251 A.3d 724
D.C.
2021

GEORGE L. BAILEY, Appellant, v. UNITED STATES, Appellee.

Nos. 21-CO-26 & 21-CO-27

District of Columbia Court of Appeals

JUN 3 2021

BEFORE: GLICKMAN and DEAHL, Associate Judges, and NEBEKER, Senior Judge.

1984 FEL 3542, 1995 FEL 1484

PUBLISHED ORDER

This is an appeal from a denial of compassionate release under D.C. Code § 24-403.04(a) (2021 Supp.). The compassionate release statute, passed amidst the ongoing coronavirus pandemic, permits prisoners to seek modification of their terms of imprisonment if they meet certain eligibility criteria and the trial court “determines [they are] not a danger to the safety of any other person or the community.” D.C. Code § 24-403.04(a). Appellant George Bailey—who is currently serving a 95-year sentence for his participation in three murders during a “crack-house” raid1—sought a modification of his prison term to permit his early release under that statute. While the government agreed that he met the eligibility criteria for compassionate release, it maintained he failed to establish that he was non-dangerous. The trial court agreed and denied his motion for release.

Bailey appealed and now moves for summary reversal. The government cross-moves for summary affirmance. We grant Bailey‘s motion for summary reversal, deny the motion for summary affirmance, and remand the case for reconsideration of the compassionate release petition in light of the principles articulated below. We take the further step of publishing this order to provide guidance on two legal issues raised in this appeal that are of recurring concern in the compassionate release context and which we have not resolved to date.

I. The Prisoner Bears the Burden of Proving Non-Dangerousness by a Preponderance of the Evidence

The first question is who bears the burden on whether a prisoner is non-dangerous and what, precisely, that burden is. The statute is silent on those points. While Bailey acknowledges that it is his burden to establish his non-dangerousness, he contends he need do so only by a preponderance of the evidence, while further arguing the trial court erroneously held him to a higher standard of proof. The government does not dispute that a prisoner need only demonstrate their non-dangerousness by a preponderance of the evidence, but neither does it expressly concede the point, instead assuming it only arguendo. We conclude that it is the prisoner‘s burden to establish they are non-dangerous by a preponderance of the evidence. We reverse because it appears from the record before us that the trial court may have applied a higher standard of proof in denying Bailey‘s compassionate release motion.

The preponderance-of-the-evidence standard applies here for two main reasons. First, the preponderance standard is the “default rule.” CIGNA Corp. v. Amara, 563 U.S. 421, 444 (2011).2 Where no standard is specified in the statute and due process does not compel a different result, it ordinarily applies. See id.; Raphael v. Okyiri, 740 A.2d 935, 957 (D.C. 1999) (“[A] party with the burden of persuasion on an issue must ordinarily establish the relevant facts by a preponderance of the evidence[]” and “[e]xceptions to this standard are uncommon“) (citing Price Waterhouse v. Hopkins, 490 U.S. 228, 253 (1989)). Elsewhere when the Council has intended a higher standard of proof to apply to dangerousness determinations affecting release decisions, it has clearly said so. See, e.g., D.C. Code § 23-1325(b)–(c) (2012 Repl.) (specifying clear and convincing standard for release pending sentencing and appeal); D.C. Code § 23-1322(b)(2) (authorizing pretrial detention where court finds “by clear and convincing evidence that no condition or combination of conditions will reasonably assure . . . the safety of any other person and the community“). Its failure to expressly deviate from this default rule here is strong reason to adhere to it. See generally 1618 Twenty-First Street Tenants’ Ass‘n v. The Phillips Collection, 829 A.2d 201, 206 (D.C. 2003) (Council‘s failure to give “bona fide” any “special meaning” in statutory provision contrasts with other provisions indicating “the Council knew how to give ‘bona fide’ a special meaning when it deemed necessary“).

Second, federal courts that have addressed the issue under the federal compassionate release statute have—it appears uniformly—held that the preponderance standard applies. See, e.g., United States v. Grasha, 489 F. Supp. 3d 403, 406 (W.D. Pa. 2020); United States v. Shryock, No. CR 95-345-RSWL-16, 2020 WL 7773887, at *1 (C.D. Cal. Dec. 30, 2020); United States v. Kazanowski, No. 15-cr-00459-DKW-5, 2020 WL 3578310, at *3 (D. Haw. July 1, 2020). Because the District‘s compassionate release statute is “modeled after [the] federal” one,3 and is intended to “align” with the use of federal compassionate release following the First Step Act of 2018,4 this federal guidance is highly persuasive. See Minor v. United States, 623 A.2d 1182, 1186 n.8 (D.C. 1993) (“We may look to decisions interpreting an identical, or substantially identical, federal statute as persuasive authority in interpreting its local counterpart.“).

Having now clarified the pertinent standard, we are concerned that the trial court may have erroneously imposed a heightened standard of proof in assessing whether Bailey established his non-dangerousness. Although, as the government stresses, the trial court considered whether “on balance” certain factors weighed for or against granting relief, the trial court‘s order does not say or indicate it applied the preponderance standard to the ultimate issue of dangerousness. Instead, the order employs language arguably implying otherwise in the absence of a clear statement to that effect. For instance, the court stated that it “must be confident” in Bailey‘s non-dangerousness before sanctioning his release, that it must “ensure that the record” demonstrated Bailey‘s non-dangerousness, and that it must have “sufficient comfort” in that conclusion before ordering his release. Those articulations are in serious tension with a preponderance standard, and that is enough to justify a remand for the trial court to clarify its ruling now that we have made the appropriate standard clear.

The government counters that, while perhaps ambiguous or inartful, these various articulations are not plainly inconsistent with a preponderance standard, and urges us to interpret the trial court‘s order as employing a preponderance standard based on the principle that “[t]rial judges are presumed to know the law.” Saidi v. United States, 110 A.3d 606, 613 (D.C. 2015). We are sympathetic to the point, particularly here where the trial court exhibited a great deal of care and thoughtfulness throughout its 15-page order. If this court had ever previously held that the preponderance standard applied in this context, or if the trial court had expressly acknowledged it as controlling its determination, we would surely excuse some apparent deviations from that standard as mere imprecisions. But that is not the backdrop here: (1) the statute does not articulate a clear standard; (2) before today, neither had we in this precise context; and (3) the trial court never plainly articulated a preponderance standard as guiding its determination.5 Indeed, it is hard to reconcile the government urging us to assume the trial court applied the correct preponderance standard while at the same time it remains agnostic as to whether that is in fact the appropriate standard. It would like us to presume—in the face of substantial contrary evidence—that the trial court rigorously followed a principle that the government itself is apparently unprepared to acknowledge exists. The very uncertainty of the principle, until today, precludes us from presuming it was followed.

The government has not asked us to find any error on this score harmless and so we vacate the trial court‘s order and remand the case for reconsideration. In doing so, we reject Bailey‘s argument that the evidence is so one-sided as to compel the conclusion that he is not dangerous as a matter of law. See Johnson v. United States, 398 A.2d 354, 364 (D.C. 1979) (recognizing that “the facts may leave the trial court with but one option it may choose without abusing its discretion, all the others having been ruled out[]“); Wright v. United States, 508 A.2d 915, 920 (D.C. 1986) (indicating that the possibility Johnson recognizes is “conceivably” true but “rare” in practice). The record here does not legally compel a finding that Bailey is not dangerous. On remand the trial court should expressly determine whether appellant has demonstrated by a preponderance of the evidence that, if released, he will not be a danger to the safety of any other person or the community.

II. Not All §§ 3142(g) and 3553(a) Factors Are Relevant to Dangerousness

The second dispute concerns the interplay between the compassionate release statute and its cross-references: 18 U.S.C. §§ 3142(g) and 3553(a), listing factors pertinent to pre-trial detention and to sentencing, respectively. Specifically, Bailey contends the trial court erred when it considered statements of the deceased victims’ family members—expressing their desire for Bailey to remain imprisoned—as he maintains those impact statements were not relevant to his dangerousness. In his view, while the compassionate release statute incorporates the 18 U.S.C. §§ 3142(g) and 3553(a) factors, a trial court can consider those factors only subject to a limiting principle that they must bear on the prisoner‘s present or future dangerousness. The government disagrees that there is any such limiting principle and, even if there were, it maintains the trial court complied with it here.6 In its view, victim impact statements are a relevant consideration under 18 U.S.C. § 3553(a)(1)—requiring courts to consider “the nature and circumstances of the offense“—and because those factors are expressly incorporated into the compassionate release statute they are necessarily relevant to this inquiry.

We agree with Bailey on the principle that the §§ 3142(g) and 3553(a) factors inform compassionate release decisions only insofar as they are relevant to a prisoner‘s present or future dangerousness; not all of them are relevant to that inquiry. However, we agree with the government that victim impact statements can inform that dangerousness assessment. We detect no error in the trial court‘s limited consideration of victim impact statements in this case.

To parse the competing views, we begin with the text of the compassionate release statute:

[T]he court shall modify a term of imprisonment imposed upon a defendant if it determines the defendant is not a danger to the safety of any other person or the community, pursuant to the factors to be considered in 18 U.S.C. §§ 3142(g) and 3553(a) and evidence of the defendant‘s rehabilitation while incarcerated.

D.C. Code § 24-403.04(a) (emphases added).

In Bailey‘s view, the first italicized phrase limits the second. While the compassionate release statute instructs a court to consult the §§ 3142(g) and 3553(a) factors, that consultation is with a limited purpose: assessing the prisoner‘s dangerousness. The government counters that § 24-403.04(a) contains no express limitation on which of the §§ 3142(g) and 3553(a) factors a trial court may consider, and it further stresses that contrasts with the federal statute, which provides that courts should consider “the factors set forth in section 3553(a) to the extent that they are applicable.” 18 U.S.C. § 3582(c)(1)(A) (emphasis added). Because our statute contains no such limitation, the government argues its omission demonstrates the §§ 3142(g) and 3553(a) factors are relevant in toto to compassionate release decisions. The government reasons that to hold otherwise would render the § 3553(a) factors all but irrelevant to compassionate release decisions, contrary to the D.C. Council‘s express intent that they be considered.

We agree with Bailey that the §§ 3142(g) and 3553(a) factors are relevant to compassionate release decisions only insofar as they inform the determination of a prisoner‘s present or future dangerousness. That is plainly set forth in the statute as the baseline determination that a trial court must make—determining whether “the defendant is not a danger“—a phrase which limits the proper consideration of statutory factors.

Certain factors in §§ 3142(g) and 3553(a) clearly have no relevance to present or future dangerousness and therefore cannot be given any weight in a compassionate release decision. Take, for example, § 3142(g)(4)‘s “inquiry into the source of the property . . . as collateral to secure a bond” and whether it will “reasonably assure the appearance of the” defendant. While it is easy to see the relevance of that inquiry to a pretrial detention determination—sufficient collateral may provide assurances that a defendant will return for their court dates—the inquiry is irrelevant to a determination of dangerousness. There is not even a mechanism for prisoners to post bonds as collateral in exchange for compassionate release, making this § 3142(g)(4) consideration plainly inapposite. Other examples include the need for general deterrence in § 3553(a)(2)(B),7 and “the need to provide restitution to any victims” under § 3553(a)(7). While those factors are relevant in the initial sentencing determination outlined by § 3553, neither is relevant to a later assessment of the prisoner‘s dangerousness if released. Where a defendant is eligible for early release and found to be non-dangerous, there is simply no room in the statutory scheme for concerns about general deterrence and victim restitution to trump those determinations.

Those factors instead risk derailing the dangerousness inquiry. A trial court that determines a prisoner is both eligible for early release and non-dangerous—yet believes early release would provide inadequate general deterrence in light of the need to prevent others from committing the same crime—is not free to deny compassionate release. The statute provides a clear instruction in that instance: the court “shall modify [the] term of imprisonment imposed upon a defendant if it determines the defendant is not a danger” and is eligible for modification. D.C. Code § 24.403.04(a) (emphasis added). General deterrence is beside the point to the statutory calculus. We note this is a recent change to the statute—one made since the trial court issued its order in this case—as the previous iteration provided only that a trial court may modify a sentence upon a finding of dangerousness or eligibility. See D.C. Law 23-274, § 1203(b) (Apr. 27, 2021) (permanent legislation); D.C. Act 24-30, § 706(b)(2) (Mar. 17, 2021) (emergency legislation). We do not address whether, under that prior iteration of the statute, courts retained some discretion to deny compassionate release where they found a prisoner eligible and non-dangerous, as use of the word “may” would suggest. The trial court here did not purport to exercise such discretion; it simply found that Bailey had not demonstrated his non-dangerousness. Nor does the government rest any part of its argument on that discretionary hook. Instead, the government defends the trial court‘s ruling on the basis that all §§ 3142(g) and 3553(a) factors are necessarily relevant to dangerousness under the compassionate release statute, a proposition we reject.

Still, we discern no error here. The trial court, in denying compassionate release, made fleeting reference to two of the deceased victims’ family members’ wishes that Bailey not be released. Bailey argues that was error because those wishes are inapposite to his dangerousness. Not so. When considering a prisoner‘s dangerousness, it is appropriate to weigh “the nature and circumstances” of their underlying offense(s), 18 U.S.C. § 3553(a)(1), as informing their present and future dangerousness. “What‘s past is prologue.”8 The nature of a prisoner‘s past crimes—including the lasting harms those crimes have inflicted on victims and their family members and the fears those persons may continue to harbor—have predictive value in informing the degree of harm the prisoner might inflict if they reoffend. To be sure, a court ought to be careful not to become untethered from the ultimate inquiry into a prisoner‘s dangerousness when considering this § 3553(a)(1) factor. A court would run afoul of the statute if it found a prisoner eligible and non-dangerous and—for any reason, including concern for the victims’ wants—did not modify their sentence. But that is not what occurred here. The trial court did not prioritize these victims’ desires over the statutory command that Bailey be released if found to be non-dangerous. Instead, it briefly noted those desires as evidence of the “far reaching and still ever-present consequences” Bailey‘s crimes inflicted upon his victims. That comment on the impact of Bailey‘s past crimes did not stray from concerns related to Bailey‘s present and future dangerousness.

We do not attempt a comprehensive listing of which factors in §§ 3142(g) and 3553(a) are and are not relevant to the issue of dangerousness. While we have identified a few that we think are plainly inapposite, relevance will often depend on

context. Take § 3553(a)(1)‘s concern for the nature and circumstances of the underlying offense, which will often (we think rightly) play a prominent role in the compassionate release inquiry. A trial court might make both proper and improper use of that factor; the demarcation line between the two is whether it remains tethered to concerns of present and future dangerousness. If a trial court homes in on the gravity of past offenses as informing the likelihood and consequences of a potential re-offense, that remains consistent with the statute‘s overarching concern with dangerousness. If, however, a trial court weighs the severity of past offenses in a manner that is divorced from dangerousness—for example, if it finds the prisoner non-dangerous but undeserving of early release in light of past crimes9—then it has strayed from the statutory mandate. What matters is not some abstract categorization of the §§ 3142(g) and 3553(a) factors into buckets of those relevant and irrelevant to dangerousness; it is instead whether the trial court remains moored to the baseline determination of dangerousness. Here, we think the trial court‘s nod to the enduring harm that Bailey‘s crimes have had on his victims’ families did not tread over that line.

III. Conclusion

On consideration of appellant‘s motion for summary reversal, appellee‘s opposition and cross-motion for summary affirmance, appellant‘s unopposed motion for leave to late-file his lodged reply and opposition, appellee‘s lodged reply, and the records on appeal, it is

ORDERED that appellant‘s motion for leave is granted, and the lodged pleadings are hereby filed. It is

FURTHER ORDERED that appellant‘s motion for summary reversal is granted and appellee‘s cross-motion for summary affirmance is denied. See generally Watson v. United States, 73 A.3d 130, 131 (D.C. 2013) (citing Oliver T. Carr Mgmt., Inc. v. Nat‘l Delicatessen, Inc., 397 A.2d 914, 915 (D.C. 1979)). It is

FURTHER ORDERED and ADJUDGED that the order on appeal is reversed and the matter remanded to the trial court for further consideration of appellant‘s compassionate release motion consistent with this order. It is

FURTHER ORDERED that the mandate shall issue forthwith. See D.C. App. R. 41(b).

PER CURIAM.

Copies emailed:

Honorable James A. Crowell IV

Honorable Juliet J. McKenna

Director, Criminal Division

Copies e-served:

Samia Fam, Esquire

Public Defender Service

Elizabeth Trosman, Esquire

Assistant United States Attorney

cml

Notes

2
While CIGNA concerned a civil statute and made clear that a preponderance standard is the default rule in civil cases, its rationale applies to criminal cases when, like here, the legislature has placed an evidentiary burden on the defendant rather than the government. In those circumstances there are no due process concerns counseling in favor of a heightened standard of proof; because the defendant bears the burden on the question, any due process concerns would only further counsel against a heightened evidentiary standard. Cf. Walton v. Arizona, 497 U.S. 639, 650 (1990) (holding that a defendant‘s due process rights “are not violated by placing on him the burden of proving mitigating circumstances sufficiently substantial to call for leniency“), overruled on other grounds by Ring v. Arizona, 536 U.S. 584, 609 (2002).
3
COVID-19 Response Supplemental Emergency Amendment Act of 2020: Hearing on B23-733, D.C. Council 27th Legis. Meeting (Apr. 7, 2020) (statement of Councilmember Charles Allen).
4
See D.C. Res. 23-399, COVID-19 Response Supplemental Emergency Declaration Resolution of 2020, § 2(f) (Apr. 7, 2020).
5
We acknowledge that “on balance” resembles a preponderance standard, but given the court‘s other statements invoking heightened standards, we think it insufficiently clear to carry the load the government seeks to place on it.
6
The government also contends that Bailey abandoned this argument in the trial court. We do not reach that question. Because this is a recurring issue in compassionate release appeals we think it prudent to address the merits first and, because we detect no reversible error, we do not reach the abandonment question.
7
While § 3553(a)(2)(B) speaks only in terms of “adequate deterrence“—without distinguishing between general and specific deterrence—it is clear that it reflects a concern with general deterrence, particularly in light of § 3553(a)(2)(C)‘s more focused concern with specific deterrence. See United States v. Russell, 600 F.3d 631, 637 (D.C. Cir. 2010) (“A condition that takes into account only the magnitude of the harm from defendant‘s offense may serve the general deterrence goals of § 3553(a)(2)(B), but may be more or less severe than required to serve § 3553(a)(2)(C)‘s goal of preventing this defendant from repeating his offense.“).
8
WILLIAM SHAKESPEARE, THE TEMPEST, act 2, sc. 1.
9
The trial court appeared to make improper use of the § 3553(a) factors in Griffin v. United States, No. 20-CO-769 (D.C. June 3, 2021)—also issued today in a published order accompanying this one—when it denied compassionate release, in part, because the result would be that “Defendant‘s sentence would not adequately reflect the seriousness of the offense, promote respect for the law, nor provide just punishment.” Any consideration of those factors must be tied to an assessment of the prisoner‘s present dangerousness. For instance, a trial court might gauge the appropriateness of the original sentence as a rough benchmark reflecting society‘s assessment of the prisoner‘s dangerousness if released “prematurely“; but it may not stray into considerations unrelated to dangerousness, like whether early release would diminish society‘s interests in general deterrence and respect for law. We affirmed in Griffin because the trial court‘s consideration of § 3553(a) factors unrelated to dangerousness had no effect on its decision, id. at 2, but highlight it as an example of the misapplication of the nature and circumstances of the prisoner‘s past offenses in the inquiry under the compassionate release statute.
Read the detailed case summary