Commentary on Interim

Purposes of the Bail Reform Act and This Rule

In 2018, the General Assembly enacted legislation to reform the system under which courts subject defendants to pretrial conditions of release (the “Bail Reform Act” or “Acf). Codified in Chapter 21, Title 11 of the Delaware Code, the Bail Reform Act encourages the use of non-monetary conditions of release when those conditions reasonably assure the defendant’s appearance at court proceedings, public safety, and the integrity of the judicial process. In particular, the Act was intended to prevent defendants from being subjected to excessive financial conditions of release, traditionally referred to as money bail. By this means, the Act sought to reduce the unnecessary pretrial incarceration of defendants who are not wealthy enough to pay money bail, as well as reduce the resulting loss in employment, the pressure to plead guilty, the economic toll on non-affluent defendants and their families, and other substantial harm that results from the excessive use of money bail.

To accomplish those goals, this rule requires courts to impose the least restrictive conditions necessary to reasonably assure the defendant’s appearance in court, public safety, and the integrity of the judicial process. The rule also recognizes that the risk that the defendant will fail to appear is different from the threat that a defendant will commit harm if released pending trial. Finally, the Act and this rule reflect the overriding principle that, given the presumption of innocence underlying the American system of criminal justice, there is a crucial difference between pretrial conditions of release and post-conviction sentences, and thus the court shall not treat the former the same as the latter.

Overview of the Pretrial Assessment

Role of the pretrial assessment in accomplishing the purposes of the Act and this rule. The Bail Reform Act and this rule seek to increase the reliability and equity of decisions about conditions of pretrial release. To that end, the Act requires the use of an empirically developed pretrial assessment instrument. Consistent with that requirement, a diverse group of constituencies, with the aid of qualified academic and professional advisors, developed the “DELPAT.” The DELPAT takes into account factors relevant to whether defendants are, compared to other defendants, at greater risk of failing to appear at trial, endangering public safety, or obstructing justice, and is therefore designed to provide a reliable basis for setting conditions of release. The pretrial assessment was based on consideration of nationally available models and has been tested preliminarily for reliability and validity and to ensure that the assessment is not affected by bias based on race, gender, wealth, or other inappropriate grounds. It is designed to: (i) improve the reliability of pretrial decisions by ensuring that relevant factors are given consistent and measured weight, based on empirical testing and professional input; (ii) increase equity by ensuring that relevant factors are given consistent weight in all like cases, so that defendants are subject to equal treatment; and (iii) reduce discrimination against poor defendants.

Pretrial assessment score. To aid in the reliable, equitable, and efficient determination of discretionary conditions of release, the pretrial assessment measures the defendant’s risk of pretrial failure, which is defined as either (i) failing to appear after notice of a court proceeding or (ii) incurring a new criminal arrest while the current case is pending. The assessment has two scales that address these two distinct types of failure: a Failure to Appear (“FTA”) scale and a New Criminal Arrest (“NCA”) scale. The scores of these two scales are combined to produce the final pretrial assessment score,

Risk of failure to appear. The FTA scale assesses the risk that the defendant will fail to appear after notice of a court proceeding. The FTA scale employs four risk factors that were found to correlate with an increased risk of non-appearance:

(i) at least one prior sentence that included probation supervision in the past 10 years;

(ii) total number of prior FTA warrants in the past year;

(iii) total number of prior FTA warrants in the past 10 years; and

(iv) the current arrest includes at least one charge of Uniform Crime Reporting code “larceny” or “stolen vehicle.”

Factors (i) and (iv) each receive one point, if applicable. Factors (ii) and (iii) each receive one point if one prior FTA warrant is present and two points if two or more prior FTA warrants are present. Thus, a person’s FTA score will range from zero to six. There are three FTA scale risk levels: low (zero to one), moderate (two to three), and elevated (four to six).

Risk of new criminal arrest. The NCA scale assesses the risk that the defendant will be arrested for committing a new crime while the current case is pending. The NCA scale employs seven risk factors that were found to correlate with an increased risk of a new criminal arrest before trial:

(i) another pending criminal case;

(ii) prior convictions;

(iii) at least one prior misdemeanor arrest in the past two years;

(iv) at least one prior sentence that included probation supervision;

(v) age at first arrest;

(vi) at least one prior FTA warrant; and

(vii) at least one prior violent conviction within the past five years.

Each risk factor that is present receives one point, except that factor (ii) receives two points in the case of multiple convictions and factor (iii) receives two points. Thus, a person’s NCA score will range from zero to nine. There are four NCA scale risk levels: very low (zero to one), low (two to four), moderate (five to six), and elevated (seven to nine).

Testing of the Statutory Factors and FTA and NCA Scales

An important proviso: defendants at an elevated risk in the large sample used to validate the pretrial assessment were still likely to appear in court and to not get arrested. The testing of the pretrial assessment demonstrated that during the period before trial, most defendants (i) appeared in court and (ii) did not get arrested. That is, even as to the category of defendants deemed to be at an elevated risk of nonappearance or incurring a new criminal arrest, most defendants appeared as required at court proceedings and were not arrested before trial. Thus, in none of the tiers does the defendant’s pretrial assessment score indicate that the defendant is more likely than not to fail to appear or get arrested. Instead, the score indicates only that the defendant presents a greater risk of failing to appear or being arrested compared to individuals in lower tiers.

Testing and other consideration of statutory factors. Section 2105(b) of the Bail Reform Act requires the court to consider certain factors when determining the defendant’s risk of pretrial failure. To implement that requirement, the professionals developing the pretrial assessment tested many of these factors as part of its development. Some of the statutory factors were not included in the pretrial assessment because statistical analysis determined that they were not good predictors of the defendant’s risk of failing to appear or getting arrested. Some of the other factors could not be tested because of a lack of reliable and consistent data. Because the Act and this rule were designed to improve the consistent, reliable, and equitable determination of conditions of release, the use of factors that are not subject to reliable and consistent data risks defeating this key purpose. Other factors have been considered, and the empirical literature suggests they are associated with bias on the basis of race, gender, or wealth, and therefore the use of them would be inconsistent with the purposes of the Act and this rule,

Statiitoiy factors included in the FTA and NCA scales or elsewhere in this rule. Consistent with section 2105(b), the pretrial assessment, the domestic violence assessment, and this rule have fully or substantially taken into account many of these statutory factors and determined their appropriate weight, thereby obviating the need for courts to consider the factors on an inconsistent, ad hoc basis. For that reason, the court may not give additional weight to any factor fully or substantially included in the pretrial assessment, the domestic violence assessment, or this rule absent a special finding that there is a compelling reason indicating that the pretrial assessment, the domestic violence assessment, and this rule do not adequately account for the factor. Factors in this category needing a special fmdmg include:

(i) the nature and circumstances of the crime charged;

(ii) whether a firearm was used or possessed;

(iii) the possibility of statutory mandatory imprisonment;

(iv) the defendant’s record of convictions;

(v) the defendant’s history of amenability to lesser sanctions;

(vi) the defendant’s history of breach of release; and

(vii) the defendant’s record of appearances at court proceedings or of flight to avoid prosecution or failure to appear at court proceedings;

Statiitoiy factors tested and excluded for not being good predictors of the defendant’s risk of pretrial failure. Several of the section 2105(b) factors were tested and found to lack a sufficiently strong correlation with the defendant’s risk of pretrial failure. The rule therefore prohibits the court from giving weight to these factors. Factors in this category include:

(i) the defendant’s employment;

(ii) the defendant’s custody status at the time of the offense; and

(iii) the defendant’s length of residence in the community.

Suspect statutory factors not tested but presenting a risk of racial, gender, or wealth bias. Several of the section 2105(b) factors were not tested due to the lack of reliable and consistent data or preexisting evidence of racial, gender, or wealth bias in the academic literature. The rule prohibits giving weight to these suspect factors except by making special findings that considering the factor does not create disparhies based on race, gender, or wealth. Factors in this category include:

(i) the defendant’s family ties;

(ii) the defendant’s financial resources; and

(iii) the defendant’s character and mental condition.

Special note on the defendant’s character and mental condition. The Bail Reform Act refers to the defendant’s “character and mental condition.” These terms refer to separate concepts. The term “character” is subjective, has no reliable benchmark, and overlaps with many factors already taken into account by a defendant’s criminal record, which is a major input into the pretrial assessment. For that reason, any consideration of that factor is suspect. The term “mental condition” is vague and may also lead to biased decisionmaking, such as where a court uses the term to refer to the defendant’s perceived intelligence. That said, there may be circumstances that call for consideration of the defendant’s mental condition, in the more discrete sense of a case where the defendant is suffering from a mental illness or condition that indicates that the defendant poses a substantial risk to public safety, or where the defendant is currently under the influence of alcohol or drugs. Given the potential for bias, however, the rule instructs the court not to consider the defendant’s mental condition unless the defendant has objectively documented mental health issues relevant to the defendant’s risk to public safety or currently exhibits mental health or substance abuse issues relevant to the defendant’s risk to public safety. In the case of a defendant who is unable to knowingly and intelligently participate in presentment proceedings because of incapacitation as a result of the consumption of alcohol or the use of drugs, the court shall follow the separate statutory procedures and standards contained in 11 Del. C. § 1909.

Domestic Violence

Domestic violence assessment. In cases where the State contends that the defendant has committed domestic or intimate partner violence, the domestic violence assessment, when available, helps to measure the risk that the defendant may seriously injure or kill the alleged victim. This domestic violence assessment includes the following questions that a first responder is to ask the alleged victim at the time of the incident:

(i) Has the defendant ever used a weapon against you or threatened you with a weapon?

(ii) Has the defendant threatened to kill you or your children?

(iii) Do you think the defendant might try to kill you?

(iv) Does the defendant have a gun or can the defendant get one easily?

(v) Has the defendant ever tried to choke you?

(vi) Is the defendant violently or constantly jealous or does the defendant control most of your daily activities?

(vii) Have you left the defendant or separated after living together or being married?

(viii) Is the defendant unemployed?

(ix) Has the defendant ever tried to commit suicide?

(x) Do you have a child that the defendant knows is not his or hers?

(xi) Does the defendant follow or spy on you or leave threatening messages?

(xii) Is there anything else that worries you about your safety? If so, what worries you?

Trigger of referral protocol. Questions in the domestic violence assessment are grouped into two sets, plus a general catch-all question that serves as an override, the answers to which may trigger a referral protocol that requires the first responder to inform the alleged victim of the high danger assessment and offer the alleged victim the opportunity to be screened by a hotline counselor for assistance. The first set consists of questions (i) through (iii), the second set contains questions (iv) through (xi), and question (xii) is the general catch-all question. The referral protocol is triggered by either (i) an affirmative response to any one of the first set of questions; (ii) an affirmative response to at least four of the second set of questions; or (iii) a discretionary decision by the first responder, based on the alleged victim’s response to the catch-all question or any other reason, to trigger the referral protocol.

Conditions of Release for Failure to Appear

Although it is critical that there be incentives and consequences that address the possibiUty that a defendant will not appear for trial, the risk that the defendant will fail to appear at scheduled court proceedings is categorically different from the threat that a defendant will commit harm if released pending trial. For that reason, the rule discourages the use of monetary conditions where the only risk is that the defendant will fail to appear, and the court can require a secured bond only if the State makes a special showing and the court makes special findings that there is a demonstrated and specific flight risk in the current case. For example, if a defendant with a warrant for his or her arrest purchases a one-way ticket to a foreign country, those circumstances may justify the imposition of a secured bond to assure the defendant’s appearance in court. In addition, the court has the authority to make special findings of this kind on its own motion. In most cases, however, the court should not require a secured bond at first appearance where the main risk is that the defendant will fail to appear. Instead, the rule encourages the use of notification measures and other non-monetary condifions of release to assure the defendant’s appearance. If the defendant has already failed to appear in the current case, by contrast, the court may under Rule 5.4 impose more intensive conditions of release in its discretion without ordering a new pretrial assessment. When, despite being given a chance to have conditions of release not involving monetary terms, a defendant fails to appear, the defendant’s breach will subject him to the imposition of monetary conditions, in the discretion of the court, as is generally true when a defendant breaches conditions of release.

Conditions of Release for Risk to Public Safety

The most serious threat to be addressed by conditions of release is when the defendant poses a risk of harm to the general public or a specific person. Consistent with the seriousness of that threat, the pretrial assessment provides for a higher score based on the defendant’s risk to public safety, and the domestic violence assessment provides for a secured conditions of release bond where there is a heightened risk that the defendant may harm the alleged victim. In addition, the rule identifies certain serious offenses that signal a heightened risk to public safety. These signal offenses involve very serious offenses, such as homicides, crimes of violence, sex crimes, gun crimes, and high level drug offenses. A number of these signal offenses specifically involve domestic violence and therefore supplement the protections of the domestic violence assessment in addressing this important area. As to these signal offenses, the court has discretion to impose more stringent conditions of release than suggested by the initial recommended response, including monetary conditions. The rule also addresses the risks posed by recidivist impaired drivers, by providing for conditions of release tailored to the unique threats posed by these defendants.

In addition, the rule allows for the court to require more intensive conditions of release where the State makes a special showing and the court makes special findings that releasing the defendant with less intensive conditions of release would pose a substantial danger to public safety. Likewise, the court has the authority to make special findings of this kind on its own motion. Given the breadth of the signal offenses, the rule contemplates that these “residual” departures will be relatively rare.

Conditions of Release for Obstruction of Justice

The Bail Reform Act recognizes that some defendants pose the danger of undermining the rule of law by intimidating witnesses or taking other steps that obstruct justice and the ability of the judicial system to hold a fair trial. These situations are case-specific, not susceptible to being addressed by a pretrial assessment instrument, and therefore this rule sets forth standards by which the State can make a special showing and the court can make special findings to address when a defendant has taken specific actions that show this threat to be substantial and justify more intensive conditions of release than the initial recommended response. As with non-appearance and public safety risks, the court also has the authority to make special findings of this kind on its own motion.

Downward Departures from the Initial Recommended Response

The rule provides for procedures where the court wishes to depart downward from the initial recommended response. The court may impose less intensive conditions of release than the initial recommended response without making “special findings” so long as the court finds that less intensive conditions will adequately assure the defendant’s appearance at court proceedings, public safety, and that the defendant does not obstruct justice. For statistical tracking purposes, the court still must record its decision by selecting at least one judicial response code in the Delaware Criminal Justice Information System.

Limitations on Release to Pretrial Services

Consistent with the risk that overuse of supervision by Pretrial Services may cause many of the same harms as the overuse of money bail, and consistent with the presumption of innocence guaranteed by the United States and Delaware Constitutions, the rule discourages the court from confusing release to Pretrial Services with either a post-conviction sentence involving probation and parole or participation in a diversion program. Specifically, if the court releases the defendant to Pretrial Services, it must either impose the standard conditions of release used by Pretrial Services or first consult with the Department of Correction. In addition, the rule discourages the use of types of conditions of release that are either suspect in the context of pretrial release or resource intensive: alcohol or drug testing, curfews, and electronic monitoring.

Consideration of the Defendant’s Financial Circumstances

Due to concerns about creating disparities based on race, gender, or wealth, the rule prohibits giving weight to the defendant’s financial circumstances in setting the conditions of release except by making special findings. Put plainly, in initially determining whether to impose monetary conditions of release, a defendant’s wealth should not bias that decision. But, when the court decides to impose monetary conditions of release, equity requires the court to consider the defendant’s wealth and ability to satisfy those monetary conditions of release. The rule therefore directs the court to provide defendants with a form to provide financial information so that the conditions of release can fully take into account the defendant’s means in setting appropriate monetary conditions of release.

Sua Sponte Departures from the Initial Recommended Response

As previously noted, the court may depart downward from the initial recommended response on its own initiative. And as noted, the rule also allows for the court to depart upward on its own initiative, provided that the court explains its reasons for doing so (and that the court makes the same special findings that it must make when the State requests more intensive conditions of release). But consistent with the defendant’s right to seek review of her conditions of release, the court must offer the defendant an opportunity to contest that upward departure.