistic” that there could finally be a break in his friend’s murder.

Wikswo was less heartened, saying he doubts the case will ever be solved unless someone comes forward with new information.

As investigators take another look at Levine’s murder, what remains certain for both Wikswo and Schnurle is the sorrow over the friend they lost and the unfulfilled potential of a budding physicist.

“It was just this huge sense of loss, not only for the friend and the person, but just the knowledge and his brilliant mind,” Schnurle said. “It was a loss to everyone, all of us — who knows what he would have done?”