3-5 PAGES APA STYLE AI & PLAGARISM FREE

Texas A&M University School of Law Texas A&M University School of Law

Texas A&M Law Scholarship Texas A&M Law Scholarship

Faculty Scholarship

10-2023

The Ever-Shifting Ground of Pretrial Detention Reform The Ever-Shifting Ground of Pretrial Detention Reform

Jenny E. Carroll Texas A&M University School of Law, [email protected]

Follow this and additional works at: https://scholarship.law.tamu.edu/facscholar

Part of the Criminal Law Commons, Criminal Procedure Commons, and the Law Enforcement and

Corrections Commons

Recommended Citation Recommended Citation Jenny E. Carroll, The Ever-Shifting Ground of Pretrial Detention Reform, 19 Ann. Rev. L. & Soc. Sci. 75 (2023). Available at: https://scholarship.law.tamu.edu/facscholar/2053

This Article is brought to you for free and open access by Texas A&M Law Scholarship. It has been accepted for inclusion in Faculty Scholarship by an authorized administrator of Texas A&M Law Scholarship. For more information, please contact [email protected].

Annual Review of Law and Social Science

The Ever-Shifting Ground of Pretrial Detention Reform Jenny E. Carroll School of Law, University of Alabama, Tuscaloosa, Alabama, USA; email: [email protected]

Annu. Rev. Law Soc. Sci. 2023. 19:75–91

First published as a Review in Advance on July 12, 2023

The Annual Review of Law and Social Science is online at lawsocsci.annualreviews.org

https://doi.org/10.1146/annurev-lawsocsci-120122- 092655

Copyright © 2023 by the author(s). This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See credit lines of images or other third-party material in this article for license information.

Keywords

bail reform, pretrial detention, criminal law

Abstract

In the past six decades, pretrial detention systems have undergone waves of reform. Despite these efforts, pretrial jail populations across the country continue to swell. The causes of such growth in jail populations are diffi- cult to pinpoint, but some are more readily apparent: Fear over rising crime rates, judicial reluctance to release accused persons, and monetary burdens associated with release have all contributed to increased detention pretrial across criminal legal systems in the United States. This article examines var- ious pretrial detention reform efforts and highlights the need for greater research in the area.

75

1. INTRODUCTION

The past six decades have brought waves of pretrial detention reform in the United States (Baradaran Baughman 2017). Colloquially, these are often referred to as bail reform, though the changes have involved more than bail alone. Bail is the monetary condition an accused person must comply with to avoid detention while their criminal case is pending. A bail amount set by a judge or other state actors could be imagined as a sort of deposit on liberty. A person posts an amount of money, the bail, and risks forfeiting that money if they fail to appear at future court proceedings or to comply with other conditions of release. Although many reform efforts have focused on reducing or eliminating monetary bail, such a focus is not exclusive. Pretrial deten- tion does not always pivot on the ability to post bail, and accused persons are often released on conditions beyond bail (Carroll 2021). Nonmonetary conditions of pretrial release may include requirements to remain in school or employed, random drug testing or electronic monitoring, compliance with no contact orders, or enrollment in counseling programs. Recent reform efforts have also attempted to address such nonmonetary conditions. As a result, this article considers not only bail reform but, more broadly, movements to reform pretrial detention mechanisms.

To those working within criminal legal systems, the urgency of such reform is plain. Despite relatively steady declines in arrest and crime rates during this time, the nation’s jail populations have increased (Carson 2022). Of the half-million people currently in jail, scholars estimate as many as 60% are pretrial detainees (Garrett 2022). And, much like in criminal legal systems, the burden of pretrial detention falls more heavily on marginalized communities, disproportionately holding people who are Black, Latinx, or poor; have behavioral health needs; and often have some combination of these identities. Unlike detention that is a result of a sentence, pretrial detention comes before rather than after assessment of guilt. This is significant not only for the obvious rea- son that those held pretrial are not convicted persons but also for the less immediately apparent reason that they have not yet benefitted from the process associated with assessments of guilt. As discussed in Section 2, compared to trial processes, the process that accompanies pretrial deten- tion decisions is truncated at best and illusory at worst. Pretrial detention decisions are often made quickly, with limited opportunity or at times incentive for advocacy, and without extensive consid- eration of the impact of the conditions or detention on the individual or their community. All this suggests that more work is left to be done to reform pretrial detention systems. Yet even in the face of needed and urgent change to these systems, identifying the success of reform movements that have already occurred is a more complex proposition. Indeed, few studies have examined the quantitative and qualitative outcomes of reform movements around pretrial detention systems.

This article provides an overview of such movements. Like most overviews, it highlights select efforts, particularly focusing onmodern pretrial reform efforts and considering what remains to be done. Other scholars, including myself, have turned their attention to specific aspects of reforms and have imagined future waves of change in pretrial detention systems (Baradaran Baughman 2017, 2020; Carroll 2020a,b,c, 2021; Garrett 2022). It is clear in looking at the history of pretrial detention reform that much work remains and that scholars play a critical role in working with reform activists to identify and document pretrial procedures and imagine possible revisions. It is my hope that this article will serve as a catalyst for further work in this area.

With that goal in mind, this article proceeds in four sections. Section 2 describes what I have termed government-based reform efforts. These are reforms implemented by government rather than citizen actors. Such efforts include statutory, law enforcement and prosecutorial, and judi- cial reforms. Section 3 examines community-based reform movements. Such community-based movements are often entwined with government-based reform efforts, though they deserve sepa- rate attention because they can also behave independently to push less formal changes in systems.

76 Carroll

These include court watching and bail funds movements, as well as efforts to recenter pretrial safety considerations in terms of the impact on the community of detaining the accused person as opposed to just the impact of releasing the accused person. Section 4 considers the current rev- olution in machine-generated assessments of pretrial risk and the criticisms of such assessment systems. These three sections, by exploring past and ongoing reform efforts, lay the necessary groundwork for the normative assessment of current pretrial detention reform in Section 5. This last section builds on the critiques of existing reform movements to identify developing and/or undertheorized aspects of pretrial detention systems in the hopes of supporting future works.This includes consideration of litigation, government, and community-based reform efforts.

2. GOVERNMENT-BASED REFORM

Historically, pretrial reform movements have come in waves. Each of these waves has varied in form, design, and effect. And, like much in legal reform, government-based reform in the pretrial arena was and is responsive to community-led efforts. The 1960s and 1970s saw a community- led push for uniformity in pretrial detention decision making and targeted analysis of the risk accused persons posed. These movements in turn spawned statutory reform, in the form of early uniform bail statutes [including the federal Bail Reform Act (BRA)], and functional changes, in the form of efforts to calculate risk posed by individual defendants through targeted analysis of set factors and identities (Duker 1977). Both sought to reduce disparities in decision making by defining when bail (as opposed to detention) was appropriate. They also considered how risk assessments were conducted to ensure that pretrial conditions—monetary or otherwise—or de- tention decisions promoted the state’s twin goals of ensuring future court attendance and safety of the community at large while the criminal case was pending (Gouldin 2018). In time, bail statutes and methods of calculating the risks that accused individuals posed pretrial evolved, and so too did reform movements.

From the creation of the BRAs at the federal level to the election of progressive prosecu- tors at the local level, government-based changes around pretrial detention have been driven by community-led movements that have sought to call attention to failures and inequities in pretrial detention systems.Other government-based changes are the product of litigation efforts on behalf of detained individuals. Despite their responsive nature, these government-based reforms are also distinct from community-led movements, as they often create formalistic and systematic changes. This section explores examples of such reforms.

2.1. Legislative Reform

The Federal Bail Reform Act of 1966 (hereafter 1966 BRA) (18 U.S.C. §3141) was the product of the first wave of pretrial detention reform. The Vera Institute was central to sparking this first wave of reform in the 1960s. Through the Manhattan Bail Project, researchers at the Vera Institute demonstrated that limited and targeted questions could accurately predict an individual’s probability of success on pretrial release (Katzive 1968). Using a questionnaire, the Project relied on the existence of identified factors to inform pretrial detention decision making (Ares et al. 1963, Botein 1965). The use of a uniform questionnaire on individuals prior to pretrial detention hearings sought to eliminate inconsistent pretrial detention decisions, but it was also efficient, as judges were informed of the assessment results at the time of the pretrial detention hearing. The Project served as an early prototype for pretrial risk assessment that was both individualized and efficient.

At its core, the Bail Project criticized both the arbitrary nature of bail decisions and the harm of pretrial detention on poor populations. It urged a refocus on alternatives to detention and

www.annualreviews.org • Pretrial Detention Reform 77

individualized assessments of risk of flight pretrial (Ares et al. 1963). There was broad consensus that existing systems detained people because they were poor rather than because they posed a true risk (Foote 1956; Kennedy 1963, 1964).

Passed in response to these concerns, the 1966 BRA created a presumption of release, on either personal recognizance or unsecured bond, andwas focused onmitigating flight risk.To accomplish this goal, the Act urged judicial decisionmakers to impose conditions of release designed to ensure that the accused would return for future court appearances. President Johnson noted that the 1966 BRA granted judges discretion to detain “dangerous people” while creating a “flexible set of conditions” that would allow others to be released ( Johnson 1966).

As Garrett (2022) notes, in the second wave of pretrial detention reform, legislative efforts quickly shifted with the passage of the 1966 BRA away from concerns about flight and toward concerns about dangerousness.The 1984 BRA reflected the shift that was already occurring across jurisdictions. It permitted defendants to be held without bond upon a showing by clear and con- vincing evidence that they posed a danger or were a flight risk [Federal Bail Reform Act of 1984, 18 U.S.C. § 3142(e)]. In United States v. Salerno (1987), the US Supreme Court upheld the provi- sion, holding that the BRA’s no-bail policy did not violate the Eighth Amendment’s prohibition on excessive bail or the Fifth Amendment’s due process requirements [United States v. Salerno (1987)]. The court held that the government’s interest in safety was significant, and if a judge determined that no pretrial conditions could assure it, then a defendant could be held preventively pretrial.

Across the nation, states had largely already followed suit (or did soon after the decision in Salerno), creating a multitiered statutory approach to pretrial detention (Garrett 2022). First, the risk the accused person posed was assessed. Second, conditions of release—monetary or otherwise—were set, or a defendant was detained because they presented too grave a risk to allow pretrial liberty. This approach served as a precursor to today’s assessment mechanisms and the challenges they present.

The assessment of risk contemplated after the 1984 BRA was, in theory, based on objective data designed to assess risk of flight and dangerousness, as opposed to poverty or other aspects of identity (Gouldin 2018). But the process was fundamentally flawed, and as Section 4 notes, these flaws were carried forward in future assessment techniques (Mayson 2019). Risk assessments were often based on criminal history, arrest records, and the current basis of the arrest—all of which, as discussed in Section 4, carry their own embedded biases (Gouldin 2018, Mayson 2019).

Relying on these limited data, pretrial screeners and judges assessed the flight risk an individual posed and their risk for future dangerousness if released (Gouldin 2018, Mayson 2018). Those deemed too dangerous were held. For those who posed only some flight risk or danger, judges set conditions of release that often included monetary bail. The difficulty is that setting conditions for those who posed only some risk often resulted in their detention when they were unable to post bail and/or comply with nonmonetary conditions of release. Their resulting detention was a product of their poverty or marginalized existence, not the risk they posed. In this, the second wave of reform replicated the very concerns the first wave had sought to correct and thus drove the third wave of reform (Mayson 2018).

This third wave of legislative reform did not jettison the risk calculation that characterized its predecessors; it did, however, attempt to create a more nuanced calculus of that risk (Gouldin 2018, Mayson 2018). The creation and adoption of the risk assessment tools that characterize this wave of reform are discussed in more detail in Section 4, but other significant aspects of this reform movement are worth noting here.

First, in the past 15 years, states have adopted not just risk assessment tools but also a vari- ety of legislative reforms that have altered the type of bail that may be imposed, including the elimination of cash bail either for designated offenses or altogether [725 Ill. Comp. Stat. Ann.

78 Carroll

5/110-6-6.1 2023; Ariz. R. 7.3(b)(2)(A)-(B) (2017); In re Kenneth Humphrey (2021)], or the stan- dard for pretrial detention hearings [725 Ill. Comp. Stat. Ann. 5/110-6-6.1 (2023); Ala. Constit. §16 (2022); N.M.R.A. 5-409 2017]. By and large, such state-based reform efforts reflect goals to reduce pretrial detention and the use of cash bail systems to govern release. This path to reform has not, however, been linear.

Efforts to eliminate cash bail in California have wavered (Thompson 2022). In addition, there remain few data about whether jurisdictions eliminating cash bail are more likely to impose non- monetary conditions of release that can carry costs themselves and can be criminogenic in that released individuals may be subject to new charges for violating nonmonetary conditions of re- lease (Carroll 2021).The elimination of cash bail altogether or for particular categories of offenses also does not alleviate due process and equal protection concerns (Carroll 2020a, 2021). First, the imposition of nonmonetary conditions of release or a decision to detain still requires some mod- icum of due process, given its impact on a defendant’s liberty (Carroll 2020a, 2021). The precise articulation of what process is constitutionally required remains elusive, with the Supreme Court offering only the guidance that the condition imposed must advance the state’s articulated inter- ests [Stack v. Boyle (1951); Carroll 2020a]. Second, decisions that affect pretrial detention, even if cash bail has been eliminated, may disproportionately impact the same marginalized populations as bail (Baradaran Baughman 2017, Carroll 2021, Heaton et al. 2017, Wiseman 2018). Investiga- tory decisions, charging decisions, and the imposition of nonmonetary conditions may all suffer the same bias (Mayson 2018, 2019) and have the same outsized impact on particular populations as earlier, now-banned bail decisions (Heaton et al. 2017).

Importantly, in considering the hazards of these reform efforts, some jurisdictions have at- tempted to curtail the disproportionate impact of bail in particular by requiring more process. This includes a right of representation, discovery, and individualized judicial rulings stating find- ings and the basis for the imposition of particular conditions in bail hearings. Illinois, for example, has adopted this model [725 Ill. Comp. Stat. Ann. 185 (2021); N.M.Constit. Art. 2, Sec. 13 (2021); In re Kenneth Humphrey (2021)]. The Supreme Court in Salerno (1987) suggested that a robust ad- versarial hearing in which the government bore the burden of establishing the need for detention quieted constitutional concerns. Most states, however, do not impose procedural requirements including a right to written factual findings or a right to appeal (Garrett 2022). However, even in jurisdictions that impose procedural requirements statutorily, few data exist to suggest that such robust hearings in fact occur. One study that attempted to track bail hearing procedures in Georgia after bail reform found inconsistencies within and among jurisdictions and barriers to data collection (Woods et al. 2020). Simply put, the existence of 51 jurisdictions (50 states and the federal system), with individualized and often idiosyncratic state actors within each system, renders pretrial detention proceedings irregular at best, obscure at worst, and often both: Each judge in a particular systemmay behave singularly with regard to those who appear in their courts, resulting in inconsistent conditions imposed across defendants in a particular jurisdiction.

Consider my own efforts to track pretrial detention hearings in Alabama prior to the COVID-19 pandemic (Carroll 2021), which revealed similar barriers and inconsistencies to those reported in the Georgia study (Woods et al. 2020). Although pretrial detention hearings were often short (less than a minute—48 seconds on average from my observations) and perfunctory, some judges appeared more willing to consider defense recommendations, whereas others ar- rived at the hearing with the pretrial detention order apparently completed, imposing conditions of release including bail (Carroll 2021). Within several counties, judges took different positions regarding appropriate bail amounts and standard release conditions (Carroll 2021). In addition, other procedural protections were followed sporadically; for example, in one county, some judges insisted on counsel being present, whereas others proceeded without counsel if the defendant did

www.annualreviews.org • Pretrial Detention Reform 79

not object. Such observations are consistent with those reported in other jurisdictions (Heaton 2021).

2.2. Executive Reform

Although legislative reforms around pretrial detention have been significant, particularly in the third wave of pretrial detention reform movements, they are not the only form of government- based reform. Recently, executive actors have engaged in their own forms of criminal legal system reform, particularly on local levels, that have impacted pretrial detention. From mayoral deci- sions to discontinue arrests for marijuana possession in New York City (Mueller 2018), to sheriffs exercising discretion to release pretrial detainees during COVID to alleviate jail overcrowding (Accardia 2020), to the election of self-identified progressive law enforcement officers (Dowd 2022, Taylor 2022,Westervelt 2019), executive actors have played a recent and significant role in pretrial detention reform (Mueller 2018, Taylor 2022).

Progressive prosecutors have played a significant role in this reform effort. Their rise in pop- ularity signals a community coalescing around less rather than more detention (Taylor 2022, Westervelt 2019). Although community-based reforms are discussed in the next section, these reform movements have undoubtedly influenced the election of progressive prosecutors.

One of the most notable of such prosecutors is Larry Krasner of Philadelphia, who ran on a platform of reforming (among other aspects of the city’s criminal legal system) pretrial deten- tion (Kreps 2021, Taylor 2022). Krasner opted for a categorical approach designating particular offenses for which his office would not seek pretrial detention (Melamed 2019,Taylor 2022). In ad- dition, he announced noncharging policies for some types of offenses and offenders (Taylor 2022). Those that fell within these categories would either be offered alternative dispositions or face no charges. Krasner was not the only progressive prosecutor elected in the last two decades, but his efforts to reform a large urban office and to place pretrial detention at the center of the undertak- ing garnered national attention and fueled further conversations around the role of prosecutorial discretion and the value of pretrial detention (Ouss & Stevenson 2023, Taylor 2022).

Although Krasner was an outspoken proponent of reform, the success of his efforts in reducing pretrial detention is difficult to assess, particularly in comparison to nonprogressive prosecutors (Palmer 2020, Taylor 2022). There is little question that Krasner sparked an important conver- sation about pretrial detention and prosecutors’ role in burgeoning jail populations. In addition, his policies suggest an awareness that for some offenders, pretrial detention would be inappropri- ate regardless of the amount of evidence gathered against them. As intuitive as this may seem to some, such a position was a sea change, at least to official policy. This is not to say that individual prosecutors in other jurisdictions do not adopt similar release positions in particular cases or may make similar decisions not to charge particular offenders or offenses. However, use of prosecuto- rial discretion in this way is difficult to track. Krasner’s adoption of an office-wide policy created a previously absent transparency around the use of such discretion.

In many ways, however, Krasner’s policies were limited reforms. Rather than implementing a global policy of no cash bail, he limited his no-bail policy to a narrow swath of offenses for which defendants with no criminal record were likely to be released without bail regardless of the state’s position (Palmer 2020). In addition, he couched the terms of his pretrial reform solely within the control of his office (Taylor 2022). His office made charging decisions, after all. If a prosecutor wanted to seek bail, they could seek an alternative charge for the alleged conduct that would allow them to circumvent the new policy. Indeed, one of the most common critiques of Krasner around his use of charging discretion was that he had not removed enough members of the prior admin- istration to effectuate his policies, particularly in the beginning of his first term (Taylor 2022). As a result, prosecutors from former administrations remained at the forefront of charging decisions

80 Carroll

and simply adjusted according to Krasner’s new policy to charge outside of the designated list of offenses to achieve pretrial detention (Taylor 2022). One critique suggested that prosecutors were charging more aggressively in light of Krasner’s reform efforts than prior administrations had (Taylor 2022). Unfortunately, data are difficult to collect, as charging discretion is not subject to public scrutiny and few data exist around cases that are never charged, so the claim is difficult to prove or disprove. If it is in fact true, however, it raises concerns about the effectiveness of pretrial detention reform that is prosecutorally driven (Palmer 2020, Taylor 2022).

Further early examination of the effect of Krasner’s tenure suggested replication of biases com- mon in other prosecutors’ offices (Palmer 2020). Reliance on prior criminal records, including noncharged encounters with police, as a threshold qualifier for no bail resulted in the exclusion of overpoliced and overprosecuted populations. Finally, his office’s no-bail policy did not extend to nonmonetary conditions (Carroll 2021). Public defenders reported that even as the office re- ceived praise for Krasner’s no-bail policy, accused persons were routinely subjected to conditions of release that carried monetary consequences, such as fees associated with uranalysis, electronic monitoring, or mandated counseling (Carroll 2021).

Despite the criticisms of progressive prosecution policies, one study found that the imposition of bail had little effect on flight risk (Ouss & Stevenson 2023). Defendants who were subjected to cash bail as a condition of release prior to Krasner’s administration had roughly the same failure- to-appear rate as those released without bail under Krasner’s policy. This would suggest that bail, for this limited category of individuals, did not serve the strong function of encouraging appear- ance as its proponents had claimed but did increase pretrial detention rates. Further study into the impact of Krasner’s and other progressive executive actors’ policies would fill a critical gap in pretrial detention literature.

2.3. Judiciary-Based Reform

Thus far, this section has examined reform from executive and legislative actors. In addition, judges have altered pretrial detention systems in response to or instead of reform effectuated by other governmental actors. These judicial alterations occur on micro and macro levels.On a micro level, judges in individual cases have modified or altered pretrial calculus to permit release or detention based on facts significant to single cases.On amacro level, judges have issued rulings in response to impact or class action litigation that has altered pretrial detention policies or the implementation mechanisms of such policies. Macro-level changes are easier to track, as courts issue orders that affect large populations or whole systems and appellate processes create detailed records of pro- ceedings [In re Kenneth Humphrey (2021);O’Donnell v.Harris County Texas (2017, 2018)].That said, even macro-level changes can elude observation when parties enter into settlement agreements that are sealed or otherwise not made public. On the micro level, judicial action can be especially difficult to track, because individual decisions may not be published and, because of the nature of pretrial decision making, conditions of release or detention decisions are transient, replaced by sentencing orders or release upon acquittal. Even when records exist, exercise of judicial discre- tion in calculating conditions of release, including bail, can be opaque and difficult to quantify. As noted above, pretrial detention hearings often happen quickly (Carroll 2021).

All this highlights lingering questions about the efficiency or effectiveness of pretrial deten- tion reformwithout requirements of judicial transparency. Put another way, discretionary decision making by judges is frequently lauded as a mechanism to ensure “right” outcomes. Yet such deci- sionmaking is subject to bias and decision fatigue,whereby a judge, having heard somany pleas for release, becomes unsympathetic or simply too tired to apply proper scrutiny to each case (Danziger et al. 2011, Tokson 2015,Wistrich et al. 2015). Alternatively, the judge may be significantly more lenient with individuals they find sympathetic than those with whom they have little in common,

www.annualreviews.org • Pretrial Detention Reform 81

leading to inconsistent release patterns.This is significant not only because it can signal embedded bias and create uneven decision-making models but also because it suggests that those who can conform with judicial expectations are more likely to benefit from judicial discretion than those who cannot (Gouldin 2018, 2020; Mayson 2018). Such behavior has been studied in the context of verdicts and sentencing, though less work has been done in the context of pretrial detention (Farrell & Givelber 2010, Kalven & Zeisel 1966, Starr & Rehavi 2013). Beyond this, local incen- tives may push judges toward detention rather than release when in doubt (Carroll 2021, Gouldin 2018). Judges are rarely praised for allowing accused people to live within their community with- out incident. They are, however, routinely criticized for allowing pretrial release of individuals who commit heinous crimes while on bail. This negative publicity effect often counsels toward detention (Carroll 2021,Gouldin 2018). Finally, judges may not have much in common with those subject to their pretrial detention decisions. As a result, they may have trouble understanding or sympathizing with claims that even seemingly small financial burdens create insurmountable obstacles to release for marginalized defendants (Gouldin 2018).

3. COMMUNITY-BASED REFORM

One of the biggest influences on pretrial detention reform has been the rise of community-based movements. Unlike their government-based counterparts, such efforts have been driven by col- lective action to work within existing criminal legal systems. This section focuses on several such actions, namely, the creation of community bail funds and community-based alternatives to de- tention and the effect of court watching on pretrial detention decisions. In addition, this section considers the possibility of still more community involvement, such as that proposed by Appleman (2012) for bail juries. This is far from a complete accounting of community-driven reform of pretrial detention systems in practice or in theory. Still, it is a start, and others have offered far more detailed descriptions of this important aspect of pretrial detention reform (Simonson 2014, 2017a,b).

3.1. Community Bail Funds

As broad public awareness around policing tactics has grown, so have community bail funds. The organization of such funds gives community actors a share in collective decision making around pretrial detention (Simonson 2017a,b). Functionally, they allow individuals to pool resources to provide bail for pretrial detainees. They have taken various forms, from purely locally organized endeavors that post limited amounts of bail to national movements to bail out particular individ- uals, such as Mama’s Day Bail Outs or bailouts for persons accused of particular types of crimes (Simonson 2017a,b). These funds are nongovernmental agencies and are often volunteer based. In this, they are a form of direct community action against pretrial detention.

Despite their growing popularity, these funds have also been criticized as perpetuating existing pretrial detention decisions. Early critiques raised concern that bail funds rendered bail more acceptable by creating a safety valve for individuals unable to afford bail (Simonson 2017a,b). Such a critique reasoned that judges aware of such bail funds, particularly targeted funds, would simply raise bail amounts to frustrate reform efforts. It is difficult to assess if this prediction has occurred without uniform data on bail decisions before and after the presence of such funds. The existence of such funds has certainly raised public awareness of the impact of even small bail amounts on indigent defendants. Their popularity suggests that this awareness favors release but also signals faith in released individuals to return to court and remain safely within communities.

Another dilemma in community bail funds is that they tend to rely on the ability to recycle available funds. Individuals who benefit from bail funds agree to return for all future court

82 Carroll

appearances and abide by conditions of release (Simonson 2017a,b). If such terms are met, at the conclusion of the court proceedings, the posted bail money is returned to the bail fund for use to gain the release of another individual. In this, bail funds can create reliable cycles of release. Not all jurisdictions return bail funds even following successful termination of the period of pretrial detention, however. In Alabama, for example, a convicted defendant’s bail is used to pay off fines, fees, and restitution, before any remainder can be returned. As a result, bail funds in jurisdictions with similar practices cannot recycle funds fully, if at all. This reality may contribute to a perverse incentive to increase bail amounts to exhaust existing community bail funds, though again, no studies tracking this trend exist.

3.2. Court Watching

Although not targeted directly at pretrial detention reform, court watching has emerged as a mechanism to monitor court proceedings and push for transparency, accountability, and consis- tency among formal actors (Simonson 2014). I am aware of no longitudinal studies on the effect of court watching efforts, other than anecdotal accounts in scholarly papers and court watching projects (Siegler & Zunkel 2020). However, proponents of court watching have argued that the presence of community-based witnesses to judicial action can affect decision making (Siegler & Zunkel 2020). Beyond this, others have argued that court watching may increase involvement in other reform efforts, as court watchers are motivated by what they witness as observers, and may heighten a sense of ownership and involvement in criminal legal processes (Simonson 2014, 2017a,b, 2019).Certainlymore empirical research into the impact of this community-based reform effort is needed.

3.3. Community-Based Alternatives to Pretrial Detention

A third area of community-based reform is in the push for conditions of release that allow in- dividuals to remain in their communities pretrial. I have written elsewhere, as have others, that such community-based alternatives to detention can carry significant financial and emotional costs (Carroll 2021,Weisburd 2022). In addition, these conditions often relinquish degrees of autonomy for those subject to them. Section 5 addresses low- or no-cost alternatives that may accomplish more effective reform of pretrial detention rates. In this section, however, it is worth noting that such reform efforts around community-based alternatives to pretrial detention have served an ad- ditional function besides simply creating avenues of release: They have also pushed back against binary definitions of safety in which the calculation is weighted toward the effect of release on the community’s safety as opposed to the effect of detention on that safety (Carroll 2020b,c; 2021). By acknowledging and pushing for return of individuals to their communities, the effect of pre- trial detention of an individual on the community as a whole is recentered. To the extent that reform movements have historically sought to address the harm created by pretrial detention, these reforms in particular define that harm more broadly.

3.4. Bail Juries

In her 2012 article “Justice in the Shadowlands,” Appleman (2012) noted that decisions around pretrial detention are often purposely obscured. Risk assessments are opaque by nature, with limited information about the basis of their calculations or interpretation of those calculations. Likewise, the Supreme Court has done little to explain what process is required around the im- position of pretrial detention or conditions of release (Carroll 2020a). For their parts, states have adopted various procedures that render pretrial detention hearings more efficient than rigorous

www.annualreviews.org • Pretrial Detention Reform 83

(Carroll 2021). As a result, they often forgo fundamental protections that lend legitimacy and fairness to resulting decisions (Carroll 2020a, Gouldin 2018).

Among possible solutions she proposes, Appleman (2012) suggests pretrial detention juries. It is a small part of her paper, but it is a significant reform suggestion. This section has considered community-based movements to drive pretrial detention reform. These movements have enjoyed tremendous success both in terms of their own efforts and in terms of pushing governmental actors toward systematic reform as described in Section 2. Appleman’s suggestion of a pretrial detention jury is unique in that it would offer the opportunity for direct and formalized community input into pretrial detention decisions, as jurors would set conditions of release.

I do not mean to suggest, nor do I think Appleman would suggest, that such a reform is a panacea. It would carry its own unique risks that critics of juries are fond of noting (Abramson 1998; Carroll 2012, 2015). Lack of representation on petit juries and bias in jury selection and among jurors themselves would be present in such juries (Carroll 2015). Beyond this, exclusion from jury service of populations based on prior conviction or housing insecurity in many juris- dictions would likewise be replicated in bail juries. Such risks, however, could be mitigated by expanding juror eligibility and by other procedural protections, such as opportunities for effective counsel and robust adversarial hearings, as the court contemplated in Salerno.

Bail juries may also carry the same unintended consequence as jury trials. They would render pretrial detention hearings less efficient and, in the process, would encourage their avoidance. In the trial context, this has created a coercive system of pleas (ironically, driven partly by high pretrial detention rates), but one cannot help but wonder if jurors asked to set conditions of release might be willing to weigh the harms of pretrial detention against risks created by individuals in the community and the harm suffered by those who care for them (Appleman 2012). Certainly, a recent study by Stevenson &Mayson (2022), in which individuals were asked to rank the harm of detention against harm related to crime, suggests that, at least among their interview subjects, the harm of detention weighs heavily even in the face of grave harm. In short, even when confronted with serious accusations against an individual, lay decision makers (such as jurors), unlike their judicial counterparts, were more willing to release because they viewed the deprivation of liberty as a significant burden.

4. THE RISE OF THE MACHINE: PRETRIAL ASSESSMENT TOOLS

One of the biggest areas of pretrial detention reform has been in the development and adop- tion of pretrial assessment tools, specifically machine-generated assessment tools (Mayson 2019). In many ways, these tools are a continuation of the early assessment questionnaires used by the Manhattan Bail Project and the Vera Institute in the 1960s and 1970s. Like these early assessment tools, machine-based assessments are designed to accurately calculate risk of flight and/or safety concerns while reducing bias (Garrett 2022, Gouldin 2020, Mayson 2019). Such machine-based assessment tools rely on algorithms to generate a score that correlates with the potential risk for each individual. Kentucky was the first state to adopt and uniformly use pretrial assessment tools [Public Safety and Offender Accountability Act, H.B. 463, Gen. Assemb., Reg. Sess. Ky. (2011)].

The touted benefit of machine-based pretrial assessment tools was that they would reduce bias and increase accuracy in the calculation of risk that might be sufficient to support pretrial detention (Mayson 2019). Despite this claim, concerns surrounding the use and construction of such tools surfaced quickly. First, although the machine-based tools could generate risk scores, they could not protect against inconsistent use of the scores (Gouldin 2020). Decision makers— whether pretrial officers or judges—proved inconsistent in their use of scores.

84 Carroll

Second, scholars noted initially that such tools were black boxes, with system creators claiming that algorithms used to generate scores were proprietary (Mayson 2019). Even as transparency increased surrounding the information that these tools relied on to generate scores, new concerns arose. Algorithms relied on similar data as previous assessment tools and so replicated bias embed- ded in these inquiries (Mayson 2019). Reliance on data points such as criminal history and prior arrests without charge created a class of “dangerous defendants” and embedded bias (Eaglin 2017, Gouldin 2018, Mayson 2018). As Mayson’s (2019) work showed, these machine-based assessment tools carried the bias of their creators in terms of both the data they collected and the risk con- struction and calculation. The questions they chose to ask and the weight they gave answers to those questions reflected their own perceptions of risk. ProPublica’s study of one such algorithm confirmed this. In a 2016 article, “Machine Bias,” Angwin et al. (2016) analyzed COMPAS (Cor- rectional Offender Management Profiling for Alternative Sanctions) in the context of sentencing and found racial and gender bias in its assessment of risk.

Perhaps such bias,whether from construction or application, is inherent in systems that attempt to predict risk based on past behavior (Mayson 2019). Such predictions by their nature will be imperfect and inevitably rely on the predictors’ concerns, whether those concerns surface in the construction of the tool itself or in its application (Gouldin 2018).Efforts to respond to critiques of bias also suggest thatmodification of such toolsmay help ameliorate bias in construction.Likewise, adoption of uniform standards of interpretation of results between and among jurisdictions may help reduce bias in application. Such standards, however, come at the cost of discretion. To the extent that discretion is viewed as a mechanism to adjust otherwise rigid systems, its loss may be significant—though, as discussed above, discretion can certainly be a double-edged sword in decision making and frequently is identified as its own source of bias in application.

Such concerns about poor construction or application of the risk assessment tool may sug- gest that increased procedural protections within pretrial detention hearings may remedy some concerns (Carroll 2020a,b,c). Appointment of counsel, rigorous adversarial hearings that include access to discovery, written findings by decision makers, and the opportunity to appeal pretrial detention and conditions of release each serve to safeguard against bias and ensure consistent and transparent application of pretrial decision-making devices (Carroll 2020a,b). In addition, in- creased inquiry into the construction and application of machine-based pretrial decision making is needed.

5. RESOLVING CONTINUING PROBLEMS IN THE PRETRIAL DETENTION ARENA

Whereas the prior three sections discussed current reform efforts, this section considers lingering issues around pretrial detention and reform efforts. These movements have focused primarily on either the mechanisms by which detention decisions are made or bail as a barrier to release. As discussed above, both warrant attention. However, this focus has obscured additional significant barriers to release and potential sources of bias in systems. This section considers three of these: nonmonetary conditions of pretrial release, conditions of detention themselves, and the lack of procedural protections during pretrial hearings.

5.1. Nonmonetary Conditions of Pretrial Release

As discussed above, reform efforts that have focused on bail have made significant headway in reducing reliance on bail, restricting predatory bonding practices, and creating community-based bail funds. Each of these serves to reduce disproportionate impacts of poverty on pretrial

www.annualreviews.org • Pretrial Detention Reform 85

detention. Socioeconomic conditions, after all, are not indicia of higher risk, even as they increase the probability of detention among marginalized populations.Without detracting from the value of these reform efforts, this focus on bail often ignores the costs of nonmonetary conditions of pretrial release—monetary and otherwise (Carroll 2021). This is a mistake if reform efforts are serious about the goal of reducing pretrial detention.

Nonmonetary conditions of release vary widely, from seemingly minor requirements—such as returning to court for all future hearings—to more complex requirements—such as submis- sion to monitoring (Carroll 2021, Weisburd 2022). These conditions carry various types of costs for different defendants. Some of these conditions are routinely imposed even in jurisdictions that follow no monetary bail policies. In my own work, I requested copies of blank pretrial detention/release orders from courts throughout the country (Carroll 2021). Many of these or- ders showed “prechecked” conditions of release that were always imposed on accused persons, including a requirement to return for all court appearances, to submit to drug testing if ordered by pretrial service officers, and to commit no future crimes/have no future arrests. Contact with selected clerks in jurisdictions in which orders did not have “prechecked” conditions revealed that judges did routinely order such conditions. The imposition of “routine” conditions, no matter how minor, raises questions about the individuality of the assessment (Carroll 2021). That these do not appear to carry any monetary costs (a contested proposition) or that they appear minor does not alter the fact that they are clearly not related in any way to a discovered risk (Carroll 2021, Weisburd 2022). They are imposed no matter how good the accused’s prior record or how disproportionate a burden they create.

Even relatively minor conditions of release, such as requirements to return to court for future court appearances, may create burdens on individuals who lack access to reliable transportation, public or private; who have inflexible work schedules or child or eldercare obligations; or who are economically marginal (Carroll 2021). In addition, those with housing insecurity may fail to receive mailed notices of future appearances. Providing free text notices of future court appear- ances and grace periods that allow individuals to report within designated periods (often a week) have successfully reduced failures to appear in jurisdictions using them (Carroll 2021, Cyrus 2022, Fishbane et al. 2020, Herian & Bornstein 2010). Providing free parking, transportation, and child and elder care may also help reduce costs created by appearance requirements (Carroll 2021). Fi- nally, reduction in the required number of appearances for indigent defendants may help reduce lost wages and other costs associated with appearance.

More significant conditions of release carry correspondingly higher burdens. In many jurisdic- tions, required drug testing and electronic monitoring carry financial costs—costs that at times are higher than imposed bail amounts (Carroll 2021, Weisburd 2022). Such conditions are also invasive, as they constantly monitor defendants and at times their family members and close asso- ciates (Weisburd 2022). Those subject to electronic monitoring may be required to demonstrate that they have reliable high-speed Internet service in their home. This is not only costly but im- possible to obtain in some rural areas. In response, some jurisdictions have agreed to waive or reduce costs connected to these conditions, but this is the exception, not the rule. Finally, some conditions of release, such as required no-contact orders, are criminogenic. Even if an individual is found not guilty of the underlying offense that generated the condition of release, they may still suffer conviction for violating the condition imposed (Carroll 2021). Although these nonmonetary conditions do permit individuals to avoid detention, the costs they impose and, in some circum- stances, their disconnect with risk raise serious questions about their constitutionality and the disproportionate impact they replicate on marginalized defendants. Like monetary conditions of release, nonmonetary conditions may be imposed disproportionately on marginalized defendants and may create barriers to pretrial release.

86 Carroll

5.2. Conditions of Detention and Their Effect on the Detained

As pretrial detention numbers continue to grow across the nation, jail populations far exceed that which they were designed to house. Increased rates of detention and delays in time to trial exacerbate this problem. The COVID-19 pandemic made it worse (Carroll 2020a, Garrett & Kovarsky 2022). Courts shut down in response to public health orders and speedy trial rights were suspended, swelling jail populations and rendering short terms of detention into indefinite ones (Garrett & Kovarsky 2022). These crowded conditions render jails unsafe spaces.

During the COVID-19 pandemic, jails became viral epicenters, with infection spreading among detained persons and to communities beyond (Carroll 2020b,c; Garrett & Kovarsky 2022). COVID-19 was hardly the first pandemic to ravage the vulnerable and crowded populations of pretrial detainees. Prior to this most recent global health crisis, MRSA (methicillin-resistant Staphylococcus aureus) virus and hepatitis C, to name just two, demonstrated the ease and persis- tence of infection in jail and the shortage of preventive and curative care (Carroll 2020b). For those with preexisting health conditions, including mental health and behavioral needs, jails often fail to provide even routine and established care, creating space instead for decompensation away from functional baselines (Carroll 2020b,c).

From a public health standpoint, jails are a disaster in themaking (Carroll 2020b).This is in part due to the nature of jails—for example, they tend to be crowded; have common ventilation systems, toilets, and showers; restrict personal hygiene products; and have minimal budgets for services such as health care (Carroll 2020b). Beyond this, their populations tend to be more transient than prison populations (Carroll 2020b, Garrett & Kovarsky 2022). Individuals in jail are serving short sentences in some jurisdictions, are awaiting transport to prison facilities or sentencing, or are pretrial detainees. This means not only increased exposure to contagions and mental health triggers, with constantly changing populations, but also fewer opportunities and arguably lower incentives for the state to create continuity of care for those housed in local jail facilities (Carroll 2020b,c).

The harm of pretrial detention, however, is not limited to public health impacts, although re- cently more attention has focused on this impact. Studies have repeatedly documented impacts of pretrial detention on accused persons (Dobbie et al. 2018, Heaton et al. 2017, Lowenkamp et al. 2013, Scully 2021). Even brief periods of pretrial detention can affect an individual’s ability to de- fend against a charge and may incentivize accepting a plea offer regardless of guilt (Carroll 2021, Gouldin 2020,Mayson 2018).Pretrial detention can also carry tremendous personal and economic impacts (Heaton et al. 2017). Those detained prior to their trials lose jobs, educational opportu- nities, homes, financial independence, and custody of children (Carroll 2021). They lose access to the very communities that might support them in times of crisis. Their communities in turn lose them. Detained individuals suffer the physical, emotional, mental, and economic harms of their detention, but so do those who depend on them (Carroll 2021,Heaton et al. 2017). Although the current literature sketches the impacts of pretrial detention on those detained, studies within their communities are rare. The failure to account for this impact not only has fueled community- based reform movements but also points to a gap in criminal legal systems’ understanding of (or perhaps empathy for) the full impact of pretrial detention. If the state is to take seriously the goal of achieving safety through its pretrial detention regime, it must also take seriously the harm that regime creates for those it holds and the community left to suffer their absence.

All of these issues are likely to be exacerbated, as claims of rising crime rates since June 2020 have spurred conversations about increasing pretrial detention rates even further (Bates 2022). Although the reality of rising crime rates—and more specifically the types of crime rising—is a contested proposition, many state officials appear to be focusing on the need to detain more individuals pretrial, rather than fewer (Akinnibi 2022). Given the lack of opportunities to contest

www.annualreviews.org • Pretrial Detention Reform 87

conditions outside of civil impact litigation (which can present its own problems in a transient population such as pretrial detainees) individuals detained pretrial may face significant obstacles to challenging their conditions of confinement (Kovarsky 2020).

5.3. Due Process Failures

Finally, litigation-based reform efforts have attempted to raise claims surrounding insufficiency of process in pretrial detention hearings. This litigation strategy has met with limited success. Given the Supreme Court’s failure to interpret the Eighth Amendment excessive bail prohibition as offering relief for pretrial detention when defendants cannot post bail (Colgan 2018), courts are looking increasingly to the Due Process Clause to offer relief for prolonged periods of pretrial detention (Carroll 2020a). The difficulty is that the court has not clearly established what process the accused is due during pretrial detention hearings (Carroll 2020a).

The court’s decision in Rothgery v. Gillespie County (2008) set some procedural standards, though, as Garrett (2022) noted, these standards, including appointment of counsel, are frequently overlooked in pretrial detention hearings. Cases interpreting bail offer little support. As noted above, Salerno permitted pretrial detention after an assessment of risk and a rigorous adversarial hearing, but it failed to define what such a hearing would look like [United States v. Salerno (1987)]. Salerno did not overturn Stack v. Boyle’s (1951) requirement that the conditions of release be tied to compelling and articulated state interests, though little has been done to force the state to es- tablish this nexus between conditions and state interests. More recently, in O’Donnell v. Harris County (2017), the district court declined to adopt a substantive due process claim with regard to bail schedules, though this area of litigation may yet prove promising. Not unrelatedly, litigants have also raised equal protection claims, which may provide some traction with regard to existing systems’ disparate impact on marginalized populations (Carroll 2020a).

Clearly, as cases wind their way through state and federal courts, questions around sufficiency of process and inequities in current systems will persist (Garrett 2022). This creates several possibili- ties, including that such systems will voluntarily adopt or legislatures will impose on them process that may reduce concerns about pretrial detention systems and may create the possibility of mean- ingful relief when pretrial release is denied or rendered unattainable by the court’s imposition of conditions. Increased procedural requirements may reduce the number of detention requests filed by the state and may push for more nuanced assessment of actual risk. Put another away, requiring a hearing may mean that the state—as prosecutors or judges—may be disinclined to trigger pro- cess requirements. In short, they may forgo a pretrial detention request.The unanswered question is whether jettisoning pretrial detention in most cases will actually increase recidivism and/or fail- ures to appear in court. It is difficult to answer this question with any certainty in the abstract, though data in jurisdictions that have reduced pretrial detention do not suggest increased rates of reoffense or flight (Lowenkamp et al. 2013,Ouss & Stevenson 2023).Within juvenile delinquency systems, some jurisdictions have begun to experiment with eradicating pretrial detention of juve- niles accused of crimes within juvenile criminal legal systems (Hildebrand 2021). Although these programs are relatively nascent, emerging crime data from such jurisdictions demonstrate neither a spike in failure-to-appear warrants nor new crime. Although juvenile systems are unique from adult criminal legal systems, the success of these programs certainly raises their possible replica- tion in an adult context. Beyond this, community benefit gleaned from reduced detention rates may offset risk in the aggregate.

6. CONCLUSION

A tremendous amount of work remains to be done in the context of pretrial detention. Although the reform movements including those described above have sought to reduce bias in pretrial

88 Carroll

detention decision making and to increase opportunities for release of accused persons, efforts have failed to alleviate the burgeoning pretrial detention population in the United States. In addition, even as pretrial assessment tools have become more sophisticated, they still rely on dated risk criteria and fail to shift the conversation about safety away from detention and toward community integration. Finally, concepts of safety remain binary—the accused person versus the community. Yet communities suffer and may be less rather than more safe when individuals are detained pretrial. As reform movements progress, these concerns will continue to require redress.

Throughout this article, I have identified gaps in the literature that might document the impact of new and long-established reform efforts on detention rates and on safety within communities. Beyond this, increased research into the impact of conditions of release and/or detention as well as into alternative reform possibilities is needed. Small changes, such as provisions for free parking or transportation, child or elder care, or contemporaneous text notice of hearings, may increase appearance rates and benefit the larger community. Yet little research has been done to document their impact.

Given the current political climate, the urgency of reform in the area of pretrial detention is unlikely to abate. The need for greater research with an eye toward documenting both what has worked (and why) and where alternative solutions might present is equally urgent.

DISCLOSURE STATEMENT

I am a former public defender who represented marginalized defendants in the systems described in my work. This professional experience informs my scholarship.

ACKNOWLEDGMENTS

The author would like to thank AdamSteinman, Judith Resnik,BrandonGarrett,LaurynGouldin, and Russell Gold for their helpful thoughts and comments on the topic of bail reform.

LITERATURE CITED

Abramson J. 1998. Two ideals of jury deliberation.Univ. Chicago Legal Forum 1998(1):125–60 Accardia M. 2020. Morgan jail releases nonviolent inmates to protect against COVID-19.Hartselle Enquirer,

April 19. https://hartselleenquirer.com/2020/04/19/morgan-jail-releases-nonviolent-inmates- to-protect-against-covid-19/

Akinnibi F. 2022. How bail reform, crime surge mix in an angry debate. Washington Post, Novemb. 4. https://www.washingtonpost.com/business/how-bail-reform-crime-surge-mix-in-an-angry- debate/2022/11/04/1ae90d48-5c60-11ed-bc40-b5a130f95ee7_story.html

Angwin J, Larson J, Mattu S, Kirchner L. 2016. Machine bias. ProPublica, May 23. https://www.propublica. org/article/machine-bias-risk-assessments-in-criminal-sentencing

Appleman LI. 2012. Justice in the shadowlands: pretrial detention, punishment, & the Sixth Amendment. Washington Lee Law Rev. 69:1297–369

Ares C, Rankin A, Sturz H. 1963. TheManhattan Bail Project: an interim report on the use of pre-trial parole. NYU Law Rev. 38:67–95

Baradaran Baughman S. 2017. The Bail Book: A Comprehensive Look at Bail in America’s Criminal Justice System. Cambridge, UK: Cambridge Univ. Press

Baradaran Baughman S. 2020. Dividing bail reform. Iowa Law Rev. 105:947–1025 Bates J. 2022. Eric Adams wants “dangerousness” factored into New York’s bail laws. Advocates say it

will only bring more bias. Time, Febr. 10. https://time.com/6146431/eric-adams-bail-reform- dangerousness/

Botein B. 1965. The Manhattan Bail Project: its impact on criminology and the criminal process. Tex. Law Rev. 43:319–31

www.annualreviews.org • Pretrial Detention Reform 89

Carroll JE. 2012. The jury’s second coming.Georgetown Law J. 100:657–707 Carroll JE. 2015. The jury as democracy. Ala. Law Rev. 66:825–70 Carroll JE. 2020a. The due process of bail.Wake Forest Law Rev. 54:757–93 Carroll JE. 2020b. Pretrial detention in the time of COVID-19.Northwest. Univ. Law Rev. Online 115:59–87 Carroll JE. 2020c. Safety, crisis, and criminal law. Ariz. State Law J. 51:769–809 Carroll JE. 2021. Beyond bail. Fla. Law Rev. 73:143–97 Carson E. 2022. Prisoners in 2021. Stat. Tables, Bur. Justice Stat.,Washington, DC. https://bjs.ojp.gov/sites/

g/files/xyckuh236/files/media/document/p21st.pdf Colgan BA. 2018.The excessive fines clause: challenging the modern debtors’ prison.UCLA Law Rev. 65:2–77 Cyrus R. 2022. The alternative to cash bail is a simple reminder. American Prospect, Sept. 6. https://prospect.

org/justice/the-alternative-to-cash-bail-is-a-simple-reminder/ Danziger S, Levav J, Avnaim-Pesso L. 2011. Extraneous factors in judicial decisions. PNAS 108:6889–92 Dobbie W, Goldin J, Yang CS. 2018. The effects of pre-trial detention on conviction, future crime, and

employment: evidence from randomly assigned judges. Am. Econ. Rev. 108:201–40 DowdT. 2022.Care about policing? You should be watching these sheriff races.Vice News,Novemb. 8.https://

www.vice.com/en/article/epzkjm/midterms-sheriffs-elections Duker W. 1977. The right to bail: a historical inquiry. Ala. Law Rev. 42:33–120 Eaglin JM. 2017. Constructing recidivism risk. Emory Law J. 67:59–122 Farrell A, Givelber D. 2010. Liberation reconsidered: understanding why judges and juries disagree about

guilt. J. Crim. Law Criminol. 100:1549–86 Fishbane A,Ouss A, Shah AK. 2020.Behavioral nudges reduce failure to appear for court.Science 370:eabb6591 Foote C. 1956. The coming constitutional crisis in bail: I.Univ. Pa. Law Rev. 113:959–99 Garrett BL. 2022. Models of bail reform. Fla. Law Rev. 74:879–931 Garrett BL, Kovarsky L. 2022. Viral injustice. Calif. Law Rev. 110:117–78 Gouldin LP. 2018. Defining flight risk.Univ. Chicago Law Rev. 85:677–742 Gouldin LP. 2020. Reforming pretrial decision-making.Wake Forest Law Rev. 55:857–906 Heaton P. 2021. Enhanced public defense improves pretrial outcomes and reduces racial disparities. Indiana

Law J. 96:701–50 Heaton P,Mayson S, Stevenson M. 2017. The downstream consequences of misdemeanor pretrial detention.

Stanford Law Rev. 69:711–94 HerianMT, Bornstein BH. 2010. Reducing failure to appear in Nebraska: a field study.Nebraska Lawyer, Sept.

https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1008&context=publicpolicyfacpub Hildebrand SS. 2021. Reviving the presumption of youth innocence through presumption of release: a

legislative framework for abolition of juvenile pretrial detention. Pa. State Law Rev. 125:695–735 In re Kenneth Humphrey, 482 P.3d 1008 (Ca. 2021) Johnson L. 1966. Remarks at the signing of the Bail Reform Act of 1966. Speech, June 22. https://www.

presidency.ucsb.edu/documents/remarks-the-signing-the-bail-reform-act-1966 Kalven H, Zeisel H. 1966. The American Jury. Chicago: Univ. Chicago Press Katzive MC. 1968. New areas for bail reform. Rep., Vera Inst. Justice, Brooklyn, NY. https://www.vera.org/

downloads/publications/1497.pdf Kennedy R. 1963.Report of the Attorney General’s Committee on Poverty and the Administration of Criminal Justice.

Rep., Dep. Justice, Washington, DC Kennedy R. 1964. Testimony on bail legislation before the Subcommittees on Constitutional Rights and Improvements

in Judicial Machinery of the Senate Judiciary Committee. Testimony, 88th Congr. Kovarsky L. 2020. Pandemics, risks, and remedies. Va. Law Rev. Online 106:71–98 Kreps D. 2021. ‘Philly D.A.’: Larry Krasner plans radical bail reform in clip from PBS docuseries. Rolling

Stone, Febr. 5. https://www.rollingstone.com/culture/culture-news/philly-da-larry-krasner- docuseries-pbs-clip-1123869/

LowenkampCT,VanNostrandM,Holsinger A. 2013.The hidden costs of pretrial detention. Rep., Arnold Found., Houston. https://craftmediabucket.s3.amazonaws.com/uploads/PDFs/LJAF_Report_hidden- costs_FNL.pdf

Mayson SG. 2018. Dangerous defendants. Yale Law J. 127:490–568

90 Carroll

Mayson SG. 2019. Bias in, bias out. Yale Law J. 128:2218–300 Melamed S. 2019. Philly DA Larry Krasner stopped seeking bail for low-level offenses. Philadelphia In-

quirer, Febr. 19. https://www.inquirer.com/news/philly-district-attorney-larry-krasner-money- bail-criminal-justice-reform-incarceration-20190219.html

Mueller B. 2018. New York City will end marijuana arrests for most people.New York Times, June 19. https:// www.nytimes.com/2018/06/19/nyregion/nypd-marijuana-arrests-new-york-city.html

O’Donnell v. Harris County Texas, 251 F. Supp. 3d 1052 (S.D. Tex. 2017) O’Donnell v. Harris County Texas, 892 F.3d 147 (5th Cir. 2018) Ouss A, Stevenson M. 2023. Does cash bail deter misconduct? Am. Econ. J. Appl. Econ. 15:150–82 Palmer C. 2020. Tensions boiling over between Philadelphia District Attorney and bail reform advocates.

Philadelphia Inquirer, July 29. https://www.inquirer.com/news/philadelphia/philadelphia-da-larry- krasner-cash-bail-reform-advocates-20200729.html

Rothgery v. Gillespie County, 544 U.S. 191 (2008) Scully S. 2021. Criminal justice reform means reforming the mental health system. NAMI Blog, March

5. https://www.nami.org/Blogs/NAMI-Blog/March-2021/Criminal-Justice-Reform-Means- Reforming-the-Mental-Health-System

Siegler A, Zunkel E. 2020. Rethinking federal bail advocacy to change the culture of detention. Champion 44:46

Simonson J. 2014. The criminal court audience in a post-trial world.Harvard Law Rev. 127:2173–232 Simonson J. 2017a. Bail nullification.Mich. Law Rev. 115:585–638 Simonson J. 2017b. Democratizing criminal justice through contestation and resistance.Northwest. Univ. Law

Rev. 111:1609–24 Simonson J. 2019. The place of “the people” in criminal procedure. Columbia Law Rev. 119:249–308 Stack v. Boyle, 342 U.S. 1 (1951) Starr SB, Rehavi MM. 2013. Mandatory sentencing and racial disparity: assessing the role of prosecutors and

the effects of Booker. Yale Law J. 123:2–80 Stevenson MT, Mayson SG. 2022. Pretrial detention and the value of liberty. Va. Law Rev. 108:709–82 Taylor K-Y. 2022. Larry Krasner and the limits of “law and order.”The New Yorker,Novemb. 16.https://www.

newyorker.com/news/the-political-scene/larry-krasner-and-the-limits-of-law-and-order Thompson D. 2022. California stalls scaled down bail reform after years of delay. U.S. News, Sept. 1.

https://www.usnews.com/news/best-states/california/articles/2022-09-01/california-stalls- scaled-down-bail-reform-after-years-delay

Tokson M. 2015. Judicial resistance and legal change.Univ. Chicago Law Rev. 82:901–73 United States v. Salerno, 481 U.S. 739 (1987) Weisburd K. 2022. Punitive surveillance. Va. Law Rev. 108:147–221 Westervelt E. 2019. San Francisco elects Chesa Boudin as new district attorney. NPR, Novemb. 20. https://

www.npr.org/2019/11/20/781358722/san-francisco-elects-chesa-boudin-as-new-district- attorney

Wiseman SR. 2018. Bail and mass incarceration.Ga. Law Rev. 53:235–80 Wistrich AJ, Rachliski JJ, Guthrie C. 2015. Heart versus head: Do judges follow the law or their feelings? Tex.

Law Rev. 93:855–923 Woods A, Mayson SG, Sudeall L, Armstrong G, Potts A. 2020. Boots and bail on the ground: assessing the

implementation of misdemeanor bail reforms in Georgia.Ga. Law Rev. 54:1235–96

www.annualreviews.org • Pretrial Detention Reform 91