As a consequence, city juries have become a deterrent of sorts to prosecutors, who are willing to accept weaker pleas or tolerate dismissals rather than waste the city’s time and money on cases involving defendants who are clearly guilty, but who have been charged on evidence that is anything less than overwhelming. Naturally, a competent defense attorney or public defender understands that in most cases, a jury trial is the last thing a city prosecutor wants, and he uses this leverage when he bargains for his client.

For the detectives, the decision to plea or dismiss a case is the flashpoint in their ongoing love-hate relationship with the state’s attorney’s office. True, thinks a detective, these people are on our side. True, they’re working to put bad guys in prison at half the salary they might get at an outside firm. True, they’re looking for the same justice we are. But brotherly feelings are out the window when a young assistant state’s attorney, two years out of the University of Baltimore School of Law, gives up on a drug murder that took three weeks to develop. When that happens, the chip goes right back on the shoulder: I busted my ass to get reluctant witnesses into the grand jury, and what for? Just so this goof with pinstripes and a power tie could dump it on the stet docket? Hell, he didn’t even have balls enough to pick up a phone and call me, much less ask how the damn file might be salvaged.

Some of the cases are weak and should be dumped, no doubt about that. Some of them arrived at the courthouse as viable prosecutions, only to self-destruct once the witnesses started backing up. Any homicide detective knows that most basic truth: Shit happens. But he also believes that too many borderline cases, and even a few that are healthy, manage to slip through the cracks, particularly with less experienced attorneys.

A good detective will excuse some of it as understandable and inevitable. As is true elsewhere, the Baltimore state’s attorney’s office is chronically understaffed and underfunded; its trial division is manned by a core of competent veterans and a slew of recent arrivals-younger lawyers who have worked their way up to felony violence after a few years in the district courts. Some will be good trial lawyers, some could go either way, and a few are genuinely dangerous in a courtroom. A detective hopes for a competent prosecutor, but he understands that the system runs by triage. The homicides are parceled out with an eye toward keeping the major cases-those involving true victims or those in which the defendant is suspected or charged with multiple crimes-in the hands of veteran attorneys. The hope is that in the most critical cases, the prosecutor will not be outclassed or intimidated by the coterie of experienced defense attorneys who by private retainer or court appointment always gravitate to city homicide cases.

Every detective also understands the need to take pleas on at least two-thirds or more of the viable murder prosecutions. Although most everyone outside the legal system regards “plea bargain” as a dirty word, those who make their living at the courthouse recognize it as a structural necessity. Without plea agreements, the system would lurch to a halt, with cases waiting for courtrooms the way commuter flights wait for runways in Atlanta. Even with the current ratio of pleas to trials, the delay between a murder indictment and the court trial averages between six and nine months.

But in a detective’s mind, there is a vast difference between a good plea and a bad one. Second-degree and thirty is always a respectable plea, except for truly evil acts such as, say, child abuse cases or robbery murders. If the case is borderline, second-degree and twenty isn’t too shabby, although it’s not exactly the iron fist of justice when you consider that the parole board puts most of them back on the street after about seven to ten. In a true manslaughter case-a domestic murder that was the act of fear or impulse, though it could in no way be called an accident-anything from two to ten is reasonable. But what’s hardest for a detective to swallow is a prosecutor allowing a particularly bad murder to go as second-degree, or calling a murder a manslaughter, or a manslaughter an accidental. Even in those instances, most detectives won’t speak their piece unless they’re asked, and the prosecutors don’t usually ask. In the homicide unit, the time-worn philosophy is that it’s on the prosecutor; you did your job, fuck him if he won’t do his. Occasionally, however, a detective will cross the emotional boundary.

Worden, for instance, has been known to say something to a young prosecutor who’s giving up on a file too quickly, or seems afraid to take a decent case into court. Landsman will sometimes do the same, and Edgerton, if you give him a chance, will tell a prosecutor how to try the case and then write out the closing argument. A lot of men in homicide carry around a case or two that still burns them. Garvey, for one, still isn’t saying much to the ASA who turned the Myeisha Jenkins murder into a second-degree plea-Myeisha, who was all of nine when her mother let her boyfriend beat the child to death and dump her on the shoulder of the Baltimore-Washington Parkway. Garvey told the lawyer he was a piece of shit for taking that plea, told it to him in such a way that the man didn’t even try to argue.

If he cares enough about a case, a detective can lobby or even argue for a particular strategy. But in the end, decisions about the legal approach to a case are not his to make. From crime scene to conviction, the courthouse is the only part of the process in which the detective becomes a passive participant, a player wholly dependent on the decisions of others. A detective is there to testify and otherwise serve the lawyers in any way he can. The lawyers, meanwhile, regard that service with varying amounts of appreciation. Some prosecutors consult the investigators on evidence and presentation, asking the opinions of veteran detectives who have been through the process more often than the attorneys. Others view the detectives as little more than props and gofers, responsible for showing up on time with the right evidence and the right witnesses.

Homicide detectives are further distanced from their cases because, as witnesses, they are sequestered and therefore prohibited from attending court and listening to other witnesses. Detectives in Baltimore spend 90 percent of their court time sitting on hard wooden benches in corridors, or running bags of evidence between the courtroom and the prosecutor’s office, or chasing down a witness who’s supposed to testify in the afternoon session but hasn’t shown up, or maybe bullshitting with the secretaries upstairs in the Violent Crimes Unit. Court time for a detective is a strange limbo, a period of nonexistence that is only briefly interrupted when he is called to testify.

The stand is the last point in the process in which a detective’s expertise counts for something. In most cases, the testimony of civilian witnesses-primed and prepared by the prosecutor before trial-will produce the most critical evidence. But in every case, the testimony of the detective, concerning the crime scene, the discovery of witnesses, the statements made by the defendant, lays the groundwork for the prosecution’s case. Among prosecutors, there is a theory that says a detective’s performance on the stand can rarely win a case, but it can be enough to wreck a prosecution.

Before taking the oath, a detective who knows his business makes a point of reading through the case file. After all, it’s been six months and a lot of bodies between the arrest and trial. In 1987, a city detective-no longer in the homicide unit-responded to a prosecutor’s question with an elaborate description of the crime scene and subsequent investigation. After a minute or two, he saw that the prosecutor was making strange faces. Even the defendant looked a little curious.