Davidson A/K/A Muhammad v. United States

Notice: This opinion is subject to formal revision before publication in the Atlantic and Maryland Reporters. Users are requested to notify the Clerk of the Court of any formal errors so that corrections may be made before the bound volumes go to press.

DISTRICT OF COLUMBIA COURT OF APPEALS

Nos. 24-CO-295 & 24-CO-296

WILLIAM D. DAVIDSON, A.K.A. MUHAYMIN MUHAMMAD, APPELLANT,

V.

UNITED STATES, APPELLEE.

Appeal from the Superior Court of the District of Columbia (1982-FEL-006288 & 1982-FEL-007489)

(Hon. Craig Iscoe, Motions Judge) (Argued September 23, 2025 Decided January 22, 2026)

Areeba Jibril, Public Defender Service, with whom Mikel-Meredith Weidman and Jaclyn Frankfurt, Public Defender Service, were on the briefs, for appellant.

Katelyn B. Benton, Assistant United States Attorney, with whom Edward R. Martin, Jr., United States Attorney at the time the brief was filed, and Chrisellen R. Kolb and Christopher Macomber, Assistant United States Attorneys, were on the brief, for appellee.

Before BLACKBURNE-RIGSBY, Chief Judge, and EASTERLY and SHANKER, Associate Judges.

BLACKBURNE-RIGSBY, Chief Judge: Appellant Muhaymin Shabazz Muhammad 1 appeals the trial court’s denial of his motion to reduce his sentence under the Incarceration Reduction Amendment Act (IRRA), D.C. Code § 24-403.03, which he filed on July 7, 2023. Mr. Muhammad has served forty-three years of his sixty-five years to life sentence for a series of violent crimes he committed over a four-week period in 1982, including first-degree murder, rape, and armed robbery, when Mr. Muhammad was eighteen years old. Mr. Muhammad challenges the trial court’s finding of his current dangerousness under the IRAA, particularly its reliance on his failure to complete sex offender treatment, which the Bureau of Prisons (BOP) never determined he needed, and the court’s reliance on his 2016 infraction for “stalking” two female prison employees. Additionally, Mr. Muhammad contends that the trial court erred in holding him to an elevated burden of proof to “guarantee” his non-dangerousness as opposed to the IRAA’s preponderance of the evidence standard. 2 For the reasons discussed below, we vacate the trial court’s decision and remand for further proceedings consistent with this opinion.

1 Mr. Muhammad changed his name from William Davidson after his religious conversion. We refer to him by his current name. 2 Mr. Muhammad also argues that the trial court abused its discretion when it relied on factors set out in the D.C. Voluntary Sentencing Guidelines, including the nature and seriousness of the underlying offenses and principles of deterrence, in its interests of justice analysis. At oral argument, the government conceded that the

I. Underlying Offenses and Incarceration History

Mr. Muhammad’s actions in October and November of 1982 resulted in two cases: (1) 1982 FEL 006288, involving the first-degree murder while armed of David Ball on October 16, 1982, and (2) 1982 FEL 007489, involving five counts arising from offenses committed against five victims between October 31 and November 8, 1982.3 Mr. Muhammad entered a global guilty plea resolving these two cases, which included an Alford plea to the charge of manslaughter for the shooting and killing of David Ball and guilty pleas for one count of first-degree murder, two counts of rape, and one count of armed robbery. Mr. Muhammad contends that his criminal behavior was caused by his drug use, particularly his use of PCP, noting that he had no history of involvement in the criminal legal system prior to his arrest. Mr. Muhammad received an aggregate sentence of sixty-five years to life.

During his forty-three years in prison, Mr. Muhammad has completed more than 1,600 hours of educational programming, most notably drug education courses, a drug abuse treatment program, and Narcotics Anonymous and Alcoholics Anonymous classes. Mr. Muhammad was not referred to nor did he take any sex

trial court erred in its analysis based on our opinions in Doe v. United States, 333 A.3d 893 (D.C. 2025), and Riley v. United States, 338 A.3d 1 (D.C. 2025). 3 In 1982 FEL 007489, Mr. Muhammad was indicted on forty-five different counts for his involvement in over ten separate criminal events, including kidnapping, rape, and robbery, but those were dropped as part of his plea deal.

4 offender treatment courses; however, he was placed on a waitlist for sexual rehabilitative training in December of 2015—although the record is unclear as to why or how. 4 In addition to personal development courses, Mr. Muhammad also obtained his GED and took business-related and financial courses, while maintaining employment, at various points, in food services, mechanical work, paint shop work, laundry services, and barbering. Mr. Muhammad also “found his faith” and built community within his faith at a number of facilities. Using the Prisoner Assessment Tool Targeting Estimated Risks and Needs (PATTERN), BOP determined that Mr. Muhammad presented a “minimum” level of dangerous recidivism as of 2023.

During his forty-three years of incarceration, Mr. Muhammad has accrued twenty-seven disciplinary infractions, the most recent of which was in 2017. Only one of these infractions was for violence—a Level 200 fighting infraction from 1999—which Mr. Muhammad contends was for action he took in self-defense. He

4 At the evidentiary hearing, Mr. Muhammad’s counsel stated that “[Mr. Muhammad] has put himself on waitlist and has remained on that waitlist,” but she did not explain the circumstances surrounding this belated decision to seek sex offender treatment. According to Mr. Muhammad’s presentence report, he was recommended to “be under the care of a psychologist for what is clearly disturbance in his behavior that merits professional care,” and his treatment plan included that he “undergo psychological counseling” in addition to drug abuse therapy. It is unclear if Mr. Muhammad ever received this recommended psychological counseling, as it is not mentioned in his initial IRAA motion, his brief, or elsewhere in the record.

5 has never received a Level 100 infraction, which is the highest level of sanction in the BOP.

One infraction, issued December 12, 2016, for “stalking,” involved Mr. Muhammad’s behavior towards a female Cook Supervisor from September 7 to October 13, 2016, during which time he filed five reports accusing her of stealing money, not preparing his meals properly, and lying to him, while calling her derogatory names. These reports caused the employee to feel “very upset” and that “her safety would be in danger if [Mr. Muhammad] is released to general population,” as characterized in the investigating officer’s report. This infraction followed an earlier incident, in which Mr. Muhammad received “verbal counseling” for “inappropriate emails” that he sent to a female Reentry Affairs Coordinator, beginning on May 17, 2016, addressing her as “Dear Beloved” and informing her how she needed to do her job. After receiving the Coordinator’s response to these emails, Mr. Muhammad complained to another staff member, stating: “I have two life sentences, I don’t give a damn about slapping an officer or what’s going to happen to me.” 5 Following this incident, Mr. Muhammad received verbal

5 On a later page in the DHO Report, this line is instead quoted as: “I have two life sentences. I can give a damn about slapping an officer or what’s going to happen to me.” As the trial court and parties consistently rely on the “I don’t give a damn” version, we do so here as well.

6 counseling for “behavior exhibited around female staff which harasses, alarms, and annoys staff.”

The Discipline Hearing Officer (DHO) Report, which followed both the verbal counseling and the formal infraction for Mr. Muhammad’s harassment of the Coordinator and the Cook Supervisor, respectively, concluded that “there is sufficient evidence of a threat to the safety of staff . . . based on [Mr. Muhammad’s] extensive history of taking hostages, rape and threatening staff, specifically female staff.” The report also referenced Mr. Muhammad being placed in the Special Housing Unit “for his constant behavior towards female staff.”

II. IRAA Motion Hearing

The trial court held an evidentiary hearing on Mr. Muhammad’s motion on December 19, 2023. At the hearing, Mr. Muhammad sought to emphasize his maturity, rehabilitation, and lack of dangerousness by referencing his long length of imprisonment, his young age at the time of his offenses, his record of nonviolence while incarcerated, his decades of sobriety, his faith, and his remorse for the offenses he committed at age eighteen. He also pointed to his “minimum” PATTERN score and a letter from an expert psychologist, Dr. Shauna Keller, confirming Mr. Muhammad’s low risk of sexual recidivism. Mr. Muhammad connected the

7 educational programming and UNICOR employment programming he completed in prison with his hopes of working as a barber following his release.

In discussing the 2016 “stalking” infraction during the hearing, Mr. Muhammad’s counsel took issue with the trial court “cherry picking” Mr. Muhammad’s conduct to find a connection between those incidents and his underlying offenses, clarifying that there was nothing sexual in the “stalking” incidents with staff. 6 Moreover, counsel stated that the stalking infraction was the culmination of Mr. Muhammad filing a number of BP-29s, which are grievances towards staff, and that “[t]hey happen to be women in this case . . . however, there’s really no indication that it was because they were women or that they were targeted other than that they happened to be in particular roles where [he] was having issues or perceived injustices with them doing their job.” Counsel also noted that “female staff in the BOP represent about 30 percent of the population” and indicated that these situations were not the exclusive times that Mr. Muhammad had worked with staff in the BOP, including women. Additionally, counsel reiterated that Mr. Muhammad had received positive reviews from other staff, including his counselor

6 Mr. Muhammad also addressed the trial court’s concerns about a separate infraction for making a sexual proposal to staff, indicating that the accusation was misattributed to Mr. Muhammad, as confirmed by the reporting officer. The government did not contest this assertion.

8 at a recent facility who indicated that Mr. Muhammad was “someone that [the staff] turned to to help resolve conflicts.”

Mr. Muhammad’s counsel argued that sex offender treatment is not appropriate for all persons convicted of decades-old sex offenses and referenced as support Dr. Keller’s conclusion in her report that other types of programming were more beneficial for his current situation. Counsel also pointed out that Mr. Muhammad’s January 2023 individualized plan, created by BOP, did not include sex offender treatment programming as a goal.

Mr. Muhammad’s counsel also presented his reentry plan, which included temporary and transitional housing opportunities in the District for at least a year and a barbering program that he planned to take, as well as other vocational job training opportunities. Mr. Muhammad was present but did not testify at the hearing. Instead, his counsel read extensively from a letter that Mr. Muhammad wrote to the court, in which he stated his desire to have a positive impact on his community and engage with youth in some unspecified way.

The trial court denied Mr. Muhammad’s IRAA motion in a written order, finding that Mr. Muhammad had not satisfied his burden of demonstrating that he is no longer a danger to the community and that the interests of justice warrant a sentence reduction.

III. Discussion

A. Standard of Review

“This court ‘review[s] the denial of an IRAA motion for abuse of discretion, but consider[s] questions of statutory construction de novo.”’ Doe v. United States, 333 A.3d 893, 898 (D.C. 2025) (alterations in original) (citing Bishop v. United States, 310 A.3d 629, 641 (D.C. 2024)). In reviewing for abuse of discretion, we “must determine whether the decision maker failed to consider a relevant factor, whether [the decision maker] relied upon an improper factor, and whether the reasons given reasonably support the conclusion.” Id. (citation modified). “The abuse-of-discretion standard includes review—generally described as de novo review—to determine that the discretion was not guided by erroneous legal conclusions.” Id. (citation modified) (quoting Welch v. United States, 319 A.3d 971, 975 (D.C. 2024)).

B. IRAA Statutory Structure

The D.C. Council enacted the IRAA in 2016 to “ensur[e] that all juvenile offenders serving lengthy prison terms have a realistic, meaningful opportunity to obtain release based on their diminished culpability and their maturation and rehabilitation.” Doe, 333 A.3d at 898 (quoting Williams v. United States, 205 A.3d

10 837, 846 (D.C. 2019)). For an IRAA motion, “[t]he burden of proof is on the movant.” See Doe, 333 A.3d at 900 (quoting Bishop, 310 A.3d at 636). Pursuant to the IRAA, a trial court “shall reduce a term of imprisonment imposed upon a defendant for an offense committed before the defendant’s 25th birthday” if: (1) certain threshold requirements, including the length of time served, are met, and (2) after considering eleven factors, the court finds that “the defendant is not a danger to the safety of any person or the community and that the interests of justice warrant a sentence modification.” D.C. Code § 24-403.03(a); see Doe, 333 A.3d at 899. The eleven factors that the court “shall consider” are:

(1) The defendant’s age at the time of the offense;

(2) The history and characteristics of the defendant;

(3) Whether the defendant has substantially complied with the rules of the institution to which the defendant has been confined, and whether the defendant has completed any educational, vocational, or other program, where available;

(4) Any report or recommendation received from the United States Attorney;

(5) Whether the defendant has demonstrated maturity, rehabilitation, and a fitness to reenter society sufficient to justify a sentence reduction;

(6) Any statement, provided orally or in writing, provided pursuant to § 23-1904 or 18 U.S.C. § 3771 by a victim of the offense for which the defendant is imprisoned, or by a family member of the victim if the victim is deceased;

(7) Any reports of physical, mental, or psychiatric examinations of the defendant conducted by licensed health care professionals;

(8) The defendant’s family and community circumstances at the time of the offense, including any history of abuse, trauma, or involvement in the child welfare system;

(9) The extent of the defendant’s role in the offense and whether and to what extent another person was involved in the offense;

(10) The diminished culpability of juveniles and persons under age 25, as compared to that of older adults, and the hallmark features of youth, including immaturity, impetuosity, and failure to appreciate risks and consequences, which counsel against sentencing them to lengthy terms in prison, despite the brutality or cold- blooded nature of any particular crime, and the defendant’s personal circumstances that support an aging out of crime; and

(11) Any other information the court deems relevant to its decision.

D.C. Code § 24-403.03(c).

Notably, the Council amended the IRAA in 2019 to remove “the nature of the offense” as a standalone consideration under factor two; however, the Council emphasized that courts still “consider the facts and circumstances surrounding the underlying offense through their review of the various factors and evidence.” Report on Bill No. 23-0127 before the Committee on the Judiciary and Public Safety, Council of the District of Columbia, at 18-19 (Nov. 23, 2020). This court discussed

12 the implications of this removal in Doe, which we note was decided after the trial court’s decision in this case. 333 A.3d at 900, 906-11.

C. Lack of Sex Offender Treatment

Mr. Muhammad raises three arguments as to how he believes the trial court abused its discretion. We address each in turn. First, Mr. Muhammad argues that the trial court erred in finding that he was still dangerous because it based its ruling, in large part, on his lack of sex offender treatment and discounted Dr. Keller’s expert opinion and other evidence indicating he did not require sex offender treatment. We agree that the trial court’s underlying reliance on Mr. Muhammad’s lack of sex offender treatment in its evaluation of his current dangerousness was not given the explanation it requires. We hold that, as the record stands, with only the unchallenged expert testimony of Dr. Keller, indicating that he did not need sex offender treatment, but that he instead benefitted from other treatment he received, there was an insufficient basis for the trial court to find that Mr. Muhammad needed sex offender treatment in order to prove his non-dangerousness under the IRAA factors.

The trial court focused its inquiry on whether Mr. Muhammad had sufficiently shown that he had sought sex offender treatment or, in the alternative, whether he had sufficiently explained why it was not available to him at any point in his

13 incarceration. Notwithstanding the lack of evidence in the record, the trial court concluded that Mr. Muhammad “provided no evidence that, until quite recently, he ever sought to enroll in a sex offender program.” 7

Mr. Muhammad relies on Bishop for the proposition that a lack of programming opportunities might excuse programming shortcomings, and “mandate[] a different calculus”—referring to (c)(3)’s “where available” language. Bishop, 310 A.3d at 642, n.7. Mr. Muhammad contends that a lack of evidence as to whether the programming was unavailable to him from 1982 to 2015 should be read in his favor that it was not available, just as it was not available from 2015 to the present. The government, meanwhile, argues that a lack of evidence of its unavailability should be read in favor of its presumed availability. Mr. Muhammad further asserts that his “pre-2015 state of mind about sex offender treatment was not relevant to [his] current dangerousness.” 8 Rather than addressing the factual question of availability, the trial court found that both (c)(3) and (c)(5) weighed

7 The trial court did not seek to enlarge the record on this point, which it had the ability to do under Section 24-403.03(b)(2), at its discretion. Such enlargement may well have provided additional information indicating that Mr. Muhammad needed sex offender treatment and that it was made available to him, but, in its absence, we cannot speculate. 8 Mr. Muhammad is quoted in Dr. Keller’s report as professing that he “do[es]n’t need [mental health] treatment,” while claiming that he “could benefit from whatever [program] they have to offer.”

14 against release because Mr. Muhammad could “show[] no effort to enroll in sex-offender treatment from 1982 to 2015, shortly before he was charged with stalking.”

Although Mr. Muhammad also argued that the question of the treatment’s availability and his own efforts to enroll in it were essentially moot because he did not require sex offender treatment in the first place, the trial court was unpersuaded by Mr. Muhammad’s BOP individualized programming plan, PATTERN score, and Dr. Keller’s report to the extent that they indicated that sex offender treatment was not necessary for his rehabilitation under (c)(3) or that they demonstrated his maturity, rehabilitation, and fitness to reenter society, under (c)(5).

“[T]he trial court [i]s under no obligation to accept [an expert’s] testimony.” Henny v. United States, 321 A.3d 621, 631 (D.C. 2024) (citing United States v. McNeil, 434 F.2d 502, 504 (D.C. Cir. 1970) (“The weight to be given any expert opinion admitted in evidence . . . is exclusively for the [factfinder].”)). And “[w]here there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous.” Johnson v. United States, 232 A.3d 156, 168 (D.C. 2020) (citing Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 574 (1985)). But as in Henny, the trial court’s disinclination to find that Dr. Keller’s assessment sufficiently showed that Mr. Muhammad did not require sex offender treatment reflects a “failure [by the court] to make sufficient findings . . . on the

15 weight given to an expert’s testimony.” Henny, 321 A.3d at 631 (internal quotation marks omitted) (quoting A.C. v. N.W., 160 A.3d 509, 518 n.13 (D.C. 2017)).

The trial court acknowledged that Dr. Keller concluded that “participation in sex offender-specific treatment is not warranted in Mr. Muhammad’s case,” but it indicated its concern that Dr. Keller did not explain to the court’s satisfaction how she came to this conclusion. While Dr. Keller’s report lays out in some detail how she came to score Mr. Muhammad on two assessment tools, which resulted in his receiving low recidivism scores, the trial court took issue specifically with her conclusion that “[Mr. Muhammad’s] prominent risk factors would be better addressed through his participation in mental health or re-entry support groups” as opposed to sex offender-specific treatment. The court also expressed concern that Dr. Keller was not called as a witness by either party. However, the trial court did find that Dr. Keller’s forensic mental health evaluation of Mr. Muhammad’s risk of recidivism “weigh[ed] slightly in favor of granting the [IRAA] motion” under factor (c)(7), despite her absence at the evidentiary hearing. Her absence, according to the trial court, meant that “the Court did not have the opportunity to inquire of Dr. Keller about how she came to her conclusions that Mr. Muhammad, who was indicted on 45 charges related to the stalking and sexual assaulting of women, and especially who again was accused of stalking two women in 2016, would [not] benefit from this type of [sex offender] treatment.” Additionally, the trial court was concerned

16 that Dr. Keller’s conclusions were based solely on a two-hour telephone conversation with Mr. Muhammad. 9 The trial court concluded that Dr. Keller’s report “does not demonstrate to the Court that Mr. Muhammad would not be a danger upon release.”

We see no support in the record for the trial court’s dangerousness finding based on its assumption that Mr. Muhammad needed sex offender treatment. The court made repeated comments to that effect, stating, for example, that “[s]ex offender treatment can reduce an inmate’s likelihood of reoffending” and noting that “Mr. Muhammad’s lack of plans for sex offender treatment, and his failure even to recognize the need for such treatment, is not likely to provide Mr. Muhammad the support he needs to avoid reoffending if released.” On the record before us, there is insufficient basis to find that Mr. Muhammad needed sex offender treatment in order to prove his non-dangerousness under the IRAA factors. See Doe, 333 A.3d at 898.

D. The 2016 “Stalking” Infraction

Mr. Muhammad next argues that the trial court erroneously relied on his 2016 stalking infraction and its surrounding circumstances in evaluating his current

9 The trial court stated that Dr. Keller’s conclusions are “based solely on a two-hour telephone conversation with Mr. Muhammad.” However, while Dr. Keller had only one two-hour conversation, the report lists fourteen document-based sources of information that Dr. Keller also reviewed in making her assessment.

17 dangerousness under (c)(3) and (c)(5). The court, he contends, “dr[e]w an unsubstantiated analogy between the 2016 infraction and Mr. Muhammad’s underlying convictions,” based on the name of the infraction—“stalking,” without focusing on the benign nature of the infraction. Mr. Muhammad asserts that the infraction was based on his repeated filing of administrative grievances and did not relate to any sexual, violent, or physical conduct by Mr. Muhammad against staff. The fact that this conduct was related to complaints filed by two women staff members is, according to Mr. Muhammad, pure chance. Moreover, Mr. Muhammad argued that this infraction occurred more than seven years ago and was accounted for in his most recent “minimum” PATTERN score and Dr. Keller’s assessment.

We agree that the trial court incorrectly analogized this infraction to Mr. Muhammad’s prior crimes. At the IRAA hearing, the judge observed that “when I see stalking it seems to me in some ways similar to following somebody in a vehicle, smashing into a vehicle and then sexually assaulting them,” before asking for clarification concerning whether there was a sexual component to the “stalking” infraction. Appellant’s counsel was permitted to read into the record the full description of the infraction as detailed by the BOP, that is, “stalking another person through repeated behaviors which harasses, alarms, or annoys the person after

18 having been previously warned to stop such conduct.” 10 Despite this clarification, the trial court continued to perceive a connection between the infraction and Mr. Muhammad’s convictions for rape and murder, which, it noted, began with the stalking of victims, specifically women driving alone in D.C. In its review of the DHO Report, the trial court paid special attention to: (1) the derogatory language used by Mr. Muhammad in emails sent to both female staff members, (2) Mr. Muhammad’s comment that he “do[es]n’t give a damn about slapping an officer or what’s going to happen to me” following the first incident, (3) the extended duration of both instances of the “stalking” behavior over the course of several months, as opposed to deriving from a single outburst, (4) the fear of physical attack that Mr. Muhammad’s actions instilled in the victim of the second incident, and (5) the fact that these events occurred approximately thirty-four years into Mr. Muhammad’s incarceration when he was about fifty-two years old. However, the trial court did not acknowledge that the basis for the infraction was Mr. Muhammad’s repeated filing of administrative complaints, as opposed to physically stalking or attacking the women employees, and ultimately concluded that “Mr. Muhammad’s infraction

10 In addition to expounding on the “stalking” infraction, Mr. Muhammad’s counsel was also given the opportunity to address two other infractions that caused the trial court concern: a “sexual proposal” infraction and a “making threats of bodily harm” infraction. The trial court did not refer to these additional infractions in its opinion, nor did it question counsel’s explanations for why the infractions did not affect Mr. Muhammad’s dangerousness determination.

19 for ‘stalking,’ . . . raises significant concerns, especially in light of [his] convictions for two rapes, one attempted rape, one first degree murder, and one armed robbery, all of which he committed after sta[l]king the victims by driving on the beltway and stalking women who were driving alone.” (Additionally, after discussing the stalking infraction, the trial court noted “[f]urther” that “despite having been convicted of raping two different women and attempting to rape another, Mr. Muhammad has never completed any sex offender programming”; the trial court elsewhere stated that “[t]he [c]ourt is concerned that Mr. Muhammad remains a danger to the community due to his 2016 infraction for stalking two women and his lack of sex offender treatment, even though he pleaded guilty to serious sex offenses against five different women”; and the court commented at the hearing that “[the] offenses that concern me are the infractions . . . for stalking, sexual proposals, and threats and threatening bodily harm. Those are not Category 100 infractions, but they are serious ones and they relate in many ways to the offenses . . . .”)

Because we are vacating and remanding on other grounds, we note that while the trial court may permissibly consider this piece of Mr. Muhammad’s disciplinary history in its analysis, such consideration should acknowledge the circumstances of the infraction—a single, nonviolent, nonsexual infraction from 2016.

E. Standard of Proof

Finally, Mr. Muhammad argues that the trial court erroneously held him to a heightened standard of proof to demonstrate his non-dangerousness. His argument rests on three assertions: (1) that the trial court dismissed his “minimum” PATTERN score because that score did not “guarantee” that he would reoffend, (2) that the trial court dismissed Dr. Keller’s report because it was offered as a written submission unsupported by live testimony, and (3) that the trial court dismissed Mr. Muhammad’s letter to the court expressing his remorse because it was offered as a written submission and not as live testimony. We are not persuaded by Mr. Muhammad’s argument that the trial court applied an incorrect standard of proof. Moreover, as we have now made clear in Welch, which was issued after the trial court’s decision in this case, the correct standard under the IRAA is preponderance of the evidence, 319 A.3d at 977, and trial courts are encouraged to expressly acknowledge and employ this standard when making their rulings on IRAA motions.

Mr. Muhammad notes that the trial court’s opinion does not identify the standard it is applying, which, he argues, prevents this court from presuming that the correct standard was applied. Mr. Muhammad claims that while this court has not decided the burden of proof by which an IRAA petitioner must establish non-dangerousness, “there is no question that the burden is no higher than preponderance

21 of the evidence.” Mr. Muhammad bases his contention that the trial court applied a standard higher than preponderance of the evidence on a single use of the word “guarantee” in the trial court’s statement that “a PATTERN score is no guarantee of how Mr. Muhammad will behave outside of prison.” The government contends this was simply “an inartful use” that merely signifies the trial court’s reasoning that “the PATTERN score alone (after discrediting Dr. Keller and Muhammad’s letter) did not satisfy the defendant’s burden of establishing a lack of danger.”

In Welch, a similar question of standard of proof was presented to this court, on the basis of the trial court’s assessment that Mr. Welch “ha[d] not reached ‘the highest level of rehabilitation and maturity’” and on its description of his release plan as “weak.” Welch, 319 A.3d at 977. There, as here, the trial court did not specify the evidentiary standard it had applied, but both parties agreed that the correct standard was preponderance of the evidence. Id. On appeal, we determined that the trial court properly weighed the evidence and had not applied a higher standard than preponderance of the evidence. Id.; see also Doe, 333 A.3d at 900 n.6 (“The preponderance-of-the-evidence standard is the default rule and generally applies unless a statute specifies a different standard.” (citing Bailey v. United States, 251 A.3d 724, 729 (D.C. 2021) (per curiam) (applying a preponderance of the evidence standard under the compassionate release statute))). Indeed, we held that “[a]bsent any indication to the contrary, we presume that the trial judge knew the

22 proper standard of proof to apply and did in fact apply it.” Id. (quoting In re C.T., 724 A.2d 590, 597 (D.C. 1999)).

Mr. Muhammad contends that his case is more similar to Bailey, a compassionate release case, where we decided that the trial court had erroneously held the petitioner to a higher standard than preponderance of the evidence. 251 A.3d at 730-31. In Bailey, as here and in Welch, the trial court did not specifically identify the standard it was applying, and we concluded that the court had “employ[ed] language arguably implying” that the standard was higher than preponderance of the evidence. Id. at 730. However, what set Bailey apart was the trial court’s repeated statements that it “must be confident” in Mr. Bailey’s non-dangerousness, that it must “ensure that the record” demonstrated his non-dangerousness, and that it must have “sufficient comfort” in its conclusion before ordering his release. Id. As the trial court did not expressly articulate the controlling standard and because no previous holding existed applying the preponderance standard in this context, we could not “excuse [these] apparent deviations from that standard as mere imprecisions.” Id.

The instant case is distinguishable from Bailey and more akin to Welch, where the IRAA petitioner identified only two instances of language that he contended demonstrated that the trial court required a higher burden of proof. Welch, 319 A.3d

23 at 977. We did not find these two instances of language that allegedly required a higher standard to be persuasive and held that they were both “made in the context of evaluating the evidence submitted by Mr. Welch as to specific factors, not assessments of the weight of the evidence as a whole.” Id. Similarly, we do not find that a single use of the word “guarantee” here is sufficient to overcome the presumption that the trial court applied the correct standard. See Welch, 319 A.3d at 977. Mr. Muhammad further argues that the trial court showed its reliance on a higher burden of proof through its impermissible preference for spoken, as opposed to written, testimony, even though the IRAA expressly permits the inclusion of affidavits or other written material and does not require live testimony.

We are not convinced that references to outstanding questions of fact and credibility are indicative of an elevated burden of proof. However, we note that in considering whether Mr. Muhammad’s letter was sufficient to meet his burden of demonstrating maturity, rehabilitation, and a fitness to reenter society, the trial court appears to have penalized him for not testifying—after explicitly telling him he did not need to do so—given its statement that “Mr. Muhammad’s decision not to testify made it more difficult for the Court to assess whether his letter, which may have been drafted or edited by others, was an accurate depiction of his feelings.” The trial court, on remand, should consider the letter on its own terms for the information it contains in weighing this factor.

IV. Conclusion

The trial court’s dangerousness finding is based on an unsubstantiated assumption that Mr. Muhammad needed and failed to obtain sex offender treatment. In the absence of support in the record for this assumption, we conclude that the trial court failed to properly exercise its discretion. 11 As the government does not dispute that this error was not harmless, we vacate and remand the case to the trial court to reconduct its IRAA analysis consistent with this opinion. See Riley, 338 A.3d at 10; see also Wheeler v. United States, 930 A.2d 232, 246 (D.C. 2007) (holding that government has the burden of persuasion in harmless error analysis).

So ordered.

11 See Brooks v. United States, 993 A.2d 1090, 1093 (D.C. 2010) (“The trial court must make an informed choice drawn from a firm factual foundation.” (citation modified)). As the government conceded, we also conclude the trial court erred in its interests of justice analysis by relying on factors like the “seriousness of [Mr. Muhammad’s] offense” because they “were not only outside of what IRAA enumerates, but that run contrary to IRAA’s purpose.” Riley, 338 A.3d at 9.