Call it the CSI list: fingerprints, gunshot residue, ballistics, toxicology, bite patterns - the full rundown of forensic methods used by prosecutors to link defendants to crime scenes.
Public perception and generations of prosecutors suggest that all of those forensic methods produce rock-solid scientific evidence against criminal defendants. And one by one, Patrick Kent, chief of the forensics division at the state public defender's office, is trying to destroy those certainties.
Kent has enjoyed success by attacking the validity of gunshot residue and - just last month in a Baltimore County murder case - fingerprints. Predictably, prosecutors are no fans, insisting that the science he is trying to undermine has stood the test of time.
"We have always had to prove that science is reliable," said Sharon Holback, director of forensic science investigations for the Baltimore state's attorney's office. "Patrick Kent did not invent that rule of law."
Just as predictably, Kent is unruffled by those who dismiss him.
"Quite candidly, my job is not to be popular," he said in his Baltimore office last week. "My job is to ensure the integrity of the criminal justice system, and ultimately, if that litigation causes organizations or entities to be upset at me, then so be it."
Impassioned and tireless - he cannot remember his last vacation - he says he is on a mission to make sure that the forensic evidence used by prosecutors is, indeed, science and not merely speculation or opinion dressed up as science.
"Literally, every forensic science is going to be systemically litigated," said Kent, 41. He says he does not want to make it harder for prosecutors to lock up the bad guys. He wants to make it harder for them to lock up the good guys.
"The reality is and has been that people are wrongfully convicted and sent to jail," he said. "It's not a question of if that happens. The real question is how many people are in jail based on bad forensic science. And the answer, I'm certain, is a very high number."
Nationwide, only a handful of public defender's offices have a forensics division, but Kent seemed destined for such a job. A native of Syracuse, N.Y., he started as a pre-med major at St. Lawrence College in New York before switching to psychology, giving him a background in hard sciences and an appreciation for the rigors of research.
After college, Kent spent several years as a caseworker at a shelter for homeless and runaway youths in Syracuse, where he became well acquainted with the problems of mental health and abusive homes.
"Ultimately, what that experience left me with was what goes wrong with society is society," he said. "We like to blame the individual, but frankly, they're just the manifestation of a dysfunctional society."
While other caseworkers at the shelter worked a 9-to-5 shift, Kent drove a van around the city on nights and weekends to deliver counseling and other services directly to teens. The work was rewarding, but Kent realized that he wanted to effect change on a broader scale. He decided the law could be his vehicle for accomplishing that.
He got his law degree at the University of Baltimore and then joined the Maryland public defender's office, where he thought he could best help those who have been left behind by society.
It was in law school that he came to realize that the science he'd learned in college had no relation to the science on display in courtrooms.
"Having been exposed to the hard sciences, and then seeing forensic science, it's jarring," he said. "You can't help but say that this is nothing remotely like the science that we think of in the traditional sense."
Fueled by an unending tide of Diet Pepsi, Kent works out of a 12th-floor office in Baltimore, a block from the Clarence M. Mitchell Jr. Courthouse. He's so focused on his cases that, after four years as head of the forensics division, he hasn't found time yet to decorate or even hire a secretary. Six pots that line the windowsill are filled with dirt but no plants.
The only art on the walls are Magic Marker drawings by his 5-year-old son, Zachary, including one showing father and son each holding a magnifying glass. His wife, Karen Kent, is a clinical psychologist, a profession that does not lend itself as easily to representation by Magic Marker.
Zachary was a mad scientist for Halloween - his idea. The boy's love for science is a reflection of his father's obsession with it. But Kent's reverence is for honest science. He has little patience with practitioners who call their methods science when they are not.
He says methods such as fingerprint analysis, ballistics tests to match bullets to weapons and gunshot residue examination are accepted because they've been around for decades, not because they've been scientifically proven.
"The passage of time was the validator, not that there was an underlying scientific basis," Kent said. "They call themselves 'forensic science' for a reason. They want the power of science."
But, he added, they want that power without accepting the rigor that comes with science.
Prosecutors counter that defense attorneys are trying to have it both ways: They want to use forensic evidence when it clears their clients, but attack it when it links defendants to crimes. Baltimore State's Attorney Patricia C. Jessamy said more people have been released from prison because of forensic science, such as DNA tests, than those who have been convicted because of bad science - a number she puts at zero.
"They're using science to exonerate people," she said of public defenders. "How do you then say it's bad science, when you're relying on it to get people out of jail?"
Forensic science is on trial in every case, prosecutors say, as defense attorneys are free to cross-examine expert witnesses and present their own evidence. Methods that are deemed unreliable by the scientific community, prosecutors say - such as the use of lie detectors - are not introduced in court.
"We wouldn't present evidence if we thought it was unreliable," said Leo Ryan, deputy state's attorney for Baltimore County, where Kent challenged the validity of fingerprint evidence in the murder trial of 23-year-old Bryan Keith Rose.
A judge ruled that an analysis linking Rose's fingerprints to prints found on a stolen car was inadmissible because the error rate in fingerprint comparison is unknown. The ruling cannot be appealed and does not make precedent, but it raises questions about a long-established forensic science and could prompt challenges in other cases.
It wasn't the first time that Kent has caused this sort of trouble. Two years ago, in a case supervised by Kent, a Baltimore circuit judge found that the city police lab's practice of labeling "two-element" gunshot residue particles as unique fell short of accepted scientific standards. Since then, prosecutors have introduced only the more exacting "three-element" standard in court.
There is more to come, Kent promises. His office is preparing a case on tool mark identification - the markings that a gun leaves on a bullet when fired. When guns were made by hand, such markings were unique. But with mass-produced weapons, does each still have a unique marking?
"I think science can really inform, but it can also mislead," said Jeff Gilleran, a public defender in Kent's office. "What we're trying to do is open the doors of the crime lab and let some light in."
Kent says he is only in search of the truth. If prosecutors want to use forensic evidence in the courtroom, he says, then they must go back to their crime labs and tell them to validate their work. And until that happens, he'll continue to ask uncomfortable questions. He has yet to receive satisfactory answers.
"I used to have trouble sleeping before," he said. "I have more trouble now."