Printer Friendly

Managerial justice and mass misdemeanors.

IV. A QUALITATIVE ACCOUNT OF MANAGERIAL MISDEMEANOR JUSTICE: THE DISPOSITION PROCESS

The official rules of criminal procedure and criminal law do not define how the criminal justice system actually operates in practice. Legal scholars have explored how the real-world operation of criminal law is shaped by the charging and sentencing discretion of prosecutors and judges as well as by the inclinations and structural capacities of defendants and defense attorneys. (110) In this Part, I propose something more. The formal substantive and procedural rules of criminal law do not make out an instruction manual by naming a clearly established end goal and specifying the precise means to secure that goal. Rather, substantive and procedural rules are simply the tools available in the contested and always-underspecified endeavor of social control. Frontline legal actors must decide not only how they ought to use those tools, but also exactly what the social control ends of criminal law are in the first place.

A key insight from the scholarship of organizational sociology is that the "practical circumstances" of work shape how front-line actors in an organization make sense of their goals in the first instance. (111) The practical circumstances of everyday work encompass the concrete setting of daily tasks, the exigent demands, the situations of choice, and the available information and resources to perform tasks (including investigative capacity and, most importantly, the amount of time). The era of mass misdemeanors has produced a certain set of practical circumstances in criminal courts which, coupled with the subject matter of misdemeanor justice, shape how courtroom actors come to understand both what the animating question of their daily work is and how that question can be adequately addressed with available resources.

Courtroom actors are constrained by the costs of obtaining information relative to their resources, by their capacity to absorb information, and by the unavailability of information." (112) With high caseloads and the constant pressure to resolve the hundreds of new cases that arrive daily, not only do courtroom actors adapt their understanding of the animating question of their daily work, but they also decide what information is relevant to that task. They must consider whether an investment in searching for more information would be profitable," given how they have defined their task. (113) In the managerial model, records of prior encounters and prior convictions play an important role in processing cases because marks of past encounters are perceived as a reliable signal of the defendant's overall governability. These records are more reliable than, say, the minimal investigation the actors will have the time or inclination to perform with respect to a specific allegation, or self-representations by defendants or even complaining witnesses. Records of prior encounters are accessible and costless to consult. Furthermore, these records speak to what has emerged as the animating moral question of misdemeanor punishment: whether this person is a persistent or occasional rule breaker.

A. The Practical Circumstances of Arraignment

In New York City over 57% of all misdemeanor and violation cases reach a disposition at arraignment. (114) Early and rapid disposition is an established feature of misdemeanor justice in New York City: over the past thirteen years the percentage of subfelony cases with a disposition at arraignment has fluctuated between a high of 65.5% and a low of 57.9%. (115)

A typical arraignment courtroom may have between 100 and 200 cases to be arraigned during a shift that has about six hours of operational court time (day shifts run from 9:30 a.m. to 5 p.m. with about 1.5 hours for lunch; night shifts run from about 5:30 p.m. to 1 a.m. with typically about 1.5 hours of downtime for dinner). The prosecution, judge, and defense receive two pieces of essential paperwork before arraignment as part of the defendant's criminal court file, the complaint, which is the criminal court charging document with the charges and a brief description of the facts that make out the offense, and the defendant's rap sheet--called the NYSID sheet. (116)

Prosecutors typically flip through the paperwork contained in the file for somewhere between one and five minutes before marking down a plea offer or recommendation on the front flap of the file if an offer is going to be made at arraignment. (117) In most boroughs the policy is to not make plea offers at arraignment on certain types of cases, for example, any case that involves a complainant (especially domestic violence cases) or driving under the influence. In those cases the prosecutor will make a bail recommendation note on the file.

Defense attorneys meet their clients for the first time at arraignment, either in a small, caged-in interview room separated by metal grating off the holding cells in the case of arraignment of online arrests, or in the hallway in the case of DAT arraignments. Sometimes interviews are very short, if for example the attorney tells the client the offer will be an ACD and the client readily accepts. And sometimes interviews go for ten to twenty minutes. (118) Attorneys are focused on getting essential information about the arrest circumstances, but, more importantly, they are focused on getting information relevant to the bail application. Defense attorneys know that a client is better placed to fight a case successfully if she is not being held at Rikers Island (the New York City jail) on bail. (119)

Attorneys are also focused on speed. The longer all parties take doing their part to move arraignments along, the fewer people who are arraigned during that shift, which means defendants sit in the holding pens longer while waiting to see the judge. During a night shift, if defense attorneys do not make it through the arraignment load, those defendants will have to wait another eight hours in custody before seeing a judge. Since the majority of defendants whose cases continue past arraignment are released on their own recognizance, minimizing pre-arraignment detention is a driving concern. (120)

Some arraignment shifts have defense attorneys dedicated to "disposable" misdemeanor cases, which are likely to go to disposition at arraignment. (121) The fact that clerks and paralegals can estimate which cases will be disposable simply by looking at the charging documents and rap sheets indicates that something other than facts relating to innocence and guilt is driving this disposition process. One reason why clerks can designate large numbers of cases as disposable is that they know that prosecutors will not even seek conviction in a substantial number of first-arrest cases, or they understand that the standard offers will be readily accepted at arraignment.

The standard offer for many cases that do not involve a complainant and represent a first arrest is an ACD (which is significant because non-complainant cases constitute the majority of misdemeanor arrests). (122) Sometimes the ACD is conditioned upon the defendant's completing a short "program" or a few days of community service. (123) This offer policy does not in practice distinguish between guilty and innocent defendants because it is very difficult to do so at arraignment. The only factual information the actors in the system have at this point is the limited paperwork in the file described above. Prosecutors and judges rarely make an attempt to make this distinction even with the limited information they do have. Whether prosecutors and judges assume all defendants are guilty, are uncertain, or are unconcerned, the managerial tactic is indistinguishable: offering conditional dismissals allows the state to mark defendants for a limited time to see if the person cycles back into the criminal justice system. (124)

Sometimes the defense and prosecution argue about the facts of innocence and guilt, and sometimes the facts raised in those discussions affect some aspect of the offer. Yet a defense attorney can rarely, if ever, overcome at arraignment the imperative to exert some marking from the encounter. As one supervising ADA explained: "There are very few outright dismissals at arraignments, at that point we have our police paperwork and version of events, so we can't just dismiss the case." (125) The only difference between an ACD and an "outright dismissal" is the ability to impose a temporary mark on the defendant; otherwise they are legally equivalent dismissals (after the adjournment period). The mark records the fact of the prior encounter so that the judge or prosecutor can update the assessment of the defendant if he is arrested again during the adjournment period. It also puts the defendant on notice. As one judge explained an ACD disposition, it is "a low maintenance form of probation, you don't have to report because you monitor yourself." (126)

The following two stories illustrate that prosecutors often quickly agree not to seek a criminal conviction in cases where there is weak evidence. They do, however, often insist on an ACD disposition that accomplishes a limited-term marking in lieu of granting an outright dismissal.

In April 2011, in the Desk Appearance Ticket arraignment courtroom, John, a young, black man dressed in business casual, was being arraigned on charges of theft of services. (127) The arresting officer accused him of using a special MetroCard that provides discounted rates to people with disabilities for which he was not eligible. Because John had been given a DAT, he had time before arraignment to procure a letter from his employer, a social service agency, stating that one of his job duties included accompanying disabled people on outings. (128) John explained to his defense attorney, who explained to the judge and supervising arraignment ADA, that, on the day he was arrested, he was with a group of disabled people, and he was helping them swipe their cards. He insisted that he swiped a regular MetroCard, and he claimed that the arresting officer must have mistaken which light went off on the turnstile when he went through because they all went through the turnstile in close succession (different color lights illuminate for different types of MetroCards, such as student, senior, or disabled discount cards).

After a brief discussion the prosecutor agreed to offer a "straight ACD" instead of the initial offer of an ACD with one day of community service. Explaining why she would not move for an outright dismissal after seeing the employer's letter indicating the young man worked for an organization supervising disabled people she stated:

   I can tell you that we don't dismiss cases. I mean we do, but we
   have to have proof that he is not guilty. The offer was an ACD with
   one-day community service and he did bring proof that he works for
   that organization, so I dropped the community service. I still
   don't have proof that he was not using the disabled card on that
   day of arrest, so I gave him the benefit of the doubt with the ACD.
   (29)


Ray, an attorney for the largest public defender organization in New York City, recounted the following story about a white man in his mid-thirties who had been arrested for possessing oxycodone. The arresting officer had recovered the oxycodone from his pocket, not a prescription container:

   The attorney speaking to the client determines and discovers that
   the client has a prescription for oxycodone.... We sent the
   individual home at 10:00 in the morning to go get the prescription
   to bring back to show to the court. Because we knew we were going
   to have to have this argument. And he made it back by about 12:00
   in the afternoon. And between 12:00 and 1:00 we argued about the
   fact that this case should be dismissed. And the District Attorney's
   office, ... determined that--well, the prescription that he was
   issued for whatever his ailment was for oxycodone was issued in July
   of 2010 and he was arrested in November of 2010 or December of 2010
   --and he didn't have it in the bottle. And the prescription bottle
   says "no refills." ... In their perspective, because he had a
   prescription bottle that was issued in July of 2010 that said no
   refills that he must have been doing something illegal with carrying
   the pill in his pocket.

      We argued for an hour until finally the judge decided to--you'll
   excuse my expression--"grow a set," and just dismiss the case....
   It took an hour to get past that ... that alleged perspective from
   the DA's office. And as we exited the courthouse--because this is
   what stuck in my mind--exit the courthouse, the DA was still
   arguing with me about that case. How it shouldn't have been
   dismissed because you don't know; he had no refills that were
   supposed to be issued on that. (130)


In both cases the prosecution was willing to offer an ACD, but not to dismiss the case outright. The judge was not willing to do so either in the first case and finally agreed to do so in the second case after extended argument that made clear the prosecution was not willing to do so on its own motion. To state the obvious, I have no way of knowing if the claims of these defendants are true. I did not conduct any more of an investigation than the people in the courthouse that day deciding the fate of those cases at the first court appearance. My point is that prosecutors and judges are not necessarily maximizing punitive response or even attempting to secure conviction automatically from the police determination to make an arrest. Instead, they display what I called in Part I the presumption of need for social control: seeking, even in the context of limited facts indicative of guilt in the specific case at bar, some ability to track the person for later encounters.

The regular course of conduct in misdemeanor court is often at odds with administrative determinations of guilt. Prosecutors seek, and judges almost always grant, the same ACD disposition in cases where there is most likely sufficient evidence to show guilt at trial. In June 2011, a group of six men were brought in front of a judge on assorted violation charges, including disorderly conduct, open container of alcohol in public, and loud music. The judge in this courtroom is widely known as one of the most punitive judges in criminal court in this borough. All the men had been held in central bookings since early that morning because they had been picked up on outstanding warrants for not appearing on their summonses at the appointed time. As soon as the court officer finished reading the charges for the six men, the judge immediately granted an ACD to all defendants en masse; neither the defense, nor the prosecution, nor the judge even discussed the matter. (131) Defendants arrested for narcotics possession or marijuana possession are regularly granted ACDs--sometimes en masse with groups of four to six defendants huddled around a single attorney--with almost no conversation whatsoever between the defense and prosecution. (132) Again, in these cases I have no idea whether all or some of these defendants were factually innocent, or if there were viable search and seizure legal issues. But despite the fact that prosecutors often have positive drug field tests and reliable accounts from the arresting officers for these cases, they make no attempt to convict. Instead, they quickly offer a conditional dismissal without further inquiry, a disposition that marks the defendant for a limited amount of time.

Note that all actors could agree on this disposition even as they remained subjectively far apart on their personal determinations of factual guilt. John, the young man accused of swiping a disabled-discounted MetroCard, insisted to his attorney he was innocent. The defense attorney urged John to accept the ACD offer, even with one day of community service. John had shown his attorney the employment letter to prove that he worked for the social service agency, but the letter also indicated that he was paid over $30,000 a year in that job. At that salary he would not qualify for public defender services. The defense attorney explained to John that he would have to hire a private attorney to come back to fight the case if he wanted to be acquitted at trial, and most likely that would take over a year of court appearances. Furthermore, the arresting officer signed a supporting deposition saying he saw the defendant swipe a card that indicated by the lights of the turnstile that it was a MetroCard for a disabled person. The attorney warned his client it was far from assured that a judge would believe the defendant over the arresting officer at trial. (133)

The defendant also told his attorney that he had lost his job because of this arrest. His defense attorney explained that, in terms of employment collateral consequences, taking the ACD today was a better bet than fighting the case. If he took the ACD today, it would only be open to be viewed on the publicly accessible WebCrims database for six months, and it would indicate the case was adjourned in contemplation of dismissal. If he chose to fight the case, it would remain an "open" matter on the website for the entire period of time it would take to push the case to trial. It is worth noting that the defense attorney was not exaggerating in his estimation of how long it would take to fight the case: in that borough the mean docket age at disposition of cases that proceed to jury trial is well over 400 days. (134) If the defendant lost at trial of course he would then be marked with a permanent criminal record and all the attendant collateral consequences, all for a case where the prosecution offered a conditional dismissal at arraignment. The miniscule trial rates for misdemeanor cases make clear that few defendants find it in their interest to take this chance.

Not all cases disposed of at arraignment terminate in a form of dismissal. In 2011, about 44% of arraignment dispositions were ACDs or dismissals at arraignment, and around 51% were convictions of some type (violations or misdemeanors). (135) It is not infrequent for jail sentences to be imposed at an arraignment plea. Again, the defendant's record largely dictates both the incentives to take the plea at arraignment and the sentence.

In February 2011, a thirty-something, black man was arraigned for petit larceny. (136) He was accused of stealing a package of Cracker Barrel cheese. Because of his recent arrest history he was deemed subject to "Operation Spotlight," a citywide initiative targeting persistent misdemeanor recidivists. (137) The DA's policy in Operation Spotlight cases is to not engage in plea bargaining (at least at arraignment), and instead to recommend a plea to the top charge and the maximum statutorily allowed sentence, which in this case was a conviction for a Class A misdemeanor and one year of jail. Judges can, and often do, make their own offers on these cases, but they are limited to sentencing within the range of the top charge. (138) In this case the judge offered a plea to the Class A misdemeanor with five days of jail, and the defendant readily accepted.

I have no idea if this defendant was in fact guilty of stealing the Cracker Barrel cheese. I do know that had the defendant not accepted the judge's jail offer, the judge most likely would have set bail because of the defendant's substantial recent arrests and bench warrant history. The defendant would have then been held in custody for a longer time than the offered sentence because a person does two-thirds of city time (so four days on a five-day sentence), and an in-custody case is typically adjourned for at least five days to allow the prosecution the statutory time to convert the charging document. (139) This defendant, like most defendants who plead to misdemeanor crimes with jail time at arraignments, was already marked with a prior criminal conviction. Therefore, taking this plea and sentence at arraignment allowed him to minimize jail time, even if he incurred another conviction mark. The incentive structure illustrated here helps us make sense of the quantitative trends documented in Part V, which show that the probability of conviction on a new misdemeanor arrest increases substantially with the number of convictions the defendant bears at the time of arrest.

Even the jail sentence entailed an element of marking. Any portion of a day spent in custody is credited as a full day toward sentence time, including the day of arrest. This defendant was arrested on a Tuesday, sentenced on a Wednesday, and therefore would be released two days after the plea. A defendant arrested on a Thursday and sentenced to five days on a Friday will be immediately released because if a release date falls on a Saturday or Sunday the defendant is released on the Friday prior, as the New York City Department of Corrections does not release on weekends. However, this does not mean that the noted jail sentence is meaningless. It sets a floor, and, as one judge explained it, "most judges conform to the folkways of the system," which is "this mechanical ratcheting of sentence length," often in five-day increments for each subsequent conviction. (140) As one defense attorney explained, even when a jail sentence of a specific length is in fact equivalent to a time served sentence, many prosecutors will seek the day-denominated jail sentence because it is "a note to a future prosecutors not to offer less time on the next case." (141)

B. The Uses of Adjournments

Cases survive past arraignment for various reasons. The DA's office may have a policy of not making offers at arraignment in certain types of cases, for example as discussed above, if the case involves a complaining witness or driving while intoxicated. Sometimes the defendant is in a position where she cannot take the standard offer at arraignment without significant collateral consequences that outweigh the cost of protracted adjudication, such as immigration consequences or a potential violation of parole or probation triggered by a conviction. Some defendants decide to fight the case either because they are claiming innocence, there was a potential legal defect in the stop, search, or arrest, or simply because the defendant or the defense attorney thinks the offer made at arraignment is not fair.

Plea offers often get better with later court appearances. Sometimes, however, prosecutors threaten higher offers if the arraignment offer is not accepted or make "one time only" offers at the second or third court appearance. This practice discourages not only trial as a means of settling factual disputes, but also the efforts of defense attorneys, who often have upward of eighty pending cases at once, to perform basic factual or legal research before urging clients to accept a plea. For example, in Manhattan certain designated quality-of-life cases are assigned to a specialized courtroom where the apparent policy is to propose a more punitive plea than the arraignment offer after the prosecutor has declared "ready" for trial, or sometimes at any later court appearances. (142)

The majority of defendants whose cases continue past arraignment are "out," the colloquial term indicating the defendant is not being held at Rikers Island on bail. Bail has been set in about 9% to 10% of misdemeanor cases in recent years (and in about 21 % to 24% of the cases surviving arraignment); in the remainder of cases defendants are released on their own recognizance. (143) The overwhelming majority of misdemeanor defendants are indigent, so despite the fact that the average bail in a misdemeanor case is about $ 1000 or less, only about 10% of defendants with bail can make it at arraignment and another 27% make bail after arraignment. (144) Defense attorneys summarize this dynamic as "bail means jail," meaning if bail is set, most defendants will remain in custody until disposition of the case. Bail changes the dynamic of a case that continues past arraignment. Defendants are much more likely to take a plea to get out of jail than they would if they were outside fighting the case. (145) This is so because the time a defendant would wait inside to push a case to trial is usually much longer than the jail term the defendant would be facing if he agreed to take a plea. In recent years, around 80% of the jail sentences imposed for misdemeanor convictions were less than 30 days, whereas the mean docket age of misdemeanor cases commencing jury trial is over 400 days, and the mean docket age of misdemeanor cases commencing bench trial is over 350 days. (146) Even if a defendant feels there is a legal or factual issue to be litigated, there are strong incentives to accept quick disposition as opposed to pushing those legal and factual disputes to formal adjudication, especially if the defendant is being held on bail.

In November 2010, Ted, a young, black man, was arrested at his home for being in bed with a woman who allegedly was the complainant on his open domestic violence case during a random NYPD home visit of addresses listed on active domestic violence orders of protection. The case was en route to 30.30 speedy trial dismissal because the complainant had not participated in the prosecution since the arrest and the case could not be proved without her testimony. Yet, during the speedy trial time, there remains in effect a "full" order of protection prohibiting the defendant from any contact with the complainant. Ted was arrested for contempt of a court order, and bail was set at his arraignment for this new arrest.

The order of protection listed the address where they found the woman and Ted together as the complainant's address, which Ted was prohibited from visiting. The defense attorney argued to the judge that this address was the defendant's legal residence and produced documents to show the defendant was the only person listed on the lease. Presumably, the complainant was living with the defendant at the time of the incident (and most likely immediately after the incident uninterrupted) and so gave this address to court officials making up the order of protection. (147) The defense attorney argued that the complaint should not be considered "converted" for purposes of stopping the 30.30 clock because the arresting officer had no firsthand information about the woman's identity. (148) He was relying on hearsay, namely the address listed on the order of protection, from which he inferred the woman was the complainant.

The judge summarily rejected the defense attorney's argument, merely saying that the issue of the woman's identity was an issue for trial. The ruling meant Ted would be held in custody until the disposition of the case because he could not afford bail. Ted eventually pled out on both the underlying case, which was clearly on track for dismissal prior to the new arrest, and the contempt case. Again, I have no idea what the true underlying facts in this case were, but neither did any of the legal actors in the system. This case illustrates that the structure of incentives, and not necessarily the legal or factual merits of the case, often drives disposition.

For those cases that continue past arraignment, case adjudication is just as rapid and resource constrained as disposition at arraignment. All-purpose courtrooms processing misdemeanor cases may have between 40 and 150 cases calendared each day, with thousands of open dockets. (149) Although cases are often adjourned for four to six weeks between court appearances if the defendant is "out," both the defense attorney and the prosecution have limited time and resources to devote to additional investigation of adjourned cases. ADAs' caseloads often number between 100 and 200, and defense attorneys can have as many as 80 to 150 open misdemeanor cases at a time. (150) Furthermore, aside from minor assault in the third degree, the largest misdemeanor arrest categories (marijuana possession, drug possession, turnstile jumping, etc.) are those in which there is no complaining witness. These cases largely turn on conflicting accounts between the arresting officer and the defendant.

In March 2012, Trevon was arrested and given a DAT for theft of services--otherwise known as turnstile jumping, a Class A misdemeanor that carries the potential penalty of one year in jail. (151) At his April arraignment the prosecution offered a plea to disorderly conduct and one day of community service. Trevon told his defense attorney that he, in fact, swiped his "unlimited" MetroCard to enter the subway platform on the day of arrest. He also presented the attorney with the unlimited MetroCard and proof that he purchased it from the bookstore of his community college with his student identification. Trevon and his attorney rejected the arraignment offer because Trevon insisted he was innocent of turnstile jumping and because pleading guilty would carry significant collateral consequences for him.

At the time of arrest Trevon had an open felony case for possession of a controlled substance in another borough--I will call it Borough C. His March misdemeanor arrest occurred during the final stretch of a one-year alternative drug treatment program; upon its successful completion, his felony case would be dismissed and sealed. The program required him to attend regular office visits with a social worker, group therapy, and support sessions, as well as to give urine for a drug test once a week, to be in school or employed, and to go to monthly court appearances. According to Trevon, he had successfully completed all these requirements and was slated to graduate from the drug treatment program and have his felony case dismissed the week after his new misdemeanor arrest. He was also slated to graduate from a local community college with his associate's degree. However, because of this new arrest, the court in Borough C would not dismiss the felony case until the theft of services case in Borough B was resolved.

Trevon's defense attorneys subpoenaed the swipe history of Trevon's MetroCard and also verified the purchase of the unlimited MetroCard with a reprinted receipt from the college bookstore. The felony court in Borough C kept Trevon's felony matter open for an additional five months as he attempted to fight his misdemeanor arrest in Borough B. During this time he was required to continue his monitoring and program attendance. Furthermore, according to Trevon, he was turned down for a position at a local transit authority because he had an open felony matter and was told to return when his felony case had been dismissed and sealed, which was contingent on the misdemeanor being resolved. In October the felony court in Borough C dismissed the felony case, finding he had been in substantial compliance and that the theft of services case was weak in light of the evidence of Trevon's MetroCard swipe history.

The defense attorney in Borough B presented the MetroCard evidence to the ADA assigned to his case. He asked her to dismiss the case outright because it showed both that the young man was in possession of an unlimited MetroCard and that he had in fact swiped the card at the location and time of arrest. The supervising prosecutor in the specialized quality-of-Iife misdemeanor courtroom refused, telling the defense attorney that the arresting officer insisted that the young man attempted to jump the turnstile, but then, according to the officer, noticed the police mid-jump and only after that swiped the MetroCard. She offered a plea to a Class B misdemeanor with a sentence of time served.

Trevon eventually made fourteen court appearances over a period of eight months, spending most days waiting between two to four hours for his case to be called, only for it to be adjourned again for a later date. Eventually in December 2012 the case was set for trial. On the first day of the trial the prosecution offered the defendant a plea to a disorderly conduct violation and one day of community service, which the defendant declined. After two days of a bench trial, which included four witnesses--the arresting officer, a Metropolitan Transportation Authority records custodian, Trevon's college bookstore manager, and Trevon--he was acquitted. Watching the two supervising DAs and the law student who had tried the case walk out of the courtroom, Trevon said:

   It's hard not to hate the DA after that. I mean a lot of black
   youth like me just take it ... we don't take the case to court,
   don't want to fight them petty charges, so many people I know just
   don't want to go through the system, don't want to get up and come
   back to court so take those little charges, take time served to go
   home and get it over with. That's why officers get way with petty
   stuff, they throw a lot of cases at you and you get used to it.
   (152)


This rare trial case is the exception that proves the rule. In almost all cases, administrative factfinding is the first and final venue of factfinding. Defendants and defense attorneys are not overestimating the costs of invoking adversarial due process. Pushing a misdemeanor case to trial involves significant time, willingness to make numerous court appearances, and the costs of having an open, pending criminal matter readily accessible to the public and potential employers. Prosecutors often offer, as they did in this case, dispositions that secure some marking of the defendant but minimal formal sanction, and defendants risk serious costs--namely, a permanent criminal record--if they seek to vindicate their full factual innocence. When, as in this case, the charge comes down to a credibility contest between the arresting officer and the defendant in front of a judge, it is a significant gamble for defendants to attempt vindication by trying the facts in an adjudicative venue. It should be noted this was not an obvious win. The judge made a number of legal rulings against the defense in the case and, in a post-trial discussion with the judge, prosecutor, and defense, the judge indicated he thought this was a "tough case."

Sometimes prosecutors will keep cases open knowing they will eventually be dismissed under the speedy trial clock, and will make no attempt to prosecute the case in the interim. This practice serves the goal of marking and risk management. The statutory time allowed to prosecute a case is therefore not only a guarantee to the defendant; it is a tool for the state to monitor and sometimes to punish the defendant.

In the following case, the defense attorney's client was accused of domestic violence by a woman who was his girlfriend at the time of arrest. They had subsequently broken up. The woman, however, was pregnant with the defendant's child, and she refused to participate in the prosecution and requested that the court modify the temporary order of protection from a full to a limited order. According to the defense attorney, the woman wanted the man to be present at the birth of their child so they could immediately and automatically establish paternity, and so that he could assist in child care and pay child support. The defendant was on parole for life and according to the defense attorney the ADA would not make an offer to anything below a "letter" misdemeanor because of his parole status. The defense attorney did not want her client to plead to a misdemeanor criminal offense because that would open him up to reincarceration on a parole violation, most likely for a time between one to three years:

So that case was "ready" forever, I've had it for four months now probably, and it was on for hearings and trial last week and they finally announced "not ready" so the clock started ticking.

[Q: So why did they announce "not ready"?]

Well, she said the officer wasn't available but she told me off the record too that she would rather let the case 30.30 out than agree to a limited order of protection, for example, which is what my client and his ex want, or dismiss it.

....

[Q: Why?]

She would rather let it 30.30 than ever agree to change the order of protection to a limited one or offer a violation or dismiss it.... They came down on the time, they wouldn't budge on the letter.... (153)

When asked why she thought an ADA would agree to let a case become dismissed, but not agree affirmatively to dismiss, the defense attorney stated:

   I mean because it covers, they see it as covering their backs, you
   know they're not on record as agreeing to dismiss a case, they're
   not on record as agreeing to adjust the order of protection to be
   limited, you know, on the off chance that something happens again,
   it's not their fault, they tried to prosecute it they ran out of
   time because of the statute. (154)


These examples demonstrate what I have largely observed in my fieldwork. prosecutors are not always conviction or sentence maximizers; sometimes they are risk minimizers. The overarching imperative is to secure some disposition that allows for a period of monitoring in order to keep track of law enforcement contacts over time--it is not to determine guilt and impose punishment for specific acts. The tools of dismissal and noncriminal violation convictions sort the population flowing through the court by marking them for limited periods of time, and that can provide leverage to impose more serious sanctions if there are subsequent criminal justice contacts. Defendants have a strong incentive to accept those dispositions early. Sometimes the disposition is accepted because it appears to the defendant as a genuinely fair offer given the conduct at issue. Sometimes it is the only rational choice given the structure of incentives.

Misdemeanor courts operating under the managerial model therefore often generate nonconviction dispositions for early encounters, even when prosecutors and judges have sufficient evidence to prove guilt at trial. But the primary targets of quality-of-life policing--namely, young men of color living in highly policed neighborhoods--can quickly use up their early chances and transition into more serious marks. Over the long term, these individual encounters add up and can eventually result in a criminal conviction because there is an additive logic to the managerial model. The following Subpart explicates this additive logic.

C. The Additive Imperative: Building upon Criminal Records

If a defendant with a nonconviction disposition from a prior arrest is brought back to criminal court on a new arrest, the offer on the new case will go up along one vector or another--the seriousness of the mark, the conditions he must satisfy to be granted the disposition, or the formal sentence. This additive imperative is so widely practiced that it is rarely explained.

There are a variety of ways prosecutors discuss it, but all prosecutors whom I asked explicitly about how they formed offers or recommendations consistently expressed some version of the additive logic. Jill, a supervisor in the DA's office, explained, "We try to build on prior cases." (155) Al, another longtime prosecutor said, "Our offers are progressive, first the ACD, then the violation, then the misdemeanor, etc. etc., etc." (156) Another supervising ADA with over two decades of experience explained, "We do progressive pleas, we think everyone deserves one bite at the apple, an ACD is a dismissal but one way to phrase it is it involves a six-month probationary period." (157) Prosecutors must decide who "deserves another bite at the apple" and who does not in a very rapid procession of cases. They face these cases with severe resource constraints (time being the most limited resource, but also investigative capacity) and uncertainty (conflicting accounts of the events from police paperwork, the defendant, and the complainant). Under these conditions the prior record of the defendant becomes one of the most important determinants of the outcome.

Even if early marks, such as the ACD, have a limited lifespan, they can be "built upon" if a person is rearrested before the expiration of the mark. New York City's current policing strategies rely on making very large numbers of stops (between 685,000 and 500,000 in recent years), summonses and citations (over 500,000 in 2012), and arrests for low-level offenses (over 230,000 in 20 1 2). (158) Certain individuals, namely young men of color, tend to have a lot of police contact over short periods of time because, among other reasons, these policing tactics are highly spatially concentrated. These individuals may use up their bites at the apple quickly. Leslie, a public defender, explained the dynamics of intensive policing practices in certain neighborhoods, combined with the additive logic of the managerial misdemeanor justice system, and the incentive structure of lower courts as follows:

   [D]on't forget a person can be stopped and searched fifteen times
   before they're arrested. And often are. People are getting stopped
   and searched their whole lives. And then they get an arrest. And
   then they get an ACD. And then they get a second arrest. And it's a
   discon [disorderly conduct]. And then once your fingerprint even
   reflects contact with the system you're in a different posture....
   You get a discon and then you get a misdemeanor, and then you get
   jail time.... But I think that the discon resolution is underrated,
   in terms of the effect that it has on people's lives. Especially
   for young people getting arrested ... because a discon appears on
   your rap sheet. So you think a discon is no big deal--it's a
   violation, it's not a crime. But it appears. And it will turn into
   a misdemeanor if you are at all at risk at having the increased
   police contact--which lots of our clients are. So if you have a
   bullshit arrest, right, but you decide to take a discon to get it
   over with. Next time you don't have a bullshit arrest, and you have
   less truth--less real exclusionary power or whatever. And you can't
   litigate it, because you're going to lose. You're not going to get
   the discon because you took it on your bullshit arrest, which you
   should have litigated ... especially for, like I said, people who
   are getting stopped and searched all the time.

   ... [T]hat's how a criminal record builds, and that's how the
   population that's affected by unreasonable and unlawful searches
   and seizures is getting crushed by it. Because they have no
   leverage, because they've taken so many pleas. Good pleas, bad
   pleas. They were guilty, they weren't. The main thing is to get out
   of jail. (159)


This quote illustrates a number of important features of the managerial model in misdemeanor justice. First, the people targeted by the policing techniques that emerged in the mid-1990s have frequent low-level contacts. Second, there is a range in the formality and intensity of those criminal justice contacts, but the records and marks created by the formal contacts have profound implications for how a person is treated in later encounters. Third, the incentives created by the costs of dispositions relative to the process costs of court proceedings largely structure outcomes. Once a defendant is in a certain posture vis-a-vis the managerial system as a result of his prior marks, it is much harder to push the adjudicative framework because a new set of constraints pops up. A person with a criminal record has a significantly diminished likelihood of being believed in the abbreviated administrative factfinding pervasive in misdemeanor courts. A person with a criminal record also has significantly diminished incentives to withstand process costs because she already has the mark of a conviction, and many defendants would rather take another conviction than stay in jail or come back to court for months to fight the case.

V. A Quantitative Story of Managerial Misdemeanor Justice: The Import of Prior Marks

This Part uses quantitative data to illustrate the managerial modality in misdemeanor justice, and to distinguish it from patterns we would expect from a more adjudicative model in felony justice, by showing the relative import of prior marks in predicting outcomes. I use a unique data set of criminal arrests and dispositions obtained from the New York State Division of Criminal Justice Services. (160) I slice these data in two ways.

In Subpart A, I present the criminal justice trajectories of two different cohorts of people who enter the misdemeanor justice system. These two cohorts represent the only two coherent groups of people who both newly enter the low end of the criminal justice system each year and are reliably traceable over time. (161) They also represent two divergent paths of misdemeanor case disposition: dismissal and conviction. The dismissal cohort tracks the population of misdemeanor arrestees without prior criminal convictions arrested in 2003 and 2004 whose cases terminated in an adjournment in contemplation of dismissal specific to marijuana offenses. (162) I will call this cohort the MJACD cohort for the remainder of the Article. The conviction cohort tracks the population of misdemeanor arrestees without prior criminal convictions at the time of a 2003 or 2004 misdemeanor arrest whose cases terminated in a first-time conviction for a misdemeanor criminal offense. I will call this cohort the misdemeanor conviction cohort for the remainder of the Article. 1 track their criminal justice trajectories from the date of the initiating disposition (any time between 2003 and 2004) until June 2011, which means each individual was under observation for somewhere between 6.5 and 8.5 years. (163)

I use their trajectories to highlight certain empirical regularities about misdemeanor justice during the age of mass misdemeanors. First, insofar as these cohorts are representative of the people arrested each year for a misdemeanor offense without a prior criminal record, the majority of this population stays on the low end of the criminal justice system as opposed to transitioning up to felony convictions. Second, a significant number of the people have repeated encounters with the criminal justice system and many are arrested but not convicted multiple times. I also use this data to set up the logit analysis and to provide some descriptive sense of the individuals whose criminal justice encounters constitute the micro-level data for the logit models.

Although the MJACD cohort has a rate of rearrest about eleven percentage points higher than the misdemeanor conviction cohort, it has a criminal conviction rate about twenty percentage points lower than the misdemeanor conviction cohort. As a proportion of cohort members with one or more later arrest, the misdemeanor conviction cohort has a higher percentage of individuals than the MJACD cohort of members later convicted exclusively of misdemeanor crimes. Interestingly, the percentage of arrest events allocated between felony and misdemeanor offenses is quite similar between the MJACD and misdemeanor conviction cohorts--respectively, about 71% and 69% of each cohort s later arrests are for misdemeanor offenses and about 27% and 28% are for felony offenses. However, the two cohorts diverge substantially on later misdemeanor convictions from misdemeanor arrests; yet they have strikingly similar patterns of felony conviction from felony arrests. These results set up the question for the next Subpart: are the divergent misdemeanor conviction patterns between the cohorts produced by different attributes of their constituent members, or by the fact that the conviction cohort consistently encountered the misdemeanor justice system with the mark of a prior misdemeanor conviction?

In Subpart B, I slice the data another way, switching the unit of analysis from individuals over time to the arrest event, in order to see if the differences documented in the cohort study withstand controlling for measured differences. I ask if the mark of a misdemeanor conviction changes the probability of conviction for a misdemeanor offense and if a similar relationship holds in the felony context. I combine all arrest events for the two cohorts and model the likelihood of on-par conviction" separately for felony and misdemeanor arrests using logit models, and display estimated probabilities of on-par conviction so that the impact of prior convictions can be compared between misdemeanor and felony models. (164)

These administrative data contain only limited information that can be used to control for defendant and case differences, such as ethnicity, race, age, and arrest charge. In addition to including those variables in the models, I also include the actual number of recorded arrests for each defendant, which includes all dismissed and sealed cases linked to the defendant's NYSID number. (165) Accounting for the actual number of recorded arrests the defendant had experienced at the time of disposition, in addition to the number of prior convictions and open cases readily visible to prosecutors and judges, is the best available variable for getting at the frequency of police contact. These administrative data contain no information about the legal characteristics of the case, such as the strength of the evidence or circumstances of the arrest. These are all limitations on this analysis. (166)

Results consistent with the dominance of the managerial model in misdemeanor court would show a statistically and substantively significant effect of prior misdemeanor convictions on the likelihood of conviction from a later misdemeanor arrest. This is both because there is a direct effect of the conviction on how criminal court actors treat the case and because conviction is associated with certain defendant characteristics that lead to an initial conviction and also to subsequent convictions. (167) Insofar as the entire logic of the managerial model is to create and store records about certain behaviors (e.g., having frequent police contact, failing to make court appearances, and failing to perform assigned community service) and those facts about the person are used to determine case outcomes, we would expect that the effect of the mark on likelihood of later misdemeanor conviction to reflect both the direct marking effect and the fact that the mark is associated with certain general defendant characteristics.

I believe that the different cuts of the quantitative analysis and the comparison of the effect of prior convictions between the felony and misdemeanor contexts, combined with the qualitative data presented in Part IV about the actual manner and time frame during which misdemeanor cases are resolved, triangulate my argument. That is, defendants' records, instead of legal and factual characteristics of specific cases, play a more important role in determining later conviction in the misdemeanor context than in the felony context.

A. Cohort Description and Trajectories

The MJACD cohort is limited to those individuals with no prior criminal convictions whose misdemeanor arrest in 2003 or 2004 terminated in a marijuana adjournment in contemplation of dismissal. (168) In recent years, almost half of all misdemeanor arrests have terminated in some form of dismissal, but the MJACD group is the only coherent dismissal cohort that can be reliably tracked over time. (169) This dismissal cohort is also theoretically important because marijuana offenses have made up the largest category of arrest and arraignment charges in New York City in recent years. (170)

The misdemeanor conviction cohort includes all those persons with no prior criminal convictions whose misdemeanor arrest in 2003 or 2004 terminated in a first-time criminal conviction for a misdemeanor offense. The Statistical Appendix displays descriptive data on the cohorts.

1. Rearrest and conviction of cohorts by individual members

Figure 11 displays the percentage of each cohort achieving selected arrest and conviction outcomes after entering the cohort as a proportion of the total cohort. This Figure shows that the misdemeanor conviction cohort has a lower rate of rearrest compared to the MJACD cohort--approximately 60% of the misdemeanor conviction cohort has one or more later criminal arrests after entering the cohort compared to about 70% of the MJACD cohort. However, they have a higher rate of reconviction--about 41% of the misdemeanor conviction cohort has one or more later criminal convictions after entering the cohort compared to 21% of the MJACD cohort. (171) The misdemeanor conviction cohort also has a slightly higher felony arrest rate--37% has one or more felony arrests after their first criminal conviction for a misdemeanor offense, whereas 33% of the MJACD cohort has one or more felony arrests after the date of the arrest that led to the MJACD disposition. (172)

Figure 12 displays the percentage of each cohort achieving selected conviction outcomes after entering the cohort as a proportion of the cohort experiencing one or more later criminal arrests. Considering that portion of the cohort that experiences one or more later arrests for a criminal offense, a much higher proportion of the misdemeanor conviction cohort goes on to be convicted of a criminal offense relative to the MJACD cohort--69% versus 29%. Figure 12 also shows that, conditional on one or more later arrests, a higher percentage of the misdemeanor conviction cohort goes on to be convicted exclusively of misdemeanor crimes in comparison to the MJACD cohort: 45% of the misdemeanor conviction cohort goes on to be convicted only of misdemeanor offenses compared to 14% of the MJACD cohort. The absolute difference between the two cohorts with respect to experiencing one or more felony convictions is about nine percentage points.

The cohorts are similar with respect to the low proportions of individuals who are later convicted of a violent felony offense (VFO), as defined by New York State. The category of violent felony offenses includes most major street crimes, such as murder, most classes of manslaughter and rape, most classes of sexual assault and abuse, kidnapping, most classes of robbery and burglary, simple felony assault (in addition to specific provisions against assaulting a police or peace officer and gang assault), and most classes of weapons possession. (173) Figure 13 displays the violent felony arrest and conviction rates for each cohort.

It is notable that both groups exhibit a very similar rate of attrition from VFO arrest to VFO conviction--between 15.4% and 14.5% of both cohorts are arrested at some point for a VFO but only about 4% of both cohorts are ever convicted of a VFO during the 6.5 to 8.5 years under observation in my study. The similar rates of attrition for VFOs are in marked contrast to the divergent rates of attrition from misdemeanor arrest to misdemeanor conviction.

2. Re arrest and conviction of cohorts by arrest events

Another way to approach the trajectories of these two cohorts is to shift the analysis from the unit of arrestee to the unit of arrest, and to examine the distribution of arrest and conviction events for each cohort. Data used to generate these Figures reveal that the distributions of arrest types of the MJACD and misdemeanor conviction cohorts are very similar: between 69 and 71% of all arrest events are for misdemeanor crimes, and between 27% and 28% of all arrest events are for felony crimes, respectively for each cohort. (174) Figure 14 compares the disposition patterns from all arrest events experienced by both cohorts after the initiating arrest separately for misdemeanor and felony arrests.

The dissimilar disposition pattern from misdemeanor arrests between the MJACD cohort and the misdemeanor conviction cohort is striking. Whereas 12% of the MJACD cohort's later misdemeanor arrests resulted in a misdemeanor conviction, 54% of the misdemeanor conviction cohort's later misdemeanor arrests resulted in a misdemeanor conviction. The MJACD cohort experienced many more dismissals from subsequent misdemeanor arrests than the misdemeanor conviction cohort: 52% of the MJACD cohort's later misdemeanor arrests result in no conviction while only about 26% for the misdemeanor conviction cohort result in no conviction.

The cohorts had fairly similar on-par conviction patterns from felony arrests--between 19% and 20% of the two cohorts' felony arrests resulted in felony convictions. The felony arrest outcomes diverged in the distribution of convictions for low-level crimes: the misdemeanor conviction cohort's felony arrests were much more likely to result in a misdemeanor conviction than the MJACD cohort's felony arrests.

The following Subpart will explore this divergence in later misdemeanor conviction rates between the cohorts by modeling the conviction event and controlling for measured factors about the arrestee and arrest type. Using logit models, I ask if the mark of a prior misdemeanor conviction increases the likelihood of a later on-par conviction from a misdemeanor arrest and how that differs from the effect of a prior felony conviction on the likelihood of a later on-par conviction from a felony arrest.

B. Models of Misdemeanor and Felony Conviction

The following Figures depict the estimated probability of on-par conviction separately for misdemeanor and felony arrests as derived from a series of logit models explained and reported in the Statistical Appendix.

Figure 15 graphs the probability of conviction on a misdemeanor offense for a twenty-three-year-old black male with zero prior felony convictions, ten prior arrests, and no other current open case arrested for a controlled substance misdemeanor over the range from zero to ten misdemeanor convictions. It shows that the absolute probability of conviction for a misdemeanor crime rises dramatically, from about 14% to 78% over the range of having zero prior misdemeanor convictions to having ten or more prior misdemeanor convictions, with the number of arrests and other variables held constant. Furthermore, the biggest single jump in predicted probability is from having zero misdemeanor convictions to having one conviction, a 90% increase in the level of predicted probability of conviction (an absolute difference of about thirteen percentage points). That is, there is an increase in the probability of conviction for a misdemeanor crime from a misdemeanor arrest of about 90% associated with moving from zero to one prior misdemeanor on the record at the time of disposition of a person otherwise identical on measured characteristics. There is another significant jump in predicted probability from having one prior misdemeanor conviction to having two prior convictions, an increase of about 71% in the relative probability of conviction (an absolute difference of about nineteen percentage points). Subsequent convictions increase the probability of conviction by smaller intervals.

Figure 16 displays the predicted probability of conviction for a felony crime from a felony arrest for a similar person, a twenty-three-year-old black male with ten prior arrests, no other current open case, and one prior misdemeanor conviction. This Figure shows a much smaller increase in the predicted probability of on-par conviction from a felony arrest for a person otherwise identical on measured characteristics associated with moving from zero to one prior felony conviction at the time of disposition: an increase of about 10% (an absolute difference of three-and-a-half percentage points). A person otherwise identical on measured characteristics with two prior felony convictions actually has a slightly lower predicted probability of felony conviction from a felony arrest compared to a person with a single prior felony conviction.

The different results of the open cases variable between the misdemeanor and felony models suggest that records of law enforcement contact play a more important role in case disposition in the former than the latter. The presence of one other open case instead of zero at the time of disposition is associated with an increase in predicted probability of on-par conviction of 58% in the misdemeanor model, whereas in the felony model it is associated with an increase in predicted probability of on-par conviction of about 3%. (176)

The next set of Figures was also generated using the logit models reported in the Statistical Appendix. These models exploit dummy variables of prior arrests, which allows the relationship with the likelihood of conviction to be nonlinear. Figure 17 shows the adjusted predictions of probability of conviction for six separate levels of prior misdemeanor convictions, zero through five or more, evaluated for a modal arrestee--a black male with no other current open case and zero prior felony convictions--for a misdemeanor controlled substances arrest. The Figure shows the predicted probability of misdemeanor conviction over the number of arrests, two through fifteen or more.

A striking result is that the probability of conviction does not increase as significantly over the number of arrests as it does over the number of prior misdemeanor convictions. For example, the predicted probability of a misdemeanor conviction from a misdemeanor arrest increases by about 38% as an otherwise identical person with zero prior misdemeanor convictions moves from arrest two to arrest eight (an absolute increase of about four percentage points). Yet two people otherwise identical on measured characteristics on their second arrest achieve an increase in predicted probability of conviction of approximately 96% (an absolute increase of over sixteen percentage points), just by assuming one has a single prior misdemeanor conviction and the other has no prior misdemeanor convictions.

This same relationship does not obtain in the felony conviction models. Figure 18 shows a very tightly clustered set of lines plotting the level of predicted probability of conviction for a felony offense from a felony arrest moving from zero to three prior felony convictions, and only a slightly steeper rise over the arrest number.

The results of these models suggest that the mark of a misdemeanor conviction is associated with a higher probability of conviction from later misdemeanor arrests, even after controlling for the number of prior recorded (but not necessarily visible on rap sheet) arrests, open cases, and certain other measured defendant and case characteristics. It is consistent with the managerial model that this effect could be due both to a direct signaling effect of the marks and to a selection effect in which defendants who have certain traits, such as failing to show up to court appearances, tend to get convicted and reconvicted because they do not satisfy the conditions required to earn conditional dismissals offered for early arrest events. Note that the relationship between traits such as tardiness and conviction is dependent on the operation of the managerial model; absent the practice of creating and consulting records about those behaviors, those traits would not result in the observed trajectories of repeated convictions.

However, it would not be consistent with the managerial model if the effect of the prior misdemeanor convictions were due to a specific type of selection, namely defendants' propensity to be arrested under undeniable factual or justifiable legal circumstances. There are several pieces of evidence that make that type of selection an unlikely explanation of the results. First, we would not expect such a significant divergence in the effect of prior on-par convictions across the felony and misdemeanor contexts unless a certain type of person is stopped and arrested lawfully and consistently generates strong evidence of guilt only when engaging in crimes at the misdemeanor level, but not at the felony level. Second, we would not expect such a substantively small effect of number of actual prior arrests (recorded under the NYSID but not necessarily visible on rap sheet) in comparison to the effect of marks visible to prosecutors and judges such as prior convictions and open cases. Third, the data we have on the selection mechanism in misdemeanor courts make it unlikely that courts consistently distinguish between the factual and legal traits of cases, including the fact that over fifty-seven percent of misdemeanor cases reach disposition within twenty-four hours after arrests on the basis of very minimal paperwork about the case, the limited time that legal actors have to devote to legal research and factual investigation, the explicitly stated policy of prosecutors and judges of offering progressively higher pleas for each subsequent case, and the relative costs and incentives defendants face in taking cases to trial. In light of these data, it is much more likely that the relatively small association between prior on-par convictions and the likelihood of felony conviction is a result of the relative importance of legal and factual difference for determining dispositions, even if it is via an administrative or inquisitorial system.

VI. POLICY IMPLICATIONS OF THE MANAGERIAL MODEL: WHAT IS TO BE DONE?

This is the part of a law review article where the author offers her policy prescriptions or recommendations for new legal and procedural rules. But because this Article has been dedicated to establishing an empirical claim and developing a theoretical model--and not diagnosing a problem--we need to start by identifying the policy issues raised by the preceding analysis. What, we might ask following William James, "is the truth's cash-value in experiential terms?" (177)

A number of policy concerns flow from the account of managerial justice presented here. In this Part, I puzzle through just three classes of problems: process costs and accuracy; racial, class, or space-based injustice; and the social role of criminal law. I give fair warning to the reader: this Part does only what it promises--puzzles through these concerns without laying out substantial defenses of precise policy prescriptions. I offer a preliminary pass at indexing the legal, social, and political implications of managerial misdemeanor justice. In so doing, I hope to do justice to the complexity of the legal system I have carefully studied, the institutional environment in which it operates, and the various important interests implicated in the interplay between the two. In the end, I hope to provide the groundwork for meaningful reform efforts grounded in a nuanced understanding of the object of reform.

The first obvious cashed-out implication of the managerial model is that legal determinations in misdemeanor courts might not accurately track guilt and innocence in fact. Because the managerial model of administering criminal justice largely differentiates its treatment of defendants on the basis of marks of prior encounters, it is bound to produce higher rates of type I (false conviction) and type II (false acquittal) errors than an adjudicative model of criminal justice that largely differentiates its treatment of defendants on the basis of judicially or administratively determined facts pertinent to guilt and innocence.

Indeed much of the debate about justice in lower courts in the legal literature focuses on what to make of type I and type II errors. Josh Bowers, for example, argues that we should worry less about type I errors in pursuit of abstract notions of justice than about minimizing the very real and serious burdens defendants bear in the process of trying to vindicate their innocence. He proposes that instead of making it harder for innocent defendants with extensive records to plead guilty, we should make it easier because "[i]n low-stakes cases plea bargaining is of near-categorical benefit to innocent defendants, because the process costs of proceeding to trial often dwarf plea prices." (178) Bowers maintains that insufficient winnowing at earlier stages of the criminal process, from arrest to charging, is the true source of injustice for misdemeanor defendants. Thus he concludes that taking away the possibility of factually false pleas from defendants who have been pushed into the deeper reaches of the criminal process only exacerbates their burdens.

Others resist this conclusion. Stephanos Bibas, for example, concedes that Bowers's suggestion may indeed "maximize the satisfaction of innocent defendants' preferences." (179) Yet he argues that we must nonetheless not "give in to the punishment assembly line, to make it speedier and more efficient and surrender any pretense of doing justice." (180) Along similar lines, Alexandra Napatoff argues that in most large urban misdemeanor criminal justice systems "aggregate decision-making dominates each stage of the process," such that the entire criminal process has become "dissociated from the core culpability concerns of criminal law." Therefore, she argues that "it needs either to embrace the individuated model more fully, or relinquish the punitive moral mantle of criminal law and admit that it is attempting to do something different." (181) These are all legitimate concerns insofar as one holds that the primary value of criminal courts lies in adjudicating factual guilt and innocence. Within that evaluative framework, the weight of these concerns is tied to unknown empirical regularities of factual errors in misdemeanor adjudication.

The data that I have collected in this study cannot be used to estimate the rates of type I and type II errors. I cannot say, based on any independent investigation of arrest and crime circumstances, what percentage of misdemeanor defendants plead guilty to crimes they did not commit, or what percentage of misdemeanor defendants were not convicted of crimes they did commit. It seems reasonable to conclude, based on the dynamics of the managerial model I have presented, that the error types are distributed unevenly among different sorts of defendants according to the marks they bear from prior encounters at the time of disposition. Type II errors are most likely among defendants with no criminal record because the prosecution's policy is often to quickly grant conditional dismissals for these defendants. Type I errors are more likely higher among defendants with prior misdemeanor convictions because they face higher barriers to convince prosecutors and judges of their factual innocence, and because they are less inclined to bear process costs to seek vindication since they already have a criminal record. (182)

Many defendants who are factually guilty are offered conditional dismissals precisely because actors in the system are content to monitor their later criminal justice contacts rather than incur the process costs necessary to convict them. And many innocent defendants plead guilty precisely because they are trying to avoid process costs. Over thirty years ago, Malcolm Feeley noted this tension and called it "the dilemma of lower courts." The dilemma, as he explained it, is as follows: "Expanded procedures, designed to improve the criminal process are not invoked because they might be counterproductive. Efforts to slow the process down and make it truly deliberative might lead to still harsher treatment of defendants and still more time loss for complainants and victims." (183)

So what is to be done?

The dilemma of lower courts constrains the promise of procedural reforms. Added procedures are unlikely to alter the basic calculus of process costs versus short-run disposition costs that defendants face in lower criminal courts, as defendants usually retain the option of invoking or abjuring their process rights. Furthermore, insofar as adding procedural rights is intended to increase adjudicative accuracy, it is not clear that merely providing for them would substantially change the patterns of work in courts. As I have argued, courtroom actors are not merely disregarding due process values, but actually are operating under a fundamentally different model of criminal law administration in which factual adjudication is ancillary to managerial goals.

Of course, certain legal and policy options could advance various values, including reducing the process burdens on defendants and minimizing type I or type II errors. (184) For example, New York State, like many other states, imposes a variety of monetary charges on individuals convicted of both criminal and noncriminal offenses. (185) The amounts may seem small, but to indigent defendants they are often unmanageable. (186) Waiving these fees for indigent defendants would reduce some of the cascading burdens resulting from a minor criminal conviction, which collectively far outweigh any countervailing social benefit, such as generating revenue or the punitive value of imposing additional costs on those convicted of minor crimes. Reforming bail practices could also address both unnecessary burdens on defendants and the rate of type I errors, especially for indigent defendants. If, for example, judges exercised greater discretion to expunge warrants for excused absences, and if rap sheets reflected the length of time the defendants were absent, and if he returned voluntarily, then prosecutors and judges could differentiate the mark of a bench warrant and more accurately attune bail practices to precise signals about reliability in making court appearances. (187) The funding for the essential roles in criminal court has also not kept pace with the influx of new cases in many large jurisdictions. (188) Increasing the number of judges and funding for defense organizations and prosecutors' offices would reduce the resource limitations that drive some of the high costs of invoking adversarial and formal process. Enhancing the institutional capacity of courts may allow them to revert to a more adjudicative model, and possibly reduce the rates of type I and type II errors.

But it is far from obvious that moving misdemeanor courts away from a managerial model toward an adjudicative model is desirable, even if it reflects more faithfully the textbook model of criminal courts. Technical questions of institutional reform quickly give way to questions of social values.

Many people find false convictions self-evidently problematic, for persistent misdemeanants and new entrants to the criminal justice system alike. But our normative assessment of conditional dismissals for factually guilty defendants may indeed turn on whether the defendant is a new entrant or frequent flyer in the criminal justice system. We may, upon reflection, endorse the substantive penal logic of the managerial model: what we ought to do with minor crimes is not necessarily punish the act, but rather assess the person over time to see if he persistently disregards legal rules. Perhaps the punishment of minor crimes implicates moral and social values distinct from those at play in the punishment of serious crimes.

Emile Durkheim famously distinguished penal law from what he called "restitutory law" (such as civil, commercial, procedural, and administrative law). (189) Restitutory laws ensure smooth functioning of the division of labor, and therefore they are still enforced by the state's coercive power to sanction. (190) But because the rules embodied by restitutory law are about the coordination and organization of social tasks, they "do not strike us with the force of sacred entities." (191) As a society, we do not call for expiation when there is an established infraction; we are content with reestablishing the status quo. The defining characteristic of penal law was, for Durkheim, that it addressed acts that had a particular status in our collective conscious, reflecting deep-seated, sacred values. He famously said, "[W]e should not say that an act offends the common consciousness because it is criminal, but that it is criminal because it offends that consciousness." (192) Penal law was thus distinguishable from other classes of law not by the level of harmfulness of the acts it proscribes, but by the nature of the collective sentiments that motivate it, and that it functions to reinforce.

Whether or not we accept Durkheim's overarching theory of punishment, his distinction offers an important insight in pushing us to move beyond juridical definitions in making sense of the actual operations of legal institutions. Different components of the criminal justice system, although under the same juridical category of criminal law, might embody divergent approaches to the practice of punishment because of the social meaning of the legal infractions at issue. Many of the criminal prohibitions addressed in misdemeanor court are acts that straddle the line between restitutory and penal, occupying a liminal status between coordination rules and foundational moral values. A much higher proportion of the acts addressed in misdemeanor court than in felony court simply do not offend widely held moral sentiments that have a central role in producing the possibility of social cohesion. Consequently, it should not be surprising that our social practice of punishment diverges when we address crimes that have a different status in our social collective.

The managerial model cannot be reduced to an organizational response to unmanageable caseload pressures. It also reflects a substantive moral posture toward the punishment of minor crimes. That posture involves not only a grading of punishments for persistent offenders, which is obviously the case in the punishment of felony offenses as well. It also reflects a grading of the judicial effort addressed to the offender in any capacity, including in determining whether or not he committed the proscribed act.

As I have shown, managerial misdemeanor courts are doing something with the population of defendants who are granted dismissals or noncriminal dispositions. They are just not always attempting to convict and impose a formal punishment. Instead, these courts are using the instruments of criminal procedure to monitor and check up on defendants' later behavior, building records about what type of person is involved. Thus, under the managerial model, type I errors are not errors at all; they are part of the very logic of social control that this component of the penal system uses to deal with shallow-end offenders. It is not self-evident, at least to me, that what we want from misdemeanor courts is perfect adjudicative accuracy. Considered abstractly, the managerial approach--only engaging the heavy machinery of criminal law's capacity to permanently mark a person and impose a formal punishment where there is some indication the person persistently flouts the rules--is an appealing principle for administering rules that sit between restitutory and penal.

Yet a number of problems are generated by the operation of a system that is both managerial and administrative/inquisitorial inside of a system that is ostensibly adjudicative and adversarial. One problem is that people are classified as persistent rule breakers, and thereby they become at risk for transitioning to a criminal conviction, by acquiring a set of marks that are proposed to defendants as conditional dismissals and that hold themselves out to have that official legal meaning (which of course is precisely what makes the disposition such an attractive offer early in the case's life course). However, these marks are reinterpreted at a later time--at the point of a subsequent arrest--in a different light to mean at least an indication of ungovernability, and at most a signal equivalent to a guilty plea. Prosecutors' interpretation of these marks and their resultant charging decisions and plea offers are often determinative of the outcome of the case. As discussed in Parts I and IV, managerial misdemeanor courts also are functionally administrative systems where trials are rarely viable routes to vindicate legal issues or factual innocence. There are very limited and informal leverage points to challenge the charging and plea decision of prosecutors. (193)

This raises the question of whether the managerial system in its real-world incarnation accurately and fairly identifies persistent lawbreakers. The qualitative and quantitative data presented in Parts IV and V make it clear that the current operations of lower criminal courts can be said, at best, to only imperfectly do so. As shown in Figure 17, the probability of conviction for a misdemeanor crime from a misdemeanor arrest increases only slightly as the number of arrests rises, but increases substantially with each subsequent criminal conviction. The transition from nonconviction dispositions to convictions is, for the overwhelming majority of misdemeanor cases, largely a factor of the temporal proximity of arrests. And the likelihood of arrest is a function of not only an individual's conduct, but also of policing practices. Those that are brought into the misdemeanor justice system, and those who stand the highest risk of being rearrested, are not a random sample of rule breakers or even persistent rule breakers. It is a sample systematically biased by certain social facts, some of which raise fundamental concerns of racial and class inequities. As I showed in Part II, young black and Hispanic men make up the majority of misdemeanor arrestees.

Therefore, even if the criminal courts fairly and impartially execute the managerial modality over all defendants irrespective of class, race, or immigration status, questions of fairness extend to the mechanisms that select people for misdemeanor arrests. There is substantial evidence that the underlying behaviors of some of the largest arrest categories, such as marijuana and narcotics possession, are fairly evenly distributed across racial and class groups. (194) But these groups face vastly unequal risk of arrest because of the social realities of where and how drugs are sold and used, and because of the density and form of policing in different spaces. (195) Some criminal conduct might be more likely to translate into an arrest because it occurs in communities where police have become an established institution of not only social control but also interpersonal and household control, or because of the conscious or unconscious biases of police officers shaping their discretion to make an arrest for low-level conduct. (196) Other types of behaviors may be unequally distributed. (197) Research of a different type would be needed to apportion the relative contributions of those different factors to the clearly documented concentration of misdemeanor arrests among black, Hispanic, and low-income communities.

This brings me to the third set of concerns that flow from this account of managerial justice. The ostensible objectives of these policing practices, much like the reforms that have led to mass incarceration, are to reduce the incidence of violence and social harm. It may well be that the people living in conditions of "social insecurity and marginality," whose life prospects have been circumscribed "in the wake of the twofold retrenchment of the labor market and the welfare state," are more likely to commit misdemeanor crimes. (198) And it may well be that we have good reason for deploying a high number of police officers to poor and minority neighborhoods because they suffer disproportionately from high crime. And these policing techniques may indeed be effective in creating order and repressing serious crime, although that question is hotly debated among social scientists who have studied the issue. (199) Insofar as the techniques are effective, the crime reduction benefits from these policing strategies accrue to the residents of these neighborhoods.

But the costs of these strategies fall on the same people. And the costs are tremendous. The residents inside these communities are the ones who come to have criminal records that hinder their employment and housing prospects, endure lost days of work and child care, and face interminable demands to go back and forth to court to deal with arrests and summonses for low-level infractions. They increasingly feel disrespected and oppressed by a police presence designated for their safety.

As long as we, as a society, are comfortable with securing social control and order primarily with the tools of criminal law and punishment, this will be the case.

This, I believe, is the primary problem identified by my analysis, not type I or type II errors, or even the failure of lower criminal courts to live up to a due process adjudicative ideal. The crucial problems raised by mass misdemeanors--just as with mass incarceration--are political and social questions. To what extent are we, as a political community, comfortable relying on the instrumentalities of criminal law as the primary public social control mechanisms over spaces that have been devastated by economic structural change and the retrenchment of the welfare state?

CONCLUSION

As we move from what is to be done to how can it be accomplished, history is an indispensible resource. The dominant social-ordering role presently accorded criminal justice institutions in the United States neither is natural nor was its development inevitable. It took a historical process to get to a place where we accept the current capacious operation of the penal state. It involved innumerable political battles fought at the local, state, and federal levels to direct resources to police and prisons while neglecting other social welfare institutions. It involved intentional social movements to shift our cultural posture toward the appropriate roles of penal and welfare policies. Scholars of mass incarceration have carefully traced the historical process by which "we arrive at a political moment where indefinite solitary confinement in a concrete box is sound policy, but cash assistance to poor parents 'has corrupted their souls and stolen their future.'" (200) If we find ourselves uncomfortable with the vast operations of managerial misdemeanor justice extending over poor minority communities, we must understand the political and cultural trends that brought us to accept such a social role for this and other criminal justice institutions.

Other spaces in our cities are ordered with other institutions of social control, such as schools, good jobs, and families. Building up these institutions is a possibility. We can imagine a world in which we call on the state to dedicate the same amount of attention and resources to supporting schools, good jobs, and families as it currently dedicates to building prisons and extending policing. But such a world will not be secured merely through new criminal rules and procedure. It demands a broad movement of social and political dimensions.

Issa Kohler-Hausmann, Law Research Fellow, Georgetown University Law Center; Ph.D. Candidate, New York University Department of Sociology. I am deeply grateful to my advisers on this research, Jeff Manza, David Garland, Malcolm Feeley, David Greenberg, and Lynn Haney. I am also indebted to the following people for generous feedback and comments: Rachel Barkow, Jonathan Bearak, Josh Bowers, Ryan Bubb, Paul Butler, David Cole, Malcolm Feeley, David Greenberg, Jennifer Herwig, Gregg Klauss, Alvin Klevorick, Kay Levine, Gerard E. Lynch, Allegra McLeod, Sasha Natapoff, Mike Seidman, Larry Solum, Matthew Stephenson, David Super, Josh Teitelbaum, Florencia Torche, Tom Tyler, and James Whitman, in addition to the Stanford Law Review, and, with special gratitude, Martin LaFalce.

(1.) The literature on mass incarceration is extensive. For a small selection of the important and substantial literature on the causes and consequences of mass incarceration, see generally Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness (2010); Paul Butler, Let's Get Free: A Hip-Hop Theory of Justice (2009); Mass Imprisonment: Social Causes and Consequences (David Garland ed., 2001); Bruce Western, Punishment and Inequality in America (2006); James Forman! Jr., Racial Critiques of Mass Incarceration: Beyond the New Jim Crow, 87 N.Y.U. L. Rev. 21 (2012); Marie Gottschalk, Hiding in Plain Sight: American Politics and the Carceral State, 11 Ann. Rev. Pol. Sci. 235 (2008); Tracey L. Meares, Mass Incarceration: Who Pays the Price for Criminal Offending?, 3 Criminology & Pub. Pol'y 295 (2004); Dorothy E. Roberts, The Social and Moral Costs of Mass Incarceration in African American Communities, 56 Stan. L. Rev. 1271 (2004); and Louis Michael Seidman, Hyper-Incarceration and Strategies of Disruption: Is There a Way Out?, 9 Ohio St. J. Crim. L. 109 (2011).

(2.) The era of mass incarceration might more accurately be called the era of mass conviction and correctional supervision, as parole and probation populations have grown at an even faster rate than the incarcerated population. From 1980 to 2008 the number of people under any form of correctional supervision--including jail, prison, parole, or probation--increased from just under 2 million to over 7.3 million. After peaking in 2008, the correctional population began to decline for the first time in over thirty years, dipping below 7 million in 2011. See Lauren E. Glaze, U.S. Dep't of Justice, Correctional Populations in the United States, 2011 (2012), available at http://www.bjs.gov/content/pub/pdf7 cpus11.pdf; Michelle S. Phelps, The Paradox of Probation: Community Supervision in the Age of Mass Incarceration, 35 Law & Pol'y 51 (2013).

(3.) There is a dearth of comparable and reliable data on subfelony arrests and case filings because states have so many different statutory provisions that define subfelony offenses, entities that cite for these offenses, and administrative configurations of lower courts. One of the only reliable sources that collects data from many states, the National Center for State Courts, produced from a sample of sixteen states a conservative estimate of 5.9 million misdemeanor fillings in these jurisdictions, compared to 1.4 million felony fillings in 2009. Robert C. LaFountain et al., Nat'l Ctr. for State Courts, Examining the Work of State Courts: An Analysis of 2009 State Court Caseloads 23 (2011), available at http://www.courtstatistics.org/flashmicrosites/csp/images/csp2009.pdf. As I use the term, "subfelony" includes misdemeanor criminal offenses and unclassified misdemeanors, infractions and violations, which are not classified as misdemeanor criminal offenses.

(4.) For a discussion of how mass incarceration has challenged our understandings of the dynamics of criminal procedure, punishment, and the role of criminal law in our democracy, although in different ways, see William J. Stuntz, The Collapse of American Criminal Justice 2-3 (2011) (arguing that the dysfunctional operation of America's criminal justice system, especially our unprecedented levels of incarceration and its racial skew, must be understood in the context of historical changes in criminal justice institutional design and in the content of substantive criminal law, both of which deteriorated as the locus of political control over criminal justice gravitated away from the local); Jeffrey Fagan & Tracey L. Meares, Punishment, Deterrence and Social Control: The Paradox of Punishment in Minority Communities, 6 Ohio St. J. Crim. L. 173, 180 (2008) (proposing a series of mechanisms through which America's criminal justice policies leading to mass incarceration produce the dual perverse effects of unhinging the imposition of legal punishment from its intended social role of deterrence and undermining respect for the law); Louis Michael Seidman, Criminal Procedure as the Servant of Politics, 12 Const. Comment. 207, 207-09 (1995) (arguing that the construction of a set of "elaborate and detailed constitutional protections for criminal defendants" in the United States has done little to check the march of policies that have brought about mass incarceration, and perhaps has even enabled it); and William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1, 3-6 (1997) (arguing that real-world impact of criminal procedure doctrines is not merely, or even mostly, a function of the content of formal rules and legal remedies but rather is shaped by the context of crime rates and the adaptive decisions of prosecutors and legislatures).

(5.) Although New York is not a representative American city in many respects, its law enforcement experiment is widely looked to as a national model for crime control, and the system of misdemeanor justice that has emerged there represents a model of criminal administration distinct from anything discussed in ordinary courses on criminal law.

(6.) As I will explain in Part I, these models are poles on a continuum, not mutually exclusive alternatives. Although others have used the term "managerial" in other contexts, I stake out a new meaning here. For example, Judith Resnik uses the term to indicate the overly administrative mindset of many federal judges, whose focus on docket control and case supervision, she argues, has grown at the cost of impartial deliberation. See Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374, 376-78 (1982). In sociological scholarship the term has been used to indicate a mode of governance. For scholars such as Stanley Cohen, Malcom Feeley, and Jonathan Simon, the term "managerial" indicates a trend in methods of social control of deviance and crime that is oriented to the regulation of populations and control of behavior, without necessarily attempting to rehabilitate offenders. See Stanley Cohen, Visions of Social Control: Crime, Punishment, and Classification 145-48 (1985); Jonathan Simon, Poor Discipline: Parole and the Social Control of the Underclass, 1890-1990, at 109, 137 (1993); Malcolm M. Feeley & Jonathan Simon, The New Penology: Notes on the Emerging Strategy of Corrections and Its Implications, 30 Criminology 449, 452, 455 (1992). My distinct usage is explained in Part I.

(7.) See infra Figures 6-7; see also discussion infra Part III.

(8.) William J. Stuntz, Plea Bargaining and Criminal Law's Disappearing Shadow, 117 Harv. L. Rev. 2548, 2549 (2004).

(9.) By emergent property I mean "patterns of system behavior arising not directly from initial states, from the capacities of individual agents, or from rules, but from agents' interaction." Alec Ewald, Collateral Consequences as Complex Systems 10 (July 2013) (unpublished manuscript) (on file with author) (using concepts from the natural sciences, such as "complex systems," to understand how a pattern of action or a collection of practices around collateral criminal conviction consequences comes into being without a precise, prior large-scale plan of the individual-level actors involved in the system).

(10.) Debra Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 Colum. L. Rev. 551, 586 (1997). Livingston notes that the critique prior to the due process revolution that "assembly-line justice meted out in lower criminal courts for offenses like drunkenness, disorderly conduct, vagrancy, gambling, and prostitution ... was itself one of the surest signs that the criminal sanction was being misapplied," was part of the argument for expanding constitutional and procedural controls over police discretion during the Warren Court era. See id. at 585-86.

(11.) Roscoe Pound, Criminal Justice in America 190 (1945). Other scholars made similar observations. In 1951, Samuel Dash conducted a study of Chicago's municipal courts to assess their operations twenty years after a reform effort to improve administration in Illinois. He concluded that, in the administration of misdemeanor cases, "[s]peed and the resulting careless handling of facts remain important evils," and "[a]long with the hurried atmosphere is the confusion which dominates most of the stages of the proceedings." Samuel Dash, Cracks in the Foundation of Criminal Justice, 46 U. ILL. L. Rev. 385, 388 (1951). That same year Caleb Foote began his well-known study of the handling of vagrancy-type criminal cases by Philadelphia magistrates' courts, concluding that "[procedural due process does not penetrate to the world inhabited by the 'bums' of Philadelphia." See Caleb Foote, Vagrancy-Type Law and Its Administration, 104 U. Pa. L. Rev. 603, 604 (1956). Foote observed, for example, four defendants who "were tried, found guilty and sentenced in the elapsed time of seventeen seconds from the time that the first man's name was called by the magistrate through the pronouncing of sentence upon the fourth defendant." Id. at 605.

(12.) See, e.g., John M. Junker, The Right to Counsel in Misdemeanor Cases, 43 Wash. L. Rev. 685, 685 (1967) ("[A] large majority of the [persons] charged with non-traffic misdemeanors must, if they are financially unable to hire an attorney, face the bewildering, stigmatizing and (especially at this level) assembly-line criminal justice system without the assistance of counsel." (footnotes omitted)); Ralph H. Nutter, The Quality of Justice in Misdemeanor Arraignment Courts, 53 J. Crim. L., Criminology & Police Sci. 215, 216 (1962) ("The physical conditions in Los Angeles Municipal arraignment courts are not conducive to either justice or individual attention.... Like assembly line workers in a factory, all parties operate under a climate which makes it appear that nothing may be permitted to interfere with the smooth operation of the line."); see also President's Comm'n ON Law Enforcement & Admin, of Justice, Task Force Report: The Courts 31 (1967), available at https://www.ncjrs.gov/App/publications/Abstract.aspx7icfrl47397 (describing urban lower courts as having a system of "assembly line" justice sentencing in 1967).

(13.) See 407 U.S. 25, 36(1972).

(14.) Id.

(15.) See Malcolm M. Feeley, The Process Is the Punishment: Handling Cases in a Lower Criminal Court 13 (paperbacked. 1992).

(16.) See id. ("[T]he assembly-line metaphor ignores the complexity of the criminal process, and the casualness and confusion characteristic of decision making in the lower criminal courts."), id. at 160 ("[W]hat is abundantly clear when listening to prosecutors and defense attorneys negotiate settlements is that it is difficult to articulate the factors considered in assessing the worth of a case, not because decision making in the court is arbitrary, ad hoc, or embarrassingly simple, but because it is extremely complex.").

(17.) See id at 25 ("After looking at the courts and failing to find full-fledged adversarial proceedings, they too quickly conclude that all concern for justice has given way to the pressures of heavy caseloads, organizational security, and bureaucratic self-interest.").

(18.) See id. at xxiv, 31.

(19.) See id. at 31, 192, 241-42. Hence the title The Process Is the Punishment, which captured his argument that the costs from an arrest for a minor crime were often felt before a formal punishment was imposed.

(20.) See David Garland, The Culture of Control: Crime and Social Order in Contemporary Society 142 (2001).

(21.) See Am. Bar Ass'n, Standards for Criminal Justice: Collateral Sanctions and Discretionary Disqualification of Convicted Persons 7-9 (3d ed. 2004), available at http://www.abanet.org/crimjust/standards/collateralsanctionwithcommentary.pdf; Jason A. Cade, The Plea-Bargain Crisis for Noncitizens in Misdemeanor Court, 34 CARD0Z0 L. Rev. 1751, 1758-70 (2013) (documenting the extensive immigration consequences flowing from even low-level criminal justice encounters and analyzing how those costs are exacerbated by the practices in lower criminal courts); James Jacobs & Tamara Crepet, The Expanding Scope, Use, and Availability of Criminal Records, 11 N.Y.U. J. LEGIS. & Pub. Pol'Y 177, 177-79 (2008) (conceptualizing the social significance of a criminal record as a "negative c.v." and documenting the expansion of records production, the proliferation of means of circulating and accessing criminal records, and the resultant social and legal implications); Jeremy Travis, Invisible Punishment: An Instrument of Social Exclusion, in INVISIBLE Punishment: The Collateral Consequences of Mass Imprisonment 15, 15-16 (Marc Mauer & Meda Chesney-Lind eds., 2002) (reviewing the varied collateral consequences of criminal convictions in different social arenas, including housing, welfare benefits, and voting).

(22.) While not all scholars have invoked that precise metaphor, a number of recent publications charge misdemeanor courts with mechanical processing of cases and categorical conviction. See John D. King, Beyond "Life and Liberty": The Evolving Right to Counsel, 48 Harv. C.R.-C.L. L. Rev. 1 (2013). King argues that the constitutional rule limiting the right to counsel in misdemeanor cases to those in which actual jail time is imposed is not defensible, because, "[h]istorical and contemporary accounts of how low-level crimes are actually adjudicated support Justice Douglas's 1972 characterization of the system as one of 'assembly-line justice.'" Id. at 21. Josh Bowers argues that prosecutors fail to exercise what he calls "equitable discretion" in charging and seeking convictions for petty offenses. See Josh Bowers, Legal Guilt, Normative Innocence, and the Equitable Decision Not to Prosecute, 110 Colum. L. Rev. 1655 (2010). Bowers continues that it is "institutionally expedient for prosecutors to adopt indiscriminate petty crime charging policies .... Charging thereby becomes mechanical and somewhat categorical when and where it should be most contextualized." Id. at 1704. He also argues that "[a] charge leads almost inevitably and quickly to some adjudication of guilt." Id. at 1709. M. Chris Fabricant conceptualizes quality-of-life policing and the resulting treatment in criminal courts from misdemeanor arrests as collective punishment. See M. Chris Fabricant, War Crimes and Misdemeanors: Understanding "Zero-Tolerance" Policing as a Form of Collective Punishment and Human Rights Violation, 3 Drexel L. Rev. 373, 377-78 (2011). With respect to the misdemeanor criminal process, he notes: "Though often described as 'assembly-line justice' or 'fast food justice,' an auction is perhaps a more accurate, if jarring, metaphor for understanding the arraignment process and its degradation of the core due process value of individualized justice." Id. at 403 (footnotes omitted). Alexandra Natapoff conceptualizes misdemeanor criminal courts in large urban centers as engaging in what she calls "informal aggregation," in which

   pleading petty offenders in bulk without individuated procedures
   effectively creates a "no fault" conviction regime in which the
   fact of arrest is sufficient to induce settlement. While
   professional participants know that the process is "assembly line,"
   the system straddles the no-fault-culpability distinction by
   maintaining the fiction that defendants are receiving individuated
   justice based on personal culpability.


Alexandra Natapoff, Aggregation and Urban Misdemeanors, 40 Fordham Urb. L.J. 1043, 1080-81 (2013) [hereinafter Natapoff, Aggregation and Urban Misdemeanors] (footnote omitted); see also Alexandra Natapoff, Misdemeanors, 85 S. Cal. L Rev 1313 1374 (2012).

(23.) Various authors have combined both forms of the critique. 1 make no attempt to classify authors claims systematically between these two different interpretations because my aim is merely to show that the assembly-line understanding of misdemeanor justice is prevalent, and that it can encompass two conceptually distinct notions, which in turn generate very different hypotheses about courts.

(24.) Max Weber, The Methodology of the Social Sciences 90 (Edward A. Shils & Henry A. Finch eds. & trans., 1949) (discussing the concept of an "ideal type").

(25.) Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev. 353, 363 (1978).

(26.) Id. at 369, 370. Fuller actually argues that item (3) is a sort of emergent property of the other features of adjudication, such that "issues tried before an adjudicator tend to become claims of right or accusations of fault," by virtue of how the process organizes the participants' proofs and reasoning as claims according to some cogent principle. Id. at 369 (emphasis added).

(27.) Issa Kohler-Hausmann, Misdemeanor Justice: Control Without Conviction, 119 Am. J. Soc. 351 (2013) (identifying three penal techniques of social control prevailing in misdemeanor justice--marking, procedural hassle, and performance--and showing how they are deployed to the ends of monitoring, testing, and regulating misdemeanor defendants irrespective of formal legal outcomes).

(28.) See Feeley & Simon, supra note 6, at 452, 457 (describing "new penology," which is concerned with techniques to identify, classify, and manage groupings sorted by dangerousness, and marked by "cost-effective forms of custody and control" and "technologies to identify and classify risk").

(29.) In sketching these features of the managerial model, my aim here is to illuminate the logic of the model, not to evaluate its accuracy or fairness.

(30.) Gerard E. Lynch, Our Administrative System of Criminal Justice, 66 Fordham L. Rev. 2117, 2121 (1998); see also Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869 (2008); Stephanos Bibas, Plea Bargaining Outside the Shadow of Trial, 117 Harv. L. Rev. 2464 (2003); David E. Patton, Federal Public Defense in an Age of Inquisition, 122 Yale L.J. 2578 (2013); Stuntz, supra note 8, at 2549. All of these articles address in different ways how substantive criminal law, namely expansive definitions of offenses and increased sentences and mandatory minimums, have shifted the criminal justice system away from one where adjudicatory decisions are made by judges or juries, and discuss the important implications of these shifts for prosecutorial practices and plea bargaining.

(31.) Lynch, supra note 30, at 2121-23 ("[S]ubstantive evaluation of the evidence and assessment of the defendant's responsibility is not made in court at all, but within the executive branch, in the office of the prosecutor."); see also Gerard E. Lynch, Screening Versus Plea Bargaining: Exactly What Are We Trading Off?, 55 Stan. L. Rev. 1399, 1403-04 (2003) ("Most plea negotiations, in fact, are primarily discussions of the merits of the case, in which defense attorneys point out legal, evidentiary, or practical weaknesses in the prosecutor's case, or mitigating circumstances that merit mercy.... [W]hat radically distinguishes [the administrative plea bargaining model] from the adversarial litigation model embodied in textbooks ... is that the prosecutor, rather than a judge or jury, is the central adjudicator of facts (as well as replacing the judge as arbiter of most legal issues and of the appropriate sentence to be imposed)." (emphasis omitted)).

(32.) Barkow, supra note 30, at 873-74 ("[individuals who make investigative and advocacy decisions should be separated from those who make adjudicative decisions, the latter of which should be defined to include some of the most important prosecutorial decisions today, including charging, the acceptance of pleas, and the decision whether or not to file substantial assistance motions.").

(33.) See Lynch, supra note 30, at 2120 ("One seldom sees an acknowledgment, in the debates over the superiority or inferiority of the adversarial system, that for most defendants the primary adjudication they receive is, in fact, an administrative decision by a state functionary, the prosecutor, who acts essentially in an inquisitorial mode."); see also Barkow, supra note 30, at 882-83, 887 ("As a result of ... pressures and costs of exercising trial rights, the trial is an insufficient check on prosecutorial power. With his or her power to choose from a range of federal criminal laws, to exercise significant leverage over defendants to obtain pleas and cooperation, and to control the sentence or sentencing range through charging decisions, the prosecutor combines enforcement and adjudicative power.... [Federal prosecutors now do not merely enforce the law, they make key adjudicative decisions as well.").

(34.) Lynch describes the negotiations between the federal prosecutor's office and white-collar defendants as adjudicatory:

   [P]rosecutors ... see themselves as public officials making a
   decision that is in substantial part adjudicatory.... [They] are
   not seeking simply to maximize the amount of jail time that can be
   extracted from their adversaries, regardless of guilt or innocence;
   rather, they undertake to determine, in response to the defendant's
   arguments, whether the evidence truly demonstrates guilt, and if
   so, what sentence is appropriate.


Lynch, supra note 30, at 2127. There may be, and I think there is, a strong managerial trend in some felony contexts but my point is conceptual--an administrative/inquisitorial operational model does not necessarily erode the adjudicative approach to criminal law.

(35.) Here it is clear how my usage of the term managerial differs from Resnik's usage. Her claim was not that judges have abandoned the enterprise of adjudicating cases and seeking to do substantive justice between the parties, but rather that judges have departed in a problematic fashion from their key role as neutral and detached arbitrators of legal and factual questions by "descending] into the trenches to manage the case." Resnik, supra note 6 at 391.

(36.) Criminal court actors did not necessarily set out to intentionally limit the ability of defendants to invoke individual rights in the criminal process, as is possibly the case in the administration of other legal schemes that have adopted a managerial approach, such as those used to distribute public welfare benefits. See David A. Super, Are Rights Efficient? Challenging the Managerial Critique of Individual Rights, 93 Calif. L. Rev 1051 1060 (2005).

(37.) Herbert L. Packer, Two Models of the Criminal Process, 113 U. Pa. L. Rev. 1, 5-6 (1964) (describing the Crime Control Model and the Due Process Model as abstractions' that "permit[] us to recognize explicitly the value choices that underlie the details of the criminal process").

(38.) Id. at 4-6.

(39.) Id. at 9.

(40.) See id. at 13-14.

(41.) Id. The model's faith in these screening mechanisms can be described as the "presumption of guilt," which for Packer means "[t]he supposition ... that the screening processes operated by police and prosecutors are reliable indicators of probable guilt." Id. at 11.

(42.) Id. at 15.

(43.) It is of course possible that most prosecutors, judges, and, to some extent, even defense attorneys do actually assume that the majority of defendants hauled into misdemeanor courts are factually guilty of the offenses they are accused of committing (or would affirm such a claim if explicitly pressed). And this subjective assessment might be what enables prosecutors and judges to adopt a functional managerial model despite their inculcation into the professional roles of an adversarial system. However, it is also possible that the adjudicative question, "What happened in this case?" has become largely divorced from the practical imperative facing criminal court actors in their daily work--namely, "What should I do with this case?" Actors can decide the latter question without fundamentally settling the former, and they may simply fail to even consider the former question much in the course of their regular activities. Either possibility is consistent with my account of the managerial model because it is a model not of the motivations of criminal actors, but of the recurring patterns of dispositions and the logic by which criminal law is deployed and used to produce those patterns. I thank Gerard Lynch for pushing me on this point, even though he may disagree with my ultimate conclusion.

(44.) For more on broken windows enforcement and its corresponding increase in record-keeping, see Bernard E. Harcourt, Illusion of Order: The False Promise of Broken Windows Policing 100-03 (2001). Harcourt has stressed this aspect of quality-of-life policing, suggesting that insofar as this form of policing has decreased street crime, it is most likely not through the mechanism that the broken windows thesis posits, cementing social norms of order and respect for the law, but rather through the "enhanced power of surveillance offered by a policy of aggressive misdemeanor arrests." Id. at 103.

(45.) See infra Figure 6.

(46.) See K. Babe Howell, Broken Lives from Broken Windows: The Hidden Costs of Aggressive Order-Maintenance Policing, 33 N.Y.U. Rev. L. & Soc. Change 271, 284 (2009) (showing the substantial variation in misdemeanor arrests over the course of the week, with arrests typically peaking on Wednesdays when police staffing for patrols is at its peak and plummeting on Sundays when police staffing is at its weekly low).

(47.) Note that Figures showing arrests count those that took place within the designated year(s). Similarly, Figures showing dispositions count those that took place within the designated year(s), whereas the associated arrests could have taken place any year.

(48.) See Vincent E. Henry, The COMPSTAT Paradigm: Management Accountability in Policing, Business and the Public Sector 4 (3d prtg. 2011); Hans Toch & J. Douglas Grant, Am. Psychological Ass'n, Police as Problem Solvers: How Frontline Workers Can Promote Organizational and Community Change (2d ed. 2005); Shaila K. Dewan, New York's Gospel of Policing by Data Spreads Across U.S., N.Y. Times (Apr. 28, 2004), http://www.nytimes.com/2004/04/28/nyregion/new-york-sgospel-of-policing-by-data-spreads- across-us.html?pagewanted=all&src=pm (describing the use of CompStat by the New York City Police Department and its spread to other police departments).

(49.) Over the first two years of the Giuliani administration and Bratton's tenure as police commissioner, they released a succession of police directives that laid out a series of policing goals and operational plans for achieving those goals. Not surprisingly, the first three dealt with violent crime, youth crime and violence, and the narcotics trade. N.Y.C. Police Dep't, Police Strategy No. 1: Getting Guns off the Streets of New York (1994); N.Y.C. Police Dep't, Police Strategy No. 2: Curbing Youth Violence in the Schools AND ON THE STREETS (1994); N.Y.C. POLICE DEP'T, POLICE STRATEGY NO. 3: DRIVING DRUG Dealers out of New York (1994). The fourth directive dealt with domestic violence. N.Y.C. Police Dep't, Police Strategy No 4: Breaking the Cycle of Domestic Violence (1994) [hereinafter Police Strategy No. 4], The fifth directive presented the new quality-of-life policing regime. N.Y.C. Police Dep't, Police Strategy No. 5: Reclaiming the Public Spaces of New York (1994) [hereinafter Police Strategy No. 5]. The other directives released over these first formative years addressed corruption and the organizational integrity of the New York City Police Department (NYPD), and traffic enforcement. N.Y.C. Police Dep't, Police Strategy No. 7: Rooting Out Corruption; Building Organizational Integrity in the New York Police Department (1995); N.Y.C. Police Dep't, Police Strategy No. 8: Reclaiming the Roadways of New York (1995). These priorities were not separable concerns, but deeply intertwined: the understandings of major crime and minor crime, organizational capacity, management structure, and crime fighting were all part of a wholesale reformist effort linked together by a common vision and diagnosis of underlying problems. It should be noted that intensive use of stop and frisk really expanded in the recent decade and was not a central part of the first wave of tactical reforms.

(50.) Police Strategy No. 5, supra note 49, at 5-10. Police Strategy No. 5: Reclaiming the Public Spaces of New York announced a comprehensive new policing approach in establishing new enforcement priorities for quality-of-life offenses, coupled with extensive operational and tactical reforms. The directive held out abatement of certain quality-of-life issues as an intrinsically beneficial project--to bring New York back to a "society of civility"--and listed the specific conditions the police would target: noise conditions such as loud music, loud clubs and discos, motorcycles, and car alarms; illegal double parking blocking traffic; prostitution; aggressive panhandling; squeegee cleaners; graffiti; illegal peddling and vending; aggressive bicyclists; and public drunkenness. Id; see also id. at 3.

(51.) See id. at 7 ("Police Strategy No. 5: Reclaiming the Public Spaces of New York will emerge as the linchpin of efforts now being undertaken by the New York Police Department to reduce crime and fear in the city.").

(52.) For a discussion of the broken windows hypothesis, see James Q. Wilson & George L. Kelling, Broken Windows, Atlantic Monthly, Mar. 1982, at 29, 31-32. Police Strategy No. 5: Reclaiming the Public Spaces of New York makes explicit reference to Wilson and Kelling's article, describing its "thesis that unaddressed disorder is a sign that no one cares and invites both further disorder and more serious crime." Police Strategy No. 5, supra note 49, at 6.

(53.) Some have argued that the precise tactics used by the NYPD did not conform to the model of broken windows as expressed in Wilson and Kelling's original writing, or that the NYPD always used quality-of-life policing as a pretext to engage more serious crime and was never expressly concerned with quality of life as such. See, e.g., Franklin E. Zimring, The City That Became Safe: New York's Lessons for Urban Crime and Its Control 129-31 (2012); Jeffrey Fagan & Garth Davies, Street Stops and Broken Windows: Terry, Race, and Disorder in New York City, 28 Fordham Urb. L.J. 457, 464-75 (2000). There is evidence, however, that the police meant what they said in numerous public comments and documents, promising both to address certain low-level offenses as public-order issues as a means of improving the experience of living in New York City, and to engage low-level offenses in an effort to reduce serious street crime. See Police Strategy No. 5, supra note 49, at 5-6; see also William Bratton & Peter Knobler, Turnaround: How America's Top Cop Reversed the Crime Epidemic 228 (1998); Rudolph W. Giuliani, Mayor, N.Y.C., Inauguration Speech (Jan. 2, 1994) (transcript available at http://www.nyc.gov/html/ records/rwg/html/96/inaug.html); Rudolph W. Giuliani, Mayor, N.Y.C., State of the City Address (Jan. 11, 1995) (transcript available at http://www.nyc.gov/html/records/rwg/html/ 96/city95.html). There is also significant evidence to show that as time passed misdemeanor arrests and summonses became institutionalized as performance metrics inside the NYPD and incentivized by their quantitative management system, irrespective of the quality-of-life or crime-reducing benefit of the activities. See, e.g., Floyd v. City of N.Y., No. 08 Civ. 1034 (SAS), 2013 WL 4046209, at *13-14 (S.D.N.Y. Aug. 12, 2013); John A. Eterno & Eli B. Silverman, The Crime Numbers Game: Management by Manipulation 195 (2012).

(54.) Note that Figures show the number of arrest events, not unique individuals.

(55.) Jeffrey Fagan et al., Street Stops and Broken Windows Revisited: The Demography and Logic of Proactive Policing in a Safe and Changing City, in Race, Ethnicity, and Policing: New and Essential Readings 329, 337 (Stephen K. Rice & Michael D. White eds., 2010); Amanda Geller & Jeffrey Fagan, Pot as Pretext: Marijuana, Race, and the New Disorder in New York City Street Policing, 7 J. Empirical Legal Stud. 591, 591-93, 620-22 (2010). Fagan et al., supra, argues that geographic concentration of the intensity of quality-of-life enforcement, at least stop-and-frisk activity, is not entirely explained by prior index crime rates.

(56.) Data from Richard Rosenfeld, Curators' Professor, Univ. of Mo.-St. Louis & Robert Fomango, Assistant Professor, Ariz. State Univ. (on file with author); INFOSHARE Online, http://www.infoshare.org/main/public.aspx (last visited Feb. 19, 2014). The only exception is the Midtown South Precinct. It has a high population-adjusted misdemeanor arrest rate because it has a relatively low residential population, but covers the city's transportation hubs and a very dense commercial sector including Times Square, and because it is a tourist center is it highly policed.

(57.) Precisely how these tactics would impact serious street crime was understood on a variety of levels and was indeed the subject of some debate among the NYPD brass. For example, Jack Maple, deputy police commissioner under Bratton and one of the innovators of CompStat, argued:

   [W]e needed to be more selective about who we were arresting on
   quality-of-life infractions. When a team of cops fills up a van
   with arrestees, the booking process can take those cops out of
   service for a whole day in some cities. The public can't afford to
   lose that much police protection for a bunch of first-time
   offenders, so the units enforcing quality-of-life laws must be sent
   where the maps show concentrations of crime or criminals, and the
   rules governing the stops have to be designed to catch the sharks,
   not the dolphins.


Jack Maple & Chris Mitchell, The Crime Fighter: Putting the Bad Guys Out of Business 155-56 (1999).

(58.) See Harcourt, supra note 44, at 103 (arguing that despite proponents' focus on community-level mechanisms such as enforcing shared norms of order, the individual-level mechanisms, such as "the opportunity for checking records, fingerprints, DNA, and other identifying characteristics," were probably more important).

(59.) See N.Y.C. Police Dep't, New York City Police Department Patrol Guide Manual [section][section] 208-209, 214; supra note 49.

(60.) The arrest charges are reviewed by a representative from the district attorney's office and can be changed before arraignment, but the initial police arrest charge is determinative of what type of arrest procedures are instituted.

(61.) N.Y. Crim. Proc. Law [section] 160.10 (McKinney 2013). This statute defines the offenses that are fingerprintable, namely penal-law-defined felonies and classified misdemeanors, and also provides for printing for a limited number of violations, such as loitering for the purposes of prostitution. See id

(62.) Crim. Proc. [section][section] 160.10, 160.20-.40; see also Lisa Lindsay, Criminal Court of the City of New York Annual Report 29 (Justin Barry ed., 2012) (describing the arrest to arraignment procedure and the process of taking fingerprints and transmitting them to court).

(63.) Crim. Proc. [section] 160.10.

(64.) For example, turnstile jumping is a noncriminal violation, for which police officers can issue a summons and which is punishable by a fine of up to twenty-five dollars or ten days in jail. N.Y. Comp. Codes R. & Regs. tit. 21, [section][section] 1050.4, 1050.10 (2013) (establishing turnstile jumping as a violation of the New York City Transit Rules of Conduct). Turnstile jumping is also a violation of New York Penal Law section 165.15, and classified as a Class A misdemeanor. N.Y. Penal Law [section] 160.10 (McKinney 2013). The same conduct could also support an arrest for criminal trespass, id. [section] 140.05, which is a violation, or criminal trespass in the third degree, id. [section] 140.10, which is a Class B misdemeanor.

(65.) See Lindsay, supra note 62, at 29.

(66.) Article 150 of the New York Criminal Procedure Law governs all appearance tickets; what the police (and the public) call summonses are a subset of appearance tickets. See Crim. Proc. [section] 150.10. Desk Appearance Tickets, which are discussed in the next paragraph, are another form of appearance tickets governed by this statute. The NYPD Patrol Guide provides guidelines on the issuance of summonses. N.Y.C. POLICE DEP'T, supra 59, [section] 209.

(67.) There is even some latitude for arrest procedures for certain felonies classified as lower level. Crim. Proc. [section] 140.40.

(68.) See Police Strategy No. 5, supra note 49, at 13. DAT is the colloquial term for the appearance tickets authorized by section 150.10.

(69.) See supra Table 2. Section 150.20 of the New York Criminal Procedure Law authorizes DAT arrests for all arrestable offenses (defined under Crim. Proc. [section] 140.10) except a Class A, B, C, or D felony or a violation of sections 130.25, 130.40, 205.10, 205.17, 205.19, or 215.56 of the New York Penal Law. Crim. Proc. [section] 150.20.

(70.) Lindsay, supra note 62, at 24. There were 73,522 DAT arrests and 283,900 online arrests in New York City in 2012. Id.

(71.) Police Strategy No. 5, supra note 49, at 13.

(72.) Bratton & Knobler, supra note 53, at xv (emphasis added).

(73.) Id. at 36.

(74.) See Police Strategy No. 5, supra note 49, at 13, 36-38, 41, 49-50; see also Bratton & Knobler, supra note 53, at 229.

(75.) Until 2010 the police maintained a database of every person stopped, questioned, or frisked even if no arrest was made. The police actively fought against the legislation that eventually put a stop to that practice. See, e.g., Joseph Ax & Jackie Frank, Lawsuit over NYPD's "Stop and Frisk" Data Can Move Ahead, Reuters (Dec. 20, 2012), http://www.reuters.com/assets/print?aid=USBRE8BJlEV20121220; Al Baker, Lawsuit Challenges Stop-and-Frisk Database, N.Y. Times City Room Blog (May 19, 2010), http://cityroom.blogs.nytimes.com/2010/05/19/lawsuit-challenges-stop-and-frisk-database, Bob Herbert, Op-Ed., Watching Certain People, N.Y. Times (Mar. 2, 2010), http://www.nytimes.com/2010/03/02/opinion/02herbert.html; Clifford Krauss, State Legislators Agree to Restore Arrests for Minor Offenses, N.Y. Times (Nov. 11, 1995), http://www.nytimes.com/1995/ll/ll/nyregion/state-legislators-agree-to-restore-arrests-for minor-offenses.html; Josh Robin, Governor Signs Legislation to Eliminate NYPD's Stop and Frisk Database, NY1 (Jul. 16, 2010), http://www.nyl.com/content/top_stories/122173/ governor-signs-legislation-to-eliminate-nypd-s-stop-and-ffisk-database. Other examples of new record-keeping included the "Domestic Incident Report" (DIR), a form and accompanying database that tracked perpetrator and victim information in suspected domestic violence disputes. See Police Strategy No. 4, supra note 49, at 14-18. This information not only is used by police to determine if a crime has occurred, but also is an important source for the court in setting bail, making offers, and determining punishments.

(76.) See supra Figure 1. Misdemeanor arrests have recently declined for the first time in years. This phenomenon, however, is driven almost exclusively by decreases in marijuana and trespass arrests. One explanation for this decline is the significant amount of public pressure, media attention, and litigation around marijuana arrests, stop-and-frisk tactics, vertical sweeps in public housing, and the Clean Halls program, which collectively produced the majority of the marijuana and trespass arrests. See supra Figure 2.

(77.) Police Strategy No. 5, supra note 49, at 37; see also Lindsay, supra note 62, at 35 (noting that over the past decade the low has been about 510,000 and the high about 649,000).

(78.) See infra Figure 6; see also 2012 Crime Statistics by County, N.Y. St. Division Crim. Just. Services, http://www.criminaljustice.ny.gov/crimnet/ojsa/countycrimestats.htm (last visited Feb. 19,2014).

(79.) Although the absolute number grew precipitously, the percent of individuals arrested each year for misdemeanor crimes with no prior criminal convictions grew by only about ten percentage points, going from about 60% in 1990 to just over 69% in 2012. See supra Figure 7.

(80.) Dev ah Pager, Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration 4-5 (2007) (demonstrating that the "negative credential" of a felony record has significant effects on the labor market prospects of young men of color); see also James B. Jacobs, Mass Incarceration and the Proliferation of Criminal Records, 3 U. St. Thomas L.J. 387, 420 (2006) (describing a criminal record as a "negative curriculum vitae" and discussing the implications of the extensive creation and circulation of these records).

(81.) I primarily focus on their use and meaning inside of the judicial apparatus instead of within the larger social and economic realm; but sometimes I address the latter issue when it is relevant to the types of incentives that defendants face in misdemeanor court.

(82.) The expressive aspect of punishment includes communication of the offender's diminished status to others in the social community. Harold Garfinkel, for example, conceptualized criminal court proceedings as a "status degradation ceremony." Harold Garfinkel, Conditions of Successful Degradation Ceremonies, 61 Am. J. Soc. 420, 420-24 (1956).

(83.) See JAMES Q. WHITMAN, HARSH JUSTICE: CRIMINAL PUNISHMENT AND THE Widening Divide Between America and Europe 117-18 (2003) ("[T]he Napoleonic Criminal Code of 1810.... distinguished] between full-throttle 'crimes' on the one hand, and mere debts and 'contraventions' on the other.... [T]hese distinctions are of central importance ... for a European law that finds ways to convict offenders without stigmatizing them as 'criminals.'").

(84.) Josh Bowers, Punishing the Innocent, 156 U. Pa. L. Rev. 1117, 1122, 1128 (2008) (emphasis omitted).

(85.) Bowers, supra note 22, at 1704-09 (arguing that for many low-level criminal cases--what in New York City criminal court vernacular are called "disposable cases"--the "prosecutors' initial decisions of what and whether to charge are somewhat dispositive on the question of whether the defendant will ultimately end up with some type of conviction--even if some equitable play remains in the punishment joints").

(86.) Note that although the absolute number of misdemeanor cases that prosecutors declined to prosecute over the past five years has consistently been about two to four times higher than the absolute number of declined felony cases, the misdemeanor decline-to-prosecute rate has fluctuated between a level that is 13% and 51% higher than that of the felony decline-to-prosecute rate. See Data from N.Y. State Div. of Criminal Justice Servs. (on file with author) (providing arrest disposition data for 2007-2012); see also Dispositions of Adult Arrests, N.Y. ST. DIVISION CRIM. JUST. SERVICES, http://www.criminaljustice.ny.gov/crimnet/ojsa/dispos/index.htm (last visited Feb. 19, 2014).

(87.) A judge can authorize a marijuana ACD (MJACD) without the prosecutor's consent if the defendant has never been convicted of any crime; if the defendant has a prior criminal conviction the prosecutor must consent to the MJACD. See N.Y. Crim. Proc. Law [section][section] 170.55-.56 (McKinney 2013). The details of the disposition are discussed below.

(88.) See infra Figure 10. Almost every misdemeanor case conviction disposition is the result of a plea: there were only 533 trial verdicts for misdemeanor and violation cases in 2012, as compared with over 220,000 misdemeanor case dispositions and over 500,000 summonses filed. Lindsay, supra note 62, at 35, 51; supra Figure 8; see also Data from N.Y. State Div. of Criminal Justice Servs. (on file with author) (providing misdemeanor disposition data for 1980-2012). The maximum jail sentence for a violation is fifteen days in jail. See N.Y. Penal Law [section] 70.15(4) (McKinney 2013).

(89.) New York case law holds that a person should be arraigned within twenty-four hours of arrest and provides that a defendant arrested without a warrant and held in excess of twenty-four hours prior to arraignment must be released unless the state can provide an adequate explanation for the delay. See People ex rel. Maxian v. Brown, 570 N.E.2d 223, 22325 (N.Y. 1991) (per curiam) (interpreting N.Y. Crim. Proc. Law [section] 140.20(1)).

(90.) Interview with "Nathalie," Supervisor, Early Case Assessment Bureau, in N.Y.C., N.Y. (July 25, 2012). Both felony and misdemeanor arrests are processed in the same ECAB offices and, by and large, under the same time pressures. Id. This Article contains excerpts from many confidential interviews conducted by the author. Because she promised anonymity to her interview subjects in exchange for their candor, the author has assigned her interviewees fictitious names. The Stanford Law Review has not reviewed the author's interview notes or field notes for accuracy or for any other purpose.

(91.) See Crim. Proc. [section][section] 170.55-.56.

(92.) Cases are only adjourned after a shorter period if the defense counsel makes a compelling argument as to why immediate sealing is necessary, such as to avoid severe immigration ramifications.

(93.) See Crim. Proc. [section][section] 160.50, 160.55.

(94.) WebCrims, N.Y. St. Unified Ct. Sys., https://iapps.courts.state.ny.us/ webcrim attorney/Defendant Search (last visited Feb. 19, 2014).

(95.) Sealing will not happen if the court issues a "do not seal" order upon the motion of the prosecution, or its own motion, but the defense is entitled to be heard on such motions. See Crim. Proc. [section] 160.50(1). If there are no conditions attached to the ACD, the defendant is not required to come back to court after accepting it.

(96.) See supra Figure 10. For example, a judge could dismiss a criminal complaint (the charging instrument in most misdemeanor cases) on the ground that it is "defective" (i.e., facially insufficient or for lack of jurisdiction), Crim. Proc. [section][section] 170.30, 170.35, or in the interest of justice, id. [section] 170.40.

(97.) See Crim. Proc. [section] 30.30; see also N.Y. Penal Law [section] 70.15 (McKinney 2013). In order to declare "ready," the prosecution must be able to proceed to trial, which means, inter alia, having a jurisdictionally sufficient accusatory instrument. People v. Colon, 443 N.Y.S.2d 305, 307-08 (Crim. Ct. 1981) (holding that the prosecution cannot be ready for trial if they have not "converted the complaints to jurisdictionally sufficient informations within" the statutory limit), rev'd on other grounds, 450 N.Y.S.2d 136 (App. Term 1982), rev'd, 453 N.E.2d 548 (N.Y. 1983), superseded by statute, Act of Aug. 1, 1984, ch. 670, 1984 N.Y. Laws 3019, as recognized in People v. Bolden, 578 N.Y.S.2d 914 (App. Div 1992).

(98.) See, e.g., People v. Cortes, 604 N.E.2d 71, 75 (N.Y. 1992) (holding that, in determining whether the prosecution has declared readiness within the permissible period, courts should "subtract[] any periods of delay that are excludable under the terms of the statute").

(99.) See Crim. Proc. [section] 160.50. Accordingly, after sealing, the arrest should not appear on any future rap sheets and no employer conducting a background check should be able to see evidence of the arrest.

(100.) Lindsay, supra note 62, at 40.

(101.) See id. at 51; Data from N.Y. State Div. of Criminal Justice Servs. (on file with author) (providing arrest disposition data for 2002-2012).

(102.) See supra Figure 9.

(103.) For discussion of violations and infractions, see Part II above.

(104.) See N.Y. Penal Law [section] 240.20 (McKinney 2013). For example, defendants regularly plead to disorderly conduct if the sole arrest charge was misdemeanor possession of a controlled substance.

(105.) See N.Y. Crim. Proc. Law [section] 170.56 (McKinney 2013). Provided the person has no prior criminal convictions, the DCJS is directed to destroy the fingerprints and de-link the New York State Identification Number upon sealing of these convictions. Id.

(106.) Prosecutors can, and in some boroughs often do, demand a permanent waiver of sealing as a condition of violation pleas. The district attorney (DA) can also make a motion to block sealing in "the interests of justice" to the court within a specified time period after disposition. See Crim. Proc. [section] 160.55(1). In addition, section 160.55 exempts certain violation and infraction convictions from eventual sealing. See id. [section] 160.55. The mandatory court surcharge of $120 is imposed for all violation convictions. Penal [section] 60.35(l)(iii). Quite frequently defendants request that civil judgment be entered on this surcharge, which will be, like any other civil judgment, thereafter reflected on the person's credit report. See Crim. Proc. [section] 420.40.

(107.) Prosecutors, and some judges, regularly search for sealed cases in their own internal files or in court files. This practice still continues despite a Court of Appeals holding in In re Katherine B. v. Cataldo, 833 N.E.2d 698 (N.Y. 2005), that "law enforcement agency" as used in section 160.50(l)(d)(ii) of the New York Criminal Procedure Law does not include a prosecutor--and that therefore judges may not issue ex parte unsealing orders to prosecutors seeking information about sealed cases. Id. at 702-03. Whether or not using information about cases sealed under section 160.55 to form plea and sentence offers violates the sealing law has never been litigated, but this practice is even more common since prosecutors and judges have easy access to that information in the Office of Court Administration database.

(108.) In certain instances a misdemeanor arrest can result in a felony conviction because the prosecutor charges a higher-order offense after reviewing the facts alleged by the police or--more likely--the arrest charge is eligible for a "felony bump up" because of prior convictions, such as in the case of weapons possession. See, e.g., Penal [section] 265.02(1). A felony conviction from a misdemeanor arrest is fairly rare in practice: only about 450 dispositions were convictions for felony offenses when the top arrest charge was a misdemeanor (out of over 226,000 dispositions) in 2012. See Data from N.Y. State Div. of Criminal Justice Servs. (on file with author) (providing arrest disposition data for 2002-2012).

(109.) It does provide for a "certificate of relief from disabilities," subject to various qualifying restrictions, that ostensibly mitigates some of the collateral consequences, but in practice the effects of such certificates are limited. See N.Y. Correct. Law [section][section] 701, 703 (McKinney 2013).

(110.) See, e.g., Barkow, supra note 30, at 871; Lynch, supra note 30, at 2136-40 (arguing that expansive substantive criminal prohibitions and high statutorily authorized sentences give prosecutors wide discretion); Stuntz, supra note 8, at 2569 ("The bodies of law, state and federal, that claim to define crimes and sentences do not really do what they claim. Instead, those bodies of law define a menu--a set of options law enforcers may exercise, or a list of threats prosecutors may use to induce the plea bargains they want. The menu says little about what options are exercised or what threats are used.").

(111.) See, e.g., Michael Lipsky, Street-Level Bureaucracy: Dilemmas of the Individual in Public Services (1980) (discussing how resource limitation restricts not only how front-line workers can carry out assigned tasks but also how they come to define these tasks); D.H. Zimmerman, The Practicalities of Rule Use, in Understanding Everyday Life: Toward the Reconstruction of Sociological Knowledge 223 (Jack Douglas ed., 1970) (arguing that rules inside an organization should not be thought of as having "stable, operational meanings invariant to the exigencies of actual situations of use, and distinct from the practical interests ... of the rule user").

(112.) LIPSKY, supra note 111, at 29.

(113.) Id.

(114.) Data from Chief Clerk of N.Y.C. Criminal Court (on file with author).

(115.) Id

(116.) See N.Y. Crim. Proc. Law [section] 100.10 (McKinney 2013) (defining the charging instruments that can commence an action in a local criminal court). The prosecution's file also contains additional paperwork, such as, among other things, the police arrest report, which often contains more details of the occurrence, the ECAB screening sheet, and statement or notice of intent to offer identification statements required under Crim. Proc. [section] 710.30, and a narcotics fact sheet if it is a drug arrest.

(117.) Technically a prosecutor makes an "offer" only when the proposed plea is below the top charge and otherwise makes a "recommendation" to a sentence within the range of the top charge, as judges are free to sentence to anything within the range of the top charge. I will use the term "offer" to cover any proposed plea or sentence from a prosecutor, just to streamline the language.

(118.) The general description of the misdemeanor arraignment process is based on the author's two years of fieldwork and personal experience working as a defense attorney. Nearly identical descriptions of the misdemeanor arraignment experience from experienced public defenders can be found in Fabricant's and Howell's articles. See Fabricant, supra note 22, at 403-04; Howell, supra note 46, at 294-96.

(119.) See generally Jamie Fellner, The Price of Freedom: Bail and Pretrial Detention of Low Income Nonfelony Defendants in New York City (2010).

(120.) See N.Y.C. Criminal Justice Agency, Annual Report 2011, at 18 (2012), available at http://www.cjareports.org/reports/annual11.pdf.

(121.) For further explanation, see Bowers, supra note 22, at 1709-12 (discussing the distinction between "real" and "disposable" cases).

(122.) See supra Table 2.

(123.) Program is a catchphrase for an assortment of classes, therapeutic interventions, informational sessions, and social services. Examples of programs offered in first-arrest cases include a class for shoplifters lasting several hours and a half-day introduction to drug treatment.

(124.) See supra note 43 and accompanying text.

(125.) Interview with "Ellane," Supervising Assistant Dist. Att'y, Borough^ Dist. Atty's Office, in N.Y.C., N.Y. (June 7, 2012).

(126.) Interview with Judge "Henry," Borough A Criminal Court, in N.Y.C., N.Y. (Mar. 16, 2011).

(127.) See N.Y. Penal Law [section] 165.15(3) (McKinney 2013). Theft of services, in this instance, really meant farebeat.

(128.) See supra Part II.A (explaining a DAT versus an online arraignment).

(129.) Interview with "Zena," Supervising Assistant Dist. Att'y, Borough A Dist. Atty's Office, in N.Y.C., N.Y. (Apr. 13, 2011).

(130.) Interview with "Ray," Supervising Pub. Defender, in N.Y.C., NY (Apr 11 2011).

(131.) Field Notes (June 12, 2011).

(132.) Again, this is typical if it is a first arrest; an ACD for a narcotics possession charge often involves the condition of one day of community service.

(133.) Prosecutors almost invariably reduce the arrest charge to a Class B misdemeanor before trial to ensure a bench trial and not a jury trial. New York City misdemeanor defendants are at a particular disadvantage in seeking to vindicate their innocence in front of a jury because section 340.40(2) of the New York Criminal Procedure Law exempts New York City criminal courts from the statewide rule providing for jury trial rights for all misdemeanors, limiting the right only to Class A misdemeanors. See N.Y. Crim. Proc. Law [section] 340.40(2) (McKinney 2013); see also N.Y. Penal Law [section] 70.15 (McKinney 2013).

(134.) See Lindsay, supra note 62, at 52.

(135.) N.Y.C. Criminal Justice Agency, supra note 120, at 16. The remaining 5% were other types of dispositions. Id.

(136.) See Penal [section] 155.25.

(137.) Misdemeanor defendants with two or more prior prosecuted arrests within the last twelve months and two or more misdemeanor convictions (at least one within the last twelve months) will have their rap sheets marked by the clerk's office before arraignment as qualifying for "Operation Spotlight." See Freda F. Solomon, N.Y.C. Criminal Justice Agency, Operation Spotlight: Year Four Program Report, at i (2007), available at http://www.cjareports.org/reports/spotlight4.pdf.

(138.) This means that absent some insufficiency in the charging document, the judge cannot on her own accord reduce the offense grade for purposes of a plea bargain. Judge "Henry" explained the practical necessity of judges making offers on Operation Spotlight cases as a function of both docket pressures (if the offer is one year, all defendants have an incentive to take everything to trial because they have nothing to lose) and substantive justice (one year in jail is not a fair sentence for petit larceny even if the defendant did it three times this year). Judge "Henry" was a bureau chief ADA prior to becoming a judge and said:

   The old joke is that ADAs become more liberal once they are on the
   bench and the former defense attorneys become more conservative. It
   is not about the person, but rather the situation you face ... in
   Operation Spotlight, when you're an ADA you implement the DA's
   policy of asking for an A misdemeanor and the year in Spotlight
   cases, but when you're judge you undercut that offer.


Interview with Judge "Henry," supra note 126.

(139.) See N.Y. Crim. Proc. Law [section] 170.70 (McKinney 2013) (stating that misdemeanor defendants in custody pending disposition for more than five days (excluding Sunday) without the prosecution having converted the complaint to an information must be released unless good cause shown for delay).

(140.) Interview with Judge "Alfred," Borough A Criminal Court, in N.Y.C., N.Y. (March 16,2011).

(141.) Field Notes (June 7, 2012).

(142.) This courtroom is also called the "Bench Trial Part." It is almost exclusively staffed by judicial hearing officers who are not legally authorized to take pleas to Class A misdemeanors, and who can only conduct bench trials. However, many of the cases are Class A misdemeanors, which implicate a constitutional right to demand a jury trial. Standard practice is to keep the cases as Class A misdemeanors during the pendency of the case in order to gain the benefit of the longer speedy trial time allowed for these charges, but then to reduce the charge to a Class B misdemeanor on the trial date to ensure a bench trial.

(143.) See N.Y.C. Criminal Justice Agency, supra note 120, at 11, 18; N.Y.C. Criminal Justice Agency, Annual Report 2010, at 11, 18 (2011), available at http://www.cjareports.org/reports/annuallO.pdf; N.Y.C Criminal Justice Agency, Annual Report 2009, at 11,18 (2010), available at http://www.cjareports.org/reports/annual09.pdf.

(144.) See Fellner, supra note 119, at 13, 22 (reporting numbers provided to Human Rights Watch from the New York Criminal Justice Agency for 2008); see also N.Y.C. Criminal Justice Agency, supra note 120, at 23-25 (reporting bail statistics citywide and by borough for all continued cases, including felonies).

(145.) See Fellner, supra note 119, at 31-33; Mary T. Phillips, N.Y.C Criminal Justice Agency, Research Brief No. 14, Bail, Detention, & Nonfelony Case Outcomes (2007), available at https://www.ncjrs.gov/App/publications/Abstract.aspx? id=240693.

(146.) Data from N.Y. State Div. of Criminal Justice Servs. (on file with author). Judges typically try to set quicker trial schedules if the defendant is being held in custody, but it is very difficult to move a case to trial in less than 30 days after arrest.

(147.) As Jeannie Suk notes, the criminal law seeks to reorder intimate lives with the order of protection. See generally Jeannie Suk, Criminal Law Comes Home, 116 Yale L.J. 2 (2006). Yet it is only sporadically successful in doing so. In my interviews with domestic violence defendants, the preponderance openly admitted to regular contact with the person named on the order, and a significant number continued to live together with an active full order of protection.

(148.) See supra note 139.

(149.) See Lindsay, supra note 62, at 16. In the busiest boroughs the open misdemeanor/violation dockets in each criminal courthouse can be as large as 11,500 to 13,500 at a time. Id.

(150.) The DA of Borough A has an informal guideline limiting an ADA's caseload in misdemeanor court to 120 cases, but many ADAs recount having up to 200 open cases at a time. ADAs often have higher caseloads than defense attorneys because they do not personally appear in court in most of their open matters; there is usually an ADA assigned to each courtroom that has "notes" or directions from the assigned ADA about how to deal with each case calendared that day.

(151.) N.Y. Penal Law [section] 165.15(3) (McKinney 2013).

(152.) Interview with "Trevon" in N.Y.C., N.Y. (Dec. 17, 2012).

(153.) Interview with "Alice," Pub. Defender, in N.Y.C., N.Y. (Apr. 5, 2011).

(154.) Id.

(155.) Field Notes (Sept. 30, 2010).

(156.) Interview with "Al," Supervising Assistant Dist. Att'y, in N.Y.C., N Y (Feb 16 2011).

(157.) Interview with "Carol," Supervising Assistant Dist. Att'y, in N.Y.C., N.Y. (June 7, 2012).

(158.) Lindsay, supra note 62, at 35; Eric T. Schneiderman, N.Y. State Office of the Att'y Gen., A Report on Arrests Arising from the New York City Police Department's Stop-and-Frisk Practices 5 (2013), available at http://www.ag.ny.gov/ pdfs/OAG_REPORT_ON_SQF_PRACTICES_NOV_2013.pdf; supra Figure 1.

(159.) Interview with "Leslie," Supervising Pub. Defender, in N.Y.C, N Y (Apr 5 2012).'

(160.) Data were provided by the DCJS to the author in the form of micro-level arrest incidents and de-identified individual ID numbers. The analysis, opinions, findings, and conclusions expressed in this Article are those of the author alone and not those of the DCJS. Neither New York State nor the DCJS assumes liability for its contents or use thereof.

(161.) The methodological upshot of the sealing laws discussed in Part III is that we cannot trace the criminal justice paths of all people who enter the misdemeanor justice system unless there is a disposition that allows the state to maintain fingerprints. Cf. N.Y. CRIM. Proc. Law [section][section] 160.50, 160.55 (McKinney 2013).

(162.) This is the only dismissal group that can be reliably tracked over time. See supra Part III; cf. Crim. Proc. [section] 160.50. For a fuller discussion of the cohort, see Issa Kohler_Hausmann, Managerial Justice and Mass Misdemeanors: Statistical Appendix, Stan. L. Rev. (Mar. 2014), http://www.stanfordlawreview.org/sites/default/files/KohlerHausmann_66_Stan_L_Rev_Statistical_Appendix.pdf [hereinafter Statistical Appendix],

(163.) See Statistical Appendix, supra note 162.

(164.) "On-par conviction" means a misdemeanor conviction for the misdemeanor arrest models and a felony conviction for the felony arrest models. See id.

(165.) See id.

(166.) As in any nonexperimental study, omitted variables (e.g., quantum of evidence, legal circumstances of stop, and unobservable personal characteristics) may bias the value of the coefficients if those variables are correlated with both an included predictor variable and the outcome variable.

(167.) Direct causal effects of the mark include, for example, the phenomenon of prosecutors deciding to make nonconviction plea offers in cases where defendants have no prior conviction, but declining to do so in otherwise identical cases where defendants have a prior conviction.

(168.) See N.Y. Crim. Proc. Law [section][section] 170.56, 210.46 (McKinney 2013).

(169.) The New York Criminal Procedural Law allows each person to receive only one MJACD and therefore permits the state's criminal justice record-keeping agency to maintain fingerprints linked to a unique identifying number even if the person has no prior criminal conviction. It is not permitted to do so if the case terminates in other forms of dismissal. See id [section] 170.56. I am therefore able to track all arrest events, including sealed records, for this group of people whose arrest did not lead to a conviction.

(170.) Marijuana offenses, N.Y. Penal Law [section] 221.10 (McKinney 2013), were the largest arrest category in New York City over the time period covered by the data (constituting between 17% and 23% of all misdemeanor arrests between the years of 2003 and 2011). See supra Figure 2. The MJACD is the most common outcome from a marijuana arrest. In the arrest years of my cohort, 2003 and 2004, 58% and 66%, respectively, of all arrests where the top charge was a marijuana offense resulted in MJACDs.

(171.) Note the MJACD cohort was "at risk" for rearrest for slightly longer as their cohort-entering date was, on average, earlier than the cohort-entering date of the misdemeanor conviction cohort.

(172.) Cross tabs of the number of misdemeanor and felony arrests following the initiating event are available upon request from the author.

(173.) The definition of "violent felony offense" tracks sections 70.02 and 70.04 of the New York Penal Law, which set forth the offenses defined as violent felony offenses for the purposes of sentencing and statutorily required record-keeping by the DCJS. Issa Kohler-Hausmann & Jamie Fellner, Human Rights Watch, A Red Herring: Marijuana Arrestees Do Not Become Violent Felons 32 (2012), available at http://www.hrw.org/ sites/default/files/reports/us_mjll 12webwcover.pdf. The variable used here tracks that statutory definition and adds what the DCJS codes as "VFO-like class A-1 offenses not in VFO." Id. (internal quotation marks omitted).

(174.) Cross tabs of arrest and disposition types with cell and row percentages available upon request from author.

(175.) I chose to evaluate the model for defendant and case characteristics that were common in the dataset; the relevant patterns are the same over a range of other tested values of independent variables (such as at the mean of all variables).

(176.) Predicted probability evaluated at same demographic and age values as the above Figures, and assuming zero prior misdemeanor convictions for the misdemeanor models and zero prior felony convictions for the felony models.

(177.) William James, Pragmatism: A New Name for Some Old Ways of Thinking 92 (Hackett Publ'g Co. 1981) (1907).

(178.) Bowers, supra note 84, at 1132.

(179.) Stephanos Bibas, Exacerbating Injustice, 157 U. Pa. L. Rev. PENNumbra 53, 54 (2008), http://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=1259&context= faculty_scholarship.

(180.) Id. at 57.

(181.) Natapoff, Aggregation and Urban Misdemeanors, supra note 22, at 146-47.

(182.) Bowers has suggested that type 1 errors, or what he calls "the innocence problem in plea bargaining," are almost exclusively present in this latter set of cases: low-level arrest charges for recidivists. See Bowers, supra note 84, at 1121.

(183.) Feeley, supra note 15, at 241.

(184.) While these reforms are inspired by my research in New York City, they highlight practices widespread in other major urban jurisdictions that produce the same dynamics.

(185.) In New York State, the combined fees for a noncriminal violation conviction exceed $120; for a misdemeanor conviction they are $200. N.Y. Penal Law [section] 60.35 (McKinney 2013). Sociologists have documented how the expansion of legal financial obligations--monetary sanctions in the form of fees, fines, and surcharges--"contributes to the accumulation of disadvantage" in various ways, including "by reducing family income; by limiting access to opportunities and resources such as housing, credit, transportation, and employment; and by increasing the likelihood of ongoing criminal justice involvement." Alexes Harris et al., Drawing Blood from Stones: Legal Debt and Social Inequality in the Contemporary United States, 115 Am. J. Soc. 1753, 1756 (2010); see also Alexes Harris et al., Courtesy Stigma and Monetary Sanctions: Toward a Socio-Cultural Theory of Punishment, 76 Am. Soc. Rev. 234, 237 (2011).

(186.) Many defendants request civil judgment to be entered on these fees, see N.Y. Crim. Proc. Law [section] 420.40(5) (McKinney 2013), which in turn seriously injures their credit rating for years. The collateral consequences of civil judgment on these fees are massive for indigent defendants because credit scores are now so instrumental in securing many essential services--including housing, phone service, and bank accounts and opportunities including employment and education loans.

(187.) One of the most significant factors that judges consider in setting bail is whether the defendant has prior bench warrants and how many. In New York State, that factor is listed in the bail statute. See id [section] 510.30(2)(a)(vi). The number of prior bench warrants was also one of the most common factors listed by judges in my interviews and in other research on what influences prosecutors recommendations and judges' decisions. See Fellner, supra note 119, at 42. In addition, the New York Criminal Justice Agency--the entity that interviews defendants before arraignment and makes release recommendations to the court--assigns significant weight to prior bench warrants. Research has shown that, among the defendants who fail to appear at a required court appearance (about 15% of misdemeanor defendants in 2011), about one-half of these defendants voluntarily return to court within thirty days of the bench warrant. N.Y.C. Criminal Justice Agency, supra note 120, at 27-28.

(188.) See Robert C. Boruchowitz et al., Minor Crimes, Massive Waste: The Terrible Toll of America's Broken Misdemeanor Courts 14 (2009) (noting the many jurisdictions where lack of funding for indigent defense organizations makes it structurally impossible for attorneys to provide effective assistance); Adam M. Gershowitz & Laura R. Killinger, The State (Never) Rests: How Excessive Prosecutor Caseloads Harm Criminal Defendants, 105 Nw. U. L. Rev. 261, 264 (2011) (arguing that prosecutors' caseloads contribute to both prosecutorial error and injustice to defendants).

(189.) Emile Durkheim, The Division of Labour in Society 29 (W.D. Halls trans., MacMillan Publishers Ltd. 1984) (1893).

(190.) Durkheim states that restitutory laws

   do not reflect any very acute feelings, nor even in most cases any
   kind of emotional state. For, since they determine the manner in
   which the different functions should work together in the various
   combinations of circumstances that may arise, the objects to which
   they relate are not ever-present in the consciousness.


Id. at 82.

(191.) David Garland, Punishment and Modern Society: A Study in Social Theory 56 (1990).

(192.) Durkheim, supra note 189, at 40.

(193.) For a discussion of the problematic issues brought about by a fundamentally administrative criminal justice system operating inside the legal and organizational framework set up for an adversarial system, and a reform proposal to address those issues in federal prosecutors' offices, see Barkow, supra note 30, at 876-84, 895-906.

(194.) See Kohler-Hausmann & Fellner, supra note 173, at 12.

(195.) This issue has been studied and debated in the extensive literature on the war on drugs. See, e.g., Alexander, supra note 1, at 98 & n.22, 99 & nn.23-25, 100. There is an extensive criminology literature debating the extent to which racial disparities in drag arrests can be accounted for by relevant legal variables or police deployment decisions guided by citizen complaints or crime incidents. Compare, e.g., Katherine Beckett et al, Race, Drugs, and Policing: Understanding Disparities in Drug Delivery Arrests, 44 CRIMINOLOGY 105, 105 (2006) (concluding that "race shapes perceptions of who and what constitutes Seattle's drag problem, as well as the organizational response to that problem" (italics omitted)), with Stephen D. Mastrofski, Race, Policing, and Equity, 11 Criminology & Pub. Pol'y 593, 594-95 (2012) (finding that citizen calls for service regarding drag-related activity and reported crimes account for a significant portion of disparities in drag arrests via the mechanism of police deployment).

(196.) See, e.g., Alice Goffman, On the Run: Wanted Men in a Philadelphia Ghetto, 74 Am. Soc. Rev. 339, 348-49 (2009) (describing how members of highly policed communities come to rely on the police as a tool in interpersonal relationships); Suk, supra note 147, at 69 (noting how the practices surrounding arrests, prosecution, and orders of protection issued for misdemeanor domestic violence cases "reflect a view of using criminal law to control space and family arrangements").

(197.) For example it is likely that turnstile jumping, one of the larger arrests categories, is more common among people in poverty.

(198.) Loic Wacquant, Punishing the Poor: The Neoliberal Government of Social Insecurity 58, 61 (2009).

(199.) See, e.g., David F. Greenberg, Studying New York City's Crime Decline: Methodological Issues, 31 Just. Q. 154, 182; Bernard E. Harcourt & Jens Ludwig, Broken Windows: New Evidence from New York City and a Five-City Social Experiment, 73 U. Chi. L. Rev. 271, 314-16 (2006); Steven F. Messner et al., Policing, Drugs, and the Homicide Decline in New York City in the 1990s, 45 Criminology 385, 404-07 (2007); Richard Rosenfeld et al., The Impact of Order-Maintenance Policing on New York City Homicide and Robbery Rates: 1988-2001, 45 Criminology 355, 377-79 (2007).

(200.) Julilly Kohler-Hausmann, Tangled Webs: A Case for Intertwining U.S. Carceral and Welfare History 28 (2013) (unpublished manuscript) (on file with author) (quoting E. Clay Shaw, Jr., Representative from Florida and coauthor of the Personal Responsibility and Work Opportunity Act, on the day the House passed this welfare reform bill in 1996); see also Michele Landis Dauber, The Sympathetic State: Disaster Relief and the Origins of the American Welfare State 16 (2013). The proper role of government in addressing social needs has been debated since the founding of the Republic. As Michele Landis Dauber shows, the rhetoric surrounding our public understandings about the proper role of government reflects the outcome of a lot of political work, and not the upshot of natural and obvious distinctions between, say, natural and man-made disaster or deserving and undeserving victims:

   [W]hether a particular set of circumstances mandates rejection or
   relief has always been, over the last two centuries, a question of
   moral and political judgment, not an exact science. From its
   earliest days the history of disaster relief--whether relief was
   ultimately dispensed or not--has been one of claims making and
   argument.


Id.

Table 1
Adjudicative vs. Managerial Models of Criminal Law Administration

                       Adjudicative          Managerial

What is criminal       Enforcing             Providing an
law good for?          substantive           opportunity to
                       proscriptions         identify rule
                       against bad conduct   breakers

What is the role       Selecting the right   Sorting on
of the criminal        people for            governability and
process?               punishment            testing people over
                                             time to establish
                                             their propensity to
                                             follow rules

What is the primary    Plea bargaining or    Plea bargaining
mode of processing     adversarial
cases?                 proceedings

Table 2
Misdemeanor Arrests by Offense--2012

Top Arrest Charge            Arrests    % of Total

PL 221 Marijuana              45,574        19.2%
PL 165 Other Theft *          36,925        15.6%
PL 120 Assault                35,068        14.8%
PL 220 Controlled
   Substances                 25,224        10.6%
PL 155 Larceny                24,679        10.4%
PL 140 Trespass/Burglary      13,337         5.6%
PL 265 Weapons                 8,477         3.6%
VTL 1192 Driving While
  Intoxicated                  7,712         3.3%
PL 145 Mischief                7,444         3.1%
PL 240 Public Order            6,357         2.7%
PL 205 Escape                  5,715         2.4%
PL 230 Prostitution            3,462         1.5%
PL 130 Sex Offenses            1,063         0.4%
Other
Total Misdemeanor Arrests     15,820         6.7%
                             236,857       100.0%

Source: New York State Division of Criminal Justice Services.

* Other theft includes theft of services
(e.g., turnstile jumping).

Table 3

Dismissal         Marking Period on        Court Records Seated
                  Rap Sheet                After the Marking
                                           Period

Decline to        None, case seals as      Yes
prosecute         soon as the DCJS
                  receives order from
                  court; arrest should
                  not print on rap
                  sheet.

30.30 dismissal   Case remains open        Yes
                  and visible on rap
                  sheet and in public
                  records for as long
                  as it is open
                  matter; it seals
                  only after dismissal
                  order.

Other dismissal   Case remains open        Yes
                  and visible on rap
                  sheet and in public
                  records for as long
                  as it is open
                  matter; it seals
                  only after dismissal
                  order.

ACD Dismissial

Marijuana ACD     Case remains open        Yes, unless
                  and visible on rap       defendant is
                  sheet and in public      rearrested during
                  records during a         the adjournment or
                  one-year                 violates conditions
                  adjournment;             imposed by the
                  fingerprints are not     court; DA can
                  destroyed even if        restore the case to
                  there is no prior        the calendar or,
                  conviction.              more likely in the
                                           case of a new
                                           arrest, make a less
                                           lenient offer on the
                                           new arrest.

ACD               Case remains open        Same as above.
                  and visible on rap
                  sheet and in public
                  records during a
                  six-month
                  adjournment
                  (one-year
                  adjournment for
                  family offenses);
                  fingerprints are
                  destroyed if there
                  is no prior
                  conviction.

Conviction

Violation         One year if there is     No, unless defendant
                  a "conditional           violates the
                  discharge" sentence;     condition of the
                  or immediate sealing     conditional
                  if it is a "time         discharge sentence
                  served" sentence         or unless a judge
                  once surcharge is        enters a "do not
                  paid (or civil           seal" order. Select
                  judgment entered).       violations do not
                                           seal. See supra note
                                           106.

Misdemeanor       Permanent record.        No, never sealed.
or felony

Figure 4
Racial and Ethnic Composition of Misdemeanor Arrest Events--2012

Other 9,690        4%
Unknown 2,326      1%
White 31,475      13%
Black 113,869     48%
Hispanic 79,497   34%

Source: New York State Division of Criminal Justice Services.

Note: Table made from pie chart.

Figure 10
Misdemeanor Case Dispositions

Acquitted                    212-0.1%

Other *                      1,913-0.8%

Convicted Misdemeanor        44,278-19,6%

Convicted Noncriminal
Violation/Infraction         65,057-28.7%

Convicted-Felony             454-0.2%

Convicted-Sentence Pending   1,943-0.9%

Declined to Prosecute        17,304-7,6%

Dismissed-ACD                67,689-29.9%

Dismissed-Other              27.634-12.2%

Source: New York State Division of Criminal Justice Services.

* Other dispositions include youthful offender adjudication,
diverted and dismissed, covered by another case, and other
unspecified dispositions.

Note: Table made from pie chart.

Figure 11
Arrest and Conviction Outcomes by Cohort

                          MJACD cohort     Misdemeanor conviction
                                                   cohort

One or more later             70%                   60%
Criminal arrests

Only misdemeanor              37%                   23%
arrests

One or more                   33%                   37%
felony arrests

One or more later             21%                   41%
criminal convictions


Source: New York State Division of Criminal Justice Services.

* Cohort members with one or more felony arrests may also have
had one or more misdemeanor or violation arrests.

Note: Table made from bar graph.

Figure 12
Conviction Results of Cohort Members with One or More
Subsequent Arrests

                         MJACD cohort     Misdemeanor conviction
                                                 cohort

One or more later            29%                 69%
criminal convictions

Only misdemeanor             14%                 45%
convictions

One or more                  15%                 24%
felony convictions *

Source: New York State Division of Criminal Justice Services.

* Cohort members with one or more felony arrests may also have had
one or more misdemeanor or violation arrests.

Note: Table made from bar graph.

Figure 13
Violent Felony Arrest and Conviction Rates of Cohort Members

                    MJACD cohort     Misdemeanor conviction
                                             cohort

No VFO arrests          85%                    86%

One or more
VFO arrests             15%                    14%

One or more
VFO convictions         4%                     4%

Source: New York State Division of Criminal Justice Services.

Note: Table made from bar graph.
COPYRIGHT 2014 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:IV. A Qualitative Account of Managerial Misdemeanor Justice: The Disposition Process through Conclusion, with footnotes, p. 653-693
Author:Kohler-Hausmann, Issa
Publication:Stanford Law Review
Date:Mar 1, 2014
Words:26663
Previous Article:Managerial justice and mass misdemeanors.
Next Article:The Exergen and Therasense effects.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters